List of Contributors
Numbers in parentheses indicate the page(s) on which the authors’ contributions begin.
Steven A...
111 downloads
3777 Views
55MB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
List of Contributors
Numbers in parentheses indicate the page(s) on which the authors’ contributions begin.
Steven A. Abrams (811)
USDA/ARS Children’s Nutrition Research Center, Houston, Texas 77030, USA John S. Adams (341, 1379)
Division of Endocrinology, Diabetes, and Metabolism, University of California, 8700 Beverly Blvd, Los Angeles, Cedars-Sinai Medical Center, Los Angeles, California 90048, USA Judith E. Adams (967)
Clinical Radiology, Imaging Science and Biomedical Engineering, Stopford Building, The University, Manchester M13 9PT, United Kingdom Luciano Adorini (631, 1511, 1833)
BioXell SpA, 20132 Milano, Italy Paul H. Anderson (711)
Institute of Medical and Veterinary Science, Adelaide, South Australia, Australia Gerald J. Atkins (711)
Hanson Institute, Adelaide, South Australia, Australia Jane E. Aubin (649)
Department of Molecular and Medical Genetics, Faculty of Medicine, University of Toronto, Toronto, Ontario M5S 1A8, Canada Isabelle Bailleul-Forestier (599)
INSERM E110, Université Paris VII, IFR58, Cordeliers Biomedical Institute, 75270 Paris Cedex 06, France Julia Barsony (363)
Laboratory of Cellular Biochemistry and Biology, NIDDK/NIH, Bethesda, Maryland 20892-0850, USA Thomas K. Barthel (219)
Department of Biochemistry and Molecular Biophysics, University of Arizona, Tucson, Arizona 85721, USA
Norman H. Bell (789)
Department of Medicine, Medical University of South Carolina, Charleston, South Carolina 29425, USA Ariane Berdal (599)
INSERM E110, Université Paris VII, IFR58, Cordeliers Biomedical Institute, 75270 Paris Cedex 06, France Joel J. Bergh (751)
Department of Biological Sciences, University of Delaware, Newark, Delaware 19716, USA Jacqueline L. Berry (1293)
University of Manchester, Vitamin D Research Group, Department of Medicine, Manchester Royal Infirmary, Manchester M13 9WL, UK Daniel D. Bikle (609)
Endocrine Research Unit, Veterans’ Affairs Medical Center, University of California-San Francisco, San Francisco, California 94121-1598, USA John P. Bilezikian (1355)
Department of Medicine, College of Physicians and Surgeons, Columbia University, New York, New York 10032, USA Ernst Binderup (1489)
Biological Research, Leo Pharma, DK-2750 Ballerup, Denmark Lise Binderup (1489)
Biological Research, Leo Pharma, DK-2750 Ballerup, Denmark Nicholas J. Bishop (803)
Academic Department of Child Health, University of Sheffield, Sheffield Children’s Hospital, Sheffield S10 2TH, United Kingdom Ilse Bogaerts (135)
Laboratory Analytische Chemie, Van Evenstraat 4; B-3000 Leuven, Belgium
xiv Ricardo L. Boland (883)
Department de Biología, Bioquímica & Farmacía, Universidad Nacional del Sur, San Juan 670, (8000) Bahía Blanca, Argentina Adele L. Boskey (477)
Mineralized Tissues Laboratory, Hospital for Special Surgery, Affiliated with Weil College of Cornell Medical School, New York, New York 10021, USA Roger Bouillon (135, 429, 1763)
Legendo, Onderwijs en Navorsing 902, U.Z. Gasthuisberg, Herestraat 49, B-3000 Leuven, Belgium Barbara D. Boyan (575)
Department of Biomedical Engineering at Georgia Tech and Emory University, Georgia Institute of Technology, Atlanta, Georgia 30332, USA Philippe Brachet (1779)
INSERM U 643, Centre Hospitalier Universitaire, 30 bd Jean Monnet, 44093 Nantes, France Alex J. Brown (1313, 1449)
Renal Division, Washington University School of Medicine, St. Louis, Missouri 63110, USA Edward M. Brown (551)
Division of Endocrinology, Diabetes and Hypertension, Brigham and Women’s Hospital, Harvard Medical School, Boston, Massachusetts 02115, USA Carsten Carlberg (313)
Department of Biochemistry, University of Kuopio, PFIN-70211 Kuopio, Finland Geert Carmeliet (429)
Legendo, Onderwijs en Navorsing 902, U.Z. Gasthuisberg, Herestraat 49, B-3000 Leuven, Belgium Thomas O. Carpenter (1049)
Department of Pediatrics, Yale University, School of Medicine, New Haven, Connecticut 06520-8064, USA Marie-Claire Chapuy (1085)
INSERM Unit 403, Faculty Laennec and Department of Rheumatology and Bone Disease, Edouard Herriot Hosptial, Lyon, France Fredriech K. W. Chan (1355)
Department of Medicine, Queen Elizabeth Hospital, Hong Kong Tai C. Chen (1599)
Vitamin D, Skin, and Bone Research Laboratory, Boston University Medical Center, Boston, Massachusetts 02188, USA Sylvia Christakos (721)
Department of Biochemistry and Molecular Biology, University of Medicine and Dentistry of New Jersey, New Jersey Medical School, Newark, New Jersey 07103, USA
LIST
OF
CONTRIBUTORS
Margaret Clagett-Dame (1543)
Department of Biochemistry, University of Wisconsin–Madison, Madison, Wisconsin 53706, USA Thomas L. Clemens (899)
Department of Cell Biology and Physiology, Department of Internal Medicine, University of Cincinnati School of Medicine, Cincinnati, Ohio 45276, USA Je-Yong Choi (327)
Department of Biochemistry, Kyungpook National University, Daegu, Korea Fredric L. Coe (1339)
Nephrology Section, The University of Chicago Pritzker School of Medicine, Chicago, Illinois 60637, USA Kay Colston (1663)
Department OGEM, St. George’s Hospital Medical School, London, SW17 0RE, United Kingdom Juliet E. Compston (951)
Department of Medicine, Level 5, University of Cambridge, School of Clinical Medicine, Addenbrooke’s Hospital, Cambridge, England CB2 2QQ, United Kingdom Nancy E. Cooke (117)
Department of Genetics and Medicine, University of Pennsylvania, Philadelphia, Pennsylvania 19104-6149, USA Clara Crescioli (1833)
Endocrinology Unit, Department of Clinical Physiopathology, University of Florence, Florence 50139, Italy Heide S. Cross (1709)
Department of Pathophysiology, Medical, University of Vienna, A-1090 Vienna, Währingergürtel 18-20, Austria Michael Danilenko (1635)
Department of Clinical Biochemistry, Faculty of Health Science, Ben-Gurion University of the Negev, Beer-Sheva 84105, Israel Jean-Luc Davideau (599)
INSERM E110, Université Paris VII, IFR58, Cordeliers Biomedical Institute, 75270 Paris Cedex 06, France Michael Davies (1293)
Vitamin D Research Group, University Department of Medicine, Manchester Royal Infirmary, Manchester, M13 9WL, United Kingdom Hector F. DeLuca (3, 1543)
Department of Biochemistry, University of Wisconsin–Madison, Madison, Wisconsin 53706, USA Marie B. Demay (341)
Endocrine Unit, Department of Medicine, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts 02114, USA
LIST
OF
xv
CONTRIBUTORS
Puneet Dhawan (721)
Department of Biochemistry and Molecular Biology, University of Medicine and Dentistry of New Jersey, New Jersey Medical School, Newark, New Jersey 07103, USA Carlos Encinas Dominguez (219)
Department of Biochemistry and Molecular Biophysics, University of Arizona, Tucson, Arizona 85721, USA Diane R. Dowd (291)
Department of Pharmacology–W334, Case Western Reserve University, Cleveland, Ohio 44106, USA Marc K. Drezner (1159)
Department of Medicine, Endocrinology, Diabetes, and Metabolism Section, University of Wisconsin–Madison, Madison, Wisconsin 53792, USA Thomas W. Dunlop (313)
Department of Biochemistry, University of Kuopio, PFIN-70211 Kuopio, Finland Adriana S. Dusso (1313)
Renal Division, Washington University School of Medicine, St. Louis, Missouri 63110, USA Richard Eastell (1101)
University of Sheffield Clinical Sciences Centre, Northern General Hospital, Sheffield South Yorkshire S5 7AU, United Kingdom Michael J. Econs (1189)
Indiana University School of Medicine, Department of Medicine and Medical and Molecular Genetics, Indianapolis, Indiana 46202, USA John Eisman (193)
Bone and Mineral Program, Garvan Institute of Medical Research, St. Vincent’s Hospital, Sydney, New South Wales 2010, Australia Sol Epstein (1253)
Division of Endocrinology, Diabetes, and Bone Diseases, Department of Medicine, Mount Sinai School of Medicine, New York, New York 10029, USA Erik Fink Eriksen (1805)
Osteoporosis Team of Lilly Research Laboratories, Lilly Corp Center, Indianapolis, Indiana 46285, USA Luis M. Esteban (193)
Bone and Mineral Program, Garvan Institute of Medical Research, St. Vincent’s Hospital, Sydney, New South Wales 2010, Australia Dan Faibish (477)
Mineralized Tissues Laboratory, Hospital for Special Surgery, Affiliated with Weil College of Cornell Medical School, New York, New York 10021, USA
Yue Fang (1121)
Department of Internal Medicine, Genetic Laboratory, Erasmus Medical Centre, NL-3015 GE Rotterdam, The Netherlands Mary C. Farach-Carson (751)
Department of Biological Sciences, University of Delaware, Newark, Delaware 19716, USA Murray J. Favus (1339)
Section of Endocrinology, University of Chicago, Chicago, Illinois 60637, USA David Feldman (1207, 1679)
Stanford University School of Medicine, Division of Endocrinology, Stanford, California 94305-5103, USA David Findlay (711)
Department of Orthopedic Surgery and Trauma, University of Adelaide, Adelaide 5000, South Australia, Australia Christian Frank (313)
Department of Biochemistry, University of Kuopio, PFIN-70211 Kuopio, Finland Leonard P. Freedman (263)
Department of Molecular Endocrinology and Bone Biology, Merck Research Laboratories, West Point, Pennsylvania 19486-0004, USA Ryuji Fujiki (305)
University of Tokyo, Institute of Molecular and Cellular Biosciences, 1-1-1 Yayoi-cho, Bunkyo-ku, Tokyo, 113-0032, Japan Masafumi Fukagawa (1821)
Division of Nephrology and Dialysis Center, Kobe University School of Medicine, 7-5-2 Kusunoki-cho, Chuo-ku, Kobe 650-0017, Japan Robert F. Gagel (687)
Section of Endocrinology, University of Texas M.D. Anderson Cancer Center, Houston, Texas, USA Emmanuel Garcion (1779)
INSERM U 646, 10 rue André Boquel, 49100 Angers, France Edith M. Gardiner (193)
Bone and Mineral Program, Garvan Institute of Medical Research, St. Vincent’s Hospital, Sydney, New South Wales 2010, Australia Marielle Gascon-Barré (47)
Département de Pharmacologie, Faculté de Médecine, Université de Montréal, and Centre de recherche de l’Université de Montréal, Montréal, Québec H2X 1P1, Canada Edward Giovannucci (1617)
Harvard School of Public Health, Department of Nutrition, Boston, Massachusetts 02115, USA Henning Glerup (1805)
Aarhus Kommunehospital, Dept V, Noerrebrogade 44, DK-8000 Aarhus C, Denmark
xvi Francis H. Glorieux (1197)
Genetics Unit, Shriners Hospital for Children, Departments of Surgery, Pediatrics, and Human Genetics, McGill University, Montréal, Québec H3G 1A6, Canada Wagn O. Godtfredsen (1489)
Medicinal Chemistry, Leo Pharma, DK-2750 Ballerup, Denmark David Goltzman (737)
Department of Medicine, McGill University and McGill University Health Center, Montréal, Québec H3A 1A1, Canada Soraya Gutierrez (327)
Departamento de Biología Molecular, Universidad de Concepción, Concepción, Chile Conny Gysemans (1763)
Legendo, Onderwijs en Navorsing 902, U.Z. Gasthuisberg, Herestraat 49, B-3000 Leuven, Belgium Bernard P. Halloran (823)
Division of Endocrinology, Veterans Affairs Medical Center, San Francisco, California 94121, USA Carol A. Haussler (219)
Department of Biochemistry and Molecular Biophysics, University of Arizona, Tucson, Arizona 85721, USA Mark R. Haussler (219)
Department of Biochemistry and Molecular Biophysics, University of Arizona, Tucson, Arizona 85721, USA Robert P. Heaney (773)
Creighton University, Omaha, Nebraska 68131, USA Johan Heersche (649)
Faculty of Dentistry, University of Toronto, Toronto M5S 1A8, Ontario, Canada Helen L. Henry (69)
Department of Biochemistry, University of California–Riverside, Riverside, California 92521, USA Pamela A. Hershberger (1741)
Department of Pharmacology, University of Pittsburgh, Pittsburgh, Pennsylvania 15213, USA Martin Hewison (1379)
Division of Medical Sciences, The University of Birmingham, Queen Elizabeth Medical Centre, Birmingham, B15 2TH, United Kingdom Richard A. Heyman (1557)
X-Ceptor Therapeutics, San Diego, California 92121, USA Kanji Higashio (665)
Research Center for Genomic Medicine, Saitama Medical School, Saitama, 350-1241, Japan
LIST
OF
CONTRIBUTORS
Michael F. Holick (37, 1511, 1791)
Vitamin D, Skin, and Bone Research Laboratory; Department of Medicine; Endocrinology, Nutrition and Diabetes Section; Boston Medical Center and Boston University School of Medicine, Boston, Massachusetts 02118, USA Bruce W. Hollis (931)
Departments of Pediatrics, Biochemistry, and Molecular Biology, Medical University of South Carolina, Charleston, South Carolina 29425, USA Ronald L. Horst (15)
U.S. Department of Agriculture, Agricultural Research Service, National Animal Disease Center, Ames, Iowa 50010-0070, USA Jui-Cheng Hsieh (219)
Department of Biochemistry and Molecular Biophysics, University of Arizona, Tucson, Arizona 85721, USA Karl L. Insogna (1049)
Department of Medicine, Yale University School of Medicine, New Haven, CT 06520-8064, USA Elizabeth T. Jacobs (219)
College of Medicine, University of Arizona, Tucson, Arizona 85721, USA Amjad Javed (327)
Department of Cell Biology, University of Massachusetts Medical School, Worcester, Massachusetts 01655, USA Glenville Jones (1423)
Department of Biochemistry, Queen’s University, Kingston, Ontario K7L 3N6, Canada Candace S. Johnson (1741)
Department of Pharmacology & Therapeutics, Roswell Park Cancer Institute, Buffalo, New York 14263, USA Peter W. Jurutka (219)
Department of Biochemistry and Molecular Biophysics, University of Arizona, Tucson, Arizona 85721, USA Mehmet Kahraman (1405)
Department of Chemistry, Johns Hopkins University, Baltimore, Maryland 21218-2685, USA S. Kaleem Zaidi (327)
Department of Cell Biology, University of Massachusetts Medical School, Worcester, Massachusetts 01655, USA Heidi J. Kalkwarf (839)
Division of General and Community Pediatrics, Children’s Hospital Medical Center, Cincinnati, Ohio 45229, USA
LIST
OF
xvii
CONTRIBUTORS
Shigeaki Kato (305)
University of Tokyo, Institute of Molecular and Cellular Biosciences, 1-1-1 Yayoi-cho, Bunkyo-ku, Tokyo, 113-0032, Japan Anne-Marie Kissmeyer (1489)
Biological Research, Leo Pharma, DK-2750 Ballerup, Denmark Hirochika Kitagawa (305)
University of Tokyo, Institute of Molecular and Cellular Biosciences, 1-1-1 Yayoi-cho, Bunkyo-ku, Tokyo, 113-0032, Japan Lilia M. C. Koberle (1355)
Health Sciences Department, Federal University, Sao Carlos, Brazil H. Phillip Koeffler (1727)
Hematology/Oncology Division, Cedars-Sinai Medical Center, Los Angeles, California 90048, USA Ruth Koren (761)
Felsenstein Medical Research Center, Beilinson Campus, Rabin Medical Center, Petah Tikva 49100, Israel Barbara E. Kream (703)
Department of Medicine, University of Connecticut Health Center, Farmington, Connecticut 06030-1850, USA Richard Kremer (737)
Department of Medicine, McGill University and McGill University Health Center, Montréal, Québec H3A 1A1, Canada Aruna V. Krishnan (1679)
Stanford University School of Medicine, Division of Endocrinology, Stanford, California 94305-5103, USA Noboru Kubodera (1525)
Department of Product Planning, Chugai Pharmaceutical Co. Ltd., 2-1-9 Kyobashi, Chuo-ku, Tokyo, 104-8301, Japan Rajiv Kumar (515)
Departments of Medicine, Biochemistry and Molecular Biology and Mayo Proteomics Research Center, Divisions of Nephrology, and Endocrinology, Diabetes, Metabolism, and Nutrition, Mayo Clinic and Foundation, Rochester, Minnesota 55905-0002, USA Kiyoshi Kurokawa (1821)
Research Center for Advanced Science and Technology, The University of Tokyo, Tokyo, Japan Christopher J. Laing (117)
Department of Genetics and Medicine, University of Pennsylvania, Philadelphia, Pennsylvania 19104-6144, USA Jacques Lemire (1753)
Pediatric Nephrology, University of California-San Diego, La Jolla, California 92093-0831, USA
Frédéric Lézot (599)
INSERM E110, Université Paris VII, IFR58, Cordeliers Biomedical Institute, 75270 Paris Cedex 06, France Yan Chun Li (721, 871, 1511)
Department of Medicine/GI Section, University of Chicago, Chicago, Illinois 60637, USA Jane B. Lian (327)
Department of Cell Biology, University of Massachusetts Medical School, Worcester, Massachusetts 01655, USA Alexander C. Lichtler (703)
Department of Genetics and Developmental Biology, The University of Connecticut Health Center, Farmington, Connecticut 06030, USA Paul Lips (1019)
Department of Endocrinology, Vrijie University Medical Center, Amsterdam, 1007 MB, The Netherlands Yan Liu (721)
Department of Biochemistry and Molecular Biology, New Jersey Medical School, Newark, New Jersey 07103, USA Paul MacDonald (291)
Department of Pharmacology–W334, Case Western Reserve University, Cleveland, Ohio 44106, USA Hubert Maehr (1511)
BioXell, Inc., Hoffmann-La Roche, Inc., Nutley, New Jersey 07110-1199, USA Mario Maggi (1833)
Andrology Unit, Department Clinical Physiopathology, University of Florence, 50139 Florence, Italy Peter J. Malloy (1207)
Stanford University School of Medicine, Division of Endocrinology, Stanford, California 94305-5103, USA David J. Mangelsdorf (863)
Department of Pharmacology, Howard Hughes Medical Institute, University of Texas Southwestern Medical Center at Dallas, Dallas, Texas 75390-9050, USA Chantal Mathieu (1763)
Legendo, Onderwijs en Navorsing 902, U.Z. Gasthuisberg, Herestraat 49, B-3000 Leuven, Belgium Brian May (85)
School of Molecular and Biomedical Science, University of Adelaide, South Australia 5005, Australia Andrew P. Mee (1293)
Vitamin D Research Group, University Department of Medicine, Manchester Royal Infirmary, Manchester, M13 9WL, United Kingdom
xviii Pierre J. Meunier (1085)
INSERM Unit 403, Faculty Laennec and Department of Rheumatology and Bone Disease, Edouard Herriot Hospital, Lyon, France Toshimi Michigami (851)
Department of Environmental Medicine, Osaka Medical Center and Institute for Maternal and Child Health, Osaka, Japan Martin Montecino (327)
Departamento de Biología Molecular, Universidad de Concepción, Concepción, Chile Dino Moras (279)
Département de Biologie et de Génomique Structurales, CNRS/INSERM/Université Louis Pasteur 1, BP 10142, 67404 Illkirch Cedex, France Roberta Morosetti (1727)
Pediatric Oncology Division, Catholic University of Rome, Rome, Italy Howard A. Morris (711)
Hanson Institute, Adelaide, South Australia, Australia Daniel L. Motola (863)
Department of Pharmacology, Howard Hughes Medical Institute, University of Texas Southwestern Medical Center at Dallas, Dallas, Texas 75390-9050, USA Josephia Muindi (1741)
Department of Medicine, Roswell Park Cancer Institute, Buffalo, New York 14263, USA Shigeo Nakajima (851)
Department of Developmental Medicine (Pediatrics), D-5, Osaka University Graduate School of Medicine, Osaka 565-0871, Japan Tally Naveh-Many (537)
Minerva Center for Calcium and Bone Metabolism, Hebrew University Hadassah Medical Center, Ein Karem, Jerusalem, 91120, Israel Philippe Naveilhan (1779)
INSERM U 643, Centre Hospitalier Universitaire, 30 bd Jean Monnet, 44093 Nantes, France Isabelle Neveu (1779)
INSERM U 643, Centre Hospitalier Universitaire, 30 bd Jean Monnet, 44093 Nantes, France Anthony W. Norman (381)
Department of Biochemistry, University of California, Riverside, Riverside, California 92521-0129, USA Anders Nykjaer (153)
Institute of Medical Biochemistry, University of Aarhus, Ole Worms Allee, DK-8000 Aarhus C, Denmark James O’Kelly (1727)
Hematology/Oncology Division, Cedars-Sinai Medical Center, Los Angeles, California 90048, USA
LIST
OF
CONTRIBUTORS
John L. Omdahl (85)
Office of Research, University of New Mexico School of Medicine, Albuquerque, New Mexico 87131-5166, USA Peter Ordentlich (1557)
X-Ceptor Therapeutics, San Diego, California 92121, USA Keiichi Ozono (851)
Department of Developmental Medicine (Pediatrics), D-5, Osaka University Graduate School of Medicine, Osaka 565-0871, Japan A. Michael Parfitt (497, 1029)
Division of Endocrinology and Center for Osteoporosis and Metabolic Bone Disease, University of Arkansas for Medical Science, Little Rock, Arkansas 72205, USA Donna M. Peehl (1679)
Stanford University School of Medicine, Division of Endocrinology, Stanford, California 94305-5103, USA Sara Peleg (1471)
Department of Endocrine Neoplasia and Hormonal Diseases, Unit 435, University of Texas, M.D. Anderson Cancer Center, Houston, Texas 77030-4009, USA Xiaorong Peng (721)
Department of Biochemistry and Molecular Biology, University of Medicine and Dentistry of New Jersey, New Jersey Medical School, Newark, New Jersey 07103, USA John M. Pettifor (1065)
Department of Pediatrics, Chris Hani Baragwanath Hospital, Mineral Metabolism Research Unit, P.O. Bertsham, Johannesburg, Gauteng 2013, South Africa J. Wesley Pike (167, 1207, 1403, 1543)
Department of Biochemistry, University of Wisconsin–Madison, Madison, Wisconsin 53706, USA Elizabeth A. Platz (1617)
Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland 21205, USA Lori A. Plum (1543)
Department of Biochemistry, University of Wisconsin–Madison, Madison, Wisconsin 53706, USA Shirwin Pockwinse (327)
Department of Cell Biology, University of Massachusetts Medical School, Worcester, Massachusetts 01655, USA Huibert A.P. Pols (1121, 1571)
Department of Internal Medicine, Erasmus Medical Centre, NL-3015 GD Rotterdam, The Netherlands
LIST
OF
xix
CONTRIBUTORS
Anthony A. Portale (453, 823)
Department of Pediatrics, University of California – San Francisco, San Francisco, California 94121, USA Gary H. Posner (1405)
Department of Chemistry, Johns Hopkins University, Baltimore, Maryland 21218-2685, USA Mehrdad Rahmaniyan (789)
Department of Medicine, Medical University of South Carolina, Charleston, South Carolina 29425, USA Amiram Ravid (761)
Felsenstein Medical Research Center, Beilinson Campus, Rabin Medical Center, Petah Tikva 49100, Israel G. Satyanarayana Reddy (15, 1511)
Brown University, Department of Chemistry, Providence, Rhode Island, USA Jörg Reichrath (1791)
The Saarland University Hospital, Department of Dermatology, Kirrberger Str., 66421 Homburg/Saar, Germany Timothy A. Reinhardt (15)
U.S. Department of Agriculture, Agricultural Research Service, National Animal Disease Center, Ames, Iowa 50010-0070, USA Alfred A. Reszka (263)
Department of Molecular Endocrinology and Bone Biology, Merck Research Laboratories, West Point, Pennsylvania 19486-0004, USA B. Lawrence Riggs (1101)
Division of Endocrinology, Mayo Clinic and Mayo Foundation, Rochester, Minnesota 55905, USA Natacha Rochel (279)
Département de Biologie et de Génomique Structurales, CNRS/INSERM/Université Louis Pasteur 1, BP 10142, 67404 Illkirch Cedex, France Mishaela R. Rubin (1355)
Department of Medicine, College of Physicians and Surgeons, Columbia University, New York, New York 10027, USA Andrew F. Russo (687)
Department of Physiology and Biophysics, University of Iowa, Iowa City, Iowa 52242, USA Philip Sambrook (1239)
Institute of Bone & Joint Research, University of Sydney, Royal North Shore Hospital, Sydney 2065, Australia Adina E. Schneider (1253)
Division of Endocrinology, Diabetes, and Bone Diseases, Department of Medicine, Mount Sinai School of Medicine, New York, New York 10029, USA
Gary G. Schwartz (1599)
Comprehensive Cancer Center, Wake Forest University School of Medicine, Winston-Salem, North Carolina 27157, USA Zvi Schwartz (575)
Department of Periodontics, Hebrew University Hadassah Faculty of Dental Medicine, Jerusalem, Israel Jiali Shen (327)
Department of Cell Biology, University of Massachusetts Medical School, Worcester, Massachusetts 01655, USA Nirupama K. Shevde (167, 1543)
Department of Biochemistry, University of Wisconsin–Madison, Madison, Wisconsin 53706, USA Rafal R. Sicinski (1543)
Department of Chemistry, University of Warsaw, ul. L. Pasteura 2, Warsaw 02-093, Poland Justin Silver (537)
Minerva Center for Calcium and Bone Metabolism, Hebrew University Hadassah Medical Center, Ein Karem, Jerusalem, 91120, Israel Eduardo A. Slatopolsky (1313, 1449)
Renal Division, Washington University School of Medicine, St. Louis, Missouri 63110, USA Bonny L. Specker (839)
Martin Program in Human Nutrition, South Dakota State University, Brookings, South Dakota 57007, USA René St-Arnaud (105, 1197)
Genetics Unit, Shriners Hospital for Children, Montréal, Quebec H3G 1A6, Canada Gary S. Stein (327)
Department of Cell Biology, University of Massachusetts Medical School, Worcester, Massachusetts 01655, USA Janet L. Stein (327)
Department of Cell Biology, University of Massachusetts Medical School, Worcester, Massachusetts 01655, USA Paula H. Stern (565)
Department of Molecular Pharmacology, Northwestern University Feinberg School of Medicine, Chicago, Illinois 60611, USA George P. Studzinski (1635)
UMD-New Jersey Medical School, Newark, New Jersey 07103, USA Tatsuo Suda (665)
Research Center for Genomic Medicine, Saitama Medical School, Hidaka-shi, Saitama 350-1241, Japan
xx
LIST
Amelia L.M. Sutton (291)
Department of Pharmacology–W334, Case Western Reserve University, Cleveland, Ohio 44106, USA Peter Tebben (515)
Departments of Medicine, Biochemistry and Molecular Biology and Mayo Proteomics Research Center, Divisions of Nephrology, and Endocrinology, Diabetes, Metabolism, and Nutrition, Mayo Clinic and Foundation, Rochester, Minnesota 55905-0002, USA Harriet S. Tenenhouse (453)
Departments of Pediatrics and Human Genetics, McGill University and Montréal Children’s Hospital Research Institute, Montréal, Québec, H3Z 2Z3, Canada Michelle L. Thatcher (219)
Department of Biochemistry and Molecular Biophysics, University of Arizona, Tucson, Arizona 85721, USA Susan Thys-Jacobs (1355)
Department of Medicine, College of Physicians and Surgeons, Columbia University, New York, New York 10027, USA Dwight A. Towler (899)
Department of Medicine, Washington University School of Medicine, St. Louis, Missouri 63110, USA Donald L. Trump (1741)
Department of Medicine, Roswell Park Cancer Institute, Buffalo, New York 14263, USA André G. Uitterlinden (1121)
Genetic Laboratory, Department of Internal Medicine, Erasmus Medical Centre, NL-3015 GD Rotterdam, The Netherlands Milan R. Uskokovi´c (1511)
BioXell, Inc., Nutley, New Jersey 07110-1199, USA Sami Väisänen (313)
Department of Biochemistry, University of Kuopio, PFIN-70211 Kuopio, Finland Hugo Van Baelen (135)
Legendo, Onderwijs en Navorsing 902, U.Z. Gasthuisberg, Herestraat 49, B-3000 Leuven, Belgium Sophie Van Cromphaut (429)
Legendo, Onderwijs en Navorsing 902, U.Z. Gasthuisberg, Herestraat 49, B-3000 Leuven, Belgium Johannes P.T.M. van Leeuwen (1571)
Department of Internal Medicine, Erasmus MC, 3000 DR Rotterdam, The Netherlands
OF
CONTRIBUTORS
Joyce B.J. van Meurs (1121)
Genetic Laboratory, Department of Internal Medicine, Erasmus Medical Centre, NL-3015 GD Rotterdam, The Netherlands Andre J. van Wijnen (327)
Department of Cell Biology, University of Massachusetts Medical School, Worcester, Massachusetts 01655, USA Christel Verboven (135)
Laboratory Analytische Chemie, Van Evenstraat 4; B-3000 Leuven, Belgium Reinhold Vieth (995)
Pathology and Laboratory Medicine, Mount Sinai Hospital, Toronto, Ontario, M5G 1X5, Canada Robert H. Wasserman (411)
Department of Biomedical Sciences, VRT-08-20, Cornell University, Ithaca, New York 14853, USA JoEllen Welsh (1663)
Department Biological Science, University Notre Dame, Notre Dame, Indiana 46556, USA G. Kerr Whitfield (219)
Department of Biochemistry and Molecular Biophysics, University of Arizona, Tucson, Arizona 85721, USA Michael P. Whyte (913)
Center for Metabolic Bone Disease and Molecular Research, Shriners Hospital for Children, St. Louis, Missouri 63131, USA and Division of Bone and Mineral Diseases, Washington University School of Medicine at Barnes-Jewish Hospital, St. Louis, Missouri 63110, USA. Thomas Willnow (153)
Division of Molecular Cardiovascular Research, Max-Delbrueck-Center for Molecular Medicine, Robert-Roessle-Strasse 10, D-13125 Berlin, Germany Didier Wion (1779)
INSERM U318, Centre Hospitalier Michallon, 38043 Grenoble cedex 09, France Hisataka Yasuda (665)
Center for Experimental Medicine, Institute of Medical Science, University of Tokyo, Tokyo 108-8639, Japan Daniel Young (327)
Department of Cell Biology, University of Massachusetts Medical School, Worcester, Massachusetts 01655, USA Chi Zhang (291)
Department of Pharmacology–W334, Case Western Reserve University, Cleveland, Ohio 44106, USA
Preface to the 2nd Edition
Those interested in the vitamin D field will not be surprised that this second edition is considerably larger than the first edition. A great deal of progress has been made since the first edition was published in 1997. However, our goal in planning this updated version remains the same. We have endeavored to provide investigators, clinicians, and students with a comprehensive, definitive, and up-to-date compendium of the diverse scientific and clinical aspects of vitamin D, each area covered by experts in the field. Our hope for the second edition is that this book will continue to serve as both a resource for current researchers, as well as a guide to assist those in related disciplines to enter the realm of vitamin D research. We hope that this book will illuminate the vitamin D field and help investigators identify areas where new research is needed as well as educate them about what is currently known. We believe that the first edition succeeded in stimulating interactions between researchers and clinicians from different disciplines and that it facilitated collaborations. As we move from basic science and physiology to the use of vitamin D and its analogs as pharmacological agents to treat various diseases, the need for cross-collaborations between researchers and clinicians from different disciplines will increase. We hope that this new volume will continue to be a valuable resource that plays a role in this advancement and stimulates and facilitates these interactions. Enormous progress in the study of vitamin D has been made in the approximately eight years since the first edition was written and we hope that this book has contributed in some way to this progress. The first edition proved to be highly valuable to its readers and the chapters have been cited frequently as authoritative reviews of the field. However, it became clear to us that the time was ripe to organize a second edition. Building on the original, we hope this second edition
will incorporate all of the progress made in the field since the first edition was published so that our objective of an up-to-date compendium containing everything you wanted to know about vitamin D will continue. The second edition is essentially a new and reinvigorated book. We have changed the symbol on the cover to reflect its updated content and the field's continued evolution into the molecular world. This new edition now includes 104 chapters. In order to cover the growth of new information on vitamin D, we have had to publish this new edition in two volumes. In addition, the book has undergone some major remodeling. There are 33 completely new chapters and 18 other chapters have had major changes in authorship and are totally rewritten. While approximately half of the chapters have some of the same authors, all have had major updates and many have new co-authors with new perspectives. We have endeavored to attract the leading investigators in each field to author the chapter covering their area. We are especially pleased with the roster of authors who have written for the second edition. They really are the leaders in their respective fields. We wish to express our thanks to Tari Paschall, Judy Meyer, and Sarah Hajduk as well as the rest of the Elsevier/Academic Press staff for their expertise and indispensable contribution to bringing this revised edition to fruition. Most of all, we thank the authors for their contributions. We hope that our readers will find this updated volume useful and informative and that it will contribute to the burgeoning growth of the vitamin D field. DAVID FELDMAN J. WESLEY PIKE FRANCIS H. GLORIEUX
Preface to the 1st Edition
Our reasons for deciding to publish an entire book devoted to vitamin D can be found in the rapid and extensive advances currently being made in this important field of research. Enormous progress in investigating many aspects of vitamin D, from basic science to clinical medicine, has been made in recent years. The ever-widening scope of vitamin D research has created new areas of inquiry so that even workers immersed in the field are not fully aware of the entire spectrum of current investigation. Our goal in planning this book was to bring the diverse scientific and clinical fields together in one definitive and up-to-date volume. It is our hope that this compendium on vitamin D will serve as both a resource for current researchers and a guide to stimulate and assist those in related disciplines to enter this field of research. In addition, we hope to provide clinicians and students with a comprehensive source of information for the varied and extensive material related to vitamin D. The explosion of information in the vitamin D sphere has led to new insights into many different areas, and in our treatment of each subject in this book we have tried to emphasize the recent advances as well as the established concepts. The classic view of vitamin D action, as a hormone limited to calcium metabolism and bone homeostasis, has undergone extensive revision and amplification in the past few years. We now know that the vitamin D receptor (VDR) is present in most tissues of the body and that vitamin D actions, in addition to the classic ones, include important effects on an extensive array of other target organs. To cover this large number of subjects, we have organized the book in the following manner: Section I, the enzymes involved in vitamin D metabolism and the activities of the various metabolites; Section II, the mechanism of action of vitamin D, including rapid, nongenomic actions and the role of the VDR in health and disease; Section III, the
effects of vitamin D and its metabolites on the various elements that constitute bone and the expanded understanding of vitamin D actions in multiple target organs, both classic and nonclassic; Section IV, the role of vitamin D in the physiology and regulation of calcium and phosphate metabolism and the multiplicity of hormonal, environmental, and other factors influencing vitamin D metabolism and action; and Sections V and VI, the role of vitamin D in the etiology and treatment of rickets, osteomalacia, and osteoporosis and the pathophysiological basis, diagnosis, and management of numerous clinical disorders involving vitamin D. The recent recognition of an expanded scope of vitamin D action and the new investigational approaches it has generated were part of the impetus for developing this volume on vitamin D. It has become clear that in addition to the classic vitamin D actions, a new spectrum of vitamin D activities that include important effects on cellular proliferation, differentiation, and the immune system has been identified. This new information has greatly expanded our understanding of the breadth of vitamin D action and has opened for investigation a large number of new avenues of research that are covered in Sections VII and VIII of this volume. Furthermore, these recently recognized nonclassic actions have led to a consideration of the potential application of vitamin D therapy to a range of diseases not previously envisioned. This therapeutic potential has spawned the search for vitamin D analogs that might have a more favorable therapeutic profile, one that is less active in causing hypercalcemia and hypercalciuria while more active in a desired application such as inducing antiproliferation, prodifferentiation, or immunosuppresion. Since 1α,25-dihydroxyvitamin D [1,25(OH)2D] and its analogs are all presumed to act via a single VDR, a few years ago most of us probably would have thought that it was impossible to achieve a
xxiv separation of these activities. Yet today, many analogs that exhibit different profiles of activity relative to 1,25(OH)2D have already been produced and extensively studied. The development of analogs with an improved therapeutic index has opened another large and complex area of vitamin D research. This work currently encompasses three domains: (1) the design and synthesis of vitamin D analogs exhibiting a separation of actions with less hypercalcemic and more antiproliferative or immunosuppressive activity, (2) the interesting biological question of the mechanism(s) by which these analogs achieve their differential activity, and (3) the investigation of the potential therapeutic applications of these analogs to treat various disease states. These new therapeutic applications, from psoriasis to cancer, from immunosuppression to neurodegenerative diseases, have drawn into the field an expanded population of scientists and physicians interested in vitamin D. Our goal in editing this book was to create a comprehensive resource on vitamin D that would be of use to a mix of researchers in different disciplines. To achieve this goal, we sought authors who had contributed greatly to their respective fields of vitamin D research. The book has a large number of chapters to accommodate
PREFACE
TO THE
1ST EDITION
many contributors and provide expertise in multiple areas. Introductory chapters in each section of the book are designed to furnish an overview of that area of vitamin D research, with other chapters devoted to a narrowly focused subject. Adjacent and closely related subjects are often covered in separate chapters by other authors. While this intensive style may occasionally create some redundancy, it has the advantage of allowing the reader to view the diverse perspectives of the different authors working in overlapping fields. In this regard, we have endeavored to provide many crossreferences to guide the reader to related information in different chapters. We express our thanks to Jasna Markovac (Editor-inChief), for encouraging us to develop this book and guiding us through the process; to Tari Paschall (Acquisitions Editor) and the Academic Press staff for their diligence, expertise, and patience in helping us complete this work. Most of all, we thank the authors for their contributions that have made this book possible. DAVID FELDMAN J. WESLEY PIKE FRANCIS H. GLORIEUX
Abbreviations
AA AC ACE ACF ACTH ADH ADHR ADP AHO AI AIDS Aj.AR ALP ANG II ANP APC APD AR ARC ATP ATRA AUC Bmax BARE bFGF BFU BGP BLM BMC BMD BMI BMP BMU bp BPH
arachadonic acid adenyl cyclase angiotensin converting enzyme activation frequency adrenocorticotropin antidiuretic hormone (vasopressin) autosomal dominant hypophosphatemic rickets adenosine diphosphate Albright’s hereditary osteodystrophy adequate intake acquired immunodeficiency syndrome adjusted apposition rate alkaline phosphatase angiotensin II atrial natriuretic peptide antigen presenting cell aminohydroxypropylidene bisphosphonate androgen receptor activator recruited cofactor adenosine triphosphate all-trans-retinoic acid area under the curve maximum number of binding sites bile acid response element basic fibroblast growth factor burst-forming unit bone Gla protein (osteocalcin) basal lateral membrane bone mineral content bone mineral density body mass index bone morphogenetic protein basic multicellular unit base pairs benign prostatic hyperplasia
BSA BUA [Ca2+]i
bovine serum albumin bone ultrasound attentuation internal calcium ion molar concentration CaBP calcium binding protein CAD coronary artery disease CaM calmodulin cAMP cyclic AMP CaSR or CaR calcium sensing receptor CAT chloramphenicol acetyltransferase CBG corticosteroid-binding globulin CBP competitive protein binding assay CC chief complaint CDCA chenodeoxycholic acid CDK or Cdk cyclin-dependent kinase cDNA complementary DNA CDP collagenase-digestible protein Cdx-2 caudal-related homeodomain protein CFU colony-forming unit cGMP cyclic GMP CGRP calcitonin gene-related peptide CHF congestive heart failure CK-II casein kinase-II CLIA competitive chemiluminescence immunoassay cM centimorgans Cm. Ln. cement line CNS central nervous system CPBA competitive protein binding assays cpm counts per minute CRE cAMP response element CREB cAMP response element binding protein CRF chronic renal failure CsA cyclosporin A CSF colony-stimulating factor
xxvi CT CTR CTX CVC CYP CYP24 DAG DBD DBP DBP DC DCA DCT DEXA or DXA 7-DHC DHEA DHT DIC DMSO DR DRIP DSP E1 E2 EAE EBT EBV EC EC50 or ED50 ECaC ECF EDTA EGF ELISA EMSA EP1 ER ERE ERK Et FACS FAD FCS FDA FFA FIT
ABBREVIATIONS
calcitonin or computerized tomography calcitonin receptor cerebrotendinous xanthomatosis calcifying vascular cells cytochrome P450 cytochrome P450, 24-hydroxylase diacylglycerol DNA binding domain diastolic blood pressure vitamin D binding protein dendritic cell deoxycholic acid distal convoluted tubule dual energy X-ray absorptiometry 7-dehydrocholesterol dehydroepiandrosterone dihydrotachysterol or dihydrotestosterone disseminated intravascular coagulation dimethyl sulfoxide direct repeat vitamin D receptor interacting protein dental sialoprotein estrone estradiol experimental autoimmune encephalitis electron-beam computed technology Epstein-Barr virus endothelial cells effective concentration (dose) to cause 50% effect epithelium calcium channel extracellular fluid ethylenediaminetetraacetic acid epidermal growth factor enzyme-linked immunosorbent assay electrophoretic mobility shift assay PG receptor-1 estrogen receptor or endoplasmic reticulum estrogen response element extracellular signal-regulated kinase endothelin fluorescence-activated cell sorting or sorter flavin adenine dinucleotide fetal calf serum U.S. Food and Drug Administration free fatty acid Fracture Intervention Trial
FMTC FP FRAP FS FSK FXR g g G0, G1, G2 GAG GC-MS G-CSF GDNF GFP GFR GH GHRH GIO GM-CSF GnRH GR GRE GRTH HAT HDAC HEK HHRH HIV HNF HPI HPLC HPV hr HRE HSA Hsp HSV HVDRR HVO IBMX IC50 ICA ICMA IDBP IDDM IDM IFN
familial medullary thyroid carcinoma formation period fluorescence recovery after photobleaching Fanconi syndrome forskolin farnesol X receptor gram acceleration due to gravity gap phases of the cell cycle glycosaminoglycan gas chromatography-mass spectrometry granulocyte colony-stimulating factor glial cell-derived neurotrophic factor green fluorescent protein glomerular filtration rate growth hormone growth hormone-releasing hormone glucocorticoid induced osteoporosis granulocyte-macrophage colonystimulating factor gonadotropin-releasing hormone glucocorticoid receptor glucocorticoid response element generalized resistance to thyroid hormone histone acetyltransferase histone deacetylase human embryonic kidney hereditary hypophosphatemic rickets with hypercalciuria human immunodeficiency virus hepatocyte nuclear factor history of present illness high-performance liquid chromatography human papilloma virus hour hormone response element human serum albumin heat-shock protein herpes simplex virus hereditary vitamin D-resistant rickets hypovitaminosis D osteopathy isobutylmethylxanthine concentration to inhibit 50% effect intestinal calcium absorption immunochemiluminometric assay intracellular vitamin D–binding protein insulin-dependent diabetes mellitus infants of diabetic mothers interferon
xxvii
ABBREVIATIONS
Ig IGFBP IGF-I, -II IGF-IR IL i.m. IMCal i.p. IP3 IRMA IU IUPAC i.v. JG JNK Kd Km kb kbp kDa KO LBD LCA LDL Li. Ce. LIF LNH LOD LPS LT LXR M M MAPK Mab MAR MAR MARRS MCR M-CSF MEN2 MGP MHC min MLR Mlt MR MRI mRNA MTC NADH
immunoglobulin IGF binding protein insulin-like growth factor type I, II IGF-I receptor interleukin (e.g., IL-1, IL-1β, etc.) intramuscular intestinal membrane calcium binding complex intraperitoneal inositol trisphosphate immunoradiometric assay international units International Union of Pure and Applied Chemists intravenous juxtaglomerular c-Jun NH2-terminal kinase dissociation constant Michaelis constant kilobases kilobase pairs kilodaltons knock out ligand binding domain lithocholic acid low-density lipoprotein lining cell leukemia inhibitory factor late neonatal hypocalcemia logarithm of the odds lipopolysaccharide leukotriene liver X receptor mitosis phase of cell cycle molar mitogen-activated protein kinase monoclonal antibody matrix attachment region mineral apposition rate membrane-associated rapid response steroid metabolic clearance rate macrophage colony-stimulating factor multiple endocrine neoplasia type 2 matrix Gla protein major histocompatibility complex minute mixed lymphocyte reaction mineralization lag time mineralcorticoid receptor magnetic resonance imaging messenger ribonucleic acid medullary thyroid carcinoma nicotinamide adenine dinucleotide
NADPH
nicotinamide adenine dinucleotide phosphate NAF nuclear accessory factor NBT nitroblue tetrazolium NcAMP nephrogenous cAMP NCP noncollagen protein NGF nerve growth factor NHANES III National Health and Nutrition Examination Survey III NHL Non-Hodgkin’s lymphoma NIDDM non-insulin-dependent diabetes mellitus NIH National Institutes of Health NK cell natural killer cell NLS nuclear localization signal NMR nuclear magnetic resonance NPT sodium/phosphate cotransporter NR nuclear receptor Ob osteoblast Oc osteocalcin or osteoclast OCIF osteoclastogenesis inhibitory factor (same as OPG) OCT 22-oxacalcitriol ODF osteoclast differentiation factor (same as RANKL) 1α-OHD3 1α-hydroxyvitamin D3 25OHD3 25-hydroxyvitamin D3 1,25(OH)2D3 1α,25-dihydroxyvitamin D3 24,25(OH)2D3 24,25-dihydroxyvitamin D3 OHO oncogenic hypophosphatemic osteomalacia Omt osteoid maturation time OPG osteoprotegerin OPN osteopontin OSM oncostatin M OVX ovariectomy Pi inorganic phosphate PA2 phospholipase A2 PAD peripheral arterial vascular disease PAM pulmonary alveolar macrophage PBL peripheral blood lymphocyte PBS phosphate-buffered saline PC phophatidyl choline PCNA proliferating cell nuclear antigen PCR polymerase chain reaction PCT proximal convoluted tubule PDDR pseudovitamin D deficiency rickets PDGF platelet-derived growth factor PEIT percutaneous ethanol injection therapy PHEX phosphate regulating gene with homologies to endopeptidases on the X chromosome PG prostaglandin
xxviii PHA PHP PIC PKA PKC PKI PLA2 PLC PMA PMCA PMH p.o. poly(A) PPAR PR PRA PRL PSA PSI PT PTH PTHrP PTX PUVA QCT QSAR 9-cis-RA RA RA RANK RANKL RAP RAR RARE RAS RBP RCI RDA RFLP RIA RID RNase ROCs ROS RPA RRA RT-PCR RXR
ABBREVIATIONS
phytohemagglutinin pseudohypoparathyroidism preinitiation complex protein kinase A protein kinase C protein kinase inhibitor phospholipase A2 phospholipase C phorbol 12-myristate 13-acetate plasma membrane calcium pump past medical history oral polyadenosine peroxisome proliferator-activated receptor progesterone receptor plasma renin activity prolactin prostate-specific antigen psoriasis severity index parathyroid parathyroid hormone parathyroid hormone-related peptide parathyroidectomy psoralen-ultraviolet A quantitative computerized tomography quantitative structure-activity relationship 9-cis-retinoic acid retinoic acid rheumatoid arthritis receptor activator NF-κB receptor activator NF-κB ligand receptor-associated protein retinoic acid receptor retinoic acid response element renin–angiotensin system retinol-binding protein relative competitive index recommended dietary allowance restriction fragment length polymorphism radioimmunoassay receptor interacting domain ribonuclease receptor operated calcium channels reactive oxygen species ribonuclease protection assay radioreceptor assay reverse transcriptase-polymerase chain reaction retinoid X receptor
RXRE SBP SD SDS SE SEM SH SHBG SLE SOS Sp1 SPF SRC-1 SSCP SV40 SXA t1/2 T3 T4 TBG TBP TC TF TFIIB TG TGF TIO TK TLR TmP or TmPi TNF TPA TPN TPTX TR TRAP TRAP TRP TRE TRE TRH Trk TSH TSS UF US USDA UTR UV VDDR-I
retinoid X receptor response element systolic blood pressure standard deviation sodium dodecyl sulfate standard error standard error of the mean social history sex hormone–binding globulin systematic lupus erythematosus speed of sound selective promoter factor 1 sun protection factor steroid receptor coactivator-1 single strand conformational polymorphism simian virus 40 single energy X-ray absorptiometry half-time triiodothyronine thyroxine thyroid-binding globulin TATA binding protein tumoral calcinosis tubular fluid general transcription factor IIB transgenic transforming growth factor tumor-induced osteomalacia thymidine kinase toll-like receptor tubular absorptive maximum for phosphorus tumor necrosis factor 12-O-tetradecanoylphorbol-13-acetate total parenteral nutrition thyroparathyroidectomized thyroid hormone receptor tartrate-resistant acid phosphatase thyroid hormone receptor associated proteins transient receptor potential thyroid hormone response element TPA response element thyrotropin-releasing hormone tyrosine kinase thyrotropin transcription start site ultrafiltrable fluid ultrasonography U.S. Department of Agriculture untranslated region ultraviolet vitamin D-dependent rickets type I (see PDDR)
xxix
ABBREVIATIONS
VDDR-II VDR VDRE VDRL VEGF VERT VICCs VSMC
vitamin D-dependent rickets type II (see HVDRR) vitamin D receptor vitamin D response element vitamin D receptor ligand vascular endothelial growth factor Vertebral Efficacy with Risedronate Therapy studies voltage-insensitive calcium channels vascular smooth muscle cell
VSSCs WHI WRE WSTF XLH XRD ZEB
voltage-senstive calcium channels Women’s Health Initiative Wilms’ tumor gene, WT1, responsive element Williams syndrome transcription factor X-linked hypophosphatemic rickets X-ray diffraction zinc finger, E box-binding transcription factor
Approximate Normal Ranges for Serum Values in Adultsa Measure
SI Units
Ionized calcium Total calcium Phosphorous, inorganic 25(OH)D 1,25(OH)2D
1.12–1.32 mmol/L 2.17–2.52 mmol/L 0.77–1.49 mol/L 24.9–169.5 nmol/L 60–108 pmol/L
Conventional Units
Conversion Factorb
4.5–5.3 mg/dL 8.7–10.1 mg/dL 2.4–4.6 mg/dL 10–68 ng/mL 25–45 pg/mL
0.2495 0.2495 0.3229 2.496 2.40
Approximate Normal Ranges for Serum Values in Childrena Measure Ionized calcium Total calcium Phosphorous, inorganic 25(OH)D 1,25(OH)2D aNormal
SI Units
Conventional Units
Conversion Factorb
1.19–1.29 mmol/L 2.25–2.63 mmol/L 1.23–1.62 mol/L 34–91 nmol/L 65–134 pmol/L
4.8–5.2 mg/dL 9.0–10.5 mg/dL 3.8–5.0 mg/dL 14–37 ng/mL 27–56 pg/mL
0.2495 0.2495 0.3229 2.496 2.40
ranges differ in various laboratories and these values are provided only as a general guide. factor X conventional units = SI units
bConversion
Useful Equivalencies of Different Units Vitamin D Calcium Phosphorus
1 µg = 40 IU 1 mmol = 40 mg 1 mmol = 30 mg
CHAPTER 1
Historical Perspective HECTOR F. DELUCA Department of Biochemistry, University of Wisconsin-Madison, Madison, Wisconsin
I. Discovery of the Vitamins II. Discovery That Vitamin D Is Not a Vitamin III. Isolation and Identification of Nutritional Forms of Vitamin D
IV. Discovery of the Physiological Functions of Vitamin D V. Discovery of the Hormonal Form of Vitamin D References
I. DISCOVERY OF THE VITAMINS
animals whereas purified materials could not [3]. Funk had found similar results for the prevention of neuritis and reasoned that there were “vital amines” present in foods from natural sources and actually provided the basis for the term “vitamins” used later to describe essential micronutrients [5].
A. Early Nutritional Views The field of nutrition was largely dominated in the nineteenth century by German chemists, led by Justus von Liebig [1]. They taught that adequacy of the diet could be described by an analysis of protein, carbohydrate, fat, and mineral. Thus, a diet containing 12% protein, 5% mineral, 10–30% fat, and the remainder as carbohydrate would be expected to support normal growth and reproduction. This view remained largely unchallenged until the very end of the 19th century and the beginning of the 20th century [2–5]. However, evidence opposing this view began to appear. One of the first was the famous study of Eijkman who studied prisoners in the Dutch East Indies maintained on a diet of polished rice [6]. A high incidence of the neurological disorder beri-beri was recorded in these inmates. Eijkman found that either feeding whole rice or returning the hulls of the polished rice could eliminate beri-beri. Eijkman reasoned that polished rice contained a toxin that was somehow neutralized by the rice hulls. Later, a colleague, Grijns [7], revisited the question and correctly demonstrated that hulls contained an important and required nutrient that prevented beri-beri. Other reports revealed that microorganic nutrients might be present. The development of scurvy in mariners was a common problem. This disease was prevented by the consumption of limes on British ships (hence, the term “Limey” to describe British sailors) and sauerkraut and fruits on other ships. This led Hoist and Frohlich to conclude that scurvy could be prevented by a nutrient present in these foods [8]. Experiments by Lunin, Magendie, Hopkins, and Funk showed that a diet of purified carbohydrate, protein, fat, and salt is unable to support growth and life of experimental animals [2–5]. This suggested that some unknown or vital factor present in natural foods was missing from the purified diets. Hopkins developed a growth test in which natural foods were found to support rapid growth of experimental VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX,
B. McCollum and Osborne and Mendel’s Discovery of Vitamin A and B Complex A key experiment demonstrating essential micronutrients was one carried out at the Wisconsin Agricultural Experiment Station, engineered by Stephen Moulton Babcock and carried out by E. B. Hart supported by McCollum and Steenbock [9]. Herds of dairy cows were maintained on a diet composed individually only of corn, oats, or wheat or were fed a mixture of all of these grains, all receiving the same amount of carbohydrate, protein, fat, and salts and all providing equal analysis according to the German chemists [1]. The animals on the corn diet did very well, produced milk in large amounts, and reproduced normally. Those on the wheat diet failed to thrive and soon were unable to reproduce or lactate. The oat group was found to be intermediate between the corn and wheat groups, and the mixture approximated the growth and reproduction found with corn. Yet all these diets had the same proximate analysis. The conclusion of the Wisconsin Experiment Station study was that there are unknown nutrients present in corn and not found in wheat that are essential for life and reproduction. This led E. B. Hart, Chairman of Biochemistry at Wisconsin, to conceive that a search for these nutrients must begin. Professor McCollum was asked to search for these nutrients using small experimental animals. McCollum and Davis demonstrated there was present in butter fat a substance that prevented xerophthalmia and was also required for growth. They termed this “a lipin-soluble growth factor” [10]. McCollum later named this factor “vitamin A” [11]. Copyright © 2005, Elsevier, Inc. All rights reserved.
4 This substance was absent from lard and other fats but was found in large amounts in cod liver oil. In constructing the diets, McCollum obtained the carbohydrates and salts from milk whey, which, unknown to him, supplied the vitamin B complex group of micronutrients that permitted him to observe a vitamin A deficiency. McCollum at Wisconsin [11] and Osborne and Mendel [12] at the Connecticut Experiment Station carried out experiments in which cod liver oil was used as a source of fat in the diet, but the minerals were supplied from pure chemicals mixed to approximate the mineral composition of milk. Starch or sugar was used as the carbohydrate. These animals developed a different group of symptoms, namely, neuritis, which could be cured by the provision of the milk components. McCollum and Osborne and Mendel correctly concluded that this activity was due to a different micronutrient called “vitamin B.” This ushered in the concept of the organic micronutrients known as vitamins.
C. History of Rickets The disease rickets was very likely known in antiquity but was described in the 15th century as revealed by later writings. Whistler first provided a clear description of rickets in which the skeleton was poorly mineralized and deformed [13]. Rickets undoubtedly in ancient times appeared only on rare occasions and hence was not considered a problem. However, at the end of the 19th century, the industrial revolution had taken place: A highly agrarian population had become urbanized, and smoke from the industrial plants polluted the atmosphere. Thus, in low-sunlight countries such as England, rickets appeared in epidemic proportions. In fact, it was known as “the English disease” [14]. Some reports of the beneficial action of cod liver oil had appeared. However, they were not given scientific credence. With the discovery of the vitamins, Sir Edward Mellanby in Great Britain began to reason that rickets might also be a disease caused by a dietary deficiency [15]. Mellanby fed dogs a diet composed primarily of oatmeal, which was the diet consumed where the incidence of rickets was the highest (i.e., Scotland). McCollum inadvertently maintained the dogs on oatmeal indoors and away from ultraviolet light. The dogs developed severe rickets. Learning from the experiments of McCollum, Mellanby provided cod liver oil to cure or prevent the disease. Mellanby could not decide whether the healing of rickets was due to vitamin A known to be present in the cod liver oil or whether it was a new and unknown substance. Therefore, the activity of healing rickets was first attributed to vitamin A.
HECTOR F. DELUCA
D. Discovery of Vitamin D McCollum, who had moved to Johns Hopkins from Wisconsin, continued his experiments on the fat-soluble materials. McCollum used aeration and heating of cod liver oil to destroy the vitamin A activity or the ability to support growth and to prevent xerophthalmia [16]. However, cod liver oil treated in this manner still retained the ability to cure rickets. McCollum correctly reasoned that the activity in healing rickets was due to a new and heretofore unknown vitamin that he termed “vitamin D.” On the basis of the experiments of McCollum and of Mellanby, vitamin D became known as an essential nutrient.
II. DISCOVERY THAT VITAMIN D IS NOT A VITAMIN At the same time that Sir Edward Mellanby was carrying out the experiments in dogs, Huldshinsky [17] and Chick et al. [18] independently found that rickets in children could be prevented or cured by exposing them to sunlight or to artificially induced ultraviolet light. Thus, the curious findings were that sunlight and ultraviolet light somehow equaled cod liver oil. These strange and divergent results required resolution. Steenbock and Hart had noted the importance of sunlight in restoring positive calcium balance in goats [19]. At Wisconsin, with McCollum carrying out experiments in small experimental animals (i.e., rats), Steenbock was required to work with larger animals. Steenbock then began to study goats because they would consume less materials and could serve as better experimental animals than cows. Steenbock began to study the calcium balance of lactating goats and found that those goats maintained outdoors in the sunlight were found to be in positive calcium balance, whereas those maintained indoors lost a great deal of their skeletal calcium to lactation [19]. Steenbock and Hart, therefore, noted the importance of sunlight on calcium balance. This work then undoubtedly led Steenbock to realize that the ultraviolet healing properties described by Huldschinsky might be related to the calcium balance experiments in goats. By irradiating the animals and diets, Steenbock and Black found that vitamin D activity could be induced and rickets could be cured [20]. A similar finding was reported soon thereafter by Hess and Weinstock [21]. Steenbock then traced this to the nonsaponifiable fraction of the lipids in foods [22]. He found that ultraviolet light activated an inactive substance to become a vitamin D–active material. Thus, ultraviolet light could be used to irradiate foods, induce vitamin D activity, and fortify foods to eliminate rickets as a major
5
CHAPTER 1 Historical Perspective
medical problem. This discovery also made available a source of vitamin D for isolation and identification.
III. ISOLATION AND IDENTIFICATION OF NUTRITIONAL FORMS OF VITAMIN D From irradiation of mixtures of plant sterols, Windaus and colleagues isolated a material that was active in healing rickets [23]. This substance was called “vitamin D1,” but its structure was not determined. Vitamin D1 proved to be an adduct of tachysterol and vitamin D2, and thus vitamin D1 was actually an error in identification. The British group led by Askew was successful in isolating and determining the structure of the first vitamin D, vitamin D2 or ergocalciferol, from irradiation of plant sterols [24]. A similar identification by the Windaus group confirmed the structure of vitamin D2 [25]. Windaus and Bock also isolated the precursor of vitamin D3 from skin, namely, 7-dehydrocholesterol [26]. Furthermore, 7-dehydrocholesterol was synthesized [27] and converted to vitamin D3 (cholecalciferol) as identified by the Windaus group [28]. Thus, the structures of nutritional forms of vitamin D became known (Fig. 1). Windaus and Bock, having isolated 7-dehydrocholesterol from skin, provided the presumptive evidence that vitamin D3 is the form of vitamin D produced in skin, a discovery that was later confirmed by the chemical identification of vitamin D3 in skin by Esvelt et al. [29] and of a previtamin D3 in skin by Holick et al. [30]. Synthetic vitamin D as produced by the irradiation process replaced the irradiation of foods as a means of fortifying foods with vitamin D and was also rapidly applied to a variety of diseases including rickets and tetany and in the provision to domestic animals such as chickens, cows, and pigs. Windaus’ group provided chemical syntheses of the vitamin D compounds, confirming their structures and
thus ending the era of the isolation and identification of nutritional forms of vitamin D and making them available for the treatment of disease. For his contributions, Windaus received the 1938 Nobel Prize in chemistry.
IV. DISCOVERY OF THE PHYSIOLOGICAL FUNCTIONS OF VITAMIN D A. Intestinal Calcium and Phosphorus Absorption Besides bone mineralization, the earliest discovered function of vitamin D is its important role in the absorption and utilization of calcium. The first report of this finding was in the early 1920s by Orr and colleagues [31]. Kletzien et al. [32] demonstrated that vitamin D plays an important role in the utilization of calcium from the diet, and a number of experiments were carried out on the utilization of calcium and phosphorus from cereal diets. Nicolaysen was responsible, however, for demonstrating unequivocally the role of vitamin D in the absorption of calcium and independently of phosphorus from the diet [33]. Nicolaysen also followed the early work of Kletzien et al. [32] in which animals adapted to a low calcium diet were better able to utilize calcium than were animals on an adequate calcium diet. This work was confirmed by Nicolaysen, who postulated the existence of an “endogenous factor” that would inform the intestine of the skeletal needs for calcium [34]. This endogenous factor later proved to be largely the active form of vitamin D, 1,25-dihydroxyvitamin D3 [1,25(OH)2D3] [35]. Strong support for this concept was provided by the studies of Ribovich et al. [36] that showed animals maintained on a constant exogenous source of 1,25(OH)2D3 are unable to change intestinal calcium transport in response to changes in dietary calcium levels.
B. Mobilization of Calcium from Bone
FIGURE 1 Nutritional forms of vitamin D.
For many years, investigators have attempted to show that vitamin D plays a role directly on the mineralization process of the skeleton. However, early work by Rowland and Kramer [37], later work by Lamm and Neuman [38], and more recent work by Underwood and DeLuca [39] demonstrated very clearly that vitamin D does not play a significant role in the actual mineralization process of the skeleton but that the failure to mineralize the skeleton in vitamin D deficiency is due to inadequate levels of calcium and phosphorus in the plasma. Thus, the action of vitamin D in mineralizing
6 the skeleton and in preventing hypocalcemic tetany is the elevation of plasma calcium and phosphorus [40]. These discoveries laid to rest the concept of a role of vitamin D in mineralization. However, Carlsson [41] and Bauer et al. [42] were the first to realize that a major function of vitamin D is to induce the mobilization of calcium from bone when required. Thus, in animals on a low-calcium diet, the rise in serum calcium induced by vitamin D is the result of actual mobilization of calcium from bone [43]. This important function is known to be essential for the provision of calcium to meet soft-tissue needs, especially those of nerves and muscle, on a minute-to-minute basis when it is in insufficient supply from the diet. It is likely that the function of vitamin D in mobilizing calcium from bone is an osteoclastic-mediated process [44]. It is clear, however, that both vitamin D and parathyroid hormone are required for this function [45]. Furthermore, it is clear that vitamin D plays an important role in osteoclasticmediated bone resorption [46], which is certainly the first event in bone remodeling and an essential event in bone modeling [47].
C. Renal Reabsorption of Calcium and Phosphorus A final site of vitamin D action to elevate plasma calcium is in the distal renal tubule. Although experiments were suggestive of a role for vitamin D in increasing renal tubule reabsorption of calcium, a clear demonstration of this did not occur until the 1980s at the hands of Yamamoto et al. [48]. The renal tubule reabsorbs 99% of the filtered calcium even in the absence of vitamin D. However, reabsorption of the last 1% of the filtered load requires both vitamin D and parathyroid hormone. Thus, these agents work in concert in the renal reabsorption of calcium as well as in the mobilization of calcium from bone. Both agents are required to carry out this function.
D. Discovery of New Functions of Vitamin D With discovery of the receptor for the vitamin D hormone (described in Section V,G below) came the surprising result that this receptor could be found in a variety of tissues not previously appreciated as targets of vitamin D action. It localizes in the distal renal tubule cells, enterocytes of the small intestine, bone lining cells, and osteoblasts in keeping with its known role in calcium metabolism [49,50]. However, its appearance
HECTOR F. DELUCA
in tissues such as parathyroid gland, islet cells of the pancreas, cells in bone marrow (i.e., promyelocytes), lymphocytes, and certain neural cells raised the question of whether the functions of vitamin D might be broader than previously anticipated [49,50]. As a result of those findings, new functions of vitamin D have been found. For example, vitamin D plays a role in causing differentiation of promyelocytes to monocytes and the subsequent coalescing of the monocytes into multinuclear osteoclast precursors and ultimately into active osteoclasts [51,52]. Suppression of parathyroid cell growth and suppression of parathyroid hormone gene expression represent other new vitamin D actions [53,54]. In keratinocytes of skin, vitamin D appears to play a role in suppression of growth and in cellular differentiation [55]. Likely, discoveries of many new functions of 1,25(OH)2D3 will be made and are well on their way, as described in later chapters of this volume.
V. DISCOVERY OF THE HORMONAL FORM OF VITAMIN D A. Early Work of Kodicek The true pioneer of vitamin D metabolism was Egan MA Kodicek working at the Dunn Nutritional Laboratory in Cambridge U.K. Kodicek used a bioassay at first to study the fate of the vitamin D molecule and found that much vitamin D was converted to biologically inactive products [56]. Clearly, however, this approach of assaying vitamin D activity following administration of known doses of vitamin D was of limited value in determining metabolism.
B. Radiolabeled Vitamin D Experiments Professor Kodicek then began to synthesize radiolabeled vitamin D2. Unfortunately, the degree of labeling was not sufficient to permit the administration of truly physiological doses of vitamin D. Nevertheless, Professor Kodicek continued investigations into this important area. At the conclusion of 10 years of work, he concluded that vitamin D was active without metabolic modification and that the metabolites that were found were biologically inactive [57]. This conclusion was reached even as late as 1967, when it was concluded that vitamin D3 itself was the active form of vitamin D in the intestine [58]. However, chemical synthesis of vitamin D3 of high specific activity in the laboratory of the author proved to be of key importance in the demonstration of biologically active metabolites [59].
7
CHAPTER 1 Historical Perspective
By providing a truly physiological dose of vitamin D, it could be learned that the vitamin D itself disappeared and instead polar metabolites could be found in the target tissues before those tissues responded [60]. The polar metabolites proved to be more biologically active and acted more rapidly than vitamin D itself [61]. Thus, presumptive evidence of conversion of vitamin D to active forms had been obtained as early as 1967.
C. Isolation and Identification of the Active Form of Vitamin D By 1968, the first active metabolite of vitamin D was isolated in pure form and chemically identified as 25-hydroxyvitamin D3 (25OHD3) [62]. Its structure was confirmed by chemical synthesis [63] that provided it for study to the medical and scientific world. For a couple of years, 25OHD3, was visualized as the active form of vitamin D. However, when it was synthesized in radiolabeled form, it was found to be rapidly metabolized to yet more polar metabolites [64]. By this time, the Kodicek laboratory reawakened their interest in metabolism of vitamin D and began to study the metabolism of lα-tritium-labeled vitamin D [65]. Furthermore, polar metabolites of vitamin D were found by Haussler, Myrtle, and Norman [66]. The Wisconsin group labeled these metabolites as peak 5 [64], the Norman group called it peak 4B [66], and Lawson, Wilson, and Kodicek described it as peak P [65]. Kodicek et al. claimed that the metabolite of vitamin D found in intestine was deficient in tritium at the 1-position [65]. However, Myrtle et al. reported that peak 4B did not lose its tritium [67]. Thus, the suggestion of a modification at the 1-position could not be confirmed. The DeLuca group, however, isolated the active metabolite from intestines of 1600 chickens given radiolabeled vitamin D, and, by means of mass spectrometric techniques and specific chemical reactions, the structure of the active form of vitamin D in the intestine was unequivocally demonstrated as 1,25(OH)2D3 [68]. Of great importance was the finding of Fraser and Kodicek that the peak P metabolite could be produced by homogenates of chicken kidney and that anephric animals are unable to produce the peak P metabolite [69]. They correctly concluded that the site of synthesis of the active form of vitamin D is the kidney. The Wisconsin group then chemically synthesized both 1α25(OH)2D3 [70] and 1β,25(OH)2D3 [71] and provided unequivocal proof that the active form is 1α,25(OH)2D3. Furthermore, this group was able to synthesize lαOHD3, an important
analog that assumed great importance as a therapeutic agent throughout the world [72].
D. Proof That 1,25(OH)2D3 Is the Active Form of Vitamin D Proof that 1,25(OH)2D3 and not 25OHD3 is the active form was provided by experiments in which anephric animals respond to 1,25(OH)2D3 by increasing intestinal absorption of calcium and bone calcium mobilization, whereas animals receiving 25OHD3 at physiological doses did not [73–75]. Furthermore, the experiment of nature, namely, vitamin D–dependency rickets type I, an autosomal recessive disorder, provided final proof [76]. This disease could be corrected by physiological doses of synthetic 1,25(OH)2D3, whereas large amounts of vitamin D3 or 25OHD3 were needed to heal the rickets. The exact defect in this disease is now clearly known and is described elsewhere in this volume. 25OHD3 at pharmacological doses likely acts as an analog of the final vitamin D hormone, 1,25(OH)2D3 (Fig. 2).
E. Discovery of the Vitamin D Endocrine System Immediately after the identification of 1,25(OH)2D3 as the active form of vitamin D came studies in which it could be shown that animals on a low-calcium diet produce large quantities of 1,25(OH)2D3, whereas those on a high-calcium diet produce little or no 1,25(OH)2D3 [77]. A reciprocal arrangement was found for the metabolite 24R,25(OH)2D3. Thus, when calcium is needed, production of 1,25(OH)2D3 is markedly stimulated and the 24-hydroxylation degradation reaction is suppressed. When adequate calcium is present, production of 1,25(OH)2D3 is shut off and the 24-hydroxylation reaction is turned on. This discovery also satisfactorily provided evidence that 1,25(OH)2D3 is the likely endogenous factor originally described by Nicolaysen et al. [34]. The next important step was the demonstration that it is parathyroid hormone that activates 1α-hydroxylation of 25OHD3 in the kidney [78]. Thus, parathyroidectomy eliminates the hypocalcemic stimulation of 1α-hydroxylation and suppression of 24-hydroxylation, whereas administration of parathyroid hormone restores that capability. Fraser and Kodicek also provided evidence that, in intact chickens, injection of parathyroid hormone stimulated the 1α-hydroxylation reaction [79]. Thus, the basic vitamin D endocrine system was largely discovered and reported in the early 1970s, being completed by 1974.
8
HECTOR F. DELUCA
OH
Liver
Kidney
Microsomes (Mitochondria)
Mitochondria
HO
HO Vitamin D3
OH
HO 25-hydroxyvitamin D3
OH 1α,25-dihydroxyvitamin D3
FIGURE 2 Activation of the vitamin D molecule.
F. Other Metabolites of Vitamin D During the course of identification of 1,25(OH)2D3, 21,25(OH)2D3 was reported as a metabolite, as was 25,26(OH)2D3 [80,81]. However, the identification of 21,25(OH)2D3 was in error and was corrected to 24,25(OH)2D3, with the correct stereochemistry as 24R,25(OH)2D3 [82]. Over the late 1970s and early 1980s, as many as 30 metabolites of vitamin D were identified [83]. These are covered in other chapters in this volume. Of great importance was the use of the fluoro derivatives of vitamin D to illustrate that the only activation pathway of vitamin D is 25-hydroxylation followed by 1-hydroxylation [84]. Thus, 24-difluoro25OHD3 supported all known functions of vitamin D for at least two generations of animals [85]. 24-Difluoro25OHD3 cannot be 24-hydroxylated. Furthermore, other fluoro derivatives such as 26,27-hexafluoro-25OHD3 [86] and 23-difluoro-25OHD3 [87] are all fully biologically active, illustrating that 26-hydroxylation, 24-hydroxylation, and 23-hydroxylation are not essential to the function of vitamin D.
G. Discovery of the Vitamin D Receptor Zull and colleagues provided evidence that the function of vitamin D is blocked by transcription and protein inhibitors [88]. Thus, it became clear very early that a nuclear activity is required for vitamin D to carry out its functions. This work confirmed and extended the earlier work of Eisenstein and Passavoy [89]. With the discovery of the active forms of vitamin D came new attention to the idea that vitamin D may work through a nuclear mechanism. Thus, Haussler et al. reported vitamin D compounds to be associated with chromatin [66]. However, these experiments did not
exclude the possibility that the vitamin D compounds might be bound to the nuclear membrane. The first clear demonstration of the existence of a vitamin D receptor was at the hands of Brumbaugh and Haussler [90]. Furthermore, the experiments of Kream et al. [91] provided strong and unequivocal evidence of the existence of a nuclear receptor for 1,25(OH)2D3. Intense efforts toward purification of the receptor and its study appeared with the knowledge that it is a receptor protein with a molecular weight of approximately 55,000. In 1987, a partial cDNA sequence for the chicken vitamin D receptor was determined [92]. This was followed by isolation of the full coding sequence for the human [93] and rat [94,95] receptors. Cloning of the cDNAs encoding the vitamin D receptor in a human and mouse permitted the isolation of the gene encoding the vitamin D receptor [96–98]. The human gene was completely described and the mouse promoter was isolated and shown to be a TATA-less Sp1driven promoter [98]. The human gene appears to have alternate promoters [96,97]. Two groups have prepared receptor Null mutant mice, permitting extensive experiments with vitamin D receptorless animals [99,100]. From a historical point of view, one of the most important discoveries was vitamin D-dependency rickets type II [101], which is now known to be due to a defect in the receptor gene [102,103] (discussed in Chapter 11). This discovery essentially provided receptor knockout experiments in humans, allowing unequivocal proof of the essentiality of the vitamin D receptor for the functions of vitamin D. The nature of the receptor and how it functions are described in subsequent chapters along with current thinking on the molecular mechanism of action of 1,25(OH)2D3. However, the discovery of vitamin D responsive elements in the osteocalcin, osteopontin, preproparathyroid, and 24-OHase genes represent important
CHAPTER 1 Historical Perspective
historical developments [104–107]. This led to a consensus sequence and, most important, the development of the 3, 4, 5-rule of Umesono et al. [108]. It is now clear that vitamin D–responsive elements represent two imperfect repeat sequences separated by three nonspecified nucleotides. The vitamin D receptor will bind to these response elements, but it requires the presence of another nuclear factor, which proved to be the retinoid-X receptor (RXR) [109,110]. It is quite clear that the vitamin D receptor forms a heterodimer on the vitamin D responsive elements with the RXR protein on the 5′ arm of the responsive element and the vitamin D receptor on the 3′ segment [111]. The work of Rosenfeld, Glass, and colleagues [112] has demonstrated that the RXR protein when complexed with the vitamin D receptor on the responsive elements will not accept an RXR ligand, thus acting as a silent partner. Details of what is known concerning the role of the vitamin D receptor in transcription are described fully in subsequent chapters.
Acknowledgment This work was supported in part by a fund from the Wisconsin Alumni Research Foundation.
References 1. von Liebig J 1957 Animal Chemistry or Organic Chemistry in Its Application to Physiology and Pathology. In: Glass HB (ed) A History of Nutrition. The Riverside Press, Cambridge, MA. 2. Lunin N 1881 Über die Bedeutung der anorganischen Salze für die Ernährung des Tieres. Z Physiol Chem 5:31–39. 3. Hopkins G 1912 Feeding experiments illustrating the importance of accessory food factors in normal dietaries. J Physiol 44:425–460. 4. Magendie F 1816 Ann Chim Phys. 3:86–87. In: Glass HB (ed) 1957 A History of Nutrition by McCollum EV. The Riverside Press, Cambridge MA. 5. Funk C 1911 The chemical nature of the substance that cures polyneurities in birds produced by a diet of polished rice. J Physiol (London) 43:395–402. 6. Eijkman C 1897 Ein beri-beri Anliche der Hühner. Virchows Arch 148:523–527. 7. Grijns G 1957 Concerning polyneuritis geneeskundig gallinarum tijschrift voor ned indie 1901. In: McCollum EV, Glass HB (eds) A History of Nutrition. Houghton Mifflin, Boston, p. 216. 8. Hoist A. Frolich T 1907 Experimental studies relating ship-beriberi to scurvy. II. On the etiology of scurvy. J Hyg 7: 634–671. 9. Hart EB, McCollum EV, Steenbock H, Humphrey GC 1911 Physiological effect on growth and reproduction of rations balanced from restricted sources. Wis Agric Exp Sta Res Bull 17:131–205. 10. McCollum EV, Davis M 1913 The necessity of certain lipins in the diet during growth. J Biol Chem 25:167–175.
9 11. McCollum EV, Simmons N, Pitz W 1916 The relation of the unidentified dietary factors, the fat-soluble A and watersoluble B of the diet to the growth promoting properties of milk. J Biol Chem 27:33–38. 12. Osborne TB, Mendel LB 1917 The role of vitamins in the diet. J Biol Chem 31:149–163. 13. Whistler D 1645 De morbo puerli anglorum, quem patrio ideiomate indigenae vocant “the rickets” (Lugduni, Batavorum 1645). As cited by Smerdon GT,1950 Daniel Whistler and the English Disease. A translation and biographical note. J Hist Med 5:397–415. 14. Hess A 1929 The history of rickets. In: Rickets, Including Osteomalacia and Tetany. Lea & Febiger, Philadelphia, pp. 22–37. 15. Mellanby E 1919 An experimental investigation on rickets. Lancet 1:407–412. 16. McCollum EV, Simmonds N, Becker JE, Shipley PG 1922 An experimental demonstration of the existence of a vitamin which promotes calcium deposition. J Biol Chem 53:293–298. 17. Huldshinsky K 1919 Heilung von Rachitis durch Künstlich Hohen-sonne. Deut Med Wochenschr 45:712–713. 18. Chick H, Palzell EJ, Hume EM 1923 Studies of rickets in Vienna 1919–1922. Medical Research Council, Special Report No. 77. 19. Steenbock H, Hart EB 1913 The influence of function on the lime requirements of animals. J Biol Chem 14:59–73. 20. Steenbock H, Black A 1924 Fat-soluble vitamins. XVII. The induction of growth-promoting and calcifying properties in a ration by exposure to ultraviolet light. J Biol Chem 61:405–422. 21. Hess AF, Weinstock M 1924 Antirachitic properties imparted to lettuce and to growing wheat by ultraviolet irradiation. Proc Soc Exp Biol Med 22:5–6. 22. Steenbock H, Black A 1925 Fat-soluble vitamins. XXIII. The induction of growth-promoting and calcifying properties in fats and their unsaponifiable constituents by exposure to light. J Biol Chem 64:263–298. 23. Windaus A, Linsert O 1928 Vitamin D1 . Ann Chem 465:148. 24. Askew FA, Bourdillon RB, Bruce HM, Jenkins RGC, Webster TA 1931 The distillation of vitamin D. Proc R Soc 8107:76–90. 25. Windaus A, Linsert O, Luttringhaus A, Weidlich G 1932 Crystalline-vitamin D2. Ann Chem 492:226–241. 26. Windaus A, Bock F 1937 Über das Provitamin aus dem Sterin der Schweineschwarte. Z Physiol Chem 245:168–170. 27. Windaus A, Lettre H, Schenck F 1935 7-Dehydrocholesterol. Ann Chem 520:98–107. 28. Windaus A, Schenck F, von Werder F 1936 Über das antirachitisch wirksame Bestrahlungs-produkt aus 7-Dehydrocholesterin. Hoppe-Seylers Z Physiol Chem 241:100–103. 29. Esvelt RP, Schnoes HK, DeLuca HF 1978 Vitamin D3 from rat skins irradiated in vitro with ultraviolet light. Arch Biochem Biophys 188:282–286. 30. Holick MF, MacLaughlin JA, Clark MB, Holick SA, Potts JT Jr, Anderson RR, Blank IH, Parrish JA, Elias P 1980 Photosynthesis of previtamin D3 in human skin and the physiologic consequences. Science 210:203–205. 31. Orr WJ, Holt LE Jr, Wilkins L, Boone FH 1923 The calcium and phosphorus metabolism in rickets, with special reference to ultraviolet rat therapy. Am J Dis Children 26:362–372. 32. Kletzien SWF, Templin VM, Steenbock H, Thomas BH 1932 Vitamin D and the conservation of calcium in the adult. J Biol Chem 97:265–280. 33. Nicolaysen R 1937 Studies upon the mode of action of vitamin D. III. The influence of vitamin D on the absorption of calcium and phosphorus in the rat. Biochem J 31:122–129. 34. Nicolaysen R, Eeg-Larsen N, Malm OJ 1953 Physiology of calcium metabolism. Physiol Rev 33:424–444.
10 35. Boyle IT, Gray RW, Omdahl JL, DeLuca HF 1972 Calcium control of the in vivo biosynthesis of 1,25-dihydroxyvitamin D3: Nicolaysen’s endogenous factor. In: Taylor S (ed) Endocrinology 1971. Heinemann Medical Books, London, pp. 468–476. 36. Ribovich ML, DeLuca HF 1975 The influence of dietary calcium and phosphorus on intestinal calcium transport in rats given vitamin D metabolites. Arch Biochem Biophys 170:529–535. 37. Rowland J, Kramer B 1921 Calcium and phosphorus in the serum in relation to rickets. Am J Dis Children 22:105–119. 38. Lamm M, Neuman WF 1958 On the role of vitamin D in calcification. Arch Pathol 66:204–209. 39. Underwood JL, DeLuca HF 1984 Vitamin D is not directly necessary for bone growth and mineralization. Am J Physiol 246:E493–E498. 40. DeLuca HF 1967 Mechanism of action and metabolic fate of vitamin D. Vit Horm 25:315–367. 41. Carlsson A 1952 Tracer experiments on the effect of vitamin D on the skeletal metabolism of calcium and phosphorus. Acta Physiol Scand 26:212–220. 42. Bauer GCH, Carlsson A, Lindquist B 1955 Evaluation of accretion, resorption, and exchange reactions in the skeleton. Kungl Fysiograf Sallskapets I. Lund Forhandlingar 25:3–18. 43. Blunt JW, Tanaka Y, DeLuca HF 1968 The biological activity of 25-hydroxycholecalciferol, a metabolite of vitamin D3. Proc Natl Acad Sci USA 61:1503–1506. 44. Morony S, Capparelli C, Lee R, Shimamoto G, Boone T, Lacey DL, Dunstan CR 1999 A chimeric form of osteoprotegerin inhibits hypercalcemia and bone resorption induced by IL-1β, TNF-α, PTH, PTHrP, and 1,25(OH)2D3. J Bone Miner Res 14:1478–1485. 45. Garabedian M, Tanaka Y, Holick MF, DeLuca HF 1974 Response of intestinal calcium transport and bone calcium mobilization to 1,25-dihydroxyvitamin D3 in thyroparathyroidectomized rats. Endocrinology 94:1022–1027. 46. Raisz LG, Trummel CL, Holick MF, DeLuca HF 1972 1,25-Dihydroxycholecalciferol: A potent stimulator of bone resorption in tissue culture. Science 175:768–769. 47. Frost HM 1966 Bone Dynamics in Osteoporosis and Osteomalacia. Henry Ford Hospital Surgical Monograph Series. Thomas, Springfield, IL. 48. Yamamoto M, Kawanobe Y, Takahashi H, Shimazawa E, Kimura S, Ogata E 1984 Vitamin D deficiency and renal calcium transport in the rat. J Clin Invest 74:507–513. 49. Stumpf WE, Sar M, Reid FA, Tanaka Y, DeLuca HF 1979 Target cells for 1,25-dihydroxyvitamin D3 in intestinal tract, stomach, kidney, skin, pituitary and parathyroid. Science 206:1188–1190. 50. Stumpf WE, Sar M, DeLuca HF 1981 Sites of action of 1,25 (OH)2 vitamin D3 identified by thaw-mount autoradiography. In: Cohn DV, Talmage RV, Matthews JL (eds) Hormonal Control of Calcium Metabolism. Excerpta Medica, Amsterdam, pp. 222–229. 51. Suda T, Takahashi N, Martin TJ 1992 Modulation of osteoclast differentiation. Endocr Rev 13:66–80. 52. Suda T 1992 The role of 1α,25-dihydroxyvitamin D3 in the myeloid cell differentiation. Proc Soc Exp Biol Med 191:214–220. 53. Demay MB, Kiernan, MS, DeLuca HF, Kronenberg HM 1992 Sequences in the human parathyroid hormone gene that bind the 1,25-dihydroxyvitamin D3 receptor and mediate transcriptional repression in response to 1,25-dihydroxyvitamin D3. Proc Natl Acad Sci USA 89:8097–8101. 54. Silver J 1994 Regulation of parathyroid hormone production by 1α,25-(OH)2D3 and its analogues. Their therapeutic usefulness in secondary hyperparathyroidism. In: Vitamin D and
HECTOR F. DELUCA
55.
56.
57. 58. 59. 60. 61. 62. 63. 64. 65.
66. 67. 68.
69. 70. 71. 72. 73.
Its Analogues. The Second International Forum on Calcified Tissue and Bone Metabolism. Chugai Pharmaceutical, Tokyo, pp. 60–63. Smith EL, Walworth NC, Holick MF 1986 Effect of 1α,25-dihydroxyvitamin D3 on the morphologic and biochemical differentiation of cultured human epidermal keratinocytes grown in serum-free conditions. J Invest Dermatol 86:709–714. Kodicek E 1956 Metabolic studies on vitamin D. In: Wolstenholme GWE, O’Connor CM (eds) Ciba Foundation Symposium on Bone Structure and Metabolism. Little, Brown, and Co., Boston, pp. 161–174. Kodicek E 1960 The metabolism of vitamin D. In: Umbreit W, Molitor H (eds) Proceedings of the Fourth International Congress of Biochemistry. Pergamon, London, Vol. 11, pp. 198–208. Haussler MR, Norman AW 1967 The subcellular distribution of physiological doses of vitamin D3. Arch Biochem Biophys 118:145–153. Neville PF, DeLuca HF 1966 The synthesis of [1,2-3H]vitamin D3 and the tissue localization of a 0.25 µg (10 IU) dose per rat. Biochemistry 5:2201–2207. Lund J, DeLuca HF 1966 Biologically active metabolite of vitamin D3 from bone, liver, and blood serum. J Lipid Res 7:739–744. Morii H, Lund J, Neville PF, DeLuca HF 1967 Biological activity of a vitamin D metabolite. Arch Biochem Biophys 120:508–512. Blunt JW, DeLuca HF, Schnoes HK 1968 25-Hydroxycholecalciferol. A biologically active metabolite of vitamin D3. Biochemistry 7:3317–3322. Blunt JW, DeLuca HF 1969 The synthesis of 25-hydroxycholecalciferol. A biologically active metabolite of vitamin D3. Biochemistry 8:671–675. DeLuca HF 1970 Metabolism and function of vitamin D. In: DeLuca HF, Suttie JW (eds) The Fat-Soluble Vitamins. Univ. of Wisconsin Press, Madison, pp. 3–20. Lawson DEM, Wilson PW, Kodicek E 1969 Metabolism of vitamin D. A new cholecalciferol metabolite, involving loss of hydrogen at C-l, in chick intestinal nuclei. Biochem J 115:269–277. Haussler MR, Myrtle JF, Norman AW 1968 The association of a metabolite of vitamin D3 with intestinal mucosa chromatin in vivo. J Biol Chem 243:4055–4064. Myrtle JF, Haussler MR, Norman AW 1970 Evidence for the biologically active form of cholecalciferol in the intestine. J Biol Chem 245:1190–1196. Holick MF, Schnoes HK, DeLuca HF, Suda T, Cousins RJ 1971 Isolation and identification of 1,25–dihydroxycholecalciferol. A metabolite of vitamin D active in intestine. Biochemistry 10:2799–2804. Fraser DR, Kodicek E 1970 Unique biosynthesis by kidney of a biologically active vitamin D metabolite. Nature 228: 764–766. Semmler EJ, Holick MF, Schnoes HK, DeLuca HF 1972 The synthesis of 1α,25-dihydroxycholecalciferol—A metabolically active form of vitamin D3. Tetrahedron Lett 40:4147–4150. Paaren HE, Schnoes HK, DeLuca HF 1977 Synthesis of 1α,25-dihydroxyvitamin D3. J Chem Soc Chem Commun, 890–892. Holick MF, Semmler EJ, Schnoes HK, DeLuca HF 1973 1αHydroxy derivative of vitamin D3: A highly potent analog of 1α,25-dihydroxyvitamin D3. Science 180:190–191. Boyle IT, Miravet L, Gray RW, Holick MF, DeLuca HF 1972 The response of intestinal calcium transport to 25-hydroxy and 1,25-dihydroxyvitamin D in nephrectomized rats. Endocrinology 90:605–608.
11
CHAPTER 1 Historical Perspective
74. Holick MF, Garabedian M, DeLuca HF 1972 1,25Dihydroxycholecalciferol: Metabolite of vitamin D3 active on bone in anephric rats. Science 176:1146–1147. 75. Wong RG, Norman AW, Reddy CR, Coburn JW 1972 Biologic effects of 1,25-dihydroxycholecalciferol (a highly active vitamin D metabolite) in acutely uremic rats. J Clin Invest 51:1287–1291. 76. Fraser D, Kooh SW, Kind HP, Holick MF, Tanaka Y, DeLuca HF 1973 Pathogenesis of hereditary vitamin D dependent rickets: An inborn error of vitamin D metabolism involving defective conversion of 25-hydroxyvitamin D to 1α,25dihydroxyvitamin D. N Engl J Med 289:817–822. 77. Boyle IT, Gray RW, DeLuca HF 1971 Regulation by calcium of in vivo synthesis of 1,25-dihydroxycholecalciferol and 21,25-dihydroxycholecalciferol. Proc Natl Acad Sci USA 68:2131–2134. 78. Garabedian M, Holick MF, DeLuca HF, Boyle IT 1972 Control of 25-hydroxycholecalciferol metabolism by the parathyroid glands. Proc Natl Acad Sci USA 69:1673–1676. 79. Fraser DR, Kodicek E 1973 Regulation of 25-hydroxycholecalciferol-1-hydroxylase activity in kidney by parathyroid hormone. Nature (New Biol) 241:163–166. 80. Suda T, DeLuca HF, Schnoes HK, Ponchon G, Tanaka Y, Holick MF 1970 21,25-Dihydroxycholecalciferol. A metabolite of vitamin D3 preferentially active on bone. Biochemistry 9:2917–2922. 81. Suda T, DeLuca HF, Schnoes HK, Tanaka Y, Holick MF 1970 25,26-Dihydroxycholecalciferol, a metabolite of vitamin D3 with intestinal calcium transport activity. Biochemistry 9:4776–4780. 82. Tanaka Y, Frank H, DeLuca HF, Koizumi N, Ikekawa N 1975 Importance of the stereochemical position of the 24-hydroxyl to biological activity of 24-hydroxyvitamin D3. Biochemistry 14:3293–3296. 83. DeLuca HF, Schnoes HK 1983 Vitamin D: Recent advances. Annu Rev Biochem 52:411–439. 84. Brommage R, DeLuca HF 1985 Evidence that 1,25dihydroxyvitamin D3 is the physiologically active metabolite of vitamin D3. Endocr Rev 6:491–511. 85. Brommage R, Jarnagin K, DeLuca HF, Yamada S, Takayama H 1983 1- but not 24-hydroxylation of vitamin D is required for skeletal mineralization in rats. Am J Physiol 244: E298–E304. 86. Tanaka Y, Pahuja DN, Wichmann JK, DeLuca HF, Kobayashi Y, Taguchi T, Ikekawa N 1982 25-Hydroxy-26,26,26,27,27,27hexafluorovitamin D3: Biological activity in the rat. Arch Biochem Biophys 218:134–141. 87. Nakada M, Tanaka Y, DeLuca HF, Kobayashi Y, Ikekawa N 1985 Biological activities and binding properties of 23,23difluoro-25-hydroxyvitamin D3 and its lα-hydroxy derivative. Arch Biochem Biophys 241:173–178. 88. Zull JE, Czarnowska-Misztal E, DeLuca HF 1965 Actinomycin D inhibition of vitamin D action. Science 149:182–184. 89. Eisenstein R, Passavoy M 1964 Actinomycin D inhibits parathyroid hormone and vitamin D activity. Proc Soc Exp Biol Med 117:77–79. 90. Brumbaugh PF, Haussler MR 1973 lα,25-Dihydroxyvitamin D3 receptor: Competitive binding of vitamin D analogs. Life Sci 13:1737–1746. 91. Kream BE, Reynolds RD, Knutson JC, Eisman JA, DeLuca HF 1976 Intestinal cytosol binders of 1,25-dihydroxyvitamin D3 and 25-hydroxyvitamin D3. Arch Biochem Biophys 176: 779–787. 92. McDonnell DP, Mangelsdorf DJ, Pike JW, Haussler MR, O’Malley BW 1987 Molecular cloning of complementary
93.
94. 95. 96.
97. 98. 99.
100.
101.
102.
103.
104.
105.
106.
107.
DNA encoding the avian receptor for vitamin D. Science 235:1214–1217. Baker AR, McDonnell DP, Hughes M, Crisp TM, Mangelsdorf DJ, Haussler MR, Pike JW, Shine J, O’Malley BW 1988 Cloning and expression of full-length cDNA encoding human vitamin D receptor. Proc Natl Acad Sci USA 85: 3294–3298. Burmester JK, Maeda N, DeLuca HF 1988 Isolation and expression of rat 1,25-dihydroxyvitamin D3 receptor cDNA. Proc Natl Acad Sci USA 85:1005–1009. Burmester JK, Wiese RJ, Maeda N, DeLuca HF 1988 Structure and regulation of the rat 1,25-dihydroxyvitamin D3 receptor. Proc Natl Acad Sci USA 85:9499–9502. Miyamoto K, Kesterson RA, Yamamoto H, Nishiwaki E, Tatsumi S, Taketani Y, Morita K, Pike JW, Takeda E 1997 Structural organization of the human vitamin D receptor chromosomal gene and its promoter. Mol Endocrinol 11:1165–1179. Jehan F, DeLuca HF 1997 Cloning and characterization of the mouse vitamin D receptor promoter. Proc Natl Acad Sci USA 94:10138–10143. Jehan F, DeLuca HF 2000 The mouse vitamin D receptor is mainly expressed through an Sp1-driven promoter in vivo. Arch Biochem Biophys 377:273–283. Yoshizawa T, Handa Y, Uematsu Y, Takeda S, Sekine K, Yoshihara Y, Kawakami T, Arioka K, Sato H, Uchiyama Y, Masushige S, Fukamizu A, Matsumoto T, Kato S 1997 Mice lacking the vitamin D receptor exhibit impaired bone formation, uterine hypoplasia and growth retardation after weaning. Nat Genet 16:391–396. Li YC, Pirro AE, Amling M, Delling G, Baron R, Bronson R, Demay MB 1997 Targeted ablation of the vitamin D receptor: An animal model of vitamin D-dependent rickets type II with alopecia. Proc Natl Acad Sci USA 94:9831–9835. Brooks MH, Bell NH, Love L, Stern PH, Orfei E, Queener SF, Hamstra AJ, DeLuca HF 1978 Vitamin D-dependent rickets Type II. Resistance of target organs to 1,25-dihydroxyvitamin D. N Engl J Med 298:996–999. Hughes MR, Malloy PJ, Kieback DG, Kesterson RA, Pike JW, Feldman D, O’Malley BW 1988 Point mutations in the human vitamin D receptor gene associated with hypocalcemic rickets. Science 242:1702–1706. Wiese RJ, Goto H, Prahl JM, Marx SJ, Thomas M, Al-Aqeel A, DeLuca HF 1993 Vitamin D–dependency rickets Type II: Truncated vitamin D receptor in three kindreds. Mol Cell Endocrinol 90:197–201. Demay MB, Gerardi JM, DeLuca HF, Kronenberg HM 1990 DNA sequences in the rat osteocalcin gene that bind the 1,25-dihydroxyvitamin D3 receptor and confer responsiveness to 1,25-dihydroxyvitamin D3. Proc Natl Acad Sci USA 87:369–373. Noda M, Vogel RL, Craig AM, Prahl J, DeLuca HF, Denhardt DT 1990 Identification of a DNA sequence responsible for binding of the 1,25-dihydroxyvitamin D3 receptor and 1,25-dihydroxyvitamin D3 enhancement of mouse secreted phosphoprotein 1 (Spp1, osteopontin) gene expression. Proc Natl Acad Sci USA 87:9995–9999. Demay MB, Kiernan MS, DeLuca HF, Kronenberg HM 1992 Sequences in the human parathyroid hormone gene that bind the 1,25-dihydroxyvitamin D3 receptor and mediate transcriptional repression in response to 1,25-dihydroxyvitamin D3. Proc Natl Acad Sci USA 89:8097–8101. Zierold C, Darwish HM, DeLuca HF 1995 Two vitamin D response elements function in the rat 1,25-dihydroxyvitamin D 24-hydroxylase promoter. J Biol Chem 270:1675–1678.
12 108. Umesono K, Murakami KK, Thompson CC, Evans RM 1991 Direct repeats as selective response elements for the thyroid hormone, retinoic acid, and vitamin D3 receptors. Cell 65:1255–1266. 109. Yu VC, Delsert C, Andersen B, Holloway JM, Devary OV, Naar AM, Kim SY, Boutin J-M, Glass CK, Rosenfeld MG 1991 RXRβ: A coregulator that enhances binding of retinoic acid, thyroid hormone, and vitamin D receptors to their cognate response elements. Cell 67:1251–1266. 110. Munder M, Herzberg IM, Zierold C, Moss VE, Hanson K, Clagett-Dame M, DeLuca HF 1995 Identification of the porcine intestinal accessory factor that enables DNA
HECTOR F. DELUCA
sequence recognition by vitamin D receptor. Proc Natl Acad Sci USA 93:2796–2799. 111. Jin CH, Pike JW 1996 Human vitamin D receptor-dependent transactivation in Saccharomyces cerevisiae requires retinoid X receptor. Mol Endocrinol 10:196–205. 112. DiRenzo J, Soderstrom M, Kurokawa R, Ogliastro M-H, Ricote M, Ingrey S, Horlein A, Rosenfeld MG, Glass CK 1997 Peroxisome proliferator-activated receptors and retinoic acid receptors differentially control the interactions of retinoid X receptor heterodimers with ligands, coactivators, and corepressors. Mol Cell Biol 17: 2166–2176.
CHAPTER 2
Vitamin D Metabolism RONALD L. HORST AND TIMOTHY A. REINHARDT U.S. Department of Agriculture, Agricultural Research Service, National Animal Disease Center, Ames, Iowa
G. SATYANARAYANA REDDY Brown University, Department of Chemistry, Providence, Rhode Island
I. Introduction II. Vitamin D Metabolism III. Vitamin D Toxicity
IV. Species Variation in Vitamin D Metabolism and Action V. Conclusions References
I. INTRODUCTION
II. VITAMIN D METABOLISM
In 1919, when the field of experimental nutrition was still in its infancy, Sir Edward Mellanby conducted a classic experiment that for the first time associated the supplementation of various growth-promoting fats with the prevention of rickets [1]. He credited the cure to the presence of a fat-soluble substance called vitamin A. McCollum et al. [2], however, later discovered that the factor responsible for healing rickets was distinct from vitamin A. McCollum named this new substance vitamin D. It was also during this period when scientists realized that there were two antirachitic factors with distinct structures. As discussed by Norman [3], the first factor to be identified was designated vitamin D2 (also known as ergocalciferol), whereas the structure of vitamin D3 (cholecalciferol) became evident some 4 to 5 years later. Vitamins D3 and D2 are used for supplementation of animal and human diets in the United States. Vitamin D3 is the form of vitamin D that is synthesized by vertebrates, whereas vitamin D2 is the major naturally occurring form of the vitamin in plants. Animals that bask in the sun such as amphibia, reptiles, and birds therefore synthesize sufficient endogenous vitamin D3 to meet their daily needs. However, herbivores may have evolved utilizing vitamin D2 as their predominant source. This chapter focuses on the general control and function of key enzymes involved in the regulation of vitamin D2 and vitamin D3 metabolism. Species differences in vitamin D metabolism, as well as vitamin D toxicity, are also discussed. The reader is also directed toward a number of additional reviews regarding vitamin D metabolism and action [3–8]. This chapter gives an overview of vitamin D metabolism; critical steps are discussed in further detail in the subsequent chapters of this section. Metabolism of vitamin D analogs is covered in Chapter 81.
A. Overview
VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX,
Vitamin D refers to a group of compounds that possess antirachitic activity. Technically vitamin D is classified as a secosteroid. Secosteroids are those in which one of the rings has been broken; in vitamin D, the 9,10 carbon–carbon bond of ring B is broken, and it is indicated by the inclusion of “9,10-seco” in the official nomenclature. There are several known nutritional forms of vitamin D, however, the best known examples are cholecalciferol (vitamin D3), which is produced in the skin, and ergocalciferol (vitamin D2), which is derived from plant tissues (Fig. 1). Therefore, when reference is made to vitamin D, the lack of a subscript usually implies either vitamin D2 or vitamin D3. The vitamin Ds are named according to the rules of the International Union of Pure and Applied Chemists (IUPAC) for steroid nomenclature [9]. Because vitamin D is derived from a steroid, the structure retains its numbering from the parent compound cholesterol. Configurations at asymmetric centers are designated by using the R and S notation applying the sequence-rule procedure [10]. Configuration of the double bonds are notated E for “entgegen” or trans, and Z for “zuzammen” or cis [11]. Thus the official name of vitamin D3, by relation to cholesterol, is 9,10-seco(5Z,7E)5,7,10(19) cholestatriene-3β-ol, and the official name of vitamin D2 is 9,10-seco(5Z,7E)-5,7,10(19), 22-ergostatetraene-3β-ol. Contemporary views categorize vitamin D3 not as a vitamin but, rather, as a prosteroid hormone. This concept is supported by the fact that in mammals vitamin D3 is derived from 7-dehydrocholesterol (the precursor of cholesterol) present in the skin. The direct action of sunlight on 7-dehydrocholesterol results in cleavage Copyright © 2005, Elsevier, Inc. All rights reserved.
16
RONALD L. HORST, TIMOTHY A. REINHARDT, AND G. SATYANARAYANA REDDY
FIGURE 1
Important nutritional forms of vitamin D.
of the B ring of the steroid structure that on thermoisomerization yields vitamin D3 (see Chapter 3). The significance of vitamin D as a prosteroid hormone became clearer in 1967 when Morii et al. [11a] isolated a new metabolite of vitamin D3 from rats that was as effective as vitamin D3 in healing rickets, raising blood calcium, and increasing intestinal calcium transport. This compound acted more rapidly than vitamin D3, requiring only 8 to 10 hr after oral administration to initiate its response. This metabolite was identified as 25-hydroxyvitamin D3 (25OHD3) [11b]. The liver was demonstrated to be important in the production of this most abundant circulating form of vitamin D3 that, under normal conditions, is present at 20 to 50 ng/ml [4]. Shortly following the discovery of 25OHD3, a number of laboratories showed that this metabolite is specifically hydroxylated at the lα-position in the kidney to yield 1,25-dihydroxyvitamin D3 [1,25(OH)2D3] [11c–13]. The latter metabolite is now generally accepted as the hormonally active form of vitamin D3. Its importance is evidenced by the genetic disorder pseudo vitamin D–deficiency rickets (PDDR), which is caused by mutations in the 1α-hydroxylase gene. PDDR results in the inability to produce 1,25(OH)2D leading to severe rickets (see Chapters 71 and 72). In normal human plasma, 1,25(OH)2D3 circulates at approximately 1000-fold lower concentrations than 25OHD3 and is generally present at 20 to 65 pg/ml [14]. This simplistic picture outlined for vitamin D3 activation is complicated by the fact that vitamin D3 can be oxidatively metabolized to a variety of products. Most of these numerous metabolites have no identifiable
biological function, and indeed many have been isolated from animals fed abnormally high amounts of vitamin D3. Nevertheless, the evidence collected to date indicates that 25-hydroxylated vitamin D3 metabolites are preferentially metabolized at the side chain. In particular, carbon centers C-23, C-24, and C-26 are readily susceptible to further oxidation. Figure 2 illustrates products of these oxidative pathways. As indicated, these pathways are shared by both 25OHD3 and 1,25(OH)2D3, and their physiological importance is still a matter of controversy. For example, there is evidence that 24,25-dihydroxyvitamin D3 [24,25(OH)2D3] may function to stimulate bone mineralization [15,16], suppress parathyroid hormone (PTH) secretion [17], and maintain embryonic development [18]. For the most part, however, the C-24 hydroxylation and other side-chain modifications are generally considered to be catabolic in nature and play a key role in maintaining vitamin D homeostasis [19]. Although these side-chain oxidative pathways yield metabolites that are considered “nonfunctional,” the presence of these compounds in circulation could pose serious problems in the analysis for 25OHD3 and 1,25(OH)2D3 [20]. Further complicating the issue of understanding vitamin D activation, catabolism, and metabolite analysis is the presence of vitamin D2. Vitamin D2 has been shown to contribute significantly to the overall vitamin D status in humans and other mammals consuming supplemental vitamin D2 [21–23]. Vitamin D2 can also be metabolized in a similar fashion to produce several metabolites analogous to the vitamin D3 endocrine system, including the hormonally
17
CHAPTER 2 Vitamin D Metabolism
OH
−H2 O
OH (1α)
OH
(3β)HO
OH
OH(1α)
(3α)HO
1,25(OH)2D3
OH(1α)
1,25(OH)2-3-epi-D3 −H2O
OH
OH (1α)
FIGURE 2 Pathways of vitamin D3 metabolism.
active form of vitamin D2, 1,25-dihydroxyvitamin D2 [1,25(OH)2D2] [24]. Simple inspection of the side chain, however, would imply that differences between metabolism of vitamin D2 and vitamin D3 may exist. The presence of unsaturation at carbon centers C-22/C-23, along with the additional methyl group at C-24, would seem to preclude the existence of the same metabolic pathways for the two vitamins. Figure 3 outlines some of the known pathways of vitamin D2 metabolism that have been shown to date. Deviations in the vitamin D2 and vitamin D3 pathways are discussed in detail in the following sections.
B. 25-Hydroxylase The 25-hydroxylation of vitamin D is the initial step in vitamin D activation. The enzyme responsible for production of this metabolite is located in the liver (see Chapter 4). Extrahepatic sources of 25-hydroxylation have been described [25]; however, experiments with hepatectomized rats provided evidence that the liver is the major, if not the sole, physiologically relevant site of 25-hydroxylation of vitamin D [26]. Subsequent studies
also described the existence of the 25-hydroxylase in both liver mitochondria and microsomes [27–31]. In early work, the microsomal enzyme was described as an enzyme of low capacity and high affinity and, therefore, the enzyme of greatest physiological importance [31]. In contrast, the mitochondrial enzyme was described as a high-capacity, low-affinity enzyme thought to be relevant only under conditions of high vitamin D concentration such as vitamin D toxicity [32]. Early evidence that the microsomal enzyme was the physiologically relevant enzyme came from experiments that suggested this enzyme could be regulated by vitamin D status [31]. It is now clear that liver production of 25-hydroxyvitamin D (25OHD) is not significantly regulated. 25OHD production is primarily dependent on substrate concentration. An important consequence of this lack of physiological regulation of 25OHD is that measurement of blood 25OHD is an excellent measure of vitamin D nutritional status. The purification and cloning of putative liver 25-hydroxylases have been reviewed several times [33–37]. Examination of the literature shows that most of the focus is on the mitochondrial 25-hydroxylase
18
RONALD L. HORST, TIMOTHY A. REINHARDT, AND G. SATYANARAYANA REDDY
FIGURE 3 Pathways of vitamin D2 metabolism.
designated CYP27A1, which is a cytochrome P450 capable of C-26(27) hydroxylation of sterols involved in bile acid synthesis and the 25-hydroxylation of vitamin D3. The rat, rabbit, and human enzymes have been cloned [38–41]. The CYP27A1 clone has been expressed
in COS cells [39,42], and its activity has been isolated from the mitochondria of these cells. The expressed enzyme was found to 27-hydroxylate cholestanetriol and 25-hydroxylate vitamin D3. However, CYP27A1 does not 25-hydroxylate vitamin D2 [39]. Rather, CYP27A1
CHAPTER 2 Vitamin D Metabolism
was found to 24-hydroxylate and 26(27)-hydroxylate vitamin D2. These activities could explain the presence of 24-hydroxyvitamin D2 (24OHD2), 1,24-dihydroxyvitamin D2 [1,24(OH)2D2], and 24,26-dihydroxyvitamin D2 [24,26(OH)2D2] [43–45] in the plasma of rats and cows. Since rats fed vitamin D–deficient diets and supplemented with physiological amounts of vitamin D2 have 25OHD2 as their predominant monohydroxylated vitamin D2 metabolite in the plasma [44] and targeted disruption of CYP27A1 does not decrease serum 25OHD3 [46], CYP27A1 is likely not the physiologic enzyme responsible for the 25-hydroxylation of vitamin D. The rat liver microsomal 25-hydroxylase (CYP2C11) has also been studied, but it has been shown to be male-specific [47]. Data have also been presented indicating that microsomes do [48] or do not [49] possess 25-hydroxylase activity. Therefore, conclusions regarding the importance of the CYP27A1 and other ostensible microsomal 25-hydroxylases require additional research. Data obtained studying pig liver 25-hydroxylation of vitamin D3 [36,37,48,50,51] suggest that a third liver 25-hydroxylase exists that is microsomal in origin. In the pig, this enzyme, CYP2D25, is present equally in males and females and is markedly different from CYP27A1 and CYP2C11 based on a terminal amino acid sequence [51]. Most important is the finding that this pig microsomal enzyme 25-hydroxylates vitamin D2 and vitamin D3 equally. The 25-hydroxylation of vitamin D is not yet completely understood. Several enzymes may play a role in the 25-hydroxylation of vitamin D. Whether one enzyme is more physiologically relevant than others remains to be determined. Studies in primary cultures of pig hepatocytes suggest that both CYP2D25 and CYP27A1 can play a role in 25-hydroxylation of vitamin D3 [51]. Nevertheless, it is clear that mammals can use vitamin D2 as a sole source of vitamin D. Therefore, any 25-hydroxylase proclaimed as the key enzyme(s) in the 25-hydroxylation of vitamin D must be capable of 25-hydroxylating vitamin D2 as well as vitamin D3. If mammals, for example, possessed an enzyme with specificity for the vitamin D2 side chain, this enzyme may be missed due to the almost exclusive use of vitamin D3 or vitamin D3 analogs as substrates for the 25-hydroxylating reaction.
C. 1α-Hydroxylase (CYP27B1) In the late 1960s, 25OHD3 was believed to be the metabolically active form of vitamin D. However, the presence of a more polar metabolite, which accumulated
19 in the intestinal mucosa chromatin of chicks administered 3H-labeled vitamin D3, suggested a new candidate for the active form of vitamin D [52]. Subsequent work by Lawson et al. [53] showed that during the formation of this metabolite the 1α-3H was lost. This led them to suggest that the new metabolite had an oxygen function inserted at C-1 in addition to the hydroxyl group at C-25. The enhanced biological activity of this new metabolite was evident before its structure could be determined [54–56]. Fraser and Kodicek [56] demonstrated that nephrectomy abolished production of the new metabolite, and this active vitamin D compound was synthesized by kidney mitochondria. In 1971, three laboratories identified the active form of vitamin D as 1,25(OH)2D3 [11,12,57]. Subsequently, the vitamin D2 form was also isolated and identified [24]. The CYP27B1 is located in the inner mitochondrial membrane of the proximal convoluted tubule cells of the kidney [58] and is discussed in detail in Chapter 5. Extrarenal sites of 1α-hydroxylation have been reported in bone, liver, placenta, macrophages, and skin [59]. The physiological significance of these sites on systemic calcium metabolism is in doubt, as nephrectomy and/or severe renal failure results in very low to undetectable circulating 1,25(OH)2D3 levels [60]. The regulation of 1,25(OH)2D3 production is reciprocally regulated with respect to 24,25(OH)2D3 [61]. Hypocalcemia caused by calcium-deficient diets, vitamin D deficiency, or pathological factors results in increased production of 1,25(OH)2D3 [61–67]. This hypocalcemic-mediated induction of 1,25-dihydroxyvitamin D [1,25(OH)2D] production is secondary to increased PTH. Administration of PTH to animals results in increased 1,25(OH)2D3 production [64,67,68]. PTH treatment in vitro induces CYP27B1 in renal slices [63] and cultured kidney cells [67,69] and is cAMP dependent [63,67,69,70]. Thyroparathyroidectomy (TPTX) or parathyroidectomy (PTX) results in the loss of the ability to synthesize 1,25(OH)2D3. In humans, acute administration of PTH or primary hyperparathyroidism results in increased production of 1,25(OH)2D [14,71], which is evidenced by elevations in plasma 1,25(OH)2D. However, in animal studies where PTH was administered chronically to goats and calves, a transient rise in plasma 1,25(OH)2D was observed followed by a rapid decline to nearly undetectable levels [72,73]. These results could be attributed to hypercalcemic feedback on the renal CYP27B1. When plasma calcium in these animals reached 13 mg/dl, 1,25(OH)2D production appeared to cease. This same group conducted similar experiments in rats and showed that chronic PTH infusion did not result in a reduction of plasma 1,25(OH)2D3, but rather a
20
RONALD L. HORST, TIMOTHY A. REINHARDT, AND G. SATYANARAYANA REDDY
modest rise [74]. Clearly, there are varying degrees of direct calcium-mediated feedback on the renal CYP27B1. It is possible that species and age affect the set points at which plasma calcium becomes a direct negative regulator of 1,25(OH)2D3 production. In contrast to the indirect role of plasma calcium in inducing 1,25(OH)2D3 production, the role of plasma phosphate appears more direct. As plasma phosphate declines, animals shift from 24,25(OH)2D production to increased 1,25(OH)2D3 production [75,76]. Since phosphate-deficient animals are hypercalcemic, serum PTH is down and therefore cannot be providing the signal to increase 1,25(OH)2D3 production. Furthermore, TPTX phosphate-deficient animals produce 1,25(OH)2D3 similarly to intact phosphate-deficient animals [66]. Gray [77] demonstrated that hypophysectomy abolished the increase in plasma 1,25(OH)2D concentrations that normally accompanied dietary phosphate deprivation. Gray demonstrated that growth hormone or triiodothyronine replacement to hypophysectomized rats restored elevations in plasma 1,25(OH)2D associated with low dietary phosphorus, therefore suggesting a permissive role of these hormones in regulation of the renal CYP27B1 during phosphorus deficiency. A direct negative effect of 1,25(OH)2D on its own production has been reported. The inhibitory effect of 1,25(OH)2D3 on renal CYP27B1 activity occurs both in vivo and in vitro [64,78]. This repressive activity is probably indirect [79]. In vivo, this effect may be partially mediated through the ability of 1,25(OH)2D3 to inhibit PTH secretion [80]. Similarly, Beckman et al. [81] showed that vitamin D toxicity mildly inhibited renal CYP27B1 activity while low-calcium diets significantly induced the CYP27B1. They further demonstrated that administration of toxic doses of vitamin D to animals fed a calcium-deficient diet reduced CYP27B1 activity by 90%. This result occurred in spite of the fact that these animals were hypocalcemic and had serum PTH levels equal to those of control animals receiving calcium-deficient normal vitamin D diets. These data suggest that high plasma concentration of vitamin D metabolites may act directly to suppress CYP27B1 activity. There are many additional factors such as calcitonin (CT), acidosis, sex steroids, prolactin, growth hormone, glucocorticoids, thyroid hormone, and pregnancy that are potential regulators of 1,25(OH)2D production. One of the most recent and interesting is the requirement for the endocytic receptor megalin in the proximal tubular cells to allow uptake of 25OHD for 1α-hydroxylation [82]. Discussion of these is beyond the scope of this general review of vitamin D metabolites. The characteristics and regulation of the CYP27B1 are described further in several reports [37,83–85].
The cloning of the renal CYP27B1 has been achieved [86–89]. Studies with CYP27B1 knockout mice [90,91] suggest that these animals have abnormalities similar to those observed in PDDR. These knockout models will undoubtedly provide new insight into the functions of 1,25(OH)2D and are the subject of further review in Chapter 7.
D. 24-Hydroxylase (CYP24A1) The 24-hydroxylation of 25OHD3 and 1,25(OH)2D3 to form 24,25(OH)2D3 [92] and 1,24,25(OH)3D3 [93,94] is the primary mechanism and the first step in a metabolic pathway to inactivate and degrade these vitamin D metabolites. It now appears that CYP24A1 is ubiquitous and may be present in every cell and tissue that contains the vitamin D receptor (VDR). In the kidney, CYP24A1 is found on the inner mitochondrial membrane of the renal tubules [95]. The primary regulators of renal CYP24A1 activity are PTH and 1,25(OH)2D3. Normal and TPTX animals receiving injections or infusions of 1,25(OH)2D3 show marked increases in both renal CYP24A1 mRNA levels and activity [63,74,96,97]. Administration of PTH partially or completely blocks expression of CYP24A1 mRNA and activity in these animals [63,68,74,96,97]. PTH acts on the kidney via adenylate cyclase and cAMP, and it has been shown that infusions of cAMP in vivo block l,25(OH)2D3mediated inductions of the renal CYP24A1 [96,98]. Animals on calcium-deficient diets have elevated plasma 1,25(OH)2D concentrations, which are accompanied by suppressed or undetectable renal CYP24A1 activity [78,96], as well as reduced VDR concentrations [99]. The reasons for the inability of 1,25(OH)2D3 to up-regulate renal CYP24A1 during calcium deficiency are not clear. Iida et al. [100] have proposed that the down-regulation of renal VDR during calcium deficiency may be responsible for preventing the l,25(OH)2D3mediated induction of renal CYP24A1. In vivo studies by Reinhardt and Horst [74], however, suggest that under these conditions PTH is probably the more important mediator of renal CYP24A1 regulation rather than downregulation of VDR. In their experiments, Reinhardt and Horst [74] showed that 1,25(OH)2D3 treatment of animals on normal calcium diets resulted in significant up-regulation of renal CYP24A1 as well as VDR. However, when PTH was infused simultaneously with 1,25(OH)2D3, VDR up-regulation was still observed (albeit to a lesser degree), whereas CYP24A1 up-regulation was completely blocked. The importance of PTH in preventing the l,25(OH)2D3-mediated up-regulation of the renal CYP24A1 is also apparent
21
CHAPTER 2 Vitamin D Metabolism
by observation in aged rats. With advancing age, renal PTH receptors are down-regulated [101,102], while VDR remains unchanged [103]. The reduction in renal PTH receptors makes the kidney less responsive to PTH [64], which is associated with significant elevations in CYP24A1 mRNA [103,104]. These data suggest that renal responsiveness to PTH, not a decline in VDR, is the major physiological regulator of the renal CYP24A1. In the intestine, 1,25(OH)2D3 is the primary regulator of CYP24A1. In vivo administration of 1,25(OH)2D3 rapidly induces intestinal CYP24A1 activity [105]. This activity peaks by 6 hr postinjection, and rapidly declines thereafter to control levels 24 hr postinjection. Time-course experiments show that CYP24A1 mRNA peaks 4 to 6 hr post-injection and then rapidly disappears [104]. This is in contrast to the renal CYP24A1 mRNA, which peaks 12 to 24 hr post-l,25(OH)2D3 treatment and declines much more slowly. Shinki et al. [96] proposed that the intestinal CYP24A1 was 100 times more sensitive to 1,25(OH)2D3 stimuli than the renal CYP24A1. However, they examined CYP24A1 mRNA only 3 hr after a 1,25(OH)2D3 dose. Because renal CYP24A1 requires an additional 6 to 12 hr to reach peak expression, they likely underestimated the true sensitivity of the kidney to a 1,25(OH)2D3 dose. In contrast to their effect in the kidney, TPTX, PTH administration, or cAMP infusion does not affect intestinal expression of CYP24A1 induced by 1,25(OH)2D3 [96]. Animals fed low-calcium diets, with the associated secondary hyperparathyroidism and high plasma 1,25(OH)2D3 concentrations, have marked inductions of both intestinal CYP24A1 mRNA and activity [96,105]. Another contrast between intestinal and renal CYP24A1 expression is seen in the aging rat model. Intestinal CYP24A1 mRNA and activity decline or change very little in the aged animal. This contrasts to the largely increased expression of renal CYP24A1 observed in the aged animal [103]. Calcitonin has been shown to be a potent suppressor of intestinal CYP24A1 expression [106]. In these experiments, Beckman et al. [81,106] showed that vitamin D toxicity was a potent inducer of CYP24A1 mRNA and enzyme expression in both intestine and kidney. They also showed that if the hypervitaminosis D3-induced hypercalcemia was prevented by feeding low-calcium diets, the intestinal CYP24A1 expression was enhanced four-fold over hypercalcemic animals receiving the same toxic doses of vitamin D3 but consuming a normal calcium diet. These observations prompted the examination of the possibility that CT released in response to the hypercalcemia may have suppressed the induced expression of the intestinal CYP24A1. In their series of experiments, Beckman et al. [106] clearly demonstrated that CT was a potent suppressor of intestinal CYP24A1 activity. Conceivably, the CT-mediated suppression of
CYP24A1 activity could enhance l,25(OH)2D-mediated activities by prolonging its half-life. This could exacerbate conditions that manifest hypercalcemia, such as hypervitaminosis D, by preventing 24-hydroxylation and catabolism of active vitamin D metabolites. The CYP24A1 has been purified [107,108] and cloned [107,109] and the clone has been expressed [107,109]. Analysis of the amino acid sequence of the rat and human CYP24A1s showed that the sequences were 90% similar. The 21-amino acid heme binding region was found to be 100% identical [109]. Ohyama et al. [110] isolated the gene encoding the rat CYP24A1. This single-copy gene was approximately 15 kb and was composed of 12 exons. Several putative vitamin D response elements have been identified and are currently under study. Details of the purification and cloning of the CYP24A1 have been reviewed previously [35], and additional review of the molecular analysis and regulation of the CYP24A1 can be found in Chapter 6 in this book.
E. Physiological Role of CYP24A1 The major site for 24-hydroxylation appears to be the kidney. This is based on the observation that nephrectomy reduced or eliminated plasma 24,25(OH)2D3 [111]. However, nephrectomy also eliminates the production of 1,25(OH)2D3, the primary stimulator of CYP24A1. Therefore, the possibility remains that 24,25(OH)2D3 may reappear in plasma of nephrectomized subjects treated with therapeutic doses of 1,25(OH)2D3. Some studies suggest that 25OHD3 is the primary substrate for CYP24A1 [112]; however, it is now generally accepted that CYP24A1 is distributed throughout the body and that 1,25(OH)2D3, rather than 25OHD3, is the preferred substrate for CYP24A1 [96,113,114]. Napoli et al. [115] and Napoli and Horst [116] identified the formation of 24-oxo-1,25(OH)2D3 and 24-oxo-l,23,25-trihydroxyvitamin D3 [24-oxo-1,23, 25(OH)3D3] from intestinal homogenates incubated with physiological amounts of 1,25(OH)2D3. The formation of 24-oxo-1,23,25(OH)3D3 from 1,25-(OH)2D3 was enhanced by treatment of experimental animals with exogenous 1,25-(OH)2D3. They suggested that 24-hydroxylation, followed by C-23 oxidation, most likely represents a mechanism for terminating the cellular action of 1,25(OH)2D3. In a review, Haussler [117] proposed a model for the cellular action of 1,25(OH)2D3 in which he suggested that receptor-mediated, selfinduced catabolism of 1,25(OH)2D3 modulates the action of 1,25(OH)2D3. The work of Lohnes and Jones [118], Reddy and Tserng [119], and Makin et al. [120] provided further support for this proposal by showing
22
RONALD L. HORST, TIMOTHY A. REINHARDT, AND G. SATYANARAYANA REDDY
the ubiquitous presence of catabolic pathways initiated by CYP24A1 in 1,25(OH)2D3 target tissues and the complete destruction of 1,25(OH)2D3 by these pathways. In fact, CYP24A1 has been shown to do more than just initiate this catabolic cascade. Akiyoshi-Shibata et al. [121] expressed the rat CYP24A1 cDNA in Escherichia coli. They found that this enzyme not only 24-hydroxylates but catalyzes the dehydrogenation of the 24-OH group and performs 23-hydroxylation resulting in 24-oxo-l,23,25(OH)3D3 production. Only the cleavage at C-23/C-24 resulting in the 24,25,26,27tetranor-1OH,23COOHD3 was not demonstrated. However, it is now recognized from the work of several groups that CYP24A1 is capable of the complete catabolism of 25OHD3 and 1,25(OH)2D3 via a fivestep reaction process that includes 24-hydroxylation, 24-oxidation, 23-hydroxylation, side-chain cleavage, and subsequent production of the final degradative product, calcitroic acid [121–125]. Direct evidence that self-induced metabolism of 1,25(OH)2D3 suppresses the action of 1,25(OH)2D3 on target cells was reported by Pols et al. [126,127] and Reinhardt and Horst [128,129]. Both laboratories showed that ketoconazole inhibited l,25(OH)2D3induced metabolism. This inhibition resulted in increased specific accumulation of 1,25(OH)2D3 in target cells and a significant increase in the cellular half-life of l,25(OH)2D3-occupied VDR [129]. A result of blocking the self-induced metabolism of 1,25(OH)2D3 was up-regulation of the VDR. Reinhardt and Horst [128] extended these studies by demonstrating that selfinduced metabolism of 1,25(OH)2D3 in target cells limits the response of target cells to a primary 1,25(OH)2D3 stimulus by reducing occupancy of VDR by 1,25(OH)2D3 and by preventing VDR up-regulation. Additionally, their data showed that entry of 1,25(OH)2D3 into the cell is restricted due to extensive metabolism of the 1,25(OH)2D3. In whole-cell VDR assays, hormone was degraded so rapidly that VDR binding was prevented. Reinhardt et al. [130] confirmed the inhibitory effects of self-induced induction of CYP24A1 on the cellular action of 1,25(OH)2D3 in vivo by demonstrating that ketoconazole potentiates the 1,25(OH)2D3 up-regulation of VDR in rat intestine and bone. Clearly, one of the primary roles of CYP24A1 catabolic pathway is terminating the actions of 1,25(OH)2D3. Recent work using CYP24A1 knockout mice has provided additional evidence for the role of CYP24A1 in the regulation of 1,25(OH)2D activity via catabolism [19]. These mice were unable to clear 1,25(OH)2D, as was evidenced by abnormally high circulating concentrations of this metabolite. These data provided support to the contention that the primary role of CYP24A1 is
initiating the catabolic pathway and terminating the actions of 1,25(OH)2D. The role of CYP24A1 as an enzyme responsible for the production of a biologically active compound, i.e., 24,25(OH)2D3, has been controversial. The work of St-Arnaud et al. [19] has also addressed part of this issue. Deficient mineralization of intramembranous bone was found in CYP24A1-ablated mice. The genetic cross of CYP24A1-ablated mice with VDR-ablated mice rescued the defective bone phenotype, strongly suggesting that the high concentrations of 1,25(OH)2D acting through VDR and not 24,25(OH)2D was the cause of the bone defect. Constitutive expression of CYP24A1 in transgenic rats surprisingly resulted in low plasma 24,25(OH)2D and 25OHD with no effect on plasma 1,25(OH)2D [131,132]. These rats developed albumineria and hyperlipidemia and suffered from reduced bone mass. The authors demonstrated that excreted albumin appeared to compete for the binding and reabsorption of the DBP-25-OHD3 complex with megalin, resulting in a loss of 25OHD3 into the urine and subsequent reduction of plasma 24,25(OH)2D3. Supplementation of these rats with 25OHD3 prevented the bone loss without changing plasma 1,25(OH)2D.
F. Other Vitamin D3 Derivatives Functionalized at C-24 In a series of experiments conducted by Wichmann et al. [133,134], a number of 24-hydroxylated derivatives were isolated from plasma of chicks made toxic with vitamin D3. These metabolites included 24OHD3, 23,24,25-trihydroxyvitamin D3 [23,24,25(OH)3D3], and 24,25,26-trihydroxyvitamin D3 [24,25,26(OH)3D3]. These metabolites have not been described in animals receiving physiological amounts of vitamin D3, and their biological significance is unknown; however, it is likely that 23,24,25(OH)3D3 and 24,25,26(OH)3D3 are metabolites of 24,25(OH)2D3. 24,25(OH)2D3 is also the probable precursor to the formation of the side chain cleavage product, 25,26,27-tri-norvitamin D324-carboxylic [135]. This metabolite has been shown to be a product of in vitro kidney perfusion using 25OHD3 as substrate. The analogous pathway, however, could not be demonstrated using 1,25(OH)2D3 as substrate (S. Reddy, personal communication, 1996). Another metabolite isolated in the experiments of Wichmann et al. [133] was 23-dehydro-25OHD3. The immediate precursor, site(s), and biological activity of this compound are unknown. Plausible sources of the 23-dehydro compound are dehydration of 24,25(OH)2D3
CHAPTER 2 Vitamin D Metabolism
or 23,25-dihydroxyvitamin D3 [23,25(OH)2D3]. It is not certain if any of the metabolites are important under physiological conditions.
