Laser Capture Microdissection: Methods and Protocols, 2nd Edition (Methods in Molecular Biology Vol 755)

Methods

in

Molecular Biology™

Series Editor John M. Walker School of Life Sciences University of Hertfordshire Hatfield, Hertfordshire, AL10 9AB, UK



For further volumes: http://www.springer.com/series/7651

wwwwwwwwwwwwwwwww

Laser Capture Microdissection Methods and Protocols Second Edition

Edited by

Graeme I. Murray Department of Pathology, University of Aberdeen, Aberdeen, UK

Editor Graeme I. Murray, MB ChB, PhD, DSc, FRCPath Department of Pathology University of Aberdeen Aberdeen, UK [email protected]

ISSN 1064-3745 e-ISSN 1940-6029 ISBN 978-1-61779-162-8 e-ISBN 978-1-61779-163-5 DOI 10.1007/978-1-61779-163-5 Springer New York Heidelberg London Dordrecht Library of Congress Control Number: 2011931522 © Springer Science+Business Media, LLC 2011 All rights reserved. This work may not be translated or copied in whole or in part without the written permission of the publisher (Humana Press, c/o Springer Science+Business Media, LLC, 233 Spring Street, New York, NY 10013, USA), except for brief excerpts in connection with reviews or scholarly analysis. Use in connection with any form of information storage and retrieval, electronic adaptation, computer software, or by similar or ­dissimilar methodology now known or hereafter developed is forbidden. The use in this publication of trade names, trademarks, service marks, and similar terms, even if they are not identified as such, is not to be taken as an expression of opinion as to whether or not they are subject to proprietary rights. Printed on acid-free paper Humana Press is part of Springer Science+Business Media (www.springer.com)

Preface Laser microdissection techniques have revolutionized the ability of researchers in general, and pathologists in particular, to carry out molecular analysis on specific types of normal and diseased cells and to fully utilize the power of current molecular technologies, including PCR, microarrays, and proteomics. The primary purpose of the second edition of this volume of Methods in Molecular Biology is to provide the reader with practical advice on how to carry out tissue-based laser microdissection successfully in their own laboratory using the different laser microdissection systems that are available and to apply a wide range of molecular technologies. The individual chapters encompass detailed descriptions of the individual laser-based microdissection systems. The downstream applications of the laser microdissected tissue described in the book include PCR in its many different forms as well as gene expression analysis, including the application to microarrays and proteomics. The editor is especially grateful to all the contributing authors for the time and effort they have put into the individual chapters. The series editor John Walker has provided expert guidance through the editorial process while colleagues at Springer have been very helpful in dealing with all the publication related issues. Aberdeen, UK

Graeme I. Murray

v

wwwwwwwwwwwwwwwww

Contents Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

v xi

  1 Laser Capture Microdissection: Methods and Applications . . . . . . . . . . . . . . . . . . 1 Kristen DeCarlo, Andrew Emley, Ophelia E. Dadzie, and Meera Mahalingam   2 Laser Microdissection for Gene Expression Profiling . . . . . . . . . . . . . . . . . . . . . . 17 Lori A. Field, Brenda Deyarmin, Craig D. Shriver, Darrell L. Ellsworth, and Rachel E. Ellsworth   3 Gene Expression Using the PALM System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 Jian-Xin Lu and Cheuk-Chun Szeto   4 Immunoguided Microdissection Techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 Michael A. Tangrea, Jeffrey C. Hanson, Robert F. Bonner, Thomas J. Pohida, Jaime Rodriguez-Canales, and Michael R. Emmert-Buck   5 Optimized RNA Extraction from Non-deparaffinized, Laser-Microdissected Material . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67 Danny Jonigk, Friedrich Modde, Clemens L. Bockmeyer, Jan Ulrich Becker, and Ulrich Lehmann   6 Laser Capture Microdissection for Analysis of Gene Expression in Formalin-Fixed Paraffin-Embedded Tissue . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77 Ru Jiang, Rona S. Scott, and Lindsey M. Hutt-Fletcher   7 MicroRNA Profiling Using RNA from Microdissected Immunostained Tissue . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85 Clemens L. Bockmeyer, Danny Jonigk, Hans Kreipe, and Ulrich Lehmann   8 Profiling Solid Tumor Heterogeneity by LCM and Biological MS of Fresh-Frozen Tissue Sections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95 Donald J. Johann, Sumana Mukherjee, DaRue A. Prieto, Timothy D. Veenstra, and Josip Blonder   9 Amplification Testing in Breast Cancer by Multiplex Ligation-Dependent Probe Amplification of Microdissected Tissue . . . . . . . . . . . 107 Cathy B. Moelans, Roel A. de Weger, and Paul J. van Diest 10 Detection and Quantification of MicroRNAs in Laser-Microdissected Formalin-Fixed Paraffin-Embedded Breast Cancer Tissues . . . . . . . . . . . . . . . . . . 119 Sarkawt M. Khoshnaw, Des G. Powe, Ian O. Ellis, and Andrew R. Green 11 Laser Capture Microdissection Applications in Breast Cancer Proteomics . . . . . . . 143 René B.H. Braakman, Theo M. Luider, John W.M. Martens, John A. Foekens, and Arzu Umar

vii

viii

Contents

12 Proteomic Analysis of Laser Microdissected Ovarian Cancer Tissue with SELDI-TOF MS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Isabelle Cadron, Toon Van Gorp, Philippe Moerman, Etienne Waelkens, and Ignace Vergote 13 LCM Assisted Biomarker Discovery from Archival Neoplastic Gastrointestinal Tissues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Patricia A. Meitner and Murray B. Resnick 14 Purification of Diseased Cells from Barrett’s Esophagus and Related Lesions by Laser Capture Microdissection . . . . . . . . . . . . . . . . . . . . . Masood A. Shammas and Manjula Y. Rao 15 Laser Microdissection of Intestinal Epithelial Cells and Downstream Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Benjamin Funke 16 Application of Laser Microdissection and Quantitative PCR to Assess the Response of Esophageal Cancer to Neoadjuvant Chemo-Radiotherapy . . . . . . Claus Hann von Weyhern and Björn L.D.M. Brücher 17 Oligonucleotide Microarray Expression Profiling of Contrasting Invasive Phenotypes in Colorectal Cancer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Christopher C. Thorn, Deborah Williams, and Thomas C. Freeman 18 Evaluation of Gastrointestinal mtDNA Depletion in Mitochondrial Neurogastrointestinal Encephalomyopathy (MNGIE) . . . . . . . . . . . . . . . . . . . . . Carla Giordano and Giulia d’Amati 19 Laser Microdissection for Gene Expression Study of Hepatocellular Carcinomas Arising in Cirrhotic and Non-Cirrhotic Livers . . . . . . . . . . . . . . . . . . Maria Tretiakova and John Hart 20 Laser Capture Microdissection of Pancreatic Ductal Adeno-Carcinoma Cells to Analyze EzH2 by Western Blot Analysis . . . . . . . . . . . . . . . . . . . . . . . . . Aamer M. Qazi, Sita Aggarwal, Christopher S. Steffer, David L. Bouwman, Donald W. Weaver, Scott A. Gruber, and Ramesh B. Batchu 21 Laser-Capture Microdissection of Renal Tubule Cells and Linear Amplification of RNA for Microarray Profiling and Real-Time PCR . . . . . . . . . . . Susie-Jane Noppert, Susanne Eder, and Michael Rudnicki 22 Subcellular Renal Proximal Tubular Mitochondrial Toxicity with Tenofovir Treatment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . James J. Kohler and Seyed H. Hosseini 23 Application of Laser-Capture Microdissection to Study Renal Carcinogenesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Kerstin Stemmer and Daniel R. Dietrich 24 Laser-Capture Microdissection and Transcriptional Profiling in Archival FFPE Tissue in Prostate Cancer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Ajay Joseph and Vincent J. Gnanapragasam 25 Quantitative Analysis of the Enzymes Associated with 5-Fluorouracil Metabolism in Prostate Cancer Biopsies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Tomoaki Tanaka 26 Microdissection of Gonadal Tissues for Gene Expression Analyses . . . . . . . . . . . . Anne Jørgensen, Marlene Danner Dalgaard, and Si Brask Sonne

155

165

181

189

197

203

223

233

245

257

267

279

291

301 307

Contents

27 Duplex Real-Time PCR Assay for Quantifying Mitochondrial DNA Deletions in Laser Microdissected Single Spiral Ganglion Cells . . . . . . . . . . . . . . . Adam Markaryan, Erik G. Nelson, and Raul Hinojosa 28 Neuronal Type-Specific Gene Expression Profiling and Laser-Capture Microdissection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Charmaine Y. Pietersen, Maribel P. Lim, Laurel Macey, Tsung-Ung W. Woo, and Kai C. Sonntag 29 Region-Specific In Situ Hybridization-Guided Laser-Capture Microdissection on Postmortem Human Brain Tissue Coupled with Gene Expression Quantification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . René Bernard, Sharon Burke, and Ilan A. Kerman 30 UV-Laser Microdissection and mRNA Expression Analysis of Individual Neurons from Postmortem Parkinson’s Disease Brains . . . . . . . . . . Jan Gründemann, Falk Schlaudraff, and Birgit Liss 31 Transcriptome Profiling of Murine Spinal Neurulation Using Laser Capture Microdissection and High-Density Oligonucleotide Microarrays . . . . . . . . . . . . . Shoufeng Cao, Boon-Huat Bay, and George W. Yip 32 Probing the CNS Microvascular Endothelium by Immune-Guided Laser-Capture Microdissection Coupled to Quantitative RT-PCR . . . . . . . . . . . . Nivetha Murugesan, Jennifer Macdonald, Shujun Ge, and Joel S. Pachter 33 Laser-Capture Microdissection for Factor VIII-Expressing Endothelial Cells in Cancer Tissues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Tomoatsu Kaneko, Takashi Okiji, Reika Kaneko, Hideaki Suda, and Jacques E. Nör 34 Laser-Capture Microdissection and Analysis of Liver Endothelial Cells from Patients with Budd–Chiari Syndrome . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Selcuk Sozer and Ronald Hoffman 35 Laser-Capture Microdissection of Hyperlipidemic/ApoE−/− Mouse Aorta Atherosclerosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Michael Beer, Sandra Doepping, Markus Hildner, Gabriele Weber, Rolf Grabner, Desheng Hu, Sarajo Kumar Mohanta, Prasad Srikakulapu, Falk Weih, and Andreas J.R. Habenicht 36 Gene Expression Profiling in Laser-Microdissected Bone Marrow Megakaryocytes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Kais Hussein 37 Specific RNA Collection from the Rat Endolymphatic Sac by Laser-Capture Microdissection (LCM): LCM of a Very Small Organ Surrounded by Bony Tissues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Kosuke Akiyama, Takenori Miyashita, Ai Matsubara, and Nozomu Mori 38 The Use of Laser Capture Microdissection on Adult Human Articular Cartilage for Gene Expression Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Naoshi Fukui, Yasuko Ikeda, and Nobuho Tanaka 39 Laser-Capture Microdissection of Developing Barley Seeds and cDNA Array Analysis of Selected Tissues . . . . . . . . . . . . . . . . . . . . . . . . . . . . Johannes Thiel, Diana Weier, and Winfriede Weschke

ix

315

327

345

363

375

385

395

405

417

429

441

449

461

x

Contents

40 Quantitative RT-PCR Gene Expression Analysis of a Laser Microdissected Placenta: An Approach to Study Preeclampsia . . . . . . . . 477 Yuditiya Purwosunu, Akihiko Sekizawa, Takashi Okai, and Tetsuhiko Tachikawa Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 491

Contributors Sita Aggarwal  •  Pennington Biomedical Research Center, Louisiana State University System, Baton Rouge, LA, USA Kosuke Akiyama  •  Department of Otolaryngology, Faculty of Medicine, Kagawa University, Kagawa, Japan Giulia d’Amati  •  Department of Experimental Medicine, Sapienza University, Rome, Italy Ramesh B. Batchu  •  Laboratory of Surgical Oncology & Developmental Therapeutics, Department of Surgery, Wayne State University, Detroit, MI, USA; John D Dingell VA Medical Center, Detroit, MI, USA Boon-Huat Bay  •  Department of Anatomy, Yong Loo Lin School of Medicine, National University of Singapore, Singapore Jan Ulrich Becker  •  Institute of Pathology, Medizinische Hochschule Hannover, Hannover, Germany Michael Beer  •  Institute for Vascular Medicine, Friedrich Schiller University of Jena, Jena, Germany René Bernard  •  Charité Campus Mitte – Universitätsmedizin Berlin, Centrum für Anatomie, Institut für Integrative Neuroanatomie, Berlin, Germany Josip Blonder  •  Laboratory of Proteomics and Analytical Technologies, SAIC-Frederick, Inc., National Cancer Institute at Frederick, Frederick, MD, USA Clemens L. Bockmeyer  •  Institute of Pathology, Medizinische Hochschule Hannover, Hannover, Germany Robert F. Bonner  •  Pathogenetics Unit and Laser Microdissection Core, Laboratory of Pathology, Center for Cancer Research, National Cancer Institute, National Institutes of Health, Bethesda, MD, USA David L. Bouwman  •  Department of Surgery, Wayne State University, Detroit, MI, USA René B.H. Braakman  •  Department of Medical Oncology, Center for Translational Molecular Medicine, and Cancer Genomics Centre, Erasmus MC Rotterdam, Rotterdam, The Netherlands Björn L.D.M. Brücher  •  Comprehensive Cancer Center, University of Tübingen, Tübingen, Germany Sharon Burke  •  Molecular and Behavioral Neuroscience Institute, Ann Arbor, MI, USA Isabelle Cadron  •  Division of Gynecological Oncology, Department of Obstetrics and Gynecology, University Hospitals Leuven, Katholieke Universiteit Leuven, Leuven, Belgium

xi

xii

Contributors

Shoufeng Cao  •  Department of Anatomy, Yong Loo Lin School of Medicine, National University of Singapore, Singapore Ophelia E. Dadzie  •  Dermatopathology Section, St John’s Institute of Dermatology, St. Thomas’ Hospital, London, UK Marlene Danner Dalgaard  •  Department of Growth and Reproduction, Rigshospitalet, Copenhagen, Denmark Kristen DeCarlo  •  Boston University School of Medicine, Boston, MA, USA Brenda Deyarmin  •  Windber Research Institute, Windber, PA, USA Paul J. van Diest   •  Department of Pathology, University Medical Centre Utrecht, Utrecht, The Netherlands Daniel R. Dietrich  •  Human and Environmental Toxicology, University of Konstanz, Konstanz, Germany Sandra Doepping  •  Institute for Vascular Medicine, Friedrich Schiller University of Jena, Jena, Germany Susanne Eder   •  Department of Internal Medicine IV (Nephrology and Hypertension), Functional Genomics Research Group, Center of Internal Medicine, Medical University Innsbruck, Innsbruck, Austria Ian O. Ellis  •  Department of Histopathology, School of Molecular Medical Sciences, University of Nottingham and Nottingham University Hospitals Trust, Nottingham, UK Darrell L. Ellsworth  •  Windber Research Institute, Windber, PA, USA Rachel E. Ellsworth  •  Translational Breast Research, Clinical Breast Care Project, Windber Research Institute, Windber, PA, USA Andrew Emley  •  Dermatopathology Section, Department of Dermatology, Boston University School of Medicine, Boston, MA, USA Michael R. Emmert-Buck  •  Pathogenetics Unit and Laser Microdissection Core, Laboratory of Pathology, Center for Cancer Research, Gaithersburg, MD, USA, Lori A. Field  •  Windber Research Institute, Windber, PA, USA John A. Foekens  •  Department of Medical Oncology, Center for Translational Molecular Medicine, and Cancer Genomics Centre, Erasmus MC Rotterdam, Rotterdam, The Netherlands Thomas C. Freeman  •  Roslin Institute, University of Edinburgh, Edinburgh, UK Naoshi Fukui  •  Clinical Research Center, National Hospital Organization, Sagamihara Hospital, Kanagawa, Japan Benjamin Funke  •  Institute of Pathology, University Hospital Heidelberg, Heidelberg, Germany; Department of Anaesthesiology, University Hospital Heidelberg, Heidelberg, Germany Shujun Ge  •  Blood-Brain Barrier Laboratory, Department of Cell Biology, University of Connecticut Health Center, Farmington, CT, USA Carla Giordano  •  Department of Experimental Medicine, Sapienza University, Rome, Italy Vincent J. Gnanapragasam  •  Translational Prostate Cancer Group, Hutchison MRC Research Centre, University of Cambridge, Cambridge, UK

Contributors

xiii

Rolf Grabner  •  Institute for Vascular Medicine, Friedrich Schiller University of Jena, Jena, Germany Andrew R. Green  •  Department of Histopathology, School of Molecular Medical Sciences, University of Nottingham and Nottingham University Hospitals Trust, Nottingham, UK Scott A. Gruber  •  John D Dingell VA Medical Center, Wayne State University, Detroit, MI, USA Jan Gründemann  •  Wolfson Institute for Biomedical Research, University College London, London, UK Andreas J.R. Habenicht  •  Institute for Vascular Medicine, Friedrich Schiller University of Jena, Jena, Germany Jeffrey C. Hanson  •  Pathogenetics Unit and Laser Microdissection Core, Laboratory of Pathology, Center for Cancer Research, National Cancer Institute, National Institutes of Health, Bethesda, MD, USA John Hart  •  Department of Pathology, University of Chicago, Chicago, IL, USA Markus Hildner  •  Institute for Vascular Medicine, Friedrich Schiller University of Jena, Jena, Germany Raul Hinojosa  •  Section of Otolaryngology – Head and Neck Surgery, Department of Surgery, University of Chicago, Chicago, IL, USA Ronald Hoffman  •  Tisch Cancer Institute, Department of Medicine, Mount Sinai School of Medicine, New York, NY, USA; Myeloproliferative Disorder Research Consortium, New York, NY, USA Seyed H. Hosseini  •  Science Department, Georgia Perimeter College, Clarkston, GA, USA Desheng Hu  •  Institute for Vascular Medicine, Friedrich Schiller University of Jena, Jena, Germany Kais Hussein  •  Institute of Pathology, Hannover Medical School, Hannover, Germany Lindsey M. Hutt-Fletcher  •  Department of Microbiology and Immunology, Center for Molecular and Tumor Virology and Feist-Weiller Cancer Center, Louisiana State University Health Sciences Center, Shreveport, LA, USA Yasuko Ikeda  •  Clinical Research Center, National Hospital Organization, Sagamihara Hospital, Kanagawa, Japan Ru Jiang  •  Department of Microbiology and Immunology, Center for Molecular and Tumor Virology and Feist-Weiller Cancer Center, Louisiana State University Health Sciences Center, Shreveport, LA, USA Donald J. Johann  •  Medical Oncology Branch, Center for Cancer Research, National Cancer Institute, National Institutes of Health, Bethesda, MD, USA Danny Jonigk   •  Institute of Pathology, Medizinische Hochschule Hannover, Hannover, Germany Anne Jørgensen  •  Department of Growth and Reproduction, Rigshospitalet, Copenhagen, Denmark Ajay Joseph  •  Translational Prostate Cancer Group, Hutchison MRC Research Centre, University of Cambridge, Cambridge, UK Reika Kaneko  •  Applied Molecular Medicine, Niigata University Graduate School of Medical and Dental Sciences, Chuo-Ku, Niigata, Japan

xiv

Contributors

Tomoatsu Kaneko  •  Cariology, Operative Dentistry and Endodontics, Niigata University Graduate School of Medical and Dental Sciences, Niigata, Japan Ilan A. Kerman  •  University of Alabama at Birmingham, Department of Psychiatry and Behavioral Neurobiology, Birmingham, AL, USA Sarkawt M. Khoshnaw  •  Department of Histopathology, School of Molecular Medical Sciences, University of Nottingham and Nottingham University Hospitals Trust, Nottingham, UK James J. Kohler  •  Department of Pediatrics, Laboratory of Biochemical Pharmacology, Emory University School of Medicine, Decatur, GA, USA Hans Kreipe   •  Institute of Pathology, Medizinische Hochschule Hannover, Hannover, Germany Ulrich Lehmann  •  Institute of Pathology, Medizinische Hochschule Hannover, Hannover, Germany Maribel P. Lim  •  Laboratory of Cellular Neuropathology, Department of Psychiatry, McLean Hospital, Harvard Medical School, Belmont, MA, USA Birgit Liss  •  Institute of Applied Physiology, University of Ulm, Ulm, Germany Jian-Xin Lu   •  Department of Medicine & Therapeutics, Prince of Wales Hospital, The Chinese University of Hong Kong, Shatin, Hong Kong, China Theo M. Luider  •  Department of Neurology and Laboratory of Clinical and Cancer Proteomics, Erasmus MC Rotterdam, Rotterdam, The Netherlands Jennifer Macdonald  •  Blood-Brain Barrier Laboratory, Department of Cell Biology, University of Connecticut Health Center, Farmington, CT, USA Laurel Macey  •  Department of Psychiatry, McLean Hospital, Harvard Medical School, Belmont, MA, USA Meera Mahalingam  •  Dermatopathology Section, Department of Dermatology, Boston University School of Medicine, Boston, MA, USA Adam Markaryan  •  Section of Otolaryngology – Head and Neck Surgery, Department of Surgery, University of Chicago, Chicago, IL, USA John W.M. Martens  •  Department of Medical Oncology, Center for Translational Molecular Medicine, and Cancer Genomics Centre, Erasmus MC Rotterdam, Rotterdam, The Netherlands Ai Matsubara  •  Department of Otolaryngology, Faculty of Medicine, Kagawa University, Kagawa, Japan Patricia A. Meitner  •  COBRE Center for Cancer Research Development, Rhode Island Hospital, Providence, RI, USA Takenori Miyashita  •  Department of Otolaryngology, Faculty of Medicine, Kagawa University, Kagawa, Japan Friedrich Modde  •  Institute of Pathology, Medizinische Hochschule Hannover, Hannover, Germany Cathy B. Moelans  •  Department of Pathology, University Medical Centre Utrecht, Utrecht, The Netherlands Philippe Moerman  •  Department of Pathology, University Hospitals Leuven, Katholieke Universiteit Leuven, Leuven, Belgium Sarajo Kumar Mohanta  •  Institute for Vascular Medicine, Friedrich Schiller University of Jena, Jena, Germany

Contributors

xv

Nozomu Mori  •  Department of Otolaryngology, Faculty of Medicine, Kagawa University, Kagawa, Japan Sumana Mukherjee  •  Medical Oncology Branch, Center for Cancer Research, National Cancer Institute, National Institutes of Health, Bethesda, MD, USA Nivetha Murugesan  •  Blood-Brain Barrier Laboratory, Department of Cell Biology, University of Connecticut Health Center, Farmington, CT, USA Erik G. Nelson  •  Section of Otolaryngology – Head and Neck Surgery, Department of Surgery, University of Chicago, Chicago, IL, USA Susie-Jane Noppert  •  Department of Internal Medicine IV (Nephrology and Hypertension), Functional Genomics Research Group, Center of Internal Medicine, Medical University Innsbruck, Innsbruck, Austria Jacques E. Nör  •  Cariology, Restorative Sciences, and Endodontics, School of Dentistry, University of Michigan, Ann Arbor, MI, USA; Department of Biomedical Engineering, College of Engineering, University of Michigan, Ann Arbor, MI, USA; Comprehensive Cancer Center, University of Michigan, Ann Arbor, MI, USA Takashi Okai  •  Department of Obstetrics and Gynecology, Showa University School of Medicine, Tokyo, Japan Takashi Okiji  •  Cariology, Operative Dentistry and Endodontics, Niigata University Graduate School of Medical and Dental Sciences, Niigata, Japan Joel S. Pachter  •  Blood-Brain Barrier Laboratory, Department of Cell Biology, University of Connecticut Health Center, Farmington, CT, USA Charmaine Y. Pietersen  •  Laboratory of Cellular Neuropathology, Department of Psychiatry, McLean Hospital, Harvard Medical School, Belmont, MA, USA Thomas J. Pohida  •  Pathogenetics Unit and Laser Microdissection Core, Laboratory of Pathology, Center for Cancer Research, National Cancer Institute, National Institutes of Health, Bethesda, MD, USA Des G. Powe  •  Department of Histopathology, School of Molecular Medical Sciences, University of Nottingham and Nottingham University Hospitals Trust, Nottingham, UK DaRue A. Prieto  •  Laboratory of Proteomics and Analytical Technologies, SAIC-Frederick, Inc., National Cancer Institute at Frederick, Frederick, MD, USA Yuditiya Purwosunu  •  Department of Obstetrics and Gynecology, Showa University School of Medicine, Tokyo, Japan; Department of Obstetrics and Gynecology, University of Indonesia, Cipto Mangunkusumo National Hospital, Jakarta, Indonesia Aamer M. Qazi  •  Department of Surgery, John D Dingell VA Medical Center, Wayne State University, Detroit, MI, USA Manjula Y. Rao  •  Department of Neurology, Center on Human Development and Disability, University of Washington, Seattle, WA, USA Murray B. Resnick  •  Department of Pathology, Rhode Island and The Miriam Hospital, Alpert Medical School, Brown University, Providence, RI, USA Jaime Rodriguez-Canales  •  Pathogenetics Unit and Laser Microdissection Core, Laboratory of Pathology, Center for Cancer Research, National Cancer Institute, National Institutes of Health, Bethesda, MD, USA

xvi

Contributors

Michael Rudnicki  •  Department of Internal Medicine IV (Nephrology and Hypertension), Functional Genomics Research Group, Center of Internal Medicine, Medical University Innsbruck, Innsbruck, Austria Falk Schlaudraff  •  Institute of Applied Physiology, University of Ulm, Ulm, Germany Rona S. Scott  •  Department of Microbiology and Immunology, Center for Molecular and Tumor Virology and Feist-Weiller Cancer Center, Louisiana State University Health Sciences Center, Shreveport, LA, USA Akihiko Sekizawa  •  Department of Obstetrics and Gynecology, Showa University School of Medicine, Bunkyo-ku, Tokyo, Japan Masood A. Shammas  •  Department of Medical Oncology, Harvard (Dana Farber) Cancer Institute and VA Boston Healthcare System, Boston, MA, USA Craig D. Shriver   •  Walter Reed Army Medical Center, Washington, DC, USA Si Brask Sonne  •  Department of Biology, University of Copenhagen, Copenhagen, Denmark Kai C. Sonntag  •  Department of Psychiatry, McLean Hospital, Harvard Medical School, Belmont, MA, USA Selcuk Sozer  •  Research Institute for Experimental Medicine (DETAE), Istanbul University, Istanbul, Turkey; Tisch Cancer Institute, Department of Medicine, Mount Sinai School of Medicine, New York, NY, USA Prasad Srikakulapu  •  Institute for Vascular Medicine, Friedrich Schiller University of Jena, Jena, Germany Christopher S. Steffer  •  Department of Surgery, Wayne State University, Detroit, MI, USA Kerstin Stemmer  •  Human and Environmental Toxicology, University of Konstanz, Konstanz, Germany; Department of Internal Medicine, Metabolic Diseases Institute, University of Cincinnati, Cincinnati, OH, USA Hideaki Suda  •  Pulp Biology and Endodontics, Department of Restorative Sciences, Graduate School of Medical and Dental Sciences, Tokyo Medical and Dental University, Bunkyo-ku, Tokyo, Japan Cheuk-Chun Szeto  •  Department of Medicine & Therapeutics, Prince of Wales Hospital, The Chinese University of Hong Kong, Shatin, Hong Kong, China Tetsuhiko Tachikawa  •  Department of Oral Pathology, Showa University School of Dentistry, Tokyo, Japan Nobuho Tanaka  •  Clinical Research Center, National Hospital Organization, Sagamihara Hospital, Kanagawa, Japan Tomoaki Tanaka  •  Department of Urology, Osaka City University Graduate School of Medicine, Osaka, Japan Michael A. Tangrea  •  Pathogenetics Unit and Laser Microdissection Core, Laboratory of Pathology, Center for Cancer Research, National Cancer Institute, National Institutes of Health, Bethesda, MD, USA Johannes Thiel  •  Leibniz-Institut für Pflanzengenetik und Kulturpflanzenforschung, Gatersleben, Germany

Contributors

xvii

Christopher C. Thorn  •  Department of Academic Surgery, St. James’s University Hospital, Leeds, UK Maria Tretiakova  •  Department of Pathology, University of Chicago, Chicago, IL, USA Arzu Umar  •  Netherlands Proteomics Center, Erasmus MC Rotterdam, Rotterdam, The Netherlands; Department of Medical Oncology, Center for Translational Molecular Medicine, and Cancer Genomics Centre, Erasmus MC Rotterdam, Rotterdam, The Netherlands Toon Van Gorp  •  Division of Gynecological Oncology, Department of Obstetrics and Gynecology, University Hospitals Leuven, Katholieke Universiteit Leuven, Leuven, Belgium; Division of Gynaecological Oncology, Department of Obstetrics and Gynaecology, MUMC+, GROW – School for Oncology and Developmental Biology, Maastricht, The Netherlands Timothy D. Veenstra  •  Laboratory of Proteomics and Analytical Technologies, SAIC-Frederick, Inc., National Cancer Institute at Frederick, Frederick, MD, USA Ignace Vergote  •  Division of Gynecological Oncology, Department of Obstetrics and Gynecology, University Hospitals Leuven, Katholieke Universiteit Leuven, Leuven, Belgium Etienne Waelkens  •  Department of Molecular Cell Biology, University Hospitals Leuven, Katholieke Universiteit Leuven, Leuven, Belgium Donald W. Weaver  •  Department of Surgery, Wayne State University, Detroit, MI, USA Gabriele Weber  •  Institute for Vascular Medicine, Friedrich Schiller University of Jena, Jena, Germany Roel A. de Weger  •  Department of Pathology, University Medical Centre Utrecht, Utrecht, The Netherlands Diana Weier  •  Leibniz-Institut für Pflanzengenetik und Kulturpflanzenforschung, Gatersleben, Germany Falk Weih  •  Institute for Vascular Medicine, Friedrich Schiller University of Jena, Jena, Germany Winfriede Weschke  •  Leibniz-Institut für Pflanzengenetik und Kulturpflanzenforschung, Gatersleben, Germany Claus Hann von Weyhern  •  Comprehensive Cancer Center, University of Tübingen, Tübingen, Germany Deborah Williams  •  MRC Harwell, Oxford, UK Tsung-Ung W. Woo  •  Laboratory of Cellular Neuropathology, Department of Psychiatry, McLean Hospital, Harvard Medical School, Belmont, MA, USA George W. Yip  •  Department of Anatomy, Yong Loo Lin School of Medicine, National University of Singapore, Singapore

wwwwwwwwwwwwwww

Chapter 1 Laser Capture Microdissection: Methods and Applications Kristen DeCarlo, Andrew Emley, Ophelia E. Dadzie, and Meera Mahalingam Abstract Laser microdissection is a nonmolecular, minimally disruptive method to obtain cytologically and/or phenotypically defined cells or groups of cells from heterogeneous tissues. It is a versatile technology and allows the preparation of homogenous isolates of specific subpopulations of cells from which RNA/DNA or protein can be extracted for RT-polymerase chain reaction (PCR), quantitative PCR, Western blot analyses, and mass spectrophotometry. Key words: DNA analysis, Laser capture microdissection, Melanoma, PCR, Proteomics, RNA analysis

1. Introduction The molecular analysis of DNA, RNA, and protein derived from diagnostic tissue, has revolutionized pathology and led to the identification of a broad range of diagnostic and prognostic markers (1). Analysis of critical gene expression and protein patterns in normal developing and diseased tissue progression requires the microdissection and extraction of a microscopic homogeneous cellular subpopulation from its complex tissue milieu (2). However, the reliability of tests based on tissue or cell extracts often depends crucially on the relative abundance of the cell population in question (1). Therefore, a prerequisite for modern molecular research is the capability of preparing pure samples without a large number of “contaminating” cells (1, 3). Laser capture microdissection (LCM) offers a simple, one-step process that provides scientists with a fast and dependable method of preserving and isolating single cells, or clusters of cells, from tissue sections by direct microscopic visualization (2, 4, 5). Graeme I. Murray (ed.), Laser Capture Microdissection: Methods and Protocols, Methods in Molecular Biology, vol. 755, DOI 10.1007/978-1-61779-163-5_1, © Springer Science+Business Media, LLC 2011

1

2

K. DeCarlo et al.

1.1. History

The need to isolate specific cells from complex tissues in order to carry out accurate molecular assays has been argued for decades (6). In the 1970s, Lowry and Passonneau pioneered a procedure for biochemical microanalysis, which utilized “freehand” microdissection of specific cell types under a microscope (6, 7). At the same time, several papers described different techniques that were also based on manual dissection (under microscope control) using razor blades, needles, or fine glass pipettes to isolate the cells of interest (6). An obvious shortcoming is that manual microdissection is time consuming, tedious, and does not allow for precise control of the material effectively selected (6, 7). A significant technological advance was proposed by Shibata in 1993 who suggested selective ultraviolet radiation fractionation, a procedure which utilized an ultraviolet laser beam to destroy the DNA of all undesired components of the tissue, while the cells of interest were protected by a specific dye (6–8). Unfortunately, this technique is only useful for analytes that are susceptible to degradation by UV-light, such as DNA (7). Subsequent improvements of this procedure led to the development of more sophisticated techniques that enabled isolation of single cells (6). The LCM system was developed during the mid-1990s by Dr. Emmert-Buck and colleagues at the National Institutes of Health (NIH), Bethesda, ML, USA (9). The system was initially developed for the analyses of solid tumors, and was later commercialized by Arcturus Engineering (Sunnyvale, CA, USA) as the PixCell system (6, 9). The PixCell series is currently the most widely used laser-based microdissection system, its development propelled by its integration into the “cancer genome anatomy project” (CGAP) sponsored by the National Cancer Institute (NCI) (1, 9). Multiple generations of this instrument (PixCell II; Arcturus Engineering, Mountain View, CA, USA) are currently on the market (1). Arcturus has also recently commercialized a new system (VeritasTM microdissection) that combines their LCM system, based on infrared laser, with UV laser cutting possibilities, the latter ideal for nonsoft tissues, and capturing large numbers of cells (6, 10).

2. Overview 2.1. Principle

The LCM system by Arcturus (PixCell II) is based on the selective adherence of visually targeted cells and tissue fragments to a special thermoplastic film made of an ethylene vinyl acetate (EVA) membrane activated by a low energy infrared laser pulse (1, 6). The system consists of an inverted microscope, a solid-state nearinfrared laser diode, a laser control unit, a joystick controlled microscope stage with a vacuum chuck for slide immobilization, a charge coupled device camera, and a color monitor. The LCM

1  Laser Capture Microdissection: Methods and Applications

3

microscope is usually connected to a personal computer for additional laser control and image archiving (1). The thermoplastic membrane used for transfer of selected cells is manufactured on the bottom surface of a plastic support cap, which acts as an optic for focusing the laser (1, 11). It has a diameter of approximately 6 mm and fits on standard 0.5 ml microcentrifuge tubes to facilitate further tissue processing (1). The cap is suspended on a mechanical transport arm and placed on the desired area of the mounted tissue sections (1). After visual selection of the desired cells, laser activation leads to focal melting of the EVA membrane, which has its absorption maximum near the wavelength of the laser (1). The polymer melts only in the vicinity of the laser, and expands into the section filling small hollow spaces present in the tissue (1, 11). Properly melted polymer spots have a dark outer ring and a clear center, indicating that the polymer has melted and is in direct contact with the slide (Fig. 1) (11). The polymer then resolidifies within milliseconds (ms) and forms a composite with the tissue (1). A dye incorporated into the polymer serves two purposes: first, it absorbs laser energy, preventing damage to the cellular constituents, and second, it aids in visualizing areas of melted polymer (11). The adherence of the tissue to the activated membrane exceeds the adhesion to the glass slide and allows for selective removal of the desired cells (1). Laser pulses between 0.5 and 5 ms in duration repeated multiple times across the cap surface, allow for rapid isolation of large numbers of cells (1). Lifting the cap then shears the selected cells from the heterogeneous tissue section (1, 11). The minimum diameter of the laser beam (7.5 mm) has been reduced in the newer generation machine. Under standard working

Fig. 1. LCM polymer bubbles. Properly melted polymer bubbles have a dark outer ring, indicating the polymer has melted and is in direct contact with the slide. (a) Larger spots can be created by increasing the power and spotsize of the laser to 100 mW and 30 mm, respectively. (b) Smaller spots can be created by decreasing the power and spotsize of the laser to 30 mW and 10 mm, respectively.

4

K. DeCarlo et al.

conditions, the area of the polymer melting corresponds exactly to the laser spot size. Also, since most of the energy is absorbed by the membrane, the maximum temperatures reached by the tissue upon laser activation are in the range of 90°C for several milliseconds, thus leaving biological macromolecules intact (1). The short laser pulse durations used (0.5–5.0 ms), the low laser power levels required (1–100 mW), the absorption of the laser pulse by the dye-impregnated polymer, and the long elapsed time (0.2 ms) between laser pulses combine to prevent any significant amount of heat deposition at the tissue surface which might compromise the quality of the tissue/cells utilized in later laboratory analyses (1, 9, 11). 2.2. Tissue Fixation, Sectioning, and Staining

Laser-based microdissection techniques have been applied to a wide range of tissues, prepared with a variety of methods, and utilizing a diverse range of biological samples (9). However, the procedures used in the preparation of tissue or cells for microdissection vary with the intended purposes and the analytes sought (7). Tissue specimens are typically either fixed in aldehyde-based fixatives (e.g., 10% formalin) or snap frozen (12). Formalin-fixation (10% buffered formaldehyde) is the standard for morphologic preservation of tissue, and has been used in histology laboratories for decades because of its low cost and rapid, complete penetration of tissue (7, 11). Although formalinfixed tissues are well preserved for histopathological evaluation, the quality of the macromolecules is severely compromised (12). It is an “additive” fixative that creates cross-links between itself and proteins, and between nucleic acids and proteins (6). This cross-linking interferes with recovery of nucleic acids and proteins, as well as the amplification of DNA and RNA by polymerase chain reaction (PCR) (6, 7). As a consequence of these crosslinks, the nucleic acids isolated from these specimens are highly fragmented, especially as fixation time is increased (6). This problem often occurs when using archival material, especially since pathology laboratories did not pay much attention to fixation times in the past (6). Fortunately, it has been shown that shorter lengths of DNA, up to approximately 200 base pairs, are recoverable by PCR after extraction from formalin fixed-paraffin embedded (FFPE) tissue (7). Ethanol-based fixatives offer the best RNA preservation by fixing tissues through dehydration without creating chemical links (6, 7). However, it has been found that sectioning with alcoholbased fixatives is more difficult (13). Therefore, the use of alcohol fixatives is only feasible if microdissection is considered as one of the possible options for processing the sample from the start (6). In the case of histological preparations, it is certainly better to utilize samples that have been snap-frozen and stored in liquid

1  Laser Capture Microdissection: Methods and Applications

5

nitrogen at 80°C or colder (6, 7). Frozen sections do not undergo cross-linking due to fixatives, and as a result, yield high quality messenger RNA (mRNA) and proteins (6, 12). However, freezing and cryostat sectioning can significantly disrupt the histological architecture of the tissue (12). This is a major problem since LCM is accomplished through identification of cells by morphological characteristics (11). The main goal of tissue preparation is to ensure that both the morphology of the tissue and molecules of interest are preserved (9). Recently, methods have been developed for the extraction and amplification of RNA from FFPE tissue sections (14). Like fresh tissue, mRNA amplification by nested RT-PCR (reversetranscriptase PCR) has been reported for single cells isolated from FFPE tissue through LCM (1, 15). Similarly, there has been a development of protocols which permit the extraction and mass spectrometric analysis of proteins from FFPE tissues (9). However, even though new technologies are being developed to reverse cross-linking for extraction of sufficient quantities of nucleic acids and proteins, high-quality yield of RNA and proteins is best achieved with frozen or ethanol-fixed tissue (11). The ability to effectively break the cross-links in nucleic acid caused by formalin could allow the utilization of a wealth of archived FFPE tissue for RNA expression and genomic analysis (7). Optimal LCM is achieved with tissue sections cut at a thickness of 2–15 mm (11). Tissue sections thinner than 5 mm may not provide full cell thickness, necessitating multiple microdissections in order to obtain an adequate number of cells for a given assay (11). Tissue sections thicker than 15  mm may not microdissect completely, leaving integral cellular components adhering to the slide (11). Ideally, staining should provide an acceptable morphology to allow the selection of target cells without interfering with the macromolecules of interest, or subsequent molecular techniques (6). Therefore, tissue sections should be exposed to the dye solution for the briefest period of time (9, 11). Minimal staining times limit potential protein alterations, and reduce the risk of chemical modification due to contact with reagents (9, 11). Sections can be stained satisfactorily by a few seconds exposure to the dye solution, followed by removal of excess dye with rapid washing (9). Examples of LCM-compatible stains are hematoxylin and eosin (most commonly used for examination of histologic sections), methylene blue, Wright-Giemsa, and toluidine blue (7, 11). In our experience, eosin staining is not necessary for visualization of cells. Specimens can also be stained immunohistochemically or with fluorescent labels, allowing the investigator to target cells based on the presence of specific antigens (7, 9). Stained sections are dehydrated and kept without a coverslip (6).

6

K. DeCarlo et al.

3. Protocols 3.1. Evolution of LCM Protocols

Due to the infancy of LCM technology, protocols have been constantly changing. Our own experience confirms this. At the outset, tissue slides were cut in 7–10-mm sections and mounted on uncharged slides. However, we found that 5 mm sections allowed for better procurement of cells, particularly in melanoma samples (Fig. 2). It is our contention that in this entity, thinner tissue sections allowed the melted polymer to more effectively penetrate tissue samples, thus enhancing yield. Similarly, modifications in LCM power and spotsize have led to more efficient tissue retrieval. Initially, the PixCell IIe LCM machine was used with a power ranging from 70 to 100 mW and a spotsize of 10 mm. However, after numerous trials, we found that a power of 80  mW and a spotsize of 7.5 mm were most effective in optimizing our yield.

3.2. Factors Affecting Yield of DNA, RNA, and Protein

These include quality of sample, time of preservation before microdissection, type of preservation, fixation method, and efficiency of microdissection (2). In our experience, fixation is the most critical step to ensure a high-quality yield of DNA, RNA, or protein (11). Quality of fixation is dependent on the length of time for fixative penetration in the tissue, temperature of fixation, and tissue size (2). In contrast to DNA, mRNA and protein are more sensitive to fixation, are quickly degraded, and require stringent RNase and proteinase-free conditions during specimen handling and preparation (1, 2). Therefore, the longer the fixative takes to penetrate the tissue, the greater the chance of RNA or protein degradation due to these ubiquitous RNases and proteinases (2). As a result, tissue microdissection is currently more widely employed in the analysis of DNA, as opposed to RNA and proteins, which are much more sensitive to degradation and fixation (6). In general, one set of microdissected cells is used for the downstream analysis of only one type of molecule (2). Each class of molecule requires different solubilization schemes, extraction buffers, and denaturing temperatures. For example, a population of 10,000 microdissected cells could be solubilized in denaturing buffer at 70°C for downstream protein analysis, while a second set of 100 cells could be treated with proteinase K at 65°C for downstream DNA analysis (2). Captured cells are detached from the cap membrane by proteinase digestion, and standard singlestep PCR protocols can be applied if enough cells have been collected (1). As can be seen, it is often necessary to microdissect many more cells than necessary based solely on DNA, RNA, or protein content of a cell (11). Examples of cellular yield required for DNA, RNA, and protein analyses are greatly varied, and range from 100 to 2,000 cells for DNA, 5,000–10,000 cells for RNA, and up to 4,000–200,000 cells for protein analyses (2).

1  Laser Capture Microdissection: Methods and Applications

7

Fig. 2. LCM of melanoma. (a) Melanoma nested in heterogeneous tissue section prior to LCM (40× magnification). (b) Melanoma after LCM. (c) Melted polymer bubble containing melanoma cells extracted from the heterogeneous tissue section.

3.3. Utility 3.3.1. Advantages

Perhaps the most relevant advantage of LCM is its speed while maintaining precision and versatility (1). LCM provides a reliable method to procure pure, precise populations of target cells from a wide range of cell and tissue preparations via microscopic visualization (16). The LCM system is applicable to normal glass slides (along with a wide range of other preparations), allowing routinely prepared material to be used after removal of the coverslip (6). Conventional techniques for molecular analysis are based on whole tissue dissociation and therefore introduce inherent contamination problems, thus reducing the specificity and sensitivity to subsequent molecular analysis, while requiring a high level of manual dexterity. LCM on the other hand, is a “no touch” technique that does not destroy adjacent tissues following initial microdissection. This allows several tissue components to be sampled sequentially from the same slide (e.g., normal and atypical cells) (1, 6, 16). LCM creates no chemical bonds to the target tissue so molecules in LCM-transferred cells are not degraded when compared to the original tissue slide (17). Furthermore, LCM isolates cells via firm adherence to the cap, reducing tissue loss, where other microdissection techniques require the removal of the isolated cells with the help of a needle tip or a microcapillary (1). The LCM technique is easily documented via a database program able to record images of both captured cells and residual tissue before and after microdissection. This diagnostic record is critical for maintaining an accurate record of each dissection, and for correlating histopathology with subsequent molecular analysis (6, 16). A final, critical advantage of LCM is its application to FFPE material, one of the most widely practiced methods for clinical sample preservation and archiving. Recent discoveries show promising advances in the use of FFPE tissues with LCM and subsequent molecular analysis. Collections of FFPE tissues comprise an invaluable resource for retrospective molecular studies of diseased tissues, including translational studies of cancer development (7, 14).

8

K. DeCarlo et al.

3.3.2. Disadvantages

The few limitations of LCM mostly reflect the difficulties of microdissection in general (1). Cell identification is performed in conjunction with a pathologist, and is based upon the morphological characteristics of the cells of interest (11). However, sections for microdissection are dehydrated and kept without a coverslip, making visualization of certain samples difficult due to decreased cellular detail (6, 18). This sometimes makes precise dissection of cells from complex tissues very difficult. However, this problem can be circumvented by special stains, in particular immunohistochemical stains, which help highlight cell populations to be isolated or avoided (1). Unfortunately, standard immunohistochemical staining protocols require several hours, which can lead to further degradation of RNA and protein by RNases and proteinases, respectively (1, 2, 6). Fixation, dehydration, and staining of tissue sections also makes “live-cell analysis” (18) impossible. Another problem occasionally encountered in LCM is failure to remove selected cells from the slide (1). This can result from a lack of adherence of the cells to the EVA membrane, usually because of incomplete dehydration or a laser setting that is too low for complete permeation of the melted polymer into the section (1, 6). On the other hand, increased adherence of the section to the slide can prevent the removal of the targeted cells (1). As a result, isolation of large numbers of cells (e.g., for protein analysis) from many sections can require considerable time (2, 18). Older machines face problems related to a minimum laser spot size of 7.5 mm, which imposes restrictions on the precision of LCM recovery and makes it difficult to isolate cells of interest without contamination. The more recent generation of LCM machines, capable of dissecting cells at single cell level, have overcome these limitations (1).

3.4. Clinical Applications

LCM has significantly enhanced the molecular analysis of pathological processes as it offers a simple and efficient technique for procuring a homogeneous population of cells from their native tissue via direct microscopic visualization. LCM makes it possible to analyze cellular function between neighboring, intermingling, and morphologically identifiable cells within complex tissues and organs (17). Overall, LCM is applicable to molecular profiling of tissue in normal and disease states; this includes correlations of cellular molecular signatures within specific cell populations and the comparison of different cellular elements within a single tissue microenvironment (11). LCM-based molecular analysis is being used in many fields of research, including the study of normal cell biology, as well as in vivo genomic and proteomic states such as the profiling of cultured intervertebral disc cells, molecular analysis of skeletal cell differentiation, and gene expression in testicular cell populations (16, 19–23). Other studies focusing on the molecular analysis of

3.4.1. An Overview

1  Laser Capture Microdissection: Methods and Applications

9

histopathological lesions and disease processes include mapping genetic alterations associated with the progression of premalignant cancer lesions (breast cancer and their lymph node metastases, ovarian cancer, and prostate cancer); analyses of gene expression patterns in multiple sclerosis, atherosclerosis, and Alzheimer’s disease plaques; diagnosis of infectious micro-organisms; and the analysis of genetic abnormalities in utero from selected fetal cells in maternal fluids (1, 2, 9, 16). LCM is currently being used in the Cancer Genome Anatomy Program (CGAP) and exemplifies the molecular advances that LCM offers, as it allows researchers to catalog genes that are expressed in human tissue as normal cells undergo premalignant changes and further develop into invasive and metastatic cancer. Changes in expressed genes or alterations in cellular DNA corresponding to a specific disease state can be compared within or between individual patients, as a large number of microdissected cDNA libraries (produced from microdissected normal and premalignant tissue RNA) have been produced and published on the CGAP web page. This catalog of gene expression patterns has the potential to provide clues to etiology and, hopefully, contribute to early diagnostic detection and more accurate diagnosis of disease, followed by therapies tailored to individual patients (11, 16, 17). LCM has been applied to genomic analyses such as studies of X-chromosome inactivation patterns to assess clonality, promoter hypermethylation, restriction fragment length polymorphisms, and single strand conformation polymorphism analysis for assessment of mutations in critical genes such as p53 and K-RAS. Novel uses include cancer chemoprevention, biomarker discovery, and live and rare cell isolation. LCM has been used for biomarker discovery in various human tissue types and organ systems. In these studies, LCM is used in combination with DNA transcriptome profiling to identify differentially expressed genes (24). Intermediate endpoint biomarkers, used to monitor the success of chemoprevention, have been successfully developed for prostate cancer, cervical carcinoma, and adenomas for colorectal cancer (24). Finally, LCM has been applied to the study of live and rare cell populations. Remarkably, LCM has no influence on the viability, metabolism, and proliferation rate of isolated living cells where even an entire living organism (such as the nematode Caenorhabditis elegans) can be successfully transferred without compromising the biological composition or viability of the organism. Live cell LCM isolation equipment is available from several manufacturers (25). Finally, LCM is being used to isolate rare cells. In this rapidly developing method, rarely occurring cells are identified with automated scanning software, immediately followed by computer-controlled LCM (25).

10

K. DeCarlo et al.

3.4.2. DNA Analysis

Microdissection is now an established technique used to collect homogeneous cell populations for the analysis of genetic alterations at the DNA level (1). With the advent of efficient analytical methods for small amounts of biological material, LCM is applied in pathological diagnosis, classification, and treatment of tumors. It even plays a major role in gene mutation studies where a homogenous tumor cell population is necessary for accurate genomic analysis (1).

3.4.3. RNA Analysis

LCM offers several other advantages for mRNA analysis as compared to other laboratory techniques such as mRNA in situ hybridization or immunohistochemistry. Microdissection of purified cells, in combination with methods such as real-time quantitative RT-PCR, allows for a precise determination of cell-specific gene expression (25). Furthermore, LCM is an efficient technique that allows sampling of large numbers of cells without significant RNA degradation where tissue dehydration may even inhibit the activity of tissue RNases, thereby maintaining the tissue integrity during specimen handling and preparation (1). Gene expression analysis is critical in uncovering the patterns related to neoplastic transformation, however, the simultaneous detection of multiple different messages is preferable over the examination of single or few expressed genes. Therefore, microdissected cells are used in conjunction with cDNA array hybridization or serial analysis of gene expression to reveal the differences in gene expression profiles of normal and neoplastic cells, or to show altered gene expression patterns at various stages of cancer progression. LCM is also an essential tool in this process, as mRNA from microdissected lesions is subsequently used as the precursor for creating cDNA and expression libraries from purified cell populations (1).

3.4.4. Proteomic Analyses

Proteomics aims to establish the complete set of proteins or the “proteome” that are important in normal cellular physiology. The normal proteome is compared to a disease state proteome such as cancer using a variety of analyses including western blotting, high-resolution two-dimensional polyacrylamide gel electrophoresis (2-D PAGE), and mass spectrometry and peptide sequencing. Proteomics is a complementary approach to gene expression studies and provides supplementary information not obtained through genome or transcriptome analysis (24, 25). Deciphering alterations in proteomic profiles using LCM techniques offers the advantage of studying physiological relationships unique to protein analysis, thereby offering the potential to identify novel diagnostic and therapeutic targets.

3.4.5. Singe Cell Analysis

LCM has been applied to the isolation of single cells for the analysis of specific targets such as the identification of point mutations

1  Laser Capture Microdissection: Methods and Applications

11

in oncogenes such as RAS and the amplification of expressed gene sequences by RT-PCR. Additionally, microdissected single cells can be used as a template for whole genome amplification, the generation of expression libraries, or probes for expression profiling with cDNA arrays (1). Saurez-Quian et al. has modified the LCM protocol specifically for single cell capturing. In this technique, a cylinder covered with EVA polymer membrane has replaced the large cap surface. This decreases the contact area with the tissue and increases the accuracy of procuring a homogenous cell population (1). 3.5. Specific Diagnostic Applications 3.5.1. Tumors

3.5.2. Clonality Studies

The identification of genetic mutations is paramount in the pathological diagnosis, classification, and treatment of tumors. Loss of heterozygosity (LOH) analysis has been pivotal in cancer research for mapping of tumor suppressor genes, localization of putative chromosomal “hot spots,” and the study of sequential genetic changes in preneoplastic lesions. Microdissection has become a key technique used in LOH studies, since pure populations of tumor cells are necessary, and contamination by even a few unwanted cells may result in inappropriate amplification (via PCR) of the “lost” second allele present in noncancer tissue. LOH studies preformed via microdissection have shown that the frequencies of genetic alterations have been largely under-­ estimated such that there may even be heterogeneity present within a single tumor where some genetic changes occur early in tumorigenesis (24). Furthermore, LCM has been applied to the study of protein alterations in preneoplastic lesions and their tumor counterparts in an effort to elucidate novel tumor-specific alterations in peptide products of cancer cells. From these proteomic studies, distinct protein expression patterns have successfully classified normal, premalignant and malignant cancer cells collected using LCM from human tissues (24). Recently, it has become possible to use smaller samples of cells (not more than 20–100 dissected cells per PCR) obtained via microdissection, allowing a more refined study of preneoplastic lesions in addition to neoplastic lesions. This has been made possible using a combination of microdissection with primer extension preamplification and whole genome amplification techniques, thereby opening a whole new frontier in cancer research (24). Assessing clonality via DNA analyses using LCM has played an instrumental role in identifying the multiple endocrine neoplasia type 1 gene (MEN1), and will hopefully uncover the genetic basis underlying other cancer types. In the case of MEN1, LOH analysis of 200 microdissected endocrine tumors narrowed the interval of the genetic aberration to 300 kb. This LOH information from LCM analysis was used in conjunction with haplotyping

12

K. DeCarlo et al.

and newly identified polymorphic markers, and led to the identification of a new tumor suppressor gene responsible for MEN1 (1, 24). LCM, in conjunction with DNA analyses, has the ability to distinguish the presence of two clonal populations in the same tumor site. Fend et  al. have demonstrated this in malignant nonHodgkin’s lymphoma, where two phenotypically and morphologically distinct cell populations were present in the same tumor. In this study, LCM was used to procure homogenous samples of the two populations from immunostained slides. Subsequent sequencing of rearranged immunoglobulin genes confirmed the presence of two unrelated clones in all cases. LCM played a pivotal role in this study, as PCR analysis of DNA obtained from whole sections was not able to detect the biclonal composition of the tumors (1). 3.5.3. Clinical Applications of LCM in Dermatopathology

LCM has broadened the role of dermatopathology in molecular diagnosis and has greatly enhanced the understanding of the pathogenesis of inherited skin diseases (9, 26, 27). The examination of precancerous lesions by LCM has been applied in the study of melanomagenesis, as it is widely believed that benign nevi undergo genetic alterations that progressively lead to melanoma development. LCM is used to assess the incidence of genetic gains and losses in tumors and preneoplastic lesions, and in doing so, has the potential to uncover the molecular events associated with the transformation of banal nevi into malignant melanoma formation (28–30).

Nevi Versus Melanoma

From a histopathological perspective, melanoma development is tracked by a series of melanocyte transitions from easily characterized precursors. However, from a genetic perspective, these transitions are poorly understood (30). LCM, therefore, has the potential to shed light on the genetic profiles of melanocytes as they undergo these morphological transitions, hopefully uncovering the molecular events that lead to melanomagenesis. Using LCM to dissect distinct populations of nevic aggregates in association with melanoma, we have been able to show that banal nevic aggregates might serve as precursor lesions (31).

Clonality in Cutaneous T-cell Lymphoma

Analysis of T-cell gene rearrangement in cutaneous T-cell lymphoma (CTCL) has led to the discovery that the earliest manifestation of CTCL may be “clonal dermatitis.” Clonal dermatitis is a chronic form of dermatitis that contains a dominant T-cell clone but does not show the typical histologic features diagnostic for CTCL. Significantly, approximately 25% of clonal dermatitis cases develop into CTCL within 5 years, where the same clone is present in both the clonal dermatitis and the CTCL lesions, indicating that the clonal dermatitis clone is a precursor to the CTCL. LCM is ideal for this type of study as the often-sparse lymphocytic

1  Laser Capture Microdissection: Methods and Applications

13

infiltrate can be specifically captured. Furthermore, these studies allow unprecedented investigations into the molecular pathogenesis of CTCL, which will hopefully lead to early disease detection and help guide gene therapy (32). LCM is also able to demonstrate genetically different clones or gene mutations limited to one specific neoplastic population. This has been an important tool in cutaneous lymphoma lesions containing a mixed B- and T-cell population. Using microdissection followed by genotypic analysis, Gallardo et  al. established that the lesion of interest in the case study was cutaneous B-cell lymphoma with a dual B- and T-cell genotype. Conventional methods were not able to make this distinction, therefore illustrating the usefulness of LCM in clinical diagnostics (33). Infectious Diseases

LCM allows for the isolation of pure cell populations which can be screened through PCR for infectious agents depending on the clinical and histological suspicion. LCM also plays an important role in routine histopathologic diagnostics and has been applied to the diagnosis of infectious diseases such as borreliosis, herpes simplex virus infection, herpes zoster, Epstein–Barr virus infection, Myobacterium tuberculosis, and many others (5, 34).

4. Conclusions Tissue-based laser microdissection is a powerful technique, which combines morphology, histochemistry, and sophisticated downstream molecular analysis (35). High speed, easy handling, and good control and documentation of dissected tissue make LCM an ideal tool for the rapid collection of larger amounts of tissue. Further technological advances such as touch-screen cell annotation, automated cell microdissection, and cell recognition software are leading to the next generation machines with enhanced microdissection capabilities. The ability of LCM to visualize and capture specific populations of cells has made LCM an important diagnostic tool, not just in dermatopathology. References 1. Fend F, Raffeld M (2000) Laser capture microdissection in pathology. J Clin Pathol 53, 666–672 2. Espina V, Heiby M, Pierobon M et al (2007) Laser capture microdissection technology. Expert Rev Mol Diagn 7, 647–657 3. Burgemeister R (2005) New aspects of laser capture microdissection in research and routine. J Histochem Cytochem 53, 409–412

4. Agar NS, Halliday GM, Barnetson RS et  al (2003) A novel technique for the examination of skin biopsies by laser capture microdissection. J Cutan Pathol 30, 265–270 5. Yazdi AS, Puchta U, Flaig MJ et  al (2004) Laser-capture microdissection: Applications in routine molecular dermatopathology. J Cutan Pathol 31, 465–470 6. Esposito G (2007) Complementary techniques: Laser capture microdissection—increasing

14

K. DeCarlo et al.

specificity of gene expression profiling of cancer specimens. Adv Exp Med Biol 593, 54–65 7. Eltoum IA, Siegal GP, Frost AR (2002) Microdissection of histologic sections: Past, present, and future. Adv Anat Pathol 9, 316–322 8. Shibata D (1993) Selective ultraviolet radiation fractionation and polymerase chain reaction analysis of genetic alterations. Am J Pathol 143, 1523–1526 9. Murray GI (2007) An overview of laser capture microdissection technologies. Acta Histochem 109, 171–176 10. Veritas Microdissection System. MDS Analytical Technologies. http://www.moleculardevices. com/pages/instr uments/veritas.html. Accessed 11 July 2008 11. Espina V, Wulfkuhle JD, Calvert VS et  al (2006) Laser-capture microdissection. Nat Protoc 1, 586–603 12. Ahram M, Flaig MJ, Gillespie JW, et al (2003) Evaluation of ethanol-fixed, paraffin-embedded tissues for proteomic applications. Proteomics 3, 413–421 13. Bostwick DG, al Annouf N, Choi C (1994) Establishment of the formalin-free surgical pathology laboratory. Utility of an alcoholbased fixative. Arch Pathol Lab Med 118, 298–302 14. Gianni L, Zambetti M, Clark K et  al (2005) Gene expression profiles in paraffin-embedded core biopsy tissue predict response to chemotherapy in women with locally advanced breast cancer. J Clin Oncol 23, 7265–7277 15. Schutze K, Lahr G (1998) Identification of expressed genes by laser-mediated manipulation of single cells. Nat Biotechnol 16, 737–742 16. Bonner RF, Emmert-Buck M, Cole K et  al (1997) Laser capture microdissection: Molecular analysis of tissue. Science 278, 1481–1483 17. Simone NL, Bonner RF, Gillespie JW et  al (1998) Laser-capture microdissection: Opening the microscopic frontier to molecular analysis. Trends Genet 14, 272–276 18. Brignole E (2000) Laser-capture microdissection: Isolating individual cells for molecular analysis. Mod Drug Discovery 3, 69–70 19. Gruber HE, Mougeot JL, Hoelscher G et al (2007) Microarray analysis of laser capture microdissected-anulus cells from the human intervertebral disc. Spine 32, 1181–1187 20. Benayahu D, Socher R, Shur I (2008) Application of the laser capture microdissection technique for molecular definition of skeletal cell differentiation in  vivo. Methods Mol Biol 455, 191–201

21. Sluka P, O’Donnell L, McLachlan RI et  al (2008) Application of laser-capture microdissection to analysis of gene expression in the testis. Prog Histochem Cytochem 43, 173–201 22. Shukla CJ, Pennington CJ, Riddick AC et  al (2008) Laser-capture microdissection in prostate cancer research: establishment and validation of a powerful tool for the assessment of tumour-stroma interactions. BJU Int 101, 765–774 23. Harrell JC, Dye WW, Harvell DM et  al (2008) Contaminating cells alter gene signatures in whole organ versus laser capture microdissected tumors; a comparison of experimental breast cancers and their lymph node metastases. Clin Exp Metastasis 25, 81–88 24. Domazet B, MacLennan G, Lopez-Beltran A et al (2008) Laser capture microdissection in the genomic and proteomic era: targeting the genetic basis of cancer. Int J Clin Exp Pathol 1, 475–488 25. Ladanyi A, Sipos F, Szoke D et al (2006) Laser microdissection in translational and clinical research. Cytometry A 69A, 947-960 26. Bergman R (2008) Dermatopathology and molecular genetics. J Am Acad Dermatol 58, 452–457 27. What is a Dermatopathologist? The American Society of Dermatopathology. http://www. asdp.org/about/dermatopathologist.cfm. Accessed 1 March 2009 28. Boni R, Zhuang Z, Albuquerque A et  al (1998) Loss of heterozygosity detected on 1p and 9q in microdissected atypical nevi. Arch Dermatol 134, 882–883 29. Maitra A, Gasdar AF, Moore TO et al (2002) Loss of heterozygosity analysis of cutaneous melanoma and benign melanocytic nevi: laser capture microdissection demonstrates clonal genetic changes in acquired nevocellular nevi. Hum Pathol 33, 191–197 30. Hussein MR (2004) Genetic pathways to melanoma tumorigenesis. J Clin Pathol 57, 797–801 31. Dadzie OE, Yang S, Emley A et al (2009) RAS and RAF mutations in banal melanocytic aggregates contiguous with primary Cutaneous melanoma: clues to melanomagenesis. Br J Dermatol 160, 368–375 32. Woody GS (2001) Analysis of clonality in cutaneous T cell lymphoma and associated diseases. Ann NY Acad Sci 941, 26–30 33. Gallardo F, Pujol RM, Bellosillo D et al (2006) Primary cutaneous B-cell lymphoma (marginal zone) with prominent T-cell component

1  Laser Capture Microdissection: Methods and Applications and aberrant dual (T and B) genotype; diagnostic usefulness of laser-capture microdissection. Br J Dermatol 154, 162–166 34. Zhu G, Xiao H, Mohan VP et al (2003) Gene expression in the tuberculous granuloma:

15

analysis by laser capture microdissection and real-time PCR. Cell Microbiol 5, 445–453 35. Curran S, McKay JA, McLeod HL, et  al (2000). Laser capture microscopy. Mol Pathol 53, 64 –68

wwwwwwwwwwwwwww

Chapter 2 Laser Microdissection for Gene Expression Profiling Lori A. Field, Brenda Deyarmin, Craig D. Shriver, Darrell L. Ellsworth, and Rachel E. Ellsworth Abstract Microarray-based gene expression profiling is revolutionizing biomedical research by allowing expression profiles of thousands of genes to be interrogated in a single experiment. In cancer research, the use of laser microdissection (LM) to isolate RNA from tissues provides the ability to accurately identify molecular profiles from different cell types that comprise the tumor and its surrounding microenvironment. Because RNA is an unstable molecule, the quality of RNA extracted from tissues can be affected by sample preparation and processing. Thus, special protocols have been developed to isolate researchquality RNA after LM. This chapter provides detailed descriptions of protocols used to generate micro­ array data from high-quality frozen breast tissue specimens, as well as challenges associated with formalin-fixed paraffin-embedded specimens. Key words: Laser microdissection, Gene expression, Microarray, Frozen tissue, FFPE, Molecular signature, Breast cancer

1. Introduction Tumorigenesis is a complex process, involving structural changes at multiple chromosomal locations and altered expression of numerous genes and proteins. Early efforts to identify genes involved in cancer development evaluated single genes with known or putative roles in cellular processes such as growth, proliferation, angiogenesis, and apoptosis. While these efforts have resulted in the identification of more than 350 genes (1), additional genes of unknown or presumably unrelated function likely play critical roles in cancer development and progression (2). cDNA microarrays, which allow quantitative, large-scale analysis of gene expression, provide a global approach to identifying

Graeme I. Murray (ed.), Laser Capture Microdissection: Methods and Protocols, Methods in Molecular Biology, vol. 755, DOI 10.1007/978-1-61779-163-5_2, © Springer Science+Business Media, LLC 2011

17

18

L.A. Field et al.

genes involved in tumorigenesis and metastasis without a priori knowledge of the underlying molecular pathways (3). Microarrays have been used to develop molecular signatures that correlate with tumor characteristics or outcomes and are being used in clinical diagnostic tests to guide treatments for patients with breast cancer (4, 5). Despite the successful development of clinical assays and the publication of hundreds of microarray-based papers, the majority of microarray studies have used RNA isolated from tissue by homogenization or manual microdissection. Because the majority of human tumors are highly heterogeneous, with numerous cell types comprising the primary tumor and surrounding microenvironment, laser microdissection (LM) is necessary to isolate specific cells. For example, RNA isolated from laser-microdissected breast tumor cells will be free from contamination from normal epithelial, stromal, and vascular cells, which could compromise the accuracy of the resulting gene expression profiles. Because RNA is sensitive to degradation, isolation of RNA after LM requires a defined protocol that includes careful cleaning of all equipment with RNase inhibitors, special histological stains, and rapidity (less than 30 min) in cutting, mounting, and microdissecting the tissues. In this chapter, we present protocols for performing microarray analysis using RNA isolated after LM and describe alternate protocols for gene expression analysis of formalin-fixed paraffin-embedded (FFPE) archival specimens.

2. Materials 2.1. Tissue Sectioning, Staining, and Laser Microdissection

1. Membrane-based laser microdissection slides (W. Nuhsbaum, McHenry, IL). 2. Disposable microtome blades, HP35n, noncoated (Thermo Fisher Scientific, Pittsburgh, PA). 3. 0.5 ml PCR tubes (Eppendorf, Hauppauge, NY). 4. RNaseZap® (Applied Biosystems, Carlsbad, CA). 5. Nuclease-free water (Applied Biosystems). 6. LCM Staining Kit (Applied Biosystems) – store cresyl violet at 4°C. 7. 50% ethanol. 8. 75% ethanol. 9. 95% ethanol. 10. 100% ethanol. 11. Xylene (used only for FFPE samples).

2  Laser Microdissection for Gene Expression Profiling

19

12. Tissue-Tek® Cryomold® Standard, 25 × 20 × 5 mm (Electron Microscopy Sciences, Hatfield, PA). 13. Cryomatrix optimal cutting temperature (OCT) compound (Thermo Fisher Scientific). 2.2. RNA Isolation from Frozen Tissue

1. RNAqueous®-Micro kit (Applied Biosystems). 2. Nuclease-free water. 3. 100% ethanol. 4. Agilent RNA 6000 Pico kit (Agilent Technologies, Santa Clara, CA). 5. Agilent 2100 Bioanalyzer (Agilent Technologies). 6. RNaseZap®.

2.3. Amplification and Fragmentation of RNA from Frozen Tissue

1. MessageAmp™ Biosystems).

II

aRNA

Amplification

kit

(Applied

2. GeneChip® Eukaryotic Poly-A RNA Control kit (Affymetrix, Santa Clara, CA). 3. 75 mM Bio-11-UTP (Applied Biosystems). 4. Nuclease-free water. 5. 5× Fragmentation buffer, component of the GeneChip® Sample Cleanup Module (Affymetrix). 6. Agilent RNA 6000 Pico kit. 7. Agilent RNA 6000 Nano kit (Agilent Technologies, Santa Clara, CA). 8. Agilent 2100 Bioanalyzer. 9. NanoDrop ND-1000 Spectrophotometer (Thermo Fisher Scientific) – Note: the current model is the NanoDrop 2000.

2.4. Hybridization of aRNA to Microarrays

1. GeneChip® Expression 3¢ Amplification reagents containing 20× Eukaryotic Hybridization Controls and Control Oligo­ nucleotide B2 (Affymetrix). 2. Herring Sperm DNA (Promega, Madison, WI). 3. Bovine serum albumin (BSA) (Invitrogen, Carlsbad, CA). 4. MES hydrate (Sigma-Aldrich, St Louis, MO). 5. MES sodium salt (Sigma-Aldrich). 6. 5 M NaCl (Sigma-Aldrich). 7. 0.5 M EDTA (Sigma-Aldrich). 8. Tween 20 (Promega). 9. DMSO (Sigma-Aldrich). 10. Nuclease-free water.

20

L.A. Field et al.

11. GeneChip® Human Genome U133A 2.0 Arrays (HG U133A 2.0) (Affymetrix). 12. Hybridization oven (Affymetrix). 2.5. Washing, Staining, and Scanning Microarrays

1. Bovine serum albumin. 2. Streptavidin phycoerythrin (SAPE) (Invitrogen). 3. Goat IgG (Sigma-Aldrich). 4. Biotinylated antistreptavidin (Vector Laboratories, Burlingame, CA). 5. 20× SSPE (Sigma-Aldrich). 6. 5 M NaCl. 7. Tween 20. 8. Nuclease-free water. 9. Tough Spots (T-SPOTS; Diversified Biotech, Boston, MA). 10. Fluidics Station (Affymetrix). 11. Scanner (Affymetrix).

2.6. Quantitative Real-Time Polymerase Chain Reaction of RNA from Frozen Tissue or FFPE

1. High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems). 2. TaqMan® Universal PCR Master Mix (Applied Biosystems). 3. TaqMan® Gene Expression Assays (Applied Biosystems). 4. FirstChoice® Human Brain Reference RNA (Applied Biosystems). 5. iCycler iQ™ PCR plates (Bio-Rad Laboratories, Hercules, CA). 6. iCycler iQ™ thermal seals (Bio-Rad Laboratories). 7. iCycler iQ™ real-time PCR detection system (Bio-Rad Laboratories).

2.7. RNA Isolation from Formalin-Fixed Paraffin Embedded Specimens

1. RecoverAll™ total nucleic acid isolation kit (Applied Biosystems).

2.8. Commercial Vendor Information

1. Affymetrix – http://www.affymetrix.com.

2. 100% ethanol.

2. Agilent Technologies – http://www.agilent.com. 3. Applied Biosystems – http://www.appliedbiosystems.com. 4. Bio-Rad Laboratories – http://www.bio-rad.com. 5. Diversified Biotech – http://divbio.com/. 6. Electron Microscopy Sciences – http://emsdiasum.com/ microscopy/. 7. Eppendorf – http://www.eppendorf.com.

2  Laser Microdissection for Gene Expression Profiling

21

8. Invitrogen – http://www.invitrogen.com. 9. Promega – http://www.promega.com. 10. Sigma-Aldrich – http://www.sigmaaldrich.com. 11. Thermo Fisher Scientific – http://www.thermofisher.com. 12. Vector Laboratories – http://vectorlabs.com/. 13. W. Nuhsbaum – http://www.nuhsbaum.com/.

3. Methods RNA is extremely susceptible to degradation by RNase enzymes in the environment. To generate high-quality microarray or quantitative real-time polymerase chain reaction (qRT-PCR) data, it is critical to obtain RNA of the highest possible quality by preventing RNase contamination during tissue collection and processing, RNA isolation, and downstream applications. Several general precautions should be taken when working with RNA in the laboratory. All equipment and laboratory benches should be thoroughly cleaned with RNaseZap® and then rinsed with nuclease-free or deionized water. All pipette tips, tubes, reagents, and other consumables must be RNase-free. Pipette tips should contain barriers and should be changed each time you pipette, even if you are pipetting the same reagent, to avoid potential cross-contamination between samples and to prevent RNase contamination. For most procedures, it is advisable to use nuclease-free, hydrophobic, nonstick tubes to minimize loss of sample that may otherwise adhere to the tube walls. Gloves should be worn at all times and changed frequently, especially after coming into contact with liquids or surfaces that may be contaminated with RNases. 3.1. Sectioning and Staining

3.1.1. Frozen Tissue

To prevent RNA degradation, tissue sectioning, staining, and LM must be performed as quickly as possible (typically within 30 min). In our laboratory, two individuals perform these steps and process one slide at a time. The LCM Staining Kit employs a novel staining procedure that avoids exposing the tissue sections to pure water at any step, thus minimizing the potential for RNA degradation. 1. In the bottles provided with the LCM Staining Kit, prepare 95, 75, and 50% ethanol solutions by diluting 100% ethanol with nuclease-free water. Add the dehydration beads to the bottle labeled 100% ethanol and add absolute ethanol. Do not use the ethanol in this container to make any of the diluted solutions.

22

L.A. Field et al.

2. Clean the staining containers included in the LCM Staining Kit with RNaseZap®. For FFPE samples, a glass staining dish should also be cleaned. Spray the containers generously with RNaseZap® and allow them to sit for 10  min. Rinse twice with distilled water and then perform a final rinse with nuclease-free water. Allow the containers to dry under a hood and then fill with the appropriate solutions. 3. Set the temperature of the cryostat to −30°C. 4. Clean the knife holder (not the knife blade itself) with 100% ethanol and treat the brushes that will be used to manipulate the tissue sections with RNaseZap®. 5. Cool the specimen and brushes in the cryostat. 6. Inside the cryostat, remove the frozen OCT-embedded tissue from its cryomold and mount securely to the metal specimen stage with OCT compound, orienting the tissue according to regions of interest (see Note 1). 7. Using a fresh disposable blade, shave OCT from the block until the tissue becomes visible. Set the cutting thickness to 8 mm. 8. Section the tissue and use a small brush to straighten out the newly cut sections. 9. Manipulate sections onto the foil slides (see Note 2). Perform staining under a hood used only for RNA procedures. Change all containers and blade surfaces between each patient sample. 10. Cut sections at 8  mm and mount onto a membrane-based laser microdissection slide. 11. Wash slide in 95% ethanol for 30 s. 12. Wash in 75% ethanol for 30 s. 13. Wash in 50% ethanol for 30 s (see Note 3). 14. Pipette cresyl violet (~50  ml) onto the slide to completely cover the tissue sections; allow the slide to sit for 15 s. 15. Rinse in 95% ethanol for 5 s. 16. Rinse in 100% ethanol for 5 s. 17. Rinse in a second container of 100% ethanol for 30  s (see Note 4). 18. Allow slide to air dry. 3.1.2. Formalin-Fixed Paraffin-Embedded Tissue

1. Fill the clean staining dish with nuclease-free water and warm on a hot plate to the desired temperature for the paraffin being used (typically 37–42°C). Change the water bath between each sample.

2  Laser Microdissection for Gene Expression Profiling

23

2. Cut sections at 8 mm and lay out ribbon onto the warm water bath. 3. Mount sections onto a membrane-based LM slide. 4. Place slides in an incubator set at 56°C for 15  min (see Note 5). 5. Wash in xylene for 1  min; repeat twice for a total of three washes. 6. Wash in 95% ethanol for 30 s. 7. Wash in 75% ethanol for 30 s. 8. Wash in 50% ethanol for 30 s. 9. Pipette Cresyl Violet stain onto the slide using enough volume to cover the sections; allow to sit for 15 s. 10. Rinse in 95% ethanol for 5 s. 11. Rinse in 100% ethanol for 5 s. 12. Rinse in 100% ethanol for 30 s. 3.2. Laser Microdissection

1. Use a cover-slipped H&E section to orient the tissue for microdissection. Estimate the number of cells – in our experience, ~10,000 cells usually yields sufficient RNA for downstream applications. 2. Locate the area on the cresyl violet-stained section to be microdissected (see Note 6). 3. Pipette 60 ml of Lysis solution for OCT-embedded tissues, or 60  ml of digestion buffer for FFPE-embedded tissues, into the cap of a clean 0.5 ml Eppendorf tube. Place the cap into the cap holder apparatus of the laser microdissection system. 4. Microdissect the area of interest (Fig. 1) and drop the sample into the buffer (see Note 7). 5. Add the remaining 40 ml of Lysis solution (OCT tissues) or 340  ml of digestion buffer (FFPE tissues) to the tube and carefully close the lid.

3.3. RNA Isolation from Frozen Tissue

1. Before first use, add 10.5 ml of 100% ethanol to Wash solution 1 and 22.4  ml of 100% ethanol to Wash solution 2/3 and mix well (see Note 8). 2. On first use, thaw the Pico Ladder on ice, centrifuge briefly, and transfer to an RNase-free tube. Heat-denature the ladder for 2 min at 70°C in a heat block, then immediately place on ice. Add 90 ml of nuclease-free water, pipette up and down several times, and flick the tube to mix. Briefly centrifuge the tube and aliquot 5–10 ml to RNase-free tubes. Store at −70°C (see Note 9).

24

L.A. Field et al.

Fig. 1. Staining and laser microdissection of formalin-fixed paraffin-embedded breast tissue containing a ductal carcinoma in situ (DCIS). (a) Standard hematoxylin and eosin (H&E) stain of the DCIS on a glass slide. (b) DCIS on a foil slide stained with Cresyl Violet. (c) Breast tissue after removal of the DCIS by laser microdissection.

3. Place a tube containing nuclease-free water (at least 50 ml per sample) in a heat block at 95°C. 4. Prewarm an air incubator to 42°C. 5. Thaw LCM Additive and 10× DNase I buffer on ice. 6. Flick the tube containing the microdissected sample several times and centrifuge briefly. Place the sample in the 42°C incubator for 30 min (see Note 10). 7. Approximately 6–7  min prior to completion of the 30-min incubation, prewet the Micro Filter by adding 30 ml of Lysis solution to the filter, which is placed in a Micro-Elution Tube (Micro Filter Cartridge Assembly). After 5  min, centrifuge the Micro Filter Cartridge Assembly for 30 s at 16,000 rcf to remove the Lysis solution from the filter. 8. Remove the microdissected sample from the 42°C incubator, vortex on maximum speed by pulsing three times, and centrifuge briefly. Add 3  ml of LCM Additive, mix by vortexing, and centrifuge briefly. 9. Add 52 ml of 100% ethanol and mix completely into the sample by pipetting up and down (see Note 11). Transfer the sample to the center of the filter in the Micro Filter Cartridge Assembly. Centrifuge for 1 min at 10,000 rcf (see Note 12). 10. Add 180 ml of Wash solution 1 to the filter and centrifuge for 1 min at 10,000 rcf. 11. Add 180 ml of Wash solution 2/3 and centrifuge for 30 s at 16,000 rcf. Repeat this step one time. 12. Remove the filter from the collection tube and discard the flow-through. Recap the assembly and centrifuge for 1 min to remove trace amounts of liquid. 13. Remove the filter containing the sample and place in a new Micro Elution Tube. 14. Add 10  ml of nuclease-free water heated to 95°C in step 1 above to the center of the filter (see Note 13). Incubate the

2  Laser Microdissection for Gene Expression Profiling

25

assembly for 5 min at room temperature, then centrifuge for 1 min at 16,000 rcf to elute the RNA. Repeat this step with a second 10 ml volume of 95°C nuclease-free water, incubate, and centrifuge. 15. Remove the filter and place the sample on ice. 16. Add 2 ml of 10× DNase I buffer and 1 ml of DNase I to the sample and mix by gently flicking the tube. Centrifuge briefly and incubate for 20 min in a heat block at 37°C. During the incubation, remove the DNase Inactivation Reagent from the freezer and thaw at room temperature. 17. Remove the sample from the heat block. Vigorously vortex the DNase Inactivation Reagent and add 2.3 ml to the sample. Gently tap the side of the tube to mix and incubate for 2 min at room temperature. After 1 min, vortex the sample, tap the tube to move all contents to the bottom, and continue the incubation for 1 min. 18. Centrifuge the sample for 1 min 30 s at 16,000 rcf to pellet the DNase Inactivation Reagent. Transfer the supernatant containing the RNA to a new tube without disturbing the pellet, then place the RNA on ice (see Note 14). 3.4. Assessing RNA Integrity

1. Remove an aliquot of the Pico Ladder from the freezer and thaw on ice. Remove the Pico Gel Matrix, Pico Dye Concentrate, Pico Conditioning Solution, and Pico Marker from 4°C and allow the reagents to warm to room temperature for at least 30  min. Ensure that the Dye Concentrate is shielded from light (see Note 15). 2. Add 550 ml of Gel Matrix to a Spin Filter and centrifuge for 10  min at 1,500 rcf. Aliquot 65  ml of filtered gel into the tubes provided with the kit (produces seven to eight tubes of filtered gel). The filtered gel may be stored at 4°C for up to 2 months. 3. Vortex the tube of Dye Concentrate for 10 s and then centrifuge briefly. Add 1 ml of Dye Concentrate to a tube of filtered gel (warmed to room temperature), vortex for 10 s, then centrifuge for 10 min at 16,000 rcf. One tube of gel-dye mix can be used to run two chips per day. 4. Transfer 1.25–1.5  ml of each RNA sample into a 0.65-ml tube. Heat the sample for 2 min in a heat block at 65–70°C. Place on ice for ~5 min to cool, then centrifuge briefly to collect the RNA at the bottom of the tube. 5. Start the 2100 Expert Software and turn on the Bioanalyzer. Place an electrode cleaner containing 350 ml of nuclease-free water in the instrument and close the lid (see Note 16). On the instrument menu, select “Assays,” “Electrophoresis,”

26

L.A. Field et al.

“RNA,” and finally “Eukaryotic Total RNA Pico Series.” Select the number of samples (from 1 to 11) to be assayed. Enter the sample information and any additional comments pertaining to that sample. 6. Place the Pico Chip on the chip priming station (ensuring that the base plate is on “C”) and pull the syringe back to 1  ml. Add 9  ml of gel-dye mix to the well labeled with an encircled “G” (see Note 17). Close the chip priming station until you hear a click, then press the syringe down until it is secured beneath the syringe clip. After 30 s, release the clip, wait 5 s, and pull the syringe back to the 1 ml mark. 7. Add 9 ml of gel-dye mix to the two remaining wells marked “G.” Add 9  ml of Pico Conditioning Solution to the well marked “CS.” Add 5 ml of Pico Marker to the ladder well and to each well that will contain an RNA sample. Add 6 ml of Pico Marker to any empty sample wells. 8. Add 1 ml of diluted Pico Ladder to the ladder well and 1 ml of sample to the appropriate sample well. After loading all wells, vortex the chip using the manufacturer-supplied vortex for 1 min at 2,400 rpm. During this time, remove the electrode cleaner from the instrument. Place the Pico Chip on the Agilent 2100 Bioanalyzer and begin the run by pressing “Start” (see Notes 18 and 19) (Fig. 2). 3.5. Amplification of RNA from Frozen Tissue

When using small amounts of RNA for gene expression analysis, it is often necessary to first amplify the RNA to generate sufficient material for hybridization to the microarray. For RNA isolated

Fig. 2. Electropherogram of total RNA isolated from frozen breast tissue collected via laser microdissection using the RNAqueous®-Micro kit. The RNA (RIN = 8.7) was assayed on the Bioanalyzer using a Pico Chip. The 18S rRNA and 28S rRNA peaks are visible near 2,000 and 4,000 nucleotides (nt), respectively.

2  Laser Microdissection for Gene Expression Profiling

27

from laser-microdissected tissues, two rounds of amplification are normally required. All frozen reagents for the amplification protocol should be thawed on ice; enzymes should be stored at −20°C immediately prior to and after use. All master mixes should be prepared in excess (generally ~5%) to avoid running short of master mix when working with large numbers of samples. 1. Completely thaw the Poly-A Control Stock on ice, then add 2 ml to a small tube. Add 38 ml of nuclease-free water and mix well by vortexing or flicking the tube. Centrifuge briefly to collect the liquid at the bottom of the tube. This is the first dilution and can be stored at −80°C for up to 6  weeks (or eight freeze–thaw cycles). 2. Remove 2 ml of the first dilution and place in a new tube. Add 98 ml of nuclease-free water to make the second dilution. Mix well and centrifuge briefly. 3. Combine 2 ml of the second dilution with 98 ml of nucleasefree water to make the third dilution. Mix well and centrifuge. 4. Combine 2 ml of the third dilution with 18 ml of nuclease-free water to prepare the fourth dilution. Mix well and centrifuge. 5. Combine 2 ml of the fourth dilution with 18 ml of nucleasefree water to prepare the fifth dilution. Mix well and centrifuge (see Note 20). 6. Using the estimated RNA concentration obtained from the Bioanalyzer, calculate the volume of sample containing 10 ng of RNA (see Note 21). Transfer this volume to a 0.2 ml PCR tube and adjust the total volume to 9 ml with nuclease-free water. If the volume needed for 10 ng of RNA is greater than 9 ml, transfer this amount to a hydrophobic, nonstick microcentrifuge tube, and centrifuge in a vacuum concentrator until the volume is £9 ml. Transfer the concentrated sample to a 0.2-ml PCR tube and adjust the volume to 9 ml with nuclease-free water. 7. Flick the tubes to mix and centrifuge briefly to collect the liquid at the bottom of the tube. 3.6. First Round Amplification

1. Add 2 ml of the fifth dilution of the Poly-A Controls to each sample containing 10  ng of RNA (see Note 22). Flick the tubes to mix and centrifuge briefly. 2. Add 1 ml of Oligo(dT) primer to each sample, flick the tubes to mix, and centrifuge briefly. Incubate samples for 10 min at 70°C in a thermal cycler. 3. Remove samples from the thermal cycler, centrifuge briefly, and place on ice.

28

L.A. Field et al.

4. In a small tube, prepare a master mix containing the following for each sample: (a) 2 ml 10× first strand buffer. (b) 4 ml dNTP mix. (c) 1 ml RNase inhibitor. (d) 1 ml ArrayScript™. Vortex the tube to mix and centrifuge briefly to collect the contents at the bottom of the tube. Add 8 ml of the master mix to each sample, flick the tubes to mix, and centrifuge. Incubate samples for 2  h at 42°C in an air incubator or hybridization oven, then centrifuge briefly, and place on ice. 5. Prepare a master mix on ice containing the following reagents for each sample: (a) 63 ml nuclease-free water. (b) 10 ml 10× second strand buffer. (c) 4 ml dNTP mix. (d) 2 ml DNA polymerase. (e) 1 ml of RNase H. Vortex to mix and centrifuge briefly to collect the master mix at the bottom of the tube. Add 80 ml of master mix to each sample, flick the samples to mix, and centrifuge briefly. Incubate the samples in a precooled thermal cycler for 2 h at 16°C (see Note 23), then centrifuge briefly and place on ice. 6. Place a tube containing at least 30 ml of nuclease-free water per sample in a heat block set to 50–55 ° C. For each sample, place a filter inside a cDNA Elution tube. Note: add 24 ml of 100% ethanol to the Wash buffer before using for the first time. 7. Transfer the samples from the 0.2 ml tubes to 1.5 ml microcentrifuge tubes. Add 250  ml of cDNA Binding buffer to each sample, mix by pipetting up and down and then flicking the tubes several times. Centrifuge samples briefly, then transfer each sample to the filter of a cDNA Filter Cartridge. Centrifuge samples for 1 min at 10,000 rcf, then discard the flow-through. 8. Add 500 ml of Wash buffer to each filter. Centrifuge for 1 min at 10,000 rcf and discard the flow-through. 9. Centrifuge the cDNA Filter Cartridges for 1 min at 10,000 rcf to remove any residual liquid from the filter. Transfer filters to new cDNA Elution tubes and discard the old tubes. 10. Add 10 ml of nuclease-free water warmed to 50–55°C to the center of each filter. Incubate for 2 min at room temperature. Elute samples by centrifuging for 1 min 30 s at 10,000 rcf. Repeat this step using a second 10 ml volume of warm nuclease-free water.

2  Laser Microdissection for Gene Expression Profiling

29

11. Discard filters and place tubes containing the eluted cDNA on ice. 12. Prepare the in vitro transcription (IVT) master mix at room temperature. Note that for the first round of amplification, the IVT reactions contain only unmodified dNTPs. For each sample include: (a) 4 ml T7 ATP. (b) 4 ml T7 CTP. (c) 4 ml T7 GTP. (d) 4 ml T7 UTP. (e) 4 ml T7 10× reaction buffer. (f) 4 ml T7 enzyme mix. Vortex the master mix and centrifuge briefly to collect the contents at the bottom of the tube. Aliquot 24 ml of master mix to each sample, flick the tubes to mix, and centrifuge briefly. Incubate samples for 14 h in an air incubator or hybridization oven at 37°C. 13. Place a tube containing nuclease-free water in a heat block at 50–55°C – we recommend heating at least 120 ml of nucleasefree water per sample. 14. For each sample, place an aRNA Filter Cartridge in an aRNA Collection Tube. 15. Remove the IVT reactions from the incubator. Add 60 ml of nuclease-free water to each sample, mix by flicking the tube, and centrifuge briefly. Add 350 ml of aRNA Binding buffer followed by 250 ml of 100% ethanol to each sample. Mix the samples by pipetting up and down at least five times, then transfer each sample to an aRNA Filter Cartridge. Centrifuge samples for 1 min at 10,000 rcf, then discard the flow-through and remount the filter on the collection tube. 16. Add 650 ml of wash buffer to each Filter Cartridge and centrifuge for 1 min at 10,000 rcf. Discard the flow-through and place the Filter Cartridge back inside the collection tube. Centrifuge samples for an additional 1 min at 10,000 rcf to remove residual wash buffer. Discard the flow-through and place the Filter Cartridge in a new collection tube. 17. Apply 100 ml of nuclease-free water warmed to 50–55°C to the center of each filter. Incubate at room temperature for 2 min, then centrifuge for 1 min at 10,000 rcf to elute the aRNA. 18. Remove 3 ml of the aRNA and transfer to a small tube. Heat the samples for 2 min in a heat block at 65–70°C. Place samples on ice to cool, then centrifuge the samples briefly, and return to ice.

30

L.A. Field et al.

Fig. 3. Electropherograms of RNA isolated from frozen breast tissue following one and two rounds of amplification. (a) Total RNA amplified using the MessageAmp™ II aRNA Amplification kit and assayed on the Bioanalyzer using a Pico Chip. (b) Second round aRNA assayed using a Nano Chip. The majority of the second round aRNA product should be >500 nucleotides (nt) in length.

19. Run 1 ml of the first round aRNA samples from step 18 on the Bioanalyzer using a Pico Chip following the instructions outlined above (Fig. 3a). Use 1.5 ml of the remaining aRNA to measure the concentration of each sample on a NanoDrop ND-1000 Spectrophotometer. 3.7. Second Round Amplification

1. Calculate the volume of first round aRNA (using the concentration obtained on the NanoDrop) needed to obtain 1 mg of starting material for the second round of amplification

2  Laser Microdissection for Gene Expression Profiling

31

(see Note 24). If this volume exceeds 10 ml for any sample, concentrate those samples in a vacuum concentrator to less than 10 ml. In a 0.2-ml PCR tube, adjust the volume of all samples to 10 ml using nuclease-free water. 2. Add 2  ml of second round primers to each aRNA sample. Flick the tubes to mix and centrifuge briefly. Place samples in a thermal cycler heated to 70°C for 10 min, then centrifuge briefly and place on ice. 3. Prepare a master mix containing the following for each sample: (a) 2 ml 10× first strand buffer. (b) 4 ml dNTP mix. (c) 1 ml RNase inhibitor. (d) 1 ml ArrayScript™. Vortex the master mix and centrifuge briefly. Add 8  ml of master mix to each sample and flick the tubes to mix. Centrifuge briefly and incubate for 2  h at 42°C in an air incubator or hybridization oven. 4. Following incubation, centrifuge the samples briefly and place on ice. Add 1 ml of RNase H to each sample, flick the tubes to mix, and centrifuge briefly to collect the contents at the bottom of the tube. Incubate samples for 30 min at 37°C in an air incubator or hybridization oven, then centrifuge briefly and place on ice. 5. Add 5 ml of the Oligo(dT) primer to each sample, flick the tubes to mix, and centrifuge briefly. Incubate samples for 10 min at 70°C in a thermal cycler, then centrifuge and place on ice. 6. Prepare a master mix on ice for the second strand synthesis that includes the following for each sample: (a) 58 ml nuclease-free water. (b) 10 ml 10× second strand buffer. (c) 4 ml dNTP mix. (d) 2 ml DNA polymerase. Vortex to mix and add 74 ml to each sample. Flick the tubes to mix and centrifuge briefly. Incubate samples for 2 h in a thermal cycler that has been precooled to 16°C. Remove samples from the thermal cycler, centrifuge briefly and place on ice. 7. Purify the cDNA following the exact procedure outlined above.

32

L.A. Field et al.

8. Prepare a master mix at room temperature that contains for each sample: (a) 4 ml T7 ATP. (b) 4 ml T7 CTP. (c) 4 ml T7 GTP. (d) 2.6 ml T7 UTP. (e) 1.4 ml biotin-11-UTP. (f) 4 ml T7 reaction buffer. (g) 4 ml T7 enzyme mix. Vortex and centrifuge briefly to collect the contents at the bottom of the tube. Add 24 ml of master mix to each sample, flick the tubes to mix, and centrifuge briefly. Incubate the samples for 14 h at 37°C in an air incubator or hybridization oven. 9. Purify the second round, labeled aRNA using the same procedure for aRNA purification outlined above. 10. Run the second round, labeled aRNA on the Bioanalyzer using the Agilent RNA 6000 Nano kit. Reagent and sample preparation for the Nano Chip is very similar to that for the Pico Chip with minor exceptions. Warm all refrigerated Nano reagents to room temperature. Prepare the filtered gel and gel-dye mix using the Nano Gel Matrix and Nano Dye Concentrate as outlined above. 11. Thaw the Nano Ladder on ice. Flick the tube several times and centrifuge briefly. Transfer 2.5 ml of the Nano Ladder to a new tube. Prepare 5–10 ml aliquots of the remaining ladder and store at −20°C for future use. Place 3 ml of each aRNA sample in a small tube. Heat the samples and ladder for 2 min in a heat block at 65–70°C. Place the samples on ice to cool, then centrifuge briefly to collect any condensation at the bottom of the tube. 12. Start the 2100 Expert Software and clean the electrodes as previously described. Select the “Eukaryotic Total RNA Nano Series Assay” and select the number of samples (from 1 to 12) that will be run. Enter the sample information. 13. Load the Nano Chip using the same procedure for loading the Pico Chip, but note that the Nano Chip does not use Conditioning Solution, allowing 12 samples to be run (Fig. 3b). 14. Measure the concentration of 1.5 ml of the remaining second round aRNA on the NanoDrop. 3.8. Fragmentation of Labeled aRNA

1. For each sample, transfer 15 mg of labeled aRNA to a 0.2-ml PCR tube. The aRNA yield after two rounds of amplification typically exceeds 1,500  ng/ml and often is ~2,000  ng/ml; therefore, you should not have to vacuum-concentrate the samples prior to fragmentation. Add nuclease-free water to

2  Laser Microdissection for Gene Expression Profiling

33

Fig. 4. Electropherogram of fragmented, second round aRNA assayed on the Bioanalyzer using a Nano Chip. The aRNA fragments should be ~35 to 200 nucleotides (nt) in length.

each sample to bring the total volume to 24 ml. Flick the tubes to mix and centrifuge briefly. 2. Add 6 ml of 5× Fragmentation buffer to each sample, flick the tubes to mix, and briefly centrifuge. Fragment the aRNA samples by incubating for 35 min at 94°C in a thermal cycler. Place the samples on ice to cool, then centrifuge briefly to collect condensation at the bottom of the tube. 3. Run the fragmented samples on a Bioanalyzer Nano Chip (see Note 25) (Fig. 4). 3.9. Hybridization of Fragmented aRNA to Microarrays

Microarrays for gene expression analysis are available from several commercial vendors, including Affymetrix, Agilent Technologies, and Illumina. In this chapter, we provide protocols for hybridization of labeled, fragmented aRNA to the Affymetrix HG U133A 2.0 arrays. The protocol is the same for all eukaryotic arrays manufactured by Affymetrix; however, the starting amount of fragmented aRNA and volume of reagents used in certain steps will vary across the different array formats. 1. Thaw the 20× Eukaryotic Hybridization Controls on ice. Mix well by flicking the tube or vortexing gently and then centrifuge briefly to collect the contents at the bottom of the tube. Prepare aliquots containing the volume used during a typical hybridization set-up (we usually hybridize eight samples at a time) and store at −20°C. 2. Prepare 1 l of 12× MES stock buffer by dissolving 64.61 g of MES hydrate and 193.3 g of MES sodium salt in 800 ml of nuclease-free water. When completely dissolved, adjust

34

L.A. Field et al.

the volume to 1,000  ml with nuclease-free water and filter through a 0.22-mm filter. The pH should be between 6.5 and 6.7. Store at 4°C wrapped in foil to shield from light. Discard the solution if it turns yellow. 3. Prepare 50 ml of 2× hybridization buffer by combining: (a) 8.3 ml 12× MES stock buffer. (b) 17.7 ml 5 M NaCl. (c) 4.0 ml 0.5 M EDTA. (d) 19.9 ml nuclease-free water. (e) 100 ml 10% Tween 20. Filter the solution through a 0.22-mm filter and store wrapped in foil at 4°C. 4. Approximately 1  h before setting up the hybridization, remove the HG U133A 2.0 arrays from the refrigerator and allow them to equilibrate to room temperature. Thaw the Control Oligonucleotide B2, 20× Eukaryotic Hybridization Controls, Herring Sperm DNA, and BSA on ice. Set the temperature of the hybridization oven to 45°C. 5. Transfer 20  ml (10  mg) of fragmented aRNA to a 1.5-ml microcentrifuge tube. 6. Prepare a 1× solution of hybridization buffer by combining equal volumes of 2× hybridization buffer and nuclease-free water; vortex to mix well. 7. Heat the 20× Eukaryotic Hybridization Controls for 5 min at 65°C in a heat block before adding to the hybridization master mix. 8. Prepare the hybridization master mix (at 5% excess) by combining for each sample: (a) 3.3 ml Control Oligonucleotide B2. (b) 10 ml 20× Eukaryotic Hybridization Controls. (c) 2 ml Herring Sperm DNA. (d) 2 ml BSA. (e) 100 ml 2× hybridization buffer. (f) 20 ml DMSO. (g) 42.7 ml nuclease-free water. Vortex to mix well, centrifuge briefly, and add 180 ml of the master mix to each fragmented sample (from step 2 above). Vortex the samples, centrifuge briefly, and incubate for 5 min at 99°C in a heat block. 9. Place the arrays face down on a laboratory tissue. Insert a pipette tip into the top right septum on the back of the array to permit air to vent when filling the array. Fill the arrays with 160 ml of 1× hybridization buffer, remove the pipette tip vent,

2  Laser Microdissection for Gene Expression Profiling

35

and incubate the arrays for 10 min at 60 rpm in the hybridization oven set to 45°C. 10. Transfer the samples from 99°C to a 45°C heat block. Incubate for 5 min, then centrifuge the samples for 5 min at 16,000 rcf. 11. Remove the arrays from the hybridization oven. Insert the pipette tip vent and remove the hybridization buffer. Load 130 ml of the sample into the array, making sure that a bubble is present that can freely move when the array is slowly tilted from side to side. Incubate the arrays for 16 h at 45°C in the hybridization oven while rotating at 60 rpm (see Notes 26 and 27). 3.10. Washing, Staining, and Scanning of Microarrays

1. Prepare 250 ml of 2× stain buffer by combining: (a) 41.7 ml 12× MES stock buffer. (b) 92.5 ml 5 M NaCl. (c) 2.5 ml 10% Tween 20. (d) 113.3 ml nuclease-free water. Filter through a 0.22-mm filter, wrap in foil to shield from light, and store at 4°C. 2. Reconstitute the goat IgG by adding 150 mM NaCl to make a 10  mg/ml solution. For example, add 1  ml of 150  mM NaCl to 10 mg of lyophilized goat IgG. Aliquots should be stored at −20°C, but once thawed for use, store at 4°C. 3. Add 1 ml of nuclease-free water to reconstitute the biotinylated antistreptavidin antibody to 0.5 mg/ml. Gently pipette up and down to mix, then store at 4°C. 4. Prepare 1 l of Wash A solution by combining: (a) 300 ml 20× SSPE. (b) 699 ml deionized water. (c) 1 ml 10% Tween 20. Filter through a 0.22-mm filter and store at room temperature. 5. Prepare 1 l of Wash B solution by combining: (a) 83.3 ml 12× MES stock buffer. (b) 5.2 ml 5 M NaCl. (c) 1 ml 10% Tween 20. (d) 910.5 ml deionized water. Filter through a 0.22-mm filter, cover with foil, and store at 4°C. 6. Prepare a master mix (in 5% excess) of the SAPE Solution (stains 1 and 3) by combining for each sample: (a) 600 ml 2× stain buffer. (b) 48 ml BSA. (c) 12 ml SAPE (vortex well before pipetting).

36

L.A. Field et al.

(d) 540 ml of deionized water. Vortex to mix. For each sample, aliquot 600 ml of the SAPE Solution into two 1.5 ml microcentrifuge tubes. 7. Prepare a master mix for the Antibody Solution (Stain 2) in 5% excess. For each sample, combine: (a) 300 ml 2× stain buffer. (b) 24 ml BSA. (c) 6 ml IgG. (d) 3.6 ml biotinylated antistreptavidin. (e) 266.4 ml deionized water. Mix well by vortexing and aliquot 600 ml of the Antibody Solution master mix into a 1.5-ml microcentrifuge tube for each sample. 8. After 16  h, remove the arrays from the hybridization oven. Insert the pipette tip vent and remove the sample solution from the arrays. Fill the arrays with 160 ml of Wash A solution. 9. Start the GeneChip® Operating System (GCOS) or Affymetrix® GeneChip® Command Console® (AGCC) and enter the sample and experiment information. 10. Replace the appropriate water bottles on the fluidics station with Wash A solution and Wash B solution. Prime all modules on the fluidics station. 11. Select the appropriate wash and stain protocol for your arrays and press “Run.” Load the arrays and staining tubes onto the fluidics station. The protocol will automatically run until staining of the arrays is complete (usually ~1 h and 15 min). 12. Remove the arrays from the fluidics station. Place Tough Spots over the septa on the back of the arrays to prevent leakage during scanning. Once the scanner has warmed up (10– 15 min), scan the arrays using the GCOS or AGCC software. When scanning is complete, zoom in on each scanned image and check the entire image for any abnormalities. The scanned images can now be analyzed to generate signal intensities for the probes on the arrays as well as a QC report. 13. Once the wash and stain protocol is complete, replace the Wash A solution and Wash B solution with deionized water. Run the Shutdown protocol on all modules, and when finished, turn off the fluidics station. 3.11. Quantitative Real-Time Polymerase Chain Reaction Assays Using RNA from Frozen Tissue

Gene expression differences identified by microarray analysis are frequently confirmed using qRT-PCR. Due to the low yield of RNA following laser microdissection of small tissue sections, RNA is amplified for a single round prior to performing qRTPCR. To amplify the RNA, follow the procedure detailed above; however, it is not necessary to add the Poly-A controls.

2  Laser Microdissection for Gene Expression Profiling

37

Note that there are several different chemistries available for performing qRT-PCR, including TaqMan® probes, Molecular Beacons, Scorpions®, or SYBR® Green. The following protocol uses TaqMan® Gene Expression Assays, which utilize the same PCR conditions so there is no need to design PCR primers or optimize PCR conditions. In order to achieve accurate results from qRT-PCR, be careful and precise when performing each step in the protocol, especially pipetting. 1. Reverse transcribe the aRNA (one round of amplification) using the High-Capacity cDNA Reverse Transcription Kit. Up to 2 mg of RNA (or aRNA) can be reverse transcribed (see Note 28). In addition, reverse transcribe an appropriate amount of FirstChoice® Human Brain Reference RNA or other reference RNA which will be used to calibrate the relative levels of gene expression in the samples of interest. 2. If necessary, vacuum concentrate the samples to £10 ml and increase the volume to 10  ml with nuclease-free water in a 0.2-ml PCR tube. 3. Prepare the reverse transcription master mix (in 5% excess) containing the following for each sample: (a) 2 ml 10× RT buffer. (b) 0.8 ml 25× dNTP mix. (c) 2 ml 10× RT random primers. (d) 1 ml MultiScribe Reverse Transcriptase. (e) 1 ml RNase inhibitor. (f) 3.2 ml nuclease-free water. Vortex to mix, centrifuge briefly, and add 10 ml of the master mix to each sample. 4. Flick the sample tubes to mix and centrifuge briefly to collect the contents at the bottom of the tube. Incubate the samples in a thermal cycler using the following program: 10 min at 25°C, 2 h at 37°C, 5 s at 85°C, and then hold at 4°C. 5. When the reverse transcription reaction is complete, add nuclease-free water to adjust the concentration of the samples and the reference to 5 ng/ml. For example, if 1 mg of aRNA was reverse transcribed in a 20-ml reaction, add 180 ml of water to bring the concentration to 1 mg/200 ml or 5 ng/ml. 6. Mix the diluted samples well and centrifuge briefly. Place on ice or store at −20°C. 7. Perform the qRT-PCR using the TaqMan® Gene Expression Assay of interest (see Note 29). These assays can be run on a variety of real-time instruments, including the Bio-Rad iCycler™. Turn the iCycler™ on and allow it to warm up for at

38

L.A. Field et al.

least 15 min. Enter the plate set-up sample information, select and load the appropriate fluorophore (FAM-490), and save the file. 8. qRT-PCR reactions should be performed in duplicate, but if sufficient starting material is available, triplicate or quadruplicate reactions are recommended. The following protocol is based on 10 ng of cDNA per reaction (see Note 30). Pipette 2  ml of cDNA (5  ng/ml) into the appropriate well of an iCycler™ PCR plate. 9. Prepare a master mix containing the following per reaction: (a) 2.5 ml TaqMan® Gene Expression Assay (20×). (b) 20.5 ml nuclease-free water. (c) 25 ml TaqMan® Universal PCR Master Mix. When running a full 96-well plate, make a master mix for 100 reactions (excess of four reactions). Vortex the master mix and centrifuge briefly to bring the contents to the bottom of the tube. Add 48 ml of master mix to each sample well. Place a thermal seal on the plate, centrifuge the plate for ~10 s at ~3,000 rcf, and then load the plate on the iCycler™. Set the reaction volume to 50  ml and run the following protocol: 95°C for 10 min, 50 cycles of 95°C for 15 s followed by 60°C for 1 min, then hold at 4°C. 10. Average the Ct values for each sample and the reference (see Note 31). Calculate the relative transcript levels for each sample using the Comparative Ct Method (see Note 32). 3.12. RNA Isolation from Formalin-Fixed Paraffin-Embedded Specimens

Archived FFPE tissues represent a valuable source of molecular information for the study of cancer (6–8). Although formalin fixation and paraffin embedding preserves tissue structure and cellular morphology, protein–protein and protein–nucleic acid cross links form during the preservation process, which chemically modify and damage the nucleic acids. RNA is particularly susceptible to fragmentation by FFPE, often making RNA isolated from FFPE tissues unusable for molecular analysis. The following protocol is specifically designed to recover RNA from FFPE samples that can be used for downstream applications such as qRT-PCR. 1. Add 42 ml of 100% ethanol to the Wash 1 concentrate and 48  ml of 100% ethanol to the Wash 2/3 concentrate. Mix well. 2. Place the laser microdissected FFPE samples in 400  ml of digestion buffer. Add 4 ml of protease to each sample and mix the tubes gently ensuring that the samples are entirely immersed in buffer. Incubate the samples in a heat block set to 50°C for 3 h (see Note 33).