G. 23-Hydroxylation The discovery of a C-23 oxidative pathway emerged much later than the other pathways and was ushered in by the identification of 23(S),25(R)25OHD3-26,23lactone [136,137], 23(S),25(R)1,25(OH)2D3-26,23lactone [138], and their respective precursors 23(S),25(OH)2D3 and 1,23(S),25(OH)3D3 [139,140]. To date, there has been no specific 23-hydroxylase identified for the vitamin D system. Rather, like other side-chain modifications, 23-hydroxylation is likely carried out by CYP24A1 [122]. The compound, 25OHD3-26,23-lactone, can be detected in plasma from normal rats, pigs, and chicks [140,141]. However, in several species, this metabolite is not expressed unless animals are consuming excessive amounts of vitamin D3 [142]. This metabolite has unique activity in that it is three- to fivefold more competitive than 25OHD3 for binding to the plasma vitamin D-binding protein (DBP) [136]. It was, therefore, the first modification of 25OHD3 that led to enhanced binding to the plasma DBP. The metabolite, 1,25(OH)2D3-26,23-lactone, has also been demonstrated under normal conditions [143], with elevated plasma concentrations occurring during exogenous administration of pharmacological amounts of 1,25(OH)2D3 [144]. The major locus for formation of C-23 hydroxylated derivatives appears to be the kidney. Horst and Littledike [142] and Napoli et al. [140] demonstrated that nephrectomy eliminated or greatly impaired the biosynthesis of 25OHD326,23-lactone when animals were treated with excess vitamin D3 or 25OHD3. They showed that this response was due to the inability of the animals to synthesize 23(S),25(OH)2D3. However, when 23(S),25(OH)2D3 was given to nephrectomized animals, the synthesis of 25OHD3-26,23-lactone was restored. These data suggested that C-23-hydroxylation occurred predominantly, but not exclusively, in the kidney, whereas extrarenal tissues are quantitatively important in the pathway leading to 25OHD3-26,23-lactone synthesis, which includes formation of the lactone intermediates 23,25,26(OH)3D3 [145] and 25OHD3-26,23-lactol [146]. Although ambiguities remain regarding the biological effects of C-24 oxidation, 23-hydroxylation appears to clearly be a deactivation event. 23-Hydroxylation is the first side-chain modification of 25OHD3 noted to substantially reduce its affinity for the plasma DBP [147]. 23-Hydroxylation of 1,25(OH)2D3 also leads to its
23 increased plasma clearance and reduced VDR binding and biological activity [148]. The role of 23-hydroxylation as a primary oxidation event for the further metabolism of 25OHD3 and 1,25(OH)2D3 is relatively minor to its role in the further metabolism of vitamin D3 metabolites that have been previously oxidized at C-24. In other words, very little production of 23,25(OH)2D3 or 1,23,25(OH)3D3 would be expected under normal conditions. Rather, the convergences of the C-24 and C-23 oxidative pathways would lead predominantly to the formation of 24-keto-1,23,25(OH)2D3 and 24-keto-1,23,25(OH)3D3 [116], which subsequently cleave to form C-23 acids [119,149]. Therefore, as indicated in Fig. 2, the C-23 oxidative pathway can lead to two different patterns of side-chain modifications for both 25OHD3 and 1,25(OH)2D3. One pathway, which is relatively minor under physiological conditions and more predominant during hypervitaminosis D3, leads through 23-hydroxylation to formation of the lactones, whereas the other more physiologically significant pathway leads through 24-hydroxylation to 23-hydroxylation of 24-keto metabolites. Other oxidized C-23 metabolites that have been identified include 23-keto derivatives of 25OHD3 and 1,25(OH)2D3. 23-Keto-25-hydroxyvitamin D3 was synthesized in vitro from 23(S)25(OH)2D3 and 23(R),25(OH)2D3 and has unique properties in that it binds with twofold higher affinity than 25OHD3 for binding sites on the plasma DBP [150]. This affinity should be compared to that for 23(S),25(OH)2D3, which binds with 6- to 10-fold less affinity. 23-Keto25OHD3 is also about fourfold more competitive than 25OHD3 for binding to the VDR. 23-Ketonization is, therefore, the first example of a side-chain modification enhancing the affinity of 25OHD3 for the VDR. This high affinity of 23-keto-25OHD3 for VDR prompted biosynthesis of 23-keto-l,25(OH)2D3 to determine if this modification might enhance binding of 1,25(OH)2D3 to VDR. Horst et al. [151] prepared this metabolite by incubating 23-keto-25OHD3 in kidney homogenates prepared from vitamin D–deficient chicks. The major metabolite was isolated and identified as 23-keto-1,25(OH)2D3 and was shown to possess about 40% the activity of 1,25(OH)2D3 for VDR binding. 23-Ketonization of 1,25(OH)2D3, therefore, reduced the affinity of 1,25(OH)2D3 to VDR rather than increased binding. 23-Keto metabolites do not appear to be synthesized under physiologic conditions, as Napoli et al. [115] could not demonstrate the presence of these metabolites from rat intestinal homogenates incubated with 1,25(OH)2D3 or from intestinal extracts from rats dosed with 1,25(OH)2D3.
24
RONALD L. HORST, TIMOTHY A. REINHARDT, AND G. SATYANARAYANA REDDY
H. C-26 Hydroxylation 26-Hydroxylation of 25OHD3 and 1,25(OH)2D3 produces 25,26(OH)2D3 [152] and 1,25,26(OH)3D3 [153], respectively. The natural product was originally assigned the 25(R) configuration. However, Partridge et al. [154] gave the assignment as 25(S). Ikekawa et al. [155] later discovered that the naturally occurring 25,26(OH)2D3 actually existed as a mixture of 25(S) and 25(R) isomers. Although this assignment seems somewhat trivial, it was important in unraveling a controversy that existed regarding the physiological precursor to the in vivo synthesis of 25OHD3-26,23-lactone. Hollis et al. [156] demonstrated that 25,26(OH)2D3 isolated from in vivo sources could act as a precursor to the formation of the 25OHD3-26,23-lactone. Subsequent research, however, suggested that synthetic 25(S),26(OH)2D3 (which at the time was thought to be the natural configuration) did not act as precursor to the formation of 25OHD3-26,23-lactone [157], but synthetic 25(R),26(OH)2D3 could act as a precursor [140,158]. As naturally occurring 25,26(OH)2D3 is a mixture of the R and S isomers, this research validated the conclusion of Hollis et al. [156] suggesting that formation of 25OHD3-26,23-lactone could indeed proceed through 25,26(OH)2D3. This pathway has been shown to be relatively minor [159,160], with the major pathway to 25OHD3-26,23-lactone synthesis proceeding through 23(S),25(OH)2D3 [140,145,158]. The major locus for the 26-hydroxylase is unknown. Blood concentrations of 25,26(OH)2D3 are not depressed in nephrectomized humans or pigs [142,161,162]. Therefore, production of these metabolites must take place at extrarenal sources. 26-Hydroxylase activity has, however, been demonstrated in microsomes isolated from rat and pig kidneys [163]. The only extrarenal source was reported in liver mitochondria [164]. The physiological role of the C-26 oxidative pathway remains elusive. However, 25,26(OH)2D3 and 1,25,26(OH)3D3 have been shown to possess biological activity with regard to stimulating bone calcium resorption and intestinal calcium absorption, albeit to a lesser degree than either 25OHD3 or 1,25(OH)2D3 [152,165]. Therefore, it seems unlikely that 26-hydroxylation is essential for calcium uptake from the gut or release of calcium from bone.
I. C-3 Epimerization Reddy et al. [159,160] have reported the metabolism of 1,25(OH)2D3 in primary cultures of neonatal human keratinocytes and rat osteosarcoma cells into the novel A-ring modified metabolite, 1,25(OH)2-3-epi-D3.
This epimer is formed as a result of the change in the orientation of the C-3 hydroxy group from β to α. Other investigators also confirmed this finding [166,167]. Epimerization of hydroxy groups is a wellknown phenomenon in bile acid metabolism [168] and the reaction is conducted by bile acid hydroxysteroid dehydrogenase. Figure 4 outlines the proposed pathways of C-3 epimerization of 1,25(OH)2D3 as described by Reddy et al. [159]. Although there are two potential pathways for the production of 1,25(OH)2-3-epi-D3 from 1,25(OH)2D3, the most likely pathway is through keto intermediates. The C-3 epimerization has been shown to play a major role in hormone activation and inactivation in other steroid systems [169]. Indeed, 1,25(OH)2-3-epi-D3 binds to the cellular 1,25(OH)2D3 receptor (VDR) with less affinity [170,171] and has minimal activity at activating intestinal calcium transport and bone calcium resorption than 1,25(OH)2D3 [170]. On the other hand, 1,25(OH)2-epi-D3 is equipotent to 1,25(OH)2D3 at suppressing parathyroid hormone secretion in bovine parathyroid cells [171] and at inhibiting keratinocyte proliferation [170,172]. While 1,25(OH)2-3-epi-D3 does undergo side-chain metabolism [159,171], its conversion to C-23 and C-24 oxidized metabolites occurs at a slower rate than 1,25(OH)2D3 [171]; therefore, enhanced metabolic stability of the 1,25(OH)2-3-epi-D3 has been proposed as a possible explanation for the high in vitro activity in spite of its reduced binding affinity for VDR [171]. Thus, the enzyme(s) responsible for C-3 epimerization appears to play an important role not only in the regulation of intracellular concentration of 1,25(OH)2D3 but also in the formation of metabolites with a different biological activity profile in specific target tissues. These differences in intracellular concentration of 1,25(OH)2D3 and its metabolites from one tissue to another may be one possible explanation for the well-known tissuespecific actions of 1,25(OH)2D3.
J. Unique Aspects of Vitamin D2 Metabolism The 24 position of vitamin D2, in contrast to the similar position in vitamin D3, can be considered to be highly reactive. It is a tertiary carbon as well as an allylic position, and the formation of a reactive intermediate (radical, cation) at this position would be highly stabilized. The proximity of this reactive center to the 25 position would afford the possibility of C-24-hydroxylation of vitamin D2, but the presence of the C-24 methyl would preclude further oxidation to C-24-keto compounds as is known to occur in vitamin D3 metabolism. Jones et al. [43] were the first to demonstrate C-24 oxidation when they isolated 24OHD2
25
CHAPTER 2 Vitamin D Metabolism
FIGURE 4
Possible pathways of epimerization at C-3 of 1,25(OH)2D3 (adapted from [159]).
from the plasma of male rats treated with 100 IU of radiolabeled vitamin D2. Engstrom and Koszewski [173] have determined that production of 24OHD2 can occur in liver homogenates from a variety of species, and actually exceeds the formation of 25OHD2. Horst et al. [44] have shown that the concentration of 24OHD2 in plasma was about 20% that of 25OHD2 in rats receiving physiological doses of vitamin D2, and was equivalent to 25OHD2 in rats receiving pharmacological doses of vitamin D2. They also demonstrated that 1α-hydroxylation of 24OHD2 to form 1,24(OH)2D2 represented a minor but significant pathway for vitamin D2 activation. In their experiments, they determined that 1,24(OH)2D2 rivaled both 1,25(OH)2D2 and 1,25(OH)2D3 in biopotency. Both 24OHD2 and 25OHD2 and their 1α-hydroxylated metabolites can undergo subsequent hydroxylation to form 24,25-dihydroxyvitamin D2 [24,25(OH)2D2] and 1,24,25-trihydroxyvitamin D2 [1,24,25(OH)3D2]. The formation of 1,24,25(OH)3D2 represented an unequivocal deactivation of the vitamin D2 molecule [174]. Conversely, the comparable vitamin D3 metabolite, 1,24,25(OH)3D3, maintains significant biological activity and must undergo further side-chain oxidation to be rendered totally inactive [174]. Although vitamin D2 is known to undergo side-chain oxidation, only recently has evidence emerged suggesting that vitamin D2 (like vitamin D3) undergoes side-chain cleavage.
The paucity of information regarding vitamin D2 sidechain metabolism is primarily due to the lack of appropriate radiolabeled and cold substrates. Zimmerman et al. [175] have used [9,11-3H]-1,25-(OH)2D2 and cold 1,25(OH)2D2 as substrates to determine if 1,25-(OH)2D2 undergoes side-chain cleavage. Similar experiments were done using [3α-3H]-25OHD2 as substrate [176]. In these metabolism experiments it became clear that aqueous-soluble metabolites were being produced from 1,25(OH)2D2 [175] and 25OHD2 [176]. Furthermore, Zimmerman et al. [175] demonstrated that calcitroic acid was the major aqueous-soluble metabolite being produced from cell culture and kidney perfusion experiments. In a more recent study Horst et al. [177] studied the metabolism of 1,25(OH)2D3 and 1,25(OH)2D2 using a purified rat CYP24A1 system. As expected, 1,23(OH)2-24,25,26,27-tetranor D and calcitroic acid were the major lipid- and aqueous-soluble metabolites, respectively, when 1,25(OH)2D3 was used as substrate. However, when 1,25-(OH)2D2 was used as substrate, 1,24,25(OH)3D2 was the major lipid-soluble metabolite with no evidence for the production of either 1,23(OH)2-24,25,26,27-tetranor D or calcitroic acid. Apparently, the CYP24A1 was able to 24-hydroxylate 1,25(OH)2D2, but was unable to cause further changes that would result in side-chain cleavage. When the analog 1,25-(OH)2-22-ene-D3 was used as substrate, CYP24A1 was again able to effect 24-hydroxylation
26
RONALD L. HORST, TIMOTHY A. REINHARDT, AND G. SATYANARAYANA REDDY
but not side-chain cleavage (R. Horst, S. Reddy, and J. Omdahl, 2003, unpublished data). Sunita Rao et al. [178] demonstrated that 1,25(OH)2-22-ene-D3 could be metabolized to calcitroic acid by RWLue-4 cells and rat kidney. They suggested that the 1,25(OH)2-22-ene-D3 was first hydroxylated at C-24, followed by further oxidation to 1,25(OH)2-24-oxo-22-ene-D3 prior to side-chain, double-bond reduction to form 1,25(OH)224-oxo-D3. The 1,25(OH)2-24-oxo-D3 was then further metabolized to calcitroic acid, presumably by CYP24A1. The compound 1,25(OH)2D4 [a.k.a. 22,23 dihydro1,25-(OH)2D2] has also been shown to undergo sidechain oxidation similar to that of 1,25(OH)2D2 in vitro [179] and metabolized to calcitroic acid in vivo [180]. These results suggest that metabolism of 1,25(OH)2D2 to calcitroic acid clearly involves enzymes other than CYP24A1. The compounds 25,26-dihydroxyvitamin D2 [25,26(OH)2D2] and 1,25,26-trihydroxyvitamin D2 [1,25,26(OH)3D2] have been chemically synthesized [181,182]. The metabolite 25,26(OH)2D2 has also been tentatively, but not exhaustively, identified from in vivo sources [141,183], but it could not be demonstrated in kidneys perfused with 25OHD2 [184]. Similarly, 1,25,26(OH)3D2 could not be demonstrated [185]. However, it is clear that 26-hydroxylation does occur when vitamin D2 metabolites have been previously 24-hydroxylated. For example, when 24,25(OH)2D2 and 1,24,25(OH)3D2 were used as precursors, the metabolites 24,25,26(OH)3D2 and 1,24,25,26(OH)4D2, respectively, were produced in significant amounts [184,185]. Similarly, Koszewski et al. [45] and Jones et al. [186] demonstrated that C-26 hydroxylation was the major metabolic pathway for the further metabolism of 24OHD2 and 1,24(OH)2D2. Oxidation at C-24, therefore, appears to be a prerequisite for C-26 oxidation of vitamin D2 compounds. A similar situation also appears to exist for C-28 oxidation as demonstrated by Reddy and Tserng [184,185]. In their experiments, they isolated 24,25,28(OH)3D2 and 1,24,25,28(OH)4D2 from rat kidney perfusions and, through the use of various substrates, were able to show C-28 hydroxylation of vitamin D2 metabolites occurs only after C-24 hydroxylation [187].
III. VITAMIN D TOXICITY A. Overview The first documented reports of vitamin D intoxication were made in the late 1920s by Kreitmeir and Moll [188] and Putscher [189]. These cases resulted from the ingestion of large quantities of vitamin D in
the diet. Vitamin D intoxication, however, has never been reported following prolonged sunlight exposure. Holick et al. [190] suggested that nature has provided various control points that prevent the overproduction of vitamin D3 by the skin. The most important point of control is the diversion of vitamin D3 production from 7-dehydrocholesterol to non–biologically active overirradiation products such as lumisterol and tachysterol. In addition, these authors suggested that skin pigmentation and latitude were also significant determinants (albeit to a lesser degree) that limit the cutaneous production of vitamin D3. As discussed earlier, once vitamin D is in circulation, the conversion to 25OHD is relatively uncontrolled. Normally, 25OHD circulates at 30 to 50 ng/ml in most species [141]. However, when vitamin D is given in excess, plasma 25OHD can be elevated to concentrations of 1000 ng/ml or greater [191,192], while plasma 1,25(OH)2D remains at or below normal concentrations [193]. When circulating at very high concentrations, 25OHD can compete effectively with 1,25(OH)2D for binding to the VDR. Therefore, during vitamin D toxicosis, 25OHD can induce actions usually attributed to 1,25(OH)2D [193]. High circulating 25OHD can, therefore, explain how humans with low circulating concentrations of 1,25(OH)2D can show signs of vitamin D toxicity [193], and why anephric humans [who are incapable of producing 1,25(OH)2D] can become vitamin D toxic [194]. Clinical aspects of vitamin D toxicity are discussed in Chapter 78. Although it is generally accepted that 1,25(OH)2D is reduced during hypervitaminosis, a notable exception to this generalization is the ruminant [195,196]. Horst and co-workers [195,197] have shown that vitamin D3 intoxication initiated by giving 15 million IU of vitamin D3 intramuscularly (i.m.) results in significant elevations in plasma 1,25(OH)2D3. In contrast, pigs given the same i.m. dose showed a reduction in plasma 1,25(OH)2D3, as was observed in other species [197]. Therefore, elevations in plasma 1,25(OH)2D may play a significant role in the pathogenesis of vitamin D toxicity in ruminants.
B. Differences in Toxicity between Vitamins D2 and D3 Most research dealing with utilization of vitamins D2 and D3 assumes that these two forms are equally potent in most mammals. However, when large and potentially toxic doses were administered orally to rhesus monkeys [198] and horses [199], or were used in treating childhood osteodystrophy [200], vitamin D2
27
CHAPTER 2 Vitamin D Metabolism
presented fewer hypercalcemic side effects than vitamin D3. In addition, 1OHD2, which is as effective as the 1OHD3 in standard bioassays, was shown to be five- to 15-fold less toxic than 1OHD3 [201]. Studies by Horst et al. [44] provide some insight into the difference between vitamin D2 and vitamin D3 toxicity. They demonstrated that under physiological conditions the predominant monohydroxylated form of both vitamins D2 and D3 is 25OHD. In the vitamin D2–dosed rats, 24OHD2 accounted for approximately 20% of the monohydroxylated metabolites, whereas 24OHD3 could not be detected in the vitamin D3–dosed rats. When a modest superphysiological dose (800 IU/day) of vitamin D3 was given to rats, 25OHD3 remained the predominant metabolite in vitamin D3–dosed rats and was present at 26.3 ng/ml. Under these conditions, there was still no evidence for the presence of 24OHD3. However, when the same amount of vitamin D2 was given, the concentrations of 24OHD2 (14.1 ng/ml) nearly matched those of 25OHD2 (15.9 ng/ml). Interestingly, the combined concentrations of 24OHD2 and 25OHD2 in the vitamin D2–dosed animals (~30 ng/ml) was similar to the 25OHD3 concentration (26.3 ng/ml) in the vitamin D3–dosed rat. In standard assays, 25OHD2 and 25OHD3 are equipotent at displacing 3H-1,25(OH)2D3 from the calf thymus VDR. However, 24OHD2 has been shown to have at least a twofold lower affinity for binding to the calf thymus VDR (R. L. Horst, unpublished data, 1996). Therefore, the reduced toxicity of vitamin D2 is probably a result of diverting metabolism away from the production of 25OHD2 in favor of 24OHD2, which has a relatively limited affinity for VDR (a step necessary for the initiation of a biological response). Further differences between vitamins D2 and D3 were noted in their ability to up-regulate the VDR. Beckman et al. [202] found that VDR was significantly more enhanced in animals fed excess vitamin D3 relative to those animals receiving an equivalent amount of vitamin D2. Increased VDR would potentially accentuate toxic side effects by enhancing the responsiveness of intestinal tissue to the elevated 25OHD.
C. Factors Affecting Toxicity The severity of the effects and pathogenic lesions in vitamin D intoxication depend on such factors as the type of vitamin D (vitamin D2 versus vitamin D3), the dose, the functional state of the kidneys, and the composition of the diet. Vitamin D toxicity is enhanced by a rich dietary supply of calcium and phosphorus, and it is reduced when the diet is low in calcium and phosphorus [203,204]. Toxicity is also reduced when the
vitamin is accompanied by high intakes of vitamin A or by thyroxine injections [205]. The route of administration also influences toxicity. Parenteral administration of 15 million IU of vitamin D3 in a single dose caused toxicity and death in many pregnant dairy cows [195]. On the other hand, oral administration of 20 to 30 million IU of vitamin D2 daily for 7 days resulted in little or no toxicity in pregnant dairy cows [206]. Napoli et al. [207] have shown that rumen microbes are capable of metabolizing vitamin D to the inactive 10-keto-19-norvitamin D3. Parenteral administration would circumvent the deactivation of vitamin D by rumen microbes and may partially explain the difference in toxicity between oral and parenteral vitamin D. Various measures have been used in human medicine for treatment of vitamin D toxicity. These measures are mainly concerned with management of hypercalcemia. Vitamin D withdrawal is obviously indicated. It is usually not immediately successful, however, owing to the long plasma half-life of vitamin D (5 to 7 days) and 25OHD (20 to 30 days). This is in contrast to the short plasma half-life of 1OHD3 (1 to 2 days) and 1,25(OH)2D3 (4 to 8 hr). Because intestinal absorption of calcium contributes to hypercalcemia, a prompt reduction in dietary calcium is indicated. Sodium phytase, an agent that reduces intestinal calcium absorption, has also been used successfully in vitamin D toxicity management in monogastrics [208]. This treatment would be of little benefit to ruminants because of the presence of rumen microbial phytases. There have also been reports that CT [209], glucagon [210], etidronate [211], and glucocorticoid therapy [212] reduce serum calcium levels or prevent the calcinosis resulting from vitamin D intoxication (see Chapter 78).
IV. SPECIES VARIATION IN VITAMIN D METABOLISM AND ACTION Most concepts of vitamin D metabolism and function have been developed with the rat and/or chick as experimental models. Studying vitamin D metabolism is hampered by the paucity of data on the normal circulating levels of vitamin D metabolites in birds, mammals, and reptiles under normal conditions. Most recent research has focused on the analysis of 25OHD and 1,25(OH)2D as indicators of vitamin D status or aberrant physiological states. Table I summarizes the concentrations of the two metabolites that have been reported for several species by various laboratories. Close inspection of the information suggests that some mammals (mole rat, wild wood vole, horse, and wild wood mouse) and aquatic species (lamprey, carp,
28
RONALD L. HORST, TIMOTHY A. REINHARDT, AND G. SATYANARAYANA REDDY
TABLE I Plasma 25-Hydroxyvitamin D and 1,25-Dihydroxyvitamin D Concentrations in Several Species of Animals Concentration Species
25OHD (ng/ml) 1,25(OH)2D (pg/ml) Ref.
Human Rhesus monkey Rhesus monkey Marmoset Marmoset Wild woodmouse Wild bank vole Mole rat Lamprey Shark Leopard shark Horned shark Carp Bastard halibut Atlantic cod Bullfrog (mature) Soft-shelled turtle Turkey Chicken Cow Sheep Pig Horse aND,
32 188 50 90 64 <5 <5 <2 NDa ND 56 33 ND ND <2 2 16 26 27 43 27 76 6
31 207 95 400 640 <10 <5 17 274 87 3 6 174 192 59 21 12 52 21 38 36 60 19
[162] [231] [232] [232] [233] [234] [234] [213] [235] [235] [236] [236] [235] [235] [237] [235] [235] [141] [141] [141] [141] [141] [238]
Not done.
halibut, and bullfrog) appear to have very low or undetectable concentrations of 25OHD, and yet these animals appeared to be normal with no evidence of vitamin D deficiency. It is questionable whether some of these species have a requirement for vitamin D. The damara mole rat, for example, is a subterranean herbivore that in its natural habitat has no access to any obvious source of vitamin D and consumes a diet of roots and tubers [213]. These animals exhibit a high apparent calcium absorption efficiency (91%) and, like the horse and rabbit, actually use renal calcium excretion as the major regulator of calcium homeostasis [214,215]. In studies with rabbits consuming adequate amounts of calcium, it is very difficult to develop any overt or histological signs of vitamin D deficiency, and vitamin D may play a minor, if any, role in normal day-to-day functions in these animals [216].
In the wild, most animals do not have a dietary need for vitamin D, as sufficient vitamin D3 can be synthesized in the skin on irradiation by sunlight. However, indoor confinement of humans and other animals has resulted in the diet becoming the main source of vitamin D, leading to considerable research to determine the amount of dietary vitamin D required to substitute for lack of exposure to sunlight. Photochemically produced vitamin D3 enters the circulation and becomes immediately available, whereas dietary vitamin D3 may undergo modifications prior to becoming available for use by the body. One species where significant modification of vitamin D occurs before absorption is the ruminant. Within 24 hr, as much as 80% of vitamin D can undergo metabolism in vitro in rumen incubation media [217]. At least four metabolites are produced by the rumen microbes [217,218]. Two of these metabolites have been identified as the cis (5Z) and trans (5E) isomers of 10-keto-19-norvitamin D3 (Fig. 2) [207]. The trans isomer has also been identified in cow plasma (R. L. Horst, unpublished data, 1983). Neither compound has agonistic activity with regard to promoting bone calcium resorption [219] or intestinal calcium absorption [207]. Rather, this novel metabolism is likely a detoxification process, as evidenced by the ability of ruminants to tolerate large oral doses of vitamin D3 that would be toxic if given parenterally. The presence of the rumen, therefore, represents a major control point in vitamin D metabolism that may differ from monogastrics. Such a control point may have survival value, because the ruminant evolved as a grazing animal with the opportunity for long periods of sunlight exposure, as well as consumption of large quantities of irradiated plants. If left uncontrolled, such a combination might result in vitamin D toxicity. Shortly after the discovery of vitamin D, it seemed apparent that vitamins D2 and D3 had similar biological activities in most mammals and that birds and New World monkeys discriminated against vitamin D2 in favor of vitamin D3 [220,221]. More recent research, fostered by the discovery of sensitive analytical techniques and the availability of high specific activity 3H-labeled vitamin D species, indicated that differences in the metabolism of vitamins D2 and D3 in mammals are perhaps widespread. Most notable were the apparent discrimination against vitamin D2 by pigs [222], cows [218], and humans [223] and the apparent preference for vitamin D2 by rats [222,224]. Vitamin D and its metabolites are transported in the blood of vertebrates attached to a specific protein commonly known as the vitamin D binding protein or DBP [225]. Baird et al. [226] have shown that protein binding increases the solubility of steroids and that the
29
CHAPTER 2 Vitamin D Metabolism
metabolic clearance rate of steroids is in part dependent on their binding to specific plasma proteins. Affinity of metabolites to the plasma transport proteins may, therefore, provide a means for determining which species would utilize vitamin D2 poorly. For example, if the binding protein showed lower affinity toward 25OHD2 relative to 25OHD3, then one would predict that 25OHD2 would be removed from the circulation faster than 25OHD3. This is indeed the case for the chick. Hoy et al. [227] showed that chick discrimination against vitamin D2 was probably a result of enhanced clearance of the vitamin D2 metabolites 25OHD2 and 1,25(OH)2D2, and that the enhanced clearance was associated with weaker binding of these vitamin D2 metabolites (relative to the vitamin D3 forms) to DBP. In one of the most comprehensive studies reported to date, Hay and Watson [228] studied the affinities of DBP for 25OHD2 and 25OHD3 in 63 vertebrate species. They found that the DBP in fish, reptiles, and birds discriminated against 25OHD2 in favor of 25OHD3, which is consistent (at least in birds) with the discrimination against vitamin D2. One notable exception to this hypothesis, however, is the New World monkey. Hay and Watson [228] found that in New World monkeys, the plasma transport protein has equal affinity for 25OHD2 and 25OHD3, which is inconsistent with the well-documented discrimination against vitamin D2. Factors other than affinity of the binding protein for 25OHD are, therefore, important in determining how efficiently the different forms of vitamin D can be utilized by animals. Another example of species discrimination against the different vitamin D forms is in the rat. However, in this species, discrimination is against vitamin D3 in favor of vitamin D2 [222]. The rat DBP is known to have equal affinity for 25OHD2 and 25OHD3, but a lower affinity for vitamin D2 relative to vitamin D3 [229]. Reddy et al. [230] suggested that the lower affinity for vitamin D2 resulted in its enhanced availability for liver 25-hydroxylation. Hence, in the presence of DBP, more 25OHD2 was made relative to 25OHD3 when equal amounts of vitamin D2 or vitamin D3 substrate were perfused into rat livers. This observation is consistent with the higher circulating concentrations of 25OHD2 observed in acute experiments with vitamin D–deficient rats dosed with equal amounts of vitamins D2 and D3 [222]. In the experiments conducted by Reddy et al. [230], if binding protein was eliminated from the perfusion media, equal amounts of 25OHD2 and 25OHD3 were synthesized. Collectively, these data suggest that discrimination against the different forms of vitamin D could likely result from variations in the affinity of DBP for the parent compound and/or one or
more of their metabolites. Regardless of the mechanism for discrimination, it appears that these differences are present to afford animals the most efficient utilization of the most abundant antirachitic agents available in their environment.
V. CONCLUSION Vitamin D metabolism still remains an exciting area of research with much more to be learned. Critical questions remain unanswered regarding complete elucidation of the vitamin D2 metabolic pathway, and species differences between vitamins D2 and D3 metabolism are still virtually unexplored. The introduction of vitamin D analogs has also resulted in a totally different set of issues regarding their metabolism, tissue kinetics, mechanism of action, and potential therapeutic uses.
References 1. Mellanby E 1919 An experimental investigation on rickets. Lancet I 4985:407–412. 2. McCollum EV, Simmonds N, Becker JE, Shipley PG 1922 Studies on experimental rickets. XXI. An experimental demonstration of the existence of a vitamin which promotes calcium deposition. J Biol Chem 53:293–312. 3. Norman AW 1979 Vitamin D: The calcium homeostaic hormone. In: Darby WJ (ed) Nutrition: Basic and Applied Science. Academic Press, New York. 4. Napoli JL, Horst RL 1984 Vitamin D metabolism. In: Kumar R (ed) Vitamin D: Basic and Applied Aspects. Martinus Nijhoff, Boston, pp. 91–123. 5. Lawson DEM 1978 Vitamin D. Academic Press, London. 6. Kodichek E 1974 The story of vitamin D from vitamin to hormone. Lancet I 7853:325–329. 7. Haussler MR, McCain TA 1977 Basic and clinical concepts related to vitamin D metabolism and action. N Engl J Med 297:974–983, 1041–1050. 8. Deluca HF, Schnoes HK 1976 Metabolism and action of vitamin D. Annu Rev Biochem 45:631–666. 9. IUPAC 1969 IUPAC Commission on the Nomenclature of Organic Chemistry and IUPAC–IUB Commission on Biochemical Nomenclature. Revised tentative rules for nomenclature of steroids. Biochem J 113:5–28. 10. Cahn RS, Ingold C, Prelog V 1966 Specification of molecular chirality. Angew Chem, Int Ed Engl 5:385–415. 11. Blackwood JE, Gladys CL, Loening KL, Petrarca AE, Rush JE 1968 Unambiguous specification of stereoisomerism about a double bond. J Am Chem Soc 90:509–510. 11a. Morii H, Lund J, Neville PF, DeLuca HF 1967 Biological activity of a vitamin D metabolite. Arch Biochem Biophys 120:508–512. 11b. Blunt JW, DeLuca HF, Schnoes HK 1968 25-Hydroxycholecalciferol: A biologically active metabolite of vitamin D3. Biochemistry 7:3317–3322. 11c. Norman AW, Myrtle JF, Midgett RJ, Nowicki HG 1971 1,25-Dihydroxycholecalciferol: Identification of the
30
12.
13.
14.
15. 16. 17.
18. 19.
20. 21.
22. 23.
24. 25. 26. 27. 28. 29.
RONALD L. HORST, TIMOTHY A. REINHARDT, AND G. SATYANARAYANA REDDY
proposed active form of vitamin D3 in the intestine. Science 173:51–54. Lawson DEM, Fraser DR, Kodicek E, Morris HR, Williams DH 1971 Identification of 1,25-dihydroxycholecalciferol, a new kidney hormone controlling calcium metabolism. Nature 230:228–230. Holick MF, Schnoes HK, DeLuca HF, Suda T, Cousins RJ 1971 Isolation and identification of 1,25-dihydroxycholecalciferol. A metabolite of vitamin D active in intestine. Biochemistry 10:2799–2804. Broadus AE, Horst RL, Lang R, Littledike ET, Rasmussen H 1980 The importance of circulating 1,25-dihydroxyvitamin D in the pathogenesis of hypercalciuria and renal-stone formation in primary hyperparathyroidism. N Engl J Med 302:421–426. Bordier P, Rasmussen H, Marie P, Miravet L, Gueris J, Ryckwaert A 1978 Vitamin D metabolism and bone mineralization in man. J Clin Endocrinol Metab 46:284–294. Ornoy A, Goodwin D, Noff D, Edelstein S 1978 24,25Dihydroxyvitamin D is a metabolite of vitamin D essential for bone formation. Nature 276:517–519. Canterbury JM, Lerman S, Claflin AJ, Henry H, Norman A, Reiss E 1977 Inhibition of parathyroid hormone secretion by 25-hydroxycholecalciferol and 24,25-dihydroxycholecalciferol in the dog. J Clin Invest 78:1375–1383. Henry HL, Norman AW 1978 Vitamin D: Two dihydroxylated metabolites are required for normal chicken egg hatchability. Science 201:835–837. St-Arnaud R, Arabian A, Travers R, Barletta F, Raval-Pandya M, Chapin K, Depovere J, Mathieu C, Christakos S, Demay MB, Glorieux FH 2000 Deficient mineralization of intramembranous bone in vitamin D-24-hydroxylase-ablated mice is due to elevated 1,25-dihydroxyvitamin D and not to the absence of 24,25-dihydroxyvitamin D. Endocrinology 141:2658–2666. Horst RL, Littledike ET, Riley JL, Napoli JL 1981 Quantitation of vitamin D and its metabolites and their plasma concentrations in five species of animals. Anal Biochem 116:189–203. Hartwell D, Hassager C, Christiansen C 1987 Effect of vitamin D2 and vitamin D3 on the serum concentrations of 1,25(OH)2D2 and 1,25(OH)2D3 in normal subjects. Acta Endocrinol 115:378–384. Hollis BW, Pittard WB 1984 Relative concentrations of 25-hydroxyvitamin D2/D3 and 1,25-dihydroxyvitamin D2/D3 in maternal plasma at delivery. Nutr Res (N Y) 4:27–32. Reinhardt TA, Horst RL, Orf JW, Hollis BW 1984 A microassay for 1,25-dihydroxyvitamin D not requiring high performance liquid chromatography: Application to clinical studies. J Clin Endocrinol Metab 58:91–98. Jones G, Schnoes HK, DeLuca HF 1975 Isolation and identification of 1,25-dihydroxyvitamin D2. Biochemistry 14:1250–1256. Tucker G, Gaganon R, Haussler M 1973 Vitamin D3-25hydroxylase: Tissue occurrence and apparent lack of regulation. Arch Biochem Biophys 155:47–57. Ponchon G, DeLuca HF 1969 Metabolites of vitamin D3 and their biological activity. J Nutr 99:157–167. Bhattacharyya MH, DeLuca HF 1973 The regulation of rat liver calciferol-25-hydroxylase. J Biol Chem 248: 2969–2973. Bhattacharyya MH, DeLuca HF 1973 Comparative studies on the 25-hydroxylation of vitamin D3 and dihydrotachysterol. J Biol Chem 248:2974–2977. Bjorkhem I, Holmberg I 1979 On the 25-hydroxylation of vitamin D3 in vitro studied with a mass fragmentographic technique. J Biol Chem 254:9518–9524.
30. Bjorkhem I, Holmberg I 1978 Assay and properties of a mitochondrial 25-hydroxylase active on vitamin D3. J Biol Chem 253:842–849. 31. Madhok TC, DeLuca HF 1979 Characteristics of the rat liver microsomal enzyme system converting cholecalciferol into 25-hydroxycholecalciferol. Evidence for the participation of cytochrome P-450. Biochem J 184:491–499. 32. Bjorkhem I, Holmberg I, Oftebro H, Pedersen JI 1980 Properties of a reconstituted vitamin D3 25-hydroxylase from rat liver mitochondria. J Biol Chem 255:5244–5249. 33. Okuda K 1992 Vitamin D3 hydroxylase—its occurrence, biological importance and biochemistry. In: Ruckpaul K, Rein H (eds) Cytochrome P-450 Dependent Biotransformation of Endogenous Substrates: Frontiers in Biotransformation, Vol. 6. Akademic Verlag, Berlin, pp.114–147. 34. Okuda K 1994 Liver mitochodrial p450 involved in cholesterol catabolism and vitamin activation. J Lipid Res 35:361–372. 35. Okuda K-I, Usui E, Ohyama Y 1995 Recent progress in enzymology and molecular biology of enzymes involved in vitamin D metabolism. J Lipid Res 36:1641–1652. 36. Postlind H, Axen E, Bergman T, Wikvall K 1997 Cloning, structure, and expression of a cDNA encoding vitamin D3 25-hydroxylase. Biochem Biophys Res Commun 241: 491–497. 37. Wikvall K 2001 Cytochrome P450 enzymes in the bioactivation of vitamin D to its hormonal form (review). Int J Mol Med 7:201–209. 38. Andersson S, Davis DL, Dahlbach H, Jornvall H 1989 Cloning, structure, and expression of the mitochondrial cytochrome P-450 sterol 26-hydroxylase, a bile acid biosynthetic enzyme. J Biol Chem 264:8222–8229. 39. Guo Y-D, Strugnell S, Back DW, Jones G 1993 Transfected human liver cytochrome P-450 hydroxylates vitamin D analogs at different side-chain positions. Proc Natl Acad Sci USA 90:8668–8672. 40. Cali J, Russell D 1991 Characterization of human sterol 27-hydroxylase. J Biol Chem 266:7774–7778. 41. Usui E, Noshiro M, Okuda K 1990 Molecular cloning of cDNA for vitamin D3 25-hydroxylase from rat liver mitochondria. FEBS Lett 262:135–138. 42. Usui E, Noshiro M, Ohyama Y, Okuda K 1990 Unique property of liver mitochondrial P450 to catalyze the two physiologically important reactions involved in both cholesterol catabolism and vitamin D activation. FEBS Lett 274: 175–177. 43. Jones G, Schnoes HK, Levan L, DeLuca HF 1980 Isolation and identification of 24-hydroxyvitamin D2 and 24,25-dihydroxyvitamin D2. Arch Biochem Biophys 202:450–457. 44. Horst RL, Koszewski NJ, Reinhardt TA 1990 1α-Hydroxylation of 24-hydroxyvitamin D2 represents a minor physiological pathway for the activation of vitamin D2 in mammals. Biochemistry 29:578–582. 45. Koszewski NJ, Reinhardt TA, Napoli JL, Beitz DC, Horst RL 1988 24,26-Dihydroxyvitamin D2: A unique physiological metabolite of vitamin D2. Biochemistry 27:5785–5790. 46. Rosen H, Reshef A, Maeda N, Lippoldt A, Shpizen S, Triger L, Eggertsen G, Bjorkhem I, Leitersdorf E 1998 Markedly reduced bile acid synthesis but maintained levels of cholesterol and vitamin D metabolites in mice with disrupted sterol 27-hydroxylase gene. J Biol Chem 273: 14805–14812. 47. Hayashi S, Usui E, Okuda K 1988 Sex-related difference in vitamin D3 25-hydroxylase of rat liver mitochondria. J Biochem (Tokyo) 103:863–866.
CHAPTER 2 Vitamin D Metabolism
48. Hosseinpour F, Wikvall K 2000 Porcine microsomal vitamin D(3) 25-hydroxylase (CYP2D25). Catalytic properties, tissue distribution, and comparison with human CYP2D6. J Biol Chem 275:34650–34655. 49. Saarem K, Bergseth S, Oftebro H, Petersen JI 1984 Subcellular localization of vitamin D3 25-hydroxylation in human liver. J Biochem (Tokyo) 259:10936–10946. 50. Axen E, Bergman T, Wikvall K 1994 Microsomal 25-hydroxylation of vitamin D2 and vitamin D3 in pig liver. J Steroid Biochem Mol Biol 51:97–106. 51. Hosseinpour F, Ibranovic I, Tang W, Wikvall K 2003 25-Hydroxylation of vitamin D3 in primary cultures of pig hepatocytes: evidence for a role of both CYP2D25 and CYP27A1. Biochem Biophys Res Commun 303:877–883. 52. Haussler MR, Myrtle JF, Norman AW 1968 The association of a metabolite of vitamin D3 with intestinal mucosa chromatin in vivo. J Biol Chem 243:4055–4064. 53. Lawson DEM, Wilson PW, Kodicek E 1969 Metabolism of vitamin D. A new cholecalciferol metabolite, involving loss of hydrogen at C-1, in chick intestinal nuclei. Biochem J 115:269–277. 54. Kodicek E, Lawson DEM, Wilson PW 1970 Biological activity of a polar metabolite of vitamin D3. Nature 228: 763–764. 55. Haussler MR, Boyce DW, Littledike ET, Rasmussen H 1971 A rapidly acting metabolite of vitamin D3. Proc Natl Acad Sci USA 68:177–181. 56. Fraser DR, Kodicek E 1970 Unique biosynthesis by kidney of a biologically active vitamin D metabolite. Nature 228:764–766. 57. Holick MF, Schnoes HK, Deluca HF 1971 Identification of 1,25-dihydroxycholecalciferol, a form of vitamin D3 metabolically active in the intestine. Proc Natl Acad Sci USA 68:803–804. 58. Paulson SK, Deluca HF 1985 Subcellular location and properties of rat renal 25-hydroxyvitamin D3-1α-hydroxylase. J Biol Chem 260:11488–11492. 59. Armbrecht HJ, Okuda K, Wongsurawat N, Nemani RK, Chen ML, Boltz MA 1992 Characterization and regulation of the vitamin D hydroxylases. J Steroid Biochem 43: 1073–1081. 60. Gray RW, Weber HP, Dominguez JH, Lemann J, Jr. 1974 The metabolism of vitamin D3 and 25-hydroxyvitamin D3 in normal and anephric humans. J Clin Endocrinol Metab 39:1045–1056. 61. DeLuca HF 1974 Vitamin D: The vitamin and the hormone. Fed Proc 33:2211–2219. 62. Boyle I, Gray R, Deluca HF 1971 Regulation by calcium of in vivo synthesis of 1,25-dihydroxycholecalciferol and 21,25-dihydroxycholecalciferol. Proc Natl Acad Sci USA 68:2131–2137. 63. Armbrecht HJ, Wongsurawat N, Zenser TV, Davis BB 1984 Effect of PTH and 1,25-(OH)2D3 on renal 25(OH)D3 metabolism, adenylate cyclase, and protein kinase. Am J Physiol 246:E102–E107. 64. Armbrecht HJ, Wongsurawat N, Zenser TV, Davis BB 1982 Differential effects of parathyroid hormone on the renal 1,25-dihydroxyvitamin D3 and 24,25-dihydroxyvitamin D3 production of young and adult rats. Endocrinology 111:1339–1344. 65. Horst RL, Eisman JA, Jorgensen NA, DeLuca HF 1977 Adequate response of plasma 1,25-dihydroxyvitamin D to parturition in paretic (milk fever) dairy cows. Science 196:662–663.
31 66. Hughes MR, Brumbaugh PF, Haussler MR, Wergedal JE, Baylink DJ 1975 Regulation of serum 1α,25-dihydroxyvitamin D3 by calcium and phosphate in the rat. Science 190:578–580. 67. Henry HL 1979 Regulation of the hydroxylation of 25-hydroxyvitamin D3 in vivo and in primary cultures of chick kidney cells. J Biol Chem 254:2722–2729. 68. Omdahl JL 1978 Interaction of the parathyroid and 1,25dihydroxyvitamin D3 in the control of renal 25-hydroxyvitamin D3 metabolism. J Biol Chem 253:8474–8487. 69. Henry HL 1981 Insulin permits parathyroid hormone stimulation of 1,25-dihydroxyvitamin D3 production in cultured kidney cells. Endocrinology 108:733–735. 70. Armbrecht HJ, Forte LR, Wongsurawat N, Zenser TV, Davis BB 1984 Forskolin increases 1,25-dihydroxyvitamin D3 production by rat renal slices in vitro. Endocrinology 114:644–649. 71. Aarskog D, Aksnes L 1980 Acute response of plasma 1,25-dihydroxyvitamin D to parathyroid hormone. Lancet 1(8164):362–363. 72. Hove K, Horst RL, Littledike ET, Beitz DC 1984 Infusions of parathyroid hormone in ruminants: Hypercalcemia and reduced plasma 1,25-dihydroxyvitamin D concentrations. Endocrinology 114:897–903. 73. Hustmyer FG, Beitz DC, Goff JP, Nonnecke BJ, Horst RL, Reinhardt TA 1994 Effects of in vivo administration of 1,25-dihydroxyvitamin D3 on in vitro proliferation of bovine lymphocytes. J Dairy Sci 77:3324–3330. 74. Reinhardt TA, Horst RL 1990 Parathyroid hormone down-regulates 1,25-dihydroxyvitamin D receptors (VDR) and VDR messenger ribonucleic acid in vitro and blocks homologous up-regulation of VDR in vivo. Endocrinology 127:942–948. 75. Henry HL, Midgett RJ, Norman AW 1974 Regulation of 25-hydroxyvitamin D3-1-hydroxylase in vivo. J Biol Chem 249:7584–7592. 76. Tanaka Y, DeLuca HF 1973 The control of 25-hydroxyvitamin D metabolism by inorganic phosphorus. Arch Biochem Biophys 154:566–574. 77. Gray RW 1987 Evidence that somatomedins mediate the effect of hypophosphatemia to increase serum 1,25-dihydroxyvitamin D3 levels in rats. Endocrinology 121:504–512. 78. Armbrecht HJ, Zenser TV, Davis BB 1982 Modulation of renal production of 24,25- and 1,25-dihydroxyvitamin D3 in young and adult rats by dietary calcium, phosphorus, and 1,25-dihydroxyvitamin D3. Endocrinology 110: 1983–1988. 79. Brenza HL, DeLuca HF 2000 Regulation of 25-hydroxyvitamin D3 1α-hydroxylase gene expression by parathyroid hormone and 1,25-dihydroxyvitamin D3. Arch Biochem Biophys 381:143–152. 80. Slatopolsky E, Lopez-Hilker S, Delmez J, Dusso A, Brown A, Martin KJ 1990 The parathyroid-calcitriol axis in health and chronic renal failure. Kidney Int 38:S41–S47. 81. Beckman MJ, Johnson JA, Goff JP, Reinhardt TA, Beitz DC, Horst RL 1995 The role of dietary calcium in the physiology of vitamin D toxicity: Excess dietary vitamin D3 blunts parathyroid hormone induction of kidney 1-hydroxylase. Arch Biochem Biophys 319:535–539. 82. Leheste JR, Melsen F, Wellner M, Jansen P, Schlichting U, Renner-Muller I, Andreassen TT, Wolf E, Bachmann S, Nykjaer A, Willnow TE 2003 Hypocalcemia and osteopathy in mice with kidney-specific megalin gene defect. FASEB J 17:247–249.
32
RONALD L. HORST, TIMOTHY A. REINHARDT, AND G. SATYANARAYANA REDDY
83. Kumar R 1984 Metabolism of 1,25-dihydroxyvitamin D3. Physiol Rev 64:478–504. 84. Henry HL 1992 Vitamin D hydroxylases. J Cell Biochem 49:4–9. 85. Portale AA, Miller WL 2000 Human 25-hydroxyvitamin D-1α-hydroxylase: cloning, mutations, and gene expression. Pediatr Nephrol 14:620–625. 86. Fu GK, Portale AA, Miller WL 1997 Complete structure of the human gene for the vitamin D 1α-hydroxylase, P450c1α. DNA Cell Biol 16:1499–1507. 87. Glorieux FH, St-Arnaud R 1998 Molecular cloning of (25-OH D)-1α-hydroxylase: an approach to the understanding of vitamin D pseudo-deficiency. Recent Prog Horm Res 53:341–349; discussion 350. 88. Shinki T 1998 Cloning and expression of 25-hydroxyvitamin D3-1α-hydroxylase gene. Nippon Rinsho 56:1440–1446. 89. Yoshida T, Yoshida N, Nakamura A, Monkawa T, Hayashi M, Saruta T 1999 Cloning of porcine 25-hydroxyvitamin D3 1α-hydroxylase and its regulation by cAMP in LLC-PK1 cells. J Am Soc Nephrol 10:963–970. 90. Panda DK, Miao D, Tremblay ML, Sirois J, Farookhi R, Hendy GN, Goltzman D 2001 Targeted ablation of the 25-hydroxyvitamin D 1α-hydroxylase enzyme: evidence for skeletal, reproductive, and immune dysfunction. Proc Natl Acad Sci USA 98:7498–7503. 91. Dardenne O, Prud’homme J, Arabian A, Glorieux FH, St-Arnaud R 2001 Targeted inactivation of the 25-hydroxyvitamin D3-1α-hydroxylase gene (CYP27B1) creates an animal model of pseudovitamin D-deficiency rickets. Endocrinology 142:3135–3141. 92. Holick MF, Schnoes HK, DeLuca HF, Gray RW, Boyle IT, Suda T 1972 Isolation and identification of 24,25-dihydroxycholecalciferol, a metabolite of vitamin D3 made in the kidney. Biochemistry 11:4251–4255. 93. Kumar R, Schnoes HK, DeLuca HF 1978 Rat intestinal 25-hydroxyvitamin D3- and 1α,25-dihydroxyvitamin D324-hydroxylase. J Biol Chem 253:3804–3809. 94. Kumar R, Harnden D, DeLuca HF 1976 Metabolism of 1,25-dihydroxyvitamin D3: Evidence for side-chain oxidation. Biochemistry 15:2420–2423. 95. Iwata K, Yamamoto A, Satoh S, Ohyama Y, Tashiro Y, Setoguchi T 1995 Quantitative immunoelectron microscopic analysis of the localization and induction of 25-hydroxyvitamin D3 24-hydroxylase in rat kidney. J Histochem Cytochem 43:255–262. 96. Shinki T, Jin CH, Nishimura A, Nagai Y, Ohyama Y, Noshiro M, Okuda K, Suda T 1992 Parathyroid hormone inhibits 25-hydroxyvitamin D3-24-hydroxylase mRNA expression stimulated by 1α,25-dihydroxyvitamin D3 in rat kidney but not in intestine. J Biol Chem 267:13757–13762. 97. Nishimura A, Shinki T, Jin CH, Ohyama Y, Noshiro M, Okuda K, Suda T 1994 Regulation of messenger ribonucleic acid expression of 1α,25-dihydroxyvitamin D3-24-hydroxylase in rat osteoblasts. Endocrinology 134:1794–1799. 98. Shigematsu T, Horiuchi N, Ogura Y, Miyahara T, Suda T 1986 Human parathyroid hormone inhibits renal 24-hydroxylase activity of 25-hydroxyvitamin D3 by a mechanism involving adenosine 3′,5′-monophosphate in rats. Endocrinology 118: 1583–1589. 99. Goff JP, Reinhardt TA, Beckman MJ, Horst RL 1990 Contrasting effects of exogenous 1,25-dihydroxyvitamin D [1,25-(OH)2D] versus endogenous 1,25-(OH)2D, induced by dietary calcium restriction, on vitamin D receptors. Endocrinology 126:1031–1035.
100. Iida K, Shinki T, Yamaguchi A, DeLuca HF, Kurokawa K, Suda T 1995 A possible role of vitamin D receptors in regulating vitamin D activation in the kidney. Proc Natl Acad Sci USA 92:6112–6116. 101. Hanai H, Liang CT, Cheng L, Sacktor B 1989 Desensitization to parathyroid hormone in renal cells from aged rats is associated with alterations in G-protein activity. J Clin Invest 83:268–277. 102. Hanai H, Brennan DP, Cheng L, Goldman ME, Chorev M, Levine MA, Sacktor B, Liang CT 1990 Downregulation of parathyroid hormone receptors in renal membranes from aged rats. Am J Physiol 259:F444–F450. 103. Johnson JA, Beckman MJ, Pansini-Porta A, Christakos S, Bruns ME, Beitz DC, Horst RL, Reinhardt TA 1995 Age and gender effects on 1,25-dihydroxyvitamin D3-regulated gene expression. Exp Gerontol 30:631–643. 104. Matkovits T, Christakos S 1995 Variable in vivo regulation of rat vitamin D-dependent genes (osteopontin, Ca,Mg-adenosine triphosphatase, and 25-hydroxyvitamin D3 24-hydroxylase): Implications for differing mechanisms of regulation and involvement of multiple factors. Endocrinology 136: 3971–3982. 105. Goff JP, Reinhardt TA, Engstrom GW, Horst RL 1992 Effect of dietary calcium or phosphorus restriction and 1,25-dihydroxyvitamin D administration on rat intestinal 24-hydroxylase. Endocrinology 131:101–104. 106. Beckman MJ, Goff JP, Reinhardt TA, Beitz DC, Horst RL 1994 In vivo regulation of rat intestinal 24-hydroxylase: Potential new role of calcitonin. Endocrinology 135: 1951–1955. 107. Ohyama Y, Noshiro M, Okuda K 1991 Cloning and expression of cDNA encoding 25-hydroxyvitamin D3 24-hydroxylase. FEBS Lett 278:195–198. 108. Ohyama Y, Hayashi S, Okuda K 1989 Purification of 25-hydroxyvitamin D3 24-hydroxylase from rat kidney mitochondria. FEBS Lett 255:405–408. 109. Chen K-S, DeLuca HF 1995 Cloning of the human 1α,25dihydroxyvitamin D-3 24-hydroxylase gene promoter and identification of two vitamin D-responsive elements. Biochim Biophys Acta 1263:1–9. 110. Ohyama Y, Noshiro M, Eggertsen G, Gotoh O, Kato Y, Bjorkhem I, Okuda K 1993 Structural characterization of the gene encoding rat 25-hydroxyvitamin D3 24-hydroxylase. Biochemistry 32:76–82. 111. Horst RL, Littledike ET, Gray RW, Napoli JL 1981 Impaired 24,25-dihydroxyvitamin D production in anephric human and pig. J Clin Invest 67:274–280. 112. Taniguchi T, Eto TA, Shiotsuki H, Sueta H, Higashi S, Iwamura T, Okuda KI, Setoguchi T 2001 Newly established assay method for 25-hydroxyvitamin D3 24-hydroxylase revealed much lower Km for 25-hydroxyvitamin D3 than for 1α,25-dihydroxyvitamin D3. J Bone Miner Res 16:57–62. 113. Jones G, Tenenhouse HS 2002 1,25(OH)2D, the preferred substrate for CYP24. J Bone Miner Res 17:179–181. 114. Burgos-Trinidad M, DeLuca HF 1991 Kinetic properties of 25-hydroxyvitamin D- and 1,25-dihydroxyvitamin D-24hydroxylase from chick kidney. Biochim Biophys Acta 1078:226–230. 115. Napoli JL, Pramanik BC, Royal PM, Reinhardt TA, Horst RL 1983 Intestinal synthesis of 24-keto-1,25-dihydroxyvitamin D3. A metabolite formed in vivo with high affinity for the vitamin D cytosolic receptor. J Biol Chem 258:9100–9107. 116. Napoli JL, Horst RL 1983 C(24)- and C(23)-oxidation, converging pathways of intestinal 1,25-dihydroxyvitamin D3
33
CHAPTER 2 Vitamin D Metabolism
117. 118. 119. 120.
121.
122.
123.
124.
125. 126.
127.
128. 129.
130.
131.
132.
metabolism: Identification of 24-keto-1,23,25-trihydroxyvitamin D3. Biochemistry 22:5848–5853. Haussler MR 1986 Vitamin D receptors: Nature and function. Annu Rev Nutr 6:527–562. Lohnes D, Jones G 1987 Side chain metabolism of vitamin D3 in osteosarcoma cell line UMR-106. Characterization of products. J Biol Chem 262:14394–14401. Reddy GS, Tserng KY 1989 Calcitroic acid, end product of renal metabolism of 1,25-dihydroxyvitamin D3 through C-24 oxidation pathway. Biochemistry 28:1763–1769. Makin G, Lohnes D, Byford V, Ray R, Jones G 1989 Target cell metabolism of 1,25-dihydroxyvitamin D3 to calcitroic acid: Evidence for a pathway in kidney and bone involving 24-oxidation. Biochem J 262:173–180. Akiyoshi-Shibata M, Sakaki Y, Ohyama Y, Noshiro M, Okuda K, Yabusaki Y 1994 Further oxidation of hydroxycalcidiol by calcidiol 24-hydroxylase. Eur J Biochem 224:335–343. Sakaki T, Sawada N, Komai K, Shiozawa S, Yamada S, Yamamoto K, Ohyama Y, Inouye K 2000 Dual metabolic pathway of 25-hydroxyvitamin D3 catalyzed by human CYP24. Eur J Biochem 267:6158–6165. Sakaki T, Sawada N, Nonaka Y, Ohyama Y, Inouye K 1999 Metabolic studies using recombinant Escherichia coli cells producing rat mitochondrial CYP24. CYP24 can convert 1α,25-dihydroxyvitamin D3 to calcitroic acid. Eur J Biochem 262:43–48. Inouye K, Sakaki T 2001 Enzymatic studies on the key enzymes of vitamin D metabolism; 1α-hydroxylase (CYP27B1) and 24-hydroxylase (CYP24). Biotechnol Annu Rev 7:179–194. Beckman MJ, Tadikonda P, Werner E, Prahl J, Yamada S, DeLuca HF 1996 Human 25-hydroxyvitamin D3-24-hydroxylase, a multicatalytic enzyme. Biochemistry 35:8465–8472. Pols HAP, van Leeuwen JPTM, Schilte JP, Visser TJ, Birkenhager JC 1988 Heterologous up-regulation of the 1,25-dihydroxyvitamin D3 receptor by parathyroid hormone (PTH) and PTH-like peptide in osteoblast-like cells. Biochem Biophys Res Commun 156:588–594. Pols HAP, Birkenhager JC, Schilte JP, Visser TJ 1988 Evidence that the self-induced metabolism of 1,25-dihydroxyvitamin D3 limits the homologous up-regulation of its receptor in rat osteosarcoma cells. Biochim Biophys Acta 970:122–129. Reinhardt TA, Horst RL 1989 Self-induction of 1,25-dihydroxyvitamin D3 metabolism limits receptor occupancy and target tissue responsiveness. J Biol Chem 264:15917–15921. Reinhardt TA, Horst RL 1989 Ketoconazole inhibits selfinduced metabolism of 1,25-dihydroxyvitamin D3 and amplifies 1,25-dihydroxyvitamin D3 receptor up-regulation in rat osteosarcoma cells. Arch Biochem Biophys 272:459–465. Reinhardt TA, Horst RL, Engstrom GW, Atkins KB 1988 Ketoconazole potentiates 1,25-(OH)2D3-directed upregulation of 1,25-(OH)2D3-receptors in rat intestine and bone. In: Norman AW, Schaefer K, Grigoleit H-G, Herrath DV (eds) Vitamin D: Molecular, Cellular and Clinical Endocrinology. Walter de Gruyter, Berlin, pp. 233–234. Hosogane N, Shinki T, Kasuga H, Taketomi S, Toyama Y, Suda T 2003 Mechanisms for the reduction of 24,25-dihydroxyvitamin D3 levels and bone mass in 24-hydroxylase transgenic rats. FASEB J 17:737–739. Kasuga H, Hosogane N, Matsuoka K, Mori I, Sakura Y, Shimakawa K, Shinki T, Suda T, Taketomi S 2002 Characterization of transgenic rats constitutively expressing
133.
134.
135.
136.
137.
138.
139. 140.
141.
142. 143.
144.
145. 146.
147.
vitamin D-24-hydroxylase gene. Biochem Biophys Res Commun 297:1332–1338. Wichmann J, Schnoes HK, DeLuca HF 1981 Isolation and identification of 24(R)-hydroxyvitamin D3 from chicks given large doses of vitamin D3. Biochemistry 20: 2350–2353. Wichmann JK, Schnoes HK, DeLuca HF 1981 23,24, 25-Trihydroxyvitamin D3, 24,25,26-trihydroxyvitamin D3, 24-keto-25-hydroxyvitamin D3, and 23-dehydro-25-hydroxyvitamin D3: New in vivo metabolites of vitamin D3. Biochemistry 20:7385–7391. DeLuca HF, Schnoes HK 1979 Recent developments in the metabolism of vitamin D3. In: Norman AW (ed) Vitamin D: Basic Research and Its Clinical Application. Walter de Gruyter, Berlin, pp. 445–458. Horst RL 1979 25-OHD3-26,23-lactone: A metabolite of vitamin D3 that is 5 times more potent than 25-OHD3 in the rat plasma competitive protein binding radioassay. Biochem Biophys Res Commun 89:286–293. Wichmann JK, DeLuca HF, Schnoes HK, Horst RL, Shepard RM, Jorgensen NA 1979 25-Hydroxyvitamin D3 26,23lactone: A new in vivo metabolite of vitamin D. Biochemistry 18:4775–4780. Ohnuma N, Norman AW 1982 Production in vitro of 1α,25dihydroxyvitamin D3-26,23-lactone from 1α,25-dihydroxyvitamin D3 by rat small intestinal mucosa homogenates. Arch Biochem Biophys 213:139–147. Napoli JL, Horst RL 1983 (23S)-1,23,25-Trihydroxycholecalciferol, an intestinal metabolite of 1,25-dihydroxycholecalciferol. Biochem J 214:261–264. Napoli JL, Pramanik BC, Partridge JJ, Uskokovic MR, Horst RL 1982 23S,25-Dihydroxyvitamin D3 as a circulating metabolite of vitamin D3. Its role in 25-hydroxyvitamin D326,23-lactone biosynthesis. J Biol Chem 257:9634–9639. Horst RL, Littledike ET 1982 Comparison of plasma concentrations of vitamin D and its metabolites in young and aged domestic animals. Comp Biochem Physiol 73B: 485–489. Horst RL, Littledike ET 1980 25-OHD3-26,23 lactone: Demonstration of kidney-dependent synthesis in the pig and rat. Biochem Biophys Res Commun 93:149–154. Ishizuka S, Ohba T, Norman AW 1988 1,25-(OH)2D3-26,23lactone is a major metabolite of 1,25-(OH)2D3 under physiological conditions. In: Norman AW, Schafer K, Grigoleit H-G, Herrath DV (eds) Vitamin D: Molecular, Cellular and Clinical Endocrinology. Walter de Gruyter, Berlin, pp. 143–144. Ishizuka S, Ishimoto S, Norman AW 1984 Isolation and identification of 1α,25-dihydroxy-24-oxovitamin D3, 1α,25dihydroxyvitamin D3 26,23-lactone, and 1α,24(S),25-trihydroxyvitamin D3: In vivo metabolites of 1α,25-dihydroxyvitamin D3. Biochemistry 23:1473–1478. Napoli JL, Horst RL 1982 23,25,26-Trihydroxycholecalciferol. A 25-hydroxycholecalciferol-26,23-lactone precursor. Biochem J 206:173–176. Yamada S, Nakayama K, Takayama H, Shinki T, Takasaki Y, Suda T 1984 Isolation, identification, and metabolism of (23S,25R)-25-hydroxyvitamin D3 26,23-lactol. A biosynthetic precursor of (23S,25R)-25-hydroxyvitamin D3 26,23-lactone. J Biol Chem 259:884–889. Horst RL, Pramanik BC, Reinhardt TA, Shiuey SJ, Partridge JJ, Uskokovic MR, Napoli JL 1982 Binding properties of 23S,25dihydroxyvitamin D3: An in vivo metabolite of vitamin D3. Biochem Biophys Res Commun 106:1006–1011.
34
RONALD L. HORST, TIMOTHY A. REINHARDT, AND G. SATYANARAYANA REDDY
148. Horst RL, Wovkulich PM, Baggiolini EG, Uskokovic MR, Engstrom GW, Napoli JL 1984 (23S)-1,23,25-Trihydroxyvitamin D3: Its biologic activity and role in 1,25-dihydroxyvitamin D3 26,23-lactone biosynthesis. Biochemistry 23:3973–3979. 149. Esvelt RP, Schnoes HK, DeLuca HF 1979 Isolation and characterization of 1α-hydroxy-23-carboxytetranorvitamin D: A major metabolite of 1,25-dihydroxyvitamin D3. Biochemistry 18:3977–3983. 150. Horst RL, Reinhardt TA, Pramanik BC, Napoli JL 1983 23-Keto-25-hydroxyvitamin D3: A vitamin D3 metabolite with high affinity for the 1,25-dihydroxyvitamin D specific cytosol receptor. Biochemistry 22:245–250. 151. Horst RL, Reinhardt TA, Napoli JL 1982 23-Keto-25hydroxyvitamin D3 and 23-keto-1,25-dihydroxyvitamin D3: Two new vitamin D3 metabolites with high affinity for the 1,25-dihydroxyvitamin D3 receptor. Biochem Biophys Res Commun 107:1319–1325. 152. Suda T, DeLuca HF, Schnoes HK, Tanaka Y, Holick MF 1970 25,26-Dihydroxycholecalciferol, a metabolite of vitamin D3 with intestinal calcium transport activity. Biochemistry 9:4776–4780. 153. Reinhardt TA, Napoli JL, Pramanik B, Littledike ET, Beitz DC, Partridge JJ, Uskokovic MR, Horst RL 1981 1α,25,26-Trihydroxyvitamin D3: An in vivo and in vitro metabolite of vitamin D3. Biochemistry 20:6230–6235. 154. Partridge JJ, Shiuey SJ, Chadha HK, Baggiolini ET, Bolount JE, Uskokovic MR 1981 Synthesis and structure proof of a vitamin D3 metabolite, 25(S),26-dihydroxycholecalciferol. J Am Chem Soc 103:1253–1255. 155. Ikekawa N, Koizumi N, Ohshima E, Ishizuka S, Takeshita T, Tanaka Y, DeLuca HF 1983 Natural 25,26-dihydroxyvitamin D3 is an epimeric mixture. Proc Natl Acad Sci USA 80:5286–5288. 156. Hollis BW, Roos BA, Lambert PW 1980 25,26Dihydroxycholecalciferol: A precursor in the renal synthesis of 25-hydroxycholecalciferol-26,23-lactone. Biochem Biophys Res Commun 95:520–528. 157. Napoli JL, Horst RL 1981 25,26-Dihydroxyvitamin D3 is not a major intermediate in 25-hydroxyvitamin D3-26,23-lactone formation. Arch Biochem Biophys 212:754–758. 158. Tanaka Y, DeLuca HF, Schnoes HK, Ikekawa N, Eguchi T 1981 23,25-Dihydroxyvitamin D3: A natural precursor in the biosynthesis of 25-hydroxyvitamin D3-26,23-lactone. Proc Natl Acad Sci USA 78:4805–4808. 159. Reddy GS, Muralidharan KR, Okamura WH, Tserng KY, McLane JA 2001 Metabolism of 1α,25-dihydroxyvitamin D3 and its C-3 epimer 1α,25-dihydroxy-3-epi-vitamin D3 in neonatal human keratinocytes. Steroids 66:441–450. 160. Siu-Caldera ML, Sekimoto H, Weiskopf A, Vouros P, Muralidharan KR, Okamura WH, Bishop J, Norman AW, Uskokovic MR, Schuster I, Reddy GS 1999 Production of 1α,25-dihydroxy-3-epi-vitamin D3 in two rat osteosarcoma cell lines (UMR 106 and ROS 17/2.8): Existence of the C-3 epimerization pathway in ROS 17/2.8 cells in which the C-24 oxidation pathway is not expressed. Bone 24:457–463. 161. Horst RL, Shepard RM, Jorgensen NA, DeLuca HF 1979 The determination of 24,25-dihydroxyvitamin D and 25,26-dihydroxyvitamin D in plasma from normal and nephrectomized man. J Lab Clin Med 93:277–285. 162. Shepard RM, Horst RL, Hamstra AJ, DeLuca HF 1979 Determination of vitamin D and its metabolites in plasma from normal and anephric man. Biochem J 182: 55–69.
163. Bergman T, Postlind H 1991 Characterization of mitochondrial cytochromes P-450 from pig kidney and liver catalysing 26-hydroxylation of 25-hydroxyvitamin D3 C27 steroids. Biochem J 276:427–432. 164. Napoli JL, Okita RT, Masters BS, Horst RL 1981 Identification of 25,26-dihydroxyvitamin D3 as a rat renal 25-hydroxyvitamin D3 metabolite. Biochemistry 20:5865–5871. 165. Tanaka Y, Schnoes HK, Smith CM, DeLuca HF 1981 1,25,26-Trihydroxyvitamin D3 isolation, identification and biological activity. Arch Biochem Biophys 210: 104–109. 166. Astecker N, Reddy GS, Herzig G, Vorisek G, Schuster I 2000 1α,25-Dihydroxy-3-epi-vitamin D3 a physiological metabolite of 1α,25-dihydroxyvitamin D3: its production and metabolism in primary human keratinocytes. Mol Cell Endocrinol 170:91–101. 167. Masuda S, Kamao M, Schroeder NJ, Makin HL, Jones G, Kremer R, Rhim J, Okano T 2000 Characterization of 3-epi-1α,25-dihydroxyvitamin D3 involved in 1α,25-dihydroxyvitamin D3 metabolic pathway in cultured cell lines. Biol Pharm Bull 23:133–139. 168. Hylemon PB 1985 Metabolism of bile acids in intestinal microflora. In: Danielsson H, Sjoevall J (eds) Sterols and Bile Acids, Elsevier, pp. 331–344. 169. Penning TM, Bennett MJ, Smith-Hoog S, Schlegel BP, Jez JM, Lewis M 1997 Structure and function of 3α-hydroxysteroid dehydrogenase. Steroids 62:101–111. 170. Norman AW, Bouillon R, Farach-Carson MC, Bishop JE, Zhou LX, Nemere I, Zhao J, Muralidharan KR, Okamura WH 1993 Demonstration that 1β,25-dihydroxyvitamin D3 is an antagonist of the nongenomic but not genomic biological responses and biological profile of the three A-ring diastereomers of 1α,25-dihydroxyvitamin D3. J Biol Chem 268: 20022–20030. 171. Brown AJ, Ritter C, Slatopolsky E, Muralidharan KR, Okamura WH, Reddy GS 1999 1α,25-dihydroxy-3-epi-vitamin D3, a natural metabolite of 1α,25-dihydroxyvitamin D3, is a potent suppressor of parathyroid hormone secretion. J Cell Biochem 73:106–113. 172. Schuster I, Astecker N, HE, Hertig G, Reddy GS, Schmid J, Vorisek G 1997 Vitamin D metabolism in human keratinocytes and biological role of products. In: Norman AW, Bouillon R, Thomasset M (eds) Vitamin D, Chemistry, Biology and Clinical Applications of the Steroid Hormone. Walter de Gruyter, New York, pp. 551–558. 173. Engstrom GW, Koszewski NJ 1989 Metabolism of vitamin D2 in pig liver homogenates: Evidence for a free radical reaction. Arch Biochem Biophys 270:432–440. 174. Horst RL, Reinhardt TA, Ramberg CF, Koszewski NJ, Napoli JL 1986 24-Hydroxylation of 1,25-dihydroxyergocalciferol: An unambiguous deactivation process. J Biol Chem 261:9250–9256. 175. Zimmerman DR, Reinhardt TA, Kremer R, Beitz DC, Reddy GS, Horst RL 2001 Calcitroic acid is a major catabolic metabolite in the metabolism of 1α-dihydroxyvitamin D2. Arch Biochem Biophys 392:14–22. 176. Rao LS, Ray R, Holick MF, Horst RL, Uskokovic MR, Reddy GS 2000 Metabolism of [3α-3H] 25-hydroxyvitamin D2 in kidneys isolated from normal and vitamin D2-intoxicated rats. J Nutr Sci Vitaminol (Tokyo) 46:222–229. 177. Horst RL, Omdahl JA, Reddy S 2003 Rat cytochrome P450C24 (CYP24) does not metabolize 1,25-dihydroxyvitamin D2 to calcitroic acid. J Cell Biochem 88: 282–285.
CHAPTER 2 Vitamin D Metabolism
178. Sunita Rao D, Balkundi D, Uskokovic MR, Tserng K, Clark JW, Horst RL, Satyanarayana Reddy G 2001 Double bond in the side chain of 1α,25-dihydroxy-22-ene-vitamin D3 is reduced during its metabolism: studies in chronic myeloid leukemia (RWLeu-4) cells and rat kidney. J Steroid Biochem Mol Biol 78:167–176. 179. Byford V, Strugnell S, Coldwell R, Schroeder N, Makin HL, Knutson JC, Bishop CW, Jones G 2002 Use of vitamin D4 analogs to investigate differences in hepatic and target cell metabolism of vitamins D2 and D3. Biochim Biophys Acta 1583:151–166. 180. Tachibana Y, Tsuji M 2001 Study on the metabolites of 1α,25-dihydroxyvitamin D4. Steroids 66:93–97. 181. Williams DH, Morris DS, Gilhooly MA, Norris AF 1982 Synthesis of 25(R)-hydroxycholecalciferol-26,23(S)-lactone and of 25(S),26-dihydroxyergocalciferol. In: Norman AW, Schaefer K, Herrath DV, Grigoleit H-G (eds) Vitamin D: Chemical, Biochemical and Clinical Endocrinology of Calcium Metabolism. Walter de Gruyter, Berlin, pp. 1067–1072. 182. Mazur Y, Segev D, Jones G 1982 Synthesis and determination of absolute configuration of kidney metabolites of vitamin D2. In: Norman AW, Schaefer K, Herrath DV, Grigoleit H-G (eds) Vitamin D: Chemical, Biochemical and Clinical Endocrinology of Calcium Metabolism. Walter de Gruyter, Berlin, pp. 1101–1106. 183. Hay AWM, Jones G 1979 The elution profile of vitamin D2 metabolites from Sephadex LH-20 columns. Clin Chem 25:473–475. 184. Reddy GS, Tserng KY 1990 24,25,28-Trihydroxyvitamin D2 and 24,25,26-trihydroxyvitamin D2: Novel metabolites of vitamin D2. Biochemistry 29:943–949. 185. Reddy GS, Tserng KY 1986 Isolation and identification of 1,24,25-trihydroxyvitamin D2, 1,24,25,28-tetrahydroxyvitamin D2, and 1,24,25,26-tetrahydroxyvitamin D2: New metabolites of 1,25-dihydroxyvitamin D2 produced in rat kidney. Biochemistry 25:5328–5336. 186. Jones G, Byford V, Makin H, Kremer R, Rice R, deGraffenried LA, Knutson JC, Bishop CW 1996 Antiproliferative activity and target cell catabolism of the vitamin D analog 1α24(S)-(OH)2D2 in normal and immortalized human epidermal cells. Biochem Pharmacol 52:133–140. 187. Rao DS, Siu-Caldera ML, Uskokovic MR, Horst RL, Reddy GS 1999 Physiological significance of C-28 hydroxylation in the metabolism of 1α,25-dihydroxyvitamin D2. Arch Biochem Biophys 368:319–328. 188. Kreitmeir H, Moll T 1928. Hypervitaminose durch grosse Dosen Vitamin D Munch. Med. Wochenschr 75:637–639. 189. Putscher W 1929 Über Vigantolschädigung der Niere einum Kinde. Z Kinderheilkd 48:269. 190. Holick MF, MacLaughlin JA, Doppelt SH 1981 Regulation of cutaneous previtamin D3 photosynthesis in man: Skin pigment is not an essential regulator. Science 211:590–593. 191. Clark MB, Potts JT 1977 25-Hydroxyvitamin D3 regulation. Calcif Tissue Res 22(Supplement):29–34. 192. Shepard RM, DeLuca HF 1980 Plasma concentrations of vitamin D3 and its metabolites in the rat as influenced by vitamin D3 or 25-hydroxyvitamin D3 intakes. Arch Biochem Biophys 202:43–53. 193. Hughes MR, Baylink DJ, Jones PG, Haussler MR 1976 Radioligand receptor assay for 25-hydroxyvitamin D2/D3 and 1α,25-dihydroxyvitamin D2/D3: Application to hypervitaminosis D. J Clin Invest 58:61–70. 194. Counts SJ, Baylink DJ, Shen FH, Sherrard DJ, Hickman RO 1975 Vitamin D intoxication in an anephric child. Ann Intern Med 82:196–200.
35 195. Littledike ET, Horst RL 1982 Vitamin D3 toxicity in dairy cows. J Dairy Sci 65:749–759. 196. Reinhardt TA, Conrad HR 1980 Mode of action of pharmacological doses of cholecalciferol during parturient hypocalcemia in dairy cows. J Nutr 110:1589–1594. 197. Horst RL, Reinhardt TA 1983 Vitamin D metabolism in ruminants and its relevance to the periparturient cow. J Dairy Sci 66:661–678. 198. Hunt RD, Garcia FG, Walsh RJ 1972 A comparison of the toxicity of ergocalciferol and cholecalciferol in Rhesus monkeys (Masaca mulatta). J Nutr 102:975–986. 199. Harrington DD, Page EH 1983 Acute vitamin D3 toxicosis in horses: Case reports and experimental studies of the comparative toxicity of vitamins D2 and D3. J Am Vet Med Assoc 182:1358–1369. 200. Hodson EM, Evans RA, Dunstan CR, Hills E, Wong SYP, Rosenberg AR, Roy LP 1985 Treatment of childhood renal osteodystrophy with calcitriol or ergocalciferol. Clin Nephrol 24:192–200. 201. Sjoden G, Smith C, Lingren U, DeLuca HF 1985 1α-hydroxyvitamin D2 is less toxic than 1α-hydroxyvitamin D3 in the rat. Proc Soc Exp Biol Med 178:432–436. 202. Beckman MJ, Horst RL, Reinhardt TA, Beitz DC 1990 Up-regulation of the intestinal 1,25-dihydroxyvitamin D receptor during hypervitaminosis D: A comparison between vitamin D2 and vitamin D3. Biochem Biophys Res Commun 169:910–915. 203. Mortensen JT, Brinck P, Binderup L 1993 Toxicity of vitamin D analogues in rats fed diets with standard or low calcium contents. Pharmacol Toxicol 72:124–127. 204. Hines TG, Jacobson NL, Beitz DC, Littledike ET 1985 Dietary calcium and vitamin D: Risk factors in the development of atherosclerosis in young goats. J Nutr 115: 167–178. 205. Payne JM, Manston R 1967 The safety of massive doses of vitamin D3 in the prevention of milk fever. Vet Rec 81:214–216. 206. Hibbs JW, Pounden WD 1955 Studies on milk fever in dairy cows. IV. Prevention by short-time, prepartum feeding of massive doses of vitamin D. J Dairy Sci 38:65–72. 207. Napoli JL, Sommerfeldt JL, Pramanik BC, Gardner R, Sherry AD, Partridge JJ, Uskokovic MR, Horst RL 1983 19-Nor-10-ketovitamin D derivatives: Unique metabolites of vitamin D3, vitamin D2, and 25-hydroxyvitamin D3. Biochemistry 22:3636–3640. 208. Recker RR, Schoenfeld JL, Slatopolsky R, Goldsmith R, Brickman A 1979 25-Hydroxyvitamin D (calcidiol) on renal osteodystophy: Long term results of a six center trial. In: Norman AW, Schaefer K, Herrath DV, Grigoleit HG, Coburn JW, DeLuca HF, Mawer EB, Suda T (eds) Vitamin D: Basic Research and Its Clinical Application. Walter de Gruyter, Berlin, pp. 869–875. 209. West TET, Joffe M, Sinclair L, O’Riordan JLH 1971 Treatment of hypercalcemia with calcitonin. Lancet, 675–678. 210. Ulbrych-Jablonska A 1972 The hypocalcemic effect of glucagon in cases of hypercalcemia. Helv Paediatr Acta 27:613–615. 211. Kingma JG, Roy PE 1990 Effect of ethane-I-hydroxyl, I-diphosphonate on arterial calcinosis induced by hypervitaminosis D: a morphologic investigation. J Exp Path (Oxford) 71:145–153. 212. Streck WF, Waterhouse C, Haddad JG 1979 Glucocorticoid effects in vitamin D intoxication. Arch Intern Med 139: 974–977.
36
RONALD L. HORST, TIMOTHY A. REINHARDT, AND G. SATYANARAYANA REDDY
213. Buffenstein R, Skinner DC, Yahav S, Moodley GP, Cavaleros M, Zachen D, Ross FP, Pettifor JM 1991 Effect of oral cholecalciferol supplementation at physiological and supraphysiological doses in naturally vitamin D3-deficient subterranean damara mole rats (Cryptomys damarensis). J Endocrinol 131:197–202. 214. Cheeke PR, Amberg JW 1973 Comparative calcium excretion by rats and rabbits. J Anim Sci 37:450–454. 215. Schryver HF, Hintz HF, Lowe JE 1974 Calcium and phosphorus in the nutrition of the horse. Cornell Vet 64:493–515. 216. Bourdeau JE, Schwer-Dymerski DA, Stern PH, Langman CB 1986 Calcium and phosphorus metabolism in chronically vitamin D-deficient laboratory rabbits. Miner Electrolyte Metab 12:176–185. 217. Sommerfeldt JL, Horst RL, Littledike ET, Beitz DB 1979 In vitro degradation of cholecalciferol in rumen fluid. J Dairy Sci 62(Supplement 1):192. 218. Sommerfeldt JL, Napoli JL, Littledike ET, Beitz DC, Horst RL 1983 Metabolism of orally administered [3H]ergocalciferol and [3H]cholecalciferol by dairy calves. J Nutr 113:2595–2600. 219. Stern PH, Horst RL, Gardner R, Napoli JL 1985 10-Keto or 25-hydroxy substitution confer equivalent in vitro boneresorbing activity to vitamin D3. Arch Biochem Biophys 236:555–558. 220. Hunt RD, Garcia FG, Hegsted DM, Kaplinsky N 1967 Vitamin D2 and vitamin D3 in New World primates: Influence on calcium absorptions. Science 158:943–945. 221. Steenbock H, Kletzien SWF, Haplin JG 1932 Reaction of chicken to irradiated ergosterol and irradiated yeast as contrasted with natural vitamin D of fish liver oils. J Biol Chem 97:249–264. 222. Horst RL, Napoli JL, Littledike ET 1982 Discrimination in the metabolism of orally dosed ergocalciferol and cholecalciferol by the pig, rat and chick. Biochem J 204:185–189. 223. Sebert JL, Garabedian M, deChasteigner C, Defrance D 1991 Comparative effects of equal doses of vitamin D2 and vitamin D3 for the correction of vitamin D deficiency in the elderly. In: Norman AW, Bouillon R, Thomasset M (eds) Vitamin D: Gene Regulation, Structure–Function Analysis and Clinical Application. Walter de Gruyter, Berlin, pp. 765–766. 224. Reddy GS, Ray R, Holick MF 1990 Serum 1,25-dihydroxyvitamin D2 levels are two times higher than 1,25-dihydroxyvitamin D3 levels in vitamin D-deficient rats, dosed acutely with equal amounts of vitamin D2 and vitamin D3. J Bone Miner Res 5(Supplement 2):S265. 225. Haddad JG, Walgate J 1976 25-Hydroxyvitamin D transport in human plasma. J Biol Chem 251:4803–4809. 226. Baird DT, Horton R, Longcope C, Tait JF 1969 Steroid hormone dynamics under steady state conditions. Recent Prog Horm Res 25:611–664. 227. Hoy DA, Ramberg CF Jr., Horst RL 1988 Evidence that discrimination against ergocalciferol by the chick is the
228. 229. 230.
231.
232.
233.
234.
235.
236.
237.
238.
result of enhanced metabolic clearance rates for its monoand dihydroxylated metabolites. J Nutr 118:633–638. Hay AWM, Watson G 1977 Vitamin D2 in vertebrate evolution. Comp Biochem Physiol 56B:375–380. Horst RL, Reinhardt TA, Beitz DC, Littledike ET 1981 A sensitive competitive protein binding assay for vitamin D in plasma. Steroids 37:581–592. Reddy GS, Napoli JL, Hollis BW 1984 Preferential 25-hydroxylation of vitamin D2 vs vitamin D3 by rat liver is dependent upon vitamin D-binding protein. Calcif Tissue Int 36:524 (Abstract). Arnaud SB, Young DR, Cann C, Reinhardt TA, Henrickson R 1985 Is hypervitaminosis D normal in the rhesus monkey? In: Norman AW, Schaefer K, Grigoleit H-G, Herrath DV (eds) Vitamin D. Chemical, Biochemical and Clinical Update. Walter de Gruyter, Berlin, pp. 585–586. Shinki T, Shiina Y, Takahashi N, Tanioka Y, Koizumi H, Suda T 1983 Extremely high circulating levels of 1α,25dihydroxyvitamin D3 in the marmoset, a New World monkey. Biochem Biophys Res Commun 114:452–457. O’Loughlin PD, Morris HA 1991 Duodenal transport of calcium in the marmoset. In: Norman AW, Bouillon R, Thomasset M (eds) Vitamin D. Gene Regulation, Structure– Function Analysis and Clinical Application. Walter de Gruyter, Berlin, pp. 345–346. Shore RF, Hayes ME, Balment RJ, Mawer EB 1988 Serum 25(OH)D3 and 1,25(OH)2D3 levels in wild and laboratorybred wood mice and bank voles. In: Norman AW, Schaefer K, Grigoleit H-G, Herrath DV (eds) Vitamin D. Molecular, Cellular and Clinical Endocrinology. Walter de Gruyter, Berlin, pp. 633–634. Kobayashi T, Takeuchi A, Okano T 1991 An evolutionary aspect in vertebrates from the viewpoint of vitamin D3 metabolism. In: Norman AW, Bouillon R, Thomasset M (eds) Vitamin D. Gene Regulation, Structure–Function Analysis and Clinical Application. Walter de Gruyter, Berlin, pp. 679–680. Glowacki J, Deftos LJ, Mayer E, Norman AW, Henry H 1982 Chondrichthyes cannot resorb implanted bone and have calcium-regulating hormones. In: Norman AW, Schaefer K, Herrath DV, Grigoleit H-G (eds) Vitamin D. Chemical, Biochemical and Clinical Endocrinology of Calcium Metabolism. Walter de Gruyter, Berlin, pp. 613–615. Sundell K, Bishop JE, Björnsson BT, Horst RL, Hollis BW, Norman AW 1991 Organ distribution of a high affinity receptor for 1,25(OH)2 vitamin D3 in the Atlantic cod (Gadus morhua), a marine teleost. In: Norman AW, Bouillon R, Thomasset M (eds) Vitamin D. Gene Regulation, Structure– Function Analysis and Clinical Application. Walter de Gruyter, Berlin, pp. 114–115. Maenpaa PH, Koskinen T, Koskinen E 1988 Serum profiles of vitamins A, E and D in mares and foals during different seasons. J Anim Sci 66:1418–1423.
CHAPTER 3
Photobiology of Vitamin D MICHAEL F. HOLICK
I. II. III. IV.
Vitamin D, Skin, and Bone Research Laboratory; Department of Medicine; Endocrinology, Nutrition and Diabetes Section; Boston Medical Center and Boston University School of Medicine, Boston, Massachusetts
Introduction Historical Perspective Photobiology of Vitamin D Role of Sunlight and Dietary Vitamin D in Bone Health, Overall Health, and Well-Being
I. INTRODUCTION Vitamin D is neither a vitamin nor a hormone when adequate exposure to sunlight is available to promote the synthesis of vitamin D in the skin [1]. First produced in ocean-dwelling phytoplankton and zooplankton, vitamin D has been made by life forms on earth for almost 1 billion years [2]. Although the physiological function of vitamin D in these lower life forms is uncertain, it is well recognized that the photosynthesis of vitamin D became critically important for landdwelling vertebrates that required a mechanism to increase the efficiency of absorption of dietary calcium. The major physiological function of vitamin D in vertebrates is to maintain extracellular fluid concentrations of calcium and phosphorus within a normal range. Vitamin D accomplishes this by increasing the efficiency of the small intestine to absorb dietary calcium and phosphorus, and by stimulating the mobilization of calcium and phosphorus stores from the bone [3].
II. HISTORICAL PERSPECTIVE It was the lack of appreciation of the beneficial effect of sunlight in preventing rickets that caused the widespread endemic outbreak of this bone-deforming disease in the 17th through 19th centuries. As early as 1822, Sniadecki [4] had suggested that the high incidence of rickets that occurred in Warsaw, Poland, in the early 1800s was likely caused by the lack of adequate exposure to sunlight. This was based on his observations that whereas children who lived in Warsaw had a very high incidence of rickets, children who lived in the rural areas outside of the city did not suffer the same malady. Seventy years later Palm [5] also recognized that lack of sunlight was a common denominator associated with the high incidence of rickets in children VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX
V. Sunlight, Vitamin D, and Skin Cancer VI. Conclusion Acknowledgment References
living in the inner cities when compared to children living in underdeveloped countries. Both Sniadecki [4] and Palm [5] encouraged systematic sunbathing as a means of preventing and curing rickets, but their recommendations went unheeded. Huldschinsky [6] was the first to clearly demonstrate that it was exposure of the skin to ultraviolet radiation that was responsible for the antirachitic activity of sunlight that had been postulated by Sniadecki [4] and Palm [5]. He exposed four rachitic children to radiation from a mercury arc lamp and reported dramatic reversal of rickets within 4 months. He further reasoned that the radiation responsible for producing the antirachitic factor was the same radiation that caused melanization of the skin. The intimate relationship between sunlight and its antirachitic properties in humans was finally demonstrated in 1921 when Hess and Unger [7] took rachitic children and exposed them to sunlight on the roof of a New York City hospital and demonstrated that within 4 months the rachitic lesions had healed. The concept that rickets was caused by a nutritional deficiency was supported by the observation in 1827 by Bretonneau, who treated a 15-month-old child with acute rickets with cod liver oil and noted the incredible speed at which the patient was cured [8]. His student Trousseau used liver oils from a variety of fish and aquatic mammals including herring, whales, and seals for the treatment of rickets and osteomalacia. In the early 20th century, Mellanby [9] conducted classic studies where he demonstrated that rachitic beagles that were fed cod liver oil were cured of their bone-deforming disease. Originally, it was thought that the active antirachitic factor in cod liver oil was vitamin A. However, McCollum et al. [10] heated cod liver oil to destroy the vitamin A activity and demonstrated that the preparation maintained its antirachitic activity. He, therefore, coined the term “vitamin D” for the antirachitic factor. Copyright © 2005, Elsevier, Inc. All rights reserved.
38 The appreciation that exposure to sunlight could cure rickets in children prompted Steenbock and Black [11] and Hess and Weinstock [12] in 1924 to expose a variety of foods and other substances including vegetable oils, animal feed, lettuce, wheat, and grasses to ultraviolet radiation. They found that this process imparted antirachitic activity to all of the substances. These observations led Steenbock [13] to conclude that there would be great utility in irradiation of food substances for the prevention and cure of rickets in children. This concept initially led to the addition of provitamin D (ergosterol or 7-dehydrocholesterol) to milk and its subsequent irradiation to impart antirachitic activity. Once vitamin D (D without a subscript refers to either D2 or D3) was inexpensive to produce, vitamin D was directly added to milk. This simple concept led to the eradication of rickets as a significant health problem in the United States and other countries that followed this practice.
MICHAEL F. HOLICK
SUN
SUN
SUN H HO 7-DEHYDROCHOLESTROL
SUN H
HO
LUMISTEROL3 SUN HO
PREVITAMIN3 CH3
H BLOOD VESSEL
OH TACHYSTEROL3
DBP HO SUN
VITAMIN3
SUN
SUN
III. PHOTOBIOLOGY OF VITAMIN D A. Photosynthesis of Vitamin D3 When human skin is exposed to sunlight, it is the solar ultraviolet B photons with energies between 290 and 315 nm that are responsible for causing the photolysis of 7-dehydrocholesterol (provitamin D3; the immediate precursor in the cholesterol biosynthetic pathway) to previtamin D3 [1,14] (Fig. 1). This photochemical process occurs in the plasma membrane of skin cells; as a result, the thermodynamically unstable cis,cis isomer of previtamin D3 is rapidly transformed by a rearrangement of double bonds to form vitamin D3 [15] (Fig. 2). Approximately 50% of previtamin D3 is converted to vitamin D3 within 2 hr. As vitamin D3 is formed in the membrane, its open flexible structure is likely jettisoned from the plasma membrane into the extracellular space. Once vitamin D3 enters the extracellular fluid space, it is attracted to the vitamin D binding protein (DBP) in the circulation, and thus enters the dermal capillary bed.
B. Regulation of Vitamin D Synthesis by Sunlight In the late 1960s, a theory was popularized that human skin pigmentation evolved to play a critically important role in regulating the cutaneous photosynthesis of vitamin D3. Loomis [16] speculated that melanin pigmentation in humans evolved for the purpose of preventing excessive production of vitamin D3 in the skin due to chronic excessive exposure to
OH SUPRASTEROL I
HO SUPRASTEROL II OH 5,6-TRANSVITAMIN D
FIGURE 1 Photochemical events that lead to the production and regulation of vitamin D3 in the skin. Reprinted with permission from Holick [1]. DBP is the plasma vitamin D binding protein.
sunlight especially in peoples who live near the equator. Although it is likely that increased pigmentation was important for the prevention of skin cancer, there is little evidence that skin pigmentation evolved to prevent vitamin D intoxication. The reason for this is that once previtamin D3 is formed in the skin, it either can isomerize to vitamin D3 or can absorb solar ultraviolet B radiation and isomerize into biologically inactive photoisomers lumisterol and tachysterol [17] (Fig. 1). Furthermore, when vitamin D3 is made, it can either enter into the circulation or absorb solar ultraviolet photons and isomerize to at least three photoproducts including suprasterol I, suprasterol II, and 5,6-transvitamin D3 [18] (Fig. 1). Thus, sunlight itself can regulate the total output of vitamin D3 in the skin by causing the photodegradation of previtamin D3 and vitamin D3. This is the likely explanation for why there are no reported cases of vitamin D intoxication–induced hypercalcemia from chronic excessive exposure to sunlight. Melanin, on the other hand, provides the body with an effective natural sunscreen. Thus, increased melanin pigmentation will reduce the efficiency of the sun-mediated photosynthesis of previtamin D3 [19]. This is particularly
39
CHAPTER 3 Photobiology of Vitamin D
OH
s-trans,s-cis-PRE-D3
SUN
C A
D
B
HO
HEXANE
PRO-D3
19 C D 10 9 3 5 6 7 HO
T1/2= 91 HRS
s-cis,s-cis-PRE-D3 25°C
CH3
SUN HO
VITAMIN D3 T1/2= 8 HRS
SKIN
HO PRO-D3
HO s-cis,s-cis-PRE-D3
FIGURE 2
Photolysis of provitamin D3 (pro-D3) into previtamin D3 (pre-D3) and its thermal isomerization to vitamin D3 in hexane and in lizard skin. In hexane, pro-D3 is photolyzed to s-cis,s-cis-pre-D3. Once formed, this energetically unstable conformation undergoes a conformational change to the s-trans,s-cis-pre-D3. Only the s-cis, s-cis-pre-D3 can undergo thermal isomerization to vitamin D3. The s-cis,s-cis conformer of pre-D3 is stabilized in the phospholipid bilayer by hydrophilic interactions between the 3/3-hydroxyl group and the polar head of the lipids, as well as by van der Waals interactions between the steroid ring and side-chain structure and the hydrophobic tail of the lipids. These interactions significantly decrease the conversion of the s-cis,s-cis conformer to the s-trans,s-cis conformer, thereby facilitating the thermal isomerization of s-cis,s-cis-pre-D3 to vitamin D3. Reprinted with permission from Holick et al. [15].
important in Blacks who live in northern latitudes and who ingest very little dietary vitamin D. It is the likely explanation for why Blacks have lower circulating concentrations of 25-hydroxyvitamin D3 (25OHD) and are more prone to developing vitamin D deficiency [20] (see Chapter 33).
C. Influence of Latitude, Season, and Time of Day on Vitamin D Synthesis It was recognized at the beginning of the 20th century that the incidence of rickets was much higher during the winter and early spring months and that rickets was less prevalent during the summer and autumn [21]. An evaluation of the effect of season on the cutaneous production of vitamin D3 in Boston revealed that during the summer months of June and July, 7-dehydrocholesterol was most efficiently converted to previtamin D3 [22] (Fig. 3). There was a gradual decline in the production of previtamin D3 after August, and there was essentially no previtamin D3 formed in human skin
after November (Fig. 3). Previtamin D3 photosynthesis commenced in the middle of March. To evaluate the influence of latitude on the cutaneous production of vitamin D3, similar studies were conducted in Edmonton, Canada (52°N), Los Angeles (34°N), and San Juan, Puerto Rico (18°N). In Edmonton, the photosynthesis of previtamin D3 essentially ceased by mid-October and did not resume until mid-April. However, in Los Angeles and San Juan, the production of previtamin D3 in the skin occurred throughout the year (Fig. 3). The cutaneous production of vitamin D3 was also evaluated in Boston every hour from the time the sun rose to the time the sun set in the middle of the month on cloudless days for a full year. It was found that during the summer, sunlight was capable of producing vitamin D3 in the skin from 0700 to as late as 1700 hr Eastern Standard Time (EST) (Fig. 4). However, as the zenith angle of the sun increased in the spring and autumn, previtamin D3 photosynthesis in the skin began at approximately 0900 and ceased at approximately 1600 EST [1].
40
MICHAEL F. HOLICK
D. Influence of Sunscreen Use, Melanin, Clothing, Glass, and Plastics on Vitamin D Synthesis
25
20
15
10
5
0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Month
FIGURE 3
Photosynthesis of previtamin D3 after exposure of 7-dehydrocholesterol (7-DHC) to sunlight. Measurements were as follows: in Boston (42°N) after 1 hr () and 3 hr () and total photo products (previtamin D3, lumisterol, and tachysterol) after 3 h in Boston (); in Edmonton, Canada (52°N), after 1 hr (); in Los Angeles (34°N) () and Puerto Rico (18°N) in January (). Reprinted with permission from Webb et al. [22].
Boston Edmonton Bergen
A
% Previtamin D3 formation
8
Any substance such as melanin, clothing, or a sunscreen that absorbs ultraviolet B radiation will reduce the cutaneous production of vitamin D3 [1,23–25]. Sunscreen use is now widely accepted as an effective method for diminishing the damaging effects due to chronic excessive exposure to sunlight. Sunscreens are of great benefit because they can prevent sunburn, skin cancer, and skin damage associated with exposure to sunlight. The solar radiation that is most responsible for causing damage to the skin is the ultraviolet B (UVB) radiation. Because the major function of sunscreens is to absorb solar UVB radiation on the surface of the skin before it penetrates into the deeper viable layers, sunscreen use diminishes the total number of UVB photons that can reach the 7-dehydrocholesterol stores in the viable epidermis to form previtamin D3. Topical application of a sunscreen will substantially diminish or completely prevent the cutaneous production of previtamin D3 [25]. The topical application of
8
C
6
6
4
4
2
2
0
0 Jan
Mar
May
Jul
Sep
8:00
Nov
Buenos Aires Johannesburg Cape Town Ushuaia
10
% Previtamin D3 formation
% Previtamin D3 formation
B
8 6 4 2
10:00 12:00 14:00 16:00 18:00
Hour of Day
Month 10
Boston Edmonton Bergen
% Previtamin D3 formation
% Photoproducts from 7-DHC
30
Buenos Aires Johannesburg Cape Town Ushuaia
D
8 6 4 2 0
0 Jan
Mar
May
Jul
Month
Sep
Nov
8:00
10:00 12:00 14:00 16:00 18:00
Hour of Day
FIGURE 4 Influence of season, time of day, and latitude on the synthesis of previtamin D3 in Northern (A and C) and southern hemispheres (B and D). The hour indicated in C and D is the end of the 1-hr exposure time. (Reproduced with permission from Chen TC 1998 The Photobiology of Vitamin D. In: Holick MF (ed) Vitamin D—Physiology, Molecular Biology and Clinical Applications. Humana Press, Totowa, NJ, pp. 17–37.)
41
CHAPTER 3 Photobiology of Vitamin D
Serum Vitamin D (nmol/L)
80
Without Sunscreen
60 40
With Sunscreen
20 0
−2
0
2
4
8 6 Days
10
12
14
16
FIGURE 5 Effect of sunscreen on vitamin D status. Circulating concentrations of vitamin D were measured in young adults after application of a cream either with a sun protection factor of 8 or without sunscreen (topical placebo cream) following a single exposure to one minimal erythemal dose of simulated sunlight. Reprinted with permission from Matsuoka et al. [23].
60
Serum 25-Hydroxyvitamin D (ng/ml)
Serum Vitamin D3 (ng/ml)
a sunscreen with a sun protection factor (SPF) of 8, followed by whole-body exposure to one minimal erythemal dose of simulated sunlight, prevented any significant increase in the blood levels of vitamin D3 in healthy young adult volunteers. On the other hand, without the sunscreen, there was a marked 10- to 20-fold increase in circulating concentrations of vitamin D3 [23] (Fig. 5). Similarly, since clothing absorbs ultraviolet B radiation, the wearing of clothing on body surfaces prevents the cutaneous production of vitamin D3 in those covered surfaces [24] (Fig. 6). Thus, in cultures that require the covering of almost the entire body with
clothing, such as Bedouins living in the Negev Desert, despite the intense sunlight environment, the Bedouin women have an increased risk of vitamin D deficiency and osteomalacia and their children are more prone to developing vitamin D–deficiency rickets [26]. Although sunscreens and clothing significantly diminish the cutaneous production of vitamin D, for children and young adults there is little concern about their developing vitamin D deficiency by practicing sun protection. The main reason is that the casual everyday limited exposure of sunlight to the face and hands is sufficient to provide the vitamin D requirement. However, elderly people, who are often very concerned about their health and appearance, will consistently apply a topical sunscreen on all sun-exposed areas and/or wear clothing over most sun-exposed areas before going outdoors. Since aging significantly decreases the capacity of the skin to produce vitamin D3 because of the marked decline in 7-dehydrocholesterol in the epidermis, chronic sunscreen use by elderly people can increase the risk of vitamin D deficiency [1,23] (Fig. 7). Because glass and Plexiglas efficiently absorb most if not all ultraviolet B photons, exposure of the skin to sunlight through glass or Plexiglas will not produce any vitamin D3 (Fig. 8).
80 60 40 20 0
None
Summer
Autumn
Summer and Sunscreen
PA IL 40
30
20
10
0 Without Sunscreen
FIGURE 6 Effect of clothing and sunscreen on vitamin D status. Circulating concentrations of vitamin D were measured in human subjects who wore either no clothing, summer-type clothing, autumn-type clothing, or summer-type clothing and sunscreen 24 hr after a whole-body exposure to one minimal erythemal dose of ultraviolet B radiation. Reprinted with permission from Matsuoka et al. [24].
50
FIGURE 7
With Sunscreen
Effect of chronic sunscreen use on vitamin D status. Circulating concentrations of 25-hydroxyvitamin D were measured in adults from Pennsylvania (PA) and Illinois (IL) who always wore a sunscreen or never wore a sunscreen. Reprinted with permission from Matsuoka et al. [25].
42
MICHAEL F. HOLICK
% Previtamin D3 Synthesis
7 6
5.98%
5 4 3 2 1 0
No Shield
ND Glass
ND Plexiglas
ND Plastic
FIGURE 8 Prevention of provitamin D3 formation by UV radiation by common light-shielding materials. The effects of glass, plastic, or Plexiglas (DuPont Chemical Company, Memphis, TN) placed between the simulated sunlight source and the provitamin D3 were measured. ND, Not detectable. Reprinted with permission from Holick [1].
IV. ROLE OF SUNLIGHT AND DIETARY VITAMIN D IN BONE HEALTH, OVERALL HEALTH, AND WELL-BEING Vitamin D deficiency is finally being recognized as an epidemic for all age groups in most industrialized countries [27–35]. Ninety to 95% of our vitamin D requirement comes from exposure to sunlight [36]. Most experts agree that in the absence of sun exposure, 1000 IU of vitamin D is required to satisfy the body’s vitamin D need [37]. If children and adults received sensible exposure to sunlight during the spring, summer and fall, they are able to store vitamin D in their body fat and call upon it during the winter when the sun is incapable of producing vitamin D in the skin. However, because of the increased concern about the risk of skin cancer from sun exposure and because both children and adults have fewer outdoor activities, there is a resurgence of vitamin D deficiency for the population at large. Nesby-O’Dell et al. [38] reported that 42% of African-American women aged 15–49 years were found to be vitamin D–deficient throughout the United States at the end of the winter. Tangpricha et al. reported that 32% of healthy young men and women aged 18–29 years were vitamin D deficient at the end of the winter in the Boston area [31]. Sullivan et al. [35] reported that 48% of young girls aged 9–11 years were vitamin D deficient at the end of the winter and 17% were vitamin D deficient at the end of the summer in Maine. Vitamin D deficiency has also become a major health concern for pregnant women and infants who receive their sole nutrition from breastfeeding [54].
Rickets, which was considered to be a disease of the 19th century, has become much more common, especially in children of color who are solely breastfed [52]. For preadolescent and adolescent children, vitamin D deficiency will prevent them from attaining their genetically programmed peak bone mass. For adults, it will precipitate and exacerbate osteoporosis, increase risk of fracture, and cause the painful bone disease osteomalacia [39–44]. It is now also recognized that insufficient exposure to sunlight and vitamin D deficiency increases risk of many common cancers, including colon, breast, prostate, ovarian, and esophageal cancers and non-Hodgkin’s and Hodgkin’s lymphoma [45–49]. In addition, it increases risk of children developing type 1 diabetes [50] and also puts adults at increased risk of developing hypertension and having heart disease [51–55].
V. SUNLIGHT, VITAMIN D, AND SKIN CANCER There is no question that chronic excessive exposure to sunlight and sun-burning experiences, especially during childhood and young adulthood, increases risk of squamous and basal cell carcinoma and wrinkling [56]. There is, however, little evidence that sensible limited exposure to sunlight to satisfy the body’s requirement for vitamin D will substantially increase risk of developing either basal or squamous cell carcinoma [36]. Furthermore, if basal or squamous cell carcinoma develops and is detected early, this type of cancer is easily curable. Melanoma, on the other hand, is often a deadly skin cancer. Although it has been suggested that melanoma is associated with sun exposure, it should be noted that most melanomas occur on non—sun-exposed areas, i.e., back of leg, abdomen, etc., and rarely occur on the face [57]. Furthermore, a recent study in Scandinavia suggested that melanoma is associated with increased number of nevi (moles), color of the hair, i.e., red and blonde, and number of sun-burning experiences [58]. They also reported that there may be an association with frequent exposure to a tanning bed. However, the data were not statistically significant and the authors concluded that there was only a suggestion of an association. To put this into perspective, it is estimated that approximately 1200 people per year will die in the United States of squamous and basal cell carcinomas that are clearly associated with excessive sun exposure. However, the lack of adequate sun exposure and/or adequate vitamin D nutrition puts at risk more than 150,000 lives for dying of many common cancers [59,60]. There has been no estimate of the toll that vitamin D
CHAPTER 3 Photobiology of Vitamin D
deficiency has in mortality rates due to heart disease and complications of type 1 diabetes, which has now been clearly linked to vitamin D deficiency.
VI. CONCLUSION The skin has a huge capacity to produce vitamin D [61]. Children and adults exposed to natural or artificial ultraviolet B radiation can satisfy their vitamin D requirement [62–66]. Exposure of a healthy adult in a bathing suit to one minimal erythemal dose (a light pinkness to the skin) of sunlight or tanning bed UV radiation is equivalent to taking between 10,000 IU and 20,000 IU of vitamin D [36,61]. Because melanin is such an effective sunscreen, African Americans require five to 50 times the exposure that a white person requires to satisfy their body’s vitamin D requirement [19]. Although aging markedly diminishes the capacity of the skin to produce vitamin D3, because its capacity is so high, elders exposed to sunlight can still raise their blood levels of 25(OH)D into a satisfactory range [67]. Thus, there needs to be a reexamination of the message that any exposure to sunlight requires some type of sunscreen or sun protection. This extreme position has pervaded the psyche of the population at large. It unfortunately has put many people at risk for vitamin D deficiency and many of the serious chronic diseases that have been associated with inadequate sun exposure and vitamin D deficiency. Sensible exposure to sunlight, typically no more than 5 to 10 minutes a day between 10 A.M. and 3 P.M. during seasons when vitamin D can be produced in the skin will satisfy most people’s vitamin D requirement [60,61]. Taking a multivitamin containing 400 IU of vitamin D will satisfy approximately 40% of the body’s requirement. Thus, additional supplementation and/or foods that contain vitamin D are required to satisfy the 1000 IU of vitamin D that most experts believe is necessary to raise the blood levels of 25(OH)D above 30 ng/ml, which is considered to be a healthy level.
Acknowledgment This work was supported in part by the following grants: General Clinical Research Center M01RR00533, R01 AR36963 and the ITA’s UV Foundation.
References 1. Holick MF 1994 McCollum Award Lecture, Vitamin D—New horizons for the 21st century. Am J Clin Nutr 60:619–630.
43 2. Holick MF 1989 Phylogenetic and evolutionary aspects of vitamin D from phytoplankton to humans. In: Pang PKT, Schreibman MP (eds) Vertebrate Endocrinology: Fundamentals and Biomedical Implications, Vol. 3. Academic Press, Orlando, FL, pp. 7–43. 3. Holick MF 1995 Vitamin D: Photobiology, metabolism, and clinical applications. In: DeGroot LJ, Besser M, Burger HG, Jameon JL, Loriaux DL, Marshall JC, Odell WD, Potts JT, Rubenstein AH (eds) Endocrinology, 3rd ed. Chapter 59. Saunders, Philadelphia, pp. 990–1013. 4. Sniadecki J 1939 cited by W Mozolowski, Jerdrzej Sniadecki (1768–1883) on the cure of rickets. Nature 143:121. 5. Palm TA 1980 The geographic distribution and etiology of rickets. Practitioner 45:270–279, 421–442. 6. Huldschinsky K 1919 Curing rickets by artificial UV-radiation (Heilung von Rachitis durch kunstliche Hohensonne). Deut Med Wochenschr 45:712–713 (in German). 7. Hess AF, Unger LF 1921 Cure of infantile rickets by sunlight. JAMA 77:39. 8. Mayer J 1957 Armand Trousseau and the arrow of time. Nutr Rev 15:321–323. 9. Mellanby T 1918 The part played by an accessory factor in the production of experimental rickets. J Physiol 52:11–14. 10. McCollum EF, Simmonds N, Becker JE, Shipley PG 1922 Studies on experimental rickets; and experimental demonstration of the existence of a vitamin which promotes calcium deposition. J Biol Chem 53:293–312. 11. Steenbock H, Black A 1924 The reduction of growth-promoting and calcifying properties in a ration by exposure to ultraviolet light. J Biol Chem 61:408–411. 12. Hess AF, Weinstock M 1924 Antirachitic properties imparted to inert fluids and green vegetables by ultraviolet irradiation. J Biol Chem 62:301–313. 13. Steenbock H 1924 The induction of growth-prompting and calcifying properties in a ration exposed to light. Science 60:224–225. 14. MacLaughlin JA, Anderson RR, Holick MF 1982 Spectral character of sunlight modulates photosynthesis of previtamin D3 and its photoisomers in human skin. Science 216:1001–1003. 15. Holick MF, Tian XQ, Allen M 1995 Evolutionary importance for the membrane enhancement of the production of vitamin D3 in the skin of poikilothermic animals. Proc Natl Acad Sci USA 92:3124–3126. 16. Loomis F 1967 Skin-pigment regulation of vitamin D biosynthesis in man. Science 157:501–506. 17. Holick MF, MacLaughlin JA, Doppelt SH 1981 Factors that influence the cutaneous photosynthesis of previtamin D3. Science 211:590–593. 18. Webb AR, deCosta BR, Holick MF 1989 Sunlight regulates the cutaneous production of vitamin D3 by causing its photodegradation. J Clin Endocrinol Metab 68:882–887. 19. Clemens TL, Adams JS, Henderson SL, Holick MF 1982 Increased skin pigment reduces the capacity of the skin to synthesize vitamin D. Lancet 1:74–76. 20. Bell NH, Greene A, Epstein S, Oexmann MJ, Shaw W, Shary J 1985 Evidence for alteration of the vitamin D endocrine system in Blacks. J Pediatr 76:470–473. 21. Kassowitz M 1987 Tetany and autointoxication in infants (Tetani and autointoxication in kindersalter). Wien Med Presse 97:139 (in Dutch). 22. Webb AR, Kline L, Holick MF 1988 Influence of season and latitude on the cutaneous synthesis of vitamin D3: Exposure to winter sunlight in Boston and Edmonton will not promote vitamin D3 synthesis in human skin. J Clin Endocrinol Metab 67:373–378.
44 23. Matsuoka LY, Ide L, Wortsman J, MacLaughlin JA, Holick MF 1987 Sunscreens suppress cutaneous vitamin D3 synthesis. J Clin Endocrinol Metab 64:1165–1168. 24. Matsuoka LY, Wortsman J, Dannenberg MJ, Hollis BW, Lu Z, Holick MF 1992 Clothing prevents ultraviolet-B radiationdependent photosynthesis of vitamin D3. J Clin Endocrinol Metab 75:1099–1103. 25. Matsuoka LY, Wortsman J, Hanifan N, Holick MF 1988 Chronic sunscreen use decreases circulating concentrations of 25-hydroxyvitamin D: A preliminary study. Arch Dermatol 124:1802–1804. 26. Sedrani SH, Al-Arabi KM, Abanny A, Elidrissy A 1990 Frequency of vitamin D deficiency rickets in Riyadh. In: Study of Vitamin D Status and Factors Leading to Its Deficiency in Saudi Arabia. King Saudi Univ. Press, Riyadh, pp. 281–285. 27. Gloth FM, Gundberg CM, Hollis BW, Haddad HG, Tobin JD 1995 Vitamin D deficiency in homebound elderly persons. JAMA 274:1683–1686. 28. Malabanan A, Veronikis IE, Holick MF 1998 Redefining vitamin D insufficiency. Lancet 351:805–806. 29. Lips P 2001 Vitamin D deficiency and secondary hyperparathyroidism in the elderly: consequences for bone loss and fractures and therapeutic implications. Endocr Rev 22:477–501. 30. Chapuy MC, Preziosi P, Maaner M, Arnaud S, Galan P, Hercberg S, et al. 1997 Prevalence of vitamin D insufficiency in an adult normal population. Osteopor Int 7:439–443. 31. Tangpricha V, Pearce EN, Chen TC, Holick MF 2002 Vitamin D insufficiency among free-living healthy young adults. Am J Med 112:659–662. 32. Kreiter SR, Schwartz RP, Kirkman HN, Charlton PA, Calikoglu AS, Davenport M 2000 Nutritional rickets in African American breast-fed infants. J Pediatr 137:2–6. 33. Welch TR 2000 Vitamin D deficient rickets: the reemergence of a once-conquered disease J Pediatr 137:143–145. 34. Thomas MK, Lloyd-Jones DM, Thadhani RI, Shaw AC, Deraska DJ, Kitch BT 1998 Hypovitaminosis D in medical inpatients. N Engl J Med 338:777–783. 35. Sullivan SS, Rosen CJ, Chen TC, Holick MF 2003 Seasonal changes in serum 25(OH)D in adolescent girls in Maine. J Bone Mine Res (Proceedings for Twenty-Fifth Annual Meeting of the American Society for Bone and Mineral Research), S407. 36. Holick MF 2003 Vitamin D: a millennium perspective. J Cell Biochem 88:296–307. 37. Heaney RP, Davies KM, Chen TC, Holick MF, Barger-Lux MJ 2003 Human serum 25-hydroxycholecalciferol response to extended oral dosing with cholecalciferol. Am J Clin Nutr 77:204–210. 38. Nesby-O’Dell S, Scanlon KS, Cogswell ME, Gillespie C, Hollis BW, Looker AC 2002 Hypovitaminosis D prevalence and determinants among African American and white women of reproductive age: third national health and nutrition examination survey, 1988–1994. Am J Clin Nutr 76:187–192. 39. Holick MF 2001 Sunlight “D”ilemma: risk of skin cancer or bone disease and muscle weakness. Lancet 357:4–6. 40. Glerup H, Mikkelsen K, Poulsen L, Hass E, Overbeck S, Andersen H et al. 2000 Hypovitaminosis D myopathy without osteomalacic bone involvement. Calcif Tissue Int 66:419–424. 41. Chapuy MC, Schott AM, Garnero P, Hans D, Delmas PD, Meunier J, et al. 1996 Healthy elderly French women living at home have secondary hyperparathyroidism and high bone turnover in winter. J Clin Endocrinol Metab 81:1129–1133. 42. Malabanan AO, Turner AK, Holick MF 1998 Severe generalized bone pain and osteoporosis in a premenopausal black female: Effect of vitamin D replacement. J Clin Densitometr 1:201–204.
MICHAEL F. HOLICK
43. Rimaniol J, Authier F, Chariot P 1994 Muscle weakness in intensive care patients: initial manifestation of vitamin D deficiency. Intensive Care Med 20:591–592. 44. Bischoff HA, Stähelin HB, Dick W, Akos R, Knecht M, Salis C, et al. 2003 Effect of vitamin D and calcium supplementation on falls: a randomized controlled study. J Bone Miner Res 18:343–351. 45. Garland CF, Garland FC, Shaw EK, Comstock GW, Helsing KJ, Gorham ED 1989 Serum 25-hydroxyvitamin D and colon cancer: Eight-year prospective study. Lancet 18:1176–1178. 46. Garland CF, Garland FC, Gorham ED, Raffa J 1992 Sunlight, Vitamin D, and Mortality from Breast and Colorectal Cancer in Italy. Biologic Effects of Light. Walter de Gruyter, New York, pp. 39–43. 47. Hanchette CL, Schwartz GG 1992 Geographic patterns of prostate cancer mortality. Cancer 70:2861–2869. 48. Ahonen MH, Tenkanen L, Teppo L, Hakama M, Tuohimaa P 2000 Prostate cancer risk and prediagnostic serum 25-hydroxyvitamin D levels (Finland). Cancer Causes Control 11:847–852. 49. Grant WB 2002 An ecologic study of dietary and solar ultraviolet-B links to breast carcinoma mortality rates. Am Cancer Soc 94:272–281. 50. Hypponen E, Laara E, Jarvelin M-R, Virtanen SM 2001 Intake of vitamin D and risk of type 1 diabetes: a birth-cohort study. Lancet 358:1500–1503. 51. Rostand SG 1979 Ultraviolet light may contribute to geographic and racial blood pressure differences. Hypertension 30:150–156. 52. Li YC, Kong J, Wei M, Chen ZF, Liu SQ, Cao LP 2002 1,25dihydroxyvitamin D3 is a negative endocrine regulator of the renin–angiotensin system. J Clin Invest 110: 229–238. 53. Scragg R, Jackson R, Holdaway IM, Lim T, Beaglehole R 1990 Myocardial infarction is inversely associated with plasma 25-hydroxyvitamin D3 levels: a community-based study. Int J Epidemiol 19:559–563. 54. Zitterman A, Schulze Schleithoff S, Tenderich C, Berthold H, Koefer R, Stehle P 2003 Low vitamin D status: a contributing factor in the pathogenesis of congestive heart failure? J Am Coll Cardiol 41:105–112. 55. Holick MF 2002 Sunlight and vitamin D, both good for cardiovascular health [editiorial] J Gen Intern Med 17: 733–735. 56. Housman TS, Feldman SR, Williford PM, Fleischer Jr AB, Goldman ND, Acostamadiedo JM 2003 Skin cancer is among the most costly of all cancers to treat for the Medicare population. J Am Acad Dermatol 48:425–429. 57. Garland CF, Garland FC, Gorham ED. Rising trends in melanoma. An hypothesis concerning sunscreen effectiveness. Ann Epidemiol 3:103–110. 58. Veierod MB, Weiderpass E, Thorn M, Hansson J, Lund E, Armstrong B, Adami H-O 2003 A prospective study of pigmentation, sun exposure, and risk of cutaneous malignant melanoma in women. J Nat Can Inst 95:1530–1538. 59. Grant WB 2002 An estimate of premature cancer mortality in the U.S. due to inadequate doses of solar ultraviolet-B radiation. Cancer 70:2861–2869. 60. Holick MF, Jenkins M 2003 The UV Advantage. iBooks, New York 61. Holick MF 2002 Vitamin D: the underappreciated D-lightful hormone that is important for skeletal and cellular health. Curr Opin Endocrinol Diabetes 9:87–98. 62. Chel VGM, Ooms ME, Popp-Snijders C, Pavel S, Schothorst AA, Meulemans CCE, et al. 1998 Ultraviolet irradiation corrects vitamin D deficiency and suppresses
CHAPTER 3 Photobiology of Vitamin D
secondary hyperparathyroidism in the elderly. J Bone Miner Res 13:1238–1242. 63. Chuck A, Todd J, Diffey B 2001 Subliminal ultraviolet-B irradiation for the prevention of vitamin D deficiency in the elderly: a feasibility study. Photochem Photoimmun Photomed 17:168–171. 64. Lund B, Sorensen OH 1979 Measurement of 25-hydroxyvitamin D in serum and its relation to sunshine, age, and vitamin D intake in the Danish population. Scand J Clin Lab Invest 39:23–30.
45 65. Adams JA, Clemens TL, Parrish JA, Holick MF 1982 Vitamin-D synthesis and metabolism after ultraviolet irradiation of normal and vitamin-D-deficient subjects. N Engl J Med 306:722–725. 66. Koutkia P, Lu Z, Chen TC, Holick MF 2001 Treatment of vitamin D deficiency due to Crohn’s disease with tanning bed ultraviolet B radiation. Gastroenterology 121:1485–1488. 67. Reid IR, Gallagher DJA, Bosworth J 1985 Prophylaxis against vitamin D deficiency in the elderly by regular sunlight exposure. Age Ageing 15:35–40.
CHAPTER 4
The Vitamin D 25-Hydroxylase MARIELLE GASCON-BARRÉ
Département de Pharmacologie, Faculté de Médecine, Université de Montréal, and Centre de recherche de l’Université de Montréal, Montréal, Québec Canada
I. Introduction II. Hepatic Uptake of Vitamin D III. The Monooxygenases Active in the Hydroxylation of Vitamin D at C-25 IV. The Microsomal Enzymes V. The Mitochondrial Enzyme
VI. Cerebrotendinous Xanthomathosis VII. Ontogeny of the Vitamin D 25-Hydroxylases VIII. Sex Differences in the Hydroxylation of Vitamin D at C-25 IX. Conclusions References
I. INTRODUCTION
mitochondrial enzymes, believed by many to be more physiologically relevant than the mitochondrial entity. In this chapter, we review the most relevant research area on the D 25-hydroxylases and address the specificity and regulation of each enzyme in the context of its dynamic functioning, including uptake, ontogeny, sex-related differences, and hepatic and intrahepatic regionalization.
25-Hydroxyvitamin D (25OHD)* is the first hydroxylated metabolite of vitamin D (D) and the immediate precursor of the fully active and hormonal form of the vitamin, 1α,25-dihydroxyvitamin D [1,25(OH)2D]. It was discovered by DeLuca and his group, who rapidly identified the liver as the first site of activation of D3 [1–3]. Over the past 35 years, the enzyme systems involved in the C-25 hydroxylation of D3, D2, and several of their analogs have been the object of intense studies by groups in North America, Europe, and Japan. The research has allowed the identification of two intrahepatic organelles, the smooth endoplasmic reticulum (microsomes) and the mitochondrium, as sites possessing fully active but distinct D 25-hydroxylases. The mitochondrial enzyme has been cloned [4–6] and its identity as a D3 25-hydroxylase established with certainty. Moreover, its presence and activity has been positively identified in all species studied including the human [7]. The microsomal enzyme received the attention of early workers in the field. It has been identified clearly in the pig, where the enzyme has been cloned and clearly shown to hydroxylate D3 and D2 at C-25 [8]. Lately, a new microsomal D 25-hydroxylase species has also been cloned and its gene transcript shown to be present in mice and humans [9]. The latter is also reported to be active in the C-25 hydroxylation of both D3 and D2. In most species, early work has shown that the microsomal D3 25-hydroxylase is an enzyme also active in the oxidation of several endogenous and exogenous substances and, based on the enzyme kinetics of the respective microsomal and *The
term vitamin D includes both vitamin D3 and vitamin D2.
VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX
II. HEPATIC UPTAKE OF VITAMIN D A. The First Prerequisite Step toward Its Activation The hepatic sequestration of vitamin D is the first prerequisite step toward its activation and subsequent C-25 hydroxylation by the liver. Under normal conditions, the fractional hepatic uptake of D3 during a single pass across the rat and dog liver lies between 40 and 60% [10–12], an uptake higher than that observed for all hydroxylated metabolites of the vitamin [10,13,14]. Its hepatic clearance is estimated to be 357 ml min−1 in dogs [11]. Studies in which total uptake has been investigated for periods varying from 18 sec to 70 min have revealed that the liver does not accumulate significantly more D3 than that observed during the first pass across the organ [13,15–19] suggesting a steady state liver/ serum concentration ratio of 0.4 to 0.6 at physiological concentrations of the vitamin. In addition, data available indicate that there is no stringent regulation of uptake by the vitamin D status [13]. Both hepatocytes and nonparenchymal liver cells are able to sequester D3 and the mitochondrial D3 25-hydroxylase gene transcript has been found in sinusoidal endothelial, stellate, and Kupffer cells of the liver [20]. In a 1981 study, Dueland et al. [16], however, found that only hepatocytes were able to transform the vitamin into 25OHD3. Copyright © 2005, Elsevier, Inc. All rights reserved.
48 The uptake of D3 is not regioselective within the hepatic acinus and not perturbed by the destruction of either the periportal (proximal) or perivenous (distal or pericentral) region [21], indicating that its extraction takes place according to the concentration gradient along the acinus. This observation suggests that the intrahepatic concentration of the vitamin should be higher in periportal than in perivenous hepatocytes. This notion may be important in light of the reported intraacinar localization of the mitochondrial D3 25-hydroxylase, the kinetic parameters of the mitochondrial and microsomal enzymes, and the influence of regioselective liver diseases on the biosynthesis of 25OHD3 (see Chapter 75 for review). In addition, the hepatic extraction of D3 has been shown to be independent of its hepatic venous or arterial route of delivery [22,23], indicating that circulating vitamin D of endogenous or exogenous origin should have similar hepatic availability. In addition, manipulation of the diet in order to obtain different proportions of putative D3 carrier proteins or lipoproteins in vivo was found not to significantly affect the hepatic extraction of D3 when evaluated in isolated perfused liver preparations [22], an observation also made in vitro by Ziv et al. [19]. However, Haddad et al. [15] found that chylomicron remnants or LDL optimized the hepatic uptake of the vitamin in vitro.
III. THE MONOOXYGENASES ACTIVE IN THE HYDROXYLATION OF VITAMIN D AT C-25 Early studies rapidly established that the liver was the site of hydroxylation of D3 at C-25 [3,22,24]. The first experimental evidence obtained indicated that the D3 25-hydroxylase was localized in both the microsomal and mitochondrial fractions of liver homogenates [25]. However, it was soon reported that the enzyme was instead found almost entirely in liver microsomes and that mitochondria exhibited no or very low activity toward D3 [26]. It was also rapidly established that oxygen-18 from molecular oxygen was incorporated at C-25, strongly suggesting that the microsomal enzyme was a mixed-function oxidase [27]. Firm demonstration for the presence of a mitochondrial D3 25-hydroxylase only came several years later [7]. In 1978, studies by Suda’s group using isolated–perfused rat liver preparations demonstrated that the activation of 1αOHD3 was not regulated (Km 2.0 µM) compared to that of the natural substrate, and that two distinct Km values could be defined for the C-25 hydroxylation of D3 (5.6 nM and 1 µM), suggesting the participation of more than one D3 25-hydroxylase [28].
MARIELLE GASCON-BARRÉ
Evaluation of enzyme activity revealed that 25OHD3 production was also present in extrahepatic tissue [29,30]. At that time, support for the extrahepatic presence of a D3 25-hydroxylase came from the in vivo studies of Ponchon et al. [2] and Olson et al. [3], who reported that hepatectomy markedly reduced but did not eliminate the formation of 25OH[3H]D3 after [3H]D3 injection. They, indeed, observed that 10% of the radioactivity appearing in the plasma of intact rats was found as [3H]25OHD3 in hepatectomized animals 4 hr after injection of the parent compound. These observations have now been confirmed at the molecular level with the demonstration of the presence of the mitochondrial D3 25-hydroxylase in numerous tissues and organs (Table I). In the pig, the microsomal D3 25-hydroxylase gene transcript and its associated protein have also been shown to be present not only in the liver but also in several other organs including the kidney [31,32], whereas the gene expression of a mouse and human microsomal D 25-hydroxylase has been reported to be mainly present in the liver and testis (Table II) [9]. The contribution of the extrahepatic enzymes to the circulating concentrations of 25OHD under normal physiological conditions is not, however, presently known, but several laboratories have clearly shown that the mitochondrial and the porcine microsomal enzymes are active in extrahepatic tissues. These observations indicate that the enzymes participate in the local production of 25OHD and possibly, as well, in the production of 1α,25(OH)2D as both enzymes have been reported to also harbor 1α-hydroxylase activity as indicated in Tables III and IV.
IV. THE MICROSOMAL ENZYMES DeLuca’s group was the first to report on the subcellular localization of the D3 25-hydroxylase. Indeed, in the mid-1970s, Bhattacharyya and DeLuca [26,33,34] observed that the D3 25-hydroxylase was present in the microsomal fraction of rat liver and that it was influenced by the vitamin D status and required the presence of a cytosolic factor for optimum activity. They also noticed that the 25-hydroxylation of the synthetic compound dihydrotachysterol (DHT3) was not regulated compared to that of the natural substrate D3 [34]. In 1979, Madhok and DeLuca [35] confirmed the presence of the enzyme in rat liver microsomes and soon after, purification and reconstitution of the enzyme was achieved in Björkhem’s [36] and DeLuca’s [37] laboratories. It was shown to be a cytochrome P450 (P450, CYP) mixed-function oxidase requiring NADPH and NADPH cytochrome c/P450 reductase. Yoon and DeLuca [37] also reported that a soluble
49
CHAPTER 4 The Vitamin D 25-Hydroxylase
TABLE I Expression Sites of the Mitochondrial CYP27A1 Species
Location
Humana
Liver (adult and fetal)
Mouse
Rabbit
Rat
Reference
Cali and Russell, 1991 [6] Gascon-Barré et al. [141] Kidney (adult and fetal) Gascon-Barré et al., 2001 [141] HepG2 hepatoma cells Guo et al., 1993 [87] Fibroblastsc Cali and Russell 1991 [6] Macrophages Hansson et al., 2003 [165] Stroma cells of the ovary Jabara et al., 2003 [166] Arterial endothelial cells Javitt et al., 2002 [167] Keratinocytes Lehmann et al., [168] Parathyroid glands Schuessler et al. [169] Parathyroid tumors Correa et al., [170] Intestine (fetal jejunum Theodoropoulos et al. [120] and colon) Liver Ichikawa et al., 1995 [171] Duodenum Calvaria Lung Skin Long bone Spleen Osteoblasts (primary culture) Liver Andersson et al., 1989 [4] Duodenum Adrenal Lung Kidney Spleen Liverd Usui et al., 1990 [5] Twisk et al., 1995c [118] Liver sinusoidal Theodoropoulos et al. [20] endothelial, stellate and Kupffer cells Intestine Theodoropoulos et al. [119] Ovaries Su et al., 1990 [86] Kidney Mullick et al., 1995 [91] Axén et al., 1995 [112]
cytosolic factor was necessary for full reconstitution of enzyme activity (see Table III). In 1983, Andersson et al. [38] purified to homogeneity a rat microsomal cytochrome P450 active on the C-25 hydroxylation of the bile acid intermediates 5β-cholestane-3α,7α,12α-triol (cholestane-triol, C-triol), 5β-cholestane-3α,7α-diol (C-diol) as well as D3, 1αOHD3 but not D2. A year later, Hayashi et al. [39] purified and partially sequenced a P450 active on the 25-hydroxylation of D3 and suggested that an inhibitor was removed from microsomes during the purification steps as previously suggested by others [36,40]. Moreover, contrary to earlier observations, the reconstituted enzyme did not require the presence of a cytoplasmic factor. The partial amino acid sequence obtained indicated that the enzyme was different from P450 (PB-1) and P450 (MC-1) (CYP2B1 and 1A1, respectively, according to P450 nomenclature [41]). The same group later obtained an N-terminal amino acid sequence and reported that their purified P450 was also active on several endo- and xenobiotics as already reported by others [38,42] as well as D3, 1αOHD3 but
TABLE II Species CYP2D25 Pig
CYP2R1 Mouse
In addition to these sites, enzyme activity has been shown in a human macrophages [172] and b bovine endothelial cells [173]. c SV40 transformed; d in situ hybridization illustrates a preferential perivenous localization within the acinus [118].