2  Laser Microdissection for Gene Expression Profiling

39

3. Add 480 ml of isolation additive to each sample and vortex to mix. The solution should be white and cloudy (see Note 33). 4. Add 1.1 ml of 100% ethanol and mix by carefully pipetting the sample up and down. The sample should become clear after mixing (see Note 33). 5. Place a filter cartridge inside a collection tube provided with the kit. Apply 700 ml of the sample to the filter, centrifuge for 1 min at 10,000 rcf, then discard the flow-through and return the filter to the same collection tube. 6. Apply an additional 700 ml of sample to the filter and centrifuge for 1 min at 10,000 rcf. Continue this process until the entire sample has been passed through the filter. 7. Add 700  ml of Wash 1 solution to the Filter Cartridge and centrifuge for 30 s at 10,000 rcf. Discard the flow-through. 8. Add 500 ml of Wash 2/3 solution to the filter and centrifuge for 30 s at 10,000 rcf. Discard the flow-through. Centrifuge the Filter Cartridge assembly an additional 30 s at 10,000 rcf to remove residual liquid from the filter. 9. Treat the samples in the filters with DNase by combining for each sample: (a) 6 ml 10× DNase buffer. (b) 4 ml DNase. (c) 50 ml nuclease-free water. Add this solution to the center of each filter and incubate for 30 min at room temperature. 10. Add 700  ml of Wash 1 to the filters, incubate for 1  min at room temperature, then centrifuge for 30  s at 10,000 rcf. Discard the flow-through. 11. Add 500 ml of Wash 2/3 to the filters and centrifuge for 30 s at 10,000 rcf. Discard the flow-through. Add an additional 500  ml of Wash 2/3 to the filters, centrifuge for 1  min at 10,000 rcf, then transfer the filters to new collection tubes. 12. Add 30 ml of nuclease-free water heated to 95°C to the center of each filter (see Note 33). Incubate at room temperature for 1 min and then centrifuge for 1 min at 16,000 rcf. Repeat this step using a second 30  ml volume of heated nuclease-free water. 13. Read the concentration of the eluted samples on the NanoDrop and run the samples on a Bioanalyzer Pico Chip as described in Subheading 3.2.4 (Fig. 5). 14. Store samples at −80°C.

40

L.A. Field et al.

Fig. 5. Electropherogram of total RNA isolated from formalin-fixed paraffin-embedded breast tissue following laser microdissection. RNA was isolated using the RecoverAll™ Total Nucleic Acid Isolation Kit and assayed on the Bioanalyzer using a Pico Chip. Note that the 18S rRNA and 28S rRNA are completely degraded and thus not distinguishable.

3.13. Quantitative Real-Time Polymerase Chain Reaction Assays Using RNA from FFPE

RNA extracted from FFPE tissues is usually not conducive to amplification using an oligo(dT) primer due to fragmentation and degradation. Therefore, qRT-PCR is performed directly on RNA from FFPE tissues – there is no RNA amplification step. 1. Reverse transcribe the FFPE RNA as described. 2. Following reverse transcription of the FFPE RNA to cDNA, perform qRT-PCR as outlined for frozen tissue, but increase the number of PCR cycles to 60 cycles. 3. Average the Ct values for each sample and the reference as described above. Calculate the relative transcript levels for each sample using the Comparative Ct Method.

4. Notes 1. Flash-frozen tissues can only be sectioned after embedding in OCT. To embed a flash-frozen tissue sample in OCT, place the tissue and plastic cryomold in the cryostat set to −30°C. Cover the bottom of the cryomold with OCT, orient the tissue so that the region of interest is facing up, then fill the mold with additional OCT. Allow the compound to solidify – the block will turn white in color when completely frozen. 2. The foil slides should be at room temperature before applying a tissue section. Tissues will not properly adhere to cold slides. Note: If you decide to place more than one tissue section on

2  Laser Microdissection for Gene Expression Profiling

41

a slide to maximize resources, keep the slide in the cryostat while placing the additional tissue section(s) onto the slide. If the additional section does not adhere to the cold slide, warm the area of the slide where the tissue will be deposited by pressing your thumb to the reverse side. Work as quickly as possible. 3. If the OCT is not completely dissolved after the 50% ethanol wash, make sure that the solution is mixed properly. Shake the original container, add fresh solution, and wash the slide again for 30 s. 4. When staining multiple slides per patient sample, change the second 100% ethanol solution when it is noticeably purple in color, indicating that it is no longer a 100% solution. A 100% solution prevents leaching of stain from the slide and allows for faster drying. 5. Incubation at 56°C heat-fixes the section to the slide. The section will fall off the slide during staining if not properly fixed to the slide. 6. Because the slide has no cover-slip, histologic features may not be easily identified. Cellular details can be better visualized by pipetting 100% ethanol onto the tissue section. 7. Large microdissected areas can be lifted directly from the slide with a clean pair of forceps and placed into 100 ml of buffer. We generally perform LM at 4× magnification (after careful inspection at 10×) when working with frozen breast tissue. If the laser tends to drift from the user-defined cut lines, calibrate the laser periodically. With Laser Microdissection LMD software version 4.4, the laser control settings we typically use at 4× are aperture, 15; intensity, 44; speed, 6 for clean cuts; offset, 18; bridge, medium; aperture differential, 6. 8. The lysis solution and wash solutions may be stored at 4°C or room temperature, but be sure to warm the reagents to room temperature before beginning the protocol. 9. The Pico Ladder should only be heated when first diluted and aliquoted and may not run properly if heated multiple times. 10. Completely submerge the foil containing the microdissected tissue in the lysis solution to ensure complete cell lysis and inactivation of endogenous RNases. If this is not achieved by centrifugation, push the foil into the buffer using a pipette tip. 11. To isolate both large and small RNA species including tRNAs or microRNAs, add 129  ml of 100% ethanol rather than 52 ml. 12. When transferring the sample to the filter, be sure not to transfer large pieces of foil that may block the flow of liquid through the filter.

42

L.A. Field et al.

13. To increase accuracy when pipetting hot liquids, prewet the pipette tip before pipetting the 95°C water and applying it to the filter. 14. To easily remove the DNase Inactivation Reagent from the RNA and maximize RNA recovery, centrifuge the sample to pellet the Reagent. Transfer ~20 ml of RNA to a new tube, centrifuge the tube containing the Reagent for an additional 30 s to 1 min, then transfer the remaining RNA using a small bore pipette tip (10XL tips work well) without disturbing the DNase Inactivation Reagent pellet. 15. When running a Pico or Nano Chip, ensure that the reagents have not expired and that no more than 2 months have passed since the gel was filtered. Expired reagents can adversely affect the run and can cause aberrations to the baseline of the electropherogram. 16. When cleaning the Bioanalyzer electrodes prior to running a Pico or Nano Chip, fill the electrode cleaner with nucleasefree water, not RNaseZap®. The Pico Chip is very sensitive and even minute amounts of RNaseZap® can interfere with proper function. 17. When loading the Pico or Nano Chips, do not use the blowout function of the pipette as this may introduce air bubbles that may interfere with the chip running properly. 18. Most RNA samples isolated from frozen tissue via laser microdissection have a RIN (RNA Integrity Number) greater than 7, which is acceptable for downstream applications. Samples with a lower RIN (between 6 and 7) may be usable. Repeat the RNA isolation if significant degradation has occurred. Remove the sample from the study if subsequent isolations do not show improved RNA quality. 19. The pin set on the Bioanalyzer will need periodic maintenance – remove and clean thoroughly with RNaseZap®, then rinse thoroughly with nuclease-free water. 20. The final volume of this dilution may be adjusted based on the number of samples that will be amplified at one time. For example, if amplifying ten or more samples, prepare 40 ml of the fifth dilution by mixing 4 ml of the fourth dilution with 36 ml of nuclease-free water. 21. We typically use 10 ng of RNA as input for the first round of amplification; however, we have amplified as little as 2  ng with more than sufficient yields of second round aRNA. 22. If the starting amount for amplification is greater or less than 10 ng, adjust the volume of the Poly-A Controls proportionately. For example, if starting with 5 ng of RNA, add 1 ml of the fifth dilution to the sample.

2  Laser Microdissection for Gene Expression Profiling

43

23. Ensure that the thermal cycler lid is also cooled to 16°C. Exposure of the reaction to higher temperatures will lead to inefficient second-strand cDNA synthesis and will compromise aRNA yield. If the lid temperature cannot be adjusted to 16°C, turn the lid temperature off or incubate with the lid off. 24. We have started the second round of amplification with as little as 350 ng of first round aRNA with successful results. Any remaining aRNA from the first round of amplification can be stored at −80°C for future use. 25. Because the samples have just been heated to 94°C, it is not necessary to heat denature them at 70°C. The heat-denatured ladder can be used in this run without reheating. The same gel-dye mix that was prepared to run the Nano Chip for the second round aRNA can be used to run the Nano Chip for the fragmented aRNA if both chips are run on the same day. 26. Before placing the arrays in the hybridization oven overnight, cover each septum with a small piece of tape to prevent the sample from leaking out of the array. 27. Arrange the arrays in the hybridization oven with a balanced configuration to prevent undue stress to the motor. 28. We typically reverse transcribe sufficient aRNA to run all of the TaqMan® Gene Expression Assays that we have defined for the study so that additional reverse transcription reactions on the same sample are not necessary. 29. We recommend using TaqMan® Gene Expression Assays with the suffix “_ml” if possible, as these assays amplify regions spanning exon junctions and will not amplify genomic DNA that may contaminate the RNA sample. For “_g1” or “_s1” assays, include a control with no reverse transcriptase when performing the reverse transcription reaction and subsequent PCR to ensure that amplification is due to the presence of RNA transcripts and not genomic DNA contamination. 30. 10 ng of cDNA usually produces good results; however, when working with certain low abundance transcripts, the starting amount of cDNA may need to be increased and the volume of water in the master mix adjusted accordingly. 31. In some cases, the chosen reference RNA may not express a particular transcript of interest and an alternative reference RNA will need to be selected. 32. To calculate relative transcript levels using the Comparative Ct Method: (a) Determine the DCt as follows:

DC t = C t target - C t reference

44

L.A. Field et al.

where the target is the gene of interest and the reference represents an endogenous control such as actin or GAPDH. (b) Calculate the DDCt using the following formula:

DDC t = DC t test sample - DC t calibrator (c) The fold change relative to the calibrator is 2- DDC t . As an alternative to the Comparative Ct Method, relative levels of gene expression can be determined using the Relative Standard Curve Method. The following document describes both the Comparative Ct Method and the Relative Standard Curve Method: Applied Biosystems. Guide to Performing Relative Quantitation of Gene Expression Using Real-Time Quanti­tative PCR, available at: h t t p : // w w w 3 . a p p l i e d b i o s y s t e m s . c o m / c m s / g r o u p s / mcb_support/documents/generaldocuments/cms_042380.pdf.

33. This protocol has been modified by the manufacturer. The current RecoverAll™ protocol reduces the volume of digestion buffer from 400 to 100 ml, shortens the incubation from 3 h at 50°C to 15 min at 50°C followed by 15 min at 80°C, reduces the volume of isolation additive from 480 to 120 ml and the volume of 100% ethanol from 1.1 ml to 275 ml, and lowers the temperature of the eluant from 95°C to room temperature (22–25°C).

Acknowledgments This work was supported by the United States Department of Defense (Military Molecular Medicine Initiative MDA W81XWH05-2-0075, protocol #01-20006) and was performed under the auspices of the Clinical Breast Care Project, a joint effort of many investigators and staff members. The opinion and assertions contained herein are the private views of the authors and are not to be construed as official or as representing the views of the Department of the Army or the Department of Defense. References 1. Collins, F. S. and Barker, A. D. (2007) Mapping the cancer genome. Pinpointing the genes involved in cancer will help chart a new course across the complex landscape of human malignancies. Sci. Am. 296, 50–57. 2. Manolio, T. A., Brooks, L. D., and Collins, F. S. (2008) A HapMap harvest of insights into the

genetics of common disease. J. Clin. Invest. 118, 1590–1605. 3. Ellsworth, R. E., Seebach, J., Field, L. A., Heckman, C., Kane, J., Hooke, J. A., et  al. (2009) A gene expression signature that defines breast cancer metastases. Clin. Exp. Metastasis 26, 205–213.

2  Laser Microdissection for Gene Expression Profiling 4. Perou, C. M., Sorlie, T., Eisen, M. B., van de Rijn, M., Jeffrey, S. S., Rees, C. A., et  al. (2000) Molecular portraits of human breast tumours. Nature 406, 747–752. 5. Sørlie, T., Perou, C. M., Tibshirani, R., Aas, T., Geisler, S., Johnsen, H., et  al. (2001) Gene expression patterns of breast carcinomas distinguish tumor subclasses with clinical implications. Proc. Natl. Acad. Sci. USA 98, 10869–10874. 6. Ellsworth, R. E., Ellsworth, D. L., Deyarmin, B., Hoffman, L. R., Love, B., Hooke, J. A., et  al. (2005) Timing of critical genetic changes in human breast disease. Ann. Surg. Oncol. 12, 1054–1060.

45

7. Becker, T. E., Ellsworth, R. E., Deyarmin, B., Patney, H. L., Jordan, R. M., Hooke, J. A., et  al. (2008) The genomic heritage of lymph node metastases: implications for clinical management of patients with breast cancer. Ann. Surg. Oncol. 15, 1056–1063. 8. Ellsworth, R. E., Hooke, J. A., Love, B., Kane, J. L., Patney, H. L., Ellsworth, D. L., et al. (2008) Correlation of levels and patterns of genomic instability with histological grading of invasive breast tumors. Breast Cancer Res. Treat. 107, 259–265.

wwwwwwwwwwwwwww

Chapter 3 Gene Expression Using the PALM System Jian-Xin Lu and Cheuk-Chun Szeto Abstract The PALM Robot MicroBeam laser microdissection system can isolate specified cells from complex tissues section, in a rapid and precise manner. Combined with other methods, PALM may be used for gene expression elucidating the role of specialized cell type in physiological and pathological activity. This chapter describes the application of the PALM MicroBeam system to isolate RNA from cells in a complex tissue for subsequent gene expression analysis. Protocols show the steps from preparation of tissue samples to the final quantitative results. The process is articulated in several steps, each of which requires optimal choices in order to obtain reliable data from a limited number of cells (500–10,000 cells). Furthermore, the notes regarding tissue preparation, microdissection of the interested cells, are also emphasized. Key words: Gene expression, PALM, Microdissection, RNA, Real-time quantitative PCR

1. Introduction The PALM® Robot MicroBeam laser microdissection system (P.A.L.M. Carl Zeiss, Bernried, Germany) provides a convenient, efficient, and precise method of selecting specified cell populations from complex tissues for subsequent analysis of their RNA, DNA, or protein content. Microdissection of cells defined under the microscope ensures to obtain pure samples of cells of interest for downstream molecular applications. Molecular biological techno­ logy developments (e.g., PCR) allow analyzing gene expression from only limited amounts of cells. Thereby, the combination of microdissection and PCR may allow the possibility of assessing the role of specialized cell type or tissue in the normal physiologic condition or under disease process (1–3). Thus, the details of the disease’s pathogenesis, diagnostic and prognostic accuracy, and ultimately targeted therapeutics may be elucidated much clearer.

Graeme I. Murray (ed.), Laser Capture Microdissection: Methods and Protocols, Methods in Molecular Biology, vol. 755, DOI 10.1007/978-1-61779-163-5_3, © Springer Science+Business Media, LLC 2011

47

48

J.-X. Lu and C.-C. Szeto

In principle, PALM system removes coherent cell fields by applying a pulsed UVA laser through an inverted microscope to allow laser ablation of cells and tissue on a tissue section, either manually or automatically (4). It belongs to laser cutting microdissection. Briefly, three steps are involved: first, the tissue is mounted on a membrane-cover slide and is viewed by a computer. Subsequently, the pulsed ultraviolet A (UVA) laser beam (focal spot <1  mm) impacts the samples and break molecular bonds resulting in “cutting.” At last, the interested cells are catapulted into the cap of a tube overlying the membrane slide. This procedure ensures that no unwanted elements reach the specimen. Since the laser in the process is only directed at the sample for about 1 ns, it does not transfer any heat. Thus, it avoids the potential risk of molecule modification of interest by especially the heating and cooling of the thermoplastic membrane usually used in laser capture microdissection. The process is completely contact and contamination-free and guarantees the best possible preservation of the material (5). This chapter mainly describes the application of the PALM MicroBeam laser microdissection system in gene expression analysis. The following protocol is specifically intended for gene expression of glomerular and tubulointerstitial in kidney biopsy. However, this could easily be adapted to other samples. It is well known that the original quality and subsequent handling of tissue used as a template source are fundamentally important. For RNA extraction, which would rapidly degrade due to the widespread presence of ribonuclease, it is especially important. We prefer to use fresh, unfixed tissues to develop a robust methodology.

2. Materials 1. Cryostat (Leica Microsystems, Wetzlar, Germany). 2. Disposable microtome blades (Leica Microsystem). 3. Tweezers. 4. Brushes. 5. Barrier pen. 6. Needle. 7. Dry ice. 8. PALM Laser-MicroBeam System (P.A.L.M. Carl Zeiss, Bernried, Germany). 9. A microcentrifuge tube with adhesive cap (Carl Zeiss). 10. 0.17 Poly-ethylene-naphthalene (PEN) Membrane Slide (Carl Zeiss, PALM Microlaser Technologies, Bernried, Germany).

3  Gene Expression Using the PALM System

49

11. RNase-free water (Ambion, Austin, TX). 12. RNase Away (Invitrogen Life Technologies, Carlsbad, CA). 13. Optimal cutting temperature (OCT; Tissue-Tek) medium. 14. RNaseZap (Sigma, St. Louis, MO). 15. Fresh desiccant. 16. 50% ethanol. 17. 70% ethanol. 18. 100% ethanol. 19. Cresylviolet solution. 20. Guanidine thiocyanate containing lysis buffer. 21. RNAqueous®-Micro Kit (Applied Biosystems, USA). 22. High-Capacity cDNA Reverse Transcription Kit with RNase Inhibitor (Applied Biosystems, USA). 23. TaqMan® Gene Expression Assays (Applied Biosystems, USA). 24. TaqMan Universal PCR Master Mix (Applied Biosystems, USA). 25. ABI Prism 7900HT Fast Real-Time PCR System (Applied Biosystems, Foster City, CA).

3. Methods The methods described below include (1) preparation of tissue samples, (2) staining or nonstaining of tissue sections, (3) laser microdissection and pressure catapulting, (4) RNA extraction, (5) cDNA reverse transcription, and (6) real-time quantitative polymerase chain reaction (RT-QPCR). 3.1. Preparation of Tissue Samples

It should be noted that RNase-free is most essential during hand­ ling samples, especially when nanogram amounts are processed (see Note 1). After surgical resection, samples are dissected and snap-frozen in liquid nitrogen. Frozen tissue is stored at −80°C until further processing (see Note 2). 1. Precool the temperature of cryostat to −17°C. Clean the knife holder (not the knife blade itself) with 100% ethanol and RNaseZap. 2. Cool the tissue platform (chuck), PEN membrane slide, and brushes in the cryostat (see Note 3). 3. Inside the cryostat, attach and orient the frozen OCTembedded tissue to the chuck with OCT media. 4. Install a fresh disposable blade into the blade holder.

50

J.-X. Lu and C.-C. Szeto

5. When the OCT media get harden, carefully pulse forward to move the specimen toward the blade and cutting surface. 6. Set the cutting thickness to 8–10 mm. 7. Section the tissue. Use the small brushes to straighten out folded, curled, or wrinkled sections by pressing them against the cutting surface (see Note 4). 8. Attach the sections onto the central area of a precleaned PEN membrane slide (see Note 5). 9. Put slides on dry ice (see Note 6). 10. Proceed to the tissue staining protocol or store slides in a tightly closed box with desiccant at −80°C for up to 4 weeks depending on the tissue (see Note 7). 3.2. Staining or Nonstaining of Tissue Sections for Laser Microdissection

1. Fix the tissue section slide for 30 s each in decreasing (100, 70, 50%) concentrations of ethanol consecutively. 2. Mark the area to be stained with a barrier pen and add 200 mL Cresylviolet solution for 15–30 s (see Note 8). 3. Drain off stain and dehydrate tissue 30 s each in increasing (50, 70, 100%) concentrations of ethanol consecutively (see Note 9). 4. Add 500 mL 100% ethanol on the slide and place slides in a slide box with fresh desiccant and store at room temperature until ready to perform PALM (see Note 10). 5. Without staining(see Note 11): add to the tissue section slide 70% ethanol for 2  min. Incubate in100% ethanol for 60  s. Air-drying the tissue for 3 min for laser microdissection.

3.3. Laser Microdissection and Laser Pressure Catapulting

Tissue sections are laser microdissected following the manufacturer’s protocol for the PALM Laser-MicroBeam System. Approximately 20–30 glomerulus and 20 randomly selected tubulointerstitial area were isolated from each specimen within 60–120 min. 1. Turn on the computer and PALM system. 2. Click PALM RoboSoftware (PALM Microlaser Technologies). It enables the observation of the microscopic image on a computer screen. The image is overlaid with a graphical user interface enabling the user to perform laser manipulation of tissue directly on the screen (see Note 12). 3. Put the tissue section slide on microstat. The slide can move on the robotized microscope stage. 4. Mount a microcentrifuge cap moistened with 30  mL lysis buffer upside down just above the tissue section. Adjusting the distance between the cap and tissue properly (see Note 13).

3  Gene Expression Using the PALM System

51

Fig. 1. Glomerular image. (a) Image before laser cutting. (b) Image of split line between required and nonrequired material. (c) Image after laser capturing. (d) Image of selected tissue floating in the cap.

5. First observe tissue under 5× and 10× magnification and then cut in 20× (Figs. 1 and 2). 6. Under direct visual control, areas of interest in the histological specimens were selected by moving the computer mouse. The adjusted laser beam cuts the contour of the selected areas sample along predrawn lines, so that a clearly split line between required and nonrequired material is generated (see Note 14) (Fig. 2). 7. Following isolation of cells, a high-energy pulse of the focused laser beam just below the focal plane of the tissue specimen is used to create a pressure wave to separate the targeted tissue and capture it into the microcentrifuge cap. The isolated tissue was then laser catapulted into the cap for the subsequent RNA isolation with blank left (see Note 15) (Fig. 2). 8. Check the slide again before completion (see Note 16). 9. Take off the cap carefully (see Note 17). Lysis buffer is added into the microfuge tube, vortex briefly, and centrifuged for 5 min to spin down cells from the lid. The samples are stored at −70°C for future RNA extraction. 10. Turn off the PALM system and computer.

52

J.-X. Lu and C.-C. Szeto

Fig. 2. Tubulointerstitial image. (a) Image before laser cutting. (b) Image of split line between required and nonrequired material. (c) Image after laser capturing. (d) Image of selected tissue floating in the cap.

3.4. RNA Extraction

RNA isolation is performed using a RNAqueous®-Micro Kit (Applied Biosystems, USA) according to the manufacturer’s instructions (see Note 18). In brief: 1. Drop sample into 100  mL lysis solution and incubate for 30 min at 42°C. 2. Prewet the filter with 30 mL of lysis solution for ³5 min. 3. Add 3 mL of LCM additive to the lysate and mix. 4. Add 100% ethanol to recover all RNA. 5. Pass the lysate mixture through a prepared Micro Filter Cartridge Assembly. 6. Wash with 180 mL of wash solution #1 and twice with 180 mL wash solutions #2 and #3. Discard the flow-through and centrifuge the filter for 1 min. 7. Elute the RNA in 2× 8–10 mL of elution solution. 8. Add 1/10 volume of 10× DNase I buffer and 1 mL of DNase I. Incubate 20 min at 37°C. 9. Add 2 mL or 1/10 volume DNase inactivation reagent, mix well, and leave at room temperature for 2 min. 10. Pellet the DNase inactivation reagent and transfer the RNA to a fresh tube stored at −80°C.

3  Gene Expression Using the PALM System

53

The purified RNA is used for quantitative real-time PCR analysis. Therefore, the quality and quantity of purified RNA are greatly concerned (see Note 19). 3.5. cDNA Reverse Transcription

The method of reverse transcription and RT-QPCR was described in our previous studies (6, 7). cDNA synthesis is performed according to the protocol of High-Capacity cDNA Reverse Transcription Kits (Applied Biosystems, USA). 1. Prepare the 10 mL total RT master mix on ice: 2 mL 10× RT buffer, 0.8 mL 25× dNTP mix (100 mM), 2 mL 10× RT random primers, 1 mL MultiScribe™ reverse transcriptase, 1 mL RNase inhibitor, and 3.2 mL nuclease-free water. 2. Add 10 mL RNA sample and mix well. Place the plate or tubes on ice until you are ready to load the thermal cycler. 3. Reverse transcription was performed at 25°C for 10 min and 37°C for 120 min, followed by inactivation reaction at 85°C for 5 s. The resulting cDNA was stored at −20°C until use.

3.6. Real-Time Quantitative Polymerase Chain Reaction

Gene expression was performed by ABI Prism 7900HT Fast RealTime PCR System (Applied Biosystems) following the manufacturer’s instruction. For RT-QPCR, universal master mix, primer and probe set, cDNA, and H2O were mixed to make a 5 mL reaction volume (see Note 20). Each sample was analyzed in triplicate. RT-QPCR was performed at 50°C for 2  min, 95°C for 10  min, 40 cycles at 95°C for 15  s, and 60°C for 1  min. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH, Applied Biosystems) was used as a housekeeping gene to normalize the mRNA expression level of each target gene. Results were analyzed with Sequence Detection Software version 1.7 (Applied Biosystems). In order to calculate the differences of expression level for each target gene among samples, the DDCT method was used to assess the relative quantitation according to the manufacturer’s manual (8).

4. Notes During the whole procedure, there are many important notes used to facilitate techniques and improve outcome. 1. Use special table for RNA laboratory only. Use the RNaseZap® Solution to clean all the materials that will come in contact with the tissue (e.g., tweezers, brushes, working place, and pippets). Make sure all reagents are RNase-free. Wear clean disposable gloves while working with RNA and change frequently. Do not reuse solutions or pour them back into the original containers.

54

J.-X. Lu and C.-C. Szeto

2. To optimize recovery of RNA from microdissected specimens, appropriate methods of initial tissue processing are crucial. Particular attention is required for tissue preparation encompassing tissue fixation, embedding, and histological staining. It is better to freeze tissue specimens immediately after resection. Some author found formalin fixation and paraffin embedding may lead to rapid RNA degradation (9). Therefore, snap-frozen tissues are best choice while formalin-fixed and paraffin embedded tissue be considered only in case snap-frozen materials unavailable. 3. The microscope slide should be kept in the cryostat. Warm the back space where tissue section will be deposited with fingers. All slides were treated with ultraviolet irradiation and RNase Away before use. PEN membrane slide facilitates the laser catapulting notwithstanding its expensive price. 4. Keep a slide box on dry ice to store the finished slides during the sectioning step. Discard slides with folds or wrinkled sections. Use a new disposable blade for each tissue when cutting more than one tissues. 5. Usually, 6–8 sections can be positioned on one membrane slide and two PEN membrane slides for one sample. 6. It is very important to keep the slides on dry ice or in cryostat. Drying or thawing of the tissue prior to staining and dehydration will result in deterioration of morphology and RNA. 7. The tissues which are low in RNases such as brain may be stored longer. It is recommended to proceed to the tissue staining protocol as quickly as possible. 8. It is often necessary to stain tissue sections so that discrete structures within the tissue can be discerned. To isolate RNA from PALM samples, it is important to select proper stain reagents that minimize RNA degradation during staining. Besides cresylviolet, acridine orange stain reagents (10) and Nuclear Fast Red (11) could also be considered. Generally, the incubation period should be as short as possible in any staining procedure. 9. The movement of slide should be careful and soft preventing from tissue gliding during washing and dehydration steps. The drop should be added to the side of slide instead of the tissue directly. 10. Usually, the tissue is ready for laser microdissection after 3 min of air-drying. It is recommended that PALM be carried out immediately after staining for best RNA quality and yields. 11. If the tissue is easy to distinguish in microscope, it is no use to stain the tissue since any staining may also destroy RNA (Fig. 2 shows image without staining).

3  Gene Expression Using the PALM System

55

12. The software allows saving the images of tissue when needed. 13. If the distance between the cap and tissue is too close, it may cause the lysis buffer drop on tissue during laser cutting. On the contrary, too far from each other may fail in capture the selected tissue in the cap. 14. During cutting, adjust the UV energy power and focus when needed. Usually, UV energy power is 60–80  mW, focus is 60–70. 15. By adjusting focus of microscope, we can confirm whether the selected tissue is captured or not by the image of selected tissue floating in the cap (Fig.  2). In case the tissue is too adherent to be captured by the laser, use the precleaned needle to help cut the tissue. 16. Since sometimes some tissues have not been captured into the cap and scattered on the slide, it is necessary to check the slide again. Reload the tissues if needed. 17. Avoid sprinkling the lysis buffer outside. 18. The method of RNA extraction can be crucial. Different RNA isolation assay may have got different quantity and quality of RNA. It is better to test the recovery rate of RNA samples of known concentration before including a new method or kit in the RNA extraction process. 19. The amount of microdissected cells, type of cells, and tissue processing affect the amount of isolated RNA (3, 12). We should capture selected cells as more as possible (13). 20. Use of 384-well plate can save the sample since it only needs total reaction volume as low as 5 mL. Useful Web sites http://www.zeiss.de/microdissection: Technical support for the use of the PALM laser systems. http://www.ambion.com/techlib/index.html: Technical resources about the work with gene expression. References 1. Murray, G.I. (2007). An overview of laser microdissection technologies. Acta Histochem. 109, 171–6. 2. Wang. G., Lai, F., Lai, K., et  al. (2009). Discrepancy between intrarenal messenger RNA and protein expression of ACE and ACE2 in human diabetic nephropathy. Am J Nephrol. 29, 524–31. 3. Blakey, G., Laszik, Z. (2004). Laser-assisted microdissection of the kidney: fundamentals and applications. J Mol Histol. 35, 581–7.

4. Pinzani, P., Orlando, C., Pazzagli, M. (2006). Laser-assisted microdissection for real-time PCR sample preparation. Mol Aspects Med. 27, 140–59. 5. Schutze, K., Becker, B., Bernsen, M., Bjornsen, T., Broksch, D., Bush, C., Clement-Sengewald, A., et al. Tissue microdissection, laser pressure catapulting. In: Bowtell, D., Sambrook, J. (Eds.), DNA microarrays: A Molecular Cloning Manual. Cold Spring Harbor Laboratory Press, Cold Spring, Harbor, NY, 2003:331–56.

56

J.-X. Lu and C.-C. Szeto

6. Chan, R., Lai, F., Li, E., et al. (2007). Intrarenal cytokine gene expression in lupus nephritis. Ann Rheum Dis. 66, 886–92. 7. Chan, R., Lai, F., Li, E., et  al. (2004). Expression of chemokine and fibrosing factor messenger RNA in the urinary sediment of patients with lupus nephritis. Arthritis Rheum. 50, 2882–90. 8. Guide to Performing Relative Quantitation of Gene Expression Using Real-Time Quantitative PCR. Foster City, Applied Biosystems, 2004. 9. Vincek, V., Nassiri, M., Knowles, J., Nadji, M., Morales, A. (2003). Preservation of tissue RNA in normal saline. Lab Invest. 83, 137–8.

10. LCM Staining Kit. Foster City, Applied Biosystems, 2008. 11. Burgemeister, R., Gangnus, R., Haar, B., Schütze, K., Sauer, U. (2003). High quality RNA retrieved from samples obtained by using LMPC (laser microdissection and pressure catapulting) technology. Pathol Res Pract. 199, 431–6. 12. Cohen, C., Gröne, H., Gröne, E., Nelson, P., Schlöndorff, D., Kretzler, M. (2002). Laser microdissection and gene expression analysis on formaldehyde-fixed archival tissue. Kidney Int. 61, 125–32. 13. Espina, V., Wulfkuhle, J., Calvert, V., et  al. (2006). Laser-capture microdissection. Nat Protoc.1, 586–603.

Chapter 4 Immunoguided Microdissection Techniques Michael A. Tangrea, Jeffrey C. Hanson, Robert F. Bonner, Thomas J. Pohida, Jaime Rodriguez-Canales, and Michael R. Emmert-Buck Abstract Over the past 15 years, laser-based microdissection has improved the precision by which scientists can procure cells of interest from a heterogeneous tissue section. However, for studies that require a large amount of material (e.g., proteomics) or for cells that are scattered and difficult to identify by standard histological stains, an immunostain-based, automated approach becomes essential. In this chapter, we discuss the use of immunohistochemistry (IHC) and immunofluorescence (IF) to guide the microdissection process via manual and software-driven auto-dissection methods. Although technical challenges still exist with these innovative approaches, we present here methods and protocols to successfully perform immuno-based microdissection on commercially available laser dissection systems. Key words: Automation, Immunohistochemistry, Microdissection

1. Introduction Immunohistochemistry (IHC) was developed over 60 years ago for in situ measurements of proteins in histological sections (1). Based on the recognition of target cell antigens by specific primary antibodies, IHC has become a staple for molecular analysis of tissue sections in the research laboratory and clinic. The more recent integration of this commonly used technique with laser microdissection has resulted in a powerful combination that is enabling new studies and is amenable to further automation to improve dissection throughput (2–5). The first such method developed was called immunoguided laser capture microdissection or immuno-LCM (4) (Fig.  1a).

Graeme I. Murray (ed.), Laser Capture Microdissection: Methods and Protocols, Methods in Molecular Biology, vol. 755, DOI 10.1007/978-1-61779-163-5_4, © Springer Science+Business Media, LLC 2011

57

58

M.A. Tangrea et al.

Fig. 1. Schematic representation of immunoguided microdissection techniques. Initial immunostaining step represented by the antibody complex attached to a labeling enzyme (circles).

This approach combines immuno-stained tissue with LCM, allowing the investigator to select specific cell types by visually aiming a laser at stained target cells. This procedure is helpful for identifying cell populations that would otherwise not be recognized by standard histological stains, such as hematoxylin and eosin (H&E), and to subsequently dissect a specific subset of cells using a molecular marker. Immuno-LCM can now be performed in an automated fashion using newly developed software packages available for several commercial dissection instruments. These instruments identify target cells by image processing and then automatically dissect the tissue using a motorized stage and a coordinate system. Both manual and automated immuno-LCM can utilize either IHC or immunofluorescence (IF) to guide the dissection process (4–5, 11). As a specific example of an automated immuno-LCM system, the ArcturusXT (Life Technologies) incorporates an image analysis software package called AutoScanXT, which specifically identifies stained areas with minimal supervision by the investigator (Fig. 1b). The use of image analysis software improves the speed of the microdissection process by selecting targets based on a threshold set by the instrument’s operator. This chapter provides the materials, protocols, and related references that will permit investigators to successfully perform immunoguided microdissection. However, the reader should note that these technologies and associated molecular analysis methods are not static, but are being continually refined and improved. In particular, on-going studies to assess the effects of immunostaining and microdissection on biomolecular recovery

4  Immunoguided Microdissection Techniques

59

and methods to improve extraction techniques are currently areas of active investigation by our group and others (6–11). Moreover, there are often subtle differences in the dissection conditions that are required for each tissue type and target cell. Nonetheless, the protocols described below are amenable to most dissection studies­ and serve as a good starting template.

2. Materials 2.1. Immuno­ histochemistry

The following materials are for an IHC protocol that we use to label epithelial cells using monoclonal antibodies against cytokeratins. 1. IHC horizontal staining tray (Thermo Scientific, Waltham, MA). 2. Vertical staining tubs and racks (Simport). 3. Rice steamer (Black & Decker), used for antigen retrieval. 4. Citrate buffer, pH 6.0 (DAKO Carpinteria, CA) for antigen retrieval, stored at 4°C. 5. Phosphate-buffered saline (PBS): 150  mM NaCl, 25  mM NaPO4, pH 7.4. Stored at room temperature. 6. DAKO Envision+ IHC Kit (DAKO) provides the peroxidase blocking solution (0.03% H2O2), the secondary antibody polymer (either mouse or rabbit or dual), and the 3,3¢-diaminobenzidine (DAB) substrate solution (see Note 1). 7. Mouse Anti-Cytokeratin AE1/AE3 diluted 1:50 in Zymed Antibody Diluent (Invitrogen, Carlsbad, CA). Reagents stored at 4°C, but the dilution is made fresh prior to staining and used at room temperature (see Note 2). 8. DAB Enhancer (Invitrogen) stored at room temperature. 9. Deionized water (diH2O). Stored at room temperature. 10. Graded alcohols (70, 95, and 100% ethanol). Stored at room temperature. 11. Xylenes for drying and dewaxing formalin-fixed, paraffinembedded (FFPE) tissue. Stored at room temperature.

2.2. Immuno­ fluorescence

1. Reagents from IHC protocol described above, plus secondary Goat Anti-Mouse antibody with Alexa Fluor® 488 (Invitrogen).

2.3. Immuno-LCM

1. CapSure® Macro LCM caps (Life Technologies, Carlsbad, CA). 2. LCM machine; ArcturusXT, Veritas, or PixCell II (Life Technologies).

60

M.A. Tangrea et al.

2.4. LCM Image Recognition Software

1. CapSure® Macro LCM caps (Life Technologies). 2. LCM machine; ArcturusXT (Life Technologies) with optional AutoScanXT software.

3. Methods Described below are protocols for tissue immunostaining, immuno-LCM, and immuno-LCM using the AutoScanXT software. The initial immunostaining step is essential to successful microdissection using these approaches. A strong staining with minimal background enables precise procurement of the targets of interest. It is important to note that downstream molecular analysis must be considered when utilizing these immuno-stained microdissection methods. DNA is the most robust biomolecule following IHC; however, yields are typically reduced; yet the overall quality of the DNA is unaffected (6, 11–14). RNA is the most susceptible to the staining process and is often degraded in the initial staining steps (6, 15). If RNA were the downstream molecule of interest, it would be necessary to shorten the staining protocol times significantly and add RNase inhibitors. However, it is highly dependent on the abundance of the RNA of interest and it has been our experience that only small RNA targets are recoverable following standard IHC. Better results for RNA extraction have been obtained with the use of modified protocols for immunofluorescence staining (16, 17). On the contrary, proteins are analyzable via SDS–PAGE and mass spectrometry methods, but mild lysis buffer methods, such as 2D-PAGE, presents a challenge to recover the protein content of the tissue as harsher buffers and treatments are more effective (6, 15). Despite these few caveats, the combination of immunoguided dissection tools with the burgeoning field of molecular biology enables researchers to pursue tissue-based scientific questions that would otherwise not be feasible. 3.1. Immuno­ histochemistry: Frozen Tissues

1. A tissue specimen is sectioned utilizing a cryostat machine (Leica) at 8-mm thick sections. The tissue is then placed on positively charged glass slides. Tissue sections can be stored at −80°C for several days prior to microdissection. 2. To begin IHC, place the tissue slides in a 70% ethanol bath for 2 min at room temperature. Then transfer the slides to a diH2O bath. 3. Proceed with the endogenous peroxidase block from the DAKO Envision+ kit, for 10–40  min at room temperature (see Note 3). 4. Wash the slides in PBS bath (3×, 2 min each wash).

4  Immunoguided Microdissection Techniques

61

5. Incubate slides with diluted primary antibody for 30 min at room temperature (see Note 4). 6. Wash the slides in PBS bath (3×, 2 min each wash). 7. Incubate slides with pre-diluted DAKO Envision+ secondary antibody for 30 min at room temperature. 8. Wash the slides in PBS bath (3×, 2 min each wash). 9. Apply DAKO Envision+ DAB solution for 5  min at room temperature. 10. Wash in diH2O bath (3×, 2 min each wash). 11. Apply the Zymed DAB Enhancer for 3  min at room temperature. 12. Wash in diH2O bath (3×, 2 min each wash) (see Note 5). 13. Dehydrate the sample in graded alcohols (ethanol 70, 95, and 100%) for 2 min each followed by xylenes for 3×, 2 min each. 3.2. Immuno­ fluorescence: Frozen Tissue

1. Section the tissue specimen using a cryostat (Leica) at 8-mm thickness and place the section on positively charged glass slides. Tissue sections can be stored at −80°C for several days prior to microdissection. 2. To begin IHC, place the tissue slide in a 70% ethanol bath for 2  min at room temperature. Then transfer the slide to a diH2O bath. 3. Wash the slides in PBS bath (3×, 2 min each wash). 4. Incubate slide with diluted primary antibody for 30  min at room temperature (see Note 4). 5. Wash the slides in PBS bath (3×, 2 min each wash). 6. Incubate slide with secondary antibody with Alexa Fluor 488 for 30 min at room temperature. 7. Wash in diH2O bath (3×, 2 min each wash). 8. Dehydrate the sample in graded alcohols (ethanol 70, 95, and 100%) followed by xylenes for 2 min each.

3.3. Immuno­ histochemistry: Formalin-Fixed Paraffin-Embedded Tissues

1. For FFPE tissue, an antigen retrieval step is typically required prior to incubation with antibodies (18). A number of antigen retrieval approaches are possible, including the use of heat induced epitope retrieval (HIER) (19) and the use of enzymes, such as trypsin or pepsin (20–26). HIER will be described here (see Notes 6 and 7). 2. For HIER, a citrate buffer (300  ml) is pre-heated in a rice steamer for 40 min. 3. The FFPE tissue is dewaxed in xylenes 3×, 2  min for each incubation. The tissue is then rehydrated through graded alcohols (100, 95, and 70% for 2 min each) to diH2O prior to placement in the hot citrate bath.

62

M.A. Tangrea et al.

4. Place the rehydrated tissue slides in the hot citrate bath for 25 min. 5. Remove entire citrate bath containing the slides from the streamer and allow cooling at room temperature for 25 min. 6. Proceed with the IHC protocol (Subheading 3.1) as stated above beginning with step 3. 3.4. Immuno-LCM

1. These instructions assume the use of a PixCell II instrument, although they are easily adaptable to other products in the LCM line by Life Technologies. 2. Following immunohistochemistry (Subheadings  3.1 and 3.3), the dehydrated tissue is placed on the platform of the PixCell II device and the vacuum is applied (see Note 8). 3. A CapSure® Macro LCM cap (Life Technologies) is placed over the dehydrated tissue via the loading arm. 4. The IR laser is tested in an open area of the cap to obtain a laser shot “ring”. Adjust the parameters of the laser for optimal lifting. For example, for a 30-mm laser spot size, the ranges of typical parameters are: 20–30  mW power and 5–6 ms duration. 5. Under microscopic visualization, locate the immuno-stained cells of interest and fire the laser over the desired area. 6. When the dissection is complete, remove the cap from the slide and proceed with downstream molecular analysis.

3.5. LCM Image Recognition Software

1. The AutoScanXT Software Module from Life Technologies is an image analysis program that automatically identifies immuno-stained regions of interest as defined by the user (27) (see Note 9). AutoScanXT provides three levels of analysis: pixel, texture, and morphology. In this protocol, we will focus on the pixel analysis tool. 2. Following immunostaining, the dehydrated tissue is placed on the platform of the ArcturusXT device. 3. The operator utilizes the ArcturusXT software field of view to select a tissue area of interest and the appropriate magnification. Once selected, a CapSure® Macro LCM cap (Life Technologies) is then placed automatically over the dehydrated tissue. 4. Select the “AutoScanXT” icon located on the Select Tool Panel of the ArcturusXT operating software. 5. Select “Acquire Image” and a static image is captured of the area of interest (see Note 10). A new analysis file will need to be created and named by the user. 6. In the pixel analysis settings, regions of interest (ROI) are selected on the static jpeg image by the user and highlighted

4  Immunoguided Microdissection Techniques

63

Fig. 2. AutoScanXT software guided dissection of a cytokeratin AE1/AE3 IHC-stained prostate tissue. (a) Screen capture of the static jpeg image where ROIs (circles) and background (squares) as determined by the user. Inset highlights the symbols used to select the ROIs and background. The symbols are enhanced in the inset for visualization purposes. (b) Screen capture of the ROIs (circled areas) selected by the AutoScanXT analysis module. (c) Image of the LCM cap showing­ the dissected epithelial cells. (d) Image of the remaining tissue following AutoScanXT-directed dissection. All images captured at 100× magnification.

by red circles. Likewise, background areas are also selected by the user and highlighted by blue squares on the jpeg image (Fig. 2a) (see Note 11). 7. The AutoScanXT software then “learns” what the desired ROIs are and highlights the area to be dissected as indicated by empty red circles (Fig. 2b). A list of the identified regions selected by the program is then generated in the “Regions” tab (see Note 12). 8. The user verifies if this is the desired area of interest to be dissected and if so can “harvest” the area (see Note 13). At this point, the ArcturusXT machine automatically dissects the highlighted area (Fig. 2c, d) (see Note 14). 9. When the dissection is complete, move the cap to the quality control (QC) station and proceed with downstream ­molecular analysis.

64

M.A. Tangrea et al.

4. Notes 1. Several companies offer similar IHC products. 2. Cytokeratin AE1/AE3 is presented as an example; however, numerous primary antibodies against particular antigens of interest can be utilized. The antibody dilution needs to be determined empirically by the user for proper staining and can vary from 1:25 to 1:2,000. 3. The incubation time is dependent on the amount of endogenous peroxidase present in the tissue. Some tissues display more activity than others and thus longer blocking times may be required as determined by the user. 4. The primary antibody incubation is highly dependent on the relative abundance of the target antigen and the kinetics of the antibody–antigen interaction. A separate protocol that is commonly used is incubation with the primary antibody overnight at 4°C. 5. No counterstain, such as hematoxylin, is used when performing the immuno-stain procedure for the microdissection methods described, as it will decrease the signal-to-noise between the desired stained areas of the tissue and the undesired background. 6. Antigen retrieval is highly dependent on the antigen of interest and the amount of fixation. Therefore, a number of strategies and parameters may need to be tested by the user to identify the ideal protocol. 7. Immunofluorescence is also possible on FFPE tissue, although it is not discussed in detail here. The formalin fixative increases background autofluorescence of the sample, which makes the technique challenging. For a robust protocol of immunofluorescence on FFPE, refer to method described by Robertson et al. (28). 8. Immunofluorescent-stained slides may also be used for immuno-LCM. This is especially useful for proteomic studies (29). The Life Technologies microdissection device for this approach requires fluorescent capabilities. 9. It is also worth noting that other companies also offer similar image analysis software, such as the Leica LMD AutoVision Control system. 10. The AutoScanXT software only works with jpeg images. The program does not recognize tiff files and thus the user must select only jpeg images. 11. It is important to choose representative ROIs and areas of background in the static jpeg image, as this will help the

4  Immunoguided Microdissection Techniques

65

s­ oftware decipher between the two in a more comprehensive fashion. A large number of points are not necessary for a successful selection. 12. The user can either select all of the regions determined to fit the pixel criteria or individual regions of interest via the checkboxes located in the “Regions” tab. 13. If the highlighted area is incorrect. The user can increase or decrease the stringency accordingly and repeat the process. 14. For a complete step-by-step instruction of how to use the AutoScanXT, refer to the instruction manual, which can be downloaded at http://www.moldev.com.br/pages/software/ autoscanXT.html. References 1. Coons, A. H., Creech, H. J., and Jones, R. N. (1941) Immunological properties of an antibody containing a fluorescent group, Proceedings of the Society for Experimental Biology and Medicine 47, 200–202. 2. Emmert-Buck, M. R., Bonner, R. F., Smith, P. D., Chuaqui, R. F., Zhuang, Z., Goldstein, S. R., Weiss, R. A., and Liotta, L. A. (1996) Laser capture microdissection, Science 274, 998–1001. 3. Bonner, R. F., Emmert-Buck, M., Cole, K., Pohida, T., Chuaqui, R., Goldstein, S., and Liotta, L. A. (1997) Laser capture microdissection: molecular analysis of tissue, Science 278, 1481,1483. 4. Fend, F., Emmert-Buck, M. R., Chuaqui, R., Cole, K., Lee, J., Liotta, L. A., and Raffeld, M. (1999) Immuno-LCM: laser capture microdissection of immunostained frozen sections for mRNA analysis, Am J Pathol 154, 61–66. 5. Fend, F., Kremer, M., and QuintanillaMartinez, L. (2000) Laser capture microdissection: methodical aspects and applications with emphasis on immuno-laser capture microdissection, Pathobiology 68, 209–214. 6. Tangrea, M. A., Kreitman, M. S., Rosenberg, A. M., Mukherjee, S., Eberle, F. C., Jaffe, E. S., Wallis, B. S., Hanson, J. C., Chuaqui, R. F., Rodriguez-Canales, J., and Emmert-Buck, M. R. (2010) Effects of Immunohistochemistry on Biomolecules in Tissue Specimens: Impor­ tance for Expression-Based Microdissection Technologies, In American Association for Cancer Research Annual Meeting, Washington, D.C. 7. Macdonald, J. A., Murugesan, N., and Pachter, J. S. (2008) Validation of immuno-laser capture microdissection coupled with quantitative RT-PCR to probe blood-brain barrier gene

expression in situ, J Neurosci Methods 174, 219–226. 8. Liu, Y., Wu, J., Liu, S., Zhuang, D., Wang, Y., Shou, X., and Zhu, J. (2009) Immuno-laser capture microdissection of frozen prolactioma sections to prepare proteomic samples, Colloids Surf B Biointerfaces 71, 187–193. 9. Nakamura, N., Ruebel, K., Jin, L., Qian, X., Zhang, H., and Lloyd, R. V. (2007) Laser capture microdissection for analysis of single cells, Methods Mol Med 132, 11–18. 10. Fassunke, J., Majores, M., Ullmann, C., Elger, C. E., Schramm, J., Wiestler, O. D., and Becker, A. J. (2004) In situ-RT and immunolaser microdissection for mRNA analysis of individual cells isolated from epilepsy-associated glioneuronal tumors, Lab Invest 84, 1520–1525. 11. Eberle, F. C., Hanson, J. C., Killian, J. K., Wei, L., Ylaya, K., Hewitt, S. M., Jaffe, E. S., Emmert-Buck, M. R., and Rodriguez-Canales, J. (2010) Immunoguided laser assisted microdissection techniques for DNA methylation analysis of archival tissue specimens, J Mol Diagn 12, 394–401. 12. Grover, A. C., Tangrea, M. A., Woodson, K. G., Wallis, B. S., Hanson, J. C., Chuaqui, R. F., Gillespie, J. W., Erickson, H. S., Bonner, R. F., Pohida, T. J., Emmert-Buck, M. R., and Libutti, S. K. (2006) Tumor-associated endothelial cells display GSTP1 and RARbeta2 promoter methylation in human prostate cancer, J Transl Med 4, 13. 13. Hanson, J. A., Gillespie, J. W., Grover, A., Tangrea, M. A., Chuaqui, R. F., Emmert-Buck, M. R., Tangrea, J. A., Libutti, S. K., Linehan, W. M., and Woodson, K. G. (2006) Gene promoter methylation in prostate tumor-associated

66

M.A. Tangrea et al.

stromal cells, J Natl Cancer Inst 98, 255–261. 14. Rodriguez-Canales, J., Hanson, J. C., Tangrea, M. A., Erickson, H. S., Albert, P. S., Wallis, B. S., Richardson, A. M., Pinto, P. A., Linehan, W. M., Gillespie, J. W., Merino, M. J., Libutti, S. K., Woodson, K. G., Emmert-Buck, M. R., and Chuaqui, R. F. (2007) Identification of a unique epigenetic sub-microenvironment in prostate cancer, J Pathol 211, 410–419. 15. Tangrea, M.A., Mukherjee, S., Gao, B., Markey, S.P., Du, Q., Armani, M., Kreitman, M.S., Rosenberg, A.M., Wallis, B.S., Eberle, F.C., Duncan, F.C., Hanson, J.C., Chuaqui, R.F., Rodriguez-Canales, J., and EmmertBuck, M.R. (2011) Effect of immunohistochemistry on molecular analysis of tissue samples: implications for microdissection technologies, J Histochem Cytochem 59, 591–600. 16. Murakami, H., Liotta, L., and Star, R. A. (2000) IF-LCM: laser capture microdissection of immunofluorescently defined cells for mRNA analysis rapid communication, Kidney Int 58, 1346–1353. 17. Anthony, R. M., Urban, J. F., Jr., Alem, F., Hamed, H. A., Rozo, C. T., Boucher, J. L., Van Rooijen, N., and Gause, W. C. (2006) Memory T(H)2 cells induce alternatively activated macrophages to mediate protection against nematode parasites, Nat Med 12, 955–960. 18. Ramos-Vara, J. A., and Beissenherz, M. E. (2000) Optimization of immunohistochemical methods using two different antigen retrieval methods on formalin-fixed paraffin-embedded tissues: experience with 63 markers, J Vet Diagn Invest 12, 307–311. 19. Shi, S. R., Key, M. E., and Kalra, K. L. (1991) Antigen retrieval in formalin-fixed, paraffinembedded tissues: an enhancement method for immunohistochemical staining based on microwave oven heating of tissue sections, J Histochem Cytochem 39, 741–748. 20. Battifora, H., and Kopinski, M. (1986) The influence of protease digestion and duration of fixation on the immunostaining of keratins.

A comparison of formalin and ethanol fixation, J Histochem Cytochem 34, 1095–1100. 21. Huang, S. N. (1975) Immunohistochemical demonstration of hepatitis B core and surface antigens in paraffin sections, Lab Invest 33, 88–95. 22. Huang, S. N., Minassian, H., and More, J. D. (1976) Application of immunofluorescent staining on paraffin sections improved by trypsin digestion, Lab Invest 35, 383–390. 23. Jacobsen, M., Clausen, P. P., and Smidth, S. (1980) The effect of fixation and trypsinization on the immunohistochemical demonstration of intracellular immunoglobulin in paraffin embedded material, Acta Pathol Microbiol Scand A 88, 369–376. 24. Miettinen, M. (1989) Immunostaining of intermediate filament proteins in paraffin sections. Evaluation of optimal protease treatment to improve the immunoreactivity, Pathol Res Pract 184, 431–436. 25. Ordonez, N. G., Manning, J. T., Jr., and Brooks, T. E. (1988) Effect of trypsinization on the immunostaining of formalin-fixed, paraffin-embedded tissues, Am J Surg Pathol 12, 121–129. 26. Pinkus, G. S., O’Connor, E. M., Etheridge, C. L., and Corson, J. M. (1985) Optimal immunoreactivity of keratin proteins in formalinfixed, paraffin-embedded tissue requires preliminary trypsinization. An immunoperoxidase study of various tumours using polyclonal and monoclonal antibodies, J Histochem Cytochem 33, 465–473. 27. http://www.moleculardevices.com/pages/software/autoscanXT.html. AutoScanXT Software. 28. Robertson, D., Savage, K., Reis-Filho, J. S., and Isacke, C. M. (2008) Multiple immunofluorescence labelling of formalin-fixed paraffin-embedded (FFPE) tissue, BMC Cell Biol 9, 13. 29. Mouledous, L., Hunt, S., Harcourt, R., Harry, J. L., Williams, K. L., and Gutstein, H. B. (2003) Proteomic analysis of immunostained, laser-capture microdissected brain samples, Electrophoresis 24, 296–302.

Chapter 5 Optimized RNA Extraction from Non-deparaffinized, Laser-Microdissected Material Danny Jonigk, Friedrich Modde, Clemens L. Bockmeyer, Jan Ulrich Becker, and Ulrich Lehmann Abstract mRNA extraction and subsequent RT-polymerase chain reaction (PCR)-based expression analysis from laser-microdissected material is by now a well-established and reproducible method. Most routinely stored tissue samples are preserved as formalin-fixed, paraffin-embedded materials. While this allows for a convenient storage and stable preservation of nucleic acids, deparaffinization before staining for laser microdissection may result in a significant loss of mRNA quality and consequently of PCR sensitivity. We describe a method of isolating anatomic compartments from non-deparaffinized, formalin-fixed, and paraffinembedded tissues by laser-assisted microdissection which allows for a highly efficient mRNA retrieval. Key words: mRNA extraction, Non-deparaffinized sections, Laser-assisted microdissection, Quantitative RT-PCR

1. Introduction Formalin-based conservation of biological tissues has been ­introduced into the life sciences over 100 years ago – more than 60  years before the discovery of nucleic acids. The possible ­recovery of DNA and especially of the rather unstable RNA from formalin-fixed, paraffin-embedded (FFPE) materials that ­originally had been preserved for sectioning and light microscopy analysis was first described in 1991 and entered routine practice less than 10 years ago (1, 2). As the possibilities to analyze nucleic acids and proteins from FFPE tissues are limited due to fragmentation of nucleic acids and modifications of amino acids, the frozen ­storage of samples in tissue banks continues to be an efficient way

Graeme I. Murray (ed.), Laser Capture Microdissection: Methods and Protocols, Methods in Molecular Biology, vol. 755, DOI 10.1007/978-1-61779-163-5_5, © Springer Science+Business Media, LLC 2011

67

68

D. Jonigk et al.

to enable diagnostics later on. But frozen storage is an expensive and time-consuming method that requires continuous monitoring and limits the quality of subsequent light microscopy evaluation. So FFPE continues to serve as the major, cost-efficient method for biological tissue conservation (3). Microdissection in tissue slides allows for the isolation of anatomical compartments and distinct cell types (4). So quantitative and cell-type-specific mRNA expression analysis from FFPE is feasible. To ensure maximum effectiveness of RT-polymerase chain reaction (PCR)-based RNA expression analysis from FFPE samples, we present a method of laser-assisted microdissection in non-deparaffinized tissue slides combined with target-gene-­specific preamplification. Optimized fixation ensures less RNA degradation due to autolysis while skipping the time-consuming deparaffinization and subsequent staining process before laser microdissection reduces the loss in RNA quality further (5). Finally, preamplification increases the sensitivity of RT-PCR up to a 1,000-fold (6). Our method is feasible for distinct anatomical structures in a variety of tissues.

2. Materials 2.1. Tissue Fixation

1. Dissecting tools. 2. Tissue cassettes. 3. Buffered formalin (4%). 4. Graded ethanol (70 and 100%). 5. Paraffin wax.

2.2. Slide Preparation and Laser-Assisted Microdissection

1. Membrane slides for laser-assisted microdissection (Molecular Machines & Industries, Glattbrugg, Switzerland). 2. Microtome. 3. Glass slides. 4. Laser microdissection tubes (500 ml; Molecular Machines & Industries). 5. MMI CellCut Plus laser microdissection system (Molecular Machines & Industries).

2.3. RNA Isolation

1. 500-ml reaction tubes. 2. Proteinase K (20 mg/ml; Merck, Darmstadt, Germany). 3. Guanidinium thiocyanate. 4. 1 M Tris–HCl, pH 7.6. 5. ß-Mercaptoethanol.

5  Optimized RNA Extraction from Non-deparaffinized, Laser-Microdissected Material

69

6. 3 M sodium acetate, pH 5.2. 7. Roti-aqua-phenol (Carl Roth, Karlsruhe, Germany). 8. Chloroform. 9. Isopropanol. 10. Glycogen (Roche, Basel, Switzerland). 11. Ethanol. 12. DEPC-treated water. 2.4. cDNA Synthesis

1. 200-ml reaction tubes. 2. 10× RT buffer (2×; Applied Biosystems, Foster City, CA, USA). 3. dNTP mix (100 mM; Applied Biosystems). 4. RT random primers (Applied Biosystems). 5. Multiscribe reverse transcriptase (Applied Biosystems). 6. RNAse inhibitor (20 U/ml; Applied Biosystems). 7. Thermocycler.

2.5. cDNA Preamplification

1. 200-ml reaction tubes. 2. TaqMan PreAmp master mix (Applied Biosystems). 3. Assay pool (pool of RT-PCR primers with TE buffer, Applied Biosystems). 4. 100 mM Tris buffer, pH 7.6. 5. 0.5 M EDTA, pH 9.0. 6. Ultrapure water (“Ampuwa,” Fresenius Kabi, Bad Homburg, Germany). 7. Thermocycler.

2.6. Real-Time PCR

1. 500-ml reaction tubes. 2. Ultrapure water (“Ampuwa,” Fresenius Kabi). 3. Gene Expression Master Mix (Applied Biosystems). 4. 96-well plate (Applied Biosystems). 5. MicroAmp optical adhesive film (Applied Biosystems). 6. Fresco 17 centrifuge (Thermo Scientific, Waltham, MA, USA). 7. 7500 Real-Time PCR System (Applied Biosystems).

2.7. Primary Data Evaluation

1. Sequence Detection Software (Version 1.3.1, 7500 Fast System SDS software; Applied Biosystems). 2. Excel 8.0 (Microsoft, Redmond, WA, USA).

70

D. Jonigk et al.

3. Methods The following steps describe FFPE-based tissue fixation and ­subsequent mRNA analysis from laser-microdissected unstained sections: 1. Tissue fixation. 2. Slide preparation and laser-assisted microdissection. 3. RNA isolation. 4. cDNA synthesis. 5. cDNA preamplification. 6. RT-PCR. 7. Data interpretation. 3.1. Tissue Fixation

1. Prepare tissue (organ specimens, biopsies, etc.) with dissecting tools as required (scalpel, scissors, tweezers, etc.). Section the sample into thin slices and place into a tissue cassette, taking care not to fill up more than approximately 60% of the cassette’s volume (see Note 1). The maximum thickness should not exceed 5 mm (formalin penetration is about 1 mm/h). 2. Transfer cassette into formalin for 12–20 h. The ratio of formalin/tissue should ideally be 20:1 or higher. Lower ratios might result in inadequate fixation and degrading of nucleic acids. 3. Transfer cassette into graded ethanol to dehydrate (70% for 1 h, 90% for 45 min, 100% for 45 min, 100% for 1 h (no. 1), 100% for 1 h (no. 2), 100% for 1 h (no. 3); 40°C). 4. Transfer cassette into xylene (for 45 min (no. 1), for 45 min (no. 2), for 1 h). 5. Transfer cassette into paraffin wax (for 30  min (no. 1), for 30 min (no. 2), for 30 min (no. 3); 62°C). 6. Embed sample into a paraffin block.

3.2. Slide Preparation and Laser-Assisted Microdissection

1. Cut 5–10 mm thick serial sections with a fresh blade from the paraffin block and float them out on a warm water bath (46°C) (see Note 2). 2. Mount sections on nuclease-free membrane slides. You can store the prepared slides until usage for a couple of days at 4°C. Directly before laser microdissection, dry the slides at 37°C overnight. 3. To add additional stabilization, place a regular glass slide under each membrane slide. Use the MMI CellCut Plus ­system to cut out desired anatomical compartments. Set the power and magnification appropriately to the size of the ­target structure(s)

5  Optimized RNA Extraction from Non-deparaffinized, Laser-Microdissected Material

71

and the thickness of the sections (see Note 3). In our experience­, a 200× magnification works best for most tissues. You should be able to collect about 80 human glomeruli-sized structures (about 200-mm diameter each) with one microdissection tube. 3.3. RNA Isolation

1. Fill 50 ml digestion solution (4.2 M guanidinium thiocyanate, 30  mM Tris–HCl pH 7.6, 2% sodium N–lauryl sarcosine), 50 ml proteinase K (20 mg/ml), and 0.5 ml ß-mercaptoethanol into the microdissection tube (see Note 1). Vortex thoroughly and incubate at 55°C overnight. 2. Centrifuge the tube briefly. Do not centrifuge at full speed, as otherwise the adhesive lid might slip into the tube. Transfer the lysate into a regular tube without an adhesive lid. 3. Add 10 ml sodium acetate, 63 ml Roti-aqua-phenol, and 27 ml chloroform, vortex thoroughly, and leave the tube on ice for 20 min. 4. Centrifuge tube at 16,200 ´ g for 30 min. 5. Transfer 95 ml of the aqueous supernatant into a fresh tube, add 95 ml ice-cold isopropanol and 1 ml glycogen, and store at −20°C for at least one night. 6. Centrifuge the tube for 30 min at 16,200 ´ g (4°C) and discard the supernatant, taking care not to disturb the pellet. 7. Wash the pellet with 180 ml ice-cold 70% ethanol. 8. Centrifuge the tube again for 5  min (4°C) at 16,200 ´ g, discard the supernatant, and air dry the pellet for 5 min (on ice). Then, dissolve the pellet in 10 ml DEPC water. If required, RNA can be temporarily stored at −20°C, though −80°C is preferable.

3.4 cDNA Synthesis

1. Mix RNA with master mix (2 ml 10× RT buffer), 0.8 ml dNTPs mix (100  mM), 2  ml RT random primers, 1  ml multiscribe reverse transcriptase, and 1  ml RNAse Inhibitor (20  U/ml) and fill up with DEPC water to a total reaction volume of 20 ml in a 200-ml tube (see Note 2). 2. Incubate the tube in a thermocycler (thermocycler program: 25°C for 10 min, 37°C for 120 min, 85°C for 5 s). 3. If required, cDNA can be stored at −20°C, though −80°C is preferable.

3.5. cDNA Preamplification

1. Mix 6.25 ml cDNA with 12.5 ml TaqMan PreAmp master mix and 6.25 ml assay pool in a 200-ml tube (3 ml of each individual primer to be used in the RT-PCR (a maximum of 100 primers can be used) filled up to a total reaction volume of 300 ml with 1× TE buffer (5 ml 100 mM Tris buffer, pH 8.1 with 100 ml 0.5 M EDTA and 45 ml Ampuwa)) (see Note 2).

72

D. Jonigk et al.

2. Incubate the tube in a thermocycler (thermocycler program: 1 cycle of 95°C for 10 min, 14 cycles of 95°C for 20 s and 60°C for 4 min). 3. If required, the preamplified cDNA can be stored at −20°C, though −80°C is preferable (see Note 4). 3.6. Real-Time PCR

1. Dilute part of the preamplified cDNA 1/10 with Ampuwa (amount of the diluted cDNA depends on the number of desired RT-PCRs, see below) in a separate tube. The remaining preamplified cDNA should be restored at −80 C (see Note 2). 2. Prepare reaction mix for a 96-well plate: Each well has to contain 4  ml Ampuwa, 10  ml Gene Expression master mix, 5 ml diluted preamplified cDNA (see above), and 1 ml of the individual primer (total reaction volume/well: 20  ml; e.g., reaction mix for five samples: 20  ml Ampuwa, 50  ml Gene Expression master mix, 5  ml primer, and a total of 25  ml diluted preamplified cDNA). For negative controls, replace diluted and preamplified cDNA by water. 3. Cover the plate with an adhesive foil, closing up each well. 4. Centrifuge plate briefly at 530 ´ g. 5. Place 96-well plate in 7500 Real-Time PCR System and incubate (thermocycler program: 1 cycle of 50°C for 2 min, 1 cycle of 95°C for 10 min, 45 cycles of 95°C for 15 s and 60°C for 1 min; see Note 4).

3.7. Primary Data Evaluation

1. Visualize the amplification curves and CT values with the SDS (Version 1.3.1). 2. Normalize to the mean expression of endogenous controls and convert into 2T-DC values. Visually inspect the expression curves for every reaction: amplification curves with CT values below 35 and with a regular sigmoid-shaped plot are considered valid. Atypically shaped curves with double-sigmoid form, an early flat slope, or no plateau are not considered meaningful. Omit these from further analysis, even when the CT values are below 35. We selected podocyte-associated (nephrin and Wilms tumor protein) target genes as well as reference genes (POLR2A) and were able to generate reproducible (measurements were replicated twice) mRNA expression results from laser-microdissected human glomeruli by employing the methods described above. Hundred isolated glomeruli from deparaffinized tissue samples resulted in CT values that exceeded those of 50 non-deparaffinized glomeruli from the same kidney by 0.9–2.1 cycles (Fig. 1 and Table 1).

5  Optimized RNA Extraction from Non-deparaffinized, Laser-Microdissected Material

73

Fig. 1.  (a–d) Isolation of a glomerulum from the surrounding, unagitated kidney tissue by laser-assisted microdissection (the laser cut is visible in (b); (d) shows the isolated glomerulum in the adhesive cap). Original magnification: ×100. I× 71 microscope (Olympus, Germany) with CellCut Plus system for laser-assisted microdissection.

Table 1 Average CT values of selected genes from deparaffinized and non-deparaffinized specimen glomeruli from the same kidney CT value Gene

50 Non-deparaffinized glomeruli

100 Deparaffinized glomeruli

POLR2A

29.00

29.91

Nephrin

24.52

25.40

Wilms tumor protein

28.51

30.60

Therefore, microdissection of half the glomeruli in non-deparaffinized slides resulted in an expression signal about twice as strong. As laser-assisted microdissection of circumscribed structures in non-deparaffinized slides is as feasible as in deparaffinized and regularly stained slides, we propose our method as the preferred one for samples, where (immuno) histochemistry is not needed for target labeling, e.g., prominent anatomical structures, like blood vessels, airways, or epithelial layers (Figs. 2 and 3).

74

D. Jonigk et al.

Fig. 2. (a–d) Isolation of a segment of an umbilical artery by laser-assisted microdissection. Note the vascular wall from which another segment has already been isolated (the laser cut is visible in (b); (d) shows the isolated part of the vascular wall in the adhesive cap). Original magnification: ×100. I× 71 microscope (Olympus, Germany) with CellCut Plus system for laser-assisted microdissection.

Fig. 3. (a–d) Isolation of a pulmonary vein from the surrounding lung tissue by laser-assisted microdissection (the laser cut is visible in (b); (d) shows the isolated vein in the adhesive cap). Note the delicate adjacent alveolar walls. Original magnification: ×100. I× 71 microscope (Olympus, Germany) with CellCut Plus system for laser-assisted microdissection.

5  Optimized RNA Extraction from Non-deparaffinized, Laser-Microdissected Material

75

4. Notes 1. In our institute, we recommend the routine use of cut-resistant gloves that are worn under your examination gloves. While they offer only a very limited protection against punctures, the resistance against scalpel-type knives and scissors is very good. 2. All work with RNA and DNA should be performed in nucleasefree conditions. 3. The optimal settings of laser energy level and focus depend largely on the tissue that you are going to cut. It is always advisable to test your settings on a marginal area of your section before moving on to the desired targets. Remember that a significant reduction of your cutting speed is also an efficient way to increase the penetration depth of your laser beam. 4. Due to the tremendous amplification power of the PCR (multiplication up to 1013 times) and the resulting risk of contamination, a strict physical separation of reaction products (including preamplified cDNA) from all sample preparation has to be accomplished. To this end, we maintain a laboratory (pre-PCR area), where sectioning, laser-assisted microdissection, and PCR preparation are performed. Everything in this laboratory (including lab coats, pipettes, notepads, etc.) is dedicated exclusively to this room and strictly separated from the post-PCR area. Workplaces and plastic labware are regularly cleaned using a 3% hypochlorite solution. PCR products are analyzed in a separate laboratory (post-PCR area). Under no circumstances should (pre)amplified samples or equipment from this laboratory be brought back to the pre-PCR area. References 1. von Weizsacker F, Labeit S, Koch HK, Oehlert W, Gerok W, Blum HE. (1991) A simple and rapid method for the detection of RNA in formalin-fixed, paraffin-embedded tissues by PCR amplification. Biochem Biophys Res Commun 174, 176 –180. 2. Stanta G, Schneider C. (1991) RNA extracted from paraffin-embedded human tissues is amenable to analysis by PCR amplification. Biotechniques 11:304, 306, 3088. 3. Lewis F, Maughan NJ, Smith V, Hillan K, Quirke P. (2001) Unlocking the archive– gene expression in paraffin-embedded tissue. J Pathol 195, 66–71.

4. Jonigk D, Lehmann U, Stuht S, et al. (2007) Recipient-derived neoangiogenesis of arterioles and lymphatics in quilty lesions of cardiac allografts. Transplantation 84, 1335–1342. 5. Leiva IM, Emmert-Buck MR, Gillespie JW. (2003) Handling of clinical tissue specimens for molecular profiling studies. Curr Issues Mol Biol 5, 27–35. 6. Theophile K, Jonigk D, Kreipe H, Bock O. (2008) Amplification of mRNA from lasermicrodissected single or clustered cells in formalin-fixed and paraffin-embedded tissues for application in quantitative real-time PCR. Diagn Mol Pathol 17, 101–106.

wwwwwwwwwwwwwww

Chapter 6 Laser Capture Microdissection for Analysis of Gene Expression in Formalin-Fixed Paraffin-Embedded Tissue Ru Jiang, Rona S. Scott, and Lindsey M. Hutt-Fletcher Abstract A combination of laser capture microdissection and reverse transcriptase real-time quantitative PCR ­provides a powerful tool for the analysis of relative gene expression in archived tissue specimens. This chapter describes standard methodologies that can be used to determine the relative levels of gene expression in individual cells captured from formalin-fixed paraffin-embedded tissues. Key words: Formalin-fixed tissues, Laser capture microdissection, Gene expression, Reverse transcription, Real-time quantitative PCR

1. Introduction Much information about cell behavior and regulation can be revealed by examining gene expression in cells which are grown as homogeneous cultures in vitro. Sophisticated, controlled analyses of molecular changes in response to different stimuli are possible. However, gene expression within individual cells in complex tissues in vivo can be very different. Laser capture microdissection (LCM) provides the opportunity to bridge the gap between the two (1). Combined with gene arrays or real-time quantitative PCR (RT-QPCR), it can provide the invaluable comparisons which are needed to evaluate the broader biological relevance of in vitro findings. Many archival human tissues are formalin-fixed and paraffin-embedded (FFPE). This is not optimal for the preservation of RNA that is still amenable to transcription, but these tissues represent, in many cases, samples with known outcomes and as such are a valuable resource which should not be lost.

Graeme I. Murray (ed.), Laser Capture Microdissection: Methods and Protocols, Methods in Molecular Biology, vol. 755, DOI 10.1007/978-1-61779-163-5_6, © Springer Science+Business Media, LLC 2011

77

78

R. Jiang et al.

Identification of a cell type within an FFPE tissue section can be based on a combination of context and morphology. More definitive markers can be provided by immunohistochemistry. However, immunohistochemical protocols, which can require long periods of immersion in aqueous media, may result in further degradation of RNA (2). This chapter focuses on techniques for using LCM and RT-QPCR to evaluate gene expression in FFPE tissue sections stained with hematoxylin and eosin. A combination of context, morphology and concurrent analysis of genes known to be silent or expressed in a given tissue can be used for identification.

2. Materials 2.1. Section Preparation

1. Microtome. 2. Hematoxylin. 3. Eosin Y (Richard-Allan Scientific, Kalamazoo, MI). 4. Histological grade xylene. 5. 70, 95, and 100% ethanol. 6. Ultrapure RNase-free water.

2.2. Laser Capture Microdissection, RNA Isolation, and Reverse Transcription

1. PixCell IIe LCM system (Arcturus Engineering, Inc., Mountain View, CA). 2. CapSure HS LCM caps (Arcturus) coated with infrared lightabsorbing ethylene vinyl acetate. 3. CapSure Sample Preparation System with CapSure HS Alignment Tray, Incubation Block, and ExtracSure Sample Extraction Devices (Arcturus). 4. Paradise Whole Transcript RT Reagent System which includes all reagents for RNA isolation and reverse transcription with the exception of the reverse transcriptase itself (Arcturus). 5. SuperScript III Reverse Transcriptase Enzyme (Invitrogen, Carlsbad, CA). 6. RNase Zap (Ambion; Applied Biosystems, Foster City, CA) or other commercially available nuclease decontamination solution. 7. Nuclease-free PCR tubes and filtered pipette tips. 8. Tweezers. 9. Hybridization oven. 10. NanoVue (GE) or Agilent 2100 Bioanalyzer (Agilent) or similar instrument.

6  Laser Capture Microdissection for Analysis of Gene Expression¼

2.3. Real-Time Quantitative PCR

79

1. ABI Prism 7500 Fast Real-Time PCR System (Applied Biosystems). 2. FAM-labeled target probes (Integrated DNA Technologies, Coralville, IA) and VIC-labeled probe for glyceraldehyde-3phosphate dehydrogenase (GAPDH) or other housekeeping gene to be used as a reference gene (Applied Biosystems). 3. 2× TaqMan® Universal PCR Master Mix (Applied Biosystems). 4. 2× SYBR® Green Master Mix (Applied Biosystems). 5. MicroAmp Fast Optical 96-well Reaction Plates (Applied Biosystems). 6. Optical Adhesive Film (Applied Biosystems).

3. Methods RNA in formalin-fixed tissues deteriorates over time. It is often difficult to isolate sufficient quantity of intact RNA for reverse transcription from tissues that have been fixed for more than 4 or 5 years, and even tissues that have been stored for shorter periods of time can prove challenging. Thus, while RT-QPCR can be sensitive enough under optimal conditions to examine gene expression in a single cell, in practice, it is often necessary to isolate many cells from FFPE tissues to accomplish the same goal. 3.1. Preparation of Tissue for Microdissection

1. Cut 4–5 mm sections (see Note 1) of FFPE tissue onto standard glass slides. 2. Begin to deparaffinize the sections by dipping slides into two separate jars of xylene for 5 min at a time. Follow this by dipping slides ten times in 100% ethanol for 1–2 s, ten times in 95% ethanol, ten times in 70% ethanol, and ten times in ultrapure, RNase-free water. 3. Dip slides in hematoxylin for approximately 1 min and follow that by dipping slides ten times each for 1–2 s in two separate jars of ultrapure RNase-free water, one of 70% ethanol and one of 95% ethanol. 4. Dip slides in eosin Y for approximately 1 min and follow that by dipping slides twice in 95% ethanol for 1–2 s and ten times in 100% ethanol. Wash in xylene for 1 min and air dry for at least 2 min. 5. Store slides until use in a box containing desiccant.

3.2. Laser Capture Microdissection

1. Turn on all the components of the PixCell IIe system. 2. Start the archiving software and enter the study name, slide number or tissue name, and the Cap Lot # from the CapSure HS box of caps as prompted.

80

R. Jiang et al.

3. Put a slide in the middle of the stage over the illuminated area, move it to find a region containing the desired target cells, and activate the vacuum chuck to hold it in place. Increase illumination, choose the objective providing appropriate magnification, and adjust the focus. 4. Load CapSure Caps by sliding two cassettes each containing four unused Caps into the CapSure cassette module on the right of the stage. 5. Following the instrument operation instructions, use the Cap Placement Arm to lower a CapSure Cap onto the slide. Adjust the fine focus, and move the stage to inspect the region of the slide within the black ring that is visible. Laser capture can be performed on any cell within this ring. 6. Move to an area on the slide, where there are no cells. Enable the laser, and set the intensity at >0.200 V and the spot size to 7.5 mm. The spot size and intensity of the laser can be adjusted by the computer software or more easily by the Controller box. The laser cannot be seen through the oculars of the microscope and has to be visualized and focused on the computer screen. Switch to the 20× objective, and focus the beam as a sharp white targeting spot surrounded by one or two sharp pink or purple halos (it may be necessary to reduce the level of illumination to do this). Once focused, set up the laser spot size to approximate the size of the cells to be captured (often, this is 7.5 mm). Put the computer mouse arrow over the laser to mark it. This will allow you to increase illumination to visualize cells, should it be necessary, without losing sight of the location of the laser beam. 7. The power and duration of the laser pulse need to be set and will depend on the type of CapSure Cap. Tables indicating appropriate power and duration for different Caps and different laser spot sizes are provided by the manufacturer. Then, still on an area of slide where there are no cells, test fire the laser. There should be an audible beep as it is fired. Move the stage slightly to see the melted spot of polymer that should have been produced. There should be a visible thin ring around a clear area marking where the laser has been fired and melted the polymer. The manufacturer’s manual provides pictures of what the spot should look like and suggestions about what adjustments to the laser duration and power might be made to correct one that is not optimal. The power and duration values are saved with the spot size and will be recalled when the machine is next used. It is also possible to save personalized settings with the Archiving Software. 8. Move the stage to find cells to be captured that are within the black ring of the CapSure Cap. Position a cell that is wanted under the laser and fire the laser. Repeat until all cells within

6  Laser Capture Microdissection for Analysis of Gene Expression¼

81

the black ring that are wanted have been captured. Up to 3,000 cells may be accommodated by one cap, depending on the size of the cells. However, the cap cannot be usefully moved to another position on the sample and used to collect more cells. 9. Lift the CapSure Cap from the slide with the Cap Placement Arm. A picture of the area that has been captured and the cells remaining can be taken following the instructions provided for the Image Archiving Software. Cells that have been lifted can also be visualized by turning off the vacuum pump, moving the slide to an area where there is no sample, putting the pump back on, and lowering the Cap again. Adjust the focus, and scan the cap by moving the stage. An image of the cells that have been captured can also be made with the Image Archiving Software. If cells have not been captured, refer to the manufacturer’s troubleshooting guide for suggestions. 10. Lift the CapSure Cap again and move it to the Unload Station, a platform on the right of the microscope. Remove the cap from the Upload Station with the Cap insertion tool provided, taking great care not to touch the polymer surface of the Cap and in turn remove the Cap from the Cap insertion tool using tweezers, and put it in the Alignment tray with the sample facing up. The tray will hold up to 28 Caps. 11. Position an ExtracSure Device over the Cap with its central fill port facing up and push down to engage. The device seals the perimeter of the Cap, but allows the addition of enzymes to its surface through the Device fill port. 12. Add proteinase K solution to the fill port, and put a 0.5-ml microcentrifuge tube over the top of the port. The volume depends on the number of cells captured, but about 25 ml is appropriate for 1,000–3,000 cells. Incubate in the incubation block at 50°C for 16–20  h (a hybridization oven is convenient for this). Invert the device and cap and centrifuge cell extract into the microcentrifuge tube. If there is considerable debris in the extract, transfer the supernatant to a new tube. Freeze at −80°C or proceed immediately to RNA isolation. 3.3. RNA Isolation and Reverse Transcription

1. Follow the detailed instructions provided with the Paradise Whole Transcript RT Reagent System, taking particular care to clean the work area with nuclease decontamination solution. Wear disposable gloves and ensure that solutions are nuclease-free, plasticware is nuclease-free, and all forceps are washed with detergent and well-rinsed with nuclease-free water. The steps involved are, briefly, as follows: 2. Precondition the purification column with conditioning buffer using the collection tube provided to centrifuge out the buffer. Prepare the master mix of binding buffer and ethanol

82

R. Jiang et al.

solution, add to the cell extract in the microcentrifuge tube and load in no more than 210 ml volumes onto the preconditioned column. Centrifuge to bind RNA to the column membrane and repeat until all extract is loaded. 3. Wash as instructed, making sure that all wash buffer is removed before elution of RNA with 12 ml of elution buffer into a new microcentrifuge collection tube that is provided. Discard the purification column. At this point, the RNA can be stored at −80°C or DNase treatment can be done immediately. 4. Add 2 ml sDNase reaction mix to the RNA. Incubate tube in a thermal cycler at 37°C for 15 min, chill, and add sDNase stop solution. Inactivate DNase at 70°C for 10 min. Chill and centrifuge. Remove 1–2 ml to measure RNA yield by spectrophotometry. Either store the remaining RNA at −80°C or proceed directly to reverse transcription. 5. Transfer 10–2,000 ng of the RNA to be reverse transcribed to a microcentrifuge tube, and add the reagents provided as instructed. These include random primers and First Strand Synthesis mix comprising an RT master mix, RT enzyme mix, enhancer, and the SuperScript III reverse transcriptase, which is not provided in the kit and has to be purchased separately. Incubate as described in the user guide. Reverse transcription is then followed with a final nuclease digestion to remove any remaining RNA. The cDNA can be frozen for storage or used immediately for RT-QPCR. 3.4. RT-QPCR

1. Primers and probes used for RT-QPCR can be designed by a software program, such as Beacon Designer 7.0 (Premier Biosoft International) or Primer Express 2.0 (Applied Biosystems). Check the specificity of the primers using cDNA obtained from a cell line known to express the target gene and SYBR Green reagents (see below). A dilution series should be prepared for primer validation. The dissociation curves of the amplified products, melting temperatures, and profiles should be similar. Also visualize the specificity of the PCR product on an agarose gel. 2. Set up RT-QPCR in duplicate in a MicroAmp Fast Optical 96-well plate. For SYBR green reactions, use 3  ml cDNA, 15 ml TaqMan Universal PCR master mix, and a final concentration of primers of 300 nM in a total volume of 30 ml. For TaqMan reactions, use 3 ml cDNA, 15 ml TaqMan Universal PCR master mix, primers at a final concentration of 300 nM, and probes at a final concentration of 200  nM in a final volume of 30  ml (see Note 2). Seal the plate with Optical Adhesive Film. 3. For TaqMan thermocycling, run 40 cycles of 95°C for 15  s and 60°C for 1 min with an initial cycle of 50°C for 2 min and

6  Laser Capture Microdissection for Analysis of Gene Expression¼

83

95°C for 10 min. For SYBR green, after the PCR cycles, add a dissociation step of 95°C for 15 s, 60°C for 1 min, 95°C for 15 s, and 60°C for 15 s to obtain a melting curve (see Note 3). 4. Set the background immediately prior to the exponential rise in amplification. Set the threshold cycle (Ct) of target and reference genes to ten standard deviations above the background. If the relative efficiencies of amplification of target and reference genes are similar (see Note 4), the relative amounts of the target gene can be calculated by the comparative Ct method (3) and, if desired, reactions can be multiplexed. DCt represents the difference in Ct values between the target and reference genes at the same condition points for each sample. DDCt is described by the equation: DDCt = DCt (target gene) − DCt (target gene expression in a cell line chosen as an arbitrary standard). The numerical value for DDCt is then used to calculate 2−DDCt, which represents the differential expression of the target gene in the sample relative to that in the chosen arbitrary standard. If the efficiencies of amplification of target and reference genes are not the same, then reactions for each are run separately and compared to a standard curve constructed from an RNA of known concentration (see Note 4).

4. Notes 1. Sections of up to 7.5 mm can be used, but those of 4–5 mm are easier to capture with the LCM system. 2. Primer concentrations may have to be optimized for each given set and scaled to the amount of template available. 3. Ct values for amplification of reverse-transcribed RNA obtained from FFPE sections may be high because of the difficulties in isolating sufficient quantities of good-quality RNA, even when multiple cells are captured. This is particularly true for sections from tissues more than 2 or 3 years old. To ensure that the Ct values do not represent artifacts, it is important to ensure that melting temperatures of individual cycles are identical. This can either be checked using SYBR Green and comparing the dissociation curves of the amplified products, which should be the same, or by determining if the Ct values fall in the linear range previously determined for the primer set in Subheading 3.4, step 1. 4. To make valid comparisons of the expression of different genes, it is important to verify that the relative efficiencies of amplification for each primer set are similar to those of the reference gene. The absolute value of the slope of a plot of

84

R. Jiang et al.

log cDNA concentration vs. DCt (Cttarget − CtGAPDH) should be less than 0.1. If it is not, then amplification efficiencies of each primer set will have to be taken into account in the quantitation method (4).

Acknowledgments This work was supported by grants DE 016669 and CA114416. References 1. Curran, S., McKay, J. A., McLeod, H. L., Murray, G. I. (2000) Laser capture microscopy. Journal of Clinical Pathology 53, 64–68. 2. Fend, F., Emmert-Buck, M. R., Chuaqui, R., Cole, K., Lee, J., Liotta, L. A., Raffeld, M. (1999) Immuno-LCM: laser capture microdissection of immunostained frozen sections for mRNA anal­ ysis. American Journal of Pathology 154, 61–66.

3. Livak, K. J., Schmittgen, T. D. (2001) Analysis of relative gene expression data using real-time quantitative PCR and the 2(−Delta Delta C(T)) Method. Methods 25, 402–408. 4. Pfaffl, M. W. (2001) A new mathematical model for relative quantification in real-time RT-PCR. Nucleic Acids Research 29, e45.

Chapter 7 MicroRNA Profiling Using RNA from Microdissected Immunostained Tissue Clemens L. Bockmeyer, Danny Jonigk, Hans Kreipe, and Ulrich Lehmann Abstract mRNA expression profiling has been used to define molecular subtypes of human breast cancer. Also microRNAs have been investigated in these breast cancer subtypes. However, little is known regarding the microRNA signature of healthy luminal and basal breast epithelial cells. Therefore, a method is described to isolate immunostained luminal and basal breast epithelial cells in formalin-fixed paraffin-embedded tissues by laser microdissection. Employing this new methodological approach, we could identify distinct microRNA profiles of luminal and basal breast epithelial cells by real-time PCR-based profiling. Key words: Breast, Laser microdissection, miRNA, Myoepithelial, Luminal

1. Introduction MiRNA gene expression studies are in the focus of many research areas (1). Not only in cancer, but also in inflammatory or vascular disease, miRNAs are important regulators of many pathways (2–4). Real-time PCR is suitable for quantitative studies of changes in the individual miRNA expression of small amounts of miRNAs (5). Precise quantification is achieved routinely with as little as 25 pg of total RNA for most miRNAs (6). Multiplex approaches are also developed to profile all known miRNAs similar to the microarray approach, but with higher sensitivity for the detection of small amounts of RNA (7). Breast epithelial cells have a specialized phenotype maintained by the expression of a unique set of mRNA genes (8). Furthermore, in situ hybridization studies performed on healthy breast tissues

Graeme I. Murray (ed.), Laser Capture Microdissection: Methods and Protocols, Methods in Molecular Biology, vol. 755, DOI 10.1007/978-1-61779-163-5_7, © Springer Science+Business Media, LLC 2011

85

86

C.L. Bockmeyer et al.

identified an alteration of miRNA expression (9). This method is excellent for direct visualization of cell-specific miRNA transcripts. However, in contrast to real-time PCR, in situ hybridization provides no quantification, has a low-throughput and a low specificity in discriminating between closely related microRNA transcripts. The MMI CellCut plus laser microdissection system is a robust and reliable tool for contamination-free isolation of single cells or small groups of cells. It allows sample inspection at high magnification, and cells of interest are easy to dissect due to solid-state laser technology and user-friendly operation software (10). This technique is especially useful for dissection of single cells like megakaryocytes (11). Combination of this laser microdissection system and real-time PCR allows appropriate gene expression profiling of specific cell types. In this chapter, a detailed protocol is provided for microdissection of luminal and basal breast epithelial cells and subsequent RT-PCR-based profiling of miRNA transcripts.

2. Materials Manufacturers or distributors are specified only if reagents or laboratory equipment might be important for the outcome or if a source might be difficult to identify. All chemicals were purchased in analytical grade quality from Merck, Roth, J.T Baker, or Sigma. 2.1. Tissue Preparation

1. 4% Phosphate-buffered formaldehyde. 2. Graded ethanols (100, 90, 70, and 50%). 3. Xylene. 4. Paraffin. 5. Membrane slides for laser microdissection, nuclease-free (Molecular Machines & Industries, Glattburg, Switzerland). 6. Hydrogen peroxide. 7. Double-distilled H2O. 8. 50 mM Tris–HCl, pH 7.4, 0.9%NaCl (Tris buffer). 9. ZytoChem Plus (HRP) Polymer Kit (Zytomed Systems). 10. Smooth muscle actin antibody, monoclonal anti-mouse, clone 1A4 (DAKO). 11. Antibody diluent (Zytomed Systems). 12. Diaminobenzidine (Zytomed Systems). 13. Hemalaun. 14. Tubes for laser microdissection (Molecular Machines & Industries, Glattburg, Switzerland).

7  MicroRNA Profiling Using RNA from Microdissected Immunostained Tissue

2.2. RNA Extraction

87

1. Proteinase K solution (20 mg/mL, Merck). 2. Guanidinium thiocyanate. 3. 1 M Tris–HCl pH 7.6. 4. ß-Mercaptoethanol. 5. 3 M Sodium acetate. 6. Roti-aqua-phenol (Roth, Karlsruhe). 7. Chloroform. 8. Isopropanol. 9. Glycogen (Roche, Basel). 10. Ethanol. 11. DEPC H2O. 12. Digestion solution: 50  μl 4.2  M guanidinium thiocyanate/ 30  mM Tris–HCl pH 7.6/2% Sodium-N-lauryl sarcosine, 50 ml proteinase K, and 0.5 ml ß-Mercaptoethanol.

2.3. cDNA Synthesis and Preamplification

1. TaqMan microRNA RT Kit (Applied Biosystems). 2. Megaplex RT primers, human Pool A and B (Applied Biosystems). 3. Megaplex Preamp primers, human Pool A and B (Applied Biosystems). 4. Preamp mastermix (Applied Biosystems).

2.4. Real-Time PCR-Based Profiling

1. Universal PCR Mastermix (Applied Biosystems).

2.5. Equipment

1. Automatic tissue-processing device.

2. TaqMan human microRNA Array A and B (Applied Biosystems).

2. Microtome. 3. MMI CellCut plus laser microdissection system (Molecular Machines & Industries, Glattburg, Switzerland). 4. Centrifuge (Thermos scientific, Fresco 17). 5. GeneAmp PCR system 9700 – 96-well (Applied Biosystems). 6. Centrifuge (Heraeus, Multicentrifuge 3SR). 7. Sealer (Applied Biosystems). 8. Array holder for centrifugation (Applied Biosystems). 9. 7900 HT Fast Real-Time PCR system (Applied Biosystems). 10. SDS 2.3 software (Applied Biosystems). 11. RQ manager 1.2 software (Applied Biosystems).

88

C.L. Bockmeyer et al.

3. Methods The protocols described below for microRNA profiling of immunostained laser-microdissected tissue concentrate on the following steps: 1. Specimen preparation and immunohistochemistry. 2. Laser microdissection. 3. Isolation of RNA. 4. Megaplex cDNA synthesis and preamplification. 5. MicroRNA profiling. 6. Data interpretation. 3.1. Specimen Preparation (see Note 1) and Immunohistochemistry

This work has to be done under nuclease-free conditions. 1. Serial 5 mm breast tissue sections are cut with a fresh knife, floated out on a hot water bath, and finally mounted on membrane slides. Tissue sections should be positioned in the center of the slide and on the bottom side. Slides are dried at 37°C overnight. 2. Slides are deparaffinised, incubated for 10 min in 3% hydrogen peroxide, and rehydrated in 50 mM Tris buffer for 5 min. 3. Sections are incubated for 5 min with a commercial blocking solution (Reagent 1, ZytoChem Plus (HRP) Polymer Kit) and washed in Tris buffer for 5 min. 4. Slides are incubated for 1  h with SM-actin antibody (room temperature, 1:25 diluted) and washed in 50 mM Tris buffer for 5 min (see Note 2). 5. For signal intensification, reagent 2 [ZytoChem Plus (HRP) Polymer Kit] is applied for 20 min and slides are washed in Tris buffer for 5 min. 6. Slides are incubated for 30 min with secondary anti-mousepolymer antibody (Reagent 3, ZytoChem Plus (HRP) Polymer Kit) and washed in Tris buffer for 5 min. 7. Slides are incubated for 5 min in distilled water and incubated for 8 min with DAB (1:20 diluted). 8. Slides are washed in distilled water for 5  min and counterstained with hemalaun for 2 min (see Note 3). 9. Slides are dried at 37°C overnight.

3.2. Laser Microdissection

This work has to be done under nuclease-free conditions. For stabilization, one glass slide is positioned under the membrane slide, so the tissue is located between a glass slide and the membrane. It is necessary to use the 40× objective for dissection

7  MicroRNA Profiling Using RNA from Microdissected Immunostained Tissue

89

Fig. 1. Laser microdissection of immunostained breast epithelial cells. Basal cells were stained for smooth muscle actin (horseradish peroxidase, dark grey ), whereas nuclei were counterstained by hemalaun (a). First luminal cells were cut (b) out and then retrieved using adhesive isolation caps (c). Afterward, basal cells were cut out and retrieved using a second cap (d) (original magnification: ×400).

of luminal and basal breast epithelial cells. First, all luminal cells were cut and collected by the adhesive lid (Fig. 1). In our experience, one can collect about 100 acini by the adhesive membrane of the lid. In the second step, all basal cells were cut and collected by the adhesive lid of a second tube. In total, for each sample, about 3,000 luminal cells and 1,600 basal cells were dissected. Using the 40× objective, cutting speed 4, laser focus 43, and power 71, the estimated line width is optimal for dissection of luminal and basal breast epithelial cells (about 0.8 mm; see Note 4). The MMI laser has the advantages of using low pulse energy and high repetition rate. 3.3. Isolation of RNA

This work was done under nuclease-free conditions (see Note 5). 1. Microdissection tubes are filled with 50 ml digestion solution, and then the tubes are turned upside down and incubated at 55°C overnight.

90

C.L. Bockmeyer et al.

2. The tubes have to be centrifuged briefly, and the lysate has to be transferred into a normal tube without adhesive lid. This step is necessary due to the fact that microdissection tubes cannot be centrifuged for a long time at high speed because the adhesive lid might drop down into the tube during prolonged centrifugation. 3. For phenol–chloroform extraction, 100  ml lysate is mixed with 10 ml 3 M sodium acetate, 63 ml Roti-aqua-phenol, and 27 ml chloroform strongly vortexed and incubated on ice for 20 min. It is necessary to maintain the order of adding first sodium acetate, then Roti-aqua-phenol, and finally chloroform. Subsequently, the samples are centrifuged for 30 min at 16,200 × g. 4. The supernatant (95  ml) is precipitated by incubation with 95 ml ice-cold isopropanol and 1.5 ml glycogen (20 mg/mL) at −20°C for at least one night. 5. After centrifugation for 30 min (16,200 × g), the supernatant is discarded and the pellet is washed in 180 ml ice-cold ethanol (70%). 6. After centrifugation for 5 min at 16,200 × g, the supernatant is again discarded and the pellet air-dried (see Note 6). 7. The RNA pellet is dissolved in 15 ml DEPC H2O and stored at −20°C. (Note 7). 3.4. Megaplex cDNA Synthesis and Preamplification

1. 3 ml RNA is mixed with 4.5 ml megaplex–mastermix (0.8 ml megaplex RT primer, 0.2 ml dNTPs (100 mM), 0.2 ml DEPC H2O, 0.8 Ml 10× RT buffer, 0.9  ml MgCl, 0.1  ml RNaseInhibitor, 1.5  ml multiscribe reverse transcriptase) (each for human pool A and B). 2. Cycling modus (ABI Gene Amp 9700 HT): 40 cycles at 16°C for 2 min, 42°C for 1 min, and 50°C for 1 s, and last 1 cycle at 85°C for 5 min. 3. 2.5  ml cDNA is preamplified: 12.5  ml preamp mastermix, 2.5  ml megaplex–preamp primers, 7.5  ml DEPC H2O, and cDNA are mixed (each for human pool A and B). 4. Cycling modus (ABI Gene Amp 9700 HT): 1 cycle at 95°C for 10 min, 1 cycle at 55°C for 2 min, 1 cycle at 72°C for 2 min, 12 cycles at 95°C for 15 s and 60°C for 4 min, 1 cycle at 99°C for 10 min.

3.5. MicroRNA Profiling (See User Bulletin Applied Biosystems TaqMan Low-Density Array)

1. For miRNA profiling, we use the so-called TaqMan low-density arrays. One array has 384 reaction chambers therein lyophilized primers are provided. 2. For one measurement, 24  ml preamplified cDNA, 388.5  ml DEPC H2O, and 412.5  ml universal PCR mastermix are

7  MicroRNA Profiling Using RNA from Microdissected Immunostained Tissue

91

mixed. 100 ml of this sample-specific reaction mix is dispensed in each fill port. 3. Arrays are placed into special array holders and centrifuged two times at 200 ´ g for 1 min. 4. After that, the microfluidic card has to be carefully placed in the sealer to seal each reaction chamber of the array. 5. Real-time PCR is performed on the 7900 HT PCR system. Cycling modus: 1 cycle at 50°C for 2 min, 1 cycle at 94.5°C for 10 min, 45 cycles at 97°C for 30 s and 59.7°C for 1 min. 3.6. Data Interpretation

Amplification curves and CT values are generated with the software RQ manager 1.2 (Applied Biosystems). Amplification curves for every reaction are inspected visually and underwent a stringent quality control procedure: 1. Transcripts at a detectable level are defined as those with a CT value below 38 and a regular sigmoid-shaped amplification curve. 2. “No meaningful amplification curve” is defined as atypical curve with double-sigmoid form, early flat slope, or no amplification plateau (Fig.  2). This curve is omitted from further analysis, even when CT values are below 38. All miRNA transcripts are classified into three groups: (1) no detectable expression, (2) “no meaningful amplification curve” considered as not analyzable, and (3) distinct gene expression patterns. Unfortunately, up to now, no universally accepted consensus of how to evaluate amplification plots has been established. 3. For the identification of suitable endogenous controls, eight different candidate microRNAs were analyzed for variance in gene expression according to two different algorithms: NormFinder (12) and geNorm (13). Both statistical methods ranked the candidate endogenous control genes with an excellent correlation in raw stability values (Fig. 3). The most stably expressed were snRNA U6 and RNU48.

4. Notes 1. It is difficult to get healthy mammary tissues with sufficient amounts of epithelial cells. Therefore, you have to sample about ten tissue blocks from one tissue specimen. Use the best block for microdissection. 2. It is necessary to use immunohistochemistry for clear separation of luminal from basal cells. In our experience, only hemealaun staining does not provide sufficient distinction between luminal

92

C.L. Bockmeyer et al.

Fig. 2. “No meaningful amplification curves”. Amplification curves of six luminal samples each for miR-224 and miR-551b are shown. Amplification curves were defined as “no meaningful amplification curves,” when curves demonstrated a double-sigmoid form, early flat slope, or no amplification plateau.

7  MicroRNA Profiling Using RNA from Microdissected Immunostained Tissue

93

NormFinder (stability value)

2.0 miR17

1.5 RNU44

1.0 RNU24

0.5

miR191 RNU48 snRNU6

2.0

p < 0.01 r = 0.91

RNU43

2.5

3.0

3.5

geNorm (M-value) Fig.  3. Normalization of different candidate genes for qPCR of miRNA transcripts. CT values of snRNU6, RNU6B, RNU24, RNU43, RNU44, RNU48, miR-17, and miR-191 were analyzed by calculating DCT values [CT (luminal sample–basal sample); n = 5]. Relative quantities (2−DCT) for each candidate gene were measured for stability by two different calculations (geNorm and NormFinder). snRNU6 and RNU48 were most stable as indicated by a low-stability value and M-value.

and basal cells probably due to the reduced optical quality. You have to keep in mind that tissue sections are dried and not coverslipped for microdissection. We stained SMA to visualize basal cells due to the fact that this antigen does not need to be unmasked. We have no experience with heat-induced or enzymatic antigen retrieval. Immunohistochemistry was performed under PCR contamination-free conditions. In general, you have to consider that all steps after RNA isolation have to be made under PCR contamination-free conditions. 3. Slides should be stained only for 1  min with hemealaun. Otherwise, staining results may be too dark and it could be difficult to differentiate between luminal and basal cells. In our experience, it is better to use tubes with a transparent lid for laser microdissection of brown and blue stained tissues. 4. The exact values might vary slightly depending on the original installation of the laser by the manufacturer. 5. We use the conventional guanidine-phenol-chloroform method with an ethanol precipitation step. Different methods for RNA extraction and consecutive mRNA analysis are described elsewhere (see, e.g., ref. 14). You have to keep in mind that not all published protocols or commercially available kits are suitable for the isolation of very short, mature microRNAs. 6. Avoid prolonged air drying because overdried RNA pellets cannot be dissolved again. 7. For short-term storage (up to a few weeks), aqueous RNA solutions might be stored at −20°C. However, for longer

94

C.L. Bockmeyer et al.

periods of time, storage at −80°C is recommended. For longterm storage, precipitation is the best way of preserving RNA integrity. References 1. Boyd SD. (2008) Everything you wanted to know about small RNA but were afraid to ask. Lab Invest. 88, 569–78. 2. Mirnezami AH, Pickard K, Zhang L, Primrose JN, Packham G. (2009) MicroRNAs: key players in carcinogenesis and novel therapeutic targets. Eur J Surg Oncol. 35, 339–347. 3. Suárez Y, Sessa WC. (2009) MicroRNAs as novel regulators of angiogenesis. Circ Res. 104, 442–54. 4. Urbich C, Kuehbacher A, Dimmeler S. (2008) Role of microRNAs in vascular diseases, inflammation, and angiogenesis. Cardiovasc Res. 79, 581–588. 5. Tang F, Hajkova P, Barton SC, Lao K, Surani MA. (2006) MicroRNA expression profiling of single whole embryonic stem cells. Nucleic Acids Res. 34, e9. 6. Chen C, Ridzon DA, Broomer AJ, Zhou Z, Lee DH, Nguyen JT, Barbisin M, Xu NL, Mahuvakar VR, Andersen MR, Lao KQ, Livak KJ, Guegler KJ. (2005) Real-time quantification of microRNAs by stem-loop RT-PCR. Nucleic Acids Res. 33, e179. 7. Lao K, Xu NL, Yeung V, Chen C, Livak KJ, Straus NA. (2006) Multiplexing RT-PCR for the detection of multiple miRNA species in small samples. Biochem Biophys Res Commun. 343, 85–89. 8. Jones C, Mackay A, Grigoriadis A, Cossu A, Reis-Filho JS, Fulford L, Dexter T, Davies S, Bulmer K, Ford E, Parry S, Budroni M, Palmieri G, Neville AM, O’Hare MJ, Lakhani SR. (2004) Expression profiling of purified normal human luminal and myoepithelial breast

cells: identification of novel prognostic markers for breast cancer. Cancer Res. 64, 3037–3045. 9. Sempere LF, Christensen M, Silahtaroglu A, Bak M, Heath CV, Schwartz G, Wells W, Kauppinen S, Cole CN. (2007) Altered MicroRNA expression confined to specific epithelial cell subpopulations in breast cancer. Cancer Res. 67, 11612–11620. 10. Palinauskas V, Dolnik O, Valkiūnas G, Bensch S. (2010) Laser microdissection microscopy and single cell PCR of avian haemosporidians. J Parasitol.:1. 11. Theophile K, Hussein K, Kreipe H, Bock O. (2008) Expression profiling of apoptosisrelated genes in megakaryocytes: BNIP3 is downregulated in primary myelofibrosis. Exp Hematol. 36, 1728–1738. 12. Andersen CL, Jensen JL, Ørntoft TF. (2004) Normalization of real-time quantitative reverse transcription-PCR data: a model-based variance estimation approach to identify genes suited for normalization, applied to bladder and colon cancer data sets. Cancer Res 64, 5245–5250. 13. Vandesompele J, De Preter K, Pattyn F, Poppe B, et al. (2002) Accurate normalization of realtime quantitative RT-PCR data by geometric averaging of multiple internal control genes. Genome Biol 3, 34.1–34.11. 14. Votavova H, Forsterova K, Stritesky J, Velenska Z, Trneny M. (2009) Optimized protocol for gene expression analysis in formalin-fixed, paraffin-embedded tissue using real-time quantitative polymerase chain reaction. Diagn Mol Pathol. 18, 176–82.

Chapter 8 Profiling Solid Tumor Heterogeneity by LCM and Biological MS of Fresh-Frozen Tissue Sections Donald J. Johann, Sumana Mukherjee, DaRue A. Prieto, Timothy D. Veenstra, and Josip Blonder Abstract The heterogeneous nature of solid tumors represents a common problem in mass spectrometry (MS)-based analysis of fresh-frozen tissue specimens. Here, we describe a method that relies on synergy between laser capture microdissection (LCM) and MS for enhanced molecular profiling of solid tumors. This method involves dissection of homogeneous histologic cell types from thin fresh-frozen tissue sections via LCM, coupled with liquid chromatography (LC)-MS analysis. Such an approach enables an in-depth molecular profiling of captured cells. This is a bottom-up proteomic approach, where proteins are identified through peptide sequencing and matching against a specific proteomic database. Sample losses are minimized, since lysis, solubilization, and digestion are carried out directly on LCM caps in buffered methanol using a single tube, thus reducing sample loss between these steps. The rationale for the LCM-MS coupling is that once the optimal method parameters are established for a solid tumor of interest, homogeneous histologic tumor/tissue cells (i.e., tumor proper, stroma, etc.) can be effectively studied for potential biomarkers, drug targets, pathway analysis, as well as enhanced understanding of the pathological process under study. Key words: Thin fresh-frozen tissue sections, Laser capture microdissection, Liquid chromatographymass spectrometry, Solid tumor heterogeneity, Biomarker, Cancer

1. Introduction Solid tumors have a heterogeneous cellular architecture. Critical functional units include cancer cells proper and stromal elements. The histology is often complex. For instance, an epithelial tumor may contain regions of: inflammation, neovascularity, carcinoma in situ, well to poorly differentiated carcinoma, nerves, hyperplasia, etc. The tumor microenvironment is composed of both normal and modified stromal cells that serve to nurture the malignant Graeme I. Murray (ed.), Laser Capture Microdissection: Methods and Protocols, Methods in Molecular Biology, vol. 755, DOI 10.1007/978-1-61779-163-5_8, © Springer Science+Business Media, LLC 2011

95

96

D.J. Johann et al.

process. The tumor stroma is now recognized as an important area in cancer therapy, and many new therapeutic strategies target the aspects of this functional region (1). Solid tumor heterogeneity is reflective of a diversity present at the molecular level that has profound biologic and therapeutic implications (2). For instance, breast cancer is actually many different diseases with the only common characteristic being the organ of origin. Hence, the ability to directly and effectively profile solid tumors at the proteome level is essential, since proteins are the final mediators of pathologic processes and proteomics in particular can begin to characterize molecular events, such as alternative protein splicing and posttranslational modifications, which are fundamental events in physiologic/pathologic processes. Additionally, although cell culture studies are quite important, they lack a true microenvironment and thus clinical translation can be limited. Therefore, methods to deconstruct solid tumors to better enable biological understanding and biomarker discovery are needed (3, 4). Laser capture microdissection (LCM) (5) and mass spectrometry (MS) (6) are powerful independent analytical technologies. Both have been commonly used for molecular profiling of formalinfixed paraffin-embedded tissue sections (7). We have shown that an LCM-MS platform can be effectively used for proteomic profiling of thin fresh-frozen tissue sections obtained from a solid tumor in conjunction with a simple methanol-aided solubilization and digestion (8) of captured cell proteomes (9). In this chapter, we further illustrate the method employed for the proteomic profiling of a solid tumor using LCM coupled to biological MS.

2. Materials 2.1. LCM

1. TISSUE-Tek O.C.T. cryostat-mounting medium (Sakura Finetek Inc., Torrance, CA). 2. Mayer’s hematoxylin solution (Sigma, St. Louis, MO). 3. Eosin Y solution (alcohol-based) (Sigma, St. Louis, MO). 4. Scott’s Tap Water Substitute Bluing Solution (magnesium sulfate buffered with sodium bicarbonate; Fisher Scientific, Hampton, NH). 5. 100% ethanol (ethyl alcohol, absolute, 200° proof for molecular biology). 6. 70% ethanol (v/v) and 95% (v/v) were established using Milli-Q-filter with purified H2O (Millipore, Billerica, MA).

8  Profiling Solid Tumor Heterogeneity by LCM and Biological¼

97

7. Xylene. 8. CapSure® Macro LCM Caps (MDS Analytical Technologies, Sunnyvale, CA). 9. PixCell IIe, Veritas, or ArcturisXT (Arcturus Molecular Devices now MDS Analytical Technologies, Sunnyvale, CA). 10. Leica Cryostat CM 1850 UV (Leica Microsystems, Wetzlar, Germany). 11. Precleaned glass microscope slides, 25 mm × 75 mm (Fisher Scientific, Hampton, NH). 12. Membrane slide options include: (a) Pen-membrane glass slide. (b) Pen-membrane frame slide; both options available from (Arcturus Molecular Devices, now MDS Analytical Technologies, Sunnyvale, CA). 2.2. Protein Extraction and Digestion

1. Ammonium bicarbonate. 2. Sequencing grade trypsin (Promega, Madison, WI). 3. Trifluoroacetic acid (TFA) and formic acid (FA, Fluka, Milwaukee, WI). 4. HPLC grade acetonitrile (ACN; CH3CN) and methanol (MeOH; CH3OH, EM Science, Darmstadt, Germany). 5. Tris[2-carboxyethyl] phosphine (TCEP; Pierce, Rockford, IL). 6. Barnstead Nanopure water purification system (Barnstead, Dubuque, IA). 7. ZipTips packed with C18 reversed-phase resin Millipore (Billerica, MA, USA). 8. MeOH lysis buffer [50 mM NH4HCO3 with 100% MeOH (v/v 40/60)].

2.3. RP-LC-MS

1. HPLC grade ACN (CH3CN, EM Science, Darmstadt, Germany).

2.4. Computational Support for CollisionInduced Dissociation Spectra Analysis

1. Single computer workstation or a cluster computer that follows a Beowulf design model (see Note 1). 2. Software for protein database search and match to experimental mass spectrometry data (see Note 2). 3. Nonredundant human proteome database. 4. Software for reverse database creation for the assessment of a false-positive rate. 5. Software to analyze the experimental data for biologic classification and implications (see Note 3).

98

D.J. Johann et al.

3. Methods To obtain reliable and reproducible results, it is important to handle the sample rapidly and effectively during the tissue acquisition step. During a surgical procedure, once a solid tumor is ligated from its blood supply, tissue degradation and possibly frank necrosis will eventually commence. Therefore, a few simple, but deliberate steps are recommended to minimize ischemic phenomena. Most succinctly, as soon as possible, the tissue should be snap frozen in liquid nitrogen and then placed in a freezer at −80°C. Subsequently, the tissue specimen will be embedded in cryostatmounting medium (TISSUE-Tek O.C.T.). Tissue sections, usually with a slice thickness range of 8–12 mm, are then serially cut with the cryostat from the frozen tissue block. As a convenient measurement or rule of thumb, the majority of cells will have a diameter either larger or within this range. Therefore, the recommended slice thickness will aid in the homogenization/ lysis procedure, since one of the key steps of the method being presented is effective liberation of proteins from captured cells. While MS is a critical component in any bottom-up proteomic analysis, sample handling factors, such as effective lysis and digestion, are the key requisites for effective large-scale protein identification. Effective digestion of small-size LCM specimens requires optimal buffering conditions. Such conditions maintain the proteins solubilized and denatured throughout the digestion process, without unnecessary manipulations and use of reagents that might interfere with LC-MS analysis. To simplify and improve the analysis of captured cells and avoid the deficiencies associated with traditional approaches (which use detergents and chaotropes), we developed a simple two-step methanol-assisted solubilization/digestion protocol. In the first solubilization step, 20% buffered methanol is used to facilitate denaturation and solubilization of cytosolic proteins. In the second step, the solubilization/digestion is carried out in a 60% methanol buffer, targeting more hydrophobic proteins, which are insoluble in 20% buffered methanol, thus resulting in enhanced proteome coverage. A schematic of this experimental workflow is shown in Fig. 1. 3.1. Initial Pathologic Analysis (Prior to LCM)

A formal H&E with coverslip should be performed with every tenth slide. Then, prior to LCM analysis, these slides should be reviewed with a pathologist in order to properly evaluate the histology, plan LCM sessions, and guard against potential bias in the z-dimension of the tumor tissue plane.

3.2. LCM Staining

The fresh-frozen tissue slide must melt before beginning the staining protocol below. Placing the slide on the palm of your glove works well. As soon as condensate forms on the entire slide,

8  Profiling Solid Tumor Heterogeneity by LCM and Biological¼

99

Fresh-frozen tissue acquisition

Pathology review LCM design

LCM slide prep

LCM

Homogenization Lysis

LCM quality control

MeOH-based twostep tryptic digestion

LC-MS/MS

Data Analysis

Desalting

Fig. 1. LCM-MS experimental design.

the protocol below may commence. In order to ensure good visualization and tissue capture, suggested times are provided for both membrane and glass slides (see Note 4). 1. 70% ethanol, fix tissue section to slide, 15 s (membrane slide), 30 s (glass slide). 2. H2O, remove OCT, rehydrate tissue, 30 s (membrane slide), 30 s (glass slide). 3. Hematoxylin, stain nuclei, 45 s (membrane slide), 30 s (glass slide). 4. H2O, remove excess hematoxylin, 15  s (membrane slide), 30 s (glass slide). 5. Bluing solution, change hematoxylin hue, 15  s (membrane slide), 30 s (glass slide). 6. 70% ethanol, start dehydration, 15 s (membrane slide), 30 s (glass slide). 7. Eosin, stain cytoplasm (1–2 quick dips), 1–2  s (membrane slide), 2 s (glass slide). 8. 95% ethanol, dehydration, 30  s (membrane slide), 1  min (glass slide).

100

D.J. Johann et al.

9. 95% ethanol, dehydration, 30  s (membrane slide), 1  min (glass slide). 10. 100% ethanol, dehydration, 30  s (membrane slide), 2  min (glass slide). 11. 100% ethanol, dehydration, 30  s (membrane slide), 2  min (glass slide). 12. Xylene, ethanol removal, 3  min (membrane slide), 3  min (glass slide). 3.3. LCM Procedure

LCM analysis may begin on the slide(s) once they are air-dried. Laser-based dissection systems allow for dissections approaching 100% purity. Staining with hematoxylin and eosin allows microscopic visualization during microdissection and does not diminish protein recovery. Generally, we have found that depending on the type of tissue under study, approximately 5,000–50,000 cells are required to produce mass spectrometry results with acceptable numbers of protein identifications, and species diversity (see Note 5). Figure  2a–c illustrates a stepwise approach for successful LCM tissue extraction by either a PixCell IIe or Veritas system. LCM tissue extraction involves: 1. Establishing a histology area of interest (Fig. 2a). 2. Manual filling of the pattern to enable the removal of a larger amount of cells (Fig. 2b). 3. LCM extraction of the cells from the selected region (Fig. 2c).

3.4. LCM Membrane Tissue/Cell Extraction and Lysis

The sample preparation protocol for protein extraction/digestion from LCM samples captured on polymer cap is now presented, initially as a brief overview and then with full details. 1. Carefully remove the LCM polymer membrane by peeling it off the cap and then place it in a siliconized tube (conical bottom). 2. Add 50  mL of 12.5  mM hypotonic lysis buffer containing 1 mM TCEP (final concentration). 3. Incubate on dry ice for 30 min. 4. Thaw the sample in ice-cold water for 10 min. 5. Incubate the sample in a water bath for 2 h at 70°C. 6. Cool the sample on ice for 20 min. 7. Adjust buffer to 50  mM by adding 1.65  mL of 1  M NH4HCO3.

8  Profiling Solid Tumor Heterogeneity by LCM and Biological¼

101

Fig. 2. LCM workflow.

3.5. First Trypsin Digestion

Trypsin dilution. The background to this protocol is based on single-cell protein content estimates that are in the range of 0.75 pg–0.5 ng (10) (see Note 6). Prepare the dehydrated and frozen trypsin, e.g., Promega Trypsin Gold, 20-mg vial. Mix with 20 mL 50 mM NH4HCO3, yielding a concentration of 1 mg/1 mL. The availability of some tissue samples is quite limited in quantity. Therefore, this section attempts to accommodate these circumstances as well as situations with more abundant tumor tissues. 1. Rehydrate trypsin by adding 20 mL of 50 mM NH4HCO3. 2. Dilute trypsin in accordance with sample cell count per the trypsin dilution protocol (Table 1). 3. Add the appropriate volume of trypsin solution and mix for 10 min. 4. Briefly vortex the sample.

102

D.J. Johann et al.

Table 1 Recommended amount of trypsin as a function of sample cell count Cell count

Protein estimate, mg

Trypsin:protein

Trypsin for sample, mg

50

0.025

1:50

0.0005

500

0.25

1:50

0.005

5,000

2.5

1:50

0.05

15,000

7.5

1:50

0.15

1:50

0.5

50,000

25

5. Place the sample in the water bath sonicator for 5 min. 6. Transfer the sample tube to a small centrifuge and spin for ~15 s. 7. Incubate the sample digest for 6 h at 37°C with good table motion. 3.6. Second Trypsin Digestion

1. Add 60% MeOH lysis buffer. This can be achieved by adding 50 mL of MeOH to each sample. 2. Add the appropriate volume of trypsin solution and mix for 10 min. 3. Briefly (~5 s) vortex the sample. 4. Place the sample in the water bath sonicator for 5 min. 5. Transfer the sample tube to a small centrifuge and spin for ~15 s. 6. Incubate the sample digest for 6 h at 37°C with good table motion. 7. Lyophilize all samples to dryness.

3.7. Desalting Using ZipTip Columns

1. Rehydrate peptides in 20 mL 0.1% TFA by sonication in water bath for 2 min. 2. Prepare 10-mL aliquots of elution buffer 60% ACN/0.1% TFA (v/v) for each sample before beginning (to avoid contamination). Avoid drawing air through the tip during the procedure (from equilibration to elution). If you find that you make bubbles in the tip, try pulling the buffers in more slowly. 3. Set the Pipetman to 10 mL and attach the ZipTip. 4. Activate the ZipTip column by pipetting up 20  mL of 60% ACN and then discard it in the waste. Repeat it three times. 5. Equilibrate the ZipTip column by pipetting up 20 mL of 0.1% TFA and then discard it in the waste. Repeat it three times. These steps act as a gradient for the minicolumn, which activates the resin and conditions it to bind peptides.

8  Profiling Solid Tumor Heterogeneity by LCM and Biological¼

103

6. Load the peptides by pipetting the sample up and down (discarding it back into its tube). Repeat it ten times. 7. Wash the ZipTip column using the 0.1 TFA. Pipette up the buffer and then discard it in the waste. Repeat it ten times. 8. Elute the sample by pipetting the ZipTip up and down in the elution buffer [back into its tube in the 4 mL 60% acetonitrile/0.1% TFA (already aliquoted)]. Repeat it ten times. The organic phase elutes the peptides off the resin into the buffer. Now, the sample is desalted as well as concentrated. 9. Lyophilize to dryness and dissolve the peptides in 10 mL of 0.1% TFA prior to LC-MS/MS analysis. 3.8. Guidelines for LC-MS Analysis of LCM Samples

Although there are a wide variety of mass spectrometer systems and liquid chromatography platforms, the linear ion trap coupled with a reversed-phase liquid chromatography separation system is widely used in the proteomics community, and will be illustrated in this section. For our LCM-based proteomic studies, a reversed-phase column is coupled with a linear ion trap mass spectrometer (LTQ ThermoElectron, San Jose, CA) for shotgun proteomic analysis. 1. In our configuration, the solvent system is delivered by an HP 1100 pump (Agilent Technologies, Palo Alto, CA). 2. A nanoelectrospray ionization source is employed applying a voltage of 1.7 kV and a capillary temperature of 160°C. 3. The LTQ is operated in a data-dependent mode. 4. The seven most abundant peptide molecular ions detected by each MS survey scan are dynamically selected for MS/MS using collision-induced dissociation (CID) facilitated by a normalized collision energy of 35%. 5. Dynamic exclusion is employed to avoid redundant acquisition of precursor ions previously selected for fragmentation. 6. Reversed-phase liquid chromatography separations are performed with a 75-mm i.d. × 10-cm long fused silica capillary column (Polymicro Technologies, Inc., Phoenix, AZ) with a flame-pulled tip (~5–7 mm orifice). 7. The column is slurry packed in-house with 5 mm, 300 Å pore size C-18 stationary phase (Vydac, Hercules, CA) using a slurrypacking pump (model 1666, Alltech Associates, Deerfield, IL). 8. Note: The total MS run time for each sample is 180 min. (a) After injecting 5 mL of sample, the column is washed for 30 min with 98% mobile phase A (0.1% FA in H2O). (b) Peptides are then eluted using a linear step gradient from 2 to 40% mobile phase B (0.1% FA in ACN) over 90 min. (c) Then, an elution gradient of 60–98% for mobile phase B over 10 min at a constant flow rate of 0.25  mL/min is performed.

104

D.J. Johann et al.

(d) Next, the column is washed for 20 min with 98% mobile phase B. (e) Finally, the column is re-equilibrated with 2% mobile phase B for 30  min prior to subsequent loading of the next sample. 3.9. Data Processing Guidelines

As previously stated, the searching and matching of experimentally obtained spectra against a nonredundant protein database are computationally intensive, but highly parallelizable, and therefore amendable to divide and conquer strategies employing cluster computers. 1. For our LTQ-derived data, the precursor ion tolerance is set to 1.5 Da and the fragment ion tolerance to 0.5 Da. These two values effectively serve as binning parameters during data acquisition concerning parent and child (fragment) ions. 2. We require candidate peptides to possess tryptic terminus at both ends, and generally will allow for a maximum of two missed tryptic cleavages. 3. The following SEQUEST thresholds are routinely used to filter experimental peptides: (a) Delta-correlation score (dCn) ³ 0.08 (b) Charge state cross-correlation scores as follows: ³2.1 for [M + H]1+ peptides ³2.3 for [M + H]2+ peptides ³3.5 for [M + H]3+ peptides 4. The final list of protein identifications is created using a parsimony principle, reporting a minimal number of protein identifications from a pool of uniquely identified peptides. 5. Resultant raw data are routinely subjected to a false-positive rate assessment via decoy (reverse) database analysis. 6. Lastly, data are analyzed for biologic implications by Ingenuity Pathway Analysis (IPA) and the Database for Annotation, Visualization, and Integrated Discovery (DAVID).

4. Notes 1. The processing of CID spectra is computationally intensive, but highly parallelizable, and follows a classic divide and conquer paradigm. Therefore, a cluster computer solution generally offers substantial time savings that follows a linear function, depending on the number of computational elements in the cluster configuration.

8  Profiling Solid Tumor Heterogeneity by LCM and Biological¼

105

2. Commercial products include MASCOT (Matrix Science, http://www.matrixscience.com) and SEQUEST (Thermo Scientific, http://www.thermo.com). Open source solutions include (1) the X!Tandem database search engine (http://www.thegpm.org/TANDEM/index.html), (2) the Trans Proteomic Pipeline (TPP, http://tools.proteomecenter. org/wiki/index.php?title=Software:TPP), and (3) the Open Mass Spectrometry Search Algorithm (OMSSA), http://pubchem.ncbi.nlm.nih.gov/omssa/. 3. Commercial products include IPA, http://www.ingenuity. com). Public domain tools include the DAVID, http://david. abcc.ncifcrf.gov). 4. For each step in the staining protocol, a different solution bath is recommended. Through experience, this has been found to make a significant difference. 5. We have found that tissues with a compact cellular density provide greater protein yields, and thus usually require a smaller quantity of cells. However, when encountering a new tumor tissue type, a few preliminary experiments are recommended for a general estimate of protein yield. 6. In addition to the provided reference, these estimates are also cited in the following text books: Molecular Biology of the Cell, 3rd Edition by Alberts et al. and Molecular Cell Biology, 4th Edition by Lodish et al.

Acknowledgments This project was funded in whole or in part with federal funds from the National Cancer Institute, National Institutes of Health, under contract HSN261200800001E. The content of this publication does not necessarily reflect the views or policies of the Department of Health and Human Services, nor does the mention of trade names, commercial products, or organizations implies endorsement by the US Government. References 1. Mbeunkui F, Johann DJ, Jr., (2009) Cancer and the tumor microenvironment: a review of an essential relationship. Cancer Chemother Pharmacol 63: 571–82. 2. Swanton C, Caldas C, (2009) Molecular classification of solid tumours: towards pathwaydriven therapeutics. Br J Cancer100: 1517–22.

3. Johann DJ, Jr., Blonder J, (2007) Biomarker discovery: tissues versus fluids versus both. Expert Rev Mol Diagn 7: 473–5. 4. Johann DJ, Wei BR, Prieto DA, Chan KC, Ye X, Valera VA, Simpson RM, Rudnick PA, Xiao Z, Issaq HJ, Linehan WM, Stein SE, Veenstra TD, Blonder J. Combined Blood/ Tissue Analysis for Cancer Biomarker

106

D.J. Johann et al.

Discovery: Application to Renal Cell Carcinoma. Anal Chem 2010. 5. Emmert-Buck MR, Bonner RF, Smith PD, Chuaqui RF, Zhuang Z, Goldstein SR, Weiss RA, Liotta LA. (1996) Laser capture microdissection. Science 274: 998–1001. 6. Aebersold R, Mann M, (2003) Mass spectrometry-based proteomics. Nature 422: 198–207. 7. Hwang SI, Thumar J, Lundgren DH, Rezaul K, Mayya V, Wu L, Eng J, Wright ME, Han DK. (2007) Direct cancer tissue proteomics: a method to identify candidate cancer biomarkers from formalin-fixed paraffinembedded archival tissues. Oncogene 26: 65–76.

8. Blonder J, Chan KC, Issaq HJ, Veenstra TD, (2006) Identification of membrane proteins from mammalian cell/tissue using methanol-facilitated solubilization and tryptic digestion coupled with 2D-LC-MS/MS. Nat Protoc 1: 2784–90. 9. Johann DJ, Rodriguez-Canales J, Mukherjee S, Prieto DA, Hanson JC, Emmert-Buck M, Blonder J. (2009) Approaching solid tumor heterogeneity on a cellular basis by tissue proteomics using laser capture microdissection and biological mass spectrometry. J Proteome Res 8: 2310–8. 10. Wibke H, Pelargus C, Keffhalm K, Ros A, Anselmetti D. (2005) Single cell manipulation, analytics, and label-free protein detection in microfluidic devices for systems nanobiology. Electrophoresis 26: 3689–96.

Chapter 9 Amplification Testing in Breast Cancer by Multiplex Ligation-Dependent Probe Amplification of Microdissected Tissue Cathy B. Moelans, Roel A. de Weger, and Paul J. van Diest Abstract This chapter describes a method for the rapid assessment of gene copy numbers in laser-microdissected materials using multiplex ligation-dependent probe amplification (MLPA). An MLPA is a powerful multiplex PCR technique that can identify gains, amplification, or losses of up to 50 genes in a single experiment, thereby requiring only minute quantities of DNA extracted from frozen or paraffin-embedded materials. A previous study in breast cancer has shown that MLPA can detect amplifications in cases with a tumor percentage lower than 10%, but still a low tumor percentage in the tissue tested could obscure low levels of amplification due to dilution of the tumor cell population by normal cells. Laser capture microdissection allows enrichment of tumor cells by eliminating background noise from normal and preinvasive cells, thereby increasing specificity and sensitivity. This chapter describes a method for MLPA analysis using invasive breast tumor cells acquired by laser capture microdissection. This protocol can also be applied to MLPA analysis of preinvasive lesions and metastases. Key words: Multiplex ligation-dependent probe amplification, MLPA, Coffalyser, Laser microdissection, Cancer

1. Introduction Several genes have been shown to be involved in the development, progression, and response to therapy of invasive breast cancer. Among these, HER-2/neu is likely the most important proto-oncogene. Amplification of the HER2 gene is present in about 15–30% of breast carcinomas and leads to protein overexpression (1, 2). Patients having this overexpression have an overall worse prognosis (3), but respond well to the treatment with

Graeme I. Murray (ed.), Laser Capture Microdissection: Methods and Protocols, Methods in Molecular Biology, vol. 755, DOI 10.1007/978-1-61779-163-5_9, © Springer Science+Business Media, LLC 2011

107

108

C.B. Moelans et al.

trastuzumab, a recombinant humanized monoclonal anti-HER2 antibody (4, 5). Furthermore, amplification of HER2 has also been shown to be associated with resistance to conventional adjuvant chemotherapy and tamoxifen (6, 7). HER2-targeted therapy in breast cancer can aid approximately 20% of women with HER2 overexpression, but no single gene copy number assessment seems to completely explain prognosis or response to therapy of individual breast cancer patients. A simultaneous analysis of copy number changes of a variety of genes involved in prognosis and therapy response may thus be very useful for molecular profiling of individual breast cancer patients. This can be achieved by an easy-to-perform, high-throughput PCR-based technique, called multiplex ligation-dependent probe amplification (MLPA). The MLPA technique was first described in 2002 by Schouten et al. (8) and is summarized in Fig. 1. Up to 50 probe sets can be run in one reaction. MLPA has been used to assess gene copy number changes (9–11), gene expression (12, 13), and methylation (14–16). Due to the short lengths of the target sequences of hemiprobes, MLPA can not only be applied to DNA isolated from fresh-frozen materials, but is also suitable for the more fragmented DNA from paraffin-embedded materials. Depending on the quality of the DNA, 50–200 ng of DNA suffice. The ability to carry out a multiplex copy number assessment on small amounts of paraffin-embedded materials makes MLPA a very attractive method in pathology. In previous studies using whole tissue sections, we obtained very promising results with MLPA for HER2 amplification detection in comparison with immunohistochemistry (IHC) (17), fluorescence in situ hybridization (FISH), and chromogenic in situ hybridization (CISH) (18). However, the dynamic range of MLPA copy number ratios was lower than that with FISH. Furthermore, although results showed that amplification could be detected in cases with a tumor percentage lower than 10%, the sensitivity of MLPA in these cases depend on the degree of amplification, so lower levels of amplifications can be missed in case of a low tumor percentage. Obviously, MLPA is a nonmorphological method that requires proper morphological control of the input materials. In cases with a low percentage of relevant materials, laser microdissection may be necessary (19).

2. Materials 2.1. Tissue Cutting, H&E, and Microdissection

1. Uncoated glass slides (Thermo Scientific, Menzel-Gläser, Germany). 2. Hematoxylin solution (Klinipath, Duiven, The Netherlands). 3. Eosin yellow solution (Klinipath). 4. Xylene.

9  Amplification Testing in Breast Cancer by MLPA

109

Fig. 1. Principle of multiplex ligation-dependent probe amplification (MLPA). MLPA uses a mixture of hemi-probe sets that consist of two oligonucleotides, both having PCR primer sequences (x /y) on the outer ends and a sequence complementary to a part of the target sequence on the inner ends. One of the primers has a spacer (stuffer sequence) of variable length in between the PCR primer sequence and the complementary target sequence. When the complementary target sequences of both hemi-probes hybridize adjacent to each other on the target sequence, they can be ligated to each other, and subsequently amplified using the PCR primer sequences. Because the PCR primers are the same for all hemi-probe sets, they can be amplified in a single PCR, which will provide amplicons of unique and defined lengths (120 – 500  bp) due to the specific stuffer length within each probe set.

5. Pertex-mounting medium (Histolab products AB, Göteborg, Sweden). 6. PALM Liquid Cover Glass N (P.A.L.M. Microlaser Techn., Bernried, Germany): Dilute 2 ml stock resin in 10 ml dilution buffer (1/6) (see Note 1). 7. MembraneSlide 1.0 PEN (Carl Zeiss, Munchen, Germany; 1 mm). 8. 70, 85, and 100% ethanol. 9. PALM Laser Microbeam System (P.A.L.M. AG, Bernried, Germany).

110

C.B. Moelans et al.

2.2. DNA Isolation

1. Proteinase K: 10 mg/ml. Dissolve 1 g proteinase K (Sigma Aldrich Chemie BV) in 100 ml distilled water, and store 1-ml aliquot at −20°C. 2. Lysis buffer: 50  mM Tris–HCl buffer, pH 8.0 with 0.5% Tween 20. Dissolve 0.61 g Tris (Roche, 708976) in 80 ml Milli-Q water, add 20 ml Milli-Q water, and adjust pH to 8.0 with HCl (37%). Add Tween 20 (Riedel de Haën) until final concentration of 0.5%.

2.3. Multiplex Ligation-Dependent Probe Amplification

1. MLPA kit (P004-B1 for HER2, MRC-Holland, Amsterdam, The Netherlands) (see Note 2). 2. TE buffer: Dissolve 1.21 g Tris (10 mM) and 0.37 g sodium– EDTA (1  mM) in 900  ml distilled water, adjust the pH between 7.5 and 8.0 with HCl, and add the other 100  ml distilled water. 3. Labeled size standard (Applied Biosystems, Foster City, CA). 4. Deionized formamide (Amresco, OH) (see Note 3). 5. 10× EDTA buffer (Applied Biosystems). 6. Performance-optimized polymer (POP, Applied Biosystems). 7. Thermocycler and capillary sequencer. 8. Fragment analysis software (Genescan, Applied Biosystems). 9. Coffalyser MLPA analysis software (freeware at http://www. mrc-holland.nl).

3. Methods The preparation of samples, laser microdissection, and the optimum reaction conditions for MLPA is described. A set of guidelines specific for software analysis of MLPA assays is included. 3.1. Preparation of Samples and Laser Microdissection

1. Cut 2–3 mm slides (uncoated) and perform a routine H&E staining for each tissue block to be microdissected. Mark the appropriate area to be microdissected. 2. Mount 8–10  mm thick sections on sequential PALM MembraneSlides (see Note 4). 3. Bake these sections at 56°C for 1 h, deparaffinize in xylene for 10 min, and rehydrate through graded alcohols (100, 85, and 70% for 1 min each) (see Note 5). 4. Stain slides with hematoxylin solution for 10 s, rinse in tap water for 1 min, and dip briefly in eosin solution (see Note 6). 5. Dehydrate in 100% ethanol for 1 min and air dry.

9  Amplification Testing in Breast Cancer by MLPA

111

6. To improve morphology and allow larger tissue areas to be laser pressure-catapulted (20), apply PALM Liquid Cover Glass by aerosol and air dry sections for at least 15 min. 7. A microdissection system with UV laser separates the marked invasive tumor groups from their surrounding tissues. Subsequently, these groups are catapulted by laser pressure catapulting into a cap of a common microfuge tube moistened with a drop of paraffin oil (see Note 7). 3.2. DNA Extraction from LaserMicrodissected Samples

1. Add 50 ml direct lysis buffer to the tube containing the microdissected material (see Note 8). 2. Add 10 ml proteinase K (10 mg/ml), incubate at 56°C for at least 1 h, and then boil (heat inactivation) for 10 min. Cool the lysate immediately on ice (see Note 9). 3. Centrifuge for 2 min at room temperature at 20,800 rcf and carefully pipet the DNA-containing supernatant into a clean tube (see Note 10).

3.3. MLPA Assay

1. Dilute the DNA sample (preferred range 50–200  ng) with TE (10 mM Tris–HCl pH 8.2; 1 mM EDTA) to 5 ml and add to PCR tubes. Take appropriate positive/negative controls along with your samples (see Notes 11 and 12). 2. Heat for 5 min at 98°C and cool to 25°C before opening the thermocycler. 3. Add a mixture of 1.5 ml SALSA Probemix (black cap) + 1.5 ml MLPA buffer (yellow cap) to each tube (see Note 13). 4. Mix with care. Incubate for 1 min at 95°C (denaturation), followed by 16  h at 60°C (hybridization overnight) (see Note 14). 5. Prepare the Ligase-65 mix (less than 1 h before use and stored on ice): Add 3 ml Ligase-65 buffer A (transparent cap) to 3 ml Ligase-65 buffer (white cap) and 25  ml MilliQ water, mix, add 1  ml Ligase-65 thermostable enzyme (green cap), and mix again. 6. Reduce the temperature of the thermocycler to 54°C. While at this temperature, add 32 ml Ligase mix to each sample and mix well. 7. Incubate for 15  min at 54°C, and then heat for 5  min at 98°C. Place the ligation product on ice (see Note 15). 8. Mix in new PCR tubes: 4  ml SALSA PCR buffer (red cap) + 26 ml MilliQ water + 10 ml MLPA ligation reaction. 9. Prepare polymerase mix (less than 1 h before use and stored on ice): Mix 2 ml SALSA PCR primers (brown cap) with 2 ml SALSA enzyme dilution buffer and 5.5  ml MilliQ water. Add 0.5  ml SALSA polymerase (orange cap). Mix well (see Note 16).

112

C.B. Moelans et al.

Table 1 Multiplex ligation-dependent probe amplification (MLPA) PCR conditions Hybridization reaction

98°C 25°C 95°C 60°C

5 min Hold 1 min Hold

Ligation reaction

54°C 54°C 98°C 4°C

Hold 15 min 5 min Hold

PCR

60°C 95°Ca 60°Ca 72°Ca 72°C 4°C

Hold 30 s 30 s 60 s 20 min Hold

35 cycles

a

10. While the tubes are in the thermocycler at 60°C, add 10 ml polymerase mix to each tube and immediately start the PCR. PCR conditions are 35 cycles of 30 s at 95°C, 30 s at 60°C followed by 60 s at 72°C. The PCR ends with 20 min incubation at 72°C (Table 1; see Note 17). 11. Separate the amplification products by a capillary sequencing system with fragment analysis software. In our case (ABI310), we mix 22 ml deionized formamide, 0.75 ml ROX-500, and 3 ml PCR product, denature at 80°C for 5 min, and keep the plate on ice until the start of the run (see Note 18). 3.4. Software Analysis and Interpretation of MLPA Data

1. To analyze MLPA data, we recommend using Coffalyser software developed by the manufacturer of MLPA kits, especially for large numbers of samples (see Note 19). 2. Open Coffalyser, select the probemix you used (for example, P004-B1), and import your raw fragment analysis files (see Notes 20 and 21). 3. Navigate to the “Data filtering” page, where you can filter your data with the automatically generated default bin set that needs to be checked. If necessary, adjust the set bin lengths (that Coffalyser determines based on your selected reference samples) and use these as default. Start filtering the reference and sample data.

9  Amplification Testing in Breast Cancer by MLPA

113

4. Navigate to the “Data analysis” page. A short summary of all filtered data is given, including name, DNA concentration test, number of signals/probes found, number of control probes found, ligation-dependent probes found, and sex determination (M/F). Reference runs should always contain all signals! Delete reference runs that do not contain all signals (see Note 22). 5. Choose an appropriate analysis method, for example “Tumor analysis.” In this case, Coffalyser uses all probes in the selected MLPA kit for slope correction (corrects for signal strength drop with increasing length of the probes), but normalizes the data using only the reference probes present in the kit (located in stable genomic areas) (see Note 23). 6. Analyze and explore results by navigating to “MLPA results.” You will be able to look at your reference results and sample results in different sheets. The gene copy number status is marked with a color. If the MLPA ratio is >1.3 (gain), the color is green. If the ratio is <0.7 (loss), the color is red and in between 0.7 and 1.3, the color is blue. Check that the reproducibility of your reference runs is ok (all probes are marked blue). 7. In the case of HER2, gene copy number was determined by calculating the mean of all three HER2 probe peaks in duplicate (six values). As previously established (21, 22), if this value was below 1.3, the test was scored nonamplified, values 1.3–2.0 were scored as a gain, and values >2.0 as HER2amplified (see Note 24).

4. Notes 1. PALM Liquid Cover Glass, when diluted 1/6, can be kept at 4°C for 2 weeks. 2. In case MRC-Holland cannot offer an MLPA probemix for your application, you can design your own synthetic MLPA probes according to a protocol provided by the manufacturer. Synthetic probes differ from MRC-Holland probes in that the latter consist of one synthetic oligonucleotide and a clonederived one. This allows to make longer probes and to include up to 50 different probes in one MLPA reaction. As an extra control, MLPA kits often contain more than one probe per gene. 3. Currently, most capillary sequencers no longer use deionized formamide. In these cases, formamide is replaced by distilled water.

114

C.B. Moelans et al.

4. To overcome the hydrophobic nature of the membrane, it is advisable to irradiate with UV light at 254 nm for 20–30 min. The membrane gets more hydrophilic, therefore the sections obtain a better adherence. We did not observe any negative effects on MLPA results when using UV irradiation. When cutting the sections for microdissection and subsequent MLPA analysis, it is preferred not to have 0.2% bovine serum albumin (BSA) in the water bath, often used for releasing surface tension, since it can influence DNA quality and therefore MLPA results. 5. Be very careful during xylene and alcohol steps, as the section and membrane can be very fragile. 6. Hematoxylin stain can considerably inhibit PCR amplification (23). The dye may bind to DNA, and subsequently interfere with proteinase digestion, or it might influence divalent cation (Mg++) concentration that is important in maintaining polymerase activity. No deleterious effects have been described for eosin. We, therefore, advise to keep staining times as short as possible, especially for hematoxylin. 7. If DNA isolation is not performed immediately after laser microdissection, microfuge tubes can be kept at 4°C for 2 weeks, perhaps even longer (not tested). 8. Since the microdissected material is in the tip of the tube, it is advisable to add the lysis buffer to the tip and then tick hard so that the material drops to the bottom of the tube together with the lysis buffer. Another option is to use a pipette tip to move the laser-microdissected tissue to the bottom of the tube (or a new tube). We used 50 ml of lysis buffer, but it is also possible to use less or more depending on the amount of tissue you were able to microdissect. 9. If necessary, the incubation time can be increased to an overnight step; in most cases, this leads to a higher DNA concentration. Always use a lysis buffer/proteinase K proportion of 5/1. 10. The extracted DNA is tenable for 2–3 weeks at 4–10°C. This easy, fast, and cheap DNA isolation method leads to lower quality and fragmented DNA, but is sufficient for MLPA analysis due to its very short hybridizing probe sequences. We have tried different column precipitation isolation methods that also work well and generally lead to a higher DNA quality, but at the same time to a lower DNA quantity. 11. The EDTA concentration in the DNA sample should not exceed 1 mM, and the sample volume should not exceed 5 ml. The volume of the reaction is important for the hybridization speed, which is probe- and salt concentration-dependent. If the DNA concentration of a sample is very low, add 5 ml of

9  Amplification Testing in Breast Cancer by MLPA

115

DNA without TE buffer. We usually perform every sample in duplicate as an extra control. 12. Negative (and optional positive) control samples should be run simultaneously with the test samples. It is highly recommended to compare reference and tumor samples extracted by the same method and derived from the same source (e.g., blood versus blood). Use at least three reference samples in each MLPA run (we used four negative control samples per 21 samples). When using more than 21 samples, add one additional reference sample for each seven samples. Reference samples should be spread randomly over the sample plate to avoid bias, and thus minimize variation. It is also recommended to include a no-DNA (5 ml water or TE) reaction in each experiment, as it reveals contamination of water, MLPA reagents, electrophoresis reagents, or capillaries. 13. The SALSA MLPA buffer is usually frozen at −20°C, so thawing is necessary before use. Furthermore, the MLPA buffer is viscous and does not mix easily. Mix probemix and MLPA buffer by repeated pipetting just before use. 14. The hybridization time can be anywhere between 12 and 24 h (16–18 h is recommended, but hybridization should be nearly complete at 12 h). Be sure that there is no excessive evaporation, if so try a different brand of tubes or try using mineral oil on top of the ligation–PCR mix. 15. Following ligation inactivation at 98°C, samples can be stored at 4°C for up to 1 week. For longer periods, storage at −20°C is recommended. 16. Start the PCR as soon as possible after the addition of polymerase mix. 17. All volumes of the PCR can be reduced in order to save reagents. The recommended number of PCR cycles is 35; however, the number of cycles can be reduced to 30 and, in the case of small DNA amounts, the number of cycles can be increased up to 37. The sequence (5¢-3¢) of the labeled forward PCR primer is GGGTTCCCTAAGGGTTGGA and of the unlabeled reverse primer GTGCCAGCAAGATCC AATCTAGA. The PCR product can be stored in the dark at 4°C for at least 1 week. 18. The amount of the PCR product required for analysis by capillary electrophoresis depends on the instrument and fluorescent label used. The settings can be found at the MRCHolland Web site in the support section under technical MLPA protocols. 19. Coffalyser is an excel-based program, which runs on a Microsoft Office version 2003 or higher. Data normalization and correction for probe length-dependent decrease in peak area/height are built-in functions of this program.

116

C.B. Moelans et al.

20. Many more MLPA mixes are commercially available. There are kits for hereditary cancers, tumor characterization, genetic disorders, mental retardation/neurogenetics, pre- and postnatal defects, pharmacogenetics, methylation analysis, and mRNA analysis. 21. Manuals on how to create files for import can be found at the size calling and export manuals on the MRC-Holland Web site. ABIF (FSA) files coming from the ABI-310 and ABI3100 can directly be imported and will be size called during import. TXT files are also supported and can be exported from Genescan, Genemapper, Peak scanner software, LICOR software, Megabace software, and the Spectrophotometrix. After import, the electropherograms from raw runs can be visualized. 22. Coffalyser calculates the ratios of all analyzed sample runs. Several quality check points are also displayed: the number of found probes; the number of found reference probes; whether the ligation control peak (92  bp) was found; whether the sample was male or female (if a Y probe was present); whether there was enough DNA [by estimation of the relative signal of Q-fragments (60, 68, 74, and 80  bp) present in every MLPA mix]; whether the DNA was denatured completely [relative signals of denaturation fragments (88 and 96  bp) present in every kit]; the Pearson product moment correlation (PPMC) value (the correlation value found when slope correction is applied); and the median of absolute deviation (MAD) value (a measure for the stability of reference probes). When the MAD value becomes red, you may need to try a different normalization factor or adjust the reference probes for a more optimized normalization. 23. To best meet the requirements of the context of your experimental design, a number of pre-made settings are available (tumor analysis, direct analysis, control probe analysis, population analysis, methylation status analysis). Depending on expected aberrations, availability of reference probes and reference runs, and quality of the runs, a method needs to be chosen by the user. The manufacturer provides a guideline. 24. Probe signals for DNA sequences can be very high when the DNA in question is amplified at very high levels (e.g., in the case of HER2, EGFR, etc), causing other probe signals to be dwarfed. Probe signals can be reduced by the inclusion of a competitor oligo in the probemix. This competitor is identical to the left probe oligo (LPO) and to a small part (four nucleotides: TGGA) of the PCR primer sequence. The competitor competes with the LPO for the limiting number of binding sites on the DNA. It can be ligated to the right probe

9  Amplification Testing in Breast Cancer by MLPA

117

oligo (RPO), but the resulting comp-RPO ligation product cannot be amplified exponentially as the probe thus formed does not contain both primer sequences. The use of a 1:1 ratio of LPO and its corresponding competitor reduces the probe signal twofold. References 1. Ross JS, Fletcher JA, Bloom KJ et  al (2004) Targeted therapy in breast cancer: the HER-2/ neu gene and protein. Mol Cell Proteomics 3, 379–398. 2. Slamon DJ, Godolphin W, Jones LA et  al (1989) Studies of the HER-2/neu protooncogene in human breast and ovarian cancer. Science 244, 707–712. 3. Baak JPA, Chin D, Van Diest PJ et al (1991) Comparative long term prognostic value of quantitative Her2/Neu protein expression, DNA ploidy, morphometric and clinical features in paraffin-embedded invasive breast cancer. Lab Invest 64, 215–222. 4. Slamon DJ, Leyland-Jones B, Shak S et  al (2001) Use of chemotherapy plus a monoclonal antibody against HER2 for metastatic breast cancer that overexpresses HER2. N Engl J Med 344, 783–792. 5. Hudis CA (2007) Trastuzumab, mechanism of action and use in clinical practice. N Engl J Med 357, 39–51. 6. Borg A, Baldetorp B, Ferno M et al (1994) ERBB2 amplification is associated with tamoxifen resistance in steroid-receptor positive breast cancer. Cancer Lett 81, 137–144. 7. Tetu B, Brisson J, Plante V et  al (1998) p53 and c-erbB-2 as markers of resistance to adjuvant chemotherapy in breast cancer. Mod Pathol 11, 823–830. 8. Schouten JP, McElgunn CJ, Waaijer R et  al (2002) Relative quantification of 40 nucleic acid sequences by multiplex ligation-dependent probe amplification. Nucleic Acids Res 30, e57. 9. Moelans CB, de Weger RA, van Blokland MT et al (2010) Simultaneous detection of TOP2A and HER2 gene amplification by multiplex ligation-dependent probe amplification in breast cancer. Mod Pathol 23, 62–70. 10. Moelans CB, de Weger RA, and van Diest PJ (2010) Absence of chromosome 17 polysomy in breast cancer: analysis by CEP17 chromogenic in situ hybridization and multiplex ligationdependent probe amplification. Breast Cancer Res Treat 120, 1–7.

11. Vorstman JA, Jalali GR, Rappaport EF et  al (2006) MLPA: a rapid, reliable, and sensitive method for detection and analysis of abnormalities of 22q. Hum Mutat 27, 814–821. 12. Eldering E, Spek CA, Aberson HL et al (2003) Expression profiling via novel multiplex assay allows rapid assessment of gene regulation in defined signalling pathways. Nucleic Acids Res 31, e153. 13. Hess CJ, Denkers F, Ossenkoppele GJ et  al (2004) Gene expression profiling of minimal residual disease in acute myeloid leukaemia by novel multiplex-PCR-based method. Leukemia 18, 1981–1988. 14. Dikow N, Nygren AO, Schouten JP et  al (2007) Quantification of the methylation status of the PWS/AS imprinted region: comparison of two approaches based on bisulfite sequencing and methylation-sensitive MLPA. Mol Cell Probes 21, 208–215. 15. Nygren AO, Ameziane N, Duarte HM et  al (2005) Methylation-specific MLPA (MS-MLPA): simultaneous detection of CpG methylation and copy number changes of up to 40 sequences. Nucleic Acids Res 33, e128. 16. Procter M, Chou LS, Tang W et  al (2006) Molecular diagnosis of Prader-Willi and Angelman syndromes by methylation-specific melting analysis and methylation-specific multiplex ligation-dependent probe amplification. Clin Chem 52, 1276–1283. 17. Purnomosari D, Aryandono T, Setiaji K et  al (2006) Comparison of multiplex ligation dependent probe amplification to immunohistochemistry for assessing HER-2/neu amplification in invasive breast cancer. Biotech Histochem 81, 79–85. 18. Moelans CB, de Weger RA, van Blokland MT et al (2009) HER-2/neu amplification testing in breast cancer by multiplex ligation-dependent probe amplification in comparison with immunohistochemistry and in situ hybridization. Cell Oncol 31, 1–10. 19. Moelans CB, de Weger RA, Ezendam C et al (2009) HER-2/neu amplification testing in breast cancer by Multiplex Ligation-dependent

118

C.B. Moelans et al.

Probe Amplification: influence of manual- and laser microdissection. BMC Cancer 9, 4. 20. Micke P, Bjornsen T, Scheidl S et al (2004) A fluid cover medium provides superior morphology and preserves RNA integrity in tissue sections for laser microdissection and pressure catapulting. J Pathol 202, 130–138. 21. Coffa J, van de Wiel MA, Diosdado B et  al (2008) MLPAnalyzer: data analysis tool for

reliable automated normalization of MLPA fragment data. Cell Oncol 30, 323–335. 22. Bunyan DJ, Eccles DM, Sillibourne J et  al (2004) Dosage analysis of cancer predisposition genes by multiplex ligation-dependent probe amplification. Br J Cancer 91, 1155–1159. 23. Murase T, Inagaki H, and Eimoto T (2000) Influence of histochemical and immunohistochemical stains on polymerase chain reaction. Mod Pathol 13, 147–151.

Chapter 10 Detection and Quantification of MicroRNAs in Laser-Microdissected Formalin-Fixed Paraffin-Embedded Breast Cancer Tissues Sarkawt M. Khoshnaw, Des G. Powe, Ian O. Ellis, and Andrew R. Green Abstract MicroRNAs (miRNAs) are a class of small endogenous non-coding RNAs that regulate gene expression post-transcriptionally through targeting protein-coding mRNAs for cleavage or translational repression, and thus play key roles in cellular fate-determinant pathways. Both profiling and functional studies demonstrated derangement of miRNA repertoire in many human cancers, including breast tumours. Discovery of miRNAs provided new insights into cancer pathogenesis and led the scientific community to approach novel diagnostic and therapeutic strategies in cancer management. Research in this field is increasing, and the potential for miRNAs being used in clinical settings emphasises the need for high-throughput and sensitive detection techniques. In this chapter, techniques for the analysis of miRNA expression in lasermicrodissected formalin-fixed paraffin-embedded breast cancer tissues are discussed. Key words: Breast cancer, MicroRNA, Laser capture microdissection, Formalin-fixed paraffinembedded, FFPE

1. Introduction 1.1. MicroRNAs’ Biogenesis and Functions

MicroRNAs (miRNAs) are a recently described class of nonprotein-coding RNA molecules (18–25 nucleotides (nt) in length) involved in critical cellular regulatory pathways (1, 2). Mature miRNAs are produced in the nucleus as primary (pri-) miRNAs transcribed by RNA polymerase II (3). Pri-miRNAs, which range between several hundred to a thousand nucleotides in length, are processed inside the nucleus to shorter (70–85  nt) precursor (pre-) miRNAs mediated by RNase III enzyme complex Drosha/ DGCR8 (4). Pre-miRNAs are exported to the cytoplasm (5) and

Graeme I. Murray (ed.), Laser Capture Microdissection: Methods and Protocols, Methods in Molecular Biology, vol. 755, DOI 10.1007/978-1-61779-163-5_10, © Springer Science+Business Media, LLC 2011

119

120

S.M. Khoshnaw et al.

cleaved by Dicer, a second RNase III enzyme, to produce a ~22 nt temporary miRNA duplex made up of a mature miRNA sequence and its complementary sequence (6). The strand with less stable hydrogen bonding at its 5¢ end is incorporated into the RNAinduced silencing complex (RISC) to form the mature miRNA, and the other strand is degraded (7). The main subunit of RISC is the Argonaute2 (Ago2) protein, which is the catalytic endonuclease of human RISC (8). It is thought that miRNA genes constitute more than 1% of human genome, (9) and approximately one third of human genes may be targets of miRNAs (10). Down-regulation of gene expression by miRNAs involves both mRNA degradation and translational repression which are complex processes occurring through multiple mechanisms, such as (1) miRNA (one of the components of RISC complex) can combine with the 3¢UTR of the mRNA, which requires imperfect complementarity only, resulting in translational repression (11). (2) miRNAs may also bind to the open reading frame (ORF) of target mRNAs, requiring perfect or near-perfect complementarity, leading to cleavage and degradation of target mRNAs by the action of Ago2 (12). miRNAs have key regulatory functions in various cellular processes, such as proliferation, differentiation, and apoptosis (2). They play significant roles in the regulation of metabolism, development, cell cycle, carcinogenesis, cancer progression, and cancer metastasis (13, 14), and their expression is under tight control both spatially and temporally. Hence, altered expression of miRNAs is coupled with abnormal cellular behaviour, causing a range of disorders including human cancers. Understanding overall gene expression regulation has become more challenging since the discovery of miRNAs. It is estimated that a normal miRNA may have between 100 and 200 targets (15). Depending on the functional activity of the protein products of their target mRNAs, miRNAs can play oncogenic or tumour suppressor roles (16, 17). Tumour suppressor miRNAs negatively inhibit oncogenes and are down-regulated in cancer (16) while oncomiRs represent oncogenic miRNAs which are up-regulated in cancer cells, and have a role in the process of carcinogenesis through down-regulation of tumour suppressor molecules and/or molecules which are important for cellular differentiation (17). 1.2. Involvement of miRNAs in Breast Cancer and Their Potential Future Applications in Clinical Settings

Many studies have demonstrated the deregulation of miRNA profiling in breast tumours (18). miRNA expression profiles have been shown to be apparently aberrant in breast tumours compared with normal breast tissues. Iorio et al. found the expression level of miR-10b, miR-125b, miR-145, miR-21, and miR-155 to be considerably deregulated in breast cancer tissues, with miR-10b, miR-125b, and miR-145 being down-regulated and miR-21 and miR-155 up-regulated (18). miRNA expression patterns could

10  Detection and Quantification of MicroRNAs in Laser-Microdissected…

121

be used to classify human breast tumours, predict prognosis, and distinguish cancer tissues from adjacent normal tissues (18). Interestingly, breast cancer biological subtypes are shown to have distinctive miRNA expression signatures (19). This intimate involvement of miRNAs in malignant transformation, cellular invasion, and metastases in breast cancer makes them significant targets for developing a modern breast cancer molecular classification and opening avenues for more tailored treatment strategies for breast cancer patients. Dissecting the complex molecular network linking miRNAs to well-known oncogenes and tumour suppressor genes (such as H-RAS, HMGA2, LIN28, PEBP1, mucin 1, SOX4, PTPRN2, MERTK, TNC, HER2, HER3, IRS-1, P85 Beta, Raf1, ERK5, BMI1, ZEB1, ZEB2, Beta-catanin, AKT1, HMGA2, P53, RHOA, BCL-2, TPM1, PDCD4, PTEN, MASPIN, HOXD10, MCL1) has been an area of intense research during the last few years in an attempt to provide molecular explanation of how miRNAs regulate gene expression and induce cancer formation (see Note 3). Studies performed on human breast cancer tissues suggest that miRNAs could serve as future breast cancer biomarkers for earlier detection and more accurate diagnosis and prognostication. Moreover, they could unravel the molecular mechanisms of cancer pathogenesis and be targeted using a novel therapeutic approach involving synthetic oligonucleotide technologies. A comprehensive study of miRNA expression profiling in the currently defined breast cancer subclasses (20) has the potential to generate a more accurate molecular classification of breast cancer, which could guide clinicians to a more precise prognostication and treatment plan for each individual patient (see Note 4). 1.3. miRNA Extraction from Sections of Frozen and FormalinFixed ParaffinEmbedded Tissues

Compared with fresh-frozen tissue samples, formalin-fixed paraffin-embedded (FFPE) materials are more readily available and archives represent a major resource for the study of clinical samples with possible long-term follow-up data. However, mRNA from FFPE tissue is subject to damage, degradation, and alteration during fixation and processing, and is thus of limited value for gene expression analysis (21). RNA undergoes chemical modification by formalin and additional degradation during storage which is thought to be due to the occurrence of methylol cross links between RNA and protein during tissue processing (22). Nonetheless, it has been demonstrated that miRNA levels detected in total RNA extracted from FFPE are higher than those extracted from frozen cells when the amount of total RNA is identical (23). This might be attributed to the fact that generation of equivalent amounts of total RNA entails the inclusion of larger numbers of FFPE cells than fresh-frozen tissues, which could be due to the presence of residual cross links in RNA molecules which are not eliminated by proteinase K digestion,

122

S.M. Khoshnaw et al.

and hence failure of these RNA molecules being extracted (23). The possibility of existence of cross links is higher in longer RNA molecules; therefore, small RNAs are extracted more readily because they are less affected by this phenomenon (23). 1.4. Laser Capture Microdissection

Tissues are 3D structures comprising an assortment of cell subpopulations with different molecular and functional signatures. Laser capture microdissection (LCM) is an efficient and effective method for the separation of morphologically and/or phenotypically defined cell groups from complex tissues to facilitate downstream molecular analyses. LCM enables precise isolation of a particular cell subpopulation without contamination from surrounding cells; hence, it can be used to harvest cells of interest or remove unwanted cells under direct microscopic visualisation from a complex tissue sample. An application of LCM involves comparison of gene expression in normal tissues against tissues representing different stages of cancer progression. Principles, technical aspects, and applications of LCM were reviewed by Fend and Raffeld (24) and Espina et al. (25).

1.5. Aim of This Study

The aim of this work is to develop and optimise the methodology for miRNA expression analysis from microdissected FFPE breast cancer tissues.

2. Materials 2.1. Laser Microdissection

1. Positioning and ablation with Laser Microbeams (PALM) non-contact Laser catapulsion instrument (P.A.L.M. Microlaser Technologies), Carl Zeiss Ltd (Welwyn Garden City, Hertfordshire, UK). 2. PALM MembraneSlide 1.0 PEN (P.A.L.M. Microlaser Technologies GmbH), Carl Zeiss Ltd. 3. AdhesiveCap 500 opaque, Carl Zeiss Ltd. 4. 0.025% RNase-free aqueous Toluidine blue (HD Supplies, Botolph Claydon, Buckinghamshire, UK).

2.2. Total RNA Extraction and Quantification

1. miRNeasy FFPE Kit (50), QIAGEN, Inc. (Crawley, UK). 2. RecoverAll Total RNA Isol Kit FFPE 40Rxn, Applied Biosystems UK (Warrington, UK). 3. TRIzol reagent, Invitrogen, supplied by Fisher Scientific UK Ltd. (Loughborough, Leicestershire, UK). 4. Xylene, Fisher AR, 2.5 l, Fisher Scientific UK Ltd. 5. Ethanol absolute, Sigma-Aldrich (Gillingham, Dorset, UK). 6. Proteinase K solution RNA, Fisher Scientific.

10  Detection and Quantification of MicroRNAs in Laser-Microdissected…

123

7. Proteinase K digestion buffer, a component in the RecoverAll Total RNA Isol Kit FFPE 40Rxn, Applied Biosystems UK. 8. Chloroform, contains amylenes as stabiliser, ³99% (SigmaAldrich). 9. 2-Propanol, BioReagent, for molecular biology, ³99%, SigmaAldrich. 10. Glycogen UltraPure, Fisher Scientific. 11. Water, molecular biology reagent, Sigma-Aldrich. 12. Pipettor tip Microman capillary pistons, pure translucent polypropylene 1–10 ml Gilson, Fisher Scientific. 13. DNase I Amp grade Invitrogen GIBCO, 100  units, Fisher Scientific. 14. Quant-iT RiboGreen RNA Quantitation Kit, Fisher Scientific UK Ltd. (a) Quant-iT™ RiboGreen® RNA, Reagent (Component A) (b) 20× TE buffer, RNase-free (Component B) (c) Ribosomal RNA standard, 16S and 23S rRNA from, Escherichia coli (Component C) 2.3. miRNA Detection by TaqMan Real-Time PCR

1. TaqMan® MicroRNA Assays, hsa-miR-21, Applied Biosystems UK. 2. TaqMan® MicroRNA Biosystems UK.

Assays,

hsa-miR-29c,

Applied

3. TaqMan® MicroRNA Biosystems UK.

Assays,

hsa-miR-127,

Applied

4. TaqMan® Universal PCR Master Mix, No AmpErase® UNG, 1-Pack (1 × 5 ml). 5. MultiScribe™ Reverse Transcriptase, Applied Biosystems UK. 6. TaqMan “Universal PCR Master Mix, No AmpErase” UNG, 1-Pack (1 × 5 ml). Applied Biosystems UK. 7. TaqMan MicroRNA RT Kit, 200 Rxn, Applied Biosystems UK. 8. Mx3005P™ QPCR System (Agilent Technologies, South Queensferry, Scotland).

3. Methods 3.1. Samples and Cases

1. In surgically resected FFPE primary breast tumour specimens from three female breast cancer patients, “optimisation cases” were collected retrospectively with appropriate ethical approval

124

S.M. Khoshnaw et al.

Table 1 Three cases form archives of the Histopathology department of Nottingham City Hospital for methodology optimization No

Age (years)

Tumour size (cm)

Tumour stage

Tumour grade

Histological type

ER

HER2

1

46

3.6

3

3

NST

Positive

Positive

2

32

2.4

3

3

NST

Positive

Negative

3

63

5.8

3

3

NST

Negative

Negative

NST invasive non-special type

Fig. 1. Invasive breast cancer tissue.

from the archival stores of Nottingham Breast Cancer Series (Nottingham City Hospital Tumour Bank, Nottingham, UK; Table 1). 2. The average age of FFPE samples was 14 years (12–16-yearold samples), where patients were operated upon in 1993, 1995, and 1997. Selection of these cases was random, with the proviso that tissue samples from each case contained large amounts of invasive breast cancer cells (Fig. 1). H&E slides were produced from the FFPE blocks, and the presence of invasive tissue in each case was confirmed by Dr Z Hodi, consultant pathologist at Nottingham City Hospital. 3.2. Methodology Optimisation

Since the amount of total RNA in microdissected FFPE tissues is tiny, extraction of an amount of RNA sufficient for reliable miRNA detection requires investigation and optimisation of three vital aspects of methodology:

10  Detection and Quantification of MicroRNAs in Laser-Microdissected…

125

1. Thickness of FFPE samples mounted on PALM membrane slides to undergo LCM: 10 and 20 mm thick FFPE sections were investigated (see Note 1). 2. Amount of tissue to be microdissected: After mounting 10 and 20 mm thick sections on PALM membrane slides, a range of different microdissected areas (1 × 103–1 × 107  mm2) were investigated. Tissue macrodissection was also performed in which 5 × 107 mm2 was dissected with a sterile needle from tissue sections (see Note 1). 3. Choosing a total RNA extraction kit among three commercially available kits: Three different kits (Qiagen, RecoverAll, and TRIzol) were used to extract total RNA from full-face and microdissected FFPE tissues (see Note 2). To answer each of the above three questions, the outcomes of two core aspects of the procedure were taken into consideration: 1. Quantity of total RNA extract. For downstream analysis of miRNAs, 100 ng total RNA input is required for application on the human microRNA Microarrays (Agilent Technologies), (Publication number: 5990-4944EN), which contain probes for 723 humans and 76 human viral miRNAs from the Sanger database v.10.1. This technique was developed by Wang et al. (26). The labelling, probe design, and hybridization procedure used in this technique are simple and sensitive. The dephosphorylation and labelling of total RNA in the same sample tube are easy, and very low RNA input is needed. For miRNA profiling, 100 ng of tissue total RNA is dephosphorylated for 30  min at 37°C. This miRNA profiling assay directly utilises total RNA, and thus minimises the predictable sample losses occurring in more complicated procedures which require RNA size fractionation or many purification steps. In an ideal quantitative assay, each step should proceed in reproducibly high yield and be unaffected by minute variations from the standard procedure. These objectives are made possible when no amplification or separation steps are applied that can bring in sample-dependent variations and if both labelling and hybridization attain stable end points close to equilibrium, negligibly reliant on reaction kinetics and concentrations. Labelling reaction in this technique adds precisely one fluorophore to each miRNA in reproducibly high yield under circumstances that are insensitive to tiny differences in pCp-dye or miRNA concentration. Hybridization is permitted to progress towards equilibrium, and the probe melting temperature (Tm) matching ensures that nearly all miRNAs are hybridised at equilibrium. The number of hybridised labelled targets can be accurately measured via a calibrated scanner (26).

126

S.M. Khoshnaw et al.

2. miRNA detection sensitivity limit. In order to determine limits of sensitivity of miRNA detection in laser-microdissected tissues, the methodology was optimised via testing the techniques’ sensitivity to detect different miRNAs with varying expression levels in breast cancer. Therefore, three miRNAs which are differentially expressed in invasive breast tumours (high, intermediate, and low expression) were selected for methodology optimisation. It was previously shown that among 157 miRNAs, miR-21 was the most abundantly expressed miRNA in a total of five pairs of matched advanced breast tumour tissue specimens, and the level of miR-21 was much higher in the tumour tissues than in matched normal tissues (27). Recently, Yan et al. performed an extensive miRNA expression profiling study on eight primary human breast cancer tissues along with the adjacent normal tissues. They used an miRNA microarray containing 435 mature human miRNA oligonucleotide probes, and observed that 9 miRNAs (miR-21, miR-365, miR-181b, let-7f, miR-155, miR-29b, miR-181d, miR-98, and miR-29c) were up-regulated more than twofold in breast cancer compared with normal adjacent tissues while seven miRNAs (miR-497, miR-31, miR-355, miR-320, rno-mir-140, miR-127, and miR-30a-3p) were down-regulated more than twofold (28). The median of original signal of miR-21, miR-29c, and miR-127 were 10,768.5, 5,126.0, and 1,377.0, respectively. For methodology optimisation, specific probes for miR-21 as highly expressed, miR-29c as intermediately expressed, and miR-127 as an miRNA with low expression were utilised. 3.3. Sample Preparation for RNA Extraction

1. Full-face (gross) sections: 10 mm thick full-face FFPE sections were cut from the three optimisation cases. This was followed by immediate RNA extraction. 2. Microdissected tissues: (a) Mounting FFPE sections on PALM membrane slides and preparing them for LCM: As paraffin sections adhere more readily to hydrophilic surfaces, PALM membrane slides were irradiated with UV light at 254 nm for 30 min to overcome the hydrophobic nature of the membrane. Moreover, UV treatment sterilises the membrane and destroys potentially contaminating nucleic acids. The slides were used immediately after the UV treatment. 10 and 20 mm thick sections were cut using an RNase-free microtome from case three and mounted on RNase-free PALM membrane slides. For each block, the topmost few sections were not used. Sections were deparaffinised with xylene, rehydrated in RNase-free 100% ethanol, rinsed in RNase-free water and stained with 0.025% RNase-free aqueous toluidine blue.

10  Detection and Quantification of MicroRNAs in Laser-Microdissected…

127

(b) LCM: PALM non-contact Laser catapulsion instrument was used according to the manufacturer’s instructions to microdissect ten different cellular areas ranging from 1 × 103 to 1 × 107 mm2 from breast tissue sections (Fig. 2).

Fig. 2. Invasive breast cancer tissues pre- and post-laser capture microdessection (LCM) from one sample (Master Index number: 4,063). a1, b1, c1, and d1 are Pre-LCM and a2, b2, c2, and d2 are Post-LCM.

128

S.M. Khoshnaw et al.

The microdissectates were collected in PALM adhesive caps. A total of 45.5 ml proteinase K and 155.5 ml proteinase K digestion buffer were added to each sample and incubated at 55°C overnight (29) prior to RNA extraction. 3. Macrodissected tissues (needle macrodissection): Macro­ dissection of invasive breast cancer tissues, as judged by a serial H&E section, was performed with a sterile needle and placed into sterile microfuge tubes. The area macrodissected was approximately 5 × 107 mm2. 3.4. Total RNA Extraction

Three different commercial RNA extraction kits were used according to manufacturer’s instructions. 1. miRNeasy FFPE Kit, Qiagen: (a) For gross sections, four 10 mm thick sections were deparaffinised in xylene and washed by 100% ethanol. The tissues/microdissectates were resuspended in 150 ml buffer PKD, and 10 ml proteinase K was added. Samples were incubated at 55°C for 15  min, and then at 80°C for 15 min. Subsequently, 320 ml buffer RBC was added to each sample. (b) The lysate was transferred to a gDNA Eliminator spin column and centrifuged for 30 s at ³8,000 × g. A total of 1,120 ml ethanol was added. The sample was transferred to an RNeasy MinElute® spin column (700 ml at a time) and centrifuged for 15 s at ³8,000 × g. (c) This was repeated until the entire sample had passed through the spin column. 500 ml buffer RPE was added to the RNeasy MinElute spin column and centrifuged for 15 s at ³8,000 × g. 500 ml buffer RPE was again added to the RNeasy MinElute spin column and centrifuged for 2 min at ³8,000 × g. (d) The spin columns were centrifuged at full speed for 5 min with their lids open. 20 ml RNase-free water was added directly to the spin column membrane and centrifuged for 1 min at full speed to elute the RNA. 2. RecoverAll Total RNA Isol Kit FFPE: (a) For gross sections, three 20  mm thick sections were deparaffinised in xylene and washed by 100% ethanol. 400 ml digestion buffer and 4 ml protease were added to each tissue/microdissectate sample and incubated for 3 h at 50°C. 480 ml Isolation Additive was added followed by 1.1 ml 100% ethanol. 700 ml of the sample/ethanol mixture was pipetted onto a Filter Cartridge and centrifuged at 10,000 × g for 30–60 s.

10  Detection and Quantification of MicroRNAs in Laser-Microdissected…

129

(b) Subsequently, 700 ml of Wash 1 was added to the Filter Cartridge and centrifuged for 30  s at 10,000 × g. This was followed by adding 500 ml of Wash 2/3 to the Filter Cartridge and centrifuging for 30 s at 10,000 × g. 60 ml DNase mix was added to each sample and incubated for 30 min at room temperature, and then 700 ml of Wash 1 was added to the Filter Cartridge. The Filter Cartridge was washed twice with 500 ml of Wash 2/3 and centrifuged for 1 min at 10,000 × g. RNA was eluted with 30 ml nuclease-free water and centrifuged for 1  min at maximum speed. The elution step was repeated, and the volume of collected eluate (which contained the RNA) was close to 60 ml. 3. TRIzol reagent: (a) For gross sections, three 10  mm thick sections were deparaffinised in xylene by incubation at 65°C for a total of 20 min substituting xylene twice. Tissue samples were washed twice by 100% ethanol and incubated in 45.5 ml proteinase K and 155.5 ml proteinase K digestion buffer at 55°C overnight (29). 1 ml Trizol was added to each sample, and the samples were vortexed. 200 ml of chloroform was added to each sample. Samples were shaken vigorously for 15  s and left at room temperature for 1 min to settle, and then spinned at 18,894 × g for 15 min at 4°C. (b) The aqueous layer was carefully taken without touching the interphase. This was added to 650  ml isopropanol with 0.5  ml RNase-free glycogen. Samples were left at room temperature for 30  min and centrifuged at 12,000 × g for 15 min at 4°C. (c) The medium was taken, and the pellet was washed twice by adding 1 ml of 75% ethanol and spinning at 12,000 × g for 15 min at 4°C. (d) Finally, ethanol was removed and RNA was dissolved in 12 ml of nuclease-free water. For microdissectates samples, 10 and 20  mm thick sections were mounted on PALM membrane slides. Sections were dried at room temperature overnight, deparaffinised with xylene (5 min twice) and 100% ethanol (5  min twice), rinsed in RNase-free water, and then stained with 0.025% aqueous Toluidine blue. Trizol, chloroform, and isopropanol were used in different volumes as compared to full-face sections, as follows: Trizol – 200 ml, chloroform – 40 ml, and isopropanol – 300 ml.

130

S.M. Khoshnaw et al.

3.5. DNase Treatment of RNA Extracts

3.6. Total RNA Quantification

DNase treatment is one of the steps in Qiagen and RecoverAll RNA extraction kit protocols, while this is not included in the TRIzol reagent procedure. Therefore, the quantification of total RNA was repeated after performing DNase treatment for the RNA extracted from the three optimisation cases via TRIzol reagent. Briefly, 6 ml total RNA was used and increased to 10 ml by adding 1 ml DNase I Amplification Grade (1 U/ml), 1 ml 10× DNase I buffer, and 2  ml DEPC-treated water. This was gently vortexed and spinned briefly, and incubated at room temperature for 15 min. After a brief centrifuge, 1 ml of 25 mM EDTA was added to each sample on ice. Gentle mixing and a brief spinning were performed, and samples were incubated at 65°C for 10 min. 1. Total RNA in microdissected tissues is very small; hence, a sensitive technique is required to accurately quantify RNA in these samples. NanoDrop® ND-1000 UV–Vis Spectro­ photometer instrument was initially used to measure the quantity and quality of extracted total RNA. Detection limit for this method is 2–3,000 ng/ml for RNA, but when used on the microdissected tissue samples, results were not reproducible and the method was deemed inaccurate. 2. Therefore, RiboGreen reagent was used according to manufacturer’s instructions to accurately quantify total RNA in microdissected tissues. Low- and high-range standard curves were initially used in this technique. Detection limit of lowrange standard curve is 0.001–0.05 ng/ml and of high-range standard curve is 0.02–1 ng/ml. Low- and high-range standard curves were produced to match the value of RNA in the samples (Fig. 3). TE buffer (10 mM Tris–HCl, 1 mM EDTA, pH 7.5) was used to dilute Quant-iT™ RiboGreen® reagent 200-fold for the high-range assay. RNA extracted from gross tissues was diluted 1:1,500 and that from microdissected tissues 1:25 to remain within detection limits of the highrange standard curve.

Fig. 3. High- and low-range standard curves for RiboGreen reagent produced in our laboratory.

10  Detection and Quantification of MicroRNAs in Laser-Microdissected…

131

3. To produce a high-range standard curve, 16S and 23S ribosomal RNAs were used to produce a 2 ng/ml RNA stock solution in TE. Using serial dilutions, three other RNA ­concentrations (1, 0.2 and 0.04  ng/ml) were prepared to ­produce the curve. 10 ml of the diluted RiboGreen reagent was added to 10 ml of the diluted RNA samples and 10 ml of each of the four ribosomal RNA concentrations (2, 1, 0.2 and 0.04 ng/ml). 4. Fluorescence signal was measured in samples using Mx3005P™ QPCR System (Agilent Technologies). 3.7. miRNA Detection by TaqMan Real-Time PCR

1. Three mature miRNAs (miR-21, miR-29c, and miR-127) were quantified using TaqMan® MicroRNA Assays, according to the manufacturer’s instructions. 2. TaqMan miRNA assays use the stem-loop method to detect the expression level of mature miRNAs (30, 31). Total RNA extract was transcribed into cDNA in a 15  ml volume reaction. All PCRs were performed in a final volume of 20 ml and were performed in duplicate for each cDNA sample. 3. The number of PCR cycles needed to reach the threshold cycle (Ct) was identified in duplicate for each cDNA and an average taken. Briefly, 10 ng total RNA (in 5 ml) was mixed with 3  ml RT primer and 7  ml master mix per each reverse transcription (RT) reaction (15 ml) which was carried out as follows: 16°C for 30 min; 42°C for 30 min; 85°C for 5 min; and then held at 4°C. Amplification of each target miRNA was performed in duplicate using a 1.33  ml aliquot of each first-strand cDNA reaction in a 20 ml total reaction volume along with 1 ml TaqMan miRNA assay (20×), 10 ml TaqMan 2× Universal PCR Master Mix-No AmpErase UNGa, and 7.67  ml nuclease-free water. An miRNA amplification was performed using Mx3005P® QPCR System (Agilent Technologies) following manufacturer’s instructions with the following incubation times: 95°C for 10 min followed by 40 cycles at 95°C for 15 s and 60°C for 60 s. 4. The real-time PCR results were analysed using MxPro™ QPCR Software. Two negative controls were run in duplicate for each sample, in which RNase-free water was substituted for cDNA which showed no amplification.

3.8. miRNA Profiling from Full-Face (Gross) FFPE Sections

1. Total RNA was extracted from full-face FFPE sections of the three optimisation cases using three RNA extraction kits (Qiagen, RecoverAll, and TRIzol). 2. Total RNA was quantified using RiboGreen reagent, a highly sensitive RNA detection method. Total RNA was highest using TRIzol reagent (608.38  ng/1  mm tissue thickness/

132

S.M. Khoshnaw et al.

Table 2 RiboGreen (RG) reagent was used to measure the quantity of total RNA extracted from the three optimization cases RNA (ng)/1 mm thick section/ whole volume Ct of elutea (miR-21)

Cases

Total section thickness (mm)

Qiagen (DT)

Case 1 Case 2 Case 3

40 40 40

186.30

18.41 18.73 19.47

24.53 25.09 25.34

26.22 27.36 27.56

Ambion (DT)

Case 1 Case 2 Case 3

60 60 60

292.52

20.07 19.76 20.73

26.59 26.14 27.18

28.21 28.24 28.67

Invitrogen (no DT)

Case 1 Case 2 Case 3

30 30 30

608.38

19.79 18.815 20.73

26.14 25.51 26.98

27.98 27.96 28.63

Invitrogen (with DT)

Case 1 Case 2 Case 3

30 30 30

91.86

19.48 18.52 19.32

25.14 24.9 24.96

27.63 26.96 27.65

RNA extraction kit

Ct Ct (miR-29c) (miR-127)

Mature miR-21, miR-29c, and miR-127 were quantified in the total RNA using TaqMan® MicroRNA Assays Ct threshold cycle value, DT DNase treatment a 18 ml was used to elute RNA in Qiagen kit, 60 ml in Ambion kit, and 12 ml in Invitrogen kit

whole elute), and lowest using the Qiagen kit (186.30 ng/1 mm tissue thickness/whole elute; Table 2). 3.9. miRNA Detection and Quantification

1. RNA samples were analysed in duplicate to quantify miR-21, miR-29c, and miR-127 using RT-PCR (Table  2). All the samples showed earliest amplification for miR-21 and latest amplification for miR-127. 2. The Ct values for miR-21, miR-29c, and miR-127 were comparable in RNAs extracted by Qiagen and TRIzol kits while Ct values in the RNA extracted with RecoverAll kit were slightly higher (Table 2). 3. DNase treatment reduced the quantity of RNA by about sevenfold; however, the three miRNAs were amplified in DNasetreated samples at almost the same Ct values compared to non-DNase-treated samples (Table 2). 4. To confirm reproducibility of our results, RNA extraction from gross tissues of the three optimisation cases and quantification of the three miRNAs from the RNA extracts were repeated, and the results were comparable (Table 2, Fig. 4).

10  Detection and Quantification of MicroRNAs in Laser-Microdissected…

133

Fig.  4. miRNA quantification from RNA extracted from full-face sections of the three optimisation cases. RNA was extracted using three different kits. Qiagen and Invitrogen showed comparable results while Ambion showed slightly later miRNA amplification.

134

S.M. Khoshnaw et al.

3.10. miRNA Profiling from LaserMicrodissected FFPE Tissues

For laser-microdissected tissues, RecoverAll RNA extraction kit was not investigated because RNA extracts by this kit from fullface sections showed later amplification of miRNAs compared to Qiagen and TRIzol kits (see Note 2).

3.11. Total RNA Extraction and Quantification

Ten different areas (1 × 103, 5 × 103, 5 × 104, 2 × 105, 5 × 105, 1 × 106, 3 × 106, 5 × 106, 7 × 106, and 1 × 107 mm2) were microdissected from case three of the optimisation cases and RNA extracted using two alternative commercially available kits (Qiagen and TRIzol). Qiagen kit was used to extract total RNA from 10 mm thick FFE sections of case three of the optimisation cases, where the RNA quantity in microdissectates was very low, <1 ng/ml (20 ng/ whole elute, 20 ml) in all the samples (Fig. 5). TRIzol reagent: 1. Using TRIzol reagent, total RNA was extracted with and without DNase treatment. 2. Total RNA extracted from microdissected tissues was higher using TRIzol reagent compared to Qiagen kit. 3. In DNase-treated samples, total RNA quantity extracted was approximately 8 ng/ml (90 ng/whole elute after DNase treatment, 11 ml) in the largest area (1 × 107 mm2: Table 3, Fig. 6). 4. DNase treatment did not significantly reduce total RNA extract in most samples. 5. In non-DNase-treated samples from 10  mm thick sections, the amount of RNA was slightly higher in 7 × 106 and 5 × 106  mm2 areas and was indeed lower in 1 × 107  mm2 area compared to DNase-treated samples (Table 3, Fig. 6).

Fig.  5. Total RNA in laser-microdissected samples extracted by Qiagen kit (ng/whole elute of RNA, 20 ml), as quantified by NanoDrop instrument vs. RiboGreen reagent.

10  Detection and Quantification of MicroRNAs in Laser-Microdissected…

135

Table 3 RNA quantification from different thicknesses and different areas of laser-microdissected tissues using RiboGreen reagent

Laser-micro dissected area (mm2)

Qiagen kit (DT) 10 mm sections RNA (ng/the whole elute, 20 ml)

Invitrogen kit (DT) 10 mm sections RNA (ng/the whole elute, 11 ml)

50 million

Invitrogen kit (no DT) 10 mm sections RNA (ng/the whole elute, 12 ml)

Invitrogen kit (no DT) 20 mm sections RNA (ng/the whole elute, 12 ml)

722.61

10 million

18.68

89.15

64.24

574.41

7 million

7.84

69.67

76.98

309.20

5 million

3.57

48.99

55.66

328.49

3 million

12.60

18.64

47.11

122.72

1 million

13.11

12.70

19.76

32.28

500,000

12.07

16.16

26.75

19.90

200,000

3.86

17.26

27.86

14.54

50,000

4.39

15.33

35.81

24.59

5,000

3.74

15.61

31.65

23.78

12.25

35.85

22.27

1,000

 

DT DNase treatment

3.12. iRNA Detection Sensitivity

Qiagen kit: 1. The same amount of total RNA (5  ng) from each microdissected sample was used as a template for reverse transcription. 2. RNA samples extracted from microdissected tissues were analysed in duplicate to quantify mature miR-21. 3. All the different microdissected areas showed miR-21 amplification at a Ct greater than 26. 4. Microdissected areas <3 × 106 mm2 showed miR-21 amplification at a Ct greater than 34 while microdissected areas >3 × 106 mm2 amplified at a Ct < 31 (Table 4). Low amount of total RNA extract meant that the two other miRNAs (miR-29c and miR-127) could not be quantified in these samples. 5. miRNA detection sensitivity gradually decreases with decreasing the amount of tissues used. TRIzol reagent: 1. The same amount of total RNA (5 ng) from each microdissected sample was used for the reverse transcriptase step to produce cDNA.

136

S.M. Khoshnaw et al.

Fig. 6. RNA (ng/whole elute) quantification (by RiboGreen reagent ) from 10 mm (DNase-treated vs. non-DNase-treated) and 20 mm (non-DNase-treated) thick sections in a range of microdissected tissue areas. Invitrogen kit used to extract RNA. DT DNase treatment.

2. In DNase-treated samples, RNA samples extracted from microdissected tissues were analysed in duplicate to quantify miR-21, miR-29c, and miR-127. In these samples, miR-21 started to amplify earlier compared to samples extracted by Qiagen kit (Table 4). 3. In non-DNase-treated samples, RNA samples extracted from microdissected tissues were analysed in duplicate to quantify miR-21, miR-29c, and miR-127. In 10  mm non-DNasetreated samples, most of the samples showed earlier amplification of miRNAs compared to samples extracted by Qiagen or TRIzol reagent with DNase treatment (Table 4). 4. miRNA detection sensitivity gradually decreases with decreasing the amount of tissues used. Among the three total RNA extraction kits, TRIzol produced the largest quantity of RNA and is comparable to Qiagen in terms of miRNA detection sensitivity. It is not recommended to include DNase treatment because miRNAs show earlier amplification in non-DNase-treated samples; moreover, in microdissected tissues,

10  Detection and Quantification of MicroRNAs in Laser-Microdissected…

137

Table 4 Quantification of miR-21, miR-29c, and miR-127 from RNA extracted from different areas of laser-microdissected tissues and different tissue thicknesses with and without DNase treatment (DT)

miRNA

Laser-micro Qiagen kit Ct in dissected area 10 mm sections (mm2) (DT)

miR-21

50 million

Invitrogen kit Ct in 10 mm sections (DT)

Invitrogen kit Invitrogen kit Ct in 10 mm Ct in 20 mm sections (no DT) sections (no DT) 19.62

10 million

25.99

21.65

20.75

21.05

7 million

29.28

21.91

20.73

20.04

5 million

30.35

21.52

20.85

20.46

3 million

28.88

26.21

21.96

22.20

1 million

35.28

24.68

22.78

22.23

500,000

34.7

25.99

24.48

24.24

200,000

34.51

26.85

26.23

25.40

50,000

38.75

28.73

28.54

29.06

5,000

36.78

36.13

32.8

33.06

36.32

34.63

34.46

1,000 miR-29c 50 million

26.81

10 million

28.07

27.2

27.43

7 million

28.42

26.72

26.17

5 million

27.26

27.08

26.52

27.78

28.59

3 million 1 million

28.62

500,000

32.18

30.75

200,000

32.6

No Ct

50,000

35.1

35.33

5,000

36.83

38.70

1,000

No Ct

No Ct (continued)

138

S.M. Khoshnaw et al.

Table 4 (continued)

miRNA

Laser-micro Qiagen kit Ct in dissected area 10 mm sections (mm2) (DT)

Invitrogen kit Ct in 10 mm sections (DT)

miR-127 50 million

Invitrogen kit Invitrogen kit Ct in 10 mm Ct in 20 mm sections (no DT) sections (no DT) 28.34

10 million

28.7

28.51

30.83

7 million

29.36

28.58

29.38

5 million

28.71

28.90

29.57

3 million

32.91

29.77

31.95

1 million

31.44

31.08

31.47

500,000

32.29

32.62

32.64

200,000

32.67

34.65

33.80

50,000

34.84

37.69

36.85

5,000

No Ct

No Ct

No Ct

1,000

No Ct

No Ct

No Ct

Ct threshold cycle value

the amount of total RNA is not significantly different in DNasetreated and non-DNase-treated samples, especially when larger areas of microdissected tissues are considered (see Note 2). 3.13. Comparing Different Tissue Section Thicknesses

Quantification of the three miRNAs from microdissected breast cancer tissues from the same case, extracted using TRIzol reagent, was repeated using different section thicknesses: 10, 20, and 30 mm. The 30 mm thickness was practically inappropriate because it was difficult to cut on a microtome and the sections did not adhere to the PALM membrane slide during deparaffinization of sections. Therefore, the 20 and 10 mm thick sections were studied for total RNA extract and miRNA detection sensitivity: 1. Total RNA extracted from 10 and 20  mm thick sections: Amount of RNA was more than double in 20 vs. 10 mm thick sections in samples >1 × 106 mm2 (Table 3, Fig. 6). 2. miRNAs’ detection in total RNAs: Cts at which miRNAs started to amplify, where comparable between 10 and 20 mm thick sections. Indeed, amplification was later in 20 mm thick sections in some of the samples which was against expectation (Table 4, Fig. 7), but 20 mm thick sections are recommended, since they produce considerably higher amounts of total RNA that is crucially important in microdissected tissues which

10  Detection and Quantification of MicroRNAs in Laser-Microdissected…

139

Fig. 7. miR-21, miR-29c, and miR-127 quantification from 10 mm (DNase-treated vs. non-DNase-treated) and 20 mm (non-DNase-treated) thick tissue sections in a range of microdissected tissue areas. Trizol kit was used to extract RNA. DT DNase treatment.

140

S.M. Khoshnaw et al.

contain tiny amounts of RNA. Moreover, using thicker sections reduces the area required to be microdissected for successful miRNA profiling (see Note 1). 3.14. Identifying the Area of Breast Tissue to be Microdissected

1. Table 3 and Fig. 6 show that 5 × 106 mm2 of breast cancer tissues dissected from 20 mm sections is the amount of tissue from which the required amount of RNA for miRNA microarray (100 ng) can be extracted when RNA is extracted by TRIzol reagent without excessive microdissection, and that it is sufficient amount of tissue to allow reliable miRNA detection (see Note 1). 5 × 107 mm2 of FFPE tissue was macrodissected with a needle which produced 722.61 ng/whole elute volume (12 ml) in a 10 mm thick FFPE section. miR-21 was amplified at Ct 19.62, miR-29c at 26.81, and miR-127 at 28.34, respectively. In these experiments, it was ensured that in each sample, the exact amount of total RNA (5  ng) was used per reverse transcription reaction to quantify miRNAs (Table 4, Fig. 7). 2. In contrast to previously published results (32, 33), our data reveals that miRNA detection sensitivity depends on the amount of tissue used and that miRNA detection sensitivity gradually decreases with decreasing the amount of tissues when the same amount of total RNA is used for cDNA production (Table 4, Fig. 7). miRNAs cannot be accurately detected and quantified by extracting miRNA from a very limited number of microdissected cells from FFPE tissue sections (see Note 1).

4. Notes 1. For sensitive and reliable miRNA detection from laser-microdissected breast cancer tissues, it is imperative that a sufficient amount of tissue is microdissected, and we recommend this to be at least 5 × 106 mm2 tissues from 20 mm thick FFPE sections. Further technical refinements would optimise the amount of tissue needed according to the age of FFPE samples. 2. Among the three commercial RNA extraction kits (Qiagen, RecoverAll, and TRIzol), TRIzol is the most sensitive kit in isolating miRNAs from microdissected FFPE tissues. 3. A comprehensive miRNA expression profiling in normal, preinvasive, invasive, and metastatic microdissected breast tissues provides novel insights into how miRNAs might propel normal breast epithelial cells into becoming metastatic cancer cells. 4. To bring miRNAs closer to clinical settings in the management of breast cancer, future studies should spot the key miRNAs involved in the processes of carcinogenesis and cancer progression in breast tissues.

10  Detection and Quantification of MicroRNAs in Laser-Microdissected…

141

Acknowledgements We would like to thank the University of Nottingham and Breakthrough Breast Cancer Research Centre for funding this work. The authors express thanks to Dr Z Hodi for reviewing the cases. References 1. Reinhart, B. J., Slack, F. J., Basson, M., Pasquinelli, A. E., Bettinger, J. C., Rougvie, A. E., Horvitz, H. R., and Ruvkun, G. (2000) The 21-nucleotide let-7 RNA regulates developmental timing in Caenorhabditis elegans, Nature 403, 901–906. 2. Houbaviy, H. B., Murray, M. F., and Sharp, P. A. (2003) Embryonic stem cell-specific MicroRNAs, Dev Cell 5, 351–358. 3. Lee, Y., Kim, M., Han, J., Yeom, K. H., Lee, S., Baek, S. H., and Kim, V. N. (2004) MicroRNA genes are transcribed by RNA polymerase II, EMBO J 23, 4051–4060. 4. Denli, A. M., Tops, B. B., Plasterk, R. H., Ketting, R. F., and Hannon, G. J. (2004) Processing of primary microRNAs by the Microprocessor complex, Nature 432, 231–235. 5. Yi, R., Qin, Y., Macara, I. G., and Cullen, B. R. (2003) Exportin-5 mediates the nuclear export of pre-microRNAs and short hairpin RNAs, Genes Dev 17, 3011–3016. 6. Hutvagner, G., McLachlan, J., Pasquinelli, A. E., Balint, E., Tuschl, T., and Zamore, P. D. (2001) A cellular function for the RNA-interference enzyme Dicer in the maturation of the let-7 small temporal RNA, Science 293, 834–838. 7. Gregory, R. I., Chendrimada, T. P., Cooch, N., and Shiekhattar, R. (2005) Human RISC couples microRNA biogenesis and posttranscriptional gene silencing.(see comment), Cell 123, 631–640. 8. Meister, G., Landthaler, M., Patkaniowska, A., Dorsett, Y., Teng, G., and Tuschl, T. (2004) Human Argonaute2 mediates RNA cleavage targeted by miRNAs and siRNAs, Mol Cell 15, 185–197. 9. Lim, L. P., Glasner, M. E., Yekta, S., Burge, C. B., and Bartel, D. P. (2003) Vertebrate microRNA genes, Science 299, 1540. 10. Lewis, B. P., Burge, C. B., and Bartel, D. P. (2005) Conserved seed pairing, often flanked by adenosines, indicates that thousands of human genes are microRNA targets, Cell 120, 15–20.

11. Pillai, R. S., Bhattacharyya, S. N., Artus, C. G., Zoller, T., Cougot, N., Basyuk, E., Bertrand, E., and Filipowicz, W. (2005) Inhibition of translational initiation by Let-7 MicroRNA in human cells, Science 309, 1573–1576. 12. Yekta, S., Shih, I. H., and Bartel, D. P. (2004) MicroRNA-directed cleavage of HOXB8 mRNA, Science 304, 594–596. 13. Lee, R. C., Feinbaum, R. L., and Ambros, V. (1993) The C. elegans heterochronic gene lin-4 encodes small RNAs with antisense complementarity to lin-14. Cell 75, 843–854. 14. Hayashita, Y., Osada, H., Tatematsu, Y., Yamada, H., Yanagisawa, K., Tomida, S., Yatabe, Y., Kawahara, K., Sekido, Y., and Takahashi, T. (2005) A polycistronic microRNA cluster, miR-17-92, is overexpressed in human lung cancers and enhances cell proliferation, Cancer Research 65, 9628–9632. 15. Krek, A., Grun, D., Poy, M. N., Wolf, R., Rosenberg, L., Epstein, E. J., MacMenamin, P., da Piedade, I., Gunsalus, K. C., Stoffel, M., and Rajewsky, N. (2005) Combinatorial microRNA target predictions, Nat Genet 37, 495–500. 16. Johnson, S. M., Grosshans, H., Shingara, J., Byrom, M., Jarvis, R., Cheng, A., Labourier, E., Reinert, K. L., Brown, D., and Slack, F. J. (2005) RAS is regulated by the let-7 microRNA family, Cell 120, 635–647. 17. Zhu, S., Si, M. L., Wu, H., and Mo, Y. Y. (2007) MicroRNA-21 targets the tumor suppressor gene tropomyosin 1 (TPM1), J Biol Chem 282, 14328–14336. 18. Iorio, M. V., Ferracin, M., Liu, C.-G., Veronese, A., Spizzo, R., Sabbioni, S., Magri, E., Pedriali, M., Fabbri, M., Campiglio, M., Menard, S., Palazzo, J. P., Rosenberg, A., Musiani, P., Volinia, S., Nenci, I., Calin, G. A., Querzoli, P., Negrini, M., and Croce, C. M. (2005) MicroRNA gene expression deregulation in human breast cancer, Cancer Res 65, 7065–7070.

142

S.M. Khoshnaw et al.

19. Blenkiron, C., and Miska, E. A. (2007) miRNAs in cancer: approaches, aetiology, diagnostics and therapy, Hum Mol Genet 16 Spec No 1, R106-113. 20. Sorlie, T., Tibshirani, R., Parker, J., Hastie, T., Marron, J. S., Nobel, A., Deng, S., Johnsen, H., Pesich, R., Geisler, S., Demeter, J., Perou, C. M., Lonning, P. E., Brown, P. O., BorresenDale, A. L., and Botstein, D. (2003) Repeated observation of breast tumor subtypes in independent gene expression data sets, Proc Natl Acad Sci U S A 100, 8418–8423. 21. Srinivasan, M., Sedmak, D., and Jewell, S. (2002) Effect of fixatives and tissue processing on the content and integrity of nucleic acids, Am J Pathol 161, 1961–1971. 22. Masuda, N., Ohnishi, T., Kawamoto, S., Monden, M., and Okubo, K. (1999) Analysis of chemical modification of RNA from formalin-fixed samples and optimization of molecular biology applications for such samples, Nucleic Acids Res 27, 4436–4443. 23. Li, J., Smyth, P., Flavin, R., Cahill, S., Denning, K., Aherne, S., Guenther, S. M., O’Leary, J. J., and Sheils, O. (2007) Comparison of miRNA expression patterns using total RNA extracted from matched samples of formalin-fixed paraffin-embedded (FFPE) cells and snap frozen cells, BMC Biotechnology 7, 36. 24. Fend, F., and Raffeld, M. (2000) Laser capture microdissection in pathology, J Clin Pathol 53, 666–672. 25. Espina, V., Heiby, M., Pierobon, M., and Liotta, L. A. (2007) Laser capture microdissection technology, Expert Rev Mol Diagn 7, 647–657. 26. Wang, H., Ach, R. A., and Curry, B. (2007) Direct and sensitive miRNA profiling from low-input total RNA, RNA 13, 151–159.

27. Si, M. L., Zhu, S., Wu, H., Lu, Z., Wu, F., and Mo, Y. Y. (2007) miR-21-mediated tumor growth, Oncogene 26, 2799–2803. 28. Yan, L.-X., Huang, X.-F., Shao, Q., Huang, M.-Y., Deng, L., Wu, Q.-L., Zeng, Y.-X., and Shao, J.-Y. (2008) MicroRNA miR-21 overexpression in human breast cancer is associated with advanced clinical stage, lymph node metastasis and patient poor prognosis, RNA, rna.1034808. 29. Green, A. R., Krivinskas, S., Young, P., Rakha, E. A., Paish, E. C., Powe, D. G., and Ellis, I. O. (2009) Loss of expression of chromosome 16q genes DPEP1 and CTCF in lobular carcinoma in situ of the breast, Breast Cancer Res Treat 113, 59–66. 30. Chen, C., Ridzon, D. A., Broomer, A. J., Zhou, Z., Lee, D. H., Nguyen, J. T., Barbisin, M., Xu, N. L., Mahuvakar, V. R., Andersen, M. R., Lao, K. Q., Livak, K. J., and Guegler, K. J. (2005) Real-time quantification of microRNAs by stem-loop RT-PCR, Nucleic Acids Res 33, e179. 31. Lao, K., Xu, N. L., Yeung, V., Chen, C., Livak, K. J., and Straus, N. A. (2006) Multiplexing RT-PCR for the detection of multiple miRNA species in small samples, Biochem Biophys Res Commun 343, 85–89. 32. Tang, F., Hajkova, P., Barton, S. C., Lao, K., and Surani, M. A. (2006) MicroRNA expression profiling of single whole embryonic stem cells, Nucleic Acids Research 34, e9. 33. Mestdagh, P., Feys, T., Bernard, N., Guenther, S., Chen, C., Speleman, F., and Vandesompele, J. (2008) High-throughput stem-loop RT-qPCR miRNA expression profiling using minute amounts of input RNA, Nucl. Acids Res., gkn725.

Chapter 11 Laser Capture Microdissection Applications in Breast Cancer Proteomics René B.H. Braakman, Theo M. Luider, John W.M. Martens, John A. Foekens, and Arzu Umar Abstract Breast cancer tissues are characterized by cellular heterogeneity, representing a mixture of, e.g., healthy epithelial ducts, invasive or in situ tumor cells, surrounding stroma, infiltrating immune cells, blood vessels, and capillaries. As a consequence, protein extracts from whole tissue lysates also represent a variety of cell types present in the tissues under examination. This, however, seriously hampers the analysis of tumor cell-specific signals, which is of interest when performing biomarker discovery-type of studies. Therefore, laser capture microdissection is a perfect tool to isolate a relatively pure population of cells of interest, such as tumor cells. In this chapter, we describe the use of the PALM MicroBeam system for laser microdissection and pressure catapulting. Protocols are provided for sectioning, staining, microdissection, sample preparation, and mass spectrometric analysis of snap frozen breast cancer tissue. Key words: Breast cancer, Laser capture microdissection, nLC, MALDI, Mass spectrometry, Proteomics, Biomarker discovery

1. Introduction High throughput, state-of-the-art “omics” techniques are becoming widely used in cancer research and biomarker discovery studies (1–3). The secret of success of all these “omics” studies, however, is not merely technology based, but very heavily leans on the source that is being investigated. In cancer research, the basis of all work is the use of (tumor) tissue specimens from individuals of whom clinical follow-up information is available. Although tissue specimens provide a perfect source for translational research,

Graeme I. Murray (ed.), Laser Capture Microdissection: Methods and Protocols, Methods in Molecular Biology, vol. 755, DOI 10.1007/978-1-61779-163-5_11, © Springer Science+Business Media, LLC 2011

143

144

R.B.H. Braakman et al.

tumor tissues, in particular breast cancer, are very heterogeneous in cell type and composition. Tissue heterogeneity can seriously hamper downstream data interpretation acquired from a mixture of cells. In this respect, development of laser capture microdissection (LCM) technology has filled this gap by providing a means of selecting and enriching for specific cell populations of interest from their natural environment prior to subsequent molecular analysis (4). LCM has put an important step forward in both genomics and proteomics cancer biomarker discovery studies. Two types of LCM-based technologies are commonly used: the thermoplastic film contact based Arcturus system (Molecular Devices) (5) and the noncontact laser pressure catapulting PALM system (Carl Zeiss) (6). Comparative proteomics of LCM-derived breast cancer cells has been performed using 2DE (7, 8), or LC-MS/MS (9) approaches, which has resulted in the identification of proteins involved in breast cancer metastasis (7, 9), and prognosis (8). These studies have shown that proteomics technologies have advanced in such a way that they can contribute to biomarker discovery. However, major drawbacks of these studies are that they either required the use of large amounts of starting material (varying from 42 to 700 mg protein), or they yielded few identifications (50–76 proteins) in the case where little starting material was used. Clinical samples, on the other hand, are usually available in very small quantities. To allow comprehensive analysis of minute amounts of material (<1  mg), a robust proteomics platform that offers exceptional sensitivity and dynamic range of detection in a high throughput, accurate, and reproducible fashion is imperative. Recently, we have demonstrated the applicability of nanoLC-FTICR MS in combination with the accurate mass and time (AMT) tag approach for proteomic characterization of <5,000 LCM-derived breast cancer cells (10, 11). These studies showed that proteome coverage was improved compared with conventional techniques. Furthermore, we showed that large-scale protein identification, as well as quantitation is feasible from minute amounts of clinically relevant samples using nLC-FTICR MS in combination with AMT tag database searching. Most importantly, comparative proteome analysis of LCM-derived breast cancer cells resulted in the identification of a putative protein profile predicting tamoxifen therapy resistance in breast cancer (10). In this chapter, we describe the protocols used for sectioning, staining, microdissection, sample preparation, and mass spectrometric analysis of snap frozen breast cancer tissue.

11  Laser Capture Microdissection Applications in Breast Cancer Proteomics

145

2. Materials 1. Cryostat. 2. TissueTek O.C.T. compound. 3. Lint-free wipes, Wipe. 4. SuperFrost Plus glass slides (Menzel-Gläser). 5. PALM 1 mm polyethylene naphthalate (PEN) membrane glass slides (Carl Zeiss, Thornwood, NY). 6. UV incubator for glass slides. 7. Tap water. 8. Water, mLC/MS grade (Biosolve Chemicals, Valkenswaard, the Netherlands). 9. Hematoxylin. 10. Eosin. 11. Xylene. 12. Ethanol, tech grade. 13. Acetonitrile, mLC/MS grade (Biosolve Chemicals). 14. Trifluoroacetic acid, 1 ml ampules (TFA). 15. Acetic acid. 16. Halt protease and phosphatase inhibitor cocktail, EDTA free (Pierce, Thermo Fischer Scientific). 17. Aluminum foil. 18. Pertex mounting medium (Histolab). 19. PALM® MicroBeam system (Carl Zeiss). 20. PALM® adhesive caps, opaque (Carl Zeiss). 21. Protein LoBind tubes, PCR clean (Eppendorf, Hauppauge, NY). 22. Safe-lock tubes (Eppendorf ). 23. Trypsin Gold, MS grade (Promega, Madison, MI), reconstituted in 50 mM acetic acid stable for 1 year at −80°C. 24. dl-Dithiothreitol (DTT, Sigma-Aldrich, St Louis, MO). 25. Iodoacetamide, single use ampules (IAA, Perbio Biosciences, Cramlington, UK). 26. RapiGest S.P. surfactant (Waters), store at RT, reconstitute fresh as required in 50  mM NH4HCO3, stable at +4°C for 1 week, or −20°C for 1 month (working solution as follows: 10% (v/v) TFA, 50  mM acetic acid, 50  mM NH4HCO3,

146

R.B.H. Braakman et al.

100 ng/mL trypsin in 50 mM acetic acid, 500 mM DTT in 50 mM NH4HCO3, 0.1% (w/v) Rapigest in 50 mM NH4HCO3, 325 mM iodoacetamide and 50 mM NH4HCO3). 27. Thermomixer with exchangeable thermoblock (Eppendorf). 28. Paper pH strips. 29. 2,5-Dihydrobenzoic acid (Bruker Daltonik, Bremen, Germany), 1 mg/ml in 0.1% TFA, stable for 1 week at +4°C. 30. Peptide calibration standard (Bruker Daltonik). 31. AnchorChip™ 400 mm target plate (Bruker Daltonik).

3. Methods The methods section described below outlines (1) tissue preparation, (2) staining, (3) laser microdissection and pressure catapulting (LMPC), (4) sample preparation for proteomics applications, and (5) mass spectrometry. 3.1. Tissue Preparation 3.1.1. Tumor Tissue

Primary breast tumor tissues are available through our liquid N2 tissue bank. Studies in the lab involving tumor tissues are approved by the Medical Ethics Committee (MEC 02.953) of the Erasmus Medical Center Rotterdam, The Netherlands, and are performed in accordance with the Code of Conduct of the Federation of Medical Scientific Societies in the Netherlands. For experimental purposes, tissues are stored at −80°C (mid-term) or on dry ice (short-term).

3.1.2. Tissue Cryosectioning

Tissue cryosections are prepared in the cryostat (see Note 1). Set the microtome blade at −23°C and the tissue holder at −25°C as a starting point, and adjust the temperature if needed. Place tissue on the holder using TissueTek mounting material but do not fully embed the tissue (see Note 2). Cut 5 mm sections and mount on regular glass slides for hematoxylin/eosin (HE) stain. These slides are kept at room temperature (overnight if desired) for drying prior to HE stain. For microdissection, cut 8 mm sections and place on PEN covered glass slides that were previously UV sterilized for 30 min (see Note 3). Dry the slides for 30 s at room temperature and then fix for 30  s in 70% ice cold ethanol (see Note 4). After fixation, slides are rinsed in 100% ethanol and dried at room temperature for 1–2  min. At this point slides can be stored at −80°C in a slide container wrapped in aluminum foil, or used for subsequent staining.

3.2. Staining

We use hematoxylin/eosin (HE) stained tissue sections for microscopic examination to determine which areas should be microdissected.

11  Laser Capture Microdissection Applications in Breast Cancer Proteomics

147

Printouts of digital scans of these slides serve as a guide during microdissection of hematoxylin-only stained tissues. 3.2.1. HE Staining of Control Sections

Tissue sections on regular glass slides are stained using a standard hematoxylin/eosin staining procedure. Hematoxylin is filtered before first staining round. Tissue sections are fixed and rehydrated in consecutive 1-min washes in 70% and 50% ethanol and a final wash in tap water. Sections are stained for 30 s in hematoxylin, and bluing is developed by subsequent washing in running tap water for 5 min. Counter stain with eosin for 1–2 s and quickly rinse off excess of stain in tap water. De-hydrate sections by consecutive washes in 50%, 70%, and 2× 100% ethanol for 1 min, and one final 100% wash for 5 min, followed by two consecutive washed in xylene for 1 min with the final third wash for 5 min. Remove slides from the xylene solution and place on a tissue wipe. Add one drop of mounting solution (Pertex) on top of the section. Cover the section with a cover slip in such a way that no air bubbles are trapped between the tissue and the cover slip and no Pertex is on top of the cover slip, which may blur microscopic examination. Let the slides dry properly in the fume hood prior to microscopic examination.

3.2.2. Hematoxylin Staining for Proteomics Applications

Defrost PEN slides that were stored at −80°C at room temperature prior to staining (see Note 5). Filter hematoxylin, add protease and phosphatase inhibitors to the staining solutions, and keep solutions at room temperature (see Note 6, and Fig. 1). Stain slides in multislide holder containers in a volume of 20 ml. Rehydrate sections in tap water for 15  s, stain with hematoxylin for 30  s, wash in tap water for 15  s, dehydrate sections by consecutive washes in 50%, 70%, 95%, and 100% ethanol for 15 s, and a final 30-s wash in 100% ethanol. Air-dry the slides and immediately proceed with microdissection (see Note 7).

Fig. 1. Hematoxylin stain of 8 mm breast cancer tissue cryosection. Staining was performed in the presence of protease and phosphatase inhibitors in aqueous solutions stored either at room temperature (left panel ) or +4°C (right panel ).

148

R.B.H. Braakman et al.

3.3. Laser Microdissection and Pressure Catapulting

Laser microdissection and pressure catapulting (LMPC) described in this section is performed using the PALM® Laser-Microbeam system, which has a pulsed air-cooled UV-nitrogen laser. The PALM® system includes an inverted microscope with attached color CCD camera, computer-controlled X–Y stage, and LMPC capture unit. Prior to microdissection, make sure that the microscope is set up for bright field, that the cap is properly centered, and that the laser marker is positioned correctly. In the PALM® system, the UV laser cuts the PEN membrane, to which the tissue is attached, below the focal plane of the tissue specimen and the loose element is subsequently pulsed with a single, or multiple if necessary, laser shots to catapult the tissue into the cap that is placed right above the tissue, using the so-called laser pressure catapulting (LPC) option. For capturing cells, we use opaque PALM® adhesive caps (see Note 8 and Fig. 2).

3.3.1. Instrument Parameters and Software Settings

Correct instrument parameters need to be determined for each magnification to be used. For 8 mm breast cancer tissue sections, we find the following parameters work best in our hands at 20× magnification: Laser energy: 60 Laser focus: 63 LPC energy delta: 18

Fig. 2. Laser microdissection of breast cancer tissues. Tissue morphology is poorly visible using regular PALM caps (top left panel ), and improved using adhesive PALM caps (top right panel ). Area of interest is cut using the Close&Cut or RoboLPC settings (bottom left panel ), and catapulted into the cap (bottom right panel ).

11  Laser Capture Microdissection Applications in Breast Cancer Proteomics

149

LPC focus delta: 4 RoboLPC joint distance: 6 mm Different software settings are available for microdissection. We either use “Cut” and manual LPC or, if possible, semi-automated “RoboLPC” functions (see Note 9). For microdissection, encircle the area of interest by creating an element, start microdissection through “Cut,” adjust focus, laser, and LPC settings if necessary, and check whether the element is indeed captured on the cap (see Fig.  2). Similarly, “RoboLPC” cuts the element of interest but leaves the end un-cut (RoboLPC joint distance). The tissue is then automatically catapulted into the cap by a high energy pulse on the joint. We typically perform RoboLPC in series of ten elements per area of 100,000 mm2, and repeat this until the desired total area is dissected. Continue collecting elements until the desired area is reached. For breast cancer tissue, we typically collect ~500,000 mm2 (see Note 10). To keep track of all collected areas, we use the auto image option to collect images of all elements before (and after) microdissection. When finished, check the slide at 5× magnification for any elements that did not reach the cap, and catapult those areas into the cap a second time (see Note 11). 3.3.2. Sample Storage

3.4. Sample Preparation for Proteomics Applications

Once LMPC is completed, remove the capture cap and add 10–20 ml RapiGest lysis buffer (in 50  mM ammonium bicarbonate) to suspend captured tissues. Carefully transfer suspension, without damaging the silicon layer of the adhesive cap, to a protein LoBind Eppendorf microfuge tube (see Note 12). Store the samples at −80°C until further use. 1. Defrost samples at room temperature, put samples on ice when defrosted. 2. Add 0.1% Rapigest to obtain the desired buffer volume. 3. Sonicate the samples in a Branson cell disruptor at 70% amplitude for 1 min. 4. Incubate at 95°C for 5 min. 5. Add DTT to obtain a concentration of 5 mM (see Note 13) (reduction and alkylation is only performed when proceeding with nLC-MS, for MALDI-MS it is better omitted). 6. Incubate at 60°C for 30 min (reduction and alkylation is only performed when proceeding with nLC-MS, for MALDI-MS it is better omitted). 7. Spin down briefly and add IAA work solution to a final concentration of 15 mM (reduction and alkylation is only performed when proceeding with nLC-MS, for MALDI-MS it is better omitted).

150

R.B.H. Braakman et al.

8. Place in the dark for 30 min at room temperature (reduction and alkylation is only performed when proceeding with nLCMS, for MALDI-MS it is better omitted). 9. Add 100 ng trypsin and incubate at 37°C in a thermomixer shaking at 650 rpm for 4 h. 10. Add 10% TFA to a final concentration of 0.5%. 11. Incubate at 37°C for 30 min. 12. Centrifuge at 14,000 rpm for 15 min. 13. Transfer supernatant to microfuge tubes or HPLC vials and store at −80°C. 3.5. Mass Spectrometry

3.5.1. MALDI-FTICR MS

Global proteome analysis can be performed on tryptic digests from microdissected cells. For biomarker discovery, proteomes of two or multiple groups are typically compared using bioinformatics tools. Multiple mass spectrometry approaches are available for biomarker discovery type studies, of which we will describe protocols for matrix-assisted laser desorption/ionization (MALDI)-Fourier transform ion cyclotron resonance (FTICR) mass spectrometry, and nanoscale liquid chromatography (nLC)-MS. For global proteome profiling by MALDI-FTICR MS, we typically use one-tenth of the total tryptic digest sample volume. This methodology is sensitive enough to obtain peptide signals in the MS spectrum from as little as 100–300 cells, depending on the type of sample that is measured. Spotting procedure: 1. Mix equal volumes of DHB matrix with tryptic digest into a safe-lock eppendorf tube (see Note 14). 2. Vortex and spin down briefly. 3. Spot 1 ml of DHB/sample mix onto an AnchorChip target plate in duplicate or triplicate. 4. Let the spotted sample crystallize, avoid using a blow dryer (see Note 15). 5. Measure the samples using a MALDI-FTICR mass spectrometer (e.g., Apex IV Qe (Bruker Daltonik)), accumulating 100 scans of ten shots each at 60% laser power, and acquiring spectra in the mass range of 800–4,000 m/z.

3.5.2. nLC-MS

For in-depth proteome profiling, tryptic digests are analyzed by nLC-MS (e.g., using an Ultimate 3000 system coupled to LTQ Orbitrap XL™). In our hands, a minimum number of ~2,000, but preferably >4,000, microdissected breast cancer cells is necessary to obtain reproducible MS/MS data. Samples are loaded onto a C18 trap column, desalted with 0.1% TFA, switched online with the analytical C18 column, and separated using a formic

11  Laser Capture Microdissection Applications in Breast Cancer Proteomics

151

acid/acetonitrile gradient. The general rule applies that longer nLC columns (with narrow inner diameter) and longer gradient time result in higher number of protein identifications. MS/MS identifications are searched against protein databases for peptide and subsequent protein matches.