Human
Expression Sites of the Microsomal CYP2D25 and CYP2R1 Location
Reference
Liver Kidney Adrenals Brain Heart Intestine Lung Muscle Spleen Thymus
Hosseinpour and Wikvall [52]
Liver Testis Kidney Brain Epididymis Lung Spleen Heart Muscle Skin Liver
Cheng et al. [9]
Cheng et al. [9]
50
MARIELLE GASCON-BARRÉ
TABLE III Species
Sex
Substrate
The Microsomal Vitamin D 25-Hydroxylases Hydroxylation Site
Characteristics
Reference
Km: 180 nMa (D-depleted) 444 nMa (D-repleted) Km: 44 nMa 360 nMa 108 to 205 pmol 30 min−1 0.75 nmol P450−1b
Delvin et al., 1978 [174]
Rat
D3
C-25
Rat
D3 DHT3 D3
C-25 C-25 C-25
D3 D3 1αOHD3 D2 Benzphetamine D3
C-25 C-25 C-25 C-25 Demethylation C-25
D3 1αOHD3 D2 C-triole C-diol f Ethylmorphine D3
C-25 C-25 C-25 C-25 C-25 Demethylation C-25
D3 1αOHD3 Testosterone
Rat
Dehydroepiandrosterone Benzphetamine D3
C-25 C-25 C-2α C-16α C-16α
7.68 pmol hr −1 mg pro−1b,c 4.3 pmol min −1 nmol P450−1b 3 pmol min −1 nmol P450−1b 2 pmol min −1 nmol P450−1b 921 pmol min −1 nmol P450−1b 3.5–4 pmol hr −1106 hepatocytesd Km: 4 µM (D-depleted) 6 µM (D-repleted) 335 pmol min−1 nmol P450−1b 1000 pmol min−1 nmol P450−1b <10 pmol min−1 nmol P450−1b 2600 pmol min−1 nmol P450−1b 360 pmol min−1 nmol P450−1b 21,000 pmol min−1 nmol P450−1b 2.3 nmol min−1 mg pro−1b (152 pmol min−1 nmol P450) 0.21 nmol min−1 nmol P450−1b 1.73 nmol min−1 nmol P450−1b 9.34 nmol min−1 nmol P450−1b 8.36 nmol min−1 nmol P450−1b 14.6 nmol min−1 nmol P450−1b
Demethylation C-25
15.6 nmol min−1 nmol P450−1b 195 pmol min−1 nmol P450−1b
Rat
C-triol Testosterone D3 C-triol* C-diol** D3
C-25 C-16α C-25 C-25 C-25 C-25
580 pmol min−1 nmol P450−1b 1283 pmol min−1 nmol P450−1b 9 pmol min−1 nmol P450−1b 90 pmol min−1 nmol P450−1b 188 pmol min−1 nmol P450−1b 335 pmol min−1 mg pro−1b (167 pmol min−1 nmol P450−1)
D3
C-25
1αOHD3 1αOHD3 D3
C-25 C-25 C-25
6 pmol min−1 mg pro−1b (4 pmol min−1 nmol P450−1) 0.152 nmol min mg pro−1a 0.053 nmol min mg pro−1a 37 pmol min 106 hepatocytesd 29 pmol min 106 hepatocytesd
D3
C-25 C-25 C-25
Rat (phenobarbital pretreated) Rat Beef
Rat
Rat
Rat Rat
Rat Rat (D depleted) (1,25(OH)2D3treated) Rat Human
,
1αOHD3
0.46 pmol hr1 mg pro−1a,c 0.40 pmol hr1 mg pro−1a 0.8–4.7 pmol min−1 mg proa
Madhok and DeLuca, 1979 [35] Björkhem et al., 1979 [36]
Yoon and DeLuca, 1979 [37] Hiwatashi et al., 1980 [42]
Dueland et al., 1981 [16]
Andersson et al., 1983 [38]
Hayashi et al., 1984 [39] Hayashi et al., 1986 [43]
Andersson and Jörnvall, 1986 [45]
Dählback and Wikvall, 1987 [46]
Saarem and Pedersen, 1987 [47] Benbrahim et al., 1988 [175]
Thierry-Palmer et al., 1995 [49]
Continued
51
CHAPTER 4 The Vitamin D 25-Hydroxylase
TABLE III Species
Sex
Pig
,
Piglet (normal) Piglet (PDDR) Piglet (normal) Piglet (PDDR) Piglet (PDDR)
,
Castrated
COS cells (pCYP2D25)
Substrate
Hydroxylation Site
D3 D2 1αOHD3 1αOHD2 25OHD3
C-25 C-25 C-25 C-25 C-1α
1αOHD3 D3
C-25
D2 1αOHD3 25OHD3 1αOHD3 25OHD3 25OHD3 D3 D2
pCYP2D25
m/hCYP2R1
The Microsomal Vitamin D 25-Hydroxylases—Cont’d
,
Characteristics 188 pmol min−1 nmol P450−1 e 98 pmol min−1 nmol P450−1e 756 pmol min−1 nmol P450−1 f 842 pmol min−1 nmol P450−1 f 6 pmol min−1 nmol P450−1e 6 pmol min−1 nmol P450−1e 3 pmol min−1 nmol P450−1 f 1 pmol min−1 nmol P450−1 f 760 pmol min−1 nmol P450−1a 200 pmol min−1 nmol P450−1a Km: 100 nM 110 pmol min−1 nmol P450−1a 475 pmol 48 h−1 mg cell pro−1g 45 pmol 48 h−1 mg cell pro−1g 57 pmol min−1 nmol CYP2D25−1 12 pmol min−1 nmol CYP2D25−1 45 pmol min−1 nmol CYP2D25−1 ND ND
C-25 C-1α C-25 C-1α C-26 C-25 C-25
Reference Axén et al. [51]
Axén et al. [53]
Hosseinpour and Wikvall [52] Hosseinpour et al. [32] Hosseinpour et al. [32] Araya et al. [57]
Cheng et al. [9]
a Liver f Partially
microsomes. b Reconstituted system. cIn the presence of a cytosolic factor. d Freshly isolated hepatocytes. e Partially purified liver microsomal P450. purified kidney microsomal P450. g Kidney pCYP2D25. *5β-cholestane-3α,7α,12α-triol; **5β-cholestane, 3α,7α,12α-diol. ND, Not determined.
TABLE IV Species
Sex
Rat (untreated)
Substrate D3 1αOHD3 DHT3 C-triol* D3 D3
The Mitochondrial Vitamin D 25-Hydroxylase Hydroxylation Site
Rat
D3 D2 D3
C-25 C-25 C-25 C-26(27) C-25 C-25 D3 C-triol* C-25 C-25 C-25 C-25 C-25
Rat
D3
C-25
(Phenobarbital treated) (Rachitagenic diet) Human Human Rat Human
and —
D3
C-25 Rabbit
D3 1αOHD3 25OHD3 C-triol*
C-25 C-25 C-1α C-27
Characteristics 0.02 nmol min−1 nmol P450−1b 0.116 nmol min−1 nmol P450−1b 0.01 nmol min−1 nmol P450−1b 8.7 nmol min−1 nmol P450−1b 0.03 nmol min−1 nmol P450-1b 0.064 nmol min−1 nmol P450−1b 0.16 nmol min−1 nmol P450−1b 40 nmol min−1 nmol P450−1b Km: 10−5 Ma Km: 2 × 10−5 Ma,c 10 pmol min−1 mg pro−1a 2 pmol min−1 mg pro−1a 0.093 nmol hr −1 mg pro−1a 0.435 nmol hr −1 mg pro−1a 550 pmol min −1 mg pro−1a (305 pmol min nmol P450−1a) 525 pmol min nmol P450−1a 350 pmol min nmol P450−1a 395 pmol min nmol P450−1b 1200 pmol min nmol P450−1b < 10 pmol min nmol P450−1b 2500 pmol min nmol P450−1b
Reference Björkhem et al., 1980 [110]
Oftebro et al., 1981 [83] Saarem et al., 1984 [84] Saarem and Pedersen, 1985 [85] Holmberg et al., 1986 [176] Saarem and Pedersen, 1987 [35] Dählback and Wikvall, 1987 [46]
Dählback and Wikvall, 1988 [111]
Continued
52
MARIELLE GASCON-BARRÉ
TABLE IV Species
Sex
Substrate
The Mitochondrial Vitamin D 25-Hydroxylase—Cont’d Hydroxylation Site
Rat
1αOHD3
C-25
Rat Rat
hCYP27A1
C-triol* D3 1αOHD3 C-triol* D2 D3 DHD3 1αOHD2 1αOHD3 1αOHDHT3 D3 1αOHD3 25OHD3 25OHD3 C-triol* 25OHD3
hCYP27A1
D3
C-27 C-25 C-25 C-27 C-25 C-25 C-25 C-25 C-25 C-25 C-25 C-25 C-1α C-27 C-27 C-1α C-27 C-25
25OHD3
C-1α
1αOHD3
C-25 (major) C-24 (minor) C-26 (minor) C-24 (major) C-26 (minor) C-24 (major) C-26 (major) C-25 (minor) C-26 (major) C-25 (minor)
Cos cellsc
E. colid
HepG2
1αOHD2 1αOHD4 24-epi1α OHD3 ∆22-1αOHD3
Characteristics
Reference
Km: 54 µMb 3.8 nmol min−1 nmol P450−1b Km: 6.3 µMb 0.36 nmol min−1 nmol P450−1b 1.4 nmol min−1 nmol P450−1b 36 nmol min−1 nmol P450−1b N/Db 342 pmol hr−1106 cellsb 344 pmol hr−1106 cellsb 16 pmol hr−1106 cellsb 1328 pmol hr−1106 cellsb 1383 pmol hr−1106 cellsb 72 pmol min−1 nmol P450−1b 189 pmol min−1 nmol P450−1b 24 pmol min−1 nmol P450−1b 19 pmol min−1 nmol P450−1b 4337 pmol min−1 nmol P450−1b 5 pmol hr nmol hr min−1 CYP27A1−1 4.5 pmol hr nmol hr min−1 CYP27A1−1 Km = 3.2 µM Vmax = 0.27 mol CYP27A1−1 min−1 Km = 3.5 µM Vmax = 0.021 mol CYP27A1−1 min−1
Masumoto et al., 1988 [80] Okuda et al., 1988 [76] Ohyama et al., 1991 [81]
Guo et al., 1993 [87]
Axén et al., 1994 [99]
Araya et al., [57] Sawada et al., [101]
Byford et al., [105]
C-24 (major) C-26 (major) C-25 (minor)
a Mitochondria; b Reconstituted system. c In the presence of a cytosolic factor. **5β-cholestane, α,7α,12α-diol. N/D, Not detectable.
not D2 [43], further demonstrating the broad spectrum of activity of the enzyme. Sex-related differences have also been observed in the microsomal C-25 hydroxylation of D3. Indeed, Hayashi et al. [43] isolated a D3 25-hydroxylase exhibiting the same N-terminal amino acid sequence as CYP2C11 [44], a P450 known to exist in male rat liver microsomes but not in those of females. This observation may explain why microsomes (or P450 purified from the microsomal fraction) obtained from female livers have consistently been shown, as indicated in Table III, to exhibit significantly less activity in the C-25 hydroxylation of D3 than preparations
d Transfected
with hCYP27. *5β-cholestane-3α,7α,12α-triol;
obtained from their male counterparts [43,45–48] except, possibly, at low substrate concentrations as reported by Thierry-Palmer et al. [49]. Whether Yoon and DeLuca [37] and Andersson et al. [38] reported on the same isoenzyme as Hayashi et al. [43] is not known, as no amino acid sequence was proposed in the former studies, but Andersson and Jörnvall [45] later proposed a partial sequence suggestive of the malespecific testosterone 16α-hydroxylase, which is now known to be a P450 compatible with the male-specific CYP2C11. Using male pig microsomes, Segura-Aguilar [50] reported that the C-25 hydroxylation of D3 is also
CHAPTER 4 The Vitamin D 25-Hydroxylase
catalyzed by a peroxidase as indicated by the use of t-butyl hydroperoxide as electron donor in the absence of NADPH and NADPH-P450 reductase. The Km of the enzyme was found to be 30 µM, a concentration much higher than the reported values in the presence of NADPH and NADPH-P450 (see Table III).
A. The Porcine CYP2D25 Purification of the porcine microsomal D3 25-hydroxylase was achieved between 1992 and 1994 by Axén et al. [31,51]. At the time, the protein was only found in liver and kidney. However, the monoclonal antibody raised against the porcine enzyme also recognized a protein of similar molecular weight in human liver but, to date, the pig homolog of the enzyme has not been cloned in any other species. The purified enzyme was shown to catalyze the C-25 hydroxylation of vitamin D3 and vitamin D2 and both compounds competitively inhibited the C-25 hydroxylation of one another. Moreover, ketoconazole, a nonselective P450 inhibitor, also inhibited the hydroxylation of D3 as well as of D2 [51]. The enzyme has been shown to have a high affinity for D compounds and converts D3 into 25OHD3 with a reported Km of 100 nM, a concentration considered to be within the physiological range. Its order of efficiency at saturating concentrations (25 µM) is reported to be 1αD3 > D3 > D2 with 750, 200 and 110 pmol/nmol P450/min respectively [52]. A year later, Axén [31] showed that the enzyme also catalyzed the hydroxylation of 25OHD3 at C-1α, indicating that the enzyme was not only a D3 25-hydroxylase but also a microsomal 1α-hydroxylase. Interestingly, the enzyme was also shown to exhibit similar 1α-hydroxylation activity in the liver of normal pigs and of pigs suffering from pseudovitamin D deficiency rickets (PDDR) (see Chapter 71), but not in the kidney where the 1α-hydroxylation was significantly decreased by the PDDR genotype (Table III) [53]. The molecular cloning of the porcine hepatic microsomal D3 25-hydroxylase was achieved by Wikvall and his group in 1997 [8], and the gene was subsequently shown to map to chromosome 5 [54]. The gene transcript codes for a protein of 500 amino acids and has a predicted molecular weight of 56,374. At that time, Northern blot analysis revealed that the gene was found in high abundance in the liver, but the analysis also showed a clearly detectable signal in the kidney. Subsequent studies using RT-PCR revealed that the transcript is expressed in several tissues including the intestine, lung, thymus, adrenals, and the brain (Table II) [52]. Transfection in simian COS cells results in the synthesis of an enzyme that is recognized by a monoclonal antibody raised against purified D 25-hydroxylase and
53 catalyzes the C-25 hydroxylation of D. The gene transcript exhibits a 70-80% identity with members of the CYP2D family of P450 and has been termed CYP2D25. CYP2D25 has unique properties among the CYP2D enzymes. Indeed, although it metabolizes the drug tolterodine, a muscarinic receptor antagonist and a substrate for the human CYP2D6, into its 5-hydroxymethylated metabolite, tolterodine also inhibits the 25-hydroxylation of D3 while quinidine, an inhibitor of CYP2D6, does not significantly inhibit the C-25 hydroxylation of D3 [52]. Moreover, it was found that the residues in substrate recognition site 3 (SRS-3) are important determinants for its function as a D 25-hydroxylase but not for its activity in the metabolism of tolterodine [55]. Surprisingly, cyclosporine A, a known inhibitor of the mitochondrial D3 25-hydroxylase, was also shown to significantly inhibit the C-25 hydroxylation activity of CYP2D25 [52]. Phorbol 12-myristate 13-acetate (PMA) down-regulates the expression of CYP2D25 and the subsequent C-25 hydroxylation of 1αOHD3, suggesting the involvement of protein kinase C in the regulation of CYP2D25 [56]. The molecular cloning of the microsomal D3 25-hydroxylase in pig kidney was achieved by Hosseinpour et al. in 2002 [32]. Genomic Southern analysis suggests the presence of a single gene for CYP2D25 in the pig and the DNA sequence and deduced peptide sequence of the renal enzyme are homologous to those of the hepatic CYP2D25. Immunocytochemistry indicates that the enzyme is expressed almost exclusively in cells of the cortical tubules as illustrated in Fig. 1. The kidney enzyme has been shown to catalyze not only the hydroxylation of D3 at C-25 but also that of 25OHD3 at C-1α [32,57]. Moreover, recombinant CYP2D25 not only hydroxylates D3 at C-25 but also converts 25OHD3 into both 1α,25(OH)2D3 and 25,26(OH)2D3 with rates of conversion of 57, 12, and 45 pmol/nmol CYP2D25/min, respectively [57]. Microsomes obtained from human livers have been shown to hydroxylate 1αOHD3 at C-25. They, however, exhibit a wide range of activities (0.7 to 4.7 pmol/ min/mg protein). Despite a high degree of sequence identity between CYP2D25 and the human CYP2D6 (77% in its primary structure), no correlation between the microsomal CYP2D6 content and the capacity to hydroxylate 1αOHD3 at C-25 could be demonstrated [52]. In addition, the same group showed that CYP2D6 overexpressed in insect microsomes did not reveal any C-25 hydroxylation activity. Collectively, these observations clearly indicate that CYP2D6 is not a D 25-hydroxylase in human and definitive identification of the microsomal D3 25-hydroxylase, in species other than the pig, will require further kinetic, immunological, and molecular characterization. However, as indicated later, the recent identification of
54
MARIELLE GASCON-BARRÉ
A
chromosome 11, and the mouse CYP2R1 is reported to be 89% identical in sequence to the human enzyme. The enzyme is sexually dimorphic, is present in highest abundance in the liver and testis (Table II), and is active on both D2 and D3 compounds including 1αOHD3. Its sequence identity with other microsomal D 25-hydroxylases (CYP2D25, CYP2C11) has been reported to vary between 33% and 39% [9]. Biochemical characterization of the enzyme has, however, not yet been published and hence its substrate specificity is not yet known.
B
V. THE MITOCHONDRIAL ENZYME A. The C27 Sterol 27-Hydroxylase
FIGURE 1 Localization of the microsomal D3 25-hydroxylase in pig kidney. Immunohistochemical localization of CYP2D25 in sections of pig kidney incubated with polyclonal antibodies raised against the hepatic CYP2D25. Immunostaining (in red) was seen in the cells of proximal tubules but not in distal tubules (A and B). No staining was detected in glomeruli (B). Control sections not treated with primary antibodies, but otherwise treated in the same way, did not show immunostaining (not shown in figure). Reproduced from Biochimica et Biophysica Acta, 2002, volume 1580, 140 [32], by copyright permission of Elsevier Science B.V.
an orphan receptor of the CYP2 family as a putative D 25-hydroxylase may soon elucidate the role of the endoplasmic reticulum in the hepatic activation of vitamin D in humans.
B. The Human and Mouse CYP2R1 Using a cDNA library from CYP27A1 null mouse liver, Cheng et al. [9] have identified an evolutionarily conserved orphan cytochrome P450 termed CYP2R1 and demonstrated that the biochemical properties of the enzyme were consistent with a D 25-hydroxylase. The CYP2R1 gene transcript has characteristic sequence features associated with cytochromes P450 of the endoplasmic reticulum. It is conserved through evolution from fish to human [58], where it is located on
Bile acid biosynthesis represents the most important pathway in the metabolism and excretion of cholesterol. Although several enzymes participate in the oxidation of the cholesterol molecule, the C27 sterol hydroxylase catalyzes the first and probably the ratelimiting step in the oxidation of the side chain in the “acidic” bile acid biosynthesis pathway [59]. In accordance with IUPA sterol nomenclature [60] and the stereochemistry of the initial reaction catalyzed by the enzyme [61–63], Cali and Russell [6] termed the enzyme sterol 27-hydroxylase (EC 1.14.13.15), although the enzyme exhibits low substrate specificity and has been referred to as the 26-hydroxylase in most early publications [64]. CYP27A1 has now been adopted as the gene symbol. After considerable debate on the subcellular localization of the 27-hydroxylase [44,65,66], the enzyme was established in 1973 to be located in the inner membrane matrix of liver mitochondria and shown to display properties similar to that of P450, catalyzing hydroxylation in the mitochondrial membrane of the adrenal glands [67,68]. These observations were later confirmed by others [69–71]. Thus, sterol 27-hydroxylase is a mixed-function oxidase requiring the two mitochondrial-specific protein cofactors, ferrodoxin (adrenodoxin), an iron–sulfur protein [72], and ferrodoxin (adrenodoxin) reductase, a FAD-containing enzyme [73]. The enzyme catalyzes the C-27 hydroxylation of 5β-cholestane-3α,7α,12β-triol [70], 5β-cholestane3α,7α-diol [65], cholesterol [74–76], and 7-ketocholesterol [77], as well as several other substrates involved in bile acid biosynthesis. Wikvall [78] purified the enzyme to apparent homogeneity from rabbit liver mitochondria. Okuda et al. [76] using female rat liver mitochondria achieved final purification of the 27-hydroxylase and noticed that the enzyme exhibited no activity toward xenobiotics using benzphetamine, 7-ethoxycoumarin, and benzo[a]pyrene as substrates.
55
CHAPTER 4 The Vitamin D 25-Hydroxylase
In 1975, Björkhem et al. [79] observed the presence of the 27-hydroxylase in human liver mitochondria, and a few years later, the same group was the first to report that rat and human mitochondria catalyzed the C-25 hydroxylation of D3 but, at that time, attributed the D3 25-hydroxylation activity to an enzyme different from the cholestane-triol 27-hydroxylase [7].
B. The Mitochondrial D3 25-Hydroxylase Is the C27 Sterol 27-Hydroxylase Masumoto et al. [80] purified to homogeneity the D3 25-hydroxylase from rat liver mitochondria and characterized it as a P450 catalyzing the C-25 hydroxylation of both D3 and 1αOHD3. These researchers proposed that a single enzyme was involved in the C-27 hydroxylation of 5β-cholestane-3α,7α,12α-triol and in the C-25 hydroxylation of D3, although the preparation exhibited much lower activity toward D3 compounds than toward the C-27 hydroxylation of bile acid intermediates as illustrated in Table IV [76]. Ohyama et al. [81] pursued the hypothesis they earlier put forward [80] and found that D3 competitively inhibited the 27-hydroxylation of C-triol while C-triol inhibited D3 25-hydroxylation. They concluded, based on the criteria proposed by Haldane and Dixon [82], that both substrates were catalyzed at a common active site of a single protein. Meanwhile Oftebro et al. [83] and Saarem et al. [84,85] confirmed the original observation of Björkhem et al. [7] that a D3 25-hydroxylase was present in human liver mitochondria and that C27 sterol 27-hydroxylation as well as D3 25-hydroxylation activities were located probably exclusively in the inner mitochondrial membrane matrix [84,85]. Cloning of the rabbit [4], rat [5,86], human [6,87], and pig [8] CYP27A1 (chromosomal location: 2q33qter [88]) was achieved between 1989 and 1997. Cloning of the rabbit 27-hydroxylase by Andersson and co-workers [4] illustrated the hydrophobic nature of the protein (with 36% of the amino acids being either aromatic or hydrophobic side chains), and the presence of a conserved cysteine residue at position 444 that is believed to be the ligand for the heme ion in P450 enzymes. Andersson’s work also confirmed earlier biochemical observations indicating extrahepatic enzyme activity with the demonstration that CYP27A1 mRNAs were present in multiple tissues as compiled in Table I. The predicted amino acid sequence of the rabbit [4], rat [5], and human [6] mitochondrial D3 25-hydroxylase contains a mitochondrial specific presequence of 32–36 amino acids on which the destination of the hemoprotein is dependent, as illustrated by Sakaki et al. [89] using a modified CYP27A1 whose
mitochondrial targeting signal had been replaced by a microsomal N-terminal targeting signal. The construct, indeed, resulted in the microsomal localization of a fully active enzyme, providing adequate electron transfer was made available to the enzyme. The mature protein contains 444 to 501 amino acids depending on the species involved and the rat enzyme molecular weight has been estimated to be 51,182 [5]. Northern analyses demonstrate two mRNA sizes of 1.9 and 2.3 kb for CYP27A1 in liver and fibroblast cells [6,86]. Characterization of the 2.3-kb mRNA indicated that the sequence was identical to the 1.9-kb mRNA in its protein-coding region, except for a 400-nucleotide extended sequence at its 5′ end [90]. The rat CYP27A1 gene contains 11 exons of 80-415 nucleotides that are separated by 10 introns of 83 bases to 10kb [91]. The gene has been shown to be regulated by several hormones as will be discussed later [90–94]. The protein sequence of the human enzyme has been reported to be 72% identical to the rat and 81% identical to the rabbit CYP27A1 [5,6]. Heterologous cell systems confirmed the requirement for adrenodoxin and adrenodoxin reductase and not NADPH P450 reductase [4] as well as the identity of sterol 27-hydroxylase and D3 25-hydroxylase. Indeed, Usui et al. [95] prepared an expression plasmid encoding the rat liver D3 25-hydroxylase and transfected it in simian COS cells. They observed both sterol C-27 and D3 C-25 hydroxylation activities in a solubilized mitochondrial fraction supplemented with adrenodoxin and NADPH-adrenodoxin reductase. Akiyoshi-Shibata et al. [96] and Sakaki et al. [89] expressed the cDNA encoding the precursor protein of the rat liver D3 25-hydroxylase in Saccharomyces cerevisiae and found that the mitochondrial fraction exhibited both D3 25-hydroxylase and C-triol 27-hydroxylase activities. These data clearly confirmed the identity of the two enzymes and ruled out the possibility that the simian cell line contained endogenous 27-hydroxylase.
C. Specificity of the CYP27A1 CYP27A1 is a high-capacity enzyme involved in the hydroxylation of bile acid intermediates, but its substrate specificity is broad and exceeds the field of bile acid biosynthesis. As discussed ealier, the purified enzyme has been shown to be active on both the C-25 and C-27 hydroxylation of cholesterol [44,64,68], and to hydroxylate 5β-cholestane-3α,7α-diol and 5β-cholestane- 3α,7α,12α-triol at C-27. The enzyme has also been shown to carry out the oxidation of the product of the reaction, 5β-cholestane-3α,7α,12α,27-tetrol, to
56 3α,7α,12α-trihydroxy-5β-cholestanoic acid [97], as well as to substitute for the classical pathway of bile acid biosynthesis involving an NAD-coupled alcohol dehydrogenase. In a manner similar to other mitochondrial P450, CYP27A1 contains a proline-rich sequence located in the amino-terminal region of the molecule that serves to reduce the tendency of the polypeptide to misfold prior to heme binding [98]. The rabbit, rat, human, pig, and bovine enzyme hydroxylates D3 compounds at C-25 as well as at many other positions on the side chain of the molecule [5,86,87,99–101]. Indeed, CYP27A1 has been shown to catalyze the C-1α, 24, 25, 26, and 27 hydroxylation of D3, D2 and related compounds [87,100–105]. In addition, the pig and human kidney CYP27A1 catalyzes the formation of a metabolite expected to be, according to its mass spectrometric properties, 4β,25(OH)2D3 [57], a metabolite already reported by others in the rat [106,107]. While D3 substrates were found to be preferably hydroxylated at C-25, several groups have observed that when the substrate exhibits the ergocalciferol side chain such as that found in D2 or 1αOHD2, a predominance of 24-hydroxy metabolites occurs with some production of 26- and 27-hydroxylated products. These metabolites have also been found in vivo in rat, cow, chicken, and human [87,102,103,108,109]. Surprisingly, D2 is not hydroxylated at position 25 while 1αOHD2 is only poorly hydroxylated as indicated in Table IV. In addition, several laboratories have now shown that the enzyme prefers 1α-hydroxylated analogs of D3 over their nonhydroxylated counterparts, and this includes the natural substrate D3, which is hydroxylated at C-25 several times less efficiently than 1αOHD3 (see Table IV) [81,87,110,111]. The observation that the 27-hydroxylase is far more efficient at hydroxylating compounds exhibiting a 1α-hydroxyl group than toward the natural substrate is in full accordance with Ohyama et al. [81], who observed that the activity of the enzyme toward bile acid intermediates was proportional to the number of hydroxyl groups on the nucleus. Björkhem [64] also raised the hypothesis that the binding of nonpolar substrates such as vitamin D3 or cholesterol to the active site of the enzyme may differ from the corresponding binding of more polar substrates such as 5β-cholestane-3α,7α,12α-triol. Jones and co-workers [104] have also shown that lengthening the side chain of the vitamin D molecule is quite well tolerated in that the enzyme is able to continue to efficiently hydroxylate D3 compounds including its analogs. These studies led to the proposal of a model of interaction between enzyme and substrate whereby the 25-hydroxylase appears to be directed to its terminal hydroxylation site by the distance from the end of the side chain. These data were also taken to
MARIELLE GASCON-BARRÉ
suggest that the substrate binding pocket of CYP27A1 can tolerate a variety of steroidal shapes either by accommodating only the terminal carbons of the side chain, or by having a specific cleft for the side chain within a broader pocket for the vitamin D/steroidal ring structure. Using several side-chain analogs of D3, the same group showed that in the CYP27A1-containing hepatoma cell HepG2, 1αOHD3 is mainly hydroxylated at C-25 with only minor C-24 and C-26 hydroxylations. 1αOHD2, however, is mainly hydroxylated at C-24 with minor hydroxylation at C-26 while both 1αOHD4 and ∆22-1αOHD3 are mainly hydroxylated at C-24 and C-26 with only minor production of their 25-hydroxylated products. These observations led the authors to conclude that the C-24 methyl group plays a crucial role in the determination of the CYP27A1directed C-24 hydroxylation. They also concluded, based on their studies with 1αOHD4 and ∆22-1αOHD3, that the C22–C23 double bond also contributes to the altered metabolism observed with D2 or D2-derived compounds [105]. Lately, Wikvall and co-workers [57,99,112] have reported that the 27-hydroxylase purified from pig and rabbit livers as well as recombinant human CYP27A1 expressed in Escherichia coli or monkey COS cells was also able to catalyze the 1α-hydroxylation of 25OHD3, albeit at a much lower rate than that observed for the conversion of D3 into 25OHD3. Interestingly, 25OHD3 1α-hydroxylase activity has been reported previously in pig [113] and fish [114] as well as in fetal and adult rat livers [115,116]. Whether or not the reported activity was due to the 27-hydroxylase is unknown as the adult rat liver 1α-hydroxylation activity was found not in the mitochondrium but in the microsomal fraction [116].
D. Regulation of CYP27A1 and Enzyme Activity CYP27A1 mRNA t1/2 has been reported to be between 13 and 24 hr in the rat liver as well as in rat hepatocytes [20,117]. In addition, many hormones, endogenous products, hepatocyte nuclear factors (HNFs), and xenobiotics have been shown to regulate the gene encoding CYP27A1 as well as its gene product. They include bile acids, growth hormone, IGF-1, glucocorticoids, insulin, 1,25(OH)2D3, thyroxine, cyclosporine A, protease inhibitors, fibrates, lipopolysaccharides (LPSs), TNFα, IL-1β, HNF1, HNF4α, and the physiological state of the animal [20,86,91,93,94,118–126]. In the rat [92,93] but not the rabbit [127], CYP27A1 is highly sensitive to the prevailing concentrations of bile acids with increases in enzyme activity, steady-state
CHAPTER 4 The Vitamin D 25-Hydroxylase
mRNA levels, and the rate of gene transcription following interruption of the enterohepatic circulation of bile acids [93,118] while increases in bile acid concentrations are known to down-regulate CYP27A1 [93,126,128,129]. 7α(OH)-4-Cholesten-3-one is an inhibitor of CYP27A1 in the pig and, in a manner similar to the microsomal CYP2D25, PMA down-regulates the expression of CYP27A1, an observation suggesting the participation of protein kinase C in the regulation of its transcription [56]. Bile acids have been reported to suppress CYP27A1 transcription by decreasing its binding to HNF-1, a transcription factor expressed in liver, intestine, pancreas, and kidney [129,130]. Similarly, HNF4α, which is a strong stimulator of CYP27A1 gene transcription, is also down-regulated by bile acids. It has, therefore, been postulated that the inhibitory effect of bile acids on CYP27A1 could be mediated through a repression of HNF4α gene transcription and protein expression rather than through a direct effect on the bile acid response element (BARE). Cholic acid has also been reported to affect CYP27A1 mRNA stability [122]. In accordance with the latter observations, Bolt et al. [131] have shown that cholestasis induced by bile-duct ligation in rats significantly inhibited the 25-hydroxylation of D3 and that addition of bile salts to liver homogenates also had an inhibitory effect on the enzyme. By contrast, biliary cirrhosis does not seem to directly affect the total hepatic output of 25OHD3 in vivo. Indeed, Plourde et al. [132] observed by directly sampling the hepatic effluent after intraportal D3 injection that the 25-hydroxylation of D3 was similar in dogs with interruption of the enterohepatic circulation by choledococystostomy anastomosis and biliary cirrhosis secondary to bile-duct ligation. These observations still leave open any definitive conclusion as to the effect of bile salts on the C-25 hydroxylation of D3 by the mitochondrial and/or microsomal enzyme(s) in vivo (see Chapter 75). The role of insulin in CYP27A1 was illustrated when bile acid production via the sterol 27-hydroxylase was shown to be inhibited up to 58% after 24 hr incubation of hepatocytes with 140 nmol/liter insulin [94]. The decrease in enzyme activity could be explained by a concomitant reduction in CYP27A1 (–62%) mRNA level as well as in transcriptional activity (–75%) [94]. Thus physiological concentrations of insulin seem to down-regulate CYP27A1 gene transcription through a direct effect of the hormone on the hepatocytes [94], although the exact locus of interaction of insulin within the gene has not been identified. Avadhani’s group first reported regulation of CYP27A1 by pituitary-regulated steroids, growth hormone, and the diurnal rhythm with a twofold increase in enzyme activity observed in the mid-dark compared
57 to the mid-light period [93]. They subsequently showed that CYP27A1 is regulated by glucocorticoids as illustrated by a sevenfold induction in mRNA levels in isolated hepatocytes exposed to dexamethasone [117]. The response to the drug is now known to be due to a dexamethasone response element located in the promoter region of the CYP27A1 gene, which has been mapped to positions between −792 and −1095 bp [122,126]. In addition to glucocorticoids up-regulating the enzyme, other drugs known to regulate CYP27A1 include cyclosporine A, protease inhibitors, and fibrates (see also Chapters 73 and 74). At the enzyme level, cyclosporine A has been shown to inhibit the C-25 hydroxylation of both D3 and cholesterol through a direct inhibition of the gene encoding the sterol 27-hydroxylase [133,134]. Indeed, a cyclosporine A response element has been identified on the CYP27A1 gene promoter at 1087–4000 bp upstream of the putative ATG start site [122]. Protease inhibitors have also been shown to decrease the C-25 hydroxylation of D3 in a human hepatocyte cell line, suggesting that patients on antiviral therapy may have an increased risk of 25OHD insufficiency [123]. The fibrate hypolipidemic drugs, on the other hand, have been reported to decrease bile acid synthesis in rat hepatocytes and to suppress CYP27A1 mRNA and enzyme activity by transcriptional and posttranscriptional mechanisms involving the peroxisome proliferator–activated receptor α (PPARα) [124]. The presence of regulatory mechanisms related to the vitamin D status as modulators of the 25-hydroxylation of the vitamin has long been a widely accepted concept [30,33]. In addition, in the early 1980s, Baran and Milne [135] and Bell et al. [136] put forward the hypothesis that 1,25(OH)2D3 might be an inhibitor of the D3 25-hydroxylase. Later studies indicated that 1,25(OH)2D3 administration influenced the in vivo handling of D3 by accelerating its metabolism as well as by increasing its metabolic and biliary clearances, but the molecular mechanisms involved in the regulation of CYP27A1 by the D status were investigated much later [12,137–139]. Axén et al. [112] first reported that CYP27A1 located in kidney and liver was affected by 1,25(OH)2D3 administration, but they showed that CYP27A1 mRNA was decreased to a greater extent in the kidney than in the liver. On the other hand, Theodoropoulos et al. [20,119] investigated the effect of D3 nutritional status as well as that of 25OHD3 and 1,25(OH)2D3 on the rat hepatic and intestinal CYP27A1 and its associated enzyme activity. The hepatic CYP27A1 gene transcript was not found to be sensitive to the D3 or 25OHD3 status but was highly sensitive to 1,25(OH)2D3 exposure with a 60% decrease in its steady-state mRNA levels
58
MARIELLE GASCON-BARRÉ
and a parallel decrease in its protein and enzyme activity (Figs. 2 and 3) [20]. The effect of 1,25(OH)2D3 was shown to be mediated by a significant decrease in the CYP27A1 transcription rate, whereas its mRNA t1/2 remained unchanged. By contrast, intestinal CYP27A1 mRNA levels were shown to be sensitive to D3 and 25OHD3 as well as 1,25(OH)2D3 exposure. 1,25(OH)2D3 administration resulted in a decrease in both intestinal CYP27A1 mRNA t1/2 and transcription rate, and CYP27A1 enzyme activity was shown to be significantly decreased after 1,25(OH)2D3 administration in vivo [119]. In addition, in the human fetal intestine kept in primary culture, 1,25(OH)2D3 was shown to decrease CYP27A1 mRNA levels by 29% after 48 hr of incubation in the presence of the hormone [120]. Collectively, these data indicate that 1,25(OH)2D3
significantly influences CYP27A1 mRNA levels and enzyme activity in the major D-metabolizing organs as well as in target organs. It also illustrates that organspecific differences are present in the regulation of the CYP27A1 gene transcript and response to D3 or its metabolites. Interestingly, Chen and Chiang [126] have suggested that the vitamin D receptor (VDR) (which has now been shown to be present in the liver [140]) could be a mediator in the inhibition of CYP27A1 by conjugated bile acids.
E. Intraacinar Localization of CYP27A1 Several studies have indicated that the CYP27A1 gene transcript is heterogeneously distributed within the
A CYP27A
18S
B CYP27A/18S (Arbitary Units)
1.0
0.8
0.6
0.4
0.2
0.0 0
1
3
5
7
Length of repletion with 1,25 (OH)2D3 (days)
FIGURE 2
Influence of 1,25(OH)2D3 administration to vitamin D–depleted rats on the hepatic D3 25-hydroxylase gene transcript rCYP27A1. Steady-state levels of hepatic rCYP27A1 mRNA levels in hypocalcemic D-depleted rats repleted with 1,25(OH)2D3 after 1 to 7 days of repletion (28 pmol/day). (A) Representative Northern blot analyses of the CYP27A1 gene transcript. (B) Densitometric analyses. Data are presented as means ± SEM. Statistically significant differences between group means were analyzed by ANOVA, with individual contrasts evaluated by the Bonferroni post hoc test. Main effect P <0.001. Statistically significant different from D-depleted rats: *P<0.0001, **P<0.0001; n = 3 animals/group. Reproduced from the American Journal of Physiology (Endocrinology and Metabolism), 2003, volume 284, p. E143 [20] by copyright permission of the American Physiological Society.
59
CHAPTER 4 The Vitamin D 25-Hydroxylase
A
D
G
B
E
H
C
F
I
FIGURE 3 Influence of 1,25(OH)2D3 administration to vitamin D depleted rats on the hepatic intra-acinar distribution of the D3 25-hydroxylase gene transcript CYP27A1. Representative photomicrograph of rat liver sections obtained after in situ hybridization with an rCYP27A antisense riboprobe. A 10-cycle RT-PCR amplification was used. Liver sections were obtained from D-depleted rats (A and B), normal rats (D and E), and 7-day 1,25(OH)2D3-repleted rats (28 pmol/day) (G and H). Negative in situ hybridization control sections with the rCYP27A sense riboprobe are presented in C, F, and I for livers obtained from D-depleted, normal, and 1,25(OH)2D3-repleted rats, respectively. Reproduced from the American Journal of Physiology (Endocrinology and Metabolism), 2003, volume 284, p. E144 [20] by copyright permission of the American Physiological Society.
hepatic acinus with higher perivenous than periportal mRNA levels in both rat and human livers [118,141]. In the human liver, in situ hybridization illustrates a clear CYP27A1 label in hepatocytes that increases in intensity in the perivenous region of the hepatic acinus [141]. In addition, regulation studies in the rat have shown increased CYP27A1 gene transcription, mRNA levels, and production of 27-hydroxycholesterol as well as a recruitment of hepatocytes expressing CYP27A1 toward the mid- to periportal region of the liver acinus when the enterohepatic circulation was interrupted by a bile salt sequestrant [93,118]. These data clearly indicate that CYP27A1 is subject to feedback regulation (in intensity and intrahepatic regionalization) at the transcription level by bile acids returning via the portal blood [93,118]. A study on the effect of the vitamin D status has, however, shown that in the vitamin D–depleted rat liver, 1,25(OH)2D3 decreased CYP27A1 in both the periportal and perivenous regions (Fig. 3) [20]. Thus the impact of 1,25(OH)2D3 on the CYP27A1
gene transcript is more widely distributed across the hepatic acinus than that of bile acids.
VI. CEREBROTENDINOUS XANTHOMATOSIS Mutations that affect either the level of expression of CYP27A1 or its primary sequence lead to cerebrotendinous xanthomatosis (CTX) [142,143]. CTX is an autosomal recessive inherited inborn error of bile acid biosynthesis leading to defects in sterol metabolism and storage characterized by xanthomatosis, atherosclerosis, progressive neurological dysfunction, and dementia. In these patients, however, both abnormal [143–146] and normal [147] vitamin D and/or calcium metabolism have been reported, which gives credence to the presence of more than one enzyme active in the 25-hydroxylation of D in humans, or to a contra-up-regulation of CYP27A1 in cases of mild mutations [148]. Indeed, some CTX
60 patients have been shown to retain residual 27-hydroxylase activity [44], while the human cell line Hep3B has been shown to efficiently activate 1αOHD3 at C-25 despite the absence of detectable CYP27A1 mRNA when evaluated by Northern analysis [149,150]. Rosen et al. [151] were the first to report on a mouse with a disrupted CYP27A1 (CYP27A1−/−). Surprisingly, despite several biochemical markers indicating disruption of bile acid and cholesterol metabolism, no CTXrelated pathological abnormalities were observed in the mice, an observation also reported by others [152]. In addition, CYP27A1−/− mice were shown to have supranormal 25OHD3 (106 compared to 43 ng/ml in CYP27A1+/+) and normal 1,25(OH)2D3 serum concentrations. The latter observations suggest that in the mouse, as in many other species, D3 25-hydroxylase(s) other than CYP27A1 are active in the endogenous synthesis of 25OHD3, an observation that has been confirmed by the identification of CYP2R1 as a D 25hydroxylase in the liver of CYP27A1−/− mice [9]. These observations also suggest that, in the CYP27A1−/− mice, an up-regulation of these enzyme(s) takes place. However, marked species differences in the mechanisms controlling the synthesis of bile acid and cholesterol metabolism have been documented [77,152–154]. Indeed, in contrast to patients suffering from CTX, CYP27A1−/− mice exhibit a dramatic increase in CYP3A activity. This leads to a marked up-regulation in the sidechain oxidation of bile acid biosynthesis, an increase not observed in CTX despite accumulation of cholestane 3α,7α,12α-triols [154]. The increase in CYP3A is secondary to the elevation of toxic bile acid intermediates in the CYP27A1−/− mouse which are potent activators of the xenobiotic/pregnane X receptor (PXR) in the mouse but only weak activators of PXR in humans. The increase in PXR acts as a feed-forward regulatory loop by which toxic bile acid intermediates activate PXR and induce their own metabolism through a PXRmediated induction of CYP3A. Because of the reported differences in bile acid and cholesterol metabolism between a mouse and human with disrupted CYP27A1, the CYP27A1−/− mouse may, therefore, not be an adequate model of the CTX-related defects in D3 anabolism observed in patients stricken with the disease. These observations also illustrate that D 25-hydroxylases other than CYP27A are most likely present in human as suggested by the putative role of the microsomal CYP2R1 in the C-25 hydroxylation of the vitamin and the capacity of CYP3A4 to act as a D 25-hydroxylase [9,58]. In additional support for this hypothesis, a study reported by Byford et al. [105] has indicated that the human CYP27A1-negative cell line Hep 3B is able to transform [3H]1αOHD4 into its 25-hydroxylated
MARIELLE GASCON-BARRÉ
metabolite at low nanomolar concentration, further suggesting the presence of a D 25-hydroxylase other than CYP27A1 in human.
VII. ONTOGENY OF THE VITAMIN D 25-HYDROXYLASES The fetal rat liver can adequately sequester D3 [155]. Moreover, when compared to values obtained in the late fetal period, D3 uptake decreased significantly during the first 2 weeks post partum and then dramatically increased (sixfold) between days 14 and 22 post partum when it reached an uptake capacity similar to that previously observed in the adult rat. The liver of 18- (3 days before birth), 19-, and 22-day-old rat fetuses hydroxylates D3 at C-25 and the value was shown to remain unchanged at day 1 after birth [156,157]. In one study [156], the microsomal 25-hydroxylase activity was shown to sharply increase from day 2 to 11 and to remain unchanged until day 60 of chronological age, whereas in another study [157], maximal enzyme activity was reached only near weaning. These authors also noticed that a cytosolic factor was necessary for optimum 25-hydroxylase activity in fetuses, neonates, and adult rats. In that context, it has been suggested that cytosol contains a cellular D binding protein [158] that could facilitate access of the substrate to the enzyme as proposed by Tsankova et al. [159]. In the pig, expression of the microsomal CYP2D25 has been shown to be similar in the newborn and adult liver, but its expression was found to be lower in the newborn than in the adult kidney [32]. The data suggest a tissue-specific developmental regulation of CYP2D25 in the pig. In contrast, the CYP27A1 gene transcript can easily be detected in the human fetal liver, kidney, jejunum, and colon. Moreover, the CYP27A1 gene transcript was shown to be regulated by 1,25(OH)2D3 in the fetal intestine [120,141]. However, in 16- to 20-week-old human fetuses, the level of the CYP27A1 gene transcript was found to be approximately 40% lower in the fetal liver than in the adult liver [141], but the level of expression of the fetal human kidney CYP27A1 gene transcript was similar to that of the adult kidney [141]. In human, 25OHD has been shown to be lower in cord than in maternal serum, although a significant relationship was observed between the two sites [160]. Premature infants have low circulating 25OHD concentrations compared to infants born at term [161,162], which rapidly increases, however, in the subsequent 4 days after birth [162], indicating fully operational D 25-hydroxylase activity in preterm infants.
61
CHAPTER 4 The Vitamin D 25-Hydroxylase
VIII. SEX DIFFERENCES IN THE HYDROXYLATION OF VITAMIN D AT C-25 Despite the observation that the circulating 25OHD concentrations are not significantly different in males and females, several studies indicate that both the microsomal and the mitochondrial enzyme activities may be gender specific and/or regulated by sex hormones. Indeed, CYP2C11, which exhibits D3 25-hydroxylase activity, is known to be male specific in the rat and rabbit but not in human [88]. Not surprisingly, the enzyme activity reported in the rat has been largely shown to be higher in males than females [45,46]. In addition, Hayashi et al. [48] concluded, based on immunoprecipitation techniques, that the D3 25-hydroxylase found in the male rat liver was a different entity from that found in the female, suggesting that a microsomal P450 distinct from CYP2C11 is also able to catalyze the 25-hydroxylation of vitamin D in rodents. By contrast, the mitochondrial enzyme CYP27A1 has been shown in the human and rat liver to be more abundant in females than in males with a 43% higher expression in female than in male rat livers, although the regulation mechanisms of the enzyme by sex hormones is not presently known [47,141,163]. In addition, Saarem and Pedersen [47] observed that both enzyme activities increased when estradiol was injected into male rats while testosterone decreased the activities of both enzymes in female rats. Addya et al. [163] also reported that the relative 25-hydroxylase levels were reduced in castrated females but increased in castrated males. Although regulatory studies have not yet been carried out on the microsomal CYP2R1, investigation of the levels of its gene transcript in male and female mouse livers indicates the absence of sexual dimorphism [9].
IX. CONCLUSIONS In liver, both microsomes and mitochondria harbor monooxygenase systems active on the 25-hydroxylation of the endogenous substrate D3, but also on several analogs of the vitamin. This allows the design of molecules with therapeutic potential that can be activated by a single hydroxylation step by the liver. The mitochondrial sterol 27-hydroxylase is a multifunctional enzyme exhibiting high capacity with a reported Km for D3 several times higher (micromolar range) than the circulating steady-state concentration of the substrate in vivo. The dynamic liver regulation of enzyme distribution associated with uptake of the substrate favoring selective acinar accumulation seems
to allow, however, the necessary conditions within the hepatocyte microenvironment for efficient transformation of D into 25OHD in vivo. In addition, the presence of CYP27A1 in multiple extrahepatic tissues and organs is suggestive of an enzyme system displaying pleiotrophic functions that are able to support the C-25 hydroxylation of the vitamin in target organs. Whether or not these sites participate in the toxicity of the vitamin, or act, in a tissue-specific manner, for the in situ delivery of 25OHD to the renal and/or the extra-renal 1α-hydroxylase as observed in vitro, remains to be investigated. The microsomal D 25-hydroxylase was originally thought to be a high-specificity low-capacity enzyme. It has now been shown that the enzymes described to date also have low substrate specificity and exhibit regulation by sex hormones and preferential intraacinar localization to the perivenous area (CYP2C11) [164]. The porcine hepatic and kidney CYP2D25 has been characterized and shown to be a high-affinity enzyme and to hydroxylate D3 and D2 compounds at both C-25 and C-1α. The human and mouse CYP2R1 has also been reported to activate D3 and D2 compounds. However, the relative importance of the microsomal and mitochondrial enzyme in the C-25 hydroxylation of vitamin D is not presently known. Indeed, Hosseinpour et al. [56] in a study in pig hepatocytes that possess both CYP2D25 (microsomes) and CYP27A1 (mitochondria) activities found that both enzymes equally participate in the 25-hydroxylation of D3, whereas Cheng et al. [9] reported that CYP27A1 was more abundantly expressed than CYP2R1 (microsomes) in the mouse liver, but the comparative contribution of each enzyme to the formation of 25OHD is not yet known. In conclusion, the liver is essential to the normal homeostasis of vitamin D. Indeed, it activates the vitamin of endogenous and exogenous origin, and efficiently exports 25OHD to the systemic circulation where its concentration represents the best marker of the vitamin D nutritional status. The presence of D microsomal and mitochondrial 25-hydroxylases (which also exhibit C-1α activity) in extrahepatic tissues also raises the hypothesis of an auto/para/intracrine system of D action in D target organs.
X. ADDENDUM While this book was in press, the biological relevance of CYP2R1 as a D 25-hydroxyalse was illustrated by the elucidation by Cheng et al. [177] of a molecular defect in the CYP2R1 gene which was found to be responsible
62 for low serum 25OHD and the classical symptoms of rickets. The affected individual had a normal CYP27A1 gene but was found to be homozygous for a transition mutation in exon 2 of the CYP2R1 gene which resulted in the elimination of D 25-hydroxylase enzyme activity. These observations establish CYP2R1 as a biologically important D 25-hydroxylase and illustrate, for the first time, that a mutation in the CYP2R1 gene which abolishes enzyme activity can lead to a rare molecular autosomal recessive disorder, a selective 25OHD deficiency.
References 1. Blunt JW, DeLuca HF, Schnoes HK 1968 25-hydroxycholecalciferol. A biologically active metabolite of vitamin D3. Biochemistry 7:3317–3322. 2. Ponchon G, Kennan AL, DeLuca HF 1969 “Activation” of vitamin D by the liver. J Clin Invest 48:2032–2037. 3. Olson EB Jr., Knutson JC, Bhattacharyya MH, DeLuca HF 1976 The effect of hepatectomy on the synthesis of 25-hydroxyvitamin D3. J Clin Invest 57:1213–1220. 4. Andersson S, Davis DL, Dahlback H, Jornvall H, Russell DW 1989 Cloning, structure, and expression of the mitochondrial cytochrome P-450 sterol 26-hydroxylase, a bile acid biosynthetic enzyme. J Biol Chem 264: 8222–8229. 5. Usui E, Noshiro M, Okuda K 1990 Molecular cloning of cDNA for vitamin D3 25-hydroxylase from rat liver mitochondria. FEBS Lett 262:135–138. 6. Cali JJ, Russell DW 1991 Characterization of human sterol 27-hydroxylase. A mitochondrial cytochrome P-450 that catalyzes multiple oxidation reaction in bile acid biosynthesis. J Biol Chem 266:7774–7778. 7. Bjorkhem I, Holmberg I 1978 Assay and properties of a mitochondrial 25-hydroxylase active on vitamin D3. J Biol Chem 253:842–849. 8. Postlind H, Axén E, Bergman T, Wikvall K 1997 Cloning, structure, and expression of a cDNA encoding vitamin D3 25-hydroxylase. Biochem Biophys Res Commun 241: 491–497. 9. Cheng JB, Motola DL, Mangelsdorf DJ, Russell DW 2003 De-orphanization of cytochrome P450 2R1. A microsomal vitamin D 25-hydroxylase. J Biol Chem 278: 38084–38093. 10. Gascon-Barré M, Huet PM 1982 Role of the liver in the homeostasis of calciferol metabolism in the dog. Endocrinology 110:563–570. 11. Gascon-Barré M, Vallières S, Huet PM 1986 Influence of phenobarbital on the hepatic handling of [3H]vitamin D3 in the dog. Am J Physiol (Gastrointest Liver Physiol) 251:G627–G635. 12. Gascon-Barré M, Gamache M 1991 Contribution of the biliary pathway to the homeostasis of vitamin D3 and of 1,25-dihydroxyvitamin D3. Endocrinology 129:2335–2344. 13. Rojanasathit S, Haddad JG 1976 Hepatic accumulation of vitamin D3 and 25-hydroxyvitamin D3. Biochim Biophys Acta 421:12–21. 14. Gascon-Barré M, Vallières S, Huet PM 1986 Uptake of the hormone 1,25-dihydroxyvitamin D3 by the dog liver. Can J Physiol Pharmacol 64:699–702.
MARIELLE GASCON-BARRÉ
15. Haddad JG, Jennings AS, Choon Aw T 1988 Vitamin D uptake and metabolism by perfused rat liver: Influences of carrier proteins. Endocrinology 123:498–504. 16. Dueland S, Holmberg I, Berg T, Pedersen JI 1981 Uptake and 25-hydroxylation of vitamin D3 by isolated rat liver cells. J Biol Chem 256:10430–10434. 17. Dueland S, Helgerud P, Pedersen JI, Berg T, Drevon CA 1983 Plasma clearance, transfer, and distribution of vitamin D3 from intestinal lymph. Am J Physiol 245:E326–E331. 18. Gascon-Barré M, Elbaz H, Therrien-Ferland D 1985 Sequestration and microsomal C-25 hydroxylation of [3H]vitamin D3. Metabolism 34:244–250. 19. Ziv E, Bar-On H, Silver J 1985 Vitamin D3 uptake by the isolated perfused rat liver from lipoprotein fractions is separate from cholesterol and triglyceride uptake. Eur J Clin Invest 15:95–99. 20. Theodoropoulos C, Demers C, Petit JL, Gascon-Barré M 2003 High sensitivity of the rat hepatic vitamin D3-25 hydroxylase CYP27A to 1,25-dihydroxyvitamin D3 administration. Am J Physiol 284:E138–E147. 21. Gascon-Barré M, Vallières S, Benbrahim N 1992 C-25 hydroxylation of vitamin D3 in the periportal and perivenous region of the hepatic acinus. Am J Physiol (Endocrinol Metab) 262:E810–E817. 22. Gascon-Barré M, Huet PM, St-Onge Brault G, Brault A, Kassissia I 1988 Liver extraction of vitamin D 3 is independent of its venous or arterial route of delivery. Studies in isolated-perfused rat liver preparations. J Pharmacol Exp Ther 245:975–981. 23. Horsting M, DeLuca HF 1969 In vitro production of 25-hydroxycholecalciferol. Biochem Biophys Res Commun 36:251–256. 24. Ponchon G, DeLuca HF 1969 The role of the liver in the metabolism of vitamin D. J Clin Invest 48:1273–1279. 25. DeLuca HF 1971 Metabolism and mechanism of action of 25-hydroxycholecalciferol. In: Nicholes G Jr., Wasserman RH (eds) Cellular Mechanisms for Calcium and Homeostasis. Academic Press, New York, pp. 426–430. 26. Bhattacharyya MH, DeLuca HF 1974 Subcellular location of rat liver calciferol-25-hydroxylase. Arch Biochem Biophys 160:58–62. 27. Madhok TC, Schnoes HK, DeLuca HF 1978 Incorporation of oxygen-18 into the 25-position of cholecalciferol by hepatic cholecalciferol 25-hydroxylase. Biochem J 175: 479–482. 28. Fukushima M, Nishii Y, Suzuki M, Suda T 1978 Comparative studies on the 25-hydroxylations of cholecalciferol and 1α-hydroxycholecalciferol in perfused rat liver. Biochem J 170:495–502. 29. Tucker GI, Gagnon RE, Haussler MR 1973 Vitamin D3-25hydroxylase: Tissue occurrence and apparent lack of regulation. Arch Biochem Biophys 155:47–57. 30. Bhattacharyya MH, DeLuca HF 1974 The regulation of calciferol-25-hydroxylase in the chick. Biochem Biophys Res Commun 59:734–741. 31. Axén E 1995 Purification from pig kidney of a microsomal cytochrome P450 catalyzing 1α-hydroxylation of 25-hydroxyvitamin D3. FEBS Lett 375:277–279. 32. Hosseinpour F, Norlin M, Wikvall K 2002 Kidney microsomal 25- and 1α-hydroxylase in vitamin D metabolism: catalytic properties, molecular cloning, cellular localization and expression during development. Biochim Biophys Acta 1580:133–144. 33. Bhattacharyya MH, DeLuca HF 1973 The regulation of rat liver calciferol-25-hydroxylase. J Biol Chem 248:2969–2973.
CHAPTER 4 The Vitamin D 25-Hydroxylase
34. Bhattacharyya MH, DeLuca HF 1973 Comparative studies on the 25-hydroxylation of vitamin D3 and dihydrotachysterol. J Biol Chem 248:2974–2977. 35. Madhok TC, DeLuca HF 1979 Characteristics of the rat liver microsomal enzyme system converting cholecalciferol into 25-hydroxycholecalciferol. Biochem J 184:491–499. 36. Björkhem I, Hansson R, Holmberg I, Wikvall K 1979 25-Hydroxylation of vitamin D3 by a reconstituted system from rat liver microsomes. Biochem Biophys Res Commun 90:615–622. 37. Yoon PS, DeLuca HF 1980 Resolution and reconstitution of soluble components of rat liver microsomal vitamin D3-25hydroxylase. Arch Biochem Biophys 203:529–541. 38. Andersson S, Holmberg I, Wikvall K 1983 25-hydroxylation of C27-steroids and vitamin D3 by a constitutive cytochrome P-450 from rat liver microsomes. J Biol Chem 258:6777–6781. 39. Hayashi S, Noshiro M, Okuda K 1984 Purification of cytochrome P-450 catalyzing 25-hydroxylation of vitamin D3 from rat liver microsomes. Biochem Biophys Res Commun 121:994–1000. 40. Holmberg I 1984 Inhibition of reconstituted vitamin D3 25hydroxylase by a protein fraction from rat liver microsomes. Biochem Biophys Res Commun 123:1209–1214. 41. Nelson DR, Kamataki T, Waxman DJ, Guengerich FP, Estabrook RW, Feyereisen R, Gonzalez FJ, Coon MJ, Gunsalus IC, Gotoh O, Okuda K, Nebert DW 1993 The P450 superfamily: Update on new sequences, gene mapping, accession numbers, early trivial names of enzymes, and nomenclature. DNA Cell Biol 12:1–51. 42. Hiwatashi A, Ichikawa Y 1980 Purification and organspecific properties of cholecalciferol 25-hydroxylase system: Cytochrome P-450D25-linked mixed function oxidase system. Biochem Biophys Commun 97:1443–1449. 43. Hayashi S, Noshiro M, Okuda K 1986 Isolation of a cytochrome P-450 that catalyzes the 25-hydroxylation of vitamin D3 from rat liver microsomes. J Biochem 99:1753–1763. 44. Okuda KI 1994 Liver mitochondrial P450 involved in cholesterol catabolism and vitamin D activation. J Lipid Res 35:361–372. 45. Andersson S, Jornvall H 1986 Sex differences in cytochrome P-450-dependent 25-hydroxylation of C27-steroids and vitamin D3 in rat liver microsomes. J Biol Chem 261: 16932–16936. 46. Dählback H, Wikvall K 1987 25-hydroxylation of vitamin D3 in rat liver: Roles of mitochondrial and microsomal cytochrome P-450. Biochem Biophys Res Commun 142: 999–1005. 47. Saarem K, Pedersen JI 1987 Sex differences in the hydroxylation of cholecalciferol and of 5β-cholestane-3α,7α,12αtriol in rat liver. Biochem J 247:73–78. 48. Hayashi SI, Usui E, Okuda K 1988 Sex-related difference in vitamin D3 25-hydroxylase of rat liver microsomes. J Biochem 103:863–866. 49. Thierry-Palmer M, Free AL, Nagappan PR, Butler AR 1995 Cholecalciferol 25-hydroxylation is similar in liver microsomes from male and female rats when cholecalciferol concentration is low. J Nutr 125:104–111. 50. Segura-Aguilar J 1996 Peroxidase activity of liver microsomal vitamin D 25-hydroxylase and cytochrome P450 1A2 catalyzes 25-hydroxylation of vitamin D3 and oxidation of dopamine to aminochrome. Biochem Mol Med 58:122–129. 51. Axén E, Bergman T, Wikvall K 1994 Microsomal 25-hydroxylation of vitamin D2 and vitamin D3 in pig liver. J Steroid Biochem Mol Biol 51:97–106.
63 52. Hosseinpour F, Wikvall K 2000 Porcine microsomal vitamin D3 25-hydroxylase (CYP2D25): Catalytic properties, tissue distribution and comparison with human CYP2D6. J Biol Chem 275:34650–34655. 53. Axén E, Harmeyer J, Wikvall K 1998 Renal and hepatic 1α-hydroxylation of 25-hydroxyvitamin D3 in piglets suffering from pseudo vitamin D-deficiency rickets, type I. Biochim Biophys Acta 1407:234–242. 54. Ciobanu DC, Gill RK, Rothschild MF, Bell NH 2000 Rapid communication: Porcine vitamin D-25-hydroxylase maps to chromosome 5′. J Anim Sci 78:3193–3194. 55. Hosseinpour F, Hidestrand M, Ingelman-Sundberg M, Wikvall K 2001 The importance of residues in substrate recognition site 3 for the catalytic function of CYP2D25 (vitamin D 25-hydroxylase). Biochem Biophys Res Commun 288:1059–1063. 56. Hosseinpour F, Ibranovic I, Tang W, Wikvall K 2003 25-Hydroxylation of vitamin D3 in primary cultures of pig hepatocytes: evidence for a role of both CYP2D25 and CYP27A1. Biochem Biophys Res Commun 303:877–883. 57. Araya Z, Hosseinpour F, Bodin K, Wikvall K 2003 Metabolism of 25-hydroxyvitamin D3 by microsomal and mitochondrial vitamin D3 25-hydroxylases (CYP2D25 and CYP27A1): a novel reaction by CYP27A1. Biochim Biophys Acta 1632:40–47. 58. Nelson DR 2003 Comparison of P450s from human and fugu: 420 million years of vertebrate P450 evolution. Arch Biochem Biophys 409:18–24. 59. Bjorkhem I 1985 Mechanism of bile acid biosynthesis in mammalian liver. In: Danielsson H, Sjovall J. (eds) Sterols and Bile Acids. Elsevier Science, New York, pp. 231–278. 60. Popjak G, Edmond J, Anet FAL, Easton NR, Jr. 1977 Carbon-13 NMR studies on cholesterol biosynthesized from [13C]mevalonates. J Am Chem Soc 99:931–935. 61. Danielsson H 1960 On the oxidation of 3α,7α,12α-trihydroxycoprostane by mouse and rat liver homogenates. Acta Chem Scand 14:348–352. 62. Shefer S, Cheng FW, Batta AK, Dayal B, Tint GS, Salen G, Mosbach EH 1978 Stereospecific side chain hydroxylations in the biosynthesis of chenodeoxycholic acid. J Biol Chem 253:6386–6392. 63. Atsuta Y, Okuda K 1981 On the stereospecificity of cholestanetriol 26-monooxygenase. J Biol Chem 256: 9144–9146. 64. Björkhem I 1992 Mechanism of degradation of the steroid side chain in the formation of bile acids. J Lipid Res 33:455–471. 65. Björkhem I, Gustafsson J 1973 Omega-hydroxylation of steroid side-chain in biosynthesis of bile acids. Eur J Biochem 36:201–212. 66. Cronholm T, Johansson G 1970 Oxidation of 5β-cholestane3α,7α,12α-triol by rat liver microsomes. Eur J Biochem 373–381. 67. Taniguchi S, Hoshita N, Okuda K 1973 Enzymatic characteristics of CO-sensitive 26-hydroxylase system for 5β-cholestane-3α,7α,12α-triol in rat-liver mitochondria and its intramitochondrial localization. Eur J Biochem 40:607–617. 68. Bjorkhem I, Gustafsson J 1974 Mitochondrial omegahydroxylation of the cholesterol side chain. J Biol Chem 249:2528–2535. 69. Okuda KI, Weber P, Ullrich V 1977 Photochemical action spectrum of the CO-inhibited 5β-cholestane-3α,7α,12α-triol 26-hydroxylase system. Biochem Biophys Res Commun 74:1071–1076.
64 70. Sato R, Atsuta Y, Imai Y, Taniguchi S, Okuda KI 1977 Hepatic mitochondrial cytochrome P-450: isolation and functional characterization. Proc Natl Acad Sci USA 74:5477–5481. 71. Pedersen JI, Oftebro H, Vanngard T 1977 Isolation from bovine liver mitochondria of a soluble ferredoxin active in a reconstituted steroid hydroxylation reaction. Biochem Biophys Res Commun 76:666–673. 72. Okamura T, John ME, Zuber MX, Simpson ER, Waterman MR 1985 Molecular cloning and amino acid sequence of the precursor form of bovine adrenodoxin: Evidence for a previously unidentified COOH-terminal peptide. Proc Natl Acad Sci USA 82:5705–5709. 73. Hanukoglu I, Gutfinger T 1989 cDNA sequence of adrenodoxin reductase. Identification of NADP-binding sites in oxidoreductases. Eur J Biochem 180:479–484. 74. Ugele B, Locher M, Burger HJ, Gebhardt R 1986 Is there a heterogeneity of liver parenchyma in taurocholate uptake? In: Paumgartner G, Stiehl A, Gerok W (eds) Bile Acids and the Liver. MTP Press Ltd., Boston, pp. 153–160. 75. Pedersen JI, Godager HK 1978 Purification of NADPHferredoxin reductase from rat liver mitochondria. Biochim Biophys Acta 525:28–36. 76. Okuda K, Masumoto O, Ohyama Y 1988 Purification and characterization of 5β-cholestane-3α,7α,12α-triol 27-hydroxylase from female rat liver mitochondria. J Biol Chem 263:18138–18142. 77. Lyons MA, Maeda N, Brown AJ 2002 Paradoxical enhancement of hepatic metabolism of 7-ketocholesterol in sterol 27-hydroxylase-deficient mice. Biochim Biophys Acta 1581:119–126. 78. Wikvall K 1984 Hydroxylations in biosynthesis of bile acids. Isolation of a cytochrome P-450 from rabbit liver mitochondria catalyzing 26-hydroxylation of C27-steroids. J Biol Chem 259:3800–3804. 79. Björkhem I, Gustafsson J, Johansson G, Persson B 1975 Biosynthesis of bile acids in man. Hydroxylation of the C27-steroid side chain. J Clin Invest 55:478–486. 80. Masumoto O, Ohyama Y, Okuda K 1988 Purification and characterization of vitamin D 25-hydroxylase from rat liver mitochondria. J Biol Chem 263:14256–14260. 81. Ohyama Y, Masumoto O, Usui E, Okuda K 1991 Multi-functional property of rat liver mitochondrial cytochrome P-450. J Biochem 109:389–393. 82. Dixon M, Webb EC 1979 The investigation of enzyme specificity. In: Dixon M, Webb EC (eds), Enzymes. Longman, London, pp. 232–234. 83. Oftebro H, Saarem K, Bjorkhem I, Pedersen JI 1981 Side chain hydroxylation of C27-steroids and vitamin D3 by a cytochrome P-450 enzyme system isolated from human liver mitochondria. J Lipid Res 22:1254–1264. 84. Saarem K, Bergseth S, Oftebro H, Pedersen JI 1984 Subcellular localization of vitamin D3 25-hydroxylase in human liver. J Biol Chem 259:10936–10960. 85. Saarem K, Pedersen JI 1985 25-hydroxylation of 1α-hydroxyvitamin D-3 in rat and human liver. Biochim Biophys Acta 840:117–126. 86. Su P, Rennert H, Shayiq RM, Yamamoto R, Zheng Y-M, Addya S, Strauss JFI, Avadhani NG 1990 A cDNA encoding a rat mitochondrial cytochrome P450 catalyzing both the 26-hydroxylation of vitamin D3: Gonadotropic regulation of the cognate mRNA in ovaries. DNA Cell Biol 9:657–665. 87. Guo YD, Strugnell S, Back DW, Jones G 1993 Transfected human liver cytochrome P-450 hydroxylates vitamin D analogs at different side-chain positions. Proc Natl Acad Sci USA 90:8668–8672.
MARIELLE GASCON-BARRÉ
88. Nebert DW, McKinnon RA 1994 Cytochrome P450: Evolution and functional diversity. In: Boyer JL, Ockner RK (eds) Progress in Liver Diseases. WB Saunders, Philadelphia, pp. 63–97. 89. Sakaki T, Akiyoshi-Shibata M, Yabusaki Y, Ohkawa H 1992 Organella-targeted expression of rat liver cytochrome P450c27 in yeast. Genetically engineered alteration of mitochondrial P450 into a microsomal form creates a novel functional electron transport chain. J Biol Chem 267: 16497–16502. 90. Shayiq RM, Avadhani NG 1992 Sequence complementarity between the 5′-terminal regions of mRNAs for rat mitochondrial cytochrome P-450c27/25 and a growth hormone– inducible serine protease inhibitor. A possible gene overlap. J Biol Chem 267:2421–2428. 91. Mullick J, Addya S, Sucharov C, Avadhani NG 1995 Localization of a transcription promoter within the second exon of the cytochrome P-450c27/25 gene for the expression of the major species of two-kilobase mRNA. Biochemistry 34:13729–13742. 92. Wiese RJ, Uhland-Smith A, Ross TK, Prahl JM, DeLuca HF 1992 Up-regulation of the vitamin D receptor in response to 1,25-dihydroxyvitamin D3 results from ligand-induced stabilization. J Biol Chem 267:20082–20086. 93. Vlahcevic ZR, Jairath SK, Heuman DM, Stravitz RT, Hylemon PB, Avadhani NG, Pandak WM 1996 Transcriptional regulation of hepatic sterol 27-hydroxylase by bile acids. Am J Physiol 270:G646–G652. 94. Twisk J, Hoekman FM, Lehmann EM, Meijer P, Mager WH, Princen HMG 1995 Insulin suppresses bile acid synthesis in cultured rat hepatocytes by down-regulation of cholesterol 7α-hydroxylase and sterol 27-hydroxylase gene transcription. Hepatology 21:501–510. 95. Usui E, Noshiro M, Ohyama Y, Okuda K 1990 Unique property of liver mitochondrial P450 to catalyze the two physiologically important reactions involved in both cholesterol catabolism and vitamin D activation. FEBS Lett 274:175–177. 96. Akiyoshi-Shibata M, Usui E, Sakaki T, Yabusaki Y, Noshiro M, Okuda K, Ohkawa H 1991 Expression of rat liver vitamin D3 25-hydroxylase cDNA in Saccharomyces cerevisiae. FEBS Lett 280:367–370. 97. Holmberg-Betsholtz I, Lund E, Bjorkhem I, Wikvall K 1993 Sterol 27-hydroxylase in bile acid biosynthesis. Mechanism of oxidation of 5β-cholestane-3α,7α,12α,27-tetrol into 3α,7α,12α-trihydroxy-5β-cholestanoic acid. J Biol Chem 268:11079–11085. 98. Kusano K, Kagawa N, Sakaguchi M, Omura T, Waterman MR 2001 Importance of a prolife-rich sequence in the amino-terminal region for correct folding of mitochondrial and soluble microbial p450s. J Biochem 129:271–277. 99. Axén E, Postlind H, Sjoberg H, Wikvall K 1994 Liver mitochondrial cytochrome P450 CYP27 and recombinant-expressed human CYP27 catalyze 1α-hydroxylation of 25-hydroxyvitamin D3. Proc Natl Acad Sci USA 91: 10014–10018. 100. Arlazoroff A, Roitberg B, Werber E, Shidlo R, Berginer VM 1991 Epileptic seizure as a presenting symptom of cerebrotendinous xanthomatosis. Epilepsia 32:657–661. 101. Sawada N, Sakaki T, Ohta M, Inouye K 2000 Metabolism of vitamin D3 by human CYP27A1. Biochem Biophys Res Commun 273:977–984. 102. Horst RL, Koszewski NJ, Reinhardt TA 1990 1α-Hydroxylation of 24-hydroxyvitamin D2 represents a minor physiological pathway for the activation of vitamin D2 in mammals. Biochemistry 29:578–582. 103. Mawer EB, Jones G, Davies M, Still PE, Byford V, Schroeder NJ, Makin HLJ, Bishop CW, Knutson JC 1998
65
CHAPTER 4 The Vitamin D 25-Hydroxylase
104.
105.
106. 107.
108. 109. 110. 111. 112.
113.
114. 115.
116. 117.
118.
119.
Unique 24-hydroxylated metabolites represent a significant pathway of metabolism of vitamin D2 in humans: 24-hydroxyvitamin D2 and 1,24-dihydroxyvitamin D2 detectable in human serum. J Clin Endocrinol Metab 83:2156–2166. Dilworth FJ, Scott I, Green A, Strugnell S, Guo YD, Roberts EA, Kremer R, Calverley MJ, Makin HLJ, Jones G 1995 Different mechanisms of hydroxylation site selection by liver and kidney cytochrome P450 species (CYP27 and CYP24 involved in vitamin D metabolism. J Biol Chem 270: 16766–16774. Byford V, Strugnell S, Coldwell R, Schroeder N, Makin HLJ, Knutson JC, Bishop CW, Jones G 2002 Use of vitamin D4 analogs to investigate differences in hepatic and target cell metabolism of vitamins D2 and D3. Biochim Biophys Acta 1583:151–166. Thierry-Palmer M, Gray TK, Napoli JL 1988 Ring hydroxylation of 25-hydroxycholecalciferol by rat renal microsomes. J Steroid Biochem 29:623–628. Rao DS, Dayal R, Siu-Caldera ML, Horst RL, Uskokovic MR, Tserng KY, Reddy GS 1999 Isolation and identification of 4,25-dihydroxyvitamin D2: A novel A-ring hydroxylated metabolite of vitamin D2. J Steroid Biochem Mol Biol 71:63–70. Jones G, Schnoes HK, Levan L, DeLuca HF 1980 Isolation and identification of 24-hydroxyvitamin D2 and 24,25-dihydroxyvitamin D2. Arch Biochem Biophys 202:450–457. Wichmann JK, Schnoes HK, DeLuca HF 1981 Isolation and identification of 24(R)-hydroxyvitamin D3 from chicks given large doses of vitamin D3. Biochemistry 20:2350–2353. Bjorkhem I, Holmberg I, Oftebro H, Pedersen JI 1980 Properties of a reconstituted vitamin D3 25-hydroxylase from rat liver mitochondria. J Biol Chem 255:5244–5249. Dählback H, Wikvall K 1988 25-Hydroxylation of vitamin D3 by a cytochrome P-450 from rabbit liver mitochondria. Biochem J 252:207–213. Axén E, Postlind H, Wikvall K 1995 Effects on CYP27 mRNA expression in rat kidney and liver by 1α,25dihydroxyvitamin D3, 1α-hydroxylase activity. Biochem Biophys Res Commun 215:136–141. Hollis BW 1990 25-hydroxyvitamin D3–1α-hydroxylase in porcine hepatic tissue: Subcellular localization to both mitochondria and microsomes. Proc Natl Acad Sci USA 87:6009–6013. Takeuchi A, Okano T, Kobayashi T 1991 The existence of 25-hydroxyvitamin D3-1α-hydroxylase in the liver of carp and bastard halibut. Life Sci 48:275–282. Takeuchi A, Okano T, Sekimoto H, Kobayashi T 1994 The enzymatic formation of 1α,25-dihydroxyvitamin D3 from 25-hydroxyvitamin D3 in the liver of fetal rats. Comp Biochem Physiol [C] 109C:1–7. Negrea LA, Slotapolsky E, Dusso AS 1995 1,25-dihydroxyvitamin D synthesis in rat liver microsomes. Horm Metabol Res 27:461–464. Stravitz RT, Vlahcevic ZR, Russell TL, Heizer ML, Avadhani NG, Hylemon PB 1996 Regulation of sterol 27-hydroxylase and an alternative pathway of bile acid biosynthesis in primary cultures of rat hepatocytes. J Steroid Biochem Mol Biol 57:337–347. Twisk J, Hoekman MFM, Mager WH, Moorman AFM, De Boer PAJ, Scheja L, Princen HMG, Gebhardt R 1995 Heterogenous expression of cholesterol 7α-hydroxylase and sterol 27-hydroxylase genes in the rat liver lobulus. J Clin Invest 95:1235–1243. Theodoropoulos C, Demers C, Mirshahi A, Gascon-Barré M 2001 1,25-dihydroxyvitamin D3 downregulates the rat
120.
121.
122.
123. 124.
125. 126. 127.
128.
129.
130. 131. 132. 133.
134.
135.
intestinal vitamin D3-25-hydroxylase CYP27A. Am J Physiol 281:E315–E325. Theodoropoulos C, Demers C, Delvin E, Ménard D, GasconBarré M 2003 Calcitriol regulates the expression of the gene encoding all three vitamin D3 hydroxylases and the drug metabolizing enzyme CYP3A4 in the human fetal intestine. Clin Endocrinol 58:489–499. Memon RA, Moser AH, Shigenaga JK, Grunfeld C, Feingold KR 2001 In vivo and in vitro regulation of sterol 27-hydroxylase in the liver during the acute phase response. Potential role of hepatocyte nuclear factor-1. J Biol Chem 276:30118–30126. Segev H, Honigman A, Rosen H, Leitersdorf E 2001 Transcriptional regulation of the human sterol 27-hydroxylase gene (CYP27) and promoter mapping. Atherosclerosis 156:339–347. Cozzolino M, Vidal M, Arcidiacono MV, Tebas P, Yarasheski KE, Dusso AS 2003 HIV-protease inhibitors impair vitamin D bioactivation to 1,25-dihydroxyvitamin D. AIDS 17:513–520. Post SM, Duez H, Gervois PP, Staels B, Kuipers F, Princen HMG 2001 Fibrates suppress bile acid synthesis via peroxisome proliferator—Activated receptor-α-mediated downregulation of cholesterol 7α-hydroxylase and sterol 27-hydroxylase expression. Arterioscler Thromb Vasc Biol 2001:1840–1845. Araya Z, Tang W, Wikvall K 2003 Hormonal regulation of the human sterol 27-hydroxylase gene CYP27A1. Biochem J 372:529–534. Chen W, Chiang JYL 2003 Regulation of human sterol 27-hydroxylase gene (CYP27A1) by bile acids and hepatocyte nuclear factor 4α (HNF4α). Gene 313:71–82. Araya Z, Sjoberg H, Wikvall K 1995 Different effects on the expression of CYP7 and CYP27 in rabbit liver by cholic acid and cholestyramine. Biochem Biophys Res Commun 216: 868–873. Nakajima S, Hsieh JC, MacDonald PN, Galligan MA, Haussler CA, Whitfield GK, Haussler MR 1994 The C-terminal region of the vitamin D receptor is essential to form a complex with a receptor auxiliary factor required for high affinity binding to the vitamin D-responsive element. Mol Endocrinol 8:159–172. Rao YP, Vlahcevic ZR, Stravitz RT, Mallonee DH, Mullick J, Avadhani NG, Hylemon PB 1999 Down-regulation of the rat hepatic sterol 27-hydroxylase gene by bile acids in transfected primary hepatocytes: possible role of hepatic nuclear factor 1α. J Steroid Biochem Mol Biol 70:1–14. Cereghini S 1996 Liver-enriched transcription factors and hepatocyte differentiation. FASEB J 10:267–282. Bolt MJG, Sitrin MD, Favus MJ, Rosenberg IH 1981 Hepatic vitamin D 25-hydroxylase: inhibition by bile duct ligation or bile salts. Hepatology 1:436–440. Plourde V, Gascon-Barré M, Willems B, Huet PM 1988 Severe cholestasis leads to vitamin D depletion without perturbing its C-25 hydroxylation in the dog. Hepatology 8:1577–1585. Princen HMG, Meijer P, Wolthers BG, Vonk RJ, Kuipers F 1991 Cyclosporine A blocks bile acid synthesis in cultured hepatocytes by specific inhibition of chenodeoxycholic acid synthesis. Biochem J 275:501–505. Dahlback-Sjoberg H, Bjorkhem I, Princen HMG 1993 Selective inhibition of mitochondrial 27-hydroxylation of bile acid intermediates and 25-hydroxylation of vitamin D3 by cyclosporin A. Biochem J 2933:203–206. Baran DT, Milne ML 1983 1,25-Dihydroxyvitamin D-induced inhibition of 3H-25 hydroxyvitamin D production by the rachitic rat liver in vitro. Calcif Tissue Int 35:461–464.
66 136. Bell NH, Shaw S, Turner RT 1984 Evidence that 1,25-dihydroxyvitamin D3 inhibits the hepatic production of 25-hydroxyvitamin D in man. J Clin Invest 74: 1540–1544. 137. Clements MR, Davies M, Hayes ME, Hickey CD, Lumb GA, Mawer EB, Adams PH 1992 The role of 1,25-dihydroxyvitamin D in the mechanism of acquired vitamin D deficiency. Clin Endocrinol 37:17–27. 138. Haddad P, Gascon-Barré M, Brault G, Plourde V 1986 Influence of calcium or 1,25-dihydroxyvitamin D3 supplementation on the hepatic microsomal and in vivo metabolism of vitamin D3 in vitamin D-depleted rats. J Clin Invest 78:1529–1537. 139. Halloran BP, Bikle DD, Levens MJ, Castro ME, Globus RK, Holton E 1986 Chronic 1,25-dihydroxyvitamin D3 administration in the rat reduces the serum concentration of 25-hydroxyvitamin D by increasing metabolic clearance rate. J Clin Invest 78:622–628. 140. Gascon-Barré M, Demers C, Mirshahi A, Néron S, Zalzal S, Nanci A 2003 The normal liver harbors the vitamin D nuclear receptor in nonparenchymal and biliary epithelial cells. Hepatology 37:1034–1042. 141. Gascon-Barré M, Demers C, Ghrab O, Theodoropoulos C, Lapointe R, Jones G, Valiquette L, Ménard D 2001 Expression of CYP27A, a gene encoding a vitamin D-25 hydroxylase in human liver and kidney. Clin Endocrinol 54:107–115. 142. Cali JJ, Hsieh CL, Francke U, Russell DW 1991 Mutations in the bile acid biosynthetic enzyme sterol 27-hydroxylase underlie cerebrotendinous xanthomatosis. J Biol Chem 266:7779–7783. 143. Leitersdorf E, Reshef A, Meiner V, Levitzki R, Schwartz SP, Dann EJ, Berkman N, Cali JJ, Klapholz L, Berginer VM 1993 Frameshift and splice-junction mutations in the sterol 27-hydroxylase gene cause cerebrotendinous xanthomatosis in Jews of Moroccan origin. J Clin Invest 91:2488–2496. 144. Berginer VM, Shany S, Alkalay D, Berginer J, Dekel S, Salen G, Tint GS, Gazit D 1993 Osteoporosis and increased bone fractures in cerebrotendinous xanthomatosis. Metab Clin Exp 42:69–74. 145. Leitersdorf E, Safadi R, Meiner V, Reshef A, Bjorkhem I, Friedlander Y, Morkos S, Berginer VM 1994 Cerebrotendinous xanthomatosis in the Israeli Druze: molecular genetics and phenotypic characteristics. Am J Hum Genet 55:907–915. 146. Berginer VM, Shany S, Alkalay D, Berginer J, Dekel S, Salen G, Tint GS, Gazit D 1993 Osteoporosis and increased bone fractures in cerebrotendinous xanthomatosis. Metabolism 42:69–74. 147. Kuriyama M, Fujiyama J, Kubota R, Nakagawa M, Osame M 1993 Osteoporosis and increased bone fractures in cerebrotendinous xanthomatosis [Letter to editor]. Metab Clin Exp 42:1497. 148. Sawada N, Sakaki T, Kitanaka S, Kato S, Inouye K 2001 Structure-function analysis of CYP27B1 and CYP27A1. Studies on mutants from patients with vitamin D–dependent rickets type I (VDDR-I) and cerebrotendinous xanthomatosis (CTX). Eur J Biochem 268:6607–6615. 149. Guo YD, Strugnell S, Jones G 1991 Identification of a human liver mitochondrial cytochrome P-450 cDNA corresponding to the vitamin D3-25-hydroxylase. J Bone Miner Res 6:S120. 150. Strugnell S, Byford V, Makin HLJ, Moriarty RM, Gilardi R, LeVan LW, Knutson JC, Bishop CW, Jones G 1995 1α,24(S)dihydroxyvitamin D2: A biologically active product of 1α-hydroxyvitamin D2 made in the human hepatoma Hep3B. Biochem J 310:233–241.
MARIELLE GASCON-BARRÉ
151. Rosen H, Reshef A, Maeda N, Lippoldt A, Shpizen S, Triger L, Eggertsen G, Björkhem I, Leitersdorf E 1998 Markedly reduced bile acid synthesis but maintained levels of cholesterol and vitamin D metabolites in mice with disrupted sterol 27-hydroxylase gene. J Biol Chem 273:14805–14812. 152. Honda A, Salen G, Matsuzaki Y, Batta AK, Xu G, Leitersdorf E, Tint GS, Erickson SK, Tanaka N, Shefer S 2001 Side chain hydroxylations in bile acid biosynthesis catalyzed by CYP3A are markedly up-regulated in CYP27−/− mice but not in cerebrotendinous xanthomatosis. J Biol Chem 276: 34579–34585. 153. Björkhem I, Araya Z, Rudling M, Angelin B, Einarsson C, Wikvall K 2002 Differences in the regulation of the classical and the alternative pathway for bile acid synthesis in human liver. No coordinate regulation of CYP7A1 and CYP27A1. J Biol Chem 277:26804–26807. 154. Goodwin B, Gauthier KC, Umetani M, Watson MA, Lochansky MI, Collins JL, Leitersdorf E, Mangelsdorf DJ, Kliewer SA, Repa JJ 2003 Identification of bile acid precursors as endogenous ligands for the nuclear xenobiotic pregnane X receptor. Proc Natl Acad Sci USA 100: 223–228. 155. Martial J, Plourde V, Gascon-Barré M 1985 Sequestration of vitamin D3 by the fetal and neonatal rat liver. Biol Neonate 48:21–28. 156. Plourde V, Haddad P, Gascon-Barré M 1985 Microsomal C-25 hydroxylation of [3H]-vitamin D3 by the fetal and neonatal rat liver. Pediatr Res 19:1206–1209. 157. Thierry-Palmer M, Cullins S, Rashada S, Gray TK, Free A 1986 Development of vitamin D3 25-hydroxylase activity in rat liver microsomes. Arch Biochem Biophys 250:120–127. 158. Miller ML, Ghazarian JG 1980 Studies on vitamin D3 metabolism. Discrete liver cytosolic binding proteins for vitamin D3 and 25-hydroxyvitamin D3. Biochem Biophys Res Commun 96:1619–1625. 159. Tsankova V, Visentin M, Cantoni L, Carelli M, Tacconi MT 1996 Peripheral benzodiazepine receptor ligands in rat liver mitochondria: Effect on 27-hydroxylation of cholesterol. Eur J Pharmacol 299:197–203. 160. Delvin EE, Glorieux FH, Salle BL, David L, Varenne JP 1982 Control of vitamin D metabolism in preterm infants: feto-maternal relationships. Arch Dis Child 57:754–757. 161. Hillman LS, Hoff N, Salmons S, Martin L, McAlister WH, Haddad J 1985 Mineral homeostasis in very premature infants: Serial evaluation of serum 25-hydroxyvitamin D, serum minerals, and bone mineralization. J Pediatr 106:970–980. 162. Salle BL, Glorieux FH, Delvin EE, David LS, Meunier G 1983 Vitamin D metabolism in preterm infants. Serial serum calcitriol values during the first four days of life. Acta Paediatr Scand 72:203–206. 163. Addya A, Zheng YM, Shayiq RM, Fan J, Avadhani NG 1991 Characterization of a female-specific hepatic mitochondrial cytochrome P-450 whose steady-state level is modulated by testosterone. Biochemistry 30:8323–8330. 164. Bühler R, Lindros KO, Nordling A, Johansson I, IngelmanSundberg M 1992 Zonation of cytochrome P450 isozyme expression and induction in rat liver. Eur J Biochem 204: 407–412. 165. Hansson M, Ellis E, Hunt MC, Schmitz G, Babiker A 2003 Marked induction of sterol 27-hydroxylase activity and mRNA levels during differentiation of human cultured monocytes into macrophages. Biochim Biophys Acta 1593:283–289. 166. Jabara S, Christenson LK, Wang CY, McAllister JM, Javitt NB, Dunaif A, Strauss JF 2003 Stromal cells of the
CHAPTER 4 The Vitamin D 25-Hydroxylase
167. 168.
169.
170.
171.
human postmenopausal ovary display a distinctive biochemical and molecular phenotype. J Clin Endocrinol Metab 88:484–492. Javitt NB 2002 25R,26-hydroxycholesterol revisited: Synthesis, metabolism, and biologic roles. J Lipid Res 43: 665–670. Lehmann B, Tiebel O, Meurer M 1999 Expression of vitamin D3 25-hydroxylase (CYP27) mRNA after induction by vitamin D3 or UVB radiation in keratinocytes of human skin equivalents. A preliminary study. Arch Dermatol Res 291:507–510. Schuessler M, Astecker N, Herzig G, Vorisek G, Schuster I 2001 Skin is an autonomous organ in synthesis, two-step activation and degradation of vitamin D3: CYP27 in epidermis completes the set of essential vitamin D3-hydroxylases. Steroids 66:399–408. Correa P, Segersten U, Hellman P, Åkerström WG 2002 Increased 25-hydroxyvitamin D3 1α-hydroxylase and reduced 25-hydroxyvitamin D3 24-hydroxylase expression in parathyroid tumors—New prospects for treatment of hyperparathyroidism with vitamin D. J Clin Endocrinol Metab 87:5826–5829. Ichikawa F, Sato K, Nanjo M, Nishii Y, Shinki T, Takahashi N, Suda T 1995 Mouse primary osteoblasts express vitamin D3 25-hydroxylase mRNA and convert 1α-hydroxyvitamin D3 into 1α,25-dihydroxyvitamin D3. Bone 16:129–135.
67 172. Bjorkhem I, Andersson O, Diczfalusy U, Sevastik B, Xiu RJ, Duan C, Lund E 1994 Atherosclerosis and sterol 27-hydroxylase: Evidence for a role of this enzyme in elimination of cholesterol from human macrophages. Proc Natl Acad Sci USA 91:8592–8596. 173. Reiss AB, Martin KO, Javitt NB, Martin DW, Grossi EA, Galloway AC 1994 Sterol 27-hydroxylase: high levels of activity in vascular endothelium. J Lipid Res 35:1026–1030. 174. Delvin EE, Arabian A, Glorieux FH 1978 Kinetics of liver microsomal cholecalciferol 25-hydroxylase in vitamin Ddepleted and -repleted rats. Biochem J 172:417–422. 175. Benbrahim N, Dubé C, Vallières S, Gascon-Barré M 1988 The calcium ionophore A23187 is a potent stimulator of the vitamin D3-25 hydroxylase in hepatocytes isolated from normocalcemic vitamin D-depleted rats. Biochem J 255: 91–97. 176. Holmberg I, Berlin T, Ewerth S, Bjorkhem I 1986 25 hydroxylase activity in subcellular fractions from human liver. Evidence for different rates of mitochondrial hydroxylation of vitamin D2 and D3. Scand J Clin Lab Invest 46: 785–790. 177. Cheng JB, Levine MA, Bell NH, Mangelsdorf DJ, Russel DW 2004 Genetic evidence that the human CYP2R1 enzyme is a key vitamin D 25-hydroxylase. Proc Natl Acad Sciences (USA) 101:771–775.
CHAPTER 5
The 25-Hydroxyvitamin D 1α-Hydroxylase HELEN L. HENRY
Department of Biochemistry, University of California at Riverside, Riverside, California
I. Occurrence and Characteristics of 25OHD3-1α-Hydroxylase II. Characteristics of the Proteins Involved in the 1α-Hydroxylation of 25OHD3 III. Cytochrome P4501α: Cloning and Gene Structure
IV. Regulation of kidney 1α-Hydroxylase Activity and Gene Expression V. Summary References
I. OCCURRENCE AND CHARACTERISTICS OF 25OHD3 1α-HYDROXYLASE
exclusively in the proximal tubules. In situ hybridization studies of cultures of embryonic mouse kidneys confirm the presence of 25OHD3 1α-hydroxylase (CYP1α) in tubular epithelium, but not collecting ducts or glomeruli [8].
A. The Kidney as the Source of Circulating 1α,25(OH)2D3 It is now well accepted that vitamin D is a precursor of the sterol hormone 1α,25-dihydroxyvitamin D3 [1α,25(OH)2D3]. The general pathway of production of 1α,25(OH)2D3 is shown in Fig. 1. It has been appreciated for some time [1,2] that the kidney is the major site of production of circulating 1α,25(OH)2D3, although as described later and discussed more thoroughly elsewhere in this volume, many other tissues and cell types have been shown to produce 1α,25(OH)2D3 from 25-hydroxyvitamin D3 (25OHD3). Within the kidney, it was established early on by microdissection studies that in the fetal rabbit [3] and in the vitamin D–deficient rat [4] and chick [5], the proximal tubules are the region of the most robust activity of the 1α-hydroxylase. With the cloning of the cDNA for the cytochrome P450 component of the 1α-hydroxylase (see Section III) has come the ability to measure its mRNA and protein levels along the nephron. Since these determinations are more sensitive than the measurement of enzyme activity, localization studies can now be carried out under conditions of vitamin D sufficiency and normal mineral status. Thus in vitamin D–sufficient mice and humans, mRNA and/or protein has been identified by in situ hybridization or immunohistochemical staining in the more distal portions of the nephron along with relatively low expression in the proximal tubules [6,7]. These observations suggest that, while the 1α-hydroxylase occurs throughout the nephron, its regulation varies such that the effects of vitamin D status and abnormal phosphorus metabolism (see Section IV) occur primarily if not VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX
B. Extrarenal Production of 1α,25(OH)2D3 Along with the demonstration that the kidney contained the enzymatic capability to produce 1α,25(OH)2D3 other work was suggesting that this metabolic step was largely confined to this organ. Physiological observations such as studies in nephrectomized animals [9] and clinical experience with patients suffering from chronic renal failure [10,11] strongly supported this concept. Thus, the kidney was the focus of investigation of the 1α-hydroxylase for the decade following its identification as the site of 1,25(OH)2D3 synthesis. In terms of the regulation of serum levels of 1α,25(OH)2D3 , the kidney remains the focus of study and most of the remainder of this chapter is devoted to the renal production of 1α,25(OH)2D3. Some extrarenal sites of 1α,25(OH)2D3 will be mentioned briefly here and discussed more thoroughly in Chapter 79. It should be noted that in general the extrarenal 1α-hydroxylation of 25OHD3 does not respond to the regulatory influences involved in calcium homeostasis that control the renal enzyme activity. In the early 1980s the first reports appeared that suggested that cells outside the kidney could produce 1α,25(OH)2D3 from 25(OH)D3. These reports described studies in calvarial cells [12,13] and were followed by those of 1α-hydroxylase and 24R-hydroxylase activity in chondrocytes [14,15]. These descriptions of 1α-hydroxylase activity in bone were followed by the demonstration that keratinocytes could also produce 1α,25(OH)2D3 [16,17]. Copyright © 2005, Elsevier, Inc. All rights reserved.
70
HELEN L. HENRY
FIGURE 1 Basic metabolic pathway for the major circulating vitamin D metabolites. The two renal hydroxylases that metabolize 25OHD3 are, in general, regulated in a reciprocal fashion to one another.
During this same time period, the hypercalcemia of sarcoidosis was traced to the production of 1α,25(OH)2D3 by pulmonary macrophages [18], and there followed many reports of the production of 1α,25(OH)2D3 by cells derived from the hematopoietic system [19–24]. These sites of the autocrine and/or paracrine production and action of 1α,25(OH)2D3 are discussed in more detail in Chapter 79. Reports that placental decidual cells [25,26] also contain 1α-hydroxylase activity were complicated by the fact the production of 1,25(OH)2D3 by trophoblastic tissues can produce 1α,25(OH)2D3 by a nonenzymatic process [27]. With the availability of the nucleotide and amino acid sequences for CYP1α, the presence of both the message and the protein in placental tissue [28,29] has now been confirmed. Other tissues in which evidence for CYP1α transcription and/or expression exists include the adrenal medulla, brain, pancreas, colon [29], parathyroid glands [30], prostate cells [31], and non-small-cell lung carcinoma [32].
C. Species Distribution The 1α-hydroxylase, not surprisingly, occurs widely among vertebrate species. None of the species examined in the original species comparison [33] have since been shown to lack the renal 1α-hydroxylase.
Since there are no reports of 1α-hydroxylase activity in invertebrates it can be tentatively concluded that the ability to produce 1α,25(OH)2D3 arose early in vertebrate evolution. There has been a special interest in the occurrence of vitamin D and its metabolites in fish, due in part to the historical importance of cod liver oil as a source of vitamin D and in part to the question of how fish produce vitamin D under conditions of limited ultraviolet light. Plasma levels of vitamin D metabolites have been measured in several species of fish [34–36]. Fish liver and kidney homogenates have been reported to convert 25OHD3 to 1α,25(OH)2D3 [33,36,37]. Both phytoplankton and zooplankton were reported to have abundant quantities of provitamin D and vitamin D and were suggested to be the dietary source of vitamin D for fish [38], although these authors have concluded that vitamin D and its metabolites are unimportant to mineral metabolism in the freshwater fish tilapia [39]. Thus, although there is no question that fish contain vitamin D and its metabolites and 1α-hydroxylase activity, the physiological role of these in fish remains unclear. Another interesting comparative aspect of vitamin D metabolism is posed by the naked mole rat. This nocturnal, cave-dwelling animal with no access to ultraviolet light nevertheless has the ability to synthesize both 1α,25(OH)2D3 and receptors for the hormone [40,41]. Thus, although these animals are usually vitamin D
71
CHAPTER 5 The 25-Hydroxyvitamin D 1α-Hydroxylase
ultimately derived from NADPH, generated intramitochondrially from tricarboxylic acid cycle intermediates. In vitro assays of 1α-hydroxylase activity in isolated mitochondria have used either malate [42] or other citric acid cycle intermediates [43] as a source of reducing equivalents. The current model for mitochondrial steroid hydroxylases involves the sequential transfer of electrons from NADPH to NADPH-ferredoxin reductase, a 50-kDa protein that, in the case of adrenal and gonadal steroid hydroxylases, is loosely associated with the inner mitochondrial membrane [44]. The electrons are then passed to ferredoxin, an 11-kDa mitochondrial matrix protein that shuttles between the NADPH-ferredoxin reductase and the terminal component of the hydroxylase machinery, cytochrome P450 (so named because of its distinct spectral characteristics when carbon monoxide is bound to it). As discussed below, each of these three proteins has been the subject of study in the context of the 1α-hydroxylation of 25OHD3.
deficient, they maintain the enzymatic machinery necessary to produce the active hormonal form and the receptor necessary to respond to it.
II. CHARACTERISTICS OF THE PROTEINS INVOLVED IN THE 1α-HYDROXYLATION OF 25OHD3 A. Mixed-Function Oxidases Enzymes that hydroxylate endogenous steroids in a stereospecific manner are called mixed-function oxidases because they reduce one atom of molecular oxygen to water and one atom to the hydroxyl group to be stereospecifically incorporated into the steroid (Fig. 2). Such reactions are important in the pathways of the production of androgens, estrogens, progestins, mineralocorticoids, and glucocorticoids. Indeed, the first step in the production of these steroid hormones, the cleavage of six carbons from the side chain of cholesterol, results from two successive hydroxylations at C-20 and C-22. Some mixed-function oxidases are microsomal (e.g., the 17α-hydroxylase/C-17-C-20 lyase and the 21-hydroxylase), and some (e.g., cholesterol side-chain cleavage and the 11β-hydroxylase) are located in the inner mitochondrial membrane. The 25OHD3 1α-hydroxylase belongs to the latter group of steroid hydroxylases. For mitochondrial mixed-function oxidases, the electrons for the reduction of molecular oxygen are
1. CYTOCHROME P450
The evidence that the 1α-hydroxylase is a member of the class of cytochrome P450-dependent steroid hydroxylases was obtained in the mid-1970s and consists of the following: (1) Solubilized preparations of chick kidney mitochondria were shown to contain cytochrome P450 and to carry out 1α-hydroxylation of 25OHD3 [45]; (2) more directly, it was shown that inhibition of 1α-hydroxylase activity in isolated chick kidney mitochondria by carbon monoxide was reversed specifically by light of 450 nm [42].
R–OH + H2O NADPH
FPox
NHI–Fe
2+
P450–Fe
3+
R Reductase
NADP
FPred
Ferredoxin
NHI–Fe 3+
Cytochrome P450
P450–Fe 2+
P450–Fe 2+−O2
O2
CO P450–Fe 2+−CO
FIGURE 2 Electron transport chain for mitochondrial steroid hydroxylases. The general class of mitochondrial mixed-function oxidases, of which the 25OHD3 1α-hydroxylase is a member, consists of three components in or associated with the inner mitochondrial membrane. Cytochrome P450 reduces molecular oxygen to water and to the hydroxyl group to be incorporated into the steroid; it confers the stereospecificity of the hydroxylation reaction. In its reduced form, cytochrome P450 binds carbon monoxide and in this form absorbs light of 450 nm. The flavoprotein (FP) is NADPH-ferredoxin reductase, and the nonheme iron (NHI) protein is ferredoxin. Ox, oxidized; red, reduced.
72
HELEN L. HENRY
There have been sporadic reports of the partial purification of CYP1α. An apparently electrophoretically pure bovine kidney cytochrome P450, with a molecular weight of 57,000, retained putative 1α-hydroxylase activity [46], but there was no confirmation of the identity of the product of 25OHD3 hydroxylation. In the early 1990s, there were two reports purporting to have purified the 1α-hydroxylase cytochrome P450 from mitochondria from either vitamin D–replete or vitamin D–deficient chick kidneys [47,48]. The purified proteins had molecular weights of 59,000 and 57,000, but amino acid compositions and N-terminal sequences, when compared with the four mammalian CYP1α sequences that are now known, make it unlikely that either of these proteins was indeed the 1α-hydroxylase cytochrome P450. More recently, a 54-kDa protein with 1α-hydroxylase activity and appropriate spectral characteristics was isolated from chick kidney [49] but no sequence information about this protein was given. In another study with pig kidney mitochondria, 1α-hydroxylase activity was partially purified [50] and separated from steroid side chain–hydroxylating enzymes [51], but in neither report were any physical characteristics of the protein given. In summary, although the primary sequence of CYP1α has been deduced from cloned cDNA from several mammalian species, purified preparations of the protein itself are still not routinely obtained. The primary sequence of the cytochrome P450 that catalyzes the 1α-hydroxylation of 25OHD3, as deduced from its cDNA sequence, reveals that it is structurally related to the mitochondrial sterol side chain hydroxylases and therefore it has been given the systematic name of CYP27B1 (see [52] for a listing of human cytochrome P450s). In this chapter, the protein and its nucleic acid sequences will be referred to by its familiar and functionally more descriptive name of CYP1α. 2.
FERREDOXIN
a. Characteristics of Mitochondrial Ferredoxin Vertebrate mitochondrial ferredoxins are nonheme iron– sulfur proteins of 114–128 amino acids, which on electrophoresis generally migrate at 11–12 kDa. cDNA analysis indicates a mitochondrial leader sequence of 58–62 amino acids, depending on the species [53]. The ferredoxin involved in the mitochondrial hydroxylation of endogenous steroids is a matrix protein that shuttles between NADPH-ferredoxin reductase and cytochrome P450 to deliver, one at a time, the two electrons required for the reduction of molecular oxygen. Ferredoxins from mammalian adrenal glands (adrenodoxins) have been studied most extensively [54,55], although the protein has been isolated from several other steroidogenic tissues [56].
As is the case for ferredoxins in other steroidogenic tissues, chick kidney ferredoxin is a mitochondrial matrix protein [57]. This submitochondrial localization is consistent with its function as an electron-carrying shuttle between the NADPH-ferredoxin reductase and CYP1α. Reconstitution assays in a number of laboratories have demonstrated the absolute requirement of 1α-hydroxylase activity for ferredoxin [58–59]. Chick kidney ferredoxin has been overexpressed in Escherichia coli and shown to be approximately twice as active as bovine adrenal ferredoxin in supporting 1α-hydroxylation in a reconstituted assay system [60], further illustrating the similarities that the 1α-hydroxylase shares with other mitochondrial mixed-function oxidases that hydroxylate endogenous steroids. cDNA cloning has shown that the pig kidney and adrenal ferredoxin share identical amino acid sequences [61]; similarly, in the chick kidney and testis ferredoxins share the same primary structure [62,63]. These observations suggest that a single gene encodes ferredoxin for the various tissues in which it is expressed. Southern blot analysis of bovine [64] and chick genomic DNA [62] supports the existence of a single ferredoxin gene in these species. In the human, two ferredoxin genes have been identified, but their coding regions for the mature protein are identical [65]. Taken together the available evidence indicates that a single ferredoxin protein serves as the electron shuttle for all mitochondrial cytochrome P450s that hydroxylate endogenous steroids in the species that have been examined thus far. b. Regulation of Mitochondrial Ferredoxin Levels Since the rate of hydroxylation of a steroid by cytochrome P450 will be governed to some extent by the supply of electrons to reduce molecular oxygen, a regulatory component at this step is possible, although such a process would clearly lack some specificity in cells that can carry out more than one hydroxylation. Mammalian ferredoxins and their mRNAs have been shown to be developmentally and hormonally regulated [66–69]. For example, hypophysectomy decreases and replacement treatment with adrenocorticotropin (ACTH) increases adrenal ferredoxin mRNA levels in the rat [67,70]. In a study of several cell types, ferredoxin mRNA was stimulated by appropriate pituitary tropic hormones, possibly through a cAMP-dependent mechanism, and ferredoxin mRNA levels decreased in the testis and adrenal during fetal development [66]. Chick kidney mitochondrial ferredoxin mRNA is up-regulated approximately 40% in the vitamin D– deficient state [62]. Although kidney mitochondrial ferredoxin is also presumably required for the 24Rhydroxylation of vitamin D, which is, in general, regulated reciprocally to the 1α-hydroxylase, 1α-hydroxylase activity in vitamin D deficiency is much higher than is
73
CHAPTER 5 The 25-Hydroxyvitamin D 1α-Hydroxylase
24R-hydroxylase activity in the vitamin D–replete chick. Therefore, increased mRNA levels of ferredoxin mRNA in vitamin D deficiency are not likely to be involved in the reciprocal regulation of these two enzymes, but rather are related to the increased demand of maximal 1α-hydroxylase activity brought about by vitamin D deficiency. Ferredoxin phosphorylation has been suggested as a posttranslational mechanism for the regulation of its steroid hydroxylation activity in the adrenal gland and kidney. Although adrenal ferredoxin can be phosphorylated in vivo by cAMP-dependent protein kinase, no convincing evidence of phosphorylated adrenal ferredoxin in vivo has appeared, nor has a role been established for such a phosphoferredoxin in the physiological regulation of adrenal or gonadal steroidogenesis. On the other hand, ferredoxin from primary cultures of chick kidney cells has been shown to exist as a phosphoprotein and to be rapidly dephosphorylated by treatment of the cells with the phorbol ester 12-O-tetradecanoylphorbol-13-acetate (TPA) [57]. This dephosphorylation is accompanied by decreased 1α-hydroxylase activity, supporting for a role of ferredoxin phosphorylation and dephosphorylation in the regulation of 1-hydroxylation of 25OHD3. However, the molecular mechanisms involved in such a process remain to be determined. 3. NADPH-FERREDOXIN REDUCTASE
The flavoprotein NADPH-ferredoxin reductase, which accepts electrons from intramitochondrially generated NADPH, has been purified from pig kidney mitochondria [71] and does not appear to differ significantly in size (~52 kDa), enzymatic characteristics, or immunologically from the reductases from other mammalian tissues that carry out hydroxylations in the production of endogenous steroids such as the corpus luteum and the adrenal gland [72,73]. cDNAs for bovine and human adrenal NADPH-ferredoxin reductase have been cloned and shown to code for a mitochondrial targeting sequence in addition to the mature protein [74].
B. Accessory Proteins In addition to the three components of the enzymatic machinery that hydroxylates 25OHD3 at the 1α position, certain other cellular proteins have been implicated in the process of moving the substrate from the circulation to the site of metabolism in the inner mitochondrial membrane. Although the steroid hormones and their precursors are lipophilic and might be thought to simply diffuse through the cellular membranes
to the site of their metabolism, specific proteins to direct their intracellular movements are quite likely. One group of candidates for this type of protein, currently termed the intracellular vitamin D binding proteins, or IDBPs, was originally found in vitamin D(and other steroid hormone)-resistant New World primates [75]. The constitutive expression of binding proteins for 25OHD3 in these animals was initially thought to be the causative factor in their vitamin D resistance and/or the response to it, leading to high circulating levels of the hormone. More recently, these proteins have been cloned and sequenced and found to be homologous to human heat shock proteins of the hsp-70 family [76,77], and they are now thought to play an active role in the delivery of the substrate 25OHD3 to the 1α-hydroxylase, rather than acting to sequester the substrate as originally thought. In addition, these proteins have been shown to bind DNA and may play a function in the regulation of 24R-hydroxylase gene expression (see Chapter 21). A somewhat more definitive role in the uptake of 25OHD3 from the circulation and delivery to the 1α-hydroxylase has been proposed for the endocytotic receptor megalin, a member of the low-density lipoprotein receptor family [78]. These authors initially observed abnormal excretion of 25OHD3 in megalin knockout mice and showed that complexes of the vitamin D binding protein, DBP, and 25OHD3 are filtered by the glomerulus and reabsorbed by megalin in the proximal tubule of the kidney. In megalin knockout mice, the mRNA levels for both the 1α-hydroxylase and the 24R-hydroxylase were altered in a way consistent with severe vitamin D deficiency (that is, elevated 1α-hydroxylase and decreased 24R-hydroxylase activity; see Section IV), confirming the inability of the kidney to produce 1α,25(OH)2D3 to regulate these two enzymes [79,80]. See Chapter 10 for a more thorough discussion of the role of megalin in vitamin D metabolism.
III. CYTOCHROME P4501α: CLONING AND GENE STRUCTURE Reports of the cloning of the cDNA encoding the cytochrome P450 catalyzing the 1α-hydroxylation of 25(OH)D3 appeared from a number of laboratories in 1997. The mouse cDNA was cloned from kidneys of mice lacking the VDR, which resulted in very high expression levels of 1α-hydroxylase mRNA [81]. At approximately the same time, the rat kidney [82,83], human keratinocyte [84], and human kidney [85] sequences for the 1α-hydroxylase cytochrome P450 were published. These were followed by the porcine kidney
74 coding sequence [86]. For the mouse, rat, and human sequences, the identification of the cloned cDNA as that encoding the 1α-hydroxylase cytochrome P450 was confirmed by transfection into cells capable of supporting steroid hydroxylation (cell lines derived from kidney or testis having endogenously expressed ferredoxin and ferredoxin reductase) and demonstration that the product formed from added 25(OH)D3 was 1α,25(OH)2D3. The porcine sequence was identified on the basis of its 88% homology with the other three mammalian sequences. The sequences of cDNAs encoding the mammalian CYP1αs predict proteins with 501 to 508 amino acids. The structures of the human [85,87,88] and mouse [89] CYP27B1 genes have been determined. The mammalian CYP27B1 gene is divided into nine exons, seven
HELEN L. HENRY
of roughly equal size (170–200 bp encoding 57–67 amino acids each) and two smaller ones (26 and 37 amino acids). The putative sterol binding region (Fig. 3) is encoded by exon VI and the heme-binding region, centering around the cysteine residue that is invariant in all mitochondrial P450s that hydroxylate endogenous steroids, is encoded by exon VIII. The intron size ranges from approximately 80 to 660 bp, so altogether, the CYP27B1 gene is small in size relative to some other members of the mitochondrial cytochrome P450s, although the overall intron/exon organization is highly conserved among the genes coding for these proteins. The gene for CYP1α has been mapped to human chromosome 12 [83,87] and linkage analysis shows that it lies
FIGURE 3 Deduced amino acid sequences of CYP1α from several mammalian species. The human sequence is given in its entirety and in the other sequences, identical amino acids are indicated by dashes. Box A is the putative sterol binding site and Box B is the heme-binding region, containing the cysteine that is invariant in all cytochrome P450s.
75
CHAPTER 5 The 25-Hydroxyvitamin D 1α-Hydroxylase
very close to the defect for human vitamin D–dependent rickets (VDDR type I) (see Chapter 71). In the mouse, the CYP1α gene is on chromosome 10, the region equivalent to the localization of the human gene [89].
IV. REGULATION OF KIDNEY 1α-HYDROXYLASE ACTIVITY AND GENE EXPRESSION In numerous whole-animal and cell-culture model systems it is now well established that the two most important regulators of the activity of the 1α-hydroxylase are 1α,25(OH)2D3 itself and parathyroid hormone. In addition, there has been evidence for the involvement of calcitonin, dietary mineral levels, and hormones of other endocrine systems in the regulation of renal 25OHD3 metabolism.
A. 1α,25(OH)2D3 It was recognized at the time of the localization of the production of 1,25(OH)2D3 in the kidney that, in both avian and mammalian species, vitamin D–deficient animals have higher 1α-hydroxylase activity than do vitamin D–replete animals [1,90–92]. The ability of 1,25(OH)2D3 to replicate this effect on 1α-hydroxylase has been demonstrated in chick [93,94] and mammalian [95] cells in vitro. Both in vivo and in cell culture, the activity of the 24-hydroxylase is regulated by 1α,25(OH)2D3 in a reciprocal fashion to the 1α-hydroxylase. When 1,25(OH)2D3 is removed from the cell culture medium [93] or when its administration in vivo is terminated [90], 1α-hydroxylase activity returns to the high levels characteristic of the vitamin D–deficient state, indicating the ready reversibility of the inhibitory effect of 1,25(OH)2D3 on kidney 1α-hydroxylase activity. In a wide variety of
experimental conditions the effect of 1α,25(OH)2D3 on 1α-hydroxylase activity observed in intact animals in vivo can be duplicated by experiments in primary cultures of kidney cells. From these types of experiments, however, the mechanism of the down-regulation of 1α-hydroxylase activity by 1α,25(OH)2D3 has not been elucidated. The availability of the cDNA and genomic sequences for CYP1α has opened up new avenues of exploration of this question. There is no question that steadystate CYP1α mRNA levels are down-regulated by 1α,25(OH)2D3 in the kidneys of intact animals [96,97] and that this down-regulation requires the presence of the VDR [81,96]. When steady-state CYP1α mRNA levels are measured in cultured cells, however, the results are mixed and may depend on the cell type used, as shown in Table I. In fact, of the cell types tested, only in the MCT cells did 1α,25(OH)2D3 downregulate the endogenous expression of the mRNA for the 1α-hydroxylase cytochrome P450. Of the nonresponsive cells, both the LLC-PK1 and the HKC-8 cells have been reported to express 1α-hydroxylase activity [98,99]. However, in none of the studies in Table I were data regarding the effect of 1α,25(OH)2D3 on 1α-hydroxylase enzymatic activity presented under the same conditions in which the cells were apparently unresponsive. Thus, it is not known whether the failure to see down-regulation of CYP1α mRNA by 1α,25(OH)2D3 was due to dissociation of its effects on the transcription and activity of the cytochrome P450 or whether these cells, for unknown reasons, failed to respond to 1α,25(OH)2D3 at both the RNA and protein levels. In addition to studies of the effects of 1α,25-dihydroxyvitamin D3 on the steady-state levels of CYP1α mRNA, several laboratories have prepared and studied reporter constructs using the 5′-flanking region of the gene. In one report such a construct was active in cells derived from proximal renal tubules but not in other
TABLE I Summary of Effects of 1α,25(OH)2 D3 on Steady-State CYP1α mRNA levels Whole Animals Species Rat Mouse Mouse-VDR-KO
Effect of 1α,25(OH)2D3 ↓ ↓ →
Cell Culture Reference
Cell Line
Effect of 1α,25(OH)2D3 treatment
Reference
[96,97] [96] [81,96]
MCT cells LLC-PK1 AOK-B50 HKC-8
↓ → → →
[96,151] [86] [152] [152]
Cell types referred to are MCT:SV-40-transformed mouse proximal tubule cells [151]; LLC-PK1; AOK-B50, a derivative of LLC-PK1 stably transfected with opossum PTH/PTHrP receptor [153]; and HKC-8, SV-40-transformed human kidney cell line [154].
76
HELEN L. HENRY
TABLE II
Summary of Effects of 1α,25(OH)2D3 on CYP1α Promoter Activity
Promoter (species-kb)
Cell type
Effect
Reference
m-1.7 m-1.6 h-1.1 h-0.5
AOK-B50 OK AOK-B50 MCT
→ ↓ → ↓
[155] [156] [88] [151]
Cell types referred to are the same as in Table 1, with the addition of the opossum kidney cell line, OK.
cell types [100]. A summary of the results of this type of study is shown in Table II. No definitive vitamin D regulatory element has been identified from sequence analysis of the promoter region of mammalian CYP1α genes, although functional analysis suggests that such a regulatory site may lie within 0.5 kb of the transcriptional start site [96]. As is the case with studies of the regulation by 1α,25(OH)2D3 of endogenous CYP1α mRNA levels, it is clear from Table II that studies of the regulation of promoter activity in various cell types have yielded mixed results and have not as yet revealed whether 1α,25(OH)2D3 down-regulates 1α-hydroxylase activity through a decrease in transcription of its mRNA, through an indirect transcriptional mechanism, or through posttranscriptional and/or posttranslational processes. It is possible that more than one mechanism is involved and that the model system employed will play a role in which is revealed.
B. Parathyroid Hormone In whole animal studies, parathyroidectomy diminishes and administration of PTH increases 1α-hydroxylase activity in kidney tissue, in both avian and mammalian species [101–105]. PTH has also been shown to stimulate 1α-hydroxylase activity in cultured avian [106–108] and mammalian [109,110] kidney cells. The stimulatory effect of PTH on 1α-hydroxylase activity is mediated at least in part by the cAMP signaling pathway, as its effects can be mimicked by forskolin [109,111]. Although specific proteins with altered phosphorylation states in response to PTH have been identified [112], it has not yet been determined whether they play a physiological role in the regulation of the 1α-hydroxylation of 25OHD3. In addition to the cAMP signaling pathway, protein kinase C is also involved in the regulation of the renal
metabolism of 25OHD3 by both the 1α-hydroxylase and the 24-hydroxylase. In chick kidney cells, the phorbol ester protein kinase C (PKC) activator TPA decreases 1α-hydroxylase activity transiently [57] and over a 4-hr incubation period [113]. The latter decrease in 1α-hydroxylase activity is associated with increased 24-hydroxylase activity. The abilities of TPA and 1,25(OH)2D3 to decrease 1α-hydroxylase activity in cultured chick kidney cells are additive, indicating that they operate through distinct mechanisms. In perfused rat kidney proximal tubules, on the other hand, PKC activation by TPA resulted in a transient increase in 1α-hydroxylase activity and a sustained increase when the calcium ionophore A23187 was added with TPA [114]. PTH has been reported to stimulate the PKC pathway in rat renal cells [115]. The relative physiological importance of the cAMP and PKC signaling pathways in the control of 1α-hydroxylase activity has not yet been fully resolved. Furthermore, the fact that TPA activation of PKC decreases 1α-hydroxylase activity in chick kidney cells, both in the short term and over longer times, while increasing its activity in rat preparations, suggests that the involvement of PKC in the regulation of 1α-hydroxylase may be more complex than initially appreciated. The availability of reagents with which to study the regulation of steady-state CYP1α mRNA levels and promoter activity has led to the demonstration that PTH exerts at least part of its effects on 1α-hydroxylase activity at the transcriptional level. In addition, the role of cyclic AMP in this process has been confirmed in these studies, which are summarized in Tables III and IV. In every cell type studied, PTH stimulated the endogenous levels of CYP1α mRNA and, when tested, this effect was enhanced by cyclic AMP. Only in the MCT cells, however, did 1α,25(OH)2D3 attenuate the effect of PTH, similar to the results seen when the ability of 1α,25(OH)2D3 alone to down-regulate CYP1α mRNA was tested (Table I). PTH also stimulated the TABLE III Summary of the Regulation of Steady-State CYP1α mRNA Levels by PTH, cAMP, and Forskolin Effect of Cell line LLC-PK1 AOK-B50 HKC-8 MCT
PTH
cAMP/FSK
1α,25(OH)2D3a
Reference
↑ ↑ ↑ ↑
↑ ↑ nd nd
→ → → ↓
[86] [152] [152] [96]
a The effect of 1α,25(OH) D refers to whether the steroid attenuates 2 3 (↓) or has no effect on (→) stimulation of CYP1α mRNA levels by PTH or cAMP. nd, Not determined.
77
CHAPTER 5 The 25-Hydroxyvitamin D 1α-Hydroxylase
activity of a variety of CYP1α promoter constructs and this effect was attenuated, although only modestly in two cases, by 1α,25(OH)2D3. The observation that the effect of 1α,25(OH)2D3 is not particularly robust may account for the lack of effect on endogenous levels of 1α,25(OH)2D3 mRNA. Taken together, the results very strongly indicate that PTH stimulation of 1α-hydroxylase activity in renal cells is mediated, at least in part, by a transcriptional effect on CYP1α gene expression. Furthermore, they support the concept that a balance between PTH and 1α,25(OH)2D3 determines this level of expression. That this transcription effect may not tell the whole story, however, is suggested by a report that in adult rats, PTH administration increases CYP1α mRNA levels in the kidney severalfold, but has no effect on 1α-hydroxylase activity. Clearly the regulation of the renal production of 1α,25(OH)2D3 is complex, occurs at more than one step, and is an area that deserves more detailed investigation.
C. Other Regulators of Renal 1α,25(OH)2D3 Synthesis
levels of this message and directly influences the transcriptional activity of the CYP1α gene promoter. Shinki et al. [97] showed increased steady-state CYP1α mRNA levels when normocalcemic rats were injected with calcitonin. Interestingly, this increase in CYP1α mRNA was preceded by a marked decrease in VDR mRNA, leaving open the possibility that part of the effect of calcitonin is to render the kidney 1α-hydroxylase less sensitive to feedback inhibition by 1α,25(OH)2D3. In a second report of the stimulation of steady-state levels of CYP1α mRNA by calcitonin in rat kidney [96] it was also observed that simultaneous administration of 1α,25(OH)2D3 abrogated this effect in intact rats but not in the VDR-KO mouse model that was also used. A direct stimulatory effect of calcitonin on CYP1α mRNA levels in the MCT cells, a cell line derived from mouse proximal tubules [96], and in LLC-PK1 cells [122] has been reported. Finally, the human CYP1α promoter fused to CAT reporter is up-regulated by calcitonin in MCT cells, further underscoring a role for this calcium-regulating hormone in the renal metabolism of 25OHD3. 2. DIETARY PHOSPHORUS
1. CALCITONIN
The role of calcitonin in the regulation of renal vitamin D metabolism has a long and somewhat controversial history. Although there is a substantial number of reports suggesting that calcitonin plays a role in the regulation of vitamin D metabolism of the intact animal [97,116–118], evidence that these effects are exerted directly on the renal cell that carries out the 1α-hydroxylation of 25OHD3 has been not been consistently obtained. Thus, when added to isolated renal cell preparations in vitro calcitonin has been reported to stimulate [119], inhibit [120], or have no effect [121] on 1α-hydroxylase activity. More recent studies measuring CYP1α mRNA levels, however, strongly suggest that calcitonin does stimulate the steady-state
In mammalian animal models, low dietary phosphorus leads to increased serum concentrations of 1α,25(OH)2D3 and increased 1α-hydroxylase activity measured in renal preparations [123–126]. In humans as well, dietary phosphorus is inversely correlated with serum levels of 1α,25(OH)2D3 [127–130]. It has been demonstrated that dietary phosphorus restriction in mice leads to an increase of renal proximal tubule production of 1α,25(OH)2D3 as well as increases in both the mRNA and protein levels of CYP1α [7], providing strong evidence that the effect of phosphorus is exerted, at least in part, on the level of transcription of CYP1α. In addition, these authors also provided evidence that reduced catabolism (due to reduced 24Rhydroxylase activity) of 1α,25(OH)2D3 may also play a role in the increased serum levels of 1α,25(OH)2D3 seen when dietary phosphorus is low. 3. OTHER STEROID HORMONES
TABLE IV Promoter (species-kb) m-1.7 m-1.6 h-1.1 h-0.3 h-0.5
Summary of Effects of PTH on CYP1α Promoter Activity Cell type
AOK-B50 OK AOK-B50 AOK-B50 MCT
PTH/FSK/ cAMP 1α,25(OH)2D3 Reference ↑ ↑ ↑ ↑ ↑
↓ (marginal) ↓ (modest) ↓ nd nd
[155] [156] [88] [157] [151]
Steroid hormones of other endocrine systems have been tested for their effects, or lack thereof, on 1,25(OH)2D3 production in vivo or in cell culture. These include estrogen and the synthetic glucocorticoid dexamethasone. The stimulatory effects of estrogen on 1α-hydroxylase activity measured in extracts of kidney tissue from chicks and Japanese quail [131–133] could not be confirmed as direct ones in cell culture [134]. Similarly, although the negative influence of glucocorticoids on bone health and the vitamin D endocrine system is well documented in a number
78
HELEN L. HENRY
of animal models and clinical studies (see Chapter 72), it is apparent that these effects cannot be attributed in a substantial way to direct alteration of vitamin D metabolism. The effects of glucocorticoids on plasma levels of 1α,25(OH)2D3 are variable [135–137], and dexamethasone exhibited a slight inhibitory effect on 1α-hydroxylase activity in cultured chick kidney cells [138]. A more recent study in mice showed that under conditions of low dietary calcium only, treatment of animals slightly reduced 1α-hydroxylase mRNA and activity in the kidney [139]. Thus, although interactions between other endocrine systems and the regulation of mineral metabolism by 1α,25(OH)2D3 undoubtedly play a role in calcium homeostasis and bone mineral metabolism, they do not apparently take place in a significant way at the level of 1α,25(OH)2D3 production in the kidney. 4. ANTIMYCOTICS
The antimycotics ketoconazole and miconazole inhibit 1α-hydroxylase in cell culture [140] and were subsequently shown to substantially decrease serum levels of 1,25(OH)2D3 and active calcium transport in rats [141]. Clinical studies have suggested that there are lowered serum 1,25(OH)2D3 levels in individuals being treated with these antifungal agents [142]. In one report, ketoconazole was used to ameliorate hypercalcemia resulting from acute active tuberculosis [143]. These effects of the antimycotics are similar to those known for other steroid hormones since they occur at the level of steroid hydroxylation [144,145]. Currently the primary interest in these agents in the vitamin D field is focused on their inhibition of the 24R-hydroxylase and the catabolic pathway for 1α,25(OH)2D3 that it initiates [146–148]. 5. THE KLOTHO GENE PRODUCT
A mouse homozygous mutation in a single gene, named klotho, displays multiple disorders that are similar to those seen in human aging [149]. In addition, these mice and a similar strain that is null for the klotho gene [150] have greatly elevated plasma 1α,25(OH)2D3. The levels of mRNA encoding CYP1α are also elevated in the kidneys of these mice. The regulatory responses to 1α,25(OH)2D3, PTH, and calcitonin appear to be intact in these mice, suggesting that there may be an alternate regulatory circuit involved in governing 1α,25(OH)2D3 hydroxylase activity.
V. SUMMARY The 25(OH)D3-1α-hydroxylase is a member of the family of cytochrome P450–dependent enzymes,
specifically those mitochondrial enzymes that metabolize endogenous steroids such as cholesterol side-chain cleavage systems in the adrenal cortex and gonads and the adrenal 11β-hydroxylase. The CYP1α activity is widely distributed in vertebrates. Although the kidney is still considered to be the primary site of circulating levels of 1α,25(OH)2D3, it is clearly produced in many other cells as well, mostly as part of paracrine/ autocrine systems. The past several years has seen considerable advances in the field of vitamin D metabolism with the cloning of CYP1α so that the tools are now available to study its regulation at the molecular level. The elucidation of these mechanisms and how they differ in the kidney and extrarenal sites of 1α,25(OH)2D3 production is just one of the exciting areas in which we can expect new ideas and information in the future.
References 1. Fraser DR, Kodicek E 1970 Unique biosynthesis by kidney of a biologically active vitamin D metabolite. Nature 288:764–766. 2. Norman AW 1971 Evidence for a new kidney produced hormone, 1,25-dihydroxycholecalciferol, the proposed biologically active form of vitamin D. Am J Clin Nutr 24:1346–1351. 3. Akiba T, Endou H, Koseki C, Sakai F 1980 Localization of 25-hydroxyvitamin D3-1α-hydroxylase activity in the mammalian kidney. Biochem Biophys Res Commun 94:313–318. 4. Kawashima H, Torikai S, Kurokawa K 1981 Localization of 25-hydroxyvitamin D3 1α-hydroxylase and 24-hydroxylase along the rat nephron. Proc Natl Acad Sci USA 78:1199–1203. 5. Brunette MG, Chan M, Ferriere C, Roberts KD 1978 Site of 1,25(OH)2 vitamin D3 synthesis in the kidney. Nature 276:287–289. 6. Zehnder D, Bland R, Walker EA, Bradwell AR, Howie AJ, Hewison M, Stewart PM 1999 Expression of 25-hydroxyvitamin D3-1α-hydroxylase in the human kidney. J Am Soc Nephrol 10:2465–2473. 7. Zhang MY, Wang X, Wang JT, Compagnone NA, Mellon SH, Olson JL, Tenenhouse HS, Miller WL, Portale AA 2002 Dietary phosphorus transcriptionally regulates 25-hydroxyvitamin D-1alpha-hydroxylase gene expression in the proximal renal tubule. Endocrinology 143:587–595. 8. Yamagata M, Kimoto A, Michigami T, Nakayama M, Ozono K 2001 Hydroxylases involved in vitamin D metabolism are differentially expressed in murine embryonic kidney: application of whole mount in situ hybridization. Endocrinology 142:3223–3230. 9. Hartenbower DL, Stella FJ, Norman AW, Friedler RM, Coburn JW 1977 Impaired vitamin D metabolism in acute uremia. J Lab Clin Med 90:760–766. 10. Mawer EB, Taylor CM, Backhouse J, Lumb GA, Stanbury SW 1973 Failure of formation of 1,25-dihydroxycholecalciferol in chronic renal insufficiency. Lancet 1:626–628. 11. Eisman JA, Hamstra AJ, Kream BE, DeLuca HF 1976 1,25Dihydroxyvitamin D in biological fluids: A simplified and sensitive assay. Science 193:1021–1023. 12. Turner RT, Puzas JE, Forte MD, Lester GE, Gray TK, Howard GA, Baylink, DJ 1980 In vitro synthesis of 1α,25dihydroxycholecalciferol and 24,25-dihydroxycholecalciferol
CHAPTER 5 The 25-Hydroxyvitamin D 1α-Hydroxylase
13.