4. Notes 1. Make sure to clean the inside of the cryostat and the cutting blade with lint-free wipes instead of normal wipes, to avoid fibers to stick to your PEN membrane glass slides. Also, avoid using a hair brush to place sections on the slides. In both cases, keratin contamination may be introduced that interferes with future proteomic applications. Use of steel needles and an antiroll plate is a good alternative. Also, use nonpowdered, latex-free gloves, preferably nitrile. 2. Do not fully embed tissues in TissueTek to avoid polymer contamination in the downstream sample preparation process, since this negatively influences mass spectrometry through ion suppression and chemical background. 3. Take notice of the expiry date for PALM® slides. The quality of slides goes down within 6  months. We have observed loosening of the PEN layer from glass slides, particularly visible as little dots or halos (air bubbles) underneath the PEN layer. Loosening of the PEN layer leads to moist that is trapped between PEN and glass, which negatively influences catapulting of tissue. Moist can be observed if the membrane is damaged by, e.g., the microtome blade; this also hampers LMPC. 4. After placing cryosections on PEN covered glass slides it is important to air dry the slides outside the cryostat for ~30 s prior to ethanol fixation. If the sections are not completely dry, they may fall off during subsequent fixing and staining, or may prove be too loose for proper microdissection. Avoid the microtome blade while placing the sections on the PEN slide, as this may damage the membrane. 5. Frozen glass slides stored in plastic multislide containers wrapped in aluminum foil are best defrosted on the bench. Make sure the foil is at room temperature before opening (3–5 min) to avoid moist on the slides. 6. Staining solutions containing protease and phosphatase inhibitor cocktail should be prepared freshly prior to use and stored at room temperature for a maximum of half a day. Storage of inhibitor containing solutions at 4°C results in poor visibility of stained tissue sections, probably due to precipitation of inhibitors. We prefer using inhibitor solution

152

R.B.H. Braakman et al.

instead of tablets, because the latter are difficult to dissolve in ethanol containing solutions. Inhibitor cocktail is added to all water-containing solutions, except for 95% ethanol, because inhibitors tend to precipitate at higher ethanol percentages. Inhibitor containing ethanol solutions can be reused for two to three slides, water needs to be replaced for each slide otherwise bluing is negatively affected. 7. Complete microdissection within 1–1.5 h from start of staining, to make sure that endogenous protease and phosphatase activity is not restored. In cases where the desired number of cells is not obtained within the available time frame, we prefer to take a new slide from the freezer containing sections from the same tissue, stain it, and perform a second dissection. 8. PALM offers a variety of tubes with special caps that facilitate easy capture of cells. Regular PALM caps have in insert halfway that reduces that distance of the tissue to the cap from ~7 mm to ~3  mm. Palm caps can be filled with lysis buffer so that catapulted cells are directly captured in the desired buffer. In our hands, we have observed that a volume of 5 ml quickly evaporates, and larger volumes stick to the side of the cap (instead of the center) and consequently drip onto the tissue section, which complicates microdissection. We prefer to use adhesive caps, which are caps filled with silicon. The advantage of these caps is that the distance between the tissue and the cap is ~1 mm, so catapulted material has a higher chance of actually reaching the cap. Catapulted cells are captured on the adhesive material without the need for using lysis buffer. Most importantly, the adhesive cap functions as a diffuser that reduces diffraction shadows, thereby providing improved visibility of tissue morphology during microdissection, as well as easy inspection of captured cells (see Fig. 2). 9. Care should be taken when using the “Cut” or “Cut&Close” function because dissected material may fly away (due to laser pressure or air flow) without reaching the cap. This problem can be avoided using RoboLPC. In addition, the latter function speeds up total dissection time because manual LPC is replaced by automated LPC. However, RoboLPC only works well if the tissue section is in focus in all areas, and if no irregularities in tissue density occur during cutting. This may be particularly critical at tumor cell-stroma borders. Therefore, we recommend adjusting focus and energy settings if necessary during RoboLPC. In addition, AutoLPC function may be used for areas that are otherwise difficult to dissect. AutoLPC does not cut the tissue but merely catapults the whole area of interest through LPC pulses, and can be performed on tissues placed on regular glass slides without PEN coating. While AutoLPC speeds up the process of microdissection even more, care must be taken that catapulted little tissue flakes are

11  Laser Capture Microdissection Applications in Breast Cancer Proteomics

153

actually captured by the cap and not scattered around the glass slide. In our hands, protein recovery is significantly reduced when using AutoLPC compared to RoboLPC. 10. Using the microdissected tissue area, the number of cells can be calculated according to the following equation, assuming that one tumor cell has a volume of 10 × 10 × 10  mm (=1,000 mm3): number of cells = area (mm2) × section thickness (mm)/cell volume (mm3). Furthermore, based on whole tissue lysates, we estimate that ~150  ng protein can be extracted from ~1,000 cells (12). 11. Elements that do not reach the cap and fall back down on the tissue are observed quite frequently when using regular PALM® caps but very rarely with adhesive caps. 12. Alternatively, close the tube of the adhesive cap and store directly at −80°C without adding any lysis buffer. Lysis buffer can than be added at a later stage during sample preparation. However, we prefer to add lysis buffer directly and transfer the captured tissues into a LoBind tube, because the adhesive cap can then be directly inspected under the microscope for remaining tissues. This particularly happens near the rim of the cap. If not all tissues are properly transferred, additional buffer can be added to the cap and transferred into the same LoBind tube. 13. Most reagents and buffers are very unstable. The processes of reduction, alkylation, and trypsin digestion require alkaline pH of 7.5–8.5. Therefore, prepare fresh stock solutions on the same day as sample preparation, keep solutions on ice prior to use, and check pH. Trypsin stock solution should be acidic (pH < 4) to suppress its activity and prevent autolysis. 14. When samples are mixed with matrix solution in Eppendorf LoBind tubes, serious polymer contamination occurs, which hampers the crystallization process resulting in irreproducible MS spectra and poor sensitivity. We recommend mixing sample with matrix solution in Eppendorf safe-lock tubes. On the other hand, if minute amounts of sample are stored at −80°C in safe-lock tubes, protein and peptide recovery are very low. A good trade-off is to store samples at −80°C in LoBind tubes, and mix in safe-lock tubes prior to spotting and crystallization. 15. If poor crystallization occurs anyway, recrystallize the spot by adding 1 ml HPLC grade water and let it air dry.

Acknowledgments The authors wish to thank Annemieke Timmermans for helpful discussions and technical assistance on tissue sectioning and staining. RBHB and AU are (partly) financially supported through

154

R.B.H. Braakman et al.

the Center for Translational Molecular Medicine, CTMM BreastCARE project 030–104, and through the Netherlands Genomics Initiative. References 1. Chen X, E Jorgenson, ST Cheung (2009) New tools for functional genomic analysis. Drug Discov Today 14:754–60 2. Hood BL, NA Stewart, TP Conrads (2009) Development of high-throughput mass spectrometry-based approaches for cancer biomarker discovery and implementation. Clin Lab Med 29:115–38 3. Kerschgens J, T Egener-Kuhn, N Mermod (2009) Protein-binding microarrays: probing disease markers at the interface of proteomics and genomics. Trends Mol Med 15:352–8 4. Emmert-Buck MR, RF Bonner, PD Smith et al (1996) Laser capture microdissection. Science 274:998–1001 5. Erickson HS, JW Gillespie, MR Emmert-Buck (2008) Tissue microdissection. Methods Mol Biol 424:433–48 6. Micke P, A Ostman, J Lundeberg et  al (2005) Laser-assisted cell microdissection using the PALM system. Methods Mol Biol 293:151–66 7. Hudelist G, CF Singer, KI Pischinger et  al (2006) Proteomic analysis in human breast cancer: identification of a characteristic protein expression profile of malignant breast epithelium. Proteomics 6:1989–2002

8. Neubauer H, SE Clare, R Kurek et al (2006) Breast cancer proteomics by laser capture microdissection, sample pooling, 54-cm IPG IEF, and differential iodine radioisotope detection. Electrophoresis 27:1840–52 9. Zang L, D Palmer Toy, WS Hancock et  al (2004) Proteomic analysis of ductal carcinoma of the breast using laser capture microdissection, LC-MS, and 16O/18O isotopic labeling. J. Proteome. Res. 3:604–12 10. Umar A, H Kang, AM Timmermans et  al (2009) Identification of a putative protein profile associated with tamoxifen therapy resistance in breast cancer. Mol Cell Proteomics 8:1278–94 11. Umar A, TM Luider, JA Foekens et al (2007) NanoLC-FT-ICR MS improves proteome coverage attainable for approximately 3000 lasermicrodissected breast carcinoma cells. Proteomics 7:323–9 12. Umar A, JC Dalebout, AM Timmermans et al (2005) Method optimisation for peptide profiling of microdissected breast carcinoma tissue by matrix-assisted laser desorption/ ionisation-time of flight and matrix-assisted laser desorption/ionisation-time of flight/ time of flight-mass spectrometry. Proteomics 5:2680–8

Chapter 12 Proteomic Analysis of Laser Microdissected Ovarian Cancer Tissue with SELDI-TOF MS Isabelle Cadron, Toon Van Gorp, Philippe Moerman, Etienne Waelkens, and Ignace Vergote Abstract The evolution of ovarian cancer will highly depend on platinum sensitivity or resistance. In an attempt to better understand this mechanism of resistance, we combined laser microdissection (LMD) of ovarian tumor cells with mass spectrometric techniques. To obtain disease-specific markers we isolated ovarian tumor cells with LMD without contamination of surrounding cells. Proteins were extracted by chemical lysis and subsequent peptide and protein information was gathered using surface-enhanced laser desorption/ ionization-time of flight mass spectrometry. Key words: Ovarian cancer, SELDI-TOF MS, Platinum resistance, Laser microdissection, Biomarkers

1. Introduction Ovarian cancer patients are mostly diagnosed with advanced stage (III/IV) disease necessitating extensive debulking surgery and platinum-based chemotherapy. Despite this, ±25% of patients will relapse during or within 6 months after initial therapy reflecting platinum-resistant disease (1). Resistance to platinum chemotherapy is multifactorial, though little is known about differences in protein expression between platinum resistant and sensitive disease (2). To date only ovarian cancer cell lines have been used to examine differentially expressed proteins, ovarian tumor tissue though seems an attractive alternative (3). The up or down regulation of these proteins is responsible for (a) an accelerated detoxification of drug substrate, (b) inhibition of apoptotic cell death, or (c) inhibiting

Graeme I. Murray (ed.), Laser Capture Microdissection: Methods and Protocols, Methods in Molecular Biology, vol. 755, DOI 10.1007/978-1-61779-163-5_12, © Springer Science+Business Media, LLC 2011

155

156

I. Cadron et al.

pathways leading to a decreased basal metabolism which helps cells to live through the duration of drug therapy (4–8). We therefore used the combination of laser microdissection (LMD) and surface-enhanced laser desorption/ionization-time of flight mass spectrometer (SELDI-TOF MS) in an attempt to identify protein profiles related to platinum resistance in ovarian cancer tissue. The advantage of laser microdissected cells is that you have a pure homogeneous cell population increasing the likelihood to find a disease-specific marker instead of a tumor-related marker due to its influence on the surrounding tissue. Several mass spectrometric techniques exist and there are several advantages of SELDI-TOF MS, being (1) you only need a small amount of sample, (2) due to the several wash steps on the array, salts (from buffers) are washed out causing less sample preparation, (3) high throughput, (4) detection of peptides in the low mass range, and (5 ) it is useful in clinical setting. In our study, we have shown that the combination of LMD and SELDI-TOF MS in ovarian cancer tissue was feasible and that differentially expressed proteins for platinum resistance can be found (9).

2. Materials 1. 2-Methyl butane, 99+% extra pure (cros Organics, Geel, Belgium; store at 4°C). 2. Liquid nitrogen. 3. Suitable embedding compound (Richard-Allan Scientific Neg50 frozen section medium, Prosan, Merelbeke, Belgium; store at room temperature). 4. Ethanol 75–96–100% (prepare fresh). 5. Microtome (Prosan, Merelbeke, Belgium). 6. LC-MS grade water (Biosolve, Valkenswaard, The Netherlands). 7. Hematoxylin stain (self-prepared, Merck Damstadt, Germany; store in dark). 8. CellCut plus system (Olympus – MMI, Hamburg, Germany). 9. Polyethylene terephthalate (PET) membrane, 1.4 mm, MMI (Olympus, Aartselaar, Belgium). 10. Tubes with adhesive lid, with diffuser 500 ml, MMI (Olympus, Aartselaar, Belgium). 11. U9 buffer (BioRad, Nazareth, Belgium; store at −20°C). 12. Nanosep MF device 0.2 mm (PALL Inc, Haasrode, Belgium). 13. Micromix5-shaker (DPC, UK). 14. Trifluoroacectic acid (TFA) (Sigma-Aldrich, Steinheim, Germany; store at 4°C and prepare fresh).

12  Proteomic Analysis of Laser Microdissected Ovarian Cancer¼

157

15. Acetonitrile (ACN) Biotech grade solvent 99.93+% (SigmaAldrich, Steinheim, Germany; store at room temperature). 16. IMAC30 buffer (0.1  M sodium phosphate, 0.5  M sodium chlorate) (BioRad, Nazareth, Belgium; store at 4°C). 17. CM10 buffer (0.1  M sodium acetate pH 4.0) (BioRad, Nazareth, Belgium; store at 4°C). 18. H50 buffer (10% ACN, 0.1% TFA) (prepare fresh). 19. Q10 buffer (Tris–HCl buffer (10–100  mM), pH 7.5–9) (BioRad, Nazareth, Belgium; store at 4°C). 20. EAM powder (SPA/CHCA) (BioRad, Nazareth, Belgium; light sensitive, store away from light; prepare fresh). 21. ProtienChip arrays (BioRad, Nazareth, Belgium; store at room temperature in a cupboard protected from moisture and direct exposure to light).

3. Methods 3.1. Collection of Tumor Specimens and Freezing of Tissue Samples

All samples were obtained after written informed consent and ethical approval from the local ethical committee. Samples were obtained during primary surgery for suspected or pathologically proven ovarian cancer, snap frozen in liquid nitrogen after prelevation (delay between prelevation and freezing is less than 30 min) and stored at −80°C until further processing. 1. Place 50 ml of isopentane in a pyrex or polypropylene beaker. 2. Immerse the beaker in a dewar or styrofoam container of liquid nitrogen. 3. Stir the isopentane until opalescent (about 2–3 min). 4. Remove the isopentane beaker from the liquid nitrogen. 5. Place OCT compound (or similar) on a specimen stub or in a cryomould and orientate the specimen within it. 6. Immerse the specimen into the cooled isopentane until frozen. 7. Place the frozen block into the cryostat for sectioning or store at −70°C until required. The specimen will require warming to its optimal cutting temperature before sectioning. 8. Biopsy is put on dry ice immediately and restored at −80°C in freezer until use.

3.2. Sectioning and Staining

Tissue samples were cut on a Prosan cryostat at −20°C and mounted on a glass slide or membrane slide depending on its use. Five micrometer sections were mounted on a glass slide and subsequently stained with hematoxylin and eosin for control and

158

I. Cadron et al.

tumor localization. Serial sections of 16 mm thickness were made for LMD and mounted on nuclease and human nucleic acid free membrane slides. These were then stained with hematoxylin only and air dried (see Note 1). Sectioning: 1. Remove and discard old microtome blade. Wipe down the knife holder and antiroll plate in the cryostat with 100% ethanol to avoid sample cross-contamination. 2. Install a new disposable microtome blade in the cryostat. 3. Set cutting thickness to 5 or 16 mm for fine cutting, 20 mm for trimming. Set specimen and knife temperature on −30°C. 4. Place a microslide box on dry ice near the cryostat to collect the mounted slides. 5. Transport the cryomould/cork with the frozen tissue from the −80°C freezer to the cryostat on dry ice. 6. Mount the specimen on the metal chucks with OCT and adjust the orientation of the cutting block so that the desired sections can be made. 7. Mount the sections on the (room temperature) PET-slides and place them immediately in the box on dry ice. 8. If cutting more than one specimen, use a new disposable microtome blade or use a new part of the blade for each one. In addition, wipe down knife holder and antiroll plate with 100% ethanol in between each specimen to avoid crosscontamination. 9. Proceed immediately to the “rapid hematoxylin procedure for frozen sections” or store at −80°C for up to 2 months. Staining: Pull slide from dry ice or −80°C storage and air dry slide for 30 s. Immediately proceed with: 10 s: Fixation in 70% alcohol. 10 s: Rinse in LC-MS grade water. 10 s: Stain in hematoxylin solution (Mayer’s). 10 s: Rinse in LC-MS grade water. 10 s: Dehydrate in 70% alcohol (see Note 2). 10 s: Dehydrate in 96% alcohol. 1 min: Dehydrate in 100% alcohol. Air dry. 3.3. Laser Microdissection

LMD was performed using the CellCut plus system (Olympus – MMI, Hamburg, Germany) (see Note 3). Briefly, a clear glass slide is pressed against the tissue on the membrane slide (sandwich

12  Proteomic Analysis of Laser Microdissected Ovarian Cancer¼

159

technique) and installed on the microscope. Areas of tumor cells are marked and cutting is performed. Cutting speed, focus, and laser energy were adjusted to ovarian cancer tissue to obtain a clear cut (see Note 4). Areas of necrosis, lymphocytic infiltration, and regions with psammoma bodies were avoided. After cutting, the specimen is picked up by the diffuser cap. In order to obtain a reliable protein profile on SELDI-TOF MS, we dissect 30,000 cells per cap (see Note 5). Once dissection is performed cells are stored at −80°C or we immediately proceed with protein extraction. 3.4. Lysis of Laser Captured Cells (see Note 6)

1. Activate fast cooling on centrifuge, thaw U9 buffer. 2. Remove sample from −80°C and pipette 50  ml of U9 lysis buffer into the microfuge tube. 3. Close cap – keep microfuge tube upside down at all times (see Note 7) and incubate at 4°C on Micromix5 for 60 min (settings: 20, 7, 60). 4. Revert microfuge tube and centrifuge at 5,000 rpm for 5 min at 4°C. 5. Add 100  ml of binding buffer (dependent on chip surface) and pipette 150 ml up and add to reservoir of Nanosep© MF device to remove gross debris. 6. Centrifuge device at 13,000 rpm for 3 min at 4°C. 7. Add 100 ml of binding buffer (dependent on chip surface) to reservoir of Nanosep© MF device. 8. Centrifuge at 13,000 rpm for 3 min at 4°C. 9. Remove the reservoir and close Nanosep© Tube. Store at −80°C until MS analysis.

3.5. SELDI-TOF MS Analysis

Protein lysates were analyzed on copper-coated immobilized metal affinity capture array (IMAC30), weak cation exchanger array (CM10), hydrophobic or reversed phase array (H50), and strong anion exchanger array (Q10). *Prepare energy absorbing matrix, e.g. CHCA 20% (see Note 8): 1. Tap the tube to ensure that contents fall to the bottom of the tube before reconstitution. 2. Add 100 ml of TFA and 100 ml of ACN 100%. 3. Vortex for 15 min. 4. Centrifuge at 10,000 rpm for 1 min at room temperature. 5. Take 50  ml of this solution in a new microfuge tube, add 100 ml of TFA 1% and 100 ml of ACN 100%. *Place arrays of the same type in a bioprocessor and handle them according to the array specific protocol.

160

I. Cadron et al.

3.5.1. IMAC 30 Array

1. Add 50 ml of 0.1 M copper sulfate and incubate for 5 min at room temperature on Micromix5 (setting 20/7/5). 2. Repeat once. 3. Remove previous and add 200 ml of ultrapure LC-MS grade water. Incubate for 2 min at room temperature on Micromix5 (setting 20/7/2). 4. Repeat once. 5. Remove previous and add 200  ml of neutralization buffer (0.1 M sodium acetate buffer pH 4). Incubate for 5 min at room temperature on Micromix5 (setting 20/7/5). 6. Remove previous and add 200 ml of ultrapure LC-MS grade water. Incubate for 2 min at room temperature on Micromix5 (setting 20/7/2). 7. Repeat once. 8. Remove previous and add 200 ml of binding buffer (=0.1 M sodium phosphate, 0.5 M NaCl pH 7) and incubate for 5 min at 4°C on Micromix5 (settings 20/7/5). 9. Repeat once. 10. Remove previous and add 100  ml of sample to each well, incubate for 1 h at 4°C on Micromix5 (settings 20/7/60). 11. Remove sample and add 200 ml of binding buffer (=0.1 M sodium phosphate, 0.5 M NaCl pH 7) and incubate for 5 min at 4°C on Micromix5 (settings 20/7/5). 12. Repeat twice. 13. Wash the chips twice with ultrapure LC-MS grade water and air dry the arrays for 15 min. 14. Apply 1 ml of energy absorbing matrix solution to each spot and air dry for 5 min. 15. Repeat once.

3.5.2. CM10 Array

1. Add 200 ml of CM10 binding buffer and incubate for 5 min at room temperature on Micromix5 (setting 20/7/5). 2. Repeat once. 3. Remove previous by shaking and add 100  ml of sample to each well, incubate for 1  h at 4°C on Micromix5 (settings 20/7/60). 4. Remove sample and add 200  ml of CM10 binding buffer, incubate for 5 min at 4°C on Micromix5 (settings 20/7/5). 5. Repeat twice. 6. Wash the chips twice with ultrapure LC-MS grade water and air dry the arrays for 15 min.

12  Proteomic Analysis of Laser Microdissected Ovarian Cancer¼

161

7. Apply 1 ml of energy absorbing matrix solution to each spot and air dry for 5 min. 8. Repeat once. 3.5.3. H50 Array

1. Prewash the chips by adding 50 ml of 50% ACN in ultrapure LC-MS grade water and incubate for 5 min at room temperature on Micromix5 (settings 20/7/5). 2. Repeat once. 3. Add 200  ml of H50 binding buffer, incubate for 5  min at room temperature on Micromix5 (settings 20/7/5). 4. Repeat once. 5. Remove previous by shaking and add 100  ml of sample to each well, incubate for 1  h at 4°C on Micromix5 (settings 20/7/60). 6. Remove sample and add 200 ml of H50 binding buffer, incubate for 5 min at 4°C on Micromix5 (settings 20/7/5). 7. Repeat twice. 8. Wash the chips twice with ultrapure LC-MS grade water and air dry the arrays for 15 min. 9. Apply 1 ml of energy absorbing matrix solution to each spot and air dry for 5 min. 10. Repeat once.

3.5.4. Q10 Array

1. Add 200 ml of Q10 binding buffer and incubate for 5 min at room temperature on Micromix5 (setting 20/7/5). 2. Repeat once. 3. Remove binding buffer by shaking and add 100 ml of sample to each well. Incubate for 1 h at 4°C on Micromix5 (setting 20/7/5). 4. Remove samples and add 200  ml of Q10 binding buffer. Incubate for 5 min at 4°C on Micromix5 (setting 20/7/5). 5. Repeat twice. 6. Wash the chips twice with ultrapure LC-MS grade water and air dry the arrays for 15 min. 7. Apply 1 ml of energy absorbing matrix solution to each spot and air dry for 5 min. 8. Repeat once.  nalyze the arrays using the protein chip reader of the SELDIA TOF MS.

162

I. Cadron et al.

4. Notes 1. Use the conventional H&E stained and glass mounted slide for tumor evaluation. Depending on the amount of tumor in the tissue sample you can then determine the number of slides needed on the membrane slide to obtain sufficient cells. 2. Dehydration is important for LMD as the transfer efficacy of the laser beam will decrease when the tissue is still moist and the presence of water within the tissue sample will not allow good capture of the cell. Proper dehydration entails the use of graded ethanol. A final 100% ethanol step is of great importance, but it will only be effective if 100% ethanol that has been freshly dispensed, is used. 3. The advantage of UV above infrared lasers is their accuracy in cutting. UV lasers have the possibility (due to the shorter wavelength) to focus in the submicron range. Initially, there were some uncertainties regarding the influence of UV lasers on the cellular macromolecules. However, several studies (10, 11) showed that the UV A range (320–400 nm) does not yield any damage to the nucleic acids or other biomolecules as their peak of absorption is outside this region. The heating caused by the laser does not cause changes at DNA, RNA, or protein level either. 4. Determining optimal conditions for cutting will highly depend on the structure and the thickness of the tissue used. Preferably use a test sample, which is equal to your study samples, on which you can try out the settings. Use the drawing method to make a meander throughout your tissue sample and pull down the speed as low as possible. While cutting adjust the focus and laser energy until you obtain a clear cut without burning margins. Once this has been determined, increase the cutting speed until satisfactory. It is possible you have to make small adjustments between samples in order to obtain a clear cut with pickup of the area after one cut. 5. Some authors (12–14) published results with LMD of very small amounts of cells. We were only able to identify reliable peaks on SELDI-TOF MS from 10,000 cells onward. This number, however, will highly depend on the consecutive methods used for protein extraction and analysis. 6. The use of complete protease inhibitor tablets is commonly used in proteomic studies. We do not use this routinely as some of these protease inhibiting peptides can mask peptides of interest in the study sample and compete with binding to the chip surface. Furthermore, we compared profiles using different lysis buffers and lysis conditions showing that this can cause alterations in the protein profile (9).

12  Proteomic Analysis of Laser Microdissected Ovarian Cancer¼

163

7. With the CellCut plus system laser microdissected cells are lifted on the cap of the tube, meaning that for lysis the tube needs to be reversed upside down. Initially, we had difficulties in finding an appropriate system to obtain the tubes upside down during shaking. Therefore, we use an Eppendorf pipettor tip rack that we turned upside down and covered with a sealing film to ± immobilize the tubes. 8. Bio-Rad supplies ProteinChip EAM as dried powder in a tube. The energy absorbing matrix will co-crystallize with the sample and yield peaks that do not interfere with sample spectrum. CHCA is preferably used for peptide analysis (1–15  kDa), while SPA is used for the higher molecular weight range. References 1. Heintz APM, Odicino F, Maisonneuve P, Quinn MA, Benedet JL and Creasman WT et  al. (2006) Carcinoma of the ovary. FIGO 6th Annual Report on the Results of Treatment in Gynaecological Cancer. Int J Gynaecol Obstet 95 (suppl. 1), S161–192. 2. Martin LP, Hamilton TC, Schilder RJ. (2008) Platinum resistance: the role of DNA repair pathways. Clin Cancer Res 14,1291–1295. 3. Cadron I, Van Gorp T, Timmerman D, Amant F, Waelkens E, Vergote I. (2009) Application of proteomics in ovarian cancer: which sample should be used? Gynecol Oncol 115,497–503. 4. Yan XD, Pan LY, Yuan Y, Lang JH, Mao N. (2007) Identification of platinum resistant associated proteins through proteomic analysis of human ovarian cancer cells and their platinum resistant sub lines. J Proteome Res 6,772–780. 5. Stewart JJ, White JT, Yan X, Collins S, Drescher CW, Urban ND, et al. (2006) Proteins associated with cisplatin resistance in ovarian cancer cells identified by quantitative proteomic technology and integrated with mRNA expression levels. Mol Cell Proteomics 5,433–443. 6. Chien J, Aletti G, Baldi A, Catalano V, Muretto P, Keeney GL, et  al. (2006) Serine proteaseHtrA1 modulates chemotherapy induced cytotoxicity. J Clin Invest 116,1994–2004. 7. Le Moguen K, Lincet H, Deslandes E, HubertRoux M, Lange C, Poulain L, et  al. (2006) Comparative proteomic analysis of cisplatin sensitive IGROV1 ovarian carcinoma cell line and its resistant counterpart IGROV1-R10. Proteomics 6,5183–5192.

8. Song J, Shih IeM, Salani R, Chan DW, Zhang Z. (2007) Annexin XI is associated with cisplatin resistance and related to tumor recurrence in ovarian cancer patients. Clin Cancer Res 13, 6842–6849. 9. Cadron I, Van Gorp T, Amant F, Vergote I, Moerman Ph, Waelkens E, Daemen A, Van de Plas R, De Moor B, Zeillinger R. (2009) The use of laser microdissection and SELDI-TOF MS in ovarian cancer tissue to identify protein profiles. Anticancer Res. 29, 1039–1045. 10. Schutze K, Clement-Sengewald A, Catch and move-cut or fuse. Nature 1994; 368:667–669. 11. Schultze K, Lahr G, Burgemeister R. (2000) The force of focused light in different areas of cell and molecular biology. J Mol Med 78, B21. 12. Melle C, Ernst G,Schimmel B, Bleul A, Thieme H, Kaufmann R, Mothes H, Settmacher U, Claussen U, Halbhuber KJ, Von Eggeling F. (2005) Discovery and identification of alfa defnsins as low abundant tumor derived serum markers in colorectal cancer. Gastroenterology 129, 66–73. 13. Diaz JL, Cazares LH, Corica A, Semmes OJ. (2004) Selective capture of prostatic basal cells and secretory epitheial cells for proteomic and genomic analysis. Urologic Oncology 329–336. 14. Kwapiszewska G, Meyer M, Bogumil R, Bohle RM, Seeger W, Weissmann N, Fink L. (2004) Identification of proteins in laser microdissected small cell numbers by SELDI-TOF and Tandem MS, BMC Biotechnology 4, 30.

wwwwwww

Chapter 13 LCM Assisted Biomarker Discovery from Archival Neoplastic Gastrointestinal Tissues Patricia A. Meitner and Murray B. Resnick Abstract Expression array analysis of epithelial mRNA to identify biomarkers of premalignant and malignant conditions in the gastrointestinal (GI) tract is an area of intense study. Archived formalin-fixed paraffinembedded (FFPE) tissues documenting these changes are readily available and should be a valuable resource for retrospective analysis. Laser capture microdissection of defined areas of epithelial cells at different stages of neoplastic progression is described together with methods for prequalification of RNA in FFPE tissue blocks selected for analysis. Paradise reagents specifically designed for isolation and amplification of RNA from FFPE archival tissue specimens are used to prepare probes for the human X3P microarray from Affymetrix. Key words: Laser capture microdissection, Formalin fixed paraffin embedded, Neoplasia, Gastrointestinal epithelium, RNA, Gene expression, DNA microarray, Real-time PCR

1. Introduction Most gastrointestinal epithelial cancers arise from preinvasive neoplastic precursor lesions. For example, the metaplastic condition Barrett’s esophagus progresses through low and high grade dysplasia to invasive adenocarcinoma. Laser capture microdissection (LCM) of neoplastic epithelium followed by RNA extraction, reverse transcription, and microarray analysis for expression of genes associated with the neoplastic transformation may lead to discovery of novel diagnostic and prognostic biomarkers. Ideally, RNA is best obtained from fresh frozen tissue since RNA in archived formalin-fixed paraffin-embedded (FFPE) blocks of tissue is subject to cross-linking and fragmentation (1, 2). In recent

Graeme I. Murray (ed.), Laser Capture Microdissection: Methods and Protocols, Methods in Molecular Biology, vol. 755, DOI 10.1007/978-1-61779-163-5_13, © Springer Science+Business Media, LLC 2011

165

166

P.A. Meitner and M.B. Resnick

years, many reports have focused on improving the yield and quality of RNA from FFPE tissues and success has allowed researchers to access the considerable and valuable archive of millions of tissue blocks typically stored in pathology archives indefinitely. These resources usually have valuable patient history and treatment outcomes available together with morphological and histopathological descriptions. Identification of diagnostic and prognostic gene signatures for various diseases including cancer has now been made utilizing RNA extracted from FFPE tissues by both RT-PCR (3–12) and global gene expression by microarray analysis (13–17). The need to control the fixation and extraction procedures or at least prequalify the extent of degradation of the extracted RNA before attempting gene array analysis is becoming clear (18–20). By using LCM, it is now possible to isolate RNA from pure populations of cells at different stages of disease progression from archived FFPE tissue blocks often from the same patient (Fig. 1). Since we first published our use of the Paradise system reagents (Molecular Devices Corp. Sunnyvale, CA) to extract RNA of sufficient quality for amplification and microarray analysis from laser captured cells (21) several other reports have appeared in the literature (22–28). We describe here reagents and protocols used successfully to laser capture microdissect cells from archived FFPE tissue and to extract RNA in order to query the human genome for expression of genes associated with disease progression in GI epithelium. The methodology utilizes the Paradise Reagent system from Molecular Devices and a custom expression array developed on the cDNA Affymetrix platform (U133-X3P). This array has a built in 3¢ bias of probe sets incorporated into the array design which can be interrogated with labeled product from the 3¢ ends

Fig. 1. Laser-capture microdissection of the esophageal epithelial cells. Esophageal epithelium stained with Histogene staining solution (a), marked for capture (b), after capture (c), and captured cells after microdissection (d). For LCM, the 6–7-mm sections are air-dried and not cover slipped resulting in poor optical quality of the photomicrographs.

13  LCM Assisted Biomarker Discovery from Archival Neoplastic Gastrointestinal Tissues

167

of expressed genes. Biased 3¢ ends of expressed genes are generated and amplified by one or two rounds of linear in vitro T7 bacteriophage oligo dT-driven transcription (29). Introduction of such a bias did not have a major impact on overall transcript profile provided the bias was present in the chip design also (30). We have found that with careful selection and prequalification of tissue blocks we can successfully access the genomic profile of gastrointestinal epithelial cells from FFPE archival tissue (21, 24).

2. Materials 2.1. Instrumentation

1. ArcturusXT LCM system (Molecular Devices Corp., Sunnyvale, CA). 2. Scanscope CS (Aperio Technologies Inc., Vista, CA). 3. 2100 Expert BioAnalyzer (Agilent Technologies, Santa Clara, CA). 4. UV/vis spectrophotometer for small volumes, e.g., Nanodrop 2000 (Thermo Fisher Scientific, Wilmington, DE) or Eppendorf BioPhotometer (Fisher Scientific, Pittsburg, PA). 5. Refrigerated microfuge capable of developing 16,000 × g. 6. Microtome capable of cutting 6 mm sections. 7. Stratagene MX4000 QPCR instrument (Stratagene/Agilent Technologies) or similar. 8. iCycler PCR instrument (Bio-Rad Laboratories, Hercules, CA) or similar.

2.2. Materials

1. Size ten sterile individually wrapped scalpel blades. 2. UltraPURE DNAse, RNAse-free distilled water (Invitrogen Corp., Carlesbad, CA), 3. RNaseZap spray and wipes (Ambion/Applied BioSystems, Austin, TX). 4. Snap top microfuge tubes, DNase RNAse-free, 1.5 and 0.5 ml. 5. Sterile blunt and needle-nose forceps. 6. Sterile nuclease-free filter tips 10, 20, 200, and 1,000  ml (Rainin Instruments, Oakland, CA). 7. Micropipets size 2, 10, 100, 200, and 1,000  ml (Rainin Instruments). 8. Nanochip and PicoChip kits for the 2100 Expert BioAnalyzer (Agilent Technologies). 9. For LCM using the ArcturusXT instrument you will need Capsure Macro and HS caps (Molecular Devices). 10. Molecular sieves (Sigma-Aldrich, Milwaukee, WI).

168

P.A. Meitner and M.B. Resnick

2.3. Slides and Stains for LCM

1. Histogene LCM staining kit (Molecular Devices). 2. Superfrost plain glass slides with clipped corners (Fisher Scientific, Pittsburg, PA). These slides fit into the plastic slide jars (Evergreen Scientific, Los Angeles, CA) supplied with the staining kit together with 50 ml tube racks from Fisher for holding the slide jars. 3. For UV laser cutting as well as IR LCM you can also use PEN membrane glass slides (Molecular Devices).

2.4. Kits for Extraction, Purification, and Post LCM Amplification for Gene Chip Analysis

1. Paradise FFPE system reagents for 1.5 rounds of amplification (Molecular Devices) (see Note 1). 2. Paradise Plus Reagent Quality Assessment kit (Molecular Devices). 3. Paradise Whole Transcript RT kit (Molecular Devices). 4. Superscript (tm) III (Invitrogen Corp.). 5. BioArray High Yield RNA Transcript Labeling kit (Enzo Life Sci. Inc., Farmingdale, NY). 6. RNase-Free DNase (Qiagen Inc., Valencia, CA). 7. Brilliant SYBR Green master mix reagents (Stratagene/ Agilent).

3. Methods 3.1. Selection of Study Cases and Pathologist Collaboration

For LCM studies it is best to have the close collaboration of a pathologist or a person with histological expertise, available to review and identify cells or regions of interest in the tissue sections. Because of the degradation of RNA in fixed tissues one must capture up to ten times more cells to yield the same amount of RNA as can be isolated from fresh frozen tissues. Projects in which biopsy material is studied require multiple sections for LCM. Three-dimensional patterns of cells of interest change as one cuts through the paraffin block so it is essential to review the sections before proceeding.

3.2. Prequalification of Samples for Laser Capture and Microarray Analysis

It is important to observe RNA precautions throughout these and subsequent procedures as RNA can be degraded at any time by exposure to nucleases ubiquitously present on all surfaces (see Note 2). The procedures from laser capture through extraction, transcription, amplification, and chip analysis or PCR are time consuming and the reagents are expensive so it is imperative to have good quality starting material. The following preliminary tests should be performed on the FFPE tissue blocks in order to prequalify samples for chip analysis and to increase the likelihood of a successful outcome (see Note 3).

13  LCM Assisted Biomarker Discovery from Archival Neoplastic Gastrointestinal Tissues 3.2.1. Check for Presence of RNA

169

Because of the potential for RNA degradation in routinely collected and processed FFPE tissues it is important to check that the blocks contain total RNA (see Note 4). After discarding the first 30  mm of sections, cut sufficient 6  mm sections to yield 100 mm2 tissue from each case. Stain using the procedure selected for visualizing the cells for laser capture (see Subheading 3.3.1). Using a scalpel blade, scrape the stained tissue into a 0.5-ml microfuge tube. Extract and purify the total RNA using a Paradise Reagent Quality Assessment kit (Molecular Devices). Remove any remaining genomic DNA by on column treatment with RNAse-Free DNAse and quantify the RNA by 2100 Bioanalyzer using an RNA 6000 Nano LabChip kit. Reject any block that yields <500  ng RNA/100  mm2 (see Note 5). FFPE-degraded RNA often has a RIN of 5 or less by Bioanalyzer analysis (Fig. 2). Continue to the next step to assess the RNA degradation in your sample. a

RIN = 9.6

Gel view

RIN = 5.3 3’:5’= 4.6

b

Gel view

RIN = 1.8 3’:5’=106

c

Gel view Fig. 2. Evaluation of total RNA by 2100 Bioanalyzer. (a) Ideal trace showing 18S and 28S peaks of ribosomal RNA. (b) Typical trace obtained with FFPE-derived RNA shows a shift to shorter fragments with a reduction in ribosomal peaks. The gel view often appears as a smear. Based on the 3¢:5¢ ratio this RNA is acceptable for LCM. (c) Trace showing highly fragmented RNA with disappearance of ribosomal peaks. This RNA is not acceptable for LCM.

170

P.A. Meitner and M.B. Resnick

3.2.2. Check for RNA Degradation

Continue with the Paradise Reagent Quality Assessment kit protocol and perform the first strand synthesis. Check for degradation by performing real-time PCR using primers for the 3¢ and 5¢ ends of b actin as recommended in the Paradise protocol (see Note 6). We found the RNA to be acceptable for subsequent analysis if the 3¢ ends to 5¢ ends ratio for b actin < 15. If you are using pairs of blocks (normal and diseased) then both members of the pair must be acceptable. The RNA generated at this step can be retained for subsequent gene confirmation by QPCR. Pairs of blocks that pass these QC measures can now be used for LCM (see Note 7).

3.3. Laser Capture Microdissection

The Paradise FFPE Reagent System protocol should be followed without modification. The users’ manual is available online (see Note 1). The protocol has three sections: preparation and staining of tissue sections for LCM (see Subheading 3.3.1); RNA extraction, isolation, and purification from laser captured cells (see Subheading  3.3.3); and amplification of RNA for microarray analysis (see Subheading 3.3.4). Reagents are provided in the kit for each section. Subheading 3.3.2 describes the LCM procedures we use with the kit.

3.3.1. Preparation and Staining of Tissue Sections for Laser Capture Microdissection

1. Cut 6–7 mm sections (see Note 8) onto plain clipped corner glass slides (see Note 9), air dry for 1  h and hold at room temperature in a dessicator (no vacuum) or slide box over Drierite for up to 5 days prior to staining. 2. Stain under RNA-free conditions with Histogene staining solution. 3. Dehydrate through graded alcohols and xylene (Table 1) (see Note 10). 4. At this point one of the slides can be coverslipped and scanned by Aperio Scanscope for final review by the pathologist at his/ her desktop monitor prior to microdissection. The Scanscope allows transmission of images via the internet and should be located adjacent to the LCM instrument. Images can be accessed on the pathologist’s office computer at a remote location where, when necessary, they can be annotated to indicate the exact location of the cells for dissection (Fig. 3). These images can then be archived as part of the experimental documentation. 5. Hold the remaining stained slides in xylene for up to 1.5 h prior to LCM.

3.3.2. Microdissection of Cells of Interest

The interface of the ArcturusXT is very intuitive and will be described briefly as the instrument at your disposal may be different. Microdissection is accomplished by placing a CapSure cap bearing a layer of thermolabile transfer film over the tissue of interest and

13  LCM Assisted Biomarker Discovery from Archival Neoplastic Gastrointestinal Tissues

171

Table 1 Dehydration and staining for LCM Xylene

2 min

Xylene

2 min

100% ethanol

2 min

95% ethanol

1 min

75% ethanol

1 min

Water

<30 s

Drain but do not dry Add three drops of Histogene Stain to the wet slide; be sure to cover the entire section Stain for 15 s to a light stain Water

<30 s

75% ethanol

1 min

95% ethanol

1 min

Anhydrous ethanol

1 min

Xylene

2 min

Xylene

2 min

Air dry

5 min

If you are using the PEN membrane slides for LCM, keep the exposure to xylenes to less than 2 min per rinse or the membrane will separate from the glass

using the instrument’s lasers to melt the film into the tissue. The cells are then lifted together with the cap for further procedures. The ArcturusXT instrument is constructed on a Nikon Eclipse Ti-E inverted research microscope platform with a computercontrolled robotic drive mechanism for moving the stage and placing the Capsure cap on the slide. The cells are visualized by CCD camera on an LCD screen and the slide is oriented such that the lasers can be precisely fired at the cells. The ArcturusXT has two lasers: one is a low energy solid-state near IR laser (810 nm) which melts the film into the surface of the tissue and is used for dissecting individual types of cells or single cells, the other is a high energy solid-state, passive Q-switched, diode-pumped UV laser (355  nm) which cuts around the area of interest and can then be adhered to the cap and removed in its entirety. When using the high energy UV laser bear in mind that RNA in cells adjacent to the laser cut may be degraded by the laser so use caution when dissecting areas of less than 30 cells. We have found it best

172

P.A. Meitner and M.B. Resnick

Fig. 3. Aperio scans annotated to identify areas for microdissection. In the clinical setting, multiple small biopsies obtained from different sites along the esophagus are placed in the same paraffin block for sectioning onto a single slide. When diseased areas merge into normal epithelium the practiced eye of the pathologist is required to define which areas should be microdissected from which biopsy. The Barrett’s epithelium slide contained six biopsies while the HGD slide above contained 15 biopsies only one of which was high grade dysplasia.

to deparaffinize and stain no more than three slides at a time for microdissection. 1. Remove the slides from the xylene and air dry for 5  min. Incomplete removal of the xylene will melt all of the transfer film onto the slide. 2. Place caps and slides on the stage of the instrument and enter specimen data into the software. 3. Locate an appropriate area for dissection and place a Capsure cap on the slide using the software controls. It is preferable to use the IR laser to dissect out the epithelial layer of cells as it is less than 20 cells thick and cutting around it with the UV laser might damage the RNA. 4. Test fire the IR laser on an area of the slide adjacent to the area you want to capture. Using the software, adjust the energy and duration of the laser to give a melted area of 20 mm diameter. The ArcturusXT is capable of retaining three different settings for laser melt size which can be toggled by a click of the mouse. Sometimes the power needed to melt the film varies across the entire cap surface and switching between power levels facilitates the capture. 5. Capture about 5,000 cells of interest on LCM Capsure caps (see Note 11). Figure 4 shows a 2× image of an HS Capsure

13  LCM Assisted Biomarker Discovery from Archival Neoplastic Gastrointestinal Tissues

173

Fig. 4. HS Capsure cap with captured cells. Approximately 5,000 cells captured from multiple sections of several GI epithelium biopsies.

cap containing about 5,000 cells after multiple captures from several biopsies. For successful multiple captures on the same cap avoid placing film with adhered cells over the new area you wish to capture. 3.3.3. RNA Extraction, Isolation, and Purification of Total RNA from Laser Captured Cells

1. Using fine needle-nose forceps peel and transfer the film with the captured cells from the Capsure cap to a 0.5-ml microfuge tube containing 50 ml proteinase K reagent from the Paradise Reagent System kit. It is important to transfer the captured cells to the proteinase reagent as soon as possible. You may combine film from two caps in one extraction tube if needed, however, do not hold all caps until the end of the session before adding the reagent. Rather set up microfuge tubes containing 50 ml aliquots of proteinase K reagent in a rack by the LCM and transfer film to each tube as the cells are collected. Seal the tube with Parafilm and digest overnight (16 h) in a 37°C water bath. We do not recommend freezing the extract at this point; rather continue to the next step. 2. Follow the kit instructions to isolate the total RNA by MiraCol spin column. Include an on-column DNAse digestion to remove genomic DNA as directed. 3. Elute from the column with 12 ml elution buffer (EB) from the kit. At this point you may store the RNA at −80°C.

3.3.4. Amplification of RNA for Microarray Analysis

1. The yield of laser captured RNA is <500 ng so the RNA must be amplified. Several micrograms of labeled RNA are required for hybridization to Affymatrix microarrays. The Paradise

174

P.A. Meitner and M.B. Resnick

system provides comprehensive protocols and reagents take your RNA through one and a half rounds of linear amplification using T7 bacteriophage RNA polymerase-driven in vitro transcription. Until you are completely familiar with the protocol we recommend you handle no more than four samples at a time. 2. The final amplification and labeling of the dsDNA product can be accomplished using an Enzo® BioArray™ HighYield™ RNA Transcript Labeling Kit from Enzo Life Sciences. Labeled cRNA should be purified using reagents from the RNA cleaned up module of the Paradise kit above. Quantify the RNA yield by Agilent Bioanalyzer as before (see Subheading 3.2.1). The yield of labeled cRNA should be at least 10 mg. Poorer values indicate that the quality of starting material was compromised and we cannot over emphasize the need for initially checking these measures. We have used the Affymetrix GeneChip Resource at the W.M. Keck Foundation Biotechnology Resource Laboratory at Yale University, New Haven, CT, for fragmentation and hybridization. Labeled cRNA is sent on dry ice over night. We purchase the custom gene chips (Human U133_X3P expression array Affymetrix) originally designed by Arcturus Engineering (now Molecular Devices) to complement the Paradise kit product. We provide the arrays to the Keck lab for the hybridization by standard procedures (protocols are available on the Keck Center web site at http://keck.med.yale.edu/microarrays). 3.4. Confirmation of Microarray Results by Real-time PCR

When starting with partially degraded RNA derived from FFPE tissue it is of particular importance to confirm the gene chip results by QPCR. 1. Confirm your GOIs in aliquots of the RNA amplified for microarray analysis (see Subheading  3.3.4). You also may want to extract more samples to increase the total number of cases evaluated (see Note 12). 2. Where ever possible, design your gene-specific primers for realtime PCR to span an intron in order to rule out artifacts from genomic DNA contamination. Your primers should amplify about 100 bp from within 400 bases of the 3¢ end of the GOI. Rational is that the Paradise kit employs oligo dT priming for first strand synthesis and formalin-fixed RNA is often fragmented into <400 base fragments. Our lab protocol is to design gene-specific primers using Primer3 shareware and have them synthesized by Operon Technologies, Huntsville, AL. 3. We performed real-time PCR on an MX4000 real-time instrument using Brilliant SYBR Green Master Mix reagents according to the manufacturer’s instructions with the exception that the reaction volume is reduced to 25 ml.

13  LCM Assisted Biomarker Discovery from Archival Neoplastic Gastrointestinal Tissues

175

4. Perform reactions in parallel using primers annealing to the 3¢ end of human b actin gene. Use these values to normalize all your GOIs to the b actin gene. Amplification conditions for each GOI and its primers are acceptable when efficiencies are better than 90% and linear regression coefficients (RSq) are >0.990. 5. In order to directly compare between PCR runs, include a calibration (standard) curve with each run. Construct the curve from b actin amplified from serial 10× dilutions of cDNA reverse transcribed from Stratagene Reference RNA (store concentrated aliquots of this reference cDNA to avoid freeze and thaw cycles; prepare working dilutions for each PCR run). This curve is used to relate the threshold cycle to the log input amount of template used and to determine relative amounts of GOI transcripts. 6. Run your samples in duplicate. 7. Carry out the thermocycling for 45 cycles, with denaturation at 95°C for 30 s, annealing for 1 min at 55°C, and extension at 72°C for 1 min. 8. Include a dissociation temperature gradient at the end of each run to confirm the absence of high molecular weight DNA and primer dimers. In our study of Barrett’s esophagus (BE), genes of interest (GOI) were selected for confirmation from those showing a >2.5fold difference between normal BE and high grade dysplasia. Figure 5 shows an example of how our microarray results correlated with those of QPCR.

Fig. 5. Expression analysis of Barrett’s Esophagus associated high grade dysplasia. Confirmation of the expression levels of selected dysregulated genes by RT-PCR and correlation with microarray data (reproduced from ref. (24) with permission).

176

P.A. Meitner and M.B. Resnick

4. Notes 1. The complete manual for the Paradise FFPE system can be found at the Molecular Devices web site. The URL for the Paradise manual is http://www.moleculardevices.com/product_ literature/family_links.php?prodid=130. Going forward the best way to locate the manual will be to search the web on ArcturusXT. This will bring up links to the Arcturus product line of instruments and associated kits and reagents. Since we have been using these products the company producing the Arcturus line of LCM instruments has been a subsidiary of several parent companies. The manual is extensive (150 pages) and contains complete instructions for the successful extraction, isolation, and amplification for QPCR and cDNA chip analysis. 2. Wear disposable gloves at all times and change them frequently. Clean all surfaces with RNaseZap wipes including forceps, the bath used for floating paraffin sections, and the bench next to the LCM. If possible, retain a bath and microtome specifically for RNase-free procedures. Change the microtome blade with every block. Use nuclease-free (RNAse-free) plastics, tubes, pipet tips, and all other reagents. It is preferable to purchase nuclease-free water rather than use DEPC-treated water as DEPC interferes with many down-stream reactions. 3. You may find with experience that the degree of RNA degradation will be similar in all blocks taken at the same time from the same patient. This is probably due to the fact that all biopsies were collected under the same conditions and that the time of fixation was the same for all tissues in the series. In which case you may decide it is unnecessary to check all blocks for RNA integrity at this point. 4. Remove and discard at least the first 25 mm of the block surface to avoid contamination by RNases resulting from handling of the blocks. Many other factors affect the RNA integrity in FFPE blocks. Usually, it is not possible to control the way tissues are collected, fixed, and embedded in the clinical setting. Extended time of fixation is the most detrimental to the RNA integrity and if at all possible some attempt should be made to minimize this. Extended time lapse before fixation also affects RNA integrity. The longer the blocks are stored, the more the RNA will degrade and you should try to choose blocks that have been stored for less than 5 years. 5. After laser capture of FFPE cells, the amount of RNA available is so small that it cannot be quantified by absorption at 260 nm in a spectrophotometer. Use a PicoChip and the Bioanalyzer

13  LCM Assisted Biomarker Discovery from Archival Neoplastic Gastrointestinal Tissues

177

2100 Expert to quantify as well as to evaluate your RNA. The Paradise linear amplification protocol calls for 10 ng starting material. We have found better results with 15–20 ng RNA. 6. RNA integrity is often determined by assessing the 3¢:5¢ ratio of a specific transcript. We use the following two sets of b-actin specific primers, one set of primers close to the 3¢ end of the message, and the other set closer to the 5¢ end in order to amplify cDNA. 3¢ bACT-F TCCCCCAACTTGAGATGTATGAAG 3¢ bACT-R AACTGGTCTCAAGTCAGTGTACAGG 5¢ bACT-F ATCCCCCAAAGTTCACAATG 5¢ bACT-R GTGGCTTTTAGGATGGCAAG If the RNA is degraded most of the polyA tail bearing RNA from the 3¢ end will be shortened. Since we use oligo dT for our reverse transcription there will be many more short transcripts than longer ones. Thus, the ratio of the 3¢–5¢ amplified fragments provides a direct indication of RNA degradation. 7. This prequalification procedure should be followed for frozen tissue also, except that RNA may be extracted after laser capture by PicoPure RNA extraction kit reagents also from Molecular Devices. 8. For successful laser capture there must be no wrinkles in the section on the slide which could result in the cap being slanted above the tissue. Melting of the film into the cells depends upon the entire surface of the cap being in intimate contact with the tissue. 9. Before starting a study with a specific tissue type determine the type of slide to use. Different tissues adhere more or less to the slide. We have found that untreated plain glass slides work best for GI mucosa. Do not use charged Plus slides and do not bake the tissue onto the slide. You can use the “Scotch tape” test to see if your tissue is too tightly bound to the glass. Place the tape over your section and press gently. The tissue will lift off the slide if it is available for LCM. 10. It is important to use anhydrous 100% ethanol for the final alcohol rinse. Add several grams of Molecular Sieves (SigmaAldrich) to the reagent bottle and change the working reagent daily. Change the xylene in the first wash after every four slides. Successful laser capture by IR laser and adherence to transfer film is dependent upon complete desiccation of the tissue. Any moisture will interfere. We have found that laser capture on humid days is difficult for this reason and if you have this problem we recommend locating the instrument in a small room with a dehumidifier. 11. Two types of capture caps are available for use with the Arc-turusXT instrument. The manufacturer recommends Capsure Macro caps

178

P.A. Meitner and M.B. Resnick

for capturing large numbers (>5,000) of cells. The entire surface of the transfer film on the Macro cap is available for use. However, the surface of the film is in direct contact with the tissue on the slide and sometimes results in nonspecific adherence of loose fragments or dirt from the slide to the cap. The HS caps on the other hand have spacer bars that hold the cap above the tissue and are recommended for capturing <100 cells. We have found that a cleaner capture of the large numbers of cells required for FFPE extraction is obtained using HS caps and the Macro cap setting on the ArcturusXT instrument so that the entire cap area is available for cell capture. 12. Often there is not enough amplified product after expression array analysis to use for confirmation of a large number of GOIs by QPCR a later time. If the FFPE-RNA passed the 3¢5¢ QC test (see Subheading 3) the fragmentation of RNA should be minimal and since QPCR amplicons are usually £100 bp, we have found that for QPCR at a later time you can substitute random hexamers for the oligo dTs in order to generate more cDNA template. The Paradise whole transcript RT kit from Molecular Devices works well to extract RNA from cells from LCM as well as the whole sections of tissue it is designed for. For whole sections, cut several 15 mm sections (rolls) from your block into a 1.5-ml microfuge tube. Remove the paraffin by two washes in xylene followed by one wash in 75% ethanol. Remove excess solvent from the pellet with a clean pipet tip and invert the tube to dry for 10 min. Proceed to digest with proteinase K and follow the kit protocol.

Acknowledgments This work was supported by NIH grant P20RR17695, awarded by the National Center for Research Resources, Institutional Development Award Program. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Center For Research Resources or the National Institutes of Health. References 1. Krafft AE, Duncan BW, Bijwaard KE et  al (1997) Optimization of the isolation and amplification of RNA from formalin-fixed, paraffin-embedded tissue: The Armed Forces Institute of Pathology Experience and Literature Review. Mol Diagn 2:217–230

2. Masuda N, Ohnishi T, Kawamoto S et  al (1999) Analysis of chemical modification of RNA from formalin-fixed samples and optimization of molecular biology application for such samples. Nucleic Acids Res 27:4436–4443

13  LCM Assisted Biomarker Discovery from Archival Neoplastic Gastrointestinal Tissues 3. Specht K, Richter T, Müller U et  al (2001) Quantitative gene expression analysis in microdissected archival formalin-fixed and paraffin-embedded tumor tissue. Am J Pathol 158:419–429 4. Cronin M, Pho M, Dutta D et  al (2004) Measurement of gene expression in archival paraffin-embedded tissues: development and performance of a 92-gene reverse transcriptase-polymerase chain reaction assay. Am J Pathol 164:35–42 5. Ma, XJ, Wang, Z, Ryan PD et  al (2004) A two-gene expression ratio predicts clinical outcome in breast cancer patients treated with tamoxifen. Cancer Cell 5:607–616 6. Gloghini A, Canal B, Klein U et  al (2004) RT-PCR analysis of RNA extracted from Bouin-fixed and paraffin-embedded lymphoid tissues. J Mol Diagn 6:290–296 7. Gianni L, Zambetti M, Clark K et al (2005) Gene expression profiles in paraffin-embedded core biopsy tissue predict response to chemotherapy in women with locally advanced breast cancer. J Clin Oncol 23:7265–7277 8. Makino H, Uetake H, Danenberg K et  al (2008) Efficacy of laser capture microdissection plus RT-PCR technique in analyzing gene expression levels in human gastric cancer and colon cancer. BMC Cancer 8:210 9. Jiang R, Gu X, Nathan CO et al (2008) Lasercapture microdissection of oropharyngeal epithelium indicates restriction of Epstein-Barr virus receptor/CD21 mRNA to tonsil epithelial cells. J Oral Pathol Med 37:626–633 10. DeCarlo CA, Escott NG, Werner J et  al (2008) Gene expression analysis of interferon kappa in laser capture microdissected cervical epithelium. Anal Biochem 381:59–66 11. Theophile K, Jonigk D, Kreipe H et al (2008) Amplification of mRNA from laser-microdissected single or clustered cells in formalin-fixed and paraffin-embedded tissues for application in quantitative real-time PCR. Diagn Mol Pathol 17:101–106 12. Kaneko T, Okiji T, Kaneko R et  al (2009) Gene expression analysis of immunostained endothelial cells isolated from formaldehydefixated paraffin embedded tumors using laser capture microdissection--a technical report. Microsc Res Tech 72):908–912 13. Onken MD, Worley LA, Ehlers JP et al (2004) Gene expression profiling in unveal melanoma reveals two molecular classes and predicts metastatic death. Cancer Res 64:7205–7209 14. Chung CH, Parker JS, Ely K et  al (2006) Gene expression profiles identify epithelial-tomesenchymal transition and activation of nuclear factor-(kappa)B signaling as characteristics of

179

a high-risk head and neck squamous cell carcinoma. Cancer Res 66:8210–8218 15. Scicchitano MS, Dalmas DA, Bertiaux MA et al (2006) Preliminary comparison of quantity, quality, and microarray performance of RNA extracted from formalin-fixed, paraffinembedded, and unfixed frozen tissue samples. J Histochem Cytochem 54:1229–1237 16. Haque T, Faury D, Albrecht S et  al (2007) Gene expression profiling from formalin-fixed paraffin-embedded tumors of pediatric glioblastoma. Clin Cancer Res 13:6284–6292 17. Frank M, Döring C, Metzler D et al (2007) Global gene expression profiling of formalinfixed paraffin-embedded tumor samples: a comparison to snap-frozen material using oligonucleotide microarrays. Virchows Arch 450:699–711 18. Penland SK, Keku TO, Torrice C et al (2007) RNA expression analysis of formalin-fixed paraffin-embedded tumors. Lab Invest 87:383–391 19. Linton KM, Hey Y, Saunders E et al (2008) Acquisition of biologically relevant gene expression data by Affymetrix microarray analysis of archival formalin-fixed paraffin-embedded tumours. Br J Cancer 98:1403–1414 20. Fedorowicz G, Guerrero S, Wu TD et  al (2009) Microarray analysis of RNA extracted from formalin-fixed, paraffin-embedded and matched fresh-frozen ovarian adenocarcinomas. BMC Med Genomics 2:23 21. Resnick MB, Sabo E, Meitner PA et al (2006) Global analysis of the human gastric epithelial transcriptome altered by Helicobacter pylori eradication in vivo. Gut 55:1717–1724 22. Coudry RA, Meireles SI, Stoyanova R et  al (2007) Successful application of microarray technology to microdissected formalin-fixed, paraffin-embedded tissue. J Mol Diagn 9:70–79 23. George MD, Wehkamp J, Kays RJ et al (2008) In vivo gene expression profiling of human intestinal epithelial cells: analysis by laser microdissection of formalin fixed tissues. BMC Genomics 9:209 24. Sabo E, Meitner PA, Tavares R et  al (2008) Expression Analysis of Barrett’s Esophagus Associated High Grade Dysplasia in Laser Capture Microdissected Archival Tissue. Clin Cancer Res 14:6440–6448 25. Rogerson L, Darby S, Jabbar T et al (2008) Application of transcript profiling in formalin-fixed paraffin-embedded diagnostic prostate cancer needle biopsies. BJU Int 102:364–370 26. Lassmann S, Kreutz C, Schoepflin A et  al (2009) A novel approach for reliable microarray

180

P.A. Meitner and M.B. Resnick

analysis of microdissected tumor cells from formalin-fixed and paraffin-embedded colorectal cancer resection specimens. J Mol Med 87: 211–224 27. Koh SS, Opel ML, Wei JP et  al (2009) Molecular classification of melanomas and nevi using gene expression microarray signatures and formalin-fixed and paraffin-embedded tissue. Mod Pathol 22:538–546 8. Calicchio ML, Collins T, Kozakewich HP 2 (2009) Identification of signaling systems in proliferating and involuting phase infantile hemangiomas by genome-wide tran-

scriptional profiling. Am J Pathol 174: 1638–1649 29. Upson JJ, Stoyanova R, Cooper HS et  al (2004) Optimized procedures for microarray analysis of histological specimens processed by laser capture microdissection. J Cell Physiol 201:366–373 30. Luzzi V, Mahadevappa M, Raja R et al (2003) Accurate and reproducible gene expression profiles from laser capture microdissection, transcript amplification, and high density oligonucleotide microarray analysis. J Mol Diagn 5:9–14

Chapter 14 Purification of Diseased Cells from Barrett’s Esophagus and Related Lesions by Laser Capture Microdissection Masood A. Shammas and Manjula Y. Rao Abstract Barrett’s esophageal adenocarcinoma (BEAC) arises from Barrett’s esophagus (BE), a premalignant lesion caused by acid reflux (heartburn). Although the cancer is uncommon, its incidence is rapidly rising in western countries. Like most other cancers, BEAC cells also have elevated telomerase activity which maintains telomere length and supports continued proliferation of these cells. It is not clear if telomerase is activated early at premalignant (BE) stage, because reports of telomerase activity in Barrett’s and normal esophagi have been controversial. We have shown that detection of telomerase and telomeres becomes easier and much more reliable if purified BE cells are used instead of tissue specimens. This chapter, therefore, emphasizes the importance of laser capture microdissection and provides the method to purify Barrett’s esophagus related cells, using this technique. Key words: Laser microdissection, Barrett’s esophagus, Esophageal adenocarcinoma, Esophageal metaplasia, Telomere, Telomerase

1. Introduction Cancer cells acquire diverse characteristics that determine their biological behavior. However, telomere maintenance and avoidance of apoptosis, constitute the life line of cancer cells. Barrett’s esophageal adenocarcinoma (BEAC) is one of the uncommon cancers whose incidence has been rapidly increasing and in white men has been increased almost 350% since the mid-1970s (1). The reason for this increased incidence is not known. Gastroesophageal reflux leads to the development of Barrett’s esophagus, a premalignant condition in which normal esophageal squamous epithelium is replaced by metaplastic columnar epithelium (2). Over time, the specialized intestinal metaplasia (SIM) of

Graeme I. Murray (ed.), Laser Capture Microdissection: Methods and Protocols, Methods in Molecular Biology, vol. 755, DOI 10.1007/978-1-61779-163-5_14, © Springer Science+Business Media, LLC 2011

181

182

M.A. Shammas and M.Y. Rao

Barrett’s esophagus progresses through worsening stages of dysplasia to adenocarcinoma (2). The cancer spreads early, before the onset of clinical symptoms, and therefore has a depressing outcome with a poor survival rate (3). A major interest of our laboratory is to investigate the roles of telomerase, telomeres, and homologous recombination in the etiology of Barrett’s adenocarcinoma. The expression of RNA component of telomerase (hTR) is elevated in Barrett’s esophagus and further increased in high-grade dysplasia (HGD) and adenocarcinoma (4). Consistently, the transcript levels of catalytic subunit of telomerase (hTERT) are also high in Barrett’s esophagus and higher in (HGD) and Barrett’s adenocarcinoma (5). These studies indicate that telomerase is probably activated early at Barrett’s stage during development of BEAC. However, the expression of hTR or hTERT do not measure telomerase activity, the ability to elongate telomeres, which depends on a number of factors including hTERT phosphorylation, its assembly into a holoenzyme, and interaction with proteins such as p23 and hsp90 (6, 7). Telomerase activity, the ability of telomerase to add “TTAGGG” sequences to the ends of chromosomes, as assessed by TRAP (Telomere Repeat Amplification Protocol) assay, is clearly shown to be elevated in dysplasia and esophageal adenocarcinoma (8, 9). However, controvercial and confusing results of telomerase activity in normal and Barrett’s esophagi have been reported. Yoneyama et al. (9) did not detect telomerase activity in normal esophagi whereas Bachor et al. (10) reported detection of telomerase activity in normal esophagi. Although a number of studies indicate that telomerase is active in Barrett’s esophagus (8, 10, 11), change in telomerase activity in BE relative to normal esophagi has not been observed (8, 10). The controversy of telomerase activity in normal mucosa and precancerous lesions of esophagus, which is less in biopsies and more in surgical specimens, probably arose because tissue extracts, containing varying proportions of contaminating connective tissue cells were used for these studies (10). Unwanted cells, which in some cases may represent a large fraction of the extract, dilute the lysate, and may also contribute factor/s to alter the activity being evaluated. Consistently, it has been reported that detection of telomerase activity in human skin becomes easier if dermis is removed from epidermal cells (12). Similar complexities are expected in genomics, proteomics, and expression analyses, conducted in tissue specimens. Laser capture microdissection can be used to purify specific diseased cells, suitable for subsequent analyses of DNA, RNA, and enzymatic activity (13). Collaud et al. (2010) have evaluated the impact of various staining techniques on purification of diseased cells by LCM and propose that staining with methyl green is more appropriate for subsequent activity-based protein profiling (14). We compared telomerase activity in tissue extracts and epithelial cells purified from normal and Barrett’s

183

14  Purification of Diseased Cells from Barrett’s Esophagus and Related Lesions…

Fig. 1. Isolation of esophageal epithelial cells by LCM. Panel I shows epithelial cells in a microscopic field before LCM; panel II is the same microscopic field after LCM and shows that epithelial cells have been removed; panel III shows the captured epithelial cells.

Shammas et al. 2008 Tissue Extracts

Telomerase Activity (TPG Units)

250

Cells Isolated By Laser Capture

200

Barrett’s

150 100 50

0 Lanes:

Normal 1

Barrett’s

Barrett’s 0.1X

2

3

Barrett’s 0.1X

Normal

4

5

6

Fig.  2. Telomerase activity in tissue extract vs. LCM purified cells. Three normal and three BE surgical specimens were processed for the evaluation of telomerase activity in LCM and tissue extracts. Each surgical specimen was cut into two portions; one processed for LCM purification of epithelial cells and the other used for making tissue extract. Telomerase activity is shown in tissue extract, equivalent of 0.6 mg protein, of normal esophagus (lane 1); tissue extract, equivalent of 0.6  mg protein, of Barrett’s esophagus (lane 2 ); diluted tissue extract, equivalent of 0.06 mg protein, of Barrett’s esophagus (lane 3 ); lysate of LCM purified normal esophageal epithelial cells, equivalent of 0.6 mg protein (lane 4 ); lysate of LCM purified Barrett’s esophagus cells, equivalent of 0.6  mg protein (lane 5 ); diluted lysate, equivalent of 0.06  mg protein, of LCM purified Barrett’s esophagus cells (lane 6).

esophagi, using LCM (Fig. 1) (15). Using the same amount of protein, we showed that telomerase activity is ~tenfold higher in LCM purified cells relative to tissue extracts of the same samples (Fig. 2). Mucosal tissue extracts are contaminated by a variety of cell types including smooth muscle, connective tissue, nerve fibers, and blood vessels, which may substantially dilute telomerase activity. The low telomerase activity detected in these extracts could also be attributed to possible inhibitors of telomerase or PCR. If this

184

M.A. Shammas and M.Y. Rao

is true, then dilution of the extracts should increase telomerase activity. However, the dilution of the samples did not increase but further reduced telomerase activity in a manner consistent with the loss of activity by dilution. The data shown in Figs. 1 and 2 may not confirm that Barrett’s esophagi have elevated telomerase activity, in general, because of insufficient sample size. However, the results do emphasize the importance of the purification of specific cells to be studied and suggest that activity assays conducted in tissue extracts may not be very accurate and reliable. Diseased cells related to Barrett’s esophagus and various stages leading to adenocarcinoma can be purified from fresh-frozen tissue sections, using the laser capture microdissection (LCM) technique as following. Frozen tissue sections  8  mm in size are cut and mounted onto glass slides. The sections are defrosted, fixed in 70% ethanol for 3 min, and stained with hematoxylin. The stained sections should be examined under microscope by an experienced gastrointestinal pathologist who identifies the target areas including specific disease states, using standard histopathological criteria (16). Briefly, the stages of disease are identified as below. SIM is defined as specialized columnar epithelium which contains well formed goblet cells. LGD is characterized by substantial dysplastic cytological changes. The crowded glands can be seen with hyperchromatic, polarized, and enlarged nuclei which vary in size and shape. Both the nucleoli and mitosis are prominent in the nuclei. Architectural distortions are minimal to mild at this stage of disease. HGD is associated with both the architectural and cytological dysplastic changes. Dysplastic glands show various growth patterns such as papillary, cribriform, back-to-back forms, anastomosing, and villiform. The glands frequently show mitoses but sometimes also necrosis. The nuclei are crowded, overlapped, and depolarized. HGD is not associated with frank invasion. In BEAC, dysplastic glands cause the break down of the basement membrane and are present within and beyond the laminar propria. Once the pathologist identifies the targeted glands, the areas are marked and the slide repositioned for LCM using the Pixcell II LCM System (Arcturus Inc., Mountain View, CA ), according to the standard protocol (17). The captured cells are placed in appropriate buffers for the analysis of DNA, RNA, or protein.

2. Materials 2.1. Laser Capture Microdissection Slide Preparation

1. 70, 95, and 100% ethanol. 2. Xylene, mixed, ACS reagent. 3. Complete, mini protease inhibitor cocktail tablets (Roche Corp. Tucson, AZ) to be added in all solutions except xylene. 4. Hematoxylin solution, Mayer’s (Sigma, St Louis, MO).

14  Purification of Diseased Cells from Barrett’s Esophagus and Related Lesions…

185

5. Eosin Y solution (Sigma). 6. Deionized water. 7. 95% ethanol (Sigma). 8. Xylenes (Sigma). 2.2. Laser Capture Microdissection

1. LCM machine (e.g., PixCell II, Molecular Devices, Sunnyvale, CA). 2. LCM caps (Molecular Devices).

2.3. DNA/RNA Isolation

1. Forceps. 2. Microcentrifuge tubes. 3. QIAamp DNA Micro kit (Qiagen Inc., Valencia, CA). 4. Picopure RNA isolation kit (Molecular Devices, Inc., Sunny vale, CA). 5. RiboAmp Plus kit (Molecular Devices, Inc., Sunnyvale, CA).

3. Methods 3.1. Paraffin Embedded and Frozen Sections Slide Preparation

For paraffin-embedded sections, start at step 1 and for frozenembedded sections, start at step 4. Place the sections in the following solutions: 1. Fresh xylenes (to deparaffinize the sections) – 5 min. 2. Fresh xylenes – 5 min. 3. 100% ethanol – 15 s. 4. 95% ethanol – 15 s. 5. 70% ethanol – 15 s. 6. Deionized water – 15 s. 7. Mayer’s hematoxylin (10%) – 30  s (or other appropriate stain). 8. Deionized water – rinse (×2) – 15 s. 9. 70% ethanol – 15 s. 10. Eosin Y(10%) – 5 s. 11. 95% ethanol – 15 s. 12. 95% ethanol – 15 s. 13. 100% ethanol – 15 s. 14. 100% ethanol – 15 s. 15. Xylenes (to ensure dehydration of the section) – 60 s. 16. To completely remove xylenes, air-dry for approximately 2 min. 17. The tissue is now LCM ready.

186

M.A. Shammas and M.Y. Rao

3.2. DNA Isolation

1. After laser capture, the polymer on the cap is peeled off using forceps and kept in a microcentrifuge tube. Laser-captured tissue attached to the polymer can be lysed by lysis buffer from QIAamp DNA Micro kit.

3.3. RNA Isolation

1. After laser capture, the polymer on the cap is peeled off using forcep and kept in a microcentrifuge tube. Captured tissue attached to the polymer can be lysed by lysis buffer from Picopure RNA isolation kit.

3.4. RNA Isolation for Microarray

1. After laser capture, the polymer on the cap is peeled off using forceps and kept in a microcentrifuge tube. Laser-captured tissue attached to the polymer can be lysed by lysis buffer from RiboAmp Plus kit. Using this kit, RNA can be isolated and amplified from frozen tissue samples. For paraffin-embedded tissue, Paradise Plus reagents can be used.

3.5. Laser Capture Microdissection

1. Paraffin-embedded tissue sections on slides are stored at room temperature under desiccation. Slides with frozen tissue sections are stored at −70°C before laser capture. When ready to process, deparaffinize the slides. For frozen slides, dip the slides in cold acetone to remove OCT (embedding medium) and for fixation. Stain with hematoxylin and eosin and dehydrate with ethyl alcohol and xylene for LCM. 2. Before laser capture experiment, air dry the slides with xylene for 2 min. 3. The slides should be without mounting medium and coverslips. 4. Place slide on the microscope stage and load the caps into the arm of the LCM machine. 5. Place the arm of LCM machine in such a way that it sits on the tissue of the slide. 6. Look at the monitor and adjust the position of arm in such a way that area of interest falls under the cap. Area should be inspected by a pathologist. 7. Take the picture before laser capture. 8. Start the laser and cut the target tissue. 9. Lift the cap and take a picture after laser capture. 10. Take a picture of cap where laser captured tissue was attached. 11. Using forcep, peel off the polymer of the cap and put it in a microcentrifuge tube with specific lysis buffer for DNA or RNA evaluation. It is a better way as compared to putting whole cap on the microfuge tube because whole cap requires more buffer (see Note 1).

14  Purification of Diseased Cells from Barrett’s Esophagus and Related Lesions…

187

4. Note 1. No nonspecific binding of tissue is observed if abovementioned protocol is used. Laser spot size can be increased if a larger area needs to be dissected. Laser intensity can also be increased if tissue is not coming off the slide. References 1. Devesa SS, Blot WJ, Fraumeni JF, Jr. (1998) Changing patterns in the incidence of esophageal and gastric carcinoma in the United States. Cancer. 83:2049–2053. 2. Spechler SJ, Goyal RK. (1986) Barrett’s esophagus. N Engl J Med. 315:362–371. 3. Blot WJ, McLaughlin JK. (1999) The changing epidemiology of esophageal cancer. Semin Oncol. 26:2–8. 4. Morales CP, Lee EL, Shay JW. (1998) In situ hybridization for the detection of telomerase RNA in the progression from Barrett’s esophagus to esophageal adenocarcinoma. Cancer. 83:652–659. 5. Lord RV, Salonga D, Danenberg KD, et  al. (2000) Telomerase reverse transcriptase expression is increased early in the Barrett’s metaplasia, dysplasia, adenocarcinoma sequence. J Gastro intest Surg. 4:135–142. 6. Aisner DL, Wright WE, Shay JW. (2002) Telomerase regulation: not just flipping the switch. Curr Opin Genet Dev. 12:80–85. 7. Cong YS, Wright WE, Shay JW. (2002) Human telomerase and its regulation. Microbiol Mol Biol Rev. 2002;66:407–425. 8. Barclay JY, Morris A, Nwokolo CU. (2005) Telomerase, hTERT and splice variants in Barrett’s oesophagus and oesophageal adenocarcinoma. Eur J Gastroenterol Hepatol. 17:221–227. 9. Yoneyama K, Aoyama N, Koizumi H, Tamai S. (1998) Telomerase activity in esophageal carcinoma and lesions unstained with Lugol’s solution. Nippon Rinsho. 56:1181–1185. 10. Bachor C, Bachor OA, Boukamp P. (1999) Telomerase is active in normal gastroin­ testinal mucosa and not up-regulated in

­ recancerous lesions. J Cancer Res Clin p Oncol. 125:453–460. 11. Going JJ, Fletcher-Monaghan AJ, Neilson L, et al. (2004) Zoning of mucosal phenotype, dysplasia, and telomerase activity measured by telomerase repeat assay protocol in Barrett’s esophagus. Neoplasia. 6:85–92. 12. Harle-Bachor C, Boukamp P. (1996) Telomerase activity in the regenerative basal layer of the ­epidermis inhuman skin and in immortal and carcinoma-derived skin keratinocytes. Proc Natl Acad Sci USA. 93:6476–6481. 13. Emmert-Buck MR, Bonner RF, Smith PD, Chuaqui RF, Zhuang Z, Goldstein SR, Weiss RA, Liotta LA. (1996) Laser capture microdissection. Sci. 274(5289):998–1001. 14. Collaud S, Wiedl T, Cattaneo E, Soltermann A, Hillinger S, Weder W, Arni S. (2010) Lasercapture microdissection impairs activity-based protein profiles for serine hydrolase in human lung ­adenocarcinoma. Biomol Tech. 21(1): 25–28. 15. Shammas MA, Qazi A, Batchu RB, et  al. (2008) Telomere maintenance in laser capture microdissection-purified Barrett’s adenocarcinoma cells and effect of telomerase inhibition in  vivo. Clin Cancer Res. 14: 4971–4980. 16. Huang Q, Yu C, Klein M, Fang J, Goyal RK. (2005) DNA index determination with Automated Cellular Imaging System (ACIS) in Barrett’s esophagus: comparison with CAS 200. BMC Clin Pathol. 5:7. 17. Ohyama H, Zhang X, Kohno Y, et al. (2000) Laser capture microdissection-generated target sample for high-density oligonule‑tide array hybridization. Biotechniques. 29:530–536.

wwwwwww

Chapter 15 Laser Microdissection of Intestinal Epithelial Cells and Downstream Analysis Benjamin Funke Abstract The intestinal epithelium is at the front line when it comes to preserving mucosal immune homeostasis. There is growing evidence that the epithelium plays a crucial role in the pathogenesis of inflammatory bowel disease. Laser captured microdissection techniques offer a promising approach to investigating the underlying molecular mechanisms. This would require reliable protocols for the extraction of high quality RNA from intestinal mucosa samples acquired by laser microdissection. However, such protocols are not around. Therefore our objective was to establish a feasible protocol which supports the study of the involvement of intestinal epithelium in the pathogenesis of IBD. Key words: Laser captured microdissection, Compartment specific, Intestinal epithelial cells, Lamina propria, Inflammatory bowel disease, DcR3, RIG-I

1. Introduction Laser captured microdissection (LCM) technologies offer a promising new approach to growing areas of scientific study such as transcriptomics where pure samples are required in order to produce reliable results (1, 2). Homogeneous tissue preparation is a prerequisite for modern molecular analyses, both in qualitative and in quantitative terms. Isolation of high-quality RNA is the most critical step in gene expression analysis because RNA degradation is part of the survival mechanism and ubiquitous in all cells (3). Several protocols are available for tissue preparation, but none of these safeguards the integrity of the RNA during LCM, especially when intestinal mucosal tissue is involved. Endogeneous RNase activity requires the presence of water (4).

Graeme I. Murray (ed.), Laser Capture Microdissection: Methods and Protocols, Methods in Molecular Biology, vol. 755, DOI 10.1007/978-1-61779-163-5_15, © Springer Science+Business Media, LLC 2011

189

B. Funke

Thus, we developed a reliable method that maintains RNA integrity (Fig. 3a) and focuses on stabilization and dehydration of samples during microdissection process. Before interpreting mRNA expression profiles, we verified whether our method delivers biologically meaningful results. Gene expression analysis of cytokeratin 20 (CK20), a marker known to be typically expressed in intestinal epithelial cells was prominent in the microdissected epithelial fraction (Fig. 1). In addition we were able to demonstrate excessive transcriptional expression levels of vimentin (VIM) in the lamina propria of the ileal mucosa (Fig. 1), a characteristic marker of this compartment. Up to now, our experiments resulted in the identification of the retinoic acid-inducible gene I (RIG-I) as one of the genes which is significantly downregulated within the epithelial layer in Crohn’s disease (CD) one of the major forms of inflammatory bowel disease (5). RIG-I is a noteworthy pattern recognition receptor similar to NOD2 which is indentified as a susceptibility gene for CD (6). Furthermore, we could demonstrate a significant over-expression of the soluble decoy receptor 3 (DcR3) in the epithelial compartment of patients suffering from CD (7). Both findings may be considered as additional pieces for solving the puzzle of the complex pathogenesis of inflammatory bowel diseases.

30

25

20 fold change

190

lamina propria

15

epithelium

10

5

0 CK20

VIM

Fig.  1. LMPC is compartment-specific. Epithelial and lamina propria fractions of the human ileum were obtained by our laser microdissection protocol. RNA was isolated and expression levels were estimated using quantitative RT-PCR. The diagram shows the relative expression levels of cytokeratin 20 (typically expressed in intestinal epithelial cells) and vimentin (usually expressed in the lamina propria). The mRNA of cytokeratin 20 is prominent in microdissected epithelial fragments. In contrast, there is an excessive transcriptional expression of vimentin in the lamina propria of the ileum in comparison to the epithelium.

15  Laser Microdissection of Intestinal Epithelial Cells and Downstream Analysis

191

2. Materials 2.1. Tissue Preparation

1. RNaseZap® (Ambion, Austin, TX). 2. MembraneSlide NF 1.0 PEN (Carl Zeiss MicroImaging, Germany). 3. RNAlater®-ICE (Ambion, Austin, TX). 4. Nuclease-free water. 5. Series of graded alcohols (50, 75, 95, and 100%). 6. 1% (w/v) cresyl violet acetate: Dissolve cresyl violet acetate (Sigma-Aldrich, Germany) at a concentration of 1% (w/v) in ACS-grade 100% ethanol at room temperature with agitation for several hours to overnight. Filter the stain through a 0.2-mM filter unit prior to use. 7. Xylene.

2.2. Lasermicrod­issection and Pressure Catapulting (LMPC)

1. AdhesiveCaps, Germany).

opaque

(Carl

Zeiss

MicroImaging,

2.3. RNA-Extraction

1. peqGOLD TriFast™ (PeqLab Biotechnologie, Germany). 2. 2 M sodium acetate buffer, pH 4. 3. Pellet Paint® Co-Precipitant (Novagen). 4. Absolutely RNA® Nanoprep Kit (Stratagene, La Jolla, CA). 5. RNA 6000 Nano Chip Kit (Agilent Technologies, Waldbronn, Germany).

2.4. Quantitative RT-PCR

1. REVERSE-IT™ max 1st strand synthesis kit (Abgene, Hamburg, Germany). 2. iTaq SYBR Green Supermix with ROX (Bio-Rad Laboratories, Hercules, USA).

2.5. Microarray Analysis

1. MessageAmpTM II-Biotin Enhanced kit (Ambion, Austin, TX). 2. Affymetrix® HG-U133 Plus 2.0 GeneChips® oligonucleotide arrays (Affymetrix UK Ltd., High Wycombe, United Kingdom).

3. Methods 3.1. Tissue Preparation

Clean cryostat thoroughly (e.g., RNaseZap®) and adjust to −22°C. Cut freshly frozen mucosa samples into 18 mm sections. Mount these sections on precooled RNAse and DNAse-free

192

B. Funke

membrane slides. For better adhesion warmth the appropriate slide area (i.e., with your thumb). Allow 2 min air drying at −22°C and incubate the slides for 10 min at −22°C in RNAlater®-ICE. For prestaining fixation dip the slides at room temperature five to seven times in a decreasing series of graded alcohols (95, 75, and 50%). To stain the tissue dip slides for 15  s in 1% cresyl violet acetate. For poststaining fixation dip the slides in an increasing series of graded alcohols (50, 75, 95, and 100%) and incubate for 5  min in xylene. Gently tap the slide on absorbent surface to remove excess ethanol or cresyl violet. Subsequently, allow airdrying for about 10 min. 3.2. Lasermicrodissection and Pressure Catapulting (LMPC)

1. These instructions assume the use of Microbeam LMPC System (Carl Zeiss MicroImaging, Munich, Germany). Apply the RoboLPC method (Fig. 2) according to manufacturer’s instructions to microdissected and capture 100,000– 250,000 cells (approximately 10  mm2 fragment area for epithelium and 20  mm2 for Lamina propria). For sample collection use AdhesiveCaps, opaque. Because of the time consuming procedure you can leave slides and AdhesiveCaps in an exsiccator overnight to continue with laser micrdissection next day.

3.3. RNA-Extraction and DNAse Treatment

1. Use the RNA purification system peqGOLD TriFast according to manufacturer’s instructions. To increase the RNA yield wash out the phenol phase with another 500  ml of 2  M sodium acetate, pH 4. For RNA precipitation use 2 ml Pellet Paint® co-precipitant. Avoid complete drying of the pellet. Dissolve pellet in 20 ml nuclease free water. Perform a second purification and DNA digestion step with an Absolutely RNA® Nanoprep Kit according to the manufacturer’s instructions (12  ml elution volume). Check RNA quality (RNA intergrity number) and quantity on an Agilent 2100 Bioanalyzer with a RNA 6000 Nano microfluidics chips according to the manufacturer’s instructions (Fig.  3, see Notes 1 and 2).

3.4. Quantitative RT-PCR

1. Transcribe 100 ng total RNA (or 2 × 5 ml RNA template) into cDNA using REVERSE-IT™ max 1st strand synthesis kit according to the manufacturer’s protocol. Dilute cDNA 1:5 with nuclease free water (100 ml). These instructions assume the use of an Applied Biosystems 7300 Real-Time PCR System (Applied Biosystems, Darmstadt, Germany) to perform real-time monitoring of PCR reactions. They are easily adaptable to other PCR formats.

15  Laser Microdissection of Intestinal Epithelial Cells and Downstream Analysis

193

Fig.  2. LMPC technique (RoboLPC method, Carl Zeiss MicroImaging): demonstration of the RoboLPC method of our Microbeam LMPC System. Ileal tissues mounted on PEN-membrane slides and stained with cresyl violet acetate. Preparation of an ileal crypt (a–d) and of lamina propria cells (e–h) was documented with the embedded video system. Tissue fragments were captured and visualized in opaque AdhesiveCaps® (d, h).

194

B. Funke

Fig. 3. RNA quality. Representative electropherogram of (a) isolated total RNA (RNA 6000 Nano microfluidics chips on an Agilent 2100 BioAnalyzer, RIN: RNA integrity number) and (b) of amplified mRNA (MessageAmpTM II, ~40 mg).

2. Conduct each PCR in 20  ml reaction volume (iTaq SYBR Green Supermix with ROX, Bio-Rad Loboratories, Hercules, USA) containing 1 ml of diluted cDNA and 250 nM of each gene-specific primer. 3.5. High-Density Oligonucleotide Microarray Analysis

1. Prior to hybridization on Affymetrix® HG-U133 Plus 2.0 GeneChips® oligonucleotide arrays, amplify and biotinylate about 200  ng of total RNA isolates using the Eberwine method (8) as outlined in the manufacturer’s manual of the MessageAmpTM II-Biotin Enhanced kit. An in vitro transcription with biotinylated UTP for 14 h results approximately in a 100-fold amplification of RNA.

15  Laser Microdissection of Intestinal Epithelial Cells and Downstream Analysis

195

2. Process amplified RNA generated from each sample on an Affymetrix GeneChip® Instrument System. Add spike controls to 15 mg fragmented aRNA, and perform an over-night hybridisation for 16 h on Affymetrix® HG-U133 Plus 2.0 GeneChips®. Wash, stain and scan the arrays (Affymetrix GeneChip® Scanner 3000, protocol: EukGE-WS2v5_450 for GeneChip® fluidics station 450) (see Note 3).

4. Notes 1. RNA is an exceptionally fragile whereas ribonuclease is a very stable molecule and its activity is ubiquitous. Apply the standard precautions for handling RNA (designate a special area for RNA work only; always wear and change gloves frequently; use 100% ethanol to clean working surfaces; use sterile, disposable plastic ware; all solutions should be prepared in nuclease free water; keep RNA on ice; store RNA aliquots at −80°C…). 2. It is advisable to verify RNA quality (RIN, Agilent Bioanalyzer) before continue with expensive and time-consuming downstream analysis (Fig. 3). 3. Deposit the results of your microarray analysis in accordance with the MIAME standard (minimum information about a microarray experiment) at public repositories. Most journals require MIAME compliant data as a condition for publishing microarray based papers. Our microarray data have been deposited at ArrayExpress at the European Bioinformatics Institute (EMBL-EBI) (9–13). The access number of the repository for microarray data is E-MEXP-2083. References 1. Burgemeister R, Gangnus R, Haar B, et  al. (2003) High quality RNA retrieved from samples obtained by using LMPC (laser microdissection and pressure catapulting) technology. Pathol Res Pract. 199:431–436 2. Niyaz Y, Stich M, Sagmuller B, et al. (2005) Noncontact laser microdissection and pressure catapulting: sample preparation for genomic, transcriptomic, and proteomic analysis. Methods Mol Med. 114:1–24 3. Houseley J, Tollervey D. (2009) The many pathways of RNA degradation. Cell. 136:763–776 4. Clement-Ziza M, Munnich A, Lyonnet S, et al. (2008) Stabilization of RNA during laser capture microdissection by performing experiments under argon atmosphere or using etha-

nol as a solvent in staining solutions. Rna.14:2698–2704 5. Funke B, Lasitschka F, Roth W, et al. (2011) Selective downregulation of retinoic acidinducible gene I within the intestinal epithelial compartment in crohn’s disease. Inflamm Bowel Dis. [Epub ahead of print] 6. Ogura Y, Bonen DK, Inohara N, et al. (2001) A frameshift mutation in NOD2 associated with susceptibility to Crohn’s disease. Nature. 411:603–606 7. Funke B, Autschbach F, Kim S, et al. (2009) Functional characterisation of decoy receptor 3 in Crohn’s disease. Gut. 58:483–491 8. Van Gelder RN, von Zastrow ME, Yool A, et al. (1990) Amplified RNA synthesized from limited quantities of heterogeneous cDNA.

196

B. Funke

Proceedings of the National Academy of Sciences of the United States of America. 87:1663–1667 9. Brazma A, Hingamp P, Quackenbush J, et al. (2001) Minimum information about a microarray experiment (MIAME)-toward standards for microarray data. Nature Genetics. 2001;29:365–371 10. Brazma A, Parkinson H, Sarkans U, et  al. (2003) ArrayExpress – a public repository for microarray gene expression data at the EBI. Nucleic Acids Research. 31:68–71

11. Parkinson H, Kapushesky M, Shojatalab M, et al. (2007) ArrayExpress – a public database of microarray experiments and gene expression profiles. Nucleic Acids Research. 35:D747–750 12. Parkinson H, Sarkans U, Shojatalab M, et al. (2005) ArrayExpress – a public repository for microarray gene expression data at the EBI. Nucleic Acids Research. 33:D553–555 13. Rocca-Serra P, Brazma A, Parkinson H, et al. (2003) ArrayExpress: a public database of gene expression data at EBI. Comptes rendus biologies. 326:1075–1078

Chapter 16 Application of Laser Microdissection and Quantitative PCR to Assess the Response of Esophageal Cancer to Neoadjuvant Chemo-Radiotherapy Claus Hann von Weyhern and Björn L.D.M. Brücher Abstract Tissues are complicated three-dimensional structures, composed of different types of interacting cells. Since the cell population of interest might constitute only a minor fraction of the total tissue volume, the problem of tissue heterogeneity has been a major barrier to the molecular analysis of normal versus diseased tissue. Thus, tissue microdissection represents one of the most promising techniques in molecular pathology offering the link between morphology and genetic analysis since it was established in the early 1970s. These first applications and further developments in the techniques enable preparation of morphologically well described and circumscribed cell populations of either tumor cells or surrounding tissue or even cytology specimens without contamination of unwanted cells. Laser capture microdissection is suitable for the dissection of both paraffin embedded and fresh frozen material. Further applications of the dissected genomic material are isolation of DNA and RNA as described later on followed by PCR or RT-PCR and sequencing. Key words: Tissue microdissection, Laser capture microdissection, PCR, qRT-PCR, DEPC

1. Introduction Variable techniques of microdissection have been developed. In the 1970s, Lowry and Passonneau used a method which used microdissection freehand under a conventional microscope. In the late 1970s and early 1980s, UV laser was used for preparation of precancerous lesions from normal hepatocytes. Further developments used UV-laser instruments and pulsed 337 nm nitrogen laser for destroying unwanted tissue (1). Nowadays, laser capture microdissection (LCM) and laser capture catapulting (LPC) are

Graeme I. Murray (ed.), Laser Capture Microdissection: Methods and Protocols, Methods in Molecular Biology, vol. 755, DOI 10.1007/978-1-61779-163-5_16, © Springer Science+Business Media, LLC 2011

197

198

C.H. von Weyhern and B.L.D.M. Brücher

most used techniques in pathology (2–4). We describe which materials are suitable for microdissection, troubleshooting, and isolation of DNA and RNA from dissected specimens.

2. Materials For laser microdissection, two competitive systems are widely used at present. Both techniques consist of an inverted microscope, a laser control unit, a CCD-camera and facultative joystick, and computer with color monitor. 1. LCM (Molecular Devices, Sunnyvale, CA): a laser microbeam of low power does not cut the tissue but melts a thermoplastic ethyl vinyl acetate membrane mounted on the plastic cap of the tube. The melted membrane sticks the selected and “lasered” cells to the membrane. When the cap with the membrane is lifted up, cells stick to the membrane and lift up together (5). 2. Laser pressure microdissection (Zeiss, Gottingen, Germany): using laser pressure microdissection (LPM), the cells of interest are cut away from the remainder tissue. The system uses a 337-nm pulsed nitrogen laser. The dissected cells are collected in a cap similar to the LCM but without any thermoplastic membrane in contrast to LCM. The mechanism for catapulting away the cells of interest is a high-photon density under the tissue. 3. RNA lysis buffer 1 M Tris/HCl (pH 8.0), 0.5 M ethylenediaminetetraacetic acid, 10% sodium dodecyl sulfate, DEPC, and proteinase K.

3. Methods 3.1. Laser Microdissection

1. For microdissection, deparaffinized slides are transferred to the stage of the microscope. The cap is fixed in a transport arm and placed on the selected tissue area. 2. For dissection, the cells of interest are selected under visual control on the screen and marked by setting frames. For LCM, the positioning beam is placed on the cells and a cluster of laser beams is placed on the single cells itself or in a ring around the desired cells. If the thermoplastic membrane sticks to the tissue, lifting up the cap together with the arm transfers the cells in the tube. For LPC, single cells or cell clusters are marked by frame on the screen. By activating the laser, pulsed

16  Application of Laser Microdissection and Quantitative PCR…

199

actions of the laser catapult the cells directly in the cap of the tube which is just above of the area of interest. 3. The most important advantage of LPC in comparison to LCM is the motorized stage with recognition of the X and Y positions of the marked cells. This feature allows selection of numerable cells at one point of time and collection of cells at the end of selection process (see Note 1). Thus, a large amount of cells can be collected within a few minutes. Due to the fact that both techniques are “no touch” techniques, no manual dexterity is required in contrast to historical micromanipulators for collecting tissue. 4. Because surrounding tissue is not destroyed by the use of the laser, more than one cell type can be selected from the same slide out of the adjacent tissue. The most important disadvantage from our point of view is the optical quality in the microscope due the deparaffinized tissue without coverslip. Therefore, some experience in defining cells by shape is needed for correct selection. For dissection of uniform cells with slight differences in shape and nucleus, prestaining with immunohistochemistry may help in distinguishing cell types with high precision (6). 3.2. Extraction of Total RNA from Microdissected Formalin-Fixed, Paraffin-Embedded Tissue from Neoadjuvant-Treated Squamous Esophageal Cancer

1. As described above, microdissection of tissue is necessary for precise selection of tumor cells in abundant amount of surrounding tissue. One application of both microdissection and quantitative RT-PCR is used in esophageal squamous cell carcinoma of the esophagus after radiochemotherapy (7). The purpose of the selected study was to evaluate the expression patterns of thymidylate synthetase, thymidylate phosphatase, dihydropyrimidine dehydrogenase Her2/neu, and cyclin D1 in samples of patients with locally advanced esophageal squamous cell carcinoma followed by transthoracic microscopic and macroscopic complete tumor resection after neoadjuvant therapy. Precise dissection of tumor cells out of surrounding tissues, such as lymphoid infiltrates, foamy cells, fibroblasts, is essential for precise results as shown for cancer and stromal cells in colon cancer and gastric cancer (8). The investigative study performed by Bruecher et  al. identified Cyclin D1 expression as significant correlation with survival, local tumor recurrence and recurrence-free survival (7).

3.3. Isolation of RNA from Formalin-Fixed Paraffin-Embedded Tissues (9)

1. Dissection of formalin-fixed paraffin-embedded tissues: routine sections (5 mm) stained with hematoxylin and eosin on a microtome with a disposable blade. 2. Mounting on noncoated glass slides. 3. Deparaffinization: Two changes of xylene for 10  min, each 5  min for 100% ethanol, 96% ethanol, 90% ethanol, 70% ethanol.

200

C.H. von Weyhern and B.L.D.M. Brücher

4. H&E staining 5 min. 5. Rinsing RNase-free H2O 30 s. 6. 100% ethanol for 1 min. 7. Microdissection of single cells or groups of cells. 8. Transfer of the sterile cap with the cells inside to a sterile 1.5 ml standard tube. 9. 200 ml RNA lysis buffer. 10. Phenol and chloroform extraction. 11. Precipitation of equal volume of isopropanol. 12. Adding 0.1 volume 3 mol/l sodium acetate (pH 4.0) 1 ml of 10 mg/ml carrier glycogen at −20°C. 13. Washing of RNA pellet once in 70% ethanol and subsequent drying. 14. Resuspension of the pellet in 10 ml RNase-free water. For the study by Brücher et al. (7), serial sections of 5 mm were cut in cryostat mounted on noncoated clean glass slides and used immediately for dissection. After six slides, a 2-mm slide was made for H&E staining as control followed by another set of six slides for microdissection. The sections were dried overnight and deparaffinized in two changes in xylene for 30 min, followed by rehydration in two changes of 100% ethanol, followed by 96% ethanol, 09% ethanol, 70% ethanol, and a solution of diethyl pyrocarbonate (DEPC). Alcohol solutions with distilled water were treated 24 h before use by DEPC and were therefore RNase free. Following rehydration, hematoxylin staining for 45 s was performed by short dehydration series in 50% ethanol, 70, 96, and 100% ethanol and xylene. For the valid results, an amount of dissected tumor cells between 1,000 and 4,000 per patient were dissected. 3.4. Application of PCR Techniques on Dissected Material

Carcinomas of the upper GI-tract under multimodal therapy regimes are a very good example for the necessity of using microdissection methods. After multimodal therapy and resection of the surgical specimen, pathohistological examination is done by embedding the complete tumor and histological investigation. Most tumor regression features in carcinomas of the gastrointestinal tract show a discontinuous distribution of residual tumor cells and tumor cell nests. Prediction of the grade of regression is an arbitrary topic and little is known about predictive biomarkers now. The idea is treating patients preoperatively in multimodal fashion considering individual marker profiles. Thus, retrospective data from preoperative genetic expression profiles protein expression obtained from biopsy specimens are necessary for future response prediction. For recent results in this field, TGFbpathway in squamous cell carcinoma of the esophagus (10) was evaluated on macrodissected pretherapeutic biopsy specimens.

16  Application of Laser Microdissection and Quantitative PCR…

201

1. At least 2,000 tumor cells were picked up under microscope control. 2. TaqMan PCRs were performed in duplicate using the PCR Mastermix (Applied Biosystems/Roche Diagnostics Mannheim, Germany). 3. Amplification cycles were repeated 50 times at 95°C for 15 s followed by 60°C for 1 min. 4. TGFb1, Smad4, and Smad7 were determined by using primers from Applied Biosystems/Roche Diagnostics Mannheim, Germany (see Note 2). Normalization of mRNA-expression was done against the housekeeping gene GAPDH. 5. Results from this study significantly higher levels of Smad4 mRNA-expression in tumors with total or subtotal regression grades determined by the regression grading scheme of Becker et  al. (11). TGFb1 Smad2 and Smad7 in contrast showed no significant results as prognostic markers.

4. Notes 1. The most important advantage of LCM as well as LPC is speed, versatility, and precision. Since the laser spot is about 1–5 mm in diameter, even single cells can be picked up. 2. Primers are spanning exons 1 and 2 for TGFb, Exons 6 and 7 for Smad4, and exons 2 and 3 for Smad7. References 1. Eltoum, I.A., Siegal, G.P., Frost, A.R., (2002) Microdissection of histologic sections: past, ­present, and future. Adv Anat Pathol, 9, 316–22. 2. Fend, F., Kremer, M., Quintanilla-Martinez, L., (2000) Laser capture microdissection: methodical aspects and applications with emphasis on immuno-laser capture microdissection. Pathobiology, 68, 209–14. 3. Fend, F. Raffeld, M., (2000) Laser capture microdissection in pathology. J Clin Pathol, 53, 666–72. 4. Fend, F., Specht, K., Kremer, M., QuintanillaMartinez, L., (2002) Laser capture microdissection in pathology. Methods Enzymol, 356, 196–206. 5. Bonner, R.F., Emmert-Buck, M., Cole, K., Pohida, T., Chuaqui, R., Goldstein, S., Liotta, L.A., (1997) Laser capture microdissection: molecular analysis of tissue. Science, 278, 1481–1483.

6. Fend, F., Emmert-Buck, M.R., Chuaqui, R., Cole, K., Lee, J., Liotta, L.A., Raffeld, M., (1999) Immuno-LCM: laser capture microdissection of immunostained frozen sections for mRNA analysis. Am J Pathol, 154, 61–6. 7. Brücher, B.L.D.M., Keller, G., Werner, M., Muller, U., Lassmann, S., Cabras, A.D., Fend, F., Busch, R., Stein, H., Allescher, H.D., Molls, M., Siewert, J.R., Hofler, H., Specht, K., (2009) Using Q-RT-PCR to measure cyclin D1, TS, TP, DPD, and Her-2/neu as predictors for response, survival, and recurrence in patients with esophageal squamous cell carcinoma following radiochemotherapy. Int J Colorectal Dis, 24, 69–77. 8. Makino, H., Uetake, H., Danenberg, K., Danenberg, P.V., and Sugihara, K., (2008) Efficacy of laser capture microdissection plus RT-PCR technique in analyzing gene expression levels in human gastric cancer and colon cancer. BMC Cancer, 2008. 8, 210.

202

C.H. von Weyhern and B.L.D.M. Brücher

9. Specht, K., Richter, T., Muller, U., Walch, A., Werner, M., and Hofler, H.,(2001) Quanti­ tative gene expression analysis in microdissected archival formalin-fixed and paraffin-embedded tumor tissue. Am J Pathol, 158, 419–29. 10. Pühringer-Oppermann, F., Sarbia, M., Ott, N., Brücher B.L.D.M.., (2010) The predictive value of genes of the TGF-b1 pathway in

multimodally treated squamous cell carcinoma of the esophagus. Int J colorectal Dis 25, 515–21. 11. Becker K, Mueller JD, Schulmacher C, Ott K, Fink U, Busch R, Böttcher K, Siewert JR, Höfler H.. (2003) Histomorphology and grading of regression in gastric carcinoma treated with neoadjuvant chemotherapy. Cancer. 98, 1521–30.

Chapter 17 Oligonucleotide Microarray Expression Profiling of Contrasting Invasive Phenotypes in Colorectal Cancer Christopher C. Thorn, Deborah Williams, and Thomas C. Freeman Abstract This chapter refers to the application of laser-capture microdissection with oligonucleotide microarray analysis. The protocol described has been successfully used to identify differential transcript expression between contrasting colorectal cancer invasive phenotypes. Tissue processing, RNA extraction, quality control, amplification, fluorescent labelling, purification, hybridisation, and elements of data analysis are covered. Key words: Oligonucleotide microarray, Laser-capture microdissection, Invasion, Colorectal

1. Introduction Metastatic potential in colorectal cancer is closely associated with a number of well-recognised histopathological characteristics including the nature of the invasive margin. Comparison of tumour epithelia between these phenotypes is confounded by the presence of stromal cell populations. Post hoc attempts to account for the contribution of the stroma is complicated by the complex nature of tumour–stromal interactions as described in the paradigm of the “permissive field” (1–6). Laser-capture microdissection (LCM) mitigates these difficulties by permitting the study of rarefied cellular populations, avoiding contamination from both stroma and also clones derived from formative stages of carcinogenesis (2, 7–9). The extent of such contamination has been demonstrated in a breast cancer xenograft model where differential transcript expression between primary breast cancer and matched lymph node metastases were concordant in only 1% of genes identified by LCM and non-LCM methodologies (10).

Graeme I. Murray (ed.), Laser Capture Microdissection: Methods and Protocols, Methods in Molecular Biology, vol. 755, DOI 10.1007/978-1-61779-163-5_17, © Springer Science+Business Media, LLC 2011

203

204

C.C. Thorn et al.

High-throughput technologies allow the identification of patterns of gene expression and can indicate pathway dysregulation in carcinogenesis as well as to provide molecular markers for clinically significant tumour subtypes, not able to be distinguished by histological means. Spotted oligonucleotide microarrays are one of a variety of modern array types used for the analysis of the transcriptome. This type of microarray consists of 50–70mer oligonucleotide probes in picolitre volumes applied to a glass slide, frequently in densities that permit genome-wide analyses of the transcriptome. These “probes” covalently bond to the substrate and the oligonucleotides are complementary to the mRNA/ cDNA sequences they represent. Hybridisation ensues when one or two samples of “target” are added. The relative abundance of transcripts, which may reflect expression, is detected with the use of fluor-dye labelling. RNA acquisition of sufficient quantity and quality to facilitate successful microarray hybridisation is a challenge when samples are processed with LCM. Techniques to address many of these difficulties have been published and discuss the sources and implications of RNA degradation (11–15). Amplification techniques are always required and both linear and exponential methods have been successfully employed (16, 17). The template-switching amplification technique utilised in this protocol generates full length cDNA for hybridisation, however like many other techniques, the poly(A) tail provides the initial primer binding site. During first and second strand cDNA synthesis, 5¢ primer sites are incorporated to permit exponential amplification of ds cDNA by the polymerase chain reaction. Labelling of the amplified cDNA is achieved with the incorporation of Cy3/Cy5 labelled nucleotides, using the 5¢–3¢ DNA polymerase activity of the Klenow fragment, initiated with random primers. An advantage of the use of oligonucleotide probes is that they can be designed to be complementary to sequences towards the 3¢ end (typically within 300 bp) of complementary DNA generated from an RNA sample therefore making possible the use of partially degraded RNA. There are numerous publications describing the successful combination of LCM and transcriptomics using a variety of downstream applications in the context of colorectal carcinogenesis (18–24).

2. Materials 2.1. Tissue Processing and Laser-Capture Dissection

1. Cryo-M-Bed Embedding media (Bright, Huntingdon, UK). 2. Propan-2-ol (VWR International Limited, Poole, Dorset, UK) and liquid nitrogen.

17  Oligonucleotide Microarray Expression Profiling of Contrasting Invasive Phenotypes…

205

3. Superfrost microscope slides (Fisher Scientific, Loughborough, Leicestershire, UK). 4. Cryostat (Leica, Nussloch, Germany). 5. Cryostat stage disinfectant, i.e. Cryofect (Leica). 6. DEPC-treated glassware. 7. 70% ethanol (VWR). 8. Staining reagents: Mayer’s haematoxylin, alcoholic eosin 1%, ethanol (100%), xylene (all VWR), and nuclease-free distilled water. 9. Vacuum dessicator. 10. Pixcell II™ microdissector (Arcturus Bioscience Inc., Mountain View, CA). 11. CapSure Macro LCM caps (Arcturus). 12. PicoPure RNA extraction kit™ (Arcturus). 2.2. RNA Extraction and Quality Control

1. PicoPure RNA extraction kit™ (Arcturus). 2. 0.5 and 1.5 ml nuclease-free microcentrifuge tubes. 3. Microcentrifuge (i.e. Eppendorf, Beckmann Coulter). 4. RNase-free DNase kit (Qiagen Ltd., Crawley, West Sussex, UK). 5. Agilent 2100 BioAnalyser (Agilent Technologies UK Limited, Stockport, Cheshire, UK). 6. cDNA integrity kit (KPL, Gaithersburg, MA) (optional). 7. RNA 6000 Nano assay (Agilent) (optional).

2.3. RNA Amplification

1. 0.5 and 1.5 ml nuclease-free microcentrifuge tubes. 2. Microcentrifuge. 3. Tetrad™ thermocycler (MJResearch Inc, Reno, NV). 4. SMART™ cDNA Synthesis Kit (Clontech, Palo Alto, CA). 5. PowerScript™ reverse transcriptase (Clontech, Palo Alto, CA). 6. Advantage® 2 PCR Kit (Clontech, Palo Alto, CA). 7. RNase-free water (Ambion, Austin, TX). 8. 0.2 ml thin wall tubes (Clontech).

2.4. cDNA Labelling

1. 10× low-C dNTP mix (5 mM dGTP, dTTP, dATP, and 2 mM dCTP) (Sigma). 2. RNase-free water (Ambion, Austin, TX). 3. BioPrime DNA Labelling system (Invitrogen, Carlsbad, CA). 4. Cy3-dCTP and Amersham, UK).

Cy5-dCTP

(Amersham

Biosciences,

206

C.C. Thorn et al.

5. Tetrad™ thermocycler (MJResearch Inc, Reno, NV). 6. 0.5 and 1.5 ml nuclease-free microcentrifuge tubes. 2.5. Labelled cDNA Purification

1. Sephadex® G-50 superfine (Sigma, St. Louis, MO). 2. MilliporeColumn loader (Millipore, Billerica, MA). 3. MultiScreen HV 96-well plate (Millipore). 4. RNase-free water (Ambion). 5. Centrifuge Alignment Frame (Millipore). 6. 96-Well plate (Corning®, Corning, NY). 7. Jouan RCMB centrifuge (Thermo Electro Corp, TX). 8. Milli-Q water (Millipore). 9. 0.5 and 1.5 ml nuclease-free microcentrifuge tubes. 10. 3 M sodium acetate (pH 5.2) and 100% ethanol (VWR). 11. Microcentrifuge. 12. Ethanol 70% (VWR). 13. Vacuum drier.

2.6. Dye Quantification and Hybridisation

1. Hybridisation buffer: 40% deionised formamide, 5× standard sodium citrate, 5× Denhardt’s solution, 1 mM sodium pyrophosphate, 50  mM Tris pH 7.4, and 0.1% SDS (VWR). Prepare using 0.22 mm syringe filter. 2. Nanodrop™ 1000 (Thermo Scientific, Wilmington, DE). 3. Heating blocks. 4. Microcentrifuge. 5. Lucidea SlidePro hybridiser Amersham, UK) (see Note 13).

(Amersham

Biosciences,

6. Hamilton syringe with pst 5 needle (Hamilton, Bonaduz, Switzerland). 7. Spotted oligonucleotide microarray slides (see Note 13). 8. Agilent DNA microarray scanner (Agilent). 9. Imagene Version 5.0 (BioDiscovery, El Segundo, CA).

3. Methods The use of this combination of techniques requires careful attention to (a) the acquisition of sufficient quantities of highquality RNA, (b) the use of equal amounts of input total RNA for amplification and labelling, and (c) the maintenance of relative transcript abundances during amplification. Methods used for laser-capture of tumour epithelium, RNA extraction and quality

17  Oligonucleotide Microarray Expression Profiling of Contrasting Invasive Phenotypes…

207

control, RNA amplification, labelling and purification, cDNA hybridisation, and data analysis are described in sequence. 3.1. Tissue Processing, Laser-Capture Dissection and RNA Extraction

The acquisition of high-quality RNA from microdissected specimens provides the initial challenge. RNA degradation may occur at different rates with different transcripts and thus potentially distort results. It has been recommended that cutting, staining and dissection all are carried out within a 3–4-h time period (see Note 1). Despite advances in the retrieval of RNA from paraffin-embedded media (25), it remains the case that fresh frozen tissue provides the highest quality RNA. Results from microarray experiments require quantitative validation, typically with real-time PCR. The relatively large quantity of microdissected tissue harvested using this technique provided sufficient material for triplicate microarray hybridisations and triplicate real-time PCR analyses. 1. Tissue samples should be carefully oriented in embedding media on a cork board and snap frozen in liquid propan-2-ol (VWR) cooled in liquid nitrogen. Samples were processed within 45 min of extraction and stored at −80°C. 2. Allow samples to equilibrate in the cryostat for 10–20 min. Cut sections to a thickness of 6 mm (see Note 2) at −23°C on the cryostat and transfer to an untreated glass slide (Superfrost). Good quality transfer is essential to preserve architecture necessary to guide microdissection and folded sections should be rejected. Immediately fix in chilled 70% ethanol at 4°C for 2 min before storing on dry ice prior to staining within 1 h. Additional sections can be processed in parallel to provide additional material for the assessment of RNA integrity. 3. Haematoxylin and eosin staining is performed within 60 min of sectioning using the following optimised protocol in a fume hood (Table  1). All reagents were molecular biology grade and all glassware treated with DEPC (see Note 3). 4. Following staining, air dry the slides for 15 min in order to completely evaporate the xylene. Processed slides may be stored temporarily in a vacuum dessicator containing silica gel. Longer term storage at −80°C is not recommended due to the deleterious effects of the freeze thaw cycle on RNA integrity (15). 5. LCM is performed with the Pixcell II™ microdissector with CapSure Macro LCM caps within 2 h of staining. The standard laser pulse settings were size 15 mm, power 40–70 mW, and duration 3 ms. These parameters may require some alteration depending on transfer quality. Between 6,000 and 26,000 laser shots are required per sample, the variability relating to the quality of cellular transfer and the architecture of the tissue of interest.

208

C.C. Thorn et al.

Table 1 Protocol used for the staining of colorectal specimens prior to LCM (see Note 4) Step

Solution

Duration

1

Deionised water

10 s

2

Mayer’s haematoxylin

10 s

3

Deionised water

10 s

4

Alcoholic eosin Y 1%

15 s

5

Ethanol 95%

30 s

6

Ethanol 100%

a

30 s

7

Xylene

b

3 min

8

Xyleneb

3 min

 Freshly dispensed to prevent ambient humidity compromising efficacy of dehydration  To achieve the complete removal of water and ethanol residues from the sample

a

b

Fig. 1. Representative images of microdissection in normal mucosa (upper panel ) and the infiltrative growth pattern (lower panel ). Scale bar = 200 mm.

6. Digital photographs of representative areas may be acquired before and after LCM. Reference to these images is found to be a convenient means of quality control (Fig. 1). 7. RNA extraction using the PicoPure™ RNA isolation kit.

17  Oligonucleotide Microarray Expression Profiling of Contrasting Invasive Phenotypes…

209

1. Place CapSure Macro LCM cap with adherent microdissected cells onto a 0.5-ml microcentrifuge tube containing 50 ml extraction buffer (contains guanidine thiocyanate and Triton X-100). 2. Invert tube and incubate at 42°C for 30  min to allow digestion of microdissected cells. If multiple caps have been used, it is recommended that separate extractions are performed rather than transferring caps to the same aliquot of extraction buffer. 3. Centrifuge at 800 × g for 2  min and transfer to a fresh microcentrifuge tube, label, and store at −80°C prior to RNA isolation which can be performed in batches to minimise experimental variation. 3.2. RNA Isolation and Quality Control

A number of commercially available systems are now available for RNA isolation. Exquisite care was taken to treat all surfaces for RNase contamination and gloves were used and changed regularly throughout the procedure. 1. Remove RNA extracts from storage and thaw. 2. Pre-condition the purification column by incubating 250 ml conditioning buffer (CB) on the column membrane for 5 min at room temperature before centrifugation at 16,000 × g for 1 min to wash through. 3. Add 50  ml ethanol 70% to each extract and mix by gentle pipetting. Multiple extractions from the same biological sample may be pooled and combined with an equal volume of ethanol. Pipette sample into the purification column. 4. Centrifugation at 100 × g for 2  min binds RNA to the column. Thereafter, centrifuge at 16,000 × g for 30 s. 5. Place 100 ml wash buffer 1 (alcohol based) into the column and centrifuge for 2 min at 8,000 × g. 6. RNase-free DNase kit. (see Note 5): add 5 ml DNAse I Stock Solution to 35 ml buffer RDD and mix by inversion. Add the 40 ml aliquot to each column and incubate for 15 min at room temperature. 7. Place 100 ml wash buffer 2 (alcohol based) into the column and centrifuge for 2 min at 8,000 × g. 8. Repeat step 7 but centrifuge for 2  min at 16,000 × g. Centrifuge again at 16,000 × g for 1  min should any wash buffer remain in the column. 9. Transfer column to a fresh microfuge tube and add 11 ml elution buffer directly onto the column surface. Incubate for 1 min, centrifuge at 1,000 × g for 1 min to distribute buffer into the column and then centrifuge at 16,000 × g for 1 min. If pooling was not performed in step 3, elutants may be pooled and concentrated in a centrifugal evaporator.

210

C.C. Thorn et al.

10. Remove 1.5 ml for RNA quality control and store the remainder at −80°C. The quality and quantity of the extracted total RNA were assessed using the Agilent 2100 BioAnalyser running the RNA 6000 Nano assay. This technology quantifies total RNA using spectrophotometric principles and provides quality assurance based upon the 28S to 18S rRNA ratio (see Note 6) (Fig.  2). Analysis demonstrated a yield of 30 ng of total RNA per 1,000 shots at a concentration of 29.5 ng/ml. Samples were rejected if the 28S:18S ratio was less than 1.6 although the technique has yielded results at lower ratios (~1.2). The RNA integrity number (RIN), provided by Agilent platforms, quantifies RNA quality from the entire electrophoretic trace rather than merely the ribosomal bands and is independent of concentration (26). Total RNA samples with a RIN of over 7 have been found to yield high quality results. RNA integrity was also assessed for the presence of full length transcripts using the KPL cDNA integrity kit. 3.3. cDNA Synthesis and SMART PCR Amplification 3.3.1. First Strand Synthesis

1. For each sample, combine in a sterile 0.2 ml thin walled tube: 3 ml total RNA extract (~90 ng total RNA) (see Note 7). 1 ml of 3¢ SMART™ CDS Primer II A (12 mM)* (5¢-AAGCAGTGGTATCAACGCAGAGTACT(30)VN-3¢) 1 ml SMART II™A Oligonucleotide (12 mM)* (5¢-AAGCAGTGGTATCAACGCAGAGTACGCGGG-3¢) Vortex and spin briefly in a microcentrifuge. 2. Incubate at 72°C for 2  min in a PCR thermocycler with a heated lid, quench on ice for 2 min, and microcentrifuge to collect contents. 3. To each tube, add the following reagents from SMART™ PCR cDNA Synthesis Kit: 2 ml 5× buffer (250 mM Tris–HCl pH 8.3). 1 ml dithiothreitol (DTT) (20 mM). 1 ml 50× dNTP (10 mM). 1 ml PowerScript™ reverse transcriptase. Gently mix with a pipette and spin in a microcentrifuge. Incubate at 42°C for 1 h in a thermocycler with heated lid. Quench reaction on ice to terminate first strand synthesis.

3.3.2. Second Strand Synthesis and LongDistance PCR cDNA Amplification

1. Prepare a master mix for amplification reagents for (n + 1) reactions. For each reaction combine from Advantage® 2 PCR Kit: 10 ml 10× Advantage® 2 PCR buffer. 2 ml 50× dNTP (10 mM). 4 ml 5¢ PCR primer IIA (10 mM) (5¢-AAGCAGTGGTATCA ACGCAGAGT-3¢).

17  Oligonucleotide Microarray Expression Profiling of Contrasting Invasive Phenotypes…

211

Fig.  2. Examples of Agilent BioAnalyser output demonstrating good quality (upper), partially degraded (middle), and grossly degraded (lower ) total RNA samples.

212

C.C. Thorn et al.

2 ml 50× Advantage® 2 Polymerase Mix**. 72 ml RNase-free water. 2. To each 10 ml of cDNA template (First strand product) add 90 ml of the amplification master mix and combine contents by gently flicking tube and microcentrifuge briefly (see Note 8). 3. Load samples into a thermocycler with heated lid and subject to the following cycling programme. 95°C for 1 min. Then 19 cycles of: 95°C for 5 s. 65°C for 5 s. 68°C for 6 min. The resulting double stranded cDNA may be stored at −20°C. 3.4. cDNA Labelling

1. Prepare a 10× low-C dNTP (5 mM dATP, 5 mM dGTP, 5 mM dTTP, and 2  mM dCTP) master mix by combining: 25  ml 100 mM dNTA, 25 ml 100 mM dNTT, 25 ml 100 mM dNTG, and 10 ml 100 mM dNTC. Add 415 ml RNase-free water. Aliquot the 10× low-C dNTP mix which may be stored at −20°C. 2. Transfer a 42-ml aliquot of amplified double-stranded cDNA into a 0.2-ml thin walled tube (see Note 9). 3. Add 40  ml 2.5× Random Primer reaction buffer from BioPrime® DNA Labelling System. 4. Incubate at 95°C for 5 min and cool on ice before adding the following reagents: 5 ml 10× low-C dNTP mix. 2 ml Cy-3 or Cy-5 d-CTP fluorescent label (see Note 10). 1 ml Klenow polymerase (40U).† 10 ml RNase-free water. Incubate at 37°C for 16 h in the dark (see Note 11). The reaction is terminated by adding 5 ml stop buffer (0.5 M Na2EDTA pH 8.0) (from BioPrime® DNA Labelling System). Product is stored at 4°C in foil to prevent photo-bleaching (total of 105 ml).

3.5. Labelled cDNA Purification 3.5.1. Preparation of Purification Columns to Remove Unincorporated Cyanine Dyes

1. Load dry Sephadex® G-50 superfine gel filtration medium (Sigma) into all wells of a 45-ml MilliporeColumn loader and level off excess resin. 2. Place a MultiScreen HV 96-well plate upside-down onto the column loader. Invert both components and tap to release resin. 3. Add 300 ml RNase free water to each well and incubate for 3 h at 4°C in order to swell the dextran beads and prepare the columns (see Note 12).

17  Oligonucleotide Microarray Expression Profiling of Contrasting Invasive Phenotypes…

213

4. Place a Centrifuge Alignment Frame on top of a round-­ bottomed polypropylene 96 well plate (Corning). Pack columns by centrifugation at 1,000 rpm (910 × g) for 5 min. 5. Add 150  ml Milli-Q water to the columns and centrifuge again for 5 min at 1,000 rpm (910 × g) to remove contaminants from the resin. 3.5.2. Purification of Labelled Targets

1. Place MultiScreen HV 96-well plate columns on a new roundbottomed 96-well plate and add labelled cDNA (105 ml) to the centre of each column. Do not pool hybridisation pairs at this stage. 2. Centrifuge columns at 1,000  rpm for 5  min to collect the purified labelled cDNA. Pool the control/reference sample at this stage depending upon experimental design (Fig. 3). 3. Combine the Cy-3 and Cy-5 dye labelled targets according to the planned hybridisation pairs in a clean 1.5 ml tube. Do not attempt to normalise the samples according to dye or DNA quantification since the readings will include unincorporated dye and may thus lead to inaccurate hybridisation. 4. Add 1/10 volume of 3 M sodium acetate (pH 5.2) and 2.5 volumes of 100% ethanol. Incubate at −70°C for 30 min and centrifuge at 13,000 rpm at 4°C for 20 min to precipitate the labelled cDNA. 5. Carefully remove the supernatant and add 750  ml of 70% ethanol, centrifuge at 13,000 rpm for 5 min to remove particulate matter. Remove the supernatant, vacuum dry for 5 min, and store at −20°C.

3.6. Hybridisation

Hybridisations should be performed at least in triplicate and despite the use of lowess normalisation, dye-swaps should be undertaken in order to mitigate the unequal incorporation of the fluorescent Cy-dyes (dye-bias). 1. Resuspended amplified, labelled, and purified cDNA pellet in 20 ml hybridisation buffer. 2. Quantify labelled cDNA product using the Nanodrop™ nucleic acid quantification system for single stranded DNA and fluorescent absorption. 1 ml of aqueous sample is required although 1.5–2 ml may be required if surface tension properties or pipettor accuracy are uncertain. Outputs include Cy-dye concentrations (pg/ml) which permit an estimation of dye incorporation bias prior to for hybridisation. 3. Add an additional 230 ml hybridisation buffer to the remaining resuspended cDNA. 4. Heated at 90°C for 5 min and then at 50°C for 5 min.

214

C.C. Thorn et al.

Fig.  3. Schematic diagram demonstrating the experimental design described in the protocol.

5. Microcentrifuge at 13,000 rpm for 5 min to collect any residual precipitate and cool on ice ready for hybridisation. 6. Hybridisation was performed over 16 h at 42°C (see Note 13). 200 ml of the hybridisation co-precipitate was injected slowly, avoiding bubbles using a Hamilton syringe into each chamber of a Lucidea SlidePro hybridiser which was loaded with sequential microarrays (HGMP_Human_Hs_SGC_Av1) oligonucleotide array (ArrayExpress ID: A-MEXP-52) from the same print run. The hybridisation solution was mixed every 10 min. Washing steps were performed twice with 1× SSC and 0.2% SDS twice and then with 0.1× SSC before air drying.

17  Oligonucleotide Microarray Expression Profiling of Contrasting Invasive Phenotypes…

215

7. The slides were removed after a final washing step and scanned using the Agilent DNA microarray scanner after which the raw image data were extracted using Imagene Version 5.0. 3.7. Data Analysis

Careful quality control of biological material is the first step to the generation of reliable results from any in silico analysis. Even with robust experimental design and protocol, the findings of semiquantitative microarray experiments require validation, typically with a quantitative technique such as qPCR. It is pertinent to remember that results indicate relative transcript abundance which can derive from a variety of factors, not all relating to transcript expression (i.e. stability). There exist multiple sources of systematic bias which require consideration and microarray design includes a number of control features. Negative controls indicate the degree of nonspecific background signal present on the array whilst the positive controls provide information regarding the uniformity of hybridisation which can be affected by fluidics. Negative controls were provided by oligonucleotides representing bacterial and plant RNA, polyA oligonucleotide, and spotting buffer. Positive controls used on the platform were total human genomic DNA and ten house-keeping genes including GAPDH, ACTG1, and UQCRC2. Spiked in controls were not used in this experiment. The use of this type of control can be useful when standardisation of input total RNA amount is impractical. Oligonucleotide probes are designed to be complementary to exogenous target cDNA from bacterial or plant sources (i.e. which will not crosshybridise to human sequences). The complementary cDNA is typically added (“spiked-in”) to target samples prior to labelling to serve either as an independent positive control or, in a series of known dilutions, to assess the linearity of the hybridisation signal. Quality control spots for full length transcripts are useful with cDNA arrays but the design of oligonucleotide probes to represent the 3¢ end of mRNA species renders them unnecessary in this context. When used, a ratio of target hybridisation (5¢:3¢ end of cDNA) from 3:1 to 20:1 represents good quality amplification, depending on the housekeeper in question and the degree of amplification. Spot or probe quality control is largely automated; however, reasonable attention may need to be given to areas of the array with high background, dust, spot bleeding, spot irregularity, spot inhomogeneity, doughnut morphology, oversaturation, and other sources of artefactual signal within the constraints of practicality (27). Inaccurate spot identification, inclusion of spots with abnormal morphology (i.e. doughnuts), and distortion of background signal can lead to misleading results. Features flagged by the Imagene 5.0 software as demonstrating poor spot morphology were excluded from further analysis. A simple manual methodology

216

C.C. Thorn et al.

for assessing spot homogeneity is to assess mean vs. median pixel intensity within a spot against a pre-determined threshold. A difference of 15% may be a suitable level at which to exclude a given spot. Data were analysed using the GeneSpring® 7.2 microarray data analysis software package. Normalisation processes scale the intensity data to facilitate comparisons between arrays which can arise from variability of probe/target concentration, fluidics, or environmental factors which differ between hybridisations. The Lowess per spot and per gene normalisation protocol is engineered to eliminate intensity-dependent dye-related artefacts, where fluorescence from one dye is absorbed by the other (quenching). Per chip normalisation standardises the variations in intensity across multiple arrays, thereby enabling their direct comparison. Subsequent to image quality control and normalisation procedures, the data were filtered using three mechanisms. Low intensity data can be seen to skew the self:self hybridisation plot and an intensity threshold was determined to exclude these unreliable data (Fig. 4). Self:self hybridisations using normalised data were used to determine the appropriate cut-off for defining true differential expression. Analysis determined that at a fold change threshold of 1.7 was appropriate to detect true differential expression, consistent with previous experience in the laboratory. There is little differential expression between experimental and control samples in the majority of genes. This feature of global gene expression analysis is exploited during normalisation processes and facilitates further strategies for data quality control. Normalised gene expression was compared using the Pearson correlation co-efficient, between each tumour sample and an in silico pool of all tumour samples. Samples with a co-efficient <0.55 (50%) were rejected. This was also performed with reference to two of the technical replicates of each sample. In this case, those samples with a correlation co-efficient <0.7 (75%) were rejected. A further group was defined by the quality of hybridisation, as defined by the proportion of probes with a high quality (unflagged) signal. Hybridisations with fewer than 30% probes unflagged (10%) were excluded. Only genes identified to demonstrate differential expression over the 1.7-fold threshold in all three of these subsets of hybridisations were included in a final high-stringency dataset. At this stage, numeric data (normalised) were extracted from the software and subjected to Welch t-test analysis. Those transcripts demonstrating statistically significant differences in expression went on to be validated using real time PCR (using the same total RNA sample).

17  Oligonucleotide Microarray Expression Profiling of Contrasting Invasive Phenotypes…

217

Fig. 4. A scatterplot demonstrating the skew of self:self hybridisation data at lower intensities. A threshold was set to exclude these data.

4. Notes 1. It has been reported that quantitative real-time PCR and microarray applications are markedly more sensitive to RNA degradation than RT-PCR. RNA integrity has been demonstrated by real-time PCR analysis of housekeeper genes in non-microdissected breast cancer tissue for at least 3 h postextraction in a time-course study (14). However, others have commented that an ischaemia time of just 30 s is sufficient to significantly degrade RNA (28). RNA preservation solutions have been reported to impede thorough freezing in the embedding media and consequently may distort tissue morphology on the slide. 2. Sections have been used between 4 and 10 mm in published studies although Arcturus recommend sections of between 7 and 8 mm. Slides may be immediately stored at −70°C after

218

C.C. Thorn et al.

sectioning if required. Alternatively, slides may be stored in chilled ethanol fixative for several hours without significant RNA degradation; however, we found that sections became more difficult to microdissect after 2–3  min, possibly by reducing cell–cell dissociation. 3. DEPC degradation products can inhibit in vitro transcription with RNA polymerases and DEPC-treated water should be avoided in preference of alternative preparations, however, glassware was treated with DEPC within this protocol. Inclusion of an RNAase inhibitor in aqueous staining solutions may protect RNA from loss of gene transcripts (29). 4. LCM may be conducted without staining but dehydration is mandatory. Alternative staining reagents which have been described to provide high-quality RNA yields from microdissected tissue include methyl green, toluidine blue, and Nissl stain (30–32). Excessive dehydration results in non-specific cell lifting onto cap, reducing specificity. Inadequate dehydration or rehydration from ambient humidity reduces tissue transfer. Difficulties in achieving accurate cellular transfer are mitigated with the use of UV cutting LCM technologies. This is a faster technique and is well suited to large areas of homogeneous tissue requiring microdissection. However, disadvantages include the requirement for special section mounting and the possibility of UV damage to cut cells, more pertinent when dissecting multiple small areas. 5. A DNAase step is not strictly required when using oligonucleotide probes as it is when using cDNA probes. However, certain RNA quantification methods (using spectrophotometric data rather than ribosomal bands) will produce inaccurate estimations of RNA yields in the presence of genomic DNA. 6. Total RNA should be assessed at this stage for both quantity and quality. The use of traditional UV spectrophotometry is impractical since the RNA concentration yielded with microdissection is an order of magnitude below the detection level of the equipment (~4  ng/ml), once diluted in the required volume of solvent (50–100 ml). If total RNA concentration is very small, a parallel extraction of stained, non-microdissected tissue may be used a surrogate sample to assess RNA integrity. However, quantification is mandatory in order to permit the equalisation of input RNA for subsequent processing (this is not required when a pooled reference sample is used but is always required for test/experimental samples). The Agilent Bioanalyser 2100 is able to quantify total RNA samples from 25  ng/ml and requires only 1  ml aliquot for analysis. The NanoDrop 1000 spectrophotometer can detect total RNA samples in a volume of 1 ml from a concentration of 2 ng/ml. Regardless of the technology used, a 260/280 ratio of 1.8– 2.0 indicates a pure preparation of DNA or RNA. A ratio of

17  Oligonucleotide Microarray Expression Profiling of Contrasting Invasive Phenotypes…

219

less than 1.6 indicates contamination with protein or other organic chemicals which can inhibit downstream enzymatic reactions. The technique cannot distinguish between nucleic acid species and therefore contamination with DNA can affect results. Ratios lower than 1.6 have been found to produce reproducible microarray data. The sensitivity of quantification technologies can be enhanced up to 1,000-fold with the use of fluorescent dyes but this is impractical when using microlitre. 7. SMART amplification of 1  ng total RNA samples (cell line source) has produced reproducible microarray results in the laboratory. Many considerations may influence experimental design and are discussed comprehensively elsewhere (33). 8. With 50–1,000 ng total RNA, this protocol generates sufficient cDNA for successful hybridisation with a 2-ml aliquot of first strand synthesis product and 15–19 cycles of PCR. If the amount of first strand cDNA product is altered, adjust the volume of water accordingly. It has been noted that ex vivo material fails to amplify as well as ex vitro (cell line) material even when RNA quality is comparable. This may be due to selective mRNA degradation in clinical samples or a lower ratio of mRNA:rRNA abundance in clinical samples, giving a false impression of RNA quality from 28S:18S ratio. 9. The protocol can be adjusted to use 21 ml cDNA input cDNA for the labelling process although 42 ml cDNA appears preferable when working with clinical samples. This requires equal adjustment of the volume of random primer mix (i.e. half) but other quantities (enzyme/dye and dNTP) remain unchanged. 10. Labelling with 0.5–1  ml Cy-dye has been effective though hybridisations give slightly less signal. 11. The labelling reaction may be incubated for either 2  h or overnight (16–18 h); the latter provides higher intensity fluorescence. An increase of 10% feature identification has been observed with 10,000 spot arrays, the additional genes identified are at the lower end of the signal intensity spectrum where data may otherwise be excluded due to background noise levels. Longer incubations have been shown to increase signal intensity without distorting differential ratios. 12. Once the columns are swollen they may be stored for up to 2 weeks at 4°C, ensuring that the level of humidity is maintained by sealing with Parafilm M® or storing in a sealed plastic container containing a moist, lint-free cloth. 13. Hybridisations were conducted at a temperature 15–20°C below the Tm of the 70mer probe oligonucleotides. The relatively large hybridisation volume requires the use of a chamber (hydrophobic cover slips may be used in some circumstances with small hybridisation volumes) which precisely regulates temperature and humidity which influence

220

C.C. Thorn et al.

hybridisation quality whilst circulating the hybridisation solution to facilitate even hybridisation. The Lucidea SlidePro hybridiser has been discontinued; several manufacturers including Tecan Group Ltd. produce suitable alternatives. Spotted microarrays used in this protocol were distributed from a free resource formerly funded by the MRC which unfortunately is no longer operational. However, slide-based oligonucleotide microarrays are still available from Agilent Technologies. References 1. Wernert N (1997) The multiple roles of tumour stroma. Virchows Arch; 430: 433–43. 2. Liotta LA, Kohn EC (2001) The microenvironment of the tumour-host interface. Nature; 411: 375–9. 3. Bhowmick NA, Moses HL (2005) Tumorstroma interactions. Curr Opin Genet Dev; 15: 97–101. 4. Le NH, Franken P, Fodde R (2008) Tumourstroma interactions in colorectal cancer: converging on beta-catenin activation and cancer stemness. Br J Cancer; 98: 1886–93. 5. Nakamura T, Mitomi H, Kanazawa H et  al (2008) Tumor budding as an index to identify high-risk patients with stage II colon cancer. Dis Colon Rectum; 51: 568–72. 6. Nakahara H, Howard L, Thompson EW et al (1997) Transmembrane/cytoplasmic domainmediated membrane type 1-matrix metalloprotease docking to invadopodia is required for cell invasion. Proc Natl Acad Sci U S A; 94: 7959–64. 7. Brabletz T, Jung A, Reu S et al (2001) Variable beta-catenin expression in colorectal cancers indicates tumor progression driven by the tumor environment. Proc Natl Acad Sci U S A; 98: 10356–61. 8. Komatsu K, Kobune-Fujiwara Y, Andoh A et  al (2000) Increased expression of S100A6 at the invading fronts of the primary lesion and liver metastasis in patients with colorectal adenocarcinoma. Br J Cancer; 83: 769–74. 9. Nabeshima K, Shimao Y, Inoue T et al (2002) Immunohistochemical analysis of IQGAP1 expression in human colorectal carcinomas: its overexpression in carcinomas and association with invasion fronts. Cancer Lett; 176: 101–9. 10. Harrell JC, Dye WW, Harvell DM et al (2008) Contaminating cells alter gene signatures in whole organ versus laser capture microdissected tumors: a comparison of experimental breast cancers and their lymph node metastases. Clin Exp Metastasis; 25: 81–8.

11. Goldsworthy SM, Stockton PS, Trempus CS et al (1999) Effects of fixation on RNA extraction and amplification from laser capture microdissected tissue. Mol Carcinog; 25: 86–91. 12. Huang J, Qi R, Quackenbush J et  al (2001) Effects of ischemia on gene expression. J Surg Res; 99: 222–7. 13. Nygaard V, Hovig E (2006) Options available for profiling small samples: a review of sample amplification technology when combined with microarray profiling. Nucleic Acids Res; 34: 996–1014. 14. Ohashi Y, Creek KE, Pirisi L et al (2004) RNA degradation in human breast tissue after surgical removal: a time-course study. Exp Mol Pathol; 77: 98–103. 15. Schoor O, Weinschenk T, Hennenlotter J et al (2003) Moderate degradation does not preclude microarray analysis of small amounts of RNA. Biotechniques; 35: 1192–6, 8–201. 16. Zhao H, Hastie T, Whitfield ML et al (2002) Optimization and evaluation of T7 based RNA linear amplification protocols for cDNA microarray analysis. BMC Genomics; 3: 31. 17. Petalidis L, Bhattacharyya S, Morris GA et al (2003) Global amplification of mRNA by template-switching PCR: linearity and application to microarray analysis. Nucleic Acids Res; 31: e142. 18. Kitahara O, Furukawa Y, Tanaka T et al (2001) Alterations of gene expression during colorectal carcinogenesis revealed by cDNA microarrays after laser-capture microdissection of tumor tissues and normal epithelia. Cancer Res; 61: 3544–9. 19. Alevizos I, Mahadevappa M, Zhang X et  al (2001) Oral cancer in  vivo gene expression profiling assisted by laser capture microdissection and microarray analysis. Oncogene; 20: 6196–204. 20. Luo L, Salunga RC, Guo H et al (1999) Gene expression profiles of laser-captured adjacent neuronal subtypes. Nat Med; 5: 117–22.

17  Oligonucleotide Microarray Expression Profiling of Contrasting Invasive Phenotypes… 21. Luzzi V, Holtschlag V, Watson MA (2001) Expression profiling of ductal carcinoma in situ by laser capture microdissection and high-density oligonucleotide arrays. Am J Pathol; 158: 2005–10. 22. Miura K, Bowman ED, Simon R et al (2002) Laser capture microdissection and microarray expression analysis of lung adenocarcinoma reveals tobacco smoking- and prognosis-related molecular profiles. Cancer Res; 62: 3244–50. 23. Zhu G, Reynolds L, Crnogorac-Jurcevic T et  al (2003) Combination of microdissection and microarray analysis to identify gene expression changes between differentially located tumour cells in breast cancer. Oncogene; 22: 3742–8. 24. Thorn CC, Freeman TC, Scott N et al (2009) Laser microdissection expression profiling of marginal edges of colorectal tumours reveals evidence of increased lactate metabolism in the aggressive phenotype. Gut; 58: 404–12. 25. Hewitt SM, Lewis FA, Cao Y et  al (2008) Tissue handling and specimen preparation in surgical pathology: issues concerning the recovery of nucleic acids from formalin-fixed, paraffin-embedded tissue. Archives of pathology & laboratory medicine; 132: 1929–35. 26. Schroeder A, Mueller O, Stocker S et al (2006) The RIN: an RNA integrity number for assigning integrity values to RNA measurements. BMC molecular biology; 7: 3.

221

27. Knapen D, Vergauwen L, Laukens K et  al (2009) Best practices for hybridization design in two-colour microarray analysis. Trends in biotechnology; 27: 406–14. 28. Simone NL, Bonner RF, Gillespie JW et  al (1998) Laser-capture microdissection: opening the microscopic frontier to molecular analysis. Trends Genet; 14: 272–6. 29. Wang H, Owens JD, Shih JH et  al (2006) Histological staining methods preparatory to laser capture microdissection significantly affect the integrity of the cellular RNA. BMC Genomics; 7: 97. 30. Bahn S, Augood SJ, Ryan M et al (2001) Gene expression profiling in the post-mortem human brain–no cause for dismay. Journal of chemical neuroanatomy; 22: 79–94. 31. Betsuyaku T, Griffin GL, Watson MA et  al (2001) Laser capture microdissection and realtime reverse transcriptase/ polymerase chain reaction of bronchiolar epithelium after bleomycin. Am J Respir Cell Mol Biol; 25: 278–84. 32. Pan J, Kunkel EJ, Gosslar U et  al (2000) A novel chemokine ligand for CCR10 and CCR3 expressed by epithelial cells in mucosal tissues. J Immunol; 165: 2943–9. 33. Causton HC, Quackenbush J, Brazma A (2003) Microarray Gene Expression Data Analysis: A Beginner’s Guide. Blackwell, Oxford.

wwwwwww

Chapter 18 Evaluation of Gastrointestinal mtDNA Depletion in Mitochondrial Neurogastrointestinal Encephalomyopathy (MNGIE) Carla Giordano and Giulia d’Amati Abstract Mitochondrial neurogastrointestinal encephalomyopathy (MNGIE) is a rare disease characterized by severe gastro-intestinal (GI) dysmotility caused by mutations in the thymidine phosphorylase gene. Thymidine phosphorylase (TP) is involved in the control of the pyrimidine nucleoside pool of the cell. Reduced TP activity induces nucleotide pool imbalances that in turn affect both the rate and fidelity of mtDNA replication, leading to multiple deletions and depletion of mtDNA. By using laser capture microdissection and quantitative real-time-polymerase chain reaction technique, we showed that depletion of mitochondrial DNA (mtDNA) is the most prominent molecular defect in the gut wall of MNGIE patients. Depletion affects severely the smooth muscle cells of muscularis propria and the skeletal muscle component of the upper esophagus, while ganglion cells of the myenteric plexus show only a milder mtDNA reduction. Key words: Mitochondrial disease, mtDNA depletion, mtDNA deletion, CPEO

1. Introduction Human mitochondria are double-membrane cytoplasmic organelles that carry multiple copies of their own 16,569-bp circular mitochondrial DNA (mtDNA) (1). The number of mtDNA molecules per cell varies in different physiologic and pathologic conditions, according to the specific cell type. Point mutation and large-scale deletions of mtDNA as well as reduction of mtDNA copy number (i.e., mtDNA depletion) are the causes of a heterogeneous group of multisystem disorders characterized by reduced ATP production (2). Among these, mitochondrial neurogastrointestinal encephalomyopathy (MNGIE) is a rare disease characterized by severe

Graeme I. Murray (ed.), Laser Capture Microdissection: Methods and Protocols, Methods in Molecular Biology, vol. 755, DOI 10.1007/978-1-61779-163-5_18, © Springer Science+Business Media, LLC 2011

223

224

C. Giordano and G. d’Amati

gastro-intestinal (GI) dysmotility caused by mutations in the thymidine phosphorylase gene (TYMP ) (3). Thymidine phosphorylase (TP) is involved in the control of the pyrimidine nucleoside pool of the cell. Reduced TP activity induces nucleotide pool imbalances that in turn affect both the rate and fidelity of mtDNA replication, leading to multiple deletions and depletion of mtDNA (4). By using laser capture microdissection (LCM) and quantitative real-time-polymerase chain reaction (RT-PCR) technique, we showed that depletion of mtDNA is the most prominent molecular defect in the gut wall of MNGIE patients. Depletion affects severely the smooth muscle cells of muscularis propria and the skeletal muscle component of the upper esophagus, while ganglion cells of the myenteric plexus show only a milder mtDNA reduction (5, 6).

2. Materials 2.1. Hematoxylin and Eosin Stain

1. Mayer hematoxylin (Sigma-Aldrich, St. Louis, MO). 2. Eosin Y, disodium salt (Sigma-Aldrich, St. Louis, MO). Prepare 1% solution in distilled water, store at room temperature. 3. Phloxine B (Sigma-Aldrich, St. Louis, MO). Prepare 1% solution in distilled water, store at room temperature. 4. Eosin/Phloxine B solution: 100 ml of 1% eosin Y, 10 ml of 1% phloxine B, 780 ml of 95% alcohol, 4 ml of glacial acetic acid. Mix well, store at room temperature.

2.2. Combined Cytochrome c Oxidase/Succinic Dehydrogenase Histochemical Stain

1. Cytochrome c from horse heart (Sigma-Aldrich, St. Louis, MO). 2. 3,3¢ Diaminobenzidine tetrahydrochloride (DAB, SigmaAldrich, St. Louis, MO). DAB is carcinogenic and careful handling is required. 3. Catalase from bovine liver, lyophilized powder, activity 2,000–5,000  units/mg protein, from Sigma-Aldrich, St. Louis, MO. Store at −20°C. Make fresh as required 2% solution in distilled water. 4. Cytochrome c oxidase (COX) incubating medium: 9  ml of 0.05  M phosphate buffer pH 7.2, 20  mg cytochrome c, 20 mg DAB, 750 mg sucrose. Mix well then add 1 ml of 2% catalase solution. Adjust pH to 7.3. Store in single use aliquots (100 ml) at −20°C. 5. Nitroblue tetrazolium (NBT) solution: 25  mM KH2PO4, 10 mM NBT (Sigma-Aldrich, St. Louis, MO). Adjust pH to 7.3 with HCl. Make fresh as required.

18  Evaluation of Gastrointestinal mtDNA Depletion in Mitochondrial¼

225

6. Sodium succinate solution: 25 mM KH2PO4, 150 mM Na2HPO4, 0.2 M sodium succinate. Make fresh as required. 7. Succinic dehydrogenase (SDH) incubating medium: add 17.5 ml of sodium succinate solution to 2.5 ml of NBT solution. Adjust pH to 7.3 with HCl. Store in single use aliquots (100 ml) at −20°C (see Note 1). 2.3. Laser Capture Microdissection

1. Inverted microscope (Nikon Instruments, Kingston, UK ) equipped with MMI UV-CellCut System and MMI CellTools software (Molecular Machines & Industries, Glattbrug, Germany). 2. Membrane slides for laser microdissection (Molecular Machines & Industries, Glattbrug, Germany). 3. Tube with adhesive lid, with diffuser, 500  ml (Molecular Machines & Industries, Glattbrug, Germany).

2.4. DNA Extraction

1. PicopureTM DNA extraction kit (Arcturus, Los Altos, CA).

2.5. Quantification of mtDNA by RealTime PCR

1. 7500 Fast Real-Time PCR System (Applied Biosystem, Foster City, CA) or an equivalent quantitative PCR system. 2. MicroAmp® optical 96-well reaction plate (Applied Biosystem, Foster City, CA). 3. MicroAmp® optical adhesive film (Applied Biosystem, Foster City, CA). 4. TaqMan® Universal PCR Master mix containing AmpErase Uracyl-N-Glycosilase (UNG) (Applied Biosystem, Foster City, CA). Store at −20°C. Avoid thawing and re-freezing. 5. Forward and reverse oligonucleotide primers for amplification of a mitochondrial DNA fragment (MT-ND2 gene) (Applied Biosystem, Foster City, CA): F 5¢-CAC AGA AGC TGC CAT CAA GTA TTT-3¢, R 5¢-CCG GAG AGT ATA TTG TTG AAG AGG A-3¢ Prepare 100  mM stock solution in distilled water. Store at −20°C. 6. Forward and reverse oligonucleotide primers for amplification of a nuclear DNA fragment (FasL gene) (Applied Biosystem, Foster City, CA): F 5¢-TCT GTG AGG GAT ATA AAG ACA TGC A-3¢ R 5¢-ACG CAC CGG CAG GAA A-3¢ Prepare 100 mM stock solution in distilled water. Store at −20°C. 7. TaqMan® probe for mtDNA with the 6-carboxy fluorescein (FAM) as a fluorescent reporter dye at the 5¢ end the 6-carboxytetramethyl-rhodamine (TAMRA) as a quencher dye at the 3¢ end (Applied Biosystem, Foster City, CA): 5¢FAM-CAA GCA

226

C. Giordano and G. d’Amati

ACC GCA TCC A-TAMRA3¢. Prepare 100 mM stock solution in distilled water. Aliquot and store at −20°C. Light sensitive. 8. TaqMan® probe for nuclear DNA with the VIC fluorophore at the 5¢ end and a TAMRA quencher at the 3¢ end: 5¢VICCAT GTG GTT CTC TGT TCC-TAMRA3¢. Prepare 100 mM stock solution in distilled water. Aliquots and store at −20°C. Light sensitive.

3. Methods Hematoxylin and eosin stain allow recognition of tissue morphology under light microscopy. However, when frozen tissue is available, combined COX/SDH stain allows detection of respiratory chain deficient cells on sections. SDH is a respiratory chain complex entirely encoded by nuclear DNA, whereas 3 of the 13 subunits of COX are encoded by mtDNA (1). Combining the histoenzymatic mitochondrial activities of COX (orange) and SDH (blue) results in a brown stain. Single cells with mtDNA defects affecting mitochondrial protein synthesis (i.e., mtDNA deletions and/or depletion) lose their COX activity; however, the activity of the nuclear-encoded SDH is unchanged, and is highlighted in blue on the brown background. Quantitative real-time PCR is used to evaluate mtDNA depletion. The technique involves obtaining the ratio of an unknown variable (number of copies of mtDNA) to a known factor (number of copies of a nuclear DNA gene). By assuming each cell contains two copies of each nuclear chromosome, the ratio of nDNA to mtDNA values could be used to calculate a per cell mtDNA copy number (7). 3.1. Preparation of Specimen

1. Serial 5-mm thick sections from formalin-fixed paraffinembedded samples (see Note 2) or frozen tissue sections are placed directly on the membrane. Paraffin sections can be stored dry at room temperature before performing hematoxylin and eosin stain. Frozen sections are stored at −20°C wrapped in cling-film and allowed to dry fully before use. Avoid thawing and refreezing sections, as artifacts may occur. Freshly cut 8–10-mm thick frozen sections that are needed for combined COX/SDH staining. If sections are too thick they may come off the slide during subsequent procedures. In such a case, standard protocols can be used to increase section adhesion (i.e., use of poly-l-lysine coating). 2. Staining may be done either by immersing the membranes in a 50-ml coplin jar or by adding the incubating solution on the sections in a moisturized container (i.e., Petri dish with moistened filter paper to prevent drying).

18  Evaluation of Gastrointestinal mtDNA Depletion in Mitochondrial¼

3.2. Hematoxylin and Eosin

227

1. Deparaffinize sections in two changes of xylene, 10 min each. 2. Rehydrate in two changes of absolute alcohol, 5  min each; then 95% alcohol and 70% alcohol for 2 min. 3. Wash briefly in distilled water. 4. Stain in Mayer hematoxylin solution for 5–8 min. 5. Rinse in distilled water. 6. Wash in warm running tap water for 10–15 min. 7. Rinse in distilled water. 8. Counterstain in eosin/phloxine B solution for 15 s. 9. Dehydrate through 95% alcohol, two changes of absolute alcohol, 15 s each. 10. Clear in two changes of xylene, 5 min each. 11. Store at room temperature wrapped in cling-film.

3.3. Combined COX/ SDH Stain

1. Place a few drop of COX incubating medium on the section, ensuring that it is completely covered. Incubate for 1  h at room temperature. 2. Drain (do not rinse the section) and place a few drop of SDH incubating medium. Incubate for 1  h at 37°C in damp atmosphere. 3. Wash in distilled water for two times, dehydrate in ascending alcohol series (50%, 70%, 90%, and two changes of 100%), clear in two changes of xylene. 4. Store at room temperature wrapped in cling-film (representative COX/SDH stains are shown in Figs. 1 and 2).

3.4. Laser Capture Microdissection

1. Cell selection is performed under direct microscopic visualization. The dissection is carried out by an ultraviolet laser, which performs circumferential dissection of selected tissue areas, following precisely a drawn incision path. The microdissected tissue areas may be measured and documented. Between 100 and 200 smooth muscle cells and 15–20 skeletal muscle fibers are collected and pooled for analyses (example of tissue microdissection is shown in Fig. 2).

3.5. One-Step DNA Extraction

1. The procedure is based on the digestion of samples with proteinase K without organic extraction or use of spin columns. This allows the maximum DNA recovery from very small amount of sample. These instructions assume the use of the PicopureTM DNA extraction Kit, however, similar procedures may apply to different commercial kits (e.g., Pinpoint Slide DNA Isolation System™ from Zymo Research). It is critical to use high quality proteinase K avoiding thawing and refreezing of the proteinase K solution (see Note 3).

228

C. Giordano and G. d’Amati

a

IL

EL

b

* c

100

mtDNA copy/cell log

* 10

*

1 IL

EL

MP

Fig. 1. Evaluation of mtDNA amount on small intestinal wall of MNGIE patients and controls. (a) Combined COX/SDH stain on small intestine wall of a normal individual. The external layer (EL) of muscularis propria shows weaker stain intensity as compare to the internal layer (IL), suggesting a lower amount of mitochondria at this site (original magnification 10×). (b) Small intestine wall of a MNGIE patient before (right ), during (middle), and after (left ) laser microdissection of a ganglion cell of the myenteric plexus. Note the marked atrophy of the external layer (EL) of muscularis propria (asterisk ) (original magnification 40×). (c) Real-time PCR evaluation of mtDNA amount on microdissected tissues from small intestine wall of MNGIE patients (n = 5, open circle) and age-matched autopsy control (n = 10, black circle). In controls, the external layer of muscularis propria shows a lower amount of mtDNA copy/cell as compared to the internal layer (ratio 1/2). MNGIE patients present a significant reduction of mtDNA amount in all segment of small intestine wall, with the external layer of muscularis propria showing the lowest amount of mtDNA. MP myenteric plexus, IL internal layer, EL external layer of muscularis propria. Data are expressed as the mean value of three repeated measurements.

2. Pipette 30 ml of proteinase K extraction solution into the tube with adhesive lid. Invert the tube and shake down the extraction solution ensuring that the adhesive lid is completely covered. Seal with parafilm. 3. Incubate the inverted tube at 65°C in damp atmosphere for 16–24 h (4–5 h at 56°C are enough if using home-made buffer with proteinase K from Roche).

18  Evaluation of Gastrointestinal mtDNA Depletion in Mitochondrial¼

a

c

229

b

160 140 120 100 80 60 40 20 0

controls

COX+

COX− patients

Fig. 2. Evaluation of mtDNA amount on skeletal muscle of a MNGIE patient and controls. (a) Combined COX/SDH stain on skeletal muscle from a MNGIE patient. Note the presence of COX-deficient fibers (arrows) staining intensely dark with the SDH reaction (original magnification 20×). (b) Skeletal muscle before (top), during (middle), and after (bottom) laser microdissection of a COX-deficient muscle cell (original magnification 40×). (c) Real-time PCR evaluation of mtDNA amount on microdissected COX-positive and COX-deficient skeletal muscle fibers from two MNGIE patients and five agematched controls. Both COX-positive and COX-deficient skeletal muscle fibers from MNGIE patients show a marked reduction in mtDNA amount. The COX-deficient cells present the lowest mtDNA amount. Data are expressed as the mean value of three repeated measurements.

4. After incubation centrifuge the tubes for 1 min at 1,000 × g. 5. Inactivate proteinase K by heating at 95°C for 10 min. 6. Cool the sample to room temperature and store at 4°C (for up few days) or at −20°C avoiding thawing and refreezing (see Note 4). The sample is ready to be used in a PCR reaction (see Note 5). 3.6. Evaluation of mtDNA Copy Number by Quantitative RT-PCR

1. These instructions assume the use of the 7500 Fast Real-Time PCR System and TaqMan® Universal PCR Master mix from Applied Biosystem (Foster City, CA), however, similar procedures may apply to different RT-PCR machines. Prepare the reaction mix for the two targets separately, according to the

230

C. Giordano and G. d’Amati

Applied Biosystem 7500 Real-Time PCR system Guide. Include 900 nM of each primers, 250 nM probe, 1× TaqMan Universal Master mix and water to get a total volume of 20 ml for one reaction. Do not add template DNA at this stage, you will add when preparing the reaction plate (see later). Calculate the volume of the reaction mix based on the number of samples you wish to run in the same experiment (each sample is run in triplicate) and on the volume of template DNA that you wish to add for each sample (the total volume of 20 ml includes the template DNA, therefore you have to subtract 3–5 ml of template volume from the volume of water that you add to the reaction mix; see Note 6). It is critical to run a negative control containing water instead of sample within each experiment. Include an excess reaction volume of at least 10% to account for the loss that occurs during pipetting. 2. Thaw Master mix and probes by placing them on ice. Mix the Master mix by swirling the tube, vortex primers and probe, then centrifuge the tube briefly. 3. Mix the reaction mix components in each tube by gently pipetting. Be very precise during pipetting, since real-time PCR quantification is very sensitive to little volume differences. Avoid to create bubbles when pipetting. 4. Centrifuge the tubes briefly to remove air bubbles. 5. Place the reaction mix on ice protected from light while preparing the reaction plate. 6. Prepare the reaction plate by adding the reaction mix and the template DNA to the appropriate wells in the reaction plates (e.g., add 16 ml reaction mix + 4 ml template DNA). 7. Seal the reaction plate with optical adhesive film (adhesive side facing the reaction plate). Be sure the film completely covers all the reaction plate. Ensure good contact between the film and the entire plate surface. 8. Centrifuge the reaction plate briefly to remove air bubbles. Verify that the reaction mix is at the bottom of each well. Place the reaction plate at 4°C protected from light until you are ready to perform the run. 9. Run the Comparative CT Experiment according to the Applied Biosystem 7500 Real-Time PCR Systems software. Thermocycling conditions are 50°C for 2  min for optimal UNG enzyme activity, 95°C for 10 min for UNG denaturation and Taq polymerase activation, and 40 cycles of 95°C for 15 s and 60°C for 1 min. 10. The experiment is analyzed according to Applied Biosystem 7500 Real-Time PCR Systems software. The amount of mtDNA relative to genomic DNA is evaluated with the ­comparative Ct method (2DCt) where DCT = CT,mtDNA − CT,nuclearDNA.

18  Evaluation of Gastrointestinal mtDNA Depletion in Mitochondrial¼

231

A CT value >8 and <35 is desirable. A CT value of >35 indicate too low amount of target in the reaction. In this case, increase the amount of microdissected cells in starting sample. Alternatively, proteinase K could have not been inactivated or PCR inhibitors may be present in sample. 11. For the DCt calculation to be valid, the amplification efficiencies of the mitochondrial and nuclear DNA must be similar. We recommended generating a standard curve for mtDNA and nDNA using serial dilutions (10–10−6) of a DNA sample (see Note 7). The amplification efficiency is calculated using the slope of the regression line in the standard curve. A slope close to −3.32 indicates a 100% PCR amplification efficiency.

4. Notes 1. All solutions should be kept sterile. 2. Fixation type and time may influence DNA preservation. Formaldehyde leads to the generation of cross-linking between nucleic acids and protein, and causes DNA fragmentation (8). In addition cross-linking may inhibit the PCR amplification (8). We recommended using 10% buffered formalin at room temperature for 24–48 h. 3. We have got optimal results using proteinase K from Roche (prepare stock solution 10 mg/ml in distilled water, aliquot, and store at −20 °C), and the following DNA extraction buffer: 100 mM Tris–HCl pH 8,5; 1 mM EDTA; 1% Tween-20 (9). The extraction buffer is autoclaved and stored at 4°C. Add 20  mg proteinase K per 100  ml extraction buffer just before use. 4. Thawing and refreezing of DNA greatly affects reproducibility of results, probably because of differential nuclear and mitochondrial DNA fragmentation and/or degradation. 5. Additional purification of DNA is usually not required. Noticeably, DNA samples obtained with the single step extraction method contain significant quantities of impurities (i.e., hematoxylin) that, if present in high concentration, may inhibit PCR reaction. Purification of DNA decreases the amount of inhibitors, however, it severely reduces DNA recovery. 6. Sample impurities strikingly interfere with the accurate spectrophotometric measurement of DNA, leading to sample overestimate. Thus, we believe this method is useless to measure small quantities of not purified DNA samples. According to this specific protocol, a sample volume of at least 3–5  ml is necessary to achieve DNA amplification by RT-PCR.

232

C. Giordano and G. d’Amati

7. To set up the method we performed a standard curve for mtDNA and nDNA using serial known dilutions of a vector (kind gift from Professor Andrea Cossarizza) in which the regions used as template for the two amplifications were cloned tail to tail, to have a ratio of 1:1 of the reference molecules. Alternatively, two separated standard curves for mtDNA and nuclear DNA can be generated using serial dilution (10–106) of DNA obtained from peripheral lymphocytes (stock solution 200 ng/ml).

Acknowledgments The authors wish to thank Doctor Martina Leopizzi and Doctor Maurizia Orlandi for technical assistance. This work was supported by Fondazione Giuseppe Tomasello and Associazione Serena Tallarico per i Giovani nel Mondo. References 1. Anderson, S., Bankier, A.T., Barrell, B.G., de Bruijn, M.H., Coulson, A.R., Drouin, J., et  al. (1981) Sequence and organization of the human mitochondrial genome. Nature. 290, 457–465. 2. DiMauro, S., Schon, E.A. (2003) Mitochondrial respiratory-chain diseases. N Engl J Med. 348, 2656–2668. 3. Nishino, I., Spinazzola, A., Papadimitriou, A., Hammans, S., Steiner, I., Hahn, C.D., et  al. (2000) Mitochondrial neurogastrointestinal encephalomyopathy: an autosomal recessive disorder due to thymidine phosphorylase mutations. Ann Neurol. 47, 792–800. 4. Spinazzola, A., Marti, R., Nishino, I., Andreu, A.L., Naini, A., Tadesse, S., et  al. (2002). Altered thymidine metabolism due to defects of thymidine phosphorylase. J Biol Chem. 277, 4128–4133. 5. Giordano, C., Sebastiani, M., Plazzi, G., Travaglini, C., Sale, P., Pinti, M., et al. (2006) Mitochondrial neurogastrointestinal encephalomyopathy: evidence of mitochondrial DNA depletion in the small intestine. Gastroenterology. 130, 893–901.

6. Giordano, C., Sebastiani, M., De Giorgio, R., Travaglini, C., Tancredi, A., Valentino, M.L., et  al. (2008) Gastrointestinal dysmotility in mitochondrial neuro-gastrointestinal encephalomyopathy is caused by mitochondrial DNA depletion. Am J Pathol. 173, 2120–29. 7. Cossarizza, A., Riva, A., Pinti, M., Ammannato, S., Fedeli, P., Mussini, C., et  al. (2003) Increased mitochondrial DNA content in peripheral blood lymphocytes from HIVinfected patients with lipodystrophy. Antivir Ther. 8, 51–57. 8. Gilbert, M.T., Haselkorn, T., Bunce, M., Sanchez, J.J., Lucas, S.B., Jewell, L.D., et al. (2007) The isolation of nucleic acids from fixed, paraffin-embedded tissues—which methods are useful when? PLoS ONE. 2, p. e537. 9. He, L., Chinnery, P.F., Durham, S.E., Blakely, E.L., Wardell, T.M., Borthwick, G.M., et al. (2002). Detection and quantification of mitochondrial DNA deletions in individual cells by real-time PCR. Nucleic Acids Res. 30, e68.

Chapter 19 Laser Microdissection for Gene Expression Study of Hepatocellular Carcinomas Arising in Cirrhotic and Non-Cirrhotic Livers Maria Tretiakova and John Hart Abstract Laser microdissection (LMD) is a robust well-established technology for the isolation of chosen cell populations from surrounding tissues and cells. This technique is particularly useful to minimize bias inherent in the molecular analysis of highly heterogeneous whole tissue sections. The aim of this study was to identify the pattern of mRNA expression in hepatocellular carcinoma (HCC) arising in cirrhotic liver and compare it to the pattern of expression in HCC arising from non-cirrhotic liver. The expression profiles of the tumors were also compared to that of the surrounding liver (either cirrhotic or non-cirrhotic) from the same patient. In addition, the expression pattern of each of the four tissues were compared to normal hepatic tissue. Samples of HCC tissue and surrounding cirrhotic or non-cirrhotic parenchyma were collected at the time of resection or liver transplantation. The samples were snap frozen and stored at −80°C. The snap frozen samples were then cryosectioned and stained with hematoxylin and eosin for LMD. Hepatocytes from each sample were collected using the Leica LMD instrument. The RNA was extracted according to standard methodology and amplified. Microarray analysis was performed using the Affymetrix human genome array platform. The resulting microarray data were analyzed using Affymetrix Microarray Suite 5.0 (MAS 5.0). Results were displayed using Genespring, dChip, SAM, and GenMapp/MAPP Finder software. Validation studies on selected genes and proteins were performed utilizing RT-PCR and immunohistologic techniques. Key words: Hepatocellular carcinoma, Cirrhotic, Non-cirrhotic, Laser microdissection, Gene expression

1. Introduction Hepatocellular carcinoma (HCC) is the sixth most common cancer worldwide (626,000/year), but because of its very poor prognosis, it is the third most common cause of death from cancer (598,000/year) (1). Most HCCs develop in a background of

Graeme I. Murray (ed.), Laser Capture Microdissection: Methods and Protocols, Methods in Molecular Biology, vol. 755, DOI 10.1007/978-1-61779-163-5_19, © Springer Science+Business Media, LLC 2011

233

234

M. Tretiakova and J. Hart

cirrhosis caused by chronic liver disease. Cirrhosis is defined as a diffuse alteration of normal hepatic architecture in which bands of fibrous tissue divide the hepatic parenchyma into regenerative nodules. The risk of the development of HCC depends on the etiology of the underlying chronic liver disease. For instance, there is a relatively high risk of HCC in patients with cirrhosis due to chronic hepatitis B or C virus infection or genetic hemochromatosis. In contrast, patients with cirrhosis due to primary biliary cirrhosis or Wilson disease have a significantly lower rate of HCC development. However, the risk of HCC is higher in cirrhosis due to any cause when compared to patients without cirrhosis. This suggests that both nonspecific chronic necroinflammatory activity (due to any chronic liver disease) and the particular hepatic insult (e.g., iron overload in genetic hemochromatosis or integration of the viral genome in chronic HBV infection) are factors that promote the development of HCC (2). A morphologic sequence has been identified as the pathway to HCC in cirrhotic livers (macroregenerative nodule → dysplastic nodule → HCC), akin to the adenoma → carcinoma sequence for colonic cancers (3). In the USA, about 15% of patients with HCC have no evidence of cirrhosis. These patients tend to be younger at presentation than cirrhotic patients with HCC (4), and they more often have symptoms that can be directly related to the tumor. The tumors are also more often single and tend to be larger. The prognosis is better in patients without cirrhosis (5), due to the better general medical condition of these patients, the higher likelihood of surgical resectability, and perhaps also to less inherent biologic aggressiveness. The etiologic factors important in the development of HCC in non-cirrhotic patients have not been elucidated (6), and the morphologic precursor lesion has not been clearly identified (7). Molecular studies have identified significant differences in the genetic alterations present in non-cirrhotic HCCs when compared to cirrhotic HCCs (8, 9). For instance, comparative genomic hybridization analysis has revealed copy number gains in 8q (10) and losses of 14q21 and 10q23 (9) in non-cirrhotic HCCs but not in cirrhotic HCCs. Advances in DNA sequencing technology and the development of genomic microarray technology have recently provided powerful tools to study the expression of thousands of genes in a single experiment. Using high-density oligonucleotide microarray, Dr. Iizuka and colleagues (11) identified 89 genes that were expressed differently between HBV-HCCs with and without cirrhosis, and 8 genes that were expressed differently between HCV-HCCs with and without cirrhosis. However,

19  Laser Microdissection for Gene Expression Study of Hepatocellular Carcinomas...

235

none of these mRNA expression differences were validated at the protein level by immunohistochemistry. In this study, we first identified and compared genetic alterations between HCC with and without cirrhosis at the DNA level using LMD technique, followed by validation at the protein level by immunohistochemistry (Fig. 1).

2. Materials 1. PEN-membrane glass slides for Microsystems, Bannockburn, IL).

Leica

LMD

(Leica

2. OCT embedding media for frozen sections. 3. Plastic Coplin staining jars. 4. HistoGene staining solution (Molecular Devices, Sunnyvale, CA). 5. Ethanol (95 and 100%). 6. Diethylpyrocarbonate (DEPC)-treated water. 7. TRIzol Reagent (Invitrogen, Carlsbad, CA). 8. 0.5  ml Thin Wall Flat Cap PCR-tubes for Leica LMD (LabSource). 9. RNase ZAP (Ambion). 10. Kimwipe (or similar lint-free towel). 11. b-Mercaptoethanol (b-ME) (Sigma, St Louis, MO). 12. RNeasy Micro kit for RNA isolation (Qiagen, Valencia, CA). 13. RiboAmp® kit for linear RNA amplification (Molecular Devices). 14. Preparation of DEPC-treated water. DEPC destroys enzymatic activity by modifying –NH2, –SH, and –OH groups in RNases and other proteins. DEPC treatment is a very effective way to treat solutions that will contact RNA. Add (DEPC) at a concentration of 0.1% v/v to distilled water (1 ml to 1 l). Stir or shake. Incubate for several hours. Autoclave at least for 45  min. Aliquot DEPC-treated water and store at room temperature. 15. Preparation of RNA Lysis buffer. b-Mercaptoethanol (b-ME) must be added to buffer RLT (Qiagen) before use. Add 10 ml b-ME per 1  ml buffer RLT. Dispense in a fume hood and wear appropriate protective clothing. Buffer RLT is stable at room temperature for 1 month after addition of b-ME.

236

M. Tretiakova and J. Hart

Fig. 1. Study design for LMD and genomic microarray study of HCC in cirrhotic and noncirrhotic liver. Representative frozen section images depicting of HCC (a) in cirrhotic background (b) and HCC (c) in non-cirrhotic background (d). Laser capture microdissection (LMD) of pure HCC cell population without surrounding fibrotic tissue: before LMD (e), during laser cut (f), after laser cutting (g), and tissue collected into the PCR tube cap for RNA isolation (h). RNA integrity evaluation using Agilent

19  Laser Microdissection for Gene Expression Study of Hepatocellular Carcinomas...

237

3. Methods The methods described below outline detailed slide preparation (1), staining and dehydration (2), laser microdissection (LMD) (3), RNA isolation (4), RNA amplification (5), brief summary of genomic microarray analysis (6), and results validation (7). 3.1. Slide Preparation (see Notes 1–3)

1. Precool the cryostat to −20°C for most of the tissues or −30°C for lipid-rich tissues. 2. Remove and discard old microtome blade. Wipe down the knife holder and antiroll plate in the cryostat with 100% ethanol to avoid sample cross-contamination. 3. Install a new disposable microtome blade in the cryostat. 4. Transfer the cryomold containing the specimen from the −80°C freezer to the cryostat, transporting on dry ice if necessary. 5. Wait a minimum of 10 min for the specimen to equilibrate with the temperature of the cryostat. 6. Mount specimen to specimen holder with OCT. Start specimen trimming at 10–20 mm thickness (see Note 1). 7. Set cutting thickness to desired thickness (5–10  mm) (see Note 2). 8. Mount sections toward the center of a PEN-foil LMD glass microscope slide at room temperature. Place the slide immediately into slide box on dry ice. Do not allow slide to dry at room temperature. 9. Discard the slides with folded or wrinkled sections. If cutting more than one specimen, use a new disposable microtome blade for each one. In addition, wipe down knife holder and antiroll plate with 100% ethanol in between each specimen to avoid cross-contamination. 10. Proceed immediately to the “Staining and Dehydration” segment of the protocol or store slides with mounted tissue sections at −80°C for up to 2 months (see Note 3).

Fig. 1. (continued) 2100 Bioanalyzer shows pure RNA from LMD samples with minimal degradation on six representative samples of HCC tumors and surrounding background cirrhotic liver (i). Genomic microarray analysis (j) was performed following RNA amplification and hybridization steps. Differential gene expression was validated on protein level on liver tissue microarray containing dozens of HCC cases with cirrhotic and non-cirrhotic adjacent liver tissues (k) stained with various available antibodies using routine IHC technique (l) (reprinted from Histopathology, Tretiakova et  al., 2010;56:683–93).

238

M. Tretiakova and J. Hart

3.2. Staining and Dehydration

1. Label and fill four slide Coplin jars with 50 ml of the appropriate solutions (75% ethanol prepared using DEPC-treated water, DEPT-treated water, 95% ethanol, and 100% ethanol). 2. Remove one slide at a time from the slide box on dry ice or from the −80°C freezer, air dry slide for 30 s (s) at room temperature (see Note 4). 3. Place it in slide jar containing 75% ethanol for 30 s to fix the ­section. Use cover glass forceps to transfer slides from jar to jar. 4. Transfer the slide to jar containing DEPC water for 30 s to remove OCT. 5. Place the slides on a Kimwipe or a horizontal staining tray. Using an RNase-free pipette tip, apply approximately 100 ml of the HistoGene Staining Solution so it covers the section. Stain for 30–60 s. 6. Place the slide back in jar containing DEPC water for 30 s. 7. Transfer the slide to jar containing 95% ethanol for 30 s. 8. Transfer the slide to jar containing 100% ethanol for 30 s. 9. Air-dry glass-foil slide for 1–2 min and start LMD immediately (see Note 5).

3.3. Laser Microdissection

We used the Leica AS LMD system, a predecessor of current models LMD6000 and LMD7000, which have improved design, laser quality, and software, but utilize similar principles of operation. These UV-laser-based microdissection systems combine automated upright microscope architecture, three-dimensional optical control of the dissecting laser beam and the dissected area, noncontact tissue sampling, and motorized postdissection handling. The key steps of LMD process are as follows: 1. Start the microscope and software. 2. Insert the slide into the specimen holder with tissue section face down. 3. Insert 0.5 ml microcentrifuge tube caps into collection device. 4. Fill tube caps with 75 ml of RLT lysis buffer (RNeasy Micro Kit, Qiagen) on the collection device. 5. Select a cap, represented by a microcentrifuge tube lid. 6. Select scanning magnification 4× using right buttons of the joystick and move relevant tissue detail into the field of view. Switch to the cutting objective of your choice, perform laser calibration. 7. Draw line around target area and cut out selected target area. 8. Inspect results by switching to the cap key in the menu toolbar. Focus the image and inspect the collected material within the cap.

19  Laser Microdissection for Gene Expression Study of Hepatocellular Carcinomas...

239

9. Return to the specimen image of previous cutting position by pressing the “Slide” key. 10. Upon LMD completion carefully remove microcentrifuge tube cap with cells from collection device and close a cap. Vortex the sample for 10 s. Centrifuge tube at 3,000 × g for 1 min. Adjust volume of RLT buffer to 75 ml if some buffer evaporated during LMD (see Note 6). 11. Proceed immediately to the “RNA isolation” segment of the protocol or store homogenized by vortexing cells at −20°C for several days, or at −80°C for up to 2 months (see Note 7). 3.4. RNA Isolation Using Qiagen RNeasy ® MicroKit

The RNeasy microprocedure is designed to purify RNA from small amounts of tissues or cells such as laser-microdissected samples. RNeasy Microtechnology combines the selective binding properties of a silica-gel-based membrane with the microspin technology. The GITC-containing lysis buffer RLT stabilizes RNA during LMD. With the RNeasy Microprocedure, all RNA molecules longer than 200 nucleotides are isolated. The procedure enriches for mRNA since most RNAs <200 nucleotides (such as 5.8S rRNA, 5S rRNA, and tRNAs, which together make up 15–20% of total RNA) are selectively excluded. Manufacturer’s recommended protocol is simplified and noted below: 1. Disrupt sample and lyse with GITC-containing buffer RLT 5–30 min. 2. Homogenize sample to shear genomic DNA by vortexing, then centrifuge 2 min at 5,000 × g. 3. Add ethanol to adjust binding conditions, 1 min. 4. Apply sample to RNeasy MinElute Spin Column for absorption of RNA to membrane, 1 min. 5. Remove contaminants with simple wash steps. 6. Elute ready-to-use RNA in 10–14 ml water.

3.5. RNA Quality Evaluation: Quantity, Purity, and Integrity

RNA quality is the most important starting point for a successful microarray experiment. “Genomic microarray quality” RNA should meet the following standards: RNA quantity: The amount of total RNA should be in the range of 1–10 mg with a concentration of 0.5 mg/ml. RNA purity: A260/A280 and A260/A230 ³ 1.8. RNA integrity: Ethidium bromide staining of the RNA gel should reveal discrete 18S and 28S ribosomal RNA (rRNA) bands (i.e., no significant smearing below each band). The 28S rRNA band should be approximately twice as intense as the 18S rRNA band. The presence of a broad band at the bottom of the gel may indicate degradation.

240

M. Tretiakova and J. Hart

For RNA integrity assessment of LMD samples, we used LabChip technology. The Agilent 2100 Bioanalyzer requires only nanogram quantities of RNA sample for quality evaluation. Like a denaturing agarose gel, the Bioanalyzer fractionates the RNA molecules according to their size and can determine the amounts of 18S and 28S rRNA (see manufacturer’s recommendations). RNA quantity and purity were determined using a GeneSpec III (MiraiBio) to reliably and reproducibly determine RNA quantity and purity using 1 ml sample, which is critical for RNA samples with limited quantity. 3.6. RNA Amplification Using RiboAmp® Kit

The RiboAmp® OA RNA Amplification Kit, predecessor of improved kits such as RiboAmp® Plus and HS Plus (Molecular devices, CA), enables the production of 20–50 mg quantities of antisense RNA (aRNA) from nanogram quantities of total cellular RNA. This specially formulated kit conveniently prepares small samples for direct labeling to hybridize onto oligonucleotide arrays requiring labeled aRNA, such as GeneChip® Probe Arrays (Affymetrix). The RiboAmp OA Kit achieves high yields of aRNA with a proprietary linear amplification process using doublestranded cDNA as template in a T7 RNA Polymerase-catalyzed amplification. 5–30 ng total RNA or 500–3,000 LCM cells typically yield enough aRNA for two or more oligo array hybridizations. Therefore, the kit achieves amplifications of up to 1,000-fold in one round of amplification (Table 1). Generated aRNA can be used for a second round of amplification to generate biotin-labeled aRNA during in vitro transcription step for target microarray hybridization.

3.7. Genomic Microarray Analysis

Twenty micrograms of biotin-labeled aRNA were fragmented and hybridized to the array, washed, and stained. Microarray image files were obtained through Affymetrix GeneChip software (MAS 5.0) and GeneArray scanner (Affymetrix, Santa Clara, CA). Subsequent robust multichip analysis (RMA) was performed (12, 13). RMA is an R-based technique that analyzes directly from the Affymetrix microarray image file and is comprised of three steps: background adjustment, quantitative normalization, and summarization. In the Windows operating system, the RMA express function was used to process the data. The output is a log 2 transformed expression index data of each probe set. The data were compiled in a spreadsheet and analyzed using STATA 9.0 software (StataCorp, College Station, TX). The genomic microarray data was analyzed by hierarchical clustering and student t-test. Fisher’s exact test and Chi-square analysis were used for comparing the clinicopathologic parameters and the immunohistochemical markers between groups. Survival data were analyzed by Kaplan–Meier survival estimates and log-rank tests.

19  Laser Microdissection for Gene Expression Study of Hepatocellular Carcinomas...

241

Table 1 Protocol overview: time and temperature requirements Procedure

T (°C)

Time

Total length

1. cDNA first strand synthesis

65 4 42 4 37 95 4

5 min Hold 45 min Hold 20 min 5 min Hold

»1.5 h

2. cDNA second strand synthesis

95 4 room 37 70 4

2 min hold 5 min 10 min 5 min hold

»25 min

3. cDNA purification

Room

15–20 min

»15 min

4. In vitro transcription

42 4 37 4

4 h hold 15 min Hold

»4.3 h

5. aRNA purification

Room

15–20 min

»15 min

First round RNA amplification

3.8. Validation of Genomic Microarray Results

»7 h

The differential gene expressions by microarray technology was confirmed by means of Quantitative Real-Time-PCR (QRTPCR), which allows accurate and quantitative measurements of target DNA amplified at each cycle during exponential phase. In our study, we used ABI Prism 7700 Sequence Detection System running PCR reaction mix, which is similar to normal PCR mix except for the addition of a gene-specific probe (SYBR Green) at a final concentration of 0.2 mM. For result validation at the protein levels, an immunohistochemical staining was performed on 4  mm sections obtained from formalin-fixed, paraffin-embedded tissue microarray sections. After deparaffinization, rehydration, and antigen retrieval, tissue sections were incubated with monoclonal antibodies against various antibodies including E-cadherin and matrix metalloproteinases: MMP-1, -2, -7, and -9. For detection a biotin-free HRP enzyme-labeled polymer of EnVision plus system (Dako Cytomation, Carpenteria, CA) was used. A positive reaction was visualized with diaminobenzidine (DAB) solution followed by

242

M. Tretiakova and J. Hart

counterstaining with hematoxylin. Positive controls were selected according to the manufacturers’ recommendations, whereas negative controls were prepared by using nonimmune mouse or rabbit immunoglobulins. The stained slides were read by two pathologists independently at different times (14).

4. Notes 1. Collect five 10–20 mm trim sections for RNA isolation, place in 2 ml Eppendorf tube with 1.5 ml of DEPC-treated water to remove OCT, wash in DEPC treated water twice then add 1 ml of Trizol reagent or RLT buffer (Qiagen) for standard RNA isolation procedure. We strongly advise checking initial RNA quality of tissues prior to LMD to prevent using molecularly degraded tissues. 2. Optimal thickness of frozen sections should be experimentally defined based on good morphology and ease of laser cutting (thinner is better), as well as high RNA yields (thicker is better). Our recommendation is to use 7–8 mm sections. 3. Frequent cycling of the tissue block from −80°C to −20°C for cryosectioning may accelerate RNA degradation. For best results, cut and mount a sufficient number of sections for 2 months’ use during one cryosectioning session. Store the mounted sections at −80°C until needed. 4. Frozen sections on PEN-foil slides removed from −80°C should be warmed up prior to staining to prevent detachment of PEN-foil from glass surface by rapid temperature change. The slide is ready for staining when frost and water is completely evaporated from the slide surface. 5. Carry out the “Staining and Dehydration” protocol with only one slide at a time. Change all solutions in the plastic slide jars between each case (maximum of five slides) to avoid cross contamination. Do not reuse solutions. Do not transfer solutions back into their original bottles. If you plan to reuse jars, discard all water-based solutions upon completion of staining. Clean them with 100% ethanol, followed by distilled water, RNase ZAP, then DEPC water and allow drying completely in the hood. 6. Do not perform LMD on one slide for longer than 30 min to minimize RNA degradation. In general, cells are protected from RNases once they are transferred to the tube cap containing GITC-containing lysis buffer RLT. Remaining cells on the tissue sections are relatively protected while section is still dry, but moisture from the air will activate endogeneous RNases within

19  Laser Microdissection for Gene Expression Study of Hepatocellular Carcinomas...

243

minutes. Also, the RLT lysis buffer will start to evaporate and crystallize after 20–30 min of LMD procedure, requiring volume adjustment prior to RNA isolation. 7. Every effort should be made to minimize RNA degradation during the LMD steps. The RNase contamination could be avoided by adhering to the following recommendations throughout your experiment: (a) Always handle RNA in a manner that avoids introduction of RNases: work fast, wear disposable gloves, change them frequently and avoid touching RNase contaminated surfaces. (b) Wash scalpels, tweezers, and forceps with detergent and bake at 210°C for 4 h before use. (c) Use only RNAse-free glassware, plastic ware, pipette tips, and microcentrifuge tubes. (d) Use only RNase-free solutions, dilute or wash in DEPCtreated water only. (e) Clean work surfaces and labware with commercially available RNase decontamination solutions [RNase AWAY® (Molecular BioProducts) or RNase ZAP® (Ambion)] prior to performing reactions.

Acknowledgments We cordially thank Dr. Xinmin Li and Jian Zhou from the Functional Genomics Facility (University of Chicago) for their help in performing the Affymetrix Microarray experiments. References 1. Parkin, D. M., Bray, F., Ferlay, J., and Pisani, P. (2005) Global cancer statistics, 2002. CA Cancer. J. Clin. 55, 74–108. 2. Coleman, W. B. (2003) Mechanisms of human hepatocarcinogenesis. Curr. Mol. Med. 3, 573–588. 3. Feitelson, M. A., Sun, B., Satiroglu Tufan, N. L., Liu, J., Pan, J., and Lian, Z. (2002) Genetic mechanisms of hepatocarcinogenesis. Oncogene. 21, 2593–2604. 4. Chang, C. H., Chau, G. Y., Lui, W. Y., Tsay, S. H., King, K. L., and Wu, C. W. (2004) Longterm results of hepatic resection for hepatocellular carcinoma originating from the noncirrhotic liver. Arch. Surg. 139, 320–5; discussion 326.

5. Chen, M. F., Tsai, H. P., Jeng, L. B., Lee, W. C., Yeh, C. N., Yu, M. C., and Hung, C. M. (2003) Prognostic factors after resection for hepatocellular carcinoma in noncirrhotic livers: univariate and multivariate analysis. World J. Surg. 27, 443–447. 6. Nzeako, U. C., Goodman, Z. D., and Ishak, K. G. (1995) Comparison of tumor pathology with duration of survival of North American patients with hepatocellular carcinoma. Cancer. 76, 579–588. 7. Nzeako, U. C., Goodman, Z. D., and Ishak, K. G. (1996) Hepatocellular carcinoma in cirrhotic and noncirrhotic livers. a clinico-histopathologic study of 804 North American patients. Am. J. Clin. Pathol. 105, 65–75.

244

M. Tretiakova and J. Hart

8. Ho, M. K., Lee, J. M., Chan, C. K., and Ng, I. O. (2003) Allelic alterations in nontumorous liver tissues and corresponding hepatocellular carcinomas from Chinese patients. Hum. Pathol. 34, 699–705. 9. Kim, G. J., Cho, S. J., Won, N. H., Sung, J. M., Kim, H., Chun, Y. H., and Park, S. H. (2003) Genomic Imbalances in Korean hepatocellular carcinoma. Cancer Genet. Cytogenet. 142, 129–133. 10. Wong, N., Lai, P., Lee, S. W., Fan, S., Pang, E., Liew, C. T., Sheng, Z., Lau, J. W., and Johnson, P. J. (1999) Assessment of genetic changes in hepatocellular carcinoma by comparative genomic hybridization analysis: relationship to disease stage, tumor size, and cirrhosis. Am. J. Pathol. 154, 37–43. 1 1. Iizuka, N., Oka, M., Yamada-Okabe, H., Mori, N., Tamesa, T., Okada, T., Takemoto, N., Hashimoto, K., Tangoku, A., Hamada, K., Nakayama, H., Miyamoto, T., Uchimura, S., and Hamamoto, Y. (2003) Differential

gene expression in distinct virologic types of hepatocellular carcinoma: association with liver cirrhosis. Oncogene. 22, 3007–3014. 12. Irizarry, R. A., Hobbs, B., Collin, F., BeazerBarclay, Y. D., Antonellis, K. J., Scherf, U., and Speed, T. P. (2003) Exploration, normalization, and summaries of high density oligonucleotide array probe level data. Biostatistics. 4, 249–264. 13. Bolstad, B. M., Irizarry, R. A., Astrand, M., and Speed, T. P. (2003) A comparison of normalization methods for high density oligonucleotide array data based on variance and bias. Bioinformatics. 19, 185–193. 14. Tretiakova, M. S., Hart, J., Shabani-Rad, M. T., Zhang, J., and Gao, Z. H. (2009) Distinction of hepatocellular adenoma from hepatocellular carcinoma with and without cirrhosis using e-cadherin and matrix metalloproteinase immunohistochemistry. Mod. Pathol. 22, 1113–1120.

Chapter 20 Laser Capture Microdissection of Pancreatic Ductal Adeno-Carcinoma Cells to Analyze EzH2 by Western Blot Analysis Aamer M. Qazi, Sita Aggarwal, Christopher S. Steffer, David L. Bouwman, Donald W. Weaver, Scott A. Gruber, and Ramesh B. Batchu Abstract Pure populations of tumor cells are essential for the identification of tumor-associated proteins for the development of targeted therapy. In recent years, laser capture microdissection (LCM) has been used successfully to obtain distinct populations of cells for subsequent molecular analysis. The polycomb group (PcG) protein, enhancer of zeste homolog 2 (EzH2), a methyl-transferase that plays a key role in ­transcriptional gene repression, is frequently overexpressed in several malignant tumors. High levels of EzH2 are often associated with advanced disease stage in many solid tumors; however, its role in the pathogenesis of pancreatic ductal adeno-carcinoma (PDAC) is poorly understood. Because of the limited sample availability and the absence of in vitro amplification steps for proteins, the use of LCM for proteomics studies largely depends on highly sensitive protein detection methods. Here, we developed a faster and sensitive Western blot protocol and validated it for the detection of EzH2 in ~2,000 cells. Initially, cultured PANC-1 cells were used to optimize protein electrophoresis and western blotting conditions. Gradient gel electrophoresis in combination with optimized antibody concentrations, and a sensitive chemiluminescent assay provided a strong signal. In order to further confirm the role of EzH2 in PDAC, employing siRNA-mediated gene silencing via long lasting plasmid vectors containing shRNA, we investigated the potential role of EzH2 gene silencing in pancreatic cancer regression. Positive correlation of EzH2 expression was observed with advanced stage, serous histology, and increasing grade in pancreatic cancer patient tissues. Further EzH2 knockdown resulted in decreased cell growth and invasiveness. The findings of this study emphasize that western blotting of a LCM-generated pure population of cancer cells may be a valuable technique for the study of tumor-specific proteins. Key words: Laser capture microdissection, Pancreatic ductal adeno-carcinoma, EzH2, Immunohistochemical analysis, Western blot analysis, RNA interference

Graeme I. Murray (ed.), Laser Capture Microdissection: Methods and Protocols, Methods in Molecular Biology, vol. 755, DOI 10.1007/978-1-61779-163-5_20, © Springer Science+Business Media, LLC 2011

245

246

A.M. Qazi et al.

1. Introduction Cancer tissues are often composed of multicellular regions of both malignant and nonmalignant cells. Analysis of pure population of malignant lesions is essential for the identification of tumor-specific pathways and molecules that can offer clues as to the nature of the fundamental alterations that underlie cancer development. Identification of tumor-associated proteins will be not only useful for clinicians to guide therapy but also needed for the development of targeted therapy. Most genomic and proteomics studies require procurement of several thousand cells to provide enough material for statistically meaningful analysis. Over the years, several microdissection techniques have been developed to isolate pure populations of cancer cells, with laser capture microdissection (LCM) emerging as methods of choice (1). LCM technology has many applications, but its power lies in the selective harvesting of tumor cells from multicellular regions of histological sections (2). The laser is roughly 350 nm and has a high energy density at the small focal point of less than 1 mm that allows separation of single cell from tissue sections. Since there is no heat transfer, the laser beam does not affect nucleic acid or protein as the 350 nm wavelength of the laser falls short of the absorption spectra for these molecules. LCM has become an essential tool in genomic/proteomics research enabling the selection of a pure population of tumor cells from a tumor tissue that contains both normal and tumor cells (3, 4). For protein analysis of laser-captured cells, promising results have been obtained in the recent past for immunoassays (5), studies of two-dimensional polyacrylamide gel electrophoresis (2D-PAGE) (6) and mass spectrometry (7). Polycomb protein EzH2 (enhancer of Zeste homolog 2) is a catalytic subunit of polycomb repressive complex 2 (PRC2) which trimethylates histone H3 at lysine residue 27 (H3K27me3) often at tumor suppressor genes resulting in dense packing of chromatin often called heterochromatin (8, 9). This epigenetic modification of histones makes it less accessible to transcriptional machinery resulting in silencing of the genes. EzH2 is often overexpressed in human cancers and is associated with aggressiveness of the disease. Although enhanced expression of EzH2 has been shown to correlate with poor prognosis of pancreatic cancer (10, 11) at a molecular level, very little is known on its contribution to aggressiveness of pancreatic cancer. Further, pancreatic cancer presents considerable heterogeneity and the isolated tumor tissue invariably consists of a significant percentage of normal cells. Isolation of pure population of pancreatic cancer cells is a key step not only in understanding of molecular mechanisms specific to the tumor cells but also to formulating better therapy. Although numerous LCM studies of

20  Laser Capture Microdissection of Pancreatic Ductal Adeno-Carcinoma Cells…

247

improved techniques to isolate nucleic acids have been described, studies involving proteomics have been hampered by the inability to obtain sufficient amounts of protein for further analysis. Due to this limitation, the majority of the studies were done with proteins that have the most abundant cellular expression. Since EzH2 is overexpressed in pancreatic cancer, we standardized Western blot analysis with small number of cells employing LCM that enabled a rapid and precise method of isolating pancreatic cancer cells.

2. Materials 2.1. Collection of Surgically Resected Pancreatic Ductal Adeno-Carcinoma Cells and Preparing Specimens for Storage

1. Liquid nitrogen in appropriate transport carrier. 2. Safety glasses or face shield. 3. Freezer gloves. 4. Disposable sterile latex or nitrile surgical gloves. 5. Surgical mask. 6. Clean laboratory coat. 7. Clean protective shoes. 8. Sterile cryovials, labeled and unlabeled extras. 9. Racks for vials/containers (separate for tumor and normal). 10. Sterile disposable scalpels or scalpel blades (#10, 21, 22, 60, or single edge razor blade). 11. Sterile disposable forceps, sterile 21G needles, sterile disposable towels, or drapes. 12. Sterile dishes or plates, such as petri or cell culture dishes, sterile saline (PBS). 13. Sterile gauze. 14. Marking pen or pencil. 15. Chipped ice, flat container or tray for chipped ice, such as an autoclavable pyrex cake dish. 16. Sharps container for disposal of biohazardous sharps are some of the essential materials.

2.2. Acquisition of Target Cells with Laser Capture Microdissection (LCM)

1. Mayer’s hematoxylin. 2. Ethanol. 3. Xylene. 4. One millimeter thick polyethylene naphthalene membrane (Meiwa Shoji Ltd., Japan), adhesive. 5. Methanol. 6. Toluidine blue.

248

A.M. Qazi et al.

7. Ultrapure distilled water. 8. Protease cocktail (Invitrogen Corp., New York). 2.3. Tissue Section Fixation and Microdissection

1. Tumor tissues, fixative reagents any one, such as 10% formalin, 4% paraformaldehyde, methanol, cold acetone, 70% ethanol, 100% ethanol, methanol solution (60% v/v) absolute methanol, 30% chloroform, and 10% glacial acetic acid, paraffinblock, standard cryostat with a clean blade. 2. Superfrost Plus or 3. Poly-l-precoated glass slides.

2.4. Cell Proliferation and Invasion Assays

1. Human pancreatic tumor cell line PANC-1 (American Type Culture Collection, Manassas, VA), was grown in subconfluent monolayer cultures in DMEM medium containing 10% FBS, supplemented with 2  mM glutamine, 100  U/mL penicillin, and 100  mg/mL streptomycin. Growth curves were obtained for each cell line (treated and control groups) in triplicate using CCK-8 kit (Dojindo, Gaithersburg, MD). 2. Nontarget control and EzH2 shRNA vector pLKO.1 were obtained from Sigma-Aldrich (St. Louis, MO). The nontarget control vector sequence is 5¢ CCG GCA ACA AGA TGA AGA GCA CCA ACT CGA GTT GGT GCT CTT CAT CTT GTT GTT TTT 3¢; the EzH2-shRNA vector sequence is 5¢ CCG GCC CAA CAT AGA TGG ACC AAA TCT CGA GAT TTG GTC CAT CTA TGT TGG GTT TTT G 3¢. 3. Cell invasion assay was performed with a modified Boyden chamber using filters with a pore size of 8-mm.

2.5. SDS-PAGE and Western Blot Analysis

XCell SureLock Mini-Cell and XCell II Blot Module, Novex 4–20% Tris–Glycine Gel 1.5  mm, 15 well, iBlot Dry Blotting System, iBlot gel transfer stacks nitrocellulose, regular.

3. Methods 3.1. Collection of Surgically Resected PDAC Cells and Preparing Specimens for Storage ( see Note 1)

1. Be present in the surgery room before the completion of the surgical procedure. Bring personal protective equipment and all supplies required for the procedure including a clean lab coat. 2. Make sure that you have the completed consent form for the patient. 3. Set up supplies needed for the procedure. Have prelabeled vials arranged for storage of tumor and normal specimens. 4. All sterile dishes (prelabeled) should be placed on ice.

20  Laser Capture Microdissection of Pancreatic Ductal Adeno-Carcinoma Cells…

249

5. Transport container having liquid nitrogen should also be placed at an accessible location. Avoid accidental spillage by making sure that the liquid nitrogen vessel is placed in a secure place. 6. The surgical specimen should be transported from the surgical area to the pathology receiving room in a sterile container, which has been covered with a cap or sterile cloth. Please make sure that the tissue is not submerged in formalin. 7. Specimen should be placed on ice in order to minimize degradation and autolysis. First preference should be to snap freeze or resuspended in appropriate media within 5 min of surgical removal. 8. Samples from the resection specimen are bisected using sterile scalpel blades or scalpels or razor blades and sterile forceps. It is important that separate blades and forceps should be used for specimens from each anatomic site to avoid crosscontamination of tissue. This step has to be performed by experienced pathologist. 9. Transfer the specimens provided by the pathologist to prelabeled sterile container using a sterile needle or forceps. Again use separate forceps/needles for each specimen to avoid cross-contamination. You can place the forceps or needle on the lids of every container for repeated use. The ice can contaminate the specimen if it gets in contact. 10. Using a sterile blade, bisect the specimen into 0.5 × 0.5 × 0.5 cm or smaller samples. Samples should not exceed 0.5  cm on one dimension, as larger sizes may impair freezing or perfusion rates. Size ranges from 0.5 to 0.3 cm are optimal. The forceps/needle can be used to hold the specimen while bisecting. 11. Tumor tissue: For tumor specimen trim any normal tissue if possible. Site priority (in order of decreasing priority) (a) central area of the tumor, (b) tumor margin (leading edge if known). 12. Storage priority: After cutting tissue from each anatomic site to the appropriate size, specimens are placed into separate sterile labeled vials or containers for snap freezing (first priority), formalin fixation (second priority), and OCT media embedding (third priority). 13. Normal tissue: Site priority (in order of decreasing priority) (a) distant, grossly uninvolved squamous mucosa, (b) normal tissue adjacent to the tumor, (c) surrounding tissue size requirements: (a) maximum storage size: 0.5 × 0.5 × 0.5  cm, (b) minimum storage size: none, (c) storage priority: same as above.

250

A.M. Qazi et al.

3.2. Acquisition of Target Cells with Laser Capture Microdissection

1. Target cells from fresh-frozen tissue sections were obtained using the laser capture microdissection (LCM) technique as follows: Frozen tissue sections (8  mm) were cut, mounted onto glass slides, fixed (70% ethanol for 3 min), and stained with hematoxylin as follows: Put 8-mm thick sections, onto a slides and dip immediately into 70% ethanol for 60  s (see Note 2), dip into water five times to remove imbedding medium, stain section for 15 s in Mayer’s hematoxylin, dip into water ten times (15  s), dip into automation buffer (Biomeda, 1:10) ten times, 15 s water (dip ten times), 30 s in 70% ethanol (dip 20 times), 30  s in 96% ethanol (dip 20 times), [optional: 15  s eosin (1:2 diluted in water plus one drop of glacial acetic acid per 50 mL)], 15 s 96% ethanol (dip ten times), 30  s 96% ethanol (dip 20 times), repeat three times, 30 s xylene (dip 20 times), 3 min xylene, 3 min xylene, let section air dry (2 min under a fume hood). 2. The stained sections were examined under a microscope by an experienced gastrointestinal pathologist and the targeted area including various pancreatic cancer related lesions was identified according to the standard histopathological criteria (12). 3. The area is then marked with a marker and the slide repositioned for laser capture microdissection according to the protocol out lined below (13). The captured cells were placed in an appropriate buffer for subsequent analysis.

3.3. Laser Capture Microdissection

LCM is a widely used technique, which allows the precise identification, dissection, and harvesting of the desired cell(s) from heterogeneous cell populations. For example, a tumor specimen contains not only tumor cells, but also, nontumor cells, such as inflammatory cells, stroma cells, normal epithelial cells, etc. The presence of nontumor cells along with the tumor cells decreases the accuracy and sensitivity of molecular analysis of tumors. With a tissue microdissection technique, we can separate the tumor cells from nontumor cells and can perform the molecular analysis of pure tumor cells. The principle of LCM is based on the adherence of visually selected cells to a thermoplastic membrane, which overlies the dehydrated tissue section and is focally melted by triggering of a low energy infrared laser pulse (2). The use of immuno-histochemical stains allows the selection of cells according to phenotypic and functional characteristics. Also, complete dryness of the sections is a prerequisite for successful microdissection of the tissue, as it allows precise and thin cuts at energy levels of 70–80  ml per pulse (14). The system consists of an inverted microscope, a solid-state near infrared laser diode, a laser control unit, a joy stick controlled microscope stage with a vacuum chuck

20  Laser Capture Microdissection of Pancreatic Ductal Adeno-Carcinoma Cells…

251

for slide immobilization, a CCD camera, and a color monitor (2). Two different systems have been developed. One uses a shortpulsed infrared laser beam that fuses the targeted cell membranes of the slide-mounted tissue with a temperature-sensitive polyacetate membrane. The attached cells adhere to the membrane and can then be removed from the slide and placed directly into a DNA, RNA, or protein extraction buffer (15). The advantages of the LCM system are that it does not require manual labor, has high speed, thousands of cells of interest can be collected within few minutes, the morphology of the captured cells as well as the residual tissue is well preserved, and the danger of tissue loss is minimized (2). The other commercially available, advanced laser microdissection technology is laser microbeam microdissection (PALM; Mikrolaser Technologie, Bernried, Germany). This system cuts the tissue directly on the slide using a temperature-neutral ultraviolet (UV) laser, leaving a micrometer-sized gap around the specimen. The same laser beam then catapults the tissue into a lid containing mineral oil or an appropriate buffer for the isolation of nucleic acids. The advantage of this system over LCM is that it allows a contact free collection of desired isolated cells in a microcentrifuge tube (14, 16). 1. The frozen blocks from the rat samples were also frontally sectioned and 7-mm thick serial sections were used for laser capture microdissection. 2. The sections were collected individually on a 1.35-mm thick polyethylene naphthalene membrane (Meiwa Shoji Ltd., Japan) with a powerful adhesive and left for 1  h in the cryostat. 3. They then were stored in the freezer at −80°C. 4. The sections on the film were fixed with methanol for 3 min. 5. Stained with toluidine blue for 10 s. 6. Washed with distilled water (ultra pure distilled water, Invitrogen Corp., New York). 7. PDAC cells were microdissected from the sections on the film using the PALM Micro Beam system (PALM Microlaser Technologies AG, Bernried, Germany) equipped with a nitrogen laser (337 nm) for cutting, ablating, and collecting the tissues by laser pressure catapulting technology. 8. After microdissection, the dissected samples were ejected from the object plane without mechanical contact and catapulted directly into a microfuge cap using a single laser shot for subsequent protein extraction. Samples were collected from approximately 20 sections.

252

A.M. Qazi et al.

3.4. Tissue Section Fixation and Microdissection

1. Five to ten micrometer thick sections from a snap frozen tumor tissue can be cut on a standard cryostat with a clean blade, and are mounted on Superfrost Plus or poly- l precoated glass slides. 2. The frozen sections are thawed at room temperature for 30–60 s without drying and immersed immediately into one of the following fixatives: methanol (5  min), cold acetone (5 min), 4% paraformaldehyde (5 min), 70% ethanol (15 s) followed by acetone (5 min) (17). Also, the frozen sections can be fixed immediately with 100% ethanol for 30 min, and then stored at room temperature until microdissected (12). 3. After fixation, the slides are rinsed in 1× phosphate-buffered saline, pH 7.4, and subjected to histochemical staining (12). 4. In case of whole tumor tissue fixed in 10% formalin or in 4% paraformaldehyde from 18 to 24 h (18) or in methacarn solution (60% (v/v) absolute methanol, 30% chloroform, and 10% glacial acetic acid) for 2 h at 4°C (19), then embedded in a paraffin-block are also cut into 5–10-mm thick sections on a standard cryostat with a clean blade, and are mounted on Superfrost Plus or poly-l-precoated glass slides. 5. The sections are deparaffinized by immersing in xylene three times for 2 min, followed by 99.5% ethanol twice for 2 min. Deparaffinized sections are then subjected for histochemical staining (Fig. 1a) (19).

3.5. SDSPolyacrylamide Gel Electrophoresis

1. For the analysis of whole sections, tissue were dewaxed and dehydrated, as described for the preparation of LCM samples and scraped from the microscope slides using a clean scalpel. 2. The scraped cells were then suspended in 30 ml of reducing sample buffer per section and further diluted to obtain the appropriate protein concentration.

Fig. 1. Western blot analysis of LCM separated cells from tissue sections: (a) Immunohistochemical staining of EzH2 in control pancreatic and pancreatic cancer tissue. 1. Pancreatic normal tissue section – EzH2 negative (40×); 2. Pancreatic cancer tissue section – EzH2 Positive (40×). (b) Western blot analysis of pancreatic cancer cells and control pancreatic cells isolated by LCM for the presence of EzH2. Upper panels: 1 and 2, pancreatic cancer cells (PANC-1); 2 and 3, normal pancreatic cells. Lower panel: b-actin expression was monitored as the control.

20  Laser Capture Microdissection of Pancreatic Ductal Adeno-Carcinoma Cells…

253

3. LCM separated cells were lysed by adding 20 ml of reducing sample buffer on top of the LCM caps. Cell lysates were heat-denatured and subjected to Western blot analysis using 5 mg of protein with 4–15% SDS Tris–Glycine gels. Cut open the gel cassette pouch and rinse with deionized water. 4. Peel off the tape covering the slot on the back of the gel cassette of Invitrogen precast gradient gels, pull the comb out of the cassette and gently wash the cassette wells with 1× running buffer. 5. Insert gel cassettes into the lower buffer chamber with the help of gel tension wedge. Place one cassette behind the core and one cassette in front of the core. 6. Pull forward on the gel tension lever in a direction toward the front of the XCell SureLock™ unit until lever comes to a firm stop and the dam appear snug against the buffer core. 7. The upper buffer chamber is the void formed between the two gel cassettes should be filled with 200 mL of the running buffer to completely cover the sample wells. 8. Use a round loading tip to underlay the samples into the gel wells. 9. Samples are loaded and after electrophoresis (120  min run time at 70 V constant), remove the gel cassettes and carefully insert the Gel Knife’s beveled edge into the narrow gap between the two plates of the cassette, push up and down gently on the knife’s handle to separate the plates. Remove and discard the plate without the gel, allowing the gel to remain on the other plate. 10. Proceed to the blotting; proceed to the western transfer protocol without removing the gel from the plate. 3.6. Western Transfer of the Proteins

1. Proteins were transferred to nitrocellulose papers using iBlot dry blotting device (Invitrogen Corp, Carlsbad, CA). This buffer-free system is equipped with its own integrated power supply, utilizes pre-assembled transfer stacks with an integrated nitrocellulose membrane, and transfers proteins in a very short period of 8-min, 23 V transfer. 2. In order to obtain a concentrated spots of protein on the membrane, we used gels that are thick but not wide (2-mm thick, 15 well per/gel). 3. Membranes were blocked in Tris-buffered saline containing 0.05% Tween-20 (TBS-T) and 5% nonfat dry milk (Bio-Rad) for 1 h. 4. Probing of the blots with primary antibody overnight at 4°C with mouse monoclonal anti-b-actin (Sigma, St Louis, MO,

254

A.M. Qazi et al.

USA; 1: 5000), mouse monoclonal anti-EzH2 (Transduction Laboratories, BD Biosciences, San Diego, CA; 1: 1,000) (Fig. 1b). 5. Washed five times with PBS-T, and incubated with HRPconjugated secondary antibody (Santa Cruz, CA). 6. Enhanced chemiluminescence kit was used for the detection of EzH2 (Amersham Pharmacia, Uppsala, Sweden). The membranes were restripped for probing b-actin, used as a housekeeping control protein. 1. Cells were cultured in a humidified atmosphere with 5% CO2 at 37°C. Trypsin (0.25%)/EDTA solution was used to detach the cells from the culture flask for passing the cells. 2. Standard prototype growth curves and number of viable cells were determined for each cell line (treated and control groups) in triplicate experiments according to the CCK-8 (Dojindo, Gaithersburg, MD). 3. Growth curves were plotted as a percentage of the value of control transfected minus the value of untreated cells on day 0. Day 3 values were considered for the determination of the 50% cell proliferation inhibition (IC50) for a given treatment. Parallel manual count was also performed with trypan blue and counting by exclusion method using a hemocytometer. 4. Western blot analysis revealed a significant knock down of the cellular EzH2 protein levels accompanied by growth inhibition of PANC-1 cells (Fig. 2a, b).

b a

EzH2 β-actin

Live cells (% control)

3.7. Cell Proliferation and Invasion Assays

120 100 80 60 40 20 0

Control

c

Control

EzH2shRNA

EzH2 shRNA 0

1

2

3

4

5

Days

Fig. 2. Interference of EzH2 expression by shRNAs inhibited cell growth and invasion in PANC-1 cells: (a) the knockdown of EzH2 analyzed by Western blot analysis. Upper panel: probed for EzH2 expression. Lower panel: probed for b-actin. (b) Cell growth inhibition by EzH2-shRNA: cells were transfected with shRNA control vector or the EzH2 shRNA vector and relative cell viability was measured by colorimetric assay at the indicated times. (c) Trans-well migration assay in Boyden chamber: EzH2-shRNA and control vector transfected cells were trypsinized and seeded in Boyden trans-well chamber separated by membrane filter. Migrated cells at the lower surface of the trans-well filter were stained and counted.

20  Laser Capture Microdissection of Pancreatic Ductal Adeno-Carcinoma Cells…

255

5. Cell migration and invasion are highly integrated multistep processes that play a central role in tumor metastasis (20). Since EzH2 over expression is associated with a more aggressive tumor phenotype we performed trans-well Boyden chamber invasion assays to find out whether EzH2 over expression has any effect on tumor cell motility in PDAC. 6. EzH2 shRNA transfected PANC-1 cells (105/500 mL) were added to the upper chamber with the lower chamber filled with medium and incubated for 18 h at 37°C in a 5% CO2 humidified chamber. 7. Cells on the upper surface of the membrane were then removed using wet cotton swabs. The underside of the chamber was washed twice with PBS, and stained with 4  mM Calcein stain. 8. The migrated cells on the membrane side facing the bottom chamber were counted under a fluorescent microscope in six random fields and photographed. Cell invasion was significantly impaired by EzH2-shRNA transfection (Fig. 2c).

4. Notes 1. Tissue samples that were immediately processed and frozen at −80°C from surgically resected pancreatic cancer were used for this study. The tissue specimens were coded without use of any of patient private identifiers. The specimens of six normal, five PDAC were used for this study. Populations of normal and abnormal cells were isolated by LCM. 2. According to publications, sections are stable up to 72 h at these conditions and at 4°C. On the strength of our past experience the sections are stable for a few weeks. References 1. Emmert-Buck, M., R. Bonner, et al. (1996). “Laser capture microdissection.” Science 274: 998. 2. Fend, F. and M. Raffeld (2000). “Laser capture microdissection in pathology.” Journal of Clinical Pathology 53: 666. 3. Banks, R., M. Dunn, et  al. (1999). “The potential use of laser capture microdissection to selectively obtain distinct populations of cells for proteomic analysis-preliminary findings.” Electrophoresis 20: 689–700. 4. Espina, V., J. Wulfkuhle, et al. (2006). “Lasercapture microdissection.” Nature Protocols 1: 586–603.

5. Simone, N., A. Remaley, et  al. (2000). “Sensitive immunoassay of tissue cell proteins procured by laser capture microdissection.” American Journal of Pathology 156: 445. 6. Craven, R. and R. Banks (2001). “Laser capture microdissection and proteomics: possibilities and limitation.” Proteomics 1: 1200–1204. 7. Von Eggeling, F., H. Davies, et  al. (2000). “Tissue-specific microdissection coupled with ProteinChipÆ array technologies: applications in cancer research.” Biotechniques 29: 1066–1070.

256

A.M. Qazi et al.

8. Cao, R., L. Wang, et  al. (2002). “Role of histone H3 lysine 27 methylation in Polycombgroup silencing.” Science 298: 1039. 9. Kirmizis, A., S. Bartley, et al. (2004). “Silencing of human polycomb target genes is associated with methylation of histone H3 Lys 27.” Genes & Development 18: 1592. 10. Kleer, C. G., Q. Cao, et al. (2003). “EZH2 is a marker of aggressive breast cancer and promotes neoplastic transformation of breast epithelial cells.” Proceedings of the National Academy of Sciences of the United States of America 100: 11606–11611. 11. Bachmann, I. M., O. J. Halvorsen, et al. (2006). “EZH2 expression is associated with high proliferation rate and aggressive tumor subgroups in cutaneous melanoma and cancers of the endometrium, prostate, and breast.” Journal of Clinical Oncology 24: 268–273. 12. Balch, C., P. Yan, et al. (2005). “Antimitogenic and chemosensitizing effects of the methylation inhibitor zebularine in ovarian cancer.” Molecular Cancer Therapeutics 4: 1505. 13. Ohyama, H., X. Zhang, et al. (2000). “Laser capture microdissection-generated target sample for high-density oligonucleotide array hybridization.” Biotechniques 29: 530–536. 14. Hoffmann, M., J. Gaikwad, et  al. (2002). “Analysis of odontoblast gene expression using

a novel approach, laser capture microdissection.” Connect Tissue Res 43(2–3): 376–380. 15. Bonner, R. F., M. Emmert-Buck, et al. (1997). “Laser capture microdissection: molecular analysis of tissue.” Science 278: 1481,1483. 16. Schutze, K., H. Posl, et  al. (1998). “Laser micromanipulation systems as universal tools in cellular and molecular biology and in medicine.” Cell Mol Biol (Noisy-le-grand) 44: 735–746. 1 7. Fend, F., M. R. Emmert-Buck, et  al. (1999). “Immuno-LCM: laser capture microdissection of immunostained frozen sections for mRNA analysis.” Am J Pathol 154: 61–66. 18. Namimatsu, S., M. Ghazizadeh, et al. (2005). “Reversing the effects of formalin fixation with citraconic anhydride and heat: a universal antigen retrieval method.” J Histochem Cytochem 53: 3–11. 19. Uneyama, C., M. Shibutani, et  al. (2002). “Methacarn fixation for genomic DNA analysis in microdissected, paraffin-embedded tissue specimens.” J Histochem Cytochem 50: 1237–1245. 20. Friedl, P. and K. Wolf (2003). “Tumour-cell invasion and migration: diversity and escape mechanisms.” Nature Reviews Cancer 3: 362–374.

Chapter 21 Laser-Capture Microdissection of Renal Tubule Cells and Linear Amplification of RNA for Microarray Profiling and Real-Time PCR Susie-Jane Noppert, Susanne Eder, and Michael Rudnicki Abstract Laser-capture microdissection and transcriptional profiling have enabled compartment- and cell-specific analysis of gene expression in chronic kidney disease, thus facilitating the investigation of pathophysiological associations between glomerular, tubular, and interstitial structures. Due to the pico- and nanogram amounts of RNA isolated from LCM-captured material linear RNA amplification protocols are necessary prior to real-time PCR and microarray analysis. In this chapter, we describe the isolation of renal tubule cells from cryocut sections from routine kidney biopsies, and the isolation and linear amplification of RNA for downstream purposes. Key words: Renal, Kidney, Tubule, Microarray, Linear amplification

1. Introduction In the last decade, a molecular-based characterization of chronic renal diseases has been proposed based on studies using novel molecular analysis tools, such as microarray technology and quantitative real-time PCR (qRT-PCR) (1, 2). Using these techniques, several biomarkers have successfully been identified in the tubulointerstitial compartment which have been associated with progression of renal injury (1). This is in-line with the well known fact that prognosis of kidney disease depends more on the degree of tubulointerstitial scarring than on the degree of glomerular damage. However, gene expression between each of the compartments varies considerably (3). Since the tubulointerstitium comprises a mix of several cellular subpopulations (i.e., tubular cells, fibroblasts, leukocytes, endothelial cells), isolation of specific Graeme I. Murray (ed.), Laser Capture Microdissection: Methods and Protocols, Methods in Molecular Biology, vol. 755, DOI 10.1007/978-1-61779-163-5_21, © Springer Science+Business Media, LLC 2011

257

258

S.-J. Noppert et al.

cells of interest (e.g., renal tubule cells) may help elucidate the mechanisms of renal disease progression in more detail (4). Laser-capture microdissection (LCM) can be applied to any tissue that is accessible through biopsy and it is used to isolate cells of interest from complex, heterogeneous cell populations with high specificity (5). Briefly, a thermoplastic transfer film is placed in contact with a 5-mm tissue section. The film is precisely activated by a 980-nm gallium–arsenide laser pulse and binds strongly to the selected cells. Then the target cells can safely be removed, and RNA can easily be isolated using various isolation methods. LCM and downstream analysis by qRT-PCR have been used successfully to determine gene expression in glomeruli and specific parts of the nephron (6, 7). The limitation of LCM is the minute amount of isolated RNA which is usually in the range of picograms to nanograms. The introduction of a T-7-RNA polymerase-based linear amplification protocol enabled amplification of RNA by 1,000-fold, and performing two rounds of this amplification protocol even allowed the analysis of the gene expression of single live neurons (8, 9). Furthermore, linear amplification of RNA isolated from tubular cells has been shown to yield robust microarray results (10). Just recently, a linear preamplification method using TaqMan primers and probes has been introduced for amplification of minute amounts of RNA from organ biopsies for qRT-PCR (11). This chapter describes the isolation of renal tubule cells (RTCs) by LCM, the linear amplification of RNA for microarray hybridizations, and the preamplification of RNA for qRT-PCR.

2. Materials The protocols for tissue collection of renal biopsies suitable for LCM have recently been published in detail by others (12, 13). 2.1. Staining of RTCs for LCM

1. Staining chambers and rack. 2. NBT staining solution: Dissolve 500 mg 4-nitroblue tetrazolium (NBT) in 10 ml dimethylformamide (70%). 3. BCIP staining solution: Dissolve 500 mg 5-bromo-4-chloro3-indolyl phosphate (BCIP) p-toluidine salt in 10 ml dimethylformamide (100%). 4. Staining buffer: 12.11  g Trishydroxymethylaminomethane (Tris–HCl), 5.84 g NaCl, 10.17 g MgCl2, RNase-free water to 1,000 ml, adjust to pH 9.5 with 1N HCl; the final buffer contains 0.1 M Tris–HCl, 0.1 M NaCl, and 50 mM MgCl2. 5. RNase-free water.

21  Laser-Capture Microdissection of Renal Tubule Cells¼

259

6. 3% glycerol. 7. Xylene. 8. 100% ethanol. Staining stock solutions should be protected from light and stored at +4°C. 2.2. Laser-Capture Microdissection of Renal Tubule Cells

1. Laser-capture microscope Pixcell II™ (Arcturus, Mountain View, CA).

2.3. RNA Isolation from LCM-Captured Cells

1. PicoPure™ RNA isolation kit (Arcturus, Mountain View, CA).

2.4. Linear Amplification of RNA for Microarray Hybridization

1. Ovation® Pico WTA System (NuGEN Technologies, Inc., San Carlos, CA).

2. CapSure™ HS LCM Caps (Arcturus, Mountain View, CA).

2. 70% ethanol.