14.
15.
16. 17. 18. 19.
20.
21.
22. 23.
24.
25. 26.
27.
28.
by isolated calvarial cells. Proc Natl Acad Sci USA 77: 5720–5724. Puzas JE, Turner RT, Howard GA, Brand JS, Baylink DJ 1987 Synthesis of 1,25-dihydroxycholecalciferol and 24,25-dihydroxycholecalciferol by calvarial cells: Characterization of the enzyme systems. Biochem J 245:333–338. Pedrozo HA, Boyan BD, Mazock J, Dean DD, Gomez R, Schwartz Z 1999 TGFβ1 regulates 25-hydroxyvitamin D3, 1α- and 24-hydroxylase activity in cultured growth plate chondrocytes in a maturation-dependent manner. Calcif Tissue Int 64:50–56. Schwartz Z, Brooks B, Swain L, Del Toro F, Norman AW, Boyan BD 1992 Production of 1,25-dihydroxyvitamin D3 and 24,25-dihydroxyvitamin D3 by growth zone and resting zone chondrocytes is dependent on cell maturation and is regulated by hormones and growth factors. Endocrinology 130:2495–2504. Bikle DD, Nemanic MK, Gee E, Elias P 1986 1,25Dihydroxyvitamin D3 production by human keratinocytes. J Clin Invest 78:557–566. Bikle DD, Nemanic MK, Whitney JO, Elias PW 1986 Neonatal human foreskin keratinocytes produce 1,25dihydroxyvitamin D3. Biochem 25:1545–1548. Adams JS, Sharma OP, Gacad MA, Singer FR 1983 Metabolism of 25-hydroxyvitamin D3 by cultured pulmonary alveolar macrophages in sarcoidosis. J Clin Invest 72:1856–1860. Reichel H, Koeffler HP, Tobler A, Norman AW 1987 1α,25Dihydroxyvitamin D3 inhibits γ-interferon synthesis by normal human peripheral blood lymphocytes. Proc Natl Acad Sci USA 84:3385–3389. Reichel H, Bishop JE, Koeffler HP, Norman AW 1987 25-Dihydroxyvitamin D3 metabolism by lipopolysaccharidestimulated normal human macrophages. J Clin Endocr Metab 64:1–9. Adams JS, Singer FR, Gacad MA, Sharma OP, Hayes MJ, Vouros P, Holick MF 1985 Isolation and structural identification of 1,25-dihydroxyvitamin D3 produced by cultured alveolar macrophages in sarcoidosis. J Clin Endocr Metab 60:960–966. Mawer EB, Hayes ME 1992 Renal and extra-renal synthesis of 1,25-dihydroxy vitamin D. Prog Endocrinol 87:382–386. Cadranel J, Garabedian M, Milleron B, Guillozo H, Akoun G, Hance AJ 1990 1,25(OH)2D3 production by T lymphocytes and alveolar macrophages recovered by lavage from normocalcemic patients with tuberculosis. J Clin Invest 85: 1588–1593. Pryke AM, Duggan C, White CP, Posen S, Mason RS 1990 Tumor necrosis factor-α induces vitamin D-1-hydroxylase activity in normal human alveolar macrophages. J Cell Physiol 142:652–656. Delvin EE, Arabian A 1987 Kinetics and regulation of 25-hydroxycholecalciferol 1-α−hydroxylase from cells isolated from human term decidua. Eur J Biochem 164:659–662. Zerwekh JE, Breslau NA 1986 Human placental production of 1,25-dihydroxyvitamin D3: Biochemical characterization and production in normal subjects and patients with pseudohyperparathyroidism. J Clin Endocr Metab 62:192. Hollis BW, Iskersky VN, Chang MK 1989 In vitro metabolism of 25-hydroxyvitamin D3 by human trophoblastic homogenates, mitochondria, and microsomes: Lack of evidence for the presence of 25-hydroxyvitamin D3-1α-hydroxylase and 24R-hydroxylase. Endocrinology 125:1224–1230. Diaz L, Sanchez I, Avila E, Halhali A, Vilchis F, Larrea F 2000 Identification of a 25-hydroxyvitamin D3 1α-hydroxylase gene transcription product in cultures of human syncytiotrophoblast cells [In Process Citation]. J Clin Endocrinol Metab 85: 2543–2549.
79 29. Zehnder D, Bland R, Williams MC, McNinch RW, Howie AJ, Stewart PM, Hewison M 2001 Extrarenal expression of 25-hydroxyvitamin D3-1α−hydroxylase. J Clin End Metab 86: 888–894. 30. Segersten U, Correa P, Hewison M, Hellman P, Dralle H, Carling T, Akerstrom G, Westin G 2002 25-hydroxyvitamin D(3)-1alpha-hydroxylase expression in normal and pathological parathyroid glands. J Clin Endocrinol Metab 87:2967–2972. 31. Schwartz GG, Whitlatch LW, Chen TC, Lokeshwar BL, Holick MF 1998 Human prostate cells synthesize 1,25-dihydroxyvitamin D3 from 25-hydroxyvitamin D3. Cancer Epidemiol Biomarkers Prev 7:391–395. 32. Jones G, Ramshaw H, Zhang A, Cook R, White J, Petkovich M 1999 Expression and activity of vitamin D-metabolizing cytochrome P450s (CYP1α and CYP24) in human nonsmall cell lung carcinomas. Endocrinology 140:3303–3310. 33. Henry HL, Norman AW 1975 Presence of renal 25-hydroxyvitamin-D-1-hydroxylase in species of all vertebrate classes. Comp Biochem Physiol 50B:441–434. 34. Nahm TH, Lee SW, Fausto A, Sonn Y, Avioli LV 1979 25-(OH)D, a circulating vitamin D metabolite in fish. Biochem Biophys Res Commun 89:396–402. 35. Egaas E, Lambertsen G 1979 Naturally occurring vitamin D3 in fish products analyzed by high performance liquid chromatography using vitamin D2 as an internal standard. J Am Oil Chem Soc 56:188. 36. Takeuchi A, Okano T, Kobayashi T 1991 The existence of 25-hydroxyvitamin D3-1α-hydroxylase in the liver of carp and bastard halibut. Life Sci 48:275–282. 37. Brown PB, Robinson EH 1992 Vitamin D studies with channel catfish (Ictalurus punctatus) reared in calcium-free water. Comp Biochem Physiol 103A:213–219. 38. Rao DS, Raghuramulu N 1996 Food chain as origin of vitamin D in fish. Comp Biochem Physiol 114A:15–19. 39. Rao DS, Raghuramulu N 1999 Vitamin D3 and its metabolites have no role in calcium and phosphorus metabolism in Tilapia mossambica. J Nutr Sci Vitaminol (Tokyo) 45:9–19. 40. Buffenstein R, Sergeev IN, Pettifor JM 1993 Vitamin D hydroxylases and their regulation in a naturally vitamin D– deficient subterranean mammal, the naked mole rat (Heterocephalus glaber). J Endcrinol 138:59–64. 41. Sergeev IN, Buffenstein R, Pettifor JM 1993 Vitamin D receptors in a naturally vitamin D-deficient subterranean mammal, the naked mole rat (Heterocephalus glaber): Biochemical characterization. Gen Comp Endocrinol 90:338–345. 42. Henry HL, Norman AW 1974 Studies on calciferol metabolism. IX. Renal 25-hydroxyvitamin D3-1-hydroxylase. Involvement of cytochrome P-450 and other properties. J Biol Chem 249: 7529–7535. 43. Hagenfeldt Y, Pedersen JI, Bjorkhem I 1988 Properties of guinea-pig kidney 25-hydroxyvitamin D3-1α-hydroxylase assayed by isotope dilution-mass spectrometry. Biochem J 250:521–526. 44. Churchill PF, deAlvare LR, Kimura T 1978 Topological studies of the steroid hydroxylase complexes in bovine adrenocortical mitochondria. J Biol Chem 253:4924–4929. 45. Ghazarian JG, Jefcoate CR, Knutson JC, Orme-Johnson W, DeLuca HF 1974 Mitochondrial cytochrome P450. A component of chick kidney 25-hydroxycholecalciferol1α−hydroxylase. J Biol Chem 249:3026–3033. 46. Hiwatashi A, Nishii Y, Ichikawa Y 1982 Purification of cytochrome P-450-1α (25-hydroxyvitamin D3-1α hydroxylase) of bovine kidney mitochondria. Biochem Biophys Res Commun 105:320–327.
80 47. Burgos-Trinidad M, Ismail R, Ettinger RA, Prahl JM, DeLuca HF 1992 Immunopurified 25-hydroxyvitamin D 1α-hydroxylase and 1,25-dihydroxyvitamin D 24-hydroxylase are closely related but distinct enzymes. J Biol Chem 267:3498–3505. 48. Moorthy B, Mandel ML, Ghazarian JG 1991 Amino-terminal sequence homology of two chick kidney mitochondrial proteins immunoisolated with monoclonal antibodies to the cytochrome P450 of the 25-hydroxyvitamin D3-1α hydroxylase. J Bone Miner Res 6:199–204. 49. Wakino S, Meguro M, Suzuki H, Saruta T, Ogishima T, Ishimura Y, Shinki T, Suda T 1996 Evidence for 54-kD protein in chicken kidney as a cytochrome P450 with a high molecular activity of 25-hydroxyvitamin D3 1α-hydroxylase. Gerentology 42:67–77. 50. Gray RW, Ghazarian JG 1989 Solubilization and reconstitution of kidney 25-hydroxyvitamin D3 1α- and 24-hydroxylases from vitamin D-replete pigs. Biochem J 259:561–568. 51. Postlind H 1990 Separation of the cytochromes P-450 in pig kidney mitochondria catalyzing 1α-, 24- and 26-hydroxylations of 25-hydroxyvitamin D3. Biochem Biophys Res Commun 168:261–266. 52. Nelson DR 1999 Cytochrome P450 and the individuality of species. Arch Biochem Biophys 369:1–10. 53. Okamura T, John ME, Zuber MX, Simpson ER, Waterman MR 1985 Molecular cloning and amino acid sequence of the precursor form of bovine adrenodoxin: evidence for a previously unidentified COOH-terminal peptide. Proc Natl Acad Sci USA 82:5705–5709. 54. Sakihama N, Hiwatashi A, Miyatake A, Shin M, Ichikawa Y 1988 Isolation and purification of mature bovine adrenocortical ferredoxin with an elongated carboxyl end. Arch Biochem Biophys 264:23–29. 55. Cupp JR Vickery LE 1989 Adrenodoxin with a COOHterminal deletion (des 116-128) exhibits enhanced activity. J Biol Chem 264:1602–1607. 56. Coughlan VM, Cupp JR, Vickery LE 1988 Purification and characterization of human placental ferredoxin. Arch Biochem Biophys 264:376–382. 57. Tang C, Kain SR, Henry HL 1993 The phorbol ester 12-O-tetradecanoyl-phorbol-13-acetate stimulates the dephosphorylation of mitochondrial ferredoxin in cultured chick kidney cells. Endocrinology 133:1823–1829. 58. Yoon PS, DeLuca HF 1980 Purification and properties of chick renal mitochondrial ferredoxin. Biochemistry 19:2165–2171. 59. Driscoll WJ, Omdahl JL 1989. Characterization and N-terminal amino acid sequence of multiple ferredoxins in kidney and adrenal mitochondria. Eur J Biochem 185:181–187. 60. Tang C, Henry HL 1993 Overexpression in Escherichia coli and affinity purification of chick kidney ferredoxin [published erratum appears in J Biol Chem 1993 268(15):11474]. J Biol Chem 268:5069–5076. 61. Omdahl JL, Wilson K, Swerdlow H, Driscoll WJ 1992 Molecular cloning and immunological characterization of porcine kidney ferredoxin. Arch Biochem Biophys 293:213–218. 62. Blanchard RK, Henry HL 1996 Chick kidney ferredoxin: Complementary DNA cloning and vitamin D effects on mRNA levels. Comp Biochem Physiol [B] 114B:337–344. 63. Kagimoto K, McCarthy JL, Waterman MR, Kagimoto M 1988 Deduced amino acid sequence ot mature chick testis ferredoxin. Biochem Biophys Res Commun 155:379–383. 64. Kagimoto M, Kagimoto K, Simpson ER, Waterman MR 1988 Transcription of the bovine adrenodoxin gene produces two species of mRNA of which only one is translated into adrenodoxin. J Biol Chem 263:8925–8928.
HELEN L. HENRY
65. Chang C-Y, Wu D-A, Mohandas TK, Chung B-C 1990 Structure, sequence, chromosomal location and evolution of the human ferredoxin gene. DNA Cell Biol 9:205–212. 66. Voutilainen R, Picado-Leonard J, DiBlasio AM, Miller WL 1988 Hormonal and developmental regulation of adrenodoxin messenger ribonucleic acid in steroidogenic tissues. J Clin Endocrinol Metab 66:383–388. 67. John ME, John TS, Boggaram V, Simpson ER, Waterman MR 1986 Transcriptional regulation of steroid hydroxylase genes by corticotropin. Proc Natl Acad Sci USA 83:4715–4719. 68. Kramer RE, Anderson CM, Peterson JA, Simpson ER, Waterman MR 1982 Adrenodoxin biosynthesis by bovine adrenal cells in monolayer culture. J Biol Chem 257: 14921–14925. 69. Miller WL 1988 Molecular biology of steroid hormone synthesis. Endocr Rev 9:295–318. 70. Simpson ER, Waterman MR 1983 Regulation by ACTH of steroid hormone biosynthesis in the adrenal cortex. Can J Biochem Cell Biol 61:692–707. 71. Gnanaiah W, Omdahl JL 1986 Isolation and characterization of pig kidney mitochondrial ferredoxin: NADP+ oxidoreductase. J Biol Chem 261:12649–12654. 72. Chu JW, Kimura T 1973 Studies on the adrenal steroid hydroxylases. Molecular and catalytic properties of adrenodoxin reductase (a flavoprotein). J Biol Chem 248:2089–2094. 73. Brandt ME, Vickery LE 1992 Expression and characterization of human mitochondrial ferredoxin reductase in Escherichia coli. Arch Biochem Biophys 294:735–740. 74. Solish SB, Picado-Leonard J, Morel Y, Kuhn RW, Mohandas TK, Hanukoglu I, Miller WL 1988 Human adrenodoxin reductase: Two mRNAs encoded by a single gene on chromosome 17cen q25 are expressed in steroidogenic tissues. Proc Natl Acad Sci USA 85:7104–7108. 75. Gacad MA, Adams JS 1993 Identification of a competitive binding component in vitamin D-resistant New World primate cells with a low affinity but high capacity for 1,25-dihydroxyvitamin D3. J Bone Miner Res 8:27–35. 76. Wu S, Ren S., Chen H, Chun RF, Gacad MA, Adams JS 2000 Intracellular vitamin D binding proteins: Novel facilitators of vitamin D-directed transactivation. Mol Endocrinol 14: 1387–1397. 77. Wu S, Chun R, Gacad MA, Ren S, Chen H, Adams JS 2002 Regulation of 1,25-dihydroxyvitamin D synthesis by intracellular vitamin D binding protein-1. Endocrinology 143:4135–4138. 78. Nykjaer A, Dragun D, Walther D, Vorum H, Jacobsen C, Herz J, Melsen F, Christensen EI, Willnow, TE 1999 An endocytic pathway essential for renal uptake and activation of the steroid 25-(OH) vitamin D3. Cell 96:507–515. 79. Hilpert J, Wogensen L, Thykjaer T, Wellner M, Schlichting U, Orntoft TF, Bachmann S, Nykjaer A, Willnow TE 2002 Expression profiling confirms the role of endocytic receptor megalin in renal vitamin D3 metabolism. Kidney Int 62:1672–1681. 80. Leheste JR, Melsen F, Wellner M, Jansen P, Schlichting U, Renner-Muller I, Andreassen TT, Wolf E, Bachmann S, Nykjaer A, Willnow TE 2003 Hypocalcemia and osteopathy in mice with kidney-specific megalin gene defect. FASEB J 17:247–249. 81. Takeyama K, Kitanaka S, Sato T, Kobori M, Yanagisawa J, Kato S 1997 25-Hydroxyvitamin D3 1α-hydroxylase and vitamin D synthesis. Science 277:1827–1830. 82. Shinki T, Shimada H, Wakino S, Anazawa H, Hayashi M, Saruta T, DeLuca HF, Suda T 1997 Cloning and expression of rat 25-hydroxyvitamin D3-1α-hydroxylase cDNA. Proc Natl Acad Sci USA 94:12920–12925.
CHAPTER 5 The 25-Hydroxyvitamin D 1α-Hydroxylase
83. St-Arnaud R, Messerlian S, Moir JM, Omdahl JL, Glorieux FH 1997 The 25-hydroxyvitamin D 1α-hydroxylase gene maps to the pseudovitamin D deficiency rickets (PDDR) disease locus. J Bone Miner Res 12:1552–1559. 84. Fu GK, Lin D, Zhang MY, Bikle DD, Shackleton CH, Miller WL, Portale AA 1997 Cloning of human 25-hydroxyvitamin D1α-hydroxylase and mutations causing vitamin D-dependent rickets type 1. Mol Endocrinol 11:1961–1970. 85. Monkawa T, Yoshida T, Wakino S, Shinki T, Anazawa H, DeLuca HF, Suda T, Hayashi M, Saruta T 1997 Molecular cloning of cDNA and genomic DNA for human 25-hydroxyvitamin D3 1α-hydroxylase. Biochem Biophys Res Commun 239:527–533. 86. Yoshida T, Yoshida N, Nakamura A, Monkawa T, Hayashi M, Saruta T 1999 Cloning of porcine 25-hydroxyvitamin D3 1α-hydroxylase and its regulation by cAMP in LLC-PK1 cells. J Am Soc Nephrol 10:963–970. 87. Fu GK, Portale A, Miller WL 1997 Complete structure of the human gene for the vitamin D 1α−hydroxylase, P450c1α. DNA Cell Biol 16:1961–1970. 88. Kong XF, Zhu XH, Pei YL, Jackson DM, Holick MF 1999 Molecular cloning, characterization, and promoter analysis of the human 25-hydroxyvitamin D3-1α-hydroxylase gene. Proc Natl Acad Sci USA 96:6988–6993. 89. Panda DK, Kawas SA, Seldin MF, Hendy GN, Goltzman D 2001 25-Hydroxyvitamin D 1α-hydroxylase: Structure of the mouse gene, chromosomal assignment, and developmental expression. J Bone Miner Res 16:46–56. 90. Henry HL, Midgett RJ, Norman AW 1974 Studies on calciferol metabolism X: Regulation of 25-hydroxyvitamin D3-1-hydroxylase, in vivo. J Biol Chem 249:7584–7592. 91. Colston KW, Evans IM, Spelsberg TC, Macintyre I 1977 Feedback regulation of vitamin D metabolism by 1,25-dihydroxycholecalciferol. Biochem J 164:83–89. 92. Takahashi N, Miyaura C, Horiuchi N, Suda T 1981 Regulation and its refractoriness of 25-hydroxyvitamin D3 metabolism in vitamin D deficiency. J Nutr Sci Vitaminol 27:429–437. 93. Henry HL 1979 Regulation of the hydroxylation of 25hydroxyvitamin D3 in vivo and in primary cultures of chick kidney cells. J Biol Chem 254:2722–2729. 94. Spanos E 1983 Regulation of the renal 25OHD3 hydroxylases by vitamin-D3 metabolites and analogs in primary chick kidney cell cultures. Acta Endocrinol 104:25–26. 95. Turner RT, Bottemiller BL, Howard GA, Baylink DJ 1980 In vitro metabolism of 25-hydroxyvitamin D3 by isolated rat kidney cells. Proc Natl Acad Sci USA 77:1537–1540. 96. Murayama A, Takeyama K, Kitanaka S, Kodera Y, Kawaguchi Y, Hosoya T, Kato S 1999 Positive and negative regulations of the renal 25-hydroxyvitamin D3 1α-hydroxylase gene by parathyroid hormone, calcitonin, and 1α,25(OH)2D3 in intact animals. Endocrinology 140:2224–2231. 97. Shinki T, Ueno Y, DeLuca HF, Suda T 1999 Calcitonin is a major regulator for the expression of renal 25-hydroxyvitamin D3-1alpha-hydroxylase gene in normocalcemic rats. Proc Natl Acad Sci USA 96:8253–8258. 98. Condamine L, Vztovsnik F, Friedlander G, Menaa C, Garabédian M 1994 Local action of phosphate depletion and insulin-like growth factor 1 on in vitro production of 1,25dihydroxyvitamin D3 by cultured mammalian kidney cells. J Clin Invest 94:1673–1679. 99. Bland R, Walker EA, Hughes SV, Stewart PM, Hewison M 1999 Constitutive expression of 25-hydroxyvitamin D3-1αhydroxylase in a transformed human proximal tubule cell line: evidence for direct regulation of vitamin D metabolism by calcium. Endocrinology 140:2027–2034.
81 100. Yoshida T, Yoshino J, Hayashi M, Saruta T 2002 Identification of a renal proximal tubular cell-specific enhancer in the mouse 25-hydroxyvitamin D 1alpha-hydroxylase gene. J Am Soc Nephrol 13:1455–1463. 101. Fraser DR, Kodicek E 1973 Regulation of 25-hydroxycholecalciferol-1-hydroxylase activity in kidney by parathyroid hormone. Nature New Biol 241:163–166. 102. Baksi SN, Kenny AD 1979 Acute effects of parathyroid extract on renal vitamin D hydroxylases in Japanese quail. Pharmacology 18:169–174. 103. Booth BE, Tsai HC, Morris RC 1985 Vitamin-D status regulates 25-hydroxyvitamin-D3-1-α-hydroxylase and its responsiveness to parathyroid hormone in the chick. J Clin Invest 75:155–161. 104. Walker AT, Stewart AF, Korn EA, Shiratori T, Mitnick MA, Carpenter TO 1990 Effect of parathyroid hormone-like peptides on 25-hydroxyvitamin D-1α-hydroxylase activity in rodents. Am J Physiol 258:E297–E303. 105. Armbrecht HJ, Boltz MA, Hodam TL 2003 PTH increases renal 25(OH)D3-1alpha-hydroxylase (CYP1alpha) mRNA but not renal 1,25(OH)2D3 production in adult rats. Am J Physiol Renal Physiol 284:F1032–F1036. 106. Spanos E, Brown DJ, Macintyre I 1979 Regulation of 25(OH)-D3 metabolism by parathyroid hormone in primary chick kidney cell cultures. FEBS Lett 105:31–34. 107. Henry HL 1981 Insulin permits parathyroid hormone stimulation of 1,25-dihydroxyvitamin D3 production in cultured kidney cells. Endocrinology 108:733–735. 108. Rost CR, Bikle DD, Kaplan RA 1981 In vitro stimulation of 25-hydroxycholecalciferol 1α-hydroxylation by parathyroid hormone in chick kidney slices: Evidence for a role for adenosine 3′5′-monophosphate. Endocrinology 108:1002–1006. 109. Fukase M, Birge SJ, Rifas L, Avioli LV, Chase LR 1982 Regulation of 25 hydroxyvitamin D3-1-hydroxylase in serumfree monolayer culture of mouse kidney. Endocrinology 110: 1073–1075. 110. Kramer R, Goltzman D 1982 Parathyroid hormone stimulates mammalian 25-hydroxyvitamin D3-1α-hydroxylase in vitro. Endocrinology 110:294–296. 111. Henry HL 1985 Parathyroid hormone modulation of 25-hydroxyvitamin D3 metabolism by cultured chick kidney cells is mimicked and enhanced by forskolin. Endocrinology 116:503–507. 112. Noland TA, Henry HL 1983 Protein phosphorylation in chick kidney: Response to parathyroid hormone, cyclic AMP, calcium and phosphatidyl serine. J Biol Chem 258: 538–546. 113. Henry HL, Luntao EM 1989 Interactions between intracellular signals involved in the regulation of 25-hydroxyvitamin D3 metabolism. Endocrinology 124:2228–2234. 114. Ro HK, Tembe V, Favus MJ 1992 Evidence that activation of protein kinase C can stimulate 1,25-dihydroxyvitamin D3 secretion by rat proximal tubules. Endocrinology 131: 1424–1428. 115. Dunlay R, Hruska KA 1990 PTH receptor coupling to phospholipase C is an alternative pathway of signal transduction in bone and kidney. Am J Physiol 258:F223–F231. 116. Galante L, Colston KW, MacAuley SJ, Macintyre I 1972 Effect of calcitonin on vitamin D metabolism. Nature 238: 271–273. 117. Nesbitt T, Lobaugh B, Drezner MK 1987 Calcitonin stimulation of renal 25-hydroxyvitamin-D-1α-hydroxylase activity in hypophosphatemic mice—evidence that the regulation of calcitriol production is not universally abnormal in x-linked hypophosphatemia. J Clin Invest 79:15–19.
82 118. Horiuchi N, Takashita H, Matsumoto T, Takahashi N, Shimazawa E, Suda T, Ogata E 1979 Salmon calcitonininduced stimulation of 1α,25-dihydroxycholecalciferol synthesis in rats involving a mechanism independent of adenosine 3′:5′:-cyclic monophosphate. Biochem J 184:269–275. 119. Larkins RG, MacAuley SJ, Rapoport A, Martin TJ, Tullock BR, Macintyre I 1974 Effects of nucleotides,hormones, ions, and 1,25-dihydroxycholecalciferol on 1,25-dihydroxycholecalciferol production in isolated chick renal tubules. Clin Sci Mol Med 46:569–582. 120. Rasmussen H, Wong M, Bikle DD, Goodman DBP 1972 Hormonal control of the renal conversion of 25-hydroxycholecalciferol to 1,25-dihydroxycholecalciferol. J Clin Invest 51:2502–2504. 121. Henry HL, Noland TA, Al-Abdaly FA, Cunningham NS, Luntao EM, Amdahl LD 1982 Responses of cultured chick kidney cells to parathyroid hormone and calcitonin. In: Norman AW, Schaefer K, Herrath DV, Grigoleit H-G (eds.) Vitamin D: Chemical, Biochemical, and Clinical Endocrinology of Calcium Metabolism. Walter de Gruyter, Berlin. 122. Yoshida N, Yoshida T, Nakamura A, Monkawa T, Hayashi M, Saruta T 1999 Calcitonin induces 25-hydroxyvitamin D3 1αhydroxylase mRNA expression via protein kinase C pathway in LLC-PK1 cells. J Am Soc Nephrol 10:2474–2479. 123. Hughes MR, Brumbaugh PF, Hussler MR, Wergedal JE, Baylink DJ 1975 Regulation of serum 1alpha,25-dihydroxyvitamin D3 by calcium and phosphate in the rat. Science 190:578–580. 124. Baxter LA, DeLuca HF 1976 Stimulation of 25-hydroxyvitamin D3-1α-hydroxylase by phosphate depletion. J Biol Chem 251:3158–3161. 125. Gray RW, Napoli JL 1983 Dietary phosphate deprivation increase 1,25-dihydroxyvitamin D3 synthesis in rat kidney in vitro. J Biol Chem 258:1152–1155. 126. Rader JI, Baylink DJ, Hughes MR, Safilian EF, Haussler MR 1979 Calcium and phosphorus deficiency in rats: Effect on PTH and 1,25-dihydroxyvitamin D3. Am J Physiol 236: E118–E122. 127. Gray RW, Wilz DR, Caldas AE, Lemann J 1977 The importance of phosphate in regulating plasma 1,25-(OH)2-vitamin D levels in humans: Studies in healthy subjects, in calcium-stone formers and in patients with primary hyperparathyroidism. J Clin Endocrinol Metab 45:299–306. 128. Maierhofer WJ, Gray RW, Lemann J 1984 Phosphate deprivation increases serum 1,25-(OH)2-vitamin-D concentrations in healthy men. Kidney Int 25:571–575. 129. Portale AA, Halloran BP, Murphy MM, Morris FC 1986 Oral intake of phosphorus can determine the serum concentration of 1,25-dihydroxyvitamin D by determining its production rate in humans. J Clin Invest 77:7–12. 130. Portale AA, Halloran BP, Murphy MM, Morris RC, Jr. 1986 Oral intake of phosphorus can determine the serum concentration of 1,25-dihydroxyvitamin D by determining its production rate in humans. J Clin Invest 77:7–12. 131. Castillo L, Tanaka Y, DeLuca HF, Sunde ML 1977 The stimulation of 25-hydroxyvitamin D3-1α-hydroxylase by estrogen. Arch Biochem Biophys 179:211–217. 132. Baksi SN, Kenny AD 1980 Estradiol-induced stimulation of 25-hydroxyvitamin D3-1-hydroxylase in vitamin D-deficient Japanese quail. Pharmacology 20:298–303. 133. Pike JW, Spanos E, Coloston KW, Macintyre I, Haussler MR 1978 Influence of estrogen on renal vitamin D hydroxylases and serum 1α,25-(OH)2D3 in chicks. Am J Physiol 235: E338–E343.
HELEN L. HENRY
134. Henry HL 1981 25-OH-D3 metabolism in kidney cell cultures: Lack of a direct effect of estradiol. Am J Physiol 240:E119–E124. 135. Findling JW, Adams ND, Lemann J Jr., Gray RW, Thomas CJ, Tyrrell JB 1982 Vitamin D metabolites and parathyroid hormone in Cushing’s syndrome: Relationship to calcium and phosphorus homeostasis. J Clin Endocrinol Metab 54: 1039–1044. 136. Chesney RW, Hamstra AJ, Mazess RB, DeLuca HF, O’Reagan S 1978 Reduction of serum 1,25-dihydroxyvitamin D3 in children receiving glucocorticoids. Lancet 2:1123–1125. 137. Seeman E, Kumar R, Hunder GG, Scott M, Heath H, Riggs BL 1980 Production, degradation, and circulating levels of 1,25Dihydroxyvitamin D in health and in chronic glucocorticoid excess. J Clin Invest 66:664–669. 138. Henry HL 1986 Effect of dexamethasone on 25-hydroxyvitamin D3 metabolism in chick kidney cell cultures. Endocrinology 118:940–944. 139. Akeno N, Matsunuma A, Maeda T, Kawane T, Horiuchi N 2000 Regulation of vitamin D-1α-hydroxylase and 24-hydroxylase expression by dexamethasone in mouse kidney. J Endocrinol 164:339–348. 140. Henry HL 1985 Effect of ketoconazole and miconazole on 25-hydroxyvitamin D3 metabolism by cultured chick kidney cells. J Steroid Biochem Mol Biol 23:991–994. 141. Boass A, Toverud SU 1995 Suppression of circulating calcitriol and duodenal active Ca transport by ketoconazole in pregnant rats. Am J Physiol Endocrinol Metab 269: E934–E939. 142. Glass AR, Eil C 1986 Ketoconazole-induced reduction in serum 1,25-dihydroxyvitamin D. J Clin Endocr Metab 63: 766–769. 143. Saggese G, Bertelloni S, Baroncelli GI, Di Nero G 1993 Ketoconazole decreases the serum ionized calcium and 1,25-dihydroxyvitamin D levels in tuberculosis-associated hypercalcemia. Am J Dis Child 147:270–273. 144. Kowal J 1983 The effect of ketoconazole on steroidogenesis in cultured mouse adrenal cortex tumor cells. Endocrinology 112:1541–1543. 145. Loose DS, Kan PB, Hirst MA, Marcas RA, Feldman D 1983 Ketoconazole blocks adrenal steroidogenesis by inhibiting cytochrome P450-dependent enzymes. J Clin Invest 71: 1495–1499. 146. Zhao J, Tan BK, Marcelis S, Verstuyf A, Bouillon R 1996 Enhancement of antiproliferative activity of 1α,25-dihydroxyvitamin D3 (analogs) by cytochrome P450 enzyme inhibitors is compound- and cell-type specific. J Steroid Biochem Mol Biol 57:197–202. 147. Peehl DM, Skowronski RJ, Leung GK, Wong ST, Stamey TA, Feldman D 1994 Antiproliferative effects of 1,25-dihydroxyvitamin D3 on primary cultures of human prostatic cells. Cancer Res 54:805–810. 148. Peehl DM, Seto E, Feldman D 2001 Rationale for combination ketoconazole/vitamin D treatment of prostate cancer. Urology 58:123–126. 149. Yoshida T, Fujimori T, Nabeshima Y 2002 Mediation of unusually high concentrations of 1,25-dihydroxyvitamin D in homozygous klotho mutant mice by increased expression of renal 1α-hydroxylase gene. Endocrinology 143:683–689. 150. Tsujikawa H, Kurotaki Y, Fujimori T, Fukuda K, Nabeshima YI 2003 Klotho, a gene related to a syndrome resembling human premature aging, functions in a negative regulatory circuit of vitamin D endocrine system. Mol Endocrinol.
CHAPTER 5 The 25-Hydroxyvitamin D 1α-Hydroxylase
151. Murayama A, Takeyama K, Kitanaka S, Kodera Y, Hosoya T, Kato S 1998 The promoter of the human 25-hydroxyvitamin D3 1α−hydroxylase gene confers positive and negative responsiveness to PTH, calcitonin, and 1α,25(OH)2D3. Biochem Biophys Res Commun 249:11–16. 152. Brenza HL, DeLuca HF 2000 Regulation of 25-hydroxyvitamin D3 1α-hydroxylase gene expression by parathyroid hormone and 1,25-dihydroxyvitamin D3 [In Process Citation]. Arch Biochem Biophys 381:143–152. 153. Bringhurst FR, Juppner H, Guo J, Urena P, Potts JT Jr Kronenberg HM, Abou-Samra AB, Segre GV 1993 Cloned, stably expressed parathyroid hormone (PTH)/PTH-related peptide receptors activate multiple messenger signals and biological responses in LLC-PK1 kidney cells. Endocrinology 132:2090-2098. 154. Racusen LC, Monteil C, Sgrignoli A, Lucskay M, Marouillat S, Rhim JG, Morin JP 1997 Cell lines with extended in vitro
83 growth potential from human renal proximal tubule: characterization, response to inducers, and comparison with established cell lines. J Lab Clin Med 129:318–329. 155. Brenza HL, Kimmel-Jehan C, Jehan F, Shinki T, Wakino S, Anazawa H, Suda T, DeLuca HF 1998 Parathyroid hormone activation of the 25-hydroxyvitamin D3-1α-hydroxylase gene promoter. Proc Natl Acad Sci USA 95:1387–1391. 156. Armbrecht HJ, Hodam TL, Boltz MA 2003 Hormonal regulation of 25-hydroxyvitamin D3-1alpha-hydroxylase and 24-hydroxylase gene transcription in opossum kidney cells. Arch Biochem Biophys 409:298–304. 157. Gao XH, Dwivedi PP, Choe S, Alba F, Morris HA, Omdahl JL, May BK 2002 Basal and parathyroid hormone induced expression of the human 25-hydroxyvitamin D 1alpha-hydroxylase gene promoter in kidney AOK-B50 cells: role of Sp1, Ets and CCAAT box protein binding sites. Int. J Biochem Cell Biol 34:921–930.
CHAPTER 6
The 25-Hydroxyvitamin D 24-Hydroxylase JOHN OMDAHL BRIAN MAY
Office of Research, University of New Mexico School of Medicine, Albuquerque, NM 87131-5166 School of Molecular and Biomedical Science, University of Adelaide, South Australia 5005, Australia
I. Background II. Enzyme Structure and Function III. Cellular Expression and Regulation
I. BACKGROUND A. Enzyme Function and Regulated Expression Vitamin D is a secosteroid whose biological function is dependent upon its metabolic activation and turnover. These metabolite pathways contain specific hydroxylase enzymes that are members of the cytochrome P450 superfamily of mixed-function monooxygenases. Bioactivation of vitamin D involves the sequential actions of 25-hydroxylase and 1-hydroxylase enzymes leading to the synthesis of the hormonally active secosteroid 1,25-dihydroxyvitamin D [1,25(OH)2D]* (Fig. 1). These two enzymes are discussed in Chapter 4 (vitamin D 25-hydroxylase) and Chapter 5 (25-hydroxyvitamin D 1α-hydroxylase) and will be mentioned in this chapter only on a comparative basis to 25-hydroxyvitamin D24(R)-hydroxylase cytochrome P450c24 (CYP24), the enzyme that directs the side-chain metabolism of 25-hydroxylated vitamin D metabolites, which leads to their terminal physiological inactivation and turnover. Most cellular actions of vitamin D are mediated through the secosteroid hormone 1,25(OH)2D and involve the transcription of vitamin D–dependent genes. These regulatory processes involve the coordinated modulation and coupling of rapid signal-transduction pathways with slower acting ligand-dependent transcription factors [1]. In both cases, the secosteroid ligand binds to a ligand-specific receptor. The rapid response receptor is located in the cellular membrane of target tissues and initiates rapid signaling responses through a receptor that has been referred to as the membrane-associated receptor [e.g., mVDR or
*The term vitamin D will be used to denote vitamin D3 unless stated otherwise. VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX
IV. Molecular Aspects References
membrane-associated rapid response steroid (MSSRS) receptor complex] [2,3]. The hormone receptor for the transcription process involves the nuclear vitamin D receptor (nVDR), which is a ligand-dependent transcription factor that functions as a VDR:1,25(OH)2D heterodimeric complex with the cis-retinoic acid: RXR complex (i.e., VDR-RXR) to regulate vitamin D–dependent genes associated with development and homeostasis (see Chapters 13–17 for details). Cellular and ambient levels 1,25(OH)2D are regulated through the hormone’s synthesis and degradation. The regulated 1α-hydroxylase (i.e., P4540c1 or CYP27B1) directs the hormone’s biosynthesis. In a complementary manner, the regulatory 24(R)-hydroxylase enzyme (i.e., P450c24 or CYP24A1) mediates degradation of 1,25(OH)2D through side-chain oxidation reactions. Expression of the two enzymes is most commonly regulated in a diametric manner that functions to maintain the hormone’s level at a proper set point. For example, low-calcium states are associated with an elevation in parathyroid hormone (PTH) that functions to up-regulate P450c1 (i.e., CYP27B1) while down-regulating P450c24 (i.e., CYP24A1) expression. The associated increase in 1,25(OH)2D functions to stimulate calcium absorption and correct the low-calcium state. Control of 1,25(OH)2D overproduction and the attendant development of a hypercalcemic state is autoregulated through a negative feedback system whereby 1,25(OH)2D down-regulates its oversynthesis (i.e., suppresses CYP27B1 expression) while up-regulating expression of CYP24A1. Of particular importance is the action of CYP24A1 to catabolically inactivate 1,25(OH)2D and thereby function to counterregulate the hormone’s cellular and ambient concentration. The regulatory and expression linkages between CYP27B1 and CYP24A1 are fundamental to understanding the vitamin D regulatory pathway and will serve as a point of reference in this chapter as well as other Copyright © 2005, Elsevier, Inc. All rights reserved.
86
JOHN OMDAHL AND BRIAN MAY
Liver Liver UV light 7-dehydro cholesterol
Skin Skin
P450c25
Pre-Vitamin D
Diet
25(OH)D3 Kidney Kidney
Target Target Tissues Tissues P450c24
P450c1 Calcitroic acid
1,25(OH)2D3 24,25(OH)22D33
1,25(OH)22D33 RXR VDR
( )
P450c24
P450c1
1,24,25(OH)33D33 VDRE
Genes
P450c24 Calcitroic acid
Excretion
FIGURE 1
Activation and molecular pathways for vitamin D. Integrative schematic of vitamin D synthesis, metabolism and molecular action.
sections in this book in which various aspects of vitamin D metabolism are discussed extensively.
B. Enzyme Cellular Expression The 24(R)-hydroxylase enzyme displays a broad tissue distribution wherein it is expressed in nearly all cells. A number of examples are discussed in the regulatory section in the second half of this chapter. The major route of enzyme induction involves the combined action of 1,25(OH)2D and VDR to increase transcription of the CYP24A1 gene as discussed in the latter two sections of this chapter. Recognizing the importance of 1,25(OH)2D in the induction of CYP24A1, the enzyme is not expressed in vitamin D– deficient animals. Consequently, the cellular expression of CYP24A1 is linked tightly to the vitamin D status and coexpression of VDR. Because of the ubiquitous distribution of VDR [4,5], most cells express a basal level of CYP24A1 or respond to increased 1,25(OH)2D levels by inducing biosynthesis of the enzyme, particularly in kidney and small intestine [6–8]. Consequently, the cellular level of CYP24A1 is particularly influenced by the concentration of VDR and the level of 1,25(OH)2D, the latter reflecting both locally produced hormone [9] and hormone acquired from the circulation. Hence the broad cellular induction of CYP24A1 by 1,25(OH)2D can be envisioned as an effector arm of the negative feedback system that functions to maintain
proper hormone level and avoids the buildup of a potentially toxic secosteroid. Kidney is the major endocrine organ for 1,25(OH)2D and therefore there has been considerable interest in the regulation of renal CYP24A1 expression. Using the rat developmental model, Matkovits and Christakos [6] reported low levels of renal mRNA at birth that remained unchanged for several weeks and then increased substantially over the next 20 months, thereby establishing kidney as the major site of basal enzyme expression. In recent studies, quantitation of rat kidney mRNA levels using a real time RT-PCR protocol [10] confirmed this developmental profile for CYP24A1 mRNA expression (PH Anderson, S Iida, BK May, HA Morris, submitted). A small fall in renal CYP24A1 mRNA level was seen between 3 and 9 weeks of age followed by a continuous increase until 24 months. In a comparative measurement, CYP27B1 (i.e., 1αhydroxylase) mRNA was highest at 3 weeks of age and then decreased particularly between 3 and 15 weeks, followed by a further slight decrease over 24 months. The high production of renal 1,25(OH)2D during earlier development would ensure adequate circulating 1,25(OH)2D and hence calcium for bone growth. Serum levels of 1,25(OH)2D are determined by the expression levels of both CYP27B1 and CYP24A1 in the kidney [10]. CYP24A1 is expressed in both proximal and distal tubules [11–13], although induced enzyme expression occurs to a higher level in the proximal tubule [13].
87
CHAPTER 6 The 25-Hydroxyvitamin D 24-Hydroxylase
Mechanistic and regulatory studies of 1,25(OH)2D3 action and CYP24A1 expression have relied on a host of cell-culture models. Investigations with cultured cells have revealed induction of the CYP24A1 gene in many different cell types. Enzyme activity and/or CYP24A1 mRNA levels have been detected in numerous primary and transformed cultures that include intestinal, osteoblast, monocyte/macrophage, keratinocyte, and cancer cells [14–25]. In most of these studies, cells incubated in a serum-free medium expressed little if any 24(R)-hydroxylase activity or CYP24A1 mRNA, but enzyme and message levels were highly induced by 1,25(OH)2D treatment. In addition to 1,25(OH)2D, other more cell-type-selective regulators of CYP24A1 gene expression have been studied, and include PTH, calcitonin, phorbol ester, dexamethasone, and gammainterferon (see Section III of this chapter). The induction of CYP24A1 mRNA in primary cultures of rat renal tubular cells was prevented by actinomycin D, verifying that the increase in CYP24A1 mRNA level was due to gene transcription [15]. Cycloheximide also inhibited the induction of CYP24A1 mRNA in these cells, demonstrating a continued need for protein synthesis [15,19], which could involve both signal transduction and transcription regulatory proteins.
C. Enzyme Properties P450c24 acquired its name 25-hydroxyvitamin D 24(R)-hydroxylase based upon its action to 24(R)hydroxylate, the prehormone 25-hydroxyvitamin D3 [25OHD] to 24(R),25-dihydroxyvitamin D [24,25 (OH)2D] [26]. The P450c24 enzyme was subsequently shown to 24(R)-hydroxylate 1,25(OH)2D3 to 1,24,25 (OH)3D3 [27]. See Figure 2. The CYP24A1 in some species expresses predominantly 23(S)-hydroxylase activity [11,28]; therefore, CYP24A1 isoenzymes direct both C23 and C24 oxidative pathways. In both instances, the dominant C23 or C24 hydroxylation activity directs the initial step associated with the enzyme’s side-chain oxidative reactions associated with the inactivation of 25-hydroxylated vitamin D metabolites as discussed in the next section on enzyme pathways. The CYP24A1 enzyme is a protoporphyrin IX hemoprotein (~52 kDa) that binds oxygen and directs the regiospecific and stereospecific hydroxylation and oxidation of secosteroid substrate molecules using electrons derived from NADPH [29] (Fig. 2). The vitamin D enzyme is distinct from the 24(S)-hydroxylase enzyme involved with cholesterol metabolism [30] (i.e., CYP2R1). CYP24A1 is an inner mitochondrial membrane enzyme that is part of a mini–electron
Cyp24A1red
Cyp24A1ox
25 (OH)D3 + O2 [1,25 (OH)2D3]
24,25 (OH)2D3 + H2O [1,24,25 (OH)3D3]
FIGURE 2
Terminal reaction of CYP24A1. The use of molecular oxygen by reduced CYP24A1 to 24(R)-hydroxylate 25-hydroxyvitamin D substrates. red, reduced; ox, oxidized.
transport chain consisting of NADPH-ferredoxin reductase, ferredoxin (an iron–sulfur protein), and the terminal cytochrome CYP24A1 [31] (Fig. 2). Electrons are delivered through a specific protein:protein interaction between the enzyme and the electron donor ferredoxin [32]. Since the vitamin D 24(R)-hydroxylase enzyme is less than 40% similar to other P450s, it was assigned to a new P450 family and given the cytochrome P450 abbreviation CYP24A1. Consistent with its unique protein sequence, CYP24A1 does not react immunologically with several liver P450 enzymes or mitochondrial P450c27 (CYP27A1) that expresses 25-hydroxylase activity and is a family member with the 25-hydroxyvitamin D 1α-hydroxylase enzyme (i.e., CYP27).
D. Enzyme Pathways CYP24A1 directs the biosynthesis of 24,25(OH)2D and 1,24,25(OH)3D. These two vitamin D metabolites express less (i.e., 24,25(OH)2D) or approximately equivalent (i.e., 1,24,25(OH)3D) biological activity compared to 1,25(OH)2D. A biological role for 24,25 (OH)2D has been studied most extensively [2,3] because of its higher cellular and ambient concentration. Both metabolites are initial reactants in the C24-oxidation pathway that leads to the side-chain oxidation of 25-hydroxy metabolites to terminal C23-carboxylic acid end products [33–35] (Fig. 3). An investigation with partially purified 24(R)-hydroxylase enzyme from pig kidney documented copurification of the 23- and 24-hydroxylase activities [31], which was consistent with one enzyme expressing both activities. All catalytic activities in the C24-oxidation pathway were subsequently shown to be expressed by recombinant rat CYP24A1 [28,36]. The multifunctional enzyme initially oxidizes the 24-hydroxyl group to a 24-keto (oxo) functionality. Subsequent steps involve the 23(S)-hydroxylation of the 24-oxo metabolites and
88
JOHN OMDAHL AND BRIAN MAY
to favor an initial C23(S)- or C24(R)-hydroxylation. It is evident, therefore, that CYP24A1 is a multicatalytic enzyme [28,36–38] that is capable of directing all reactions in the side-chain oxidation of 25-hydroxylated vitamin D metabolites to cyclic lactone (C23 pathway) [38] or C23 carboxylic-acid (C24 pathway) [37], or both [28,38] end products (Fig. 3). In this regard, CYP24A1 is the central enzyme involved with determining the biological half-life of 1,25(OH)2D and its therapeutic analogs.
II. ENZYME STRUCTURE AND FUNCTION A. Cytochrome P450 Motif and Function
FIGURE 3
Multicatalytic action of CYP24A1. Reactions of the C23 and C24 oxidative pathways catalyzed by CYP24A1 with lactol and carboxylic acid end products.
oxidative cleavage between C23 and C24 followed by oxidation of the C23 hydroxyl, via a C23 aldehyde intermediate, to 24,25,26,27-tetranor-23-COOH-vitamin D or calcitroic acid from 25(OH)2D and 1,25(OH)2D substrates, respectively (Fig. 3). However, the initial hydroxylation occurs at C23 in several CYP24A1 isoforms from different species, which result in an alternate pathway that leads to a side-chain cyclic lactone as final product (Fig. 3). For example, rat CYP24A1 expresses predominantly the C24-oxidation pathway, whereas human [28] and guinea pig [11] enzymes express significant levels of C23-hydroxylase activity that results in the synthesis of both lactone and carboxylicacid end products. Expression of predominantly 23(S)or 24(R)-hydroxylase activity in different isoforms would appear to be determined by point-residue differences in the substrate binding and catalytic center that direct substrate alignment in the heme active site so as
Cytochromes P450 constitute a superfamily of hemecontaining enzymes that direct the monooxygenation of hydrophobic substrates. More than 1000 genes have been identified and divided into 70 families [39] that metabolize preferred substrates ranging from polycyclic aromatic hydrocarbons to fatty acids and steroids. Although all P450s have a similar structural core, they have unique amino acid sequences that determine the enzyme’s substrate recognition and mediate regiospecific and stereospecific monooxygenation reactions within the active site. Molecular oxygen used in the monooxygenase reactions is activated through its interaction with the iron atom in the active site hemeprotoporphyrin IX molecule. In most instances, the oxygen molecule is activated to an oxyferryl functionality that stereospecifically directs the oxygenation/ hydroxylation of substrates that enter through a surface oriented substrate-access channel and subsequently align with the heme molecule in the enzyme’s central active site.
B. Molecular Modeling Only a limited number of P450 enzymes have been crystallized and characterized by X-ray diffraction analysis [40–44]. However, because of the molecules’ conserved structural attributes, it is possible to use molecular-modeling software (e.g., Insight-II and Sybyl) to develop enzyme similarity models that closely resemble the enzyme’s native conformation. Using the Insight-II Modeler program (Accelrys Inc.) and crystallographic coordinate values for P450-BM3 and P450-2C5, it has been possible to derive a similarity model for rat P450c24 (CYP24A1) (A Annalara, S Graham, JL Omdahl, unpublished data). Most of the substrate-interactive residues in CYP24A1 are located predominantly in the B′, F, G, and I helices, and the F-G loop.
89
CHAPTER 6 The 25-Hydroxyvitamin D 24-Hydroxylase
FIGURE 4
Model of rat CYP24A1 active site. A similarity model for CYP24A1 active site that shows alignment of the side-chain C24 with the heme (red) and alignment of the A-ring (adjacent to the 1,25(OH)2D notation) with F-249 in the F-helix.
Substrate docking of 1,25(OH)2D in the CYP24A1 model placed the C24 position in close proximity (2 Å) to the heme oxyferryl group with the A-ring of the molecule oriented toward the F-helix (Fig. 4). The A-ring displayed close interactions (3 Å) with met-246 and phe-249 of the F-helix. A prominent role for phe-249 in binding 1,25(OH)2D was determined though site-directed mutagenesis as discussed in the following section. The CYP24A1 similarity model is being used in conjunction with site-directed mutagenesis to map the full spectrum of substrate-binding sites within the substrate-access channel and active site.
C. Metabolic, Kinetic, and Spectroscopic Analysis 1. METABOLIC ANALYSIS
Consistent with the P450 nature of CYP24A1, the enzyme’s activity is impeded by P450 azole inhibitors [45–47]. The CYP24A1 purified from rat kidney [29] was shown to metabolize 25(OH)D to 24,25(OH)2D and 1,25(OH)2D to 1,24,25(OH)3D. Subsequent studies using bacterial-expressed recombinant enzyme demonstrated the action of CYP24A1 to side-chain metabolize 1,25(OH)2D [37]. The role of CYP24A1 to
direct side-chain oxidation reactions beyond the initial 24(R)-hydroxylation was established using recombinant enzyme preparations from bacteria and baculovirus [28,36–38]. Most interestingly, the rat enzyme was demonstrated to metabolize all steps in the C24-oxidation pathway leading to C23-carboxylic acid end products [36,37]. In contrast to the rat enzyme, human CYP24A1 [28,38] was found to catalyze both the C24 and C23 oxidation pathways that resulted in the tetranor C23-carboxylic acid and the 26,23-lactone end products, respectively (Fig. 3). Evident from these studies was the multicatalytic activity of CYP24A1 isozymes to direct all reactions in the C24 and C23 oxidative pathways. In contrast to previous recombinant CYP24A1 studies that used enzyme membrane preparations or detergent extracts (i.e., impure enzyme), a more recent investigation succeeded in purifying bacterial recombinant rat CYP24A1 (A Annalora, JL Omdahl, unpublished data). The purified enzyme expressed all steps in the C24 oxidation pathways, which demonstrated clearly the multicatalytic activity of CYP24A1. 2. KINETIC ANALYSIS
The rate of substrate metabolism by CYP24A1 has been analyzed using cellular, detergent extracted, and
90
JOHN OMDAHL AND BRIAN MAY
purified enzyme preparations. Km values from cellular and detergent extract preparations [19,48–51] span a large range (0.01–1.5 µM) in which Km values for 1,25(OH)2D are 1/4 to 1/30 that of 25(OH)D, which supported the concept that CYP24A1 has a higher preference for 1,25(OH)2D hormone. However, a more recent study [52] with solubilized enzyme determined the Km for 25(OH)D to be 0.6 µM and that of 1,25(OH)2D was 35-fold higher (20.9 µM), which indicated an enzyme preference for 25(OH)D instead of 1,25(OH)2D. The diversity of Km values is due in part to the sequential–multicatalytic properties of CYP24A1 and the combined effect of electron transfer from the co-redox ferredoxin partner and dissociation of intermediate reactants from the enzyme. Using purified rat CYP24A1, it has been possible to focus on the first step of the multicatalytic reaction path and to regulate the rate of electron transfer to the enzyme (A Pastuezyn, JL Omdahl, unpublished data). Using these conditions and those that favor the 24(R)-hydroxylation step, we have determined Km values for 25(OH)D and 1,25(OH)2D that are less than 1 µM in which 25(OH)D had a slightly lower Km and greater turnover number (A Pastuezyn, JL Omdahl, unpublished data). Also evident from the kinetic analysis with pure enzyme was the slow reaction kinetics for CYP24A1 [i.e., turnover number (TN) of 2–20 min−1] [29], compared to the higher rate observed for another mitochondrial P450 from the bile-acid pathway (i.e., CYP27A1, TN = 40–50 min−1) [53,54]. 3. SPECTRAL ANALYSIS
CYP24A1 displays a typical type-I mitochondrial P450 CO-reduced difference spectrum with a 450-nm Soret band that shifts to 420 nm upon enzyme inactivation. Because of the presence of a heme molecule in the enzyme’s active site, it is possible to follow the absolute spectral shift from 420 to 390 nm that occurs upon addition of substrate (i.e., substrate-induced spectral shift) and determine the dissociation constant (Kd) for substrate with enzyme [55]. Such binding techniques revealed a high specificity of the enzyme for vitamin D secosteroids containing a 1α- and/or 25-hydroxyl group (A Annalora, JL Omdahl et al., submitted). Both 1α-hydroxyvitamin D and 1,25(OH)2D displayed the lowest Kd (highest binding affinity) for CYP24A1. However, only 1,25(OH)2D was capable of being 24(R)-hydroxylated [29], which demonstrated the necessity of a C25-hydroxyl group for the physiological side-chain 24(R)-hydroxylation and subsequent oxidation reactions catalyzed by CYP24A1. In a related manner, the side-chain metabolism of 1,25(OH)2D2 to calcitroic acid is impeded by the vitamin D2 side-chain
structure modifications (i.e., double bond at C22 and methyl group at C24) [35,56]. The Kd for 25(OH)D is sixfold higher than observed for 1,25(OH)2D. But because of its lower Km, higher turnover number, and 1000-fold higher ambient level, the conversion of 25(OH)D to 24,25(OH)2D is a favored reaction. In addition, the 24(R)-hydroxylated metabolite has a decreased affinity for CYP24A1 (A Annalora, JL Omdahl, submitted), which favors its dissociation from enzyme, especially in cells where electron transfer is limiting. The combined effects result in a significant serum level of 24,25(OH)2D (2–5 ng/ml) that is 1/5 to 1/10 the concentration of the parent prehormone 25(OH)D. In contrast, the ambient level of 1,24,25(OH)3D is insignificant because of the metabolite’s low cellular concentration and greater affinity to remain bound to CYP24A1 (i.e., lower Kd) for further oxidative metabolism. Introduction of a single mutation in the F-helix of CYP24A1 has a profound impact on substrate binding and perturbation of the heme spin-state in the active site [57] (A Annalora, JL Omdahl, unpublished data). Mutation of phenylalanine 249 to threonine (F249T) caused a three- to sixfold increase in the Kd for 25(OH)D and 1,25(OH)2D, respectively. Based upon the CYP24A1 similarity model, phenylalanine 249 is projected to interact with the A-ring of 1,25(OH)2D when bound in the active site (Fig. 4). Loss of this hydrophobic interaction through the F249T mutation prevents the hormone from binding properly in the substrate pocket. As a consequence, the last oxidative sequence in the C24 pathway (i.e., 24,25,26,27-tetranor-1,23(OH)2D to calcitroic acid) is suppressed because of the misalignment of substrate side-chain in the active site. Mutagenesis of phenylalanine 249 to alanine (F249A) or tyrosine (F249Y) also confounded the substrate binding. However, in this instance the mutation affected an early step in the C24 pathway involving conversion of 1,24,25(OH)3D to 24-oxo-1,25(OH)2D. It is evident, therefore, that the combination of molecular modeling and site-directed mutagenesis can be used to identify amino acid residues that are involved with the multiple catalytic activities associated with CYP24A1.
III. CELLULAR EXPRESSION AND REGULATION A. CYP24A1 Knockout and Transgenic Animals CYP24A1 cDNA clones have been isolated for rat [58], human [59], chicken [60], mouse [61], and pig [62].
91
CHAPTER 6 The 25-Hydroxyvitamin D 24-Hydroxylase
The human CYP24A1 gene is located on chromosome 20 [63], but there are no known diseases that can be attributed to this gene, although aberrant overexpression is potentially oncogenic as discussed in a later section. The exon–intron structures have been determined for the rat [64] and mouse [61] CYP24A1 genes, which are “typical genes” of length about 15 kb. The biological functions of CYP24A1 gene expression in vivo have been pursued in conventional CYP24A1 knockout mice and also in transgenic rats constitutively expressing CYP24A1. Knockout mice were generated using embryonic stem cell technology so that the heme-binding domain of the mouse CYP24A1 gene was deleted ensuring that no functional CYP24A1 protein was produced [65,66]. Homozygous null mice exhibited a high circulating level of 1,25(OH)2D in keeping with the lack of catabolism through the C23/24 oxidation pathway. There was abnormal bone histology characterized by excessive undermineralized bone matrix and this raised the possibility that a deficiency of the CYP24A1 oxidation product 24,25(OH)2D may be responsible. While 24,25(OH)2D has been considered an inactive metabolite, there is evidence for a biological function of 24,25(OH)2D in bone physiology, particularly in the development of growth plate chondrocytes [67,68]. However, the mouse phenotype was rescued when the CYP24A1 null mice were mated with vitamin D receptor (VDR) knockout mice [69], demonstrating that the bone abnormality was not due to a deficiency of 24,25(OH)2D but most likely to elevated 1,25(OH)2D, which requires VDR for its biological actions [70,71]. In a different approach, constitutive expression of CYP24A1 has been achieved in the tissues of transgenic rats with the introduction of a rat CYP24A1 cDNA sequence under the control of the viral MLV-LTR promoter [72,73]. It was anticipated that characteristics of this transgenic rat would resemble those observed in CYP27B1 knockout mice where the lack of production of 1,25(OH)2D results in hypocalcemia, hyperparathyroidism, growth retardation, and rickets [74,75]. Unexpectedly, after 8 weeks the transgenic rats developed albuminuria and hyperlipidemia with a significant reduction in circulating 25(OH)D and 24,25(OH)2D levels accompanied by a lowering of bone mineral density [72,73]. On the other hand, the plasma concentration of 1,25(OH)2D in the transgenic rats was the same as in wild type, most likely because of up-regulation of CYP27B1. Excess renal excretion of albumin is probably responsible for the reduction in plasma 25(OH)D due to its prevention of 25(OH)D reuptake by renal tubular cells [73]. Infusion of transgenic rats with 25(OH)D prevented the mineralization defect by making the circulating 1,25(OH)2D
more available [73]. It is not clear how constitutive expression of CYP24A1 alters renal function, resulting in albuminuria and marked reduction of plasma 25(OH)D. Overall, these studies are significant and suggest that 1,25(OH)2D is involved in functions other than regulation of vitamin D metabolism. Thus far, there have been no reports on the ablation or overexpression of CYP24A1 in selected tissues, which could shed light on tissue specific roles for CYP24A1.
B. Regulation of Enzyme Expression 1. KIDNEY
The most important physiological regulators of renal CYP24A1 expression in the adult are 1,25(OH)2D and PTH [76–78]. These hormones will therefore influence the production of renal 1,25(OH)2D and hence the level of circulating hormone. Renal CYP24A1 gene transcription is induced by 1,25(OH)2D in a VDR-dependent fashion. This action of 1,25(OH)2D occurs chiefly in the proximal convoluted tubules [79] and will assist in the maintenance of 1,25(OH)2D levels, in particular in the prevention of hypercalcemia. In marked contrast, the peptide hormone PTH functions through its cellular receptor to suppress renal CYP24A1 expression [50] when circulating levels of calcium are low [76–78]. This inhibition occurs through alteration of CYP24A1 mRNA stability and alterations in CYP24A1 gene transcription [62,80,81] (R Serda and JL Omdahl, unpublished). However, the PTH-mediated inhibition of CYP24A1 expression is not observed in the intestine that lacks PTH receptor [50]. PTH-differentially regulates CYP24A1 in proximal and distal convoluted tubular cells with PTH inhibiting 1,25(OH)2D-stimulated expression of CYP24A1 in proximal convoluted tubules, but potentiating the initial induction of CYP24A1 by 1,25(OH)2D in distal convoluted tubular cells [79]. It has been proposed that in distal tubules a coordinated action of PTH and 1,25(OH)2D stimulates calcium reabsorption, but when levels of 1,25(OH)2D are high, PTH stimulates the induction of CYP24A1 by 1,25(OH)2D, thereby lowering cellular 1,25(OH)2D and avoiding excessive calcium uptake by these tubules [79]. In addition to 1,25(OH)2D and PTH, the peptide hormone calcitonin [82–84] is observed to also contribute to regulation of renal CYP24A1 expression. Calcitonin is secreted by the thyroid parafollicular C cells in response to elevated serum calcium levels and is well known to have hypocalcemic actions that result from an inhibition of osteoclastic bone resorption and stimulation of renal calcium excretion [83,85]. Whether these actions contribute significantly to the
92
JOHN OMDAHL AND BRIAN MAY
maintenance of ambient calcium levels in human adults is uncertain [83]. Alternatively, calcitonin could contribute to calcium homeostasis through regulating 1,25(OH)2D levels. Renal CYP27B1 expression in response to calcitonin has been studied. For example, calcitonin induces CYP27B1 expression in kidney LLC-PK1 and MCT proximal tubule cells [86,87], but there is no effect on CYP27B1 levels in the proximal convoluted HKC-8 cells [88]. In distal parts of the nephron, calcitonin stimulates CYP27B1 under normocalcemic conditions [89]. However, the physiological rationale for any induction of CYP27B1 by calcitonin is unclear, since calcitonin is produced in response to high serum calcium. Recent studies have established for the first time that expression of the rat CYP24A1 promoter can be substantially induced by calcitonin in both transiently and stably transfected kidney HEK293 cells (X-P Gao, P Dwivedi, JL Omdahl, H Morris, and B May, submitted). These findings raise the possibility that such a stimulatory action by calcitonin could be important in vivo for modulating renal 1,25(OH)2D production. 2. INTESTINE
Intestinal cells do not express PTH or calcitonin receptors. Circulating 1,25(OH)2D stimulates intestinal calcium import, and the induction of intestinal CYP24A1 by the hormone is likely to be an important factor in determining the level of circulating secosteroid [76]. In this regard, induction of CYP24A1 in the intestine is more responsive to acute changes in 1,25(OH)2D compared with the kidney [90]. 3. BONE
An important role for CYP24A1 in bone development was suggested from the phenotype of CYP24A1 null mice that were found to exhibit abnormal intramembranous bone formation as discussed earlier. This phenotype may result from a local toxic level of 1,25(OH)2D due to the loss of CYP24A1. In earlier studies, Nishimura et al. [91] showed that 1,25(OH)2D induces expression of CYP24A1 in osteoblast cell lines and observed that there is a greater induction by 1,25(OH)2D in the less mature osteoblastic cell lines. This data has been confirmed by the quantification of CYP24A1 mRNA [10] in both rat and human osteoblast-like cell lines (P Anderson, H Morris, and B May, unpublished data). Nonstimulated levels of CYP24A1 mRNA are very low but can be increased dramatically up to 400,000-fold by 1,25(OH)2D. This 1,25(OH2)D–dependent stimulation of CYP24A1 expression appears blunted in fully differentiated, mature cells. Locally produced 1,25(OH)2D, acting in an autocrine/paracrine fashion, is now considered to
have an antiproliferative and prodifferentiation effect on many cells types, both normal and cancerous [92]. It is therefore an attractive hypothesis that 1,25(OH)2D may modulate CYP24A1 expression during osteoblast development so that low levels of 1,25(OH)2D are maintained during the proliferative phase but are increased during the mineralization phase. In support of this concept, the addition of 1,25(OH)2D to proliferative osteoblasts in culture inhibits proliferation but addition to more mature osteoblasts promotes expression of genes involved in differentiation and mineralization [93]. Osteoblasts also express PTH receptor, but there is no evidence that PTH lowers expression of 1,25(OH)2D-induced CYP24A1 as seen in the kidney [94]. In fact, in rat osteoblastic UMR-106 cells, PTH synergises with 1,25(OH)2D to initially enhance CYP24A1 mRNA and protein levels by a mechanism that involves PKA activity [94,95], although mRNA expression is lowered at later time intervals (R Serda and JL Omdahl, unpublished data). It is possible that the early synergistic increase in CYP24A1 in response to PTH and 1,25(OH)2D serves to ensure lowered cellular levels of 1,25(OH)2D commensurate with early osteoblast development. In a manner similar to PTH, the potent glucocorticoid dexamethasone markedly stimulates CYP24A1 mRNA expression and CYP24A1 activity in 1,25(OH)2Dtreated UMR-106 cells [96]. In vivo, glucorticoids markedly affect bone metabolism, albeit in a complex fashion, and an enhanced catabolism of 1,25(OH)2D in the presence of glucorticoid in osteoblasts could influence net bone resorption by counteracting the anabolic effect of 1,25(OH)2D in stimulating mineralization in these cells [96]. 4. MONOCYTES/MACROPHAGES
Immune cell types such as monocytes and macrophages express both CYP27B1 and CYP24A1 with the locally synthesized 1,25(OH)2D exerting its effects on the immune system through both autocrine and paracrine mechanisms [97]. The production of 1,25(OH)2D expression by monocytes or macrophages is markedly enhanced by exposure of cells to the cytokine INF-gamma [98]. This response mimics the autoimmune disease sarcoidosis where there is an abnormally high production of 1,25(OH)2D by diseaseactivated macrophages with associated hypercalcemia. INF-gamma-stimulated production of 1,25(OH)2D results both from up-regulation of CYP27B1 expression [99,100] and from a novel repression of CYP24A1 induction by 1,25(OH)2D that suppresses the ability of 1,25(OH)2D to control its own synthesis. The molecular aspects of this repression have been defined [101] and are discussed later.
93
CHAPTER 6 The 25-Hydroxyvitamin D 24-Hydroxylase
5. KERATINOCYTE
As mentioned, 1,25(OH)2D can influence the differentiation of many cells in culture [92] and keratinocytes are a well studied system in which locally synthesized 1,25(OH)2D acts in an autocrine/paracrine fashion to inhibit cell proliferation and to stimulate terminal keratinocyte differentiation [102,103]. As well as expressing CYP27B1 for local 1,25(OH)2D production, keratinocytes have the capacity to metabolise 1,25(OH)2D through induction of CYP24A1 [45]. At pharmacological levels, 1,25(OH)2D and its analogs can be used in the treatment of hyperproliferative diseases such as psoriasis, but their effectiveness is restricted by expression of CYP24A1, and this has led to a search for selective inhibitors of CYP24A1 [45]. It appears that a constitutive expression of CYP27B1 enzyme in keratinocytes synthesizes local 1,25(OH)2D in which this level of 1,25(OH)2D functions to finely tune CYP24A1. Interestingly, the level of CYP27B1 enzyme activity in keratinocytes markedly declines when CYP24A1 activity is induced by 1,25(OH)2D [45,104]. This inhibition of CYP27B1 enzyme activity is not due to a decrease in CYP27B1 mRNA but rather reflects an increased CYP24A1 directed catabolism of both substrate 25(OH)D and the product 1,25(OH)2D of CYP27B1 [45,104]. This finding that CYP27B1 activity is not directly regulated by 1,25(OH)2D differs from the situation in kidney, where excess 1,25(OH)2D appears to inhibit transcription of the CYP27B1 gene [105,106]. 6. CANCER CELLS
CYP24A1 may be an important player in contributing to the dysregulation of cell growth through lowering the local cellular 1,25(OH)2D concentration. In this regard, CYP24A1 has been suggested as a potential oncogene, since the gene was found coamplified with a known oncogene in human breast cancer tissue [107]. Elevated expression of CYP24A1 mRNA has been noted in highly malignant human colon carcinomas and may be due to amplification of the CYP24A1 gene [108]. A study of different prostate cancer cell lines showed that the growth inhibitory properties of 1,25(OH)2D correlated inversely with the level of CYP24A1 [109]. For example, the human prostate cancer DU-145 cell line expresses high levels of CYP24A1 in response to 1,25(OH)2D, and there is only minimal inhibition of cell proliferation. However, the antiproliferative activity of 1,25(OH)2D on these cells can be significantly restored through inhibition of CYP24A1 activity [110]. The high levels of induction of CYP24A1 observed in the prostate cancer cell lines [109] do not correlate with VDR amounts and the mechanism is not known. A significantly lower prevalence of prostate, mammary, and
colon cancers in Asia compared with Western industrialized countries has been linked to typical Asian diets with high amounts of soy products rich in the phytoestrogens, particularly isoflavonoids, suggesting that these compounds may prevent cancer [111]. Genestein, a major isoflavonoid, has been investigated recently. Administration of genestein or soy protein decreased expression of CYP24A1 mRNA in mouse colon [111] while genestein was also found to be a potent inhibitor of CYP24A1 mRNA expression in DU-145 cells [112].
IV. MOLECULAR ASPECTS In keeping with the widespread tissue distribution of CYP24A1, expression of the gene can be influenced by a broad range of regulatory agents. Molecular studies have focused on the control of CYP24A1 expression by 1,25(OH)2D, PTH, calcitonin, phorbol ester, dexamethasone, and gamma interferon, and their actions are summarized in Fig. 5. Control of CYP24A1 expression by these regulators appears to occur predominantly through modulation of gene transcription, although PTH is a notable exception in which it can alter CYP24A1 mRNA stability.
A. 1,25(OH)2D and the Rat CYP24A1 Promoter 1,25(OH)2D is the most potent inducer of CYP24A1 expression in kidney, osteoblast and other cell types. Investigations into the architecture of the rat CYP24A1 proximal promoter region have permitted a detailed understanding of the 1,25(OH)2D inductive process. Functional transcription factor binding sites have been identified that underlie basal and induced expression (see Figs. 6 and 7). Of central importance are two vitamin D response elements (VDREs) present on the non-coding strand and about 100 bp apart [113–118]. The proximal VDRE-1 at −136/−150 (5′-AGGTGAgtgAGGGCG-3′) and the distal VDRE-2 at −244/−258 (5′-GGTTCAgcgGGTGCG-3′) are DR3-type elements comprising two hexameric half-site direct repeats with a spacing of three nucleotides. Such a tandem arrangement of two VDREs remains unique to the CYP24A1 promoter and underlies transcriptional synergy between the VDREs and the high level of 1,25(OH)2D induction [117]. Electrophoretic mobility shift assays employing specific antibodies have shown that each VDRE binds VDR together with its partner RXR. Earlier experiments revealed than an oligonucleotide encompassing VDRE-1 bound the VDR/RXR heterodimer
94
JOHN OMDAHL AND BRIAN MAY
TPA (Dex) 1,25D
mVDR
calcitonin
CTR
KIDNEY
PKC
mVDR ERKs
PKA/PKC
P P AP1-like, Ets-1
P
RXR, Ets-1 other factors
PKA
Sp1, NF-Y
endonuclease
CYP24 mRNA RXR VDR
RXR VDR
VDRE-2
VDRE-1
PTHR
PTH
Induced CYP24 Gene
CYP24 protein
P mitochondria
Stat-1 VDR
NUCLEUS P Stat-1
IFN-gamma receptor
MONOCYTES/ MACROPHAGES
INF-gamma
FIGURE 5 Transcription regulation model in kidney and monocyte/macrophage cells. The transcription factors on the CYP24A1 promoter are shown for the regulatory agents phorbol ester (TPA), dexamethasone (dex), 1,25(OH)2D acting on either the nuclear VDR (nVDR) at the membrane or a novel membrane-inserted VDR (mVDRm), calcitonin acting on the calcitonin receptor (CTR), PTH acting on the PTH receptor (PTHR), and interferon gamma acting on the interferon gamma receptor. It is proposed that PTH action results in degradation of the CYP24A1 mRNA through the action of an endonuclease.
to a much lesser extent than an oligonucleotide encompassing VDRE-2 [117]. However, mutagenesis of each site within a –298 bp CYP24A1 promoter construct demonstrated that VDRE-1 contributes far more strongly to the 1,25(OH)2D inductive response in all cell types examined including COS-1, kidney HEK-293, UMR106, and ROS17/2.8 cell lines. This suggested that transcription factors bound nearby to VDRE-1 could enhance its activity.
Subsequently, a functional binding site for the ubiquitous transcription factor Ets-1 (Ets binding site, EBS; see Fig. 6) was identified downstream of VDRE-1 at −119/−128 on the noncoding strand and was required for maximal 1,25(OH)2D induction in COS-1 cells [1,119]. The EBS cooperates with VDRE-1 but not VDRE-2, presumably reflecting the close proximity of the VDRE-1 [1]. Direct protein–protein interaction between Ets-1 and VDR has been demonstrated by the
RNAP II machinery −298
RXR VDR
VDRE-2
RXR VDR
AP-1 like VDRE-1 EBS
GC
CCAAT
FIGURE 6 Cis-binding elements in CYP24A1 promoter. Functional binding sites for transcription factors within the first –298 bp of the rat CYP24A1 promoter. The arrows indicate that the sequences for VDRE-1, VDRE-2, and EBS are on the noncoding strand.
95
CHAPTER 6 The 25-Hydroxyvitamin D 24-Hydroxylase
two-hybrid assay [1]. As shown in Fig. 7B, Ets-1 bound at the EBS probably interacts with liganded VDR/RXR on VDRE-1 [1] and also with coactivator complexes [78] bound to liganded VDR/RXR. The EBS does not contribute to the low basal CYP24A1 promoter expression observed in transfected cells in the absence of added 1,25(OH)2D [119]. A repressor complex bound to unliganded VDR/RXR on the CYP24A1 promoter maintains this low expression in the absence of secosteroid [120]. The presence of this repressor complex may sterically prevent the binding of Ets-1 (Fig. 7A). When 1,25(OH)2D binds to VDR, it is proposed that the repressor complex dissociates and is replaced first by a coactivator complex with histone acetyltransferase activity that remodels
A
Basal
AP-1
Ets-1
X
X
chromatin [78,121] and then by a mediator complex, as described later. Since the EBS does not contribute to CYP24A1 promoter-directed basal expression detected in cells, it was clear that other transcription factor binding sites must be present. A GC box at −101/−114 (5′-CCCGCCCC-3′) immediately downstream of EBS and an inverted CCAAT box further downstream at –51/–62 (5′CATTGGC-3′) are functionally active (see Fig. 5; X-H Gao, P Dwivedi, and B May, unpublished data). Supershift analysis employing isolated nuclear extracts suggest that the ubiquitous factors Sp1 and NF-Y [122] bind to the GC and CCAAT box, respectively, with no binding detected of C/EBP alpha or beta to the CCAAT motif. Mutagenesis experiments established
HDAC? Repressor complex RXR VDR
VDRE-2
B
RNAP II machinery
Repressor complex RXR VDR
AP-1 like VDRE-1 EBS
weak GC
CCAAT
1,25D-induction
HAT
HAT CBP/p300
CBP/p300
p160
p160
RXR VDR
RXR VDR
VDRE-2
AP-1 like VDRE-1 EBS
RNAP II machinery
GC
CCAAT
C
RNAP II machinery
mediator complex RXR VDR
VDRE-2
RXR VDR
AP-1 like VDRE-1 EBS
Strong GC
CCAAT
FIGURE 7 Proposed binding of regulatory trans-proteins. (A) In the absence of 1,25(OH)2D, it is proposed that a repressor complex binds a VDR/RXR heterodimer on each VDRE. This complex may repress through modifying chromatin via histone deacetylase activity (HDAC) and/or preventing access to the AP-1 like binding protein and Ets-1. Low basal expression is driven by Sp1 and NF-Y bound at the GC and CCAAT box sites, respectively, through interaction with the RNA polymerase II machinery. (B) Following ligand binding to VDR, it is proposed that coactivators p160 and CBP/p300 are recruited and through histone acetyl transferase activity (HAT) modify chromatin and enhance binding of AP-1 like protein and Ets-1. Synergy may result from an interaction of all transcription factors with CBP except NF-Y and from interaction of the two coactivator complexes. (C) The two coactivator complexes are proposed to be replaced with a coactivator mediator complex that interacts with and activates RNA polymerase II and hence stimulates CYP24A1 gene transcription.
96
JOHN OMDAHL AND BRIAN MAY
(−) 1,25 (OH)2D + 1,25 (OH)2D UMR-106 cells 60x 10.00
36x
5.00
33x
38x 24x
6x
1.00 0.50
Mutated AP-1 like
Mutated EBS
Mutated GC + CCAAT
Mutated CCAAT
Mutated GC
WTpCYP24-298-luc
Mutated AP-1 like
Mutated EBS
Mutated GC + CCAAT
Mutated CCAAT
0
Mutated GC
0.25
WT pCYP24-298-luc
Relative Luciferase Activity
15.00
FIGURE 8
Mutational analysis of CYP24A1 promoter during induction of 1,25(OH)2D. Expression of transiently transfected wild type and mutant –298 bp CYP24A1 promoter constructs in UMR-106 cells. Cells were treated with 1,25(OH)2D at 10−7M. The relative luciferase activity represents the mean ±SD firefly luciferase/Renilla luciferase ratios of triplicate samples.
that the GC and CCAAT boxes predominantly are responsible for basal expression of the CYP24A1 promoter in the HEK-293 and UMR-106 cell lines examined. In UMR-106 cells, basal expression is substantially lowered when the CCAAT and GC boxes are mutated either individually or together (Fig. 8). A similar contribution from each site to basal expression is observed in stably transfected HEK-293T cells (Fig. 9). Sp1 and NF-Y could drive basal expression through interactions with the general transcription machinery (see Fig. 7A), notably with the general transcription factor TFIID bound at the TATA box of the CYP24A1 promoter. The GC and CCAAT box elements are important for induction of the CYP24A1 promoter by 1,25(OH)2D. In UMR-106 cells, when the GC and CCAAT box sites are both inactivated, 1,25(OH)2D-induced CYP24A1 promoter activity is reduced from about 60-fold to 24-fold (Fig. 9). The GC and CCAAT box sites are not responsive per se to 1,25(OH)2D, since mutagenesis of both VDREs in the CYP24A1 promoter abolishes induction by 1,25(OH)2D. In UMR-106 cells, mutagenesis of EBS has no effect on basal expression but lowers 1,25(OH)2D induction from 60-fold to 36-fold (Fig. 8). Another sequence further upstream of VDRE-1 at –163/−171, 5′-TGTCGGTCA-3′, has recently been characterized through site-directed mutagenesis of the
CYP24A1 promoter and shown to markedly stimulate 1,25(OH)2D induction, particularly in osteoblastic cell lines (P Dwivedi, J Kaplan, JL Omdahl, H Morris, and B May, in preparation). This site resembles an AP-1 site (consensus 5′-TGAG/CTCA-3′) and is referred to here as AP-1 like (as shown in Figs. 6 and 7). In UMR106 cells, inactivation of this sequence markedly lowers 1,25(OH)2D induction of the CYP24A1 promoter from 60-fold to six-fold (Fig. 8), and further analysis has established that the site cooperates with VDRE-1 and not VDRE-2. The AP-1 like sequence does not contribute to basal CYP24A1 promoter expression, perhaps because of the presence of the repressor complex bound on VDRE-1 (Figs. 8 and 7A). Within the AP-1 like motif, the sequence 5′-CGGTCA-3′ resembles a VDRE halfsite and has attracted attention previously when it was shown that mutagenesis of this sequence lowered 1,25(OH)2D induction [114,118]. Overall, the rat CYP24A1 proximal promoter (see Fig. 6) comprises a weak VDRE-1 encircled by compensatory transcription factors and a strong upstream VDRE-2 apparently devoid of nearby transcription factors. A model for the nuclear genomic action of 1,25(OH)2D that leads to stimulation of transcription of the CYP24A1 promoter can be proposed (Fig. 7). When 1,25(OH)2D binds to VDR located at each VDRE, the
97
CHAPTER 6 The 25-Hydroxyvitamin D 24-Hydroxylase
−calcitonin
50.00
+calcitonin
Luciferase Activity / ug Protein
14x
45.00
2.8x
5.00
1.2x
Mutated GC
Mutated CCAAT
WTpCYP24-298-luc
Mutated GC
Mutated CCAAT
WT pCYP24-298-luc
0
FIGURE 9 Mutational analysis of CYP24A1 promoter during induction by calcitonin. Expression of wild-type and mutant –298 bp CYP24A1 promoter constructs in stably transfected HEK-293T cells expressing the calcitonin receptor. Cells were treated overnight with calcitonin (10 nM). The fire fly luciferase activity was normalised to the amount of cellular protein and the values represent the mean ± SD of triplicate samples.
repressor complex is dissociated and a p160 coactivator is recruited on each VDRE, together with other coactivators such as CBP/p300 and SKIP/NCoA-62 [78,121], so that cooperative interactions between the coactivator complexes result in transcriptional synergy. At VDRE-1, it is envisaged that recruitment of a coactivator complex by liganded VDR/RXR is enhanced by the AP-1 like protein, by Ets-1, and also by Sp1. The coactivator complex through its intrinsic histone acetyl transferase activity could modify chromatin permitting access to factors and facilitating assembly of the transcription machinery. Finally, in a proposed two-step mechanism, the coactivator complex would be replaced by the multiprotein mediator complex DRIP (VDR interacting proteins) that interacts with and activates RNAPII [78]. There is evidence that the induction of CYP24A1 mRNA by 1,25(OH)2D can be blocked by the protein synthesis inhibitor cycloheximide, implicating the obligatory production of a regulatory protein [81,123]. This protein is apparently not VDR or RXR, suggesting that for induction to proceed optimally, synthesis of a transcription factor or a coactivator component or a protein that stabilizes CYP24A1 mRNA is required [81].
B. 1,25(OH)2D and Mitogen-Activated Protein Kinase Activities In addition to its classical genomic action in the nucleus, it is now clear that 1,25(OH)2D also functions at the plasma membrane to rapidly activate downstream signaling pathways critical for transcription of the CYP24A1 promoter [1,119]. Such a nongenomic action is also a feature of steroid hormones, which is currently receiving considerable attention [124]. The importance of phosphorylation events in the induction of the CYP24A1 promoter by 1,25(OH)2D was first suggested by studies in COS-1 cells cotransfected with VDR, where overexpression of H-Ras increased the inductive response and also potentiated the effect of exogenously expressed wild-type Ets-1 acting through the EBS [119]. This response agreed with the proposal that the functionality of Ets-1 was dependent on phosphorylation of threonine residue 38. A mutant form of Ets-1 where threonine residue 38 was mutated to alanine lowered the level of induction of the CYP24A1 promoter to that observed with mutated EBS [1]. Also, a dominant negative Ras mutant inhibited 1,25(OH)2D induction to a level greater than that observed following
98 EBS inactivation. This demonstrated that the action of Ras on the CYP24A1 promoter was likely to be channeled not only through Ets-1 but also through other components on the promoter. Since MAP kinase activities are downstream effectors of Ras, a detailed investigation was undertaken into the possible role of these kinases in the activation of the rat CYP24A1 promoter by 1,25(OH)2D in VDRtransfected COS-1 cells [1]. Rapid stimulation of ERK1/2 and ERK5 activities by 1,25(OH)2D occurred in these cells with maximal activities peaking at 5–10 min and inhibition of activity directed by a Ras dominant negative clone. These data raised the possibility that ERK1/2 and ERK5 could be involved in the regulation of the CYP24A1 promoter. Studies with specific pharmacological inhibitors and dominant negative mutants established that in 1,25(OH)2D treated COS-1 cells, inhibition of the ERK1/2 module markedly lowered induction of the CYP24A1 promoter while the p38 and JNK modules were not required [1]. Stimulation of ERK5 activity by 1,25(OH)2D was required for maximal induction in which the secosteroid’s action was effected through phosphorylation of Ets-1. Activated ERK1 phosphorylated RXRalpha at serine 260, and when COS-1 cells were transfected with the mutant RXRalphaS260A clone (in which serine was replaced by alanine) induction of the CYP24A1 promoter was inhibited. VDR was not phosphorylated by either kinase activity. These studies established a critical role in these cells for the Ras-ERK1/2-RXRalpha signaling pathway for induction of the CYP24A1 promoter with an enhancing contribution from the Ras-ERK5 pathway through phosphorylation of Ets-1. An analysis of the signaling pathways that contribute to 1,25(OH)2D induction of the CYP24A1 promoter using specific pharmacological inhibitors and dominant negative clones has revealed that the contribution of ERK modules and other signaling molecules is dependent on the cell type. For example, in HEK-293 cells 1,25(OH)2D induction is dependent on JNK activity and there is little activation of ERK1/2. The contribution to promoter induction of Src and PI3Kinase activities also is cell type dependent. PKC (see Section III,B,4) is important for 1,25(OH)2D induction in most cell types examined. Overall these studies show that the nongenomic signaling pathways activated by 1,25(OH)2D are vital for CYP24A1 promoter activity with PKC/Ras/ ERK modules of particular importance. The precise role of the signaling pathways and the downstream targets on the CYP24A1 promoter (i.e. AP-1-like sequence, VDR/RXR, Ets-1, Sp1 and CCAAT-box binding protein) in the different cell types will require an in-depth analysis. As discussed previously [1], it is interesting that whereas overexpressed Ras facilitates
JOHN OMDAHL AND BRIAN MAY
1,25(OH)2D induction of the CYP24A1 promoter activity in COS-1 cells through phosphorylation of serine 160 in RXR, in Ras-transformed keratinocytes 1,25(OH)2D-dependent phosphorylation of serine 160 in RXR attenuates 1,25(OH)2D transactivation [125]. The molecular basis for this difference in the action of phosphorylated RXR is not known. As discussed earlier, the inhibitory action of genestein on CYP24A1 mRNA expression supports the proposal that soy consumption may protect against cancer through elevating local 1,25(OH)2D production. Genestein is a well-known tyrosine kinase inhibitor and the observed inhibitory effect on CYP24A1 mRNA expression is likely to be at the nongenomic level.
C. 1,25(OH)2D and the Nongenomic Response The rapid activations of ERK1/2 and ERK5 and other signaling molecules by 1,25(OH)2D, as a prelude to CYP24A1 promoter stimulation, are consistent with the nongenomic actions reported for steroid hormones [124]. Studies of the estrogen receptor (ER) provide a model in which ER is recruited to the plasma membrane by interaction with caveolin proteins that serve as scaffolds in caveolae, discrete domains of the plasma membrane [126,127]. Signaling molecules including various G proteins, Src, PI3Kinase, and Raf are also recruited to these scaffolds so that the ER in response to ligand can rapidly modulate a variety of signaling pathways within a confined space. There is evidence for the ER interacting directly with Src, and PI3K and stimulating PKC and PKA activities through various G proteins. Studies on the mechanism underlying the nongenomic action of 1,25(OH)2D are less extensive than those of estrogen/ER pathway. There is evidence that nongenomic actions of 1,25(OH)2D include G-protein stimulation of adenylate cyclase, phospholipase C beta [77] and phospholipase C gamma, the last through the tyrosine kinases Src and PI3K [128]. By analogy with ER, it can be envisaged that the classical VDR is recruited to the plasma membrane where it activates signaling cascades in response to 1,25(OH)2D. Indeed, nongenomic activities are lost from osteoblast cultures derived from homozygous mice expressing a mutant VDR lacking the first zinc finger necessary for DNA binding [129]. Presumably conformational changes in the mutant VDR interfere with membrane localization or binding to the downstream signaling molecules (i.e., function as dominant-negative VDR). On the other hand, it is possible that a unique VDR may participate in nongenomic signaling. A novel progesterone receptor has been identified with seven putative transmembrane domains, thus resembling the classic G protein coupled
99
CHAPTER 6 The 25-Hydroxyvitamin D 24-Hydroxylase
receptors [130]. This raises the possibility that a novel membrane form of VDR, perhaps together with the classical VDR acting at the plasma membrane, contribute to nongenomic biological responses. In that regard, it is interesting that PKC activation by 1,25(OH)2D is still observed in osteoblast cultures isolated from VDR null mice where the second zinc finger of the DNA binding domain has been deleted [131,132]. Although these authors suggest that the nongenomic response is independent of the nuclear VDR activity, it is still possible that the absence of the second zinc finger alters downstream interactions necessary for PKC activation. Other evidence has shown that the ligand specificity for nongenomic activities in osteoblasts is distinct from that for nuclear gene activation [133].
lowered synergy from 7.1-fold to 3.3-fold and 1.1-fold, respectively. Hence both sites play a role, but the AP-1 like sequence plays a predominant role in TPA sensitivity. It can be proposed (see Fig. 5) that activated PKC in response to TPA further stimulates the activity of MAP kinase signaling pathways above that achieved by 1,25(OH)2D alone, leading to the phosphorylation and activation of AP-1 like protein and Ets-1 and overall enhancement of 1,25(OH)2D induction. Whether the nonresponsiveness of the CYP24A1 promoter to phorbol ester action in osteoblasts mirrors the lack of activation of MAP kinase pathways by TPA remains to be determined.
E. Glucocorticoid and Protein Kinase C D. Phorbol Ester and Protein Kinase C The stimulation of CYP24A1 expression by phorbol ester treatment and the involvement of PKC activity has been of interest for many years [76]. The synergistic action of phorbol ester and 1,25(OH)2D appears to be cell type specific. In mouse renal tubules, the phorbol ester 12-O-tetradecanoylphorbol-13-acetate (TPA), also referred to as PMA, increases CYP24A1 activity in a PKC-dependent fashion [134]. TPA also markedly increased the level of CYP24A1 mRNA in 1,25(OH)2Dtreated rat renal cells [15,135] and in rat intestinal epithelial cells [136,137]. However, synergy is not observed in osteoblast cells. TPA does not affect either CYP24A1 mRNA levels or CYP24A1 promoter activity in the presence or absence of 1,25(OH)2D [94,137,138] in rat osteoblastic UMR-106 cells, although some TPA stimulation of CYP24A1 mRNA in these cells in the presence of 1,25(OH)2D has been reported at earlier times after treatment [96]. Synergy also is not observed in rat osteoblastic ROS17/2.8 cells transfected with rat –298 bp CYP24A1 promoter constructs (P Dwivedi and B May, unpublished). The molecular mechanism by which TPA stimulates 1,25(OH)2D-induced CYP24A1 expression has attracted interest but has remained elusive [76]. Since it is well known that TPA can stimulate the activity of the transcription factor AP-1 and more recently Ets proteins [139], studies have been undertaken to determine the possible roles of the EBS and AP-1 like sequences in the CYP24A1 promoter in kidney 293T cells (P Dwivedi, B Nutschey, and B May, submitted). TPA alone gives minimal induction of the CYP24A1 promoter, but in the presence of 1,25(OH)2D there is marked synergy (about 7.1-fold) and this was localized to the first –186 bp of promoter sequence. Within this region, mutagenesis individually of the AP-1 like sequence and EBS
Dexamethasone markedly enhances 1,25(OH)2D induction of CYP24A1 mRNA in UMR-106 cells and to a lesser extent in renal LLC-PK1 cells with the effect of dexamethasone being blocked by a PKC inhibitor [96]. In this study, dexamethasone markedly increased c-fos mRNA and protein but only slightly induced the abundance of c-jun mRNA in UMR-106 cells. It was proposed that dexamethasone enhances CYP24A1 mRNA expression in osteoblast cells via activation of AP-1 in a process that requires PKC activity. It would be relevant to determine if dexamethasone acts through the AP-1 like sequence identified in the CYP24A1 promoter.
F. Parathyroid Hormone In recent studies employing opossum kidney (OK) cells that express both endogenous VDR and PTH receptors, Armbrecht et al. [140] showed that PTH (acting through cAMP) moderately increases expression of a reporter construct containing −580 bp of rat CYP24A1 promoter. The mechanism by which this cAMP-mediated increase in transcription occurs is not known but may involve increased VDR [138]. As expected, 1,25(OH)2D also stimulated promoter expression. The apparent paradox that PTH stimulates renal CYP24A1 promoter expression, yet substantially lowers CYP24A1 activity in vivo, has been resolved [62,80,81]. It has been demonstrated that PTH markedly destabilizes the CYP24A1 mRNA, reducing the half-life by fourfold as determined in AOK-B50 cells [81]. The precise mechanism underlying this cAMP-dependent degradation is not known but could involve PKA-directed phosphorylation of a CYP24A1 mRNA binding protein whose loss permits degradation by a specific endonuclease (Fig. 5). Interestingly, while the 3′ untranslated region is a common target for regulating mRNA stability, the studies
100 of Zierold et al. [80] showed that neither the 3′ nor the 5′ untranslated regions of the CYP24A1 mRNA contain the instability element, suggesting that such a site is located in the coding region. Alteration of mRNA stability is considered a rapid response mechanism for modulating gene expression. Hence in the hypocalcemic state, PTH production by the parathyroid gland will result in destabilization of renal CYP24A1 mRNA with a rapid increase in the production of 1,25(OH)2D by the kidney through decreased removal of 25(OH)D and 1,25(OH)2D. In contrast to this, it is accepted that the major effect of PTH on renal CYP27B1 expression is through increased promoter activity [140,141].
G. Calcitonin As discussed earlier, recent data have shown that calcitonin substantially increases expression of a rat CYP24A1 –298 bp promoter construct in both transiently and stably transfected kidney HEK-293 cells expressing the calcitonin receptor. This induction is strongly dependent on the GC and CCAAT box sites as demonstrated in stably transfected HEK-293 cells (Fig. 9) in which calcitonin induces the transfected promoter 14-fold. However, this induction is reduced to 2.8-fold and 1.2-fold when the CCAAT box and GC boxes are mutated, respectively. Pharmacological inhibitor studies indicated the involvement of both PKA and PKC activities in calcitonin stimulation but not basal expression. Although calcitonin is known to activate ERK1 and ERK2 in these cells [142], these ERK activities do not participate in calcitonin induction, since the ERK inhibitor PD98059 and the dominant negative clone ERK1K71R did not affect calcitonin induction. The precise roles of PKA and PKC remain to be determined, but it seems likely that they act through Sp1 and NF-Y (Fig. 5). On the basis of this new data, it is tempting to suggest that calcitonin in vivo contributes to the control of circulating calcium levels through regulating 1,25(OH)2D. Such an action of calcitonin could be important in the hypercalcemic state when a transient increase in serum calcium triggers the production of calcitonin. In this regard, rats consuming a high-calcium diet markedly increase renal CYP24A1 and VDR mRNA levels, although the hormone regulation is unclear [10].
H. Interferon (INF)-Gamma Increased production of 1,25(OH)2D by INF-gamma activated monocytes/macrophages is associated with a marked inhibition of CYP24A1 up-regulation
JOHN OMDAHL AND BRIAN MAY
by 1,25(OH)2D. A novel mechanism for this antagonistic action by INF-gamma has been elucidated [101]. Basal expression of a human 24-hydroxylase promoter construct encompassing the two VDREs was not affected by INF-gamma, but 1,25(OH)2D-stimulated expression was inhibited. These workers established a sequence of events where, following the binding of INF-gamma to its receptor on the cell surface, the transcription factor Stat1 is activated by phosphorylation and in its homodimeric phosphorylated form enters the nucleus and binds liganded VDR (Fig. 5). This interaction abrogates the binding of liganded VDR/RXR to the VDREs in the CYP24A1 promoter, thereby suppressing 1,25(OH)2D activation. It is interesting that INFgamma stimulated CYP27B1 gene expression is not inhibited by high levels of 1,25(OH)2D in these cells [99], although studies suggesting a negative regulation of CYP27B1 gene expression by 1,25(OH)2D have been reported in the kidney [143]. Perhaps a genomic-acting inhibitory mechanism exists only in kidney cells.
Acknowledgments The authors recognize the expert assistance of Stephanie Finlayson, Andrezej Pastuszyn, Rita Serda, and Andrew Annolara, who greatly facilitated the preparation and composition of this chapter. Also available were discussions with Howard Morris, Prem Dwivedi, and Satya Reddy on subject matter contained in this chapter.
References 1. Dwivedi PP, Hii CS, Ferrante A, et al. 2002 Role of MAP kinases in the 1,25-dihydroxyvitamin D3-induced transactivation of the rat cytochrome P450C24 (CYP24) promoter. Specific function for Erk1/Erk2 and Erk5. J Biol Chem 277:29643–29653. 2. Nemere I, Pietras RJ, Blackmore PF 2003 Membrane receptors for steroid hormones: Signal transduction and physiological significance. J Cell Biochem 88:438–445. 3. Farach-Carson MC, Nemere I 2003 Membrane receptors for vitamin D steroid hormones: potential new drug targets. Curr Drug Targets 4:67–76. 4. Pike JW 1991 Vitamin D3 receptors: structure and function in transcription. Annu Rev Nutr 11:189–216. 5. Hannah SS, Norman AW 1994 1 Alpha,25(OH)2 vitamin D3regulated expression of the eukaryotic genome. Nutr. Rev 52:376–382. 6. Matkovits T, Christakos S 1995 Variable in vivo regulation of rat vitamin D-dependent genes (osteopontin, Ca,Mg-adenosine triphosphatase, and 25-hydroxyvitamin D3 24-hydroxylase): implications for differing mechanisms of regulation and involvement of multiple factors. Endocrinology 136: 3971–3982.
101
CHAPTER 6 The 25-Hydroxyvitamin D 24-Hydroxylase
7. Armbrecht HJ, Boltz MA 1991 Expression of 25-hydroxyvitamin D 24-hydroxylase cytochrome P450 in kidney and intestine. Effect of 1,25-dihydroxyvitamin D and age. FEBS Lett 292:17–20. 8. Shinki T, Jin CH, Nishimura A, et al. 1992 Parathyroid hormone inhibits 25-hydroxyvitamin D3-24-hydroxylase mRNA expression stimulated by 1 alpha,25-dihydroxyvitamin D3 in rat kidney but not in intestine. J Biol Chem 267:13757–13762. 9. Zehnder D, Bland R, Williams MC, et al. 2001 Extrarenal expression of 25-hydroxyvitamin D(3)-1 alpha-hydroxylase. J Clin Endocrinol Metab 86:888–894. 10. Anderson PH, O’Loughlin PD, May BK, Morris HA 2003 Quantification of mRNA for the vitamin D metabolising enzymes CYP27B1 and CYP24 and VDR in kidney using real-time reverse transcriptase-polymerase chain reaction. J Mol Endocrinol (in press). 11. Pedersen JI, Hagenfeldt Y, Bjorkhem I 1998 Assay and properties of 25-hydroxyvitamin D3 23-hydroxylase. Evidence that 23,25-dihydroxyvitamin D3 is a major metabolite in 1,25-dihydroxyvitamin D3-treated or fasted guinea pigs. Biochem J 250:527–532. 12. Kumar R, Schaefer J, Grande JP, Roche PC 1994 Immunolocalization of calcitriol receptor, 24-hydroxylase cytochrome P-450, and calbindin D28k in human kidney. Am J Physiol 266:F477–F485. 13. Iwata K, Yamamoto A, Satoh S, et al. 1995 Quantitative immunoelectron microscopic analysis of the localization and induction of 25-hydroxyvitamin D3 24-hydroxylase in rat kidney. J Histochem Cytochem 43:255–262. 14. Henry HL, Dutta C, Cunningham N, et al. 1992 The cellular and molecular regulation of 1,25(OH)2D3 production. J Steroid Biochem Mol Biol 41:401–407. 15. Chen ML, Boltz MA, Armbrecht HJ 1993 Effects of 1,25dihydroxyvitamin D3 and phorbol ester on 25-hydroxyvitamin D3 24-hydroxylase cytochrome P450 messenger ribonucleic acid levels in primary cultures of rat renal cells. Endocrinology 132:1782–1788. 16. Matsumoto T, Kawanobe Y, Ogata E 1985 Regulation of 24,25-dihydroxyvitamin D-3 production by 1,25-dihydroxyvitamin D-3 and synthetic human parathyroid hormone fragment 1-34 in a cloned monkey kidney cell line (JTC-12). Biochim Biophys Acta 845:358–365. 17. Tomon M, Tenenhouse HS, Jones G 1990 Expression of 25-hydroxyvitamin D3-24-hydroxylase activity in Caco-2 cells. An in vitro model of intestinal vitamin D catabolism. Endocrinology 126:2868–2875. 18. Kane KF, Langman MJ, Williams GR 1996 Antiproliferative responses to two human colon cancer cell lines to vitamin D3 are differently modified by 9-cis-retinoic acid. Cancer Res 56:623–632. 19. Inaba M, Burgos-Trinidad M, DeLuca HF 1991 Characteristics of the 25-hydroxyvitamin D3- and 1,25-dihydroxyvitamin D3-24-hydroxylase(s) from HL-60 cells. Arch Biochem Biophys 284:257–263. 20. Skowronski RJ, Peehl DM, Feldman D 1995 Actions of vitamin D3, analogs on human prostate cancer cell lines: comparison with 1,25-dihydroxyvitamin D3. Endocrinology 136:20–26. 21. Makin G, Lohnes D, Byford V, Ray R, Jones G 1989 Target cell metabolism of 1,25-dihydroxyvitamin D3 to calcitroic acid. Evidence for a pathway in kidney and bone involving 24-oxidation. Biochem J 262:173–180. 22. Eil C, Liberman UA, Marx SJ 1986 The molecular basis for resistance to 1,25-dihydroxyvitamin D: studies in cells
23.
24. 25. 26.
27. 28. 29.
30.
31.
32.
33. 34.
35.
36.
37.
38. 39.
cultured from patients with hereditary hypocalcemic 1,25(OH)2D3-resistant rickets. Adv Exp Med Biol 196: 407–422. Morgan JW, Reddy GS, Uskokovic MR, et al. 1994 Functional block for 1 alpha,25-dihydroxyvitamin D3-mediated gene regulation in human B lymphocytes. J Biol Chem 269:13437–13443. Naveilhan P, Neveu I, Baudet C, et al. 1993 Expression of 25(OH) vitamin D3 24-hydroxylase gene in glial cells. NeuroReport 5:255–257. Chen ML, Heinrich G, Ohyama YI, et al. 1994 Expression of 25-hydroxyvitamin D3-24-hydroxylase mRNA in cultured human keratinocytes. Proc Soc Exp Biol Med 207:57–61. Holick MF, Schnoes HK, DeLuca HF, et al. 1972 Isolation and identification of 24,25-dihydroxycholecalciferol, a metabolite of vitamin D made in the kidney. Biochemistry 11:4251–4255. Holick MF, Kleiner-Bossaller A, Schnoes HK, et al. 1973 1,24,25-Trihydroxyvitamin D3. A metabolite of vitamin D3 effective on intestine. J Biol Chem 248:6691–6696. Beckman MJ, Tadikonda P, Werner E, et al. 1996 Human 25-hydroxyvitamin D3-24-hydroxylase, a multicatalytic enzyme. Biochemistry 35:8465–8472. Ohyama Y, Okuda K 1991 Isolation and characterization of a cytochrome P-450 from rat kidney mitochondria that catalyzes the 24-hydroxylation of 25-hydroxyvitamin D3. J Biol Chem 266:8690–8695. Lund EG, Guileyardo JM, Russell DW 1999 cDNA cloning of cholesterol 24-hydroxylase, a mediator of cholesterol homeostasis in the brain. Proc Natl Acad Sci USA 96: 7238–7243. Gray RW, Omdahl JL, Ghazarian JG, Horst RL 1990 Induction of 25-OH-vitamin D3 24- and 23-hydroxylase activities in partially purified renal extracts from pigs given exogenous 1,25-(OH)2D3. Steroids 55:395–398. Pikuleva IA, Cao C, Waterman MR 1999 An additional electrostatic interaction between adrenodoxin and P450c27 (CYP27A1) results in tighter binding than between adrenodoxin and p450scc (CYP11A1). J Biol Chem 274:2045–2052. Reddy GS, Tserng KY 1989 Calcitroic acid, end product of renal metabolism of 1,25-dihydroxyvitamin D3 through C-24 oxidation pathway. Biochemistry 28:1763–1769. Esvelt RP, Schnoes HK, DeLuca HF 1979 Isolation and characterization of 1 alpha-hydroxy-23-carboxytetranorvitamin D: a major metabolite of 1,25-dihydroxyvitamin D3. Biochemistry 18:3977–3983. Horst RL, Reinhardt TA, Ramberg CF, Koszewski NJ, Napoli JL 1986 24-Hydroxylation of 1,25-dihydroxyergocalciferol. An unambiguous deactivation process. J Biol Chem 261:9250–9256. Akiyoshi-Shibata M, Sakaki T, Ohyama Y, et al. 1994 Further oxidation of hydroxycalcidiol by calcidiol 24-hydroxylase. A study with the mature enzyme expressed in Escherichia coli. Eur J Biochem 224:335–343. Sakaki T, Sawada N, Nonaka Y, Ohyama Y, Inouye K 1999 Metabolic studies using recombinant Escherichia coli cells producing rat mitochondrial CYP24. CYP24 can convert 1α,25-dihydroxyvitamin D3 to calcitroic acid. Eur J Biochem 262:43–48. Sakaki T, Sawada N, Komai K, et al. 2000 Dual metabolic pathway of 25-hydroxyvitamin D3 catalyzed by human CYP24. Eur J Biochem 267:6158–6165. Peterson JA, Graham SE 1998 A close family resemblance: The importance of structure in understanding cytochromes P450. Structure 6:1079–1085.
102 40. Poulos TL, Finzel BC, Howard AJ 1987 High-resolution crystal structure of cytochrome P450cam. J Mol Biol 195:687–700. 41. Ravichandran KG, Boddupalli SS, Hasermann CA, Peterson JA, Deisenhofer J 1993 Crystal structure of hemoprotein domain of P450BM-3, a prototype for microsomal P450’s. Science 261:731–736. 42. Hasemann CA, Ravichandran KG, Peterson JA, Deisenhofer J 1994 Crystal structure and refinement of cytochrome P450terp at 2.3 A resolution. J Mol Biol 236:1169–1185. 43. Brown AJ, Finch JL, Lopez-Hilker S, et al. 1990 New active analogues of vitamin D with low calcemic activity. Kidney Int Suppl 29:S22–S27. 44. Wester MR, Johnson EF, Marques-Soares C, et al. 2003 Structure of a substrate complex of mammalian cytochrome P450 2C5 at 2.3 Å resolution: Evidence for multiple substrate binding modes. Biochemistry 42:6370–6379. 45. Schuster I, Egger H, Astecker N, et al. 2001 Selective inhibitors of CYP24: Mechanistic tools to explore vitamin D metabolism in human keratinocytes. Steroids 66:451–462. 46. Schuster I, Egger H, Bikle D, et al. 2001 Selective inhibition of vitamin D hydroxylases in human keratinocytes. Steroids 66:409–422. 47. Peehl DM, Seto E, Feldman D 2001 Rationale for combination ketoconazole/vitamin D treatment of prostate cancer. Urology 58:123–126. 48. Bayanov AA, Brunt AR 1999 Role of hypoxia and constitutionally different resistance to hypoxia/stress as the determiners of individual profile of cytochrome P450 isozyme activity. Gen Pharmacol 33:355–361. 49. Tenenhouse HS, Yip A, Jones G 1988 Increased renal catabolism of 1,25-dihydroxyvitamin D3 in murine X-linked hypophosphatemic rickets. J Clin Invest 81:461–465. 50. Shinki T, Jin CH, Nishimura A, et al. 1992 Parathyroid hormone inhibits 25-hydroxyvitamin D3-24-hydroxylase mRNA expression stimulated by 1 alpha,25-dihydroxyvitamin D3 in rat kidney but not in intestine. J Biol Chem 267:13757–13762. 51. Burgos-Trinidad M, DeLuca HF 1991 Kinetic properties of 25-hydroxyvitamin D- and 1,25-dihydroxyvitamin D24-hydroxylase from chick kidney. Biochim Biophys Acta 1078:226–230. 52. Taniguchi T, Eto TA, Shiotsuki H, et al. 2001 Newly established assay method for 25-hydroxyvitamin D3 24hydroxylase revealed much lower Km for 25-hydroxyvitamin D3 than for 1α,25-dihydroxyvitamin D3. J Bone Miner Res 16:57–62. 53. Pikuleva IA, Bjorkhem I, Waterman MR 1997 Expression, purification, and enzymatic properties of recombinant human cytochrome P450c27 (CYP27). Arch Biochem Biophys 343:123–130. 54. Pikuleva IA, Puchkaev A, Bjorkhem I 2001 Putative helix F contributes to regioselectivity of hydroxylation in mitochondrial cytochrome P450 27A1. Biochemistry 40: 7621–7629. 55. Schenkman JB, Remmer H, Estabrook RW 1967 Spectral studies of drug interaction with hepatic microsomal cytochrome. Mol Pharmacol 3:113–123. 56. Horst RL, Omdahl JA, Reddy S 2003 Rat cytochrome P450C24 (CYP24) does not metabolize 1,25-dihydroxyvitamin D2 to calcitroic acid. J Cell Biochem 88:282–285. 57. Omdahl JL, Bobrovnikova EV, Annalora A, Chen P, Serda R 2003 Expression, structure–function, and molecular modeling of vitamin D P450s. J Cell Biochem 88:356–362. 58. Ohyama Y, Noshiro M, Okuda K 1991 Cloning and expression of cDNA encoding 25-hydroxyvitamin D3 24-hydroxylase. FEBS Lett 278:195–198.
JOHN OMDAHL AND BRIAN MAY
59. Chen KS, Prahl JM, DeLuca HF 1993 Isolation and expression of human 1,25-dihydroxyvitamin D3 24-hydroxylase cDNA. Proc Natl Acad Sci USA 90:4543–4547. 60. Jehan F, Ismail R, Hanson K, DeLuca HF 1998 Cloning and expression of the chicken 25-hydroxyvitamin D3 24-hydroxylase cDNA. Biochim Biophys Acta 1395:259–265. 61. Akeno N, Saikatsu S, Kawane T, Horiuchi N 1997 Mouse vitamin D-24-hydroxylase: molecular cloning, tissue distribution, and transcriptional regulation by 1α,25-dihydroxyvitamin D3. Endocrinology 138:2233–2240. 62. Zierold C, Reinholz GG, Mings JA, Prahl JM, DeLuca HF 2001 Regulation of the porcine 1,25-dihydroxyvitamin D3-24-hydroxylase (CYP24) by 1,25-dihydroxyvitamin D3 and parathyroid hormone in AOK-B50 cells. Arch Biochem Biophys 381:323–327. 63. Hahn CN, Baker E, Laslo P, et al. 1993 Localization of the human vitamin D 24-hydroxylase gene (CYP24) to chromosome 20q13.2—>q13.3. Cytogenet Cell Genet 62: 192–193. 64. Ohyama Y, Noshiro M, Eggertsen G, et al. 1993 Structural characterization of the gene encoding rat 25-hydroxyvitamin D3 24-hydroxylase. Biochemistry 32:76–82. 65. St Arnaud R, Arabian A, Glorieux FH 1996 Abnormal bone development in mice deficient for the vitamin D 24-hydroxylase gene. J Bone Miner Res 11:S126. 66. St Arnaud R 1999 Targeted inactivation of vitamin D hydroxylases in mice. Bone 25:127–129. 67. Boyan BD, Sylvia VL, Dean DD, Schwartz Z 2001 24, 25-(OH)2D3 regulates cartilage and bone via autocrine and endocrine mechanisms. Steroids 66:363–374. 68. Schwartz Z, Ehland H, Sylvia VL, et al. 2002 1α,25-dihydroxyvitamin D3 and 24R,25-dihydroxyvitamin D3 modulate growth plate chondrocyte physiology via protein kinase Cdependent phosphorylation of extracellular signal-regulated kinase 1/2 mitogen-activated protein kinase. Endocrinology 143:2775–2786. 69. St Arnaud R, Arabian A, Travers R, et al. 2000 Deficient mineralization of intramembranous bone in vitamin D-24hydroxylase-ablated mice is due to elevated 1,25-dihydroxyvitamin D and not to the absence of 24,25-dihydroxyvitamin D. Endocrinology 141:2658–2666. 70. Haussler MR, Whitfield GK, Haussler CA, et al. 1998 The nuclear vitamin D receptor: biological and molecular regulatory properties revealed. J Bone Miner Res 13:325–349. 71. Issa LL, Leong GM, Eisman JA 1998 Molecular mechanism of vitamin D receptor action. Inflamm Res 47:451–475. 72. Kasuga H, Hosogane N, Matsuoka K, et al. 2002 Characterization of transgenic rats constitutively expressing vitamin D-24-hydroxylase gene. Biochem Biophys Res Commun 297:1332–1338. 73. Hosogane N, Shinki T, Kasuga H, et al. 2003 Mechanisms for the reduction of 24,25-dihydroxyvitamin D3 levels and bone mass in 24-hydroxylase transgenic rats. FASEB J 17:737–739. 74. Panda DK, Miao D, Tremblay ML, et al. 2001 Targeted ablation of the 25-hydroxyvitamin D 1α-hydroxylase enzyme: Evidence for skeletal, reproductive, and immune dysfunction. Proc Natl Acad Sci USA 98:7498–7503. 75. Dardenne O, Prud’homme J, Arabian A, Glorieux FH, St Arnaud R 2001 Targeted inactivation of the 25-hydroxyvitamin D(3)-1(α)-hydroxylase gene (CYP27B1) creates an animal model of pseudovitamin D-deficiency rickets. Endocrinology 142:3135–3141. 76. Omdahl JL, May BK 1997 The 25-hydroxyvitamin D 24-hydroxylase. In: Feldman D, Glorieux F, Pike J (eds) Vitamin D. Academic Press, San Diego pp. 69–85.
103
CHAPTER 6 The 25-Hydroxyvitamin D 24-Hydroxylase
77. Omdahl JL, Morris HA, May BK 2002 Hydroxylase enzymes of the vitamin D pathway: Expression, function, and regulation. Annu Rev Nutr 22:139–166. 78. Sutton AL, MacDonald PN 2003 Vitamin D: More than a “bone-a-fide” hormone. Mol Endocrinol 17:777–791. 79. Yang W, Friedman PA, Kumar R, et al. 1999 Expression of 25(OH)D3 24-hydroxylase in distal nephron: coordinate regulation by 1,25(OH)2D3 and cAMP or PTH. Am J Physiol 276:E793–E805. 80. Zierold C, Mings JA, DeLuca HF 2001 Parathyroid hormone regulates 25-hydroxyvitamin D3-24-hydroxylase mRNA by altering its stability. Proc Natl Acad Sci USA 98: 13572–13576. 81. Zierold C, Mings JA, DeLuca HF 2003 Regulation of 25hydroxyvitamin D3-24-hydroxylase mRNA by 1,25-dihydroxyvitamin D3 and parathyroid hormone. J Cell Biochem 88:234–237. 82. Pondel M 2000 Calcitonin and calcitonin receptors: bone and beyond. Int J Exp Pathol 81:405–422. 83. Brown EM, MacLeod RJ 2001 Extracellular calcium sensing and extracellular calcium signaling. Physiol Rev 81:239–297. 84. Inzerillo AM, Zaidi M, Huang CL 2002 Calcitonin: The other thyroid hormone. Thyroid 12:791–798. 85. Jones G, Strugnell SA, DeLuca HF 1998 Current understanding of the molecular actions of vitamin D. Physiol Rev 78:1193–1231. 86. Yoshida N, Yoshida T, Nakamura A, et al. 1999 Calcitonin induces 25-hydroxyvitamin D3 1α-hydroxylase mRNA expression via protein kinase C pathway in LLC-PK1 cells. J Am Soc Nephrol 10:2474–2479. 87. Murayama A, Takeyama K, Kitanaka S, et al. 1998 The promoter of the human 25-hydroxyvitamin D3 1α-hydroxylase gene confers positive and negative responsiveness to PTH, calcitonin, and 1α,25(OH)2D3. Biochem Biophys Res Commun 249:11–16. 88. Bland R, Zehnder D, Hughes SV, et al. 2001 Regulation of vitamin D-1α-hydroxylase in a human cortical collecting duct cell line. Kidney Int 60:1277–1286. 89. Shinki T, Ueno Y, DeLuca HF, Suda T 1999 Calcitonin is a major regulator for the expression of renal 25-hydroxyvitamin D3-1α-hydroxylase gene in normocalcemic rats. Proc Natl Acad Sci USA 96:8253–8258. 90. Henry HL 2001 The 25(OH)D3/1α,25(OH)2D3-24(R)-hydroxylase: A catabolic or biosynthetic enzyme? Steroids 66:391–398. 91. Nishimura A, Shinki T, Jin CH, et al. 1994 Regulation of messenger ribonucleic acid expression of 1α,25-dihydroxyvitamin D3-24-hydroxylase in rat osteoblasts. Endocrinology 134:1794–1799. 92. Ylikomi T, Laaksi I, Lou YR, et al. 2002 Antiproliferative action of vitamin D. Vitam Horm 64:357–406. 93. Gurlek A, Kumar R 2001 Regulation of osteoblast growth by interactions between transforming growth factor-beta and 1α,25-dihydroxyvitamin D3. Crit Rev Eukaryot Gene Expr 11:299–317. 94. Armbrecht HJ, Hodam TL 1994 Parathyroid hormone and 1,25-dihydroxyvitamin D synergistically induce the 1,25dihydroxyvitamin D-24-hydroxylase in rat UMR-106 osteoblast-like cells. Biochem Biophys Res Commun 205:674–679. 95. Armbrecht HJ, Hodam TL, Boltz MA, et al. 1998 Induction of the vitamin D 24-hydroxylase (CYP24) by 1,25-dihydroxyvitamin D3 is regulated by parathyroid hormone in UMR106 osteoblastic cells [see comments]. Endocrinology 139: 3375–3381. 96. Kurahashi I, Matsunuma A, Kawane T, Abe M, Horiuchi N 2002 Dexamethasone enhances vitamin D-24-hydroxylase
97. 98.
99. 100.
101.
102. 103. 104. 105.
106.
107. 108.
109.
110.
111.
112.
expression in osteoblastic (UMR-106) and renal (LLC-PK1) cells treated with 1α,25-dihydroxyvitamin D3. Endocrine 17: 109–118. Griffin MD, Xing N, Kumar R 2003 Vitamin D and its analogs as regulators of immune activation and antigen presentation. Annu Rev Nutr 23:117–145. Dusso AS, Kamimura S, Gallieni M, et al. 1997 γ-Interferoninduced resistance to 1,25-(OH)2 D3 in human monocytes and macrophages: A mechanism for the hypercalcemia of various granulomatoses. J Clin Endocrinol Metab 82:2222–2232. Monkawa T, Yoshida T, Hayashi M, Saruta T 2000 Identification of 25-hydroxyvitamin D3 1α-hydroxylase gene expression in macrophages. Kidney Int 58:559–568. Overbergh L, Decallonne B, Valckx D, et al. 2000 Identification and immune regulation of 25-hydroxyvitamin D-1-α-hydroxylase in murine macrophages. Clin Exp Immunol 120:139–146. Vidal M, Ramana CV, Dusso AS 2002 Stat1-vitamin D receptor interactions antagonize 1,25-dihydroxyvitamin D transcriptional activity and enhance stat1-mediated transcription. Mol Cell Biol 22:2777–2787. Bikle DD, Ng D, Tu C, Oda Y, Xie Z 2001 Calcium- and vitamin D-regulated keratinocyte differentiation. Mol Cell Endocrinol 177:161–171. Bikle DD, Tu CL, Xie Z, Oda Y 2003 Vitamin D regulated keratinocyte differentiation: Role of coactivators. J Cell Biochem 88:290–295. Xie Z, Munson SJ, Huang N, et al. 2002 The mechanism of 1,25-dihydroxyvitamin D3 autoregulation in keratinocytes. J Biol Chem 277:36987–36990. Murayama A, Takeyama K, Kitanaka S, et al. 1999 Positive and negative regulations of the renal 25-hydroxyvitamin D3 1α-hydroxylase gene by parathyroid hormone, calcitonin, and 1α,25(OH)2D3 in intact animals. Endocrinology 140:2224–2231. Murayama A, Kim M, Unno K, Takeyama K, Kato S 2001 A novel mechanism of the vitamin D dependent transcriptional repression through the human 25-hydroxyvitamin D3 1α-hydroxylase in nVDRE. J Bone Miner Res 16:S229. Albertson DG, Ylstra B, Segraves R, et al. 2000 Quantitative mapping of amplicon structure by array CGH identifies CYP24 as a candidate oncogene. Nat Genet 25:144–146. Bareis P, Bises G, Bischof MG, Cross HS, Peterlik M 2001 25-hydroxy-vitamin D metabolism in human colon cancer cells during tumor progression. Biochem Biophys Res Commun 285:1012–1017. Miller GJ, Stapleton GE, Hedlund TE, Moffat KA 1995 Vitamin D receptor expression, 24-hydroxylase activity, and inhibition of growth by 1α,25-dihydroxyvitamin D3 in seven human prostatic carcinoma cell lines. Clin Cancer Res 1:997–1003. Ly LH, Zhao XY, Holloway L, Feldman D 1999 Liarozole acts synergistically with 1α,25-dihydroxyvitamin D3 to inhibit growth of DU 145 human prostate cancer cells by blocking 24-hydroxylase activity. Endocrinology 140: 2071–2076. Kallay E, Adlercreutz H, Farhan H, et al. 2002 Phytoestrogens regulate vitamin D metabolism in the mouse colon: Relevance for colon tumor prevention and therapy. J Nutr 132: 3490S–3493S. Farhan H, Wahala K, Adlercreutz H, Cross HS 2002 Isoflavonoids inhibit catabolism of vitamin D in prostate cancer cells. J Chromatogr B Anal Technol Biomed Life Sci 777:261–268.
104 113. Ohyama Y, Ozono K, Uchida M, et al. 1994 Identification of a vitamin D-responsive element in the 5′-flanking region of the rat 25-hydroxyvitamin D3 24-hydroxylase gene. J Biol Chem 269:10545–10550. 114. Hahn CN, Kerry DM, Omdahl JL, May BK 1994 Identification of a vitamin D responsive element in the promoter of the rat cytochrome P450(24) gene. Nucleic Acids Res 22:2410–2416. 115. Zierold C, Darwish HM, DeLuca HF 1994 Identification of a vitamin D-response element in the rat calcidiol (25-hydroxyvitamin D3) 24-hydroxylase gene. Proc Natl Acad Sci USA 91:900–902. 116. Zierold C, Darwish HM, DeLuca HF 1995 Two vitamin D response elements function in the rat 1,25-dihydroxyvitamin D 24-hydroxylase promoter. J Biol Chem 270:1675–1678. 117. Kerry DM, Dwivedi PP, Hahn CN, et al. 1996 Transcriptional synergism between vitamin D-responsive elements in the rat 25-hydroxyvitamin D3 24-hydroxylase (CYP24) promoter. J Biol Chem 271:29715–29721. 118. Ohyama Y, Ozono K, Uchida M, et al. 1996 Functional assessment of two vitamin D-responsive elements in the rat 25-hydroxyvitamin D3 24-hydroxylase gene. J Biol Chem 271:30381–30385. 119. Dwivedi PP, Omdahl JL, Kola I, Hume DA, May BK 2000 Regulation of rat cytochrome P450C24 (CYP24) gene expression. Evidence for functional cooperation of rasactivated Ets transcription factors with the vitamin D receptor in 1,25-dihydroxyvitamin D3-mediated induction. J Biol Chem 275:47–55. 120. Dwivedi PP, Muscat GE, Bailey PJ, Omdahl JL, May BK 1998 Repression of basal transcription by vitamin D receptor: evidence for interaction of unliganded vitamin D receptor with two receptor interaction domains in RIP13∆1. J Mol Endocrinol. 20:327–335. 121. Barry JB, Leong GM, Church WB, et al. 2003 Interactions of SKIP/NCoA-62, TFIIB, and retinoid X receptor with vitamin D receptor helix H10 residues. J Biol Chem 278:8224–8228. 122. Mantovani R 1999 The molecular biology of the CCAATbinding factor NF-Y. Gene 239:15–27. 123. Zierold C, Mings JA, Prahl JM, Reinholz GG, DeLuca HF 2002 Protein synthesis is required for optimal induction of 25-hydroxyvitamin D3-24-hydroxylase, osteocalcin, and osteopontin mRNA by 1,25-dihydroxyvitamin D3. Arch Biochem Biophys 404:18–24. 124. Falkenstein E, Tillmann HC, Christ M, Feuring M, Wehling M 2000 Multiple actions of steroid hormones—a focus on rapid, nongenomic effects. Pharmacol Rev 52:513–556. 125. Solomon C, White JH, Kremer R 1999 Mitogen-activated protein kinase inhibits 1,25-dihydroxyvitamin D3-dependent signal transduction by phosphorylating human retinoid X receptor α. J Clin Invest 103:1729–1735. 126. Razandi M, Alton G, Pedram A, et al. 2003 Identification of a structural determinant necessary for the localization and function of estrogen receptor α at the plasma membrane. Mol Cell Biol 23:1633–1646. 127. Razandi M, Oh P, Pedram A, Schnitzer J, Levin ER 2002 ERs associate with and regulate the production of caveolin: implications for signaling and cellular actions. Mol Endocrinol 16:100–115. 128. Buitrago C, Gonzalez PV, De Boland AR 2002 Nongenomic action of 1α,25(OH)2-vitamin D3. Activation of muscle cell PLC gamma through the tyrosine kinase c-Src and PtdIns 3-kinase. Eur J BioChem 269:2506–2515.
JOHN OMDAHL AND BRIAN MAY
129. Erben RG, Soegiarto DW, Weber K, et al. 2002 Deletion of deoxyribonucleic acid binding domain of the vitamin D receptor abrogates genomic and nongenomic functions of vitamin D. Mol Endocrinol. 16:1524–1537. 130. Hammes SR 2003 The further redefining of steroid-mediated signaling. Proc Natl Acad Sci USA 100:2168–2170. 131. Wali RK, Kong J, Demay MB, et al. 2001 The Vitamin D receptor is not required for the rapid activation of PKC and rise in intracellular calcium induced by 1,25-dihydroxyvitamin D3 in mouse osteoblasts. J Bone Miner Res 16:S229. 132. Wali RK, Kong J, Sitrin MD, Bissonnette M, Li YC 2003 Vitamin D receptor is not required for the rapid actions of 1,25-dihydroxyvitamin D3 to increase intracellular calcium and activate protein kinase C in mouse osteoblasts. J Cell BioChem 88:794–801. 133. Norman AW, Henry HL, Bishop JE, et al. 2001 Different shapes of the steroid hormone 1α,25(OH)2-vitamin D3 act as agonists for two different receptors in the vitamin D endocrine system to mediate genomic and rapid responses. Steroids 66:147–158. 134. Mandla S, Boneh A, Tenenhouse HS 1990 Evidence for protein kinase C involvement in the regulation of renal 25-hydroxyvitamin D3–24-hydroxylase. Endocrinology 127: 2639–2647. 135. Armbrecht HJ, Chen ML, Hodam TL, Boltz MA 1997 Induction of 24-hydroxylase cytochrome P450 mRNA by 1,25-dihydroxyvitamin D and phorbol esters in normal rat kidney (NRK-52E) cells. J Endocrinol 153:199–205. 136. Koyama H, Inaba M, Nishizawa Y, Ohno S, Morii H 1994 Protein kinase C is involved in 24-hydroxylase gene expression induced by 1,25(OH)2D3 in rat intestinal epithelial cells. J Cell BioChem 55:230–240. 137. Armbrecht HJ, Boltz MA, Hodam TL, Kumar VB 2001 Differential responsiveness of intestinal epithelial cells to 1,25-dihydroxyvitamin D3—role of protein kinase C. J Endocrinol 169:145–151. 138. Yang W, Hyllner SJ, Christakos S 2001 Interrelationship between signal transduction pathways and 1,25(OH)2D3 in UMR106 osteoblastic cells. Am J Physiol Endocrinol Metab 281:E162-E170. 139. Reddy SP, Vuong H, Adiseshaiah P 2003 Interplay between proximal and distal promoter elements is required for squamous differentiation marker induction in the bronchial epithelium: role for ESE-1, Sp1, and AP-1 proteins. J Biol Chem 278:21378–21387. 140. Armbrecht HJ, Hodam TL, Boltz MA 2003 Hormonal regulation of 25-hydroxyvitamin D3-1α-hydroxylase and 24-hydroxylase gene transcription in opossum kidney cells. Arch Biochem Biophys 409:298–304. 141. Gao XH, Dwivedi PP, Choe S, et al. 2002 Basal and parathyroid hormone induced expression of the human 25-hydroxyvitamin D 1α-hydroxylase gene promoter in kidney AOK-B50 cells: role of Sp1, Ets and CCAAT box protein binding sites. Int J Biochem Cell Biol 34: 921–930. 142. Raggatt LJ, Evdokiou A, Findlay DM 2000 Sustained activation of Erk1/2 MAPK and cell growth suppression by the insert-negative, but not the insert-positive isoform of the human calcitonin receptor. J Endocrinol 167:93–105. 143. Takeyama K, Kitanaka S, Sato T, et al. 1997 25-Hydroxyvitamin D3 1α-hydroxylase and vitamin D synthesis. Science 277:1827–1830.
CHAPTER 7
Mutant Mouse Models of Vitamin D Metabolic Enzymes RENÉ ST-ARNAUD
Genetics Unit, Shriners Hospital for Children, and McGill University, Montreal (Quebec), Canada
I. Introduction II. Hepatic 25-Hydroxylation and cyp27A1 Knockout III. 24-Hydroxylation and cyp24A1 Knockout
I. INTRODUCTION A. Note on Cytochrome P450 Nomenclature Since 1987, efforts have been made to encourage scientists worldwide to avoid “home-made” designations for naming cytochrome P450s in order to avoid confusing the nomenclature system and the scientific literature. This chapter will follow the guidelines provided in the relevant publications [1,2] and the recently updated names listed on the Cytochrome P450 Homepage Web site (http://drnelson.utmem.edu/mouse.master.table.html). Human genes will be italicized in capital letters, while mouse genes will be in lower case italics. Proteins are referred to in capital, nonitalicized characters (Table I).
B. Vitamin D Metabolism and Mutant Mouse Models of Vitamin D Metabolic Enzymes Vitamin D, produced endogenously in the skin upon exposure to ultraviolet light (sunlight) or ingested in the diet must be metabolized twice to be activated and function as a key regulator of mineral ion homeostasis. Vitamin D, bound to the vitamin D binding protein, DBP, is transported to the liver where vitamin D 25-hydroxylases (CYP27A1 and/or CYP2R1) add a hydroxyl group on carbon 25 to produce 25-hydroxyvitamin D [25(OH)D] (see Chapter 4). The 25(OH)D metabolite also circulates in the bloodstream bound to DBP (see Chapter 8). It must be further hydroxylated in the kidney to gain hormonal bioactivity. Hydroxylation at position 1α by the enzyme 25-hydroxyvitamin D1α-hydroxylase (CYP27B1) converts 25(OH)D to 1α,25-dihydroxyvitamin D [1,25(OH)2D], the active, hormonal form of vitamin D that regulates mineral homeostasis, skeletal homeostasis, and cellular differentiation (see Chapter 5). Among several target genes, the 1,25(OH)2D hormone induces in target cells the VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX
IV. 1α-Hydroxylation and cyp27B1 Knockout V. Summary and Perspectives References
expression of the gene encoding the key effector of its catabolic breakdown: 25-hydroxyvitamin D-24-hydroxylase (CYP24A1). This ensures attenuation of the 1,25(OH)2D biological signal inside target cells and helps regulate vitamin D homeostasis. All vitamin D hydroxylases characterized to date belong to the superfamily of cytochrome P450 enzymes. They are heme-containing, mixed-function oxidases that use molecular oxygen as a terminal electron acceptor. They require the accessory electron transfer proteins, ferredoxin and ferredoxin reductase, to accept reducing equivalents from nicotinamide adenine dinucleotide phosphate (NADPH) and stereospecifically hydroxylate vitamin D metabolites. Molecular genetics–based research has allowed dramatic progress in our understanding of vitamin D metabolism and function. Several mice strains, deficient for the vitamin D metabolic enzymes CYP27A1, CYP27B1, and CYP24A1, have been engineered. This chapter will review the phenotype of the mutant strains, highlight the key findings derived from the analysis of the strains, and discuss the future avenues for research.
II. HEPATIC 25-HYDROXYLATION AND cyp27A1 KNOCKOUT Until recently, only one cytochrome P450 molecule had been cloned and shown to hydroxylate vitamin D at position 25. This was the bifunctional CYP27A1 sterol 27-hydroxylase that derives its name for its ability to both 27-hydroxylate the side chains of cholesterolderived intermediates involved in bile acid biosynthesis and 25-hydroxylate vitamin D [3]. However, several pieces of experimental evidence argued for the existence of a second, physiologically relevant vitamin D 25-hydroxylase enzyme (reviewed in [4]). First, mice with a disrupted cyp27A1 gene have reduced bile acid Copyright © 2005, Elsevier, Inc. All rights reserved.
106
RENÉ ST-ARNAUD
TABLE I Cytochrome P450 Nomenclature Used Common name
Human gene symbol
Mouse gene symbol
Protein
CYP27A1 CYP2R1 CYP27B1 CYP24A1
cyp27A1 cyp2R1 cyp27B1 cyp24A1
CYP27A1 CYP2R1 CYP27B1 CYP24A1
Vitamin D-25-hydroxylase (mitochondrial) Vitamin D-25-hydroxylase (microsomal) 25-Hydroxyvitamin D-1α-hydroxylase 25-Hydroxyvitamin D-24-hydroxylase
synthesis, but normal vitamin D metabolite levels [5]. Second, patients with the inherited disease cerebrotendinous xanthomatosis, caused by mutations in CYP27A1 [6,7], exhibit normal vitamin D metabolism and no obvious 25(OH)D or 1,25(OH)2D deficiency [8]. Third, early studies using perfused rat liver revealed kinetics of vitamin D metabolism that supported two 25-hydroxylase activities: a high-affinity, low-capacity enzyme and a low-affinity, high-capacity form (now presumed to be CYP27A1) [9]. The subcellular localizations of these two enzymatic activities are thought to differ, CYP27A1 being a mitochondrial enzyme [10,11], while the high-affinity, low-capacity enzyme is associated with the endoplasmic reticulum (microsomes) [12]. Next, the promoter of CYP27A1 is regulated by bile acids, but not by vitamin D metabolites [13], a finding that is not consistent with the weak but demonstrated regulation of the hepatic 25-hydroxylase observed following vitamin D intake in previously vitamin D–depleted animals [14]. Last, cells transfected with an expression vector for CYP27A1 are not able to 25hydroxylate vitamin D2 [15]. The 25-hydroxy-D2 metabolite is synthesized in vivo, however, suggesting that an enzyme distinct from CYP27A1 is involved in its synthesis.
A. Phenotype of cyp27A1-Deficient Mice The CYP27A1 enzyme is important for bile acid biosynthesis and regulation of cholesterol homeostasis [10,11]. Mutations in the human gene cause cerebrotendinous xanthomatosis (CTX), a lipid storage disorder leading to premature atherosclerosis and progressive neurological dysfunction [6,7]. Mice deficient for cyp27A1 have been engineered [5]. Homozygous cyp27A1−/− animals have larger livers and larger adrenals [16], decreased synthesis and excretion of bile acids [5], and hypertriglyceridemia [16]. The increased formation of 25-hydroxylated bile alcohols and cholestanol observed in CTX patients was not observed in cyp27A1−/− mice, and no CTX-related pathological abnormalities were evident [5]. At present, it remains
unclear whether the cyp27A1-deficient mouse is a valid animal model for CTX. Interestingly, cyp27A1-null mice had normal serum concentrations of 1,25(OH)2D and slightly elevated levels of 25(OH)D in their blood [5]. These results strengthen the hypothesis that a distinct vitamin D 25-hydroxylase enzyme plays a physiological role in vitamin D metabolism.
B. Identification of Novel Vitamin D-25-Hydroxylases An expression-based screening strategy has been used to demonstrate that CYP2R1 has vitamin D 25-hydroxylase activity. The experimental strategy used a cDNA library from cyp27A1-null mice to avoid isolating clones encoding the sterol 27-hydroxylase (CYP27A1). Expression of CYP2R1 in transfected cells led to transcriptional activation through the vitamin D receptor when vitamin D2, vitamin D3, or 1α(OH)D3 was added to cells [17]. The results obtained strongly suggest that CYP2R1 is the microsomal vitamin D 25-hydroxylase. Mice with mutations in the cyp2R1 gene are being engineered. It will prove interesting to analyze vitamin D metabolism and circulating levels of 25(OH)D and 1,25(OH)2D in cyp2R1-null mice. Based on specific activity determination, it was estimated that the microsomal enzyme is responsible for about 30% of the total 25-hydroxylation activity, while the mitochondrial enzyme is responsible for the remaining 70% [18]. This suggests a redundancy of function that could lead to no detectable vitamin D–related phenotype in cyp2R1-deficient mice. Altering the vitamin D 25-hydroxylase activity in mice in vivo would then require crossing the cyp27A1-null and cyp2R1-deficient strains and measuring circulating vitamin D metabolites levels as a function of cyp27A1 and cyp2R1 gene dosage. Purified recombinant CYP3A4 was also recently shown to 25-hydroxylate the 1α(OH)D2 substrate in vitro, with lower enzymatic activity toward 1α(OH)D3 or vitamin D2 [19]. Interestingly, CYP3A4 expression
107
CHAPTER 7 Mutant Mouse Models of Vitamin D Metabolic Enzymes
vectors were devoid of 25-hydroxylase activity in the transfection-based strategy that led to the identification of CYP2R1 as a putative microsomal vitamin D 25-hydroxylase [17]. CYP3A4 regulates the metabolism of more than half of all commonly used drugs. It remains to be determined whether CYP3A4 plays any physiological role in vitamin D metabolism or whether its activity is limited to the conversion of the prodrug 1α(OH)D2.
III. 24-HYDROXYLATION AND cyp24A1 KNOCKOUT The CYP24A1 enzyme catalyzes the addition of an hydroxyl group on carbon 24 of the vitamin D secosteroid backbone. It can utilize several substrates: 25(OH)D, to produce 24,25-dihydroxyvitamin D [24,25(OH)2D]; 1,25(OH)2D, to produce 1,24,25trihydroxyvitamin D [1,24,25(OH)3D]; and even 1α-hydroxyvitamin D [1α(OH)D], to generate 1α,24dihydroxyvitamin D [1,24(OH)2D]. The 24,25(OH)2D metabolite is the most abundant dihydroxylated metabolite in the circulation. The C-24 oxidation pathway catalyzed by the CYP24A1 molecule comprises five enzymatic steps that involve successive hydroxylation/oxidation reactions at carbons 24 and 23, followed by cleavage of the secosteroid at the C-23/C-24 bond and subsequent oxidation of the cleaved product to calcitroic acid [20,21]. The hypothesis that the main role of the C-24 oxidation pathway is attenuation of the 1,25(OH)2D biological signal inside target cells was tested in vitro using cytochrome P450 inhibitors. Blocking P450 activity by treatment of cells with ketoconazole inhibits catabolism and results in 1,25(OH)2D accumulation and extended hormone action [22]. This hypothesis was also tested and confirmed in vivo by engineering cyp24A1-deficient mice.
A. Phenotype of cyp24A1-Deficient Mice Exons 9 and 10 of the cyp24A1 gene, encoding the heme-binding domain [23], were deleted using homologous recombination in embryonic stem (ES) cells. This strategy ensured that a true null allele was created. Heterozygous animals were phenotypically normal and fertile, and the engineered mutation was transmitted to the progeny with the expected Mendelian ratio [24]. Fifty percent of cyp24A1−/− mice die before 3 weeks of age (Table II) [24,25]. Analysis of macrophage function ruled out impaired responses to infection as the cause for post-natal death. The perinatal lethality is
TABLE II
Incomplete Penetrance of the cyp24A1−/− Perinatal Lethality Phenotype
Number of cyp24A1−/− embryos 578
Alive at weaning
Dead at weaning
274 (47%)
304 (53%)
most likely a consequence of hypercalcemia secondary to hypervitaminosis D, since the inactivation of the cyp24A1 gene in mice impaired the ability of the animals to clear 1,25(OH)2D. Bolus and chronic 1,25 (OH)2D administration resulted in a marked elevation in serum 1,25(OH)2D levels in the mutant animals [24]. Chronic 1,25(OH)2D administration in cyp24A1−/− mutants resulted in histological changes consistent with hypervitaminosis D in the kidney: cortical tubular dilation, necrotic debris, and calcification (nephrocalcinosis). The inability to regulate 1,25(OH)2D and calcium homeostasis presumably leads to fatal hypercalcemia. Indeed, extremely high levels of circulating 1,25(OH)2D and calcium were measured in runted animals that died before weaning [24]. Another condition where vitamin D homeostasis is disregulated in cyp24A1-null mice is pregnancy. Circulating 1,25(OH)2D levels rise in the late stages of gestation in rodents [26]. Since gestating cyp24A1−/− females cannot regulate this increase through CYP24A1-mediated catabolism, serum 1,25(OH)2D concentrations rise to abnormally high levels in pregnant cyp24A1−/− dams [24]. This is accompanied by concomitant elevation of the circulating calcium concentration [24] that sometimes reaches fatal levels. Since half of the mutant progeny appear unaffected by the cyp24A1 deficiency, these animals most likely use alternate means of regulating vitamin D homeostasis. We have measured clearance and metabolism of labeled 1,25(OH)2D in cyp24A1−/− survivors and heterozygote controls. These experiments have confirmed that cyp24A1-null mice have impaired clearance of 1,25(OH)2D. Surprisingly, cyp24A1−/− mice appear to lack not only 24-hydroxylated metabolites but also 1,25(OH)2D-26,23-lactone [27], supporting the view that the CYP24A1 enzyme is also responsible for hydroxylating 1,25(OH)2D at C-23 and C-26 [28,29]. These surprising findings suggest that the cyp24A1null survivors adapt to the impaired vitamin D catabolism not by using an alternative catabolic route, but by limiting the synthesis of the active compound. The survival of some cyp24A1−/− mutant animals to adulthood has also allowed experiments designed to address the effect of perturbing vitamin D metabolism during development. Bone development is
108 abnormal in homozygous mutants born of homozygous females [24,30]. Bone forms via two mechanisms during embryogenesis: intramembranous bone formation and endochondral bone formation [31,32]. The intramembranous formation allows the growth of the surface of flat bones as well as the thickening of cortical bone, while endochondral formation provides for the lengthening of long bones. The major difference between the two is the presence or absence of an intermediate cartilaginous phase. Intramembranous bone formation occurs when mesenchymal precursor cells proliferate and differentiate directly into osteoblasts that produce a mucoprotein matrix, called osteoid, in which collagen fibrils are enmeshed. The osteoblasts then begin to mineralize the osteoid by depositing inorganic crystals of calcium phosphate on, between, and within the collagen fibers, forming a primary immature bone tissue called woven bone. Woven bone is progressively replaced by mature, lamellar bone. In lamellar bone, the matrix is compact and the collagen fibers are long and highly ordered. Endochondral bone formation entails the conversion of a cartilaginous template, the “anlagen,” into bone. Mesenchymal cells condense and differentiate into chondroblasts that secrete the cartilaginous matrix. The chondroblasts embed themselves within the matrix they produce and become chondrocytes. Because of the malleable nature of the cartilaginous matrix, the chondrocytes can continue to proliferate in their lacunae. The embryonic anlagen, a template of the future bone piece, is avascular. During its early development, a ring of woven bone (the bony collar) is formed by intramembranous ossification in the future midshaft area. This calcified woven bone is then invaded by vascular tissue, and this process of angiogenesis brings to the bone rudiment the precursors of the osteoclastic lineage. The osteoclasts excavate the hematopoietic bone marrow cavity while new osteoblasts are recruited to replace the cartilage scaffold with bone matrix. At the extremities of long bones (epiphyses), longitudinal growth occurs by a similar process of endochondral bone formation at the growth plates [32]. Growth-plate chondrocytes undergo several steps of maturation from proliferating chondrocytes, forming characteristic orderly columns, to nonproliferating, hypertrophic cells. The cartilaginous matrix becomes mineralized just below the hypertrophic zone of the growth plate and the chondrocytes then undergo apoptosis. Once calcified, the cartilage matrix is partially resorbed by osteoclasts. Following resorption, osteoblasts differentiate to form a layer of woven bone on top of the mineralized cartilage remnants of the longitudinal septa. This is the first activation–resorption– formation sequence of remodeling the cartilage into
RENÉ ST-ARNAUD
woven bone. Further down in the growth plate, this immature bone will undergo subsequent remodeling to replace the woven bone and calcified cartilage remnants with lamellar bone. An extensive literature demonstrates that cells from the growth plate respond to 24,25(OH)2D in vitro in a cell maturation–dependent manner (reviewed in [33]). The less differentiated cells of the resting zone respond to 24,25(OH)2D. The more mature cells of the prehypertrophic and hypertrophic compartments respond primarily to 1,25(OH)2D. Interestingly, treatment of resting zone chondrocytes with 24,25(OH)2D induces a change in maturation state [34], supporting the hypothesis that 24,25(OH)2D plays a role in cartilage development. It is interesting to note that no significant disruption of growth plate organization was noted in cyp24A1-null mice and that all cell types could readily be identified (Fig. 1), with the expression of differentiation markers such as type II collagen and type X collagen confirmed (St-Arnaud and Beaulieu, unpublished results). Our data suggest that the absence of CYP24A1 activity does not affect growth plate development and
FIGURE 1 Normal growth plate architecture in cyp24A1-null mice. Sagittal sections through the vertebrae of cyp24A1+/− (upper panel) and cyp24A1−/− (lower panel) littermates. Sections were stained by the Goldner method. See color plate.
CHAPTER 7 Mutant Mouse Models of Vitamin D Metabolic Enzymes
that 24,25(OH)2D is not required for chondrocyte maturation in vivo. It remains formally possible, however, that a redundant endocrine system may be able to compensate for the function of the 24,25(OH)2D metabolite in living animals. Histological examination of the bones from homozygous cyp24A1−/− pups born from homozygous cyp24A1-null dams revealed an accumulation of osteoid at sites of intramembranous ossification, however [24,30]. Control heterozygote littermates showed normal bone structure. Two major hypotheses can be formulated to account for the phenotype of the cyp24A1-deficient embryos from cyp24A1 mutant mice: (1) perturbation of 1,25(OH)2D catabolism through the inactivation of the C-24 oxidation pathway affects bone development; (2) metabolites of vitamin D hydroxylated at position 24 are essential for bone development. To differentiate between these possibilities, the cyp24A1-deficient animals were bred to mice carrying an inactivating mutation of the vitamin D receptor (VDR) gene [35]. If elevated 1,25(OH)2D levels, acting through the vitamin D receptor, were responsible for the observed phenotype, then mice lacking the receptor and the cyp24A1 gene should not show the aberrant intramembranous bone development. Using this elegant genetic strategy, incontrovertible evidence was obtained that expression of the VDR is required for the impaired mineralization phenotype of the cyp24A1-deficient animals: double mutant homozygotes (cyp24A1−/− and VDR−/−) showed normal intramembranous bone formation at all sites examined [24]. This demonstrates that elevated 1,25(OH)2D levels during gestation affect mineralization and suggest that impaired vitamin D metabolism during development perturbs bone formation.
B. Future Studies The cyp24A1-deficient strain represents a useful tool to address the still controversial question of the putative biological activity of vitamin D metabolites hydroxylated at the 24 position. Does it remain possible that 24,25(OH)2D exerts biological activity in specific tissues and that genetic redundancy prevented the identification of this activity in cyp24A1-null mice? Using the cyp24A1-ablated mice as a source of costochondral chondrocytes would confirm or refute the importance of local production of vitamin D metabolites in the maturation of chondrocytic cells in this culture system [36]. Published results also suggest that 24,25(OH)2D may be important during fracture repair. The circulating levels of 24,25(OH)2D increase during fracture repair
109
in chicks because of an increase in renal CYP24A1 activity [37]. When the effect of various vitamin D metabolites on the mechanical properties of healed bones was tested, the strength of healed bones was maximized in animals fed 24,25(OH)2D3 [38]. These results support a role for 24,25(OH)2D as an essential vitamin D metabolite for fracture repair. We have started to use the murine model of distraction osteogenesis [39] to compare fracture repair between wild-type mice and cyp24A1−/− mutant mice. If a defect in fracture repair is observed in cyp24A1-null mice, we will attempt rescue using 24,25(OH)2D, 1,24,25(OH)3D, 1,24(OH)2D, or other vitamin D analogs.
IV. 1α-HYDROXYLATION AND cyp27B1 KNOCKOUT The 25-hydroxyvitamin D-1α-hydroxylase enzyme (CYP27B1) converts 25(OH)D to 1,25(OH)2D, the biologically active metabolite of vitamin D. Mutations in the CYP27B1 gene cause pseudo–vitamin D deficiency Rickets (PDDR), a rare autosomal disease characterized by growth retardation, failure to thrive, rickets, and osteomalacia [40,41] (see Chapter 71). Biochemical analysis of serum from affected patients reveals hypocalcemia, secondary hyperparathyroidism, and undetectable levels of 1,25(OH)2D3.
A. Phenotype of cyp27B1-Deficient Mice An animal model of PDDR was engineered independently by two laboratories using targeted inactivation of the cyp27B1 gene in mice [42,43]. The engineered mutation was transmitted to the progeny with the expected Mendelian ratio. Heterozygous animals had no discernable phenotype and were fertile. Homozygous mutant animals were phenotypically normal at birth but exhibited retarded growth as measured by weight gain [43] and femur length at 8 weeks [44]. Serum analysis of homozygous mutant animals confirmed that they were hypocalcemic, hypophosphatemic, and had undetectable circulating levels of 1,25(OH)2D3 [42,43]. Serum parathyroid hormone levels were highly increased as early as weaning [42] and remained elevated thereafter [42,43]. The parathyroid glands also increased in size in young adult cyp27B1 homozygous mutant mice [42,43]. From a biochemical standpoint, the only difference between the cyp27B1−/− mice and patients with PDDR is that patients with PDDR have normal serum levels of 25(OH)D [45–47] and 24,25(OH)2D [48], whereas
110 elevated levels of 25(OH)D [42,43] and very low levels of 24,25(OH)2D [42] are seen in cyp27B1 mutant mice. The 1,25(OH)2D hormone is the main in vivo regulator of the expression of the CYP24A1 enzyme that catalyzes the synthesis of 24,25(OH)2D [49]. Therefore, low circulating 24,25(OH)2D levels and undetectable cyp24A1 expression in cyp27B1−/− mice would be anticipated. The discrepancy with the human disease remains to be explained but could result from species differences. This observed inhibition of cyp24A1 expression in mice, combined with the targeted ablation of the CYP27B1 enzyme, leads to a metabolic block in mutant animals and an accumulation of the unprocessed 25(OH)D substrate [42]. The expression of vitamin D–dependent genes was also analyzed in cyp27B1-null mice. Expression of the calbindin-D9k gene was completely inhibited in the intestine of mutant mice [42,43] and severely downregulated in the kidney [43]. A reduction in the levels of mRNA for calbindin-D28k and the vitamin D receptor in kidney tissue was observed, whereas renal expression of cyp24A1 was almost completely abolished [42,43]. Interestingly, expression levels of the osteocalcin and osteopontin genes in tibia were unaffected in mutant animals [42]. The osteocalcin gene promoter contains several positive and negative regulatory elements [50]. Similarly, osteopontin gene expression is regulated by many agents acting via diverse signaling pathways in specific cell types [51]. Although 1,25(OH)2D3 has been shown to regulate the expression of these genes both in vitro and in vivo [50,51], the unaltered mRNA levels for both genes measured in cyp27B1-null animals suggest that the vitamin D hormone is not the main modulator of the steady-state expression of these genes in murine bone tissue. Histological analysis of the bones from 3-week-old mutant animals confirmed the evidence of rickets [42]. At the age of 7 to 8 weeks, femurs from cyp27B1-ablated mice present a severe disorganization in the architecture of the growth plate and marked osteomalacia [42,43]. The X-ray features of the bones from cyp27B1−/− animals also matched the clinical manifestations of PDDR. Contact radiography of femurs from mutant animals revealed diffuse osteopenia (hypomineralization) and rachitic metaphyseal changes [52]. Inactivation of cyp27B1 therefore produced a bone phenotype that is identical to the human genetic disease PDDR, confirming that mutations in CYP27B1 underlie this disorder. The animal model was further used to compared efficacy of clinical rescue regimens (see later discussion). The vitamin D hormone has been shown to have immunomodulatory effects, and local synthesis of 1,25(OH)2D could play an important autocrine or paracrine role in the differentiation or function of immune cells [53–55]. Interestingly, cyp27B1-deficient
RENÉ ST-ARNAUD
mice exhibited a reduction in CD4- and CD8-positive peripheral T lymphocyte populations [43]. Large CD4positive ectopic lymph nodes were detected in the neck of the cyp27B1-null and cyp27B1+/− heterozygous animals [43]. These findings support an important role for CYP27B1 in the regulation of immune function. This phenotype has never been reported in vitamin D receptor (VDR) knockout strains [35,56,57], which raises the hypothesis that 1,25(OH)2D may not exert its immunomodulatory effects via the VDR or that CYP27B1 influences immune function via metabolites other than 1,25(OH)2D [43]. The fertility and reproductive capacity of vitamin D– deficient rats is impaired, and normal reproductive effectiveness can be rescued by normalizing calcemia in males but not females, where 1,25(OH)2D is required for reproductive function [58,59]. cyp27B1-deficient females showed uterine hypoplasia, decreased ovarian size, compromised folliculogenesis, and infertility [43]. A similar phenotype was observed in one strain of VDR-deficient mice [57]. These results establish a critical role for cyp27B1 function in female reproduction.
B. Rescue of the Phenotype 1. 1,25(OH)2D REPLACEMENT THERAPY
The treatment of choice for PDDR patients is replacement therapy with 1,25(OH)2D3 [45]. It results in rapid and complete correction of the abnormal phenotype, restoring normocalcemia, eliminating secondary hyperparathyroidism and features of rickets. The restoration of bone mineral content is equally rapid and histological evidence of healing has been documented [45]. This therapeutic regimen was used to rescue the PDDR phenotype of cyp27B1−/− mice [60]. Animals received daily subcutaneous injections of 1,25(OH)2D3. A “rescue” concentration of 500 pg per g body weight (BW) was administered from 21 days of age until 35 days of age, followed by a “maintenance” dose of 100 pg per g BW (s.c.) from 35 days of age until they reached the age of 60 days. This treatment normalized biochemical parameters, restored the biomechanical properties of bones, and cured rickets and osteomalacia, as assessed by histomorphometric analyses [60]. PDDR patients receiving 1,25(OH)2D3-replacement therapy have normal serum concentrations of the metabolite [45], whereas chronic treatment with 1,25(OH)2D3 did not normalize circulating levels of the metabolite in cyp27B1−/− mice [60]. This difference is most likely due to the induction of the CYP24A1 enzyme observed in 1,25(OH)2D3-treated cyp27B1null mutants [60], and further highlights the species differences in vitamin D metabolic pathways that was noted in the first description of the animal model for
111
CHAPTER 7 Mutant Mouse Models of Vitamin D Metabolic Enzymes
PDDR [42,43], with regard to the regulation of cyp24A1 gene expression and circulating 25(OH)D3 or 24,25(OH)2D3 metabolite levels. 2. 1α(OH)D REPLACEMENT THERAPY
Before 1,25(OH)2D3 became available as a drug, clinicians used the monohydroxylated analog 1α(OH)D3 for treating PDDR patients [61]. This compound, which requires only liver hydroxylation at the 25 position to fully mimic 1,25(OH)2D3, is still used in Japan and some European countries. We have used a rescue regimen similar to the treatment described earlier to treat cyp27B1-deficient mice with 1α(OH)D3. The treatment was as effective as the injection protocol with 1,25(OH)2D3 to normalize calcemia, phosphatemia, and reverse the secondary hyperparathyroidism in cyp27B1−/− mice, without deleterious effects in heterozygous controls. The replacement therapy with 1α(OH)D3 also normalized growth and corrected the bone phenotype of the mutant mice, effectively curing rickets and osteomalacia and normalizing histomorphometric indices. Experiments are in progress to compare the efficacy of other vitamin D analogs, such as 1α(OH)D2 or 1,24(S)(OH)2D2 [62], in rescuing the PDDR phenotype of cyp27B1 knockout animals. 3. HIGH-CALCIUM, HIGH-PHOSPHORUS, HIGH-LACTOSE DIET
Mutation in the vitamin D receptor (VDR) gene results in a second form of vitamin D–related rickets called hereditary vitamin D–resistant rickets (HVDRR),
characterized by hypocalcemia, secondary hyperparathyroidism, rickets, osteomalacia, and alopecia [63] (see Chapter 72). Several laboratories have developed a valid mouse model for this type of hereditary rickets [35,56,57]. Rescue of rickets has been successfully performed using a 2% calcium diet, 1.25% phosphorus and 20% lactose, previously reported to prevent hypocalcemia in vitamin D–deficient rats [64]. This diet completely normalized all biochemical and histomorphometric parameters in VDR-ablated animals [65,66]. We rescued the PDDR phenotype of mice deficient for the cyp27B1 gene by feeding them with the highcalcium diet [67]. Homozygous mutant and heterozygous control animals were fed the rescue diet (2% calcium, 1.25% phosphorus, and 20% lactose) from 3 weeks of age until sacrifice at 8.5 weeks of age. Blood biochemistry analysis revealed that the rescue diet corrected the hypocalcemia and secondary hyperparathyroidism. Despite the restoration of normocalcemia, cyp27B1−/− (and heterozygous control) animals fed the rescue diet initially gained weight less rapidly than control mice fed normal mouse chow. Although cyp27B1−/− mice fed the rescue diet eventually reached the same weight as control animals, the treatment did not entirely correct bone growth as femur size remained significantly smaller than control (Fig. 2). Bone histology and histomorphometry confirmed that the rickets and osteomalacia were cured. The rescue diet also restored the biomechanical properties of the bone tissue within normal parameters. These results demonstrate that
Growth rescue in cyp27B1-deficient mice *
20
**
control
* Femur length (mm)
1,25(OH)2D3 rescue High Ca2+ diet rescue 10
0 +/−
−/−
The high-calcium rescue diet only partially restores bone growth of cyp27B1−/− mice. Femurs were measured at sacrifice (60 days old). cyp27B1−/− animals had significantly reduced femur lengths, which was only partially corrected by the high-calcium diet (High Ca2+ diet rescue). Replacement therapy with 1,25(OH)2D3 (1,25(OH)2D3 rescue) normalized femur growth. *, p < 0.05; **, p < 0.01.
FIGURE 2
112 correction of the abnormal mineral ion homeostasis by feeding with a high-calcium rescue diet is effective to rescue the PDDR phenotype of cyp27B1 mutant mice. This treatment, however, does not appear as effective as 1,25(OH)2D3 replacement therapy since bone growth remained impaired. In addition to the partial rescue of femur growth, the high-calcium rescue diet increased growth plate thickness in cyp27B1−/− and cyp27B1+/− animals [67]. This phenomenon was not observed in VDR-ablated mice receiving similar treatment [65,66]. A major difference between the two strains of animals is the ability to synthesize 1,25(OH)2D3, both systemically and locally. It has been hypothesized that local production of 1,25(OH)2D3 is important for the proliferation and maturation of chondrocytes in the growth plate, and that some of these effects involve nongenomic mechanisms of action for 1,25(OH)2D3 [33]. Autocrine synthesis and nongenomic actions of 1,25(OH)2D3 would remain active in VDR-ablated animals, whereas they are affected in cyp27B1-deficient mice. This could explain the differences in the response of the two strains to the dietary manipulation. The targeted cyp27B1 mouse model that we have engineered [42] represents an invaluable tool to characterize the putative autocrine/ paracrine roles of 1,25(OH)2D3 in the growth plate under conditions of normal mineral ion homeostasis (see later discussion).
C. Future Studies Local production of 1,25(OH)2D could play an important autocrine or paracrine role in the differentiation or function of osteoblasts, chondrocytes, keratinocytes, and macrophages [33,68,69]. The targeting strategy used to engineer the cyp27B1-deficient strain, based on the Cre/lox methodology, allows tissuespecific inactivation of the targeted gene. The technology utilizes the Cre recombinase from the P1 bacteriophage. This enzyme catalyzes recombination between two 34-base-pair recognition elements, called loxP sites, causing excision of the intervening sequences. Transgenic mice lines expressing Cre under the control of tissue-specific promoters are crossed with mice harboring loxP sites to achieve excision at will (Fig. 3) (reviewed in [70]). The cyp27B1 gene was cloned from a 129 SV mouse genomic library to construct a targeting vector in which exon 8, encoding the heme binding domain [71], was flanked by a 5′-loxP recognition site and by a 3′-loxP-neo-loxP selection cassette. Following homologous recombination at the cyp27B1 locus in ES cells and injection of targeted ES cells in host blastocysts, a mouse strain in which the cyp27B1
RENÉ ST-ARNAUD
locus is “floxed” (flanked by loxP) was established. The conditional cyp27B1 allele provides an invaluable genetic tool to analyze the putative autocrine/paracrine roles that have been hypothesized for 1,25(OH)2D in various cell types. It may be argued that a significant portion of the active 1,25(OH)2D metabolite found in target tissues is derived systemically. Against the background of circulating levels of the hormone, any local drop may be too subtle an effect to be easily detectable. However, a recent finding in the field of sex steroids is that a large proportion of androgens and estrogens are synthesized locally in peripheral target tissues such as skin or adipose tissue [72]. Despite “central” steroidogenesis in the gonads, peripheral target tissues utilize “on the spot” intracellular hormone synthesis/action to regulate their function [73]. Moreover, chondrocytes are compartmentalized away from the circulation, and local synthesis of vitamin D metabolites may contribute significantly to the regulation of the proliferation, differentiation, and function of growth plate cells. The 1,25(OH)2D hormone has been shown to stimulate macrophage function. Treatment of macrophages with 1,25(OH)2D enhances macrophage chemotaxis [74] and increases phagocytic killing of bacteria [75]. Crossing the floxed cyp27B1 mice with mice expressing Cre in macrophages under the control of the macrophage scavenger receptor A gene promoter [76] would inhibit autocrine synthesis of 1,25(OH)2D by macrophages. One possible phenotype that could be predicted following this genetic manipulation would be increased susceptibility to microbial infection. The 1,25(OH)2D hormone is likely to be an autocrine or paracrine factor for epidermal differentiation [77] since keratinocytes make 1,25(OH)2D [78], express the VDR [79], and respond to 1,25(OH)2D with changes in proliferation and differentiation [79,80]. Indeed, our results show that the expression of keratinocyte differentiation markers is reduced in the skin of cyp27B1 knockout animals and that recovery of the epidermal barrier (assessed by the transepidermal water loss assay) [81] is significantly delayed in CYP27B1-deficient skin [82]. Tissue-specific inactivation of cyp27B1 in floxed mice crossed to keratin 5-Cre [83] or keratin 14-Cre transgenicmice [84] would allow examination of the contribution of systemic versus local production of 1,25(OH)2D on skin differentiation. Chondrocytes have been shown to express cyp27B1 both in vivo [71] and in vitro [36,85]. This has led investigators to hypothesize that local production of 1,25(OH)2D could play an important autocrine or paracrine role in the differentiation or function of these cells [33]. To test this putative role of the vitamin D
CHAPTER 7 Mutant Mouse Models of Vitamin D Metabolic Enzymes
113
FIGURE 3 Tissue-specific gene inactivation using the Cre/lox strategy. Floxed (flanked by loxP) mice are crossed with transgenic mice in which the Cre recombinase is under the control of a tissue-specific promoter (TSP). In this example, loxP sites (red triangles) flank exon 8 of the target gene and the tissue-specific promoter is expressed in bone tissue. Expression of the Cre transgene in bone cells will cause excision of exon 8 between the two loxP sites (bottom left), while the target gene will remain intact in other tissues due to lack of Cre expression (bottom right panel).
hormone in growth plate cells, we have initiated cyp27B1 gene ablation in chondrocytes [86] by crossing the floxed cyp27B1 strain with transgenic mice expressing the Cre recombinase under the control of the type II collagen promoter/enhancer (Col2-Cre) [87]. Because 1,25(OH)2D regulates the growth and terminal differentiation of hypertrophic chondrocytes in vitro [33,88,89], we hypothesize that these parameters will be perturbed in vivo following the genetic manipulation that we will engineer. This should be manifested as increased chondrocyte proliferation and an abnormal architecture of the growth plate, which could even affect the primary spongiosa. Neovascularization and VEGF expression could be reduced [90]. The expression of differentiation markers could be perturbed in Col2-Cre; 1α−/fl mice.
V. SUMMARY AND PERSPECTIVES Mutant mouse models have allowed significant progress in our understanding of vitamin D metabolism and function. The observation that cyp27A1−/− (mitochondrial 25-OHase) mice have normal vitamin D homeostasis suggested that a distinct cytochrome P450 enzyme could be involved in the 25-hydroxylation of vitamin D under physiological conditions. An expressionbased cloning strategy using a cDNA library prepared from the liver of cyp27A1−/− animals was then used to identify the microsomal 25-hydroxylating enzyme, CYP2R1. It will prove interesting to analyze vitamin D metabolism and circulating levels of 25(OH)D and 1,25(OH)2D in cyp2R1-null mice and cyp27A1-cyp2R1 compound mutants to determine the relative contributions
114 of the two enzymes in vivo. The cyp24A1-deficient strain will be used to test the putative role of vitamin D metabolites hydroxylated at position 24 in mammalian fracture repair. The confirmation of such a role for 24,25(OH)2D would open up interesting avenues for clinical research using vitamin D analogs. Finally, the cyp27B1-deficient mice represent a valid animal model for pseudo–vitamin D deficiency rickets. The conditional ablation strategy used to engineer these animals will help elucidate the hypothesized autocrine/ paracrine roles of 1,25(OH)2D in target tissues such as cartilage, skin, and macrophages.
References 1. McAlpine PJ, Van Cong N, Boucheix C, Pakstis AJ, Doute RC, Shows TB 1987 The 1987 Catalog of mapped genes and report of the nomenclature committee. Cytogenet Cell Genet 46:29–101. 2. Shows TB, et al., McAlpine PJ, Boucheix C, Collins FS, Conneally PM, Frezal J, Gershowitz H, Goodfellow PN, Hall JG 1987 Guidelines for human gene nomenclature. An international system for human gene nomenclature (ISGN, 1987). Cytogenet Cell Genet 46:11–28. 3. Okuda KI, Usui E, Ohyama Y 1995 Recent progress in enzymology and molecular biology of enzymes involved in vitamin D metabolism. J Lipid Res 36:1641–1652. 4. Jones G, Strugnell SA, DeLuca HF 1998 Current understanding of the molecular actions of vitamin D. Physiol Rev 78:1193–1231. 5. Rosen H, Reshef A, Maeda N, Lippoldt A, Shpizen S, Triger L, Eggertsen G, Bjorkhem I, Leitersdorf E 1998 Markedly reduced bile acid synthesis but maintained levels of cholesterol and vitamin D metabolites in mice with disrupted sterol 27-hydroxylase gene. J Biol Chem 273:14805–14812. 6. Salen G, Shefer S, Cheng FW, Dayal B, Batta AK, Tint GS 1979 Cholic acid biosynthesis: The enzymatic defect in cerebrotendinous xanthomatosis. J Clin Invest 63:38–44. 7. Cali JJ, Hsieh CL, Francke U, Russell DW 1991 Mutations in the bile acid biosynthetic enzyme sterol 27-hydroxylase underlie cerebrotendinous xanthomatosis. J Biol Chem 266: 7779–7783. 8. Kuriyama M, Fujiyama J, Kubota R, Nakagawa M, Osame M 1993 Osteoporosis and increased bone fractures in cerebrotendinous xanthomatosis. Metabolism 42:1497–1498. 9. Fukushima M, Suzuki Y, Tohira Y, Nishii Y, Suzuki M 1976 25-Hydroxylation of 1α-hydroxyvitamin D3 in vivo and in perfused rat liver. FEBS Lett 65:211–214. 10. Cali JJ, Russell DW 1991 Characterization of human sterol 27-hydroxylase. A mitochondrial cytochrome P-450 that catalyzes multiple oxidation reactions in bile acid biosynthesis. J Biol Chem 266:7774–7778. 11. Andersson D, Davis DL, Dahlback H, Jornvall H 1989 Cloning, structure and expression of the mitochondrial cytochrome P-450 sterol 26-hydroxylase, a bile acid biosynthetic enzyme. J Biol Chem 264:8222–8229. 12. Madhok TC, DeLuca HF 1979 Characteristics of the rat liver microsomal enzyme system converting cholecalciferol into 25-hydroxycholecalciferol. Evidence for the participation of cytochrome p-450. Biochem J 184:491–499.
RENÉ ST-ARNAUD
13. Vlahcevic ZR, Jairath SK, Heuman DM, Stravitz RT, Hylemon PB, Avadhani NG, Pandak WM 1996 Transcriptional regulation of hepatic sterol 27-hydroxylase by bile acids. Am J Physiol 270:G646–652. 14. Bhattacharyya MH, DeLuca HF 1973 The regulation of rat liver calciferol-25-hydroxylase. J Biol Chem 248:2969–2973. 15. Guo YD, Strugnell S, Back DW, Jones G 1993 Transfected human liver cytochrome P-450 hydroxylates vitamin D analogs at different side-chain positions. Proc Natl Acad Sci USA 90:8668–8672. 16. Repa JJ, Lund EG, Horton JD, Leitersdorf E, Russell DW, Dietschy JM, Turley SD 2000 Disruption of the sterol 27-hydroxylase gene in mice results in hepatomegaly and hypertriglyceridemia. Reversal by cholic acid feeding. J Biol Chem 275:39685–39692. 17. Cheng JB, Motola DL, Mangelsdorf DJ, Russell DW 2003 De-orphanization of cytochrome P450 2R1: A microsomal vitamin d 25-hydroxylase. J Biol Chem 278:38084–38093. 18. Bjorkhem I, Holmberg I 1978 Assay and properties of a mitochondrial 25-hydroxylase active on vitamin D3. J Biol Chem 253:842–849. 19. Gupta RP, Hollis BW, Patel SB, Bell NH 2003 CYP3A4 is a human microsomal vitamin D-25-hydroxylase. J Bone Miner Res (in press). 20. Makin G, Lohnes D, Byford V, Ray R, Jones G 1989 Target cell metabolism of 1,25-dihydroxyvitamin D3 to calcitroic acid. Evidence for a pathway in kidney and bone involving 24-oxidation. Biochem J 262:173–180. 21. Reddy GS, Tserng KY 1989 Calcitroic acid, end product of renal metabolism of 1,25-dihydroxyvitamin D3 through C-24 oxidation pathway. Biochemistry 28:1763–1769. 22. Reinhardt TA, Horst RL 1989 Ketoconazole inhibits selfinduced metabolism of 1,25-dihydroxyvitamin D3 and amplifies 1,25-dihydroxyvitamin D3 receptor up-regulation in rat osteosarcoma cells. Arch Biochem Biophys 272:459–465. 23. Ohyama Y, Noshiro M, Eggertsen G, Gotoh O, Kato Y, Bjorkhem I, Okuda K 1993 Structural characterization of the gene encoding rat 25-hydroxyvitamin D3 24-hydroxylase. Biochemistry 32:76–82. 24. St-Arnaud R, Arabian A, Travers R, Barletta F, Raval-Pandya M, Chapin K, Depovere J, Mathieu C, Christakos S, Demay MB, Glorieux FH 2000 Deficient mineralization of intramembranous bone in vitamin D-24-hydroxylase-ablated mice is due to elevated 1,25-dihydroxyvitamin D and not to the absence of 24,25-dihydroxyvitamin D. Endocrinology 141:2658–2666. 25. St-Arnaud R 1999 Targeted inactivation of vitamin D hydroxylases in mice. Bone 25:127–129. 26. Kovacs CS, Kronenberg HM 1997 Maternal-fetal calcium and bone metabolism during pregnancy, puerperium, and lactation. Endocr Rev 18:832–872. 27. Masuda S, Byford V, Arabian A, Sakai Y, Demay MB, St-Arnaud R, Jones G 2004 Altered pharmacokinetics of 1α,25-dihydroxyvitamin D3 and 25-hydroxyvitamin D3 in blood and tissues of the vitamin D-24-hydroxylase null mouse. Submitted. 28. Miyamoto Y, Shinki T, Yamamoto K, Ohyama Y, Iwasaki H, Hosotani R, Kasama T, Takayama H, Yamada S, Suda T 1997 1α,25-dihydroxyvitamin D3-24-hydroxylase (CYP24) hydroxylates the carbon at the end of the side chain (C-26) of the C-24-fluorinated analog of 1α,25-dihydroxyvitamin D3. J Biol Chem 272:14115–14119. 29. Beckman MJ, Tadikonda P, Werner E, Prahl J, Yamada S, DeLuca HF 1996 Human 25-hydroxyvitamin D3-24-hydroxylase, a multicatalytic enzyme. Biochemistry 35:8465–8472.
CHAPTER 7 Mutant Mouse Models of Vitamin D Metabolic Enzymes
30. St-Arnaud R, Glorieux FH 1997 Vitamin D and bone development. In: Feldman D, Glorieux FH, Pike JW (eds) Vitamin D. Academic Press, San Diego, pp. 293–303. 31. Zelzer E, Olsen BR 2003 The genetic basis for skeletal diseases. Nature 423:343–348. 32. Kronenberg HM 2003 Developmental regulation of the growth plate. Nature 423:332–336. 33. Boyan BD, Dean DD, Sylvia VL, Schwartz Z 1997 Cartilage and vitamin D: genomic and nongenomic regulation by 1,25(OH)2D3 and 24,25(OH)2D3. In: Feldman D, Glorieux FH, Pike JW (eds) Vitamin D. Academic Press, San Diego, pp. 395–422. 34. Schwartz Z, Dean D, Walton J, Brooks B, Boyan B 1995 Treatment of resting zone chondrocytes with 24,25-dihydroxyvitamin D3 [24,25-(OH)2D3] induces differentiation into a 1,25-(OH)2D3-responsive phenotype characteristic of growth zone chondrocytes. Endocrinology 136:402–411. 35. Li YC, Pirro AE, Amling M, Delling G, Baron R, Bronson R, Demay MB 1997 Targeted ablation of the vitamin D receptor: an animal model of vitamin D-dependent rickets type II with alopecia. Proc Natl Acad Sci USA 94:9831–9835. 36. Schwartz Z, Brooks B, Swain L, Del Toro F, Norman A, Boyan B 1992 Production of 1,25-dihydroxyvitamin D3 and 24, 25-dihydroxyvitamin D3 by growth zone and resting zone chondrocytes is dependent on cell maturation and is regulated by hormones and growth factors. Endocrinology 130:2495–2504. 37. Seo EG, Norman AW 1997 Three-fold induction of renal 25-hydroxyvitamin D3-24-hydroxylase activity and increased serum 24,25-dihydroxyvitamin D3 levels are correlated with the healing process after chick tibial fracture. J Bone Miner Res 12:598–606. 38. Seo EG, Einhorn TA, Norman AW 1997 24R,25-dihydroxyvitamin D3: An essential vitamin D3 metabolite for both normal bone integrity and healing of tibial fracture in chicks. Endocrinology 138:3864–3872. 39. Tay BK, Le AX, Gould SE, Helms JA 1998 Histochemical and molecular analyses of distraction osteogenesis in a mouse model. J Orthop Res 16:636–642. 40. Miller WL, Portale AA 1999 Genetic causes of rickets. Curr Opin Pediatr 11:333–339. 41. St-Arnaud R, Glorieux FH 2000 Hereditary defects in vitamin D metabolism and action. In: DeGroot LJ, Jameson JL (eds) Endocrinology, 4th ed., Vol. 2. W.B. Saunders, Philadelphia; pp. 1154–1168. 42. Dardenne O, Prud’homme J, Arabian A, Glorieux FH, St-Arnaud R 2001 Targeted inactivation of the 25-hydroxyvitamin D3-1α-hydroxylase gene (CYP27B1) creates an animal model of pseudovitamin D-deficiency rickets. Endocrinology 142:3135–3141. 43. Panda DK, Miao D, Tremblay ML, Sirois J, Farookhi R, Hendy GN, Goltzman D 2001 Targeted ablation of the 25-hydroxyvitamin D 1α-hydroxylase enzyme: Evidence for skeletal, reproductive, and immune dysfunction. Proc Natl Acad Sci USA 98:7498–7503. 44. St-Arnaud R, Demay MB 2003 Vitamin D biology. In: Glorieux FH, Juppner H, Pettifor JM (eds) Pediatric Bone: Biology & Diseases. Elsevier/Academic Press, San Diego, pp. 193–216. 45. Delvin EE, Glorieux FH, Marie PJ, Pettifor JM 1981 Vitamin D dependency: replacement therapy with calcitriol. J Pediatr 99: 26–34. 46. Rosen JF, Finberg L 1972 Vitamin D-dependent rickets: actions of parathyroid hormone and 25-hydroxycholecalciferol. Pediatr Res 6:552–562.
115
47. Scriver CR, Reade TM, DeLuca HF, Hamstra AJ 1978 Serum 1,25-dihydroxyvitamin D levels in normal subjects and in patients with hereditary rickets or bone disease. N Engl J Med 299:976–979. 48. Mandla S, Jones G, Tenenhouse HS 1992 Normal 24-hydroxylation of vitamin D metabolites in patients with vitamin Ddependency rickets type I. Structural implications for the vitamin D hydroxylases. J Clin Endocrinol Metab 74:814–820. 49. Shinki T, Jin CH, Nishimura A, Nagai Y, Ohyama Y, Noshiro M, Okuda K, Suda T 1992 Parathyroid hormone inhibits 25-hydroxyvitamin D3-24-hydroxylase mRNA expression stimulated by 1α,25-dihydroxyvitamin D3 in rat kidney but not in intestine. J Biol Chem 267:13757–13762. 50. Lian JB, Stein GS, Stein JL, van Wijnen AJ 1998 Osteocalcin gene promoter: Unlocking the secrets for regulation of osteoblast growth and differentiation. J Cell Biochem Suppl 31:62–72. 51. Denhardt DT, Noda M 1998 Osteopontin expression and function: role in bone remodeling. J Cell Biochem Suppl 31:92–102. 52. St-Arnaud R, Dardenne O, Glorieux FH 2001 Etiologie moléculaire des rachitismes vitamino-dépendents héréditaires. Médecine/Sciences 17:1289–1296. 53. Abe E, Miyaura C, Sakagami H, Takeda M, Konno K, Yamazaki T, Yoshiki S, Suda T 1981 Differentiation of mouse myeloid leukemia cells induced by 1α,25-dihydroxyvitamin D3. Proc Natl Acad Sci USA 78:4990–4994. 54. Nunn JD, Katz DR, Barker S, Fraher LJ, Hewison M, Hendy GN, O’Riordan JL 1986 Regulation of human tonsillar T-cell proliferation by the active metabolite of vitamin D3. Immunology 59:479–484. 55. Provvedini DM, Rulot CM, Sobol RE, Tsoukas CD, Manolagas SC 1987 1α,25-Dihydroxyvitamin D3 receptors in human thymic and tonsillar lymphocytes. J Bone Miner Res 2:239–247. 56. Van Cromphaut SJ, Dewerchin M, Hoenderop JG, Stockmans I, Van Herck E, Kato S, Bindels RJ, Collen D, Carmeliet P, Bouillon R, Carmeliet G 2001 Duodenal calcium absorption in vitamin D receptor-knockout mice: functional and molecular aspects. Proc Natl Acad Sci USA 98:13324–13329. 57. Yoshizawa T, Handa Y, Uematsu Y, Takeda S, Sekine K, Yoshihara Y, Kawakami T, Arioka K, Sato H, Uchiyama Y, Masushige S, Fukamizu A, Matsumoto T, Kato S 1997 Mice lacking the vitamin D receptor exhibit impaired bone formation, uterine hypoplasia and growth retardation after weaning. Nat Genet 16:391–396. 58. Halloran BP, DeLuca HF 1980 Effect of vitamin D deficiency on fertility and reproductive capacity in the female rat. J Nutr 110:1573–1580. 59. Kwiecinski GG, Petrie GI, DeLuca HF 1989 Vitamin D is necessary for reproductive functions of the male rat. J Nutr 119:741–744. 60. Dardenne O, Prudhomme J, Hacking SA, Glorieux FH, St-Arnaud R 2003 Rescue of the pseudo-vitamin D deficiency rickets phenotype of CYP27B1-deficient mice by treatment with 1,25-dihydroxyvitamin D3: biochemical, histomorphometric, and biomechanical analyses. J Bone Miner Res 18: 637–643. 61. Reade TM, Scriver CR, Glorieux FH, Nogrady B, Delvin E, Poirier R, Holick F, DeLuca HF 1975 Response to crystalline 1α-hydroxyvitamin D3 in vitamin D dependency. Pediatr Res 9:593–599. 62. Horst R, Prapong S, Reinhardt T, Koszewski N, Knutson J, Bishop C 2000 Comparison of the relative effects of 1,24dihydroxyvitamin D(2) [1,24-(OH)(2)D(2)], 1,24-dihydroxyvitamin D(3) [1,24-(OH)(2)D(3)], and 1,25-dihydroxyvitamin D(3)
116
63. 64. 65.
66.
67.
68.
69. 70. 71.
72. 73.
74.
75.
76.
77.
[1,25-(OH)(2)D(3)] on selected vitamin D-regulated events in the rat. Biochem Pharmacol 60:701–708. Malloy PJ, Pike JW, Feldman D 1999 The vitamin D receptor and the syndrome of hereditary 1,25-dihydroxyvitamin Dresistant rickets. Endocr Rev 20:156–188. Kollenkirchen U, Fox J, Walters MR 1991 Normocalcemia without hyperparathyroidism in vitamin D-deficient rats. J Bone Miner Res 6:273–278. Li YC, Amling M, Pirro AE, Priemel M, Meuse J, Baron R, Delling G, Demay MB 1998 Normalization of mineral ion homeostasis by dietary means prevents hyperparathyroidism, rickets, and osteomalacia, but not alopecia in vitamin D receptor-ablated mice. Endocrinology 139:4391–4396. Amling M, Priemel M, Holzmann T, Chapin K, Rueger JM, Baron R, Demay MB 1999 Rescue of the skeletal phenotype of vitamin D receptor-ablated mice in the setting of normal mineral ion homeostasis: formal histomorphometric and biomechanical analyses. Endocrinology 140:4982–4987. Dardenne O, Prud’homme J, Hacking SA, Glorieux FH, St-Arnaud R 2003 Correction of the abnormal mineral ion homeostasis with a high-calcium, high-phosphorus, highlactose diet rescues the PDDR phenotype of mice deficient for the 25-hydroxyvitamin D-1α-hydroxylase (CYP27B1). Bone 32:332–340. Hewison M, O’Riordan JLH 1997 Immunomodulatory and cell differentiation effects of vitamin D. In: Feldman L, Glorieux FH, Pike JW (eds) Vitamin D. Academic Press, San Diego, pp. 447–462. Bikle D 1997 Vitamin D and skin. In: Feldman D, Glorieux FH, Pike JW (eds) Vitamin D. Academic Press, San Diego, pp. 379–394. Jiang R, Gridley T 1997 Gene targeting: Things go better with Cre. Curr Biol 7:R321–323. Panda DK, Al Kawas S, Seldin MF, Hendy GN, Goltzman D 2001 25-hydroxyvitamin D 1α-hydroxylase: structure of the mouse gene, chromosomal assignment, and developmental expression. J Bone Miner Res 16:46–56. Chen W, Thiboutot D, Zouboulis CC 2002 Cutaneous androgen metabolism: basic research and clinical perspective. J Invest Dermatol 119:992–1007. Labrie F, Luu-The V, Lin SX, Simard J, Labrie C, El-Alfy M, Pelletier G, Bélanger A 2000 Intracrinology: role of the family of 17β−hydroxysteroid dehydrogenases in human physiology and disease. J Mol Endocrinol 25:1–16. Girasole G, Wang JM, Pedrazzoni M, Pioli G, Balotta C, Passeri M, Lazzarin A, Ridolfo A, Mantovani A 1990 Augmentation of monocyte chemotaxis by 1α,25-dihydroxyvitamin D3. Stimulation of defective migration of AIDS patients. J Immunol 145:2459–2464. Rook GA, Taverne J, Leveton C, Steele J 1987 The role of gamma-interferon, vitamin D3 metabolites and tumour necrosis factor in the pathogenesis of tuberculosis. Immunology 62:229–234. Horvai A, Palinski W, Wu H, Moulton KS, Kalla K, Glass CK 1995 Scavenger receptor A gene regulatory elements target gene expression to macrophages and to foam cells of atherosclerotic lesions. Proc Natl Acad Sci USA 92:5391–5395. Pillai S, Bikle DD, Elias PM 1988 Vitamin D and epidermal differentiation: Evidence for a role of endogenously produced
RENÉ ST-ARNAUD
78. 79. 80.
81. 82.
83.
84.
85.
86.
87.
88.
89.
90.
vitamin D metabolites in keratinocyte differentiation. Skin Pharmacol 1:149–160. Bikle DD, Nemanic MK, Whitney JO, Elias PW 1986 Neonatal human foreskin keratinocytes produce 1,25-dihydroxyvitamin D3. Biochemistry 25:1545–1548. Pillai S, Bikle DD, Elias PM 1988 1,25-Dihydroxyvitamin D production and receptor binding in human keratinocytes varies with differentiation. J Biol Chem 263:5390–5395. Smith EL, Walworth NC, Holick MF 1986 Effect of 1α,25dihydroxyvitamin D3 on the morphologic and biochemical differentiation of cultured human epidermal keratinocytes grown in serum-free conditions. J Invest Dermatol 86: 709–714. Vinson LJ, Singer EJ, Kowbler WR, Lehman MD, Masurat T 1965 The nature of the epidermal barrier and some factors influencing skin permeability. Tox Appl Pharm 7:7–19. Bikle DD, Chang S, Crumrine D, Elalieh H, Man M-Q, Choi EH, Dardenne O, Xie Z, St-Arnaud R, Feingold K, Elias PM 2004 25-hydroxyvitamin D 1 alpha hydroxylase is required for optimal epidermal differentiation and permeability barrier homeostasis. J Invest Dermatol 122:984–992. Tarutani M, Itami S, Okabe M, Ikawa M, Tezuka T, Yoshikawa K, Kinoshita T, Takeda J 1997 Tissue-specific knockout of the mouse Pig-a gene reveals important roles for GPI-anchored proteins in skin development. Proc Natl Acad Sci USA 94:7400–7405. Vasioukhin V, Degenstein L, Wise B, Fuchs E 1999 The magical touch: genome targeting in epidermal stem cells induced by tamoxifen application to mouse skin. Proc Natl Acad Sci USA 96:8551–8556. Pedrozo HA, Boyan BD, Mazock J, Dean DD, Gomez R, Schwartz Z 1999 TGFβ1 regulates 25-hydroxyvitamin D3 1αand 24-hydroxylase activity in cultured growth plate chondrocytes in a maturation-dependent manner. Calcif Tissue Int 64:50–56. St-Arnaud R, Dardenne O, Prud’homme J, Hacking SA, Glorieux FH 2003 Conventional and tissue-specific inactivation of the 25-hydroxyvitamin D-1α-hydroxylase (CYP27B1). J Cell Biochem 88:245–251. Terpstra L, Prud’homme J, Arabian A, Takeda S, Karsenty G, Dedhar S, St-Arnaud R 2003 Reduced chondrocyte proliferation and chondrodysplasia in mice lacking the integrin-linked kinase in chondrocytes. J Cell Biol 162:139–148. Gerstenfeld LC, Kelly CM, Von Deck M, Lian JB 1990 Comparative morphological and biochemical analysis of hypertrophic, non-hypertrophic and 1,25(OH)2D3 treated non-hypertrophic chondrocytes. Connect Tissue Res 24: 29–39. Kato Y, Shimazu A, Iwamoto M, Nakashima K, Koike T, Suzuki F, Nishii Y, Sato K 1990 Role of 1,25-dihydroxycholecalciferol in growth-plate cartilage: inhibition of terminal differentiation of chondrocytes in vitro and in vivo. Proc Natl Acad Sci USA 87:6522–6526. Lin R, Amizuka N, Sasaki T, Aarts MM, Ozawa H, Goltzman D, Henderson JE, White JH 2002 1α,25-dihydroxyvitamin D3 promotes vascularization of the chondro-osseous junction by stimulating expression of vascular endothelial growth factor and matrix metalloproteinase 9. J Bone Miner Res 17:1604–1612.
CHAPTER 8
Vitamin D–Binding Protein CHRISTOPHER J. LAING AND NANCY E. COOKE Departments of Genetics and Medicine, University of Pennsylvania, Philadelphia, Pennsylvania
I. Introduction II. Vitamin D Binding Protein and Its Gene III. Functional Features of Vitamin D Binding Protein
IV. Concluding Remarks References
I. INTRODUCTION
been cloned and sequenced. The amino acid structure of DBP was first predicted from these cDNA sequences. Human DBP mRNA contains 1690 nucleotides and encodes a 458-amino-acid secreted protein. There is a 16-amino-acid signal peptide, based upon alignment of the sequenced amino terminus of the mature protein [15,16] with the cDNA-predicted sequence of the primary translation product. The predicted mature DBP proteins in rat, mouse, and rabbit are each 460 amino acids long, two amino-acids longer than the human sequence, whereas that of T. scripta is 466 amino acids in length (see Section II,B,3). There is an N-linked glycosylation consensus sequence in human, rat, and mouse but not rabbit or T. scripta DBP. Periodic positioning of 28 cysteine residues predicts a characteristic secondary structure demarcated by internal disulfide bonds and defines the presence of three internally homologous domains [15] (Fig. 1). The crystal structure of DBP has now been described by several groups [17–21]. The three domains of DBP consist entirely of α-helices, 10 in domain 1, nine in domain 2, and four in domain 3. For a more detailed description of the tertiary structure of the DBP molecule, refer to Chapter 9.
Originally identified in 1959 by serum electrophoresis as a polymorphic protein, vitamin D binding protein (DBP) was referred to as the group-specific component of serum (Gc globulin) [1]. At that time its function was not known, although it became useful in population genetics [2] and forensic medicine [3]. DBP is structurally related to albumin and α-fetoprotein, and the DBP gene is a member of the linked albumin and α-fetoprotein gene family. Binding, solubilization, and serum transport of the vitamin D sterols have been considered the predominant functions of DBP, although there is now evidence that DBP may have a spectrum of biological activities, including roles in inflammation and the immune system. The following sections review the general structural and functional features of DBP. A more detailed review of how some of these structural features may explain DBP’s functional activities can be found in Chapter 9. Chapter 10 reviews the relationships among DBP, megalin, and cubilin in the regulation of a key step in vitamin D metabolism.
II. VITAMIN D BINDING PROTEIN AND ITS GENE A. Structural Features Serum DBP is a polymorphic, monomeric, serum α-globulin of approximately 58 kDa, its size being dependent upon glycosylation state [4]. The DNA sequences have been determined for the human [5,6] and rat [7] DBP structural genes. Rat and human DBP genes span 35 and 42 kb, respectively, and both contain 13 exons. The signal sequence of the protein is encoded on the first exon, whereas the last exon contains the entire noncoding 3′-untranslated region. Since the initial reports of human DBP cDNA structure [8,9], DBP cDNAs from the rat [10], mouse [11], rabbit [12], chicken [13], and a turtle, Trachemys scripta [14], have VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND, GLORIEUX
B. Evolutionary Relationships 1. CHROMOSOMAL LOCALIZATION AND LINKAGES
The human DBP gene was assigned to chromosome 4 by analysis of somatic cell hybrids and sublocalized to 4q11-q13 by in situ hybridization to metaphase chromosome spreads [22]. This sublocalization overlaps with the known positions of albumin and α-fetoprotein [23]. These three genes have been assigned to chromosome 13 in the rat [24] and to chromosome 5 in the mouse [11]. The rodent chromosomes encoding DBP are syntenic with human chromosome 4, thus demonstrating conservation of the DBP–α-fetoprotein–albumin linkage in three species. A linkage between DBP and albumin Copyright © 2005, Elsevier, Inc. All rights reserved.
118
CHRISTOPHER J. LAING AND NANCY E. COOKE
L
NH2
T
S
L
V
L
Q S
A
K
S C E
E
K
T
K N K
E
Y D R
G R E
C K
E
F
S H
L
G K
E D
F
A C C A
E
T
L
S
V
V
E
K
D P D C Y D
T
R
T
S
A
L
S
V S
L
V
L
Y
S R
K
V Q E
F
T G S
P
F G A
E
S N S
P
F
P V
G
DOMAIN I
L
E I
A S
V
Y
E R K
C C E L
C M
D N
T
P
E
T
P
F
E Q P Q H K
L
C E
A
F
R K D P
K
E
Y
W
S
T
C C S G V M S
L
Y
S
K
P
T
V C
F
A N Q F M
T G P H
A A E
Y
S
T
N
Y G
Y
S
V
L
L
S
L
P
A Q
S R
L
S N
L
I
K
L
A Q K
A
L
P
L
V D E
L
D A
T
V
K
L
C D N
L
S
T
K
T
K
E Q C C D E
F
K
S
L
K
E R
L
Q
L
K
V R N S
L
T
T
L
L
S
L
C S Q Y
A
A
Y G E
K
K
D E
T
A
H
V S
E C C K
S
L
I
N
T
I
S
E D C M A
K
E
L
P
E H
T
P
A N
DOMAIN II
A M D V K N
T
P
L
F
V C
T
Y
F
P
Q A
A
E
V D P
L
E
L
T
K
V M D K
Y
T
F
E G
L
S
K
L
T
P
E
L V
K
S
N A
K G P
L
L
K
E Q G K D
I
F
S
S
L
E
C A D Y
S
E N
T
F
T
E
Y
K
K
K
I
S C C N S
A
F
D S R K N V
L
K
A
P
L
E
I
D A
I
L
COOH
M P
D V C D P G N E D
V D C C S
T
T
C
F
E
L
S R R
L
F
V
E
P
L
L
A
E R
L
K
A
L
E
K
T
A D P
T H
K L K L
DOMAIN III N S
P
Y C D S
E
L
K N
P
FIGURE 1 Human DBP, primary structure. The amino acid sequence is indicated using single-letter amino acid abbreviations. Fewer than 25% of the amino acids are different in human, mouse, and rat DBP (shaded) [8–11]. The smaller circles mark the positions of amino acid residues present in albumin and α-fetoprotein, but absent in DBP [8]. The three internal domains defined by amino acid sequence similarities are indicated at left. Disulfide bond positions are marked, and the position of a disulfide bond present in domain 1 of DBP, but absent in both albumin and α-fetoprotein, is indicated by a distinctive line between residues 13 and 59. The positions of interruptions to the coding sequence by introns are indicated by arrowheads.
119
CHAPTER 8 Vitamin D–Binding Protein
20 kb upstream from the 5′ end of the α-fetoprotein gene [31,32]. The rat α-albumin gene is located 10 kb downstream from the α-fetoprotein locus, so that the order of the four genes from 5′ to 3′ with respect to the direction of albumin transcription is the same in the rat as in humans. Similar spatial relationships between the genes have also been reported in the mouse [33]. Thus, DBP is a member of a linked multigene family encoding at least four serum proteins, each of which is predominantly expressed in the fetal and/or adult liver (see Section III,A).
has also been noted in horses and in chickens [25,26]. A fourth gene in the DBP/albumin/α-fetoprotein family has been described by two groups and named afamin by one group [27] and α-albumin by the other [28]. α-Albumin is selectively expressed in the liver during the late stages of development, suggesting that it may be a phylogenetic intermediate between α-fetoprotein and albumin and may share some of the functions of these proteins [28,29]. Physical and meiotic mapping has determined that the organization of the gene family in humans is centromere–DBP–albumin–α-fetoprotein–α-albumin– telomere, the transcriptional orientation of DBP being opposite to the other members of the gene family (Fig. 2). The separation between DBP and the remaining members of the gene family is at least 1.5 Mb [30]. In contrast to the substantial distance between DBP and albumin, the albumin and α-fetoprotein genes are tightly linked in the mammalian species studied so far. In the rat and human, albumin is located only 14 to
2. EVOLUTION WITHIN THE GENE FAMILY
There is a striking pattern of structural similarity among the proteins and genes of the DBP, albumin, α-fetoprotein, and α-albumin family [6–12,15,16,20, 21,28]. The four mRNAs and encoded proteins have sequence similarities of approximately 40% at the nucleotide level and 25% at the amino acid level. The positioning of disulfide bond–forming cysteine residues
IN SITU LOCALIZATION DBP ALB/AFP/αALB
GENETIC LINKAGE (cM) IL8
1.9
ALB
1.0
DBP
>1.5 Mb
630 kb αALB AFP ALB
PHYSICAL MAP
(100 kb)
280 MY 300 MY HUMAN CHROMOSOME 4 560-600 MY
TRIPLE DOMAIN
SINGLE DOMAIN
FIGURE 2 The linked DBP multigene family. The chromosomal locations of DBP, albumin (ALB), α-fetoprotein (AFP), and α-albumin (αALB) as determined by in situ localization and recombination analyses are indicated [22,157]. Physical mapping has indicated the arrangement and transcriptional orientations of the genes to be as shown [30]. The evolutionary relationships among the members of the DBP/ALB/AFP-αALB multigene family, as predicted from amino acid and nucleic acid similarities, are indicated in the cladistic diagram [34–37]. Times of divergence are indicated in millions of years (MY). The physical maps and cladogram are not drawn to scale.
120 is nearly identical in all four proteins, predicting a shared three-domain secondary structure. Despite the amino acid sequence similarities between DBP and albumin, and the similarity in individual domain topologies, the orientation of the three domains is very different in the two proteins. This possibly contributes to the unique binding properties of DBP [20,21]. The carboxy-terminal domain of DBP is 124 amino acids shorter that those of albumin and α-fetoprotein. This is due to the loss of the exons homologous to 12 and 13 in the 15-exon albumin and α-fetoprotein genes, resulting in the 13-exon DBP gene [6,7]. With the exception of these two missing exons, the positions at which the introns interrupt the coding regions of the three genes are highly similar (Fig. 1). These striking structural parallels firmly establish the shared ancestry of the DBP, albumin, and α-fetoprotein genes. A number of models have been proposed to account for the evolution of the DBP multigene family and for the shared, triplicated internal domain structure. In one model, the ancestral internal domain is encoded by four exons that subsequently triplicated, creating the present gene encoding a three-domain protein. This domain triplication likely predated vertebrate evolution [34,35]. The separation of a unique DBP gene from this precursor is estimated to have occurred 560–600 million years ago [36,37] (Fig. 2). It has been postulated that albumin duplicated from the ancestral gene and began to diverge about 280 million years ago, just after the time of the amphibian/reptile divergence about 350 million years ago. The absence of a larval α-fetoprotein gene in amphibians and the existence of α-fetoprotein in chickens is consistent with this conclusion [36]. Duplication and subsequent evolution of the α-albumin and α-fetoprotein genes followed this [37]. Evidence of this divergence may be observed in a duplicated albumin gene in the adult amphibian and some reptiles [36,38], and the presence of α-fetoprotein in reptiles. It has been estimated that DBP has evolved at the slowest rate, and that α-fetoprotein and α-albumin have evolved at the fastest rates of the four proteins, although all have evolved at a more rapid rate than proteins of the globin family [36,37]. There has been significant invasion of noncoding sequences by repetitive DNA elements. Intron expansion in the DBP gene has been significantly greater compared with the other family members, providing further evidence of its longer period of separation from the ancestral gene [39]. Regardless of the precise details of this series of evolutionary events, DBP appears to be the oldest member of this ancient and well-conserved gene family. 3. DBP IN VERTEBRATE EVOLUTION
A specific protein carrier for vitamin D compounds has been identified in examples of all major vertebrate
CHRISTOPHER J. LAING AND NANCY E. COOKE
phyla [40–44]. The presence of a serum DBP in some species of cartilaginous fish (Cyprinus carpio), but not in others (Tilapia nilotica), suggests that the emergence of a discrete DBP may have occurred during the evolution of this phylum [41]. Overall the structures of the DBP gene, mRNA, and protein have been well conserved. There is over 70% amino acid homology among human, mouse, rat, rabbit, and T. scripta DBP. Furthermore, the affinity of reptilian, avian, and mammalian DBPs for the major vitamin D metabolite, 25-hydroxyvitamin D3, are all in the nanomolar range. However, despite this high degree of both structural and functional conservation, significant interspecies variations in DBP concentration and in DBP–ligand affinity exist. These variations appear to be more dependent on ecological variables such as geographic distribution and diet, than on phylogenetic distance [43,44]. Only mammalian DBP demonstrates equal affinity for metabolites of vitamin D2 and vitamin D3, suggesting that the ability to efficiently handle vitamin D2 metabolites may have conferred some advantage on emerging mammalian vertebrates [44]. An unusual circumstance has been observed in the emydid turtle T. scripta, where DBP is also able to transport thyroxine on a unique binding site. The DBP of this species constitutes the major carrier of vitamin D sterols, but its serum concentration is regulated by thyroxine. Comparative studies suggest that this dual binding is unique to members of Emydidae [45,46]. The case of emydid turtles notwithstanding, DBP is structurally and functionally highly conserved among vertebrates, and this, along with the apparent evolutionary responsiveness of its functional features to ecological factors, suggests an important physiological role for this protein.
C. Polymorphisms of DBP/Gc-Globulin Protein and Gene DBP (or group-specific component, Gc) was originally characterized in humans by serum electrophoresis as the product of two autosomal, codominant alleles (Gc1 and Gc2). Isoelectric focusing allowed the further characterization of slow (Gc1S) and fast (Gc1F) subtypes of Gc1 [47,48], resulting in six common phenotypes. The protein isoforms produced by the Gc1F and Gc1S alleles represent a mixed population containing or lacking a single N-acetylneuraminic acid on a threonine residue at position 420 [49]. In contrast, the Gc2 allele produces a single protein that contains a nonglycosylated lysine at position 420. The DBP gene contains an Alu middle repetitive element located at the end of DBP intron 8 resulting in four gene polymorphisms created by differences in the size of the Alu poly(A) tract. Each of these polymorphisms is equally distributed among
121
CHAPTER 8 Vitamin D–Binding Protein
the three common alleles [50]. Although the major alleles can be distinguished at the DNA level by restriction digestions or PCR-single-strand conformation polymorphism analysis [51], protein electrophoretic techniques remain the most usual initial approach to assigning protein isoform phenotypes. Although commonly used for forensic purposes in the past [3,52], DBP phenotypying is now more frequently used to map population dynamics and to assess phenotypic susceptibility to disease (see later discussion). In addition to the three common alleles, there are also more than 124 rare variant alleles described worldwide. The geographic occurrences of these variants often correspond to patterns of human population migrations and thus are of anthropological interest [2,53–56]. The molecular bases for some of these rare variants have been determined by sequencing of exons amplified by the polymerase chain reaction [57–59]. A number of genetic studies have suggested the presence of a DBP-null allele [60,61], but these have not been clearly distinguished from low-expressing “pseudo-silent” alleles [62]. The lack of a naturally occurring DBP-null homozygote in the human population suggested that such a genotype would be embryonically lethal. A homozygous, DBP-deficient mouse line has, however, been generated that is viable and fertile [63]. DBP polymorphisms have also been identified in nonhuman primates [64,65], rodents [66], a variety of ungulates [25,66–70], domestic cats [71], marsupials [72], and birds [26,73,74]. The high degree of polymorphic variation in the human population and its widespread occurrence among vertebrates make the DBP locus among the most polymorphic known. There is some suggestion that differences in serum concentration of DBP may be attributable to phenotypic isoform [75], and that different isoforms of DBP may exhibit affinity differences for vitamin D metabolites [76,77]. However, the biological significance, if any, of these differences remains to be established. DBP isoforms have also been reported to be associated with differences in susceptibility to a wide variety of diseases and syndromes, including multiple sclerosis [78,79], autoimmune thyroid disease [80], chronic obstructive pulmonary disease [81,82], low bone mineral density in women [83], and male osteoporosis [84]. A protein that is structurally related to DBP, and thus a possible variant of DBP, has been identified as a potential serum marker for prostate cancer [85]. The nature of this protein and its relationship to DBP await its isolation and further characterization. The associations between DBP isoform and disease are, for the most part, limited to specific geographic areas or ethnic groups. It is not clear whether they apply to the human population in general, or what, if any, the biological bases and implications of these associations may be.
III. FUNCTIONAL FEATURES OF VITAMIN D BINDING PROTEIN A. Synthesis and Turnover of DBP The circulating DBP concentration in adult mammals and birds is in the micromolar range [43,44]. DBP concentration increases in the postnatal period and can be influenced by a number of factors and processes. Most studies have failed to link vitamin D status or disturbances of mineral homeostasis with alterations to DBP concentration [86–89], although there is some evidence that there may be a correlation between ligand concentration and DBP concentration in the elderly [90]. DBP concentration increases during pregnancy [91,92], and under the influence of female sex hormones in humans and birds [93–95]. In rodents, DBP concentration increases in response to testosterone and is reported to be higher in males than in females [96]. Malnutrition results in a decrease in circulating DBP concentration [97], as does liver failure [91,98], pronounced proteinuria [99,100], and increased actin–DBP complexing resulting from tissue necrosis or damage (see Section III,C,2). The specific mechanisms by which these factors and processes regulate DBP concentration are not well understood. Like albumin and other serum proteins, DBP is produced primarily in the liver, and the increase in postnatal DBP concentration to adult levels is thought to reflect maturation of the liver protein synthesis processes [86,101]. It has been estimated that 10 mg/kg/day of DBP is produced in humans [102]. The relatively low concentration of hepatic DBP mRNA suggests that the DBP transcript, like that of albumin, may be very stable and exhibit a long cytosolic half-life [10]. Spliced DBP transcripts have been identified in a range of extrahepatic sites, although at concentrations approximately 100- to 1000-fold lower than those found in the liver [101]. The role of these extrahepatic transcripts is not known, as they are not thought to be in enough abundance to significantly influence the circulating concentration of DBP, and indeed it has not yet been determined whether they are translated. DBP production is up-regulated following its increased clearance in response to trauma or hepatic necrosis [103–105] (see Section III,C,2), raising the possibility that its synthesis is regulated by a positive feedback mechanism. A range of growth factors and steroids have been demonstrated to influence DBP production by cells in culture [106,107], although the mechanisms by which these factors regulate DBP synthesis are largely unknown. It is thought that the postnatal regulation of DBP production is largely at the level of gene transcription, as is the case for α-fetoprotein and albumin [108]. Recent work has identified a locus-control
122 region for the human DBP gene, located within its large first intron. The expression of a full-length human DBP transgene in mice is abolished by deletion of this region [109, and Hiroki et al., submitted, 2004]. Expression of the DBP gene is partially under the control of HNF-1α and HNF-1β, which bind to three cis elements near its proximal promoter. HNF-1α has a major stimulatory effect whereas HNF-1β is a transdominant inhibitor of HNF-1α stimulation [110]. Both are members of the POU-homeodomain family of transcription regulators [111]. Protein kinase C has been implicated in the regulation of α-fetoprotein and albumin gene expression in the liver; however, it has not been demonstrated to regulate DBP expression [112]. The regulatory mechanisms for DBP clearance from the circulation are not known. Clearance of DBP from the plasma in the rabbit, like that of albumin, is described by a multiexponential curve with an estimated t1/2 of 1.7 days. This is less than half that of albumin with a t1/2 of 5 days, but over five times faster than that of its major ligand, 25-hydroxyvitamin D3 (25(OH)D3), suggesting that ligand recycling is likely [113]. DBP is primarily cleared by the kidneys as a result of its uptake by megalin in the proximal tubule epithelium [102,114–116] (see Section III,B,3 of this chapter and Chapter 10 of this volume). DBP–actin complexes are cleared up to three times more rapidly than unliganded DBP, primarily by hepatic filtration [117,118] (see Section III,C,2). Its association with 25(OH)D3, on the other hand, has not been reported to affect the clearance rate of DBP [113]. It has been suggested that glycosylation state may be responsible for altering clearance mechanisms of other plasma proteins, including sex hormone–binding globulin (SHBG) [119]; however, differences in glycosylation of DBP (see Section II,C) have not been reported to influence its clearance from the circulation [113]. The structural similarities between, and shared ancestry of albumin, α-fetoprotein, and DBP originally suggested overlapping regulatory mechanisms of production and turnover. It is now clear, however, that regulation of DBP synthesis and catabolism may be quite independent, and these processes await further characterization.
B. Molecular Interactions DBP associates with a number of biological molecules, including vitamin D sterols, actin, fatty acids, and various cell membrane components. These interactions will be discussed in this section, and their proposed physiological significance considered in Section C. Some of these interactions are summarized in Table I.
CHRISTOPHER J. LAING AND NANCY E. COOKE
TABLE I Features and Binding Characteristics of DBP Features Isoelectric point Electrophoretic migration Size Plasma concentration Plasma half-life Daily production rate Altered plasma levels Increased Decreased Vitamin D sterol binding Plasma capacity Normal sterol occupancy Affinity (KD) ~nanomolar ~micromolar Clearance by Actin binding Plasma capacity Normal actin occupancy Affinity (KD) DBP-actin complexes Clearance by Cell associations Renal epithelium Cell “receptor” Internalization Physiology Lymphocytes Cell “receptor” Internalization Physiology Monocytes/macrophages Cell “receptor” Internalization Physiology Neutrophils Cell “receptor” Internalization Physiology
4.5–4.8 Post-albumin, inter-α-globulin 58 kDa, single-chain glycoprotein 4–8 µM (232–464 mg/liter) 2.5–3.0 days ~10 mg/kg Estrogen, pregnancy Nephrotic syndrome, liver disease, malnutrition mol/mol (2.4 mg D sterol/liter) <5% 25OHD, 24,25(OH)2D 1,25(OH)2D, vitamin D Renal proximal tubule epithelial cells mol/mol (270 mg/liter) Trace—? mg/liter Nanomolar Seen with tissue injuries, inflammation Hepatic phagocytic cells, sinusoidal endothelium
Megalin, cubilin Yes Protein clearance, delivery of ligand ?, IgG–Fc receptor complex ? ?, conversion to DBP-maf ? ? ?, activation of phagocytic function Chondroitin sulfate proteoglycan ?, no ?, enhancement of C5a chemotaxis
1. VITAMIN D LIGANDS
As well as having weak, nonspecific associations with lipoproteins and albumin, vitamin D sterols in the extracellular fluid form a specific, high-capacity, high-affinity association with DBP [120–122]. All naturally occurring and synthetic vitamin D sterols are bound by a single binding site, but with variable affinity [reviewed in 123–126]. Among the naturally occurring vitamin D metabolites, DBP shows the greatest affinity for 25(OH)D3 and 24,25-dihydroxyvitamin D3 (24,25(OH)2D3), followed by 1,25dihydroxyvitamin D3 (1,25(OH)2D3) and parent
123
CHAPTER 8 Vitamin D–Binding Protein
vitamin D3 [4,113,42,127,128] (Table I). The vitamin D–binding site of DBP was first localized to the amino terminus of the protein using biochemical techniques such as proteolysis [129,130] (Fig. 1). Molecular modeling and labeling studies have identified important contact points and molecular orientations of a variety of vitamin D sterols within the vitamin D–binding pocket [124]. The vitamin D–binding site of DBP imposes rigid steric restrictions on its ligands. The binding pocket of the nuclear vitamin D receptor shares this feature, which contrasts with many other ligand–protein interactions such as the estrogen receptor-binding site, within which ligands are relatively free to rotate [125]. However, the actual orientations of vitamin D ligands within the DBP and vitamin D receptor-binding pockets are very different. In the latter case, the sterol is completely surrounded by the receptor protein, whereas there is indication that the DBP binding site remains “open” and that part of the sterol molecule protrudes outside of the pocket [124–126,131] (see Chapter 15). Elucidation of the crystal structure of 25(OH)D3liganded DBP has provided a more detailed understanding of the molecular interaction between these two molecules [19,20]. These structural features will pave the way to understanding how different ligands are oriented within the vitamin D–binding space and are discussed in further detail in Chapter 9 of this volume. SHBG is known to bind estradiol and C19-androgen metabolites in different orientations resulting in distinct conformational changes in the binding protein, and possibly providing a structural explanation for differences in the biological activities of these ligands [132,133]. Whether similar binding differences are observed among the various vitamin D ligands of DBP remains to be fully determined. Vitamin D binding increases the chemical [134] and thermal [122] stability of DBP, diminishes its isoelectric point [134,135], and is now known to result in rotation of DBP domain I structures [20]. The association of vitamin D ligands with DBP is thought to facilitate their dispersal throughout the body, and their delivery to sites of metabolism and action (see Section III,C,1). 2. ACTIN
The ability of DBP to form a complex with actin was first observed during cytosolic contamination of preparations of DBP-containing extracellular fluid [136,137]. DBP binds globular, but not filamentous actin in a specific, saturable, high-affinity manner (Table I). It possesses a single binding site for actin that was originally localized to the C-terminal end of the molecule [130,138] (Fig.1), but which more recently, through elucidation of the crystal structure of DBP, has been identified as involving components
of all three DBP domains [17–19,21] (for more detail see Chapter 9). Binding of actin involves rotations of the domains and flexion of some of the intradomain extended structures within the DBP molecule [17]. The extent, if any, to which the binding of actin influences the affinity of DBP for vitamin D, or vice versa, remains unclear, although evidence thus far suggests that the binding sites are structurally and functionally independent [17,19,21,139]. DBP–actin complexes have been identified in normal human plasma and have been found to increase in concentration during pregnancy and following incidents of tissue damage and necrosis. This has prompted the proposal that DBP acts in concert with other actin-binding proteins, particularly gelsolin, as a component of an extracellular actin scavenging system [140–143] (see Section III,C,2). 3. CELL ASSOCIATIONS
DBP can associate with the plasmalemma of a wide variety of cells, including renal proximal tubule epithelium [114], lymphocytes [144,145], monocytes [144–147], neutrophils [148], yolk-sac endodermal cells [149], trophoblastic cells [150,151], and spermatocytes [152]. Despite the demonstration of DBP transcripts in a number of tissues other than liver (see Section III,A), it is generally thought that this cellsurface DBP is more likely to be acquired, at least partially, from the extracellular fluid, rather than completely produced de novo [147]. A search for a specific cellsurface DBP receptor has been thus far unsuccessful, although various nonspecific cell-surface structures have been found to interact with DBP (Table I). In the renal proximal tubule epithelium, megalin, a member of the low-density lipoprotein receptor gene family, and its co-receptor cubilin are responsible for binding a wide variety of molecules, including DBP [114,115,153]. DBP is thought to bind to the IgG-Fc receptor complex of lymphocytes [154], and an elastase-sensitive chondroitin sulfate proteoglycan has been implicated as a nonsaturable DBP-binding structure on neutrophil surfaces. This chondroitin sulfate, although not specifically identified, is thought to bind multiple DBP molecules [148,155]. Although a surface receptor on monocytes has yet to be identified, the existence of a monoclonal antibody for DBP that does not recognize monocyte-associated protein [146,156], suggests a discrete structural interaction of DBP with the cell surface. The fate and function of cell-associated DBP are not fully understood in most instances, although megalin-complexed DBP is thought to deliver 25-hydroxyvitamin D3 to intracellular metabolic enzymes (see Section III,C,1,b), and neutrophilassociated DBP is thought to be involved in regulation of chemotaxis (see Section III,C,2).
124 C. Physiological Roles of DBP DBP was originally thought to be an acute phase reactant until it was found to bind vitamin D sterols and the identity of Gc-globulin and DBP were confirmed [122,157,158]. However, the subsequent discovery of its range of seemingly unrelated ligand and cell associations has prompted the suggestion that DBP is a multifunctional protein [159]. Although actions of DBP in addition to the physiological economy of vitamin D have been well documented (see Section III,B), many of the biological roles suggested for DBP are speculative, or supported by circumstantial evidence. The reported functions of DBP fall into two main categories: those involving the physiological economy and function of vitamin D and calcium homeostasis, and those involved in inflammatory processes and the immune response. It is still not clear whether these functions of DBP are independent or are physiologically linked. 1. DBP AND THE PHYSIOLOGICAL ECONOMY FUNCTION OF VITAMIN D
AND
a. The Role of Plasma Proteins in Sterol Transport and Function The sterols, and their biologically functional subset, the steroid hormones, are a group of biological molecules that are structurally related to cholesterol. They include the sex hormones, corticosteroids, thyroid hormones, retinoids, and vitamin D molecules. These molecules function in a diverse range of tissues, which necessitates distribution throughout the body. Sterols are highly lipophilic, and as such disperse poorly in the aqueous environment of the extracellular space. They do, however, readily and rapidly diffuse across cell membranes. These shared limitations necessitate transport mechanisms that allow the dispersal of sterols through the circulation, and their controlled access to cells. These requirements are met partially by nonspecific carriers such as albumin, but more importantly by a heterogeneous group of ligandspecific protein carriers. In addition to DBP, this group of proteins includes ex-hormone binding globulin (SHBG), corticosteroid-binding globulin (CBG), thyroidbinding globulin (TBG), and retinol-binding protein (RBP). These carrier proteins, while structurally unrelated, share many common functional features, reflective of the shared biochemical and physiological features of their sterol ligands. This section summarizes these shared features, along with the changing views of the functional roles of carrier proteins in steroid physiology. The specific case of DBP and vitamin D transport is considered in Section III,C,1,b. The vast majority of circulating steroids (over 90%, and in many cases up to 99%) are found in association
CHRISTOPHER J. LAING AND NANCY E. COOKE
with their specific carrier proteins. The free hormone theory was developed to describe the manner in which these proteins disperse their ligands throughout the extracellular space and make them available to target cells [160,161]. According to this theory, the ease with which the steroids can diffuse across cell membranes means that it is the small, free (or unbound) fraction of steroid in the extracellular fluid that is biologically available to cells. As such, sterol carrier proteins are thought to function by sequestering their ligands and making them unavailable to cells (and hence ensuring their continued transit in the circulation). The association between a particular protein carrier and its ligand obeys the law of mass action, where bound and “free” species exist in the following equilibrium: Ligand + BP ↔ Ligand−BP The factors that determine the proportion of free (or bioavailable) sterol are the concentration of each of the equilibrium species and the affinity of the ligand for its carrier (as well as those of other ligands competing for the same site). The metabolic clearance rates of the sex steroids and the corticosteroids are inversely proportional to their affinities for their respective specific plasma binding proteins [162,163], providing experimental support for the free hormone theory. However, the free hormone hypothesis has been challenged by a number of observations that suggest it may not fully explain the relationship between sterols, their binding proteins, and target cells. One of the most important is the direct, tissue-specific association of binding proteins with cell membranes. A specific cell membrane receptor for SHBG has been well described in a number of cell types. The binding of specific ligands to the SHBG–receptor complex initiates second-messenger cascades resulting in alterations to cell growth and differentiation, and secretion of specific proteins [reviewed in 164–166]. Thus SHBG not only acts to deliver its ligands to target cells, but directly participates in a biologically functional process that is either independent of or that augments the classic genomic actions of steroid hormones [reviewed in 167]. It has been recognized that SHBG expression occurs in a range of extrahepatic sites, many of which are sex-steroid target cells and coexpress the SHBG receptor. This suggests that SHBG may act as an autocrine or paracrine factor [168,169]. Although less well described, and more hotly contested, evidence has been presented which suggests that CBG and RBP may also bind to cell membrane receptors in a tissue-specific fashion. The physiological consequences of such associations are still largely speculative [170–174].
CHAPTER 8 Vitamin D–Binding Protein
Our original views of sterol binding proteins ascribed to them a primarily passive role in the retention of their ligands in the circulation, thus creating a pool from which controlled cellular uptake could occur. They were thought to maintain adequate stores of their ligands in the face of intermittent supply, and to buffer against potentially toxic overload. This role for the binding proteins is certainly physiologically relevant. However, more recent descriptions of receptors and cell interactions for these proteins suggest a more diverse and active function in the physiological trafficking of sterols. The functional similarities within this structurally diverse group of binding proteins suggest that some of these themes may also be shared by DBP. b. The Case of Vitamin D–Binding Protein The free hormone theory has been suggested to describe the DBP–vitamin D–cell interaction [91,160,175], and a number of lines of experimental evidence support this idea. The metabolic clearance rates of vitamin D metabolites and analogs are inversely proportional to their affinity for DBP [123,176], and DBP has been shown to limit the uptake and action of 1,25(OH)2D3 in cells grown in vitro [177,178]. Analogs of 1,25(OH)2D3 with a low affinity for DBP stimulate vitamin D– responsive gene expression more rapidly than those with higher affinity [179], and mice not expressing DBP demonstrate a more rapid genomic response to 1,25(OH)2D3 than their wild-type littermates [63]. It has been proposed that the pathophysiology of vitamin D toxicity may involve the dislodging from DBP and resultant increased bioavailability of 1,25(OH)2D3 by the more abundant and avidly bound 25(OH)D3 [63,180]. Several experimental observations, however, suggest that the free hormone hypothesis may not fully explain the role of DBP in vitamin D transport and function. Its high plasma concentration and its high ligand affinity (Table I) ensure that DBP binds between 85% and 98% of each circulating vitamin D sterol, depending upon the metabolite, while the remainder is considered “free” or in association with low-affinity carriers [91,127,181]. On the other hand, only 5% of DBP molecules are occupied by a vitamin D sterol ligand [130]. Using a range of estimates of “free” 1,25(OH)2D3 concentration [91,127,176,181,182], the concentration of bioavailable 1,25(OH)2D3 in the extracellular fluid is still approximately an order of magnitude less than the dissociation constant of the intracellular vitamin D receptor. Thus if only the free fraction of hormone were available for cellular uptake, there would be very little likelihood of high receptor occupancy. This suggests that proteinmediated ligand uptake by cells may play an important role in the physiology of vitamin D metabolism and function.
125 Megalin–DBP complexes are internalized by the renal proximal tubular cells, and DBP is thought to undergo proteolysis after receptor and ligand scavenging. This results in the delivery of vitamin D sterols to intracellular binding sites such as metabolic enzymes and the vitamin D receptor, via specific cytosolic chaperones, members of the heat-shock protein-70 family [183]. This process constitutes an important step in the conversion of 25(OH)D3 to the biologically active 1,25(OH)2D3 [114,116] (see Chapter 10). The association of DBP with other cell types may involve further specific or nonspecific receptors (see Section III,C,2). These receptors may also be involved in ligand internalization, or may regulate cell signaling as has been described for cell surface receptors for SHBG (see Section III,C,1,a). 1,25(OH)2D3 and 24,25(OH)2D3 can exert effects in a variety of cells via rapid, nongenomic pathways thought to involve novel cell membrane receptors [184,185] (see Chapter 23). This is a theme that has also been described for sex hormones, which can stimulate cell-signaling cascades directly via sex-steroid membrane receptors, or via SHBG receptors [167]. It is tantalizing to speculate that some of the non-genomic actions of vitamin D metabolites may be mediated via interactions of DBP with cell surface receptors. The demonstration of DBP transcripts in a range of extrahepatic tissues (see Section III,A) suggests that DBP produced at these sites may be involved in functions not directly related to the dispersal of vitamin D ligands in the circulation and may act as an autocrine or paracrine factor as has been demonstrated for SHBG (see Section III,C,1,a). Further work is required to elucidate the mechanisms by which DBP regulates the cellular uptake, metabolism, and biological activities of its ligands. However, it is clear that whether the bioavailability of vitamin D is dependent upon the “free hormone” model, or protein-mediated pathways, or a combination of the two, DBP plays an important role in regulating the physiological economy of its sterol ligands. When fed a normal diet, mice in which the DBP locus has been inactivated do not demonstrate any specific phenotype that distinguishes them from wild-type littermates [63]. This “mild” phenotype is in contrast to mouse models in which the genes for the vitamin D receptor [186] or enzymes of metabolism of vitamin D [187,188] have been inactivated. Rather mild phenotypic differences have also been observed in mouse models and natural human mutations resulting in no or low expression of CBG [189,190], TBG [191,192], and RBP [193], suggesting that the mechanisms by which sterols arrive at and enter cells are likely to be tissue specific and to possess a certain element of redundancy. It has been suggested that the nonspecific carrier
126 proteins, such as albumin, may play a more than trivial role in the transport and processing of sterols [194]. However, in the case of both DBP–null [63] and RBPnull [193] mice, the absence of specific binding protein renders the animal more susceptible to derangements associated with abnormal intake or production of the respective ligand. The intermittent nature of vitamin D production or intake suggests that an important evolutionary advantage may be conferred upon organisms possessing an efficient plasma binding protein for the vitamin D sterols. That interspecies variations in DBP concentration and 25OHD3 affinity correlate more closely with ecological factors than with phylogenetic distance (see Section II,B,3) suggests that the properties of this protein have been important determinants of the conservation or activity of its vitamin D ligands during vertebrate evolution. Furthermore, there is some evidence to suggest that the concentration of DBP may be altered during times of increased calcium demand, such as at the initiation of lactation [44]. Possible links among malnutrition, DBP concentration, and the development of rachitic syndromes [97,195,196] also support the important role played by DBP in the regulation of the physiological economy of its vitamin D ligands. 2. THE ROLE OF DBP IN INFLAMMATORY PROCESSES SYSTEM
AND IN THE IMMUNE
Inflammation is a complex tissue reaction to physical injury, to infection, or to a local immune response and is characterized by changes to the local vasculature and the accumulation of fluid, biologically active proteins, and white blood cells. The pathogenesis and development of both local and systemic inflammation is intricately linked to these vascular and cellular responses. DBP has been associated with a number of steps within the inflammatory process, including the prevention of thromboembolic events in the microvasculature by actin, the stimulation of chemotaxis by phagocytic neutrophils, and the activation and stimulation of phagocytic function by macrophages. In this section, the evidence for a physiological role for DBP in the inflammatory process and in immune function is discussed. Tissue damage or necrosis, either due to or as a sequel to a primary insult, is an early component of the general inflammatory process and results in the release of intracellular proteins such as filamentous actin. Free actin in the circulation results in the development of thromboemboli in the microvasculature and the disruption of normal coagulation events [197]. This leads to tissue ischemia and multiple organ dysfunction syndrome. Actin may also be directly toxic to pulmonary
CHRISTOPHER J. LAING AND NANCY E. COOKE
endothelial cells [198]. DBP and another plasma protein, gelsolin, have been hypothesized to act in a coordinated fashion to dismantle actin filaments, sequester monomeric actin, and facilitate its clearance from the circulation [18,21,117,118,197]. In support of this hypothesis, circulating DBP concentrations are decreased and DBP–actin complexes increased in a variety of circumstances associated with tissue necrosis or cell disruption, such as fulminant hepatic necrosis [104,105,199], septic shock [200], and multiple trauma [103,201,202]. This reduced concentration of DBP is thought to be the result of the rapid removal of DBP–actin complexes in the liver [104,203] (see Section III,A). During recovery from liver disease or tissue trauma, plasma DBP concentration rises above the normal range, perhaps indicating a compensatory stimulation of DBP synthesis, which has been suggested to result from cytokine activity [102–105]. The priority for maintenance of plasma DBP concentration by the liver is similar to that of prothrombin index, and higher than that for plasma albumin concentration [98]. Nevertheless, the actin scavenging system appears to be saturable [141] and an association between low plasma DBP concentration and the development of multiple organ dysfunction has been documented in patients with fulminant hepatic necrosis and following severe injury [199,202]. An early response to tissue damage is the recruitment and activation of phagocytic cells. In the immune system, chemotaxis involves the directional migration of cells in response to an extracellular cue, usually in the form of a chemical gradient. A wide variety of chemoattractants have been identified, with varying degrees of specificity. A serum enhancer of C5a-stimulated chemotaxis of leukocytes was subsequently identified to be DBP. The mechanism(s) by which DBP enhances neutrophil response to C5a are unknown, but several studies document that such an effect exists [155,204–207]. Preincubation of DBP with neutrophils is sufficient for enhanced C5a-stimulated chemotaxis, which is apparently mediated via DBP–cell surface proteoglycan interaction [148,205,207] (see Section III,B,3). Although cell membrane–associated proteoglycans have been suggested to facilitate the uptake of their ligands [208], DBP bound to the surface of neutrophils does not appear to be internalized [155]. Some cell-surface proteoglycans are known to possess cytoplasmic domains that interact with components of the cytoskeleton [208], raising the possibility that the DBP–chondroitin sulfate complex is involved in cell signaling. It is now known that a neutrophil-associated serine protease, elastase, establishes the steady-state binding of DBP to neutrophil surfaces by cleaving and thus regulating the number of available chondroitin-sulfate binding sites. The establishment of steady-state binding is essential
127
CHAPTER 8 Vitamin D–Binding Protein
for cochemotactic activity, suggesting that the mechanism by which DBP acts as a cochemotaxin may involve dynamic proteoglycan signaling in the cytosol, or that localization of the DBP–receptor complex to active areas of the cell membrane is required [155]. Antigenic stimulants cause the activation of phagocytic function in leukocytes and other cells via several pathways. In one such pathway, cell death and inflammation increase the activity of membrane phospholipases, resulting in the accumulation of lysophospholipids and other phospholipid degradation products, which are potent stimulators of macrophage activation [209,210]. It has been demonstrated ex vivo and in vitro that this activation may occur through an intermediate macrophage-activating factor, the partially deglycosylated DBP (DBP-maf). DBP is converted to DBP-maf via the inducible, stepwise actions of B-lymphocytic β-galactosidase and T-lymphocytic sialidase [211–213]. DBP-maf activity is specific to Fcγ-mediated phagocytosis and does not appear to be involved in complementprimed phagocytic processes that can be stimulated by such factors as lipopolysaccharide [210]. DBP-maf activated macrophages demonstrate significant tumoricidal activity [214–216]. The specific stimulants for DBP-maf generation and the particular phagocytic pathways that it activates suggest that the principal role of DBP-maf may be in the immune response to neoplasia. The exact mechanism by which DBP-maf acts is unknown; however, the complete deglycosylation of DBP-maf by the removal of the terminal alphaN-acetylgalactosamine (GalNAc) moiety results in diminished macrophage-activating activity. Furthermore, the accumulation of tumor-derived alpha-N-acetylgalactomanidase in advanced cancer patients has been linked to immunosuppression. This suggests that DBPmaf may stimulate Fcγ translocation to the cell surface via GalNAc-specific lectin on the macrophage surface [217–219]. DBP maf has also been reported to stimulate bone-resorptive activities of osteoclasts in vitro [220]. It has been proposed that a defect in lymphocytic β-galactosidase and/or some other deficiency in DBPmaf generation may be involved in the pathogenesis of some osteopetrosis syndromes [221–223]. Most recently, it has been suggested that DBP–maf may have direct antiangiogenic effects on endothelial cells and be able to inhibit the angiogenic actions of a number of tumor-derived growth factors. This activity is proposed to operate via the CD36 receptor, expressed on both macrophages and endothelial cells [224,225], although the nature of this proposed activity awaits more thorough characterization. Given the seemingly diverse and integral roles played by DBP in inflammatory and immune processes, it was surprising that mice in which DBP expression had
been disrupted were viable [63]. Indeed, DBP-null mice show no obvious phenotype that can be related to deficient immune function. No differences in phagocytic cell recruitment in response to intraperitoneal treatment with thioglycollate medium [226] or to infection with Listeria monocytogenes or Toxoplasma gondii [White et al., submitted, 2004.] could be attributed to the inactivated DBP allele. DBP-null mice do not demonstrate osteopetrotic lesions [63], causing some uncertainty about the obligatory nature of DBP-maf in osteoclastic function. The reasons for these apparent discrepancies between in vitro or ex vivo actions of DBP and observations in the DBP-null mouse are as yet unclear. The processes of immunity and inflammation reflect complex genetic determinants. Strain differences in peritoneal recruitment of leukocytes have been identified in mice [226], and species differences are likely to be more profound still. Furthermore, phagocyte site specificity, as well as the nature of the target particles, may differentially determine the nature and regulation of phagocytic processes [227]. Therefore, the lack of observed phenotypic difference between DBP-null and wild-type mice in relation to inflammation and immunity may reflect species-, strain-, tissue-, and/or pathologyspecific factors. Immune defects related to the inactivated DBP allele, if any, may not become apparent without producing the appropriate specific challenges to the immune system. Alternatively, it is possible that the “non–vitamin D binding” actions of DBP may not represent an obligate role for DBP in inflammation and immunity, or may be biologically redundant. A third possibility is that they may be involved in an as-yetunidentified integrated physiological role for DBP. Further experimentation is required to resolve these points.
IV. CONCLUDING REMARKS Several important technical steps have advanced our understanding of the structural and functional features of DBP. The elucidation of the sequences of both the full-length DBP gene and its messenger RNA in a variety of vertebrates has revealed the relationship between this protein and the albumin/α-fetoprotein multigene family, and their place in vertebrate evolution. Comparisons of the primary structures of these proteins have identified differences in DBP that pointed toward a structural explanation for its unique functional activities. Descriptions of the crystal structure of DBP in complex with two of its ligands, 25(OH)D3 and actin, have extended this explanation to the tertiary structural level. The generation of a mouse line with an inactivated DBP locus has provided an important tool for the study
128 of the integrated physiology of DBP. It has enabled the demonstration of the important role played by DBP in the turnover and actions of its vitamin D ligands. Further investigation of non–vitamin D functions of DBP using this mouse model are underway. While our understanding of its actions and their molecular bases is increasing, further work is required to clarify the physiological role(s) of DBP in the whole organism. Functional similarities reported among the other sterol binding proteins may hold important clues for future investigations of the physiology of DBP.
References 1. Hirschfeld J, Jonsson B, Rasmusson M 1959 Immune electrophoretic demonstration of qualitative differences in human sera and their relation to the haptoglobins. Acta Pathol Microbiol 47:160–168. 2. Cleve H, Constans J 1988 The mutants of the vitamin D–binding protein: more than 120 variants of the Gc/DBP system. Vox Sang 54:215–225. 3. Westwood WA, Werret DJ 1986 Group specific component: a review of the isoelectric focusing methods and auxiliary methods available for the separation of its phenotypes. Forensic Sci Int 32:135–150. 4. Haddad JG, Walgate J 1976 25-Hydroxyvitamin D transport in human plasma. Isolation and partial characterisation of calcifidiol-binding protein. J Biol Chem 251:4803–4809. 5. Braun A, Kofler A, Morawietz S, Cleve H 1993 Sequence and organization of the human vitamin D–binding protein gene. Biochim Biophys Acta 1216:385–394. 6. Witke FW, Gibbs PEM, Zielinski R, Yang F, Bowman BH, Dugaiczyk A 1993 Complete structure of the human Gc gene: differences and similarities between members of the albumin gene family. Genomics 16:751–754. 7. Ray K, Wang XK, Zhao M, Cooke NE 1991 The rat vitamin D binding protein (Gc globulin) gene: structural analysis, functional and evolutionary correlations. J Biol Chem 266: 6221–6229. 8. Cooke NE, David EV 1985 Serum vitamin D binding protein is a third member of the albumin and α-fetoprotein gene family. J Clin Invest 76:2420–2424. 9. Yang F, Brune JL, Naylor SL, Cupples RL, Naberhaus KH, Bowman BH 1985 Human group-specific component (Gc) is a member of the albumin family. Proc Natl Acad Sci USA 82:7994–7998. 10. Cooke NE 1986 Rat vitamin D binding protein: determination of the full length primary structure from cloned cDNA. J Biol Chem 261:3441–3450. 11. Yang F, Bergeron JM, Linehan LA, Lalley PA, Sakaguchi AY, Bowman BH 1990 Mapping and conservation of the group-specific component gene in mouse. Genomics 7: 509–516. 12. Osawa M, Tsuji T, Yukawa N, Saito T, Takeichi S 1994 Cloning and sequence analysis of cDNA encoding rabbit vitamin D binding protein (Gc globulin). Biochem Mol Biol Int 34:1003–1009. 13. Burnside J, Sofer L 1999 Gallus gallus vitamin D binding protein cDNA. GenBank direct submission; accession number AF121350.
CHRISTOPHER J. LAING AND NANCY E. COOKE
14. Hunt JL, Licht P 1998 Identification and structural characterization of a novel member of the vitamin D binding protein family. Comp Biochem Physiol 121B:397–406. 15. Schoentgen F, Metz-Boutigue MH, Jolles J, Constans J, Jolles P 1986 Complete amino acid sequence of human vitamin D binding protein (group-specific component): evidence of a three fold internal homology as in serum albumin and alphafetoprotein. Biochim Biophys Acta 871:189–198. 16. Litwiller R, Fass R, Kumar R 1987 The amino acid sequence of the NH2-terminal portion of the rat and human vitamin D binding protein: evidence for a high degree of homology between rat and human vitamin D binding protein. Life Sci 38:2179–2184. 17. Head JF, Swamy N, Ray R 2002 Crystal structure of the complex between actin and human vitamin D binding protein at 2.5Å resolution. Biochemistry 41:9015–9020. 18. Otterbein LR, Cosio C, Graceffa P, Dominguez R 2002 Crystal structures of the vitamin D–binding protein and its complex with actin: structural basis of the actin-scavenger system. Proc Natl Acad Sci USA 99:8003–8008. 19. Swamy N, Head JF, Weitz D, Ray R 2002 Biochemical and preliminary crystallographic characterization of the vitamin D sterol- and actin-binding by human vitamin D–binding protein. Arch Biochem Biophys 402:14–23. 20. Verboven C, Rabijns A, De Maeyer M, Van Baelen H, Bouillon R, De Ranter C 2002 A structural basis for the unique binding features of the human vitamin D–binding protein. Nature Struct Biol 9:131–136. 21. Verboven C, Bogaerts I, Waelkens E, Rabijns A, Van Baelen H, Bouillon R, De Ranter C 2003 Actin–DBP: the perfect structural fit? Acta Crystallogr D Biol Crystallogr 59:263–273. 22. Cooke NE, Williard HF, David EV, George DL 1986 Direct regional assignment of the gene for vitamin D binding protein (Gc-globulin) to human chromosome 4q11–q13 and identification of an associated DNA polymorphism. Hum Genet 73:225–229. 23. Harper ME, Dugaiczyk A 1983 Linkage of the evolutionarilyrelated serum albumin and α-fetoprotein genes with q11–q22 of human chromosome 4. Am J Hum Genet 35:565–572. 24. Cooke NE, Levan G, Szpirer J 1987 The rat vitamin D binding protein (Gc-globulin) gene is syntenic with the rat albumin and α-fetoprotein genes on chromosome 14. Cytogenet Cell Genet 44:98–100. 25. Weitkamp LR, Allen PZ 1979 Evolutionary conservation of equine Gc alleles and of mammalian Gc/albumin linkage. Genetics 92:1347–2272. 26. Juneja RK, Sandberg K, Andersson L, Gahne B 1982 Close physical linkage between albumin and vitamin D binding protein (Gc) loci in chicken: a 300 million year old linkage group. Genet Res Cambridge 40:95–98. 27. Lichenstein HS, Lyons DE, Wurfel MM, Johnson DA, McGinley MD, Leidli JC, Trollinger DB, Mayer JP, Wright SD, Zukowski MM 1994 Afamin is a new member of the albumin, α-fetoprotein, and vitamin D–binding protein gene family. J Biol Chem 269:18149–18154. 28. Belanger L, Roy S, Allard D 1994 New albumin gene 3′ adjacent to the α1-fetoprotein locus. J Biol Chem 269:5481–5484. 29. Gabant P, Forrester L, Nichols J, Van Reeth T, De Mees C, Pajack B, Watt A, Smitz J, Alexandre H, Szpirer C, Szpirer J 2002 Alpha-fetoprotein, the major fetal serum protein, is not essential for embryonic development but is required for female fertility. Proc Natl Acad Sci USA 99:12865–12870. 30. Song Y-H, Naumova AK, Liebhaber SA, Cooke NE 1999 Physical and meiotic mapping of the region of human chromosome 4q11–q13 encompassing the vitamin D binding
CHAPTER 8 Vitamin D–Binding Protein
31. 32.
33. 34. 35. 36.
37. 38. 39. 40. 41. 42. 43.
44. 45.
46.
47. 48.
49.
protein DBP/Gc-globulin and albumin multigene cluster. Genome Res 9:581–587. Urano Y, Sakai M, Watanabe K, Tamaoki T 1984 Tandem arrangement of the albumin and alpha-fetoprotein genes in the human genome. Gene 32:255–261. Chevrette M, Guertin M, Turcotte B, Belanger L 1987 The rat α1-fetoprotein gene: characterization of the 5′-flanking region and tandem organization with the albumin gene. Nucleic Acids Res 15:1338–1339. The Jackson Laboratory, Bar Harbor, Maine 2003 Mouse Genome Informatics Web Site. URL: http://www.informatics.jax. org. August. Eiferman FA, Young PR, Scott RW, Tilghman SM 1981 Intragenic amplification and divergence in the mouse α-fetoprotein gene. Nature 294:713–718. Gibbs EM, Dugaiczyk A 1987 Origin of structural domains of the serum albumin gene family and a predicted structure of the gene for vitamin D–binding protein. Mol Biol Evol 4:364–379. Haefliger DN, Moskaitis JE, Schoenberg DR, Wahli W 1989 Amphibian albumins as members of the albumin, alpha-fetoprotein, vitamin D–binding protein multigene family. J Mol Evol 29:344–354. Gibbs PEM, Witke WF, Dugaiczyk A 1998 The molecular clock runs at different rates among closely related members of a gene family. J Mol Evol 46:552–561. Brown MA, Chambers GK, Licht P 1997 Purification and partial amino acid sequences of two distinct albumins from turtle plasma. Comp Biochem Physiol 118B:367–374. Nishio H, Hamdi HK, Dugaiczyk A 1999 Genomic expansion across the albumin gene family on human chromosome 4q is directional. Biol Chem 380:1431–1434. Hay AWM, Watson G 1976 The plasma transport of 25-hydroxycholecalciferol in fish, amphibians, reptiles and birds. Comp Biochem Physiol 53:167–172. Van Baelen H, Allewaert K, Bouillon R 1988 New aspects of the plasma carrier for 25-hydroxycholecalciferol in vertebrates. Ann NY Acad Sci 538:60–68. Laing CJ, Fraser DR 1999 The vitamin D system in iguanian lizards. Comp Biochem Physiol 123B:373–379. Laing CJ, Shea GM, Fraser DR 1999 Variation in the concentration and binding affinities of the plasma vitamin D–binding protein. In: Danks J, Dacke C, Flik G, Gay C (eds) Calcium Metabolism: Comparative Endocrinology. BioScientifica, Bristol, UK, pp. 93–98. Laing CJ 2000 Comparative Biology of Plasma Vitamin D Binding Protein. Ph.D. thesis. Department of Animal Science, University of Sydney, Sydney, Australia (http://adt.caul.edu.au). Licht P 1994 The relation of the dual thyroxine/vitamin D– binding protein (TBP/DBP) of emydid turtles to the vitamin D– binding proteins of other vertebrates. Gen Comp Endocrinol 94:215–224. Whitworth DJ, Hunt L, Licht P 2000 Widespread expression of the mRNA encoding a novel vitamin D/thyroxine dual binding protein in the turtle Trachemys scripta. Gen Comp Endocrinol 118:354–358. Constans J, Viau M 1977 Group-specific component: evidence for two subtypes of the Gc1 gene. Science 198:1070–1071. Hoste B 1979 Group specific component (Gc) and transferrin (Tf) subtypes ascertained by isoelectric focusing. A simple non-immunological staining procedure for Gc. Hum Genet 50:75–79. Cleve H, Patutschnick W 1979 Neuraminidase treatment reveals sialic acid differences in certain genetic variants of the Gc system (vitamin D binding protein). Hum Genet 47:193–198.
129 50. Braun A, Bichlmaier R, Muller B, Cleve H 1993 Molecular evaluation of an Alu repeat including a polymorphic variable poly(dA) (AluVpA) in the vitamin D binding protein (DBP) gene. Hum Genet 90:526–532. 51. Dissing J, Thymann M, Hopkinson D 2003 Simultaneous detection of ACP1 and GC genotypes using PCR/SSCP. Ann Hum Genet 67:81–85. 52. Thymann M 1981 Gc subtypes determined by agarose isoelectric focusing. Distribution in Denmark, and application to paternity cases. Hum Hered 31:214–221. 53. Lai LYC, Constans J, Archer GT 1990 A new cathodal Gc variant in Australia. Ann Hum Biol 17:526–532. 54. Chen LZ, Easteal S, Board PG, Summers KM, Bhatia KK, Kirk RL 1990 Albumin–vitamin D–binding protein haplotypes in Asian-Pacific populations. Hum Genet 85:89–97. 55. Kravchuk OI, Balanovskii OP, Nurbaev SD, Makarov SV, Spitsyn VA, Ginter EK 1998 Genogenography of the aboriginal population of Marii El (from data on immunobiochemical polymorphism). Genetika 34:1542–1554. 56. Yin J, Zhang G, Liu J 2000 Genetic polymorphism of groupspecific component/vitamin-D–binding protein subtypes in six populations from north-eastern China. Anthropol Anzeiger 58:193–198. 57. Kofler A, Braun A, Jenkins T, Serjeanston SW, Cleve H 1995 Characterization of mutants of the vitamin D–binding protein/ group specific component: GC aborigine (1A1) from Australian aborigines and South African blacks, and 2A9 from south Germany. Vox Sang 68:50–54. 58. Yuasa I, Kofler A, Braun A, Bichlmaier R, Kammerer S, Cleve H 1995 A novel sequence polymorphism in exon 8 of the human vitamin D–binding protein (GC) gene in an African population. Jap J Hum Genet 40:149–151. 59. Yuasa I, Kofler A, Braun A, Umetsu K, Bichlmaier R, Kammerer S, Cleve H 1995 Characterization of mutants of the vitamin D–binding protein/group-specific component: molecular evolution of GC*1A2 and GC*1A3, common in some Asian populations. Hum Genet 95:507–512. 60. Dykes D, Polesky H, Cox E 1981 Isoelectric focusing of Gc (vitamin D binding globulin) in parentage testing. Hum Genet 58:174–175. 61. Yoshifumi Y, Nishimoto H, and Ikemoto S 1989 Interstitial deletion of the proximal long arm of chromosome 4 associated with father-child incompatibility within the Gc-system: probable reduced gene dosage effect and partial piebald trait. Am J Med Genet 32:520–523. 62. Vavrusa B, Cleve H, Constans J 1983 A deficiency mutant of the Gc system. Hum Genet 65:102–107. 63. Safadi FF, Thornton P, Magiera H, Hollis BW, Gentile M, Haddad JG, Liebhaber SA, Cooke NE 1999 Osteopathy and resistance to vitamin D toxicity in mice null for vitamin D binding protein. J Clin Invest 103:239–251. 64. Cleve H, Patutschnick W 1979 Different phenotypes of the group-specific component (Gc) in chimpanzees. Hum Genet 50:217–220. 65. Tsuchida S, Nagai A, Ikemoto S 1990 Polymorphisms of the group-specific component (Gc) in three species of Tamarins (Sanguinus oedipus, S. labiatus, S. mystax). Jpn J Vet Sci 52:649–651. 66. Imawari M, Akanuma Y, Muto Y, Itakura H, Kosaka K 1980 Isolation and partial characterization of two immunologically similar vitamin D–binding proteins in rat serum. J Biochem 88:349–360. 67. Gahne B, Juneja RK 1978 Polymorphic post-albumin of cattle and horse plasma identified as vitamin D binding
130
68. 69. 70. 71. 72. 73. 74. 75. 76.
77. 78.
79. 80.
81.
82.
83.
84.
protein (Gc protein). Anim Blood Groups Biochem Genet 9:37–40. Tate ML, Dodds KG, Thomas KJ, McEwan KM 1992 Genetic polymorphism of plasminogen and vitamin D binding protein in red deer (Cervus elephus L). Anim Genet 23:303–313. Vankan DM, Bell K 1992 Genetic polymorphism of plasma vitamin D–binding protein (Gc) in Australian goats. Anim Genet 23:457–462. Shin JA, Yang YH, Kim HS, Yun YM, Lee KK 2002 Genetic polymorphism of the serum proteins of horses in Jeju. J Vet Sci 3:255–263. Juneja RK, Niini T, Larsson HEB, Sandberg K 1991 Genetic polymorphism of six plasma proteins in the domestic cat. J Hered 82:178–181. Arthur H, Bell K, Cooper DW 1998 Plasma protein polymorphisms in the tammar wallaby (Macropus eugenii). Aust J Zool 46:193–199. Juneja RK, Gahne B, Kuryl J, Gasparka J 1982 Genetic polymorphism of the vitamin D–binding protein and a pre-transferrin in chicken plasma. Hereditas 98:89–96. Shibata T, Abe T 1996 Linkage between the loci for serum albumin and vitamin D binding protein (Gc) in the Japanese quail. Anim Genet 27:195–197. Lauridsen AL, Vestergaard P, Nexo E 2001 Mean serum concentration of vitamin D-binding protein (Gc globulin) is related to the Gc phenotype in women. Clin Chem 47:753–756. Braun A, Brandhofer A, Cleve H 1990 Interaction of the vitamin D–binding protein (group-specific component) and its ligand 25-hydroxy-vitamin D3: binding differences of the various genetic types disclosed by isoelectric focusing. Electrophoresis 11:478–483. Arnaud J, Constans J 1993 Affinity differences for vitamin D metabolites associated with the genetic isoforms of the human serum carrier protein (DBP). Hum Genet 92:183–188. Arnason A, Jensson O, Skaftadottir I, Birgisdottir B, Gudmundsson G, Johannesson G 1980 HLA types, Gc protein and other genetic markers in multiple sclerosis and two other neurological diseases in Iceland. Acta Neurol Scand 62(Suppl 78):39–40. Niino M, Kikuchi S, Fukazawa T, Yabe I, Tashiro K 2002 No association of vitamin D–binding protein gene polymorphisms in Japanese patients with MS. J Neuroimmunol 127:177–179. Pani MA, Regulla K, Segni M, Hoffmann S, Hufner M, Pasquino AM, Usadel KH, Badenhoop K 2002 A polymorphism within the vitamin D–binding protein gene is associated with Graves’ disease but not with Hashimoto’s thyroiditis. J Clin Endocrinol Metab 87:2564–2567. Ishii T, Keicho N, Teramoto S, Azuma A, Kudoh S, Fukuchi Y, Ouchi Y, Matsuse T 2001 Association of Gc-globulin variation with susceptibility to COPD and diffuse panbronchitis. Eur Respir 18:753–757. Sandford AJ, Chagani T, Weir TD, Connett JE, Anthonisen NR, Pare PD 2001 Susceptibility genes for rapid decline of lung function in the lung health study. Am J Resp Crit Care Med 163:469–473. Ezura Y, Nakajima T, Kajita M, Ishida R, Inoue S, Yoshida H, Suzuki T, Shiraki M, Hosoi T, Orimo H, Emi M 2003 Association of molecular variants, haplotypes, and linkage disequilibrium within the human vitamin D–binding protein (DBP) gene with postmenopausal bone mineral density. J Bone Miner Res 18:1642–1649. Papiha SS, Allcroft LC, Kanan RM, Francis RM, Datta HK 1999 Vitamin D binding protein gene in male osteoporosis: association of plasma DBP and bone mineral density with
CHRISTOPHER J. LAING AND NANCY E. COOKE
85.
86.
87. 88. 89. 90.
91.
92.
93. 94.
95. 96. 97. 98.
99.
100. 101.
(TAAA)(n)-Alu polymorphism in DBP. Calcif Tissue Int 65:262–266. Hlavaty JJ, Partin AW, Shue MJ, Mangold LA, Derby J, Javier T, Kelley S, Stieg A, Briggman JV, Hass GM, Wu Y-J 2003 Identification and preliminary clinical evaluation of a 50.8-kDa serum marker for prostate cancer. Urology 61:1261–1265. Haddad JG, Walgate J 1976 Radioimmunoassay of the binding protein for vitamin D and its metabolites in human serum. Concentrations in normal subjects and patients with disorders of mineral homeostasis. J Clin Invest 67:1550–1560. Bouillon R, Van Baelen H, DeMoor P 1977 The measurement of vitamin D–binding protein in human serum. J Clin Endocrinol Metab 45:225–231. Imawari M, Goodman DS 1977 Immunological and immunoassay studies of the binding protein for vitamin D and its metabolites in human serum. J Clin Invest 59:432–442. Parviainen M, Harmoinen A, Visakorpi J 1982 Immunonephelometric quantification of serum vitamin D–binding globulin. Scand J Clin Lab Invest 42:571–575. Rejnmark L, Lauridsen AL, Vestergaard P, Heickendorff L, Andreasen F, Mosekilde L 2002 Diurnal rhythm of plasma 1,25-dihydroxyvitamin D and vitamin D–binding protein in postmenopausal women: relationship to plasma parathyroid hormone and calcium and phosphate metabolism. Eur J Endocrinol 146:635–642. Bikle DD, Gee E, Halloran B, Haddad JG 1984 Free 1,25dihydroxyvitamin D levels in serum from normal subjects, pregnant subjects, and subjects with liver disease. J Clin Invest 74:1966–1971. Abbas SK, Care AD, Van Baelen H, Bouillon R 1987 Plasma vitamin D–binding protein and free 1,25-dihydroxyvitamin D3 index in pregnant ewes and their fetuses in the last month of gestation. J Endocrinol 115:7–12. Fraser DR, Emtage JS 1976 Vitamin D in the avian egg. Its molecular identity and mechanism of incorporation into yolk. Biochem J 160:671–682. Cheema C, Grant BF, Marcus R 1989 Effects of estrogen on circulating “free” and total 1,25-dihydroxyvitamin D and on parathyroid–vitamin D axis in postmenopausal women. J Clin Invest 83:537–542. Harris SS, Dawson-Hughes B 1998 The association of oral contraceptive use with plasma 25-hydroxyvitamin D levels. J Am Coll Nutr 17:282–284. Bouillon R, Vandoren G, Van Baelen H, DeMoor P 1978 Immunochemical measurement of the vitamin D–binding protein in rat serum. Endocrinology 102:1710–1715. Laing CJ, Fraser DR 2002 Changes with malnutrition in the concentration of plasma vitamin D–binding protein (DBP) in growing rats. Br J Nutr 88:133–139. Schiodt FV, Bondesen S, Muller K, Rasmussen A, Hjortrup A, Kirkegaard P, Hansen BA, Tygstrup N, Ott P 1999 Reconstitution of the actin-scavenger system after orthotopic liver transplantation for end-stage liver disease: a prospective and longitudinal study. Liver Transplant Surg 5:310–317. Nyomba BL, Bouillon R, Lissens W, Van Baelen H, De Moor P 1985 1,25-Dihydroxyvitamin D and vitamin D–binding protein are both decreased in streptozotocin-diabetic rats. Endocrinology. 116:2483–2488. Auwerx J, De Keyser L, Bouillon R, De Moor P 1986 Decreased free 1,25-dihydroxycholecalciferol index in patients with the nephrotic syndrome. Nephron 42:231–235. McLeod JF, Cooke NE 1989 The vitamin D–binding protein, α-fetoprotein, albumin multigene family: detection of transcripts in multiple tissues. J Biol Chem 264:21760–21769.
CHAPTER 8 Vitamin D–Binding Protein
102. Kawakami M, Blum CB, Ramakrishan R, Dell RB, Goodman DS 1981 Turnover of the plasma binding protein for vitamin D and its metabolites in normal human subjects. J Clin Endocrinol Metab 53:1110–1116. 103. Dahl B, Schiodt FV, Rudolph S, Ott P, Kiaer T, Heslet L 2001 Trauma stimulates the synthesis of Gc-globulin. Inten Care Med 27:394–399. 104. Schiodt FV, Ott P, Tygstrup N, Dahl B, Bondesen S 2001 Temporal profile of total, bound, and free Gc-globulin after acetaminophen overdose. Liver Transplant 7:732–738. 105. Schiodt FV, Clemmesen JO, Bondesen S, Dahl B, Ott P 2001 Increased turnover of Gc-globulin in patients with hepatic encephalopathy. Scand J Gastroenterol 36:998–1003. 106. Egawa T, Ito H, Nakamura H, Yamamoto H, Kishimoto S 1992 Hormonal regulation of vitamin D–binding protein production by a human hepatoma cell line. Biochem Int 28:551–557. 107. Guha C, Osawa M, Werner PA, Galbraith RM, Paddock GV 1995 Regulation of human Gc (vitamin D–binding) protein levels: hormonal and cytokine control of gene expression in vitro. Hepatology 21:1675–1681. 108. Danan JL, Ribiero V, Lechner MC, Pompom D 1991 Expression of the rat vitamin D–binding protein gene. In: Norman AW, Bouillon R, Thomasset M (eds) Vitamin D. Gene Regulation, Structure-Function Analysis and Clinical Application. Proceedings of the Eighth Workshop on Vitamin D. Walter de Gruyter, New York, pp. 697–698. 109. Hiroki T, Song Y-H, Naumova AK, Liebhaber SA, Cooke NE 2000 A human vitamin D binding protein gene with limited flanking sequences is sufficient for liver-specific and copy number-dependent expression in transgenic mice. The Endocrine Society, 82nd Annual Meeting, Toronto, June 21–25, (poster). 110. Song Y-H, Ray K, Liebhaber SA, Cooke NE 1998 Vitamin D– binding protein gene transcription is regulated by the relative abundance of hepatocyte nuclear factors 1α and 1β. J Biol Chem 273:28408–28418. 111. Chi YI, Frantz JD, Oh BC, Hansen L, Dhe-Paganon S, Shoelson SE 2002 Diabetes mutations delineate an atypical POU domain in HNF-1alpha. Mol Cell 10: 1129–1135. 112. Roncero C, Ventura J-J, Sanchez A, Bois-Joyeux B, Mesa ML, Thomassin H, Danan J-L, Benito M, Fabregat I 1998 Phorbol esters down-regulate α-fetoprotein gene expression without affecting growth in fetal hepatocytes in primary culture. Biochim Biophys Acta 1402:151–164. 113. Haddad JG, Fraser DR, Lawson DEM 1981 Vitamin D binding protein. Turnover and fate in the rabbit. J Clin Invest 67:1550–1560. 114. Nykjaer A, Dragun D, Walther D, Vorum H, Jacobsen C, Herz J, Melsen F, Christensen EI 1999 An endocytic pathway essential for renal uptake and activation of the steroid 25-(OH) vitamin D3. Cell 96:507–515. 115. Nykjaer A, Fyfe JC, Kozyraki R, Leheste JR, Jacobsen C, Nielsen MS, Verroust PJ, Aminoff M, De la Chapelle A, Moestrup SK, Ray R, Gliemann J, Willnow TE, Christensen EI 2001 Cubilin dysfunction causes abnormal metabolism of the steroid hormone 25(OH) vitamin D3. Proc Natl Acad Sci USA 98:13895–13900. 116. Leheste JR, Melsen F, Wellner M, Jansen P, Schlichting U, Renner-Muller I, Andreassen TT, Wolf E, Bachmann S, Nykjaer A, Willnow T 2003 Hypocalcemia and osteopathy in mice with kidney-specific megalin gene defect. FASEB J 17:247–249.
131 117. Goldschmidt-Clermont P, Van Baelen H, Bouillon R, Shook TE, Williams MH, Nel AE, Galbraith RM 1988 Role of groupspecific component (vitamin D binding protein) in clearance of actin from the circulation in the rabbit. J Clin Invest 81:1519–1527. 118. Dueland S, Nenseter MS, Drevon CA 1991 Uptake and degradation of filamentous actin and vitamin D–binding protein in the rat. Biochem J 274:237–241. 119. Cousin P, Dechaud H, Grenot C, Lejeune H, Hammond GL, Pugeat M 1999 Influence of glycosylation on the clearance of recombinant human sex hormone-binding globulin from rabbit blood. J Steroid Biochem Mol Biol 70:115–121. 120. Innes Chalk KJ, Kodicek E 1961 The association of 14Clabelled vitamin D2 with rat serum proteins. Biochem J 79:1–7. 121. Rikkers H, DeLuca HF 1967 An in vivo study of the carrier proteins of 3H-vitamins D3 and D4 in rat serum. Am J Physiol 213:380–386. 122. Bouillon R, Van Baelen H, Rombauts W, DeMoor P 1976 The purification and characterization of the human serum binding protein for the 25-hydroxycholecalciferol (transcalciferin). Identity with group specific component. Eur J Biochem 66:285–291. 123. Bouillon R, Verstuyf A, Zhao J, Tan B, Van Baelen H 1996 Nonhypercalcemic vitamin D analogs: interactions with the vitamin D–binding protein. Horm Res 45:117–121. 124. Swamy N, Dutta A, Ray R 1997 Roles of the structure and orientation of ligands and ligand mimics inside the ligandbinding pocket of the vitamin D–binding protein. Biochemistry 36:7432–7436. 125. Norman AW, Ishizuka S, Okamura WH 2001 Ligands for the vitamin D endocrine system: different shapes function as agonists and antagonists for genomic and rapid response receptors or as a ligand for the plasma vitamin D binding protein. J Steroid Biochem Mol Biol 76:49–59. 126. Carlberg C 2003 Molecular basis of the selective activity of vitamin D analogues. J Cell Biochem 88:274–281. 127. Vieth R 1994 Simple method for determining specific binding capacity of vitamin D–binding protein and its use to calculate the concentration of “free” 1,25-dihydroxyvitamin D. Clin Chem 40:435–441. 128. Teegarden D, Meredith SC, Sitrin MD 1991 Determination of the affinity of vitamin D metabolites to serum vitamin D binding protein using an assay employing lipid-coated polystyrene beads. Anal Biochem 199:293–299. 129. Ray R, Bouillon R, Van Baelen H, Holick MF 1991 Photoaffinity labelling of human serum vitamin D binding protein and chemical cleavages of the labelled protein: identification of an 11.5 kDa peptide containing the putative 25-hydroxyvitamin D3 binding site. Biochemistry 30:6638–7642. 130. Haddad JG, Hu YZ, Kowalski MA, Laramore C, Ray K, Robzyk P, Cooke NE 1992 Identification of the sterol- and actin binding domains of plasma vitamin D binding protein (Gc globulin). Biochemistry 31:7174–7181. 131. Addo JK, Swamy N, Ray R 2002 The C19 position of 25-hydroxyvitamin D3 faces outward in the vitamin D sterolbinding pocket of vitamin D–binding protein. Bioorg Med Chem Lett 12:279–281. 132. Avvakumov GV, Grishkovskaya I, Muller YA, Hammond GL 2002 Crystal structure of human sex hormone-binding globulin in complex with 2-methoxyestradiol reveals the molecular basis for high affinity interactions with C-2 derivatives of estradiol. J Biol Chem 277:45219–45225.