2. SPRIPlate® 96R, Ring Magnet Plate (Beckman Coulter Genomics, Bernried Germany). 3. 100% ethanol. 4. MinElute® Reaction Cleanup Kit (Qiagen, Hilden, Germany).

2.5. Preamplification of RNA for qRT-PCR

1. High capacity cDNA Reverse Transcription Kit (Applied Biosystems, Foster City, CA). 2. TaqMan™ PreAmp Master Mix (Applied Biosystems, Foster City, CA). 3. TaqMan™ Gene Expression Assays (Applied Biosystems, Foster City, CA). 4. RNase-free water.

3. Methods This section describes the RTC-specific alkaline phosphatase staining, LCM of tubule cells, RNA isolation and linear RNA amplification for qRT-PCR, and/or microarray analysis. Tissue collection has been described in detail elsewhere (13). In brief, renal biopsies are procured using a 16-gauge needle, embedded in OCT™ (Tissue Tek™, Fisher, Pittsburgh, PA), cut into 5-mm thick sections at −20°C using a cryomicrotome. The sections are then attached to glass slides and stored at −80°C.

260

S.-J. Noppert et al.

3.1. Staining of Renal Tubule Cells for LCM

1. Perform all staining steps at room temperature. 2. Prepare the stain by combining 100 ml staining buffer with 450  ml NBT and 350  ml BCIP stock solution. The stain is photosensitive, minimize exposure to light. 3. Place the following solutions into sterilized dye chambers: Chamber 1: RNase-free water. Chamber 2: 3% glycerol. Chamber 3: 100% ethanol. Chamber 4: 100% ethanol. Chamber 5: xylene. Chamber 6: xylene. 4. Submerge sections briefly in ethanol, drain. 5. Sumberge sections in water, swirl, and drain. 6. Wet sections with stain and leave for 1–2 min, drain. 7. Submerge sections in 3% glycerol for 20 min. 8. Submerge sections in ethanol, swirl, and drain. Repeat in second chamber of ethanol. 9. Submerge sections in xylene for 20 min, drain. Repeat in second chamber. 10. Allow sections to air-dry.

3.2. Laser-Capture Microdissection of Renal Tubule Cells

1. Prepare a 0.5-ml microcentrifuge tube and add 50 ml extraction buffer (XB). 2. Place the slide with the stained sections on the microscope. 3. The tubule cells are stained violet to dark-blue. Locate and mark these target areas (Fig. 1). 4. Place the LCM CapSure™ HS Cap onto the slide. 5. Pulse the laser. Approximately 2,000–4,000 pulses are enough to obtain 1–3  ng of total RNA. In our experience, a laser pulse duration of 3 ms and a laser pulse energy of 60 mW are sufficient (see Note 1). 6. Remove the CapSure™ HS Cap and insert it onto the prepared microcentrifuge tube. 7. Invert the microcentrifuge tube and make sure that the XB covers the complete LCM Cap. 8. Incubate the tube and the cap for 30 min at 42°C in an incubation block. 9. Centrifuge the assembly at 800 × g for 2  min to collect the extract into the microcentrifuge tube. 10. Remove the LCM cap and save the microcentrifuge tube containing the cell extract. 11. Proceed to RNA isolation or store at −80°C.

21  Laser-Capture Microdissection of Renal Tubule Cells¼

261

Fig. 1. Cryosection of renal tubule cells from a routine cryocut kidney biopsy stained for alkaline phosphatase. (a) After NBT/BCIP staining alkaline phosphatase in the brush border of the tubule cells stain violet to dark blue. (b, c) The stained tubule cells are targeted, laser captured, and adhered to the transfer film. (d) Kidney biopsy section after LCM of renal tubule cells. Magnification ×100.

3.3. RNA Isolation from LCM-Captured Cells

1. For each sample prepare an RNA purification column by adding 250 ml conditioning buffer to the filter membrane, incubate for 5 min at room temperature and centrifuge the column for 1 min at 16,000 × g, discard the flowthrough liquid. 2. Pipette 10 ml of 70% ethanol into the cell extract, mix well by pipetting, and add to the prepared purification column. 3. Centrifuge for 2 min at 100 × g, then immediately centrifuge for 30 s at 16,000 × g, discard flowthrough. 4. Add 100 ml wash buffer 1 to the column and centrifuge for 1 min at 8,000 × g, discard flowthrough. 5. Add 100 ml wash buffer 2 to the column and centrifuge for 1 min at 8,000 × g. Add another aliquot of 100 ml wash buffer 2 and centrifuge for 2 min at 16,000 × g. 6. Transfer the column to a fresh tube and pipette 11 ml elution buffer directly onto the membrane inside the column. 7. Incubate the column for 1 min at room temperature. 8. Centrifuge the column for 1  min at 1,000 × g followed by 1 min at 16,000 × g. 9. The eluted RNA can be used immediately or stored at −80°C (see Note 2).

262

S.-J. Noppert et al.

3.4. Linear Amplification of RNA for Microarray Hybridization

RNA isolated from LCM captured cells is used to generate amplified cDNA in a three-step process with the Ovation Pico WTA™ system: First strand cDNA synthesis, DNA/RNA heteroduplex double strand cDNA synthesis, and SPIA isothermal linear amplification. The amplification reaction generates cDNA with sequence complementary to the original mRNA. A minimum amount of 500 pg total RNA is required and an average 15,000fold amplification is achieved resulting in microgram amounts of cDNA (see Note 3). The amplified cDNA produced is then cleaned with the MinElute™ Reaction Cleanup Kit and can undergo labeling with an appropriate dye. We use the Agilent Genomic DNA Enzymatic Labeling Kit prior to analysis with Agilent Gene Expression microarrays (see Note 4).

3.4.1. Synthesis of First Strand cDNA

1. Thaw first strand buffer mix, First strand primer mix, and water at room temperature, vortex to mix, and centrifuge briefly to spin down contents. Maintain buffer mix and primer mix on ice until required. Spin down contents of first strand enzyme mix and place on ice. 2. Combine 2 ml of first strand primer mix with 5 ml total RNA (500 pg to 50 ng), incubate at 65°C for 2 min then immediately place on ice. 3. Prepare a master mix containing 2.5 ml first strand buffer and 0.5 ml first strand enzyme mix. To allow for reagent that is lost during pipetting the master mix should include at least one more reaction than you have samples, one additional reaction for every ten samples is usually sufficient. Add 3 ml of this master mix to each sample, mix by pipetting, and spin down. Place samples into a precooled thermocycler programmed as follows: 4°C for 1 min, 25°C for 10 min, 42°C for 10 min, 70°C for 15 min, and hold at 4°C. 4. Once the program is complete and samples have cooled to 4°C briefly spin samples and place on ice. Proceed immediately with second strand synthesis.

3.4.2. Synthesis of Second Strand cDNA

1. Allow Agencourt RNAClean purification beads to warm to room temperature. 2. Thaw second strand buffer mix, vortex briefly, spin down, and place on ice. Spin down the contents of the second strand enzyme mix and place on ice. 3. Prepare a master mix containing 9.75 ml Second strand buffer mix and 0.25 ml Second strand enzyme mix for each sample plus an additional reaction to allow for volume loss during transfer. Add 10 ml of this master mix to each sample, mix by pipetting, and spin down. 4. Place samples into a precooled thermocycler programmed as follows: 4°C for 1 min, 25°C for 10 min, 50°C for 30 min, 70°C for 5 min, and hold at 4°C.

21  Laser-Capture Microdissection of Renal Tubule Cells¼

263

5. Once the program is complete and samples have cooled to 4°C briefly spin samples and place at room temperature. Proceed immediately with purification of unamplified cDNA. 6. Ensure the Agencourt RNAClean purification beads have reached room temperature, resuspend beads by inverting and tapping the tube. 7. Add 32 ml of bead suspension to each sample, mix thoroughly by pipetting up and down ten times, and incubate at room temperature for 10 min. 8. Transfer samples to a magnet plate and stand for 5  min to clear the solution of beads. Carefully remove 45  ml of the binding buffer and discard. 9. With the samples still on the magnet add 200  ml of freshly prepared 70% ethanol and allow to stand for 30 s. 10. Carefully remove the ethanol wash with a pipette and discard. 11. Repeat this wash step two more times taking care to remove as much ethanol as possible after the final wash. 12. Air dry the beads on the magnet for 15–20 min then proceed immediately with the amplification step. 3.4.3. Amplification of cDNA

1. Thaw the SPIA buffer mix and the SPIA Primer mix at room temperature, vortex briefly, spin down and place on ice. Thaw the SPIA enzyme on ice, mix by inverting the tube gently five times (minimize bubble formation), spin briefly and place on ice. 2. Prepare a master mix by sequentially combining 80 ml SPIA buffer mix, 40 ml SPIA primer mix and 40 ml SPIA enzyme mix per sample. Add 160  ml of this master mix to the samples bound to the dried beads. Use a pipette to mix each sample well and resuspend the beads. Transfer one half of each sample (80 ml) into a second tube. 3. Place samples in a precooled thermal cycler programmed as follows: 4°C for 1 min, 47°C for 60 min, 95°C for 5 min, and hold at 4°C. 4. Once the program is complete and samples have cooled to 4 C briefly spin samples and place on ice. Recombine the half reactions. 5. Transfer samples to a magnet plate for 5 min to clear the solutions of beads. Carefully remove all of the supernatant containing the eluted cDNA and transfer to a fresh tube. 6. At this stage the cDNA can be stored at −20°C or purified using the MinElute™ Reaction Cleanup Kit. Clean cDNA is ready to undergo labeling and hybridization to the oligonucleotide microarray platform of choice.

264

S.-J. Noppert et al.

3.5. Preamplification of RNA for qRT-PCR

3.5.1. Reverse Transcription Reaction

First, RNA is converted to single-stranded cDNA by a reverse transcription enzyme. Then, the preamplification process selectively amplifies a cohort of genes in a linear manner so that an expression profile for the selected genes may be determined from a small sample of cDNA. The protocol can be adapted for either 1,000-fold (ten cycles) or 16,000-fold (14 cycles) amplification of the selected genes. Before proceeding with preamplification of cDNA from the limited samples, it is vital to check the TaqMan assays for preamplification uniformity using a control cDNA from a nonlimited source (see Note 5). 1. Thaw high-capacity cDNA reverse transcription kit reagents and RNA samples on ice. 2. To allow for reagent that is lost during pipetting the master mix should include at least one more reaction than you have samples, one additional reaction for every ten samples is usually sufficient. Prepare a 2× master mix using the following quantities of reagent per reaction, mix gently and maintain on ice: 2 ml 10× RT buffer, 0.8 ml 25× dNTP mix (100 mM), 2 ml 10× RT random primers, 1 ml MultiScribe™ Reverse Transcriptase, 1 ml RNase Inhibitor, and 3.2 ml nuclease-free water. 3. Pipette 10 ml RNA into a PCR tube and add 10 ml master mix to the sample, mix by gentle pipetting and seal the tube. If necessary briefly centrifuge the samples to spin down contents to the bottom of the PCR tubes, maintain on ice until ready to transfer to the thermal cycler. 4. Create the following program in the thermal cycler you will use: 25°C for 10 min, 37°C for 120 min, 85°C for 5 s, and 4°C hold. 5. Load the samples into the thermal cycler and begin the program. 6. Once the cDNA samples have cooled remove them from the thermal cycler. The cDNA may be transferred to a RNase-free microcentrifuge tube and stored at −20°C until required or used immediately.

3.5.2. Preamplification Reaction

1. Prepare a pooled assay mix of the TaqMan gene expression assays for the genes of interest, a maximum of 100 genes can be preamplified. Thaw the 20× assays on ice and pipette 5 ml of each assay into a nuclease-free microcentrifuge tube. Add 1× TE buffer to a final volume of 500 ml. For example, if the pooled assay mix contains 5 ml each of ten assays add 450 ml TE buffer for a final volume of 500 ml. Each assay is present at a final concentration of 0.2×. The pooled assay mix can be frozen at −20°C for storage or used immediately. 2. Pipette 1–250 ng cDNA sample into a PCR tube, aim to use a consistent amount for all samples, and adjust the volume

21  Laser-Capture Microdissection of Renal Tubule Cells¼

265

with nuclease-free water to 10  ml. Add 10  ml pooled assay mix and 20 ml TaqMan® PreAmp master mix to each sample, mix gently by pipetting, and maintain on ice until ready to transfer to the thermal cycler. 3. Create the following program in the thermal cycler: 95°C for 10 min, ten cycles at 95°C for 15 s, 60°C for 4 min. 4. Load the samples into the thermal cycler and begin the program. Once cycling is complete it is important to immediately transfer the samples from the cycler onto ice to halt the reaction. 5. Transfer the preamplified sample to a nuclease-free microcentrifuge tube and dilute with 160 ml 1× TE buffer. The preamplified cDNA can be stored at −20°C or used immediately in a PCR reaction.

4. Notes 1. When more than 5,000 LCM pulses are performed we increase the laser energy to 80 mW. For glomeruli longer durations than 3 ms may be needed, e.g., 10 ms. We always take the smallest of the three sizes of the laser beam. Try to focus the laser beam exactly to the level of the section placed on the slide. The microdissection works best when you allow the cap to warm up for 2–3 min. 2. RNA quality control can be performed using Agilent Bioanalyzer RNA Pico Chips™ (both Agilent Technologies, Palo Alto, CA). However, due to the minute amounts of RNA we rarely perform this quality control step. We rather measure the expression values of certain housekeeper genes such as cyclophilin A (PPIA). 3. Although recommended by others (13), we do not perform a DNase step since we do not experience contamination by DNA and we observe a considerable degree of RNA degradation during this treatment. 4. In our experience, 90% of target genes show uniform and reproducible preamplification using the TaqMan™ PreAmp protocol, which is also reported by Applied Biosystems. Since the reason for a reliable linear preamplification is not entirely clear we strongly recommend testing the linearity of each assay on nonlimited RNA. There have been published reports of frequently used housekeeping genes such as HPRT (11) and GAPDH (14) not attaining DDCt values within the ±1.5 range. The endogenous uniformity reference recommended by Applied Biosystems is CDKN1B (Assay ID

266

S.-J. Noppert et al.

Hs00153277_m1). In kidney tissue. we have found PPIA (Assay ID Hs99999904_m1) to be a reliable reference gene. 5. The ovation™ Pico WTA System (NuGen) works best with RNA amounts >1 ng. References 1. Henger A, Kretzler M, Doran P et al (2004) Gene expression fingerprints in human tubulointerstitial inflammation and fibrosis as prognostic markers of disease progression. Kindey Int 65:904–917 2. Wenjun J, Eichinger F, Bitzer M et al (2009) Renal gene and protein expression signatures for prediction of kidney disease progression. Am J Pathol 174:2073–2085 3. Higgins J P, Wang L, Kambham N et  al (2004) Gene expression in the normal adult human kidney assessed by complementary DNA microarray. Mol Biol Cell 15: 649–656 4. Rudnicki M, Perco P, Enrich j et  al (2009) Hypoxia response and VEGF-A expression in human proximal tubular epithelial cells in stable and progressive renal disease. Lab Invest 89:337–346 5. Emmert-Buck MR, Bonner RF, Smith PD, et  al (1996) Laser capture microdissection. Science 274:998–1001 6. Kohda Y, Murakami H, Moe OW et al (2000) Analysis of segmental renal gene expression by laser capture microdissection. Kidney Int 57:321–331 7. Nagasawa Y, Takenaka M, Matsuoka Y et  al (2000) Quantitation of mRNA expression in glomeruli using laser-manipulated microdissection and laser pressure catapulting. Kidney Int 57:717–723

8. Van Gelder RN, von Zastrow ME, Yool A et al (1990) Amplified RNA synthesized from limited quantities of heterogeneous cDNA. Proc Natl Acad Sci USA 87:1663–1667 9. Eberwine J, Yeh H, Miyashiro K et al (1992) Analysis of gene expression in single live neurons. Proc Natl Acad Sci USA 89:3010–3014 10. Rudnicki M, Eder S, Schratzberger G et  al (2004) Reliability of t7-based mRNA linear amplification validated by gene expression analysis of human kidney cells using cDNA microarrays. Nephron Exp Nephrol 97: e86-e95 11. Noutsias M, Rohde M, Block A et al (2008) Preamplification techniques for real-time RT-PCR analyses of endomyocardial biopsies. BMC Mol Biol 14:3 12. Edgley AJ, Gow RM, Kelly DJ (2010) Lasercapture microdissection and pressure catapulting for the analysis of gene expression in the renal glomerulus. Methods Mol Biol 611:29–40 13. Woroniecki RP, Bottinger EP (2009) Laser capture microdissection of kidney tissue. Methods Mol Biol 466:73–82 14. Denning KM, Smyth PC, Cahill SF et  al (2007) A molecular expression signature distinguishing follicular lesions in thyroid carcinoma using preamplification RT-PCR in archival samples. Mod Pathol 20:1095–1102

Chapter 22 Subcellular Renal Proximal Tubular Mitochondrial Toxicity with Tenofovir Treatment James J. Kohler and Seyed H. Hosseini Abstract Nucleoside reverse transcriptase inhibitors (NRTIs) are drugs used in the treatment of HIV/AIDS. Despite the distinct benefits of NRTI-based therapies, tissue specific toxicity is a limiting factor. Although the mechanisms of these specific antiretroviral drug-related toxicities remain unclear, it has been hypothesized that as analogs to native nucleosides, NRTIs may potentially inhibit mammalian DNA polymerases, including mitochondrial DNA (mtDNA) polymerase g. Tenofovir disoproxil fumarate (TDF) is a nucleotide analog of adenosine monophosphate and the only NRTI that is associated with renal disease. The inherent heterogeneity of kidney tissues could affect the outcome and interpretation of molecular studies to define the mechanism(s) of tenofovir tubular toxicity. Laser-capture microdissection (LCM) provided a specific, single-cell isolation of proximal tubules from fixed heterogeneous kidney tissues. LCM-captured renal proximal tubules from transgenic mice (TGs) showed decreased mtDNA abundance with tenofovir, demonstrating a subcellular specific mitochondrial toxicity of tenofovir in an AIDS model. Key words: NRTIs, TDF, Renal proximal tubules, Mitochondrial toxicity, Renal disease, Transgenic mice, mtDNA abundance

1. Introduction Despite the distinct benefits of NRTI-based therapies, a limiting factor is toxicity. In particular, a number of in vitro and in vivo studies demonstrated cardiomyopathy and hepatic failure associated with specific NRTIs (1–4). NRTI toxicities appear to be tissuespecific (5, 6). Zidovudine and stavudine are toxic to striated muscle (2, 7). Didanosine (ddI) is toxic to liver and pancreas (4, 8). An important tenofovir toxicity is renal tubular dysfunction (9–12), but its mechanism or precise target is not elucidated. It has been hypothesized that as analogs to native nucleosides, tenofovir and other NRTIs can potentially inhibit mammalian DNA Graeme I. Murray (ed.), Laser Capture Microdissection: Methods and Protocols, Methods in Molecular Biology, vol. 755, DOI 10.1007/978-1-61779-163-5_22, © Springer Science+Business Media, LLC 2011

267

268

J.J. Kohler and S.H. Hosseini

polymerases including mitochondrial DNA polymerase g (pol g) and induce oxidative stress (13–15). One approach to determine mitochondrial toxicity is to measure changes in mtDNA abundance from target tissues following treatment with potential toxin. Tenofovir disoproxil fumarate (TDF) is an oral prodrug of tenofovir (TFV), an acyclic nucleotide analog of adenosine monophosphate widely used clinically as a nucleoside reverse transcriptase inhibitor (NRTI) for the treatment of human immunodeficiency virus type 1 (HIV-1) infection (16). Initial controlled clinical studies found TDF to be safe, with low incidence (1–3%) of TDF-associated renal impairment characterized by elevated serum creatinine or hypophosphatemia (17–19). Several reports have reassessed the renal safety (20) with a growing subset of TDF-treated patients presenting with acute renal failure (21–25). In the majority of the cases, the kidney damage was reversed with discontinuation of TDF treatment (26). These reports have raised questions about potential kidney-specific toxicity and TDFassociated increased risk of tubulopathy, which parallels increased risks associated with other related antiviral acyclic nucleoside phosphonates such as cidofovir and adefovir (27). Many of these case reports have suggested different mechanisms to explain the link between TDF and renal toxicity. Laser-capture microdissection (LCM) is utilized here to anatomically isolate renal proximal tubules from fixed tissues to define effects related to tenofovir on tubules in HIV model.

2. Materials 2.1. Animals

1. Hemizygous HIV-1 TG mice (28), compliment from Paul Klotman. Originally on FVB/n background, this TG line bred congenically to C57/BL6 (and given the trivial name “MSB”). Murine TG authenticity confirmed for each generation using dot blot analysis and real-time PCR (29). 2. Age and gender matched, including both male and female, wildtype (WT) and TGs. Each cohort (“2 × 2” TG and WT with TDF or vehicle alone) consisted of ~20 mice (see Note 1). 3. Standard rodent chow and water provided ad libitum in a 12 h light–dark, humidity- and temperature-controlled environment in Division of Animal Resources (http://www.dar. emory.edu/admin_about.html) at Emory University School of Medicine. 4. Tenofovir disoproxil fumarate (TDF, Viread™, Gilead Sciences) provided by manufacturers (compliment from Professor Raymond Schinazi, VA Medical Center, Decatur, GA, USA and Emory Center for AIDS Research Pharmacology Core).

22  Subcellular Renal Proximal Tubular Mitochondrial Toxicity with Tenofovir Treatment

269

5. Doses resemble human therapy on a mg/kg/day basis (TDF, 300 mg/60 kg human = 5 mg/kg; ~25 g mouse = 0.125 mg/ day). 6. Vehicle (buffered with 0.1 M NaOH). 2.2. Formalin-Fixed, Paraffin-Embedded (FFPE) Sections of Kidney Tissues for LCM

Formalin (Fisher Diagnostics, Middletown, VA), neutral buffered, diluted 1:10. Liquefied paraffin infiltration of fixed tissue (standard procedure).

2.3. Periodic Acid/ Schiff (PAS) Stain

Periodic acid (Sigma Aldrich, St. Louis, MO). Schiff reagent (Fisher Scientific, Pittsburgh, PA).

2.4. LCM

1. MMI Cellcut system (MMI Molecular Machines & Industries, Inc., Haslett, MI). 2. MMI membrane mounted metal slides (MMI Molecular Machines & Industries, Inc., Haslett, MI). 3. MMI isolation caps, 500-ml tubes with adhesive lid with diffuser (MMI Molecular Machines & Industries, Inc., Haslett, MI).

2.5. DNA Extraction

PicoPure™ DNA Extraction Kit (Arcturus Biosciences, Inc., Mountainview, CA).

2.6. LightCycler 480 mtDNA/nDNA Multiplex Protocol for the LCM Samples

LightCycler 480 (Roche Diagnostics Corp., Indianapolis, IN); LightCycler 96-multiwell plate (Roche Diagnostics Corp.); LightCycler 96-multiwell block (Roche Diagnostics Corp.). 1. Probes (Roche Diagnostics Corp, special order): mtDNA probes: 3¢FmCOXPR1 (3¢fluorescein-mCOXPR1-5¢AAC CAG GTG CAC TTT TAG GAG ATG ACC-F3¢), and COX5¢L670 (5¢-LC Red 670 3¢-phosphate-blockedmCOXPR2, 5¢L-AAT TTA CAA TGT TAT CGT AAC TGC CCA TGC-P3). nDNA probes: 3¢FmASPGR1 (5 flourescein-mASPGPR1-5¢GCG CTT TGG ACC TTT GGG TGT AG-F3), and ASPG5¢610 (mASPGPR2-610-5 L-GTT ACG AAA GAA CCT AGC CTC ACA GTG GT-P3¢). 2. Primers: mtDNA primers: mmtLCCOX1F (5¢-TCG TTG ATT ATT CTC AAC CAA TCA-3¢), and mmtLCCOX2R (5¢-GCC TCC AAT TAT TAT TGG TAT TAC TAT GA-3¢). nDNA primers: mnucASPG1F (5¢-GGA GGA GGC ACT TTC TCA GC-3¢), and mnucASPG2R (5¢-GAA GAC CTG CTC CCT GAA CAC-3¢). 3. Genotyping master kit (cat# 04 707 524 001, Roche Diagnostics Corp.), contains PCR-grade H2O, and 5× premix.

270

J.J. Kohler and S.H. Hosseini

3. Methods 3.1. Tissue Sectioning

1. Cut 5-mm sections on clean microtome with a clean blade. 2. Float paraffin ribbons on 35–40°C deionized water to smooth out and eliminate folds and wrinkles. 3. Mount sections on membrane mounted metal frame slides. 4. Air-dry and bake for 45 min at 60°C (paraffin must be removed from tissue sections and the tissue must be rehydrated).

3.2. PAS Staining

1. Immerse slides in xylene for 5 min. Repeat twice. 2. Immerse slides in 100% ethanol for 2 min. Repeat twice. 3. Immerse slides in 95% ethanol for 1 min. Repeat twice. 4. Immerse slides in 70% ethanol for 1 min. 5. Wash in running tap water for 2 min. 6. Immerse slides in periodic acid for 5 min. 7. Wash in distilled water for 2 min. 8. Immerse slides in Schiff reagent for 10 min. 9. Wash in running tap water for 10 min. 10. Immerse slides in 70% ethanol for 30 s. 11. Immerse slides in 95% ethanol for 30 s. Repeat twice. 12. Immerse slides in 100% ethanol for 30 s. Repeat twice. 13. Immerse slides in xylene for 30 s. Repeat twice. 14. Air-dry for 2–5 min. Ready for LCM.

3.3. LCM of Proximal Tubules from FFPE Tissue Sections Stained with PAS

H&E is the most commonly used general histochemical stain for visualization of tissues under light microscopy for LCM, with very little impact on DNA integrity from isolated cells. However, we have selected PAS staining to aid in visual delineation and selection of proximal tubules (which have glycogen-rich brush borders) from distal tubules (glycogen poor) (Fig. 1). Using standard LCM techniques, a set of proximal tubules are isolated from individual samples (one sample per slide): 1. Proximal tubules are identified by PAS-stained brush borders and selected as a cumulative group (Fig. 2). 2. To normalize each sample, isolation of proximal tubules is based upon total area (~750,000  mm2) rather than by total cell counts, as proximal tubules can vary in size. Arcturas software provides a running count of total area selected. 3. Isolated proximal tubules from each sample (~750,000 mm2) are collected in separate tubes and analyzed for mtDNA as individual samples.

22  Subcellular Renal Proximal Tubular Mitochondrial Toxicity with Tenofovir Treatment

271

Fig. 1. Histology section stained with PAS provides clear identification of proximal tubules (glycogen-rich brush border staining) from distal tubules and glomeruli.

Fig. 2. Proximal tubules selectively microdissected from formalin-fixed, paraffin-embedded (FFPE) kidney tissues using LCM. Captured images of renal tissue before (left ) and after (center ) sample microdissection along with resultant isolated proximal tubules (right, original magnification ×20). PT proximal tubules, DT distal tubules, G glomeruli.

4. Following isolation, slides are stored at 4°C until DNA extraction is confirmed (see Note 2). 3.4. DNA Extraction of LCM Isolated Proximal Tubules (see Note 3 )

1. Set culture incubator to 65°C. Time to preheat is approximately 1.5  h. Monitor to insure that the temperature has stabilized before incubating samples. 2. Pipette 155 ml of reconstitution buffer (included in the kit) into one vial of proteinase K (provided in kit). Gently pulse-vortex the tube to dissolve powder completely (excessive mixing may denature proteinase K). 3. Pipette 50 ml of the reconstituted proteinase K solution into each LCM sample tube. Close and invert the tube. Vortex briefly, shaking down the solution until it completely covers the inside surface of the LCM cap (see Note 4).

272

J.J. Kohler and S.H. Hosseini

4. Incubate the inverted tubes at 65°C overnight (~16–18  h) (see Note 5). 5. After incubation, preheat Thermomixer or water bath to 96°C. Briefly (~10 s) centrifuge the tubes (upright, not inverted) at 1,000 × g. Label new 1.5 ml microcentrifuge tubes for use in step 6 below (see Note 6). 6. To wash any residual DNA from the laser-capture (LC) tubes: (a) Pipette samples into tubes labeled in step 5, retaining the tips in the 1.5-ml tubes. (b) Using repeating pipettor, add 50 ml DNAse-free water to each LC tube, invert, and vortex briefly. (c) Using the tips from point (a) of step 6, transfer the water from each LCM tube into the corresponding 1.5-ml tube. Heat in thermomixer or water bath at 96°C for 15 min to inactivate proteinase K. 7. Samples may be stored at 4°C, or proceed to next step if performing mtDNA quantification immediately, or within ~24 h. 8. Briefly (<10 s) centrifuge samples at high speed, to pull any droplets of liquid down. 9. Place each 1.5-ml tube in the speed vacuum concentrator (speedvac), with caps completely open, and run speedvac (~2 h, or until no residual liquid remains). 10. Remove tubes and add 25 ml DNAse-free water to each tube. Store at 4°C (short term), −20°C (long term), or immediately begin mtDNA abundance protocol. 3.5. Real-Time PCR Analysis of mtDNA Abundance (see Note 7)

In a typical experiment, each well of the plate will hold 22 ml of the LCM extracted DNA and 28 ml of master mix for a total volume of 50 ml. The master mix in each well consists of: 1. 10 ml premix (genotyping master, 5×; yellow sticker on cap labeled “1”) 2. 4.0 ml of each primer (4 × 4 = 16.0 ml) 3. 0.5 ml of each probe (4 × 0.5 = 2.0 ml) Calculate master mix for the number of samples + number of calibrators and blanks + approximately [(1/15) × total # wells to fill] extra (to account for loss during plate loading). Then round up to the nearest 5 or 10. For example, for 40 samples, 4 calibrators, and 1 blank (a total of 45 wells needed), make sufficient master mix for 45 + [(1/15) × 45 = 3] = 48 wells. Round to 50. Check that the total volume will be correct by adding 500 + (4 × 200) + (4 × 25), which should equal 28 × 50 = 1,400 ml. Calibrators (see Note 8): Briefly, extract total DNA (genomic DNA) from various tissues (e.g., heart, skeletal muscle, etc.) from the organism (i.e., mouse), pool the DNA, measure the concentration,

22  Subcellular Renal Proximal Tubular Mitochondrial Toxicity with Tenofovir Treatment

273

and make serial dilutions (10−1, 10−2, 10−3, 10−4, 10−5, and 10−6) from the pooled samples. Perform light-cycler real-time PCR assay using each dilution (in triplicate). Create efficiency curves for both mtDNA and nDNA and save results as “external efficiency curves.” For each light cycler run, add proper dilutions of the calibrators based on the samples’ concentration range. For example, for LCM samples, lower concentrations calibrators are favorable. If the “Relative Quantification Method” is being used for data analysis, the “efficiency curve” for both mtDNA and nDNA is recommended to be applied to the run. However, one can use an embedded efficiency of two (Eff 2) for both mtDNA and nDNA to determine the mtDNA/nDNA ration of each sample (see Note 9). Setting up the LC480 run: 1. Remove the reagents from the −20°C storage (see Note 10). 2. Pulse-vortex and then pipette the volume needed of each component (first premix, then primers, and probes last) into a sterile microcentrifuge tube large enough to contain the entire volume of master mix (if >2 ml master mix is needed, it may be prepared in aliquots and combined in a sterile trough before loading). After all components have been added, vortex the tube briefly. 3. Load 28 ml of master mix into the wells to be used. A singlechannel pipettor, multichannel pipettor, or the Precision XS may be employed. Take appropriate precautions to ensure that precision and reliability are maintained (see Note 11). 4. Load 22 ml of each LCM sample, calibrator, and water (blank) into the appropriate wells (see Note 12). 5. Seal the plate with LC480 sealing foil using the sealing tool. 6. Place the plate in a standard swing-bucket centrifuge with appropriate rotor for multiwell plate and centrifuge at 1,500 × g for 1 min (see Note 13). Transfer the plate to the LC 480 instrument. 7. Start the LC 480 (or other real-time PCR instrument available to you) and apply the appropriate program (see Note 14). 3.6. Determining mtDNA Abundance (mtDNA/nDNA Ratio) from LightCycler 480

After the run is completed, data are analyzed based on the type of real-time PCR instrument. If LightCycler 480 (LC-480, Roche) is being used, data can be analyzed either by “Relative Quantification Analysis” module or “Absolute Quantification Analysis” module of the data analysis software (Roche). Refer to Roche instrument manual for specific details. In summary: 1. Open the sample editor and enter the sample information. 2. From “create new analysis” dialog box, choose appropriate analysis type (i.e., Relative Quant).

274

J.J. Kohler and S.H. Hosseini

Fig. 3. Changes in mtDNA abundance after TDF treatment: using “2 × 2” protocols, cohorts of TG and WT mice were treated with TDF or vehicle (5 weeks). After treatment, cardiac, kidney, and liver tissues were collected. Total DNA extractions from whole tissues were performed and mtDNA abundance (ratio of mtDNA/nDNA) determined for each cohort as normalized ratios of mtDNA/nDNA (reproduced from ref. 29 with permission from Nature Publishing Group).

Fig. 4. mtDNA abundance from LCM isolated renal proximal tubules were determined (reproduced from ref. 29 with permission from Nature Publishing Group).

3. Apply appropriate “color compensation” object to select for correct filter combination, based o the type of dye in use (see Note 15). 4. Activate the “calculate” key to calculate the CP values of all samples. 5. From the CP values, choose appropriate calibrator with similar CP value to majority of the samples (i.e., 10−2). 6. Create a new sample set containing all the samples with the chosen calibrator. 7. Apply a previously saved efficiency curve or default setting (Eff = 2) (please see Subheading 3.5 above) and analyze the samples.

22  Subcellular Renal Proximal Tubular Mitochondrial Toxicity with Tenofovir Treatment

275

8. The analyzed data will be shown as the “normalized ratio” of mtDNA to nDNA (Fig. 3). 9. For statistical analysis, transfer data to appropriate statistical analysis software (i.e., GraphPad Prism, GraphPad Software, Inc., La Jolla, CA) (Fig. 4).

4. Notes 1. The number of mice per cohort (N = 20) was determined previously to be the necessary minimum to provide statistical power for mtDNA abundance data. 2. Stained slides are stable for up to 1 week and can be reused if necessary. 3. Modified from PicoPure DNA extraction kit (KIT0103) user guide (version B, p. 11). 4. Do not press the tubes too far down into the holder or else the caps may come open and spill the solution when you remove the tubes. 5. Optionally, vortex samples occasionally during incubation (but not more than two times). 6. Do not exceed recommended force or duration, as this may dislodge the polymer cap into the tube. 7. The following protocol is designed for LightCycler 480 realtime PCR instrument (Roche) but it may be adjusted for other real-time PCR instruments. 8. Refer to the manufacturer’s (Roche) protocol regarding use of “calibrator.” 9. If the “Absolute Quantification” module is used, then a standard curve must be prepared and applied during analysis (please see the manufacturer’s protocol for details). 10. Put premix on ice immediately; all other reagents should be allowed to thaw and then put on ice. Protect probes from light (supplied in amber vials, but also keep them covered with foil). 11. Master mix is viscous, so pipette slowly and carefully. 12. Be careful not to produce air bubbles when pipetting in samples, as they will interfere with LightCycler measurements. 13. Check for air bubbles and repeat the process if necessary. 14. Refer to Roche LC 480 instrumentation guide for program protocol. 15. Since two different dyes are being detected, color compensation is required to correct for spectral overlap, also known as “crosstalk.” Please see the manufacturer’s manual for more details.

276

J.J. Kohler and S.H. Hosseini

Acknowledgments The authors would like to thank Professor William Lewis (WL) for his advice and support and Dianne Alexis in the Pathology Core Laboratory for assistance in the histological staining and tissue section protocols. This work was supported by R01 HL79867 to WL; JK is a recipient of K01 DK78513. References 1. Lewis, W., Gonzalez, B., Chomyn, A., et  al. (1992) Zidovudine induces molecular, biochemical, and ultrastructural changes in rat skeletal muscle mitochondria. J Clin Invest 89: 1354–1360. 2. Lewis, W., Grupp, I. L., Grupp, G., et  al. (2000) Cardiac dysfunction occurs in the HIV-1 transgenic mouse treated with zidovudine. Lab Invest 80: 187–197. 3. Lewis, W. (2003) Mitochondrial DNA replication, nucleoside reverse-transcriptase inhibitors, and AIDS cardiomyopathy. Prog Cardiovasc Dis 45: 305–318. 4. Sulkowski, M. S., Mehta, S. H., Torbenson, M., et al. (2005) Hepatic steatosis and antiretroviral drug use among adults coinfected with HIV and hepatitis C virus. Aids 19: 585–592. 5. Lund, K. C., Peterson, L. L. & Wallace, K. B. (2007) Absence of a universal mechanism of mitochondrial toxicity by nucleoside analogs. Antimicrob Agents Chemother 51: 2531–2539. 6. Bruggeman, L. A., Thomson, M. M., Nelson, P. J., et al. (1994) Patterns of HIV-1 mRNA expression in transgenic mice are tissue- dependent. Virology 202: 940–948. 7. Lewis, W., Haase, C. P., Raidel, S. M., et  al. (2001) Combined antiretroviral therapy causes cardiomyopathy and elevates plasma lactate in transgenic AIDS mice. Laboratory Investigation 81: 1527–1536. 8. Price, C. J., George, J. D., Marr, M. C., et al. (2006) Prenatal developmental toxicity evaluation of 2¢,3¢-dideoxyinosine (ddI) and 2¢,3¢-didehydro-3¢-deoxythymidine (d4T) coadministered to Swiss Albino (CD-1) mice. Birth Defects Res B Dev Reprod Toxicol 77: 207–215. 9. Fux, C. A., Christen, A., Zgraggen, S., et  al. (2007) Effect of tenofovir on renal glomerular and tubular function. Aids 21: 1483–1485. 10. Mocroft, A., Kirk, O., Gatell, J., et al. (2007) Chronic renal failure among HIV-1-infected patients. Aids 21: 1119–1127.

11. Cote, H. C., Magil, A. B., Harris, M., et  al. (2006) Exploring mitochondrial nephrotoxicity as a potential mechanism of kidney dysfunction among HIV-infected patients on highly active antiretroviral therapy. Antivir Ther 11: 79–86. 12. Vidal, F., Domingo, J. C., Guallar, J., et al. (2006) In vitro cytotoxicity and mitochondrial toxicity of tenofovir alone and in combination with other antiretrovirals in human renal proximal tubule cells. Antimicrob Agents Chemother 50: 3824–3832. 13. Birkus, G., Hajek, M., Kramata, P., et  al. (2002) Tenofovir diphosphate is a poor substrate and a weak inhibitor of rat DNA polymerases alpha, delta, and epsilon*. Antimicrob Agents Chemother 46: 1610–1613. 14. Lewis, W. (2003) Mitochondrial dysfunction and nucleoside reverse transcriptase inhibitor therapy: experimental clarifications and persistent clinical questions. Antiviral Res 58: 189–197. 15. Lewis, W., Copeland, W. C. & Day, B. (2001) Mitochondrial DNA depletion, oxidative stress and mutation: Mechanisms of nucleoside reverse transcriptase inhibitor toxicity. Laboratory Investigation 81: 777–790. 16. Lyseng-Williamson, K. A., Reynolds, N. A. & Plosker, G. L. (2005) Tenofovir disoproxil fumarate: a review of its use in the management of HIV infection. Drugs 65: 413–432. 17. Izzedine, H., Isnard-Bagnis, C., Hulot, J. S., et al. (2004) Renal safety of tenofovir in HIV treatment-experienced patients. Aids 18: 1074–1076. 18. Schooley, R. T., Ruane, P., Myers, R. A., et al. (2002) Tenofovir DF in antiretroviralexperienced patients: results from a 48-week, randomized, double-blind study. AIDS 16: 1257–1263. 19. Padilla, S., Gutierrez, F., Masia, M., et  al. (2005) Low frequency of renal function impairment during one-year of therapy with tenofovir-containing regimens in the real-world: a case-control study. AIDS Patient Care STDS 19: 421–424.

22  Subcellular Renal Proximal Tubular Mitochondrial Toxicity with Tenofovir Treatment 20. Nelson, M. R., Katlama, C., Montaner, J. S., et al. (2007) The safety of tenofovir disoproxil fumarate for the treatment of HIV infection in adults: the first 4 years. Aids 21: 1273–1281. 21. Barrios, A., Garcia-Benayas, T., GonzalezLahoz, J., et  al. (2004) Tenofovir-related nephrotoxicity in HIV-infected patients. Aids 18: 960–963. 22. Breton, G., Alexandre, M., Duval, X., et  al. (2004) Tubulopathy consecutive to tenofovircontaining antiretroviral therapy in two patients infected with human immunodeficiency virus1. Scand J Infect Dis 36: 527–528. 23. James, C. W., Steinhaus, M. C., Szabo, S., et al. (2004) Tenofovir-related nephrotoxicity: case report and review of the literature. Pharmacotherapy 24: 415–418. 24. Lochet, P., Peyriere, H., Le Moing, V., et al. (2005) [Assessment of renal abnormalities in 107 HIV patients treated with tenofovir]. Therapie 60: 175–181.

277

2 5. Peyriere, H., Reynes, J., Rouanet, I., et al. (2004) Renal tubular dysfunction associated with tenofovir therapy: report of 7 cases. J Acquir Immune Defic Syndr 35: 269–273. 26. Zimmermann, A. E., Pizzoferrato, T., Bedford, J., et al. (2006) Tenofovir-associated acute and chronic kidney disease: a case of multiple drug interactions. Clin Infect Dis 42: 283–290. 27. Izzedine, H., Launay-Vacher, V., Isnard-Bagnis, C., et  al. (2003) Drug-induced Fanconi’s syndrome. Am J Kidney Dis 41: 292–309. 28. Dickie, P., Felser, J., Eckhaus, M., et al. (1991) HIV-associated nephropathy in transgenic mice expressing HIV-1 genes. Virology 185: 109–119. 29. Kohler, J. J., Hosseini, S. H., Hoying-Brandt, A., et al. (2009) Tenofovir renal toxicity targets mitochondria of renal proximal tubules. Lab Invest 89: 513–519.

wwwwwww

Chapter 23 Application of Laser-Capture Microdissection to Study Renal Carcinogenesis Kerstin Stemmer and Daniel R. Dietrich Abstract Kidney cancer is characterized by significant morphological and molecular heterogeneity. Evaluation of mechanisms involved in the development and progression of kidney cancer require comprehensive analyses of genomes, transcriptomes, proteomes, and methylation profiles in normal and tumor tissue. To date, indiscriminate homogenates of tumor tissue or biopsy samples have been used as a source for DNA, RNA, or protein isolation. A major technical improvement has been the development of laser-assisted microdissection that allows the isolation of morphologically similar cells. The applications of this techno­ logy to kidney cancer research are outlined. Key words: Kidney, Kidney cancer, Microarray, RNA, Renal carcinogenesis, Transcriptional profiling

1. Introduction 1.1. Current Developments and Methodological Challenges in Kidney Cancer Research

The kidney is one of the most complex organs of the human body. Its multitude of cells can be roughly divided into functional cells of the renal parenchyma that make up over one million nephrons, the supporting cells of the connective tissue (e.g., renal fibroblasts), and the endothelial cells of the blood vessels. Each functional nephron is further subdivided into a glomerulus and various tubular segments, all of which are composed of several unique cell types with highly specific physiological functions. It is, therefore, not surprising that all functional entities and even cell types comprise different susceptibilities toward physiological changes, and toxic or carcinogenic impairments. The type and cellular origin of toxic or carcinogenic lesions are, therefore, determined not only by the chemical properties of the respective compound but also

Graeme I. Murray (ed.), Laser Capture Microdissection: Methods and Protocols, Methods in Molecular Biology, vol. 755, DOI 10.1007/978-1-61779-163-5_23, © Springer Science+Business Media, LLC 2011

279

280

K. Stemmer and D.R. Dietrich

by its basolateral or apical uptake into the epithelial cells of a distinct tubular segment and by the respective cellular capacity for metabolism and damage repair. Indeed, the vast majority of all kidney tumors result from diverse environmental causes, resulting in a heterogeneous group of kidney tumors. Renal cell carcinoma (RCC) represents 80–85% of all kidney tumors in humans. According to their histological characteristics and cell type of origin, they are subclassified into clear cell RCC, papillary RCC, chromophobe RCC, oncocytoma, and collecting duct carcinoma (1). The finding that very similar kidney tumor phenotypes can be experimentally induced in rodents by distinct chemicals, hormones, radiation, or viruses (2) allowed to study the dynamics of kidney tumor development. These studies demonstrated that once initiated, growth of distinct RCC can pass through a morphological continuum from single atypical tubules to hyperplasia to adenoma and/or carcinoma. In most cases, fulminant RCC is morphologically highly complex, consisting of different altered cell clones that grow at different rates (3). Furthermore, regional differences in vasculature, host infiltrates, and connective tissue components build up a complex microenvironment braid around the tumor cells. These tumor–microenvironment interactions are increasingly recognized to influence tumor progression including tumor angiogenesis and metastasis. Recent studies have demonstrated that tumor cells derived from discrete areas of the same neoplastic mass can vary in their relative gene expression (4, 5) and, moreover, that changes in gene expression precede the occurrence of morphological and functional alterations, e.g., invasiveness of the lesion (6). Accordingly, the complex structure of the kidney and the even more allotropic and dynamic composition of kidney tumors and their tumor microenvironments greatly exacerbate the analysis of cellular and molecular characteristics of kidney cancer. 1.2. Laser Microdissection to Study Kidney Cancer

Kidney cancer is characterized by an enormous heterogeneity in the genetic changes and in the alterations in cellular networks of molecular pathways. Evaluation of mechanisms involved in kidney cancer and its potential targets for therapy, therefore, requires comprehensive analyses of genomes, transcriptomes, proteomes, and methylation profiles in healthy and tumor tissue, as well as in the tumor microenvironment. For many years now, primarily, indiscriminate homogenates of whole tumor or biopsy samples have been used as a source for DNA, RNA, or protein isolation. However, an exact delineation of molecular changes seems challenging within those samples, since the predominant cell type can dilute the biologically relevant effects that are restricted to a limited number of cells or to a particular cell type. Furthermore, “contamination” of a tumor sample by nonneoplastic cells can tamper or render impossible sensitive molecular techniques, e.g., loss of

23  Application of Laser-Capture Microdissection to Study Renal Carcinogenesis

281

heterozygosity (LOH) analysis or high-throughput gene- or protein-expression profiling. A major technical improvement was the development of laser-assisted microdissection that allows the isolation of morphologically similar cells or even subcellular structures from a heterogeneous frozen- or paraffin-embedded tissue under direct microscopic visualization (6, 7). Because laser microdissection does not destroy the surrounding tissue, different tissue components (e.g., neoplastic and adjacent tumor microenvironment or nonneoplastic cells) can be sequentially sampled from the same slide and applied to subsequent comparative studies. Today, two different laser-based microdissection systems dominate the market: One is the infrared-laser-based laser-capture microdissection system (LCM), which melts a thermoplastic film on a sample of interest that can subsequently be lifted from the specimen. The second one is the laser microdissection and pressure catapulting system (LMPC) that uses a pulsed UV-A laser to cut and directly catapult the isolated sample into a collection tube, without impairing the biological and molecular characteristics of the specimen (Fig. 1). Over the past decade, the combination of laser microdissection with modern methods of molecular biology has provided a

Fig. 1. (a) Renal H&E stained cryosection prepared for laser microdissection, showing a basophilic atypical tubule (circled  ). Top panel: a section with limited optical resolution; Bottom panel: a section with optimal optical resolution; note the improved demarcation of tubule borders. (b) Renal H&E stained paraffin section prepared for pathologic evaluations, showing a basophilic atypical tubule.

282

K. Stemmer and D.R. Dietrich

tremendous improvement for studying the pathophysiological networks involved in kidney cancer progression. Microscopic visualization of renal tumor tissue allows a controlled isolation of morphologically enriched or even pure cell populations from different histological types or stages of renal lesions from the same sample. This is in contrast to other indirect cell sorting techniques (e.g., fluorescence-activated cell sorting (FACS) or immunomagnetic bead-based cell separation) that depend on proteolytically individualized cell suspensions and the availability of suitable antibodies. Despite the availability of hightechnology tools, the more routine immunohistochemical characterization of kidney tumors is of a high prognostic value, since it allows the discrimination between different lesion types and stages in a unique tissue sample. Thus, some investigators have combined the advantages of both techniques by using antibodystained tissue slices for microdissection, so-called immuno-(guided) laser microdissection, and subsequent molecular downstream assays. 1.2.1. Current Limitations of Laser Microdissection in Kidney Cancer Research

Laser microdissection brought great advancements in the understanding of kidney physiology and pathology, e.g., by allowing segmental renal gene expression analyses (8), protein identification in nuclei of proximal tubular cells (9), or gene expression profiling in renal preneoplastic lesions (10). However, the successful application of laser microdissection to study molecular and biochemical events in kidney cancer is mainly determined by three factors: (1) preservation and microscopic visualization of the renal tissue morphoalogy for reliable sample identification, (2) isolation of sufficient amounts of tissue in an appropriate time, and (3) maintenance of the molecular integrity of the microdissected samples. Each of the three factors has technical hurdles that need to be overcome to obtain reproducible and reliable results from the highly sensitive molecular analyses carried out subsequent to microdissection: 1. Achievement of sufficient optical resolution requires mounting of paraffin or frozen tissue sections. However, neither cover slips nor mounting media are compatible with laser microdissection, inevitably resulting in a suboptimal visualization of the dry tissue sections at high magnification. Moreover, the optical resolution of frozen sections is generally rather poor, making detailed distinction of lesions or specific cells to be microdissected a difficult matter, often requiring extensive histopathological experience and expertise (Figs. 2 and 3). Furthermore, many molecular downstream techniques require the use of cryopreserved tissue instead of formalin fixation. Thus, based on the complexity of normal renal tissue and morphological variability of kidney tumors, a well-trained and experienced toxicological

23  Application of Laser-Capture Microdissection to Study Renal Carcinogenesis

283

Fig. 2. Schematic of the process of identification, lasering, and catapulting of lesions from kidney sections for DNA, RNA, and protein isolation (modified from Carl Zeiss MicroImaging GmbH, Munich, Germany, with permission).

Fig. 3. (a) Renal H&E stained cryosection prepared for laser microdissection, showing a basophilic hyperplasia (darker circle) and an area presenting with chronic progressive nephropathy (lighter circle). (b) Renal H&E stained cryosection prepared for laser microdissection, showing an early stages of a basophilic adenoma (circled ). Note the increased basophilia of the trabecular cells of this adenoma and the presence of multiple parts of the adenoma (a second part of the adenoma is shown in the lower left hand corner outlined in green) showing up in the same plane of vision of the section.

or clinical pathologist may be required to visually discriminate between the cell populations of interest. 2. Depending on the downstream application, a sufficient number of cells or amount of tissue has to be collected for subsequent DNA, RNA, and protein isolation (11). In many cases, cells or small tissue samples have to be pooled to achieve adequate amounts of material for sample processing. It is important to realize that pooled samples despite having nearly the same

284

K. Stemmer and D.R. Dietrich

microscopic phenotype may have different molecular profiles. In addition, microscopic identification of small tissue samples or single cells and their combined microdissection is extremely time-consuming. Increasing amounts of time required for microdissecting sufficient amounts of sample may interfere with the molecular integrity and, thus, the necessary minimum quality of a sample used for subsequent analyses (see below). 3. Sophisticated and sensitive applications (e.g., methylation specific PCRs, gene expression profiling, proteomics) often require an optimal sample preparation during all processing steps (e.g., tissue harvesting, fixation, sample storage, sectioning, staining, and finally microdissection at room temperature) to achieve sufficient template integrity (10, 11). Thus, molecular analyses from laser microdissection samples means making compromises in either sample morphology or visualization or molecular integrity. Consequently, every new molecular downstream approach to study kidney cancer from microdissected tissue or cell samples requires an antecedent step-by-step protocol validation. In addition, caution is advised when generalizing protocols to different sample types. Table 1 summarizes some possible ways to resolve the most important technical demands of laser microdissection.

2. Materials 1. 2-methylbutane, Sigma-Aldrich, St Louis, MO. 2. OCT embedding medium, Sakura Finetek, Torrance, CA. 3. Standard cryotome, e.g., Thermo Fisher Scientific, Florence, KY. 4. RNAse-free Mayer’s hematoxylin, Sigma-Aldrich, St Louise, MO. 5. RNAse-free eosin Y, Sigma-Aldrich, St Louise, MO. 6. Diethyl pyrocarbonate (DEPC), Sigma-Aldrich, St Louise, MO. 7. LMPC system, Germany.

PALM

Microlaser

GmbH,

Bernried,

8. RNeasy Micro Kit, Qiagen, Hilden, Germany. 9. ß-mercaptoethanol, Sigma-Aldrich, St Louise, MO. 10. Agilent Bioanalyzer, Santa Clara, CA. 11. Two-Cycle cDNA Synthesis Kit, Affymetrix, Santa Clara, CA. 12. MEGAscript T7 Kit, Ambion, Austin, TX. 13. Gene array, Affymetrix, Santa Clara, CA.

23  Application of Laser-Capture Microdissection to Study Renal Carcinogenesis

285

Table 1 Preservation and microscopic visualization of tissue morphology Technique

Pros

Cons

Use of paraffin sections

Provides good morphology and easier cell or structure identification on serial renal sections

Formalin fixation and paraffin embedding can strongly interfere with RNA and DNA quality by cross-linking of nucleic acids to proteins Restricted number of antibodies that allow immunohistochemistry in fixed tissues without antigen retrieval and thus some disruption of tissue morphology

Allows clear identification of immunohistochemically labeled cells or lesions

Use of cryosections

No interference of fixatives with RNA and DNA

High diversity of antibodies available Excellent immunohistochemical staining and, thus, specific localization of cells or lesions via expression labeling and subsequent microdissection (see below)

Loss or reduced recognition of morphology and, thus, harder cell or structure identification on serial renal sections Despite excellent epitope or antigen recognition in immunohistochemistry poor optical distinction of fine histological details

The main caveat is the effect on yield and integrity of biomolecules, particularly RNA. Thus, staining methods must be fast, for instance by adjusting washing and staining times

Antibody staining of renal sections (immuno-guided microdissection)

Possibility to better identify particular tubular segments or distinct cell types such as tumor cells (including tumor stem cells) and components of the microenvironment

Xylene mounting of dry tissue sections

Although the evaporation time is Provides wetting of the tissue quite short, associated delays in slices and, therefore, better optical refraction for microscopy. sample microdissection may interfere with the sample quality, Xylene as mounting solution particularly RNA integrity evaporates much quicker than Xylene is classified as a harmful water and allows dry sample substance that requires conducting collection sample mounting, microscopy, and laser microdissection under a hood

Application of diffusion filters for microscopy

Some commercial laser microdissection systems include optical light diffusion filters that improves the optical resolution

Depending on the device, diffusion filters may not be included

(continued)

286

K. Stemmer and D.R. Dietrich

Table 1 (continued) Technique

Pros

Cons

Software development

The most recent generation of laser microdissection systems provide software modules for automated sample recognition, isolation and capture on serial sections

Owing to the convoluted tubular structure of the renal parenchyma, fully automated sample recognition, based on object-oriented parameters, for example, density, area, neighborhood, and shape, may not be easily applicable

Isolation of sufficient amounts of tissue Different amplification Application of sample protocols can be applied amplification protocols to generate the required to yield sufficient microgram amounts of RNA quantities of RNA for or DNA for downstream downstream analyses even from single cells applications

Sample amplification, particularly RNA amplification by in vitro transcription can result in saturation of the amplification process, an artifact that could lead to preferential amplification of certain genes and consequently to biased gene expression profiles Thus careful antecedent protocol standardization is required. Amplification techniques are not applicable for protein samples

Maintenance of the molecular sample integrity Although HOPE-fixed sections show HOPE fixation When compared to formalin formalin-like morphology, they fixation, HOPE fixation was must be carefully handled because shown to result in only marginal of their fragility RNA and DNA degradation. Since it is also suitable for subsequent paraffin embedding, it can be applied for improved sample visualization and molecular downstream analyses

3. Methods This protocol was designed to isolate different types of preneoplastic lesions from H&E stained cryosections for high-quality RNA isolation and has been successfully applied for quantitative gene expression profiling using microarrays (10). The protocol establishment is based on several optimization steps, which are discussed in detail in Stemmer et al. (11). The finding that the accuracy of RNA-based techniques, e.g., RNA amplification or quantitative gene expression analyses, critically depend on a defined quantity of starting RNA emphasizes the necessity to isolate sufficient tissue to obtain quantifiable amounts of RNA (see Note 1). Table 2 summarizes the most commonly applied RNA

23  Application of Laser-Capture Microdissection to Study Renal Carcinogenesis

287

Table 2 Required amount of microdissected tissue (lesions, bATs) to provide for sufficient quantifiable amounts of RNA when using different RNA quantification techniques Approximate amount of microdissected atypical tubules of similar size

Technique

Device

Detection range

Minimum required micro dissected area (mm2)

Microcapillary electrophoresis (MCE)

MCE Agilent RNA 6000 Nano and Pico kits (Agilent, Santa Clara, CA, USA)

Nano chip: 5–500 ng/ ml

Nano chip: 1.7 mm2

Nano chip: 31 bAT’s

Pico chip: 0.05–5 ng/ml

Pico chip: 0.016 mm2

Pico chip: <1 bAT

Fluorescence-based quantification

Fluorescence-based Quant-iT RiboGreen kit (Invitrogen, Carlsbad, CA, USA)

1 pg/ml–1 ng/ml

3.3 × 10−4 mm2

<1 bAT

Ultraviolet (UV) absorbance

NanoDrop, Fisher Thermo, Wilmington, DE, USA

2 ng/ml–3 mg/ml

0.7 mm2

13 bAT’s

quantification techniques, their detection range, the required microdissected tissue area, and approximate number of lesions (in this case basophilic atypical tubules, bAT) to provide for detectable amounts of RNA, when using 10-mm thick cryosections and a Qiagen RNeasy Micro Kit for RNA isolation. The calculations in Table 2 are based on previous findings whereby approximately 3 ng of RNA can be isolated from 18 pooled bATs with an overall area of 1 mm2. The latter, thus, corresponds to approximately 0.17 ng of RNA per bAT. 3.1. Tissue Collection and Preparation of Frozen Sections

1. Kidneys collected from sacrificed animals should be sectioned into a few replicate slices for multiple analyses. 2. Replicate samples should be quickly frozen in a beaker filled with the cryogen 2-methylbutane precooled on dry ice. 3. Kidney slices can be wrapped in aluminum foil and samples can be stored at −80°C until use.

3.2. Preparation of H&E Stained Cryosections

1. For preparation of cryosections, cryostat should be cooled to −20°C and the respective sample “thawed” from −80°C to cryostat temperature (see Note 2).

288

K. Stemmer and D.R. Dietrich

2. Since the use of tissue freezing media can interfere with laser microdissection, samples should be fixed only on a small drop of OCT embedding media in the cryostat. 3. Cryosections should be mounted onto special RNAse-free membrane-covered glass slides, such as those recommended for the respective microdissection system, e.g., LMPC system (see Note 3). Freshly cut sections should be air-dried for 20 s in the cryostat and subsequently fixed in −20°C cold ethanol (70% in RNAse-free ddH2O) for 3 min. Fixed sections should be subsequently air-dried for 10 min at RT. 4. The sections are stained with RNAse-free, ice-cold Mayer’s hematoxylin for 3 min and subsequently rinsed with RNAsefree tap (DEPC-treated) water for 3 min. 5. Counterstaining can be conducted by dipping the samples into RNAse-free, ice-cold eosin Y for 3 min. 6. Excess staining solution should be carefully removed on an absorbent surface, and the sections should be dehydrated by shortly dipping into ice-cold 70 and 100% ethanol. 7. The sections should be air-dried at room temperature and directly used for laser-assisted microdissection. 3.3. Laser Microdissection and Pressure Catapulting

The procedure below is specific to the PALM LMPC system (Carl Zeiss Microimaging GmbH, Munich, Germany). Similar procedures may applicable to laser microdissection systems from other companies. For better sample visualization, an optical light diffusion filter should be applied for microscopy. 1. For RNA isolation from microdissected kidney samples, a required minimum area of should be microdissected (see Table 2), and samples should be collected into an appropriate sampling cup of the microdissection system. 2. After completion of the sample collection, pooled samples should be immediately lysed via addition of 300  ml lysis buffer containing 1% b-mercaptoethanol (e.g., component of the RNeasy Micro Kit). 3. A maximum total of 180 min between initial tissue section staining and lysis of pooled microdissected samples should not be exceeded.

3.4. RNA Isolation from Microdissected Samples for Affymetrix Chip Hybridization

1. RNA should be isolated using, for example, RNeasy Micro Kit according to the manufacturer’s instruction. 2. RNA quality and estimate total quantity should be estimated, for example, by using an Agilent Bioanalyzer (see Note 4). RNA amplification and conversion to biotin-labeled cRNA should be carried out using, for example, the Affymetrix

23  Application of Laser-Capture Microdissection to Study Renal Carcinogenesis

289

Two-cycle Target labeling kit according to the manufacturer’s instructions. For all RNA amplification procedures, the same RNA starting quantity (5 or 10 ng RNA) must be used. 3. Typically, 15 mg biotin-labeled cRNA samples should be applied for Affymetrix chip hybridization. Hybridization should be carried out according to manufacturer’s instructions.

4. Notes 1. The most rate-limiting factor is the sensitivity of different RNA quantification techniques that largely vary in their respective detection range. 2. To avoid cryoartifacts, samples should not be refrozen at −80°C more than three times. 3. The use of 10-mm kidney slices usually provides a good compromise, i.e., sufficient optical resolution, but adequate yields of microdissected tissue. 4. For later amplification and hybridization procedures, only RNA samples with an RNA integrity number (RIN) >7.0 should be used.

References 1. Eble, J., Sauter, G., Epstein, J., and Sesterhenn, I. (2004) Pathology and Genetics of Tumors of the Urinary System and Male Genital Organs, IARC Press, WHO Classification of Tumours, Volume 7. 2. Lock, E. A., and Hard, G. C. (2004) Chemically induced renal tubule tumors in the laboratory rat and mouse: review of the NCI/NTP database and categorization of renal carcinogens based on mechanistic information, Crit Rev Toxicol 34, 211–299. 3. Dietrich, D. R., and Swenberg, J. A. (1991) Preneoplastic lesions in rodent kidney induced spontaneously or by non-genotoxic agents: predictive nature and comparison to lesions induced by genotoxic carcinogens, Mutat Res 248, 239–260. 4. Zhu, G., Reynolds, L., Crnogorac-Jurcevic, T., Gillett, C. E., Dublin, E. A., Marshall, J. F., Barnes, D., D’Arrigo, C., Van Trappen, P. O., Lemoine, N. R., and Hart, I. R. (2003) Combination of microdissection and microarray analysis to identify gene expression changes between differentially located

tumour cells in breast cancer, Oncogene 22, 3742–3748. 5. Castro, N. P., Osorio, C. A., Torres, C., Bastos, E. P., Mourao-Neto, M., Soares, F. A., Brentani, H. P., and Carraro, D. M. (2008) Evidence that molecular changes in cells occur before morphological alterations during the progression of breast ductal carcinoma, Breast Cancer Res 10, R87. 6. Ma, X. J., Dahiya, S., Richardson, E., Erlander, M., and Sgroi, D. C. (2009) Gene expression profiling of the tumor microenvironment during breast cancer progression, Breast Cancer Res 11, R7. 7. Stemmer, K., Ellinger-Ziegelbauer, H., Ahr, H. J., and Dietrich, D. R. (2007) Carcinogenspecific gene expression profiles in short-term treated Eker and wild-type rats indicative of pathways involved in renal tumorigenesis, Cancer Res 67, 4052–4068. 8. Kohda, Y., Murakami, H., Moe, O. W., and Star, R. A. (2000) Analysis of segmental renal gene expression by laser capture microdissection, Kidney Int 57, 321–331.

290

K. Stemmer and D.R. Dietrich

9. Dietrich, D. R., Heussner, A. H., O’Brien, E., Gramatte, T., Runkel, M., Rumpf, S., and Day, B. W. (2008) Propiverine-induced accumulation of nuclear and cytosolic protein in F344 rat kidneys: isolation and identification of the accumulating protein, Toxicol Appl Pharmacol 233, 411–419. 10. Stemmer, K., Ellinger-Ziegelbauer, H., Ahr, H. J., and Dietrich, D. R. (2009) Molecular

characterization of preneoplastic lesions provides insight on the development of renal tumors, Am J Pathol 175, 1686–1698. 11. Stemmer, K., Ellinger-Ziegelbauer, H., Lotz, K., Ahr, H. J., and Dietrich, D. R. (2006) Establishment of a protocol for the gene expression analysis of laser microdissected rat kidney samples with affymetrix genechips, Toxicol Appl Pharmacol 217, 134–142.

Chapter 24 Laser-Capture Microdissection and Transcriptional Profiling in Archival FFPE Tissue in Prostate Cancer Ajay Joseph and Vincent J. Gnanapragasam Abstract Prognostic markers can improve prediction of the behaviour of a cancer at the point of diagnosis. A key value of any prognostic marker is at the point of tumour diagnosis. In the context of prostate cancer, this implies profiling in the diagnostic formalin-fixed, paraffin-embedded (FFPE) transrectal ultrasoundguided (TRUS) needle biopsy. TRUS needle biopsies commonly contain both stromal and epithelial cells, and malignant glands are found as isolated foci within this tissue. Using the entire biopsy for genetic analysis inevitably results in a significant contamination of malignant cells with benign tissue. This combination of minimal tumour yields and tissue heterogeneity have so far prohibited prognostic transcript and microarray molecular studies in needle biopsies. Laser-capture microdissection (LCM) allows enriched cell populations to be accurately isolated from heterogeneous tissue, hence facilitating analysis of different components from a single tissue sample. Here, we describe its use in isolating tumour cells in archival FFPE prostate needle biopsies and subsequent application for RNA extraction and quantitative real-time PCR (QPCR). Key words: LCM, FFPE, Prostate cancer, Diagnostic needle biopsies, QPCR, TRUS, Prognostic marker, Heterogeneous tissue

1. Introduction Prostate cancer accounts for nearly a quarter (24%) of all cancer cases diagnosed in men and 13% of male cancer-related deaths in UK. A fundamental challenge in prostate cancer is the ability to predict tumour response to therapy. Molecular tissue biomarkers have shown great promise in improving prediction of clinical outcome in this regard (1, 2). The majority, however, have been investigated in surgically derived tissue when a treatment (radical prostatectomy) has already been administered. Prostate cancer, however, can be treated by different modalities including active

Graeme I. Murray (ed.), Laser Capture Microdissection: Methods and Protocols, Methods in Molecular Biology, vol. 755, DOI 10.1007/978-1-61779-163-5_24, © Springer Science+Business Media, LLC 2011

291

292

A. Joseph and V.J. Gnanapragasam

monitoring, surgery, radiotherapy, and androgen deprivation. The key value of any prognostic marker, however, is at the point of tumour diagnosis and in helping to select the optimal therapy for a patient. Transrectal ultrasound-guided (TRUS) needle biopsy of the prostate is the clinical standard for obtaining a diagnosis in prostate cancer and is the only tissue routinely available before a treatment decision. In non-surgical therapy, it is also the only tissue collected. It is, therefore, crucial that prognostic biomarkers are studied in this tissue (3). The use of needle biopsies for biomarker research, however, has been problematic because the yield of malignant glands is usually small and within a mixture of normal prostate stroma and glands. These setbacks accelerated the need for a sophisticated tool for the isolation of the specimen of interest from heterogeneous tissue. Laser-capture microdissection (LCM) allows enriched cell populations to be accurately isolated from heterogeneous tissue, hence facilitating analysis of different components from a single tissue sample (4). LCM can also be performed on formalin-fixed, paraffinembedded (FFPE) tissue, which is the clinical standard for preserving prostate biopsy tissue for histopathology. FFPE preserves tissue architecture and allows the storage of diagnostic and surplus tissue in archival banks. This resource, therefore, represents a vast repository of tissue material with a long-term clinical follow-up, which can be used for both hypothesis-driven and discovery-­based research into biomarkers. We have recently developed the use of LCM to isolate areas of pre-marked malignant or benign glands from archival FFPE prostate biopsies which were surplus to diagnostic requirements (5). This technique along with RNA extraction and amplification has allowed us to use smallvolume FFPE prostate needle biopsies for transcript expression studies. Here, we describe this technique and the optimised protocols and reagents used in our current methodology.

2. Materials 1. Formalin-fixed, paraffin-embedded (FFPE) archival diagnostic needle biopsies were obtained from a local tissue bank with ethical approval from the hospital authorities (see Note 1). 2. Polyethylene naphthalate (PEN), polyethylene terephthalate (PET), polyester (POL) membrane slides (Leica Microsystems, Milton Keynes, UK) (see Note 2). 3. Haematoxylin (Sigma Diagnostics, UK). 4. Eosin (Sigma Diagnostics, UK). 5. 0.1% Diethylpyrocarbonate (DEPC)-treated water, RNase Zap® (Sigma Life Sciences).

24  Laser-Capture Microdissection and Transcriptional Profiling…

293

6. Xylene (Sigma Diagnostics, UK). 7. 50–90% ethanol prepared in DEPC-treated water (see Note 3). 8. Leica LMD6000 integrated with DM6000 B microscope (Leica Microsystems). 9. 0.5-mL Microcentrifuge tube (Fisher Scientific, Loughborough, UK) for collecting the samples after microdissection, lysis buffer (Roche Applied Science, Burgess Hill, UK) (see Note 4). 10. High Pure RNA paraffin kit (Roche Applied Science). RNA quantity assessed using a Nanodrop spectrophotometer (Nanodrop Technologies, Detroit, MI, USA) (see Note 5). 11. Light Cycler Pre-amplification kit (Roche Applied Science). 12. High Pure PCR Clean-Up Kit (Roche Applied Science). 13. Primers were designed with the aid of primer design software “Primer 3” (Applied Biosystems) with amplicons in the range of 100–150 bp.

3. Methods LCM technology encompasses (1) a light microscope for the visualisation of tissue specimen, (2) a UV laser beam that vaporises a thermo-labile polymer thereby sectioning the adherent cells of interest, and (3) a collection device that receive the cells of interest from tissue sections (6). Tissue sections to be microdissected are attached to the surface of a UV-absorbing membrane. Upon laser firing, the membrane and the tissue in the sectioning line are photovolatilised. Since the membrane is not adherent to the slide, the dissectate (with the attached specimen) drops into the cap of the microcentrifuge tube. The specimen is then subjected to molecular extraction techniques (DNA/RNA/microRNA) for subsequent expression profiling studies. 3.1. Specimen Preparation

1. LMD6000 applications require a special PEN membrane, bound to a glass slide, for laser microdissection. Tissue sections must be mounted onto the membrane prior to laser microdissection. Mounting of tissue sections onto slides is relatively easy, since the glass provides solid support (see Note 6). 2. The formalin blocks of selected patients’ samples are sectioned into 5-mm strips using a microtome, and these strips are placed on the PEN membrane slide. 3. From each patient sample, three sections are obtained and placed on a single membrane slide (see Note 7).

294

A. Joseph and V.J. Gnanapragasam

3.2. Haematoxylin and Eosin Staining

1. The paraffin must be removed from paraffin-embedded sections prior to staining. The embedding process must be reversed to get rid of the paraffin wax from the tissue and allow watersoluble dyes to penetrate the sections. 2. The slides are “deparaffinised” and rehydrated by washing them with xylene followed by alcohol and then water. The process of deparaffinisation is carried out, as stains do not penetrate into tissues containing paraffin (see Note 3). 3. The histopathological architecture of the prostate stroma and glands need to be visualised via haematoxylin and eosin (H&E) staining for accurate LCM. 4. Xylene (C6H4(CH3)2), an aromatic hydrocarbon isomer, acts as a solvent of paraffin wax. For the efficient removal of paraffin wax, paraffin-embedded slides are placed in Coplin jars and treated with xylene twice using a shaker, for 5 min each time. This process is followed by hydrating through an alcohol gradient, 100% (twice), 95% (once), 70% (once), and 50% (once) ethanol sequentially for 5 min each. The specimen is then soaked in DEPC-treated water for 5 min. 5. The length of time necessary to counter-stain the tissues will depend upon the fixation and the type of haematoxylin employed. For attaining optimal staining by H&E for 5-mm sections, sample slides must be treated for 26–28 s and 3–5 s in haematoxylin and eosin, respectively. The slides are washed using DEPC-treated water and air-dried as quickly as possible by incubating them for 30 min at 37°C. The slides are then ready for microdissection (see Note 8).

3.3. Laser Capture Microdissection

Deparaffinised and stained slides are laser-microdissected using the Leica LMD6000 system. After staining, histopathological differences between the cells may deteriorate over time due to drying, which can pose problems in distinguishing between cell types. This may in turn lead to contamination by different population of cells during microdissection. To ensure accurate microdissection, reference slides are stained and mounted with a coverslip. These are pre-marked by a pathologist and used for parallel identification and selection of the histological areas of interest. The pattern is then used to guide microdissection on the specimen slide. Leica LMD6000 system description: 1. LMDsystem with the laser is pre-assembled on an upright research microscope Leica DM6000B with a motorized transmitted light and Hitachi analog 3CCD camera. The UV laser is set at 355 nm. 2. The instrument comprises a multislide holder that allows work with up to three slides simultaneously. It collects microdissected specimens into four 0.5-mL microcentrifuge tubes held by a collection device on a movable tray.

24  Laser-Capture Microdissection and Transcriptional Profiling…

295

3. The LMD module is coupled with Smart move control, which acts as an interface between the microscope and the screen, enabling real-time monitoring of the tissue sections. 4. The software controls the microscope as well as the laser. In addition, the software automatically controls focus, the collection device, and the positioning of the microcentifuge tubes. Other features include multicutting over the entire slide, three slide holder control and serial section cutting over all slides at all magnifications. 5. The software also has a database that allows archiving the images of the tissue sections before and after microdissection. 6. The stained slides are placed in inverted fashion into the threefold specimen tray. Smart move control allows fine and core focusing of the specimen by tuning in either direction. The LMD Smartcut series includes objective of magnification ranging from 6.3× up to 150× (see Note 9). 7. Using the Smart move control, the laser is focused on a segment of the slide free of areas of interest and autocalibrated taking into account various factors such as thickness of the section, moisture content of the specimen, etc. 8. Once the slide has been calibrated, the microscope is focused on the specimen. With the guidance of the reference slide, areas of interest are manually marked with the help of selection tool in the software module. With this tool, it is possible to precisely define the shape, size, and area of interest to be sectioned. 9. The laser follows the marked sectioning line and vaporises the membrane along with the specimen adhered to it. The free membrane and attached specimen falls by gravity into the cap of one of the previously selected microcentifuge tubes. The cap is filled with 20 mL of lysis buffer and stored immediately on ice, which prevents further degradation of nucleic acids. In this manner, different cell populations can be collected in each of the four microcentifuge tubes (Fig. 1).

Fig. 1. (a) Mounted haematoxylin and eosin (H&E) stained reference section of FFPE prostate needle biopsy. (b) Laser outline of the cancer epithelial cells to be microdissected. (c) The remaining tissue after laser-capture microdissection.

296

A. Joseph and V.J. Gnanapragasam

3.4. RNA Isolation

Laser microdissected tissue specimens present a particular challenge for molecular analysis, as nucleic acids must be purified from a very small amount of tissue. This is a particular issue for RNA which cannot be directly amplified by PCR, like other nucleic acids. In addition, working with FFPE tissues can also pose problems for RNA purification due to the degrading effect of formalin on RNA (see Note 10). Adding the lysis solution to the microcentrifuge tube helps in the immediate stabilisation of nucleic acids (7). Other factors can also affect the RNA yield including the quality of the sample, duration of preservation before microdissection, type of preservation, fixation method, and efficiency of microdissection. The collected specimens obtained by laser capture microdissection are subjected to RNA extraction using the High Pure RNA paraffin kit (Roche Applied Science). Manufacturer’s protocol is standardised and followed (see Note 11). 1. The collected samples are disrupted and homogenised during an overnight incubation with proteinase K, rather than by using stable tissue homogeniser to avoid the loss of cells. 2. Nucleic acid extraction is carried out by allowing them to bind in the presence of chaotropic salt specific to the surface of glass fibres prepacked in the High Pure filter tube. The binding process is specific for nucleic acids in general, but the binding conditions are optimized for RNA. Bound RNA is purified in a series of rapid wash and spin steps to remove contaminating cellular components. Residual DNA is digested by incubation with DNAse I (see Note 12). 3. A second incubation step with proteinase K improves the purity of RNA. Finally, low-salt elution buffer is used to elute the nucleic acids from the glass fibre. The process does not require RNA precipitation, organic solvent extraction, or extensive handling of RNA.

3.5. cDNA Synthesis and Amplification Using Ribo-SPIA® Technology

Reverse transcription in conjunction with quantitative polymerase chain reaction (QPCR) is a powerful approach for differential gene expression analysis (8). However, a major challenge in our study is that RNA obtained from FFPE sample derived from microdissected tissue is significantly degraded (9, 10). A single human cell from a tissue section contains approximately 10 pg of total RNA regardless of the type of extraction method employed (7). However, transcription profiling assays require a minimum of 5 mg of total RNA, which can only be achieved by amplification of nanogram quantities of RNA. Recent technological advancements have introduced a number of RNA amplification methods, especially PCR-based exponential amplification methods (11). In vitro transcription (IVT) is one of the most widely used RNA amplification method in which cDNA is initially synthesised,

24  Laser-Capture Microdissection and Transcriptional Profiling…

297

which in turn acts as a template for amplification using T7 RNA polymerase (12). However, this method requires an intact poly A tail. This technique, therefore, preferentially amplifies more intact RNA (intact 3¢ ends), which can lead to expression bias when FFPE-derived degraded RNA is studied. Unlike IVT, whole transcriptome amplification (WTA) does not depend on the presence of a poly A tail. Instead, universal oligonucleotide primers are used, which are more suited for the amplification of degraded RNA, thus eliminating 3¢ end bias (13). WTA systems enable the amplification of nanogram quantities of RNA, in 3–4 h using a combination of cDNA synthesis and QPCR. They, therefore, have the potential to enhance amplification efficiency while maintaining representative transcript levels for studying differential gene expression (14, 15). 1. WTA amplification of isolated RNA from FFPE tissues is performed using a kit such as Ribo-SPIA pre-amplification kit (Roche Applied Sciences). 2. The starting amounts of 5–50 ng of RNA extracted are converted into cDNA libraries and amplified by manufacturerrecommended protocols yielding approximately 5  mg of amplified product. 3. First strand cDNA synthesis is carried out using a unique DNA/RNA chimeric primer mix and reverse transcriptase (RT). DNA segment of the primer hybridises to either the 3¢ poly A tail of any polyadenylated transcript or randomly across the whole transcript. The RNA segment of the primer contains a unique sequence that can be used to incorporate a priming site for linear amplification step ahead. RT extends the 3¢ DNA end of each primer generating first strand cDNA. 4. Second strand cDNA synthesis takes place with DNA polymerase replicating the entire cDNA template, forming double stranded cDNA. The 3¢ end of the newly synthesised strand will contain a RNA–DNA heteroduplex, complementary to the unique RNA sequence in the first-strand chimeric DNA/ RNA primer. 5. The SPIA amplification step is accomplished by a combination of SPIA DNA/RNA chimeric primer, DNA polymerase and RNase H in an isothermal reaction. RNase H digests the RNA sequence from the priming sites of double-stranded cDNA exposing the single stranded DNA that is available for binding to the SPIA chimeric primer. 6. DNA polymerase initiates replication at the 3¢ end of the primer, displacing the existing forward strand. 7. Once the extension reaction begins, RNase H cleaves the RNA at the 5¢ end of the newly synthesised strand, exposing the priming site for initiation of the next round of DNA synthesis.

298

A. Joseph and V.J. Gnanapragasam

Fig. 2. QPCR data showing GAPDH Expression (Cp) on ten different samples ranging from 1998 to 2005.

8. The entire process of SPIA RNA/DNA primer binding, extension of new DNA molecules, strand displacement and RNA cleavage is repeated, resulting in rapid accumulation of cDNA complementary to the original mRNA molecule. One thousand five hundredfold amplification can be attained with a starting amount of 5 ng total RNA. 9. Amplified cDNA can be stored at −20°C for 6  months. To enable reliable determination of cDNA yield, the pre-amplified cDNA can be purified using the High Pure PCR Clean-Up kit, which can help avoid PCR inhibition in subsequent expression analysis. 10. A set of FFPE samples collected from 1999 to 2005 have been processed using this protocol and used for QPCR. Similar Cp (Crossing point) value for the housekeeping gene GAPDH were observed for all the study samples regardless of the year of collection (Fig. 2).

4. Notes 1. Formalin fixed, paraffin-embedded (FFPE) biopsies, ranging from 5 to 8 years were used for molecular profiling studies. This permits the validation of clinical biomarkers retrospectively, with a long-term clinical follow-up. 2. Polyethylene naphthalate (PEN) membrane slides are employed for the current study due to their convenience and cost-effectiveness. The membrane is mounted on the upper side of the slide.

24  Laser-Capture Microdissection and Transcriptional Profiling…

299

3. Shakers must be used to facilitate deparffinisation since they allow good penetration of the solvent (xylene). 4. Use a thin wall 0.5-mL microcentifuge tube for the accurate collection of microdissected samples. 5. The quality of the FFPE samples assessed by Nanodrop will be poor compared to fresh samples due to degradation. 6. The membrane is adhered to the central region of the slide. In order to prevent the damage of the membrane, slides must be handled with care; otherwise, this will obstruct laser action. 7. Two slides of six sections are utilized to obtain enough RNA for downstream application. 8. Stained tissues can be used directly for laser microdissection or alternatively the dried sections can be stored in containers with a desiccant at −80°C for several months. Immediately prior to laser microdissection, the container with the slides is removed from the freezer and slowly adjusted to room temperature to avoid condensation of water inside the container. This is critical for RNA quality, as water can activate RNases. 9. The laser beam is triggered from the upper part of the microscope; therefore, the slides are placed upside down to direct the specimen towards the cap of the collection tube placed underneath the slide. 10. Staining steps can also compromise the integrity of nucleic acids. To reduce the loss of RNA, many manufacturers assemble specially designed kits for laser microdissected samples. These kits contain DNase inhibitors in the lysis buffer, which allows for the purification of total RNA. 11. When working with RNA, precautions must be taken to avoid RNA cleavage by RNAse. Before beginning the procedure, all the equipment including pipettes, glassware and the work area must be cleaned once with RNase Zap and, twice with 70% ethanol diluted in RNase-free water. 12. This system is designed to amplify RNA, but large amounts of contaminating genomic DNA may amplify during the process, so DNase treatment is essential during RNA purification. References 1. Heidenreich A, Aus G, Bolla M, Joniau S, Matveev VB, Schmid HP, and Zattoni F. (2008) EAU guidelines on prostate cancer, Eur Urol 53, 68–80. 2. Verhagen PC, Tilanus MG, de Weger RA, van Moorselaar RJ, van den Tweel JG, and Boon TA. (2002) Prognostic factors in localised prostate cancer with emphasis on the application

of molecular techniques, Eur Urol 41, 363–371. 3. Schlomm T, Erbersdobler A, Mirlacher M, and Sauter G. (2007) Molecular staging of prostate cancer in the year 2007, World J Urol 25, 19–30. 4. Player A, Barrett JC, and Kawasaki ES. (2004) Laser capture microdissection, microarrays

300

A. Joseph and V.J. Gnanapragasam

and the precise definition of a cancer cell, Expert Rev Mol Diagn 4, 831–840. 5. Rogerson L, Darby S, Jabbar T, Mathers ME, Leung HY, Robson CN, Sahadevan K, O’Toole K, and Gnanapragasam VJ. (2008) Application of transcript profiling in formalinfixed paraffin-embedded diagnostic prostate cancer needle biopsies, BJU Int 102, 364–370. 6. Rodriguez AS, Espina BH, Espina V, and Liotta LA. (2008) Automated laser capture microdissection for tissue proteomics, Methods Mol Biol 441, 71–90. 7. Ladanyi A, Sipos F, Szoke D, Galamb O, Molnar B, and Tulassay Z. (2006) Laser microdissection in translational and clinical research, Cytometry A 69, 947–960. 8. Livak KJ, and Schmittgen TD. (2001) Analysis of relative gene expression data using real-time quantitative PCR and the 2(−Delta Delta C(T)) Method, Methods 25, 402–408. 9. Polacek DC, Passerini AG, Shi C, Francesco NM, Manduchi E, Grant GR, Powell S, Bischof H, Winkler H, Stoeckert CJ, Jr., and Davies PF. (2003) Fidelity and enhanced sensitivity of differential transcription profiles following linear amplification of nanogram amounts of endothelial mRNA, Physiol Genomics 13, 147–156. 10. Vermeulen J, Derveaux S, Lefever S, De Smet E, De Preter K, Yigit N, De Paepe A, Pattyn F, Speleman F, and Vandesompele J. (2009) RNA pre-amplification enables large-scale

RT-qPCR gene-expression studies on limiting sample amounts, BMC Res Notes 2, 235. 11. Seth D, Gorrell MD, McGuinness PH, Leo MA, Lieber CS, McCaughan GW, and Haber PS. (2003) SMART amplification maintains representation of relative gene expression: quantitative validation by real time PCR and application to studies of alcoholic liver disease in primates, J Biochem Biophys Methods 55, 53–66. 12. Watson JD, Wang S, Von Stetina SE, Spencer WC, Levy S, Dexheimer PJ, Kurn N, Heath JD, and Miller DM, 3rd. (2008) Complementary RNA amplification methods enhance microarray identification of transcripts expressed in the C. elegans nervous system, BMC Genomics 9, 84. 13. Tomlins SA, Mehra R, Rhodes DR, Shah RB, Rubin MA, Bruening E, Makarov V, and Chinnaiyan AM. (2006) Whole transcriptome amplification for gene expression profiling and development of molecular archives, Neoplasia 8, 153–162. 14. Dafforn A, Chen P, Deng G, Herrler M, Iglehart D, Koritala S, Lato S, Pillarisetty S, Purohit R, Wang M, Wang S, and Kurn N. (2004) Linear mRNA amplification from as little as 5 ng total RNA for global gene expression analysis, Biotechniques 37, 854–857. 15. Kurn N, Chen P, Heath JD, Kopf-Sill A, Stephens KM, and Wang S. (2005) Novel isothermal, linear nucleic acid amplification systems for highly multiplexed applications, Clin Chem 51, 1973–1981.

Chapter 25 Quantitative Analysis of the Enzymes Associated with 5-Fluorouracil Metabolism in Prostate Cancer Biopsies Tomoaki Tanaka Abstract Orotate phosphoribosyl transferase (OPRT) is the initial enzyme of 5-FU activation, in which 5-FU is converted to 5-fluorouridinemonophosphate. Dihydropyrimidine dehydrogenase (DPD) is a degrading enzyme that catabolizes 5-FU. In this study, we investigated the expression of these enzymes in normal prostate gland (NP), hormone-sensitive prostate cancer (HSPC), and hormone-refractory prostate cancer (HRPC). The prostatic tissue specimens were obtained from patients who had undergone prostate needle biopsies without any treatments or with PSA failure after initial androgen deprivation. The tissue samples derived from formalin-fixed, paraffin-embedded (FFPE) sections were prepared by laser-capture microdissection, and from them RNA was extracted. The levels of OPRT and DPD mRNA expression were examined by quantitative reverse transcriptase-polymerase chain reaction (RT-PCR). The level of OPRT mRNA expression in the HSPC or the HRPC specimens was significantly higher than that in the NP specimens. There was a significant correlation between OPRT mRNA expression levels and the tumor pathological grade. Furthermore, the OPRT/ DPD expression ratio, a powerful predictive factor to evaluate 5-FU sensitivity, in the HRPC group was significantly higher than that in the low-grade HSPC group. Thus, the quantitative evaluation for these enzymes based on phosphorylation of 5-FU may be an effective option for some prostate cancer patients, particularly HRPC group. Key words: Prostate cancer, 5-FU, Orotate phosphoribosyl transferase, Dihydropyrimidine dehydrogenase, Laser-cutting microdissection, RT-PCR

1. Introduction 5-Fluorouracil (5-FU) is a chemotherapeutic agent used against a variety of cancers, including prostate cancer (1–4). A previous report has demonstrated that the response rate of single agent 5-FU against hormone-refractory prostate cancer (HRPC) is about 30% (2). 5-FU is converted to the active metabolic forms, fluorodeoxyuridine monophosphate (FdUMP), which exerts its

Graeme I. Murray (ed.), Laser Capture Microdissection: Methods and Protocols, Methods in Molecular Biology, vol. 755, DOI 10.1007/978-1-61779-163-5_25, © Springer Science+Business Media, LLC 2011

301

302

T. Tanaka

anticancer effect through the inhibition of thymidylate synthase (TS), a key enzyme of DNA synthesis. Dihydropyrimidine dehydrogenase (DPD) is the first and rate-limiting enzyme of 5-FU catabolism and degradates 5-FU (5). Some studies have demonstrated that DPD expression was elevated in those cancer cells which had a low sensitivity to 5-FU (6, 7). In the anabolic pathway of 5-FU, orotate phosphoribosyl transferase (OPRT) is the most important enzyme for phosphorylation of 5-FU. It metabolizes 5-FU to 5-fluorouridinemonophosphate (FUMP), which predominantly inhibits RNA synthesis (1). Recent studies have suggested that the increased expression and activity of OPRT induces a high sensitivity to 5-FU in several tumor types (8–11). Although there have been some studies showing the therapeutic efficacy of 5-FU in some HRPC patients (3, 12, 13), the sensitivity to 5-FU in individual prostate cancer patients remains difficult to predict. The quantitative analysis of the expression of the above enzymes may contribute to the prediction of the efficacy of 5-FU administration and the overall outcome in prostate cancer patients (14).

2. Materials 2.1. Formalin-Fixed, Paraffin-Embedded Tissue Sections and Staining

1. Formalin: 4% (w/v) buffered paraformaldehyde phosphate. Store at 4°C. 2. Mayer’s hematoxylin and eosin (H&E) solution (Wako Pure Chemical Industries, Ltd., Japan). Store at 4°C. 3. Nuclear fast red (NFR, American MasterTech Scientific Inc., Lodi, CA, USA). Store at 4°C. 4. Xylene, store at room temperature. 5. Setup solution; 100 (w/v), 90, 80, and 70% ethanol, store at room temperature.

2.2. Laser-Capture Microdissection Apparatus

1. PALM MicroBeam (P.A.L.M. Microlaser Technologies AG, Munich, Germany).

2.3. RNA Extraction and Quantitative RT-PCR

1. RNA extraction buffer and procedure (Response Genetics, LA, CA, USA). 2. QuickPrep Micro mRNA Purification Kit (Amersham Pharmacia Biotech Inc., Piscataway, NJ). Store at 4°C. 3. Moloney murine leukemia virus (MMLV) reverse transcriptase (200 U/mL; Life Technologies, Gaithersburg, MD). Store at −20°C.

25  Quantitative Analysis of the Enzymes Associated with 5-Fluorouracil Metabolism…

303

4. 5× Moloney murine leukemia virus (MMLV) buffer (Life Technologies, Gaithersburg, MD): 250  mM/L Tris–HCl [pH 8.3], 375 mM/L KCl, and 15 mM/L MgCl2. Store at −20°C. 5. Dithiothreitol (100  mM/L; Life Technologies). Store at −20°C. 6. dNTP (each 10  mM/L; Amersham Pharmacia Biotech). Store at −20°C. 7. Random hexamers (Amersham Pharmacia Biotech): 50 OD dissolved in 550 mL of 10 mM/L Tris–HCl [pH 7.5], and 1 mM/L EDTA. Store at −20°C. 8. Bovine serum albumin (Amersham Pharmacia Biotech): 3 mg/mL in 10 mM/L Tris–HCl, pH 7.5. Store at 4°C.

3. Methods 3.1. Preparation and Characterization of Tissue Samples

1. The cancer specimens, which are collected by needle biopsy, have been subsequently fixed for 24 h in 4% (w/v) paraformaldehyde phosphate buffer and paraffin-embedded (see Note 1). 2. Formalin-fixed, paraffin-embedded (FFPE) tumor specimens and adjacent normal tissues are cut into sequential sections with a thickness of 10  mm. These sections are gradually hydrated in each glass rack filled with xylene, 100, 90, 80, and 70% (w/v) ethanol, respectively. 3. Representative tissue sections are selected by a pathologist after examination of the hematoxylin- and eosin-stained slides.

3.2. Laser Cutting Microdissection

1. The other sections are stained with nuclear fast red for a few minutes to enable visualization of histology before the microdissection procedure. 2. The laser microdissection system of PALM MicroBeam is applied to all the slides prepared on the previous step. UV-laser (337  nm wavelength), which is synchronized with inverted microscope, cuts along the delineated lines to separate cell population of interest from surrounding unwanted regions and is then used to catapult the dissected tissue into a collecting tube filled with appropriate lysis buffer.

3.3. RNA Extraction and cDNA Synthesis

1. RNA is isolated from FFPE specimens using a novel, proprietary procedure. 2. A guanidinium thiocyanate method of mRNA isolation is used (15).

304

T. Tanaka

3. Isolated mRNA is dissolved in 50 mL of 5 nM/L Tris–HCl (pH 7.5). For cDNA synthesis, 20 mL 5× Moloney murine leukemia virus (MMLV) buffer, 10 mL dithiothreitol, 10 mL dNTP (each 10 mM/L), 0.5 mL random hexamers, 2.5 mL bovine serum albumin (3 mg/mL), and 5 mL MMLV reverse transcriptase (200  U/mL) are added to a total volume of 50.5 mL. 3.4. PCR Quantification of mRNA Expression

1. Quantitation of OPRT, DPD cDNA and an internal reference cDNA (b-actin) is done using a fluorescence detection method (ABI PRISM 7700 Sequence Detection System [TaqMan]) (15). 2. This method uses a dual-labeled fluorogenic oligonucleotide probe that anneals specifically within forward and reverse primers. Laser stimulation within the capped wells containing the reaction mixture causes emission of 3¢ quencher dye (TAMRA) until the probe is cleaved by the 5¢ to 3¢ nuclease activity of the DNA polymerase during PCR extension, causing release of a 5¢ reporter dye (6FAM) (see Note 2). 3. The 25 mL PCR mixture contains 600 nM/L of each primer, 200  nM/L each of dATP, dCTP, and dGTP, 400  mM/L dUTP, 5.5 mM/L MgCl2, and 1× TaqMan buffer A containing a reference dye. The PCR conditions were 50°C for 10 s and 95°C for 10 min, followed by 42 cycles at 95°C for 15 s and 60°C for 1  min. OPRT and DPD mRNA levels are quantified as ratios between two measurements (target gene/b-actin). 4. Data analysis of quantitative RT-PCR is performed using a Student’s t-test.

4. Notes 1. The extent of fragmentation essentially depends on fixation time, and the optimal time of fixation in molecular analyses is approximately 24 h. 2. Production of an amplicon, thus, causes emission of a fluorescent signal that is detected by the system’s charge-coupled device detection camera, and the amount of signal produced at a threshold cycle within the purely exponential phase of the PCR reflects the starting copy number of the sequence of interest. Comparison of the starting copy number of the particular sequence with the starting copy number of the reference gene provides a relative gene expression level.

25  Quantitative Analysis of the Enzymes Associated with 5-Fluorouracil Metabolism…

305

References 1. Longley, D. B., Harkin, D. P., and Johnston, P. G. (2003) 5-fluorouracil: mechanisms of action and clinical strategies, Nat Rev Cancer 3, 330–338. 2. Miyake, H., Hara, I., Yamazaki, H., and Eto, H. (2005) Clinical outcome of oral uracil/tegafur (UFT) therapy for patients with hormone refractory prostate cancer, Oncol Rep 14, 673–676. 3. Bhandari, M. S., Pienta, K. J., Fardig, J., Olson, K., and Smith, D. C. (2006) Phase II trial of oral uracil/tegafur plus leucovorin in patients with hormone-refractory prostate carcinoma, Cancer 106, 1715–1721. 4. Hoshi, S., Ohyama, C., Hagisawa, S., Ono, K., Satoh, M., Saito, S., Fukuzaki, A., and Arai, Y. (2003) Complete regression of bone metastases on super bone scan, by low-dose cisplatin, UFT, diethylstilbestrol diphosphate, and dexamethasone in a patient with hormonerefractory prostate cancer, Int J Clin Oncol 8, 118–120. 5. Kubota, T. (2003) 5-fluorouracil and dihydropyrimidine dehydrogenase, Int J Clin Oncol 8, 127–131. 6. Isshi, K., Sakuyama, T., Gen, T., Nakamura, Y., Kuroda, T., Katuyama, T., and Maekawa, Y. (2002) Predicting 5-FU sensitivity using human colorectal cancer specimens: comparison of tumor dihydropyrimidine dehydrogenase and orotate phosphoribosyl transferase activities with in  vitro chemosensitivity to 5-FU, Int J Clin Oncol 7, 335–342. 7. Oguri, T., Achiwa, H., Bessho, Y., Muramatsu, H., Maeda, H., Niimi, T., Sato, S., and Ueda, R. (2005) The role of thymidylate synthase and dihydropyrimidine dehydrogenase in resistance to 5-fluorouracil in human lung cancer cells, Lung Cancer 49, 345–351. 8. Sakamoto, E., Nagase, H., Kobunai, T., Oie, S., Oka, T., Fukushima, M., and Oka, T. (2007) Orotate phosphoribosyltransferase expression level in tumors is a potential determinant of the efficacy of 5-fluorouracil, Biochem Biophys Res Commun 363, 216–222. 9. Wada, Y., Yoshida, K., Suzuki, T., Mizuiri, H., Konishi, K., Ukon, K., Tanabe, K., Sakata, Y., and Fukushima, M. (2006) Synergistic effects

of docetaxel and S-1 by modulating the expression of metabolic enzymes of 5-fluorouracil in human gastric cancer cell lines, Int J Cancer 119, 783–791. 10. Fujii, R., Seshimo, A., and Kameoka, S. (2003) Relationships between the expression of thymidylate synthase, dihydropyrimidine dehydrogenase, and orotate phosphoribosyl transferase and cell proliferative activity and 5-fluorouracil sensitivity in colorectal carcinoma, Int J Clin Oncol 8, 72–78. 11. Katsumata, K., Tomioka, H., Sumi, T., Yamashita, S., Takagi, M., Kato, F., Nakamura, R., Koyanagi, Y., Aoki, T., and Kato, K. (2003) Correlation between clinicopathologic factors and kinetics of metabolic enzymes for 5-fluorouracil given to patients with colon carcinoma by two different dosage regimens, Cancer Chemother Pharmacol 51, 155–160. 12. Birtle, A. J., Newby, J. C., and Harland, S. J. (2004) Epirubicin carboplatin and 5-fluorouracil (ECarboF) chemotherapy in metastatic hormone refractory prostate cancer, Br J Cancer 91, 1472–1476. 13. Droz, J. P., Muracciole, X., Mottet, N., Ould Kaci, M., Vannetzel, J. M., Albin, N., Culine, S., Rodier, J. M., Misset, J. L., Mackenzie, S., Cvitkovic, E., and Benoit, G. (2003) Phase II study of oxaliplatin versus oxaliplatin combined with infusional 5-fluorouracil in hormone refractory metastatic prostate cancer patients, Ann Oncol 14, 1291–1298. 14. Tanaka, T., Kawashima, H., Matsumura, K., Yamashita-Hosono, T., Yoshimura, R., Kuratsukuri, K., Harimoto, K., and Nakatani, T. (2009) Overexpression of orotate phosphoribosyl transferase in hormone-refractory prostate cancer, Oncol Rep 21, 33–37. 15. Lord, R. V., Salonga, D., Danenberg, K. D., Peters, J. H., DeMeester, T. R., Park, J. M., Johansson, J., Skinner, K. A., Chandrasoma, P., DeMeester, S. R., Bremner, C. G., Tsai, P. I., and Danenberg, P. V. (2000) Telomerase reverse transcriptase expression is increased early in the Barrett’s metaplasia, dysplasia, adenocarcinoma sequence, J Gastrointest Surg 4, 135–142.

wwwwwww

Chapter 26 Microdissection of Gonadal Tissues for Gene Expression Analyses Anne Jørgensen, Marlene Danner Dalgaard, and Si Brask Sonne Abstract Laser microdissection permits isolation of specific cell types from tissue sections or cell cultures. This may be beneficial when investigating the role of specific cells in a complex tissue or organ. In tissues with easily distinguishable morphology, a simple hematoxylin staining is sufficient, but in most cases a more specific staining is required to identify which cells to microdissect. We have established two staining protocols for frozen sections (1) Oil red O, which stains lipid droplet in fat cells and steroid-producing cells and (2) NBT BCIP, which stains cells expressing an alkaline phosphatase enzyme, such as fetal germ cells, testicular carcinoma in situ cells, and putatively also other early stem cell populations. We have applied these protocols for microdissection of rat Leydig cells, fetal human and zebrafish germ cells, and human testicular germ cell tumors, but the staining protocols could also be used in other species and for other cell types containing lipid droplets or expressing alkaline phosphatase. Both protocols ensure a morphology that enables microdissection of single cells with RNA quality sufficient for subsequent gene expression analysis. However, RNA yields after microdissection and purification are small, and therefore, two rounds of linear amplification are recommended prior to gene expression analysis. Key words: Alkaline phosphatase, Oil red O, Lipid droplets, NBT BCIP, Histological staining, Fetal gonads, Gene expression

1. Introduction Laser microdissection (LMD) is a powerful tool to isolate specific cells or tissues and thereby ensure a specific gene expression profile without noise from other cells in a tissue. The first step when setting up a LMD protocol for gene expression analysis is to ensure the highest possible RNA quality. This includes optimal tissue storage, fixation, and staining procedures (1).

Graeme I. Murray (ed.), Laser Capture Microdissection: Methods and Protocols, Methods in Molecular Biology, vol. 755, DOI 10.1007/978-1-61779-163-5_26, © Springer Science+Business Media, LLC 2011

307

308

A. Jørgensen et al.

RNA from tissues fixed in formalin or other cross-linking fixatives is more fragmented than RNA from frozen tissue. Although it is possible to use microdissected tissue from formalin-fixed, paraffinembedded tissue for microarray gene expression analysis (2, 3), frozen tissue is preferable. In our experience, the most critical factors when working with frozen tissue are to avoid repeated thawing and to limit the time in aqueous solutions throughout the procedure. To avoid initial RNA degradation, the tissue should be snap-frozen and stored at −80°C. We have previously shown that storage of cryosections in 100% ethanol at −80°C for one month did not affect the RNA quality, providing an alternative to repeated thawing of the tissue for sectioning (1). Fixation of the frozen tissues should fulfill three criteria: It should (1) retain a high RNA quality, (2) give the best possible morphology, and (3) be compatible with the staining procedure. In our hands, ice-cold acetone, modified methacarn (8:1, methanol and acetic acid) and Carnoy’s fixative (6:1:3, ethanol, acetic acid, and chloroform) gave the best RNA quality. The most optimal morphology was obtained with ice-cold acetone, 100 and 75% ethanol, but other fixatives may also be used (1, 4–6). Generally, the morphology of frozen and dehydrated tissue is poor, and it is difficult to distinguish between different cells; therefore, a staining protocol is essential for identification of specific cell types. Hematoxylin staining is sufficient if the structures for microdissection are easily distinguished in the tissue; otherwise, it is necessary with a more specific staining protocol. We optimized two different staining protocols that retain a RNA quality sufficient for subsequent microarray analysis (7) (1) Oil red O (ORO) for Leydig cells and (2) NBT BCIP for fetal germ cells and testicular carcinoma in situ (CIS) cells. Leydig cells produce steroid hormones and, therefore, contain lipid droplets that are stained by ORO (1). Fetal germ cells and CIS cells retain embryonic stem cell characteristics including expression of alkaline phosphatase (1, 8), which can be visualized using NBT BCIP. Both staining protocols are applicable only in frozen tissues, and for the ORO protocol the prestaining steps should preserve the lipid droplets. We used 60% isopropanol for ORO and 75% ethanol for NBT BCIP for fixation of tissue, as they were the least toxic alternatives compatible with our staining protocols. RNA was extracted from microdissected cells and amplified by two rounds of linear amplification, which maintains the best possible representation of the starting RNA population. We have shown that RNA from microdissected cells prepared according to the presented protocols is of a quality and quantity that allows subsequent RT-PCR and microarray analysis (7, 9).

26  Microdissection of Gonadal Tissues for Gene Expression Analyses

309

2. Materials 1. O.C.T. compound (Tissue-tek, Sakura Finetek). 2. Isopentane. 3. Diethylpyrocarbonate (DEPC; Sigma-Aldrich, St Louis, MO). 4. Ethanol. 5. Isopropanol. 6. Dimethylformamide (DMF). 7. ORO solution: 24 mL ORO stock solution (0.5 g ORO in 100 mL 99% isopropanol) with 16 mL DEPC-treated H2O. Incubate the solution for 10 min at RT. Filter through 3 W Munktell filter paper (store at RT). 8. In situ hybridization buffer (10): 100 mL 1 M Tris, pH 9.5, 100 mL 1 M NaCl, 50 mL 1 M MgCl2, DEPC-treated H2O to 1 L (store at RT). 9. NBT/BCIP solution: 7 mL NBT stock solution (75 mg NBT in 1 mL 70% v/v DMF/DEPC-treated H2O – store at −20°C) and 9 mL BCIP stock solution (50 mg BCIP in 1 mL 100% DMF) in 2 mL in situ hybridization buffer, sterile filter and use immediately. 10. Mayer’s hematoxylin. 11. Membrane slides (Molecular Machines & Industries, Glattbrugg, Switzerland). 12. Tubes with adhesive lids (Molecular Machines & Industries). 13. RNAqueous Micro kit (Applied Biosystems). 14. Amino Allyl MessageAmp II aRNA Amplification Kit (Applied Biosystem, Carlsbad, CA). 15. AMV reverse transcriptase (15  U/mL) (USB Corporation, Cleveland, OH). 16. Cy3 Mono-Reactive Dye Pack (GE Healthcare). 17. Agilent Gene Expression Hybridization Kit (Agilent Technologies, Santa Clara, CA). 18. Agilent Whole Human, Zebrafish, and Rat Genome Microarray 4 × 44K chips. 19. 5× cDNA synthesis buffer: 650 mL 1 M Tris–HCl, pH 8.3, 25 mL 1 M MgCl2, 100 mL 1 M KCl, and 225 mL DEPCtreated H2O (store at −20°C). 20. cDNA synthesis mix (for two samples): 2  mL 5× cDNA synthesis buffer, 0.5 mL 2.5 mM dNTP, 5.5 mL DEPC-treated H2O, and 0.3 mL AMV reverse transcriptase (15 U/mL).

310

A. Jørgensen et al.

3. Methods 3.1. Preparation of Tissues Prior to Staining

Tissues were embedded in O.C.T. compound and immediately frozen in −78°C isopentane (on dry ice) and stored at −80°C. 10–20  mm sections were cut on a Shandon SME cryotome (Life Sciences International Europe Ltd.), collected on RNAsefree membrane slides (Molecular Machines & industries) (see Note 1) and immediately fixed.

3.2. Fixation for NBT/ BCIP and Mayer’s Hematoxylin Staining

1. Fix freshly cut sections for 10 min in 75% ethanol.

3.3. Mayer’s Staining Protocol

1. Prior to staining, leave sections to air-dry for 5 min.

2. Store in 100% ethanol at −80°C until use.

2. Incubate sections for 10 s in In situ hybridization buffer. 3. Stain for 10 s in Mayer’s hematoxylin (see Note 2). 4. Wash for 10 s in DEPC-treated H2O. 5. Dehydrate: for 10 s in 62% ethanol, two times for 10 s in 96% ethanol, two times for 10 s in 100% ethanol. 6. Air-dry.

3.4. NBT BCIP Staining Protocol

1. Prior to staining, leave sections to air-dry for 5 min. Incubate for 10 s in In situ hybridization buffer. 2. Stain for 90–120  s in NBT BCIP staining solution (see Note 3). 3. Wash for 10 s in DEPC-treated H2O. 4. Dehydrate: for 10 s in 62% ethanol, two times for 10 s in 96% ethanol, and two times for 10 s in 100% ethanol. 5. Air-dry.

3.5. Oil Red O Staining Protocol

1. Fix freshly cut tissue sections for 5 min in 60% 2-propanol. 2. Stain sections for 5 min in ORO staining solution. 3. Wash thoroughly in 60% 2-propanol. 4. Wash thoroughly in DEPC-treated H2O (see Note 4). 5. Stain in Mayer’s hematoxylin for 1 s. 6. Wash in DEPC-treated H2O for 10 s. 7. Air-dry.

3.6. Laser Microdissection

1. We microdissected stained tissues using a MMI SmartCut (Olympus). For each sample, 500–5,000 cells were collected on the adhesive cap of collection tubes (Molecular Machines & Industries) (see Notes 5–7). Laser intensity and focus was calibrated to each individual tissue type and section thickness.

26  Microdissection of Gonadal Tissues for Gene Expression Analyses

3.7. RNA Preparation and Amplification

311

1. RNA was purified using the RNAqueous micro kit (Ambion) according to the manufacturer’s protocol for LCM samples. Except that the 30 min incubation at 42°C was performed in an inverted heating block similar to the ExtraSure device from Applied Biosystems. RNA quality was tested on a bioanalyzer 6000 Pico total RNA kit according to the manufacturer’s protocol (Agilent Technologies). 2. Samples were amplified twice using the Amino Allyl MessageAmp II aRNA Amplification Kit. Second round amplification was performed with amino allyl labeled nucleotides if samples were used for microarray analysis. Otherwise, unmodified nucleotides were used (see Note 8). 3. RNA fragment size was tested on a bioanalyzer 6000 Nano total RNA kit, showing an average fragment size of approximately 500 bases (see Note 9).