132 133. Grishkovskaya I, Avvakumov GV, Hammond GL, Catalano MG, Muller YA 2002 Steroid ligands bind human sex hormone-binding globulin in specific orientations and produce distinct changes in protein conformation. J Biol Chem 277:32086–32093. 134. Surarit R, Svasti J 1980 Effect of ligand binding on the conformation of human plasma vitamin D–binding protein (group specific component). Biochem J 191:401–410. 135. Koszelak S, McPherson A, Bouillon R, Van Baelen H 1985 Crystallisation and preliminary X-ray analysis of the vitamin Dbinding protein from human serum. J Steroid Biochem 23:1077–1078. 136. Cooke N, Walgate J, Haddad J 1979 Human serum binding protein for vitamin D and its metabolites. II. Specific, high affinity association with a protein in nucleated tissue. J Biol Chem 254:5965–5971. 137. Van Baelen H, Bouillon R, De Moor P 1980 Vitamin D–binding protein (Gc-globulin) binds actin. J Biol Chem 255:2270–2272. 138. Buch S, Gremm D, Wegner A, Mannherz HG 2002 Binding of a C-terminal fragment (residues 369 to 435) of vitamin D– binding protein to actin. Biol Chem 383:1621–1631. 139. McLeod JF, Kowalski MA, Haddad JG 1989 Interactions among serum vitamin D binding protein, monomeric actin, profilin and profilactin. J Biol Chem 264:1260–1267. 140. Emerson DL, Arnaud P, Galbraith RM 1983 Evidence of increased Gc:actin complexes in pregnant serum: a possible result of trophoblast embolism. Am J Reprod Immunol 4:185–189. 141. Goldschmidt-Clermont PJ, Galbraith RM, Emerson DL, Werner PAM, Nel AE, Lee WM 1985 Accurate quantitation of native Gc in serum and estimation of endogenous GcG-actin complexes by rocket immunoelectrophoresis. Clin Chim Acta 148:173–183. 142. Haddad JG, Harper KD, Guoth M, Pietra GG, Sanger JW 1990 Angiopathic consequences of saturating the plasma scavenger system for actin. Proc Natl Acad Sci USA 87:1381–1385. 143. Lee WM, Galbraith RM 1992 The extracellular actinscavenger system and actin toxicity. New Engl J Med 326: 1335–1341. 144. Petrini M, Galbraith RM, Werner, PAM, Emerson DL, Arnaud P 1984 Gc (vitamin D binding protein) binds to cytoplasm of all human lymphocytes and is expressed on B-cell membranes. Clin Immunol Immunopathol 31:282–295. 145. Machii T, Kimura H, Ueda E, Chujo T, Morikta T, Katagiri S, Tagawa S, Kitani T 1986 Distribution of Gc protein (vitamin D binding protein) on the surfaces of normal human lymphocytes and leukemic lymphocytes. Acta Haematol 75:26–29. 146. McLeod JF, Kowalski MA, Haddad JG 1986 Characterization of a monoclonal antibody to human serum vitamin D binding protein (Gc globulin): recognition of an epitope hidden in membranes of circulating monocytes. Endocrinology 119:77–83. 147. Guoth M, Murgia A, Smith RM, Prystowsky MB, Cooke NE, Haddad JG 1990 Cell surface vitamin D binding protein (Gc globulin) is acquired from plasma. Endocrinology 127:2313–2321. 148. DiMartino SJ, Kew RR 1999 Initial characterisation of the vitamin D binding protein (Gc-globulin) binding site on the neutrophil plasma membrane: evidence for a chondroitin sulfate proteoglycan. J Immunol 163:2135–2142. 149. Danan JL, Delorme AC, Ripoche MA, Bouillon R, Van Baelen H, Mathieu H 1985 Presence of immunoreactive vitamin D–binding protein in rat yolk sac endodermal cells. Endocrinology 117:243–247.
CHRISTOPHER J. LAING AND NANCY E. COOKE
150. Emerson DL, Werner PA, Cheng MH, Galbraith RM 1985 Presence of Gc (vitamin D–binding protein) and interactions with actin in human placental tissue. Am J Reprod Immunol Microbiol 7:15–21. 151. Nestler JE, McLeod JF, Kowalski MA, Strauss JF, Haddad JG 1987 Detection of vitamin D binding protein on the surface of cytotrophoblasts isolated from human placentae. Endocrinology 120:1996–2002. 152. Yu HM, Li XJ, Kadam AL, Cheng CY, Koide SS 1994 Human testis vitamin D binding protein involved in infertility. Arch Androl 33:119–128. 153. Moestrup SK, Verroust PJ (2001) Megalin- and cubilinmediated endocytosis of protein-bound vitamins, lipids, and hormones in polarized epithelia. Ann Rev Nutr 21:407–428. 154. Petrini M, Galbraith RM, Emerson DL, Nel AE, Arnaud P 1985 Structural studies of T lymphocyte Fc receptors. Association of Gc protein with IgG binding to Fc gamma. J Biol Chem 260:1804–1810. 155. DiMartino SJ, Shah AB, Trujillo G, Kew RR 2001 Elastase controls the binding of the vitamin D–binding protein (Gc-globulin) to neutrophils: a potential role in the regulation of C5a co-chemotactic activity. J Immunol 166:2688–2694. 156. Osawa M, Kimura A, Yukawa N, Seto Y, Saito T, Tsuji T, Takeichi S 1998 Epitope analysis of monoclonal antibodies to human Gc globulin (vitamin D–binding protein). Biochem Mol Biol Int 44:293–303. 157. Weitkamp LR, Rucknagel DL, Gershowitz H 1966 Genetic linkage between structural loci for albumin and group specific component (Gc). Am J Hum Genet 18:559–571. 158. Daiger SP, Schanfield MS, Cavalli-Sforza LL 1975 Human group-specific component (Gc) proteins bind vitamin D and 25-hydroxyvitamin D. Proc Natl Acad Sci USA 72:2076–2080. 159. White P, Cooke N 2000 The multifunctional properties and characteristics of vitamin D binding protein. Trends Endocrinol Metab 11:320–321. 160. Mendel CM 1989 The free hormone hypothesis: a physiologically based mathematical model. Endocr Rev 10:232–274. 161. Hammond GL 2002 Access of reproductive steroids to target tissues. Obstet Gynecol Clin N Am 29:411–423. 162. Siiteri PK, Murai JT, Hammond GL, Nisker JA, Raymoure WJ, Kuhn RW 1982 The serum transport of steroid hormones. Rec Prog Horm Res 38:457–503. 163. Vermeullen A 1988 Physiology of the testosterone binding globulin in man. Ann NY Acad Sci 538:103–111. 164. Nakhla AM, Leonard J, Hryb DJ, Rosner W 1999 Sex hormone–binding globulin receptor signal transduction proceeds via a G protein. Steroids 64:213–216. 165. Rosner W, Hryb DJ, Khan MS, Nakhla AM, Romas NA 1999 Androgen and estrogen signaling at the cell membrane via G-proteins and cyclic adenosine monophosphate. Steroids 64:100–106. 166. Rosner W, Hryb DJ, Khan MS, Nakhla AM, Romas NA 1999 Sex hormone–binding globulin mediates steroid hormone signal transduction at the plasma membrane. J Steroid Biochem Mol Biol 69:481–485. 167. Heinlein CA, Chang C 2002 The roles of androgen receptors and androgen-binding proteins in nongenomic androgen actions. Mol Endocrinol 16:2181–2187. 168. Hryb DJ, Nakhla AM, Kahn SM, St George J, Levy NC, Romas NA, Rosner W 2002 Sex hormone-binding globulin in the human prostate is locally synthesized and may act as an autocrine/paracrine effector. J Biol Chem 277: 26618–26622.
CHAPTER 8 Vitamin D–Binding Protein
169. Kahn SM, Hryb DJ, Nakhla AM, Romas NA, Rosner W 2002 Sex hormone-binding globulin is synthesized in target cells. J Endocrinol 175:113–120. 170. Orchinik M, Hastings N, Witt D, McEwen BS 1997 Highaffinity binding of corticosterone to mammalian neuronal membranes: possible role of corticosteroid binding globulin. J Steroid Biochem Mol Biol 60:229–236. 171. Hodam JR, Creek KE 1998 Comparison of the metabolism of retinol delivered to human keratinocytes either bound to serum retinol-binding protein or added directly to the culture medium. Exp Cell Res 238:257–264. 172. Sundaram M, Sivaprasadarao A, DeSousa MM, Findlay JBC 1998 The transfer of retinol from serum retinol-binding protein to cellular retinol-binding protein is mediated by a membrane receptor. J Biol Chem 273:336–3342. 173. Noy N Retinoid-binding proteins: mediators of retinoid action. Biochem J 348:481–495. 174. Breuner CW, Orchinik M 2002 Plasma binding proteins as mediators of corticosteroid action in vertebrates. J Endocrinol 175:99–112. 175. Bouillon R, Van Assche F, Van Baelen H, Heyns W, DeMoor P 1981 Influence of the vitamin D–binding protein on the serum concentration of 1,25-dihydroxyvitamin D—significance of the free 1,25-dihydroxyvitamin D concentration. J Clin Invest 67:589–596. 176. Dilworth FJ, Calverly MJ, Makin HLJ, Jones G 1994 Increased biological activity of 20-epi-1,25-dihydroxyvitamin D3 is due to reduced catabolism and altered protein binding. Biochem Pharmacol 47:987–993. 177. Vanham G, Van Baelen H, Tan BK, Bouillon R 1988 The effect of vitamin D analogs and of vitamin D-binding protein on lymphocyte proliferation. J Steroid Biochem 29:381–386. 178. Vargas S, Bouillon R, Van Baelen H, Raisz LG 1990 Effects of vitamin D-binding protein on bone resorption stimulated by 1,25-dihydroxyvitamin D3. Calcif Tissue Int 47:164–168. 179. Furuichi T, Kawata S, Asoh Y, Kumaki K, Ohyama Y 1998 Differential time course of induction of 1α,25-dihydroxyvitamin D3-24-hydroxylase mRNA expression in rats by 1α, 25dihydroxyvitamin D3 and its analogs. Life Sci 62:453–459. 180. Pettifor JM, Bikle DD, Cavaleros M, Zachen D, Kamdar MC, Ross FP 1995 Serum levels of free 1,25-dihydroxyvitamin D in vitamin D toxicity. Ann Intern Med 122:511–513. 181. Van Hoof HJ, Swinkels LM, Ross HA, Sweep CG, Benraad TJ 1998 Determination of non-protein-bound plasma 1,25dihydroxyvitamin D by symmetric (rate) dialysis. Anal Biochem 258:176–183. 182. Silver J, Fainaru M 1979 Transport of vitamin D sterols in human plasma: effect of excess vitamin D, 25-hydroxyvitamin D and 1,25-dihydroxyvitamin D. Eur J Clin Invest 9:433–438. 183. Adams JS, Chen H, Chun RF, Nguyen L, Wu S, Ren S-Y, Barsony J, Gacad MA 2003 Novel regulators of vitamin D action and metabolism: lessons learned at the Los Angeles Zoo. J Cell Biochem 88:308–314. 184. Larsson D, Nemere I, Sundell K 2001 Putative basal lateral membrane receptors for 24,25-dihydroxyvitamin D3 in carp and Atlantic cod enterocytes: characterization of binding and effects on intracellular calcium regulation. J Cell Biochem 83:171–186. 185. Norman AW, Olivera CJ, Silva FRMB, Bishop JE 2002 A specific binding protein/receptor for 1α,25-dihydroxyvitamin D3 is present in an intestinal caveolae membrane fraction. Biochem Biophys Res Commun 298:414–419.
133 186. Donohue MM, Demay MB 2002 Rickets in VDR null mice is secondary to decreased apoptosis of hypertrophic chondrocytes. Endocrinology 143:3691–3694. 187. St-Arnaud R, Arabian A, Glorieux FH 1996 Abnormal bone development in mice deficient for the vitamin D 24-hydroxylase gene. J Bone Miner Res 11:S126. 188. St-Arnaud R, Dardenne O, Prud’homme J, Hacking SA, Glorieux FH 2003 Conventional and tissue-specific inactivation of the 25-hydroxyvitamin D-1-hydroxylase (CYP27B1). J Cell Biochem 88:245–251. 189. Torpy DJ, Bachmann AW, Grice JE, Fitzgerald SP, Phillips PJ, Whitworth JA, Jackson RV 2001 Familial corticosteroidbinding globulin deficiency due to a novel null mutation: association with fatigue and relative hypotension. J Clin Endocrinol Metab 86:3692–3700. 190. Joyner JM, Hutley LJ, Bachmann AW, Torpy DJ, Prins JB 2003 Greater replication and differentiation of preadipocytes in inherited corticosteroid-binding globulin deficiency. Am J Physiol Endocrinol Metab 284:E1049–E1054. 191. Domingues R, Bugalho MJ, Garrao A, Boavida JM, Sobrinho L 2002 Two novel variants in the thyroxine-binding globulin (TBG) gene behind the diagnosis of TBG deficiency. Eur J Endocrinol 146:485–490. 192. Su C-C, Wu Y-C, Chiu C-Y, Won JG-S, Jap T-S 2003 Two novel mutations in the gene encoding thyroxine-binding globulin (TBG) as a cause of complete TBG deficiency in Taiwan. Clin Endocrinol 58:409–414. 193. Vogel S, Piantedosi R, O’Byrne SM, Kako Y, Quadro L, Gottesman ME, Goldberg IJ, Blaner WS 2002 Retinol-binding protein-deficient mice: biochemical basis for impaired vision. Biochemistry 41:15360–15368. 194. Baker ME 2002 Albumin, steroid hormones and the origin of vertebrates. J Endocrinol 175:121–127. 195. Soliman AT, Madina EH, Morsi MR 1996 Radiological, biochemical, and hormonal changes in malnourished children with rachitic manifestations. J Trop Pediatr 42:34–37. 196. Lulseged S, Fitwi G 1999 Vitamin D deficiency rickets: sociodemographic and clinical risk factors in children seen at a referral hospital in Addis Ababa. East Afr Med J 76: 457–461. 197. Vasconcellos CA, Lind SE 1993 Coordinated inhibition of actin-induced platelet aggregation by plasma gelsolin and vitamin D–binding protein. Blood 82:3648–3657. 198. Erukhimov JA, Tang Z-L, Johnson BA, Donahoe MP, Razzack JA, Gibson KF, Lee WM, Wasserloos KJ, Watkins SA, Pitt BR 2000 Actin-containing sera from patients with adult respiratory distress syndrome are toxic to sheep pulmonary epithelial cells. Am J Respir Crit Care Med 162: 288–294. 199. Schiodt FV, Ott P, Bondesen S, Tygstrup N 1997 Reduced serum Gc-globulin concentrations in patients with fulminant hepatic failure: association with multiple organ failure. Crit Care Med 25:1366–1370. 200. Lee WM, Reines D, Watt GH, Cook JA, Wise WC, Halushka PV, Galbraith RM 1989 Alterations in Gc levels and complexing in septic shock. Circ Shock 28:249–255. 201. Osawa M, Yukawa N, Saito T, Huang XL, Kusakabe T, Takeichi S 1998 Increased complex formation of Gc globulin with actin in plasma from human cadavers. Forensic Sci Int 96:39–45. 202. Dahl B, Schiodt FV, Ott P, Wians F, Lee WM, Balko J, O’Keefe GE 2003 Plasma concentration of Gc-globulin is associated with organ dysfunction and sepsis after injury. Crit Care Med 31:152–156.
134 203. Dahl B, Schiodt FV, Kiaer T, Ott P, Bondesen S, Tygstrup N 1998 Serum Gc-globulin in the early course of multiple trauma. Crit Care Med 26:285–289. 204. Metcalf JP, Thompson AB, Gossman GL, Nelson KJ, Koyama S, Rennard SI, Robbins RA 1991 Gc globulin functions as a co-chemotaxin in the lower respiratory tract. Am Rev Respir Dis 143:844–849. 205. Kew RR, Mollison KW, Webster RO 1992 Binding of Gc globulin (DBP) to C5a or C5a des Arg is not necessary for co-chemotactic activity. J Leukocyte Biol 58:55–58. 206. Kew RR, Fisher JA, Webster RO 1995 Co-chemotactic effect of Gc-globulin (DBP) for C5a: transient conversion into an active co-chemotaxin by neutrophils. J Immunol 155: 5369–5374. 207. Binder R, Kress A, Kan G, Herrmann K, Kirschfink M 1999 Neutrophil priming by cytokines and vitamin D binding protein (Gc-globulin): impact on C5a-mediated chemotaxis, degranulation and respiratory burst. Mol Immunol 36:885–892. 208. Belting M 2003 Heparan sulfate proteoglycan as a plasma membrane carrier. Trends Biochem Sci 28:145–151. 209. Yamamoto N, Ngwenya BZ 1987 Activation of mouse peritoneal macrophages by lysophospholipids and ether derivatives of neutral lipids and phospholipids. Cancer Res 47:2008–2013. 210. Yamamoto N, St Claire DA, Homma S, Ngwenya BZ 1988 Activation of mouse macrophages by alkylglycerols, inflammation products of cancerous tissues. Cancer Res 48:6044–6049. 211. Naraparaju VR, Yamamoto N 1994 Roles of β-galactosidase of B lymphocytes and sialidase of T lymphocytes in inflammation-primed activation of macrophages. Immunol Lett 43:143–148. 212. Yamamoto N 1996 Structural definition of a potent macrophage activating factor derived from vitamin D3-binding protein with adjuvant activity for antibody production. Mol Immunol 33:1157–1164. 213. Mohamad SB, Nagasawa H, Uto Y, Hori H 2002 Preparation of Gc protein-derived macrophage activating factor (GcMAF) and its structural characterization and biological activities. Anticancer Res 22:4297–4300. 214. Yamamoto N, Naraparaju VR, Neil RN, Suyama H, Nakazato H 1997 Therapeutic efficiency of vitamin D3-binding protein-derived macrophage activating factor for prostate, breast, and colon cancers. Proc Am Assoc Cancer Res 38:31. 215. Koga Y, Naraparaju VR, Yamamoto N 1999 Antitumor effect of vitamin D–binding protein-derived macrophage activating factor on Ehrlich ascites tumor-bearing mice. Proc Soc Exp Biol Med 220:20–26.
CHRISTOPHER J. LAING AND NANCY E. COOKE
216. Yamamoto N, Urade M 2002 Therapeutic efficacy of Gc protein derived macrophage activating factor for adenocarcinoma. Proc Am Assoc Cancer Res 43:914. 217. Ono T, Ichikawa H, Asami R, Yamamoto N 1995 Enzymatically modified Gc globulin induces the translocation of FcγRI and FcγRII from intracellular storage compartments to the cell surface in murine peritoneal macrophage. Jpn J Inflamm 15:293–299. 218. Yamamoto N, Naraparaju VR, Asbell SO 1996 Deglycosylation of serum vitamin D3–binding protein leads to immunosuppression in cancer patients. Cancer Res 56:2827–2831. 219. Mohamad SB, Nagasawa H, Uto Y, Hori H 2002 Tumor cell alpha-N-acetylgalactosaminidase activity and its involvement in GcMAF-related macrophage activation. Comp Biochem Biophys 132A:1–8. 220. Swamy N, Ghosh S, Schneider GB, Ray R 2001 Baculovirusexpressed vitamin D–binding protein-macrophage activating factor (DBP-maf) activates osteoclasts and binding of 25-hydroxyvitamin D3 does not influence this activity. J Cell Biochem 81:535–546. 221. Schneider GB, Benis KA, Flay NW, Ireland RA, Popoff SN 1995 Effects of vitamin D–binding protein-macrophage activating factor (DBP-MAF) infusion on bone resorption in two osteopetrotic mutations. Bone 16:657–662. 222. Yamamoto N, Naraparaju VR 1996 A defect in β-galactosidase of B-lymphocytes in the osteopetrotic (op/op) mouse. Immunol Lett 50:35–40. 223. Datta HK, Cook DB, Kanan RM 1999 Evidence for continuous basal generation of Gc-MAF: absence in infantile osteopetrosis and restoration after bone marrow transplant. Blood 93:4026–4027. 224. Kanda S, Mochizuki Y, Miyata Y, Kanetake H, Yamamoto N 2002 Effects of vitamin D3-binding protein–derived macrophage activating factor (GcMAF) on angiogenesis. J Nat Cancer Inst 94:1311–1319. 225. Kisker O, Onizuka S, Becker CM, Fannon M, Flynn E, D’Amato R, Zetter B, Folkman J, Ray R, Swamy N, PineShepherd S 2003 Vitamin D binding protein-macrophage activating factor (DBP-maf) inhibits angiogenesis and tumor growth in mice. Neoplasia 5:32–40. 226. White P, Liebhaber SA, Cooke NE 2002 129X1/SvJ mouse strain has a novel defect in inflammatory cell recruitment. J Immunol 168:869–874. 227. Reichner JS, Fitzpatrick PA, Wakshull E, Albina JE 2001 Receptor-mediated phagocytosis of rat macrophages is regulated differentially for opsonized particles and nonopsonized particles containing β-glucan. Immunology 104:198–206.
CHAPTER 9
New Aspects of DBP ILSE BOGAERTS AND CHRISTEL VERBOVEN Lab. Analytische Chemie, Van Evenstraat 4; B-3000 Leuven, Belgium
HUGO VAN BAELEN AND ROGER BOUILLON Legendo, Onderwijs en Navorsing 902, Gasthuisberg, Herestraat 49, B-3000 Leuven, Belgium.
I. Introduction II. Three-Dimensional Structure of Vitamin D–Binding Protein III. Three-Dimensional Structures of Vitamin D–Binding Protein in Complex with Ligands
IV. A Structural Explanation for the Unique Functions of Vitamin D–Binding Protein within Its Gene Family References
SUMMARY
N-terminal part of domain I. The crystal structure of the DBP–actin complex was determined at 2.4 Å resolution. This structure reveals that the fast-growing side of actin monomers is blocked completely through a perfect structural fit with DBP, demonstrating how DBP effectively interferes with actin-filament formation. Comparisons of the DBP structures with the structure of serum albumin, another family member, reveal a similar topology but also significant differences in overall as well as local folding. These structural differences explain the unique vitamin D3–binding and actin-binding properties of DBP.
The crystal structure of human serum vitamin D– binding protein (DBP) and the complex of hDBP and actin were solved. Despite the common genetic origin and similarities in overall gene/protein structure, the 3D structure of DBP differs substantially from that of human serum albumin. The crystal structure of DBP allowed the positioning of the surface grove in domain I responsible for binding of all vitamin D metabolites. This allows us to explain or predict the relative affinity of vitamin D metabolites and analogs for DBP. This is of particular importance since DBP affinity explains, at least in part, differences in pharmacokinetics and selective profiles of vitamin D analogs. The resolution of the DBP surface interacting with G actin explains the tight binding between these two proteins and also allows us to explain the molecular basis of the DBP actin scavenging system, which is an important protection against tissue damage–induced multiple organ failure. Finally, the binding site of DBP has been identified for FFA, but other putative interactions of DBP with macrophages (DBP-MAF) and complement need further work.
I. INTRODUCTION The human serum vitamin D–binding protein (DBP) has many physiologically important functions, ranging from transporting vitamin D3 metabolites, binding and sequestering globular actin and binding fatty acids to functioning in the immune system. The crystal structure of DBP in complex with the vitamin D3 metabolite 25-hydroxyvitamin D3 was solved, as well as the structure of DBP in complex with a vitamin D3 analog. Both structures reveal the vitamin D–binding site in the VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX
II. THREE-DIMENSIONAL STRUCTURE OF VITAMIN D–BINDING PROTEIN A. Overall Features of the Vitamin D–Binding Protein Structure The crystal structure of human DBP was solved to 2.3-Å resolution [1]. The two DBP molecules present in the asymmetric unit (further referred to as DBP-A and DBP-B) are not identical. One of the major differences between them is that the DBP-B molecules are in complex with 25OHD3,whereas the DBP-A molecules are not. As in human serum albumin (HSA) [2,3], DBP has an all α-helical structure and contains three structurally similar domains (Fig. 1). DBP and the other members of its protein family (human serum albumin, afamin, and α-fetoprotein) are postulated to have evolved from a progenitor that arose from the triple repeat of a 192-amino acid sequence [4]. This threedomain structure has been preserved in DBP; however, the third repeat is largely truncated at the C terminus. Copyright © 2005, Elsevier, Inc. All rights reserved.
136
ILSE BOGAERTS, CHRISTEL VERBOVEN, HUGO VAN BAELEN, AND ROGER BOUILLON
DOMAIN I DOMAIN I
DOMAIN II
DOMAIN II
FIGURE 1
DOMAIN III
DOMAIN II
Stereo view of the overall fold of the DBP structure.
The HSA structure [2,3] indicates that each domain consists of 10 α-helices. In the DBP structure, the first domain (residues 1–191) has this α-helical arrangement. However, the second domain (residues 192–378) has a similar topology but helix 7 is replaced by a coil folding (Figs. 2, 3), and the third domain (residues 379–458) contains only helices 1–4. The three domains of DBP do not pack in a spherical manner but adopt a rather peculiar shape with two large grooves A more detailed comparison of both DBP-A and DBP-B structures reveals differences in local and overall folding. All pairwise Cα atoms in DBP-A and DBP-B superimpose with a root mean square (rms) deviation of 1.63 Å. Superimposing all common Cα atoms of both domains I, II, and III gives, however, mean rms deviations of 0.646 Å, 0.981 Å, and 0.416 Å, respectively. The higher value for the superposition of the overall DBP-A and DBP-B structures is caused by slightly different orientations of the respective domains in both structures. This difference of DBP-B may be
caused by the presence of 25OHD3, because a similar phenomenon is observed with the HSA structure. All three HSA domains assume a different orientation upon binding of free fatty acids [5]. Differences between DBP-A and DBP-B are also observed at the temperature factor level. The main difference is a much higher B-factor of the N terminus of domain I in the DBP-A molecule (91.2 compared to 76.8 for DBP-B). 25OHD3 is bound to helices 1–6 of domain I (see later discussion); its presence explains the higher stability in this region of DBP-B compared to the same region in DBP-A. In addition to the electron density for the 25OHD3 molecule, elongated residual density is observed that is able to accommodate oleic acid, a free fatty acid (FFA). DBP has been reported to bind unsaturated FFAs such as oleic acid [6–8]. Because the DBP25OHD3 crystals were obtained from a pooled human serum sample, the FFAs were expected to cocrystallize with DBP. Although DBP can be postulated to have several different FFA binding sites analogous with HSA [5,9], only one site was detected in DBP-A, and two fatty acids were seen in DBP-B. Because the FFAs present in this crystal are acquired from plasma rather than deliberately added to the protein sample, all possible FFA binding pockets may not be occupied.
B. Comparison with the Structure of Serum Albumin, a Member of Its Gene Family
FIGURE 2
The structure of domain II illustrating its typical topology. The numbering of the helices is kept the same as the similar HSA structure, although the small helix 7 is not present in this second domain of DBP.
Based on the presence of sequence homology and nearly identical disulfide bridge pattern in DBP, human serum albumin (HSA), α-fetoprotein, and afamin, the overall folds of these proteins are believed to be homologous [11]. Despite the sequence similarity, the only function DBP shares with the other family members is its fatty acid binding ability [6,7].
137
CHAPTER 9 New Aspects of DBP
FIGURE 3 Sequence alignment of DBP sequences of various species (DBP-H: human; DBP-RB: rabbit; DBP-R: rat; DBP-M: mouse) and the human sequences belonging to the same protein family (albu-H: human serum albumin; AFAM-H: human afamin, FETA-H: human α-fetoprotein). DBP has 23% sequence identity with serum albumin, 20% with α-fetoprotein, and 21% with afamin. Conserved residues are marked in blue; identical residues, green; and homologous residues, purple. Triangles indicate the Cys residues responsible for the typical disulfide bridge pattern observed in this protein family. For ALBU-H, AFAM-H, and FETO-H, only the sequence that aligns with the DBP sequence is shown. The secondary structure elements of DBP are also shown. The α-helices, shown at the top, have the same numbering as the similar HSA structure, although α-helix 7 of domain II has a coil folding in DBP. This figure was made with ALSCRIPT [10]. See color plate.
Superposition of the respective domains of the DBP and HSA structures shows similar topologies (Fig. 4a). Although the folding within each corresponding domain shows some parallels, the global orientation of the three domains in both molecules is strikingly different (Fig. 4b), resulting in two totally different structures. Superposition of all the aligned Cα atoms of DBP and HSA [3] gives an rms deviation as high as 10.1 Å (438 atoms superimposed). A comparison between the separate domains also indicates remarkable differences. The rms deviation between the aligned Cα atoms of domain I is 4.0 Å (175 atoms superimposed), 4.2 Å for domain II (185 atoms superimposed), and 3.9 Å for domain III
(78 atoms superimposed). These rms deviations are even higher when HSA is bound to fatty acids.
III. THREE-DIMENSIONAL STRUCTURES OF VITAMIN D–BINDING PROTEIN IN COMPLEX WITH LIGANDS A. The DBP–Vitamin D Complex 1. THE VITAMIN D BINDING SITE
Residual electron density that can accommodate 25OHD3 is observed in DBP-B close to the biochemically
138
ILSE BOGAERTS, CHRISTEL VERBOVEN, HUGO VAN BAELEN, AND ROGER BOUILLON
A
B
DOM III HSA
DOM III HSA DOM I HSA
DOM I HSA
DOM I DBP
DOM I DBP
NN
N N
DOM III DBP DOM III DBP
DOM II DBP DOM II DBP
DOM II HSA
DOM II HSA
FIGURE 4 Structural differences between DBP and serum albumin. (A) Stereo representation of the superimposed domains II of DBP (light gray) and HSA (dark grey). The helices of HSA are labeled. The small helix 7 is not present in this particular DBP domain. (B) Stereo view of the overall folds of the DBP and HSA (medium gray) molecules based on the superposition of the domains II. DBP domain I is shown in dark grey, DBP domains II and III in light grey.
identified vitamin D–binding residues (35–49) [12] (Fig. 5b). Residues belonging to helices 1–6 of domain I form the complete vitamin D–binding site (Fig. 5a,b). This binding site designation was further confirmed by the elucidation of the structure of DBP in complex with the JY analog, 22-(m-hydroxyphenyl)-23,24,25,26,27pentanor vitamin D3 (Table I). The electron density map of the DBP-JY structure clearly illustrates the presence of residual density in the N-terminal part of domain I for both the DBP-A and DBP-B molecules (Fig. 5c). This density can accommodate the JY analog. In contrast with the DBP-25(OH)D3 structure (Fig. 5b), the shape of the electron density in the sidechain region of the vitamin D3 molecule illustrates the presence of an aromatic feature (Fig. 5c). In the DBP-JY structure, the vitamin D–binding regions of both DBP-A
and DBP-B have similar average B-values, which is in contrast to the DBP-25(OH)D3 structure. The vitamin D–binding site is lined predominantly by hydrophobic residues (Fig. 5b), allowing favorable interactions with the hydrophobic vitamin D3 ligand. Binding of 25(OH)D3 is further improved by hydrogen bond formation of the 25-hydroxyl with Tyr 32 (O–H…O distance 2.85 Å), and the 3-hydroxyl with Ser 76 (O…H–O distance 2.81 Å) and Met 107 (O–H…S distance 3.01 Å). For DBP-JY, the same hydrogen bonds can be formed for 3-hydroxyl, but the hydroxyl of the phenolic substituent is hydrogen bonded to Glu 15. This hydroxyl of JY is at the same position as the oxygen of a water molecule in the DBP-25(OH)D3 structure (Fig. 5b,c). However, the detailed structural features of 25(OH)D3 or JY cannot be seen in the
139
CHAPTER 9 New Aspects of DBP
density of the structures, probably because not all of the sites present in the DBP-25(OH)D3 and DBP-JY crystals are fully occupied, but it may also be ascribed to mobility of the bound ligand. The vitamin D–binding site of DBP-B in the DBP-25(OH)D3 structure is
estimated to be 55% occupied by 25(OH)D3. For the DBP-JY structure, the estimated occupancy of JY in DBP-A is 0.66, whereas that for DBP-B is 0.72. Despite the low detail in the observed electron density, the proposed orientation of the molecule in the
A
B
F24
V12 WAT WAT
V51 E8
L31
Y32
T72
L47 435
576
F36 579
M107
FIGURE 5 The conformation of vitamin D metabolites and analogs in the vitamin D–binding site of DBP. (A) Stereo view of the location of the vitamin D–binding site in the DBP molecule. Helices 1–6 of domain I of DBP-A (light blue) and DBP-B (pink) from the DBP-25OHD3 structure are superimposed based on the common Cα atoms in helix 2. The DBP-B molecule is the one in which the loop at the top and helix 1(H1) are complete. The 25OHD3 is shown in a ball-and-stick representation. This superposition illustrates that the disordered residues of the DBP-A molecule belong to the region that forms the vitamin D–binding site. Two water molecules are present in the vitamin D–binding site (red balls). (B) View of the simulated annealing Fo−Fc omit map of DBP in the DBP-25OHD3 structure showing the electron density for the omitted 25OHD3 molecule. The map is contoured at 3.0 σ. Residues closer than 4.2 Å to the ligand are shown. Two water molecules (WAT) are present in the vitamin D–binding site (red balls). (Continued)
140
ILSE BOGAERTS, CHRISTEL VERBOVEN, HUGO VAN BAELEN, AND ROGER BOUILLON
C
E15
F24 L27
V12 E8
JY T72
WAT
V51 L47
S28 L31 Y32 K35 F36
L75 S76 S79 M107 D
FIGURE 5
Cont’d (C) Same as (B), but showing the electron density for the omitted JY molecule in the DBP-JY structure. The one remaining water molecule in the vitamin D–binding site is a red ball. (D) Stereo view of the ball-and-stick representation of the modeled vitamin D3 analogs in the vitamin D–binding site. 25OHD3 is shown in gray; 1,25(OH)2D3, dark blue; JX, pink; JY, purple; ZAF, yellow; ZAG, light blue; ZAQ, red; and ZAU, green. For ZAF, ZAG, and ZAQ, the presence of the substituent allows for extra hydrophobic interactions with DBP residues, probably resulting in the somewhat higher affinity compared to 1,25(OH)2D3. See color plate.
binding site is consistent with extensive biochemical and modeling data and, therefore, is assumed to be correct. The increased binding of the JY analog to DBP compared to the binding of 25(OH)D3 could result from favorable stacking of the JY aromatic side chain with Phe 24 (Fig. 5c). The aromatic residues Tyr 32, Phe 36, and Tyr 68 further stabilize the JY aromatic side chain. The JX analog, 22-(P-hydroxyphenyl)23,24,25,26,27-pentanor vitamin D3 (Table I), has an approximately threefold higher affinity for DBP than 25(OH)D3. Modeling of JX shows that the p-hydroxyl
allows for extra stacking with Phe 24 (Fig. 5d). This hydroxyl also makes a tight hydrogen bond with Ser 28. Although the modeling of 1,25(OH)2D3 illustrates that this molecule can make the same hydrogen bonds with DBP as 25(OH)D3 (Fig. 5d), it also shows that the axial 1-hydroxyl causes steric hindrance with Met 107. This may explain the lower DBP binding affinity observed for 1,25(OH)2D3. The vitamin D–binding site is a cleft; consequently, certain parts of the vitamin D3 molecule have no interaction with DBP—for example, the β-side of the C- and D-rings of
141
CHAPTER 9 New Aspects of DBP
TABLE I Structural Formulas of Vitamin D3 Metabolites and Analogs and Their Affinity for DBP 18
R1
12
R3
17 13
11
16
D
C 14
9
15
8
7
H
6 19
5 10
4 A
1
3 HO Vitamin D analog
R2
2
R2
R1 21
25OHD
24
22 20
23
25 27
R3
H
Affinity for DBP
H
667
H
1
H
H
2111
H
H
1314
26 OH 21 1,25(OH)2D3
22 20
24 23
25 27
OH
26 OH
JX OH OH JY
21 ZAF
ZAG
21
24
22 20
23
22 20
25 27
OH
2
OH
1.5
26 OH 24
23
25 27 26 OH
21 ZAQ 21 ZAU
22 20
24 23
22 20
OH
CH2=CH
OH
H
26 OH 24
23
25 27
25 27
3.8 1
To illustrate the relative affinity of the ligands for the vitamin D–binding protein, the reported Relative Competitive Index values [13] are shown. All reported values are relative to the result for 1,25(OH)D3, which is normalized to 1. The ZAU analog has the same structural formula as 1,25(OH)D3 except for the methyl group on C-13, which is replaced by a vinyl group in the β-position.
142
ILSE BOGAERTS, CHRISTEL VERBOVEN, HUGO VAN BAELEN, AND ROGER BOUILLON
the vitamin D3 molecule (Table I, Fig. 5a–c). Analogs substituted in these parts of the molecule display affinities similar to the affinity between 1,25(OH)2D3 and DBP (Table I). Furthermore, modeling of the analogs ZAG (11β-phenyl-1α,25-dihydroxyvitamin D3), ZAF (11α-phenyl-1α,25-dihydroxyvitamin D3), ZAQ (11αvinyl-1α,25-dihydroxyvitamin D3), and ZAU (13β-vinyl18-nor-1α,25-dihydroxyvitamin D3) illustrates that large substituents, such as a vinyl group or a phenyl ring, can easily fit in the binding site (Fig. 5d). These observations confirm the proposed orientation of 25OHD3 in the vitamin D–binding site. 2. DIFFERENCES BETWEEN THE VITAMIN D–BINDING SITE AND THE VITAMIN D POCKET OF THE VITAMIN D RECEPTOR
The effects of the vitamin D3 hormone and its analogs are mediated through binding to an intracellular vitamin D receptor (VDR) [13]. The comparison of the DBP vitamin D–binding site with that of the VDR structure [14] clearly illustrates that they are dissimilar. In VDR, the vitamin D–binding site is a closed pocket formed in the inner structure of the receptor, whereas in DBP it is a cleft located at the surface of the molecule and partly in contact with the surrounding solvent. Moreover, no parallel for both of these physiologically important binding sites is found in ligand conformation, hydrogen bonding, or hydrophobic interactions. In the VDR pocket, the A-ring of the vitamin D3 ligand has a B-chair conformation (3-hydroxyl axial), whereas in the DBP cleft the A-ring has an A-chair conformation (3-hydroxyl equatorial). In both binding sites, the C5–C6–C7–C8 torsion angle of the ligand is nonplanar. The angle is 149° in DBP and −149° in VDR. The torsion angles responsible for the side-chain orientation of the vitamin D3 molecule are different as well (C13–C17–C20–C22 = −77° in DBP and 89° in VDR; C17–C20–C22–C23 = −70° in DBP and −156° in VDR). 3. BIOLOGICAL IMPLICATIONS
The active vitamin D3 hormone, 1,25(OH)2D3, is used in the treatment of renal osteodystrophy, hypoparathyroidism, and osteoporosis. Its analogs are used successfully in the treatment of psoriasis [15] and are currently being explored for the treatment of hyperproliferative disorders [16] and the prevention of autoimmune disorders [17,18]. However, treatment with 1,25(OH)2D3 is limited because of severe side effects (such as hypercalciurea, hypercalcemia, and increased bone resorption) caused by the required pharmacological doses [13]. In order to separate the antiproliferative and prodifferentiating effects from the calcemic and skeletal effects,
a whole generation of chemically modified 1,25(OH)2D3 molecules (analogs) has been synthesized, but with poor success rates. Many of the analogs, having a more selective activity profile, display a decreased binding affinity for DBP [19–21]. Recent progress in certain aspects of the functional mechanism of 1,25(OH)2D3 provides new perspectives for the rational design of vitamin D3 analogs with a more selective profile. For example, structural analysis of the ligand binding domain of VDR [14,22] revealed the three-dimensional arrangement of its vitamin D–binding pocket. The elucidation of the three dimensional architecture of DBP’s vitamin D–binding site provides a solid basis for detailed structural studies on DBP–vitamin D analog complexes that will lead to the better understanding of the selective action mechanisms of 1,25(OH)2D3 and its analogs. The information contained in the three-dimensional architectures of both the DBP and VDR vitamin D–binding sites may lead to the development of compounds with a more original chemical structure and a more tissue- or gene-specific biological action.
B. The DBP–Actin Complex 1. ROLE OF DBP IN THE ACTIN-SCAVENGER SYSTEM
Actin is a highly abundant intracellular protein present in all eukaryotic cells and has a pivotal role in muscle contraction as well as in cell movements. Actin also has an essential function in maintaining and controlling cell shape and architecture: It is the essential building block of the microfilament system, a cytoskeletal structure that complements two other cytoskeletal structures (the microtubules and the intermediate filaments). In low-salt buffers actin exists as a monomeric protein (globular actin, G-actin), but it polymerizes under physiological salt conditions into a double helical 10-nm-thick filament structure (filamentous actin, F-actin). In vivo, the equilibrium between G-actin and F-actin is controlled mainly by the action of several actin-binding proteins with distinct activities (e.g., gelsolin or profilin). In conditions involving severe cell injury, such as trauma, shock, sepsis, and fulminant hepatic necrosis, large quantities of actin are released in the systemic circulation. The presence of actin filaments in blood, leading to an increase in blood viscosity, is dangerous and may be fatal [23]. In addition, actin can promote clot formation by its ability to aggregate platelets [24]. Therefore, the presence of a protective system, preventing actin polymerization and ensuring actin’s fast elimination, is essential.
143
CHAPTER 9 New Aspects of DBP
The vitamin D–binding protein is an actin-binding protein [11,25] and acts as an actin-sequestering agent in extracellular space. DBP and gelsolin—the only other plasma protein that binds actin avidly—play a crucial role in the clearance of actin filaments from the circulation, a process known as the “actin-scavenger system” [23]. Gelsolin has three actin-binding properties: (i) it severs F-actin; (ii) it caps the end of F-actin; and (iii) it nucleates actin filament assemblies. Gelsolin will shift the equilibrium between actin assembly–disassembly toward actin depolymerization, but will basically not prevent actin reassembly. DBP, by forming a complex with G-actin, prevents the (re-)formation of actin filaments. Therefore DBP and gelsolin have complementary functions: Upon severing the actin filaments by gelsolin, the generated globular actin is locked in its monomeric state by DBP. The clearance of this DBP–actin complex is even substantially faster compared to the clearance of free DBP [26–29]. The complementary action of both proteins is further illustrated by the previous observation that both DBP and gelsolin are required to inhibit actinstimulated platelet aggregation [24]. 2. GENERAL FEATURES OF THE DBP–ACTIN STRUCTURE
The DBP–actin structure described here was solved by X-ray diffraction by Verboven et al. [30]. The final model of the DBP–actin structure converged to an R factor of 19.90% and an Rfree value of 25.35%. It contains an actin, DBP, and ATP molecule, one calcium ion, and 290 water molecules. The structure of the DBP–actin complex reveals that G-actin binds in one of the DBP grooves, mainly formed by helix 10 of domain I, helix 6 of domain II, and helix 3 of domain III, allowing DBP and actin to fit as two pieces of a jigsaw puzzle (Fig. 6). The surface area buried at the interface of this complex is as large as 3600 Å2. Numerous intermolecular hydrogen bonds, hydrophobic contacts, and electrostatic interactions (Table II) stabilize the complex. The actin monomer consists of two domains with each domain further subdivided in two subdomains (Fig. 6). Subdomain 1 comprises residues 1–32, 70–137, and 338–375; subdomain 2, residues 33–69; subdomain 3, residues 138–180 and 270–337; and subdomain 4, residues 181–269. Biochemical experiments delimited the DBP interface by the residues 360–372 of the actin subdomain 1 [31]. The DBP–actin structure demonstrates that actin residues of subdomains 1 and 3 constitute the DBP binding interface (Fig. 6, Table II). As in actin–gelsolin segment 1 (actin-GS1) [32] and in actin–gelsolin segment 4 to 6 (actin-GS4-6) [33], the binding of DBP occurs at the actin cleft formed at the
SUBDOMAIN2
SUBDOMAIN4
SUBDOMAIN1 SUBDOMAIN3
FIGURE 6 Overall architecture of the DBP–actin structure. Actin binds in one of the large grooves present in the DBP structure. The three domains of DBP and the four subdomains in actin are shown in different colors. See color plate.
interface of subdomains 1 and 3. The exposed hydrophobic residues on helix 341–349 of actin subdomain 1 are thereby occluded from solvent. Even more striking is that in all these complexes the apolar patch (actin residues 341–349) is masked by a solventexposed hydrophobic face of a helix present in gelsolin segment 1, in gelsolin segment 4, and in DBP. Upon complexation with DBP, the changes in the actin structure are restricted to some small regions rather than affecting the general fold. This is reflected in the small rms differences obtained from the superposition of the DBP–actin structure with other actin complexes: actin-GS1 [32], actin-DNase I [34], and uncomplexed actin [35]. Similar to other actin complex structures [32,33], the structural differences are mainly caused by folding changes in subdomains 2 and 4, whereas, despite their interaction with DBP, the backbone fold of subdomains 1 and 3 is hardly altered. Only the side chains of the residues at the interface adopt different orientations to ensure the favorable contacts with the neighboring DBP residues. The ATP molecule, bound in the cleft between the subdomains 2 and 4, and nearly all of the ATP surrounding residues in DBP–actin have the same orientation and
144
ILSE BOGAERTS, CHRISTEL VERBOVEN, HUGO VAN BAELEN, AND ROGER BOUILLON
TABLE II
Intermolecular Contacts between Actin and DBP Residues
α-Actin
DBP
Residue
Atom
Residue
Atom
Distance (Å)
LYS 113 LYS 113 TYR 143 TYR 143 ALA 144 SER 145 SER 145 GLY 146 GLY 146 GLY 146 GLY 146 ARG 147 ARG 147 ARG 147 ARG 147 ARG 147 ARG 147 THR 148 THR 148 TYR 166 TYR 166 TYR 166 TYR 166 GLU 167 GLU 167 GLU 167 TYR 169 TYR 169 LEU 171 HIS 173 HIS 173 MET 176 GLU 276 TYR 279 TYR 279 ASN 280 ASN 280 ASN 280 MET 283 MET 283 MET 283 LYS 284
NZ NZ CG OH O O C O CA O O NH2 NH1 NH1 NH1 NH1 NH1 CG2 CG2 CE2 OH CD2 CD2 O OE1 CG CE1 OH CD1 NE2 CD2 SD CG OH CD2 OD1 ND2 OD1 O CB CE NZ
SER 278 ASP 282 LEU 184 LEU 188 THR 180 ASN 125 THR 180 ASN 125 THR 180 PHE 183 ARG 187 GLU 122 PRO 123 THR 124 ASN 125 ILE 128 ARG 187 ARG 187 LEU 188 LEU 195 THR 198 VAL 294 PHE 298 LYS 191 SER 194 LEU 195 GLU 286 CYS 295 PHE 298 PHE 298 PRO 300 THR 398 THR 400 PHE 399 THR 400 THR 398 PHE 399 THR 400 TYR 394 PHE 399 SER 434 SER 395
OG OD2 CD1 CD2 CG2 OD1 CG2 OD1 O CB NE O O C ND2 CD1 NH2 CB CD2 CD1 CG2 CG2 CE2 NZ OG CD1 OE1 SG CD2 O CG CG2 CG2 CE2 CG2 CB N OG1 OH CB CB O
3.51 3.49 3.64 3.87 3.21 3.45 3.87 3.14 3.65 3.57 3.25 3.59 2.92 3.64 2.84 3.80 3.23 3.93 3.85 4.07 3.25 4.07 3.72 2.45 2.63 3.34 3.85 3.28 3.81 4.02 3.32 4.05a 4.20 3.67b 4.01a 3.19 3.51 3.00 3.61 3.89 3.45 3.76 (Continued)
145
CHAPTER 9 New Aspects of DBP
TABLE II Intermolecular Contacts between Actin and DBP Residues—Cont’d α-Actin
DBP
Residue
Atom
Residue
LYS 284 ASP 286 ASP 286 ASP 286 ILE 287 ILE 287 ILE 287 ILE 287 ASP 288 ILE 289 ARG 290 ARG 290 LYS 291 LYS 291 LYS 291 ALA 295 LYS 328 LYS 328 ILE 345 LEU 346 LEU 349 LEU 349 LEU 349 THR 351 THR 351 MET 355 HIS 371
NZ OD1 OD1 OD1 CD1 CG2 CG2 CD1 OD2 CD1 NH2 NH2 CD NZ NZ CB NZ NZ CG2 CD1 CD1 CD1 CD1 OG1 CG2 CE CB
ASN 397 TYR 297 ASN 440 ASN 440 PHE 117 SER 434 ASN 435 PRO 442 ARG 202 PHE 298 TYR 394 SER 434 PRO 118 TYR 120 TYR 151 VAL 121 PRO 123 GLU 127 LEU 184 LEU 188 LEU 184 LYS 185 LEU 188 LYS 185 GLN 189 LYS 287 GLN 285
β- or γ-Actin
Atom O CE2 CG ND2 CE1 O OD1 CD NE CZ OH O O O OH CG1 CA OE1 CD2 CD1 CG CA CD1 CG CD CE OE1
Distance (Å) 3.30 3.74 3.60 2.97 3.19a 3.48b 3.79b 3.26a 3.26 4.03 4.07 2.87 3.36 2.79 2.83 3.58 3.94 3.19 3.70 4.06 4.06 4.13 3.99 4.19 3.74 4.20 2.89
DBP
Residue
Atom
Residue
Atom
Distance (Å)
CYS 272 PHE 279 VAL 287 VAL 287
SG CE2 CG1 CG1
THR 400 PHE 399 SER 434 ASN 435
OG1 CD2 O OD1
3.36c 3.55 3.40 3.66
a
These interactions are absent in DBP–β-actin or DBP–γ-actin complexes. residues are mutated in β- or γ-actin and have different interactions with neighboring DBP residues. Their interactions are shown in the second part of the table. cThis interaction is not present in DBP–α-actin. bThese
146
ILSE BOGAERTS, CHRISTEL VERBOVEN, HUGO VAN BAELEN, AND ROGER BOUILLON
conformation as found for all other known ATP–actin structures. Although DBP interacts with the C-terminal part of actin, located in subdomain 1, the C-terminal actin residues 373 to 375 are disordered in DBP–actin and show no electron density. Similar to actin-GS1 [32] and actin-GS4–6 [33], residues 41 to 50 (i.e., the loop that forms a β-sheet with DNase I in the actinDNase I structure [34] are also disordered in our structure. According to limited proteolysis experiments performed with DBP [12] the actin binding site is located in its C-terminal part, and more specifically residues 350 to 403 (of domains II and III) are involved in actin binding. However, our DBP–actin structure shows that all three DBP domains interact with actin (Fig. 6, Table II). Besides residues 394–395 and 397–400, both part of the biochemically identified actin-binding region (residues 350–403), an additional 35 residues from other DBP regions also ascertain the interaction. Only minor folding differences are observed between the DBP conformations in the presence or absence of actin. The rms difference of 1.86 Å, obtained from superimposing all common α-carbons of the DBP–actin and of the DBP (PDB code 1J78) [1] structures, can mainly be ascribed to a different orientation of domain I and to a different folding in some loop regions. The domain by domain superposition gives lower rms difference values: 1.40 Å, 1.27 Å and 0.46 Å for domains I, II, and III, respectively. The larger rms difference value for domain I is caused by the interaction with actin and by crystal packing differences (Fig. 7). In order to allow the interaction between solvent-exposed hydrophobic DBP residues on the C-terminal helix of domain I (amino acid region 180–188) and the hydrophobic cleft formed by actin subdomains 1 and 3, the C-terminal part of DBP domain I has to readjust its orientation slightly (Fig. 7). In agreement with previous biochemical observations [25,36], the DBP–actin structure proves that the bound actin and the observed folding differences in domain I do not hinder the binding of vitamin D3 compounds to the vitamin D–binding site of DBP (Fig. 7). All presently missing residues in the DBP structure (98–104; 318–322; 361–364), belonging to regions that are poorly defined in the electron density map, are located opposite of the actin binding site. The DBP residues involved in actin binding are relatively well conserved between different species in all presently known DBP sequences. Together with the highly conserved amino acid sequence of actin throughout evolution (e.g., human and rabbit skeletal muscle actin are 100% identical) it explains why the
Neighboring DBP
Actin
25OHD3
Domain I DBP-25OHD3 Domain I DBP-actin
FIGURE 7
The superposition of the domains I of DBP–actin and of the DBP–25-hydroxyvitamin D3 (DBP–25OHD3) complex [1]. Domain I of DBP–actin is shown in dark grey, actin in light grey, domain I of DBP-25OHD3 in medium grey, and a neighboring DBP (at the top) present in the DBP–25OHD3 crystal is also in dark gray. Some of the interacting residues are shown in ball-and-stick representation as well as the 25-hydroxyvitamin D3. The C-terminal part of domain I (helices 7–10) of DBP–actin is shifted toward the actin to allow favorable contacts between DBP helix 10 and the cleft between actin subdomains 1 and 3. Owing to the presence of a neighboring molecule in the DBP-25OHD3 crystal, the N-terminal part of domain I (helices 1–6; i.e., the vitamin D binding site) is shifted toward this neighbor to make intermolecular hydrogen bonds and hydrophobic contacts. This superposition also illustrates that the binding of a vitamin D3 compound to DBP in the presence of actin is possible.
actin-binding property of DBP is universally observed in vertebrates. In comparison with α-actin, a twofold decrease of DBP binding affinity for the nonmuscle actin isoforms, β-actin and γ-actin, was observed [36]. The actin in this DBP–actin structure originates from rabbit muscle, i.e., α-actin. Among the α/β or α/γ sequence differences, the residues Met/Leu 176, Ala/Cys 272, Tyr/Phe 279, and Ile/Val 287 are located in the DBP binding site and the residues Asn/Thr 297, Met/Leu 299, and Thr/Ser 358 are situated in its surroundings. However, in silico replacement of these residues in the DBP–actin structure does not result in any unfavorable contacts (Table II), and there is no indication that the binding interface should fold differently to accommodate the replaced amino acids. Even, the interaction surface between β- or γ-actin and DBP, using the modeled DBP–actin β- or γ-complex for the calculations, remains similar (3609 Å2 compared to 3607 Å2 for DBP–α-actin). The lower affinity of DBP for both β- and γ-actin can only be ascribed to the loss of hydrophobic interactions between DBP residues 117 and 442 and the β or γ-actin residue 287, between DBP residue 398
147
CHAPTER 9 New Aspects of DBP
and the β- or γ-actin residue 176, and between DBP residue 400 and the β- or γ-actin residue 279 (Table II). 3. THE DBP–ACTIN BINDING INTERFACE COMPARED OTHER ACTIN COMPLEXES
TO
Several structures of G-actin in complex with an actin-binding protein have already been elucidated: actin–DNase I [34], actin–profilin [37], actin–GS1 [32], and actin–GS4-6 [33]. The actin surface buried by the complex formation with DBP is 3600 Å2, whereas the buried surfaces for the actin–DNase I, actin–GS1, actin–GS4-6, and actin–profilin complexes are 1800 Å2, 2100 Å2, 2100 Å2, and 2000 Å2, respectively. From all these complexes, DBP has the largest binding interface, indicating that the DBP–actin interaction is not merely fortuitous. The DNase I binding site of actin, consisting of residues from subdomains 2 and 4, is located opposite to the DBP binding site (subdomains 1 and 3). This corroborates the observation that the ternary complex of DBP, DNase I, and actin can be formed in vitro [25,38]. The interaction surfaces of profilin, gelsolin segment 1, and gelsolin segments 4–6 are all located at the same side of actin as the DBP binding site and are all partially overlapping. This structural comparison also demonstrates how DBP bound to actin obstructs its further interaction with profilin, as previously observed [39]. Gelsolin, consisting of six segments of 120 to 130 amino acids each (S1 to S6), has been reported to nucleate growth through the binding of two actin monomers in an appropriate orientation. DBP can prevent re-formation of actin filaments by reducing the effective G-actin concentration. The observation that DBP is capable of displacing one actin from the ternary gelsolin:actin (1:2) complex may also play a role in inhibiting the nucleation [40]. However, the current understanding of the structural organization of the gelsolin:actin (1:2) complex does not permit us to explain how DBP can remove one actin from it. For, in the proposed model of this ternary complex [33], the DBP binding interface is blocked in both actins either by segment 1 or by segment 4 of gelsolin. However, another study, involving cross-linking experiments with different gelsolin constructs, indicated that a ternary complex with the two actins in appropriate orientation for filament growth [41] is formed only with gelsolin segments 2–6, whereas with the whole gelsolin the actin monomers are in an antiparallel orientation as in the lower actin dimer. It is noteworthy that DBP, gelsolin segment 1, and gelsolin segment 4–6 have a hydrophobic helix that interacts with the hydrophobic patch in the cleft between subdomains 1 and 3 of actin. It can be postulated that other proteins binding actin at
the barbed end side and having an appropriate helix with solvent accessible hydrophobic residues, will bind to actin in such an orientation that this helix is also located in the cleft between subdomains 1 and 3. This may be the case for the N-terminal helix of the actin sequestering β-thymosins. Such an orientation of this helix would explain the cross-links between thymosin β 4 and actin: Lys 3–Glu 167 and Lys 18–Asp 1 [42]. Moreover, it was previously suggested that thymosin β 4 interacts with actin through a patch of hydrophobic residues located on the N-terminal helix [43]. This hypothesis also supports the previously suggested idea that the α3-helix of cofilin [44], an actin-severing protein, could interact with actin in a similar manner to its corresponding α-helix in gelsolin. 4. THE EXCELLENT ACTIN-SEQUESTERING PROPERTIES OF DBP
Actin monomers polymerize into double helical filaments (Fig. 8a) twisting around each other [45]. In contrast to other types of polymerization, the actin polymerization is not linear but involves a nucleation step, with a nucleus consisting of three subunits. The resulting microfilaments are polarized: the affinity for addition of new actin monomers is different (approximately 10-fold) at both ends. This can result, at intermediate actin monomer concentrations, in a polymerization of the filament at one end (the “fast growing” or “barbed” end) and a simultaneous depolymerization at the other end (the “slow growing” or “pointed” end) [46]. DBP binds G-actin at the side corresponding to the barbed or fast-growing end of actin in the F-actin filament. It thereby occludes the actin residues 166 to 169 and 286 to 289, all responsible for the longitudinal interactions between actin subunits within one F-actin strand [45]. Moreover, both regions are part of a hydrophobic pocket, which is postulated to be essential for the insertion of a hydrophobic plug from an actin subunit of the opposite actin strand. Although DBP does not make an immediate contact with actin residues 110 to 112, which are involved in interactions with an actin subunit of the opposite F-actin strand, its large volume sterically hinders the approach of the “opposite strand” subunit. The presence of DBP on G-actin therefore prevents the binding of two actin subunits: one that belongs in F-actin to the same strand, the other to the opposite strand (Fig. 8a). Hence, DBP not only blocks elongation of actin at this side, but also prohibits actin nucleation. The presence of DBP on actin seems not to hinder the interaction with other actin molecules at the pointed end side (Fig. 8a). Filament growth is, however, impossible. In conditions where the molar G-actin concentration is less than or equal to that of DBP, all G-actin subunits will be
148
ILSE BOGAERTS, CHRISTEL VERBOVEN, HUGO VAN BAELEN, AND ROGER BOUILLON
A
(-)
B
(+) FIGURE 8 The binding of DBP to actin prevents actin-filament formation. (A) Superposition of the actin from the DBP–actin complex with the central actin subunit of the F-actin model [45] (present in PDB file 1alm) [58]. The five actin subunits have different grey scales. The presence of DBP (dark grey) in the DBP–actin complex prevents the binding of two subunits, which would belong in an actin filament to opposite strands: the two bottom actin subunits shown in light grey. For clarity reasons, the actin from the DBP–actin is not shown. The + and − signs indicate the fast-growing (barbed) and slow-growing (pointed) ends of F-actin, respectively. (B) Different DBP–actin complexes with the DBP always shown in dark grey and the actin in the same grey scales as the central actin subunit and the pointed-end actin subunits of (A). The presence of the bulky DBP on all actin subunits prohibits the approach of these actin subunits to another actin subunit.
captured and blocked by DBP (Fig. 8b). In addition, the presence of DBP prevents the formation of the so-called lower dimer. In this actin dimer model, the two actin molecules are in antiparallel orientation, a prerequisite for enabling actin polymerization [47]. 5. IMPLICATIONS FOR THE ACTIN-SCAVENGER SYSTEM
The observed perfect fit between actin and DBP, as illustrated here, provides the structural basis for the important role of DBP in the extracellular actin-scavenger system. Furthermore, although DBP
and gelsolin are both present in large concentrations (µM) in blood, their capacity to scavenge actin may still be overwhelmed during massive cell injury. Indeed, the saturation of the actin-scavenger system leads to thrombi formation and microangiopathy [48], and excessive amounts of actin in the circulation may lead to a condition resembling “multiple organ dysfunction syndrome.” This syndrome was found to be associated with reduced serum DBP levels [49,50]. Moreover, DBP serum concentrations are reduced in some patients with hepatic failure [49,51]. The DBP concentrations not only have some value in predicting
149
CHAPTER 9 New Aspects of DBP
survival from hepatic failure [52,53] but also may identify high-risk patients after multiple trauma [54]. Consequently, the DBP–actin structure, revealing the DBP residues essential for the interaction with actin and for the actin sequestering, is of major importance for the development of a drug that could be applied in the pathological conditions just mentioned. Despite intensive analysis no DBP-deficient individual has been identified, leading to the suggestion that certain functions of DBP might be essential for survival. However, viable DBP knockout mice have been generated [55]. These mice were only used for analysis of DBP’s role in vitamin D metabolism and action. Although DBP knockout mice are viable in normal laboratory conditions, it might well be that the role of DBP in the actin-scavenger system is confined to cases of severe tissue damage in which DBP is essential for survival. This would be in line with our striking structural observations and the perfect conservation of DBP–actin interaction during the evolution of vertebrates. In the actin–GS1 [32], actin–GS4-6 [33], actin– profilin [37], and DBP–actin complexes and in actin covalently linked to tetramethylrhodamine-5-maleimide [56] (actin-TMR), actin filament formation is blocked at the barbed-end side. The structures of all these complexes illustrate that all these actin ligands block the cleft between subdomains 1 and 3. Even a small molecule such as TMR is able to prevent filament formation. Intracellular sequestering proteins such as thymosin β4 (5 kDa) and profilin (15 kDa) are rather small compared to their extracellular counterpart, DBP (51 kDa). These remarkable differences in size of the actin interaction surfaces between the extra- and intracellular sequestering proteins might be related to the fact that in the intracellular compartment an equilibrium between G-actin and F-actin has to be maintained. In extracellular space F-actin formation has to be prevented at any cost. Therefore, DBP, a better actin-sequestering protein, is present in high concentrations in the extracellular compartment. 6. COMPARISON OF THE THREE REPORTED DBP–ACTIN STRUCTURES
Although our DBP–actin structure was obtained from a crystal grown in a different space group (P21) to that of two other very recently reported DBP–actin structures (space group P212121; PDB codes 1kxp [35] and 1lot [57]), the structures are very similar. The Cα atoms of our DBP–actin structure and the 1lot structure superimpose with an rms difference of 0.791 Å (781 atoms superimposed), while the Cα atoms of our DBP–actin structure and the 1kxp structure superimpose
with an rms difference of 0.709 Å (777 atoms superimposed). The differences between these three structures mainly reside in the loop regions and the N and C termini and are mainly the result of different lattice contacts. In the paper of Otterbein and co-workers, the possibility of a conformational change within the C-terminal actin residues 365–375 is proposed, since in their structure (1kxp) this region is poorly defined in the electron-density map. However, the seven extra residues 365–371 in our structure assume a similar conformation as in the other reported actin structures. The 1kxp structure and our structure have Mg2+-ATP at their nucleotide-binding site, while the 1lot structure has Ca2+-ATP. The presence of a different ion does not alter the conformation of ATP or the conformations of the surrounding amino acids. Only the water coordination of the ions is different: Ca2+ has a pentagonal bipyramidal coordination with five water molecules and two O atoms from the ATP β- and γ-phosphates, compared with the nearly octahedral coordination of Mg2+, which has one less water molecule. In our DBP–actin structure and the 1lot structure, His 73 of the actin molecule is methylated, while in the 1kxp structure this is not the case. Although the peptide bond between Glu 72 and His 73 of actin has the same orientation in 1lot as in our structure, there is no hydrogen bond in 1lot between the main-chain carbonyl group of Glu 72 and the side chain of Arg 183. In 1kxp, the orientation of this peptide bond is different, making such a hydrogen bond impossible.
IV. A STRUCTURAL EXPLANATION FOR THE UNIQUE FUNCTIONS OF VITAMIN D–BINDING PROTEIN WITHIN ITS GENE FAMILY The unique functions of DBP within this protein family must arise from its distinctive fold. The different helical arrangement observed in domain I of DBP allows for the binding of vitamin D3 ligands. Superposition of helix 2 of domain I of DBP with that of HSA illustrates that helix 4 of HSA makes a closer contact with helices 2 and 3, leaving no room for a vitamin D3 molecule. Compared to HSA, α-fetoprotein, and afamin, the third domain of DBP is largely truncated at the C terminus. Moreover, the comparison of the DBP and HSA structures revealed remarkably differing orientations of the three domains in both structures [1]. The shorter domain III of DBP together with its distinctive fold allows DBP to bind actin in contrast to the other family members (Fig. 9).
150
ILSE BOGAERTS, CHRISTEL VERBOVEN, HUGO VAN BAELEN, AND ROGER BOUILLON
A
DBP domain I Actin HSA domain III
HSA domain I
DBP domain III DBP domain II HSA domain II
B
FIGURE 9
Superposition of the DBP from the DBP-actin structure with the HSA structure [2], illustrating that HSA cannot bind actin. (A) Stereoview of the overall structure of HSA (in gray) positioned on the DBP–actin structure based on the superposition of domains II of DBP and HSA. Only the actin subdomains 1 and 3 (light grey) are shown. Domain I of DBP is shown in dark grey, domain II, in grey and domain III in medium grey. The orientation of domains I–III in DBP and HSA is completely different. Consequently, the residues forming the actin-binding interface in DBP have a totally different arrangement in HSA. (B) Close-up view of (A) illustrating that owing to their different orientation, HSA helices 9 and 10 of domain I and helices 1–3 of domain III collide with parts of the actin structure.
References 1. Verboven C, Ragijns A, De Maeyer M, Van Baelen H, Bouillon R, De Ranter C 2002 A structural basis for the unique binding features of the human vitamin D-binding protein. Nat Struct Biol 9:131–136. 2. He XM, Carter DC 1992 Atomic-structure and chemistry of human serum-albumin. Nature 358:209–215. 3. Sugio S, Kashima A, Mochizuki S, Noda M, Koboyashi K 1999 Crystal structure of human serum albumin at 2.5 angstrom resolution. Protein Eng 12:439–446.
4. Cooke NE, Haddad JG 2003 In: “Vitamin D” Feldman D, Glorieux FH, Pike JW (eds) Vitamin D; 1st ed. Academic Press, Orlando, FL, pp. 87–101. 5. Curry S, Mandelkow H, Brick P, Franks N 1998 Crystal structure of human serum albumin complexed with fatty acid reveals an asymmetric distribution of binding sites. Nat Struct Biol 5:827–835. 6. Williams MH, Vanalstyne EL, Galbraith RM 1988 Evidence of a novel association of unsaturated fatty-acids with Gc (vitamin-D-binding protein). Biochem Biophys Res Commun 153:1019–1024.
CHAPTER 9 New Aspects of DBP
7. Ena JM, Esteban C, Perez MD, Uriel J, Calvo M 1989 Fattyacids bound to vitamin-D-binding protein (Dbp) from human and bovine sera. Biochem Int 19:1–7. 8. Bouillon R, Xiang DZ, Convents R, Vanbaelen H 1992 Polyunsaturated fatty-acids decrease the apparent affinity of vitamin-D metabolites for human vitamin D–binding protein. J Steroid Biochem Mol Biol 42:855–861. 9. Bhattacharya AA, Grune T, Curry S 2000 Crystallographic analysis reveals common modes of binding of medium and long-chain fatty acids to human serum albumin. J Mol Biol 303:721–732. 10. Barton GJ 1993 Alscript—a tool to format multiple sequence alignments. Protein Eng 6:37–40. 11. White P, Cooke N 2000 The multifunctional properties and characteristics of vitamin D-binding protein. Trends Endocrinol Metab 11:320–327. 12. Haddad JG, Hu YZ, Kowalski MA, Laramore C, Ray K, Robzyk P, Cooke NE 1992 Identification of the sterol-binding and actin-binding domains of plasma vitamin-D bindingprotein (Gc-globulin). Biochemistry 31:7174–7181. 13. Bouillon R, Okamura WH, Norman AW 1995 Structure– function relationships in the vitamin-D endocrine system. Endocr Rev 16:200–257. 14. Rochel N, Wurtz JM, Mitschler A, Klaholz B, Moras D 2000 The crystal structure of the nuclear receptor for vitamin D bound to its natural ligand. Mol Cell 5:173–179. 15. Ashcroft DM, Po ALW, Williams HC, Griffiths CEM 2000 Systematic review of comparative efficacy and tolerability of calcipotriol in treating chronic plaque psoriasis. Br Med J 320:963–967. 16. van den Bemd GJCM, Pols HAP, van Leeuwen JPTM 2000 Anti-tumor effects of 1,25-dihydroxyvitamin D-3 and vitamin D analogs. Curr Pharm Design 6:717–732. 17. Casteels K, Waer M, Laureys J, Valckx D, Depovere J, Bouillon R, Mathieu C 1998 Prevention of autoimmune destruction of syngeneic islet grafts in spontaneously diabetic nonobese diabetic mice by a combination of a vitamin D-3 analog and cyclosporine. Transplantation 65:1225–1232. 18. Cantorna MT, Hayes CE, DeLuca HF 1996 1,25-Dihydroxyvitamin D-3 reversibly blocks the progression of relapsing encephalomyelitis, a model of multiple sclerosis. Proc Natl Acad Sci USA 93:7861–7864. 19. Dusso AS, Negrea L, Gunawardhana S, Lopezhilker S, Finch J, Mori T, Nishii Y, Slatopolsky E, Brown AJ 1991 On the mechanisms for the selective action of vitamin-D analogs. Endocrinology 128:1687–1692. 20. Bouillon R, Allewaert K, Xiang DZ, Tan BK, Vanbaelen H 1991 Vitamin-D analogs with low affinity for the vitamin-D binding-protein—Enhanced in vitro and decreased in vivo activity. J Bone Miner Res 6:1051–1057. 21. Brown AJ 2000 Mechanisms for the selective actions of vitamin D analogues. Curr Pharm Design 6:701–716. 22. Tocchini-Valentini G, Rochel N, Wurtz JM, Mitschler A, Moras D 2001 Crystal structures of the vitamin D receptor complexed to superagonist 20-epi ligands. Proc Natl Acad Sci USA 98:5491–5496. 23. Lee WM, Galbraith RM 1992 Mechanisms of disease—the extracellular actin-scavenger system and actin toxicity. N Engl J Med 326:1335–1341. 24. Vasconcellos CA, Lind SE 1993 Coordinated inhibition of actin-induced platelet-aggregation by plasma gelsolin and vitamin-D-binding protein. Blood 82:3648–3657. 25. Van Baelen H, Bouillon R, De Moor P 1980 Vitamin-Dbinding protein (Gc-globulin) binds actin. J Biol Chem 255: 2270–2272.
151 26. Lind SE, Smith DB, Janmey PA, Stossel TP 1986 Role of plasma gelsolin and the vitamin-D-binding protein in clearing actin from the circulation. J Clin Invest 78:736–742. 27. Goldschmidt-Clermont PJ, Van Baelen H, Bouillon R, Shook TE, Williams MH, Nel AE, Galbraith RM 1988 Role of group-specific component (vitamin-D binding-protein) in clearance of actin from the circulation in the rabbit. J Clin Invest 81:1519–1527. 28. Dueland S, Nenseter MS, Drevon CA 1991 Uptake and degradation of filamentous actin and vitamin-D-binding protein in the rat. Biochem J 274:237–241. 29. Herrmanssdoerfer AJ, Heeb GT, Feustel PJ, Estes JE, Keenan CJ, Minnear FL, Selden L, Giunta C, Flor JR, Blumenstock FA 1993 Vascular clearance and organ uptake of G- and F-actin in the rat. Am J Physiol 265:G1071–G1081. 30. Verboven C, Bogaerts I, Waelkens E, Rabijns A, Van Baelen H, Bouillon R, De Ranter C 2003 Actin-DBP: the perfect structural fit? Acta Crystallogr D Biol Crystallogr 59: 263–273. 31. Houmeida A, Hanin V, Constans J, Benyamin Y, Roustan C 1992 Localization of a vitamin-D-binding protein-interaction site in the COOH-terminal sequence of actin. Eur J Biochem 203:499-503. 32. McLaughlin PJ, Gooch JT, Mannherz HG, Weeds AG 1993 Structure of gelsolin segment-1-actin complex and the mechanism of filament severing. Nature 364:685–692. 33. Robinson RC, Mejillano M, Le VP, Burtnick LD, Yin HL, Choe S 1999 Domain movement in gelsolin: a calciumactivated switch. Science 286:1939–1942. 34. Kabsch W, Mannherz HG, Suck D, Pai EF, Holmes KC 1990 Atomic structure of the actin–DNase-I complex. Nature 347:37–44. 35. Otterbein LR, Cosio C, Graceffa P, Dominguez R 2002 Crystal structures of the vitamin D-binding protein and its complex with actin: structural basis of the actin-scavenger system. Proc Nat Acad Sci USA 99:8003–8008 36. Mcleod JF, Kowalski MA, Haddad JG 1989 Interactions among serum vitamin-D binding-protein, monomeric actin, profilin, and profilactin. J Biol Chem 264:1260–1267. 37. Schutt CE, Myslik JC, Rozycki MD, Goonesekere NCW, Lindberg U 1993 The structure of crystalline profilin betaactin. Nature 365:810–816. 38. Goldschmidt-Clermont PJ, Galbraith RM, Emerson DL, Marsot F, Nel AE, Arnaud P 1985 Distinct sites on the the G-actin molecule bind group-specific component and deoxy-ribonuclease-I. Biochem J 228:471–477. 39. Goldschmidt-Clermont PJ, Vanalstyne EL, Day JR, Emerson DL, Nel AE, Lazarchick J, Galbraith RM 1986 Groupspecific component (vitamin-D binding-protein) prevents the interaction between G-actin and profilin. Biochemistry 25: 6467–6472. 40. Janmey PA, Stossel TP, Lind SE 1986 Sequential binding of actin monomers to plasma gelsolin and its inhibition by vitamin-D-binding protein. Biochem Biophy Res Commun 136:72–79. 41. Hesterkamp T, Weeds AG, Mannherz HG 1993 The actin monomers in the ternary gelsolin–2-actin complex are in an antiparallel orientation. Eur J Biochem 218:507–513. 42. Safer D, Sosnick TR, Elzinga M 1997 Thymosin β(4) binds actin in an extended conformation and contacts both the barbed and pointed ends. Biochemistry 36:5806–5816. 43. Van Troys M, Dewitte D, Goethals M, Carlier MF, Vandekerckhove J, Ampe C 1996 The actin binding site of thymosin β 4 mapped by mutational analysis. EMBO J 15:201–210.
152
ILSE BOGAERTS, CHRISTEL VERBOVEN, HUGO VAN BAELEN, AND ROGER BOUILLON
44. Fedorov AA, Lappalainen P, Fedorov EV, Drubin DG, Almo SC 1997 Structure determination of yeast cofilin. Nat Struct Biol 4:366–369. 45. Holmes KC, Popp D, Gebhard W, Kabsch W 1990 Atomic model of the actin filament. Nature 347:44–49. 46. Wegner A 1976 Head to tail polymerization of actin. J Mol Biol 108:139–150. 47. Steinmetz MO, Stoffler D, Hoenger A, Bremer A, Aebi U 1997 Actin: from cell biology to atomic detail. J Struct Biol 119:295–320. 48. Vercaeren I, Winderickx J, Devos A, Peeters B, Heyns W 1993 An effect of androgens on the length of the poly(A)-tail and alternative splicing cause size heterogeneity of the messenger ribonucleic acids encoding cystatin-related protein [corrected and republished with original paging, article originally printed in 1992 Endocrinology 131(6):2496–2502]. Endocrinology 132:2496–2502. 49. Schiodt FV, Ott P, Bondesen S, Tygstrup N 1997 Reduced serum Gc-globulin concentrations in patients with fulminant hepatic failure: association with multiple organ failure. Crit Care Med 25:1366–1370. 50. Dahl P, Schiodt FV, Klaer T, Ott P, Bondesen S, Tygstrup N 1998 Serum Gc-globulin in the early course of multiple trauma. Crit Care Med 26:285–289. 51. Schiodt FV, Bondesen S, Tygstrup N 1995 Serial measurements of serum Gc-globulin in acetaminophen intoxication. Eur J Gastroenterol Hepatol 7:635–640.
52. Lee WM, Galbraith RM, Watt GH, Hughes RD, McIntire DD, Hoffman BJ, Williams R 1995 Predicting survival in fulminant hepatic-failure using serum Gc protein concentrations. Hepatology 21:101–105. 53. Schiodt FV, Bondesen S, Petersen I, Dalhoff K, Ott P, Tygstrup N 1996 Admission levels of serum Gc-globulin: predictive value in fulminant hepatic failure. Hepatology 23:713–718. 54. Dahl B, Schiodt FV, Nielsen M, Kiaer T, Williams JG, Ott P 1999 Admission level of Gc-globulin predicts outcome after multiple trauma. Injury—Int J Care Injured 30: 275–281. 55. Safadi FF, Thornton P, Magiera H, Hollis BW, Gentile M, Haddad JG, Liebhaber SA, Cooke NE 1999 Osteopathy and resistance to vitamin D toxicity in mice null for vitamin D binding protein. J Clin Invest 103:239–251. 56. Otterbein LR, Graceffa P, Dominguez R 2001 The crystal structure of uncomplexed actin in the ADP state. Science 293:708–711. 57. Head JF, Swamy N, Ray R 2002 Crystal structure of the complex between actin and human vitamin D–binding protein at 2.5 angstrom resolution. Biochemistry 41: 9015–9020. 58. Mendelson R, Morris EP 1997 The structure of the actomyosin subfragment 1 complex: results of searches using data from electron microscopy and x-ray crystallography. Proc Natl Acad Sci USA 94:8533–8538.
CHAPTER 10
Endocytic Pathways for 25-(OH) Vitamin D3 THOMAS E. WILLNOW ANDERS NYKJAER
Division of Molecular Cardiovascular Research Max-Delbrueck-Center for Molecular Medicine D-13125 Berlin, Germany Institute of Medical Biochemistry, University of Aarhus, DK-8000 Aarhus C, Denmark
I. Introduction II. Renal Endocytosis of 25-(OH) Vitamin D3
III. Conclusions References
I. INTRODUCTION
II. RENAL ENDOCYTOSIS OF 25-(OH) VITAMIN D3
In the 1970s, Fraser and Kodicek [1] and DeLuca and co-workers [2] were the first to demonstrate that the kidney is responsible for the conversion of 25-(OH) vitamin D3 to a more polar metabolite that acts as a potent regulator of calcium transport in the intestinal mucosa. Shortly thereafter, this metabolite was identified as 1,25-(OH)2 vitamin D3 [3,4]. Conversion was shown to take place in the epithelial cells of the proximal convoluted tubules (PCT) [5,6] by the action of 25-hydroxyvitamin D3-1α-hydroxylase, a cytochrome P450 enzyme located in the mitochondria [7]. A large body of work has firmly established the pivotal role of the PCT in the conversion of 25-(OH) vitamin D3 to 1,25-(OH)2 vitamin D3, and the regulatory steps involved in this process [reviewed in 8,9]. The route, however, that delivers the precursor to PCT cells for metabolism has been less well characterized. In the circulation, 25-(OH) vitamin D3 is transported bound to the carrier vitamin D–binding protein (DBP), also known as group-specific component of serum (Gc-globulin) [10]. Similar to other steroids, 25-(OH) vitamin D3 can dissociate from its carrier and passively diffuse through the plasma membrane of target cells. In PCT, the uptake of the free vitamin is believed to proceed from the circulation via the basolateral site of the epithelium [11,12]. Recent studies have identified an alternative route for delivery of 25-(OH) vitamin D3 to the PCT. It involves endocytosis of 25-(OH) vitamin D3/DBP complexes via endocytic receptors expressed on the luminal surface of the epithelium. This chapter describes the role of endocytic pathways for 25-(OH) vitamin D3/DBP complexes in vitamin D metabolism, and the molecular components involved in this process. VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX
A. Physiology 1. 25-(OH) VITAMIN D3/DBP COMPLEXES ARE SUBJECT TO GLOMERULAR FILTRATION
Patients with nephrotic syndrome suffer from hypocalcemia due to low serum levels of 25-(OH) vitamin D3 and 1,25-(OH)2 vitamin D3. Hypovitaminosis D is caused by an increased urinary loss of 25-(OH) vitamin D3 bound to DBP, depleting the plasma vitamin D pool [13–15]. The same phenotype can be observed in nephrotic rats [16]. Because no 25-(OH) vitamin D3 or DBP is present in the urine of control subjects, increased glomerular permeability in nephrotic syndrome was held responsible for an abnormal filtration of 25-(OH) vitamin D3/DBP complexes. In the healthy kidney, 25-(OH) vitamin D3/DBP complexes were believed to be excluded from glomerular filtration. This assumption has to be revised because a number of studies now demonstrate that circulating 25-(OH) vitamin D3/DBP complexes are normally filtered through the glomerulus and retrieved by PCT cells from the primary urine. Foremost, this fact has been demonstrated in patients suffering from proximal tubular resorption deficiencies (i.e, Fanconi syndrome) [17] such as in Itai–Itai disease (cadmium-induced tubulopathy) [18], Imerslund–Grasbeck syndrome (IGS) (see Section II,C,1,b) [19] or multiple myeloma [20]. In these individuals, the glomerular filtration barrier is intact and only low-molecular-weight metabolites are lost into the urine because of the inability of the PCT to retrieve physiological ligands from the tubule lumen. Patients with tubular resorption deficiencies of various etiologies excrete DBP (Fig. 1A) and 25-(OH) Copyright © 2005, Elsevier, Inc. All rights reserved.
154
THOMAS E. WILLNOW AND ANDERS NYKJAER
A
1
B
2
3
4
5
6
7
8
9
FIGURE 1 Excretion of DBP in patients and in mice with tubular resorption deficiencies. (A) Western blot analysis demonstrates the presence of DBP in urine samples from patients with multiple myeloma (light chain disease) (lane 1), cystinosis (lane 3), Fanconi–Bickel syndrome (lane 4), and Imerslund–Grasbeck syndrome (lane 5), but not from a healthy control subject (lane 2). (Samples provided by D. Müller, Humboldt University Berlin.) (B) DBP excretion is also seen in urine samples from megalindeficient (lanes 6 and 8) but not from wild-type mice (lanes 7 and 9).
vitamin D3 [20], indicating that the vitamin–carrier complex is subject to normal filtration. Thus, the urinary loss of 25-(OH) vitamin D3/DBP in nephrotic syndrome may be explained in part by the excessive amounts of filtered high-molecular-weight proteins that saturate tubular retrieval pathways. Further evidence for a physiological role of glomerular filtration in renal vitamin D metabolism stems from the fact that glomerular filtration rates directly correlate with plasma 1,25(OH)2 vitamin D3 levels [21]. 2. MEGALIN, AN ENDOCYTIC RECEPTOR FOR 25-(OH) VITAMIN D3/DBP
The identity of the tubular retrieval pathway for 25-(OH) vitamin D3/DBP and its significance for vitamin D metabolism was uncovered through studies on megalin, an endocytic receptor abundantly expressed on the brush border (luminal) surface of the PCT cells (Fig. 2) [22,23]. Megalin is a scavenging receptor with broad ligand specificity that constitutes the main pathway for tubular reabsorption of filtered plasma proteins [24]. Mice genetically deficient for this receptor suffer from low-molecular-weight proteinuria and excrete the receptor ligands [20,25–27]. Among other proteins, megalin knockout mice lose DBP into the urine, which is normally retrieved by this receptor (Fig. 1B). No uptake of DBP is seen in megalin-deficient kidneys whereas endocytosis of DBP can easily be detected in renal tissue from control animals (Fig. 2). Both mice with an obligate or with a conditional kidney-specific megalin gene defect were used to evaluate the pathophysiological consequences of urinary
loss of DBP. These studies have highlighted the importance of tubular retrieval pathways for renal and systemic vitamin D metabolism [20,25,28,29]. Megalin-deficient mice excrete approximately 0.3 mg of DBP per day [25]. Because the same amount of protein is retrieved from the glomerular filtrate in a normal kidney in 24 hr, glomerular filtration likely accounts for most of the plasma turnover of the carrier protein [30]. Massive urinary loss of DBP in megalin knockout mice results in a concomitant loss of 25-(OH) vitamin D3 bound to the carrier protein (Table I). The total 25-(OH) vitamin D3 loss amounts to 0.3 to 0.5 ng in 24 hr [25]. On a vitamin D3–enriched diet, the plasma levels of 25-(OH) vitamin D3 and 1,25-(OH)2 vitamin D3 are reduced by more than 70% in the obligate and by 50% in the conditional knockout line (Table I). The less severe phenotype in the latter mouse model is due to the residual megalin activity that is retained in animals with kidney-specific megalin gene deletion (10% of normal levels) [29]. Owing to the dietary supplementation with vitamin D3, megalin-deficient mice can balance the constant urinary loss of 25-(OH) vitamin D3 and exhibit normal plasma calcium, phosphorus, and PTH levels. However, when placed on a vitamin D–depleted chow, the animals suffer from hypocalcemia and hyperparathyroidism (Table I) [29]. Their bones are characterized by a massive increase in osteoid surfaces and by a dramatic reduction in inactive and mineralizing surfaces and in total bone mineral content (Table II). Such features of osteomalacia as a consequence of hypovitaminosis D are also observed in mice with inactivation of the DBP gene [31], highlighting
155
Anti-DBP
Anti-megalin
CHAPTER 10 Endocytic Pathways for 25-(OH) Vitamin D3
FIGURE 2 Immunodetection of megalin and DBP in murine PCT. In wild-type (+/+) PCT cells, megalin (arrowheads) is present on the luminal cell surface where it internalizes filtered DBP, detectable in intracellular vesicles (arrows). No uptake of the carrier is seen in PCT of megalin knockout mice (−/−).
TABLE I Endocrine Parameters in Wild-Type Mice (Control) and in Animals with Obligate (Megalin−/−) or Kidney-Specific Megalin Gene Defect (Conditional) Genotypes Parameters Vitamin D–enriched diet Urine 25-OH D3 (nM/mM creatinine) Plasma 25-OH D3 (nM) 1,25-(OH)2 D3 (pM) PTH (pM) Calcium (mM) Phosphorous (mM) Vitamin D–depleted diet Plasma PTH (pM) Calcium (mM) Phosphorous (mM)
Control
n
Conditional
n
Megalin−/−
n
n.d.
8
10.9 ± 1.9
13
10.4 ± 4.1
6
—
92.8 ± 8.3 120.1 ± 7.0 52.1 ± 4.0 2.3 ± 0.04 3.2 ± 0.21
9 5 5 5 5
48.1 ± 6.4 66.4 ± 14.6 67.0 ± 6.3 2.2 ± 0.04 3.0 ± 0.08
8 4 5 5 5
27.2 ± 8.3 43.2 ± 17.5 68.5 ± 1.8 — —
4 4 3
0.0006 0.009 0.08 0.6 0.56
72.6 ± 4.0 2.2 ± 0.03 3.0 ± 0.08
10 10 10
88.3 ± 1.8 1.8 ± 0.1 3.1 ± 0.1
5 9 9
— — —
P
0.04 0.03 0.3
Data are the mean ± SEM. Statistical significance of differences between values in controls and conditional knockout mice are indicated (P). n, number of animals; n.d., not detectable.
156
THOMAS E. WILLNOW AND ANDERS NYKJAER
TABLE II Bone Morphometric Analysis of Wild Type Mice (Control) or Animals with Kidney-Specific Megalin Gene Defect (Conditional) on a Vitamin D–Depleted Diet (Mean ± SEM; n, number of animals) Parameters Bone mineral content in femur and tibia (mg) Osteoid surface (% of total surface) Inactive surfaces (% of total surface) Mineralizing surface (% of total surface)
Control (n = 5)
Conditional (n = 4)
P
50.4 ± 3.3 3.0 ± 1.7 92.4 ± 1.7 31.8 ± 6.0
29.5 ± 5.0 67.2 ± 17.9 18.7 ± 16.1 2.2 ± 0.9
< 0.01 0.02 < 0.01 < 0.01
the close functional relationship of the endocytic receptor megalin and the carrier DBP in vitamin D metabolism. Further support for a crucial role of the megalin retrieval pathway in uptake of 25-(OH) vitamin D3 comes from studies in rats treated with megalin antagonists. When receptor antagonists are infused into PCT of the rat, no conversion of 25-(OH) vitamin D3 to 1,25-(OH)2 vitamin D3 is seen, indicating that the megalin receptor pathway provides PCT cells with the precursor to produce the active hormone 1,25-(OH)2 vitamin D3 [25].
B. Cell Biology According to current hypotheses, free steroid hormones enter cells by passive diffusion through the
BN16 cells
35 Celluar uptake (% of added tracer)
plasma membrane. In contrast, steroids bound to transport proteins are considered biologically inactive and blocked from cell entry by their carriers [32]. This concept does not apply to the secosteroid 25-(OH) vitamin D3 complexed with DBP as shown in rat and mouse models [25, 29] and in cultured cells [19]. When Brown Norway rat choriocarcinoma (BN16) cells are incubated with 25-(OH) vitamin D3/DBP complexes more than 30% of the complexes are taken up within 4 hr (Fig. 3). The cellular uptake of the steroid is dependent on megalin, which is abundantly expressed in these cells, and can be blocked by a megalin antagonist, the receptor-associated protein (RAP) [33], or by anti-megalin antibodies. Very little uptake of complexed 3H-25-(OH) vitamin D3 is seen in keratinocytes that lack megalin expression (Fig. 3). Similar results have been obtained in other cell lines not expressing the megalin, indicating
30 25-OH-D3 /125I-DBP 25
3H-25-OH -D/DBP 3
20 15 10
Keratinocytes
5 0
FIGURE 3
buffer
α-megalin lgG
sheep lgG
RAP
buffer
Cellular uptake of 25-(OH) vitamin D3/DBP complexes. Brown Norway rat choriocarcinoma (BN16) cells and keratinocytes were incubated for 4 hr with 25-(OH) vitamin D3/DBP complexes labeled on either the steroid (3H-25-(OH) vitamin D3/DBP, open bars) or the protein moiety (25-(OH) vitamin D3/125I-DBP, closed bars). Cellular uptake of the carrier and the vitamin can be detected in megalin-expressing BN16 cells but not in keratinocytes that lack the receptor. Megalin-mediated uptake of complexed 25-(OH) vitamin D3 is blocked by the megalin antagonist receptor-associated protein (RAP) and by sheep anti-megalin antibodies, but not by control sheep IgG.
157
CHAPTER 10 Endocytic Pathways for 25-(OH) Vitamin D3
FIGURE 4
This immunoelectron micrograph depicts internalized DBP (arrows) in a lysosome of a murine PCT cell. (Provided by E. I. Christensen, University of Aarhus.)
that the efficient uptake of metabolites complexed with DBP requires the presence of megalin [34–37]. Like most internalized proteins, DBP is transported to lysosomes for degradation (Fig. 4) [25,38]. The intracellular route taken by 25-(OH) vitamin D3 through the endocytic compartments to the mitochondria is unknown and probably involves intracellular 25-(OH) vitamin D binding proteins (IDBP) (see Chapter 21) [39].
autoantigen in an induced glomerular nephritis model in rats, the passive Heymann nephritis [40]. In this animal model, autoantibodies directed against megalin form immunodeposits in the glomerular wall, causing severe tissue damage and renal failure [40, 41]. Megalin is a giant 600-kDa cell surface protein. Its gene covers 235,000 base pairs on the human chromosome 2q24-q31 [42] and consists of 79 exons [43]. The protein is highly conserved among species and can even be found in primitive organisms such as nematodes [44]. Sequencing of the rat and human megalin cDNA revealed a close structural similarity of this receptor with members of the LDL receptor gene family, a group of multifunctional endocytic receptors [43,45,46]. Similar to other family members, megalin is characterized by a number of structural domains required to perform endocytosis (Fig. 5). Its extracellular domain consists of 36 cysteine-rich complement-type repeats,
Complement-type repeat EGF YWTD
NH2
EGF precursor homology domain
CUB domain EGF-type repeat Amphipathic hellx
Membrane anchor NPXY motif
COOH
C. Molecular Biology Investigations in rats treated with megalin antagonists [25] or in mice with induced megalin gene defect [25,29] have established a central role of this receptor in renal uptake of 25-(OH) vitamin D3 metabolites in vivo. Recent studies have uncovered the molecular details of this pathway. Besides the receptor megalin itself, it involves co-receptors (cubilin), cellular adaptors (disabled-2), and components of the endocytic machinery (voltage-gated chloride channel-5), as well as extracellular (DBP) and intracellular 25-(OH) vitamin D3-binding proteins (IDBP). DBP and IDBP will only be mentioned briefly in this chapter because they are covered elsewhere in this textbook (Chapters 8 and 21, respectively). Other components of the pathway will be described in the following sections. 1. RECEPTORS AND CO-RECEPTORS
a. Megalin Megalin, also known as glycoprotein (gp) 330 or low-density lipoprotein (LDL) receptorrelated protein-2, was initially identified as the
NH2
Membrane
Megalin
COOH
Membrane
Cubilin
FIGURE 5 (Left panel) Structure of megalin. The extracellular domain of the receptor consists of clusters of complement-type repeats and epidermal growth factor (EGF) presursor homology domains. The clusters of complement-type repeats are the site of ligand binding. YWTD repeats in the EGF precursor homology domains are responsible for pH-dependent release of ligands. A single transmembrane domain mediates membrane attachment. NPXY motifs in the cytoplasmic tail are involved in internalization via coated pits and in cytosolic adaptor binding. (Right panel) Structure of cubilin. The cubilin polypeptide consists of 27 CUB domains, a cluster of 8 EGF repeats, and an amphipathic helix. CUB domains are involved in ligand binding. The amino-terminal amphipathic helix mediates peripheral membrane attachment.
158
bil in
ali n
Cu
α-
αM eg
organized in four clusters of seven to 11 repeats. The second cluster represents the main site of ligand binding [47]. All clusters of complement-type repeats are separated by epidermal growth factor (EGF) precursor homology domains composed of EGF-type repeats and tyrosine-tryptophan-threonine-aspartate (YWTD)containing spacer regions. The later sites are involved in pH-dependent release of ligands in endosomes [48]. A 22-amino-acid transmembrane domain mediates membrane attachment of the receptor. The cytoplasmic tail contains three asparagine-proline-X-tyrosine (NPXY) elements that represent signals for coated-pit internalization [49] and binding sites for cytosolic adapter proteins such as Disabled (Dab)-2 [50]. Megalin is expressed in many absorptive epithelia of embryonic and adult tissues, the predominant sites of expression being the yolk sac and the neuroepithelium of early embryos and the PCT and the small intestine in adults [24]. In these tissues, megalin expression is restricted to the apical cell surface and to endosomal compartments in line with a role of this receptor in apical clearance pathways [23]. Many ligands have been identified that bind to megalin in vitro and that are taken up by the receptor into cultured cells or in vivo. By and large, these ligands represent three major classes of molecules: (1) proteases and protease/inhibitor complexes, (2) lipoproteins, and (3) vitamins or hormones bound to carrier proteins [reviewed in 24,51]. Besides 25-(OH) vitamin D3/DBP, the receptor reabsorbs complexes of retinol (vitamin A) with the retinol-binding protein [26] and vitamin B12 with the carrier transcobalamin [27] from the primary urine into PCT, indicating a role as a general retrieval receptor for filtered metabolites. The function of the receptor in other epithelia is less well characterized. Forebrain defects in mice with obligate megalin gene deletion argue for an important role of the protein in brain development [52]. Megalin binds most ligands with moderate affinities (Kd 100–500 nM). Ligand binding is dependent on the presence of calcium that is required to stabilize the three-dimensional structure of the complement-type repeats [53]. DBP binds to megalin with a Kd of 108 nM regardless of whether 25-(OH) vitamin D3 is bound to the carrier or not [25]. b. Cubilin When Nykjaer et al. [19] used DBPaffinity chromatography to purify membrane proteins from renal tissues that bind this carrier, two DBPinteracting proteins were identified (Fig. 6). One was megalin, the other cubilin, a peripheral membrane protein expressed in the PCT and other epithelia [54]. Cubilin was initially identified as the receptor for uptake of intrinsic factor (IF)/vitamin B12 complexes in the
THOMAS E. WILLNOW AND ANDERS NYKJAER
DBP-Column
250
96 68 45 30 1
2
3
4
5
6
7
FIGURE 6 The kidney expresses two DBP receptors. Two proteins can be recovered by DBP-affinity chromatography from renal membrane extracts. Lanes 1–5 depict elution fractions from a DBPcolumn separated by SDS–polyacrylamide gel electrophoresis and stained with silver nitrate. Two high-molecular-weight DBP-binding proteins elute from the column, megalin and cubilin. Lanes 6 and 7 are immunoblots of the fraction in lane 4 incubated with antimegalin (lane 6) or anti-cubilin antisera (lane 7).
intestine and is therefore also known as the intrinsic factor/cobalamin receptor [55]. The human cubilin gene is located on chromosome 10p12.1 [56]. The sequence gives rise to a 460-kDa protein. So far, cubilin has been cloned from humans, rats, and dogs [56–58]. It received its name from a cluster of so called CUB domains that make up most of the receptor polypeptide (Fig. 5). CUB domains are 120 amino acids modules found in components of the complement system, such as C1r/C1s, in the EGF related sea urchin protein UEGF and in the bone morphogenic protein, BMP-1 (hence the name CUB). In addition to 27 CUB domains, cubilin possesses a short amino terminal amphipathic helix and 8 EGF-like repeats [57]. In contrast to prototype endocytic receptors such as megalin, cubilin lacks a membrane anchor and a cytosolic tail and is believed to associate with the plasma membrane via its hydrophobic amino terminal helix [59]. The protein directly binds to megalin and recycles through the endocytic compartments as a co-receptor complex (see Section II,C,1,c). In mice genetically deficient for megalin, cubilin biosynthesis and intracellular transport are significantly impaired [60]. Therefore, PCTs lacking megalin are also functionally deficient for cubilin [60, 61]. Besides in the PCT, cubilin is expressed in several absorptive epithelia including the yolk sac and the ileum.
159
CHAPTER 10 Endocytic Pathways for 25-(OH) Vitamin D3
At the ultrastructural level, cubilin with megalin colocalizes in the apical part of epithelial cells, in particular in clathrin-coated pits, in dense apical tubules, and in endosomes, in agreement with the requirement of the receptor to associate with megalin [62]. Investigations into the physiological role of cubilin have been facilitated by the identification of natural mutations in the corresponding gene in patients [63]. Defects in the cubilin gene were found in individuals suffering from megaloblastic anemia 1 or ImerslundGrasbeck syndrome (IGS), a rare recessive disorder characterized by a reduced number of red blood cells displaying an increased cell volume [63,64]. Consistent with a crucial role of endocytic receptors in tubular retrieval of filtered metabolites in the kidney, several plasma proteins have been identified that are cleared from the primary urine via cubilin. These ligands include DBP [19], transferrin [65], and albumin [61], as well as Clara cell secretory protein (CCSP), a transporter for steroid hormones [60]. As a consequence of the cubilin gene defect, patients with IGS exhibit lowmolecular-weight proteinuria and excrete ligand/carrier complexes such as 25-(OH) vitamin D3/DBP [19]. c. The Two-Receptor Model Based on studies in a number of experimental systems, Nykjaer and Willnow proposed a two-receptor model for the binding of 25-(OH) vitamin D3/DBP to the PCT cell surface and for the subsequent cellular uptake of the vitamin/carrier complexes (Fig. 7) [19,66]. According to this model, complexes of DBP and 25-(OH) vitamin D3 are reabsorbed from the glomerular filtrate by association of DBP with megalin, followed by endocytic uptake. Alternatively, the complexes are bound by cubilin first and internalized via interaction of this receptor
FIGURE 7 Two-receptor model for renal uptake of 25-(OH) vitamin D3/DBP. Details of this model are given in the text.
with megalin. Because cubilin acts as a surrogate binding site for DBP on the PCT surface, cubilin deficiency partially affects renal reabsorption of 25-(OH) vitamin D3, whereas DBP or megalin gene defects completely eliminate the uptake pathway. This two-receptor model is confirmed by the moderate vitamin D deficiency observed in cubilin-deficient patients and dogs as compared to DBP or megalin knockout mice [19,25,29,31]. 2. DISABLED-2
To recycle through the endocytic compartments, receptors require the assistance of scaffold or adaptor proteins, cytoplasmic factors that bind to the tail of the receptors and regulate their trafficking. Some adaptor proteins are common to most endocytic receptors such as AP-2 that localize receptors to the clathrin coat of endocytic vesicles [67]. Other adaptors are specific for certain receptors or even cell types. Disabled (Dab)-2 is an adaptor protein required for proper function of megalin in PCT. It harbors a phosphotyrosine-binding (PTB) domain with which it attaches to an NPXY motif in the megalin tail sequences [50]. Lack of Dab-2 expression in knockout mice results in impaired tubular endocytosis and in excretion of DBP [68]. The exact mode of action of Dab-2 still remains to be elucidated. However, findings obtained in Dab-2-deficient mice suggest a direct role of the adaptor in megalin trafficking and in endocytosis of 25-(OH) vitamin D3/DBP complexes. 3. VOLTAGE-GATED CHLORIDE CHANNEL-5
Additional support for a role of endocytic pathways in the renal metabolism of 25-(OH) vitamin D3 stems from studies in patients with Dent’s disease, a rare X-linked hereditary disorder characterized by disturbances in calcium homeostasis and bone metabolism [69]. This disease is caused by mutations in the gene encoding voltage-gated chloride channel-5 (ClC-5), a Cl− transporter expressed in endosomes of PCTs [70–72]. ClC-5 is believed to be responsible for sustaining chloride conductance required for import of H+ and acidification of the endosomes. Thus, ClC-5 defects in patients result in tubular endocytic dysfunction and in the inability to reabsorb filtered plasma proteins from the lumen of the PCT [73]. One explanation for defects in calcium homeostasis in ClC-5-deficient individuals comes from observations in mice lacking this ion channel. In ClC-5 knockout animals, endocytic malfunction results in disruption of megalin-mediated uptake of 25-(OH) vitamin D3/DBP complexes, in urinary loss of these metabolites, and in a three-fold decrease in plasma levels of 25-(OH) vitamin D3 and 1,25-(OH)2 vitamin D3 [74].
160
THOMAS E. WILLNOW AND ANDERS NYKJAER
III. CONCLUSION In recent years, a number of genetic models affecting renal vitamin D metabolism in patients and laboratory animals have been elucidated. Taken together, studies in these models have identified the existence of an endocytic pathway for 25-(OH) vitamin D3 metabolites in the PCT (Fig. 8). In this pathway, circulating 25-(OH) vitamin D3 metabolites are directed to the kidney through association with DBP, internalized from the glomerular filtrate by megalin and cubilin, two DBPbinding proteins that form a two-receptor complex, and delivered to lysosomal compartments. Recycling of the receptors megalin and cubilin requires Dab-2, a cytoplasmic adaptor for megalin. Inside the cells, DBP is degraded while 25-(OH) vitamin D3 is transferred into the cytoplasm for further activation and resecretion. The intracellular mechanisms that control the release of the vitamin D metabolite from the endocytic compartments and its transport to mitochondria, the site of hydroxylation, are unknown. However, the low solubility of 25-(OH) vitamin D3 strongly argues for the presence of intracellular transport proteins that facilitate the targeting of the steroid to the correct organelle. Such intracellular transport proteins have been identified for lipophilic compounds including vitamin A (cellular retinol–binding proteins) [75] and vitamin E (α-tocopherol transfer protein) [76]. Studies by Adams and co-workers indicate that two members of the constitutively expressed heat shock protein-70 family, designated intracellular vitamin D-binding protein-1 and -2 (IDBP-1, -2) serve similar functions for vitamin D [39,77].
FIGURE 8
The existence of an endocytic pathway for 25-(OH) vitamin D3 is intriguing as it is clearly distinct from the free diffusion of the metabolite into PCT. Uptake proceeds from the luminal rather than the basolateral side of the epithelium. Cell entry occurs in the DBP-bound, but not in the free form of the hormone. There is no doubt that this endocytic pathway plays a crucial role in the retrieval of filtered vitamin D molecules from the glomerular filtrate in order to prevent uncontrolled urinary loss of essential metabolites. The contribution of this pathway to production of 1,25-(OH)2 vitamin D3 in the kidney still remains to be clarified. However, the small fraction of free 25-(OH) vitamin D3 available for passive diffusion (0.003% of total) [78] and the large amount of precursor converted by the kidney every day (0.3 to 0.5 ng) [79] strongly argue for the existence of uptake pathways whereby the kidney gains access to the pool of bound 25-(OH) vitamin D3. The expression of megalin in other epithelia such as the small intestine, breast, and prostate that also exhibit 25-hydroxyvitamin D3-1α-hydroxylase activities may even indicate a role of this endocytic pathway in extrarenal conversion of 25-(OH) vitamin D3. In conclusion, the findings discussed in this chapter do not contradict other routes (such as free diffusion) for uptake of 25-(OH) vitamin D3 into the kidney (or other tissues). In fact, residual levels of 1,25-(OH)2 vitamin D3 are found in megalin- and DBP-deficient mice [29,31]. Whether they result from renal or extrarenal conversion of the precursor is unclear at present. Whatever the source of 1,25-(OH)2 vitamin D3 may be, this pool of the hormone is not able to sustain normal calcium and bone metabolism under dietary vitamin D restriction.
Endocytic pathway for renal uptake and conversion of 25-(OH) vitamin D3/DBP complexes. Details of this pathway are discussed in the text.
161
CHAPTER 10 Endocytic Pathways for 25-(OH) Vitamin D3
References 1. Fraser DR, Kodicek E 1970 Unique biosynthesis by kidney of a biological active vitamin D metabolite. Nature 228:764–766. 2. Gray R, Boyle I, DeLuca HF 1971 Vitamin D metabolism: the role of kidney tissue. Science 172:1232–1234. 3. Lawson DE, Fraser DR, Kodicek E, Morris HR, Williams DH 1971 Identification of 1,25-dihydroxycholecalciferol, a new kidney hormone controlling calcium metabolism. Nature 230:228–230. 4. Holick MF, Schnoes HK, DeLuca HF, Suda T, Cousins RJ 1971 Isolation and identification of 1,25-dihydroxycholecalciferol. A metabolite of vitamin D active in intestine. Biochemistry 10:2799–2804. 5. Brunette MG, Chan M, Ferriere C, Roberts KD 1978 Site of 1,25(OH)2 vitamin D3 synthesis in the kidney. Nature 276:287–289. 6. Kawashima H, Torikai S, Kurokawa K 1981 Localization of 25-hydroxyvitamin D3 1α-hydroxylase and 24-hydroxylase along the rat nephron. Proc Natl Acad Sci USA 78:1199–1203. 7. Paulson SK, DeLuca HF 1985 Subcellular location and properties of rat renal 25-hydroxyvitamin D3-1α-hydroxylase. J Biol Chem 260:11488–11492. 8. Bland R, Zehnder D, Hewison M 2000 Expression of 25-hydroxyvitamin D3-1α-hydroxylase along the nephron: new insights into renal vitamin D metabolism. Curr Opin Nephrol Hypertens 9:17–22. 9. Brown AJ, Dusso A, Slatopolsky E 1999 Vitamin D. Am J Physiol 277:F157–175. 10. Daiger SP, Schanfield MS, Cavalli-Sforza LL 1975 Groupspecific component (Gc) proteins bind vitamin D and 25-hydroxyvitamin D. Proc Natl Acad Sci USA 72:2076–2080. 11. Haddad JG 1995 Plasma vitamin D–binding protein (Gc-globulin): multiple tasks. J Steroid Biochem Mol Biol 53:579–582. 12. White P, Cooke N 2000 The multifunctional properties and characteristics of vitamin D–binding protein. Trends Endocrinol Metab 11:320–327. 13. Goldstein DA, Oda Y, Kurokawa K, Massry SG 1977 Blood levels of 25-hydroxyvitamin D in nephrotic syndrome. Studies in 26 patients. Ann Intern Med 87:664–667. 14. Barragry JM, France MW, Carter ND, Auton JA, Beer M Boucher BJ, Cohen RD 1977 Vitamin-D metabolism in nephrotic syndrome. Lancet 24:629–632. 15. Colston K, Williams NJ, Cleeve HJ 1985 Studies on vitamin D binding protein in the nephrotic syndrome. Clin Chem 31:718–721. 16. Mizokuchi M, Kubota M, Tomino Y, Koide H 1992 Possible mechanism of impaired calcium and vitamin D metabolism in nephrotic rats. Kidney Int 42:335–340. 17. Harrison HE 1953 The Fanconi syndrome. J Chron Dis 7: 346–355. 18. Teranishi H, Kasuya M, Aoshima K, Kato T, Migita S 1983 Demonstration of vitamin D–binding protein (Gc-globulin) in the urine of Itai-itai disease patients. Toxicol Lett 15:7-12. 19. Nykjaer A, Fyfe JC, Kozyraki R, Leheste JR, Jacobsen C, Nielsen MS, Verroust PJ, Aminoff M, de la Chapelle A, Moestrup SK, Ray R, Gliemann J, Willnow TE, Christensen EI 2001 Cubilin dysfunction causes abnormal metabolism of the steroid hormone 25(OH) vitamin D3. Proc Natl Acad Sci USA 98:13895–13900. 20. Leheste JR, Rolinski B, Vorum H, Hilpert J, Nykjaer A, Jacobsen C, Aucouturier P, Moskaug JO, Otto A, Christensen EI,
21.
22.
23.
24. 25.
26.
27.
28.
29.
30. 31.
32. 33. 34. 35. 36.
Willnow TE 1999 Megalin knockout mice as an animal model of low molecular weight proteinuria. Am J Pathol 155: 1361–1370. St John A, Thomas MB, Davies CP, Mullan B, Dick I, Hutchison B, van der Schaff A, Prince RL 1992 Determinants of intact parathyroid hormone and free 1,25-dihydroxyvitamin D levels in mild and moderate renal failure. Nephron 61: 422–427. Orlando RA, Farquhar MG 1993 Identification of a cell line that expresses a cell surface and a soluble form of the gp330/receptor-associated protein (RAP) Heymann nephritis antigenic complex. Proc Natl Acad Sci USA 90:4082–4086. Christensen EI, Nielsen S, Moestrup SK, Borre C, Maunsbach AB, de Heer E, Ronco P, Hammond TG, Verroust P 1995 Segmental distribution of the endocytosis receptor gp330 in renal proximal tubules. EJCB 66:349–364. Christensen EI, Willnow TE 1999 Essential role of megalin in renal proximal tubule for vitamin homeostasis. J Am Soc Nephrol 10:2224–2236. Nykjaer A, Dragun D, Walther D, Vorum H, Jacobsen C, Herz J, Melsen F, Christensen EI, Willnow TE 1999 An endocytic pathway essential for renal uptake and activation of the steroid 25-(OH) vitamin D3. Cell 96:507–515. Christensen EI, Moskaug JO, Vorum H, Jacobsen C, Gundersen TE, Nykjaer A, Blomhoff R, Willnow TE, Moestrup SK 1999 Evidence for an essential role of megalin in transepithelial transport of retinol. J Am Soc Nephrol 10:685–695. Birn H, Willnow TE, Nielsen R, Norden AG, Bonsch C, Moestrup SK, Nexo E, Christensen EI 2002 Megalin is essential for renal proximal tubule reabsorption and accumulation of transcobalamin-B12. Am J Physiol Renal Physiol 282: F408–F416. Hilpert J, Wogensen L, Thykjaer T, Wellner M, Schlichting U, Orntoft TF, Bachmann S, Nykjaer A, Willnow TE 2002 Expression profiling confirms the role of endocytic receptor megalin in renal vitamin D3 metabolism. Kidney Int 62: 1672–1681. Leheste JR, Melsen F, Wellner M, Jansen P, Schlichting U, Renner-Muller I, Andreassen TT, Wolf E, Bachmann S Nykjaer A, Willnow TE 2003 Hypocalcemia and osteopathy in mice with kidney-specific megalin gene defect. FASEB J 17: 247–249. Haddad JG, Fraser DR, Lawson DE 1981 Vitamin D plasma binding protein. Turnover and fate in the rabbit. J Clin Invest 67:1550–1560. Safadi FF, Thornton P, Magiera H, Hollis BW, Gentile M, Haddad JG, Liebhaber SA, Cooke NE 1999 Osteopathy and resistance to vitamin D toxicity in mice null for vitamin D binding protein. J Clin Invest 103:239–251. Mendel CM 1989 The free hormone hypothesis: a physiologically based mathematical model. Endocr Rev 10:232–274. Christensen EI, Gliemann J, Moestrup SK 1992 Renal tubule gp330 is a calcium binding receptor for endocytic uptake of protein. J Histochem Cytochem 40:1481–1490. Bikle DD, Gee E 1989 Free, and not total, 1,25-dihydroxyvitamin D regulates 25-hydroxyvitamin D metabolism by keratinocytes. Endocrinology 124:649–654. Vanham G, Van Baelen H, Tan BK, Bouillon R 1988 The effect of vitamin D analogs and of vitamin D–binding protein on lymphocyte proliferation. J Steroid Biochem 29:381–386. Adams JS 1984 Specific internalization of 1,25-dihydroxyvitamin D3 by cultured intestinal epithelial cells. J Steroid Biochem 20:857–862.
162 37. Manolagas SC, Deftos LJ 1980 Studies of the internalization of vitamin D3 metabolites by cultured osteogenic sarcoma cells and their application to a non-chromatographic cytoreceptor assay for 1,25-dihydroxyvitamin D3. Biochem Biophys Res Commun 95:596–602. 38. Keenan MJ, Holmes RP 1991 The uptake and metabolism of 25-hydroxyvitamin D3 and vitamin D binding protein by cultured porcine kidney cells (LLC-PK1). Int J Biochem 23:1225–1230. 39. Wu S, Ren S, Chen H, Chun RF, Gacad MA, Adams JS 2000 Intracellular vitamin D binding proteins: novel facilitators of vitamin D–directed transactivation. Mol Endocrinol 14: 1387–1397. 40. Kerjaschki D, Miettinen A, Farquhar MG 1987 Initial events in the formation of immune deposits in passive Heymann nephritis. gp330-anti-gp330 immune complexes form in epithelial coated pits and rapidly become attached to the glomerular basement membrane. J Exp Med 166:109–128. 41. Yamazaki H, Ullrich R, Exner M, Saito A, Orlando RA, Kerjaschki D, Farquhar MG 1998 All four putative ligandbinding domains in megalin contain pathogenic epitopes capable of inducing passive Heymann nephritis. J Am Soc Nephrol 9:1638–1644. 42. www.ncbi.nlm.nih.gov/cgi-bin/Entrez 43. Saito A, Pietromonaco S, Loo AK, Farquhar MG 1994 Complete cloning and sequencing of rat gp330/“megalin,” a distinctive member of the low density lipoprotein receptor gene family. Proc Natl Acad Sci USA 11 9725–9729. 44. Yochem J, Greenwald I 1993 A gene for a low density lipoprotein receptor-related protein in the nematode Caenorhabditis elegans. Proc Natl Acad Sci USA 90: 4572–4576. 45. Hjalm G, Murray E, Crumley G, Harazim W, Lundgren S, Onyango I, Ek B, Larsson M, Juhlin C, Hellman P, Davis H, Akerstrom G, Rask L, Morse B 1996 Cloning and sequencing of human gp330, a Ca2+-binding receptor with potential intracellular signaling properties. Eur J Biochem 239:132–137. 46. Willnow TE, Nykjaer A, Herz J 1999 Lipoprotein receptors: new roles for ancient proteins. Nat Cell Biol 1:E157–E162. 47. Orlando RA, Exner M, Czekay R-P, Yamazaki H, Saito A, Ullrich R, Kerjaschki D, Farquhar MG 1997 Identification of the second cluster of ligand-binding repeats in megalin as a site for receptor–ligand interactions. Proc Natl Acad Sci USA 94:2368–2373. 48. Rudenko G, Henry L, Henderson K, Ichtchenko K, Brown MS, Goldstein JL, Deisenhofer J 2002 Structure of the LDL receptor extracellular domain at endosomal pH. Science 298: 2353–2358. 49. Kibbey RG, Rizo J, Gierasch LM, Anderson RG 1998 The LDL receptor clustering motif interacts with the clathrin terminal domain in a reverse turn conformation. J Cell Biol 142:59–67. 50. Oleinikov AV, Zhao J, Makker SP 2001 Cytosolic adaptor protein Dab2 is an intracellular ligand of endocytic receptor gp600/ megalin. Biochem J 347:613–621. 51. Nykjaer A, Willnow TE 2002 The low-density lipoprotein receptor gene family: a cellular Swiss army knife? Trends Cell Biol 12:273–280. 52. Willnow TE, Hilpert J, Armstrong SA, Rohlmann A, Hammer RE, Burns DK, Herz J 1996 Defective forebrain development in mice lacking gp330/megalin. Proc Natl Acad Sci USA 93:8460–8464. 53. Fass D, Blacklow S, Kim PS, Berger JM 1997 Molecular basis of familial hypercholesterolaemia from structure of LDL receptor module. Nature 388:691–693.
THOMAS E. WILLNOW AND ANDERS NYKJAER
54. Christensen EI, Birn H, Verroust P, Moestrup SK 1998 Membrane receptors for endocytosis in the renal proximal tubules. Int Rev Cytol 180:237–284. 55. Seetharam B, Levine JS, Ramasamy M, Alpers DH 1988 Purification, properties, and immunochemical localisation of a receptor for intrinsic factor–cobalamin receptor in the rat kidney. J Biol Chem 263:4443–4449. 56. Kozyraki R, Kristiansen M, Silahtaroglu A, Hansen C, Jacobsen C, Tommerup N, Verroust PJ, Moestrup SK 1998 The human intrinsic factor vitamin B12 receptor, cubilin: molecular characterization and chromosomal mapping of the gene to 10p within the autosomal recessive megaloblastic anemia (MGA1) region. Blood 91:3593–3600. 57. Moestrup SK, Kozyraki R, Kristiansen M, Kaysen JH, Rasmussen HH, Brault D, Pontillon F, Goda FO, Christensen EI, Hammond TG, Verroust PJ 1998 The intrinsic factor– vitamin B12 receptor and target of teratogenic antibodies is a megalin-binding peripheral membrane protein with homology to developmental proteins. J Biol Chem 273:5235–5242. 58. Xu D, Kozyraki R, Newman TC, Fyfe JC 1999 Genetic evidence of an accessory activity required specifically for cubilin brush-border expression and intrinsic factor–cobalamin absorption. Blood 94:3604–3606. 59. Kristiansen M, Kozyraki R, Jacobsen C, Nexo E, Verroust PJ, Moestrup SK 1999 Molecular dissection of the intrinsic factorvitamin B12 receptor, cubilin, discloses regions important for membrane association and ligand binding. J Biol Chem 274:20540–20544. 60. Burmeister R, Boe IM, Nykjaer A, Jacobsen C, Moestrup SK, Verroust P, Christensen EI, Lund J, Willnow TE 2001 A tworeceptor pathway for catabolism of Clara cell secretory protein in the kidney. J Biol Chem 276:13295–13301. 61. Birn H, Fyfe JC, Jacobsen C, Mounier F, Verroust PJ, Orskov H, Willnow TE, Moestrup SK, Christensen EI 2000 Cubilin is an albumin binding protein important for renal tubular albumin reabsorption. J Clin Invest 105:1353–1361. 62. Sahali D, Mulliez N, Chatelet F, Dupuis R, Ronco P, Verroust P 1988 Characterization of a 280-kD protein restricted to the coated pits of the renal brush border and the epithelial cells of the yolk sac. Teratogenic effect of specific monoclonal antibodies. J Exp Med 167:213–218. 63. Aminoff M, Carter JE, Chadwick RB, Johnson C, Grasbeck R, Abdelaal MA, Broch H, Jenner LB, Verroust PJ, Moestrup SK, de la Chapelle A, Krahe R 1999 Mutations in CUBN, encoding the intrinsic factor–vitamin B12 receptor, cubilin, cause hereditary megaloblastic anaemia 1. Nat Genet 21: 309–313. 64. Imerslund O 1960 Idiopathic chronic megaloblastemic anemia in children. Acta Paediatr 49:208–209. 65. Kozyraki R, Fyfe J, Verroust PJ, Jacobsen C, Dautry-Varsat A, Gburek J, Willnow TE, Christensen EI, Moestrup SK 2001 Megalin-dependent cubilin-mediated endocytosis is a major pathway for the apical uptake of transferrin in polarized epithelia. Proc Natl Acad Sci USA 98:12491–12496. 66. Willnow TE, Nykjaer A 2002 Pathways for kidney-specific uptake of the steroid hormone 25-hydroxyvitamin D3. Curr Opin Lipidol 13:255–260. 67. Bansal A, Gierasch LM 1991 The NPXY internalization signal of the LDL receptor adopts a reverse-turn conformation. Cell 67:1195–1201. 68. Morris SM, Tallquist MD, Rock CO, Cooper JA 2002 Dual roles for the Dab2 adaptor protein in embryonic development and kidney transport. EMBO J 21:1555–1564. 69. Pook MA, Wrong O, Wooding C, Norden AG, Feest TG, Thakker RV 1993 Dent’s disease, a renal Fanconi syndrome
CHAPTER 10 Endocytic Pathways for 25-(OH) Vitamin D3
70.
71.
72.
73.
with nephrocalcinosis and kidney stones, is associated with a microdeletion involving DXS255 and maps to Xp11.22. Hum Mol Genet 2:2129–2134. Lloyd SE, Pearce SH, Fisher SE, Steinmeyer K, Schwappach B, Scheinman SJ, Harding B, Bolino A, Devoto M, Goodyer P, Rigden SP, Wrong O, Jentsch TJ, Craig IW, Thakker RV 1996 A common molecular basis for three inherited kidney stone diseases. Nature 379:445–449. Igarashi T, Gunther W, Sekine T, Inatomi J, Shiraga H, Takahashi S, Suzuki J, Tsuru N, Yanagihara T, Shimazu M, Jentsch TJ, Thakker RV 1998 Functional characterization of renal chloride channel, CLCN5, mutations associated with Dent’s Japan disease. Kidney Int 54:1850–1860. Yamamoto K, Cox JP, Friedrich T, Christie PT, Bald M, Houtman PN, Lapsley MJ, Patzer L, Tsimaratos M, van’t Hoff WG, Yamaoka K, Jentsch TJ, Thakker RV 2000 Characterization of renal chloride channel (CLCN5) mutations in Dent’s disease. J Am Soc Nephrol 11:1460–1468. Gunther W, Luchow A, Cluzeaud F, Vandewalle A, Jentsch TJ 1998 ClC-5, the chloride channel mutated in Dent’s disease, colocalizes with the proton pump in endocytotically active kidney cells. Proc Natl Acad Sci USA 95:8075–8080.
163 74. Piwon N, Gunther W, Schwake M, Bosl MR, Jentsch TJ 2000 ClC-5 Cl−-channel disruption impairs endocytosis in a mouse model for Dent’s disease. Nature 408:369–373. 75. Napoli JL 1996 Biochemical pathways of retinoid transport, metabolism, signal transduction. Clin Immunol Immunopathol 80:S52–62. 76. Ouahchi K, Arita M, Kayden H, Hentati F, Ben Hamida M, Sokol R, Arai H, Inoue K, Mandel JL, Koenig M 1995 Ataxia with isolated vitamin E deficiency is caused by mutations in the alpha-tocopherol transfer protein. Nat Genet 9:141–145. 77. Adams JS, Chen H, Chun RF, Nguyen L, Wu S, Ren SY, Barsony J, Gacad MA 2003 Novel regulators of vitamin D action and metabolism: lessons learned at the Los Angeles zoo. J Cell Biochem 88:308–314. 78. Bikle DD, Gee E, Halloran B, Kowalski MA, Ryzen E, Haddad JG 1986 Assessment of the free fraction of 25-hydroxyvitamin D in serum and its regulation by albumin and the vitamin D–binding protein. J Clin Endocrinol Metab 63:954–959. 79. Bikle DD, Siiteri PK, Ryzen E, Haddad JG 1985 Serum protein binding of 1,25-dihydroxyvitamin D: a reevaluation by direct measurement of free metabolite levels. J Clin Endocrinol Metab 61:969–975.
CHAPTER 11
The Vitamin D Receptor J. WESLEY PIKE AND NIRUPAMA K. SHEVDE Department of Biochemistry, University of Wisconsin–Madison, Madison, Wisconsin
I. II. III. IV.
Introduction Discovery of the Vitamin D Receptor Characterization of the Vitamin D Receptor Structural Gene for the Vitamin D Receptor
I. INTRODUCTION Many significant advances have been made during the past several decades in our understanding of the mechanisms through which steroid hormones function within target cells [1–6]. It is now well known that both the sex and adrenal steroids, thyroid hormone, retinoic acid (RA), and 1,25-dihydroxyvitamin D3 (1,25(OH)2D3) all bind to specific intracellular receptors. These proteins in turn transduce the environmental information carried by the small hormonal signals to the cell nucleus resulting in modifications in gene expression. Virtually all of the genes that encode nuclear receptors have now been cloned, revealing them to be part of a large gene family of proteins with highly related structural similarity. Interestingly, members of this family of nuclear receptor genes also transduce vertebrate signals generated by other forms of retinoic acid such as 9-cis-RA, fatty acids, nutritional metabolites, and invertebrate signaling molecules such as the insect molting hormone ecdysone [6] as well as the classical hormones. Perhaps most importantly, the cloning of the nuclear receptor family of genes has made available critical reagents that have led to important new insights into the structures of the proteins, an understanding of how hormonal signals activate these factors, delineation of the mechanisms by which they interact with the regulatory regions of target genes, and the discovery that numerous additional transcription factor complexes are essential for downstream gene modulation. The idea that vitamin D might function as a steroidlike hormone emerged in 1968 [7] and actually predated the documented discovery of 1,25(OH)2D3, the active hormonal form of vitamin D3 [8–10]. The lipophilic nature of the small vitamin D molecule and its capacity to localize in target tissues, as well as the fact that vitamin D responses were sensitive to transcriptional inhibitors, provided an initial basis for this hypothesis. Additional studies also defined the exquisite and highly complex nature of the regulation VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX
V. Functional Analysis of the Vitamin D Receptor VI. The Human Vitamin D–Receptor Chromosomal Gene VII. Concluding Comments References
of 1,25(OH)2D3 synthesis and production by the renal enzyme 25-hydroxyvitamin D3 1α-hydroxylase (see Chapter 5) [11]. Nevertheless, the discovery of a “binding protein” in specific target tissues that appeared to mediate nuclear localization of what eventually was determined to be the active vitamin D hormone, 1,25(OH)2D3 was perhaps the most significant [12]. Indeed, further characterization of this macromolecule over the years, and its eventual cloning in 1987 [13], confirmed the original hypothesis that 1,25(OH)2D3 was a steroid-like hormone. Its proposed mechanism in general terms is simply depicted in Fig. 1. These studies also showed that the actions of 1,25(OH)2D3 were directly linked to the gene regulatory properties of this “binding protein” or vitamin D receptor (VDR) [13–15], as alluded to in the figure. The structure and function of the VDR confirmed 1,25(OH)2D3 as an authentic steroid hormone and ushered in a new era of research on the molecular mechanisms underlying its regulation of gene expression. In this chapter, we provide an historical overview as well as contemporary summary of the VDR’s central role in mediating the actions of the vitamin D hormone. We describe the discovery of the receptor, a summary of its tissue and cellular distribution, and its biochemical characteristics. We also provide an overview of the VDR’s structural organization as well as how it functions to regulate gene expression. Finally, we describe the structural organization for the human VDR chromosomal gene. The reader is referred to the many additional chapters in the Mechanism of Action section of this book for more depth into each of the many aspects related to VDR structure, biochemistry, and biological function.
II. DISCOVERY OF THE VITAMIN D RECEPTOR Evidence for the existence of a binding protein or receptor (i.e., the VDR) was first provided by Haussler and Norman in 1969 [12]. These studies demonstrated Copyright © 2005, Elsevier, Inc. All rights reserved.
168
J. WESLEY PIKE AND NIRUPAMA K. SHEVDE
DB P-1 ,25 (O H)
2D 3
1,
25
(O H D3 )2
Target cell Nucleus R
R
mRNA
Tissue Specific Cellular Response
Mineral Homeostasis Bone Metabolism Vitamin D3 Catabolism Immunomodulation Differentiation Receptor Autoregulation
FIGURE 1
Model for the molecular mechanism of action of the vitamin D hormone. 1,25(OH)2D3 dissociates from serum vitamin D binding protein (DBP) or other carrier proteins, enters the cell, and interacts with the vitamin D receptor (VDR). Activation by the ligand leads to binding of the VDR to promoters within responsive genes and subsequent modulation of gene expression. Biological responses are indicated.
a preferential uptake of active vitamin into the intestinal chromatin fraction of chickens. They also provided further evidence that the process was saturable at low concentrations of administered active vitamin and was likely mediated by a specific protein component. Following these pioneering studies, more definitive evidence for the VDR began to emerge from the studies of Brumbaugh and Haussler [16] and Lawson and Wilson [17], who established the protein nature of the receptor via proteolytic digestion studies and equilibrium sedimentation analysis. Studies by Brumbaugh and Haussler [18] also established that the protein bound 1,25(OH)2D3 with high nanomolar affinity, an important prerequisite for mediating vitamin D action. Nevertheless, it was experiments carried out in vitro in 1975 that provided the most compelling evidence that
the protein might function as a nuclear receptor by demonstrating that cytosol-derived VDR could bind to chromatin fractions in a hormone-sensitive fashion [18]. Collectively, these findings provided the impetus for further exploration into the properties, structure, and function of the VDR.