3.8. cDNA Synthesis Protocol

1. Vacuum-dry frozen RNA samples to a maximum volume of 4 mL. 2. Add 1 mL 250 ng/mL random hexamer primers. 3. Incubate for 1 min at 80°C, 1 min on ice, and 1 min at 30°C. 4. Add 1 mL 5× cDNA synthesis buffer. 5. Add 4 mL cDNA synthesis mix. 6. Incubate for 1 h at 37°C. 7. Add 30 mL 0.1% Triton X-100. 8. Mix and spin down. 9. Incubate for 5 min at 95°C.

3.9. Labeling and Fragmentation for Microarray Analysis

Amino allyl aRNA samples were labeled with Cy3 as described in the Amino Allyl MessageAmp II aRNA Amplification protocol: “aRNA:Dye Coupling Reaction” (see Note 10). Fragmentation and hybridization was performed according to Agilent’s OneColor Microarray-Based Gene Expression protocol.

4. Notes 1. When cutting frozen sections, place two to three sections on the middle of the membrane slide as follows: hold your finger on the other side of the membrane to warm it up and melt the section to the membrane. 2. Mayer’s hematoxylin can be added to the slides by dripping 100 mL on each section. 3. NBT BCIP can be added to the slides by dripping 100 mL on each section.

312

A. Jørgensen et al.

4. ORO stained tissues should be carefully rinsed to get rid of isopropanol and excess ORO before the Mayer’s staining. 5. After staining and dehydration, sections should be completely dry before initiating the laser microdissection procedure. 6. Laser microdissection was performed on a MMI SmartCut system (Olympus). With this system, the selected cells are cut out of a membrane and collected on an adhesive cap. Because the tissue sections are placed on the opposite side of the membrane, contamination is minimized. 7. After laser microdissection, 50–100  mL lysis buffer was immediately added to the tube and inverted, so the lysis buffer covered the adhesive lid with the collected cells. The tube was then transferred to dry ice upside down and later stored at −80°C until RNA preparation. Occasionally, two to four tubes with cells from the same tissue were pooled on the same column to collect RNA from approximately 500–5,000 cells. 8. All samples should be amplified under the same conditions (length of ON in vitro transcription) and with the same number of amplifications for comparability. 9. When RNA from laser microdissected tissues have been amplified, only the 3¢ end of the genes are represented with an average fragment length of approximately 500 bases; this has to be considered in subsequent analyses. If RNA is used for microarray analysis, it is recommended to check the probe design of interesting genes, to make sure that the probe is placed in the 3¢ end of the gene. 10. The samples can be labeled with Cy5 and used for two color arrays if this is preferred. Other array platforms than Agilent may also be used; however, it should be kept in mind that probes placed in the 3¢ end of the genes are more reliable.

Acknowledgments This chapter is based on the research funded by the Danish Cancer Research Society, the Villum Kann Rasmussen Foundation, Svend Andersen’s foundation, and Kirsten and Freddy Johansen’s Foundation.

26  Microdissection of Gonadal Tissues for Gene Expression Analyses

313

References 1. Sonne SB, Dalgaard MD, Nielsen JE, HoeiHansen CE, Rajpert-De ME, Gjerdrum LM et al. (2009) Optimizing staining protocols for laser microdissection of specific cell types from the testis including carcinoma in situ. PLoS One 4, e5536. 2. Frank M, Doring C, Metzler D, Eckerle S, Hansmann ML (2007) Global gene expression profiling of formalin-fixed paraffin-embedded tumor samples: a comparison to snap-frozen material using oligonucleotide microarrays. Virchows Arch 450, 699–711. 3. Schoor O, Weinschenk T, Hennenlotter J, Corvin S, Stenzl A, Rammensee HG et  al. (2003) Moderate degradation does not preclude microarray analysis of small amounts of RNA. Biotechniques 35, 1192–1201. 4. Buckanovich RJ, Sasaroli D, O’brien-Jenkins A, Botbyl J, Conejo-Garcia JR, Benencia F et al. (2006) Use of immuno-LCM to identify the in situ expression profile of cellular constituents of the tumor microenvironment. Cancer Biol Ther 5, 635–642. 5. Cox ML, Schray CL, Luster CN, Stewart ZS, Korytko PJ, KN MK et al. (2006) Assessment of fixatives, fixation, and tissue processing on morphology and RNA integrity. Exp Mol Pathol 80, 183–191.

6. Sluka P, O’Donnell L, McLachlan RI, Stanton PG (2008) Application of laser-capture microdissection to analysis of gene expression in the testis. Prog Histochem Cytochem 42, 173–201. 7. Sonne SB, Almstrup K, Dalgaard M, Juncker AS, Edsgard D, Ruban L et al. (2009) Analysis of gene expression profiles of microdissected cell populations indicates that testicular carcinoma in situ is an arrested gonocyte. Cancer Res 69, 5241–5250. 8. Jorgensen A, Nielsen JE, Morthorst JE, Bjerregaard P, Leffers H (2009) Laser capture microdissection of gonads from juvenile zebrafish. Reprod Biol Endocrinol 7:97–103. 9. Jorgensen A, Nielsen JE, Nielsen BF, Morthorst JE, Bjerregaard P, Leffers H (2010) Expression of prostaglandin synthases (pgds and pges) during zebrafish gonadal differentiation. Comp Biochem Physiol Part A 157, 102–108. 10. Nielsen JE, Hansen MA, Jorgensen M, Tanaka M, Almstrup K, Skakkebaek NE et  al. (2003) Germ cell differentiationdependent and stage-specific expression of LANCL1 in rodent testis. Eur J Histochem 47, 215–222.

wwwwwww

Chapter 27 Duplex Real-Time PCR Assay for Quantifying Mitochondrial DNA Deletions in Laser Microdissected Single Spiral Ganglion Cells Adam Markaryan, Erik G. Nelson, and Raul Hinojosa Abstract Laser microdissection (LMD) has been used to isolate groups of cells and single cells from numerous tissues. In this chapter, we describe a technique for isolating individual spiral ganglion cells from archival formalin-fixed, celloidin-embedded (FFCE) human temporal bone sections. The DNA isolated from these single cells is suitable for analysis with a duplex real-time polymerase chain reaction (PCR) methodology to quantify the mitochondrial DNA (mtDNA) deletion level present. Key words: Presbycusis, Human temporal bones, Mitochondrial DNA deletions, Duplex real-time PCR assay, Laser microdissection

1. Introduction Somatic mtDNA mutations have been associated with age-related diseases in tissues with high energy demand such as the brain and muscles (1). Similarly, studies in our laboratory suggest that mtDNA deletions play a role in the age-related loss of hearing referred to as presbycusis (2–4). The presence of the common mtDNA deletion (CD) in archival cochlear tissues from individuals with presbycusis was initially reported by Seidman et al (5) in 1996. Subsequent investigations have demonstrated an association between the quantitatively measured CD level and the extent of cochlear element degeneration in these archival cochlear tissues (3). The deletion levels in this study were determined using a real-time PCR assay, which is detailed in this chapter. In addition to performing this assay on whole cochlear tissue sections,

Graeme I. Murray (ed.), Laser Capture Microdissection: Methods and Protocols, Methods in Molecular Biology, vol. 755, DOI 10.1007/978-1-61779-163-5_27, © Springer Science+Business Media, LLC 2011

315

316

A. Markaryan et al.

PCR assays have also been used to study the tissues of individual cochlear elements and single ganglion cells isolated by LMD (4, 6). Investigating the relationship between mtDNA deletion levels in single cells and cell loss through apoptosis may lead to a better understanding of the observed degeneration of cochlear tissues in individuals with presbycusis (7, 8). Therefore, we have utilized real-time PCR assays that estimate both the CD level and the total deletion level in the major arc of the mtDNA genome to study mtDNA isolated from single spiral ganglion cells (4).

2. Materials 2.1. FFCE Tissue Preparation and LMD

1. Director laser microdissection slides [Expression Pathology, Inc., (Gaithersburg, MD)]. 2. Ether (Acros Organics, Geel, Belgium). 3. Absolute ethanol (Acros Organics, Geel, Belgium). 4. Leica AS LMD instrument (Leica Microsystems, Wetzlar, Germany). 5. 0.5-ml PCR tubes – flat top – RNAase/DNAase Pyrogen Safe (LabSource, Chicago, IL).

2.2. DNA Isolation

1. QIAamp DNA Micro Kit (Qiagen, Valencia, CA): Tissue lysis buffer (ATL), lysis buffer (AL), elution buffer (AE), proteinase K, carrier RNA, MinElute column, wash buffers AW1 and AW2. 2. DNAse–RNAase-free water (Invitrogen, Carlsbad, CA). 3. Microfuge (Labnet International, Edison, NJ). 4. Vortex mixer (Barnstead/Thermoline, Dubuque, IA).

2.3. Preparation of mtDNA Standards and Controls

1. TOPO TA Cloning kit (Invitrogen, Carlsbad, CA). 2. Wizard Plus SV Minipreps DNA Purification System (Promega, Madison, WI). 3. 3730XL 96-capillary Automated DNA Sequencer (Applied Biosystems, Foster City, CA). 4. NanoDrop 1000 spectrophotometer (NanoDrop Technologies, Wilmington, DE). 5. DNeasy Blood and Tissue Kit (Qiagen, Valencia, CA).

2.4. Duplex Real-Time PCR Assay

1. PCR Workstation (Fisher Scientific, Waltham, MA). 2. HPLC-purified real-time PCR primers (Table 1) and IDTE nuclease-free buffer – 10  mM Tris–HCl, pH 8.0, 0.1  mM EDTA (Integrated DNA Technologies, Inc., Coralville, IA).

27  Duplex Real-Time PCR Assay for Quantifying Mitochondrial DNA Deletions…

317

Table 1 Real-time PCR primers and probes Name

Sequence

Address (nt)

D-loop forward

GCT TTC CAC ACA GAC ATC ATA ACA A

263–287

D-loop reverse

GTT TAA GTG CTG TGG CCA GAA G

338–317

D-loop probe

6FAM-AAT TTC CAC CAA ACC CC-MGB

290–306

ND4 forward

CCATTCTCCTCCTATCCCTCAAC

12087–12109

ND4 reverse

CACAATCTGATGTTTTGGTTAAACTATATTT

12170–12140

ND4 probe

VIC-CCGACATCATTACCGGGTTT-MGB

12111–12138

CD forward

CTTACACTATTCCTCATCACCCAACTAAAA A

8417–8447

CD reverse

GGAGTAGAAACCTGTGAGGAAAGG

13509–13486

CD probe

VIC-CAT TGG CAG CCT AGC ATT-MGB

8481–8482, 13460–13475

3. Real-time PCR probes (Table 1) (Applied Biosystems, Foster City, CA). 4. TaqMan Gene Expression Master Mix – supplied as a 2× solution (Applied Biosystems, Foster City, CA). 5. 384-well plates (Applied Biosystems, Foster City, CA). 6. MicroAmp Optical Adhesive Film – PCR compatible, DNA/ RNA/RNAse Free (Applied Biosystems, Foster City, CA). 7. Sorvall Legend T/RT centrifuge with microplate rotor (Kendro Laboratory Products, Inc., Asheville, NC). 8. ABI 7900HT real-time PCR instrument (Applied Biosystems, Foster City, CA).

3. Methods The DNA isolated from single ganglion cells can be analyzed with duplex real-time PCR assays developed in our laboratory. Using this approach, an estimate of the total mtDNA deletion level or the level of specific mtDNA deletions, such as the CD, are measured. A simplified diagram of the human mtDNA genome appears in Fig. 1. The D-loop region of the mtDNA genome contains the origin of heavy-chain replication and both promoter site sequences. Therefore, the D-loop must be present in all replicating mtDNA molecules and provides an estimate of the total mtDNA copy number in a sample. The mtDNA sequence encoding the NADH dehydrogenase subunit 4 (ND4) is located in the

318

A. Markaryan et al. 338 263

OH D-loop

4 D

CD

12170

N

OL

8417 8482

13509 13460

12087

X3

CO

9290

9207

Fig. 1. Diagram of the human mtDNA genome, which is circular and contains 16,569 bp. OH origin of heavy-chain replication, OL origin of light chain replication. Dashed line represents the deleted sequence of the CD. Black bars represent the real-time PCR D-loop, ND4, and CD amplicon sites. Numbers represent the nucleotide address (nt).

major arc of the mtDNA genome where deletions most frequently occur. A determination of the relative ND4 to D-loop copy number represents an estimate of the total amount of deleted mtDNA present. The D-loop and ND4 sequence copy number can be determined in a single reaction well with a duplex PCR assay rather than running the assays in two different wells. The duplex approach minimizes result variability due to pipetting inaccuracies. In addition, minute samples are conserved, and reagent costs are limited. The CD level is determined similarly by quantifying the CD break point sequence and the D-loop sequence in a duplex PCR assay. 3.1. Isolation of Single Spiral Ganglion Cells by LMD

1. Midmodiolar FFCE temporal bone sections are transferred from an 80% ethanol storage solution and placed on director laser microdissection slides in an open petri dish. The sections must be maintained with an adequate amount of 80% ethanol coverage. Irreversible wrinkling of the section will occur if it is allowed to dry out prior to celloidin removal. 2. One side of the petri dish is slightly elevated to place the slide in a gently inclined position. The celloidin is removed from the tissue by applying a solution of equal parts ethanol and ether to the elevated end on the slide and excess solution is wicked from the lower end on the slide with a paper towel. 1 ml of solution is applied at minute intervals for a total of 5 min. Extreme care is required to prevent an excessive rate of solution flow over the tissue, which may displace or distort

27  Duplex Real-Time PCR Assay for Quantifying Mitochondrial DNA Deletions…

319

Fig. 2. (a) Spiral ganglion prior to LMD. (b) Spiral ganglion after LMD removal of two spiral single ganglion cells.

the delicate structures of the cochlea, which are not attached to the slide until drying has occurred. Following celloidin removal, the sections are allowed to dry at room temperature for 30 min. 3. Areas of otic capsule bone, which were not firmly attached to the slide, are trimmed manually with a scissors. Membranous structures of the inner ear routinely maintain a firm attachment to the slide. 4. The Leica AS LMD instrument is used for LMD (6). Following proper laser calibration, tissue dissection is performed at ×200 magnification with an aperture setting of 6, a power intensity of 45, and a laser speed of 5. Areas of interest are isolated by directing the laser beam along the contours of an individual cell. Additional laser pulses are applied as needed to dislodge single cells from the surface of the slide. Micrographs of a temporal bone section demonstrating the spiral ganglion area before LMD and after the removal of two single ganglion cells are shown in Fig. 2. 5. Each LMD ganglion cell is collected in a 0.5-ml PCR tube cap containing 20 ml of tissue lysis buffer (ATL) from the QIAamp DNA Micro Kit. The tube is closed with the cap remaining in the inverted position and then centrifuged at 10,000 ´ g for 1 min to transfer the buffer and specimen from the cap into the bottom of the tube. The samples are stored at room temperature until DNA isolation is performed. 3.2. Preparation of DNA Samples

1. 10 ml of proteinase K is added to the 20 ml of tissue lysis buffer (ATL) containing the single ganglion cell sample and mixed by vortexing for 15 s. The tube is incubated at 56°C for 1 h with occasional agitation and then an additional 25 ml of tissue lysis buffer (ATL) is added. 2. 50 ml of lysis buffer (AL) containing 1 mg of carrier RNA is added and mixed for 15 s to enhance DNA binding to the column and improve the yield.

320

A. Markaryan et al.

3. 50  ml of absolute ethanol is added, mixed thoroughly by vortexing for 15  s, and incubated for 5  min at room temperature. 4. The entire lysate is transferred to the QIAamp MinElute Column and centrifuged at 4,000 ´ g for 1 min. The column is washed with 500  ml of AW1 buffer followed by an additional wash with 500 ml of AW2 buffer. The DNA is eluted from the column with 20 ml of water, instead of the elution buffer (AE) provided with the kit, and stored at –20°C (see Note 1). 5. The background level of DNA contamination during sample preparation is assessed by processing a lysis buffer sample without a ganglion cell in parallel as negative control. 3.3. Preparation of mtDNA Standards

1. PCR products representing the D-loop (417 bp), ND4 region (562 bp), and the CD (316 bp) are used to generate plasmid standards (4). The D-loop fragment and the ND4 fragment were amplified from blood DNA. The CD fragment was amplified from human temporal bone DNA. 2. All three fragments were individually cloned into the 3,956 bpPCR4-TOPO vector from the TOPO TA Cloning kit. 3. The plasmid DNA generated was purified using the Wizard Plus SV Minipreps DNA Purification System. 4. The standards were sequenced to confirm the specificity of the inserts using the 3730XL 96-capillary Automated DNA Sequencer and the DNA concentration of each was measured using the NanoDrop 1000 spectrophotometer. 5. Placenta DNA isolation is performed with the components included in the DNeasy Blood and Tissue Kit. Placenta tissue, fixed in buffered 10% formalin, is trimmed to ensure the removal of contaminating maternal components. A 20-mg tissue sample is washed in PBS and finely minced. The tissue is added to 180 ml of ATL buffer and incubated with 20 ml of proteinase K at 56°C for ~2  h until complete tissue lysis is observed. Particulate material is removed by centrifugation. The sample is applied to the silica-based membrane column and washed with the two-step buffer system. DNA is eluted from the column using 100 ml of the kit AE buffer and the concentration is determined with the NanoDrop 1000 spectrophotometer. An approximate DNA yield of 10  mg is expected from the 20 mg placenta tissue sample.

3.4. Duplex Real-Time PCR Assay

The PCR process involves a series of steps that are generated by temperature changes referred to as a thermal cycle. Initially, the temperature of the reactants is elevated to a degree that results in the dissociation of double-stranded DNA into single strands. The

321

27  Duplex Real-Time PCR Assay for Quantifying Mitochondrial DNA Deletions…

reactants are then cooled to a temperature that allows the binding, or annealing, of primers to the single-stranded DNA molecules that are specific for the DNA sequence of interest. A polymerase enzyme that has the capacity to copy DNA sequences at an elevated temperature recognizes the primer and copies the sequence. The resultant double-stranded DNA is then available for an ­additional copying cycle. With each cycle the copy number is doubled, resulting in an exponential increase in copies with multiple cycles. The real-time PCR methodology provides a means to quantify mtDNA at levels several orders of magnitude less than other techniques. Using this method, the copy number of a DNA sequence is determined based on the number of PCR cycles required to generate a detectable signal referred to as the cycle threshold (Ct). Greater copy numbers require fewer cycles for detection. The TaqMan DNA polymerase real-time PCR methodology described in this manuscript measures the amplification of specific sequences and, unlike real-time PCR methods based on measuring double-stranded DNA production, avoids potential inaccuracies due to the amplification of nonspecific templates. Figure 3 depicts a diagram of the steps involved. This approach utilizes a probe that binds specifically to the DNA sequence of interest. The probe contains a fluorescent reporter dye that does not emit light when it is adjacent to a quencher that is also attached to the probe. Copying of the DNA template is initiated by the polymerase enzyme at the primer site. As the probe binding site is reached, the reporter dye is cleaved from the probe and its fluorescence is detected and measured. The probe is subsequently displaced from the template by the polymerase enzyme and the template copy is completed. The template copy generated

R R R Primer

Q

Q

Probe

3’ 5’ Primer

5’

3’

5’

3’

5’

3’

Q

3’

5’

5’

3’

3’

5’

5’

3’

Fig. 3. Schematic diagram depicting the real-time PCR cycle. Step 1. Following the dissociation of DNA into single strands at elevated temperatures, the probe with the reporter and quencher attached binds specifically to the DNA template. The reporter does not fluoresce due its close proximity to the quencher. As the reaction temperature is reduced, specific forward and reverse primers bind to the DNA template of interest. Step 2. DNA polymerase binds to the primer site and template copying begins. When the probe is encountered, the reporter is cleaved from the probe and begins to fluoresce. Step 3. Further hydrolysis and displacement of the probe occurs as template copying is completed. The intact template copies are then available for primer and probe binding in a subsequent real-time PCR cycle.

322

A. Markaryan et al.

is then available for binding with primers and probes in an additional PCR cycle. The duplex real-time PCR assay described in this study permits the simultaneous quantification of two DNA sequences of interest in a single reaction well. This involves the use of two different primer sets and two different probes for specific each sequence. Each probe has a different reporter dye with a distinct fluorescence emission spectra. The DNA polymerase enzyme provided in the kit is uniquely engineered for optimal results with the duplex real-time PCR assay. The TaqMan Gene Expression Master Mix reagent includes all of the reactants required for the PCR process. The previously described primer and probe sequences for the D-loop region (9), the ND4 region (4, 10), and the CD (3) appear in Table 1. The probes in these duplex assays utilize FAM and VIC as reporter dyes and MGB quenchers (see Note 2). 3.4.1. Preparation of Stock Primer and Probe Solutions for the Real-Time PCR Assay (see Note 3)

The HPLC purified primers for the D-loop region, the ND4 region, and the CD are dissolved in IDTE buffer at a stock concentration of 1 mM. The probes for these regions are supplied at a concentration of 100  mM in TE (10  mM Tris–HCl, pH 8.0, 1 mM EDTA) buffer. The primers and probes are stored at −20°C. Immediately prior to the assay, the primers and probes are diluted in water to 10 mM and 5 mM, respectively.

3.4.2. Preparation of the Reaction Plate for the Duplex Real-Time PCR Assay

The reaction reagents used in each plate reaction well are identical. Therefore, the reagents are mixed and then placed in the well in one pipetting step. Preparing a reagent mix that contains all of the reaction components minimizes variability in the results between wells. In addition, this reduces the number of pipetting steps, and consequently limits the chances of contamination and pipetting errors. The standards, controls, and spiral ganglion DNA samples are added to the wells as a second and final pipetting step. The actual individual reagent volumes in each well are listed in Table  2 for the D-loop – ND4 assay and in Table  3 for the D-loop – CD assay. The volume of each reagent component used to prepare the reaction mixture is based on the values in these tables (see Note 4). The final primers and probes concentrations in the reaction mixtures are 300 nM and 100 nM, respectively. The standards are run in triplicate over a copy number range of 3 × 102–3 × 106. Serial twofold dilutions of the placenta DNA sample (6 ng/ml), which is assumed to be free of deletions, are run in triplicate as a control to determine minor differences in efficiencies between the two assays in the duplex system. The DNA samples from each single ganglion cell are run in duplicate. The DNA yield from a single ganglion cell eluted in 20  ml of water is sufficient for assays of both the major arc mtDNA deletion level and the CD level in duplicate. Following the addition of all reagents, the reaction plate is sealed with MicroAmp Optical

27  Duplex Real-Time PCR Assay for Quantifying Mitochondrial DNA Deletions…

323

Table 2 Actual reagent volume per well for duplex ND4–D-loop assay Reagent

Volume per well (ml)

TaqMan expression master mix 2×

10

ND4-VIC probe (5 mM)

0.4

D-loop – FAM probe (5 mM)

0.4

Primer ND4 forward (10 mM)

0.6

Primer ND4 reverse (10 mM)

0.6

Primer D-loop forward (10 mM)

0.6

Primer D-loop reverse (10 mM)

0.6

Water

1.8

DNA (standard/sample)

5

Total reaction volume

20

Table 3 Actual reagent volume per well for duplex CD – D-loop assay Reagent

Volume per well (ml)

TaqMan expression master mix 2×

10

CD-VIC probe (5 mM)

0.4

D-loop – FAM probe (5 mM)

0.4

Primer CD forward (10 mM)

0.6

Primer CD reverse (10 mM)

0.6

Primer D-loop forward (10 mM)

0.6

Primer D-loop reverse (10 mM)

0.6

Water

1.8

DNA

5

Total reaction volume

20

Adhesive Film. The plate is centrifuged in the Sorvall Legend T/ RT at 1,200 rpm for 2 min and each well is visually inspected to verify that the reaction mix is positioned at the bottom of the well. The reaction plate is stored briefly at 4°C until it is loaded into the ABI 7900HT real-time PCR instrument.

324

A. Markaryan et al.

3.4.3. Programming of the ABI 7900HT Real-Time PCR Instrument

The instructions for programming the ABI 7900HT real-time PCR instrument are detailed in the instruction manual. First, an Absolute Quantitation (AQ) plate document is created, which is a record of the contents in each well of the reaction plate. A name is assigned to each well of the plate, designating the sample it contains and the detectors used for the two DNA sequences that are being analyzed (see Note 5). The reporter dyes and quencher being used are entered in the drop-down menu of the document. None is selected for the nonfluorescent MGB quencher. One of three detector tasks, unknown, standard, or no-template control, is designated for both detectors in each well of the AQ plate document, which specifies how the software uses the data collected during analysis. The cycling parameters for the assay are selected prior to initiating a run on the 7900HT instrument. The following standard thermal cycling parameters are utilized: 2  min at 50°C, 10 min at 95°C, and then 40 cycles of 15 s at 95 C followed by 1 min at 60°C (see Note 6).

3.4.4. Data Analysis

The real-time PCR data generated by the ABI 7900HT real-time PCR instrument is analyzed using Sequence Detection Software version 2.3. Use of the automatic baseline and threshold features is recommended. This software automatically generates baseline and threshold values from the detectors in each well. The amplification plots for each well should be inspected to verify that the assigned baseline and threshold values are correct. If the plot of the amplification curve follows a stable base line and the assigned threshold is in the exponential phase of the curve, the values are accepted. Duplex real-time PCR assay amplification plots and standard curves for the standards appear in Fig. 4. The formula R = (1 − 2−DDCt) × 100 is used to calculate the total deletion level, where DCt represents the difference in Ct between the ND4 and D-loop sequence in the sample assay and the second D in the formula represents the difference in Ct between the ND4 and D-loop sequence in the placenta DNA assay (11). The formula R = (2−DCt) × 100 is used to calculate the CD level, where DCt represents the difference in Ct between the CD break point sequence and the D-loop sequence in the sample assay.

4. Notes 1. Unless stated otherwise, all aqueous solutions were prepared in DNAse–RNAase-free water. 2. The use of 3¢ MGB (minor groove binder) quenchers increases the probe melting temperature without increasing probe length and, therefore, allows the design of probes that are

27  Duplex Real-Time PCR Assay for Quantifying Mitochondrial DNA Deletions…

325

Fig. 4. Duplex real-time PCR assay amplification plots (top) and standard curves (bottom) for the standards. (a) D-loop. (b) ND4. (c) CD. Top: Ordinate – Signal magnitude (DRn). Abscissa – Cycle number. Horizontal line designates the cycle threshold (Ct), the cycle at which a statistically significant increase in signal magnitude is first detected. Bottom: R 2 – Correlation coefficient. The standard curve slopes for the D-loop, ND4 region, and CD are −3.31, −3.38, and −3.46, respectively, which are within the optimal range (−3.1 to −3.6) and close to the ideal slope of −3.32. Correlation coefficients of 0.99 were observed for the plasmid standards in each of the assays.

shorter and more specific than traditional TAMRA probes. In addition, this nonfluorescent quencher offers the advantage of a lower background signal, which results in improved precision. 3. All work is performed in PCR workstation hood following a 30-min sterilization period with UV light. A set of pipettes designated for hood use only, the use of aerosol-barrier tips, and sterile tubes minimize the potential for contamination. The importance of good laboratory technique cannot be underestimated. 4. Preparing a slightly larger volume of a reagent mix than needed is recommended. 5. An entry should be made for each of the 384-wells in the AQ document, including the wells that are not utilized. 6. The correct size block for the 384-well plate must be installed in the ABI 7900HT real-time PCR instrument.

326

A. Markaryan et al.

Acknowledgments This work was supported by the Bloom Temporal Bone Laboratory Endowment and a 2009 Deafness Research Foundation Grant provided by the Burch-Safford Foundation to A.M. References 1. Reeve AK, Krishnan KJ, Turnbull D (2008) Mitochondrial DNA mutations in disease, aging and neurodegeneration. Ann NY Acad Sci 1147:21–29. 2. Markaryan A, Nelson EG, Hinojosa R (2008) Detection of mitochondrial DNA deletions in the cochlea and its structural elements from archival human temporal bone tissue. Mutation Research 640:38–45. 3. Markaryan A, Nelson EG, Hinojosa R (2009) Quantification of the mitochondrial DNA common deletion in presbycusis. Laryngoscope 119:1184–9. 4. Markaryan A, Nelson EG, Hinojosa R (2010) Major arc mitochondrial DNA deletions in cytochrome c oxidase-deficient human cochlear spiral ganglion cells. Acta Oto-Laryngologica 130:780–787. 5. Seidman MD, Bai U, Khan MJ, Murphy MP, Quirk WS, Castora FJ, Hinojosa R (1996) Association of Mitochondrial DNA Deletions and Cochlear Pathology: A Molecular Biology Tool. Laryngoscope 106:777–783. 6. Markaryan A, Nelson EG, Tretiakova M, Hinojosa R (2008) Laser microdissection of cochlear structures from celloidin embedded human temporal bone tissues and detection of the mitochondrial DNA common deletion using real time PCR. Hear Res 244:1–6.

7. Nelson EG, Hinojosa R (2003) Presbycusis: A human temporal bone study of individuals with flat audiometric patterns of hearing loss using a new method to quantify stria vascularis volume. Laryngoscope 113:1672–86. 8. Nelson EG, Hinojosa R (2006) Presbycusis: A human temporal bone study of individuals with downward sloping audiometric patterns of hearing loss and review of the literature. Laryngoscope 116 (Suppl 112):1–12. 9. Sabunciyan S, Kirches E, Krause G, Bogerts B, Mawrin C, Llenos IC, Weis S (2007) Quantification of total mitochondrial DNA and mitochondrial common deletion in the frontal cortex of patients with schizophrenia and bipolar disorder. J Neural Transm 114:665–674. 10. He L, Chinnery PF, Durham SE, Blakely EL, Wardell TM, Borthwick GM, Taylor RW, Turnbull DM (2002) Detection and quantification of mitochondrial DNA deletions in individual cells by real- time PCR. Nucleic Acid Research 30(e68):1–6. 11. Yin S, Yu Z, Sockalingam R, Bance M, Sun G, Wang J (2007) The role of mitochondrial DNA large deletion for the development of presbycusis in Fischer 344 rats. Neurobiol Dis 27:370–377.

Chapter 28 Neuronal Type-Specific Gene Expression Profiling and Laser-Capture Microdissection Charmaine Y. Pietersen, Maribel P. Lim, Laurel Macey, Tsung-Ung W. Woo, and Kai C. Sonntag Abstract The human brain is an exceptionally heterogeneous structure. In order to gain insight into the neurobiological basis of neural circuit disturbances in various neurologic or psychiatric diseases, it is often important to define the molecular cascades that are associated with these disturbances in a neuronal type-specific manner. This can be achieved by the use of laser microdissection, in combination with molecular techniques such as gene expression profiling. To identify neurons in human postmortem brain tissue, one can use the inherent properties of the neuron, such as pigmentation and morphology or its structural composition through immunohistochemistry (IHC). Here, we describe the isolation of homogeneous neuronal cells and high-quality RNA from human postmortem brain material using a combination of rapid IHC, Nissl staining, or simple morphology with Laser-Capture Microdissection (LCM) or Laser Microdissection (LMD). Key words: Laser-capture microdissection, Laser microdissection, Postmortem brain, Immunohisto­ chemistry, Nissl staining, Dopamine, Parvalbumin, GABAergic, Pyramidal, Neurons, Expression profiling

1. Introduction Gene or noncoding (nc) RNA expression profiling of neuronal cell populations from postmortem brains is a quickly expanding field in neuroscience, and recent developments have introduced laserassisted microdissection to individually isolate these cells. This cell-specific based analysis enables molecular fingerprinting, such as mRNA or microRNA expression profiling, without the confounding effects of surrounding cells and/or tissue structures (1). A requirement for laser microdissection on postmortem brain material is the reliable visualization of the desired neurons. Some of the methods that are commonly used to identify specific subsets of neurons include immunohistochemistry (IHC) and Graeme I. Murray (ed.), Laser Capture Microdissection: Methods and Protocols, Methods in Molecular Biology, vol. 755, DOI 10.1007/978-1-61779-163-5_28, © Springer Science+Business Media, LLC 2011

327

328

C.Y. Pietersen et al.

Nissl staining. These procedures are conventionally optimized for the purpose of preservation of morphological details for downstream neuroanatomical analyses and typically involve a number of aqueous processing steps during a rather prolonged time interval (up to 2–3 days). As a result, they inevitably lead to significant and often complete degradation of RNA and, hence, cannot be directly adapted for laser microdissection of neurons for gene expression studies. Expression profiling from laser-isolated cells requires the acquisition of high-quality mRNA or other ncRNA molecules – e.g., for a sample purity as expressed by an absorption ratio (A260– A280) between 1.8 and 2.5 and in the case of microarray hybridization, quantities in the microgram range, usually obtained after two rounds of T7-based linear amplification, are common practice (2). Here, we first describe three methods to identify neu-

Fig. 1. Examples of neurons before (Pre, left-hand side) and after (Post, right-hand side) laser microdissection (DA) or laser-capture microdissection (Pyr and GABA) from human postmortem brain tissue. DA dopaminergic neurons, Pyr pyramidal neurons, GABA parvalbumin-expressing, g-aminobutyric acid neuron.

28  Neuronal Type-Specific Gene Expression Profiling and Laser-Capture Microdissection

329

rons of interest from postmortem brains: (1) A quick Nissl staining protocol to detect pyramidal neurons by their characteristic morphology, (2) a rapid IHC protocol to visualize parvalbumin (PV)-containing GABAergic neurons, (3), and (3) the capture of naturally pigmented neuromelanin-containing dopamine neurons (4, 5) (Fig. 1). Then, we describe two currently available laser-assisted methods to isolate neurons in postmortem brains – Laser Capture Microdissection (LCM), which makes use of a low-intensity infrared laser beam to attach neurons on a matrix (6), and Laser Microdissection (LMD) that removes cells with a highintensity UV cutting laser and collects them by gravity (7). Finally, we provide information about the quality of pre- or postamplified RNA isolated from captured cells for downstream applications such as quantitative Real-Time (qRT)-PCR or microarray hybridization, respectively.

2. Materials 2.1. Tissue Preparation

1. RNaseZap® (Ambion®, TX). 2. Microslide box (VWR, PA). 3. Microm HM 505E cryostat (Thermo Scientific, MA). 4. Plain, uncharged glass slides for LCM (MDS Analytical Technologies, CA), or Leica Frame slides PET-membrane 1.4 mm for LMD.

2.2. Morphological Identification of Pyramidal Neurons with a Nissl Stain Procedure

1. HistoGene® LCM Frozen Section Nissl Staining Kit (MDS Analytical Technologies, CA). This kit contains not only the Nissl stain but also all the appropriate ethanols, xylene, and slide jars for the process. The slide jars can be reused after every 4–6 slides after removing RNases with RNaseZap® and wiping down with absolute ethanol. 2. Roche Protector RNase Inhibitor 40 U/ml (Roche, IN). This is added to most aqueous solutions as a precautionary method to prevent degradation of RNA through RNase activity. Store at −20°C and keep on ice while working with the product. 3. Liquid Blocker Super PAP pen (Daido Sangyo, Tokyo, Japan).

2.3. Rapid Immuno­ histo­chemistry Protocol for PV-Containing GABAergic Neuron Identification

1. 0.05 M TBS (Tris-buffered saline) solution pH 7.4 (Do not add a sodium azide preservative, as this interferes with downstream processing). 2. 0.05 M TBS, 0.2% Triton X-100 solution pH 7.4. 3. 0.05M TBS, 0.2% Triton X-100, 1% BSA [Bovine Serum Albumin, Probumin® (Millipore, IL)]. Aliquot 50 ml into a Falcon tube, stored at 4°C.

330

C.Y. Pietersen et al.

4. Acetone (Fischer Scientific, NJ). 5. 30% Hydrogen peroxide. Store at 4–8°C and only add to solutions just before needed (see Subheading 3). 6. Monoclonal anti-parvalbumin primary antibody (Mouse, Sigma-Aldrich®, MO). After rehydrating the antibody, make 20-ml aliquots and store at −20°C. 7. Peroxidase AffiniPure Donkey Anti-Mouse IgG secondary antibody (Jackson ImmunoResearch Laboratories, PA). Use same storage protocol as for primary antibody. 8. Roche Protector RNase Inhibitor 40 U/ml (Roche, IN). 9. NovaRED™ SK-4800 (Vector®, CA) stored at 4°C and protected from light. When using this solution, wear appropriate personal protection equipment, such as a mask. The rinsate from this solution should also be disposed of with care, as the toxicity and carcinogenicity are unknown. 10. Molecular Sieves (EMD™, NJ). These sieves are added to the second 100% ethanol, in the dehydration series, and are to be changed every 4–6 slides. 2.4. Laser (Capture) Microdissection: LMD or L(C)M

1. LEICA AS LMD apparatus fitted with 10, 20, and 40× magnification lenses used with Eppendorf thin-walled reaction tubes with domed cap (Fisher #951010022) for cell collection. 2. ArcturusXT™ LCM System (MDS Analytical Technologies, CA) fitted with 10, 20, and 40× magnification lenses. Be sure to install the LMD or LCM equipment in a humidity- and temperature-controlled environment, with humidity set below 40%, and kept at standard room temperature. 3. The ArcturusXT™ is used with CapSure® HS LCM Caps (MDS Analytical Technologies, CA), which should be kept in a dark place until needed, and GeneAmp® thin-walled reaction tube with domed cap (Applied Biosystems, CA). Only the product category N8010611 fits perfectly around the HS cap to prevent any leakage.

2.5. RNA Extraction and Downstream Processing of Neurons after LCM or LMD

1. mirVANATM miRNA Isolation Kit (Ambion, Austin, TX) or PicoPure® RNA Isolation Kit (MDS Analytical Technologies, CA), used with the RNase-Free DNase Set (Qiagen, CA). 2. RiboAmp®HSPLUS with Turbo Labeling™ Biotin (MDS Analytical Technologies, CA), used with SuperScript™ III Reverse Transcriptase (Invitrogen, CA). When performing this protocol, use 0.5-ml tubes correlating with a 0.5-ml plate in the thermal cycler. This will increase your mRNA yield significantly.

28  Neuronal Type-Specific Gene Expression Profiling and Laser-Capture Microdissection

331

3. Methods In this protocol, the LCM method includes a staining step, either Nissl or IHC, and is specifically tailored for nitrogen-vapor flashfrozen human postmortem brain tissue. However, either staining protocol can also be combined with LMD. The parameters such as cryostat temperature, section thickness, and LCM laser specifications will vary according to the type of tissue used and the fixation method and should, therefore, be adjusted accordingly. For the IHC protocol, we noted in previous time study experiments that RNA in aqueous solutions started to degrade after 1.5 h. For this reason, our IHC protocol should be completed within 1 h. We have also added RNase Inhibitors to all solutions in both staining protocols, to further prolong RNA viability. We, therefore, recommend that the LCM/LMD capture period should follow the staining protocol directly and should also not exceed 1 h. For our LMD method, no IHC is involved, since DA neurons are neuromelanin-positive and can readily be detected by light microscopy. In general, when working with RNA, it is important to eliminate any contaminating RNase activity, and surfaces should be cleaned with RNaseZap® and wiped down with absolute ethanol. Using plasticware that is nucleic acid-free and changing your gloves regularly (especially after touching areas that are not RNase-free) will help reduce the risk of RNase contamination. 3.1. Tissue Preparation

1. Remove possible RNase-contamination on glass slide with RNaseZap®, followed by wiping it down with absolute ethanol. 2. Remove the tissue block stored in the −80°C freezer and place it in a container with dry ice for transportation. Place a microslide box in the same container. Keep the samples continuously on dry ice. 3. Adjust the cryostat temperature to −17°C (see Note 1), install a new sectioning blade, and wipe the interior and sectioning blade with 100% ethanol in a perpendicular motion away from the razor edge. Set the slice thickness to 7–8 mm (see Note 2). Do not open cryostat cover more than halfway to avoid moisture from entering the cryostat. 4. After adhering the tissue block to the platform, let the block acclimate to the cryostat environment (temperature) for ~10–15 min. 5. Mount the tissue block to the specimen holder with as little “Optimum Cutting Temperature” (OCT) as possible, only touching the bottom of the specimen. Avoid covering

332

C.Y. Pietersen et al.

the specimen with OCT along the sides. Only sections without OCT can be mounted for LCM. 6. Place specimen holder with mounted tissue such that the face of the tissue block is aligned with the blade’s edge and the thinnest part of the tissue is cut first. Bring the stage/block closer to the blade – enough to better determine how to angle the block in relation to the blade. Adjust the block as necessary. Once it appears aligned from all angles, begin cutting at 7–8 mm until the tissue reaches the blade and it cuts through the tissue. Based on where/how the blade cuts (i.e., the angle of intersection with the tissue), adjust the block such that the blade will cut the surface of the tissue evenly to produce mountable sections for LCM/LMD. 7. Use a brush to wipe away frost on the blade and stage, and pieces of tissue that cannot be mounted. Always wipe brush away from the razor edge (never along or against it!). 8. After acquiring a suitable section, adhere it to the plain glass (LCM) or the Leica Frame Slide (LMD) toward the center of the slide at room temperature and place immediately thereafter inside the cryostat or into microslide box on dry ice (see Note 3). Do not allow the slide to dry at room temperature. 9. Acquire another suitably smooth section (see Note 4), adhere it also toward the center of the same slide and place slide into the microslide box on dry ice. If cutting more than one specimen, use a new disposable microtome blade for each one. In addition, wipe down knife holder and stage with 100% ETOH in between specimens to avoid cross-contamination. 10. Cut enough sections for your entire experiment. In our case, four sections (or two slides) per case were sufficient to capture between 300–500 neurons. When you are finished sectioning, place the microslide box containing the sections into the −80°C freezer until needed for staining and/or LCM or LMD. 3.2. Morphological Identification of Pyramidal Neurons with a Nissl Stain Procedure1

1. Prepare the ethanol dehydration series (2× 75%, 2× nucleasefree water, 95%, 100%), adding 25 ml of each ethanol into the appropriate jars. 2. Add molecular sieves to the 100% ethanol jar and the xylene jar under the fume hood. 3. Place the first 75% ethanol jar into the −20°C freezer and all other jars on ice in an ice bucket.

 Protocol originally published in JoVE: Pietersen CY, Lim MP, Woo TUW (2009). Obtaining High Quality RNA from Single Cell Populations in Human Postmortem Brain Tissue. JoVE. http://www.jove.com/index/ details.stp?id=1444, doi: 10.3791/1444. 1

28  Neuronal Type-Specific Gene Expression Profiling and Laser-Capture Microdissection

333

4. Add 4  ml of RNase inhibitor to 400  ml of the HistoGene® stain in a 1.5-ml microcentrifuge tube (RNase-free) and place the tube on ice until needed. 5. Switch on the water bath and set to 42°C and place a rack containing a 50 ml Falcon tube in the water bath. 6. Remove two slides with two sections each from the −80°C freezer and defrost on a Kimwipe for 30 s or until the corners of the slides have defrosted. 7. With forceps, transfer the slides into the 75% ethanol jar that was originally placed in the −20°C freezer for 30 s. 8. After 30  s in the second jar containing nuclease-free water, place the slides on a Kimwipe. 9. Circle the sections on the slide with a PAP pen (see Note 5) and aliquot 90 ml of the HistoGene® stain with RNase inhibitor per section. Incubate at room temperature for 20 s. 10. Proceed through the dehydration series (nuclease-free water, 75%, 95%) until the final 100% ethanol dehydration step, leaving the slides in the final 100% ethanol jar containing the molecular sieves for 3 min (see Note 6). 11. Proceed to the fume hood and place the slides in the xylene jar containing the molecular sieves for 5 min, thereafter airdrying under the fume hood for 5 min (see Note 7). Proceed immediately with LCM, capturing approximately 500 cells. 3.3. Rapid Immuno­ histo­chemistry Protocol for PV-Containing GABAergic Neuron Identification

1. Prepare solutions necessary for protocol beforehand: 0.05 M TBS, 0.05 M TBS with 0.2% Triton X-100 solution, 0.05 M TBS with 0.2% Triton X-100 and BSA solution. 2. RNaseZap® and rinse with nuclease-free water all containers, including wash bottles and slide jars that will contain solutions. 3. Fill one wash bottle with TBS solution and another with nuclease-free water. Fill a zapped slide jar with the Triton X solution (refresh this solution after each use). 4. In a 1.5-ml microcentrifuge tube, aliquot 990 ml TBS solution. Cap and leave on the bench at room temperature. 5. In two 0.5-ml PCR tubes, aliquot 174 ml of the TBS/Triton X-100/BSA solution in each. Defrost one aliquot of both the primary and secondary antibodies and add 20 ml of the primary antibody to one 0.5-ml PCR tube and 20 ml of the secondary to the other 0.5-ml PCR tube containing the TBS/Triton X-100/ BSA solution. Place these two PCR tubes on 4°C until needed. 6. Add 5 ml of nuclease-free water to a new 50-ml Falcon tube and place in rack on lab bench. 7. Prepare the ethanol dehydration series (50%, 75%, 95%, 2× 100%), adding 25  ml of each ethanol into the appropriate jars. Add molecular sieves to the second 100% ethanol jar and a glass slide jar filled with xylene under the fume hood.

334

C.Y. Pietersen et al.

8. Switch on the water bath set to 42°C and place a rack containing another 50-ml Falcon tube in the water bath. 9. Remove two glass slides with two sections each from the −80°C freezer and defrost on a Kimwipe for 30 s or until the corners of the slides have defrosted. 10. Using forceps, place the slightly defrosted slides into a glass dish filled 3/4 with acetone (or just enough to cover the slides). Cover the dish and incubate for 4 min to fix the tissue. This step should take place under a fume hood. 11. During this incubation time, prepare the endogenous enzyme block solution by adding 10 ml 30% hydrogen peroxide to the previously aliquotted TBS solution in the 1.5-ml microcentrifuge tube immediately before needed. 12. Remove slides and rinse with TBS from the wash bottle, being careful not to place a direct stream onto the sections. Thereafter, rinse with TBS/Triton X-100 solution by dunking slide into previously prepared slide jar a few times. Rinse again with TBS from wash bottle. Repeat for the other slide. 13. With a Kimwipe, remove excess liquid around the sections on the slide, being careful not to touch the section itself. The excess liquid could cause dilution of the antibody in subsequent steps. Place slides on an even surface, e.g., Kimwipe or the edge of an upside-down pipette tip container lid. 14. Pipette endogenous enzyme block onto slides – just enough to cover sections sufficiently (see Note 8) – and incubate for 5 min. 15. During this incubation period, add 6 ml of RNase inhibitor to the primary antibody aliquot and TBS/Triton X/BSA solution in the 0.5-ml PCR tube from the fridge, just prior to use. 16. Rinse the slides gently with nuclease-free water from wash bottle, again not aiming the stream directly onto the sections themselves. Wipe off excess liquid as before and place on leveled surface. 17. Pipette 95 ml of the primary antibody solution onto the beginning of the slide. Gently place a glass coverslip, with its edge touching the antibody solution on the slide. Slowly lower the coverslip onto the slide, distributing the liquid evenly over both sections on the slide, and avoid any bubble formation. Repeat for the other slide and incubate for 7 min. 18. During this incubation time, prepare the secondary antibody by adding 6 ml of RNase Inhibitor to the antibody aliquot and TBS/Triton X-100/BSA solution in the 0.5-ml PCR tube previously stored in the fridge. 19. After 7 min, rinse the slides as described in step 12 and remove excess liquid (see Note 9).

28  Neuronal Type-Specific Gene Expression Profiling and Laser-Capture Microdissection

335

20. Apply the secondary antibody following the method described for the primary antibody in step 17 and incubate for 7 min. 21. During this incubation period, prepare the substrate chromogen (NovaRED) by following the instructions from the kit: Add three drops of reagent 1 to the previously prepared 5  ml nuclease-free water in the Falcon tube and mix, add two drops of reagent 2 and mix, add two drops of reagent 3 and mix. Only add hydrogen peroxide (two drops) to solution just prior to use for maximum results. 22. Place aluminum foil into a disused pipette container lid and curve aluminum foil to form sides (will look like a dish). Place Kimwipes in the middle of the “dish” and wet slightly with nuclease-free water. 23. Rinse slides as noted in step 12 and remove excess liquid. Place slides into the aluminum foil “dish” and lay down as flat as possible. 24. Pipette chromagen substrate solution onto slides, making sure to cover the sections. Use more than necessary to make sure that the sections do not dry out. Incubate for 12 min. 25. Gently rinse slides with the wash bottle containing nucleasefree water, again not directing the flow directly onto the sections, while collecting the rinsate into the aluminum foil “dish” and dispose appropriately. 26. Proceed with the dehydration series, by incubating the slides in each ethanol solution for 30 s beginning at 50%, and proceeding through to 100% ethanol. Leave the slides in the final 100% ethanol jar containing the molecular sieves for 3  min (see Note 10). 27. Proceed to the fume hood and place the slides in the xylene jar containing the molecular sieves for 2  min. Air-dry on a Kimwipe under the fume hood for 5 min. Proceed immediately with LCM, capturing approximately 350 cells. 3.4. Preparation of Tissue Slides for LMD of Dopamine Neurons

Since human midbrain dopaminergic neurons are neuromelanin positive, they are readily detectable in light microscopy (Fig. 1). Therefore, no additional staining steps are required. Prior to LMD the sectioned slides (Subheading  3.1) are dehydrated as described in Subheading 3.2, steps 10 and 11.

3.5. Laser Capture Microdissection of Pyramidal and PV GABAergic Neurons (see Footnote 1)

1. Load the slides and caps onto the Arcturus XT™ apparatus. Use the CapSure™ HS caps, but keep the program setting on Macro (see Note 11). Click on the box “Load with overview” to obtain an overview photo of each slide. 2. Adjust the brightness/focus at 2× magnification, to determine which section would be optimal for laser capture. Avoid tissue

336

C.Y. Pietersen et al.

sections with excessive folding, but rather choose sections that are intact, smooth, and stained well, especially near the region of interest. 3. Place a cap over the general area where capturing. Still at the 2× magnification, make sure that the cap rails do not rest on any folds, as this will tilt the cap. 4. Next, confirm the location of the IR laser spot manually at the 40× magnification. If the spot (red beam) does not correlate with the center of the blue cross, this can be adjusted by rightclicking on the spot and selecting “located IR spot”. 5. Save the position of the cap by clicking on the plus sign at the position function. This way, if for any reason the cap must be removed and put back on the same location, the cap will always return to precisely the same spot that it was adjusted to before. 6. Enter these values into the control box: 70 into power and 16 into duration. Unclick the “auto move stage” option and make sure that the right size symbol correlates with the symbol on the panel to the right. 7. Select the capture circle option (bottom right) to select a pyramidal/PV GABAergic neuron that you want to test capture and then activate the laser by clicking on “test IR spot”. 8. The spot made by the laser should have a crisp dark ring around the object captured. If this ring is too light, the cell was not captured. If there is a dark spot in the middle of the dark ring, then the laser strength/duration is too great, and it might have a negative effect on the RNA present in the captured cell. Make sure that the ring is big enough to encompass the cell, but small enough that it does not include unwanted tissue or other cells (see Note 12). 9. Repeat the process on different parts of the tissue within the layer that you wish to capture, to check that the spot size does not differ depending on the location. Adjust accordingly. 10. Once the laser spot has been adjusted, identify pyramidal/PV GABAergic neurons for capture (see Note 13). We identify approximately 500/350 cells respectively per sample, which results in approximately 1–25 ng of total RNA per sample (see Note 14). 11. Once all the cells have been captured, move the cap to a different part of the slide that does not contain a section. Go back to where the cells were captured and make sure that at least 80% of the neurons were removed (see Note 15). If not, do not try to recapture the same cells, but locate the area where most of the cells were captured and capture more cells in the same area. Move the cap to the QC station.

28  Neuronal Type-Specific Gene Expression Profiling and Laser-Capture Microdissection

337

12. Place the cap into the 0.5 ml-microcentrifuge tube from Applied Biosystems containing 50  ml of extraction buffer (PicoPure® RNA isolation kit). The cap has been designed to fit perfectly into this specific tube to prevent buffer from leaking. 13. Turn the assembly upside-down, making sure that extraction buffer covers the entire cap, and place it at the bottom of the 50-ml Falcon tube already present in the water bath set at 42°C. The neurons are incubated for 30 min to remove the tissue from the cap. 14. Thereafter, centrifuge the tube and cap assembly for 2 min at 800 × g. Remove the cap and store the remaining cell extract at −80°C until RNA extraction. As an extra precaution, you can reexamine the cap at the QC station on the laser-capturing apparatus to ensure that all neurons have been removed from the cap itself. 3.6. LMD of Dopamine Neurons

1. LMD is performed according to the manufacturer’s instructions. We use a LEICA AS LMD apparatus with manual or automated stage. 2. After setting up the hardware and software of the instrument place the FrameSlides with the tissue facing down and insert an Eppendorf thin-walled reaction tube with the domed cap in the tray and secure the tube in position. 3. Select the requested tube position and program, calibrate the laser, and adjust the laser to “line” setting to outline each individual cell (we capture at 40× magnification). Then, complete the capture in standard mode. While capturing use the “Move and Cut” mode to remove cells that are caught on the static of the slide. Click on the cell until it appears to have fallen (see Note 16). 4. After capturing click “no cap”. Unload and remove the tray and the slide, with the tube still in the tray, add 50 ml lysis buffer (Lysis/Binding buffer mirVana miRNA Ambion Isolation Kit) to the cap (see Note 17). Close the tube and remove from the tray; store the tube at −80°C.

3.7. RNA Extraction and Downstream Processing of Neurons after LCM or LMD

1. RNA of LCM Pyramidal/PVGABAergic neurons was isolated using the PicoPure® kit and LMD DA neurons with the mirVANATM miRNA Isolation Kit. During RNA isolation with the PicoPure® kit, follow the appendix protocol for DNase treatment of the sample, as the elimination of genomic DNA is critical for accurate downstream applications such as qRT-PCR. 2. Amplify the mRNA (two rounds) obtained from pyramidal/ GABAergic neurons with the RiboAmp®HSPLUS kit and

338

C.Y. Pietersen et al.

provided protocol. During the second strand cDNA synthesis, our lab has made one slight modification to the manufacturer’s protocol. When the sample has been transferred from the tube to the purification column, centrifuge the tube originally containing the sample and transfer the extra microliter obtained after centrifugation into the purification column. Repeat this process during the second round of amplification (see Note 18). 3. To check the (m)RNA quality, confirm the concentration and purity (A260–A280 ratio, Table 1) with the NanoDrop spectrophotometer. As the amplified mRNA from the LCM neurons was used for microarray analysis, you should run an Experion StdSens LabChip® to ensure that the transcript length exceeds 600 nucleotides required for microarray hybridization. 4. RNA obtained from LMD material was used for microarrays (4, 5, 8) and miRNA profiling with a TaqMan® qRT-PCR based on the TaqMan® Human MicroRNA A Arrays (Applied Biosystems, CA) (Sonntag, unpublished). 3.8. Downstream Applications

After viable RNA is achieved with any of the above methodologies, it can be used to evaluate the expression profiles of cellular populations via a variety of technologies available today. These include microarray technology to determine the complete gene expression profile of the cell type of interest, or qRT-PCR to investigate or validate specific genes. For example, we have used the resulting RNA from the LMD protocol for microarray profiling of postmortem dopaminergic neurons in Parkinson’s disease subjects (4, 5), and this method was also used in combination with the Nissl staining technology to determine the expression profile of hippocampal GABAergic neurons (8). We have also analyzed the microarray data obtained from the pyramidal neurons in postmortem schizophrenia subjects and controls to discover altered underlying molecular pathways in this patient population (Pietersen et al., unpublished data). Another recent application is the characterization of noncoding RNAs, such as microRNAs (miRNAs), which play an important role in regulating many aspects of gene transcription and translation (9). These molecules can be detected by sensitive highthroughput screening technologies, such as micro- or nanoarrays and qRT-PCR, which require only small amounts of RNA in the nanogram or picogram range, respectively. As miRNAs are thought to play a profound role in the fine-tuning of neuronal cell function (10, 11), the above described methodologies can provide material from single cells or a homogeneous cell population to gain insight into their molecular properties in health and disease (1).

28  Neuronal Type-Specific Gene Expression Profiling and Laser-Capture Microdissection

339

Table 1 A summary of subjects and RNA results of each neuron-type, matched for age, gender, and postmortem interval (PMI) Cell Amount of number A260/A280 RNA (mg)

ID

Diagnosis

Age Gender PMI

Assaycase

PYR1

Control

22

M

21.47 530

2.37

0.004 (69.3) Microarray/ qRT-PCR

PYR2

Control

75

M

20.25 520

2.44

0.014 (54.3) Microarray/ qRT-PCR

PYR3

Control

58

F

21.08 680

2.39

0.007 (50.1) Microarray/ qRT-PCR

PYR4

Control

71

F

20.50 530

2.47

0.012 (39.6) Microarray/ qRT-PCR

PYR5

Schizophrenia

36

M

17.97 530

2.44

0.001 (62.1) Microarray/ qRT-PCR

PYR6

Schizophrenia

62

M

10.75 630

2.34

0.002 (75.9) Microarray/ qRT-PCR

PYR7

Schizophrenia

67

F

21.80 540

2.52

0.002 (62.1) Microarray/ qRT-PCR

PYR8

Schizophrenia

88

F

13.33 550

2.47

0.002 (52.2) Microarray/ qRT-PCR

GABA1 Control

61

M

17.00 312

2.46

(47.1)

Microarray/ qRT-PCR

GABA2 Control

22

M

21.47 295

2.44

(47.4)

Microarray/ qRT-PCR

GABA3 Control

90

F

12.66 325

2.47

(26.1)

Microarray/ qRT-PCR

GABA4 Control

79

F

15.00 350

2.56

(31.2)

Microarray/ qRT-PCR

GABA5 Schizophrenia

56

M

21.83 712

2.48

(22.8)

Microarray/ qRT-PCR

GABA6 Schizophrenia

36

M

17.97 325

2.52

(24.9)

Microarray/ qRT-PCR

GABA7 Schizophrenia

93

F

  6.92 359

2.59

(25.8)

Microarray/ qRT-PCR

GABA8 Schizophrenia

55

F

22.00 400

2.53

(24.3)

Microarray/ qRT-PCR

DA1

Control

79

M

20.92 360

1.69

0.17

Microarray/ qRT-PCR

DA2

Control

78

M

21.75 390

1.73

0.22

Microarray/ qRT-PCR (continued)

340

C.Y. Pietersen et al.

Table 1 (continued) Cell Amount of number A260/A280 RNA (mg)

ID

Diagnosis

Age Gender PMI

Assaycase

DA3

Control

72

F

18.25 300

1.58

0.19

Microarray/ qRT-PCR

DA4

Control

74

F

12.17 300

1.73

0.18

Microarray/ qRT-PCR

DA5

PD

79

M

23.42 350

1.73

0.51

Microarray/ qRT-PCR

DA6

PD

72

M

26.25 330

1.78

0.76

Microarray/ qRT-PCR

DA7

PD

81

F

22.75 300

1.44

0.27

Microarray/ qRT-PCR

DA8

PD

73

F

20.97 1300

1.82

0.68

Microarray/ qRT-PCR

In all three neuron-types, four representative cases per treatment group or controls are listed. Assay represents the downstream processing used to analyze gene expression. The A260/A280 column contains sample purity ratios for postamplified mRNA for pyramidal and GABAergic neurons, but preamplified total RNA for the dopamine neurons. In the RNA amounts column, the numbers in brackets represent total amount of postamplified mRNA. No measurable preamplified RNA was detected for the GABA samples. DA dopaminergic neurons, GABA parvalbumin-expressing, g-aminobutyric acid neuron, PD Parkinson disease, Pyr pyramidal neurons

4. Notes 1. As temperature can also influence tissue smoothness, try to keep the temperature above −20°C. Temperatures lower than this can result in tissue cracking. 2. You must be able to visualize the structure you wish to capture after staining. In our case, neurons are identified via a stain. If you cut the sections too thin, you might not be able to adequately visualize these cells. If the section is too thick, cells designated for capture will not fully adhere to the cap, and part of the cell will remain on the slide, thereby decreasing RNA yield. 3. The tissue moves between different temperature states during sectioning between the dry ice (approximately −78°C), cryostat, room temperature, and eventually the −80°C freezer environments. In order to reduce the temperature gradient, do not raise the temperature of the cryostat to more than −15°C, as it could affect RNA integrity. 4. You must be able to cut the tissue into smooth sections without folding. In order to accurately capture the designated

28  Neuronal Type-Specific Gene Expression Profiling and Laser-Capture Microdissection

341

area, smooth sections are key. Too thick or too thin sections could result in folding or tearing, respectively. 5. The PAP pen helps to concentrate the staining solution onto the section. However, with the plain glass slides, it is not always necessary to use it, as the staining solution appears to cling only to the section itself. 6. The molecular sieves draw out any excess moisture that might be present in the absolute ethanol and xylene solutions. This helps to dehydrate the sections completely, which is a necessary component for LCM or LMD. 7. Check that the sections are completely dry before proceeding with LCM. If you are still not able to capture cells after dehydration and air-drying, repeat the absolute ethanol, xylene, and air-drying steps to ensure proper dehydration. 8. We have opted not to use a humid chamber during the IHC protocol, as it is not necessary for such short incubation times. However, whenever applying a solution without a coverslip onto the sections, apply enough to ensure that the sections do not dry out during the incubation time. 9. The glass coverslips should come off while rinsing. Do not manually try to remove the coverslips, as this will cause some of the tissue section to scrape off. Adding enough liquid between the slide and coverslip, either primary or secondary antibody, will ensure a smooth coverslip removal. 10. We extended the duration of the final dehydration of the sections in 100% ethanol from 30 s to 3 min, and also added molecular sieves to the jar to ensure sufficient dehydration of the tissue, to obtain adequate tissue lift and to preserve RNA integrity. Molecular sieves are also added to the xylene solution for this reason. 11. As the HS cap is made for capturing single cells, it only has a small area designated for capture, which can limit the section area from which you would like to capture. To alleviate this issue, keep the settings on the ArcturusXT™ software on Macro, while using the HS cap, to capture cells from a larger section area. 12. The spot size should be specific to your cell size, i.e., not too big that it is nonspecific and small enough to be as specific as possible. At the same time, the laser strength should still be strong enough to be able to capture the cell. Conversely, if the laser strength is too high, the spot size will enlarge and might not be specific to the cell of interest. 13. Do not include cells that have other cell types directly next to them, as these may also be captured and you will no longer have a homogeneous cell population. Also, try to limit the

342

C.Y. Pietersen et al.

amount of surrounding tissue you capture with the cell of interest by decreasing the spot size, while still ensuring adequate capture of that cell. 14. We also usually only use one section to reduce the amount of capture time, as the cap has to be readjusted for each section. Although, if you have not succeeded in capturing all cells in the allotted time, you can repeat the stain and capture from different sections from the same case. 15. Sometimes, during humid weather or particularly aqueous procedures, such as IHC, the tissue is not completely dry. In this case, and if additional dehydration does not help, use the Macro cap as it is in direct contact with the tissue and, therefore, adheres to the cells more easily. Keep in mind, however, that this cap often picks up unwanted tissue “debris” for the same reason. In this case, use the prep strips that are provided, which remove the debris from the surface of the section prior to microdissection. 16. We found that larger pieces of tissue are more likely to cling to the slide via static electricity, so avoid capturing more than one cell at a time. When cells do reattach to the slide, they are most easily removed by directing the laser toward the center of the cell. 17. We found that lysis buffer dries out quickly, so we recommend adding the buffer after capturing. 18. We recommend performing the amplification procedure using the 0.5-ml tubes provided, and subsequently, performing the thermal cycling with a corresponding 0.5-ml block to increase the RNA yield.

Acknowledgments This work was supported by NIH grants MH080272 and MH076060 (Woo) and R21NS067335 (Sonntag). References 1. Sonntag, K. C. (2010) miRNAs and deregulated gene expression networks in neurodegeneration. Brain Research 1338, 48–57. 2. McClain, K. L., Cai, Y. H., Hicks, J., Peterson, L. E., Yan, X. T., Che, S., and Ginsberg, S. D. (2005) Expression profiling using human tissues in combination with RNA amplification and microarray analysis: assessment of Langerhans cell histiocytosis Amino Acids 28, 279–90.

3. Pietersen, C. Y., Lim, M. P., and Woo, T. U. (2009) Obtaining high quality RNA from single cell populations in human postmortem brain tissue J Vis Exp. 6;pii: 1444. 4. Simunovic, F., Yi, M., Wang, Y., Macey, L., Brown, L. T., Krichevsky, A. M., Andersen, S. L., Stephens, R. M., Benes, F. M., and Sonntag, K. C. (2009) Gene expression profiling of substantia nigra dopamine neurons: further insights into Parkinson’s disease pathology. Brain 132, 1795–809.

28  Neuronal Type-Specific Gene Expression Profiling and Laser-Capture Microdissection 5. Simunovic, F., Yi, M., Wang, Y., Stephens, R., and Sonntag, K. C. (2010) Evidence for gender-specific transcriptional profiles of nigral dopamine neurons in Parkinson disease. PLoS One 5, e8856. 6. Charboneau, L., Paweletz, C. P., and Liotta, L. A. (2001) Laser capture microdissection Curr Protoc Cell Biol Chapter 2, Unit 2 5. 7. Edwards, R. A. (2007) Laser capture microdissection of mammalian tissue. J Vis Exp, 309. 8. Benes, F. M., Lim, B., Matzilevich, D., Walsh, J. P., Subburaju, S., and Minns, M. (2007)

343

Regulation of the GABA cell phenotype in ­hippocampus of schizophrenics and bipolars. Proc Natl Acad Sci U S A 104, 10164–9. 9. Bartel, D. P. (2009) MicroRNAs: target recognition and regulatory functions Cell 136, 215–33. 10. Barbato, C., Giorgi, C., Catalanotto, C., and Cogoni, C. (2008) Thinking about RNA? MicroRNAs in the brain. Mamm Genome 19, 541–51. 11. Schratt, G. Fine-tuning neural gene expression with microRNAs (2009) Curr Opin Neurobiol 19, 213–9.

wwwwwww

Chapter 29 Region-Specific In Situ Hybridization-Guided Laser-Capture Microdissection on Postmortem Human Brain Tissue Coupled with Gene Expression Quantification René Bernard, Sharon Burke, and Ilan A. Kerman Abstract This chapter describes the procedure of in situ hybridization-guided laser-capture microdissection performed on postmortem human brain tissue. This procedure permits the precise collection of brain tissue within anatomically defined brain nuclei that is enriched with mRNA. The chapter emphasizes the specific handling of postmortem tissue and preservation of RNA integrity to ensure high-quality gene profiling. Downstream procedures including mRNA amplification, gene profiling using high-density microarray chips, and confirmation with quantitative real-time polymerase chain reaction (qPCR) are described. PCR primer design and cDNA quantification required for qPCR are delineated. Key words: In situ hybridization, Postmortem, Microdissection, Laser capture, Gene expression, PCR, Microarray, Brain, Nucleus

1. Introduction Since the introduction of chip-based microarrays, gene expression profiling of human postmortem tissue has risen in popularity in the search for genetic underpinnings of many neuropsychiatric illnesses. However, studying the human brain is inherently difficult due to its heterogeneity and size. Different brain regions exhibit high levels of structural and functional differences: therefore, valid gene expression studies must possess high levels of anatomical precision. In addition, manner of death as well as postmortem delay can negatively impact mRNA integrity significantly altering gene expression measurements. In fact, these factors can alter gene expression profiles to such an extent so as to overwhelm the impact of disease on such profiles. Graeme I. Murray (ed.), Laser Capture Microdissection: Methods and Protocols, Methods in Molecular Biology, vol. 755, DOI 10.1007/978-1-61779-163-5_29, © Springer Science+Business Media, LLC 2011

345

346

R. Bernard et al.

Fig. 1. (a) Pseudocolored ISH image of a coronal brainstem section with TH mRNA (red in online version) enriched in the LC and TPH2 mRNA (green in online version) enriched in the nuclei MR and DR. (b) Klüver–Barrera stain of adjacent section displaying corresponding anatomical landmarks such as fiber tracts and nuclei. (c) Adjacent section to (b) on which LCM was performed to selectively remove LC tissue. Overlayed black dots represent TH mRNA signal from Fig. 1a to illustrate the precision and selectivity of the laser capture. (d) Adjacent section to (c) on which LCM was performed to selectively remove MR and DR tissue. Overlayed black dots represent TPH mRNA signal from Fig. 1a to illustrate the precision and selectivity of the laser capture.

For successful gene profiling, it is essential to limit the extracted tissue just to the anatomical boundaries of the area of interest and to obtain mRNA of sufficient quality that is compatible with downstream gene expression assays. Laser-capture microdissection (LCM) permits precise acquisition of discrete cell populations, such as samples from entire brain nuclei. We developed radioactive in situ hybridization (ISH) labeling of nuclei-specific markers as a guide for LCM (Fig.  1), which yields sufficient quality mRNA. In this chapter, we describe in detail the processing of human postmortem brain tissue using ISH-guided LCM for subsequent gene expression profiling using microarray and quantitative real-time polymerase chain reaction (qPCR). Application of this method has been recently validated (1, 2). 1.1. RNA Quality

RNA integrity is the single most important factor for successful gene expression profiling in postmortem tissue. There are three parameters that contribute significantly to RNA quality: postmortem

29  Region-Specific In Situ Hybridization-Guided Laser-Capture Microdissection¼

347

interval, brain pH, and the agonal state of the decedent in the final hours before death (3–5). The postmortem interval, which is defined as the time between death and postautopsy preservation of the brain, should not exceed 36  h. Brain tissue pH can be assessed with a calibrated pH meter and should be verified in multiple samples from different brain regions. The cut-off for acceptable pH is 6.6. Brain pH and agonal state are correlated parameters. Individuals who suffered from a prolonged agonal state, which includes respiratory arrest, multiorgan failure, or coma, tended to have lower brain pH values; whereas those with brief deaths, caused by accidents, cardiac events, or asphyxia, most often present a normal brain pH. Human brain cells principally depend on continuous supply of nutrition and oxygen for their energy metabolism. It is known that human brain cells are vulnerable when oxygen supply is reduced. Under such prolonged hypoxic conditions, ordinary transcriptional machinery is reprogrammed to divert the cellular resources to stress response, apoptosis, and inflammation control. Especially, anaerobic metabolic pathways, such as lactic acid cycle, are activated, resulting in an enhanced production of acid equivalents, and contribute to intra- and intercellular acidosis (6, 7). Therefore, the agonal state of the individual included in gene expression profiling studies should be assessed through determination of cause of death and detailed reconstruction of the final hours before death. Processing brain tissue for LCM, especially at room temperature, accelerates the RNA degradation. All efforts must be taken to minimize processing time at room temperature and to limit the activity of RNases. For example, we do not recommend serial processing to laser capture more than one brain region per section. Our data indicate that the second dissected area is of lower RNA quality than the first nucleus within the same section. Because of the continued decline of RNA integrity, RNA quality needs to be assessed while processing. We recommend checking RNA quality after RNA tissue extraction before proceeding with amplification. Resulting electropherograms (Fig. 2) can be assessed in different ways: 28S/18S peak ratio, RNA integrity number (RIN), and area under the curve (AUC) assessments, which are described by Schoor et  al (8). Postprocessing the gene expression data from the microarray chip deliver data that highly correlate with RNA quality: MAS5 present call (Fig.  3); 3¢/5¢ GAPDH ratio; and slopes of RNA digestion plots (3). In Table 1, we list those measures for three nuclei processed as described in this chapter. Figure 3 displays correlation analyses between different RNA integrity measures and resulting gene expression microarray detectability. The results demonstrate that RIN and Schoor ratio are significantly correlated with microarray present call numbers; whereas 18S/28S ratio is not.

348

R. Bernard et al.

Fluorescence signal

18S 30

28S 20 marker

10

pre-18S

0 20

30

40

50

60

Time (sec)

Fig. 2. Representative electrophoretic trace of a postmortem RNA sample extracted with LCM. Pre-18S, 18S, and 28S peak areas are important denominators in the assessment of RNA integrity.

2. Materials 2.1. Brain Preparation, Sectioning, and Dehydration Solutions

1. Cryostat (Leica, Wetzlar, Germany). 2. Superfrost Plus slides (Fisher Scientific, Pittsburgh, PA). 3. Stainless steel slide holding racks (for 20 slides; 75 mm long). 4. Diethylpyrocarbonate (DEPC). 5. 100% ethanol. 6. Xylene. 7. Fine-tooth saw. 8. pH meter (Corning, Cypress, CA).

2.2. LCM, RNA Isolation, and RNA Extraction

1. Autopix or Arcturus XT LCM machine (Molecular Devices, Sunnyvale, CA). 2. PicoPure RNA isolation system (Molecular Devices). 3. CapSure LCM macro caps (Molecular Devices). 4. RNAse-free DNase Set (Qiagen Valencia, CA).

2.3. RNA Quality Assessment

1. Bioanalyzer 2100 (Agilent, Santa Clara, CA).

2.4. RNA Amplification

1. Arcturus RiboAmp Plus 1.5 round kit (Molecular Devices).

2. RNA PicoGreen LabChip Assay (Agilent).

2. RNeasy mini clean-up kit (Qiagen). 2.5. Biotinylation and Hybridization for Microarray Chips

1. BioArray High Yield RNA Transcription Labeling Kit (T7) (Enzo, Farmingdale NY). 2. NanoDrop 2000 (Thermo Scientific, Wilmington, DE). 3. Affymetrix (Affymetrix, Santa Clara, CA) or Illumina (Illumina, San Diego, CA) GeneChip microarray system.

29  Region-Specific In Situ Hybridization-Guided Laser-Capture Microdissection¼

349

MAS5 Present Call

60

50

40

30 0.0

2.5

5.0

7.5

10.0

12.5

Schoor Ratio

MAS5 Present Call

60

50

40

30

0

1

2 3 4 5 6 RNA Integrity Number(RIN)

7

MAS5 Present Call

60

50

40

30 0.00

0.25

0.50 0.75 1.00 18S/28S peak ratio

1.25

Fig. 3. Correlation of RNA integrity measures obtained from Bioanalyzer electropherograms with present call detection rates from corresponding gene expression microarray chips (N = 21–23). Schoor ratio (pre-18S/28S) and RNA integrity number show significant (p < 0.05) correlations (r 2 = 0.28 and 0.33, respectively), whereas 18S/28S peak ratio did not correlate (p = 0.50; r 2 = 0.02) with MAS5 present call rates (%).

2.6. Polymerase Chain Reaction

1. Quant-IT PicoGreen dsDNA assay (Invitrogen, Carlsbad, CA). 2. CytoFluor microplate reader (Applied Biosystems, Foster City, CA). 3. iQ SYBR Green Supermix (Bio-Rad, Hercules, CA).

5.76 ± 0.31

3.8 ± 1.6

7.7 ± 1.4

Dorsal raphe

Median raphe

45.3 ± 7.5

44.3 ± 7.2

47.4 ± 11.9

9.4 ± 2.4

11.2 ± 3.8

10.8 ± 2.6

28S/total AUC

0.61 ± 0.15

0.72 ± 0.11

0.64 ± 0.19

28S/18S peak ratio

38.30 ± 6.37

49.12 ± 4.03

46.12 ± 2.62

% present call rate MAS5CALLS algorithm

1.46 ± 0.03

1.37 ± 0.03

1.51 ± 0.05

GAPDH 3¢/5¢ ratio

9.51 ± 0.56

10.85 ± 0.50

10.60 ± 0.14

RNA degradation slopes

Note that all RNA quality measures are concentration independent and that MR values are generally worse than LC and MR which is probably due to longer exposure at room temperature during LCM procedure

4.51 ± 0.91

4.58 ± 0.51

RNA integrity Pre-18S/total number (RIN) AUC

Locus coeruleus 6.6 ± 2.5

Isolated RNA conc. (ng/ml)

Schoor et al. values

Table 1 RNA quality assessments obtained from electrophoretic traces from extracted RNA and indirect RNA quality measures derived from gene expression microarray chip analysis (N = 10/nucleus; except for DR: N = 9)

350 R. Bernard et al.

29  Region-Specific In Situ Hybridization-Guided Laser-Capture Microdissection¼

351

4. iCycler system (Bio-Rad) with iQ Real-Time PCR Detection System (Bio-Rad). 5. 96-Well 0.2-ml thin-wall PCR plates (Bio-Rad). 6. Optical-quality sealing tape (Bio-Rad).

3. Methods 3.1. Tissue Preparation and In Situ Hybridization-Guided LCM

1. Frozen blocks of human brain tissue (0.75–1 cm thick) are placed on dry ice and dissected using a fine-tooth saw to generate tissue samples for subsequent cryostat sectioning. Such tissue blocks should be approximately 4 × 3 cm in size and stored immediately at −80°C. 2. Prepared tissue blocks are cut on a cryostat into 10-mm thick sections at −20°C and immediately thaw mounted onto a glass slides (one section per slide), and then stored at −80°C (see Note 1). 3. A section (adjacent to the one to be processed for LCM) needs to be treated to mark the area of interest. Several techniques can be applied (9, 10). We use digitized autoradiogram images from sections that underwent 35S-radiolabeled ISH to mark RNA of interest. Details on this method with regard to processing postmortem sections can be found here (11). The ISH processed section can then be stained, using a Luxol Fast Blue method to reveal anatomical and histological landmarks. Histostained images complement the ones with the specific mRNA marker and are used as an orientation guide enhancing the accuracy of the microdissection (Fig. 1). It is not advisable to stain the section to be processed for LCM because it compromises RNA integrity (12). 4. Dehydration of slide mounted section as preparation for LCM (see Note 2). (a) Remove from −80°C storage and leave at RT, 30 s (b) Place in 75% ethanol, 30 s (c) Place in 75% ethanol, 30 s (d) Place in 95% ethanol, 30 s (e) Place in 100% ethanol, 30 s (f) Place in 100% ethanol, 30 s (g) Place in xylene, 5 min (h) Place in xylene, 5 min (i) Air dry under hood, 20 min Use solution for a maximum of five times, change container after two sets of 20 slides have been processed.

352

R. Bernard et al.

5. The sections can now be processed for LCM. Depending on the manufacturer, these steps may vary. We use the AutoPix instrument equipped with an inverted microscope system using 4, 10, and 20× objectives. Regional microdissections are performed under 4× objective and with CapSure LCM macro caps (Arcturus). Next to the monitor with the microscope picture of the LCM image, have a second monitor (or printout) with the stained adjacent image on display, which will guide you when selecting the desired area (see Note 3). LCM settings used for human brain sections are: 50–75 mW power; 1,500–3,500 ms duration; and 200–250 mV intensity (see Note 4). When capture is completed, view captured cells on cap and the corresponding absence of cells on the tissue. Acquisition of image from the filled cap and processed section serves as a valuable quality control tool. Isolation is then initiated. 3.2. RNA Extraction

1. Place each cap into 0.5-ml tubes provided by the Arcturus PicoPure isolation kit. 2. Add 50 ml of extraction buffer to each tube. 3. Invert the tube and incubate at 42°C for 30 min while gently shaking. 4. Centrifuge at 800 × g for 2 min, collect cell extract and remove LCM cap (see Note 5).

3.3. RNA Isolation

1. Preconditioning of RNA purification column. (a) Add 250 ml conditioning buffer. (b) Incubate column for 5 min at RT. (c) Centrifuge in provided collection tube at 16,000 × g for 1 min. 2. Add 50 ml of 70% ethanol to RNA cell extract from step 3.2 and mix well by pipetting. 3. Pipette cell extract-ethanol mix (»100  ml) onto preconditioned purification columns. 4. Centrifuge at 100 × g for 2 min, followed by centrifugation at 16,000 × g for 30 s. 5. Add 100 ml of wash buffer W1 and centrifuge at 8,000 × g for 1 min. 6. Perform DNAse treatment (see Note 6). 7. Add 100 ml wash buffer W2 and centrifuge at 8,000 × g for 1 min. 8. Add another 100  ml wash buffer W2 and centrifuge at 16,000 × g for 2 min. If residual buffer remains on the column, centrifuge again at 16,000 × g for 1 min.

29  Region-Specific In Situ Hybridization-Guided Laser-Capture Microdissection¼

353

9. Transfer purification column to provided 0.5-ml tube. 10. Add 20  ml elution buffer onto membrane of purification column and incubate for 1 min at RT. 11. Centrifuge column at 1,000 × g for 1  min followed by 16,000 × g for 1 min. 12. Immediately process RNA or store at −80°C. 3.4. Agilent RNA Pico Assay and Assessment of RNA Quality

RNA integrity is crucial for the success of any gene expression profiling method. Postmortem samples will always have less than perfect quality. To know whether RNA quality is sufficient for reliable and reproducible results is imperative and there are many different methods to access RNA integrity, such as 28S/18S peak ratio, RIN, and 3¢/5¢ ratio of GAPDH expression. All these methods can tell a perfect from an imperfect sample. However, there is a gray zone at which a small amount of RNA degradation products are present, which will ultimately impact all three aforementioned measures. A recently published method (8), which we favor, indicates whether a moderately degraded RNA sample, such as from postmortem brain tissue, still produces reliable gene expression results. Next to RNA quality control, the RNA Pico assay has another purpose. It determines the amount of RNA in the sample, which is needed prior to RNA amplification reactions.

3.4.1. Agilent RNA Pico Assay

Preparing ladder 1. Thaw and spin ladder to bottom of the tube. Heat to denature for 2 min at 70°C. 2. Immediately cool the tube on ice. 3. Add 90 ml of RNase-free water and mix thoroughly. 4. Prepare aliquots in RNase-free tubes with the required amount for a typical daily use and store them at −70°C. 5. Before use, thaw ladder aliquots and keep them on ice (avoid extensive warming upon thawing process). Preparing gel matrix 6. Remove reagents from 4°C storage and bring them slowly to RT. 7. Add 550 ml of gel matrix to a spin column and centrifuge at 1,500 × g for 10 min. 8. The filtered gel can be used for up to 1 month when stored at 4°C. Preparing gel-dye mix and electrodes 9. Place 65 ml of the filtered gel matrix (at room temperature) in a 0.5-ml tube.

354

R. Bernard et al.

10. Add 1 ml of pico dye, vortex, and then centrifuge at 13,000 × g for 10 min. 11. Gel-dye mix is light sensitive and good to use for 24 h. One chip requires about 30 ml of gel-dye mix. 12. Clean electrodes by adding 400 ml of DEPC water to a clear cleaning chip, which is put into the bioanalyzer for 5  min. Open bioanalyzer lid and air dry for at least 2  min. Repeat this cleaning procedure once. 3.4.2. Loading the PicoChip

1. Place samples into 70°C water bath for 2 min. Do not heat ladder. 2. Open a new PicoChip and place in loading station. 3. Load 9 ml of gel-dye mix into a well marked with a “white G in black circle.” 4. Close top of loading station and depress plunger until it is below retaining notch. 5. Wait 30  s before releasing notch carefully and pull plunger slowly back to the 1 ml mark. Open loading station slowly. 6. Load 9  ml of gel-dye mix into two remaining wells marked with a “G.” 7. Load 9 ml of pico conditioning solution into a well marked with “CS.” 8. Load 5 ml of pico marker into the remaining 12 wells which includes “ladder.” 9. Load 1 ml of pico ladder in the well marked “ladder.” Then load 1 ml of each sample into the sample wells (see Note 7).

3.4.3. Sample Run on PicoChip

1. The chip is placed securely in the supplied IKA vortexer and vortex for 1 min. 2. Place the chip in the bioanalyzer and close the lid (see Note 8). 3. Start bioanalyzer program (Agilent 2100 Expert software) and select Eukaryote total RNA pico, name samples, and then start the sample run (see Note 9). 4. Save the data and export them to an Excel spreadsheet format. 5. Clean electrodes with distilled water and let them air dry.

3.5. RNA Quality Assessment (8)

1. In Excel, determine the beginning and ending of elution time for the 18S and 28S peak (see Note 10). 2. Calculate the AUC from the appearance of the first peak of the sample to the beginning of 18S peak. 3. Calculate the AUC of the 28S peak and the total sample AUC that ranges from the appearance of the first peak to the last. The latter is usually the 28S peak.

29  Region-Specific In Situ Hybridization-Guided Laser-Capture Microdissection¼

355

According to the algorithm determined by Schoor et  al., a sample with the following parameters delivers valid and reproducible gene expression results: 1. Less than 65% of the total signal is contained in the pre-18S area. 2. The 28S peak accounts for more than 4% of the total signal. Figure 2 shows a representative electropherogram. 3.6. RNA Amplification

For human tissue, the Agilent Pico Chip assay indicates a range from 1,000 to 10,000 pg/ml of RNA depending on LCM sample size and area. Most microarray experiments require nearly 1 mg RNA sample for detection. Amplification of the RNA is therefore required. RNA amplification is not necessary if only RT-PCR processing is desired. For amplification, we used isovolumetric sample amounts of RNA. For amplification reactions, we employed the RiboAmp Plus 1.5 round kit and the BioArray High Yield RNA Transcription Labeling Kit (T7). The end product of the RiboAmp Plus 1.5 round protocol is amplified cDNA, which can be directly used for RT-PCRs (see Note 11). When intending to perform microarray experiments, only a portion of cDNA is retained for PCR; whereas the remainder is used with the BioArray High Yield RNA kit to produce biotin-labeled amplified RNA for microarray experiments. 1. Round 1 – first strand cDNA synthesis (see Note 12). 2. Round 1 – second strand cDNA synthesis. 3. Round 1 – cDNA purification (see Note 13). 4. Round 1 – In vitro transcription. 5. Round 1 – aRNA Purification. 6. Round 2 – first strand cDNA synthesis. 7. Round 2 – second strand cDNA synthesis. 8. Round 2 – cDNA purification (see Note 14).

3.7. Biotinylation for Microarray Experiments

1. We followed the instructions from Enzo BioArray High Yield RNA Transcript Labeling Kit. This kit uses two different biotinlabeled nucleotides (Biotin–CTP and Biotin–UTP), which enables more uniform and efficient labeling. Specifically, we use only 7 ml of the Round 2 cDNA in the in vitro transcription (IVT) reaction and store the remainder of the sample at −80°C for RT-PCR. The reaction runs for 16 h and can stay at 4°C for a short length of time before purification. The biotinylated RNA needs to be cleaned up. We use the Qiagen RNeasy Kit for this step. 2. Add 40 ml of RNase-free water to each sample and then add 280 ml of buffer RLT (do not add b-mercaptoethanol to the RLT buffer). Mix by pipetting up and down.

356

R. Bernard et al.

3. Add 340 ml of 100% ethanol and mix again by pipetting up and down. 4. Load the entire volume onto a supplied spin column placed in a 2-ml collection tube. Close the lids and centrifuge at 8,000 × g for 15 s. Discard flow-through. 5. Add 700 ml buffer RW1 to the spin column. Close the lids and centrifuge at 8,000 × g for 15 s. Discard flow-through. 6. Add 500 ml buffer RPE to the spin column. Close the lids and centrifuge at 8,000 × g for 15 s. Discard flow-through. Add another 500 ml of buffer RPE and centrifuge at 8,000 × g for 2 min. 7. Place the spin column in a new 1.5-ml collection tube. Add 30 ml RNAse-free water directly onto the column. Wait 2 min and then centrifuge at 8,000 × g for 1 min to elute cleaned up RNA, which still needs to be quantified prior to array chip hybridization (see Note 15). The samples should be stored at −80°C. 3.8. Gene Chip Microarrays

1. After the biotin-labeled cRNA is prepared and purified, it can be used on Gene Chips. For the Affymetrix Expression system, the cRNA is fragmented and added to a hybridization cocktail containing probe array controls. 2. It is then hybridized to the Gene Chip during a 16 h incubation. 3. Immediately following hybridization, the array is washed and stained with a streptavidin phycoerythrin conjugate using an automated protocol on the GeneAtlas™ Fluidics Station, followed by scanning on the GeneAtlas™ Scanner. For Illumina Bead Chips, the cRNA is added directly to the hybridization cocktail and applied to the chip which is hybridized for 16–20 h at 58°C. 4. The chip is then washed and incubated with streptavidin-Cy3. The chip is again washed, dried, and scanned with an Illumina Bead Chip scanner. 5. The data is analyzed with BeadStudio software.

3.9. Quant-iT PicoGreen Assay for PCR Input Determination

Because we do not want to rely on reference housekeeping genes to determine relative abundance of gene expression, we utilized the Quant-iT PicoGreen dsDNA assay to determine nucleic acid concentration in our sample. This assay offers great sensitivity at relatively low cDNA concentrations (see Note 16). Solution preparation 1. TE buffer: 20× dilution with sterile, distilled RNAse-free water. 2. PicoGreen reagent: 200× dilution with 1× TE. 3. DNA standard: Two-step dilution (1:50, then 1:400) with 1× TE.

29  Region-Specific In Situ Hybridization-Guided Laser-Capture Microdissection¼

357

Sample processing 1. 1 ml cDNA sample. 2. 100 ml 1× TE buffer. 3. 100 ml PicoGreen reagent. Standard curve – low range 1× TE buffer (ml)

50 ng/ml DNA standard (ml)

PicoGreen reagent (ml)

Final DNA conc.

0

1,000

1,000

25 ng/ml

900

100

1,000

2.5 ng/ml

990

10

1,000

250 pg/ml

999

1

1,000

25 pg/ml

1,000

0

1,000

0

Light-protected incubation of all samples is recommended for 2–5 min. Thereafter, samples (run in triplicates) and standard solutions are assayed in final volume of 200  ml in a 96-well microplate and processed on a CytoFluor microplate reader. Samples are excited at 485 nm and measured with a 535 nm emission filter for 0.1 s. Sample concentrations were determined with the standard curve. If cDNA from amplified material is used, we recommend a two-step dilution to obtain a final concentration of 50  pg/ml, which is the input concentration for subsequent qPCR. 3.10. Polymerase Chain Reaction

1. Genomic DNA and mRNA sequences are downloaded from NCBI Entrez Gene at http://www.ncbi.nlm.nih.gov/sites/ entrez?db=gene. PCR primer pairs are designed to anneal within 500 bp of the 3¢ end to generate a single amplicon that is shared by all of the known splice variants, between 75 and 150 bp in size using Primer3 software (13). Secondary structure is minimized for each amplicon using DNA Mfold (14) (http://www.bioinfo.rpi.edu/applications/mfold/), and primers are then designed against regions lacking predicted secondary structure. 2. Amplification reactions are carried out in 96-well 0.2-ml Thin-Wall PCR plates. 3. The total volume for each real-time PCR is 20  ml – which included 5  ml of cDNA, forward and reverse primers at a final concentration of 450 nM, and 10 ml of iQ SYBR Green Supermix, containing 50  U/ml iTaq/DNA polymerase, 6  mM MgCl2, 0.4  mM of each dNTP, and 20  nM fluorescein.

358

R. Bernard et al.

4. All amplifications and fluorescence quantifications are performed in real-time on a Bio-Rad iCycler using a SYBR-488 detection protocol, at a peak excitation wavelength of 490 nm and peak emission wavelength of 530 nm. 5. To minimize amplification of nonspecific products, a hot-start PCR protocol is employed as follows: denature at 95°C for 30 s, followed by ten cycles of denaturing at 95°C for 15 s annealing at 65–60°C for 15 s, extension at 72°C for 15 s. Annealing temperature is decreased at each step of the cycle by 0.5°C from a maximum of 65°C to the final temperature of 60°C. This is followed by 35 cycles of denaturing at 95°C for 15 s, annealing at 60°C for 15 s, extension at 72°C for 15 s. Following each extension step, primer dimers are denatured by raising the temperature to 83°C for 30 s, and then fluorescence is quantified. This cycling was followed by a final extension step at 72°C for 5 min. 6. Following amplification, PCR products are denatured by sequential increases in temperature from 72°C to 95°C in 0.5°C increments. At each step, the temperature was held constant for 10 s, and fluorescence was quantified. Presence of specific amplification products was confirmed by the single peak on the melting curve, plotted as the negative derivative of fluorescence as a function of temperature.

4. Notes 1. Ideally, the cryostat is dedicated to sectioning of human specimens only to prevent contamination with human pathogens. It needs to be thoroughly cleaned before and after use with 70% ethanol. Always wear protective gloves. 2. Postmortem human brain samples, by virtue of their harvesting methodology, have less than perfect RNA quality. Therefore, it is important to prevent further RNA degradation during processing of each tissue sample. Aqueous solutions can contain RNAses, which need to be deactivated. We use the following protocol to produce RNAse-free water. (a) Add 1  ml of 0.1% diethylpyrocarbonate (DEPC) to 1,000 ml distilled water (b) Mix well and let set at room temperature overnight (c) Autoclave solution (d) Let cool to room temperature prior to use Use DEPC-treated water for making ethanol dilutions. Moisture or other contaminants on the brain section, such

29  Region-Specific In Situ Hybridization-Guided Laser-Capture Microdissection¼

359

as unevaporated xylene, contaminated ethanol solutions, fingerprints, or frost hinder the ability to perform a satisfactory LCM. 3. Next to shape, use anatomical landmarks such a tissue border, fiber tracts, and holes from blood vessels as alignment guides for your area of interest. Use roughly the same magnification as on the monitors. Newer LCM software may also permit to overlay the stained image from the adjacent section with your target section. 4. Before starting a series of collection, test fire the laser on a piece of tissue outside the area of interest. The laser focus is verified by observing black, melted rings of plastic. When slides are changed, readjustment of focus may be necessary. Keep the caps free from any contamination. Adjustments to the laser power and duration may be needed during the tissue collection process from multiple slides. In general, if the plastic does not melt increasing the power might help and if cells are not detached laser duration should be increased. Steadily increasing power settings over time to achieve cell lifting indicate that the sections are absorbing moisture. 5. It is advisable to extract only and store the sample at −80°C until all the samples for one study are captured. From RNA isolation step, all samples of one study should be processed together to minimize variability. 6. DNAse treatment is crucial step to remove genomic DNA. We use RNAse-Free DNase Set from Qiagen. (a) For one sample: Add 5  ml DNAse I Stock solution to 35 ml buffer RDD. Mix gently. According to the number of samples to be processed, calculate the needed total amount of this mix and make one large batch. (b) Pipette 40 ml DNAse incubation mix into the purification column membrane. Incubate at RT for 15 min. (c) Pipette 40 ml PicoPure RNA Kit Wash Buffer 1 (WB1) into the purification column membrane and centrifuge at 8,000 × g for 15 s. (d) Proceed with RNA isolation. 7. Perform steps 6–9 relatively fast with no breaks in between. Use the loaded chip within 5 min. 8. It is important that no bubbles are introduced during this step. Close and open the lid slowly several times to ensure the absence of any air bubble. If a message about a bad chip appears right at the beginning of the run, there are probably bubbles in a well. Lid then needs to be opened and closed three times before restarting the assay. If this does not resolve the problem, the chip may need to be rerun.

360

R. Bernard et al.

9. If after a rerun of the chip, the results are still not satisfactory, the pico head may need to be cleaned. Electrodes are gently scrubbed with a soft toothbrush, then sonicated for 10 min in RNAse-free water. Change water and repeat procedure once with water and then twice with 100% ethanol. Dry pico head at 37°C overnight before using again. 10. To calculate AUC, use the numerical approximation method called the trapezoidal rule. Divide the desired AUC into very small segmental trapezoids and add their areas together. The smallest possible time increment between two signal data points equals the height of one trapezoid. Make sure to subtract the height from the baseline first before calculating the AUC. 11. The kits contain only a slight volume coverage of components. Be careful with pipetting. Use a fresh pipette tip for each addition to each sample. This will provide a consistency between samples. You can stop at any of the points indicated in the protocol and store the samples as described. It is possible to complete all of Round 1 in one day and store the samples at −80°C overnight. 12. The kit allows you to use up to 11  ml of RNA from the LCM sample. The amount you use will depend on the concentration of the RNA determined from the Pico Chip (Subheading 3.4, step 1). For the RiboAmp Plus kit, between 10 and 40 ng of starting material is needed. All samples will have different concentrations and it is possible to adjust the volume you use so that all samples have the same starting concentration. Alternatively, one can use the same volume for all samples as long as the concentrations fall within the range indicated by the kit. 13. Be careful when adding the DNA binding buffer to your sample tube. The PCR tube will be almost full and mixing will be difficult. Be careful when centrifuging your tubes in the final step. It is possible to break the lids off if the tubes are not placed in the centrifuge correctly. Protocol directions should be strictly followed. 14. Round 2 will take less than 3 h. If you intend to start the final IVT reaction during the same day, store the samples on ice or complete Round 2 just before you begin the IVT reaction which will incubate overnight. 15. The RNA samples can be quantified with a NanoDrop 2000 UV-Vis photospectrophotometer. We use 1 ml of sample on the NanoDrop. Concentrations usually range from 500 to 2,000 ng/ml. Chip input amounts vary depending on manufacturer: While the Illumina 8 sample chip requires only 750  ng of human aRNA, the HG-U133 Plus 2.0 from Affymetrix array requires 15 mg aRNA.

29  Region-Specific In Situ Hybridization-Guided Laser-Capture Microdissection¼

361

16. The sensitivity of this assay is limited to 25 pg/ml cDNA. If no amplification step is used, one needs to collect more tissue (sections) to yield more RNA. The amounts provided in Table 1 can serve as orientation. Also consider that 50 pg/ml cDNA is needed for qPCR. References 1. Bernard, R., Kerman IA, Meng F, Evans SJ, Amrein I, Jones EG, Bunney WE, Akil H, Watson SJ, Thompson RC (2009) Gene expression profiling of neurochemicallydefined regions of the human brain by in situ hybridization-guided laser capture microdissection. J Neurosci Methods, 178:46–54. 2. Bernard, R., Kerman IA, Thompson RC, Jones EG, Bunney WE, Barchas JD, Schatzberg AF, Myers RM, Akil H, Watson SJ (2011) Altered expression of glutamate signaling, growth factor, and glia genes in the locus coeruleus of patients with major depression. Mol Psychiatry, 16: in press. 3. Atz, M., Walsh D, Cartagena P, Li J, Evans S, Choudary P, Overman K, Stein R, Tomita H, Potkin S, Myers R, Watson SJ, Jones EG, Akil H, Bunney WE Jr, Vawter MP (2007) Methodological considerations for gene expression profiling of human brain. J Neurosci Methods, 163:295–309. 4. Li, J.Z., Vawter MP, Walsh DM, Tomita H, Evans SJ, Choudary PV, Lopez JF, Avelar A, Shokoohi V, Chung T, Mesarwi O, Jones EG, Watson SJ, Akil H, Bunney WE Jr, Myers RM (2004) Systematic changes in gene expression in postmortem human brains associated with tissue pH and terminal medical conditions. Hum Mol Genet, 13:609–16. 5. Tomita, H. Vawter MP, Walsh DM, Evans SJ, Choudary PV, Li J, Overman KM, Atz ME, Myers RM, Jones EG, Watson SJ, Akil H, Bunney WE (2004) Effect of agonal and postmortem factors on gene expression profile: quality control in microarray analyses of postmortem human brain. Biol Psychiatry, 55:346–52. 6. Rehncrona, S., I. Rosén, B.K. Siesjö. (1981) Brain lactic acidosis and ischemic cell damage: 1. Biochemistry and neurophysiology. J Cereb Blood Flow Metab, 1:297–311.

7. Yates, C.M. Butterworth J, Tennant MC, Gordon A. (1990) Enzyme activities in relation to pH and lactate in postmortem brain in Alzheimer-type and other dementias. J Neurochem, 55:1624–30. 8. Schoor, O. Weinschenk T, Hennenlotter J, Corvin S, Stenzl A, Rammensee HG, Stevanović S. (2003) Moderate degradation does not preclude microarray analysis of small amounts of RNA. Biotechniques, 35:1192–6, 1198–201. 9. Greene, J.G., R. Dingledine, J.T. Greenamyre, (2005) Gene expression profiling of rat midbrain dopamine neurons: implications for selective vulnerability in parkinsonism. Neurobiol Dis, 18:19–31. 10. Torres-Munoz, J.E. Van Waveren C, Keegan MG, Bookman RJ, Petito CK. (2004) Gene expression profiles in microdissected neurons from human hippocampal subregions. Brain Res Mol Brain Res, 127:105–14. 11. Neal, C.R., Jr., H. Akil, S.J. Watson, Jr., (2001) Expression of orphanin FQ and the opioid receptor-like (ORL1) receptor in the developing human and rat brain. J Chem Neuroanat, 22:219–49. 12. Kerman, I.A. Buck BJ, Evans SJ, Akil H, Watson SJ. (2006) Combining laser capture microdissection with quantitative real-time PCR: effects of tissue manipulation on RNA quality and gene expression. J Neurosci Methods, 153:71–85. 13. Rozen, S. Skaletsky H, (2000) Primer3 on the WWW for general users and for biologist programmers. Methods Mol Biol, 132:365–86. 14. Zuker, M., (2003) Mfold web server for nucleic acid folding and hybridization prediction. Nucleic Acids Res, 31:3406–15.

wwwwwww

Chapter 30 UV-Laser Microdissection and mRNA Expression Analysis of Individual Neurons from Postmortem Parkinson’s Disease Brains Jan Gründemann, Falk Schlaudraff, and Birgit Liss Abstract Cell specificity of gene expression analysis is essential to avoid tissue sample related artifacts, in particular when the relative number of target cells present in the compared tissues varies dramatically, e.g., when comparing dopamine neurons in midbrain tissues from control subjects with those from Parkinson’s disease (PD) cases . Here, we describe a detailed protocol that combines contact-free UV-laser microdissection and quantitative PCR of reverse-transcribed RNA of individual neurons from postmortem human midbrain tissue from PD patients and unaffected controls. Among expression changes in a variety of dopamine neuron marker, maintenance, and cell-metabolism genes, we found that a-synuclein mRNA levels were significantly elevated in individual neuromelanin-positive dopamine midbrain neurons from PD brains when compared to those from matched controls. Key words: Laser microdissection, Single cell, Real-time quantitative PCR, RNA integrity number, Reverse transcription, Postmortem, Human tissue, Synuclein, Parkinson’s disease, Dopamine

1. Introduction Gene expression analysis via quantitative PCR of reverse-­ transcribed RNA (RT-qPCR) has become a routine technique. Contact free UV-laser microdissection (UV-LMD) enables RT-qPCR-based mRNA analysis of specific tissue regions, homogeneous cell pools, and even individual cells (1, 2). Besides a variety of methodological considerations that could bias RT-qPCR results in general (3, 4), cellular specificity of gene expression analysis is a desired goal to prevent tissue-sample-­ generated artifacts, in particular, when the relative number of

Graeme I. Murray (ed.), Laser Capture Microdissection: Methods and Protocols, Methods in Molecular Biology, vol. 755, DOI 10.1007/978-1-61779-163-5_30, © Springer Science+Business Media, LLC 2011

363

364

J. Gründemann et al.

target cells present in a given tissue varies dramatically – e.g., between control and disease states in the context of neurodegenerative diseases. High cellular heterogeneity, selective neuron loss, and disease-related changes in nonneuronal cells will contribute to an altered composition of the diseased brain tissue compared to that in controls. Importantly, this will confound any conclusions about specific gene expression changes in the cell type of interest. For example, one of the key pathological hallmarks of Parkinson’s disease (PD) and its animal models is the loss of dopamine-containing (DA) midbrain neurons, in particular within the substantia nigra (SN) pars compacta. In PD, typical clinical motor symptoms manifest not until approximately 75 % of these DA midbrain neurons – the most prominent cell type within the SN – are lost (5, 6). This massive loss of SN DA neurons will confound mRNA expression analysis of PD midbrain tissue when compared to controls. In addition, gene expression analysis at the level of PD midbrain tissue will be furthermore distorted by altered numbers and functional states of nonneuronal cells such as microglia, astrocytes, and local T-cells, which are known to occur in PD (7). All these factors might explain the large number of different and even contrary findings of ­tissue-based gene expression studies in PD brains, e.g., for a-synuclein (reviewed in ref. 8) – a gene that can cause familiar forms of PD when mutated (PARK1) or duplicated/triplicated (PARK4) (9). Cell-specific quantification of gene expression with single cell resolution overcomes these tissue-related limitations of gene expression data from pathological tissues and controls, since it enables the unbiased detection of cell-specific transcriptional dysregulation. Here, we describe a step-by-step protocol for UV-laser microdissection of individual neurons from frozen postmortem human midbrain tissue and subsequent reliable RT-qPCR gene expression analysis of individual cells or small cell pools (1, 10). Cell lysis and cDNA synthesis are performed in the same reaction tube without a distinct RNA isolation step to avoid RNA loss, contamination, and handling errors. We specifically focus on postmortem gene expression analysis of neuromelanin-positive (NM+) DA midbrain neurons from the SN of PD patients and respective controls and describe the robust detection of significantly higher mRNA levels of a-synuclein in NM + DA neurons from PD brains compared to controls (1). These results suggest that transcriptional dysregulation of the a-synuclein gene and elevated a-­synuclein levels not only cause rare familiar forms of PD (PARK4) (11–13) but also might additionally contribute to the risk and pathology of sporadic PD as suggested by recent genome-wide association and related expression quantitative trait loci (eQTL) studies (14–16).

30  UV-Laser Microdissection and mRNA Expression Analysis of Individual Neurons…

365

2. Materials 2.1. Handling, Fixation, and Staining of Human Brain Tissue

1. To prevent contamination, human brain tissue specimens are stored in heat-sterilized tinfoil and RNase-ExitusPlus (AppliChem, Darmstadt, Germany)-treated parafilm-sealed boxes at −80 °C. 2. Microtome blades (Leica, Nussloch, Germany, Type 819) washed for 30 s in 70 % RNase-free ethanol and whipped with RNase-ExitusPlus and isopropanol (Sigma-Aldrich, St Louis, USA) (see Note 1). 3. Ethanol dilutions (2× 75 %, 95 %, 100 % absolute, and 100 % anhydrous) are freshly prepared on each experimental day and stored in 50 ml Falcon tubes at room temperature. One tube of 75 % ethanol is kept at −20 °C. 75 % and 95 % ethanol dilutions are prepared from ethanol absolute puriss. p.a. (Sigma-Aldrich) in RNase free water (5Prime, Hamburg, Germany). Ethanol anhydrous stock is stored with molecular sieve (Merck, Darmstadt, Germany, pore size: 0.3 nm, 25 g/l). 4. 1 % cresyl violet acetate staining dye (Sigma) is diluted in 100 % ethanol absolute puriss. p.a., stored in a tinfoil-covered and parafilm-sealed Falcon tube and incubated at least overnight before use. 5. Drying box with silica gel with moisture indicator (Merck).

2.2. UV-LMD

1. A contact-free LMD microscope is needed. This protocol was successfully tested with both the Zeiss (Munich, Germany) PALM UV-LMD setup and the Leica UC-LMD6000 and 7000 setups. Heat sterilization (180 °C, 2  h) of all LMD microscope parts that are in contact with the tissue slides (i.e., slide holder) or reaction tubes (i.e., cap or tube holder) prevents RNase contamination (see Note 1). 2. PEN-membrane slides (2.0  mm, MicroDissect, Herborn, Germany) for mounting of tissue sections and laser microdissection, treated with UV-C light for 15 min. 3. RNase-free thin-walled 0.5-ml PCR reaction tubes with flat cap for cell collection and combined cell lysis and cDNA synthesis, UV-C treated for 45 mins.

2.3. Preparation of Cap-Mix for Combined Cell Lysis and cDNA Synthesis

1. Cell lysis and cDNA synthesis are performed in the same buffer (Cap-Mix), containing 0.5 % NP-40 (Roche, Basel, Switzerland, light-sensitive 10 % stock stored in aliquots in dark at +4 °C; see Note 2), 5  U SUPERase-In (Ambion, Austin TX, USA, stored in aliquots at −20 °C), 0.5  mM dNTPs (GE Healthcare, Freiburg, Germany, stored as 20 mM stock at −20 °C), 5  mM random hexamer primer (Roche,

366

J. Gründemann et al.

stored as 1 mM stock aliquots at −20 °C), 500 ng poly-inosine (Sigma-Aldrich, stored as 1 mg/ml stock at −20 °C), 2 mM Tris–HCl pH 7.4 (Sigma-Aldrich, 100  mM stock stored at −20 °C), 10 mM DTT (Invitrogen, Carlsbad, USA, stored as 100 mM stock at −20 °C) in 1× first-strand buffer (Invitrogen, 5× stock: 250 mM Tris–HCl, 375 mM KCl, 15 mM MgCl2, pH  8.3, stored in aliquots at −20 °C) at a final volume of 4.7 ml per reaction. 2. Cap-Mix sufficient for the number of samples that are collected (plus positive and negative controls) is freshly prepared on each experimental day and stored on ice in a light-­protected, RNasefree 0.5-ml single sealed reaction tube (biopure, Eppendorf, Hamburg, Germany). All components are carefully added, mixed by finger flipping, and quickly centrifuged. SUPERase-In (Ambion, Austin, USA) is added directly from −20 °C to the reaction mix. Poly-inosine, NP40, and SUPERase-In are viscous and special care has to be taken during pipetting to avoid air bubbles. If bubbles are formed, the mix is centrifuged briefly until all bubbles disappear (see Note 3). 3. 100  U SuperScript II Reverse Transcriptase (Invitrogen, stored in aliquots at −20 °C) is added to each reaction after lysis (see subheading 3.3). The enzyme aliquots are stored in a benchtop freezer (Techne, Stone, UK) at −20 °C during the experiment to avoid “freeze–thaw cycles”. 2.4. Quantitative Real-Time PCR

1. 2× QuantiTect Probe PCR Master Mix (Qiagen, Hilden, Germany). 2. 20x TaqMan Primer-Probe Assay (e.g., Applied Biosystems, Warrington, UK, a-synuclein assay number: HS00240906_m1, detects a-synuclein splice variants 112 and 140) (see Note 4). 3. cDNA for generation of standard curves to assess assay performance, e.g., SN cDNA (e.g., serial dilutions from 30 to 0.03  ng cDNA, derived from human SN tissue total RNA, Ambion, 1 mg/ml). 4. A real-time quantitative PCR (qPCR) instrument, e.g., the GeneAmp 7900HT real-time qPCR system (Applied Biosystems) used here, or a comparable instrument. 5. Suitable qPCR MicroAmp 96-well reaction plates and optical adhesive covers (e.g., both Applied Biosystems).