III. CHARACTERIZATION OF THE VITAMIN D RECEPTOR A. Tissue Distribution of the VDR The VDR was first discovered in extracts prepared from chicken intestine, perhaps as a result of the considerable role the chicken as a model system had played in delineating the function of vitamin D in
169
CHAPTER 11 The Vitamin D Receptor
calcium homeostasis. Despite this, the known actions of vitamin D extended well beyond those in the intestine and were certainly not limited to the avian species. Target tissues of vitamin D included the kidney, bone, and the parathyroid glands and accordingly, the VDR was eventually identified in those tissues in avian as well as mammalian species [19–23]. The discovery of the VDR in mammalian intestine and in most of the listed tissues, nevertheless, required efforts over several years. Kream et al. [24] provided evidence for the existence of a VDR in rat intestinal tissue. Its properties were similar to those of the chicken receptor, although equilibrium sedimentation experiments suggested that rat receptor might exhibit a smaller molecular weight. Importantly, the discovery of receptors in these as well as other tissues was facilitated by tritiumlabeled 1,25(OH)2D3, whose availability with increasing radioisotopic specific activity provided the enhanced sensitivity necessary to detect VDR levels in tissues other than the intestine where it was most abundant. Thus, high specific radioactivity 1,25(OH)2D3 was a major determinant in the discovery and subsequent characterization of the VDR. 1,25(OH)2D3 with very high specific radioactivity also enabled detection of VDR in many unexpected target tissues as well as in cultured cell lines. The search for receptors in these tissue and cell sources was prompted by emerging evidence that the biological effects of 1,25(OH)2D3 extended beyond that of calcium and phosphorus homeostasis to include a role in cellular proliferation and differentiation [25,26]. Accordingly, VDRs were discovered in tissues such as pancreas [20,27], placenta [20], pituitary [28], ovary [29], testis [30], mammary gland [31], and heart [32]. Cultured cell lines that expressed the VDR were similarly extensive and now include cells of fibroblastic, chondrocytic, osteoblastic, myoblastic, hematopoietic, and lymphopoietic origin as well as cells derived from normal kidney, intestine, and skin. Of more fundamental nature, they also include cells of both tumorigenic and nontumorigenic origin [33–36]. A partial listing of vitamin D target tissues and cells is documented in Table I. The biological role of 1,25(OH)2D3 in many of these tissues and cells is considered more extensively in many additional chapters in this book. Perhaps the most interesting biological effect of vitamin D metabolites, however, is their ability to regulate proliferation and differentiation [37]. These effects of 1,25(OH)2D3 are not unlike those of most steroid hormones. Most importantly, however, the actions highlight a potential therapeutic role for 1,25(OH)2D3 and its analogs in cancer treatment, an area considered in more depth in Chapters 84–88 and 93–97, and in the regulation of the immune system [38,39], considered in Chapters 36, 98, and 99.
TABLE I Cellular and Tissue Distribution of the VDR System Gastrointestinal Hepatic Renal Cardiovascular Endocrine Exocrine Reproductive
Immune Respiratory Musculoskeletal
Epidermis/appendage Central nervous system Connective tissue
Tissue Esophagus, stomach, small intestine, large intestine, colon Liver parenchyma cells Kidney, urethra Cardiac muscle Parathyroid gland, thyroid, adrenal, pituitary Parotid gland, sebaceous gland Testis, ovary, placenta, uterus, endometrium, yolk sac, avian chorioallantoic membrane, avian shell gland Thymus, bone marrow, B cells, T cells Lung alveolar cells Bone osteoblasts and osteocytes, cartilage chondrocytes, striated muscle Skin, breast, hair follicles Brain neurons Fibroblasts, stroma
Antiproliferative effects of vitamin D are currently being exploited in the treatment of human psoriasis, a hyperproliferative disorder of the skin [40] that is another important target of vitamin D action [37,41] outlined in more detail in Chapter 101. Indeed, the clear presence of the VDR in most tissues in avian and mammalian species indicates that the biological effects of vitamin D are likely to be very widespread. While the analysis of tissue extracts led to the discovery of many VDR-containing sites, it was the use of scintillant-enhanced autoradiography employed early on by Zile et al. [42] and Jones and Haussler [43] and later by Stumpf et al. [44,45] that led to a thorough appreciation of the wide cellular distribution of vitamin D target tissues. These studies revealed such targets as the duodenum, ileum, and jejunum of the rat intestine, stomach, kidney, skin, and pituitary as well as several neurons in the brain. While the presence of 1,25(OH)2D3 (or a metabolite of the administered compound) in these tissues through autoradiographic means does not provide direct evidence of the presence of VDR, the bulk of these targets were confirmed when an antibody to the VDR was developed some years later enabling the use of immunocytochemistry [46–49]. Most recently, studies using Northern blot [13], in situ hybridization [50], or RT-PCR [51] analyses have been employed to detect the VDR mRNA in cells and tissues.
170 The availability of probes that enabled these analyses closely followed the cloning of the VDR in 1987 and 1988 (to be discussed in Section IV). Although this RNA detection approach has been useful, it often ignores the important underlying assumption that VDR mRNA transcripts correlate directly with VDR protein. The use of these nucleic techniques has been helpful, however, in the study of the transcriptional regulation of VDR production.
B. Subcellular Distribution of the VDR Although the site of action of ligand-bound VDR in the nucleus is clear, the subcellular location of unliganded VDR has remained controversial over several decades. Early studies suggested that the unliganded VDR was cytosolic (see Fig. l and Brumbaugh and Haussler [18]). Studies in the early 1980s, however, suggested that the VDR might be loosely associated with the nuclear fraction [52,53]. Indeed, the application of immunocytochemistry supports this largely nuclear localization of the apo form of the VDR, an intracellular residence site consistent with that of most other nuclear receptors. It is important to note, however, that the presence of the unliganded VDR within the nucleus does not imply that the VDR must be bound to regulatory sites of action on 1,25(OH)2D3 responsive genes. To the contrary, recent chromatin immunoprecipitation assays to be discussed later suggest that the VDR is not bound to these regulatory sites until activated by 1,25(OH)2D3 [54,55], supporting the need to identify these alternative localization sites for VDR in the nucleus in the absence of hormone. Perhaps not surprisingly, the controversy regarding the intracellular site of residency of unliganded VDR has reemerged recently. In studies using fluorescent probes and discussed in greater detail in Chapter 22, Barsony and colleagues [57,58] provide evidence that recombinantly derived VDR distributes into both the cytoplasmic and nuclear compartments in the absence of ligand and is imported into the nucleus via several essential components, including specific nuclear import signals within the VDR molecule, interaction with RXR, and interaction with nuclear import proteins lining the nuclear pores. Moreover, the VDR actively cycles between the two compartments. With the qualification that the VDR is expressed in relatively high concentrations in the cell as a chimeric protein fused to green fluorescent protein (GFP), the data seem to support strongly active receptor movement between both cellular compartments. Given these caveats, however, it is likely that the controversy will continue.
J. WESLEY PIKE AND NIRUPAMA K. SHEVDE
C. Biochemical Properties and Organization of the VDR The VDR is expressed in low concentrations in target tissues and cultured cells, a property consistent with its role as a potent transcription regulatory molecule [58,59]. Estimates of receptor abundance range from under 200 to over 25,000 copies of VDR/cell (10 to 100 fmol/mg protein) depending on the cell type or cell line examined, and up to 1 pmol/mg protein in certain tissue extracts. Differences in VDR abundance suggest that cells with higher VDR content may be more highly responsive to 1,25(OH)2D3 than those with lower levels. Little evidence supports this hypothesis, however, and in fact recent data suggest that the expression levels of many of the comodulatory complexes discussed in Chapters 14, 16, and 17 may be more significant determinants of cellular response than the receptor concentration itself [60]. Other factors may also contribute to the level of response to 1,25(OH)2D3, including cellular capacity to internalize and metabolize 1,25(OH)2D3, differential activation events that involve the VDR directly (phosphorylation or other posttranslational modifications), and finally the accessibility and individual activation properties of specific gene targets. These, as well as additional events, are likely to contribute significantly to the biological responsiveness and cellular sensitivity to 1,25(OH)2D3. Numerous physical and functional properties of the VDR were defined following the protein’s discovery in 1973, several of which are documented in Table II. Sedimentation analysis revealed a protein of 3 to 3.7 S that exhibited an elongated shape. Gel filtration estimates of protein size ranged from 50,000 to 70,000 Da
TABLE II
Biochemical Properties of the VDR and Its RNA Transcript
VDR protein Molecular weight: 50 kDa (human) to 60 kDa (chicken) Sedimentation coefficient (S): 3.2 (human, rat) to 3.7 (chicken) Equilibrium dissociation constant: 1–2 × 10–10 M Isoelectric point: 6.2 Phosphoprotein: Serine phosphorylated at residue 208 (human) in response to 1,25(OH)2D3 Lability: Proteolytically sensitive Tissue abundance: 10 to 1000 fmol of VDR/mg protein hVDR mRNA 4.8 kb (human), 3.0 (chicken)
171
CHAPTER 11 The Vitamin D Receptor
depending on species. The most important characteristic of the VDR, however, was its capacity to bind 1,25(OH)2D3 with both high affinity and selectivity [61–65]. Accordingly, the VDR displayed an equilibrium dissociation constant (KD) of approximately 10−10 M for the natural ligand 1,25(OH)2D3 and bound this ligand’s precursors as well as less active metabolites with substantially lower affinities [63,64]. The contribution of both the 25-hydroxyl and the 1α-hydroxyl groups on the 1,25(OH)2D3 molecule to specific, high-affinity binding to VDR has been studied extensively [64] and will be discussed later. An important discovery in the late 1970s was the observation that the VDR could bind to DNA, an anticipated property of a protein that activates transcription [58]. While a more thorough understanding of VDR DNA binding emerged following the identification of specific DNA-binding sites (vitamin D response elements or VDREs) within gene promoters, as discussed later in this chapter and other chapters, the finding that the VDR bound to nonspecific DNA set the stage for the ensuing studies aimed at understanding the structural organization and function of the VDR. Perhaps the most important observation made of VDR DNA binding properties was the finding that while VDR exhibited an obvious “affinity” for DNA in the absence of 1,25(OH)2D3, a feature unlike that of most steroid receptors, its “affinity” for DNA quantitatively increased following complex formation with 1,25(OH)2D3 [65,66]. The latter property implied that the structure of the VDR or perhaps the composition of the active receptor was transformed in the presence of the hormonal ligand. Recent X-ray diffraction studies to be described later in this chapter as well as in Chapter 15 define the nature of the effects of ligand on the VDR and its DNA binding capabilities at the atomic level [67–69]. Purification of the VDR initially from chicken intestine [14,58] and later from porcine intestine [15] led to the development of monoclonal antibodies to the VDR. These reagents proved essential in further immunological characterization of the VDR, provided alternative receptor isolation methods, and ultimately offered the means whereby the VDR gene was cloned. Western blotting analysis revealed for the first time that the precise molecular mass of VDR varied from species to species, ranging from 60,000 Da in the chicken [70] to approximately 50,000 Da in humans [71]. These relative sizes were confirmed through molecular cloning. Immunological analyses of this type also revealed that the VDR comprised A and B isoforms in the chicken but a single form in mammalian-derived tissues [70]. The two forms of the chicken VDR are
now known to arise from the existence of alternative start sites in the mRNA, although the functional consequence of two such proteins with differing aminoterminal extensions remains unknown [72,73]. Finally, immunoprecipitation of the VDR from in vitro translated tissue mRNA revealed that whereas the chicken receptor protein was larger than that from mammalian cells, the mRNA transcripts were considerably smaller [71]. The differential size of these transcripts has been confirmed through the use of hybridization techniques incorporating the cDNA for the VDR as a probe. Initial insights into the structural organization of the VDR emerged using techniques such as limited proteolytic digestion together with radiolabeled 1,25(OH)2D3 and immunologic probes. Initial studies by Pike [74] suggested that the VDR epitope for a specific monoclonal antibody to the VDR lay immediately adjacent to the domain necessary for DNA binding. Additional studies by Allegretto and co-workers [75,76] employing limited digestion with trypsin led to the realization that to the l,25(OH)2D3-bound VDR could be cleaved into two domains, one a large fragment of 30 to 40 kDa that retained prebound 1,25(OH)2D3 and a second of 16 to 20 kDa that could bind DNA independently. Most importantly, the findings suggested that the two functional domains (ligand-binding and DNA-binding domains) were separated by a proteolytically sensitive “hinge.” The inability following trypsin digestion to identify the domain responsible for ligand binding of hormone-free receptor argued strongly for a ligandinduced change in receptor conformation. This finding substantiated earlier observations of increased VDR stability in tissue extracts following complex formation with ligand. Additional preliminary evidence indicated that the DNA binding domain was located amino terminal to the hormone binding domain [77]. As described in the next section, these rudimentary insights into the overall structural organization of the VDR were proven correct via the subsequent cloning of the receptor’s structural gene.
IV. STRUCTURAL GENE FOR THE VITAMIN D RECEPTOR A. Cloning of the VDR The biochemical purification of the steroid receptors in the early 1980s led to the development of immunological probes that facilitated not only further characterization of the proteins, but the cloning of their structural genes as well. As a result, the receptors for virtually all the known steroid, thyroid, and
172 vitamin hormones were cloned during the latter half of the 1980s [2,3]. Interestingly, sequence analysis of the first cloned receptor cDNAs revealed regions that exhibited significant sequence similarity with the DNA binding domain of an unrelated transcription factor TFIIIA. In addition, sequence similarities within this specific region among the several nuclear receptors suggested the possibility that related transcripts might be recovered through low-stringency hybridization screening [3]. Accordingly, this led to the recovery of numerous additional cDNAs for unknown nuclear receptors or orphan receptors. For many of these, cognate ligands, if they exist, remain currently unidentified. However, investigation of several of these unknown receptors led to the eventual discovery of novel ligands and the identification of new hormonal systems. Examples include the discovery of retinoid X receptor (RXR) and its ligand 9-cis RA [78], peroxisome proliferator-activated receptor gamma (PPARγ) and one of its ligands prostaglandin J2 [79], farnesol–X receptor (FXR) and its ligand farnesol [80], the liver X receptor (LXR) and its oxysterol ligands [81,82], and several others. Interestingly, the sequence of the DNA binding domain of the glucocorticoid receptor (GR) was first noted to exhibit a high degree of similarity with the viral oncogene v-erbA [83]. The cloning of c-erbA led to the discovery that this gene encoded the first of two thyroid hormone receptor (TR) genes [84]. The development of the anti-VDR monoclonal antibody 9A7 enabled McDonnell et al. [13] to screen randomly primed chicken intestinal cDNA expression libraries and eventually to recover a single cDNA clone that encoded a zinc finger containing DNA binding domain. This domain exhibited high homology to the DNA binding domain of GR. The presence of this domain and its reactivity to the 9A7 antibody (known to interact in or near the DNA binding domain of the VDR [65]) led to initial confidence that this and several other isolated cDNAs represented a portion of the transcript encoding the chicken VDR. Subsequent hybridization-selected in vitro translation techniques using these clones substantiated the authenticity of the cDNA clones [13]. The recovery of this cDNA allowed the subsequent rapid recovery of full-length VDR cDNA transcripts from human [85] tissue sources. The rat intestinal VDR was cloned shortly thereafter by Burmester et al. [86]. Recovery of a cDNA transcript from the human leukemia HL-60 cell line [87] revealed virtual identity to that of the original human VDR cDNA cloned by Baker et al. [85]. Sequences of the VDR have been reported from mouse [88], Japanese quail [89], and Xenopus [90] and have appeared in both the human and mouse genomic data bases [91,92]. The cloning of the VDRs from various
J. WESLEY PIKE AND NIRUPAMA K. SHEVDE
sources and species was a milestone in the vitamin D field. Perhaps as important, it provided the first direct evidence of a structural relationship between the VDR and other bona fide members of the steroid receptor family of genes [93]. The VDR is highly conserved across tissue sources and species. The overall sequence similarity of rat, mouse, and avian receptors to that of human VDR is 79, 86, and 66%, respectively. This homology rises to above 95% within specific domains such as the DNA binding domain (see Chapter 13). There is, nevertheless, considerable variability in the number of residues amino terminal to the DNA binding domain. This region varies from 21 amino acids in the human VDR (the smallest of the VDRs) to approximately 57 amino acids in the chicken protein [85,89]. In addition, the chicken gene exhibits one alternative start site (the first at nucleotide 47 and the second at nucleotide 67, which corresponds to the most 5′ ATG in the human VDR transcript identified by Baker et al. [85]) that accounts for the two forms of the VDR found in the avian species [72,73]. The human VDR cDNA sequence also presents as a transcript with two potential start sites, one lying only three codons downstream of the first [85]. Whether both initiation codons are used in this transcript is unclear. However, an important polymorphism in the VDR gene has been demonstrated in the human population [94,95]. This polymorphism is a single-base-pair transition that occurs in the first initiation site reported by Baker et al. [85] (ATG to ACG). The result of this polymorphism in humans is the production of a smaller VDR of 424 amino acids that occurs in homozygous form in approximately 37% of the human population and in heterozygous form (one allele containing both initiation sites and the second containing only the downstream site) in approximately 48% of the population [96] (Chapter 72). Whether the two proteins exhibit different biochemical properties and/or different functions remains an open question, although it does appear that the smaller form may be more active [96–98]. This has been suggested to occur via a tighter contact with the transcription factor TFIIB [96]. Association studies indicate that this polymorphism may be linked to bone mineral density [97], although these data are not particularly robust [99–101] (Chapter 68).
B. VDR Is a Member of the Intracellular Receptor Gene Family The cloning of glucocorticoid [83] and estrogen [102] receptors in 1985 and 1986 were the first of a series of successful efforts to clone the intracellular receptor genes. More than 150 cloned members of this
173
CHAPTER 11 The Vitamin D Receptor
intracellular receptor gene family [5] have been identified, making it the largest transcription factor family currently known. It is in fact noteworthy that the cloning of this family of genes was so successful that publication of the sequence of both the mouse and human genomes in 2002 did not reveal a single uncloned and unidentified functional nuclear receptor gene [91,92]. The evolutionary success of this family of transcription factors is highlighted both by its abundance and by its involvement in an incredibly wide range of biological processes that include growth, development, differentiation, and adult homeostasis. The reader is referred to other reviews on the steroid receptor family for details [1–6] as well as summaries in Chapters 13–16.
A
C. Structural Domains of the VDR 1. OVERVIEW
The cloning of the estrogen receptor (ER) in 1986 and the GR shortly thereafter revealed that these proteins were made up of distinct regions or domains, leading to their designation as domains A, B, C, D, E, and F [102]. As illustrated in Fig. 2A, the highly conserved DNA binding domain is designated the C domain and provides a central focus for the remainder of the molecules. Accordingly, the nonconserved A/B domain comprises all residues amino terminal to the DNA binding domain and the D domain or hinge region comprises residues lying downstream of C. The carboxyterminal region that contains the ligand binding
NR A/B
B
C 1
24
D 89
E/F 232
427
41 6 42 3
1
9
34
31
27
23
2
C
2
hVDR
427
44 8 45 5
7
5
37
35
0 31
27
0
E/F Domain VDR
456
40 8 41 5
8
6
33
31
1 27
23
4
TRβ
462
90 7 91 4
0
8
83
80
1 76
72
1
RARα
934
ER
53 8 54 5
45 7
43 5
38 9
34 9
PR
595
FIGURE 2 Functional domain structure of the nuclear receptor superfamily. (A) The nuclear receptors (NR) are separated into four regions designated A/B, C, D, and E/F. (B) Residue boundaries of corresponding regions within the human VDR. (C) Three regions of sequence similarity and residue boundaries within the E/F domain of several nuclear receptors. Shown in addition to VDR are thyroid receptor β, (TRβ), retinoid receptor α (RARα), progesterone receptor (PR), and estrogen receptor (ER). Functions associated with the domains include transactivation (A/B), DNA binding (C), flexible hinge (D), and dimerization, ligand binding, transactivation, and repression (E/F) (see text for details).
174 domain is termed the E or E/F domain. Three stretches of sequence similarity exist in the E domain for virtually all of the nuclear receptor gene family. The F domain exhibits extensive variability and is not conserved, much like that of the A/B domain. The domain structure of mammalian VDRs is depicted in Fig. 2B. As can be seen, the A/B domain is very short relative to other members of the nuclear receptor family, representing one of the most abbreviated of the entire gene family. The C region comprises the DNA binding domain of the VDR [103]. The D domain is a highly flexible region within the VDR and links both its DNA-binding and its hormone-binding components [103]. It is not conserved among VDRs from other species or among other members of the steroid receptor family [85,86]. This region is, however, essential for intact hormone binding, a feature unlike that of most other steroid receptors. The hinge region may be more complex than that found in other receptors in that it contains an additional segment that corresponds to a specific exon in the chromosomal gene [104]. Interestingly, however, this region can be removed without compromising 1,25(OH)2D3 binding or transcriptional function [67,105]. Finally, the E/F domain contains the 1,25(OH)2D3 binding region of the VDR [103] and, equally important, serves as a highly complex protein–protein interface for a series of additional proteins that are integral to receptor activity [106–108]. This region has been termed activation function 2 or AF-2. The ability of the VDR to bind both DNA and additional transcription factors highlights its central role in nucleating the formation of transcriptionally active complexes at specific sites of gene regulation. Since the general role of all members of the receptor gene family is similar, it is not surprising that the E region is moderately conserved across the entire gene family and contains several subdomains that are even more highly conserved, as depicted in Fig. 2C. The F domain is absent in the VDR. 2. DNA BINDING DOMAIN
The DNA binding domain of the VDR is encoded in domain C. This domain consists of two similar modules, each consisting of a zinc-coordinated finger structure. The zinc atoms are individually coordinated in a tetrahedral fashion through four highly conserved cysteine residues that serve to stabilize the finger structure itself. Importantly, the finger modules are structurally unrelated to the zinc fingers found in TFIIIA wherein the zinc atom is coordinated through two cysteines and two histidines [109,110]. Surprisingly, although the two zinc modules of the VDR appear highly related to each other structurally, they are not equivalent topologically because the chirality of the residues that
J. WESLEY PIKE AND NIRUPAMA K. SHEVDE
coordinate the zinc atoms in each module is different [110]. The function of each of these modules in DNA binding is also different. Whereas the amino-terminal module directs specific DNA binding in the major grove of the DNA binding site, the carboxy-terminal module serves as a dimerization interface for interaction with partner proteins [111,112]. It seems likely that the two exons that encode these modules (see Section VI) evolved separately, although it is possible that each derived from a common ancestral gene through duplication and then diverged later. The threedimensional structure of the DNA-binding domain of several of the receptors has been determined through both nuclear magnetic resonance (NMR) spectroscopy and X-ray crystallography. The salient features of the finger structures are the presence of carboxy-terminal α-helices essential to interaction with DNA. Thus, our understanding of the structural organization of these modules as well as the mechanisms by which they function to interact with DNA and other proteins is well advanced [113–118]. 3. LIGAND BINDING DOMAIN
The E region of the VDR represents a multifunctional domain that exerts absolute regulatory control over the DNA-binding as well as transcription-modifying properties of the VDR. Historically, this region and comparable regions found in other receptors were explored for common structural features as well as for individual functional properties. As indicated in the overview in this section, several regions of homology were identified. Since cognate ligands differ substantially, it appeared likely that these regions of similarity represented motifs integral to the receptors’ role in transactivation. Closer scrutiny revealed the presence of multiple heptad repeats that were located in the ligand binding domain (LBD) of most members of the gene family including that of the VDR [119]. Despite this and many other important biochemical and molecular biologic discoveries, it is clear that the greatest insight into LBD structure emerged as a result of X-ray crystallographic studies initiated by Moras and colleagues with the retinoid receptors RARγ and RXRα [120,121]. As of this date, the three-dimensional structures of the LBDs of most of the known ligand-binding members of the nuclear receptor family including VDR have been determined [67–69,122–126]. With the exception of RXRα, the structures of the LBDs from most of the receptors have been solved in the presence of their cognate ligands. Thus, the bulk of our information of LBD structure in the absence of hormone derives from the apo structure of RXRα. RXRs play a unique role as unliganded heterodimer partners for many members of the nuclear receptor family [5,6].
175
CHAPTER 11 The Vitamin D Receptor
three-dimensional structures of numerous members of the steroid receptor superfamily [67–69,121,122–126,128]. As with many of the steroid receptors, the threedimensional structure of the LBD of the VDR has also been determined [67–69]. As this is the subject of Chapter 15, only a brief summary of the structure of the VDR will be described here. Rochel et al. [67] first crystallized the human VDR LBD in complex with 1,25(OH)2D3 as well as other ligands [68]. A recent report by Vanhooke and colleagues has confirmed and extended this important work using the rat VDR [69]. Essential to the successful crystallography effort of both groups was the removal of a portion of the D region that prevented crystal formation. The three-dimensional structure of the LBD of the rat VDR is depicted in Fig. 3. The human and rat VDR structures revealed the presence of thirteen α-helices and three β-sheets that together form at its core a highly hydrophobic ligand binding pocket. Surprisingly, the overall volume of this pocket is much greater than that seen for many of the other steroid receptors. 1,25(OH)2D3 binds in this pocket in an extended configuration with the A ring in the β chair conformation and the 1α-OH group equatorial. The ligand is anchored by six hydrogen-bonding interactions, the 1α-OH group with Ser233, Arg270, the 3β-OH group with Tyr143 and Ser274, and the 25-OH group with His301 and His393 (using the rat VDR
Accordingly, it seems unlikely that the LBD of this receptor will faithfully mimic the other apo-receptor forms. Nevertheless, the structure has revealed the presence of 12 α-helices (H1–H12) arranged in an anti-parallel α-helical sandwich. Perhaps most interesting is the location of H12, which projects away from the body of the RXRα LBD [120]. As will be discussed hereafter and in other chapters in more detail, this projection is in contrast to the location of H12 in virtually all the liganded or holo-receptor forms wherein H12 is packed tightly against the body of the LBD [67–69,121,122–126]. This presumptive change in nuclear receptor structure in the AF-2 region in response to hormone, together with many other changes evident in the overall conformation of the receptor, appear integral to its ability to attract DNA binding partners and to recruit protein complexes essential for transcriptional activation. It is important to note here, however, that the idea that ligand induced significant changes in receptor structure emerged many years earlier from both receptor stability and proteolytic digestion studies [75,76, and references therein]. The latter analyses have been refined through the use of metabolically labeled nuclear receptors derived from in vitro translated receptor cDNAs [127] and are described in detail in Chapters 18 and 83. The reader is referred for details to the many primary publications that report the
A
DNA Binding
B
AF2
N
C 1
21
88
240
Hinge Region
423 Ligand Binding AF2
Deletion
N
C 116
165 211 240
Hinge Region
DRIP205
423 Ligand Binding
6H
N-KNHPMLMNLL KDH-C 625
637
NR Box II
FIGURE 3 Structural organization of the rat VDR in complex with l,25(OH)2D3. (A) Schematic organization of the specific functional domains of the wild-type rat VDR (upper panel) and a ligand binding domain/6x-histidine fusion protein (middle panel) used to prepare rat VDR LBD crystals as described by Vanhooke et al. [69]. Shown in the lower panel is the DRIP205 NR box II peptide that was cocrystallized with the rat VDR LBD. (B) Ribbon representation of the rat VDR LBD in complex with l,25(OH)2D3 and the DRIP205 coactivator peptide.
176 sequence numbering). Several of the LBD helices, including H9 and H10, provide a binding surface for the VDR’s DNA binding partner RXR (see later discussion). Importantly, H12 is packed against the body of the LBD such that it seals the ligand binding site and along with H3 and H4 creates the AF-2 hydrophobic cleft that serves as a docking site for transcriptional comodulators of VDR action (to be discussed later as well as in detail in Chapters 13–18). Several of the residues that form H12 make direct contact with the ligand. Importantly, the recent elucidation of the wildtype VDR from the zebrafish has confirmed that the removal of the “insertion region” located in the hinge region is without consequence to the overall structure of the receptor [129]. 4. ACTIVATION FUNCTIONS
As indicated earlier, hormone binding leads to the creation of a functional AF-2. The core of this functional domain appears to be residues associated with H12 such as human residues 416 to 423 [130,131], although early mutagenesis studies indicated more extensive involvement of components located upstream in the E/F region [7]. Indeed, H3 and H4 are also involved. The hydrophobic cleft formed by these three helices provides an essential binding site for single proteins that mediate linkage to more complex sets of factors that include the p160 histone acetyltransferase (HAT) coactivator complexes [132] and the D-receptor interacting protein (DRIP) complex (Mediator-D complex) [133], both of which are discussed hereafter and in detail in Chapters 14 and 16. Proteins such as the p160 coactivators SRC-1, SRC-2, and SRC-3, and DRIP205 interact directly within the ligand-induced H12 cleft via a short helical LxxLL motif that is common to these comodulators [134]. Indeed, studies have confirmed that LxxLL-containing peptides that bind to either the VDR or RXR function as dominant negative regulators of VDR action [135]. Vanhooke et al. [69] has provided three-dimensional structural insight into the interaction between the rat VDR and an LxxLL motif through cocrystallization of the rat VDR LBD with an LxxLL peptide derived from the second coactivator motif of DRIP205. This group observed that the LxxLL peptide folds into a short α-helix that binds to the surface groove formed by H3, H4, H5, and H12. The LxxLL interaction buries some 500 square angstroms of VDR surface area and involves direct contact with residues of VDR H3, H4, and H12 including Pro412, Leu413, and Glu416 of H12. These interactions form a “charge clamp” that has been described previously for both PPARγ and its coactivator SRC-1 [125]. The side chains of many of the LxxLL motif are buried within the pocket and surrounded by hydrophobic residues.
J. WESLEY PIKE AND NIRUPAMA K. SHEVDE
Nonspecific amino acids project directly into the solvent and away from the binding cavity. These findings support a strong structural basis for the interaction of the VDR with at least one coactivator and may lead to new insights with regard to VDR–coactivator interaction and stability. 5. LIGAND BINDING AND DIFFERENTIAL ACTIVATION
Since 1,25(OH)2D3 acts as a small molecular switch to activate the VDR, it should not be surprising that the binding of ligand would induce sweeping conformational changes in the VDR LBD. Changes in conformation are also likely to be sensitive to the overall structure of the ligand, supporting the hypothesis that synthetic analogs of natural hormonal ligands may be capable of inducing unique changes in receptor, thereby altering their biological function. Indeed, synthetic ligands of the ER as well as other nuclear receptors have been identified that produce both unique receptor conformations as identified by their crystal structures [123,136,137] and that manifest surprising tissue selectivity [138,139]. Numerous vitamin D analogs have also been created. Interestingly, although these compounds manifest unusual actions both in vivo and in vitro (see Chapters 82–88), none have demonstrated their ability to induce a conformation within the VDR that is different from that induced by 1,25(OH)2D3 via crystallography [68,69]. Thus, the molecular basis for their potentially useful therapeutic actions remains unclear.
V. FUNCTIONAL ANALYSIS OF THE VITAMIN D RECEPTOR A. Osteocalcin Regulation as a Model for VDR-Regulated Transcription Osteocalcin (OC) is a relatively abundant noncollagenous protein found in bone. Genetic ablation of the OC gene in mice suggests that this osteoblast-specific protein may contribute to the density and structural integrity of bone [140]. Despite the uncertainty surrounding the function of OC, its expression at the transcriptional level is regulated by a number of cytokines, growth factors, and systemic hormones, one of which is 1,25(OH)2D3 [141,142]. This fact together with the cloning of the human OC gene and its promoter in 1986 [143] provided the raw materials for investigation of 1,25(OH)2D3 action on gene expression at the molecular level. Initial efforts focused on determining if 1,25(OH)2D3 could stimulate the activity of the human OC gene promoter linked to a reporter gene when introduced
177
CHAPTER 11 The Vitamin D Receptor
into bone cells via transfection [103,144,145]. The results suggested that 1,25(OH)2D3 strongly induced this gene promoter, indicating that the effects were direct [146]. Kerner et al. [144] localized the cis-acting element within the promoter subsequently to a region approximately 500 bp upstream of the transcriptional start site. This study and an additional one by Ozono et al. [147] precisely mapped the first vitamin D response element (VDRE) to a directly repeated hexanucleotide sequence separated by three base pairs. Subsequent studies by others using the rat OC gene promoter resulted in a similar identification [148–150]. These studies provided the first insights into DNA sequences within genes that mediated vitamin D-inducible activity. Many genes have been explored since that time for their sensitivity to 1,25(OH)2D3, including genes for mouse osteopontin (OPN) [151], mouse calbindinD28K [152], rat calbindin D9K [153], rat [154,155] and human [156] 25-hydroxyvitamin D3 24-hydroxylase (two VDREs), human p21 [157], and a host of others. A partial list of the sequences that have been shown to mediate vitamin D action together with their locations within the promoters is documented in Table III and elsewhere in this book. These sequences reveal that a “typical” VDRE comprises two hexanucleotide repeats separated by a 3-bp spacer. Although the sequence of the “spacer” is not generally conserved, recent studies have suggested that these base pairs may influence VDR binding [158]. Nevertheless, the general half-site consensus sequence is AGGTCA or GGTTCA, although considerable variability is apparent. Chapter 18 describes many of these VDREs in more detail. It also identifies additional putative VDREs in gene promoters that are an “atypical” configuration; however, the validity of these elements in the context of their natural promoters will require additional investigation.
TABLE III Gene Rat osteocalcin Human osteocalcin Mouse osteopontin Rat calbindin-D9K Mouse calbindin-D28K Rat 24 hydroxylase Human 24 hydroxylase Rat pit 1 Human p21
The efforts to define VDREs were part of a larger effort by numerous investigators to learn about steroid receptor response elements in general, with regard to specificity as well as to receptor selectivity [159]. These overall efforts led to our current understanding that nuclear receptor binding sites can be categorized into three groups: palindromic half-sites that interact with receptors for the sex steroids; directly repeated half-sites with variable spacing that interact with receptors for 1,25(OH)2D3, retinoic acid, thyroid hormone, and a number of other ligands; and single half-sites that mediate the actions of receptors such as NGF-IB that act as monomers [6]. The reader is referred elsewhere [1–6] for a complete review of the nature of these DNA binding sites. The binding of the VDR to endogenous gene promoters in living cells has been examined by several investigators including Sierra et al. [160], Zhang et al. [161]; and Pike and colleagues [54,55] using chromatin immunoprecipitation techniques. In these assays, cultured cells are treated with 1,25(OH)2D3 for specific periods of time, and the cells are then harvested and treated briefly with formaldehyde to fix the proteins that are bound to DNA. Cellular chromatin is then isolated, sonicated to produce small random fragments of 400–500 bp, and precipitated with antibodies to the protein of interest. The coprecipitated DNA is then isolated and its abundance evaluated by PCR using primers designed to amplify specific regions of gene promoters that are of interest. The concentration of promoter DNA correlates directly with immunoprecipitated protein. Importantly, 1,25(OH)2D3 stimulated the appearance of the VDR on several genes promoters including those for OC [160] as well as Cyp24 and OPN [55]. The results of a similar experiment for Cyp24 and OPN promoters are seen in Fig. 4. Although this
Location and Sequence of Positive Natural Vitamin D Response Elements Location −460/−446 −499/−485 −757/−743 −489/−475 −198/−183 −150/−136 (Proximal) −258/−244 (Distal) −169−/155 (Proximal) −291/−277 (Distal) −67/−52 −779/−765
Nucleotide sequence GGGTGA atg AGGACA GGGTGA acg GGGGCA GGTTCA cga GGTTCA GGGTGT cgg AAGCCC GGGGGA tgtg AGGAGA AGGTGA gtg AGGGCG GGTTCA gcg GGTGCG AGGTGA gcg AGGGCG AGTTCA ccg GGTGTG AGTTCA gcga AGTTCA AGGGAG att GGTTCA
178
J. WESLEY PIKE AND NIRUPAMA K. SHEVDE
A
Mouse CYP24 promoter
VDRE2
VDRE1
−236
ORF control
−51
+1
+11055
+11252
Mouse osteopontin promoter
VDRE
−854
ORF control +1
−658
B
+5181
CYP24
1,25 (OH)2D3
−
+
+5380
Osteopontin −
+
−
+
−
+
αVDR αRXR -Ab INPUT promoter
ORF
promoter
ORF
FIGURE 4 1,25(OH)2D3-induced association of the VDR and RXR to the promoters for Cyp24 and OPN. (A) Schematic diagram of the murine Cyp24 and OPN promoters. The locations of the VDREs within the two promoters are indicated with +1 representing the start site of transcription. Primer sites utilized to amplify either the VDRE in the promoter region or the open reading frame (ORF) are indicated. (B) VDR and RXR binding to the OPN or Cyp24 promoters as assessed by chromatin immunoprecipitation. For details, see [54].
approach does not permit the exact identification of the binding site for the VDR on a target gene, the results support localization of the VDR to a region within the promoter that contains a previously identified VDRE. This type of study is consistent with virtually all previous in vitro work that has led to the identification of VDREs in the OC, Cyp24, and OPN gene promoters.
B. VDR Binding to Specific DNA Involves RXR High-affinity binding of the VDR to DNA in vitro requires two regions, the DNA binding domain and the LBD. This conclusion is drawn from numerous mutagenesis experiments that revealed that amino acid alterations either in the zinc finger modules [104,162–168] or in the VDR LBD [103,107,130,169] were capable of blocking DNA binding. Interestingly, the molecular basis for altered DNA binding in the two regions is different. In the first case, changes in the VDR DNA binding domain prevent interaction directly with DNA
(see Chapter 13 for details of specific DNA binding). In the second, however, changes in LBD residues block the ability of the receptor to form the protein dimers that are essential for the high-affinity VDRE binding [107,169]. The discovery that the VDR must form a dimer to bind DNA was suggested by the structural nature of VDREs, which were composed of two repeated halfsites. Surprisingly, however, it was found that the VDR bound to DNA not as a homodimer but rather as a heterodimer as illustrated in the cartoon in Fig. 5. Liao et al. [170] and Sone et al. [162,171,172] both observed that while VDR derived from mammalian cell extracts was fully capable of binding to DNA in vitro, receptor produced either by in vitro transcription/translation or through recombinant means in nonmammalian cells failed to produce DNA binding-competent VDR. The inactivity of the VDR could be rescued, however, through the addition of general mammalian cell extract, suggesting the presence of a VDR DNA-binding facilitator. This factor(s) was discovered in a variety of tissue
179
CHAPTER 11 The Vitamin D Receptor
CBP/p300 PCAF
DRIP
SRC1
HAT 1,25(OH)2D3
RXR
Chromatin
Basal machinery
VDR
VDRE DR3
FIGURE 5
Contemporary model for VDR- and RXR-mediated gene regulation. The VDR and RXR bind to the promoter of a target gene as a heterodimer in response to 1,25(OH)2D3. Transcriptionally active complexes that include the SRC/CBP/p300 HAT coactivator complex or the DRIP205/Mediator complex or perhaps others are subsequently recruited to facilitate changes in chromatin architecture and to enhance RNA polymerase II entry. Both may be needed to modulate gene expression.
and cell sources [172] and termed nuclear accessory factor (NAF); its presence was subsequently confirmed by others [173,174]. NAFs were also discovered for other nuclear receptors including those for thyroid hormone and retinoic acid. In 1991 and 1992, Yu et al. [175], Leid et al. [176], Zhang et al. [177], and Kliewer et al. [178] all discovered that RXR, a previously cloned member of the nuclear receptor family, was the likely NAF protein partner. Three isotypes of RXR (α, β, and γ) are known and may be present in a given cell type. Considerable effort has gone into defining the nature and function of RXR-mediated DNA binding, and the reader is referred to the many reviews that deal with this complex issue [2,5]. Nevertheless, it is now clear that both VDR and RXR subunits are necessary not only to enable high-affinity DNA binding, but also to contribute to the activation process (see Chapter 13). Jin et al. [107,169] utilized an extensive series of internal deletions of the VDR to define two regions essential for interaction with RXR. These regions coincide with the two moderately conserved subregions of homology located in the E domain of the VDR and are associated directly with H9 and H10 of the crystal structure of the RXR LBD [120]. Although there has been some disagreement with regard to domain requirements for dimerization of other receptors [179,180], few have been cocrystallized together with
RXR, which would allow a more precise understanding of dimerization at the atomic level. The interaction surfaces appear to be large, however, and are widely believed to involve H9 and H10 of each protein. Unfortunately, the deletion analysis described by Jin et al. [107] was unable to distinguish between structural abnormalities induced by the deletion and the actual dimerization domain itself. Thus, elucidation of the three-dimensional structure of VDR-RXR LBD heterodimer will be required to define precisely the interaction between VDR and RXR. Despite these in vitro studies, simple experiments designed to unequivocally demonstrate an absolute requirement for RXR in VDR function in intact cells have not been carried out, primarily because of the ubiquitous expression of endogenous RXR in mammalian cell lines. This remains important since it has been proposed that the activation of the osteopontin gene by 1,25(OH)2D3 involves only VDR homodimers [181]. It is also important because the ability of 1,25(OH)2D3 to suppress several growth factor/cytokines genes such as those for IL-2 [182], GM-CSF [183], and interferon γ [184] as well as human PTH [185] may not require RXR. This suggests that the need for RXR in vitamin D gene regulation may be gene promoter–specific. Jin and Pike [106] utilized yeast, which do not express VDR and RXR′, to recreate the VDR transcriptional
180 response unit and to test for the requirement of RXRs on a typical VDRE. While VDR exhibited little capacity to activate a chimeric gene promoter containing a VDRE derived from the OC genes in the absence of RXR, the addition of RXR dramatically increased VDR-mediated activation. Despite this study, perhaps the most definitive assessment of RXR participation has come from studies by Barsony and colleagues [186] using immunofluorescence and Pike and colleagues [54,55] using the chromatin immunoprecipitation techniques. Prufer et al. [186] discovered that both VDR and RXR colocalize in response to 1,25(OH)2D3 within the nucleus in a VDR DNA-binding domain-dependent manner as discrete complexes that are believed to represent sites of transcription. As depicted in Fig. 4, the studies by Kim et al. [55] showed unequivocally that treatment of cells with 1,25(OH)2D3 leads to the rapid colocalization of both VDR and RXR on a similar fragment of the Cyp24 gene promoter, a gene that contains typical VDREs and is believed to require a VDR/RXR heterodimer for activation. Although the resolution of binding in these intact cells does not permit direct determination of heterodimer formation, this conclusion is a reasonable one given the earlier in vitro biochemical interaction studies. Interestingly, 1,25(OH)2D3 also promotes the colocalization of both VDR and RXR to the osteopontin promoter as seen in Fig. 4 [55]. These data indicate that VDR homodimer-mediated activation of the osteopontin gene is unlikely.
C. Polarity of DNA Binding The asymmetric nature of natural VDREs (direct repeats) coupled with the heteromeric nature of the receptor activation unit (VDR-RXR) indicates that the two receptor subunits must bind to the VDRE with a defined polarity. Studies by Jin and Pike [106] and Freedman and co-workers [164] addressed this question in detail. Through the use of chimeric receptors and chimeric response elements, it is now clear that RXR binds to the upstream 5′ half-element and the VDR binds to the downstream 3′ half-element of VDREs oriented on the DNA sense strand as illustrated in Fig. 5. This organization is consistent with the relative polarity noted for both RXR-TR [180] and RXR-RAR heterodimers [187,188] bound to their respective response elements.
D. Transactivation by the VDR and RXR Involves Coregulators Studies by McDonnell et al. [103] first demonstrated that the responsiveness of a transfected human
J. WESLEY PIKE AND NIRUPAMA K. SHEVDE
OC promoter to 1,25(OH)2D3 in VDR-negative cell lines such as CV-1 or COS7 required co-introduction of VDR. This biological assay enabled an extensive examination of the domains of the VDR that are required for transactivation and revealed the not surprising crucial nature of both the DNA binding and heterodimerization functions in transactivation. They also revealed the critical role of an intact AF-2 function in the context of normal DNA binding and heterodimer function for subsequent transactivation. Interestingly, this assay also established the inactive nature of many mutant VDRs that were identified in the human syndrome of hereditary l,25(OH)2D3-resistant rickets (HVDRR) considered in more detail in Chapter 72 [104,167,168]. Activation of the VDR involves a l,25(OH)2D3dependent conformational change within the LBD, which as indicated earlier creates a functional AF-2 domain. The protein components that interact within the AF-2 region of the nuclear receptors in general include the SWI/SNF complex that functions to remodel chromatin in an ATP-dependent manner [189,190], the p300/CBP coactivator complex [191,192] that contains nuclear receptor interacting proteins SRC1, SRC2, and/or SRC3 [132] and function to acetylate histones [193, 194], mediator complex that contains the nuclear receptor interacting protein DRIP205 and functions to facilitate entry of RNA polymerase II [133,195,196], and likely other [197] complexes. The potential interactions of the SRC-1/CBP and the DRIP complexes with the VDR/RXR heterodimer are illustrated in Fig. 5. The exact role of each of these enzyme-containing complexes during transcription is unclear as is the temporal order in which they are recruited to receptor-bound promoters during activation. The VDR has been shown to interact directly with specific components of each of these complexes in vitro [197–200], and the addition of many of these factors to cells via transfection assays strongly enhances l,25(OH)2D3-induced transcription. Direct evidence that the VDR recruits many of these factors to target promoters in living cells has only recently emerged, however, through the use of chromatin immunoprecipitation assays. Zhang et al. [161] and Kim et al. [55], both have demonstrated that l,25(OH)2D3 induces the recruitment of SRC-1 to the Cyp24 promoter. Kim et al. [55] also showed that SRC-2 and SRC-3 were similarly recruited, although the temporal pattern of recruitment differed among these coactivators. This group also observed that the HAT containing integrators p300 and CBP as well as mediator component DRIP205 were also recruited to the Cyp24 promoter in response to l,25(OH)2D3 [55]. Importantly, the consequence of this recruitment in the case of DRIP205 appears to be association of RNA polymerase II [55], whereas in the case of the p160 complex they result in the acetylation of histone 4 (Kim and Pike, unpublished).
181
CHAPTER 11 The Vitamin D Receptor
of the RXR H12 region by Thompson et al. [209]. The concept that RXR is not a “silent partner” but rather a transcriptionally active partner of the VDR has been reinforced also through studies by Pathrose et al. [135], who used RXR-selective LxxLL peptides to demonstrate that blockade of RXR recruitment of coactivators also leads to inhibition of transcription induced by l,25(OH)2D3. Finally, recent studies by Bettoun et al. [210] also support the contribution of RXR to VDR-induced transcription. Thus, it seems likely that both subunits of the VDR heterodimer complex are essential to the regulation of gene expression by l,25(OH)2D3.
These studies provide clear evidence not only that the VDR interacts with these coregulators on specific genes in living cells but that the recruitment of these complexes leads to functional consequences with regard to gene promoter status and transcriptional activity. A number of proteins also interact with the VDR in a ligand-independent manner in regions outside that of AF-2. These include the protein NCoA62, a coactivator that may play a role in mRNA transcript splicing [201]; WINAC [202]; Alien [203]; hairless [204]; SUG1, a 19S proteasomal component that may play a role in VDR degradation [205]; E6-AP, an E3 ligase that may play a role in VDR processing [Pike et al., unpublished]; and corepressors such as NCoR [206] and SMRT [207]. Little is known of how these proteins interact with the VDR or of the domain(s) within the VDR that mediates the interactions. The VDR also interacts with the core promoter transcription factor TFIIB [108,208], suggesting that the biochemical mechanisms by which the VDR contributes regulatory inputs into the basal transcriptional apparatus will be multifaceted and complex. The reader is referred to Chapters 13–17 for additional details regarding the interaction of the VDR with potential comodulators and the core transcription factor TFIIB.
F. VDR Dynamics Apart from the involvement of VDR and RXR as well as numerous coregulators in transcription, recent studies have suggested that the process is highly dynamic. Indeed, general studies in cultured cells using chromatin immunoprecipitation and other assays have indicated that the binding of nuclear receptors to target gene promoters is cyclic in nature [211–213]. Accordingly, hormone-induced binding of the receptor to gene promoters exhibited cycles with a periodicity ranging from 30 to 40 min. The recruitment of coregulator complexes is also cyclic, suggesting the possibility of a spatial and temporal order to the transcriptional process that is necessary for gene regulation [214]. Whether this cycling process is fundamental to the mechanism whereby genes are activated and transcription is maintained by transcription factors is currently unclear. The binding of the VDR and RXR to the Cyp24 and OPN promoters in response to l,25(OH)2D3 is also cyclic, as seen in Fig. 6 [55]. This binding is accompanied in parallel fashion by coactivators such as the p160 family of genes, CBP, and p300 as well as
E. Direct Involvement of RXR in Transactivation One question that has emerged from many of the functional studies of VDR is whether RXR contributes directly to the recruitment of the coactivators described earlier. This is particularly relevant since RXR also contains a functional AF-2 which interacts in vitro with most of the coactivator complexes described earlier. That RXR contributes directly to VDR-mediated transactivation has been demonstrated through mutagenesis
B
CYP24
Relative intensity, ×10−3
VDR
120
RXR pol II
80 40 0
OPN 120
VDR RXR pol II
80
40
Time (min)
165
150
135
120
105
90
75
60
45
30
15
0
165
150
135
120
105
90
75
60
45
30
15
0 0
Relative intensity, ×10−3
A
Time (min)
FIGURE 6 Localization of VDR, RXR, and RNA pol II to the (A) Cyp24 and (B) OPN promoters in response to 1,25(OH)2D3 as a function of time. The results are a densitometric depiction of the PCR products detected following amplification of either Cyp24 or OPN gene promoter DNA. For details, see Kim et al. [55].
182
J. WESLEY PIKE AND NIRUPAMA K. SHEVDE
Coactivator complexes
VDRE
TATA
Ligand-driven complex assembly
1,25(OH)2D3
RXR
VDR PCAF
HAT complex
RXR VDR
CBP/p300 p160s RXR VDR
Proteasome-driven disassembly
HISTONE
VDRE
HISTONE
Coactivator complexes
DRIP complex
DRIP 205 RXR VDR VDRE
TATA
FIGURE 7
Model for the 1,25(OH)2D3-driven assembly of transcriptionally active regulatory complexes on target genes and proteasome-directed disassembly of these complexes during signal processing. The order of recruitment and the influence of the gene itself on this process are currently unknown.
by the receptor interacting component of the DRIP complex, DRIP205 [55]. While current efforts are in place to determine the fundamental role of cycling in nuclear receptor activation, it is clear that this process may well enable precise real-time cellular monitoring of ambient hormone concentrations. A model for l,25(OH)2D3-driven assembly of VDR/RXR cofactor complexes on DNA and their proteasome mediated disassembly drawn from Kim et al. [55] is seen in Fig. 7.
G. General Conclusions An understanding of the functional activity of the VDR has emerged as a direct result of biochemical, molecular biological, and structural investigation. While insights gained from the three-dimensional structure of the VDR and other nuclear receptors have been exceptionally revealing, they have benefited enormously from numerous earlier biochemical and molecular biologic studies that have provided a wealth of information regarding the general domain structure and function of the nuclear receptor family. It is, in fact, a testimony to the effectiveness of these latter approaches that such great insights have emerged and that the crystallography has in many cases been largely confirmatory. The three approaches have, however, been highly complementary and have greatly extended our understanding of nuclear receptor structure and function.
VI. THE HUMAN VITAMIN D–RECEPTOR CHROMOSOMAL GENE A. Organization of the Gene The cloning of the VDR structural gene in 1987 initiated over a decade of highly productive research on the mechanism of action of vitamin D. Interestingly, unlike many of the other members of the nuclear receptor family, which exist as products of multiple genes, the VDR itself remains the apparent product of a single gene. Indeed, the recently published sequence analysis of both the mouse and human genomes appears to confirm the existence of only one VDR [91,92]. The human VDR gene is reported to lie on chromosome 12 [215]. The initial organization of the intron–exon structure of human VDR chromosomal gene corresponding to the sequence of the VDR reported by Baker et al. [85] was determined in 1988 [104]. Additional efforts have defined the complete structure of the gene [216]. Since the VDR gene is considered in some depth in Chapter 12, it will only be described briefly in this section. Restriction mapping of several lambda clones and a series of four recovered human cosmid clones coupled to nucleotide sequence analysis of relevant portions of these clones revealed a gene spanning over 75 kb of DNA. Eight exons comprise the coding sequence of the VDR protein. The first of these is exon 2, which contains the most proximal 3 bp of the 5′ noncoding
183
CHAPTER 11 The Vitamin D Receptor
sequence, the translation start site, and nucleotide sequence that encodes the first zinc finger module. Exon 3, which lies approximately 15 kb downstream, encodes the second zinc finger module. Exons 4, 5, and 6 encode the D region or hinge. Exons 6, 7, 8, and 9 encode a portion of the hinge and the carboxy-terminal E/F region as well as approximately 3200 nucleotides of 3′ noncoding sequence. The relationship between these exons, the full-length mRNA, and the VDR protein itself is illustrated in Fig. 8. It is clear that the human chromosomal gene for the VDR is not unlike other steroid receptor genes in size and exon organization. Insight into the 5′ end of this large gene and its promoter has also emerged over the past decade, revealing the promoter and regulatory regions of the human VDR to be highly complex (Fig. 8). This seems to be untrue of the rodent VDR genes where a single promoter is likely responsible for VDR expression [217]. Two short exons lie upstream of exon 2 and account for the known 5′ noncoding sequence reported by Baker et al. [85]. These exons, termed 1a and 1c, are 77 and 81 bp in length. A promoter (P1) lies immediately upstream of exon 1a and is clearly characterized by its GC-rich nature and the absence of a TATA box. Interestingly, an exon of 121 bp not found in the originally reported sequence of Baker et al. [85] and termed exon 1b is located 4.5 kb downstream of exon 1a. Variable use of exons 1b and 1c leads to the production
−281 −204
of alternatively spliced mRNAs whose nature and function remain unknown. More recently, additional upstream exons and their associated promoters have been ascribed to the human VDR gene [218]. These additions, as well as the transcripts and protein products that they produce [219], are described more completely in Chapter 12. Since the relative abundance of the bulk of these transcripts is exceedingly low, their contribution to overall VDR expression and to the various potential forms of the VDR proteins that are produced remains controversial. Clearly, additional work will be required to establish the bona fides of these novel forms of the VDR. Nevertheless, the presence of these transcripts suggests unique and possibly tissue-selective regulation of VDR expression. The identity of regulatory control elements in the human gene as well as those in the mouse gene is only now beginning to emerge.
B. Polymorphisms within the VDR Gene Genetic polymorphisms have been identified within the human VDR gene. The first represents a C to T transition in the translation initiation site located in exon 2 [94,95]. The presence of a C in this position results in the initiation of translation at a site three codons downstream and in the production of a gene
−83 −2
FIGURE 8 Structural organization of the human chromosomal VDR gene. The human VDR gene locus (DNA) is composed of as many as 14 exons (1a–1f, 2 through 9) spanning over 75 kb of DNA. A 10-kb scale is indicated at the right. The location of exons relative to the mRNA transcript of ~4800 nucleotides (mRNA) and the encoded VDR protein of 427 amino acids (hVDR) is illustrated. With regard to the hVDR mRNA, negative numbers indicate 5′ noncoding nucleotides and positive numbers indicate protein encoding nucleotides beginning with +1 indicated by Baker et al. [85]. Numbers below the hVDR protein indicate the amino acid residue boundaries of shaded homology domains. Regions of functionality are designated A/B, C, D, and E/F as in Figs. 2 or 3.
184 product three amino acids shorter. The distribution of this polymorphism in the human population and a potential relationship between the frequency of this polymorphism and bone mineral density were discussed in an earlier section of this chapter, and additional studies have been carried out which explore this issue in greater detail [97–101]. One finding has been the observation that the smaller form of the VDR appears to be more active transcriptionally [96–98]. Additional polymorphisms, although they are not located in regions of the gene that might directly affect the structure of the protein, have also been defined within the introns located between exons 7 and 8 as well as within the 3′ noncoding region of exon 9 [220, see Chapter 68 and references therein]. These polymorphisms appear also to correlate with bone mineral density in several human populations and are hypothesized to be predictive for osteoporosis. At presence, these conclusions are highly controversial, however, and are not widely supported by numerous recent studies that have been carried out over the past several years. This topic and associated references are considered in depth in Chapter 68.
VII. CONCLUDING COMMENTS The basic elements of the mechanism of action of vitamin D have been defined. The pace of exploration into the actions of vitamin D has accelerated enormously since the mid-1980s, largely as a result of the molecular cloning of the VDR in 1987 but also as a result of the availability of cloned vitamin D target genes. As described in this chapter, we have gained considerable insight into the structure of the VDR and its compartmentalization into definable functional domains. The now-available crystal structure of the VDR LBD has placed much of this biochemical and molecular biologic work into a clearer perspective. The availability of recombinant clones has allowed investigation of the interaction of the VDR with vitamin D–inducible gene promoters and definition of VDREs. Further studies have revealed that the VDR requires a protein partner for DNA binding in the form of RXR, a central regulator of several nuclear receptors. Although additional research will be necessary, a considerable understanding of the role of receptor and its ligand in the regulation of transcription has begun to emerge. These insights are currently being utilized to gain an understanding of the tissue-selective mechanisms of action of a new generation of vitamin D analogs under consideration as therapeutic agents for a broad range of indications that include skin diseases, immunologic disorders, and cancer. The cloning of the
J. WESLEY PIKE AND NIRUPAMA K. SHEVDE
VDR also enabled the recovery of its chromosomal gene. Subsequent investigation into the nature of the human syndrome of hereditary l,25(OH)2D3-resistant rickets (HVDRR) was made possible through characterization of the VDR gene itself [104,216] and has revealed the underlying cause to be mutations in the gene that produce dysfunctional receptor protein (described in Chapter 72). This discovery, together with the genetic ablation of the VDR gene in mice (see Chapter 20 for details) that clearly mimics physiologically the syndrome of HVDRR, confirm the central role of the VDR in the regulation of mineral metabolism. In view of our current understanding of VDR action on target gene promoters, future studies are likely to focus on the downstream roles of co-regulator complexes in mediating VDR action at the individual gene promoter level. Questions that need to be answered include those that address the individual roles of coregulator complexes in transcription, the order in which they are recruited to individual promoters, and whether features of the promoter itself play a deterministic role in the recruitment process and cellular response. Elucidation of these processes will be essential to the eventual understanding of the unique actions of vitamin D analogs that are potentially useful therapeutically. Also likely to emerge from these studies will be a better understanding of the regulation of the VDR at both the transcriptional and posttranslational levels and the contribution of this regulation to the mechanism of vitamin D action. Perhaps more important than these molecular details is the likelihood that we will achieve a better understanding of how vitamin D controls directly as well as indirectly the expression of broad networks of genes that are responsible in turn for cellular and tissue activities such as those involving proliferation and differentiation. Thus, although significant progress has been made in the past few years there is still much work that needs to be done.
References 1. Beato M 1989 Gene regulation by steroid hormones. Cell 56:335–344. 2. Evans RM 1988 The steroid and thyroid hormone receptor super-family. Science 240:889–895. 3. O’Malley BW 1990 The steroid receptor superfamily: More excitement predicted for the future. Mol Endocrinol 4:363–369. 4. Beato M, Herrliche P, Schutz G 1995 Steroid hormone receptors: Many actors in search of a plot. Cell 83:851–857. 5. Mangelsdorf DJ, Evans RM 1995 The RXR heterodimer and orphan receptors. Cell 83:841–850. 6. Mangelsdorf DJ, Thummel C, Beato M, Herrliche P, Schultz G, Umesono K, Blumberg B, Kastner P, Mark M, Chambon P,
185
CHAPTER 11 The Vitamin D Receptor
7. 8.
9.
10.
11. 12. 13.
14.
15.
16.
17. 18. 19. 20.
21. 22. 23. 24. 25.
Evans RM 1995 The nuclear receptor superfamily: The second decade. Cell 83:835–839. Norman AW 1968 The mode of action of vitamin D. Biol Rev 243:4055–4064. Lawson DEM, Wilson PW, Kodicek E, Morrison HR, Williams DH 1971 Identification of 1,25-dihydroxycholecalciferol, a new kidney hormone controlling calcium metabolism. Nature 230:228–230. Norman AW, Myrtle JF, Midgett RJ, Nowicki HG, Williams V, Popjak G 1971 1,25-Dihydroxycholecalciferol: Identification of the proposed active form of vitamin D in the intestine. Science 173:51–54. Holick MF, Schnoes HK, DeLuca HF, Suda T, Cousins RJ 1971 Isolation and identification of 1,25-dihydroxycholecalciferol. A metabolite of vitamin D active in intestine. Biochemistry 10:2799–2804. Fraser DR, Kodicek E 1970 Unique biosynthesis by kidney of a biologically active vitamin D metabolite. Nature 228:764–766. Haussler MR, Norman AW 1969 Chromosomal receptor for a vitamin D metabolite. Proc Natl Acad Sci USA 62:155–162. McDonnell DP, Mangelsdorf DJ, Pike JW, Haussler MR, O’Malley BW 1987 Molecular cloning of complementary DNA encoding the avian receptor for vitamin D. Science 235:1214–1217. Pike JW, Marion SL, Donaldson CA, Haussler MR 1983 Serum and monoclonal antibodies against the chick intestinal receptor for 1,25-dihydroxyvitamin D3: Generation by a preparation enriched in a 64,000 dalton protein. J Biol Chem 258:1289–1296. Dame MC, Pierce EA, Prahl JM, Hayes CE, DeLuca HF 1986 Monoclonal antibodies to the porcine intestinal receptor for 1,25-dihydroxyvitamin D3: Interaction with distinct receptor domains. Biochemistry 25:4523–4534. Brumbaugh PF, Haussler MR 1974 1,25-Dihydroxycholecalciferol receptors in intestine. Temperature-dependent transfer of the hormone to chromatin via a specific receptor. J Biol Chem 249:1258–1262. Lawson DEM, Wilson PW 1974 Intranuclear localization and receptor proteins for 1,25-dihydroxycholecalciferol in chick intestine. Biochem J 144:573–583. Brumbaugh PF, Haussler MR 1975 Specific binding of 1,25dihydroxycholecalciferol to nuclear components of chick intestine. J Biol Chem 250:1588–1594. Brumbaugh PF, Hughes MR, Haussler MR 1975 Cytoplasmic and nuclear components for 1,25-dihydroxyvitamin D3 in chick parathyroid glands. Proc Natl Acad Sci USA 72:4871–4875. Pike JW, Gooze LL, Haussler MR 1980 Biochemical evidence for 1,25-dihydroxyvitamin D receptor macromolecules in parathyroid, pancreas, pituitary, and placental tissues. Life Sci 26:407–414. Kream BE, Jose M, Yamada S, DeLuca HF 1977 A specific high affinity binding macromolecule for 1,25-dihydroxyvitamin D3 in fetal bone. Science 197:1086–1088. Chandler JS, Pike JW, Haussler MR 1979 1,25-Dihydroxyvitamin D3 receptors in rat kidney cytosol. Biochem Biophys Res Commun 90:1057–1063. Colston K, Feldman D 1980 Nuclear translocation of the 1,25-dihydroxycholecalciferol receptor in mouse kidney. J Biol Chem 255:7510–7513. Kream BE, Yamada Y, Schnoes HK, DeLuca HF 1977 Specific cytosol-binding protein for 1,25-dihydroxyvitamin D3 in rat intestine. J Biol Chem 254:9488–9491. Abe E, Miyaura C, Sakagami H, Takeda M, Konno K, Yamazaki T, Yashiki S, Suda T 1981 Differentiation of mouse
26.
27. 28. 29. 30. 31. 32. 33. 34.
35.
36. 37.
38. 39.
40. 41. 42. 43. 44.
myeloid leukemia cells induced by 1,25-dihydroxyvitamin D3. Proc Natl Acad Sci USA 78:4990–4994. Shavit ZB, Teitlebaum SL, Reitsma P, Hall A, Pegg LE, Trial J, Kahn AJ 1983 Induction of monocytic differentiation and bone resorption by 1,25-dihydroxyvitamin D3. Proc Natl Acad Sci USA 80:5907–5911. Christakos S, Norman AW 1981 Biochemical characterization of 1,25-dihydroxyvitamin D3 receptors in chick pancreas and kidney cytosols. Endocrinology 108:140–149. Haussler MR, Manolagas SC, Deftos L 1980 Evidence for a 1,25-dihydroxyvitamin D3 receptor-like macromolecule in rat pituitary. J Biol Chem 255:5007–5010. Dokoh S, Donaldson CA, Marion SL, Pike JW, Haussler MR 1983 The ovary: A target organ for 1,25-dihydroxyvitamin D3. Endocrinology 112:200–206. Merke J, Kreusser W, Bier B, Ritz E 1983 Demonstration and characterization of a testicular receptor for 1,25-dihydroxyvitamin D3 in the rat. Eur J Biochem 130:303–308. Colston K, Hirst M, Feldman D 1980 Organ distribution of the cytoplasmic 1,25-dihydroxycholecalciferol receptor in various mouse tissues. Endocrinology 107:1916–1922. Walters MR, Wicker DC, Riggle PC 1986 1,25-Dihydroxyvitamin D3 receptor identified in the rat heart. J Mol Cell Cardiol 18:67–72. Colston K, Colston MJ, Fieldsteel AH, Feldman D 1982 1,25-Dihydroxyvitamin D receptors in human epithelial cancer cell lines. Cancer Res 42:856–859. Colston K, Colston MJ, Feldman D 1981 1,25-Dihydroxyvitamin D and malignant melanoma: The presence of receptors and inhibition of cell growth in culture. Endocrinology 108:1083–1086. Eisman JA, Martin TJ, MacIntyre I, Frampton RJ, Moseley JM, Whitehead R 1980 1,25-Dihydroxyvitamin D3 receptors in a cultured human breast cancer cell line (MCF-7). Biochem Biophys Res Commun 93:9–15. Haussler MR 1986 Vitamin D receptor: Nature and function. Annu Rev Nutr 6:527–562. Feldman D, Chen T, Hirst M, Colston K, Karasek M, Cone C 1980 Demonstration of 1,25-dihydroxyvitamin D receptors in human skin biopsies. J Clin Endocrinol Metab 51: 1463–1465. Prowedini DM, Tsoukas CD, Deftos LJ, Manolagas SC 1983 1,25-Dihydroxyvitamin D3 receptors in human leucocytes. Science 221:1181–1183. Bhalla AK, Amento EP, Clemens TL, Holick MF, Krane SM 1983 Specific high affinity receptors for 1,25-dihydroxyvitamin D3 in human peripheral blood mononuclear cells: Presence in monocytes and induction in T lymphocytes following activation. J Clin Endocrinol Metab 57:1308–1310. Krageballe K 1992 Treatment of psoriasis with calcipotriol and other vitamin D analogues. J Am Acad Dermatol 27: 1001–1008. Simpson RU, DeLuca HF 1980 Characterization of a receptorlike protein for 1,25-dihydroxyvitamin D3 in rat skin. Proc Natl Acad Sci USA 77:5822–5827. Zile M, Barsness EC, Yamada S, Schnoes HK, DeLuca HF 1976 Localization of 1,25-dihydroxyvitamin D3 in intestinal nuclei in vivo. Arch Biochem Biophys 186:15–24. Jones PG, Haussler MR 1979 Scintillation autoradiography localization of 1,25-dihydroxyvitamin D3 in chick intestine. Endocrinology 104:313–321. Stumpf WE, Sar M, Reid FA, Tanaka Y, DeLuca HF 1979 Target cells for 1,25-dihydroxyvitamin D3 in intestinal tract, stomach, kidney, skin, pituitary, and parathyroids. Science 206:1188–1190.
186 45. Stumpf WE, Sar M, Clark SA, DeLuca HF 1982 Brain target sites for 1,25-dihydroxyvitamin D3. Science 215:1403–1405. 46. Clemens TL, Garrett KP, Zhou XY, Pike JW, Haussler MR, Dempster DW 1988 Immunocytochemical localization of the 1,25-dihydroxyvitamin D3 receptor in target cells. Endocrinology 122:1224–1230. 47. Berger U, Wilson P, McClelland RA, Colston K, Haussler MR, Pike JW, Coombes RC 1988 Immunocytochemical detection of 1,25-dihydroxyvitamin D3 receptors in normal human tissues. J Clin Endocrinol Metab 67:607–613. 48. Milde P, Merke J, Ritz E, Haussler MR, Rauterberg EW 1989 Immunocytochemical detection of 1,25-dihydroxyvitamin D3 receptors by monoclonal antibodies: A comparison of four immunoperoxidase methods. J Histochem Cytochem 37:1609–1617. 49. Berger U, Wilson P, McClelland RA, Colston K, Haussler MR, Pike JW, Coombes RC 1987 Immunocytochemical detection of 1,25-dihydroxyvitamin D3 receptors in breast cancer. Cancer Res 47:6793–6799. 50. Nevah-Many T, Marx R, Keshet E, Pike JW, Silver J 1990 Regulation of 1,25-dihydroxyvitamin D3 receptor gene expression by 1,25-dihydroxyvitamin D3 in the parathyroid in vivo. J Clin Invest 86:1968–1975. 51. Mocharla H, DeTogni P, Yu XP, White P, Jilka RJ, Manolagas SC 1994 In search for an association between vitamin D receptor polymorphism and VDR mRNA expression in peripheral blood mononuclear cells of normal volunteers. J Bone Miner Res 9(Suppl. 1):S416. 52. Walters MR, Hunziker W, Norman AW 1980 Unoccupied 1,25-dihydroxyvitamin D3 receptors. Nuclear/cytosol ratios depend on ionic strength. J Biol Chem 255:6799–6805. 53. Walters MR, Hunziker W, Norman AW 1981 1,25-Dihydroxyvitamin D3 receptors: Intermediates between triiodothyronine and steroid hormone receptors. Trends Biochem Sci 6:268–271. 54. Yamamoto H, Shevde NK, Warrier A, Plum LA, DeLuca HF 2003 2-Methylene-19-nor-(20S)-1,25-dihydroxyvitamin D3 potently stimulates gene-specific DNA binding of the vitamin D receptor in osteoblasts. J Biol Chem 278: 31756–31765. 55. Kim S, Yamamoto H, Shevde NK, Pike JW 2004 1,25Dihydroxyvitamin D3 stimulates dynamic vitamin D receptor/ retinoid x receptor DNA-binding and coactivator recruitment in intact osteoblasts. J Bone Min Res 18(Suppl 2): 530 (Abst). 56. Racz A, Barsony J 1999 Hormone dependent translocation of vitamin D receptors is linked to transactivation. J Biol Chem 274:19352–19360. 57. Prufer K, Barsony J 2002 Retinoid X receptor dominates the nuclear import and export of the unliganded vitamin D receptor. Mol Endocrinol 16:1738–1751. 58. Pike JW, Haussler MR 1979 Purification of chicken intestinal receptor for 1,25-dihydroxyvitamin D3. Proc Natl Acad Sci USA 76:5488–5494. 59. Haussler MR, Pike JW, Chandler JS, Manolagas SC, Deftos LJ 1981 Molecular actions of 1,25-dihydroxyvitamin D3: New cultured cell models. Ann NY Acad Sci 372:502–517. 60. Katzenellenbogen JA, O’Malley BW, Katzenellenbogen BS 1996 Tripartite steroid hormone receptor pharmacology: interaction with multiple effector sites as a basis for the celland promoter-specific action of these hormones. Mol Endocrinol 6:119–131. 61. Mellon WS, DeLuca HF 1979 An equilibrium and kinetic study of 1,25-dihydroxyvitamin D3 binding to chicken intestinal cytosol employing high specific activity l,25-dihydroxy[3H-26,27] vitamin D3. Arch Biochem Biophys 197:90–95.
J. WESLEY PIKE AND NIRUPAMA K. SHEVDE
62. Wecksler WR, Norman AW 1980 A kinetic and equilibrium binding study of 1,25-dihydroxyvitamin D3 with its cytosol receptor from chick intestinal mucosa. J Biol Chem 255:3571–3574. 63. Kream BE, Jose JL, DeLuca HF 1977 The chick intestinal cytosol binding protein for 1,25-dihydroxyvitamin D3: A study of analog binding. Arch Biochem Biophys 179:462–468. 64. Wecksler WR, Okamura WH, Norman AW 1978 Quantitative assessment of the structural requirements for the interaction of 1,25-dihydroxyvitamin D3 with its chick intestinal mucosa receptor system. J Steroid Biochem 9:929–937. 65. Pike JW, Haussler MR 1983 Association of 1,25-dihydroxyvitamin D3 with cultured 3T6 mouse fibroblast. J Biol Chem 258:8554–8560. 66. Hunziker W, Walters MR, Bishop JE, Norman AW 1983 Unoccupied and in vitro and in vivo occupied 1,25-dihydroxyvitamin D3 intestinal receptors. Multiple biochemical forms and evidence for transformations. J Biol Chem 258:8642–8648. 67. Rochel N, Wurtz JM, Mitschler A, Klaholz B, Moras D 2000 The crystal structure of the nuclear receptor for vitamin D bound to its natural ligand. Mol Cell 5:173–179. 68. Tocchini-Valentini G, Roche N, Wurtz JM, Mitschler A, Moras D 2001 Crystal structures of the vitamin D receptor complexed to superagonist 20-epi ligands. Proc Natl Acad Sci USA 98:5491–5496. 69. Vanhooke JL, Benning MM, Bauer CB, Pike JW, DeLuca HF 2004 Molecular structure of the rat vitamin D receptor ligand binding domain complexed with 2-carbon-substituted vitamin D3 hormone analogs and an LxxLL-containing coactivator peptide. Biochemistry, 43:4101–4110. 70. Pike JW, Sleator N, Haussler MR 1987 Chicken intestinal receptor for 1,25-dihydroxyvitamin D3. Immunologic characterization and homogeneous isolation of a 60,000-dalton protein. J Biol Chem 262:1305–1311. 71. Mangelsdorf DJ, Pike JW, Haussler MR 1987 Avian and mammalian receptors for 1,25-dihydroxyvitamin D3: In vitro translation to characterize size and hormone-dependent regulation. Proc Natl Acad Sci USA 84:354–358. 72. Pike JW, Kesterson RA, Scott RA, Kerner SA, McDonnell DP, O’Malley BW 1988 Vitamin D receptors: Molecular structure of the protein and its chromosomal gene. In: Normal AW, Schaefer K, Grigoleit H-G, Herrath DV (eds) Vitamin D: Molecular, Cellular and Clinical Endocrinology, de Gruyter, Berlin, pp. 215–224. 73. Elaroussi MA, Prahl JM, DeLuca HF 1992 The avian vitamin D receptors: Primary structures and their origins. Proc Natl Acad Sci USA 91:11596–11600. 74. Pike JW, 1984 Monoclonal antibodies to chick intestinal receptors for 1,25-dihydroxyvitamin D3: Interaction and effects of binding on receptor function. J Biol Chem 259:1167–1173. 75. Allegretto EA, Pike JW 1985 Trypsin cleavage of chick 1,25dihydroxyvitamin D3 receptors. Generation of discrete polypeptides which retain hormone but are unreactive to DNA and monoclonal antibody. J Biol Chem 260: 10139–10145. 76. Allegretto EA, Pike JW, Haussler MR 1987 Immunological detection of unique proteolytic fragments of the chick 1,25-dihydroxyvitamin D3 receptor. Distinct 20-kDa DNA binding and 45-kDa hormone-binding species. J Biol Chem 262:1312–1319. 77. Allegretto EA, Pike JW, Haussler MR 1987 C-Terminal proteo-lysis of the avian 1,25-dihydroxyvitamin D3 receptor. Biochem Biophys Res Commun 147:479–485.
187
CHAPTER 11 The Vitamin D Receptor
78. Heyman RA, Mangelsdorf DJ, Dyck JA, Stein RB, Eichele G, Evans RM, Thaller C 1992 9-cis-Retinoic acid is a high affinity ligand for the retinoid X receptor. Cell 68:397–406. 79. Forman BM, Tontonoz P, Chen C, Brun RP, Speigelman BM, Evans RM 1995 15-Deoxy-∆12,14-prostaglandin J2 is a ligand for the adipocyte determination factor PPARy. Cell 83:803–812. 80. Forman BM, Goode E, Chen J, Oro AE, Bradley DJ, Perlman T, Noonan DJ, Burka LT, McMorris T, Lamph WW, Evans RM, Weinberger C 1995 Identification of a nuclear receptor that is activated by farnesol metabolites. Cell 81:687–693. 81. Janowski BA, Willy PJ, Devi TR, Falck JR, Mangelsdorf DJ 1996 An oxysterol signalling pathway mediated by the nuclear receptor LXR alpha. Nature 383:728–731. 82. Chawla A, Repa JJ, Evans RM, Mangelsdorf DJ 2001 Nuclear receptors and lipid physiology: opening the X-files. Science 294:1866–1870. 83. Hollenberg SM, Weinberger C, Ong ES, Cerelli G, Ono AE, Lebo R, Thompson EB, Rosenfeld MG, Evans RM 1985 Primary structure and expression of a functional human glucocorticoid receptor cDNA. Nature 318:635–641. 84. Weinberger C, Thompson CC, Ong ES, Lebo R, Gruol DJ, Evans RM 1986 The c-erbA gene encodes a thyroid hormone receptor. Nature 324:641–646. 85. Baker AR, McDonnell DP, Hughes MR, Crisp TM, Mangelsdorf DJ, Haussler MR, Shine J, Pike JW, O’Malley BW 1988 Molecular cloning and expression of human vitamin D receptor complementary DNA: Structural homology with thyroid hormone receptor. Proc Natl Acad Sci USA 85:3294–3298. 86. Burmester JK, Maeda N, DeLuca HF 1988 Isolation and expression of rat 1,25-dihydroxyvitamin D receptor cDNA. Proc Natl Acad Sci USA 85:1005–1009. 87. Goto H, Chen KS, Prahl JM, DeLuca HF 1992 A single receptor identical to that for intestinal T47D cells mediates the action of 1,25-dihydroxyvitamin D3 in HL-60 cells. Biochim Biophys Acta 1132:103–108. 88. Kamei Y, Kawada T, Fukuwatari T, Ono T, Kato S, Sugimoto E 1995 Cloning and sequence of the gene encoding the mouse vitamin D receptor. Gene 152:281–282. 89. Elaroussi MA, Prahl JM, DeLuca HF 1992 The avian vitamin D receptors: Primary structures and their origins. Proc Natl Acad Sci USA 91:11596–11600. 90. Li YC, Bergwitz C, Juppner H, Demay MB 1996 Cloning and characterization of the vitamin D receptor from Xenopus laevis. J Bone Miner Res 11(Suppl. 1):S161. 91. Ciana P, Vegeto E, Beato M, Chambon P, Gustafsson JA, Parker M, Wahli W, Maggi A 2002 Looking at nuclear receptors from the heights of Erice. Workshop on nuclear receptor structure and function. EMBO Rep 3:125–129. 92. Enmark E, Gustafsson JA 2001 Comparing nuclear receptors in worms, flies and humans. Trends Pharmacol Sci 22: 611–615. 93. McDonnell DP, Pike JW, O’Malley BW 1988 Vitamin D receptor: A primitive steroid receptor related to the thyroid hormone receptor. J Steroid Biochem 30:41–47. 94. Kesterson RA, Pike JW 1991 Unpublished. 95. Saijo T, Ito M, Takeda E, Huq AHM, Naito E, Yokoto I, Sone T, Pike JW, Kuroda Y 1991 A unique mutation in vitamin D receptor gene in three Japanese patients with vitamin D– dependent rickets type II: Utility of single-strand conformation polymorphism analysis for heterozygous carrier detection. Am J Hum Genet 49:668–673. 96. Jurutka PW, Remus LS, Whitfield GK, Thompson PD, Hsieh JC, Zitzer H, Tavakkoli P, Galligan MA, Dang HT, Haussler CA,
97.
98.
99.
100.
101.
102. 103.
104.
105. 106. 107. 108. 109. 110. 111.
112. 113.
Haussler MR 2000 The polymorphic N terminus in human vitamin D receptor isoforms influences transcriptional activity by modulating interaction with transcription factor IIB. Mol Endocrinol 14:401–420. Arai H, Miyamoto K, Taketani Y, Yamamoto H, Iemori Y, Morita K, Tonai T, Nishisho T, Mori S, Takeda E 1997 A vitamin D receptor gene polymorphism in the translation initiation codon: effect on protein activity and relation to bone mineral density in Japanese women. J Bone Miner Res 12:915–921. Gross C, Krishnan AV, Malloy PJ, Eccleshall TR, Zhao XY, Feldman D 1998 The vitamin D receptor gene start codon polymorphism: a functional analysis of FokI variants. J Bone Miner Res 13:1691–1699. Chiu KC, Chuang LM, Yoon C 2001 The vitamin D receptor polymorphism in the translation initiation codon is a risk factor for insulin resistance in glucose tolerant Caucasians. BMC Med Genet 2:2. Kubota M, Yoshida S, Ikeda M, Okada Y, Arai H, Miyamoto K, Takeda E 2001 Association between two types of vitamin D receptor gene polymorphism and bone status in premenopausal Japanese women. Calcif Tissue Int 68:16–22. Whitfield GK, Remus LS, Jurutka PW, Zitzer H, Oza AK, Dang HT, Haussler CA, Galligan MA, Thatcher ML, Encinas Dominguez C, Haussler MR 2001 Functionally relevant polymorphisms in the human nuclear vitamin D receptor gene. Mol Cell Endocrinol 177:145–159. Green S, Walter P, Kumar V, Krust A, Bornert JM, Argos P, Chambon P 1986 Human oestrogen receptor cDNA: Sequence expression and homology to v-erbA. Nature 320:134–139. McDonnell DP, Scott RA, Kerner RA, O’Malley BW, Pike JW 1989 Functional domains of the human vitamin D receptor regulate osteocalcin gene expression. Mol Endocrinol 3:635–644. Hughes MR, Malloy PJ, Kieback DG, Kesterson RA, Pike JW, Feldman D, O’Malley BW 1988 Point mutations in the human vitamin D receptor gene associated with hypocalcemic rickets. Science 242:1702–1705. Rochel N, Tocchini-Valentini G, Egea PF, Juntunen K, Garnier JM, Vihko P, Moras D 2001 Eur J Biochem 268:971–979. Jin CH, Pike JW 1996 Human vitamin D receptor dependent transactivation in Saccharomyces cerevisiae requires retinoid X receptor. Mol Endocrinol 10:196–205. Jin CH, Kerner SA, Hong MH, Pike JW 1996 Transcriptional activation and dimerization functions of the human vitamin D receptor. Mol Endocrinol 10:945–957. MacDonald PN, Sherman DR, Dowd DR, Jefcoat SC Jr, DeLisle RK 1995 The vitamin D receptor interacts with general transcription factor IIB. J Biol Chem 270:4748–4752. Berg JM 1989 DNA binding specificity and steroid receptors. Cell 57:1065–1068. Berg JM 1988 Proposed structure for the zinc-binding domains from transcription factor IIIA and related proteins. Proc Natl Acad Sci USA 85:99–102. Mader S, Kumar V, deVereneuil H, Chambon P 1989 Three amino acids of the oestrogen receptor are essential to its ability to distinguish an oestrogen from a glucocorticoid responsive receptor. Nature 338:271–274. Umesono K, Evans RM 1989 Determinants of target gene specificity for steroid/thyroid hormone receptors. Cell 57:1139–1146. Schwabe JWR, Neuhaus D, Rhodes D 1990 Solution structure of the DNA binding domain of the oestrogen receptor. Nature 348:458–461.
188 114. Hard T, Kellenbach E, Boelens R, Maler BA, Dahlman K, Freedman LP, Carlstedt-Duke J, Yamamoto KR, Gustafsson JA, Kaptein R 1990 Solution structure of the glucocorticoid receptor DNA binding domain. Science 249:157–160. 115. Luisi BF, Xu W, Otwinowski Z, Freedman LP, Yamamoto KR, Sigler PB 1991 Crystallographic analysis of the interaction of the glucocorticoid receptor with DNA. Nature 352: 497–505. 116. Schwabe JWR, Chapman L, Finch JT, Rhodes D 1993 The crystal structure of the oestrogen receptor DNA binding domain bound to DNA: How receptors discriminate between their response elements. Cell 75:567–578. 117. Lee MS, Kliewer SA, Provencal J, Wright PE, Evans RM 1993 Structure of the retinoid X receptor α DNA binding domain: A helix required for homodimeric DNA binding. Science 260:1117–1121. 118. Rastinejad F, Perlmann T, Evans RM, Sigler PB 1995 Structural determinants of nuclear receptor assembly on DNA direct repeats. Nature 375:203–211. 119. Forman BM, Yang CR, Au M, Casanova J, Ghysdael J, Samuels HH 1989 A domain containing leucine-zipper-like motifs mediate novel in vivo interactions between the thyroid hormone and retinoic acid receptors. Mol Endocrinol 3:1610–1626. 120. Bourguet W, Ruff D, Chambon P, Gronemeyer H, Moras D 1995 Crystal structure of the ligand binding domain of the human nuclear receptor RXRa. Nature 375:377–382. 121. Renaud JP, Natacha R, Ruff M, Vivat V, Chambon P, Gronemyer H, Moras D 1995 Crystal structure of the RARy ligand binding domain bound to all-trans-retinoic acid. Nature 378:681–689. 122. Wagner RL, Apriletti JW, McGrath ME, West BL, Baxter JD, Fletterick RJ 1995 A structural role for hormone in the thyroid hormone receptor. Nature 378:690–697. 123. Shiau AK, Barstad D, Loria PM, Cheng L, Kushner PJ, Agard DA, Greene GL 1998 The structural basis of estrogen receptor/coactivator recognition and the antagonism of this interaction by tamoxifen. Cell 95:927–937. 124. Williams S, Sigler P 1998 Atomic structure of progesterone complexed with its receptor. Nature 393:392–396. 125. Nolte RT, Wisely GB, Westin S, Cobb JE, Lambert MH, Kurokawa R, Rosenfeld MG, Willson TM, Glass CK Milburn MV 1998 Ligand binding and co-activator assembly of the peroxisome proliferator-activated receptor-gamma. Nature 395:137–143. 126. Downes M, Verdecia MA, Roecker AJ, Hughes R, Hogenesch JB, Kast-Woelbern HR, Bowman ME, Ferrer JL, Anisfeld AM, Edwards PA, Rosenfeld JM, Alvarez JG, Noel JP, Nicolaou KC, Evans RM 2003 A chemical, genetic, and structural analysis of the nuclear bile acid receptor FXR. Mol Cell 11:1079–1092. 127. Peleg S, Sastry M, Collins ED, Bishop JE, Norman AW 1995 Distinct conformational changes induced by 20-epi analogues of 1,25-dihydroxyvitamin D3 are associated with enhanced activation of the vitamin D receptor. J Biol Chem 270: 10551–10558. 128. Quack M, Carlberg C 1999 Selective recognition of vitamin D receptor conformations mediates promoter selectivity of vitamin D analogs. Mol Pharmacol 55:1077–1087. 129. Moras D 2003 Twelfth Workshop on Vitamin D Maastricht, The Netherlands, June 2003, p. 3 (Abstract). 130. Whitfield GK, Hsieh JC, Nakajima S, MacDonald PN, Thompson PD, Jurutka PW, Haussler CA, Haussler MR 1995 A highly conserved region in the hormone-binding domain
J. WESLEY PIKE AND NIRUPAMA K. SHEVDE
131.
132.
133. 134.
135.
136.
137.
138.
139.
140.
141. 142. 143. 144.
145.
of the human vitamin D receptor contains residues vital for heterodimerization with retinoid X receptor and for transcriptional activation. Mol Endocrinol 9:1166–1179. Danielian PS, White R, Lees JA, Parker MG 1992 Identification of a conserved region required for hormone dependent transcriptional activation by steroid hormone receptors. EMBO J 11:1025–1033. Xu J, O’Malley BW 2002 Molecular mechanisms and cellular biology of the steroid receptor coactivator (SRC) family in steroid receptor function. Rev Endocr Metab Disord 3:185–192. Rachez C, Freedman LP 2001 Mediator complexes and transcription. Curr Opin Cell Biol 13:274–280. McInerney EM, Rose DW, Flynn SE, Westin S, Mullen TM, Krones A, Inostroza J, Torchia J, Nolte RT, Assa-Munt N, Milburn MV, Glass CK, Rosenfeld MG 1998 Determinants of coactivator LXXLL motif specificity in nuclear receptor transcriptional activation. Genes Dev 12:3357–3368. Pathrose P, Barmina O, Chang CY, McDonnell DP, Shevde NK, Pike JW 2002 Inhibition of 1,25-dihydroxyvitamin D3dependent transcription by synthetic LXXLL peptide antagonists that target the activation domains of the vitamin D and retinoid X receptors. J Bone Miner Res 17:2196–2205. [Erratum in 2003 J Bone Miner Res 18:390.] Brzozowski AM, Pike AC, Dauter Z, Hubbard RE, Bonn T, Engstrom O, Ohman L, Greene GL, Gustafsson JA, Carlquist M 1997 Molecular basis of agonism and antagonism in the oestrogen receptor. Nature 389:753–758. Shiau AK, Barstad D, Radek JT, Meyers MJ, Nettles KW, Katzenellenbogen BS, Katzenellenbogen JA, Agard DA, Greene G 2002 Structural characterization of a subtypeselective ligand reveals a novel mode of estrogen receptor antagonism. Nature Struc Biol 9:359–364. Tzukerman MT, Esty A, Santiso-Mere D, Danielian P, Parker MG, Stein RB, Pike JW, McDonnell DP 1994 Human estrogen receptor transcriptional capacity is determined by both cellular and promoter context and mediated by two functionally distinct intramolecular regions. Mol Endocrinol 8:21–30. McDonnell DP, Clemm DL, Hermann T, Goldman ME, Pike JW 1995 Analysis of estrogen receptor function in vitro reveals three distinct classes of anti-estrogens. Mol Endocrinol 9:659–669. Ducy P, Desbois C, Boyce B, Pinero G, Story B, Dunstan C, Smith E, Bonadio J, Goldstein S, Gundberg C, Bradley A, Karsenty G 1996 Increased bone formation in osteocalcindeficient mice. Nature 382:448–452. Price PA, Baukol SA 1980 1,25-Dihydroxyvitamin D3 increases synthesis of the vitamin D-dependent bone proteins by osteosarcoma cells. J Biol Chem 225:11660–11663. Pan LC, Price PA 1986 1,25-Dihydroxyvitamin D3 stimulates transcription of bone gla protein. J Bone Miner Res 1(Suppl. 1):A20. Celeste AJ, Rosen V, Buecker JL, Kriz R, Wang EA, Wozney JM 1986 Isolation of the human gene for bone gla protein utilizing mouse and rat cDNA clones. EMBO J 5:1885–1890. Kerner SA, Scott RA, Pike JW 1989 Sequence elements in the human osteocalcin gene confer basal activation and inducible response to hormonal vitamin D3. Proc Natl Acad Sci USA 86:4455–4459. Morrison NA, Shine J, Fragonas J-C, Verkest V, McMenemy ML, Eisman JA 1989 1,25-Dihydroxyvitamin D3 responsive element and glucocorticoid repression in the osteocalcin gene. Science 246:1158–1161.
CHAPTER 11 The Vitamin D Receptor
146. Yoon K, Rutledge SJC, Buenaga RF, Rodan GA 1988 Characterization of the rat osteocalcin gene: Stimulation of promoter activity by 1,25-dihydroxyvitamin D3. Biochemistry 27:8521–8526. 147. Ozono K, Liao J, Scott RA, Kerner SA, Pike JW 1990 The vitamin D responsive element in the human osteocalcin gene: Association with a nuclear proto-oncogene enhancer. J Biol Chem 265:21881–21888. 148. Demay MB, Gerardi JM, DeLuca HF, Kronenberg HM 1990 DNA sequences in the rat osteocalcin gene that bind the 1,25-dihydroxyvitamin D receptor and confer responsiveness to 1,25-dihydroxyvitamin D3. Proc Natl Acad Sci USA 87:369–373. 149. Lian JB, Stewart C, Puchacz E, Mackowiak S, Shalhoub V, Colart D, Sambetti G, Stein G 1989 Structure of the rat osteocalcin gene and regulation of vitamin D–dependent expression. Proc Natl Acad Sci USA 86:1143–1147. 150. Terpening CM, Haussler MR, Jurutka PW, Galligan MA, Komm BS, Haussler MR 1991 The vitamin D responsive element in the rat bone gla protein gene is an imperfect direct repeat that cooperates with other cis elements in 1,25dihydroxyvitamin D3 mediated transcriptional activation. Mol Endocrinol 5:373–385. 151. Noda M, Vogel RL, Craig AM, Prahl J, DeLuca HF, Denhardt D 1990 Identification of a DNA sequence responsible for binding of the 1,25-dihydroxyvitamin D3 receptor and 1,25-dihydroxyvitamin D3 enhancement of mouse secreted phosphoprotein 1 (Supp-1 or osteopontin) gene expression. Proc Natl Acad Sci USA 87:9995–9999. 152. Gill RK, Christakos S 1993 Identification of sequence elements in the mouse calbindin-D28K gene that confer 1,25-dihydroxyvitamin D- and butyrate inducible responses. Proc Natl Acad Sci USA 90:2984–2988. 153. Darwish HM, DeLuca HF 1992 Identification of a 1,25-dihydroxyvitamin D3 response element in the 5’ flanking region of the rat calbindin D-9K gene. Proc Natl Acad Sci USA 89:603–607. 154. Ohyama Y, Ozono K, Uchida M, Shinki T, Kato S, Suda T, Yamamoto O, Noshiro M, Kato Y 1994 Identification of a vitamin D responsive element in the 5′-flanking region of the rat 25-hydroxyvitamin D3 24-hydroxylase gene. J Biol Chem 269:10545–10550. 155. Zierold C, Darwish HM, DeLuca HF 1995 Two vitamin D response elements function in the rat 1,25-dihydroxyvitamin D3–24-hydroxylase promoter. J Biol Chem 269: 1675–1678. 156. Chen KS, DeLuca HF 1995 Cloning of the human 1,25dihydroxyvitamin D3–24-hydroxylase gene promote and identification of two vitamin D responsive elements. Biochim Biophys Acta 1263:1–9. 157. Liu M, Lee M-H, Cohen M, Freedman LP 1996 Transcriptional activation of the p21 gene by vitamin D3 leads to the differentiation of the myelomonocytic cell line U937. Genes Dev 10:142–153. 158. van den Bemd GJ, Jhamai M, Staal A, van Wijnen AJ, Lian JB, Stein GS, Pols HA, van Leeuwen JP 2002 A central dinucleotide within vitamin D response elements modulates DNA binding and transactivation by the vitamin D receptor in cellular response to natural and synthetic ligands. J Biol Chem 277:14539–14546. 159. Umesono K, Murikami KK, Thompson CC, Evans RM 1991 Direct repeats as selective response elements for the thyroid hormone, retinoic acid, and vitamin D3 receptors. Cell 65:1225–1266.
189 160. Sierra J, Villagra A, Paredes R, Cruzat F, Gutierrez S, Javed A, Arriagada G, Olate J, Imschenetzky M, Van Wijnen AJ, Lian JB, Stein GS, Stein JL, Montecino M 2003 Regulation of the bone-specific osteocalcin gene by p300 requires Runx2/Cbfa1 and the vitamin D3 receptor but not p300 intrinsic histone acetyltransferase activity. Mol Cell Biol 23:3339–3351. 161. Zhang C, Dowd DR, Staal A, Gu C, Lian JB, van Wijnen AJ, Stein GS, MacDonald PN 2003 Nuclear coactivator-62 kDa/ Ski-interacting protein is a nuclear matrix-associated coactivator that may couple vitamin D receptor-mediated transcription and RNA splicing. J Biol Chem 278:35325–35336. 162. Sone T, Kerner SA, Pike JW 1991 Vitamin D receptor interaction with specific DNA. Association as a 1,25-dihydroxyvitamin D3-modulated heterodimer. J Biol Chem 266: 23296–23305. 163. Freedman LP, Towers T 1991 DNA binding properties of the vitamin D receptor zinc fingers region. Mol Endocrinol 5:1815–1826. 164. Towers T, Luisi BL, Asianov A, Freedman LP 1993 DNA target selectivity by the vitamin D receptor: Mechanism for dimer binding to an asymmetric repeat element. Proc Natl Acad Sci USA 90:6310–6314. 165. Lemon B, Freedman LP 1996 Selective effects of ligands on vitamin D3 receptor and retinoid X receptor mediated gene activation in vivo. Mol Cell Biol 16:1006–1016. 166. Nishikawa J, Kitaura M, Matsumoto M, Imagawa M, Nishihara T 1994 Difference and similarity of DNA sequence recognized by VDR homodimer and VDR/RXR heterodimer. Nucleic Acids Res 22:2902–2907. 167. Sone T, Scott R, Hughes M, Malloy P, Feldman D, O’Malley BW, Pike JW 1989 Mutant vitamin D receptors which confer hereditary resistance to 1,25-dihydroxyvitamin D3 in humans are transcriptionally inactive in vitro. J Biol Chem 264: 20230–20234. 168. Hughes MR, Malloy PJ, O’Malley BW, Pike JW, Feldman D 1991 Genetic defects of the 1,25-dihydroxyvitamin D receptor. J Receptor Res 11:699–716. 169. Nakajima S, Hsieh JC, MacDonald PN, Galligan MA, Haussler CA, Whitfield GK, Haussler MR 1994 The C-terminal region of the vitamin D receptor is essential to form a complex with a receptor auxiliary factor required for high affinity binding to the vitamin D responsive element. Mol Endocrinol 8:159–172. 170. Liao J, Ozono K, Sone T, McDonnell DP, Pike JW 1990 Vitamin D receptor interaction with specific DNA requires a nuclear protein and 1,25-dihydroxyvitamin D3. Proc Natl Acad Sci USA 87:9751–9755. 171. Sone T, McDonnell DP, O’Malley BW, Pike JW 1990 Expression of the human vitamin D receptor in Saccharomyces cerevisiae: Purification properties and generation of polyclonal antibodies. J Biol Chem 265:21997–22003. 172. Sone T, Ozono K, Pike JW 1991 A 55-kilodalton accessory factor facilitates vitamin D receptor DNA binding. Mol Endocrinol 5:1578–1586. 173. MacDonald PN, Haussler CA, Terpening CM, Galligan MA, Reeder MC, Whitfield GK, Haussler MR 1991 Baculovirusmediated expression of the human vitamin D receptor: Functional characterization, vitamin D response element interactions and evidence for a receptor auxiliary factor. J Biol Chem 266:18808–18813. 174. Ross TK, Moss VE, Prahl JM, DeLuca HF 1992 A nuclear protein essential for binding of rat 1,25-dihydroxyvitamin D3 receptor to its response elements. Proc Natl Acad Sci USA 89:256–260.
190 175. Yu V, Delsert C, Andersen B, Holloway JM, Devary OV, Naar AM, Kirn SY, Boutin JM, Glass CK, Rosenfeld MG 1991 RXR/3: A coregulator that enhances binding of retinoic acid, thyroid hormone, and vitamin D receptors to their cognate response elements. Cell 67:1251–1266. 176. Leid M, Kastner P, Lyons R, Nakshatro H, Saunders M, Zacharewski T, Chen J-Y, Staub A, Gamier J-M, Mader S, Chambon P 1992 Purification, cloning and RXR identity of the HeLa cell factor with which RAR or TR heterodimerizes to bind target sequences efficiently. Cell 68:377–395. 177. Zhang XK, Hoffman B, Tran PB, Graupner G, Pfahl M 1992 Retinoid X receptor is an auxiliary protein for thyroid hormone and retinoic acid receptors. Nature 355:441–446. 178. Kliewer SA, Umesono K, Mangelsdorf DJ, Evans RM 1992 Retinoid X receptor interacts with nuclear receptors in retinoic acid, thyroid, and vitamin D signaling. Nature 355:446–449. 179. Perlmann T, Umesono K, Rangarajan PN, Forman BM, Evans RM 1996 Two distinct dimerization interfaces differentially modulate target gene specificity of nuclear hormone receptors. Mol Endocrinol 10:958–966. 180. Perlmann T, Rangarajan PN, Umesono K, Evans RM 1993 Determinants for selective RAR and TR recognition of direct repeat HREs. Genes Dev 7:1411–1422. 181. Freedman LP, Arce V, Perez Fernandez R 1994 DNA sequences that act as high affinity targets for the vitamin D3 receptor in the absence of the retinoid X receptor. Mol Endocrinol 8:265–273. 182. Alroy I, Towers TL, Freedman LP 1995 Transcriptional repression of the interleukin-2 gene by vitamin D3: direct inhibition of NFATp/AP-1 complex formation by a nuclear hormone receptor. Mol Cell Biol 15:5789–5799. 183. Towers TL, Staeva TP, Freedman LP 1999 A two-hit mechanism for vitamin D3-mediated transcriptional repression of the granulocyte-macrophage colony-stimulating factor gene: vitamin D receptor competes for DNA binding with NFAT1 and stabilizes c-Jun. Mol Cell Biol 19:4191–4109. 184. Staeva-Vieira TP, Freedman LP 2002 1,25-Dihydroxyvitamin D3 inhibits IFN-gamma and IL-4 levels during in vitro polarization of primary murine CD4+ T cells. J Immunol 168: 1181–1189. 185. Mackey SL, Heymont JL, Kronenberg HM, Demay MB 1996 Vitamin D receptor binding to the negative human parathyroid hormone vitamin D response element does not require the retinoid x receptor. Mol Endocrinol 10:298–305. 186. Prufer K, Racz A, Lin GC, Barsony J 2000 Dimerization with retinoid X receptors promotes nuclear localization and subnuclear targeting of vitamin D receptors. J Biol Chem 275:41114–41123. 187. Kurokawa B, Yu VC, Naar A, Kyakumoto S, Han Z, Silverman S, Rosenfeld MG, Glass CK 1993 Differential orientation of the DNA-binding domain and carboxy terminal dimerization interface regulate binding site selection by nuclear receptor heterodimers. Genes Dev 7:1423–1435. 188. Sucov HM, Murakami KK, Evans RM 1990 Characterization of an autoregulated response element in the mouse retinoic acid receptor type beta gene. Proc Natl Acad Sci USA 87: 5392–5396. 189. Fyodorov DV, Kadonaga JT 2001 The many faces of chromatin remodeling: SWItching beyond transcription. Cell 106:523–525. 190. Narlikar GJ, Fan HY, Kingston RE 2002 Cooperation between complexes that regulate chromatin structure and transcription. Cell 108:475–487. 191. Chakravarti D, LaMorte VJ, Nelson MC, Nakajima T, Schulman IG, Juguilon H, Montminy M, Evans RM 1996
J. WESLEY PIKE AND NIRUPAMA K. SHEVDE
192. 193. 194.
195.
196.
197.
198.
199. 200.
201.
202.
203.
204.
205.
206.
Role of CBP/P300 in nuclear receptor signalling. Nature 383:99–103. Ogryzko VV, Schiltz RL, Russanova V, Howard BH, Nakatani Y 1996 The transcriptional coactivators p300 and CBP are histone acetyltransferases. Cell 87:953–959. Imhof A, Yang XJ, Ogryzko VV, Nakatani Y, Wolffe AP, Ge H 1997 Acetylation of general transcription factors by histone acetyltransferases. Curr Biol 7:689–692. Spencer TE, Jenster G, Burcin MM, Allis CD, Zhou J, Mizzen CA, McKenna NJ, Onate SA, Tsai SY, Tsai MJ, O’Malley BW 1997 Steroid receptor coactivator-1 is a histone acetyltransferase. Nature 389:194–198. Jiang YW, Veschambre P, Erdjument-Bromage H, Tempst P, Conaway JW, Conaway RC, Kornberg RD 1998 Mammalian mediator of transcriptional regulation and its possible role as an end-point of signal transduction pathways. Proc Natl Acad Sci USA 95:8538–8543. Rachez C, Lemon BD, Suldan Z, Bromleigh V, Gamble M, Naar AM, Erdjument-Bromage H, Tempst P, Freedman LP 1999 Ligand-dependent transcription activation by nuclear receptors requires the DRIP complex. Nature 398:824–828. Masuyama H, Brownfield CM, St-Arnaud R, MacDonald PN 1997 Evidence for ligand-dependent intramolecular folding of the AF-2 domain in vitamin D receptor-activated transcription and coactivator interaction. Mol Endocrinol 11:1507–1517. Kraichely DM, Collins JJ 3rd, DeLisle RK, MacDonald PN 1999 The autonomous transactivation domain in helix H3 of the vitamin D receptor is required for transactivation and coactivator interaction. J Biol Chem 274:14352–14358. Leo C, Li H, Chen JD 2000 Differential mechanisms of nuclear receptor regulation by receptor-associated coactivator 3. J Biol Chem 275:5976–5982. Issa LL, Leong GM, Barry JB, Sutherland RL, Eisman JA 2001 Glucocorticoid receptor-interacting protein-1 and receptor-associated coactivator-3 differentially interact with the vitamin D receptor (VDR) and regulate VDR-retinoid X receptor transcriptional cross-talk. Endocrinology 142:1606–1615. Baudino TA, Kraichely DM, Jefcoat SC Jr, Winchester SK, Partridge NC, MacDonald PN 1998 Isolation and characterization of a novel coactivator protein, NCoA-62, involved in vitamin D–mediated transcription. J Biol Chem 273: 16434–16441. Kitagawa H, Fujiki R, Yoshimura K, Mezaki Y, Uematsu Y, Matsui D, Ogawa S, Unno K, Okubo M, Tokita A, Nakagawa T, Ito T, Ishimi Y, Nagasawa H, Matsumoto T, Yanagisawa J, Kato S, 2003 The chromatin-remodeling complex WINAC targets a nuclear receptor to promoters and is impaired in Williams syndrome. Cell 113:905–917. Polly P, Herdick M, Moehren U, Baniahmad A, Heinzel T, Carlberg C 2000 VDR-Alien: a novel, DNA-selective vitamin D3 receptor-corepressor partnership. FASEB J 14:1455–1463. Hsieh JC, Sisk JM, Jurutka PW, Haussler CA, Slater SA, Haussler MR, Thompson CC 2003 Physical and functional interaction between the vitamin D receptor and hairless corepressor, two proteins required for hair cycling. J Biol Chem 278:38665–38674. Masuyama H, MacDonald PN 1998 Proteasome-mediated degradation of the vitamin D receptor (VDR) and a putative role for SUG1 interaction with the AF-2 domain of VDR. J Cell Biochem 71:429–440. Perissi V, Staszewski LM, McInerney EM, Kurokawa R, Krones A, Rose DW, Lambert MH, Milburn MV, Glass CK,
CHAPTER 11 The Vitamin D Receptor
207.
208.
209.
210.
211. 212.
Rosenfeld MG 1999 Molecular determinants of nuclear receptor-corepressor interaction. Genes Dev 13:3198–3208. Tagami T, Lutz WH, Kumar R, Jameson JL 1998 The interaction of the vitamin D receptor with nuclear receptor corepressors and coactivators. Biochem Biophys Res Commun 253:358–363. Blanco JCG, Wang I-M, Tsai SY, Tsai M-J, O’Malley BW, Jurutka PW, Haussler MR, Ozato K 1995 Transcription factor TFIIB and the vitamin D receptor cooperatively active ligand-dependent transcription. Proc Natl Acad Sci USA 92:1535–1539. Thompson PD, Remus LS, Hsieh JC, Jurutka PW, Whitfield GK, Galligan MA, Encinas Dominguez C, Haussler CA, Haussler MR 2001 Distinct retinoid X receptor activation function-2 residues mediate transactivation in homodimeric and vitamin D receptor heterodimeric contexts. J Mol Endocrinol 27:211–227. Bettoun DJ, Burris TP, Houck KA, Buck DW 2nd, Stayrook KR, Khalifa B, Lu J, Chin WW, Nagpal S 2003 Retinoid X receptor is a nonsilent major contributor to vitamin D receptor-mediated transcriptional activation. Mol Endocrinol 17:2320–2328. Shang Y, Hu X, DiRenzo J, Lazar MA, Brown M 2000 Cofactor dynamics and sufficiency in estrogen receptorregulated transcription. Cell 103:843–852. Reid G, Hubner MR, Metivier R, Brand H, Denger S, Manu D, Beaudouin J, Ellenberg J, Gannon F 2003 Cyclic, proteasomemediated turnover of unliganded and liganded ERalpha on responsive promoters is an integral feature of estrogen signaling. Mol Cell 11:695–707.
191 213. Shang Y, Myers M, Brown M 2003 Formation of the androgen receptor transcription complex. Mol Cell 9:601–610. 214. Burakov D, Crofts LA, Chang CP, Freedman LP 2002 Reciprocal recruitment of DRIP/mediator and p160 coactivator complexes in vivo by estrogen receptor. J Biol Chem 277:14359–14362. 215. Szpirer S, Szpirer C, Riviere M, Levan G, Maryen P, Cassiman JJ, Wiese R, DeLuca HF 1991 The SP1 transcription factor gene (SP1) and the 1,25-dihydroxyvitamin D receptor gene (VDR) are colocalized on human chromosomal arm 12q and rat chromosome 7. Genomics 11: 168–173. 216. Miyamoto K, Kesterson RA, Yamamoto H, Taketani Y, Nishiwaki E, Tatsumi S, Inoue Y, Morita K, Takeda E, Pike JW 1997 Structural organization of the human vitamin D receptor chromosomal gene and its promoter. Mol Endocrinol 11:1165–1179. 217. Jehan F, DeLuca HF 2000 The mouse vitamin D receptor is mainly expressed through an Sp1-driven promoter in vivo. Arch Biochem Biophys 377:273–283. 218. Crofts LA, Hancock MS, Morrison NA, Eisman JA 1998 Multiple promoters direct the tissue-specific expression of novel N-terminal variant human vitamin D receptor gene transcripts. Proc Natl Acad Sci USA 95:10529–10534. 219. Sunn KL, Cock TA, Crofts LA, Eisman JA, Gardiner EM 2001 Novel N-terminal variant of human VDR. Mol Endocrinol 15:1599–1609. 220. Morrison NA, Qi JC, Tokita A, Kelly PJ, Crofts L, Nguyen TV, Sambrook PN, Eisman JA 1994 Prediction of bone density from vitamin D receptor alleles. Nature 367:284–287.
CHAPTER 12
Vitamin D Receptor Promoter and Regulation of Receptor Expression LUIS M. ESTEBAN, JOHN A. EISMAN, AND EDITH M. GARDINER Bone and Mineral Research Program, Garvan Institute of Medical Research, Sydney, Australia
I. Introduction II. Tissue Distribution III. The Vitamin D Receptor Gene Locus
IV. Regulation of Vitamin D Receptor Expression and Abundance V. Concluding Remarks References
I. INTRODUCTION
Current understanding of this regulation is often confounded by inconsistencies between findings in different experimental systems, which may relate to the particular species, tissue, or cell type represented by the model under study. This chapter therefore describes current understanding of the complexities of VDR gene regulation, with the goal of providing a foundation for interpretation of recent findings and a basis for further exploration.
The cellular and physiological actions of 1,25dihydroxyvitamin D [1,25(OH)2D3] are mediated primarily through the nuclear vitamin D receptor (VDR) as described in Chapter 11, although there is in vitro evidence for rapid actions of the hormone that appear to be mediated by a distinct nonnuclear receptor, as described in Chapter 23. Alteration in circulating levels of 1,25(OH)2D3 is a primary mechanism by which receptorexpressing cells are subjected to physiological hormone responses, as described in Chapters 46, 49, 65, and 66. Another important regulatory mechanism, however, is variation in cellular content of functional receptor, by alteration in VDR abundance or by post-translational modification. Such regulation allows multiple target cells to have graded responses to 1,25(OH)2D3 in a given hormonal environment, depending on the VDR levels. This chapter addresses the transcriptional mechanisms by which VDR abundance is regulated. Regulation of receptor function by post-translational mechanisms is discussed in Chapter 13. The nuclear VDR gene structure is complex, with several functioning promoters and alternatively spliced 5′ coding and noncoding exons in some species. There are consensus transcription factor binding elements and response motifs in the promoter regions. In some cases, there is experimental evidence that these motifs are functional and respond to known regulators of VDR transcription. However, there are also consensus element matches that suggest previously unsuspected regulatory inputs and await experimental investigation. Moreover, in some cases the promoter regions lack consensus response elements relevant to signal transduction pathways that are known to regulate VDR expression, suggesting indirect mechanisms of regulation. Study of the transcriptional control of the VDR gene therefore provides both expected and unexpected insights into the physiological regulation and function of this receptor. VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX
II. TISSUE DISTRIBUTION Specific receptors for 1,25(OH)2D3 were initially identified in the intestine of the chick by sucrose density gradient separation and assays of binding to [3H]-1,25(OH)2D3 [1,2]. Such receptors were subsequently identified in rat and human intestine, rat and chick fetal calvaria, chick kidney and pancreas, and rat skin [3–7], with sedimentation coefficients of 3.2–3.7 S. This experimental approach failed to detect receptors in brain, lung, thymus, adipose tissue, skeletal muscle, uterus, and myocardium, leading to the initial impression that there was a selective organ distribution. In addition to these studies, which detected VDR primarily in calciotropic tissues, early reports also indicated the presence of a receptor with physicochemical properties similar to those of chicken intestine receptor in the breast cancer cell line MCF-7 [8] and in a number (23 of 33) of cultured human cancer cell lines [9], including seven of 10 malignant melanoma lines and eight of nine colonic carcinoma lines. In the 1980s monoclonal antibodies generated against VDR protein isolated from chicken and pig intestine, or against synthetic peptides from the human VDR [10–12], were used to detect VDR in protein extracts from a variety of tissues and cell lines, and to clone the chicken VDR cDNA [13,14]. Copyright © 2005, Elsevier, Inc. All rights reserved.
194 A monoclonal antibody to the chick intestinal receptor detected nuclear VDR in histological sections of normal human epithelial tissues, including liver, kidney, thyroid, adrenal, gastrointestinal tract, breast, and skin [15]. A quantitative immunoradiometric assay (IRMA) indicated the presence of VDR protein in rat tissues [16], with the highest level in proximal small intestine and colon, moderate levels in ileum and kidney, and lower amounts in tissues of the vitamin D endocrine system, e.g., bone, thyroid/parathyroid, and skin. The IRMA proved more sensitive than the earlier sedimentation and radioligand approach mentioned above, revealing even lower amounts of VDR protein in the nonclassical vitamin D target tissues lung, heart, stomach, spleen, and liver, but not in cerebrum, cerebellum, or skeletal muscle [16]. More recent studies have detected VDR in pituitary, activated B and T lymphocytes, ovary, and testes, as well as in brain [17]. The availability of even more sensitive techniques such as reverse-transcription polymerase chain reaction (RT-PCR) has made it clear that VDR is expressed even in liver, which previously was considered to have very low or even no VDR [18]. VDR mRNA and protein were very low in hepatocytes, but Kupffer, stellate, and endothelial cells expressed functional VDR that mediated a positive CYP24 promoter response to 1,25(OH)2D3, suggesting that selective hepatic cell populations are also targets for the vitamin D endocrine system [18]. Based on these findings, it appears that there are more than 50 vitamin D target tissues, with the actions of 1,25(OH)2D3 extending far beyond its role in calcium homeostasis. The levels of VDR vary greatly between tissues and even between different cell types within a given tissue. Regulation of VDR expression therefore plays a fundamental role in determining physiological vitamin D responsiveness.
III. THE VITAMIN D RECEPTOR GENE LOCUS The many biological actions of vitamin D and its derivatives and the varied distribution of its receptor indicate complex regulation, which includes precise control of VDR and its abundance in a tissue- and species-specific manner. Part one of this section is focussed on the structure of the 5′ end of the VDR gene in chicken and mammals, including human, mouse, and rat, with detailed descriptions of VDR N-terminal variants that may provide functional diversity to the vitamin D endocrine system. The second part of this section presents evidence for the complex regulation of the VDR through its promoter structure and the presence of numerous transcription factor response elements.
LUIS M. ESTEBAN, JOHN A. EISMAN, AND EDITH M. GARDINER
The third part briefly reviews evidence related to the effects of common DNA sequence polymorphisms on VDR expression.
A. VDR Gene Structure 1. STRUCTURE OF THE VDR LOCUS
VDR is widely but not ubiquitously distributed. Its transcription has been analyzed primarily in mammalian and avian species, although it is present in all vertebrates and to date cDNAs have been isolated from more than 18 species including fish, amphibians, and reptiles. The human and rat mRNAs are 4.4 kb [19,20], in contrast with the approximately 2.7 kb avian [21,22] and the two Xenopus laevis (2.2 and 1.8 kb) transcripts [23]. However the sizes of the encoded receptor proteins are somewhat comparable across species, at 422 amino acid residues in Xenopus, 424 residues in rat, 427 and 477 residues in human, and 437 and 451 residues in chicken. There are eight coding exons (exons 2 to 9 according to human VDR notation [24] and a variable number of untranslated 5′ exons. The functional start codon of the mouse locus is in exon 3 [25], which is equivalent to human and rat exon 2 and chicken exon 4 (Fig. 1). Notably, in the chicken locus, translation is initiated not in exon 4, but instead in exon 3 at one of two initiation codons, giving rise to two isoforms that differ by 14 amino acids [22]. There are two known untranslated exons upstream of the first coding exons in chicken, mouse, and rat, which are highly conserved at the nucleotide level among mammalian but not chicken loci. The human VDR locus is more complex, with six transcribed exons upstream of the initiation codon in exon 2. Alternative promoter usage and alternative splicing of exons 1A, 1B, and 1C [24] and 1D, 1E, and 1F [27] give rise to 14 different human VDR transcripts [27]. Exons 1A and 1C are equivalent to mouse exons 1 and 2 [25]. Exon 1D is of particular interest, as it contains a translation initiation codon that is in frame with the conventional human VDR open reading frame [27]. An alternatively spliced transcript has been detected that includes exons 1D and 1C and encodes an N-terminally extended receptor isoform, VDRB1, which is present in human tissues and cell lines [28] (discussed further hereafter). A second exon 1Dcontaining transcript lacking exon 1C and encoding a shorter isoform VDRB2 has been detected, but to date the existence of this putative receptor has not been confirmed. Exon 1D–containing transcripts have not been reported in other species, although the DNA sequence is evolutionarily conserved in mouse and rat [29], with
195
CHAPTER 12 VDR Promoter and Abundance
ATG
ATG hVDRB1
1A 1F
1E
1D
1B
1C
2
Human hVDR
ATG 1A
1D-like
1C
2
Rat
ATG 1
1D-like
2
3
Mouse
ATG 1
2
3
4
Chicken
FIGURE 1
Structure of the 5′ flanking region of mammalian and chicken VDR gene loci. Arrows indicate exons with the initiation codon; two start codons are present in chicken exon 3. Alternative splicing produces a number of transcripts in the human VDR; two of them, which give rise to the original VDR and the VDRB1 isoform, are shown. Dotted line on the chicken locus indicates sequence not yet determined.
an overall homology of 75% with the human sequence. Chicken exon 2 has low homology with human exon 1D (50%) and its position relative to exon 1 is not conserved compared to the relative positions of human exons 1A and 1D (Fig. 1). Primer extension analysis of mouse VDR transcripts did give rise to products with a different sequence 5′ to exon 2, perhaps originating from an upstream promoter or a differentially spliced transcript [30]. To date, there is no evidence from homology searches of the mouse and rat genome databases that the remaining human exons (1B, 1E, and 1F) are present in rodents. However, they do appear to be present in other primates, as genomic sequence from chimpanzee (Pan troglodytes) includes all the exons with a 98–100% identity to human sequences [29]. The evolutionary conservation of these exons and surrounding sequences suggests an important biological role for the N-terminal extension (whether for modulation of VDR expression or function). Future studies on other mammals may reveal whether these isoforms are a general phenomenon or limited to species evolutionarily closer to humans. 2. N-TERMINAL NUCLEAR RECEPTOR VARIANTS
The VDR is a member of the nuclear receptor (NR) superfamily of ligand-inducible transcription factors. It has a modular structure typical of the NRs, including a DNA binding domain (DBD) and a C-terminal ligand binding domain (LBD), linked by a flexible hinge
(for details see Chapter 11). There is a less wellconserved A/B domain at the N terminus, varying in length from 23 amino acids in the VDR to more than 600 amino acids in the mineralocorticoid receptor [31]. There is little or no A/B region homology among the NRs, in part due to the existence of N-terminal variants or some receptors. These isoforms can arise from separate genes or from a single gene by alternative splicing, usage of different promoters, and/or distinct translational start sites [32–38]. Such variants contribute to functional diversity. As an example, the two human progesterone receptor isoforms differ in promoter specificity [39] and ligand responses [40]. There are also functional differences, as the progesterone A receptors can dominantly inhibit the B receptors [41,42] and other steroid receptor superfamily members [43]. Such distinct functional characteristics of N-terminal variant nuclear receptors may contribute to modulation of different physiological responses. 3. VITAMIN D RECEPTOR VARIANTS
The broad range of physiological vitamin D responses indicates functional diversity, and although there is only one VDR locus in each species [on human chromosome (Chr) 12, mouse Chr 15, and rat Chr 7], there is some evidence for VDR isoforms. In flounder (Paralichthys olivaceus), two VDR subtypes originate from different transcripts, with nucleotide and predicted amino acid sequence variations distributed
196
LUIS M. ESTEBAN, JOHN A. EISMAN, AND EDITH M. GARDINER
sometimes termed VDRA [28,29]. These two human VDR isoforms have been detected together at similar levels in human kidney tissue and in kidney, intestine, and bone cell lines [28], with the identity of the larger protein confirmed using a VDRB1-specific antiserum. In addition, the VDRB1-encoding transcript was also detected in parathyroid adenoma tissue and in skin and mammary gland cell lines [27]. VDRB1 exhibits differential transactivation activity [29] that appears to be due, at least in part, to a strong ligand-independent transcriptional activation domain in the A/B region [50]. The differential accumulation of VDRB1 into nuclear speckles may also contribute to transactivation differences [28] although a mechanism for such an effect has not been established. The distinct transactivation properties of VDRB1 would be consistent with a role in diversification of the vitamin D physiological response. The relative abundance of VDR and VDRB1 in each tissue or cell type may therefore be regulated. The complexity of the 5′ flanking region of the human VDR locus suggests mechanisms that may provide such regulation (discussed in Section III,A,1 and later).
throughout the receptor [44]. Flounder VDRa shows a more limited tissue distribution than VDRb and, to date, no functional differences between the two have been reported. VDRs from chicken and quail exhibit two isoforms (A and B) that differ by 14 amino acids and are translated from a single mRNA species. The chicken VDR gene contains 3 in-frame ATG sites for translation initiation, with the third in a position in exon 4 equivalent to the mammalian start codon site (Fig. 1). The first ATG is in a suboptimal context (TCCATGT), whereas the second is a better fit for the optimal consensus sequence (AGCATGG) [22,45]. Mutational analysis indicates that the two translated avian VDR isoforms arise from the first and second ATG sites by a “leaky scanning” mechanism [22], but no differential functions have been identified for these two chicken receptors. Similarly, the physiological significance of a dominant-negative C-terminally truncated rat VDR resulting from aberrant splicing leading to retention of intron 8 is not known [46]. Variants of human VDR have been identified. A common start codon polymorphism that produces a VDR with a three-amino-acid N-terminal truncation [47] has been associated with elevated transactivation activity [48,49]. A human VDR isoform with a longer N-terminal extension is generated from an exon 1D-containing transcript, as described earlier (See Section III,A,1) [27,28]. Other than its 50-amino-acid N-terminal extension, the VDRB1 protein is identical to the originally described VDR [19] (Fig. 2), which is
A /B
DBD
B. VDR Promoter To analyze the regulation of the VDR gene, its 5′ flanking region has been cloned and its promoter activity characterized in mouse [25,30], human [24,27,51], and chicken [52]. In these species the flanking sequences
Hinge
LBD
AF-2
427 aa
50
hVDR
427 aa
ATG
hVDR B1
ATG
TGA
Fok I 1A 1D
1B
1C
2
3
4 5 6
7 8
9
FIGURE 2 Structure of the human VDR isoforms. Receptor protein domains are shown on top, exon structure and alternative 5′ splicing patterns below. Arrows indicate the start codons for VDR (exon 2) and VDRB1 isoform (exon 1D). VDRB1 differs from VDR only in its N-terminal extension of 50 amino acids. *FokI polymorphism gives rise to a protein of 424 amino acids. White boxes indicate the exons that encode the N-terminal extension of VDRB1 isoform. Striped boxes represent untranslated exons.
197
CHAPTER 12 VDR Promoter and Abundance
5′ to exon 1 (mouse and chicken) or exon 1A (humans) (see Fig. 1) are highly conserved [52]. Their properties are the focus of the following sections. 1. NONHUMAN VDR PROMOTERS
The sequence of the mouse VDR 5′ flanking DNA contains characteristics of a GC-rich promoter including four tandem Sp1 sites and two upstream CCAAT boxes in opposite orientation (Fig. 3). Such TATA-less promoters are not uncommon among the steroid/ thyroid hormone receptor superfamily [53]. A 1.5-kb segment of the mouse 5′-flanking region activated transcription of a reporter gene in rat osteosarcoma ROS 17/2.8 and mouse fibroblastic NIH3T3 cell lines [25]. Deletion analysis indicated that the sequence from 54 to 177 bp upstream of the transcription start site provides most of the promoter activity and a region 177 to 481 bp upstream appears to have a repression function [30]. The first, second, and fourth Sp1 sites between 56 and 112 bp upstream of the transcription start site are necessary for mVDR promoter activity, whereas the Sp1–3 site appears to be inhibitory [30]. The Sp1 site proximal to exon 1 (Sp1-1) (position 1435, Fig. 3) overlaps a perfect consensus site for two closely related zinc finger transcription factors, Krox-20 and Krox-24 [54]. Both factors were originally noted for their importance in the nervous system [55,56]; however, Krox20−/− mice exhibit abnormal bone formation with a defect in endochondral ossification [57]. This phenotype is consistent with Krox20 expression in a subpopulation of growth plate hypertrophic chondrocytes and in differentiating osteoblasts, two cell types that are known to be responsive to vitamin D and to express the VDR (see Chapters 32 and 33). Thus, the effect of the Krox20 deletion on bone formation may relate to a defect in VDR expression in cells normally responsive to 1,25(OH)2D3. To date, however, VDR levels in these mice and the function of the Krox binding site in the mouse VDR promoter have not been tested. The mouse VDR promoter contains an AP-1 site (position 1123, Fig. 3) that differs by one nucleotide from one that overlaps the VDRE in the human osteocalcin promoter [58]. AP-1 is a target for the stressactivated protein kinases p38 and JNK, two major MAPK (mitogen-activated protein kinase) pathways. The p38 and JNK pathways and the c-Jun component of AP-1 were required for VDR stimulation by stress activation, which further sensitized human breast cancer MCF-7 cells to vitamin D3-induced growth inhibition [59]. There is also an imperfect AP-1 site in the mouse VDR promoter (position 1281, Fig. 3), similar to a binding site for the osteoblast-specific transcription factor NMP-2 (nuclear matrix protein 2) in the rat osteocalcin promoter [60]. Although these similarities
are suggestive, its role in regulating mouse VDR transcription is not known. No consensus vitamin D-, cAMP-, glucocorticoid-, or estrogen-responsive element was identified in the mouse VDR promoter 1.5-kb region by computer analysis, although there are a putative cyclic AMP-response element binding protein/Ets transcription factor (CREB/ETF) binding site at position 944 and an AP-2 site at position 1043 (Fig. 3) [25]. Treatment with forskolin, a cyclic AMP agonist known to increase endogenous VDR protein and mRNA content (discussed further in Section IV,B,4,1) [61], resulted in a three to fivefold increase in promoter-reporter activity, although the CREB/ETF site itself was not tested for function [25]. Further analysis of the mouse promoter demonstrated the presence of two E-boxes (at positions 1034 and 1129, Fig. 3), both with the consensus sequence ‘CACCTG’ target for the ZEB transcription factor, and both mediating up-regulated VDR expression in vitro [62]. In this region there are also five potential binding sites for the transcription factor c-MYB that appear to function additively with ZEB [62]. In this context it is notable that both ZEB and VDR knockout mice exhibit growth retardation and skeletal abnormalities [63,64]. A WT1, Wilms’ tumor gene, responsive element (WRE) overlapping the Sp1–4 site (position 1387, Fig. 3) is functional and able to mediate WT1-enhanced VDR promoter activity in transient transactivation assays, consistent with elevation of endogenous VDR upon overexpression of WT1 in colon carcinoma and kidney cell lines [65,66]. The rat VDR promoter has not yet been experimentally examined, but high DNA sequence homology (over 85%) with the mouse promoter suggests it is likely to function similarly (Fig. 3). Three of the functional Sp1 sites of the mouse promoter are perfectly conserved in the rat 5′ region whereas the fourth, Sp1–2, is divergent at two nucleotides. The rat Krox20/24 site differs from the mouse sequence by one nucleotide. The chicken VDR promoter is also structurally similar to mouse, lacking a TATA box and possessing several GC-rich Sp1 sites. Deletion studies indicated that the proximal Sp1 sites are strong elements to drive gene expression, with upstream promoter sequences having a negative effect on gene expression [52]. This promoter also has an AP2 site, and a Krox20/24 site outside the proximal promoter region [52]. Thus far, the regulatory contributions of these sites to chicken VDR expression have not been elucidated. 2. HUMAN VDR PROMOTERS
Regulation of the expression of human VDR appears to be more complex than the rodent and avian VDR with six untranslated exons 5′ to the exon that
198
LUIS M. ESTEBAN, JOHN A. EISMAN, AND EDITH M. GARDINER
FIGURE 3 Alignment of the promoter regions of mouse, rat, and human VDR. The first exon is underlined (exon 1 of mouse and exon 1A of human and rat). Bent arrows indicate major transcriptional start sites. Superscript stars on shaded nucleotides indicate the beginning of the published cDNAs [19,20,26]. Additional sequence in the first exon was determined: for mVDR (mouse) by primer extension [30] and 5′ RACE (Esteban et al., unpublished); for rVDR (rat) by 5′ RACE (Esteban et al., unpublished); for hVDR (human) by 5′ RACE [24]. In the 600 bp upstream of the first exon there is 88% identity between mouse and rat and about 63% between rodents and human sequence. Alignment with chicken VDR promoter is not included because its identity with mammals is only about 40%, although the GC-rich region containing the Sp1 sites is highly conserved [30]. Specific transcription factor responsive elements are indicated only for mouse VDR which has been more extensively studied. Other sites that are also present in the human and chicken sequence [24,52] and in the promoter of human exon 1C [51] are discussed in the text. Alignment is based on mouse VDR promoter coordinates (gb#AF017779, 1574 bp) and the homologous region of rVDR promoter from gb#AC119476 and hVDR promoter from gb#AY342401 (nt 235-2019). Asterisks indicate sequence identity in all three promoters.
199
CHAPTER 12 VDR Promoter and Abundance
E-Box c-MYB
Sp1 1A
WRE
Forskolin, ATRA, 17β-E2, Phytoestrogens, Dex ATG
ATG hVDR
1F
1E 1A1D
2
1B 1C
Cdx-2
ATRA
FIGURE 4 Regulatory regions of human VDR locus. White circles indicate regions with functional promoter activity, bent arrows on exons 1A, 1C, 1D, and 1F are transcriptional start sites. Two major initiation codons on exons 1D (for VDRB1) and exon 2 (for VDR) are indicated by ATG above downward arrows. Experimentally confirmed transcription factor response elements are: WRE (WT1 responsive element), Sp1 transcription factor, Cdx-2 (caudal-related homeodomain). Also two E-boxes (ZEB responsive element) and several c-MYB responsive elements present in the promoter region 5′ to exon 1A are functional in mouse [62], but have not been tested in the human sequence. Other untested consensus response elements for other transcription factors [24] are not shown. Forskolin, ATRA (all-trans-retinoic acid), 17β-E2 (17β-estradiol), phytoestrogens, and Dex (dexamethasone) induce the promoter 5′ to exon 1C, but response elements have not been defined.
encodes the originally described initiation codon, exon 2 (Fig. 4) [24,27]. Up to 14 variant transcripts assembled from multiple combinations of the six upstream exons were initially identified in human kidney [27]. Their expression is regulated by up to three distinct promoters: a promoter upstream of exon 1A and equivalent to those described for rodents and chicken [24,27], a second promoter upstream of exon 1F [27], and a third promoter upstream of exon 1C [51]. The proximal 5′-flanking region upstream of exon 1A of the human VDR gene is highly conserved with chicken and mouse loci, both in its lack of a TATA box initiator and the presence of a GC-rich region with putative Sp1 and transcription factor binding sites (Fig. 4). As in mouse, proximal elements from 103 to 464 bp upstream of the transcription start site displayed high promoter activity, whereas elements further upstream repressed transcription [24]. Also comparable to the mouse, deletion of the Sp1 sites proximal to the start of the mRNA resulted in a 10-fold drop in promoter activity, indicating an important role in promoter function. This same promoter may also drive the expression of exon 1D transcripts, as available evidence indicates that upstream sequences within 300 bp of exon 1D lack promoter activity and may, in fact, contain a suppressor element [27]. Further studies are required to clarify whether the exon 1D promoter does overlap the 1A promoter and/or transcribed exon sequence. A few transcription factor binding sites have been identified in the human 1A promoter region. At position –3731 to –3720 bp relative to the exon 1A transcription start site, there is a polymorphic binding site for
the caudal-related homeodomain protein Cdx-2 (see Section III,C), which activates VDR gene transcription by binding to this element [67,68]. As it is an intestinespecific transcription factor, Cdx-2 may contribute to tissue-specific regulation of the VDR. As in the mouse VDR promoter, there are two E-box targets for the ZEB transcription factor and potential binding sites for the transcription factor c-MYB [62]. There are also several WREs: one overlapping the most distal Sp1 site [66] and three more putative distal sites that are able to bind WT1 protein [69]. The only functional WRE of the latter three, 308 bp upstream of the exon 1A transcription start site (Figs. 3 and 4), appears to be distinct from the WT1-response element defined for the mouse promoter. A distal promoter of the human VDR gene, more than 9 kb upstream of exons 1A and 1D, drives the expression of exon 1F–containing transcripts. Whereas exon 1A and 1D transcripts appear to be ubiquitous, the distribution of exon 1F–containing transcripts is more restricted, detected to date only in human kidney, parathyroid tissue, and the intestinal carcinoma cell line LIM 1863 [27]. As these sources represent major calciotropic target tissues, the distal promoter may require tissue- or cell-specific regulatory factors to mediate the restricted gene expression pattern. Interestingly, although not translated, exon 1F–containing transcripts are underexpressed in parathyroid adenomas and in hyperplastic glands of secondary hyperparathyroidism in comparison with normal parathyroid glands [70]. In addition to those initiating at exons 1F, 1A, and 1D, human VDR transcripts that initiate at exon 1C
200
LUIS M. ESTEBAN, JOHN A. EISMAN, AND EDITH M. GARDINER
have been detected in MCF-7 breast cancer cells [51]. These transcripts appear to arise from a third human VDR promoter upstream of exon 1C, as evidenced by the promoter activity of a 1.3 kb 5′ flanking segment in reporter gene assays [51]. Characteristic promoter features in this segment include GC-rich regions and a TATA box at –29 bp, as well as putative AP-2 and Sp1 transcription factor binding sites and several consensus glucocorticoid response elements (GREs). The GREs appear to be regulated according to cell type, as dexamethasone induced promoter activity in the SUM159PT but not in the MCF-7 breast cancer cell line [51]; however, specific functional GRE(s) in this region have not been defined. There are no consensus vitamin D (VDRE), retinoic acid (RARE), or estrogen (ERE) responsive elements in the functional 1.3-kb exon 1C promoter fragment, nor is there a clear forskolin responsive element [51]. This is somewhat surprising, as transcriptional regulation by this region is enhanced in the estrogen receptor (ER)-positive MCF-7 cell line by treatment with forskolin, retinoic acid, and 17β-estradiol. The latter response appears to be mediated by the estrogen receptor, as the response to estradiol is absent in the ER-negative SUM159PT cell line [51]. This promoter is also responsive to up-regulation by phytoestrogens in the ER positive human breast cancer cell lines T47D and MCF-7 [71]. A retinoic acid responsive region downstream of exon 1C may be relevant for regulation of endogenous VDR expression [24], and the effect of forskolin may be mediated through the AP-2 sites upstream of the human exon 1C [51], as some studies have implicated AP-2 elements in mediating cAMP responsiveness of other gene promoters [72,73]. Further studies are required to clarify the mechanisms of these 1C promoter hormone responses, which are of particular interest as there is currently no evidence to suggest that either the exon 1F or the exon 1A promoter is hormonally regulated [24,25,27,30].
1F
ATG 1A 1E 1D
Cdx-2
1B
ATG 2 1C
Fok I
C. VDR Polymorphisms Polymorphisms in the VDR locus have been studied in relation to skeletal biology, renal metabolic and immune functions, and cancer risk and responsiveness in a variety of populations worldwide. The most commonly studied polymorphisms and a more recently described polymorphism in a transcription factor binding site are briefly discussed here. A polymorphism has been described in the Cdx-2 binding site of the human promoter (see Section III,B,2), which resulted in lower promoter activity (70% of control transcription level) and impaired binding to Cdx-2 protein (Fig. 5). This single nucleotide polymorphism may have significant impact on target-cell responsiveness to 1,25(OH)2D3 by affecting VDR abundance. It has been associated with lower bone mineral density (BMD) at the lumbar spine in postmenopausal Japanese women [68]. Presence of a polymorphic FokI restriction site in exon 2 (the ‘f’ allele) actually affects receptor structure, allowing production of a normal-length human VDR (427 aa), whereas absence of the site (the ‘F’ allele) produces an N-terminally shortened 424 aa protein [47]. The ‘F’ allele has been associated with elevated transactivation activity rather than differences in VDR expression per se [48,49,74]. Specific FokI alleles have been correlated with differences in bone density, fracture incidence, nutritional rickets, risk of osteosarcoma, auto-immune hepatitis, urolithiasis, insulin sensitivity and drug clearance rates [75–81]. Restriction site polymorphisms for the enzymes BsmI and ApaI have been described in the last intron of the human VDR gene and a silent TaqI site in exon 9 [82,83] as well as a polyadenosine (A) microsatellite within the 3′ untranslated region (3′ UTR) [84]. These polymorphisms have been associated with differences in BMD in a number of genetic and population based studies [85–90], but not all have observed such associations [91,92]. Although unlikely to lead to changes in
TGA 3
45
6
78
Bsm I Apa I
9
Poly(A) Taq I microsatellite
FIGURE 5 Polymorphisms in human VDR gene. Single-nucleotide polymorphism has been identified in the Cdx-2 site, restriction site polymorphisms in exon 2, intron 8, and exon 9, and poly(A) microsatellite in 3′ UTR. Only the FokI polymorphism affects protein structure. The BsmI, ApaI, TaqI, and poly(A) polymorphisms are in linkage disequilibrium in some populations [74,84].
201
CHAPTER 12 VDR Promoter and Abundance
the receptor structure, these polymorphisms could affect mRNA accumulation and hence VDR protein level and function. A minigene representing a haplotype lacking the BsmI and ApaI sites but including a functional TaqI site (‘BBAAtt’) produced higher reporter gene activity than the ‘bbaaTT’ haplotype in transient transfection assays, suggesting that the ‘BBAAtt’ haplotype accumulated more VDR mRNA and protein [85]. Consistent with this finding, alleles ‘b’, ‘a’, and ‘T’ were associated with significantly lower VDR and higher PTH mRNA levels in patients with primary hyperparathyroidism [93]. Similarly, the poly(A) microsatellite ‘L’ (long) allele in the 3′ UTR was associated with higher VDR activity than the ‘S’ (short) allele in human fibroblast cell lines [74]. Allelespecific variation in mRNA stability or transcriptional regulation may be tissue or cell type specific, however, as the opposite pattern was observed in myeloid leukemia and prostate cancer cell lines, with a 30% reduction in RT-PCR product derived from the small ‘t’ allele [94]. Tissue specific allele effects may also explain the lack of relationship between VDR alleles and receptor density in duodenal mucosa of healthy premenopausal women, despite an association with bone mineral content (BMC) [95], and also the lack of association between VDR genotype and protein concentration in duodenum [96], peripheral blood mononuclear cells [97], and cultured skin fibroblasts [98]. Thus, although there appears to be no association between the BsmI, ApaI, and TaqI polymorphisms and VDR levels in many tissues and cell types studied, genotype may affect receptor levels in some tissues or under certain pathological conditions. Recent studies have indicated genetic interactions between various VDR alleles, with combined haplotypes sometimes regulating polymorphism effects that were not detected in single polymorphism analysis [74].
IV. REGULATION OF VITAMIN D RECEPTOR EXPRESSION AND ABUNDANCE An important mechanism for the modulation of cellular responsiveness to 1,25(OH)2D3 is mediated by the regulation of receptor abundance. Developmental regulation occurs, with receptors for 1,25(OH)2D3 appearing in the neonatal rat intestine shortly before weaning, coincident with a switch to vitamin D–dependent calcium absorption [99]. Target cell responsiveness is also regulated, with the ability of 1,25(OH)2D3 to stimulate 24-hydroxylase and inhibit collagen synthesis in mouse osteoblastic cells proportional to its receptor
content [100,101]. Receptor abundance also varies with cellular differentiation, exemplified in a series of clonal rat osteosarcoma cell lines of increasing osteoblasticlike character (ROS 24/1, ROS 2/3, and ROS 17/2.8), in which receptor levels correlated with the magnitude of 1,25(OH)2D3-associated inhibition of growth, transition from cuboid to spindle-like morphology, and suppression of colony formation in soft agar [102]. The following section focuses on a variety of studies which have shown that complex physiological signals involved in the regulation of VDR abundance vary among species and between the various 1,25(OH)2D3 target organs. To facilitate comprehension of this complex story, main findings are summarized in Table I.
A. Homologous Regulation VDR expression is affected by the presence of its own ligand, 1,25(OH)2D3, with ligand-mediated autoregulation of VDR documented in several experimental systems. Using a chicken cDNA probe, a receptor mRNA species in mouse 3T6 fibroblasts was substantially increased by prior exposure to 1,25(OH)2D3 [14]. In the pig kidney cell line LLC-PK1 there was homologous up-regulation by 1,25(OH)2D3 and other vitamin D analogs with a substantial increase in the number of 1,25(OH)2D3 receptors without altering the affinity of receptor for the hormone. Treatment of these cells with the RNA synthesis inhibitor actinomycin D showed that the increase in receptors was in part dependent on RNA synthesis, and in part on receptor occupancy [103]. Similar results were observed in rat osteosarcoma ROS 17/2 cells, in which there was a receptor increase upon treatment with 1,25(OH)2D3, with roughly 50–60% of the increase prevented by inhibition of RNA polymerase II activity with α-amanitin [104]. Similarly, human breast cancer cells increased specific binding of [3H]1,25(OH)2D3 upon hormone treatment, which was prevented by actinomycin D treatment [105]. Thus, although pharmacological inhibitors of transcription partially inhibit VDR induction, other 1,25(OH)2D3 responsive mechanisms also appeared to be involved in homologous regulation of receptor abundance. The typical response to 1,25(OH)2D3 treatment is characterized by an ascending phase, which represents hormone association with receptor and reaches a maximum at 90–120 min, followed by a rapid descending phase that is closely associated with a decrease of medium 1,25(OH)2D3 due to the metabolism of the hormone [106]. Treatment of the osteosarcoma cell line UMR-106 with ketoconazole (an inhibitor of 1,25(OH)2D3 metabolism) resulted, after a lag-time of about
202
LUIS M. ESTEBAN, JOHN A. EISMAN, AND EDITH M. GARDINER
TABLE I Brief Summary of Regulators of VDR Abundance
1,25(OH)2D3
Calcium Phosphorus
Fatty acids Glucocorticoids
Estrogens
Retinoic acid
Growth factors Cell cycle/ differentiation Growth hormone (GH) PTH
PKA activators
PKC activators
aD 3
Effect on VDR levels
Comment
Up-regulation in calcemic tissues (kidney, intestine, parathyroid gland), and no change in other tissues Up-regulation in cell lines
In normocalcemic animals
[113,114,116–120]
Mainly due to stabilization of the VDR by D3a Hypocalcemia prevents upregulation by D3 In low-phosphate diet
[124,125]
In hypocalcemia, down-regulation in kidney and parathyroid gland, and no effect in intestine In rat and chick increase of intestinal VDR. In kidney no change Parathyroid gland had no change Modulation of differentiation and increase of VDR by butyrate Species-dependent. Increased VDR in rat and decreased in mouse Intestine up-regulation Kidney down-regulation Testis, heart, lung, no change Increase of VDR in intestine and enhancement of Ca2+ absorption No change in kidney VDR increase in osteoblastic cells Up-regulation of hVDR exon 1C prom Up-regulation of VDR is tissue- and species-dependent Up-regulation of hVDR exon 1C prom Up-regulation of VDR and stimulation of proliferation. In vivo, tissue-specific Proliferating cells, VDR high Quiescent cells, VDR low Increased Ca2+ absorption and intestinal CaBP-9K, but no change in VDR Reduction in kidney VDR increase in osteoblastic cells VDR down-regulated in ROS17/2.8 and decreased up-regulatory effect of D3 VDR up-regulated by forskolin and other PKA activators Blocking Ca2+ channels reduces cAMP induction Phorbol esters down-regulate VDR in osteoblastic cells In rat kidney and epithelial cells potentiation of D3 action VDR phosphorylation
is 1,25(OH)2D3.
References
[119,127,129,130] [112,132]
With high-phosphate diet Synergistic effect of D3 and butyrate Effect probably not VDR specific, but on general development Effects are tissue-dependent and cell type-dependent
[133] [135,136,139]
Some effects of 17β-estradiol on Ca2+ are independent of D3
[155–158,162]
ER-dependent Synergistic effect of ATRA and D3
[51] [126,152,171–174]
[141,142,144] [109,143,150–152]
By EGF, insulin, IGF-I
[24,51] [179–182]
No up-regulation by D3 in differentiated cells May be IGF-I mediated effect
[100,101,185, 187] [190–192,194]
cAMP mediated By PTH
[196,198,199] [200,201]
Via cAMP elevation
[61,182,198,203]
When activated by PTH Via PKC activation
[204] [183,205]
Via PKC activation VDR-PKC cross-regulation
[207–211,214]
CHAPTER 12 VDR Promoter and Abundance
90 min, in a sharp increase of receptor accumulation, which was blocked by coincubation with cycloheximide and actinomycin D [107]. Most of these findings were based on receptor quantification by high specific activity tritiated ligand and, therefore, non-hormone-binding or unoccupied forms of the receptor may not have been characterized. Western blots using the monoclonal antibody IVG8C11, which has high cross-reactivity with porcine and other mammalian VDR molecules [11,108], avoided this potential problem and detected up-regulation of 1,25(OH)2D3 receptors in human HL-60 promyelocytic leukemic cells line at 12 hr and down-regulation at 48 and 72 hr in the continuous presence of hormone [13]. Furthermore, 1,25(OH)2D3 stabilized VDR mRNA levels in human MG-63 osteosarcoma cells, increasing its apparent half-life by 11 hr and increasing its VDR protein content [109]. However, the effect of 1,25(OH)2D3 on VDR levels appears to be cell specific, as treatment with the hormone down-regulated VDR mRNA level in the human megakaryoblastic leukemia cell line HIMeg, but caused no change in the human osteosarcoma cell line HOS-8603 [110]. Similarly, treatment with 1,25(OH)2D3 did not alter VDR protein expression in three phenotypically distinct osteosarcoma cell lines [111]. In general, these in vitro studies indicate that in many but not all cell line models, receptor abundance is increased by transcriptional and translational mechanisms. Specificity of VDR up-regulation by 1,25(OH)2D3 has also been observed in numerous in vivo studies, with outcomes highly dependent upon the tissue examined, the method by which 1,25(OH)2D3 levels were manipulated, and perhaps also upon species. Basal expression of chicken intestinal VDR appears not to be dependent upon 1,25(OH)2D3, as VDR mRNA concentrations were similar in vitamin D–deficient and normal chick duodena [112]. In vitamin D–deficient rats injected with 1,25(OH)2D3, there was significant up-regulation of renal VDR after 5 days of treatment, again with minimal effect on duodenal VDR [113]. Duration of treatment influenced results, with shorter treatment period more likely to yield an increase in VDR mRNA levels. In rat parathyroid gland there was a maximum increase of VDR mRNA at 6 hr and only a marginal increase at 24 hr [114], whereas daily dosing for 5 days did not alter the level of VDR mRNA in either the intestine or kidney of vitamin D–deficient rats [115]. Furthermore, mRNA levels of the receptor in the intestine of vitamin D-deficient rats increased 10-fold above deficiency levels at 6 and 12 hr after an intravenous dose of 1,25(OH)2D3, returning to predosing levels at 24 hr [116]. The protein response in this study was less marked, as there was only a twofold increase of the VDR at 12 hr, as detected by IRMA, with
203 no change detected by ligand-binding assay [116]. In normocalcemic rats, the administration of vitamin D resulted in a 5-fold increase in kidney receptor concentration, whereas in hypocalcemic animals, vitamin D treatment did not change receptor levels [117,118], perhaps reflecting a homeostatic increase in plasma 1,25(OH)2D3 levels in the latter group. A consistent set of observations was reported with exogenous administration of 1,25(OH)2D3 in rats fed a normal diet, which resulted in a 1.5-fold increase in duodenal and a 3-fold increase in renal VDR content [119]. By contrast, rats fed a low-calcium diet in the same study exhibited plasma 1,25(OH)2D3 concentrations similar to those treated with exogenous 1,25(OH)2D3, but duodenal VDR content was not up-regulated and renal VDR was 20–38% lower than in rats with normal calcium intake. Thus, VDR levels in kidney and intestine may be homeostatically regulated by dietary calcium, but not through the elevation of plasma 1,25(OH)2D3 levels [119] (see Section IV,B,1,a). Yet homologous VDR up-regulation was not a universal property of 1,25(OH)2D3 target tissues, as homologous up-regulation of VDR occurred under a wide variety of experimental conditions in rat kidney but not in non-calcium homeostatic VDRcontaining tissues such as lung, heart, or testis [120]. Clearly, the mechanisms of homologous up-regulation of VDR must be complex. Many studies addressing mechanism have focused on VDR transcript levels in cell line models after 1,25(OH)2D3 treatment. In addition to those studies mentioned above, increased VDR mRNA was also observed after treatment in mouse fibroblast 3T6 cells [121], the human osteosarcoma cell line MG-63 [109], and HL-60 cells [122]. Accumulation of VDR transcripts appears to occur in part through increased transcription, but also to a substantial extent through mRNA stabilization, as evidenced by studies using RNA polymerase II inhibitors [103–105]. Studies suggestive of stabilization of the VDR protein by its ligand include radio-ligand binding studies in human breast cancer and pig LLC-PK1 cell lines [105,123], but not in the rat osteosarcoma cell line ROS 17/2 [104]. More recent studies using IRMA indicated that VDR protein levels were elevated 3-fold after 8 hr of treatment with no change in mRNA levels in 3T6 mouse fibroblasts and IEC-6 rat intestinal epithelial cells, as well as in rat intestine and kidney [124]. Notably, with inhibition of protein synthesis using cycloheximide, 1,25(OH)2D3 markedly reduced degradation of previously formed receptor. Consistent findings were reported in the rat osteosarcoma ROS 17/2.8, and in the human breast cancer T47-D cell lines, in which administration of 1,25(OH)2D3 did not change receptor mRNA level, but protein did accumulate as a result of increased receptor half-life [125,126]. Thus, in addition
204 to increasing transcription and mRNA stabilization, it appears that 1,25(OH)2D3 can increase VDR content in vivo or in cell culture by stabilizing the receptor. This effect on protein stability may be due to hormoneinduced conformational changes that directly affect receptor degradation or that enhance its functional interactions with RXR, DNA, or transcriptional co-regulators and thereby prolong receptor half-life. As ligand effects on nuclear receptor complex formation, stabilization and turnover are a very active area of investigation (described in more detail in Chapters 13 to 16), it is likely that this aspect of VDR regulation could be significantly clarified in the near future.
B. Heterologous Regulation VDR expression can be modulated by numerous physiological stimuli such as dietary composition, steroid hormones and retinoids, growth factors, peptide hormones, and second messenger activators. This section will consider the effects of these agents in regulating VDR levels to clarify the interrelationships between their activities in calcium homeostasis and cell proliferation and differentiation. 1. DIETARY COMPOSITION
A primary function of the vitamin D hormone system is control of calcium and phosphorus homeostasis, in part through 1,25(OH)2D3 regulation of its own receptor, as described in Section IV,A. As the levels of 1,25(OH)2D3 in vivo are tightly linked with the levels of dietary calcium and phosphorus (discussed in detail in Chapters 24 to 26), this section will briefly discuss mechanistic aspects of the effects of dietary calcium and phosphorus on VDR expression. Fatty acids also play a role of in the regulation of VDR levels, in part through interactions between the VDR and other nuclear receptors and transcriptional cofactors, as described in Chapter 16. Recent studies that reveal a role for dietary lipids in the regulation of VDR abundance are also briefly discussed here. a. Calcium A number of studies have shown that calcium levels have no impact on the VDR levels in the intestine, but do regulate its abundance in kidney and parathyroid gland, as under conditions of severe hypocalcemia and resulting hyperparathyroidism, little or no VDR is present in these tissues. For example, in rat kidney, in the absence of vitamin D, total VDR protein was increased 2-fold by an increase in serum calcium [117], indicative of a vitamin D–independent regulatory pathway. Also, although vitamin D increased renal VDR levels 5-fold at normal serum calcium levels, this was not the case in hypocalcemic animals [117].
LUIS M. ESTEBAN, JOHN A. EISMAN, AND EDITH M. GARDINER
Complementing these VDR protein studies, dietary calcium restriction decreased renal VDR mRNA levels in mice, even after treatment with 1,25(OH)2D3, whereas with dietary calcium supplementation, treatment with 1,25(OH)2D3 caused VDR mRNA to increase 8- to 10-fold in the vitamin D-deficient group [127]. These findings are also consistent with reduced renal VDR protein content after 1,25(OH)2D3 treatment of rats with low dietary calcium intake compared to normocalcemic animals [118,119,128,129]. Thus, hypocalcemia markedly impairs homologous up-regulation of renal VDR by 1,25(OH)2D3, suggesting interaction between distinct 1,25(OH)2D3- and calcium-responsive mechanisms in this tissue. By contrast to these renal changes, duodenal VDR was not up-regulated after 1,25(OH)2D3 treatment in rats fed a low-calcium diet [118,119,129]. These findings suggest that in rodent models, down-regulation of VDR under hypocalcemic conditions is tissue-specific: intestinal VDR is present at both low and high serum calcium levels, whereas kidney VDR is absent or markedly reduced under hypocalcemic conditions, resulting in abolition of the 1,25(OH)2D3 response. A high-calcium diet prevented a fall in VDR mRNA levels in the parathyroid gland of rats with hyperparathyroidism, as did 1,25(OH)2D3 treatment [130]; this coordinate regulation of VDR expression by calcium and 1,25(OH)2D3 in kidney and parathyroid gland was not associated with changes in intestinal VDR mRNA. There are species differences, however, as there was a significant decrease in VDR mRNA in the duodenum of chicks fed a low-calcium diet in the presence of increased serum 1,25(OH)2D3 [112]. Thus, complex regulation of VDR levels under dietary deficiencies is likely to involve other factors or hormones such as parathyroid hormone for modulation of the 1,25(OH)2D3 response. Other physiological changes affecting calcium homeostasis are pregnancy and lactation, which both up-regulate intestinal calcium absorption. However, rat intestinal VDR mRNA did not change during gestation and lactation, although intestinal VDR content increased 2-fold in late pregnancy and lactation, suggesting a posttranscriptional mechanism of VDR up-regulation [131]. In general, these studies indicate that both 1,25(OH)2D3 and calcium are required for VDR production in the kidney above the basal level, whereas calcium has only a minor effect on intestinal VDR level. Downregulation of renal VDR in response to hypocalcemia may be a protective measure to block 1,25(OH)2D3mediated suppression of the 1α-hydroxylase and induction of the 24-hydroxylase, which would result in a net increase in serum 1,25(OH)2D3 levels [129].
205
CHAPTER 12 VDR Promoter and Abundance
The mechanisms by which 1,25(OH)2D3 and calcium regulate renal VDR expression levels may be indirect or posttranscriptional, as no vitamin D response elements have been detected in either mouse or human VDR promoter, and no response to 1,25(OH)2D3 has been observed in reporter-gene analyses, suggesting that 1,25(OH)2D3 does not directly activate expression of its own receptor [24,25,27,51]. b. Phosphorus Dietary phosphorus restriction upregulates VDR in a tissue-specific and time-dependent manner [132]. A low-phosphorus diet resulted in rapid increase of serum 1,25(OH)2D3 and a 3.5-fold increase in rat intestinal VDR by day 3, as quantitated by ligand binding assay, followed by a decrease to a plateau of 2.5-fold of control levels by day 5 to 10. This finding is consistent with an increase in VDR mRNA in duodenum of chicks fed a low-phosphorus diet [112]. By contrast, a low-phosphorus diet did not up-regulate kidney or splenic monocyte/macrophage VDR in rat [132]. In another study, rats on a high-phosphorus diet did not exhibit altered levels of VDR in the parathyroid gland, although PTH mRNA levels were elevated and there was prevention of both hypocalcemia and low calcitriol levels [133]. The mechanisms by which phosphorus may regulate VDR mRNA levels and functional receptor abundance have not yet been elucidated. c. Fatty Acids Understanding of the roles of dietary fatty acids in regulation of gene expression is increasing [134] and evidence that lipids regulate VDR levels is accumulating. In rats, an ovariectomy-associated increase in intestinal VDR protein declined after longterm dietary supplementation with essential fatty acids [135]. As lipids activate transcription by the nuclear receptor peroxisome proliferator-activated receptor (PPAR), their effects on VDR levels may be indirect, via increased competition by PPAR for the VDR heterodimerization partner RXR [134]. Fatty acid regulation of protein kinases, particularly PKC, may also modulate VDR action and expression levels (see Section IV,B,4). Butyrate, a short-chain fatty acid and normal constituent of colon content, enhanced nuclear VDR protein content in Caco-2 cells in a time- and dosedependent manner; its prodrug, tributyrin, significantly increased VDR-mRNA level (2.5-fold) and VDR binding activity [136]. This response may mediate the synergistic effect of butyrate compounds with 1,25(OH)2D3 in induction of differentiation in the Caco-2 cell line [137]. Similar butyrate-induced increases have been observed in some [138,139], but not all [139] cell culture models, perhaps reflecting tissue and species specificity. To date, however, there is limited evidence for butyrate regulation of VDR abundance or 1,25(OH)2D3 function in vivo.
2. STEROIDS AND RETINOIDS
In addition to 1,25(OH)2D3, other hormones that regulate VDR abundance are steroids and retinoids. Like calcitriol, these classes of hormones also act through nuclear receptors to regulate transcription of target genes that include the VDR gene. This section will focus on glucocorticoid, estrogen, and retinoid effects on VDR abundance, as each is clinically important, particularly in skeletal health. a. Glucocorticoids In prolonged exposure to elevated glucocorticoid levels such as in Cushing’s syndrome or in glucocorticoid therapy, glucocorticoid actions appear to oppose vitamin D effects (see Chapter 73). The result is often osteopenia, with major calcium homeostatic changes in intestine and bone contributing to accelerated bone loss [140]. The present section will focus on the role of glucocorticoids in the regulation of VDR abundance as a possible basis for this apparent antagonism between glucocorticoid and vitamin D actions. In vivo treatment with the glucocorticoid hydrocortisone has been reported to increase intestinal VDR mRNA in rat if administered between 15 and 17 days postpartum, although administration before day 14 or on days 19–21 postpartum did not affect VDR expression [141,142]. The sensitive period between days 15 and 17 postpartum coincides with a glucocorticoid-sensitive period of rat intestinal development, however, and may reflect a general glucocorticoid effect on intestinal maturation rather than a specific effect on VDR mRNA, as it is accompanied by an increase in actin mRNA [142]. The effect of glucocorticoids does appear to be tissue specific, as dexamethasone injection caused VDR up-regulation in rat intestine and down-regulation in kidney, but had no effect on receptor level in testis, heart, and lung [143]. Species differences may also exist, as intestinal VDR was decreased by hydrocortisone administration in mouse [144], whereas changes in VDR were not observed in chicken intestine after treatment with the glucocorticoid compound triamcinolone [145]. There is also species variation in bone cell response. In short-term rat calvarial organ cultures glucocorticoids prevented a fall in VDR levels [146] and in rat osteosarcoma cell cultures glucocorticoids stimulated 1,25(OH)2D3 receptor levels [147,148]. By contrast, in mouse osteoblastic cells glucocorticoids caused a decrease in VDR abundance [149]. Given these interspecies differences, extrapolation from animal models to humans about glucocorticoid effects on VDR is problematic. For mechanistic studies, human cells and tissue models are particularly important, even though outcomes can vary in a single cell line. In one study using MG-63 osteosarcoma cells, dexamethasone produced a time- and dose-dependent
206 decrease in VDR mRNA [150], whereas in another study dexamethasone increased the amount of VDR mRNA, although VDR protein levels were not affected [109]. Similarly, VDR protein levels in cultured human keratinocytes and fibroblasts were not altered after treatment with corticosteroids [151]. Evidence for functional interaction comes from MCF-7 breast cancer and Caov-4 ovarian cancer cell lines, in which dexamethasone augmented calcitriol effects, although VDR levels were not assessed in this study [152]. Thus, it seems that the effects of glucocorticoids on the 1,25(OH)2D3 response in human cells are tissue-specific and may not relate to VDR levels per se in all cases. Although the effects of glucocorticoids on VDR might be mediated by direct action of the glucocorticoid receptor on the promoter of the VDR gene, no glucocorticoid responsive elements (GREs) have been found in the exon 1 promoter region (1.5 kb) of the mouse or exon 1A upstream region of the human VDR locus (discussed in Section III,B) [24,25,27]. However, in the 5′ flanking region of human exon 1C there are putative GREs [51]. Dexamethasone did not induce activity of this promoter in the estrogen-dependent MCF-7 breast cancer cell line, although there was upregulation in the estrogen-independent SUM159PT breast cancer cell line [51], consistent with cell type specificity of glucocorticoid effects on VDR abundance. The presence of multiple promoters, as in the human VDR locus, or indirect regulation via other mechanisms (e.g., as in estrogen-dependent versus estrogenindependent breast cancer cell lines, or perhaps by competition for transcription cofactors; see Chapter 16) may account for the apparent differences in the effects of glucocorticoids on VDR regulation. Furthermore, glucocorticoids may interact with other hormones that regulate VDR abundance, such as PTH (see Section IV,B,3,c). b. Estrogens Actions of the steroid hormone estrogen are mediated by the estrogen receptor (ER), another member of the nuclear receptor superfamily. Reductions in estrogen levels at the menopause are associated with decreased bone mineral density, which is thought to occur as a result of calcium homeostatic changes. One such change is a reduction in intestinal calcium absorption [153] which is improved by estrogen replacement [154]. Although 17β-estradiol (E2) may positively affect intestinal calcium absorption independent of vitamin D [155], the actions attributed to estrogens might also be mediated at least in part by regulation of 1,25(OH)2D3 or VDR levels, as 1,25(OH)2D3 is the principal hormonal regulator of intestinal calcium absorption. Animal models support this concept, as estrogenstimulated increases in VDR expression and response were observed in rat duodenum [156,157]. Other studies
LUIS M. ESTEBAN, JOHN A. EISMAN, AND EDITH M. GARDINER
have found that estrogen and phytoestrogens such as genistein increase VDR transcript level, protein expression, and endogenous 1,25(OH)2D3 bioactivity in rat colonic mucosa and colon carcinoma cell lines [158,159]. Thus, some estrogen activities in the colonic mucosa such as growth suppression could be mediated, at least in part, by increased responsiveness to endogenous 1,25(OH)2D3. However, in another study intestinal VDR levels in immature rat were unaffected by estrogen, although there were increases in VDR abundance in uterus and liver but a decrease in kidney [160,161]. Thus, conclusions about the regulation of the VDR in one tissue cannot be extrapolated to other tissues. There have been a number of studies of estrogen effects on VDR abundance in bone-derived cells because of the relevance to postmenopausal osteoporosis. VDR expression in mouse bone decreased with aging and estrogen deprivation but recovered after E2 supplementation [162]. In vitro findings are consistent, as E2 treatment doubled total cellular VDR-binding capacity in the ROS 17/2.8 rat osteosarcoma cell line [163], increased VDR protein in human osteosarcoma cell lines [164], and increased VDR mRNA levels in the human MG-63 and N-976 osteoblastic cell lines [109,162]. Correlation between VDR mRNA and protein changes was not entirely consistent [109], although E2 treatment did potentiate the antiapoptotic effect of 1,25(OH)2D3 in the latter study [162]. Interaction between vitamin D and estrogen regulation has been observed in VDR knockout mice, in which VDR appears essential for the regulation of CYP19 (aromatase cytochrome P450, a key enzyme in estrogen biosynthesis) of both female and male gonads [165]. To date, estrogen responsive elements (EREs) have not been detected in the promoter region of human, mouse, or chicken VDR genes. However, the human exon 1C promoter region (see Section III,B,2) was up-regulated by 17β-estradiol and the phytoestrogens resveratrol and genistein, despite the apparent lack of an ERE in the promoter region [51,71]. As for glucocorticoids, estradiol effects on VDR abundance vary according to species and tissue. Evidence indicates that there are significant interactions between estradiol and vitamin D responses in bone and intestine, with some renal involvement as well. It is not clear whether these findings reflect a particular experimental focus on calcium homeostatic tissues, or a special role for an estrogen–vitamin D axis in calcium regulation. Studies of this interaction in other non-calcium-homeostatic tissues should resolve this question. c. Retinoic Acid Vitamin A and its active metabolites retinol and retinoic acid, known for their involvement in vision, reproduction, embryonic development, and epithelial growth and differentiation, are also
207
CHAPTER 12 VDR Promoter and Abundance
necessary for normal bone metabolism. The actions of the retinoic acids are mediated by nuclear receptors closely related to the VDR. There are two families of nuclear retinoid receptors, the retinoic acid receptor RAR, activated by all-trans-retinoic acid (ATRA) and/or 9-cis-retinoic acid (9-cis RA), and the retinoid X receptor RXR, activated specifically by 9-cis RA [166]. In addition, RXR can act as DNA binding partner for a number of other nuclear receptors, including those for the thyroid hormone (TR), as well as the VDR and the RAR [167,168]. Vitamin A and 1,25(OH)2D3 share a number of physiological effects, including stimulation of bone resorption and effects on bone cell differentiation [169] and on keratinocyte proliferation and differentiation [170]; thus, vitamin A effects on bone and other tissues may be mediated through VDR regulation. All-trans-retinoic acid produced a dose-dependent increase in 1,25(OH)2D3 binding in rat osteosarcoma cells after 24 hr that was inhibited by actinomycin D [171]. This increase in binding was due to increased VDR protein, rather than modification of receptor affinity for the hormone. Induction of VDR by retinoic acid was greater in proliferating than nonproliferating tumorigenic cells (ROS 17/2A and UMR-106M), but not in the nontumorigenic cell line RCJ 1.20 [172]. ATRA caused a reduction in VDR abundance in rat osteoblastic cells, but an increase in receptor level in mouse osteoblasts that required protein and RNA synthesis. These opposite effects of ATRA on mouse and rat cells suggest that species-dependent factors modulate the action of retinoids in mammalian cells [173]. Findings in human osteosarcoma (MG-63) and breast cancer (T47-D) cell lines were similar to those in mouse osteoblasts with an ATRA-induced increase in VDR mRNA and its half-life although, at least in the MG-63 cells, there was no change in VDR protein levels [109,126]. Other studies have focused on the synergistic effects of retinoic acid and 1,25(OH)2D3 as an indication of their interactions and cross-regulation of their activities. In breast and ovarian cancer cells (MCF-7 and NIH:OVCAR3), and in human promyeloid leukemia cells (HL-60), ATRA and vitamin D3 analogs have additive or synergistic effects with respect to inhibition of cell growth, DNA synthesis, and the induction of differentiation [152,174]. In addition, ATRA cooperates with 1,25(OH)2D3 to promote differentiation of HL60 cells to monocytes [175]. By contrast, although ATRA and 1,25(OH)2D3 each inhibited proliferation of normal human keratinocytes, in combination they had antagonistic rather than synergistic effect [176]. The action of retinoic acid may be through direct action of RAR on the VDR gene promoter, as the ability
of retinoic acid to up-regulate the VDR is mediated, at least in part, at the transcriptional level. There are sites that may confer retinoic acid responsiveness upstream and downstream of the human VDR exon 1C promoter [24,51] (see Section III,B,2), but retinoic acid response regions have not been identified in the rodent VDR loci. It appears, therefore, that the retinoic acid regulation of VDR may be tissue and species specific. It is possible that, like glucocorticoid and estrogen effects, retinoids may affect VDR abundance by indirect mechanisms such as changing levels or availability of RXR or of shared transcriptional co-regulators. If so, there may be specificity in the mechanism, as another hormone with a receptor that heterodimerizes with RXR, triiodothyronine, increased VDR content in MG63 osteosarcoma cells [177], and the progesterone analog R5020 increased VDR in the T47-D breast cancer cell line [126]. 3. PEPTIDE HORMONES
In addition to the steroid and retinoid hormones, peptide hormones also play important roles in determining VDR level. They include the polypeptide growth factors, growth hormone, and the calciotropic hormones PTH and PTHrP. This section focuses on the effects of these compounds, which act via membrane receptors. a. Growth Factors In general, 1,25(OH)2D3 inhibits cell proliferation and induces differentiation (Chapter 92), yet VDR levels are often correlated with cellular proliferation in vivo and in vitro. The level of VDR in primary mouse osteoblastic cell cultures appeared to be regulated in accordance with the rate of cell division [178], suggesting that vitamin D regulation may be modulated by cell cycle–associated changes in VDR content. Mitogenic growth factors such as epidermal growth factor (EGF), insulin, and insulin-like growth factor I (IGF-I), which stimulated the proliferation of NIH-3T3 mouse fibroblastic and MCF-7 human breast cancer cell lines and rat growth plate chondrocytes, also caused significant increases in VDR protein and mRNA levels [179,180]. Furthermore, subcutaneous administration of EGF increased intestinal VDR in the proximal segments of the small intestine but not in kidney of suckling rats [181]. By contrast, in UMR-106 osteoblastic cells, EGF reduced the number of 1,25(OH)2D3 binding sites without altering binding affinity, and attenuated PTH-induced VDR up-regulation [182]. Similar down-regulation of VDR expression by basic fibroblast growth factor (bFGF), EGF, and Ha-ras transformation was observed in mouse NIH3T3 fibroblasts and HC-11 mammary epithelial cells [183,184]. Changes in signaling by mitogenic growth factors may also be the cause of decreases in VDR in confluent cultures of the mouse MMB-1 osteoblastic cell line and
208 in mouse primary osteoblastic cell culture [100,101], although this is not a universal finding [101]. Involvement of growth factors in decreased VDR abundance during cellular differentiation is also likely [185,186], as growth factor deprivation can force cellular quiescence, with restoration of VDR expression upon cell cycle reentry [187]. In this context, although the mechanism is not known, it is interesting to note that apigenin, a flavonoid that induces cellular growth arrest with concomitant inhibition of intracellular signalling and decreased protooncogene expression, potently inhibited VDR mRNA and protein expression in human keratinocytes [188]. Effects of growth factors on VDR abundance therefore appeared to be determined by cell type or tissue origin, by the state of confluence, and by the intracellular signaling response. The differences in the effects of growth factors on VDR levels in proliferation and postconfluent cultures is consistent with a role for 1,25(OH)2D3 in control of cell cycle and progression of cells down a differentiation pathway. However, the mechanisms by which VDR abundance is controlled during cell cycle and the function of 1,25(OH)2D3 in the regulation of cellular differentiation remain as yet unanswered questions. b. Growth Hormone Postmenopausal osteoporosis is an age-related disease that is often associated with intestinal calcium malabsorption [189], and 1,25(OH)2D3 is its main hormonal regulator. Calcium malabsorption can be corrected by estrogen therapy [153] and intestinal absorption of calcium is also increased by growth hormone (GH). GH increased intestinal absorption of calcium in pigs and similarly increased intestinal calcium absorption and expression of the calcium binding protein CaBP 9K, but not plasma concentration of 1,25(OH)2D3, in aged female rats [190,191]. The stimulation of calcium absorption by GH may result, in part, from up-regulation of intestinal VDR [192] and possibly ER [193], which up-regulates intestinal VDR, as noted in Section IV,B,2,b. Although GH also increased intestinal calcium absorption and serum 1,25(OH)2D3 in pig, GH treatment did not change intestinal VDR levels though it did reduce renal VDR in this species [194]. It is still unclear whether the GH effect on VDR occurs at the level of transcription, translation, or message stability, and whether GH effects on VDR are mediated indirectly through IGF-I [191], ER [193], or changes in serum calcium [190]. c. Parathyroid Hormone Parathyroid hormone (PTH) plays a central role in the regulation of calcium homeostasis, in part by regulating the synthesis of the other major calciotropic hormone, 1,25(OH)2D3. PTH is a primary inducer of the 1α-hydroxylation of 25(OH)D3 in kidney to produce 1,25(OH)2D3 [195],
LUIS M. ESTEBAN, JOHN A. EISMAN, AND EDITH M. GARDINER
which, in turn, feeds back to control PTH production and regulate VDR in the parathyroid gland (see Chapters 5 and 30). The PTH/PTHrP receptor and VDR are both present in kidney and osteoblasts. The mechanism of PTH action is mediated through activation of the PKA and PKC signal transduction systems [196–198], both of which regulate VDR levels in vitamin D target cells (discussed in Section IV,B,4). Furthermore, PTH-induced increases in intracellular calcium may also regulate VDR levels. This diversity of PTH second messenger responses leads to differential effects on VDR levels and interactions with the vitamin D response pathway that appear to vary according to the test system. VDR content in the preosteoblastic UMR-106 rat osteogenic sarcoma cell line and the MC3T3-E1 mouse preosteoblastic line was increased by PTH in a time- and dose-dependent manner via cAMP activation [196,198,199]. By contrast, PTH or PTHrP treatment of the more mature rat osteoblastic osteosarcoma cell line ROS17/2.8 led to decreased VDR mRNA levels and 1,25(OH)2D3 binding [200,201]. The differences between these osteoblastic models may relate to the different stages of differentiation, perhaps reflecting cell type–specific variation in factors that regulate VDR levels. In concept, this would be consistent with the observations that a PTH-dependent decrease in VDR in ROS 17/2.8 was mediated by the PKA response [201], whereas a PTH-dependent increase in VDR in rat chondrocytic cultures was mediated by the PKC response pathway [202] (discussed further hereafter). Interactions in vivo are also complex, as PTH inhibited an increase in intestinal VDR levels after 5-day coinfusion of rats with 1,25(OH)2D3 [200]. There was also a partial inhibition of renal VDR levels in this study, accompanied by abolition of CYP24 response to the 1,25(OH)2D3 treatment. Tissue-specific variation in PTH effects on VDR levels presumably relate to the activation of various signaling pathways, depending on the specific target cell type and its state of differentiation (discussed further in the next section). 4. ACTIVATORS OF SECOND MESSENGERS
The various physiological inputs that affect VDR levels involve complex interactions among the intracellular signal transduction pathways. This section briefly addresses the regulation of VDR by the second messengers protein kinase A (PKA) and protein kinase C (PKC). PKA activators include forskolin or cAMP analogs, whereas PKC activators include the phorbol esters, such as phorbol myristate acetate (PMA) and phorbol-12,13-dibutyrate. Each can affect VDR levels in isolation, and the evidence suggests that the VDRmodulating activities of the PKA and PKC pathways
CHAPTER 12 VDR Promoter and Abundance
can be mutually antagonistic. Given their central role in the integration of cellular responses to extracellular molecules, these pathways are key in the mechanisms that regulate VDR levels in target cells. a. Protein Kinase A Early studies showed that cellular response to PTH was via activation of the cAMP-dependent protein kinase A (PKA) pathway, which preceded PTH-induced increases in VDR levels in UMR-106 cells [196]. Subsequent mechanistic studies in NIH3T3 cells confirmed that the adenylate cyclase activator forskolin led to a substantial time-dependent increase in VDR abundance [61]. Similar increases by forskolin have been observed in the preosteoblastic UMR-106 and MC3T3-E1 cell lines [198,199], but not in more mature osteoblastic ROS17/2.8 cells, in which forskolin caused a decrease in VDR level and suppressed homologous up-regulation of VDR [201]. PKA activation by forskolin also increased VDR mRNA in the human HOS-8603 osteosarcoma cell line and a megakaryoblastic leukemia line [110], and a prostaglandin E2-induced increase in VDR in HL-60 human promyelocytic cell line also appeared to be via cAMP [203]. The cAMP activation effect on 1,25(OH)2D3 binding in UMR-106 cells was not affected by the increase of intracellular calcium using a calcium agonist or a calcium ionophore; however, cAMP-mediated activation was reduced by use of calcium channel blockers to lower intracellular calcium [204]. The increase in VDR levels by PKA activation thus appears to involve elevation of intracellular calcium. This is particularly interesting, as intracellular calcium is increased in some cell lines by 1,25(OH)2D3, suggesting a mechanism for synergy between the 1,25(OH)2D3 and PKA responses. b. Protein Kinase C Inhibition of VDR levels in mouse NIH3T3 fibroblastic cells by basic FGF (see Section IV,B,3,a) activated protein kinase C (PKC) [183], consistent with the finding that activation of PKC by phorbol esters down-regulated VDR mRNA and protein levels in a time- and dose-dependent manner. Staurosporine, an inhibitor of PKC, blocked the effects of phorbol esters on VDR levels. Activation of PKC also overcame the stimulation of VDR expression by mitogenic serum factors [183]. Kinetics of the VDR response to PKC activation by phorbol esters was time dependent and complex, however, with a decrease and subsequent increase in VDR mRNA levels in the preosteoblastic UMR-106 rat osteosarcoma cell line [205], but a consistent time- and dose-dependent increase in the ROS17/2.8 line, representing a more mature osteoblastic stage [206]. Further, pretreatment with phorbol esters prior to 1,25(OH)2D3 treatment resulted in a synergistic 10-fold induction of VDR mRNA in the latter cell line [206]. Such synergy in regulation of VDR may explain the functional potentiation
209 of 1,25(OH)2D3 responses by PKC activation in primary rat kidney cell [207] and intestinal epithelial cell cultures [208]. In the latter, induction of the vitamin D target gene CYP24 by pharmacological doses of 1,25(OH)2D3 was potentiated by phorbol ester treatment. Similarly, 1,25(OH)2D3 induced phosphorylation of osteopontin and induced tumorigenic transformation of JB6C141.5a epidermal cells only in the presence of phorbol esters [209]. There appears to be a complex relationship between PKC and VDR actions involving cross-regulation of both molecules [210,211]. Direct modulation of the phosphorylation state of the VDR by PKC has been suggested [212,213], with phosphorylation at serine-51 by PKC altering VDR regulation of transcriptional transactivation [214]. Taken together, these observations suggest that VDR function is regulated by functional interactions between PKC and 1,25(OH)2D3. Intracellular calcium levels may play a mechanistic role in this interaction, as both PKC activation and 1,25(OH)2D3 can increase intracellular calcium. There is evidence to support this concept, as elevation in intracellular calcium by a calcium ionophore decreased VDR levels in NIH3T3 cells in a time-dependent manner, which was further decreased after further PKC activation [183]. c. PKA and PKC Interactions Simultaneous activation of PKC and PKA responses in NIH3T3 cells abolished the PKA-mediated increase in VDR protein and mRNA [61], indicating mutual antagonism between these intracellular signal transduction pathways in the regulation of VDR levels. This concept is supported by subsequent observations that PKC activation by phorbol esters in UMR-106 and ROS 17/2.8 cells blocked a PTH- and forskolin-induced up-regulation of VDR [205] and similar findings in MC3T3-E1 and UMR-106–01 osteosarcoma cells [198]. Thus, the evidence suggests cell type-specific and timedependent effects of PKC and PKA pathway activation on VDR levels, as well as interaction between these pathways. The relative strengths of the PKA and PKC signal transduction responses will, of course, vary as relevant extracellular stimuli change. The net effects of the multiple signaling inputs on VDR abundance will depend on downstream changes in VDR transcription and on VDR posttranslational modification. Changes in VDR transcription may result from activation of transcription cofactors that act on the VDR promoter, such as CREB activation by PKA after PTH stimulation. Posttranslational effects that might affect VDR function could include PKC phosphorylation of VDR, which would alter VDR transactivation activity and could have secondary effects on VDR production by homologous or heterologous pathways. The availability
210 of VDR promoter sequences and cloned promoter regions from several species will facilitate dissection of the mechanisms responsible for second-messenger effects on VDR abundance.
V. CONCLUDING REMARKS A wide range of tissues respond to 1,25(OH)2D3 with a variety of actions including calcium homeostatic changes, regulation of cellular proliferation and differentiation, and regulation of vitamin D metabolism. Abundance of the nuclear VDR is a major determinant of the strength of these 1,25(OH)2D3 responses. Factors that influence VDR content include the hormone 1,25(OH)2D3 itself, dietary calcium, phosphorus and fatty acid content, steroid and retinoid hormones, peptide hormones including growth factors and the major calciotropic hormone PTH, and other activators of secondmessenger systems. The existence of multiple promoters in the human locus and of variant VDR isoforms in human and nonhuman species suggests other mechanisms by which nuclear VDR responses to these regulators may vary. The responses to the range of regulators vary in a species-specific manner and exhibit tissue and cell-type specificity. The mechanisms that provide specificity to VDR regulation are not fully understood but may relate to the state of target cell proliferation or differentiation, cell type–specific differences in intracellular second-messenger responses to a variety of activators, and variation in the complement of downstream nuclear proteins available for the regulation of gene transcription in various target cells. There is still much work needed to elucidate the details of these mechanisms and their importance in the regulation of cell proliferation and differentiation and in the physiological vitamin D response. The complex regulation of VDR expression by this variety of effectors in fulfilling the multiple biological roles of 1,25(OH)2D3, coupled with the increasing understanding of the complex promoter structure and regulation of the VDR locus, make this gene an ideal model for further mechanistic investigation. Importantly, understanding of the regulation of VDR abundance may also have clinical utility, as target cell responses to 1,25(OH)2D3 and vitamin D analog treatment may be limited by levels of functional VDR and thus subject to enhancement by therapeutic increases in VDR availability.
References 1. Brumbaugh PF, Haussler MR 1975 Specific binding of 1α,25-dihydroxycholecalciferol to nuclear components of chick intestine. J Biol Chem 250:1588–1594.
LUIS M. ESTEBAN, JOHN A. EISMAN, AND EDITH M. GARDINER
2. Kream BE, Reynolds RD, Knutson JC, Eisman JA, DeLuca HF 1976 Intestinal cytosol binders of 1,25-dihydroxyvitamin D3 and 25-hydroxyvitamin D3. Arch Biochem Biophys 176:779–787. 3. Kream BE, Yamada S, Schnoes HK, DeLuca HF 1977 Specific cytosol-binding protein for 1,25-dihydroxyvitamin D3 in rat intestine. J Biol Chem 252:4501–4505. 4. Kream BE, Jose M, Yamada S, DeLuca HF 1977 A specific high-affinity binding macromolecule for 1,25-dihydroxyvitamin D3 in fetal bone. Science 197:1086–1088. 5. Christakos S, Norman AW 1979 Studies on the mode of action of calciferol. XVIII. Evidence for a specific high affinity binding protein for 1,25 dihydroxyvitamin D3 in chick kidney and pancreas. Biochem Biophys Res Commun 89:56–63. 6. Wecksler WR, Ross FP, Mason RS, Norman AW 1980 Biochemical properties of the 1α, 25-dihydroxyvitamin D3 cytosol receptors from human and chicken intestinal mucosa. J Clin Endocrinol Metab 50:152–157. 7. Simpson RU, DeLuca HF 1980 Characterization of a receptorlike protein for 1,25-dihydroxyvitamin D3 in rat skin. Proc Natl Acad Sci USA 77:5822–5826. 8. Eisman JA, Martin TJ, MacIntyre I, Moseley JM 1979 1,25-dihydroxyvitamin-D-receptor in breast cancer cells. Lancet 2:1335–1336. 9. Frampton RJ, Suva LJ, Eisman JA, Findlay DM, Moore GE, Moseley JM, Martin TJ 1982 Presence of 1,25-dihydroxyvitamin D3 receptors in established human cancer cell lines in culture. Cancer Res 42:1116–1119. 10. Pike JW, Marion SL, Donaldson CA, Haussler MR 1983 Serum and monoclonal antibodies against the chick intestinal receptor for 1,25-dihydroxyvitamin D3. Generation by a preparation enriched in a 64,000-dalton protein. J Biol Chem 258:1289–1296. 11. Dame MC, Pierce EA, Prahl JM, Hayes CE, DeLuca HF 1986 Monoclonal antibodies to the porcine intestinal receptor for 1,25-dihydroxyvitamin D3: interaction with distinct receptor domains. Biochemistry 25:4523–4534. 12. Tuohimaa P, Blauer M, Jaaskelainen T, Itkonen A, Lindfors M, Mahonen A, Palvimo J, Vilja P, Maenpaa PH 1992 Characterization of human 1,25-dihydroxyvitamin D3 receptor anti-peptide antibodies. J Steroid Biochem Mol Biol 43: 649–657. 13. Lee Y, Inaba M, DeLuca HF, Mellon WS 1989 Immunological identification of 1,25 dihydroxyvitamin D3 receptors in human promyelocytic leukemic cells (HL-60) during homologous regulation. J Biol Chem 264: 13701–13705. 14. McDonnell DP, Mangelsdorf DJ, Pike JW, Haussler MR, O’Malley BW 1987 Molecular cloning of complemetary DNA encoding the avian receptor for vitamin D. Science 235:1214–1217. 15. Berger U, Wilson P, McClelland RA, Colston K, Haussler MR, Pike JW, Coombes RC 1988 Immunocytochemical detection of 1,25-dihydroxyvitamin D receptors in normal human tissues. J Clin Endocrinol Metab 67:607–613. 16. Sandgren ME, Bronnegard M, DeLuca HF 1991 Tissue distribution of the 1,25-dihydroxyvitamin D3 receptor in the male rat. Biochem Biophys Res Commun 181:611–616. 17. Walters MR 1992 Newly identified actions of the vitamin D endocrine system. Endocr Rev 13:719–764. 18. Gascon-Barre M, Demers C, Mirshahi A, Neron S, Zalzal S, Nanci A 2003 The normal liver harbors the vitamin D nuclear receptor in nonparenchymal and biliary epithelial cells. Hepatology 37:1034–1042.
211
CHAPTER 12 VDR Promoter and Abundance
19. Baker AR, McDonnell DP, Hughes M, Crisp TM, Mangelsdorf DJ, Haussler MR, Pike JW, Shine J, O’Malley BW 1988 Cloning and expression of full-length cDNA encoding human vitamin D receptor. Proc Natl Acad Sci USA 85:3294–3298. 20. Burmester JK, Wiese RJ, Maeda N, DeLuca HF 1988 Structure and regulation of the rat 1,25-dihydroxyvitamin D3 receptor. Proc Natl Acad Sci USA 85:9499–9502. 21. Elaroussi MA, Prahl JM, DeLuca HF 1994 The avian vitamin D receptors: primary structures and their origins. Proc Natl Acad Sci USA 91:11596–11600. 22. Lu Z, Hanson K, DeLuca HF 1997 Cloning and origin of the two forms of chicken vitamin D receptor. Arch Biochem Biophys 339:99–106. 23. Li YC, Bergwitz C, Juppner H, Demay MB 1997 Cloning and characterization of the vitamin D receptor from Xenopus laevis. Endocrinology 138:2347–2353. 24. Miyamoto K, Kesterson RA, Yamamoto H, Taketani Y, Nishiwaki E, Tatsumi S, Inoue Y, Morita K, Takeda E, Pike JW 1997 Structural organization of the human vitamin D receptor chromosonal gene and its promoter. Mol Endocrinol 11:1165–1179. 25. Jehan F, DeLuca HF 1997 Cloning and characterization of the mouse vitamin D receptor promoter. Proc Natl Acad Sci USA 94:10138–10143. 26. Kamei Y, Kawada T, Fukuwatari T, Ono T, Kato S, Sugimoto E 1995 Cloning and sequencing of the gene encoding the mouse vitamin D receptor. Gene 152:281–282. 27. Crofts LA, Hancock MS, Morrison NA, Eisman JA 1998 Multiple promoters direct the tissue-specific expression of novel N-terminal variant human vitamin D receptor gene transcripts. Proc Natl Acad Sci USA 95:10529–10534. 28. Sunn KL, Cock TA, Crofts LA, Eisman JA, Gardiner EM 2001 Novel N-terminal variant of human VDR. Mol Endocrinol 15:1599–1609. 29. Gardiner EM, Esteban LM, Fong C, Allison SJ, Flanagan JL, Kouzmenko AP, Eisman JA 2004 Vitamin D receptor B1 and exon 1D: functional and evolutionary analysis. J Steroid Biochem Mol Biol 89–90:233–238. 30. Jehan F, DeLuca HF 2000 The mouse vitamin D receptor is mainly expressed through an Sp1-driven promoter in vivo. Arch Biochem Biophys 377:273–283. 31. Arriza JL, Weinberger C, Cerelli G, Glaser TM, Handelin BL, Housman DE, Evans RM 1987 Cloning of human mineralocorticoid receptor complementary DNA: structural and functional kinship with the glucocorticoid receptor. Science 237:268–275. 32. Kuiper GG, Gustafsson JA 1997 The novel estrogen receptor-β subtype: potential role in the cell- and promoter-specific actions of estrogens and anti-estrogens. FEBS Lett 410:87–90. 33. Tremblay GB, Tremblay A, Copeland NG, Gilbert DJ, Jenkins NA, Labrie F, Giguere V 1997 Cloning, chromosomal localization, and functional analysis of the murine estrogen receptor β. Mol Endocrinol 11:353–365. 34. Lazar MA 1993 Thyroid hormone receptors: multiple forms, multiple possibilities. Endocr Rev 14:184–193. 35. Hoopes CW, Taketo M, Ozato K, Liu Q, Howard TA, Linney E, Seldin MF 1992 Mapping of the mouse Rxr loci encoding nuclear retinoid X receptors RXRα, RXRβ, and RXRγ. Genomics 14:611–617. 36. Giguere V, Lyn S, Yip P, Siu CH, Amin S 1990 Molecular cloning of cDNA encoding a second cellular retinoic acidbinding protein. Proc Natl Acad Sci USA 87:6233–6237. 37. Dreyer C, Krey G, Keller H, Gival F, Helftenbein G, Wahli W 1992 Control of the peroxisomal β-oxidation pathway by
38.
39.
40.
41.
42.
43.
44.
45. 46.
47.
48.
49.
50.
a novel family of nuclear hormone receptors. Cell 68: 879–887. Conneely OM, Kettelberger DM, Tsai MJ, Schrader WT, O’Malley BW 1989 The chicken progesterone receptor A and B isoforms are products of an alternate translation initiation event. J Biol Chem 264:14062–14064. Kastner P, Krust A, Turcotte B, Stropp U, Tora L, Gronemeyer H, Chambon P 1990 Two distinct estrogenregulated promoters generate transcripts encoding the two functionally different human progesterone receptor forms A and B. EMBO J 9:1603–1614. Giangrande PH, McDonnell DP 1999 The A and B isoforms of the human progesterone receptor: two functionally different transcription factors encoded by a single gene. Recent Prog Horm Res 54:291–313. Vegeto E, Shahbaz MM, Wen DX, Goldman ME, O’Malley BW, McDonnell DP 1993 Human progesterone receptor A form is a cell- and promoter-specific repressor of human progesterone receptor B function. Mol Endocrinol 7:1244–1255. Tung L, Mohamed MK, Hoeffler JP, Takimoto GS, Horwitz KB 1993 Antagonist-occupied human progesterone B-receptors activate transcription without binding to progesterone response elements and are dominantly inhibited by A-receptors. Mol Endocrinol 7:1256–1265. McDonnell DP, Shahbaz MM, Vegeto E, Goldman ME 1994 The human progesterone receptor A-form functions as a transcriptional modulator of mineralocorticoid receptor transcriptional activity. J Steroid Biochem Mol Biol 48: 425–432. Suzuki T, Suzuki N, Srivastava AS, Kurokawa T 2000 Identification of cDNAs encoding two subtypes of vitamin D receptor in flounder, Paralichthys olivaceus. Biochem Biophys Res Commun 270:40–45. Kozak M 1996 Interpreting cDNA sequences: some insights from studies on translation. Mamm Genome 7:563–574. Ebihara K, Masuhiro Y, Kitamoto T, Suzawa M, Uematsu Y, Yoshizawa T, Ono T, Harada H, Matsuda K, Hasegawa T, Masushige S, Kato S 1996 Intron retention generates a novel isoform of the murine vitamin D receptor that acts in a dominant negative way on the vitamin D signaling pathway. Mol Cell Biol 16:3393–3400. Saijo T, Ito M, Takeda E, Huq AH, Naito E, Yokota I, Sone T, Pike JW, Kuroda Y 1991 A unique mutation in the vitamin D receptor gene in three Japanese patients with vitamin Ddependent rickets type II: utility of single-strand conformation polymorphism analysis for heterozygous carrier detection. Am J Hum Genet 49:668–673. Arai H, Miyamoto K, Taketani Y, Yamamoto H, Iemori Y, Morita K, Tonai T, Nishisho T, Mori S, Takeda E 1997 A vitamin D receptor gene polymorphism in the translation initiation codon: effect on protein activity and relation to bone mineral density in Japanese women. J Bone Miner Res 12:915–921. Jurutka PW, Remus LS, Whitfield GK, Thompson PD, Hsieh JC, Zitzer H, Tavakkoli P, Galligan MA, Dang HT, Haussler CA, Haussler MR 2000 The polymorphic N terminus in human vitamin D receptor isoforms influences transcriptional activity by modulating interaction with transcription factor IIB. Mol Endocrinol 14:401–420. Flanagan JL, Sunn KL, Leong GM, Fong C, Kouzmenko AP, Eisman JA, Gardiner EM 2002 Enhanced transactivation activity of vitamin D receptor B1 associated with focal nuclear accumulation and cofactor binding. J Bone Miner Res suppl 1 17:S216.
212 51. Byrne IM, Flanagan L, Tenniswood MP, Welsh J 2000 Identification of hormone-responsive promoter immediately upstream of exon 1c in the human vitamin D receptor gene. Endocrinology 141:2829–2836. 52. Lu Z, Jehan F, Zierold C, DeLuca HF 2000 Isolation and characterization of the chicken vitamin D receptor gene and its promoter. J Cell Biochem 77:92–102. 53. Gronemeyer H, Laudet V 1995 Trancription factors 3: nuclear receptors. Protein Profile 2:1173–1308. 54. Lemaire P, Vesque C, Schmitt J, Stunnenberg H, Frank R, Charnay P 1990 The serum-inducible mouse gene Krox-24 encodes a sequence-specific transcriptional activator. Mol Cell Biol 10:3456–3467. 55. Schneider-Maurouny S, Topilko P, Seitandou T, Levi G, Cohen-Tannoudji M, Pournin S, Babinet C, Charnay P 1993 Disruption of Krox-20 results in alteration of rhombomeres 3 and 5 in the developing hindbrain. Cell 75:1199–1214. 56. Topilko P, Schneider-Maunoury S, Levi G, Baron-Van Evercooren A, Chennoufi AB, Seitanidou T, Babinet C, Charnay P 1994 Knox-20 controls myelination in the peripheral nervous system. Nature 371:796–799. 57. Levi G, Topilko P, Schneider-Maunoury S, Lasagna M, Mantero S, Cancedda R, Charnay P 1996 Defective bone formationin Krox-20 mutant mice. Development 122:113–120. 58. Schule R, Umesono K, Mangelsdorf DJ, Bolado J, Pike JW, Evans RM 1990 Jun-Fos and receptors for vitamins A and D recognize a common response element in the human osteocalcin gene. Cell 61:497–504. 59. Qi X, Pramanik R, Wang J, Schultz RM, Maitra RK, Han J, DeLuca HF, Chen G 2002 The p38 and JNK pathways cooperate to trans-activate vitamin D receptor via c-Jun/AP-1 and sensitize human breast cancer cells to vitamin D3-induced growth inhibition. J Biol Chem 277:25884–25892. 60. Merriman HL, van Wijnen AJ, Hiebert S, Bidwell JP, Fey E, Lian J, Stein J, Stein GS 1995 The tissue-specific nuclear matrix protein, NMP-2, is a member of the AML/CBF/ PEBP2/runt domain transcription factor family: interactions with the osteocalcin gene promoter. Biochemistry 34: 13125–13132. 61. Krishnan AV, Feldman D 1992 Cyclic adenosine 3′,5′monophosphate up-regulates 1,25-dihydroxyvitamin D3 receptor gene expression and enhances hormone action. Mol Endocrinol 6:198–206. 62. Lazarova DL, Bordonaro M, Sartorelli AC 2001 Transcriptional regulation of the vitamin D3 receptor gene by ZEB. Cell Growth Differ 12:319–326. 63. Takagi TMH, Kondoh H, Higashi Y 1998 δEF1, a zinc finger and homeodomain transcription factor, is required for skeleton patterning in multiple lineages. Development 125:21–31. 64. Yoshizawa T, Handa Y, Uematsu Y, Takeda S, Sekine K, Yoshihara Y, Kawakami T, Arioka K, Sato H, Uchiyama Y, Masushige S, Fukamizu A, Matsumoto T, Kato S 1997 Mice lacking the vitamin D receptor exhibit impaired bone formation, uterine hypoplasia and growth retardation after weaning. Nat Gene 16:391–396. 65. Wagner KD, Wagner N, Sukhatme VP, Scholz H 2001 Activation of vitamin D receptor by the Wilms’ tumor gene product mediates apoptosis of renal cells. J Am Soc Nephrol 12:1188–1196. 66. Maurer U, Jehan F, Englert C, Hubinger G, Weidmann E, DeLuca HF, Bergmann L 2001 The Wilms’ tumor gene product (WT1) modulates the response to 1,25-dihydroxyvitamin D3 by induction of the vitamin D receptor. J Biol Chem 276: 3727–3732.
LUIS M. ESTEBAN, JOHN A. EISMAN, AND EDITH M. GARDINER
67. Yamamoto H, Miyamoto K, Li B, Taketani Y, Kitano M, Inoue Y, Morita K, Pike JW, Takeda E 1999 The caudalrelated homeodomain protein Cdx-2 regulates vitamin D receptor gene expression in the small intestine. J Bone Miner Res 14:240–247. 68. Arai H, Miyamoto KI, Yoshida M, Yamamoto H, Taketani Y, Morita K, Kubota M, Yoshida S, Ikeda M, Watabe F, Kanemasa Y, Takeda E 2001 The polymorphism in the caudalrelated homeodomain protein Cdx-2 binding element in the human vitamin D receptor gene. J Bone Miner Res 16: 1256–1264. 69. Lee TH, Pelletier J 2001 Functional characterization of WT1 binding sites within the human vitamin D receptor gene promoter. Physiol Genomics 7:187–200. 70. Correa P, Akerstrom G, Westin G 2002 Exclusive underexpression of vitamin D receptor exon 1f transcripts in tumors of primary hyperparathyroidism. Eur J Endocrinol 147:671–675. 71. Wietzke JA, Welsh J 2003 Phytoestrogen regulation of a Vitamin D3 receptor promoter and 1,25-dihydroxyvitamin D3 actions in human breast cancer cells. J Steroid Biochem Mol Biol 84:149–157. 72. Duan C, Clemmons DR 1995 Transcription factor AP-2 regulates human insulin-like growth factor binding protein-5 gene expression. J Biol Chem 270:24844–24851. 73. Johnson W, Jameson JL 1999 AP-2 (activating protein 2) and Sp1 (selective promoter factor 1) regulatory elements play distinct roles in the control of basal activity and cyclic adenosine 3′,5′-monophosphate responsiveness of the human chorionic gonadotropin-β promoter. Mol Endocrinol 13:1963–1975. 74. Whitfield GK, Remus LS, Jurutka PW, Zitzer H, Oza AK, Dang HT, Haussler CA, Galligan MA, Thatcher ML, Encinas-Dominguez C, Haussler MR 2001 Functionally relevant polymorphisms in the human nuclear vitamin D receptor gene. Mol Cell Endocrinol 177:145–159. 75. Masi L, Cimaz R, Simonini G, Bindi G, Stagi S, Gozzini A, Malentacchi C, Brandi ML, Falcini F 2002 Association of low bone mass with vitamin d receptor gene and calcitonin receptor gene polymorphisms in juvenile idiopathic arthritis. J Rheumatol 29:2225–2231. 76. Fischer PR, Thacher TD, Pettifor JM, Jorde LB, Eccleshall TR, Feldman D 2000 Vitamin D receptor polymorphisms and nutritional rickets in Nigerian children. J Bone Miner Res 15:2206–2210. 77. Ruza E, Sotillo E, Sierrasesumaga L, Azcona C, PatinoGarcia A 2003 Analysis of polymorphisms of the vitamin D receptor, estrogen receptor, and collagen Iα1 genes and their relationship with height in children with bone cancer. J Pediatr Hematol Oncol 25:780–786. 78. Vogel A, Strassburg CP, Manns MP 2002 Genetic association of vitamin D receptor polymorphisms with primary biliary cirrhosis and autoimmune hepatitis. Hepatology 35:126–131. 79. Chen WC, Chen HY, Lu HF, Hsu CD, Tsai FJ 2001 Association of the vitamin D receptor gene start codon Fok I polymorphism with calcium oxalate stone disease. BJU Int 87:168–171. 80. Chiu KC, Chuang LM, Yoon C 2001 The vitamin D receptor polymorphism in the translation initiation codon is a risk factor for insulin resistance in glucose tolerant Caucasians. BMC Med Genet 2:2. 81. Kishi S, Yang W, Boureau B, Morand S, Das S, Chen P, Cook EH, Rosner GL, Schuetz EG, Pui CH, Relling MV 2004 Effects of prednisone and genetic polymorphisms on etoposide disposition in children with acute lymphoblastic leukemia. Blood 103:67–72.
CHAPTER 12 VDR Promoter and Abundance
82. Morrison NA, Yeoman R, Kelly PJ, Eisman JA 1992 Contribution of trans-acting factor alleles to normal physiological variability: Vitamin D receptor gene polymorphisms and circulating osteocalcin. Proc Natl Acad Sci USA 89:6665–6669. 83. Faraco JH, Morrison NA, Baker A, Shine J, Frossard PM 1989 Apa I dimorphism at the human vitamin D receptor gene locus. Nucleic Acids Res 17:2150. 84. Ingles SA, Haile RW, Henderson BE, Kolonel LN, Nakaichi G, Shi CY, Yu MC, Ross RK, Coetzee GA 1997 Strength of linkage disequilibrium between two vitamin D receptor markers in five ethnic groups: implications for association studies. Cancer Epidemiol Biomarkers Prev 6:93–98. 85. Morrison NA, Qi JC, Tokita A, Kelly PJ, Crofts L, Nguyen TV, Sambrook PN, Eisman JA 1994 Prediction of bone density from vitamin D receptor alleles. Nature 367: 284–287. 86. Yamagata Z, Miyamura T, Iijima S, Asaka A, Sasaki M, Kato J, Koizumi K 1994 Vitamin D receptor gene polymorphism and bone mineral density in healthy Japanese women. Lancet 344:1027. 87. Ferrari S, Rizzoli R, Chevalley T, Slosman D, Eisman JA, Bonjour JP 1995 Vitamin-D-receptor-gene polymorphisms and change in lumbar-spine bone mineral density. Lancet 345:423–424. 88. Spector TD, Keen RW, Arden NK, Morrison NA, Major PJ, Nguyen TV, Kelly PJ, Baker JR, Sambrook PN, Lanchbury JS, Eisman JA 1995 Influence of vitamin D receptor genotype on bone mineral density in postmenopausal women: a twin study in Britain. Bri Med J 310:1357–1360. 89. Krall EA, Parry P, Lichter JB, Dawson-Hughes B 1995 Vitamin D receptor alleles and rates of bone loss: influences of years since menopause and calcium intake. J Bone Miner Res 10:978–984. 90. Grundberg E, Brandstrom H, Ribom EL, Ljunggren O, Kindmark A, Mallmin H 2003 A poly adenosine repeat in the human vitamin D receptor gene is associated with bone mineral density in young Swedish women. Calcif Tissue Int (in press). 91. Hustmyer FG, Peacock M, Hui S, Johnston CC, Christian J 1994 Bone mineral density in relation to polymorphism at the vitamin D receptor gene locus. J Clin Invest 94: 2130–2134. 92. Melhus H, Kindmark A, Amer S, Wilen B, Lindh E, Ljunghall S 1994 Vitamin D receptor genotypes in osteoporosis. Lancet 344:949. 93. Carling T, Rastad J, Akerstrom G, Westin G 1998 Vitamin D receptor (VDR) and parathyroid hormone messenger ribonucleic acid levels correspond to polymorphic VDR alleles in human parathyroid tumors. J Clin Endocrinol Metab 83:2255–2259. 94. Verbeek W, Gombart AF, Shiohara M, Campbell M, Koeffler HP 1997 Vitamin D receptor: no evidence for allelespecific mRNA stability in cells which are heterozygous for the Taq I restriction enzyme polymorphism. Biochem Biophys Res Commun 238:77–80. 95. Barger-Lux MJ, Heaney RP, Hayes J, DeLuca HF, Johnson ML, Gong G 1995 Vitamin D receptor gene polymorphism, bone mass, body size, and vitamin D receptor density. Calcif Tissue Int 57:161–162. 96. Kinyamu HK, Gallagher JC, Knezetic JA, DeLuca HF, Prahl JM, Lanspa SJ 1997 Effect of vitamin D receptor genotypes on calcium absorption, duodenal vitamin D receptor concentration, and serum 1,25(OH)2D3 dihydroxyvitamin D levels in normal women. Calcif Tissue Int 60:491–495.
213 97. Mocharla H, Butch AW, Pappas AA, Flick JT, Weinstein RS, De Togni P, Jilka RL, Roberson PK, Parfitt AM, SC M 1997 Quantification of vitamin D receptor mRNA by competitive polymerase chain reaction in PBMC: lack of correspondence with common allelic variants. J Bone Miner Res 12:726–733. 98. Gross C, Musiol IM, Eccleshall TR, Malloy PJ, Feldman D 1998 Vitamin D receptor gene polymorphisms: analysis of ligand binding and hormone responsiveness in cultured skin fibroblasts. Biochem Biophys Res Commun 242:467–473. 99. Halloran BP, DeLuca HF 1981 Appearance of the intestinal cytosolic receptor for 1,25-dihydroxyvitamin D3 during neonatal development in the rat. J Biol Chem 256:7338–7342. 100. Walters MR, Rosen DM, Norman AW, Luben RA 1982 1,25-Dihydroxyvitamin D receptors in an established bone cell line. Correlation with biochemical responses. J Biol Chem 257:7481–7484. 101. Chen TL, Li JM, Ye TV, Cone CM, Feldman D 1986 Hormonal responses to 1,25-dihydroxyvitamin D3 in cultured mouse osteoblast-like cells: modulation by changes in receptor level. J Cell Physiol 126:21–28. 102. Dokoh S, Donaldson CA, Haussler MR 1984 Influence of 1,25-dihydroxyvitamin D3 on cultured osteogenic sarcoma cells: correlation with the 1,25-dihydroxyvitamin D3 receptor. Cancer Res 44:2103–2109. 103. Costa EM, Hirst MA, Feldman D 1985 Regulation of 1,25-dihydroxyvitamin D3 receptors by vitamin D analogs in cultured mammalian cells. Endocrinology 117:2203–2210. 104. Pan LC, Price PA 1987 Ligand-dependent regulation of the 1,25-dihydroxyvitamin D3 receptor in rat osteosarcoma cells. J Biol Chem 262:4670–4675. 105. Sher E, Frampton RJ, Eisman JA 1985 Regulation of the 1,25-dihydroxyvitamin D3 receptor by 1,25-dihydroxyvitamin D3 in intact human cancer cells. Endocrinology 116:971–977. 106. Pols HAP, Schilte HP, Visser TJ, Birkenhager JC 1987 Effect of ketoconazole on metabolism and binding of 1,25dihydroxyvitamin D3 by intact rat osteogenic sarcoma cells. Biochim Biophys Acta 931:115–119. 107. Pols HAP, Birkenhager JC, Schilte JP, Visser TJ 1988 Evidence that the self-induced metabolism of 1,25-dihydroxyvitamin D3 limits the homologous up-regulation of its receptor in rat osteosarcoma cells. Biochim Biophys Acta 970:122–129. 108. Dame MC, Pierce EA, DeLuca HF 1985 Identification of the porcine intestinal 1,25-dihydroxyvitamin D3 receptor on sodium dodecyl sulfate/polyacrylamide gels by renaturation and immunoblotting. Proc Natl Acad Sci USA 82: 7825–7829. 109. Mahonen A, Maenpaa PH 1994 Steroid hormone modulation of vitamin D receptor levels in human MG-63 osteosarcoma cells. Biochem Biophys Res Commun 205:1179–1186. 110. Song LN 1996 Demonstration of vitamin D receptor expression in a human megakaryoblastic leukemia cell line: regulation of vitamin D receptor mRNA expression and responsiveness by forskolin. J Steroid Biochem Mol Biol 57:265–274. 111. Bland R, Sammons RL, Sheppard MC, Williams GR 1997 Thyroid hormone, vitamin D and retinoid receptor expression and signalling in primary cultures of rat osteoblastic and immortalised osteosarcoma cells. J Endocrinol 154:63–74. 112. Meyer J, Fullmer CS, Wasserman RH, Komm BS, Haussler MR 1992 Dietary restriction of calcium, phosphorus, and vitamin D elicits differential regulation of the mRNAs for avian intestinal calbindin-D28k and the 1,25-dihydroxyvitamin D3 receptor. J Bone Miner Res 7:441–448.
214 113. Costa EM, Feldman D 1986 Homologous up-regulation of the 1,25(OH)2 vitamin D3 receptor in rats. Biochem Biophys Res Commun 137:742–747. 114. Naveh-Many T, Marx R, Keshet E, Pike JW, Silver J 1990 Regulation of 1,25-dihydroxyvitamin D3 receptor gene expression by 1,25-dihydroxyvitamin D3 in the parathyroid in vivo. J Clin Invest 86:1968–1975. 115. Huang YC, Lee S, Stolz R, Gabrielides C, Pansini-Porta A, Bruns ME, Bruns DE, Miffin TE, Pike JW, Christakos S 1989 Effect of hormones and development on the expression of the rat 1,25-dihydroxyvitamin D3 receptor gene. Comparison with calbindin gene expression. J Biol Chem 264:17454–17461. 116. Strom M, Sandgren ME, Brown TA, DeLuca HF 1989 1,25-dihydroxyvitamin D3 up-regulates the 1,25-dihydroxyvitamin D3 receptor in vivo. Proc Natl Acad Sci USA 86:9770–9773. 117. Sandgren ME, DeLuca HF 1990 Serum calcium and vitamin D regulate 1,25-dihydroxyvitamin D receptor concentration in rat kidney in vivo. Proc Natl Acad Sci USA 87:4312–4314. 118. Zineb R, Zhor B, Odile W, Marthe RR 1998 Distinct, tissuespecific regulation of vitamin D receptor in the intestine, kidney, and skin by dietary calcium and vitamin D. Endocrinology 139:1844–1852. 119. Goff JP, Reinhardt TA, Beckman MJ, Horst RL 1990 Contrasting effects of exogenous 1,25-dihydroxyvitamin D [1,25-(OH)2D] versus endogenous 1,25-(OH)2D, induced by dietary calcium restriction, on vitamin D receptors. Endocrinology 126:1031–1035. 120. Gensure RC, Antrobus SD, Fox J, Okwueze M, Talton SY, Walters MR 1998 Homologous up-regulation of vitamin D receptors is tissue specific in the rat. J Bone Miner Res 13:454–463. 121. Mangelsdorf DJ, Pike JW, Haussler MR 1987 Avian and mammalian receptors for 1,25-dihydroxyvitamin D3: in vitro translation to characterize size and hormone-dependent regulation. Proc Natl Acad Sci USA 84:354–358. 122. Pan P, Reddy K, Lee S, Studzinski GP 1991 Differentiationrelated regulation of 1,25-dihydroxyvitamin D3 receptor mRNA in human leukaemia cells HL-60. Cell Prolif 24:159–170. 123. Costa EM, Feldman D 1987 Measurement of 1,25-dihydroxyvitamin D3 receptor turnover by dense amino acid labeling: changes during receptor up-regulation by vitamin D metabolites. Endocrinology 120:1173–1178. 124. Wiese RJ, Uhland-Smith A, Ross TK, Prahl JM, DeLuca HF 1992 Up-regulation of the vitamin D receptor in response to 1,25-dihydroxyvitamin D3 results from ligand-induced stabilization. J Biol Chem 267:20082–20086. 125. Arbour NC, Prahl JM, DeLuca HF 1993 Stabilization of the vitamin D receptor in rat osteosarcoma cells through the action of 1,25-dihydroxyvitamin D3. Mol Endocrinol 7:1307–1312. 126. Davoodi F, Brenner RV, Evans SR, Schumaker LM, Shabahang M, Nauta RJ, Buras RR 1995 Modulation of vitamin D receptor and estrogen receptor by 1,25(OH)2vitamin D3 in T-47D human breast cancer cells. J Steroid Biochem Mol Biol 54:147–153. 127. Healy KD, Zella JB, Prahl JM, DeLuca HF 2003 Regulation of the murine renal vitamin D receptor by 1,25-dihydroxyvitamin D3 and calcium. Proc Natl Acad Sci USA 100: 9733–9737. 128. Uhland-Smith A, DeLuca HF 1993 The necessity for calcium for increased renal vitamin D receptor in response to 1,25dihydroxyvitamin D. Biochim Biophys Acta 1176:321–326.
LUIS M. ESTEBAN, JOHN A. EISMAN, AND EDITH M. GARDINER
129. Beckman MJ, DeLuca HF 2002 Regulation of renal vitamin D receptor is an important determinant of 1α,25-dihydroxyvitamin D3 levels in vivo. Arch Biochem Biophys 401: 44–52. 130. Brown AJ, Zhong M, Finch J, Ritter C, Slatopolsky E 1995 The roles of calcium and 1,25-dihydroxyvitamin D3 in the regulation of vitamin D receptor expression by rat parathyroid glands. Endocrinology 136:1419–1424. 131. Zhu Y, Goff JP, Reinhardt TA, Horst RL 1997 Pregnancy and lactation increase vitamin D–dependent intestinal membrane calcium adenosine triphosphatase and calcium binding protein messenger ribonucleic acid expression. Endocrinology 139:3520–3524. 132. Sriussadaporn S, Wong MS, Pike JW, Favus MJ 1995 Tissue specificity and mechanism of vitamin D receptor up-regulation during dietary phosphorus restriction in the rat. J Bone Miner Res 10:271–280. 133. Hernandez A, Concepcion MT, Rodriguez M, Salido E, Torres A 1996 High phosphorus diet increases preproPTH mRNA independent of calcium and calcitriol in normal rats. Kidney Int 50:1872–1878. 134. Jump DB, Clarke SD, Thelen A, Liimatta M, Ren B, Badin MV 1997 Dietary fat, genes, and human health. Adv Exp Med Biol 422:167–176. 135. Leonard F, Haag M, Kruger MC 2001 Modulation of intestinal vitamin D receptor availability and calcium ATPase activity by essential fatty acids. Prostaglandins Leukot Essent Fatty Acids 64:147–150. 136. Gaschott T, Stein J 2003 Short-chain fatty acids and colon cancer cells: the vitamin D receptor—butyrate connection. Recent Results Cancer Res 164:247–257. 137. Gaschott T, Wachtershauser A, Steinhilber D, Stein J 2001 1,25-Dihydroxycholecalciferol enhances butyrate-induced p21(Waf1/Cip1) expression. Biochem Biophys Res Commun 283:80–85. 138. Costa EM, Feldman D 1987 Modulation of 1,25-dihydroxyvitamin D3 receptor binding and action by sodium butyrate in cultured pig kidney cells (LLC-PK1). J Bone Miner Res 2:151–159. 139. Maiyar AC, Norman AW 1992 Effects of sodium butyrate on 1,25-dihydroxyvitamin D3 receptor activity in primary chick kidney cells. Mol Cell Endocrinol 84:99–107. 140. Kim CH, Cheng SL, Kim GS 1999 Effects of dexamethasone on proliferation, activity, and cytokine secretion of normal human bone marrow stromal cells: possible mechanisms of glucocorticoid-induced bone loss. J Endocrinol 162: 371–379. 141. Massaro ER, Simpson RU, DeLuca HF 1983 Glucocorticoids and appearance of 1,25-dihydroxyvitamin D3 receptor in rat intestine. Am J Physiol 244:E230–E235. 142. Lee S, Szlachetka M, Christakos S 1991 Effect of glucocorticoids and 1,25-dihydroxyvitamin D3 on the developmental expression of the rat intestinal vitamin D receptor gene. Endocrinology 129:396–401. 143. Gensure RC, Fish JM, Walters MR 1993 Dexamethasone downregulates vitamin D receptors in rat kidney, unmasking a high affinity binding site. Biochem Biophys Res Commun 195:1139–1144. 144. Hirst M, Feldman D 1982 Glucocorticoids down-regulate the number of 1,25-dihydroxyvitamin D3 receptors in mouse intestine. Biochem Biophys Res Commun 105:1590–1596. 145. Wilhelm F, Norman AW 1985 Influence of triamcinolone, estradiol-17β and testosterone on 1,25-dihydroxyvitamin D3 binding performances to its chick intestinal receptor. J Steroid Biochem 23:913–918.
CHAPTER 12 VDR Promoter and Abundance
146. Manolagas SC, Anderson DC, Lumb GA 1979 Glucocorticoids regulate the concentration of 1,25-dihydroxycholecalciferol receptors in bone. Nature 277:314–315. 147. Manolagas SC, Abare J, Deftos LJ 1984 Glucocorticoids increase the 1,25(OH)2D3 receptor concentration in rat osteogenic sarcoma cells. Calcif Tissue Int 36:153–157. 148. Chen TL, Cone CM, Morey-Hilton E, Feldman D 1983 1α,25-dihydroxyvitamin D3 receptors in cultured rat osteoblast-like cells. Glucocorticoid treatment increases receptor content. J Biol Chem 258:4350–4355. 149. Chen TL, Cone CM, Morey-Holton E, Feldman D 1982 Glucocorticoid regulation of 1,25(OH)2-vitamin D3 receptors in cultured mouse bone cells. J Biol Chem 257:13564–13569. 150. Godschalk M, Levy JR, Downs RW 1992 Glucocorticoids decrease vitamin D receptor number and gene expression in human osteosarcoma cells. J Bone Miner Res 7:21–27. 151. Midorikawa K, Sayama K, Shirakata Y, Hanakawa Y, Sun L, Hashimoto K 1999 Expression of vitamin D receptor in cultured human keratinocytes and fibroblasts is not altered by corticosteroids. J Dermatol Sci 21:8–12. 152. Saunders DE, Christensen C, Williams JR, Wappler NL, Lawrence WD, Malone JM, Malviya VK, Deppe G 1995 Inhibition of breast and ovarian carcinoma cell growth by 1,25-dihydroxyvitamin D3 combined with retinoic acid or dexamethasone. Anticancer Drugs 6:562–569. 153. Gallagher JC, Riggs BL, Eisman JA, Hamstra A, Arnaud SB, DeLuca HF 1979 Intestinal calcium absorption and serum vitamin D metabolites in normal subjects and osteoporotic patients. J Clin Invest 64:729–736. 154. Gallagher JC, Riggs BL, DeLuca HF 1980 Effect of estrogen on calcium absorption and serum vitamin D metabolites in postmenopausal osteoporosis. J Clin Endocrinol Metab 51:1359–1364. 155. Ten Bolscher M, Netelenbos JC, Barto R, Van Buuren LM, Van der vijgh WJ 1999 Estrogen regulation of intestinal calcium absorption in the intact and ovariectomized adult rat. J Bone Miner Res 14:1197–1202. 156. Liel Y, Shany S, Smirnoff P, Schwartz B 1999 Estrogen increases 1,25-dihydroxyvitamin D receptors expression and bioresponse in the rat duodenal mucosa. Endocrinology 140:280–285. 157. Colin EM, Van Den Bemd GJ, Van Aken M, Christakos S, De Jonge HR, DeLuca HF, Prahl JM, Birkenhager JC, Buurman CJ, Pols HA, Van Leeuwen JP 1999 Evidence for involvement of 17β-estradiol in intestinal calcium absorption independent of 1,25-dihydroxyvitamin D3 level in the Rat. J Bone Miner Res 14:57–64. 158. Schwartz B, Smirnoff P, Shany S, Y L 2000 Estrogen controls expression and bioresponse of 1,25-dihydroxyvitamin D receptors in the rat colon. Mol Cell Biochem 203:87–93. 159. Lechner D, Cross HS 2003 Phytoestrogens and 17β-estradiol influence vitamin D metabolism and receptor expression— relevance for colon cancer prevention. Recent Results Cancer Res 164:379–391. 160. Levy J, Zuili I, Yankowitz N, Shany S 1984 Induction of cytosolic receptors for 1α,25-dihydroxyvitamin D3 in the immature rat uterus by oestradiol. J Endocrinol 100:265–269. 161. Duncan WE, Glass AR, Wray HL 1991 Estrogen regulation of the nuclear 1,25-dihydroxyvitamin D3 receptor in rat liver and kidney. Endocrinology 129:2318–2324. 162. Duque G, Abdaimi KE, Macoritto M, Miller MM, Kremer R 2002 Estrogens (E2) regulate expression and response of 1,25-dihydroxyvitamin D3 receptors in bone cells: changes with aging and hormone deprivation. Biochem Biophys Res Commun 299:446–454.
215 163. Liel Y, Kraus S, Levy J, Shany S 1992 Evidence that estrogens modulate activity and increase the number of 1,25-dihydroxyvitamin D receptors in osteoblast-like cells (ROS 17/2.8). Endocrinology 130:2597–2601. 164. Ishibe M, Nojima T, Ishibashi T, Koda T, Kaneda K, Rosier RN, Puzas JE 1995 17β-estradiol increases the receptor number and modulates the action of 1,25-dihydroxyvitamin D3 in human osteosarcoma-derived osteoblast-like cells. Calcif Tissue Int 57:430–435. 165. Kinuta K, Tanaka H, Moriwake T, Aya K, Kato S, Seino Y 2000 Vitamin D is an important factor in estrogen biosynthesis of both female and male gonads. Endocrinology 141:1317–1324. 166. Allenby G, Bocquel MT, Saunders M, Kazmer S, Speck J, Rosenberger M, Lovey A, Kastner P, Grippo JF, Chambon P, Levin A 1993 Retinoic acid receptors and retinoid X receptors: interactions with endogenous retinoic acids. Proc Natl Acad Sci USA 90:30–34. 167. Yu VC, Delsert C, Andersen B, Holloway JM, Devary OV, Naar AM, Kim SY, Boutin JM, Glass CK, Rosenfeld MG 1991 RXRβ: a coregulator that enhances binding of retinoic acid, thyroid hormone, and vitamin D receptors to their cognate response elements. Cell 67:1251–1266. 168. Kliewer SA, Umesono K, Manglesdorf DJ, Evans RM 1992 Retinoid X receptor interacts with nuclear receptors in retinoic acid, thyroid hormone and vitamin D3 signalling. Nature 355:446–449. 169. Navia JM, Harris SS 1980 Vitamin A influence on calcium metabolism and calcification. Ann N Y Acad Sci 355:45–57. 170. Bikle DD, Pillai S 1993 Vitamin D, calcium, and epidermal differentiation. Endocr Rev 14:3–19. 171. Petkovich PM, Heersche JNM, Tinker DO, Jones G 1984 Retinoic acid stimulates 1,25-dihydroxyvitamin D3 binding in rat osteosarcoma cells. J Biol Chem 259:8274–8280. 172. Lee KL, Petkovich PM, Heersche JN 1988 The effects of sodium butyrate on the retinoic acid-induced changes in 1,25-dihydroxyvitamin D3 receptors in tumorigenic and nontumorigenic bone derived cell lines. Endocrinology 122:2399–2406. 173. Chen TL, Feldman D 1985 Retinoic acid modullation of 1,25(OH)2 vitamin D3 receptors and bioresponse in bone cells: species differences between rat and mouse. Biochem Biophys Res Commun 132:74–80. 174. Dore BT, Uskokovic MR, Momparler RL 1993 Interaction of retinoic acid and vitamin D3 analogs on HL-60 myeloid leukemic cells. Leuk Res 17:749–757. 175. Brown G, Bunce CM, Rowlands DC, Williams GR 1994 Alltrans retinoic acid and 1α,25-dihydroxyvitamin D3 co-operate to promote differentiation of the human promyeloid leukemia cell line HL60 to monocytes. Leukemia 8:806–815. 176. Gibson DF, Bikle DD, Harris J 1998 All-trans retinoic acid blocks the antiproliferative prodifferentiating actions of 1,25dihydroxyvitamin D3 in normal human keratinocytes. J Cell Physiol 174:1–8. 177. Mahonen A, Pirskanen A, Maenpaa PH 1991 Homologous and heterologous regulation of 1,25-dihydroxyvitamin D-3 receptor mRNA levels in human osteosarcoma cells. Biochim Biophys Acta 1088:111–118. 178. Chen TL, Feldman D 1981 Regulation of 1,25-dihydroxyvitamin D3 receptors in cultured mouse bone cells. Correlation of receptor concentration with the rate of cell division. J Biol Chem 256:5561–5566. 179. Krishnan AV, Feldman D 1991 Stimulation of 1,25-dihydroxyvitamin D3 receptor gene expression in cultured cells by serum and growth factors. J Bone Miner Res 6:1099–1107.
216 180. Klaus G, Weber L, Rodriguez J, Fernandez P, Klein T, Grulich-Henn J, Hugel U, Ritz E, Mehls P 1998 Interaction of IGF-I and 1α,25(OH)2D3 on receptor expression and growth stimulation in rat growth plate chondrocytes. Kidney Int 53:1152–1161. 181. Bruns DE, Krishnan AV, Feldman D, Gray RW, Christakos S, Hirsch GN, Bruns ME 1989 Epidermal growth factor increases intestinal calbindin-D9k and 1,25-dihydroxyvitamin D receptors in neonatal rats. Endocrinology 125:478–485. 182. van Leeuwen JPTM, Pols HAP, Schilte JP, Visser TJ, Birkenhager JC 1991 Modulation by epidermal growth factor of the basal 1,25(OH)2D3 receptor level and the heterologous up-regulation of the 1,25(OH)2D3 receptor in clonal osteoblast-like cells. Calcif Tissue Int 49:35–42. 183. Krishnan AV, Feldman D 1991 Activation of protein kinase-C inhibits vitamin D receptor gene expression. Mol Endocrinol 5:605–612. 184. Escaleira MT, Brentani MM 1999 Vitamin D3 receptor (VDR) expression in HC-11 mammary cells: regulation by growthmodulatory agents, differentiation, and Ha-ras transformation. Breast Cancer Res Treat 54:123–133. 185. Zhao X, Feldman D 1993 Regulation of vitamin D receptor abundance and responsiveness during differentiation of HT-29 human colon cancer cells. Endocrinology 132:1808–1814. 186. Hewison M, Freeman L, Hughes SV, Evans KN, Bland R, Eliopoulos AG, Kilby MD, Moss PAH, Chakraverty R 2003 Differential regulation of vitamin D receptor and its ligand in human monocyte-derived dendritic cells. J Immunol 170: 5382–5390. 187. Segaert S, Degreef H, Bouillon R 2000 Vitamin D receptor expression is linked to cell cycle control in normal human keratinocytes. Biochem Biophys Res Commun 279:89–94. 188. Segaert S, Courtois S, Garmyn M, Degreef H, Bouillon R 2000 The flavonoid apigenin suppresses vitamin D receptor expression and vitamin D responsiveness in normal human keratinocytes. Biochem Biophys Res Commun 268:237–241. 189. Gallagher JC 1990 The pathogenesis of osteoporosis. Bone Miner 9:215–227. 190. Denis I, Thomasset M, Pointillart A 1994 Influence of exogenous porcine growth hormone on vitamin D metabolism and calcium and phosphorus absorption in intact pigs. Calcif Tissue Int 54:489–492. 191. Fleet JC, Bruns ME, Hock JM, Wood RJ 1994 Growth hormone and parathyroid hormone stimulate intestinal calcium absorption in aged female rats. Endocrinology 134: 1755–1760. 192. Chen C, Noland KA, Kalu DN 1997 Modulation of intestinal vitamin D receptor by ovariectomy, estrogen and growth hormone. Mech Ageing Dev 99:109–122. 193. Chen C, Kalu DN 1998 Modulation of intestinal estrogen receptor by ovariectomy, estrogen and growth hormone. J Pharmacol Exp Ther 286:328–333. 194. Goff JP, Caperna TJ, Steele NC 1990 Effects of growth hormone administration on vitamin D metabolism and vitamin D receptors in the pig. Domest Anim Endocrinol 7:425–433. 195. Garabedian M, Holick MF, DeLuca HF, Boyle IT 1972 Control of 25-hydroxycholecalciferol metabolism by parathyroid glands. Proc Natl Acad Sci USA 69:1673–1676. 196. Pols HAP, van Leeuwen JP, Schilte JP, Visser TJ, Birkenhager JC 1988 Heterologous up-regulation of the 1,25-dihydroxyvitamin D3 receptor by parathyroid hormone (PTH) and PTH-like peptide in osteoblast-like cells. Biochem Biophys Res Commun 156:588–594. 197. Chakravarthy BR, Durkin JP, Rixon RH, Whitfield JF 1990 Parathyroid hormone fragment [3-34] stimulates protein kinase C (PKC) activity in rat osteosarcoma and murine
LUIS M. ESTEBAN, JOHN A. EISMAN, AND EDITH M. GARDINER
198.
199.
200.
201.
202.
203.
204.
205.
206.
207.
208.
209.
210. 211.
T-lymphoma cells. Biochem Biophys Res Commun 171: 1105–1110. Krishnan AV, Cramer SD, Bringhurst FR, Feldman D 1995 Regulation of 1,25-dihydroxyvitamin D3 receptors by parathyroid hormone in osteoblastic cells: role of second messenger pathways. Endocrinology 136:705–712. van Leeuwen JPTM, Birkenhager JC, Vink-van Wijngaarden T, van den Bemd GJCM, Pols HAP 1992 Regulation of 1,25dihydroxyvitamin D3 receptor gene expression by parathyroid hormone and cAMP-agonists. Biochem Biophys Res Commun 185:881–886. Reinhardt TA, Horst RL 1990 Parathyroid hormone downregulates 1,25-dihydroxyvitamin D receptors (VDR) and VDR messenger ribonucleic acid in vitro and blocks homologous up-regulation of VDR in vivo. Endocrinology 127:942–948. Sriussadaporn S, Wong MS, Whitfield JF, Tembe V, Favus MJ 1995 Structure–function relationship of human parathyroid hormone in the regulation of vitamin D receptor expression in osteoblast-like cells (ROS 17/2.8). Endocrinology 136: 3735–3742. Klaus G, von Eichel B, May T, Hugel U, Mayer H, Ritz E, Mehls O 1994 Synergistic effects of parathyroid hormone and 1,25-dihydroxyvitamin D3 on proliferation and vitamin D receptor expression of rat growth cartilage cells. Endocrinology 135:1307–1315. Smith SJ, Green LM, Hayes ME, Mawer EB 1999 Prostaglandin E2 regulates vitamin D receptor expression, vitamin D-24-hydroxylase activity and cell proliferation in an adherent human myeloid leukemia cell line (Ad-HL60). Prostaglandins Other Lipid Mediat 57:73–85. van Leeuwen JPTM, Birkenhager JC, Schilte JP, Buurman CJ, Pols HAP 1990 Role of calcium and cAMP in heterologous up-regulation of the 1,25-dihydroxyvitamin D3 receptor in an osteoblast cell line. Cell Calcium 11:281–289. van Leeuwen JPTM, Birkenhager JC, Buurman CJ, van den Bemd GJCM, Bos MP, Pols HAP 1992 Bidirectional regulation of the 1,25-dihydroxyvitamin D3 receptor by phorbol esteractivated protein kinase-C in osteoblast-like cells: interaction with adenosine 3′,5′-monophosphate-induced up-regulation of the 1,25-dihydroxyvitamin D3 receptor. Endocrinology 130:2259–2266. Reinhardt TA, Horst RL 1994 Phorbol 12-myristate 13-acetate and 1,25-dihydroxyvitamin D3 regulate 1,25-dihydroxyvitamin D3 receptors synergistically in rat osteosarcoma cells. Mol Cell Endocrinol 101:159–165. Chen ML, Boltz MA, Armbrecht HJ 1993 Effects of 1,25dihydroxyvitamin D3 and phorbol ester on 25-hydroxyvitamin D3 24-hydroxylase cytochrome P450 messenger ribonucleic acid levels in primary cultures of rat renal cells. Endocrinology 132:1782–1788. Armbrecht HJ, Hodam TL, Boltz MA, Chen ML 1993 Phorbol ester markedly increases the sensitivity of intestinal epithelial cells to 1,25-dihydroxyvitamin D3. FEBS Lett 327:13–16. Chang PL, Prince CW 1993 1α,25-Dihydroxyvitamin D3 enhances 12-O-tetradecanoylphorbol-13-acetate-induced tumorigenic transformation and osteopontin expression in mouse JB6 epidermal cells. Cancer Res 53:2217–2220. Martell RE, Simpson RU, Taylor JM 1987 1,25-Dihydroxyvitamin D3 regulation of phorbol ester receptors in HL-60 leukemia cells. J Biol Chem 262:5570–5575. van Leeuwen JPTM, Birkenhager JC, M vdBGJC, Buurman CJ, Staal A, Bos M, Pols HAP 1992 Evidence for the functional involvement of protein kinase C in the action of 1,25-dihydroxyvitamin D3 in bone. J Biol Chem 267:12562–12569.
CHAPTER 12 VDR Promoter and Abundance
212. Desai RK, van Wijnen AJ, Stein JL, Stein GS, Lian JB 1995 Control of 1,25-dihydroxyvitamin D3 receptor–mediated enhancement of osteocalcin gene transcription: effects of perturbing phosphorylation pathways by okadaic acid and staurosporine. Endocrinology 136:5685–5693. 213. Hara H, Yasunami Y, Adachi T 2002 Alteration of cellular phosphorylation state affects vitamin D receptor–mediated
217 CYP3A4 mRNA induction in Caco-2 cells. Biochem Biophys Res Commun 296:182–188. 214. Hsieh JC, Jurutka PW, Nakajima S, Galligan MA, Haussler CA, Shimizu Y, Shimizu N, Whitfield GK, Haussler MR 1993 Phosphorylation of the human vitamin D receptor by protein kinase C. Biochemical and functional evaluation of the serine 51 recognition site. J Biol Chem 268:15118–15126.
CHAPTER 13
Nuclear Vitamin D Receptor: Structure-Function, Molecular Control of Gene Transcription, and Novel Bioactions G. KERR WHITFIELD, PETER W. JURUTKA, CAROL A. HAUSSLER, JUI-CHENG HSIEH, THOMAS K. BARTHEL, ELIZABETH T. JACOBS,* CARLOS ENCINAS DOMINGUEZ, MICHELLE L. THATCHER, AND MARK R. HAUSSLER Department of Biochemistry and Molecular Biophysics, and *Arizona Cancer Center; College of Medicine, University of Arizona, Tucson, Arizona I. Introduction II. Gene Targets and Biological Actions of the Vitamin D Receptor III. The Vitamin D Receptor as a Member of the Nuclear Receptor Superfamily IV. Structure-Function of the Vitamin D Receptor
V. Mechanisms of Vitamin D Receptor-Mediated Control of Gene Expression VI. Implications of Vitamin D Receptor-Mediated Signaling for Human Health and Disease VII. Summary and Perspectives References
I. INTRODUCTION
found to be a 3.3–3.7S protein (depending on species) that bound 1,25(OH)2D3 with high affinity (Kd = 0.1–1 nM) [8,9] and displayed a pharmacologic profile for binding various D metabolites and analogs that is consistent with their relative biological activities [10,11]. Avian intestinal VDR was shown to be a DNA-binding protein, and this property was exploited in the first purification of VDR [12]. An enriched avian VDR preparation, containing one major and several minor proteins of molecular mass 50–70 kDa [12,13], was then utilized to generate monoclonal antibodies to the VDR that cross-reacted with all species of vitamin D receptor tested [14]. Finally, via monoclonal antibody-based screening of a λGT-11 expression vector library, a partial cDNA clone of avian VDR was obtained [15], revealing the presence of the classical (Cys2–Cys2)2 zinc finger motif of the chromatinbound steroid hormone receptors [16]. Independent biochemical evidence demonstrated the importance of reduced cysteine thiol groups for DNA association [17] and implicated zinc in maintaining the DNA binding integrity of VDR [18]. Subsequently, fulllength human [19], rat [20], and avian [21] VDRs were cloned and their sequences deduced, allowing for overexpression and functional studies of the type discussed in this chapter and elsewhere in this volume. It is notable that the cloning and classification of VDR
A. Historical Aspects and Overview of the Vitamin D Receptor The existence of the vitamin D receptor (VDR) was originally appreciated when it was observed that administration of radioactive vitamin D3 at physiologic doses to rachitic chickens elicited the selective nuclear localization of a bioactive metabolite of the parent vitamin in the target intestine [1]. The binding of this D-metabolite in the nucleus was specific, saturable, and confined to the deoxynucleoprotein chromatin subfraction [1]. Subsequent salt extraction experiments resulted in the discovery of the chromosomal protein that comprises the vitamin D receptor [2]. The vitamin D metabolite associated with VDR was shown to be more polar than 25-hydroxyvitamin D3 (25(OH)D3) [3,4] and at least five times as potent and more rapidly acting than either vitamin D3 or 25(OH)D3 [4]. These results set the stage for the chemical identification [5,6] of the VDR ligand as 1α,25-dihydroxyvitamin D3 (1,25(OH)2D3) (Fig. 1), a metabolite produced mainly in the kidney [7]. When isotopically labeled 1,25(OH)2D3 could be produced in quantity, biochemical and pharmacological characterization of VDR was accomplished. VDR was VITAMIN D, 2ND EDITION FELDMAN, PIKE, AND GLORIEUX
Copyright © 2005, Elsevier, Inc. All rights reserved.
220
G. KERR WHITFIELD
VDR
ET AL .
RXR OH
HO
OH
1α,25-Dihydroxyvitamin D3
COOH
9-cis-Retinoic acid
FIGURE 1 Chemical structures of the classic natural ligands for the protein subunits of the VDR–RXR heterodimer. 1α,25(OH)2D3 is a high-affinity ligand for VDR (Kd < 1 nM) [8,9]. The vitamin A metabolite 9-cis-retinoic acid is a natural ligand for RXR with a Kd of approximately 10 nM [26].
as a DNA-binding transcription factor activated by ligand not only validated the discovery of VDR as a chromosomal protein [2], but also verified the initial report of its uncharacterized hormonal ligand localized in nuclear chromatin [1].
a heterodimer of the 1,25(OH)2D3-liganded VDR and unoccupied RXR.
B. The Vitamin D Receptor Signals as a Binary Complex with Retinoid X Receptor
A. Vitamin D Metabolism and Actions in Target Tissues
More recent research on VDR included a key demonstration [22–24] that high-affinity and specific DNA binding by VDR requires the presence of a co-receptor, now known to be any of the three retinoid X receptor (RXR) isoforms found in human tissues [25]. Furthermore, it was demonstrated that VDR-RXR heterodimer activation of transcription occurs on a DNA responsive element in the natural context of the rat osteocalcin promoter in intact bone cells, and that 9-cis-retinoic acid (9-cis RA) (Fig. 1), one of the natural ligands of the RXR partner [26], operates, at least in the context of this promoter, to suppress VDR-mediated stimulation of transcription by 1,25(OH)2D3 [27]. Therefore, RXR is the functional partner of VDR in a binary heterocomplex that is switched positively by the vitamin D hormone and negatively by 9-cis RA or a synthetic RXR-specific ligand [27–32]. Apparently, 9-cis RA attenuates 1,25(OH)2D3 responsiveness by diverting RXRs away from VDR-mediated transcription toward RXR homodimers and other RXR-dependent transcriptional pathways [33,34] (see Section V,A). From these and other studies with different vitamin D responsive genes (reviewed by Haussler et al. [35]), it appears that the active signal transduction complex that mediates the biological effects of vitamin D is
II. GENE TARGETS AND BIOLOGICAL ACTIONS OF THE VITAMIN D RECEPTOR
An overview of the formation of 1,25(OH)2D3 and its actions in vertebrates with a mineralized skeleton is presented in Fig. 2. The hormonal precursor and parent compound, vitamin D3, can either be obtained in the diet or formed from 7-dehydrocholesterol in skin via a nonenzymatic, UV light-dependent reaction. Vitamin D3 is then transported to the liver, where it is hydroxylated at the 25 position of the side chain to produce 25(OH)D3, the major circulating form of vitamin D3. The final step in the production of the hormonal form occurs mainly in the kidney, via a tightly regulated 1α-hydroxylation reaction (Fig. 2). The cytochrome P450-containing (CYP) enzymes that catalyze 25- and 1α-hydroxylations are mitochondrial CYP27A1 and CYP27B1, respectively, as well as a newly characterized microsomal 25-hydroxylase, CYP2R1 [36]. These reactions are discussed in more detail in Chapters 4–5 of this volume. As depicted in Fig. 2, 1,25(OH)2D3 can circulate to various target tissues to exert its actions which, in the conception of the authors of this chapter, are largely mediated by the nuclear VDR and its RXR heterodimeric partner. Many of the long-recognized functions of 1,25(OH)2D3 involve the regulation of calcium and phosphate metabolism, raising the blood levels of these ions to facilitate bone mineralization. Thus, in the small
221
CHAPTER 13 Nuclear Vitamin D Receptor
1,25 VDR Skin
Macrophage
Involucrin
Liver CYP27A1/2R1
Immune system
24-hydroxylated D Metabolites
Circulating 25(OH)D3
VDR Circulating 1α,25(OH)2D3
25(OH)D3 1α,25(OH)2D3 CYP27B1 −
PTH
Diet
1,25 Carbonic VDR anhydrase
1,25
CYP24
Differentiation
Vitamin D3
Kidney
Vitamin D catabolism
Keratinocyte
Ca2+
1,25
To nuclear vitamin D receptor
PO3− 4
VDR
1,25 +
VDR
Blood Ca•PO4
Parathyroid
Absorption Other 1,25(OH)2D3VDR target systems: • Central nervous • Endocrine (thyroid, pancreas, etc.) • Reproductive tract • Muscle • Some tumors
Osteoclast precursor
Reabsorption
Ca2+
1,25 VDR
Mineralization of collagen
PO3− 4 Small intestine
Bone Osteoblast
1,25 VDR
Resorption OPG decoy − 1,25 receptor RANKL VDR
+ IL-6
Osteoblast
+
RANK RANKL
Low Ca2+
PTH Synthesis − 1,25 VDR
c-fms
M-C
SF
Osteocalcin Osteopontin + Cell differentiation and fusion
PO3− 4 Ca2+ Osteoclast
FIGURE 2 Vitamin D metabolism and its biologic actions via the nuclear VDR. Depicted at the upper left are the sources of vitamin D3, which are either dietary or via a UV light-mediated photolysis of 7-dehydrocholesterol in the epidermis (see Chapter 3). The central bar (shaded) depicts the enzymatic conversion of vitamin D3 to the 25-hydroxy and then to the 1α,25-dihydroxy derivative (1α,25(OH)2D3), which is the active, hormonal form. These and subsequent hydroxylation reactions are catalyzed by CYPs (see text). The 1,25(OH)2D3 hormonal form (shown as 1,25 inside a white semicircle) then circulates to the various target tissues (selected target tissues identified in italics), where it binds to the nuclear VDR (depicted in white letters inside a black semicircle). Examples of target genes that are either transcriptionally activated (no symbol shown) or repressed (– symbol in a circle) are listed inside boxes.
intestine, 1,25(OH)2D3 binds to VDR, eliciting regulation of genes involved in calcium and phosphate absorption. In the kidney, 1,25(OH)2D3 conserves blood calcium and phosphate by promoting their reabsorption from the glomerular filtrate. The parathyroid gland also expresses VDR, and when the receptor is liganded with 1,25(OH)2D3, parathyroid hormone (PTH) synthesis is suppressed by a direct action on gene transcription [37]. This negative feedback loop, which limits the stimulation of CYP27B1 by PTH under low calcium conditions (Fig. 2), serves to curb the bone-resorbing effects of PTH in anticipation of 1,25(OH)2D3-mediated increases in both intestinal calcium absorption and bone resorption, thus preventing hypercalcemia. The direct boneresorbing effects of 1,25(OH)2D3 are mediated first in osteoblasts, where the liganded VDR promotes transcription of the gene encoding receptor activator of NF-κB ligand (RANKL) [38]. RANKL, displayed on the surface of osteoblasts, is detected by a cognate receptor
(RANK) on osteoclast precursor cells to promote their differentiation into bone-resorbing osteoclasts. This process also requires the liganding of c-fms, an osteoclast precursor surface receptor for macrophage colony stimulating factor (M-CSF). As illustrated in Fig. 2, 1,25(OH)2D3-triggered osteoclastogenesis can be inhibited by osteoprotegerin (OPG) in its capacity as a decoy receptor for RANKL [39]. Other aspects of bone biology regulated by 1,25(OH)2D3 shown in Fig. 2 include the transcriptional activation in osteoblasts of the genes encoding osteocalcin [40] and osteopontin [41], two proteins involved in bone remodeling, as well as the paracrine activation of existing osteoclasts via induction of osteoblast-derived cytokines such as IL-6 [42]. In addition to bone mineral homeostasis, 1,25(OH)2D3 also has effects on differentiation of certain cell types in skin [43] and in the immune system (Fig. 2) [44]. Examples of genes regulated in these tissues are involucrin in keratinocytes [45] and carbonic anhydrase
222 in macrophages [46]. Interestingly, the skin and the immune system are now recognized as extrarenal sites of CYP27B1 action to produce 1,25(OH)2D3 locally for autocrine and paracrine effects [47,48] (not shown in Fig. 2; see also Chapter 5). As depicted in Fig. 2 using the kidney as an example, an important mechanism by which the 1,25(OH)2D3VDR-mediated signal is terminated in all target tissues is the catalytic action of CYP24, an enzyme that initiates the process of 1,25(OH)2D3 catabolism (see Chapters 6 and 7). The CYP24 gene is transcriptionally activated by 1,25(OH)2D3 [49,50]. In addition, the 1α-hydroxylase (1α-OHase) CYP27B1 gene appears to be repressed by 1,25(OH)2D3 in a short negative feedback loop to limit the production of 1,25(OH)2D3 [51]. Sections III and IV of this volume contain a more extensive discussion of 1,25(OH)2D3 target tissues, which, as indicated in the box at the lower left of Fig. 2, include numerous cell types in addition to small intestine, bone, kidney, parathyroid, skin, and the immune system. Finally, insights into novel roles for the VDRRXR heterodimer in the colon and the hair follicle are discussed in Section VI of this chapter.
B. VDR-Regulated Genes and Vitamin D-Responsive Elements The vitamin D target cells/tissues discussed above all express significant levels of VDR and one or more of its RXR heteropartners. After liganding with 1,25(OH)2D3, the next step in the bioactivity of VDRRXR is the recognition of DNA sequences in the genes that are transcriptionally regulated. Table I displays the sequences of vitamin D-responsive elements (VDREs), which are binding sites for the VDR-RXR heterodimer in the 5′ upstream regions of a variety of vitamin D target genes. Many of the VDREs listed in Table I consist of a direct hexanucleotide repeat with a spacer of three nucleotides (DR3); such VDREs can mediate either positive (upper panel) or negative (lower panel) transcriptional regulation of target genes. In positive DR3 VDREs, VDR has been shown to occupy the 3′ halfelement, with RXR residing on the 5′ half-site [28]. Below the positive DR3 VDRE sequences in Table I is shown an “optimal” VDRE that was experimentally determined via binding of randomized oligonucleotides to a VDR-RXR heterodimer [52,53]. The conclusion from random selection is in general agreement with the repertoire of natural VDREs and defines the optimal VDRE as a direct repeat of two six-base half-elements that resemble estrogen responsive element (ERE) halfsites, i.e., AGGTCA, separated by a spacer of three nucleotides, denoted as a DR3. The highest affinity
G. KERR WHITFIELD
ET AL .
3′ (VDR) half-site is PGTTCA, where P is a purine base, and the highest affinity 5′ (RXR) half-site is PGGTCA. In addition, the random selection results suggest that the guanine at position 3 of the spacer (shown in lowercase in Table I) is important for VDR binding, an observation that is consistent with the finding [23,54] that this base is partially protected by RXR-VDR in methylation interference assays. In addition, a positive VDRE from the proximal region of the human CYP3A4 gene (central panel, Table I) represents an unusual configuration in that the two hexanucleotide half-element repeats are everted relative to one another, with a spacer of six base pairs (ER6) [55,56]. However, an examination of the 5′ halfsite complement, namely gAGTTCA, reveals that it is a perfect match to the 3′ VDR binding half-site in the optimal DR3 VDRE, including the first base pair of the spacer. Likewise, the 3′ half-site of the ER6 VDRE represents a perfect match to the 5′ RXR binding site in the optimal DR3 VDRE. Whether the ER6-bound VDR and RXR can interact with each other in the same fashion as on a DR3 element, and, indeed, whether VDR and RXR occupy the expected positions in the ER6, are questions for future research. Nevertheless, it should be emphasized that the use of an ER6 element by VDR has precedent with other receptors. In fact, this same element in the CYP3A4 gene has been shown to function as a responsive element for the pregnane X receptor (PXR), which also binds DR3 elements [57–59]. ER6-type elements also bind thyroid hormone receptor (TR)-RXR heterodimers [60,61], for which the normal responsive elements in other genes are direct repeats with a spacer of four nucleotides (DR4s) [60]. In contrast to the unique ER6 element, in which both half-elements coincide with the randomly selected sequences, the half-sites that exist in natural DR3 elements usually contain one to three bases that do not match the optimal VDRE. These variant bases occur with approximately equal distribution between the two half-sites. The variant bases, shown with single underline in Table I and tabulated below the random selection DR3, are usually purines. One notable exception is a thymidine in the third position of the 5′ (RXR) half-site, actually causing this site to match the optimal 3′ (VDR) half-site. The multiple sequence variations in natural VDREs may provide a spectrum of affinities for the VDR-RXR heterodimer, thus enabling these elements to respond to differing concentrations of the receptors (or their ligands) [62]. Another possibility, for which increasing evidence is accumulating, is that variant VDRE sequences induce unique conformations in the VDR-RXR complex, thereby promoting association of the heterodimer with distinct subsets of coactivators [63],
223
CHAPTER 13 Nuclear Vitamin D Receptor
TABLE I DNA Sequences of Selected Natural VDREs from Vitamin D-Regulated Genes Tissue/cell
Type of VDRE
5′-Half
Spacer
3′-Half
Reference
Intestine Intestine Keratinocyte Kidney Many targets Many targets Myelomonocytic line U937 Osteoblast Osteoblast Osteoblast Osteoclast precursor, macrophage Osteoclast precursor, macrophage
Positive DR3 (gene) human CYP3A4 distal rat CYP3A23 human involucrin human NPT2 rat CYP24 proximal rat CYP24 distal human p21 human osteocalcin rat osteocalcin mouse osteopontin chicken carbonic anhydrase chicken integrin β3
GGGTCA AGTTCA GGCAGA GGGGCA AGGTGA GGTTCA AGGGAG GGGTGA GGGTGA GGTTCA AGGGCA GAGGCA
gca tga tct gca gtg gcg att acg atg cga tgg gaa
AGTTCA AGTTCA GGCAGA AGGGCA AGGGCG GGTGCG GGTTCA GGGGCA AGGACA GGTTCA AGTTCG GGGAGA
[56,98,108] [56,316] [45] [317] [49] [50] [88] [318] [40,319] [41] [46] [320]
Random selection Common deviations Uncommon deviations
Optimal DR3 in positive DR3s in positive DR3s
PGGTCA –– TGG– –ACAAG
nng nna nnt
PGTTCA ––GG–G ––CAG–
[52,53]
Intestine
Positive ER6 (gene) human CYP3A4 proximal
TGAACT caaagg AGGTCA
[55,56]
Parathyroid Parathyroid Osteoblast Osteoblast Kidney Tumors, other targets Tumors, other targets (Pre-)osteoblast
Negative DR3 (gene) human PTH chicken PTH mouse osteocalcin rat bone sialoprotein PKA inhibitor rat PTHrP proximal rat PTHrP distal rat Cbfa1
GGTTCA GGGTCA GGGCA A AGGGTT ATGTTG AGGTTA GGTGGA AGTACT
[37] [321] [322] [323] [324] [301] [302] [325]
aag gga atg tat ctg ctc gag gtg
CAGACA GGGTGT AGGACA AGGTCA AGGTCA AGTGAA GGGTGA AGGTCA
Each VDRE consists of 5′ and 3′ half-elements (column 3) that bind RXR-VDR and mediate transcriptional regulation of genes (column 2) in target cells (column 1). Bases in positive VDREs that differ from the optimal sequence obtained by random selection experiments (center row) are single underlined, unless they are unique to negative VDREs, in which case they are double underlined. P denotes a purine nucleotide, i.e., A or G; n is any base.
or permitting differential actions in the context of diverse tissues [62]. When the DR3 sequences that mediate negative transcriptional regulation are examined, it is apparent that these, too, display a considerable degree of deviation from the optimal VDRE sequence. Not surprisingly, the base differences occurring in negative VDREs do not correspond to those found in positive VDREs. Those variant bases that exist only in negative VDREs are double underlined in Table I and often include pyrimidines, in addition to the predominant purine substitutions that occur in both positive and negative
VDREs (single underlined). Exactly how the uniquely negative acting bases influence the VDR-RXR heterodimer to mediate transcriptional repression is not known. However, Koszewski and colleagues [64] were able to demonstrate that the negative VDRE in the chicken PTH gene can be converted into a positive element simply by changing the most 3′ two bases from GT to CA. The success of this experiment implies that the VDRE sequence alone can determine whether the bound VDR-RXR heterodimer will activate or repress transcription. These results can also be interpreted as further evidence that the VDRE sequence can induce
224 significant conformational changes in the VDR-RXR heterodimer that are sufficient to promote binding to corepressors, rather than coactivators. Another novel aspect of negative VDREs is that the variant bases are not equally distributed between halfsites (Table I). Thus, PTH genes have unique negative variations exclusively in the 3′ element, while other negatively regulated genes have variations primarily in the 5′ half-element. The significance of this asymmetry for negative regulation, or even possibly the polarity of VDR and RXR binding to the two half-sites, is not currently known. Clues may be found by examining the all-trans-retinoic acid receptor (RAR)-RXR heterocomplex, which acquires ligand-dependent repressor activity when associated in reverse polarity (RXR on the 3′ half-site) with a direct repeat with a spacer of one nucleotide (DR1) retinoic acid responsive element (RARE) [65]. It should be noted that one negative VDRE in Table I, namely the distal VDRE in the rat parathyroid hormonerelated peptide (PTHrP) gene, has no base difference that would identify it as a negative element, i.e., all variations observed in this VDRE are also evident in positive VDREs. However, this same gene possesses a proximal VDRE that contains three base variants occurring only in negative VDREs (Table I). This observation highlights another area that is not well understood with regard to VDREs, namely the means by which multiple VDREs in the same gene function together to effect transcriptional regulation. The fact that three 1,25(OH)2D3-regulated genes (CYP24, CYP3A4, and PTHrP) with multiple VDREs have been characterized to date suggests that probing the functional interaction of two or more VDREs in the same gene is an important area for further study. The conclusion, from analyzing a number of VDREs, is that the docking sequence for VDR-RXR on DNA consists of clearly defined DR3 and ER6 motifs in the promoter region of vitamin D controlled genes. Regulation of the expression of many of these genes by 1,25(OH)2D3 in traditional vitamin D target tissues such as kidney, bone, parathyroid, skin and the hematopoietic system has been well established [66], and the combination of Fig. 2 and Table I paints a picture of the biological breadth of VDR-mediated gene control. Obviously, the 1,25(OH)2D3-VDR/RXR complex modulates the expression of a multitude of genes, and recent microarray studies [67–71] suggest that the number of vitamin D-regulated genes is much greater than the well-characterized bioresponses discussed herein. As described next, the use of VDR null animals has both validated the obligatory role of VDR in 1,25(OH)2D3 signaling, and revealed which VDR responses are the most pathophysiologically relevant.
G. KERR WHITFIELD
ET AL .
C. Consequences of VDR Ablation in Rodents and Humans Important insight into the biological impact of VDR to mediate the myriad of gene regulatory effects of 1,25(OH)2D3 discussed earlier can be gained by examining the phenotype of mice in which the VDR has been ablated (VDR knockouts), as well as the phenotype of human patients with hereditary hypocalcemic vitamin D-resistant rickets (HVDRR). HVDRR is a consequence of defects in the VDR gene on human chromosome 12 [72–75]. Clinically significant HVDRR is an autosomal recessive disorder resulting in severe bowing of the lower extremities, short stature, and often alopecia [74]. The serum chemistry in HVDRR reveals hypocalcemia, secondary hyperparathyroidism, elevated alkaline phosphatase, variable hypophosphatemia, and markedly increased 1,25(OH)2D3. Many of these sequelae are the expected manifestations of the loss of 1,25(OH)2D3 regulation in the target genes outlined above, such as PTH, RANKL, and CYP24. Most aspects of HVDRR, with the exception of alopecia, mimic those of classic vitamin D-deficiency rickets, suggesting that VDR not only mediates all of the bone mineral homeostatic actions of vitamin D, but may also participate in hair follicle function in a manner independent of vitamin D status. As detailed later in this chapter, individual natural VDR mutations observed in HVDRR have been very instructive in the elucidation of structure/ function relationships within the receptor, especially with regard to DNA, ligand, and coactivator binding. Several strains of homozygous VDR knockout mice have been created [76–78], and their phenotype has been characterized under various conditions. Somewhat surprisingly, VDR−/− mice, except for hair loss, appeared phenotypically normal until weaning, after which they began to display the suite of symptoms characteristic of HVDRR, namely hypocalcemia, hyperparathyroidism, and elevated 1,25(OH)2D3 levels [76,77]. One early observation with VDR knockout animals was that most aspects of the VDR−/− phenotype, except for alopecia, could be reversed if blood calcium levels were normalized by dietary manipulations [79]. The so-called “rescue diet” contained calcium, phosphorus, and lactose supplements to restore crucial bone mineral ions not properly assimilated after weaning [80]. The successful rescue, in relation to bone mineral homeostasis, of VDR-null mice suggests that the dominant biological action of 1,25(OH)2D3-VDR is to promote intestinal calcium and phosphate absorption, especially when there is physiologic stress from the requirement for bone mineralization under conditions of low calcium availability. With respect to alopecia, it was observed that lack of VDR prevented hair follicles from entering the anagen phase of the hair cycle [81], but transgenic
CHAPTER 13 Nuclear Vitamin D Receptor
VDR expression targeted to keratinocytes in the background of a VDR-null animal was successful in restoring hair cycling [82]. These results not only confirmed the essential role of VDR in maintaining hair (and skin) health, but also pinpointed keratinocytes as an important cellular target for VDR action in skin. More recent studies using VDR−/− animals fed the rescue diet to normalize calcium have revealed additional roles of VDR that are not simply secondary effects of maintaining serum calcium and phosphate levels. Separate studies have found that VDR−/− mice, when compared to normal mice: (a) exhibit uterine hypoplasia [77], probably linked to reduced expression of aromatase (CYP19) [83]; (b) display hyperplasia of the mammary glands [84]; (c) have blunted insulin synthesis in pancreatic beta cells [85]; (d) have reduced synthesis of macrophage IL-18 and STAT4 in T helper subset type 1 (Th1) lymphocytes, leading to an impaired Th1 response [86]; and (e) are markedly more sensitive to the skin carcinogen 7,12-dimethylbenzanthracene [87]. These exciting recent results indicate that beyond bone mineral integrity and hair cycling, VDR signaling significantly affects development in certain tissues as well as the function of the endocrine pancreas. VDR action is clearly immunomodulatory, and VDR likely mediates the reported anticancer effects of vitamin D. In the context of chemoprevention, 1,25(OH)2D3 is known to possess potent antiproliferative, prodifferentiation, proapoptotic, and/or cell cycle arrest activities in epithelial cells [56]. The molecular mechanisms underlying these effects may be related to the ability of liganded VDR to arrest cells at the G1 stage via induction of cell cycle regulatory proteins such as p21 [88] (Table I) and p27 [89], to control cell growth transcription factors such as c-myc [90] and c-fos [91], or to elicit apoptosis by repression of the Bcl-2 antiapoptotic factor [92]. However, an additional intriguing possibility is that VDR is chemopreventative by inducing detoxification enzymes, e.g., CYP3A, since several of the genes encoding these CYPs in mammals possess VDREs (Table I). Thus, as discussed in depth later in this chapter, a newly recognized action of liganded VDR is that of inducing CYPs for xenobiotic detoxification.
III. THE VITAMIN D RECEPTOR AS A MEMBER OF THE NUCLEAR RECEPTOR SUPERFAMILY A. VDR in the Context of the 48 Human Nuclear Receptors A proper understanding of the molecular actions of VDR requires an appreciation of its genetic placement within the nuclear receptor superfamily. The human
225 genome sequence [93] encodes 48 nuclear receptors [94], and interrelationships between these receptors are illustrated in Fig. 3. All human nuclear receptors, along with those from other species, have been placed into six groups according to a unified system of nomenclature [95]. As with previous analyses using a subset of human receptors [95], VDR groups with the TRs, RARs, the peroxisome proliferator-activated receptors (PPARs), the reverb orphan receptors, and the retinoic acid receptor-related orphan receptors (RORs) in a single clade that has been named group 1 [95]. As illustrated on the right side of Fig. 3, a significant feature shared by group 1 receptors is that many of them form active heterodimeric complexes with the RXRs from group 2 [96]. The closest relatives of VDR within group 1 are PXR and the constitutive androstane receptor (CAR), which, with VDR, form subgroup 1I (Fig. 3) (full designations are NR1I1 for VDR, NR1I2 for PXR, and NR1I3 for CAR) [95]. The next closest group of receptors to VDR is subgroup 1H, including the liver X receptors (LXRs) and the farnesoid X receptor (FXR). One notable feature of the 1H and 1I receptors is the ability of these receptors to recognize multiple ligands [97]. The diversity of ligands for PXR is especially broad and includes not only endogenous steroids, but also an array of other lipophilic compounds such as the secondary bile acid lithocholic acid (LCA) and the antibiotic rifampicin, as well as xenobiotics such as hyperforin, the active ingredient of St. John’s wort [97]. Until recently, VDR was thought to possess a single ligand, the 1,25(OH)2D3 hormone. However, it is now known that mammalian VDR has at least two additional ligands (Fig. 4), namely LCA and its derivative, 3-ketolithocholic acid (3-ketoLCA) [98]. This finding is of considerable interest to human medicine, since it is known that LCA is a secondary bile acid with significant carcinogenic potential [98–100], and its binding to VDR could trigger the detoxification of this ligand via the induction of CYP3A4 mediated by the VDREs in this gene (Table I). The structures of both LCA and 3-ketoLCA are shown in Fig. 4, along with the structures for prototypical PXR, FXR, and LXR ligands. It is evident when comparing these compounds that the ligand binding profile of VDR, expanded to include LCA and its 3-keto derivative, now resembles the ligand profile of the closest VDR relatives in the nuclear receptor superfamily, especially when it is noted that both PXR [101,102] and FXR [98,103] are also activated by LCA to some extent. It is now appreciated that most group 1I and 1H members, with the exception of CAR and the two LXRs, can recognize bile acid ligands. Therefore, based upon an analysis of its evolutionary position among the nuclear receptors and recent data
226
G. KERR WHITFIELD
TRα 1A TRβ RARα RARβ 1B RARγ PPARα PPARβ 1C PPARγ reverbα 1D reverbβ RORα RORβ 1F RORγ VDR PXR 1I CAR LXRβ LXRα 1H FXR HNF4 2A HNF4γ RXRα 2B RXRβ RXRγ TR2 2C TR4 TLX 2E PNR COUP COUP2 2F EAR2 DAX 0B SHP GCNF1 6A ERα 3A ERβ ERRα ERRβ 3B ERRγ GR MR 3C PR AR FTZ1 5A LRH1 NURR1 4A NOR1 NGFIB Outgroup (Cnidarian)
RXR
RXR
RXR
Group 1
RXR