3. Methods To guarantee successful UV-LMD and subsequent gene expression analysis of small cell pools and individual cells, it is essential to work in a strictly RNase-free regime. For details on RNase-free

30  UV-Laser Microdissection and mRNA Expression Analysis of Individual Neurons…

367

Fig. 1. Flowchart illustrating the experimental procedure for UV-laser microdissection and gene expression analysis of individual human dopaminergic neurons from postmortem midbrains. For details, please see text. RIN RNA integrity number.

working conditions, see Note 1. The protocol described below was used to quantify mRNA levels in NM + DA SN pars compacta neurons from human postmortem midbrain tissue blocks, provided by the German Brain Bank. However, it is applicable to various postmortem tissues from different species and can be principally adapted to many other tissues of interest. An overview of the experimental procedures is illustrated in Fig. 1. 3.1. Storage, Cryosectioning, and Staining of Human Brain Tissue

1. On the experimental day, human midbrain tissue is transferred (on dry ice) from −80 °C to the quick-freeze panel of a precooled cryostat (–35 °C) and glued with tissue freezing medium (Leica) on a specimen holder (see Note 5). After an equilibration period of 20 min at –35 °C, the cryostat is set to the optimal cutting temperature (depends on the processed tissue, for our specimens –19 °C) and further equilibrated for 45 min before 12 mm horizontal midbrain sections including the SN are cut. Chippings from the trimming procedure are collected for RNA quality and tissue pH analysis (see Fig. 1 and Note 6). 2. The brain sections are mounted on UV-C treated PENmembrane slides and allowed to thaw briefly (see Note 7). Once thawed, the slide is transferred to a 50 ml Falcon tube with 75 % ethanol at −20 °C and fixed for 2 min.

368

J. Gründemann et al.

3. The slide is removed with a sterile forceps and 0.5  ml 1 % cresyl violet is applied directly on the slide using a sterile filter syringe (0.1 mm), incubated for 1 min, then dipped briefly in 75, 95, and 100 % ethanol absolute and finally incubated for 1 min in 100 % ethanol anhydrous. 4. The fixed and stained slides, each containing several brain sections, are stored in a drying chamber containing silica gel for at least 45 min before UV-LMD. 5. Alternatively, after drying (min. 45  min) the slides can be stored at −80 °C in storage jars (containing silica gel) and used later (see Fig. 1 and Note 8). 3.2. UV-Laser Microdissection of Individual Neurons from Human Brain Samples

1. All work spaces are cleaned to ensure RNAse-free working conditions (see Note 1). 2. The slides with tissue specimens are placed on the sterile slide holder and transferred to the UV-LMD microscope. Tissue quality and staining are inspected under low and high magnifications, and only sections that allow clear identification of individual cells are used for the experiment. Optimal laser settings need to be adjusted for each individual slide/section. 3. After the brain region of interest is found (in our case SN), an UV-C treated thin-walled PCR reaction tube is placed in the cap holder and transferred to the microscope. The reaction tube cap is inspected with the cap-control function to exclude rarely occurring contaminations with dust particles. Individual cells are cut and harvested into the cap of the reaction tube. It is recommended to visually control that all laser microdissected cells are successfully harvested (cap-control function, see Fig. 2). 4. If the cap-control is positive, the cap holder is removed and 4.7 ml Cap-Mix is added to the cap immediately. Any direct contact between the cap and the pipette tip must be avoided.

3.3. Lysis and cDNA Synthesis of Individual Laser Microdissected Neurons from Human Brain Samples

1. The reaction tube is carefully removed from the cap holder and the tube is closed upside down to ensure that the CapMix remains in the cap. 2. The reaction tube is placed upside down on the cap in a preheated (72 °C) thermoblock and incubated for 2 min for cell lysis (see Note 9). 3. Afterwards, the tube is transferred onto an ice-cold metal block, again upside-down, and allowed to cool for 1 min. 4. The Cap-Mix is then spun down at 12,900 rpm in a benchtop centrifuge for 1 min at room temperature. 5. 0.3 ml SuperScript II is added directly to the Cap-Mix at the bottom of the tube after spinning.

30  UV-Laser Microdissection and mRNA Expression Analysis of Individual Neurons…

369

Fig.  2. UV-laser microdissection (UV-LMD) of individual neuromelanin-positive (NM+) dopamine midbrain neurons of human postmortem Parkinson’s disease (PD) (a) and control (b) tissue sections. Upper row : individual neurons before (left) and after UV-LMD (right ). Scale bars: 25 mm. Lower row : overview of the horizontal midbrain section containing substantia nigra after UV-LMD of 15 individual NM + neurons. Scale bars : 250 mm. Inserts : inspection of the reaction tube cap for cell collection. Scale bars: 250 mm.

6. The tube is transferred to a preheated (38 °C) thermomixer (350 rpm for 10 s every 10 min) and cDNA synthesis is carried out for at least 2  h or (recommended) overnight. For overnight incubation, all samples are spun down briefly and transferred to a preheated thermobox for final overnight cDNA synthesis (38 °C). After cDNA synthesis, samples are stored at −20 °C until further processing. 7. For each set of experiments, suitable positive (e.g., ~2  ng purified midbrain tissue RNA) and negative controls (no LMD-harvested cell in cap) are processed in parallel. 3.4. Quantitative Real-Time PCR of UV-LMD cDNA Samples and Data Analysis

1. The following procedures are all carried out in a UV-Ctreated sterile workbench. 2. Each single cell cDNA sample is diluted 1:11 by adding 50 ml molecular-biology-grade water to the tubes with cDNA in 5 ml Cap-Mix. The diluted cDNA is vortexed and spun down. The tubes are stored in ice-cold metal blocks to ensure constant cooling. Alternatively, single cell samples are purified via ethanol precipitation (see Note 10) (17). 3. A master mix for quantitative real-time PCR in 20 ml reactions is prepared by mixing 10 ml 2× QuantiTect Probe PCR Master Mix, 1 ml 20× primer/probe mix (for gene of interest, e.g., a-synuclein), and 4  ml water (molecular-biology grade) for

370

J. Gründemann et al.

each UV-LMD sample/PCR-reaction (volumes are ­multiplied by the number of samples +1). 4. 5 ml of UV-LMD sample derived cDNA (diluted or purified; for higher concentrations, see Note 10) is added to the bottom of a MicroAmp 96-well reaction plate. 15 ml of master mix is added to each cDNA sample and the plate is sealed with an optical adhesive cover. After spinning for 2  min (1,000 × g, at 4 °C), the plate is transferred to a RT-qPCR system (we use the 7900HT, Applied Biosystems) and the qPCR reaction is run using the appropriate cycling conditions (e.g., specific for our TaqMan assays: 2  min at 50 °C, 15  min at 95 °C and subsequently 50 cycles, 15 min at 94 °C and 1 min at 60 °C each, see Note 11). 5. A serial dilution of a cDNA standard is run in parallel with each experiment (e.g., 300-0.3 ng SN tissue-derived cDNA in serial-dilution steps of 10), which is used as a PCR-positive control, as well as to generate standard curves to assess assay performance and sensitivity and to calculate the cDNA concentration of the UV-LMD samples. 6. For data analysis, the detection threshold is set in the exponential phase of the qPCR amplification plot (relative fluorescence plotted against PCR cycle number, see Fig.  3). To quantify the expression of a certain gene via qPCR for a set of

Fig. 3. Real-time PCR (qPCR) quantification of a-synuclein cDNA derived from individual laser microdissected ­neuromelanin-­ positive (NM+) substantia nigra (SN) neurons of human midbrain from postmortem control (Con) and Parkinson’s disease (PD) cases. (a) a-synuclein qPCR amplification plot showing the change in relative fluorescence (DRN) per qPCR cycle for three pools of 15 dopamine neurons each, from control (grey) and PD (black) brains, as well as the qPCR amplification of standard curve cDNA (dashed lines, 30-0.03 ng cDNA, from left to right, generated from human SN tissue-derived mRNA (Ambion)). Insert: plot of threshold cycle (Ct) value vs cDNA amount (ng) for the tested standard curves. (b) Calculated a-synuclein cDNA amount per single cell (pg-equivalents of standard cDNA derived from SN-tissue per cell) for all ­analyzed UV-LMD collected dopamine neuron pools for 4 PD (grey boxes) and 5 control (black boxes) brains. Each box represents an individual UV-LMD sample. Horizontal bars represent average a-synuclein cDNA amount ± standard error of the mean per single cell for each individual brain (standard curve slope: −3.40, Yintercept: 30.0). (c) Average a-­synuclein cDNA amount per single NM+ dopaminergic midbrain neuron for all analyzed PD (274 ± 33 pg, n = 4) and control brains (46 ± 18 pg, n = 5, P < 0.001).

371

30  UV-Laser Microdissection and mRNA Expression Analysis of Individual Neurons…

samples, the same threshold value is used for all tested samples and standards. Threshold cycle (Ct) values of each sample as well as slope and Y-intercept of the standard curve are read out directly using an appropriate sequence detection software (e.g., SDS2.3, Applied Biosystems). 7. The average cDNA amount per cell in relation to the utilized standard is calculated according to the following formula:

cDNA amount per cell =

(

)

é Ct - Yintercept /slope ù û

Së

No cells ´ cDNA fraction

.



S corresponds to the serial dilution factor of the standard curve (e.g., 10 for serial dilution in steps of 10), Nocells refers to the number of harvested neurons per sample and cDNA fraction to the fraction of the UV-LMD cDNA sample used as template in the real-time PCR reaction, e.g., 5/55. The unit magnitude corresponds to the respective standard utilized, which defines the unit at the Yintercept (e.g., pg-equivalents of standard cDNA, derived from SN-tissue/cell, see Fig. 3 and Note 12).

4. Notes 1. Ribonuclease contamination is a crucial concern for successful cDNA synthesis of single laser microdissected cells or small cell pools. The ubiquitous RNase A is a highly stable and active ribonuclease, which is present on human skin as well as in the specimens and can easily contaminate any lab environment. Thus, creating and maintaining an RNase-free work environment and RNase-free solutions is essential for successfully performing RT-qPCR analysis. Therefore, we strongly recommend to follow these guidelines: (a) Always wear gloves when handling chemicals and sections/samples containing RNA. Change gloves frequently especially after touching potential sources of RNase contamination such as doorknobs, pens, pencils, and human skin. (b) If available, always use certified RNase-free tubes, pipette tips, and chemicals for all steps involved in the experiments (e.g., ethanol/staining solutions and jars for preparation of tissue sections for UV-LMD). Keep chemicals, tubes, etc. tightly sealed. (c) Treat UV-LMD (membrane) slides for 15  min with UV-light (e.g., in a sterile hood).

372

J. Gründemann et al.

(d) Heat-sterilize all metal objects (forceps, spatulas, LMD cap holder, LMD slide holder), glassware, and any other equipment that gets in contact with slides or reaction tubes during UV-LMD experiments at 220 °C overnight. (e) Clean pipettes, benches, and all other equipment that cannot be heat-sterilized with RNase decontamination solutions, e.g., RNase-ExitusPlus (AppliChem) and/or RNaseZapWipes (Ambion). 2. In our hands, NP40 seems to be less effective over time, even when continuously stored at 4 °C in the dark. Thus, we use NP40 aliquots for about three months. 3. Low-retention filter tips should be used for all pipetting steps. To avoid any RNase or DNA contamination. We recommend preparing the Cap-Mix under a sterile fume hood. 4. We recommend using qPCR amplicon sizes below 80  bp when working with tissues of significantly reduced/different RNA qualities (as assessed for example via Agilent RNA integrity number (RIN) analysis), which is often the case for human postmortem brain or other human tissues (1). 5. To reuse the specimen for several experiments, the brains are fixed on cork disks with a tissue freezing medium. These cork disks can be frozen quickly with a drop of water on the specimen holder of the cryostat and easily be removed after the experiment and stored again at −80 °C. 6. Tissue chippings of the cryosectioning procedure are used to assess overall RNA quality and tissue pH of each specimen. Transfer chippings into a liquid nitrogen precooled Falcon tube with cold sterile forceps and store at −80 °C until further usage, e.g., RNA extraction and RIN evaluation, or pH analysis. 7. LMD PEN-membrane slides with a strong iridescence must not be used. These membranes are damaged. 8. PEN-membrane slides with tissue sections can be stored in 50 ml Falcon tubes at −80 °C and reused for later experiments. To ensure that the slides stay dry, silica gel is added to the Falcon tube (see Fig. 1). A small sieve is used to separate the silica gel from the slide and to prevent contamination. For reuse, the slides are removed from −80 °C and allowed to equilibrate at −20 °C (15 min), −4 °C (15 min), and finally at room temperature (15 min) before usage. 9. Our mild lysis protocol is optimized for single UV-laser microdissected cells or small pools of individual cells from ethanol-fixed tissue sections. Please note that it is neither suited for lysis of larger microdissected tissue samples nor

30  UV-Laser Microdissection and mRNA Expression Analysis of Individual Neurons…

373

suited for lysis of single cells from PFA-fixed tissue sections, and it is not tested for lysis of individual plant cells. 10. If the undiluted cDNA reaction is used for quantitative realtime PCR analysis or microarray experiments, cDNA needs to be purified by direct in-tube precipitation to avoid qPCR detriment (17). In this case, employing a reaction tube that is suited for longer high-speed centrifugation (e.g., no thin-walled PCR tubes) is recommended to be used for harvesting of UV-LMD samples. If cDNA precipitation is not required, a maximum of 10 % of the cDNA reaction should be used for quantitative downstream analysis to avoid well-described inhibitory effects of cDNA synthesis reaction components (17). 11. Note that optimal qPCR conditions depend on the qPCR assay system, master mix, and the respective primer/probes utilized. 12. Relative quantification, i.e., normalization against a so-called “housekeeping gene”, is not recommended for RT-qPCR analysis of single cells or small pools of individual cells (18). Thus, it is crucial to use chemicals from the same stocks and lots for all experiments of a study, since even small differences in enzyme efficiencies or reagent concentrations can introduce a strong bias.

Acknowledgments We are particularly grateful to the brain donors and the support by the German BrainNet (GA28). We thank Leica Microsystems for providing the LMD6000 and Jochen Roeper for critical reading the manuscript. This work was supported by the BMBF (NGFN-Net), the DFG (SFB497), the Gemeinnützige Hertiestiftung, the Parkinson’s Disease society, and the Royal Society, UK. J.G. is supported by a PhD Studentship of the Wellcome Trust; B.L. is supported by the Alfried Krupp prize for young university teachers. References 1. Grundemann J, Schlaudraff F, Haeckel O et al (2008) Elevated alpha-synuclein mRNA levels in individual UV-laser-microdissected dopaminergic substantia nigra neurons in idiopathic Parkinson’s disease. Nucleic Acids Res 36:e38 2. Simunovic F, Yi M, Wang Y et al (2009) Gene expression profiling of substantia nigra dopamine neurons: further insights into Parkinson’s disease pathology. Brain 132:1795–1809

3. Huggett J, Dheda K, Bustin S et  al (2005) Real-time RT-PCR normalisation; strategies and considerations. Genes Immun 6:279–284 4. Bustin SA, Nolan T (2004) Pitfalls of ­quantitative real-time reverse-transcription polymerase chain reaction. J Biomol Tech 15:155–166 5. Damier P, Hirsch EC, Agid Y et al (1999) The substantia nigra of the human brain. II. Patterns of loss of dopamine-containing neurons in Parkinson’s disease. Brain 122:1437–1448

374

J. Gründemann et al.

6. Damier P, Hirsch EC, Agid Y et  al (1999) The substantia nigra of the human brain. I. Nigrosomes and the nigral matrix, a compartmental organization based on calbindin D(28K) immunohistochemistry. Brain 122: 1421–1436 7. Hirsch EC, Hunot S (2009) Neuroinflammation in Parkinson’s disease: a target for neuroprotection?. Lancet Neurol 8:382–397 8. Dachsel JC, Lincoln SJ, Gonzalez J et  al (2007) The ups and downs of alpha-synuclein mRNA expression. Mov Disord 22:293–295 9. Cookson MR (2009) alpha-Synuclein and neuronal cell death. Mol Neurodegener 4:9 10. Ramirez A, Heimbach A, Grundemann J et  al (2006) Hereditary parkinsonism with dementia is caused by mutations in ATP13A2, encoding a lysosomal type 5 P-type ATPase. Nat Genet 38:1184–1191 11. Ibanez P, Bonnet AM, Debarges B et al (2004) Causal relation between alpha-synuclein gene duplication and familial Parkinson’s disease. Lancet 364:1169–1171 12. Eriksen JL, Przedborski S, Petrucelli L (2005) Gene dosage and pathogenesis of Parkinson’s disease. Trends Mol Med 11:91–96

13. Polymeropoulos MH, Lavedan C, Leroy E et al (1997) Mutation in the alpha-synuclein gene identified in families with Parkinson’s disease. Science 276:2045–2047 14. McCormack AL, Di Monte DA (2009) Enhanced alpha-synuclein expression in human neurodegenerative diseases: pathogenetic and therapeutic implications. Curr Protein Pept Sci 10:476–482 15. Simon-Sanchez J, Schulte C, Bras JM et  al (2009) Genome-wide association study reveals genetic risk underlying Parkinson’s disease. Nat Genet 41:1308–1312 16. Satake W, Nakabayashi Y, Mizuta I et al (2009) Genome-wide association study identifies common variants at four loci as genetic risk factors for Parkinson’s disease. Nat Genet 41:1303–1307 17. Liss B (2002) Improved quantitative real-time RT-PCR for expression profiling of individual cells. Nucleic Acids Res 30:e89 18. Liss B, Franz O, Sewing S et  al (2001) Tuning pacemaker frequency of individual dopaminergic neurons by Kv4.3L and KChip3.1 transcription. EMBO J 20: 5715–5724

Chapter 31 Transcriptome Profiling of Murine Spinal Neurulation Using Laser Capture Microdissection and High-Density Oligonucleotide Microarrays Shoufeng Cao, Boon-Huat Bay, and George W. Yip Abstract Neurulation is a critical process in the formation of the central nervous system during embryonic ­development. Closure of the neural tube is driven by forces that originate from both the neuroepithelium and the surrounding tissues. In this chapter, we describe the use of laser capture microdissection to ­isolate and separately collect cells from the neuroepithelium and the underlying mesenchyme. We provide protocols for processing of samples for downstream comparison of the transcriptomes of two cell populations using high-density oligonucleotide microarrays, with an emphasis on important technical issues that are to be borne in mind when carrying out these experiments. Key words: Laser capture microdissection, Mammalian spinal neurulation, Transcriptomic analysis, High-density oligonucleotide microarray

1. Introduction Neurulation is a crucial process in the development of the central nervous system during embryogenesis (1, 2). The precursor of the brain and spinal cord is the neural tube, which is formed by elevation, apposition, and fusion of the neural folds. The importance of these processes is highlighted by congenital neural tube defects such as spina bifida and anencephaly, which occur as a consequence of failure of neural tube closure. The incidence of neural tube defects ranges from 0.5 to 2 per 1,000 pregnancies globally (3). Although more than 190 mouse mutants with neural tube defects have been identified, the complex mechanisms that regulate neural tube closure are still not well understood (4–6).

Graeme I. Murray (ed.), Laser Capture Microdissection: Methods and Protocols, Methods in Molecular Biology, vol. 755, DOI 10.1007/978-1-61779-163-5_31, © Springer Science+Business Media, LLC 2011

375

376

S. Cao et al.

Examination of the changes in the transcriptomic profiles at different stages of neural tube closure allows for the identification of candidate genes that may play important regulatory roles in this process. Mode 1 spinal neural tube closure occurs in the mouse with formation of a median hinge point between the 8- and 15-somite stages (7). During the 16- to 23-somite stages, Mode 2 closure occurs in the presence of both the median hinge point and a pair of dorsolateral hinge points. By contrast, Mode 3 closure takes place from the 24-somite stage onwards without formation of the median hinge point. Factors that drive neural tube closure arise from both the neuroepithelium and the surrounding tissues. Laser capture microdissection is a valuable tool that enables us to isolate and separately collect neuroepithelial and mesenchymal cells from the closing neural tube. In this chapter, we describe the use of laser capture microdissection and high-density oligonucleotide microarrays to compare the transcriptomes of these two cell populations at different stages of neural tube closure.

2. Materials 2.1. Harvesting, Cryosectioning, and Laser Capture Microdissection of Embryos

1. Male and female ARC(S) mice, at least 6 weeks old. 2. Cryotome, such as Leica CM3050 S (Leica, Wetzlar, Germany). 3. PixCell II Laser Capture Microdissection System (Arcturus, Mountain View, CA, USA). 4. HistoGene LCM Frozen Section Staining Kit (Arcturus). 5. CapSure Macro LCM Caps (Arcturus). 6. Tissue Tek OCT compound (Sakura Finetek Europe, Zoeterwoude, The Netherlands). 7. Tissue Tek cryomold (Sakura Finetek Europe). 8. Dulbecco’s Modified Eagle’s Medium or phosphate-buffered saline (Invitrogen, Carlsbad, CA, USA). 9. Watchmaker’s forceps (InterFocus, Haverhill, UK). 10. Dry ice or liquid nitrogen. 11. RNaseZap (Ambion, Austin, TX, USA). 12. RNAlater-ICE (Ambion). 13. Glass slides.

2.2. RNA Isolation and Analysis

1. PicoPure RNA Isolation Kit (Arcturus). 2. RNase-Free DNase Set (Qiagen, Hilden, Germany). 3. NanoDrop ND-1000 spectrophotometer Technologies, Wilmington, DE, USA).

(NanoDrop

31  Transcriptome Profiling of Murine Spinal Neurulation…

377

4. Agilent 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA). 5. RNA 6000 Nano Assay (Agilent Technologies). 2.3. Target Preparation and Hybridization for Microarray Analysis

1. GeneChip Mouse Genome 430 2.0 Array (Affymetrix, Santa Clara, CA, USA). 2. GeneChip Fluidics Station 450 (Affymetrix). 3. GeneChip Scanner 3000 7G (Affymetrix). 4. GeneChip Hybridization Oven 645 (Affymetrix). 5. GeneChip 3¢ IVT Express Kit (Affymetrix). 6. GeneChip Hybridization, Wash, and Stain Kit (Affymetrix). 7. Thermal cycler with heated lid. 8. Magnetic stand for 96-cell plate (Ambion).

3. Methods In this section, we describe methods for collecting and cryosectioning mouse embryos, isolating cells from the closing spinal neural tube and underlying mesenchyme, and processing RNA for analysis using the Affymetrix GeneChip Mouse Genome 430 2.0 Arrays. The quality of the microarray datasets obtained is highly dependent on the state of the RNA extracted from the tissues. RNase precautions must be strictly observed when handling RNA (see Note 1). 3.1. Collection and Processing of Mouse Embryos

1. Random-bred male and female ARC(S) mice are mated overnight. Noon on the day of finding a copulation plug is designated embryonic day (E) 0.5. Collection of embryos at Mode 1 neural tube closure is carried out at E8.5, Mode 2 closure at E9.0, and Mode 3 closure at E9.5. The number of somites present in each embryo is counted under a dissection microscope to stage the development of the embryo (see Note 2). To minimize the risk of RNA degradation, dissection of embryos from the uterus and placenta is performed rapidly in Dulbecco’s modified Eagle’s medium or phosphate-buffered saline that has been prechilled to 4°C (see Note 3). 2. Collected embryos are embedded in Tissue Tek OCT compound, care being taken to position the specimens to obtain transverse sections through the posterior neuropore region. Allow the OCT to solidify by placing the cryomold in a dry ice–ethanol bath or liquid nitrogen. OCT turns white when frozen. The frozen specimens can be stored in a −70°C freezer until required.

378

S. Cao et al.

3. Mount the frozen tissue block on the cryostat, cut sections of 8–10  mm thickness, and mount the sections on untreated slides (see Note 4). 4. Stain the slides using the HistoGene LCM Frozen Section Staining Kit (8). Immerse the slides in 75% ethanol followed by RNase-free water for 30 s each. Allow the sections to stain for 20  s using the HistoGene Staining Solution. The slides are then washed, for 30  s each, in RNase-free water, 75% ethanol, 95% ethanol, and 100% ethanol. Finally, leave the slides in xylene for 5 min and then allow the slides to dry for 5  min in a fume hood before proceeding to laser capture microdissection (see Note 5). 3.2. RNA Isolation

1. Neuroepithelial cells from the closing neural tube are collected by microdissection using the PixCell II LCM System and CapSure Macro LCM Caps (Fig. 1). Mesenchymal cells

Fig. 1. Laser capture microdissection in the Arcturus PixCell II LCM System. The CapSure Macro LCM Cap with a coat of thermoplastic film is positioned over the cells of interest in the tissue section (a). Application of a low-power infrared laser results in the film adhering to these cells (b). The cells are then lifted off the section and transferred to a microcentrifuge tube for extraction of RNA for downstream applications (c).

31  Transcriptome Profiling of Murine Spinal Neurulation…

379

Fig. 2. Isolation of neuroepithelium and underlying mesenchyme by laser capture microdissection. In Mode 2 closure of the neural tube, a median hinge point (asterisk) and two dorsolateral hinge points (arrows) are present in the neuroepithelium (a). To collect the neuroepithelium, a boundary (dotted line) is drawn around the neuroepithelial cells (b). (c) Shows the tissue section that remains after removal of the neuroepithelium. The success of the microdissection is confirmed by examining the LCM cap for the presence of captured neuroepithelial cells (d). The process is repeated to collect cells from the underlying mesenchyme (e–g).

lying ventral to the neuroepithelium are isolated in a similar manner (Fig. 2). 2. Insert the LCM Cap onto a 0.5-ml microcentrifuge tube containing 50 ml of extraction buffer XB from the PicoPure RNA Isolation Kit (9). Invert the tube so that all the microdissected cells are covered by the buffer and incubate for 30 min at 42°C. 3. In the meantime, condition the purification column by incubating it with 250 ml of solution CB for 5 min at room temperature. Centrifuge at 16,000 × g for 1 min to remove the buffer. 4. Add 50 ml of 70% ethanol to the extracted cells, mix well, and then transfer the mixture to the conditioned column. Centrifuge at 100 × g for 2 min and then 16,000 × g for 30 s. 5. Add 100  ml of buffer W1 to the column, and centrifuge at 8,000 × g for 1 min. 6. Remove potential genomic DNA contamination by adding 5 ml of DNase I and 35 ml of buffer RDD from the RNasefree DNase Set. Incubate for 15  min at room temperature. Add 40  ml of buffer W1 to the column, and centrifuge at 8,000 × g for 15 s. 7. Add 100  ml of buffer W2 to the column and centrifuge at 8,000 × g for 1 min. Repeat once, but centrifuge for 2 min at

380

S. Cao et al.

Fig. 3. Representative output from the Agilent Bioanalyzer. In an intact RNA sample, two peaks are seen in the electropherogram, corresponding to the 18S and the 28S ribosomal peaks. The 18S–28S rRNA ratio is approximately 1:2, and the RIN is 10. The peak on the left is the marker peak, used for aligning the samples to be analyzed with the RNA ladder. The data are displayed as a gel-like image on the right.

16,000 × g. Discard flowthrough waste and recentrifuge for 1 min at 16,000 × g to remove all residual traces of the buffer. 8. Transfer the column to a new microcentrifuge tube. Add 11 ml of the elution buffer EB to the centre of the column membrane. Incubate for 1  min at room temperature. Thereafter, centrifuge for 1  min at 1,000 × g followed by 1 min at 16,000 × g to collect the RNA. 9. Take 1 ml of the eluted sample for RNA quantification using the NanoDrop ND-1000 spectrophotometer. An A260–A280 absorbance ratio between 1.9 and 2.1 suggests the absence of contaminants in the sample. 10. Remove 1 ml of the sample to check the integrity of the RNA sample using the Agilent 2100 Bioanalyzer and the RNA 6000 Nano LabChip Kit. The microfluidic system enables the 18S and 28S rRNA bands to be detected. An approximate 18S–28S ratio of 1:2 between the two bands suggests that the RNA is intact. The Bioanalyzer software also allows an RNA Integrity Number (RIN) to be calculated. RIN is a numerical measure of RNA integrity and ranges from 1 to 10, with 1 representing the most degraded RNA and 10 being the most intact (Fig. 3). 3.3. Target Preparation for Microarray Expression Analysis

1. Target preparation is performed using the GeneChip 3¢ IVT Express Kit (10). The recommended amount of input total RNA for the kit is 100 ng in a volume of not more than 5 ml (see Note 6). However, the kit is sufficiently robust to accommodate RNA amounts ranging from 50 to 500 ng (see Note 7). 2. The RNA sample in a total volume of 5 ml (including poly-A positive controls if desired; see Note 6) is mixed with 4 ml of

31  Transcriptome Profiling of Murine Spinal Neurulation…

381

first-strand buffer mix and 1  ml of first-strand enzyme mix. The cocktail is incubated for 2  h at 42°C to synthesize the first-strand cDNA (see Note 8). 3. In the meantime, prepare the second-strand master mix, which consists of 13 ml of nuclease-free water, 5 ml of secondstrand buffer mix, and 2  ml of second-strand enzyme mix. Add the master mix to the first-strand cDNA, and incubate for 1 h at 16°C followed by 10 min at 65°C to synthesize the second-strand cDNA. 4. Prepare the in vitro transcript (IVT) master mix, containing 4 ml of IVT biotin label, 20 ml of IVT labelling buffer, and 6  ml of IVT enzyme mix. Add the IVT master mix to the double-stranded cDNA and incubate at 40°C to generate biotin-labelled aRNA. Affymetrix recommends an incubation period of 4 h if the starting amount of RNA ranges from 100 to 500  ng, or a period of 16  h if the amount of RNA lies between 50 and 250 ng. 5. Mix 10 ml of RNA binding beads with 50 ml of aRNA binding buffer concentrate. Add this to the biotin-labelled aRNA and transfer the contents to a well of a U-bottom plate. 6. To allow the aRNA in the sample to bind to the beads, add 120 ml of 100% ethanol and gently shake the plate for at least 2 min. 7. Capture the magnetic beads on a magnetic stand for 5 min. The beads will aggregate to form a pellet at the bottom of the well. Make sure that the supernatant is clear before ­discarding it. 8. Wash the beads with 100 ml of aRNA wash solution and shake for 1 min. Recapture the beads with the magnetic stand and discard the supernatant. 9. Repeat the wash and bead recapture step. 10. Shake dry the plate for 1 min to allow all remaining ethanol to evaporate from the beads. 11. To elute the aRNA, add 50 ml of aRNA elution solution that has been prewarmed to 50–60°C to the well and shake for 3 min to disperse the beads. 12. Capture the beads on a magnetic stand. Collect the supernatant, which contains the aRNA, in a microcentrifuge tube. 13. Take 1 ml of the sample to measure the aRNA concentration using the NanoDrop ND-1000 spectrophotometer. 14. Assess the aRNA profile using the Agilent 2100 Bioanalyzer and RNA 6000 Nano LabChip Kit. It is expected that most of the aRNA will be between 600 and 1,200 nt in size, with an overall range from 250 to 5,500 nt. Examination of the aRNA concentration and profile serves as a check for the preceding aRNA amplification steps.

382

S. Cao et al.

15. Add 15 mg of aRNA to 8 ml of 5× array fragmentation buffer. Top up to 40  ml with nuclease-free water and incubate the mixture for 35 min at 94°C. Thereafter, cool the reaction on ice immediately. 16. Determine the profile of the fragmented aRNA using the Agilent 2100 Bioanalyzer. The fragments should be between 35 and 200 nt, with a peak at 100–120 nt. 3.4. Target Hybridization

1. Prepare 250  ml of the hybridization cocktail using the fragmented aRNA and the reagents in the hybridization, wash, and stain Kit (11). The cocktail consists of 12.5 mg (33.3 ml) of fragmented aRNA, 4.2 ml of control oligonucleotide B2, 12.5 ml of 20× hybridization controls, 125  ml of 2× hybridization mix, 25  ml of DMSO, and 50  ml of nuclease-free water. Heat the cocktail for 5 min at 99°C, followed by 5 min at 45°C. Centrifuge for 5 min to sediment any insoluble material in the mixture. 2. In the meantime, fill each GeneChip Mouse Genome 430 2.0 Array with 200 ml of prehybridization mix and incubate for 10 min at 45°C with rotation (see Note 9). 3. Remove the Prehybridization Mix from the microarray and replace it with the hybridization cocktail. 4. Transfer the microarray to the hybridization oven and hybridize for 16 h at 45°C with rotation at 60 rpm. 5. Wash and stain the microarray using Protocol FS450_0001 in the Affymetrix Fluidics Station 450 and scan using the Affymetrix GeneChip Scanner 3000 7G (see Note 10).

4. Notes 1. To avoid RNase contamination and consequential degradation of RNA, wear disposable latex gloves when handling samples and reagents. Sterile disposable plasticware should be used. Glassware should be baked for at least 8 h at 180°C to remove any RNase. Water used in the experiment should be nuclease-free or pretreated with 0.1% diethyl pyrocarbonate (DEPC) and autoclaved. Positive air displacement pipettes should be used together with sterile RNase-free filter tips. Benchtops should be regularly wiped down with an RNase decontamination solution such as RNaseZap (Ambion). 2. The posterior neuropore length, crown–rump length, and head length can be measured using an eyepiece graticule to help in staging embryonic development (12). There are several other parameters that are useful for staging, including flexion of the embryo and the developmental status of the brain, brachial arches, and fore- and hind-limb buds (13).

31  Transcriptome Profiling of Murine Spinal Neurulation…

383

3. RNAlater (Ambion) is commonly used to prevent RNA degradation in tissues for subsequent downstream applications. However, there have been reports that its use affects tissue consistency, ease of cryosectioning, and histological morphology (14–18). In our experiments, we snap-freeze the embryos without using RNAlater. 4. To stabilize the RNA in the tissue sections and prevent its breakdown, we add prechilled RNAlater-ICE to the sections and allow the solution to permeate the cells. Unlike RNAlater, RNAlater-ICE is designed for use with frozen samples, thereby removing the risk of RNA degradation due to thawing. 5. Nuclei are stained purple by the HistoGene stain. The cytoplasm is stained pink. 6. If the RNA sample is to be spiked with poly-A RNA positive controls, the maximum volume allowed for the sample is 3 ml. If necessary, the RNA sample can be concentrated in a vacuum concentrator set at low heat. 7. The GeneChip 3¢ IVT express kit replaces the Affymetrix one-cycle and two-cycle target labelling reagents, which were discontinued in September 2009. 8. For consistency, keep the amount of starting RNA consistent for all samples that are to be compared. Use a master mix to reduce pipetting errors, and avoid pipetting small volumes below 2 ml. A thermal cycler with a heated lid is preferred for the incubation steps. Adhere to the incubation times strictly. Process all samples to be compared at the same time using the same batch of reagents. 9. The GeneChip Test3 Array can be used to assess the quality of the labelled target before hybridizing it to the Mouse Genome 430 2.0 Array. The Test3 Array is in the 400 (micro) format, and is to be filled with 80 ml of the prehybridization mix (and subsequently, of the hybridization cocktail). By contrast, the Mouse Genome Array is in the 49 (standard) format and uses a hybridization volume of 200 ml. 10. There are numerous proprietary and open-source software packages available for analysis of microarray datasets. Examples of commercial packages include GeneSpring GX, Partek Genomics Suite, and Rosetta Resolver System. Bioconductor and dChip are the two commonly used noncommercial software programs (19–22).

Acknowledgments We thank Associate Professors Ming Teh and Wei-Yi Ong for kindly allowing us to use their laser capture microdissection

384

S. Cao et al.

s­ ystems, and their colleagues for help with operating the ­instruments. We are also grateful to Ms Song-Lin Bay for her excellent assistance in preparing the diagrams for this chapter. The work was supported by Grant R-181-000-095-112 from the Academic Research Fund, Ministry of Education, Singapore (G.W.Y.). S.C. is the recipient of a graduate research scholarship from the National University of Singapore. References 1. Gilbert, S. F. (2010) Developmental biology, 9th ed. Sinauer Associates, Sunderland. 2. Sadler, T. W. (2010) Langman’s medical embryology, 11th ed. Lippincott Williams & Wilkins, Philadelphia. 3. Greene, N. D., Stanier, P., and Copp, A. J. (2009) Genetics of human neural tube defects. Hum. Mol. Genet. 18, R113-R129. 4. Copp, A. J. and Greene, N. D. (2010) Genetics and development of neural tube defects. J. Pathol. 220, 217–230. 5. Copp, A. J., Greene, N. D., and Murdoch, J. N. (2003) The genetic basis of mammalian neurulation. Nat. Rev. Genet. 4, 784–793. 6. Harris, M. J. and Juriloff, D. M. (2007) Mouse mutants with neural tube closure defects and their role in understanding human neural tube defects. Birth Defects Res. A Clin. Mol. Teratol. 79, 187–210. 7. Shum, A. S. W. and Copp, A. J. (1996) Regional differences in morphogenesis of the neuroepithelium suggest multiple mechanisms of spinal neurulation in the mouse. Anat. Embryol. (Berl. ) 194, 65–73. 8. Arcturus (2003) HistoGene LCM Frozen Section Staining Kit. Arcturus, Mountain View. 9. Arcturus (2004) PicoPure RNA Isolation Kit. Arcturus, Mountain View. 10. Affymetrix (2009) GeneChip 3’ IVT Express Kit. Affymetrix, Sant Clara. 11. Affymetrix (2008) GeneChip Hybridization, Wash and Stain Kit. Affymetrix, Santa Clara. 12. Yip, G. W., Ferretti, P., and Copp, A. J. (2002) Heparan sulphate proteoglycans and spinal neurulation in the mouse embryo. Development 129, 2109–2119. 13. Brown, N. A. and Fabro, S. (1981) Quantitation of rat embryonic development in vitro: a morphological scoring system. Teratology 24, 65–78.

14. Wang, S. S., Sherman, M. E., Rader, J. S., Carreon, J., Schiffman, M., and Baker, C. C. (2006) Cervical tissue collection methods for RNA preservation: comparison of snap-frozen, ethanol-fixed, and RNAlater-fixation. Diagn. Mol. Pathol. 15, 144–148. 15. Stemmer, K., Ellinger-Ziegelbauer, H., Lotz, K., Ahr, H. J., and Dietrich, D. R. (2006) Establishment of a protocol for the gene expression analysis of laser microdissected rat kidney samples with affymetrix genechips. Toxicol. Appl. Pharmacol. 217, 134–142. 16. Roos-van Groningen, M. C., Eikmans, M., Baelde, H. J., de, H. E., and Bruijn, J. A. (2004) Improvement of extraction and processing of RNA from renal biopsies. Kidney Int. 65, 97–105. 17. Micke, P., Ohshima, M., Tahmasebpoor, S., Ren, Z. P., Ostman, A., Ponten, F. et al. (2006) Biobanking of fresh frozen tissue: RNA is stable in nonfixed surgical specimens. Lab Invest 86, 202–211. 18. Noriega, N. C., Kohama, S. G., and Urbanski, H. F. (2009) Gene expression profiling in the rhesus macaque: methodology, annotation and data interpretation. Methods 49, 42–49. 19. Nikolova, V., Koo, C. Y., Ibrahim, S. A., Wang, Z., Spillmann, D., Dreier, R. et  al. (2009) Differential roles for membrane-bound and soluble syndecan-1 (CD138) in breast cancer progression. Carcinogenesis 30, 397–407. 20. Li, C. and Wong, W. H. (2001) Model-based analysis of oligonucleotide arrays: expression index computation and outlier detection. Proc. Natl. Acad. Sci. U. S. A 98, 31–36. 21. Hahne, F., Huber, W., Gentleman, R., and Falcon, S. (2008) Bioconductor case studies. Springer, New York. 22. Gohlmann, H. and Talloen, W. (2009) Gene expression studies using Affymetrix microarrays. Chapman & Hall, Boca Raton.

Chapter 32 Probing the CNS Microvascular Endothelium by Immune-Guided Laser-Capture Microdissection Coupled to Quantitative RT-PCR Nivetha Murugesan, Jennifer Macdonald, Shujun Ge, and Joel S. Pachter Abstract Laser-capture microdissection (LCM) allows for retrieval of distinct populations of cells from their closely surrounding neighbors in situ. As such, LCM is highly advantageous for investigating gene expression along the central nervous system (CNS) microvascular endothelium, a tissue that shows both ­considerable segmental and regional heterogeneity. Combining immunohistochemical staining of CNS microvascular endothelial cells with immunofluorescent staining of perivascular astrocytes or smooth muscle cells, immune-guided LCM, immuno-LCM, may be coupled to downstream qRT-PCR to probe varied expression of the endothelium along the CNS microvascular tree during health and disease. Immuno-LCM/qRTPCR has been used to highlight contributions of the respective segments of the CNS microvasculature to the blood–brain barrier (BBB), and can be employed to examine changes in BBB gene expression ­during pathology. Key words: LCM, Endothelial cell, BBB, Heterogeneity, qRT-PCR

1. Introduction Increasing awareness of endothelial heterogeneity (1, 2) has focused attention on the need to examine endothelial populations in their in situ positions. Indeed, not only is there segmental heterogeneity, i.e., diversity in the phenotype of endothelial cells depending vascular segment (macrovascular vs. microvascular and subtypes of microvessels vs. others) but there is also regional heterogeneity, i.e., phenotypic diversity expressed by endothelial cells of a given segmental type depending on location in the body. Perhaps nowhere is the need to consider the issue of endothelial Graeme I. Murray (ed.), Laser Capture Microdissection: Methods and Protocols, Methods in Molecular Biology, vol. 755, DOI 10.1007/978-1-61779-163-5_32, © Springer Science+Business Media, LLC 2011

385

386

N. Murugesan et al.

heterogeneity more critical than when dealing with endothelial cells of the central nervous system (CNS). With existence of the blood–brain brain barrier (BBB) in most – but not all – regions of the CNS, and an extensive microvascular tree consisting of arterioles, capillaries, and venules having different properties, the degree of varied endothelial cell gene expression in the CNS has potential for being vast (3). When challenged to examine gene expression patterns in CNS endothelial cells, it is, therefore, significant to focus analysis on particular endothelial populations. And while separating endothelial populations by crude dissection and/or biochemical means is feasible, such efforts are extremely time-consuming and run the risk of severely altering expression patterns in the process. However, the technique of laser-capture microdissection (LCM) affords opportunity to relatively quickly isolate designated CNS endothelial populations from their in situ locals and, thereby, get a physiological “snapshot” of endothelial gene expression at a point in time. The use of LCM to isolate brain microvessels was initially described by Ball et al. (4), who used H&E staining combined with detection of endogenous alkaline phosphatase activity to highlight microvascular structures. This technique was then modified by Mojsilovic-Petrovic et al. (5), who substituted use of fluorescently conjugated lectins to rapidly identify cerebral microvessels. However, a caveat of both approaches was the intimate association of astrocyte foot processes and other perivascular cells with the endothelial cells. The close juxtaposition of the different cell types precluded isolation of endothelial cells without significant contamination by adventitial cells. Moreover, as neither alkaline phosphatase activity (6, 7) nor lectin binding sites (6, 8) are expressed at high level throughout the entire vascular tree, not all endothelial beds could be stained by such methods. To mitigate these obstacles, this laboratory developed an immunoLCM (9) approach that incorporated immunohistochemical staining of endothelial cells with an antibody to CD31 (also known as PECAM-1) – a protein ubiquitously expressed by all endothelial cells (10, 11) – in conjunction with immunofluorescent staining of perivascular cells (either astrocytes, pericytes or smooth muscle cells) (Fig. 1). Resolving endothelial and perivascular cells in this way enables the respective cell types to be enriched by LCM to high purity and, when accompanied by measurements of vessel diameter, allows endothelial cell populations in different types of vessel segments to be compared (9, 12, 13). Coupling immunoLCM to quantitative RT-PCR (immuno-LCM/qRT-PCR) can, therefore, be used to probe the extent of segmental and regional endothelial cell heterogeneity in the CNS during health and disease. Though not the focus of this report, immuno-LCM can also be extended to examine the respective endothelial and epithelial layers comprising the blood–cerebrospinal fluid barrier at the level of the choroid plexus (Fig. 2).

32  Probing the CNS Microvascular Endothelium by Immune-Guided…

387

Fig. 1. Immuno-LCM of CNS microvascular endothelial cells. Microvascular endothelial cells and astrocytes of a cortical brain section were stained by immunohistochemistry with ABC-alkaline phosphatase (black) or immunofluorescence, respectively. White arrow designates a capillary targeted by LCM. (a) Tissue before LCM. (b) Tissue following laser “shot,” revealing a halo of melted plastic film surrounding capillary. (c) Tissue after LCM, showing empty space left by captured capillary, and surrounding astrocyte processes remaining behind. (d) Purified capillary fragment deposited on HS cap.

Fig. 2.  Immuno-LCM of endothelial cells and epithelial cells of the choroid plexus. The inner capillary plexus endothelium and outer choroidal epithelium of the choroid plexus were stained by immunohistochemistry with ABC-alkaline phosphatase (black) and immunofluorescence, respectively. Triangles (upper panel) and arrows (lower panel) point to areas of tissue targeted by LCM. (a) and (d) show tissues prior to LCM. (b) and (e) show tissues after LCM. (c) and (f) show purified epithelial cells and endothelial cells, respectively, deposited onto an HS cap.

2. Materials 2.1. Preparation of Brain Tissue for Immunostaining

1. Compressed CO2 in glass cylinders is used for euthanasia, as this allows the influx of gas to the induction chamber in a controlled manner. Cages are placed in the chamber, the chamber lid closed, and 100% CO2 introduced at a rate of 10–20% of the chamber volume per minute. This rate of CO2 introduction minimizes animal distress. After the animals become unconscious, the flow rate is increased to minimize the time to death. This euthanasia procedure is in accordance with measures stipulated by the Animal Care and Use Guidelines of the University of Connecticut Health Center (Animal Welfare Assurance # A3471-01). 2. Mouse brains.

388

N. Murugesan et al.

3. Sterile surgical instruments: scalpel, forceps, scissors. 4. Isopentane (2-methylbutane; Plains, NJ).

Acros

Organics,

Morros

5. Shandon Cryomatrix embedding medium (Thermo Fisher Scientific, Waltham, MA). 6. Microtome cryostat (Thermo Electron Corporation, San Jose, CA). 7. Shandon MX35 disposable microtome blades (Thermo Fisher Scientific, Waltham, MA). 8. RNase Away (Invitrogen, Carlsbad, CA). 9. Uncoated glass slides. 2.2. Immunostaining

1. Acetone (Fisher Scientific, Pittsburgh, PA). 2. Phosphate-buffered saline (PBS), pH 7.6. 3. Diethyl pyrocarbonate (DEPC; Sigma, St. Louis, MO). 4. Antibody diluting solution: All antibodies are diluted in 1× PBS containing 0.2% Tween-20. 5. PAP pen (hydrophobic marker pen). 6. Rat anti-mouse CD31 antibody, 1:10 dilution (BD Biosciences, San Jose, CA). 7. Biotinylated rabbit anti-rat IgG antibody, 1:250 dilution (Vector Labs, Burlingame, CA). 8. Alexa 488-conjugated glial fibrillary acidic protein (GFAP) antibody, 1:5 dilution (Invitrogen, Carlsbad, CA), and fluorescein isothiocyanate-conjugated alpha smooth muscle actin (FITC-a-SMA) antibody, 1:5 dilution (Sigma, St. Louis, MO). 9. Alkaline phosphatase avidin-biotinylated enzyme ABC kit (Vector Labs, Burlingame, CA). ABC Complex: Add one drop of reagent A to one drop of reagent B. Add 2.5 ml of 1× PBS. Mix well and incubate for 30  min at RT prior to use. Prewarm an aliquot to 37°C before use. 10. Alkaline phosphatase substrate (NBT/BCIP) kit IV, (Vector Labs, Burlingame, CA). NBT/BCIP substrate solution: Add one drop of reagent 1 (BCIP) to one drop of reagent 2 (NBT). Add one drop of reagent 3 (MgCl2) and 1  ml of 0.1 M Tris–HCl, pH 9.5. Mix well. 11. RNasin Plus RNase inhibitor (Promega, Madison, WI): add to all antibody and substrate solutions to a final concentration of 0.4 U/ml (1:100 dilution of stock).

32  Probing the CNS Microvascular Endothelium by Immune-Guided…

2.3. Dehydration of Tissue Sections Prior to LCM

389

1. Distilled water (DNase, RNase free; Molecular Devices, Sunnyvale, CA). 2. Ethanol (75, 95, and 100%; Fisher Scientific, Pittsburgh, PA). 3. Xylene (Fisher Scientific, Pittsburgh, PA).

2.4. LCM

1. Pixcell IIe LCM microscope equipped with epifluorescent optics (Molecular Devices, MDS Analytical Technologies, Sunnyvale, CA). 2. Capsure HS caps (Molecular Devices, MDS Analytical Technologies, Sunnyvale, CA). 3. Sterile 0.5-ml microfuge tubes.

2.5. RNA Isolation

1. TRIzol (Invitrogen, Carlsbad, CA). 2. Chloroform (Fisher Scientific, Pittsburgh, PA). 3. Glycogen (Ambion, Austin, TX). 4. Turbo DNAse (Ambion).

2.6. qRT-PCR

1. Superscript III Reverse Transcriptase (Invitrogen). 2. SYBR Green PCR Master Mix (ABI, Foster City, CA). 3. MicroAMP Fast Optical 96-well reaction plates (ABI).

3. Methods The double-labeling, immuno-LCM approach has several methodological objectives: (1) to identify the endothelial cells for capture, (2) to identify perivascular cells that one wishes to avoid, and (3) to be able to resolve different types of vascular segments. Using a combination of immunohistochemistry (for endothelial cells) and immunofluorescence (for perivascular cells) allows for all three objectives to be readily accomplished. The process for identifying endothelial cells exploits the facts that these cells both highly express CD31 throughout the vascular tree (10, 11) and possess varying amounts of endogenous alkaline phosphatase activity (6, 7). By employing an avidin– biotin complex (ABC) immunohistochemical protocol with exogenous alkaline phosphatase to detect CD31, reaction product is generated from both endogenous and exogenous enzyme sources and intense staining is generated in even the smallest capillaries. Moreover, use of NBT/BCIP as a chromogenic substrate for alkaline phosphatase results in deposition of a colored precipitate that, while relatively stable to xylene dehydration, is much more efficiently solubilized during tissue extraction than is DAB (used in conjunction with horseradish peroxidase), allowing for greater recovery of RNA and protein (14).

390

N. Murugesan et al.

3.1. Retrieval of Brain Tissue and Preparation of Tissue Sections

1. Sacrifice the animal by euthanasia and spray the head with 75% ethanol. 2. Carefully remove the whole brain from the cranium using sterile surgical instruments. 3. Immediately immerse the brain tissue into isopentane freezing medium (precooled on dry ice) for 1–2 min. 4. Remove the frozen brain tissue from isopentane with a clean spatula. 5. Embed the frozen specimen in embedding medium, on dry ice. 6. Store the embedded tissue at −80°C until ready for sectioning. 7. Prepare 7-mm frozen brain sections of desired orientation (e.g., coronal, sagittal, etc.). 8. Affix them onto uncoated clean glass slides and keep on dry ice until cutting session is completed. 9. Place the tissue sections in a clean slide box precleaned with RNase Away. 10. Store at −80°C. Tissue sections are used for LCM within 1 week of sectioning to minimize degradation of RNA with increase in storage time. If it is desired to perform LCM at later times, it is best to leave brain tissue unsectioned at −80°C. Tissue left in this way has been analyzed by immunoLCM coupled to qRT-PCR after being stored frozen for as long as 1 year.

3.2. Fixation of Frozen Tissue Sections and Quick Immunostaining for LCM

1. Remove frozen 7-mm tissue section from −80°C storage and quickly thaw to room temperature (see Note 1). 2. Fix the tissue section in 75% ethanol for 3 min. 3. Briefly air-dry the tissue and draw a water repellent circle around tissue section with PAP hydrophobic marker pen. 4. Incubate with monoclonal rat anti-mouse CD31 antibody (1:10 dilution) for 3  min at room temperature (RT). All antibodies are diluted in 1× PBS + 0.2% Tween-20. Add RNasin Plus RNase inhibitor (1:100 dilution) to all staining reagents. 5. Wash briefly by dipping the slide in 1× PBS for 5 s. 6. Incubate with biotinylated rabbit anti-rat antibody (1:250 dilution) for 2 min at RT. 7. Wash briefly by dipping the slide in 1× PBS for 5 s. 8. Add prepared avidin-biotinylated enzyme complex (ABC), prewarmed to 37°C, to the tissue for 3 min. Preparation of ABC: Add one drop of reagent A+ one drop of reagent B to 2.5 ml of 1× PBS. Mix well and incubate for 30 min at RT prior to use. Prewarm an aliquot to 37 °C before using in step 8 (see Notes 2 and 3).

32  Probing the CNS Microvascular Endothelium by Immune-Guided…

391

9. Wash briefly by dipping the slide in 1× PBS for 5 s. 10. Add prepared NBT/BCIP substrate solution and incubate until purple color develops (~5–6 min). Preparation of NBT/BCIP solution: Add one drop of reagent 1 (BCIP) + one drop of reagent 2 (NBT) + one drop of reagent 3 (MgCl2) to 1 ml of 0.1 M Tris–HCl, ph 9.5. Mix well. 11. Wash briefly by dipping the slide in 1× PBS for 5 s. 12. Add Alexa-488 GFAP antibody (1:5 dilution) or FITC-aSMA antibody (1:5 dilution) for 5 min. 13. Wash briefly by dipping the slide in 1× PBS for 5 s. 3.3. Dehydration of Tissue Sections Prior to LCM

1. Dip the immunostained slide in 75% ethanol for 10  s (see Note 4). 2. Transfer to 95% ethanol for 30 s. 3. Immerse in first 100% ethanol for 60 s. 4. Immerse in second 100% ethanol for 90 s. 5. Transfer to first xylene wash for 2 min. 6. Transfer to second xylene wash for 3 min. 7. Air-dry the slide for 5 min after the final xylene wash.

3.4. LCM

1. The dehydrated double-immunostained slide is placed on the microscope stage. 2. Vacuum seal is used to keep the slide in place. 3. An HS Capsure LCM cap is placed in position, in the path of the laser beam, above the tissue section. 4. Both bright-field and epifluorescence optics is used to visualize the CD31 positive microvessels (purple) and the green fluorescent GFAP in apposing astrocytic processes (as shown in Fig. 1). 5. A 7.5-mm laser spot size at a power range of 65–80 mW and pulse duration of 550–800 ms are used. This combination of parameters allows for efficient retrieval of microvessel sample and limits contamination from perivascular cells. 6. Microvessels typically <10 mm in diameter, mainly representing capillaries, are targeted. 7. The cap containing LCM captured material is fit onto a clean 0.5-ml microfuge tube and stored at −80°C until further analysis (see Note 5).

3.5. RNA Isolation

1. Add 100 ml TRIzol (warmed up to room temp.) to each cap and scrape out tissue well after the LCM process. Store at −80°C. 2. Pool together (up to 500 ml TRIzol total final volume) all the caps (with required amount of laser “shots”).

392

N. Murugesan et al.

3. Add 150 ml of chloroform and vortex well. 4. Keep on ice for 5 min. 5. Spin at 8,765 ´ g in the cold room for 5 min. 6. Carefully aspirate out the upper aqueous phase (~220  ml), without touching the interface and transfer it to a new tube. 7. Add another 133 ml of TRIzol to the pink organic phase and vortex. 8. Keep on ice for 5 min and spin down as in step 5. 9. Aspirate out the aqueous phase (~70–80 ml) and add to the previously collected aqueous phase in step 6. 10. Measure the total volume of the aqueous phase carefully and note it down (keep it at ~300 ml). 11. Add 4 ml of glycogen (5 mg/ml) and vortex well. 12. Add 0.15× volume of salt (2 M sodium acetate, pH 4) and vortex well. 13. Add an equal volume of isopropanol (1:1) as the aqueous phase measured in step 10. Vortex well. 14. Keep at −20°C for 30 min. 15. Spin down at 8,765 ´ g in the cold room for 15  min (see Note 6). 16. To the pellet, add 500 ml of 75% ethanol gently. Vortex well! (this will help in removing any traces of phenol if present!). 17. Spin at 8,765 ´ g for 3–5 min in the cold room. 18. Remove supernatant and air-dry the white pellet. 19. Resuspend the dry pellet in 12 ml of nuclease free H2O. 20. Proceed to DNase treatment step. 3.6. qRT-PCR

1. The isolated RNA is first treated with Turbo DNase according to manufacturer’s protocol at 37°C for 30 min to remove any contaminating genomic DNA if present. 2. DNase treatment is followed by reverse transcription and first strand cDNA synthesis using SuperScript III Reverse Trasnscriptase and manufacturer’s protocol, with a modification in the extension temperature to 42°C for 60  min. The resulting cDNA is stored at −20°C until ready for qRT-PCR. 3. Measurement of cDNA levels is performed by qRT-PCR using an ABI PRISM 7900 Sequence Detection System Version 2.3 adopting SYBR green fluorescence to detect relative amplicon amounts. A master mix consisting of each primer pair (final concentration of 250 nM), SYBR green (2× stock, ABI), and water is made for every gene of interest to be analyzed. 22.5 ml of this mix is aliquoted into each well of a 96-well Microamp qRT-PCR reaction plate.

32  Probing the CNS Microvascular Endothelium by Immune-Guided…

393

4. The prepared cDNA is then diluted appropriately, and 2.5 ml of it is then added to each well bringing the total volume in each well to a total of 25 ml. 5. Appropriate controls are used (i.e., no-template control, noreverse-transcriptase control). Each primer pair’s efficiency (E) is calculated by E = 10(−1/slope)−1, where slope refers to the slope of the standard curve from the log serial fold dilutions of template cDNA against their Ct values (15). Relative quantification is calculated using the formula: (1 + Etarget)Ct(target)/ (1 + Eref)Ct(ref) and ref: RPL-19, target: gene of interest, Ct: threshold cycle. Dissociation curves for each gene are carefully analyzed to ensure specific amplicon replication. 6. The qRT-PCR protocol used is as follows: (a) 50°C for 2 min (b) 95°C for 10 min (c) 95°C for 15  sec, 60°C for 1  min, 95°C for 15  sec; Repeat 40× (d) 95°C for 15 sec, 60°C for 15 sec, 95°C for 15 sec.

4. Notes 1. Tissue sections should be used within 1 week of sectioning to minimize degradation of RNA. Unsectioned brain tissue has been kept frozen at −80°C for up to 1 year without significant loss of qRT-PCR signal. 2. Immunohistochemistry using alkaline phosphatase rather than horseradish peroxidase is preferred, as the NBT/BCIP reaction product of the former enzyme is more soluble in TRIzol, allowing for greater recovery of RNA (this is also the case for recovery of proteins for proteomic analysis (14)). Additionally, endogenous alkaline phosphatase activity in endothelial cells amplifies the immune signal. 3. Prewarm the ABC (avidin-biotinylated enzyme complex) reagent to 37°C prior to use. This will shorten the time for colorimetric development of the reaction product. 4. Extensive dehydration of tissue sections is critical for high efficiency capture during LCM. Alcohol solutions and xylene should be changed frequently, and use of a room dehumidifier is highly recommended. 5. Loosely adhered tissue, nonspecifically stuck to the cap, can be removed using a clean Post-itTM, prior to storage. 6. If the solution appears frozen or gel-like, thaw and vortex until it liquefies completely.

394

N. Murugesan et al.

Acknowledgments This work was supported by grants RO-1-MH54718 and R21NS057241 from the National Institutes of Health, and grant PP-1215 from the National Multiple Sclerosis Society to J.S.P. References 1. Aird, W.C. (2007). Phenotypic heterogeneity of the endothelium: I. Structure, function, and mechanism. Circ. Res. 100, 158–173. 2. Aird, W.C. (2008). Endothelium in health and disease. Pharmacol. Rep. 60, 139–143. 3. Ge, S., Song, Li. and Pachter, J.S. (2005). Where is the blood-brain barrier…. really? J. Neurosci. Res. 74, 421–427. 4. Ball, H.J., McParland, B., Driussi,C. and Hunt, N.H. (2002). Isolating vessels from mouse brain for gene expression analysis using laser capture microdissection. Brain Res. Brain Res. Protoc. 9, 206–213. 5. Mojsilovic-Petrovic J., Nesic M., Pen A., Zhang W., Stanimirovic D. (2004). Development of rapid staining protocols for laser-capture microdissection of brain vessels from human and rat coupled to gene expression analyses. J. Neurosci. Methods 133, 29– 48. 6. Vorbrodt, A.W. (1988). Ultrastructural cytochemistry of blood-brain barrier endothelia. Prog. Histochem. Cytochem. 18, 1–99. 7. Ge, S. Song, L. and Pachter, J.S. (2005). Where is the blood-brain barrier…really? J. Neurosci. Res. 79, 421–427. 8. Smolkova, O., Zavadka, A., Benkston, P. and Lustyk, A. (2001). Cellular heterogeneity of rat vascular endothelium detected by HPA and GS 1 lectin-gold probes. Med. Sci. Monit. 7, 659–668. 9. Kinnecom, K. and Pachter, J.S. (2005). Selec­ tive capture of endothelial cells and perivascular

cells from brain microvessels by laser capture microdissection. Brain Res. Protoc. 16, 1–9. 10. Ilan, N. and Madri, J. (2003). PECAM-1: old friend, new partners. Curr. Opin. Cell Biol. 15, 2795–2806. 11. Feng, D., Nay, J.A., Pyne, K., Dvorak, H.F. and Dvorak, A.M. (2004). Ultrastructural localization of platelet endothelial adhesion molecule (PECAM-1, CD-31) in vascular endothelium. J. Histochem. Cytochem. 52, 87–101. 12. Macdonald, J., Murugesan, N. and Pachter, J.S. (2008). Validation of immuno-laser capture microdissection coupled with quantitative RT-PCR to probe blood-brain barrier gene expression in situ. J. Neurosci. Method. 174, 219–226. 13. Macdonald, J., Murugesan, N. and Pachter, J.S. (2009). Endothelial cell heterogeneity of blood-brain barrier gene expression along the cerebral microvasculature. J. Neurosci. Res. 88, 1457–74. 14. Lu, C., Murugesan, N., Macdonald, J., Wu, S.-L., Pachter, J.S. and Hancock, W.S. (2008). Analysis of brain microvascular endothelium using immuno-laser capture microdissection coupled to a hybrid LTQ-FT-MS proteomics platform. Electrophoresis 29, 2689–2695. 15. Pfaffl, M.W. (2001). A new mathematical model for relative quantification in real-time RT-PCR. Nuc. Acids Res. 29, e45.

Chapter 33 Laser-Capture Microdissection for Factor VIII-Expressing Endothelial Cells in Cancer Tissues Tomoatsu Kaneko, Takashi Okiji, Reika Kaneko, Hideaki Suda, and Jacques E. Nör Abstract The immunostaining based Laser-capture microdissection (LCM) method called immune-LCM allows us to quantify the mRNA. Immune-LCM has recently been introduced to enhance identification of cells carrying a particular protein from frozen tissue samples. We have recently performed the immune-LCM of formaldehyde-fixated, paraffin-embedded tissues immunostained with a monoclonal antibody Factor VIII. This method could be useful for quantitative gene expression analysis in blood vessels from tumors of patients that have been treated with antiangiogenic drugs, allowing for validation of the effect of drug on the expected targets. Such capability might be exceedingly useful for the evaluation of the bioactivity of new drugs. This method is also useful to compare gene expression patterns in tumor cells versus endothelial cells during tumor progression or tumor angiogenesis. Key words: Laser-capture microdissection, Endothelial cells, Gene expression, Immune-LC, FFPE tissues

1. Introduction Laser-capture microdissection (LCM) allows microscopic procurement of specific cell types from tissue sections that can then be used for gene expression analysis (1). In conventional LCM, frozen tissues stained with hematoxylin are normally used for molecular analysis. LCM in combination with antibody staining, called immune-LCM, has recently been introduced to enhance identification of cells carrying a particular protein (2, 3). However, the method has been limited to frozen tissue samples. We, here, describe a protocol for immune-LCM of formaldehyde-fixated, paraffin-embedded (FFPE) tissues immunostained with a monoclonal antibody Factor VIII (4). By this method, it was possible to ­compare

Graeme I. Murray (ed.), Laser Capture Microdissection: Methods and Protocols, Methods in Molecular Biology, vol. 755, DOI 10.1007/978-1-61779-163-5_33, © Springer Science+Business Media, LLC 2011

395

396

T. Kaneko et al.

the expression levels of various genes in endothelial cells and those in nonendothelial cells within surrounding tissues, by means of RT-PCR and real-time PCR. This method may be ideally suited for the analysis of relatively rare cell types within a tissue and should improve on our ability to perform differential diagnosis of pathologies as compared to conventional LCM (5).

2. Materials 1. Polyethylene naphthalate slides for LCM (Leica, Bannock­ burn, IL). 2. Mayer’s hematoxylin or Gill No.3¢ hematoxylin (SigmaAldrich, St. Louis, MO). 3. Anti-factor VIII (Lab Vision, Lab Vision, Fremont, CA). 4. Biotinylated anti-mouse IgG (Vector laboratories, Burlin­ game, CA). 5. Elite ABC kit (Vector laboratories, Burlingame, CA). 6. RNAlater® (Ambion, Austin, TX). 7. DAB substrate kit (Vector laboratories, Burlingame, CA). 8. TRIZOL® Reagent (Invitrogen, Carlsbad, CA). 9. RNeasy Mini Kit (Qiagen, Gaithersburg, MD). 10. SuperScript one-step reverse transcription PCR (RT-PCR) and Platinum Taq kit (Invitrogen, Carlsbad, CA). 11. The E-Gel® iBaseTM Power System (Invitrogen Carlsbad, CA). 12. E-GEL® (2 or 1.5% Agarose, Invitrogen, Carlsbad, CA). 13. Spectrophotometer (BioPhotometer plus, Eppendorf, Hamburg, Germany). Note: (a) Disposable gloves should always be worn to avoid contamination with RNA and RNases. (b) The use of sterile, disposable plasticware and filtered pipette tips is recommended to prevent cross-contamination with RNases.

3. Methods 3.1. Tissue Fixation and Paraffin Embedding

1. Fix tissues with 10% neutral buffered formaldehyde for 8–16 h at 4°C. 2. Dehydrate in 70% ethanol for 30 min, 90% ethanol for 1 h, and 95% ethanol for 30 min at 4°C, followed by 100% ethanol 3 × 1 h at room temperature.

33  Laser-Capture Microdissection after Immunostaining from FFPE Tissues

397

3. Immerse two times in xylene for 1 h at room temperature. 4. Immerse four times in infiltrating paraffin for 30 min at 58°C. 5. Immerse in embedding paraffin (see Note 1). 3.2. Paraffin Block Storage

FFPE blocks may be stored at room temperature, although we recommend that the blocks be stored at 4°C. The RNA from the blocks remain stable for at least 1 year and is likely stable for longer periods; however, RNA stability would be optimal the sooner the tissue can be used.

3.3. Sectioning and Mounting

1. Cut sections on a microtome with a new sterile disposable blade (6–10 mm thick). 2. Float paraffin ribbons on 43°C RNase-free (DEPC-treated) water. 3. Mount the sections on poly l-lysine coated glass foiled polyethylene naphthalate (PEN) slides for LCM. 4. Dry the slides in a 35°C incubator for 6 h.

3.4. Slide Storage

3.5. Staining 3.5.1. Staining: Nuclear Staining by Hematoxylin

We recommend that the slides be stored at 4°C and LCM be performed as soon as possible. The slides should be processed within a week after preparation. 1. Deparaffinize the slides twice with 3-min xylene washes at room temperature. 2. Wash with 100% ethanol for 3 × 30 s, 90% ethanol for 30 s, 70% ethanol for 1 min, and DEPC-treated water for 30 s at 4°C. 3. Immerse in Mayer’s hematoxylin for 20–30  s or Gill No.3¢ hematoxylin for 5–10 s at room temperature. 4. Wash with DEPC-treated water for 30 s at 4°C. 5. Air-dry the slide for 1–3 h at 4°C.

3.5.2. Staining: Immunostaining of Factor VIII

A protocol of avidin-biotin peroxidase complex (ABC) method is shown. 2–4 slides per session are easy to manage. 1. Deparaffinize the slides twice with 3-min xylene washes at room temperature. 2. Wash with 100% ethanol for 2 × 30 s, 90% ethanol for 30 s, 70% ethanol for 1 min, and RNase-free PBS for 3 × 30 s at 4°C. 3. Pretreat sections with 0.125% trypsin for 30  min at room temperature. 4. Wash three times with RNase-free PBS for 30 s at 4°C. 5. Block endogenous peroxidase activity with 3% hydrogen peroxide in methanol for 3 min at 4°C.

398

T. Kaneko et al.

6. Wash three times with RNase-free PBS for 30 s at 4°C. 7. Incubate with the primary antibody (anti-Factor VIII, diluted 1:100) for 16 h at 4°C, or for 30 min at room temperature. 8. Wash three times with RNA-free PBS for 30 s at 4°C. 9. Incubate with biotinylated anti-mouse IgG (diluted 1:500) for 1 h at 4°C. 10. Wash three times with RNA-free PBS for 30 s at 4°C. 11. Incubate with avidin–biotin-peroxidase complex (Elite ABC kit) for 1 h at 4°C. 12. Wash three times with RNA-free PBS for 30 s at 4°C. 13. Develop in diaminobenzidine-H2O2 solution (DAB substrate kit) for 3 min at room temperature. 14. Wash three times with RNA-free PBS for 30 s at 4°C. 15. Air-dry the stained slide for 1–3 h at 4°C. 3.6. Laser-Capture Microdissection (LCM) 3.6.1. Leica AS LMD System

3.6.2. LCM for Factor VIII Immunostained Cancer Tissue

The Leica AS LMD system uses a pulsed 337-nm UV laser on an upright microscope. The laser beam can be moved with a softwarecontrolled mirror system that allows us to select target cells and tissues. Target cells can be preselected on a monitor with a freehand drawing tool, and then the computer-controlled mirror moves the laser beam along the preselected path, and the target cells are excised from the section. The dissected part then falls into a PCR tube under gravity. The collection by gravity ensures quick and contamination-free processing of the dissected tissue sections. The thickness and width of the cutting line can be controlled for each tissue. A two-step dissection strategy is described, as previously informed (4, 5). 1. Dissect blood cells in factor VIII-positive capillaries to exclude these cells. 2. Dissect and collect Factor VIII-positive endothelial cells (approximately 1,500) into individual tubes filled with RNAlater® and place immediately on ice. Cells without a nucleus should be excluded (Fig. 1) (see Note 2).

3.6.3. Number of Cells to Microdissect

Dissection of approximately 50–100 endothelial cells enables gene expression analysis. If approximately 1,000 endothelial cells are collected, approximately 10 mg of RNA will be obtained.

3.7. RNA Extraction by TRIZOL® Reagent (Invitrogen)

1. Incubate the samples with 30–100 ml of TRIZOL® Reagent for 5 min at room temperature. 2. Add an equal amount of chloroform.

33  Laser-Capture Microdissection after Immunostaining from FFPE Tissues

399

Fig. 1. Immune-LCM used for retrieval of factor VIII-positive endothelial cells from FFPE tissue sections. (a) Factor VIII+ endothelial cells in the head and neck carcinoma. (b) Retrieval of factor VIII-positive endothelial cells. This step is performed after dissecting blood cells in factor VIII-positive capillaries.

3. Close sample tubes securely. Shake tubes for 30–60  s and incubate them for 5 min on ice. 4. Centrifuge the samples at 15,000 ´ g for 15 min at 4°C. 5. Transfer the aqueous phase to a fresh tube. (Following centrifugation, the mixture separates into three layers; a lower red, phenol-chloroform phase, an interphase, and a colorless upper aqueous phase. RNA is contained exclusively in the aqueous phase). 6. Add an equal amount of isopropyl alcohol (RNA precipitant) to the aqueous phase transferred, incubate at 4°C for 10 min, and centrifuge at 13,000 rpm for 15 min at 4°C. RNA will precipitate to form a white small dotted pellet on the side and bottom of the tube. The RNA precipitate is sometimes invisible after centrifugation. 7. Discard the supernatant gently. 8. Add 70–75% ethanol in DEPC-treated water and mix the sample by vortex. 9. Centrifuge at 7,000–9,000 ´ g for 5 min at 4°C. 10. Discard the supernatant gently. 11. Add 70–75% ethanol in DEPC-treated water to the RNA pellet. 12. The RNA pellet can be stored at −20°C. 3.8. RNA Cleanup with RNeasy Mini Kit

All processes can be performed at room temperature. 1. Add 10 ml of b-mercaptoethanol per 1 ml of buffer RLT (see Note 3). 2. Adjust the sample to a volume of 100  ml with RNase-free water. Add 350 ml of buffer RLT and mix briefly. 3. Add 250 ml of 100% ethanol to the diluted RNA and mix by pipetting without centrifuge.

400

T. Kaneko et al.

4. Transfer the sample (total 700  ml) to an RNeasy Mini spin column placed in a 2-ml supplied collection tube. Centrifuge for 15 s at 8,000 × g. Discard the flow-through. 5. Reuse the collection tube (see Note 4). 6. Add 350 ml of buffer RW1 to the RNeasy spin column. Close the lid gently, and centrifuge for 15 s at 12,000 ´ g to wash the spin column membrane and discard the flow-through. 7. Reuse the collection tube. 8. Add 10 ml of DNase I stock solution to 70 ml of buffer RDD that is supplied with the RNase-Free DNase Set. Mix by gently inverting the tube and centrifuge briefly to collect residual liquid from the sides of the tube. 9. Add the DNase I incubation mix (80  ml) directly to the RNeasy spin column membrane and place at room temperature for 15 min. 10. Add 350 ml of buffer RW1 to the RNeasy spin column. Close the lid gently and centrifuge for 15 s at 10,000 rpm. Discard the flow-through. 11. Add 500 ml of buffer RPE to the RNeasy spin column and centrifuge for 15  s at 10,000  rpm and discard the flowthrough. 12. Reuse the collection tube. 13. Add 500 ml of buffer RPE to the RNeasy spin column and centrifuge for 2 min at 10,000 rpm. 14. Place the RNeasy spin column in a new 2-ml collection tube, and discard the old collection tube with the flow-through, and centrifuge at 13,000 rpm for 1 min. 15. Place the RNeasy spin column in a new 1.5-ml collection tube. Add 30–50 ml of RNase-free water directly to the spin column membrane and centrifuge for 1 min at 10,000 rpm to elute the RNA. 3.9. RT-PCR for Bcl-2 and GAPDH

1. Determine RNA purity (Mo, SGman: OD260/280 = 1.9–2.1) photometrically with a spectrophotometer. 2. Primer sets of human Bcl-2 and glyceraldehyde-3-phosphate dehydrogenase (GAPDH): Bcl-2, sense, CTGCGAAGAACCTTGTGTGA antisense TGTCCCTACCAACCAGAAGG; GAPDH, sense, CATGGCCTCCAAGGAGTAAG antisense, AGGGGTCTACAGGCAACTG. cDNA synthesis and PCR amplification are done in single tubes with SuperScript one-step reverse transcriptionPCR (RT-PCR) and Platinum Taq kit. 3. Add the following to the microcentrifuge tubes placed on ice.

33  Laser-Capture Microdissection after Immunostaining from FFPE Tissues

401

4. Prepare reaction cocktails 2× Reaction mix 25 ml 1× Template RNA (20–400 ng) Sense Primer (10 mM) 1 ml Antisense Primer (10 mM) 1 ml RT/Platinum® Taq mix 1 ml DEPC-treated water to 50 ml. 5. Gently mix all the components and centrifuge for 10  s at 4°C. 6. Program the thermal cycler: (a)cDNA synthesis and pre-denaturation: Perform one cycle of the following: 45°C for 10 min 94°C for 1 min (b)PCR amplification: 30–35 cycles of: Denature, 94°C for 15 s, Anneal, 60°C for 30 s, Extend, 70°C for 1 min/ kb (c)Final extension: 1 cycle of 72°C for 5–10  min use 30  s denaturation instead of 15 s (see Note 5). 3.10. Running the Gel

The E-Gel® iBaseTM Power System and E-GEL® (2 or 1.5% Agarose) are used to run the PCR products. 1. Open the package and remove an E-GEL® cassette. 2. Slide the E-GEL® cassette into electrodes of the E-Gel® iBaseTM Power System. 3. Select the program PRE-RUN and run the gel for 2 min (see Note 6). 4. Take out the comb gently and load the 15 ml PCR products with 5 ml loading buffer. 5. Load 20 ml molecular weight markers (see Note 7). 6. Run the gel after selecting the appropriate program according to the gel types. 7. The run will stop automatically. Then, remove an E-GEL® cassette from the power system and observe with a transilluminator (Fig. 2).

4. Notes 1. Paraffin processes may be performed by using an automated tissue-processing machine. 2. The dissected samples can be stored at 4°C to −80°C until enough material is microdissected and all experimental samples are ready.

402

T. Kaneko et al.

Fig. 2. RT-PCR to evaluate Bcl-2 or GAPDH expression levels in Factor VIII-expressing endothelial cells in the head and neck carcinoma or normal mucosa adjacent to the head and neck carcinoma.

3. Buffer RLT containing b-mercaptoethanol can be stored at room temperature for up to 1 month. b-mercaptoethanol should be treated in a fume hood and appropriate protective clothing should be worn. 4. After centrifugation, carefully remove the RNeasy spin column from the collection tube so that the column does not contact the flow-through. Be sure to empty the collection tube completely. 5. The above-mentioned cycling conditions are set for the LCM analysis for Bcl-2 and GAPDH but may be altered for each experiment. The PCR-amplified samples can be stored at −20°C. 6. Do not remove the comb of the E-GEL® cassette until step 4. 7. Add 20 ml DEPC-treated water to any empty wells. References 1. Emmert-Buck MR, Bonner RF, Smith PD, Chuaqui RF, Zhuang Z, Goldstein SR, Weiss RA, Liotta LA. (1996) Laser capture microdissection. Science 274, 998–1001. 2. Fend F, Emmert-Buck MR, Chuaqui R, Cole K, Lee J, Liotta LA, Raffeld M. (1999) Immuno-LCM: laser capture microdissection

of immunostained frozen sections for mRNA analysis. Am J Pathol 154, 61–66. 3. Stoebner PE, Le Gallic L, Berthe ML, Boulle N, Lallemant B, Marque M, Gaspard C, Delfour C, Lavabre-Bertrand T, Martinez J, Meunier L. (2008) Decreased expression of thymidine phosphorylase/platelet-derived

33  Laser-Capture Microdissection after Immunostaining from FFPE Tissues endothelial cell growth factor in basal cell carcinomas. Exp Dermatol 17, 908-915. 4. Kaneko T, Zhang Z, Mantellini MG, Karl E, Zeitlin B, Verhaegen M, Soengas MS, Lingen M, Strieter RM, Nunez G, Nör JE. (2007) Bcl-2 orchestrates a cross-talk between endothelial and tumor cells that promotes tumor growth. Cancer Res 67, 9685–9693.

403

5. Kaneko T, Okiji T, Kaneko R, Suda H, Nör JE. (2009) Gene expression analysis of immunostained endothelial cells isolated from formaldehyde-fixated paraffin embedded tumors using laser capture microdissection--a technical report. Microsc Res Tech 72, 908–912.

wwwwwww

Chapter 34 Laser-Capture Microdissection and Analysis of Liver Endothelial Cells from Patients with Budd–Chiari Syndrome Selcuk Sozer and Ronald Hoffman Abstract Myeloproliferative neoplasms (MPN) are clonal hematological malignancies that are frequently ­associated with an acquired somatic mutation in JAK2 (JAK2V617F). Patients with MPN are at a high risk of developing thrombotic events. Endothelial cell (EC) abnormalities are thought to contribute to this prothrombotic state. Budd–Chiari syndrome (BCS) and portal vein thromboses have been reported to be associated with JAK2V617F positive hematopoiesis. We explored whether JAK2V617F was present in ECs within the vessels of polycythemia vera (PV) patients with BCS using laser-capture microdissection followed by nested PCR or real-time RT-PCR. The presence of JAK2V617F in both ECs and hematopoietic cells belonging to BCS patients with PV indicates that ECs from these patients are involved by the malignant process and that in this subpopulation of patients the disease may originate from a cell common to hematopoietic and endothelial cells. Key words: Myeloproliferative neoplasms, Laser-capture microdissection, Polymerase chain reaction

1. Introduction Laser-capture microdissection (LCM) is powerful tool capable of isolating cells of interest from archived paraffin-embedded tissue specimens. Immunolabelling with a specific antibody may provide an easy and effective approach with which to enhance the identification of cells belonging to a particular lineage thereby permitting the use of morphological and histological criteria as well as cell phenotype for this purpose. This method also circumvents the possibility of contamination by surrounding cells by allowing the selection of single cell. In addition, LCM permits microscopic localization of cells at the site of a pathological process and allows cells to be isolated from that site as well as sites not involved by the

Graeme I. Murray (ed.), Laser Capture Microdissection: Methods and Protocols, Methods in Molecular Biology, vol. 755, DOI 10.1007/978-1-61779-163-5_34, © Springer Science+Business Media, LLC 2011

405

406

S. Sozer and R. Hoffman

pathological process for comparative studies. We employed this technique to isolate the endothelial cells from formalin-fixed, paraffin-embedded (FFPE) liver biopsy specimens from patients with Budd–Chiari and polycythemia vera (PV). Patients with myeloproliferative neoplasms (MPN) are at a high risk of developing thrombotic events (1–3), and it is believed that endothelial cell (EC) abnormalities might contribute to this prothrombotic state (4–6). Budd–Chiari syndrome (BCS) and portal vein thrombosis have been reported to be associated with JAK2V617F-positive hematopoiesis (7, 8). Since, LCM provides a technique whereby DNA and RNA from isolated cells remain intact after isolation, the molecular characteristics of the selected cells can be further studied (9). We explored whether JAK2V617F was present in EC in the sections of archived liver biopsy tissue specimens from BCS and hepatoportal sclerosis patients with or without PV using LCM followed by nested PCR or real-time RT-PCR (10–13).

2. Materials 2.1. Reagents

1. DNA extraction kit (PicoPure DNA extraction kit, Arcturus Molecular Devices; Sunnyvale, CA). 2. RNA extraction kit (PicoPure RNA extraction kit, Arcturus Molecular Devices). 3. Mayer’s hematoxylin solution (Sigma, St Louis, MO). 4. Eosin Y solution, alcoholic (Sigma). 5. Scott’s Tap Water Substitute Blueing Solution (Fisher; Pittsburgh, PA). 6. 100% ethanol. 7. 95 and 70% ethanol (dilute 100% ethanol with nuclease-free water to obtain 95 and 70% ethanol solutions). 8. Xylene. 9. Distilled water, nuclease-free (Life Technologies, Carslbad, CA). 10. Anti-CD34 monoclonal primary antibody (clone QBEnd/10; Ventana Medical System’s CONFIRM™, Tucson, AZ). 11. Conjugated streptavidin horseradish peroxidase secondary antibody (Promega, Madison, WI). 12. Primers for JAK2V617F, GAPDH, VEGFR2, VE-cadherin, vWF, CD45, CD14, and albumin. The sequences of the primers: JAK2V617F: P1:5¢-GAT CTC CAT ATT CCA GGC TTA CAC A-3¢; P1r: 5¢-TAT TGT TTG GGC ATT GTA ACC TTC T-3¢; P2: 5¢-CCT CAG AAC GTT GAT GGC A-3¢;

34  Laser-Capture Microdissection and Analysis of Liver Endothelial Cells¼

407

P2r:5¢-ATT GCT TTC CTT TTT CAC AAG A-3¢; Pnf:5¢-AGC ATT TGG TTT TAA ATT ATG GAG TAT ATG- 3¢ and Pmr:5¢-GTT TTA CTT ACT CTC GTC TCC ACA AAA-3¢ GAPDH: 5¢-GTC TTC TCC ACC ATG GAG AAG GCT-3¢ and 5¢-CAT GCC AGT GAG CTT CCC GTT CA-3¢ KDR (VEGFR2): 5¢-CCG GAG TGA CCA AGG ATT GT-3¢ and 5¢CCG CTT TAA TTG TGT GAT TGG A-3¢ VE-cadherin: 5¢-GGC AAG ATC AAG TCA AGC GTG-3¢ and 5¢-ACG TCT CCT GTC TCT GCA TCG-3¢. vWF: 5¢-CTA TGG CTT TGT GGC CAG GAT-3¢ and 5¢CAT AAG GGT CCG AGG TCA AGG T-3¢. CD45: 5¢-CAG GCA TGG TTT CCA CAT TC-3¢ and 5¢CTA CAA ATA TTG GTT CGC TGC-3¢. CD14: 5¢-CGG CGG TGT CAA CCT AGA G-3¢ and 5¢GCC TAC CAG TAG CTG AGC AG-3¢ Albumin: 5¢-GCT GCC ATG GAG ATC TGC TTG A-3¢ and 5¢GCA AGT CAG CAG GCA TCT CAT C-3¢. 13. SuperScript™ III Carlsbad, CA).

Reverse

Transcriptase

(Invitrogen,

14. QuantiTect SYBR Green PCR Kit ( Qiagen, Valencia, CA). 15. Agarose. 16. DNA ladder; 50 bp (Fermantes). 17. Ethidium bromide(Bio-Rad, Hercules, CA). 18. Loading dye: Burnie, Ma).

Bromophenol

blue

(Fermentes,

Glen

19. 1× TAE (Tris-acetate–EDTA buffer) (Fisher, catalog number:) or prepare: First, prepare 1 l of 50× TAE of stock solution (240 g Tris base, 57.1 g glacial acetic acid, and 100 ml 0.5 M EDTA). Dissolve the ingredients in 0.9 l of distilled water. Then, take 20 ml of 50× TAE and add 980 ml of distilled water for 1 l of 1× TAE. 20. TE buffer (10 mM Tris–HCl, pH 7.0, 1 mM EDTA). 2.2. Equipment

1. Laser-capture microdissection apparatus: PixCell II and LaserCapture Microdissection system (Arcturus Molecular Devices, Sunnyvale, CA). 2. Microtome. 3. −80°C freezer. 4. Fume hood. 5. Microcentrifuge. 6. Incubation oven.

408

S. Sozer and R. Hoffman

7. PCR machine. 8. Agarose gel electrophoresis. 9. UV light box. 10. ABI Prism 7900 Sequence Detection System (Applied Biosystems, Carlsbad, CA). 2.3. Laser Microdissection Reagents

1. Uncoated, uncharged, glass microscope slides, 25 × 75  mm (Fisher). 2. CapSure Macro LCM Caps (Arcturus Molecular Devices). 3. 0.5-ml microcentrifuge tubes (Applied BioSystems). 4. RNase-free pipette tips. 5. RNase AWAY® (Life Technologies). 6. Disposable gloves. 7. Kimwipes® or similar lint-free towels. 8. Microslide box – plastic. 9. 2–20 ml pipettor. 10. 20–200 ml pipettor. 11. PrepStrip™ Tissue Preparation Strip ( Arcturus).

3. Methods The methods for selecting endothelial cells by immunological reagents are complex due to the wide spectrum of endothelial cell surface markers that have been associated with EC. Selection of a pure population of EC requires the combination of many phenotypic markers in order that a pure population of EC can be identified with some degree of certainty. Selection of EC from archived specimens is especially challenging, since tissues that are embedded in paraffin are not viable, thereby eliminating the possibility of performing functional assays. Although immunohistochemistry alone remains a powerful tool for identifying cells belonging to a particular lineage, it is limited by its inability to provide a means of isolating such cell populations to permit molecular analysis. LCM overcomes this barrier by permitting cell isolation. Using LCM for a selection also enables one to perform many molecular assays, since the DNA, RNA, and protein extracted from such cells are suitable for analysis. Therefore, we employed this technique to demonstrate whether EC from the PV patients are JAK2V617F-positive. Endothelial cells in the liver sections were easily identified following hematoxylin and eosin staining or anti-CD34 antibody staining. CD34 is an ectopeptidase expressed by human hematopoietic stem cells and progenitor cells, as well

34  Laser-Capture Microdissection and Analysis of Liver Endothelial Cells¼

409

as EC. The microdissected cells that were isolated with LCM from FFPE liver samples were processed for either DNA or RNA. The DNA of the microdissected EC was used for the detection of JAK2V617F mutation by allele-specific nested PCR. The RNA of the microdissected EC was used for the detection of transcripts characteristic of EC by real-time RT-PCR which is required indeed to be certain that the selected cells are truly EC. 3.1. Section Cutting

1. Wipe all surfaces of microtome, water bath with RNase AWAY® followed by 100% ethanol. 2. Cut 5-mm serial blocks of archived FFPE sections of liver biopsy specimens. 3. Mount the tissue sections on uncharged “non-plus” glass slides, deparaffinize the tissue before staining, and rehydrate to allow staining of the tissue elements (see Note 1). 4. Put the slides in xylene two times for 5 min and work under a fume hood (xylene is toxic, so do not inhale). 5. Transfer the slides to graded ethanol series starting from 100% ethanol for 30 s, 95% ethanol for 30 s, and 70% ethanol for 30 s (see Note 2). 6. Wash the slides with dH2O for 10 s. 7. Stain the slides either with hematoxylin and eosin staining or an anti-CD34 monoclonal primary antibody followed by staining with conjugated streptavidin horseradish peroxidase secondary antibody (see Note 3).

3.2. Hematoxylin and Eosin Staining

1. Transfer the slides to Mayer’s hematoxylin for 15 s. 2. Wash the slides with dH2O for 10 s. 3. Wash the slides with Scott’s Tap Water Substitute for 10 s. 4. 70% ethanol for 10 s. 5. Eosin Y for 10 s. 6. Ethanol series starting from 95% ethanol for 10 s, 95% ethanol for 10 s, and 100% ethanol for 30 s. 7. Treat the slides two times with xylene for 30 s (see Note 4). 8. Air-dry the slides as quickly as possible.

3.3. CD34 Staining

1. FFPE liver sections were subjected to immunostaining with the streptavidin–peroxidase technique, using the automated NexES IHC staining system (Ventana, Tucson, Arizona, USA).

3.4. Endothelial Cell Isolation by LCM

1. EC were identified by their fusiform nuclei and their location along the lining of the terminal hepatic venules following hematoxylin and eosin staining (Fig. 1), as well as their dark brown color following anti-CD34 staining (see Note 5).

410

S. Sozer and R. Hoffman

Fig. 1. Before (a) and after (b) laser capture of EC within a terminal hepatic venule identified by hematoxylin and eosin staining. The arrow shows the site of microdissection. The rest of the photo shows plates of hepatocytes arranged in trabeculae (magnification ×400).

2. Hepatocytes were identified by their morphological ­characteristics including their round centrally located nuclei and abundant cytoplasm with a trabecular arrangement. 3. Dust or debris on the slide may be removed by blotting the dried slide with a PrepStrip sample preparation strip. 4. Single-cell microdissection was achieved by adjusting the power and duration settings. The 7.5-mm spot size setting was selected and laser power and duration manually adjusted. Suggested settings for single-cell microdissection were power 45 mW and duration 650 ms using Macro caps and 20× or 40× objectives after the initial laser focus. In general, a lower pulse duration and higher power setting enabled a more precise capture of cells (see Note 6). 5. If the vessels that are microdissected for EC are filled with the blood cells, the blood cells should be microdissected on another cap and the vessel lumen should be cleaned first. Since blood cells are loosely attached to the slides as compared to the cells belonging to the tissues, during the laser capture process the blood cells can stick to the cap and contaminate the microdissected cells. If the blood cells which contain mononuclear cells are microdissected, these cells can also be used to determine the JAK2V617F status of the patient if not previously known. Since erythrocytes are enucleated, they can be ignored during the capture process (Fig. 2). 6. After microdissection, place the cap that contains the microdissected cell on a clean slide and look through the microscope to confirm morphology and number. 7. At least 10 EC and 10 hepatocytes from each biopsy specimen should be captured (Fig. 1). 8. During the collection, keep the caps that contain cells on the dry ice and as soon as the collection procedure has been completed, proceed with the isolation protocols.

34  Laser-Capture Microdissection and Analysis of Liver Endothelial Cells¼

411

Fig. 2. A hepatic venule containing blood cells before (a) and after (b) laser capture. The microdissected blood cells on the LCM cap is shown in panel (c). Cells are stained with hematoxylin and eosin (magnification ×200).

3.5. Detection of the JAK2V617F by Nested AlleleSpecific PCR

1. Insert the polymer end of the cap into the top of a 500-ml microcentrifuge tube which is filled with the extraction buffer. The sample is now ready for extraction, or alternatively the cap–tube assembly (without extraction buffer) may be stored for extraction at a later date. (Microdissected cells for DNA analysis may be stored at room temperature for up to 1 week before extraction). 2. DNA was extracted from laser-captured cells using a PicoPure™ DNA Extraction Kit according to the manufacturer’s instructions. 3. To detect the JAK2V617F mutation, nested allele-specific PCR was performed. In the first PCR, a 521-bp DNA fragment containing the V617F mutation site will be amplified from the gDNA using primers P1 and P1r. After 35 cycles consisting of 30 s at 950C 30 s at 600C, and 30 s at 720C, the PCR product is ready to undergo the second PCR (see Note 7). 4. Take 0.5 ml of the first PCR product and further amplify it by nested and allele-specific primers P2, P2r, Pnf, and Pmr following 35 cycles of 95°C for 30 s, 59°C for 25 s, and 72°C for 25 s. 5. To observe the final PCR results, prepare a 2.5% agarose gel with 1 ml of ethidium bromide (10 mg/ml). Leave the gel for at least 30 min, preferably 1 h for it to set (running DNA into a gel too soon can cause smearing of the bands). 6. Run the gel with 1× TAE buffer. 7. Load some (10 ml) of PCR product with 0.2 volumes of loading buffer (e.g., 2 ml into a 10 ml sample). Use a 50-bp marker to distinguish the wild-type JAK2 from JAK2V617F. 8. Set up the power source and run the gel at 5  V/cm (see Note 8). 9. Stop the gel when the bromophenol blue has run 3/4 the length of the gel and analyze the result on the UV light box (UV is carcinogenic and must not be allowed to shine on naked skin or eyes. Wear glasses, gloves, and long sleeves).

412

S. Sozer and R. Hoffman

Fig. 3. JAK2V617F status of LCM captured hepatocytes (H) and endothelial cells (EC) from a patient with portal vein thrombosis (PVT) due to hepatoportal sclerosis (a) and a patient with BCS and PV (b) by allele-specific nested PCR.

10. To document the result, take a picture of the gel. 11. The nested PCR product has a size of 453  bp. An allele-­ specific JAK2V617F positive has a 279-bp product and an allele-specific wild has a 229-bp product. Cells classified as homozygous for JAK2V617F contain only the 279-bp band, whereas heterozygous cells have both the 279-bp and 229-bp bands (Fig. 3). 12. Extract the total RNA from laser-captured cells using a PicroPure™ RNA Isolation Kit (Arcturus Bioscience). RNA should be extracted immediately after microdissection. Condensation in the microcentrifuge tube during storage may be a potential source of RNase contamination (perform the optional DNase-treatment step, which removes contaminating gDNA in order for PCR applications to be possible. After the second wash step, make sure the filter that contains RNA is dry and no wash buffer is present before elution of the RNA) (see Note 9). 3.6. Real-Time Reverse Transcription-PCR Assay of LaserCaptured Cells

1. Before starting RT-PCR, incubate the RNA at 70°C in TE for 1 h (see Note 10). 2. Synthesize the first-strand complementary DNA (cDNA) from total RNA with SuperScript™ III reverse transcriptase according to the manufacturer’s instructions. 3. Use 2–4 ml from the cDNA product to run real time RT-PCR. 4. The amplification and the cycling conditions of GAPDH, VE-cadherin, VEGFR-2, vWF, CD45, and CD14 are as follows: GAPDH: 30 cycles of 94°C for 1 min, 57°C for 50 s, 72°C for 1 min, and 72°C for 10 min. KDR (VEGFR2), VE-cadherin, vWF: 95°C for 10  min, 45 cycles of 95°C for 30 s, 59°C for 40 s, 72°C for 1 min. CD45: 94°C for 2 min, 31 cycles of 94°C for 1 min, 57°C for 50 s, 72°C for 1 min, and 72°C for 10 min. CD14 and albumin: 94°C for 2 min and 30 cycles of 94°C for 30 s, 55°C for 30 s, 72°C for 45 s, and 72°C for 10 min. The final PCR products were analyzed on 2% agarose gel.

34  Laser-Capture Microdissection and Analysis of Liver Endothelial Cells¼

413

Table 1 The transcripts present in the microdissected hepatocytes (H) and endothelial cells (EC) from archived liver biopsy tissue specimens of a PV patient with BCS BCS Transcripts

H

EC

VEGFR-2

(−)

(+)

vWF

(−)

(+)

VE-cadherin

(−)

(+)

CD14

(−)

(−)

CD45

(−)

(−)

Albumin

(+)

(−)

GAPDH

(+)

(+)

5. Perform the real-time reverse transcription-PCR (RT-PCR) assays with the ABI Prism 7900 Sequence Detection System and detect the PCR products with the use of SYBR green technology (QuantiTect SYBR Green PCR Kit, Qiagen). 6. Perform PCR amplification in a total volume of 20 ml; each reaction should contain 2× SYBR Green mix, 200 nM primers, and cDNA. 7. All assays should be performed in duplicate at least, and for all assays a positive control, a negative control, and a no-template control should be included. 8. Measure a cycle threshold (CT) value to assess EC or hematopoietic cell mRNA expression ( Table 1) (see Note 11).

4. Notes 1. Complete deparaffinization is essential for DNA and RNA extraction. 2. The diluted ethanol must be freshly prepared. 3. Total staining time should be as short as possible. Excessive staining prevents visualization of dry, noncoverslipped slides. 4. Stained slides can be stored in xylene for a period of time, not to exceed 5 min. Longer slide storage in xylene compromises the RNA integrity. Once removed from xylene, microdissection should be completed within 2  h to avoid partial RNA

414

S. Sozer and R. Hoffman

degradation during the staining process. Xylene dissolves the LCM cap polymer. It is important to keep the slide completely dry before cap placement for microdissection. 5. During the EC capture, the morphology of the CD34+ cells as well as their location should be taken into the consideration. The vascular wall is a complex structure with at least three different cell types: ECs, pericytes, and vascular wall resident progenitor cells that are CD34+ VEGFR2+ TIE2+ CD31− CD45+ (11, 12). These resident progenitor cells reside in a distinct zone of the vascular wall, which is localized beneath the smooth muscle and adventitial layer of human adult vascular wall, termed the vasculogenic zone. Such cells are potentially capable of differentiating into mature endothelial cells, hematopoietic and local immune cells, such as macrophages. Since the cells analyzed in the livers of patients with BCS were localized to the tunica intima of the vascular wall rather than the vasculogenic zone of the vascular wall and were negative for CD45, which is expressed in hematopoietic cells belonging to all lineages, it is unlikely that these cells were contaminated with the vascular wall resident progenitor cells. 6. Since coverslips and mounting media cannot be used during the actual microdissection, the color and detail of a given stained tissue may be lost as the slide dries. The visualization of the stained sections can be improved by adding one drop of DNase/RNase free water between the cap and the slide. It decreases the power of the laser during the cell capture. 7. Allele specific nested PCR should be performed in a very clean environment. Since the second PCR is preformed with the product of the first PCR, the augmentation of any contamination either in the first or second step has a dramatic effect on the results. 8. For the optimal adjustment of the voltage of the electrophoresis apparatus, measure the distance between the electrodes of the apparatus. If they are 10 cm apart, then run the gel at 50 V. Avoid running the gel above 5 V/cm, since the agarose may heat up and begin to melt, which might limit the resolution of the gel. 9. RNA isolation should be performed as soon as the LCM is completed. Longer incubation times increase the risk of contamination and RNase activity. 10. Degradation of tissue RNA might be due to an excessive time delay before fixation, prolonged fixation, long-term preservation, low efficiency of RNA extraction because of cross-linking with proteins, and impaired reverse transcriptase reaction by formalin-induced modification of extracted RNA. To overcome the degradation of RNA, heat treatment of RNA prior

34  Laser-Capture Microdissection and Analysis of Liver Endothelial Cells¼

415

to reverse transcription has been proposed13. For example, it was reported that the chemical modification of all four bases of RNA by fixation in phosphate-buffered formalin can be reversed to some extent by incubation in TE buffer at 70°C for 1  h, resulting in the restoration of template activity of RNA in RT-PCR. 11. The fluorescence emitted by the reporter dye above baseline signal is detected using software in real-time, recorded and represented as the CT. The CT value can be used to determine the presence or absence of the PCR product in the sample. The detection of a CT value for GAPDH represented as means of documenting that sufficient amount of RNA was available. The absence of a CT value for the gene of interest in the presence of a GAPDH signal represented a negative signal for the gene of interest, while the presence of a CT value for the gene of interest in the presence of GAPDH was indicative of a positive signal. References 1. Tefferi A, Elliott M. (2007) Thrombosis in myeloproliferative disorders: prevalence, prognostic factors, and the role of leukocytes and JAK2V617F. Semin Thromb Hemost 33, 313–320. 2. Wolanskyj AP, Schwager SM, McClure RF, Larson DR, Tefferi A. (2006) Essential thrombocythemia beyond the first decade: life expectancy, long-term complication rates, and prognostic factors. Mayo Clin Proc 159–66. 3. Finazzi G, Rambaldi A, Guerini V, Carobbo A, Barbui T. (2007) Risk of thrombosis in patients with essential thrombocythemia and polycythemia vera according to JAK2 V617F mutation status. Haematologica. 92, 135–136. 4. Robertson B, Urquhart C, Ford I, et al (2007). Platelet and coagulation activation markers in myeloproliferative diseases: relationships with JAK2 V6I7 F status, clonality, and antiphospholipid antibodies. J Thromb Haemost. 5, 1679–1685. 5. Bellucci S, Michiels JJ. (2006) The role of JAK2 V617F mutation, spontaneous erythropoiesis and megakaryocytopoiesis, hypersensitive platelets, activated leukocytes, and endothelial cells in the etiology of thrombotic manifestations in polycythemia vera and essential thrombocythemia. Semin Thromb Hemost. 32, 381–398. 6. Duda DG, Fukumura D, Jain RK. (2004) Role of eNOS in neovascularization: NO for

endothelial progenitor cells. Trends Mol Med. 10, 143–145. 7. Kiladjian JJ, Cervantes F, Leebeek FW, et  al. (2008) The impact of JAK2 and MPL mutations on diagnosis and prognosis of splanchnic vein thrombosis: a report on 241 cases. Blood. 111, 4922–4929. 8. Patel RK, Lea NC, Heneghan MA, et al. (2006) Prevalence of the Activating JAK2 Tyrosine Kinase Mutation V617F in the Budd-Chiari Syndrome. Gastroenterology. 130, 2031–2038. 9. Espina V, Wulfkuhle JD, Calvert V. et  al. (2006) Laser-capture microdissection. Nature Protocols. Nat Protoc. 1, 586–603. 10. Sozer S, Fiel MI, Schiano T, et al. (2009) The presence of JAK2V617F mutation in the liver endothelial cells of patients with Budd-Chiari syndrome. Blood. 113, 5246–5249. 11. Zengin E, Chalajour F, Gehling UM, et  al. (2006) Vascular wall resident progenitor cells: a source for postnatal vasculogenesis. Development. 133, 1543–1551. 12. Ergun S, Tilki D, Hohn HP, Gehling U, Kilic N. (2007) Potential implications of vascular wall resident endothelial progenitor cells. Thromb Haemost. 98, 930–939. 13. Masuda N, Ohnishi T, Kawamoto S, Monden M, Okubo K. (1999) Analysis of chemical modification of RNA from formalin-fixed samples and optimization of molecular biology applications for such samples. Nucleic Acids Res. 27, 4436–4443.

wwwwwww

Chapter 35 Laser-Capture Microdissection of Hyperlipidemic/ApoE−/− Mouse Aorta Atherosclerosis Michael Beer, Sandra Doepping, Markus Hildner, Gabriele Weber, Rolf Grabner, Desheng Hu, Sarajo Kumar Mohanta, Prasad Srikakulapu, Falk Weih, and Andreas J.R. Habenicht Abstract Atherosclerosis is a transmural chronic inflammatory condition of small and large arteries that is associated with adaptive immune responses at all disease stages. However, impacts of adaptive immune reactions on clinically apparent atherosclerosis such as intima lesion (plaque) rupture, thrombosis, myocardial infarction, and aneurysm largely remain to be identified. It is increasingly recognized that leukocyte infiltrates in plaque, media, and adventitia are distinct but that their specific roles have not been defined. To map these infiltrates, we employed laser-capture microdissection (LCM) to isolate the three arterial wall laminae using apoE−/− mouse aorta as a model. RNA from LCM-separated tissues was extracted and largescale, whole-genome expression microarrays were prepared. We observed that the quality of the resulting gene expression maps was compromised by tissue RNA carried over from adjacent laminae during LCM. To account for these flaws, we established quality controls and algorithms to improve the predictive power of LCM-derived microarray data. Our approach creates robust transcriptome atlases of normal and atherosclerotic aorta. Assessing LCM transcriptomes for immunity-related mRNAs indicated markedly distinctive gene expression patterns in the three laminae of the atherosclerotic aorta. These mouse mRNA expression data banks can now be mined to address a wide range of questions in cardiovascular biology. Key words: Atherosclerosis, Artery tertiary lymphoid organ (ATLO), Adventitial inflammation, Immune response, Laser-capture microdissection, Microarrays, Transcriptome atlas, Media, Plaque, Adventitia

1. Introduction Most studies of atherosclerosis to date have focused on intima lesions (plaques), which are composed of lipid-laden macrophage/ foam cells, T cells, and smooth muscle cells (SMCs) (1–5). However, it is increasingly recognized that atherosclerosis is not Graeme I. Murray (ed.), Laser Capture Microdissection: Methods and Protocols, Methods in Molecular Biology, vol. 755, DOI 10.1007/978-1-61779-163-5_35, © Springer Science+Business Media, LLC 2011

417

418

M. Beer et al.

limited to the intima but extends to the media and adjacent adventitia (6–10). We noticed that media SMCs beneath plaques – but not media segments in plaque-free areas – become activated and express a distinct set of immune response-related genes including VCAM1 and the lymphorganogenic chemokines, i.e., CXCL13 and CCL21. These data indicated that media activation may contribute to adventitial T-cell infiltration (7). Over time, chronic transmural inflammation triggers robust T- and B-cell responses and highly structured artery tertiary lymphoid organs (ATLOs) emerge in the lamina adventitia. ATLOs are newly formed aggregates of adaptive immune cells consisting of follicular dendritic cells in activated germinal centers and B cell follicles such as proliferating centrocytes and peripheral IgD+ B cells, separate T cell areas containing bona fide CD11c+/MHC-II+ dendritic cells and foxp3+ T regulatory cells, and peripheral clusters of plasma cells (9). Thus, as the disease continues to spread, distinct tissue compartments with corresponding unique innate and adaptive immune cells emerge in the diseased arterial wall. As advances in LCM methodology have allowed to examine cell–cell interactions in a variety of physiological and disease conditions ex vivo in considerable detail (11–14), we set out to employ LCM to the apoE−/− aorta (see Note 1) to obtain gene expression maps (see Note 2) of the arterial wall (Fig.  1). The long-term goal of LCM of the arterial wall and atherosclerosis is to identify differentially expressed genes in different artery compartments and to delineate disease-relevant genes. During the course of these studies, we noticed that microarrays prepared from LCM-derived media of the aorta, which consists of only 4–6 layers of SMCs, express variable amounts of neuronal mRNAs such as myelin basic protein (mbp) and proteolipid protein 1 (plp1) (9). Immunohistochemical analyses demonstrated that mbp, plp1, and other neuronal genes are exclusively expressed by neurons and their axon networks in the adventitia and are absent in the media or plaque (not shown). These data led us to systematically examine mRNA cross-contamination in LCM-derived aorta microarrays (see Note 1) and to devise algorithms to correct for these methodological inaccuracies in our model (see Note 2). Our approach yields robust transcriptome atlases of normal and diseased mouse aortae.

Fig. 1. (continued) These data indicated cross-contamination of RNA from adjacent LCM-tissues (see text and Fig. 2 below). (f) Applying strict filter conditions and a two-sided t-Test at p value P £ 0.01 for differentially expressed genes in wild-type adventitia, ATLO, and plaque and subsequent inspection of GO term cytokine activity, preferential ATLO genes are shown (top 10) as heat map. ccl11 (chemokine, C-C motif, ligand); il33, interleukin 33; cxcl13 (chemokine, C-X-C motif, ligand); ccl19 (chemokine, C-C motif, ligand); ccl8 (chemokine, C-C motif, ligand); ltb (lymphotoxin b); gdf10, growth differentiation factor; tnfsfl13b, tumor necrosis factor (ligand) superfamily member 13b (aliases: BAFF, B-cell activation factor); cxcl9 (chemokine, C-X-C motif, ligand); cxcl12 (chemokine, C-X-C motif, ligand).

35  Laser-Capture Microdissection of Hyperlipidemic/ApoE−/− Mouse…

419

Fig. 1. Approach to establish LCM-derived transcriptome atlases of normal wild-type and atherosclerotic apoE−/− mouse aorta. (a) 3D schematic view of the abdominal aorta segment used to LCM-isolate RNA using 10 mm fresh-frozen tissue sections from wild-type and apoE−/− aorta (9). (b) Dense elastin fibres denote the adventitia–media or media–plaque borders in phase-contrast optics for laser guidance on 10 mm tissue sections on a membrane slide, thereby avoiding immunohistochemical staining. (c) LCM-isolated adventitia, media, and plaque fragments are collected in Trizolcontaining Eppendorf cups to obtain 30–120 ng total RNA for each sample. (d) Following RNA extraction, RNA quality assessment of LCM-derived tissue pools of individual mice, RNA integrity is examined by RNA gel electrophoresis and its amount is quantified; high-quality RNA is used to prepare expression microarrays (see Note 3). (e) Strict filter conditions were applied and a paired two-sided t-Test at p value P £ 0.01 for differentially expressed genes was performed (see Subheading 3.4). Data are shown as heat map of wild-type adventitia and corresponding media of four individual aged mice reveal differential gene expression of neuronal genes, mbp (four probe sets) or plp1 (two probe sets) (microarray data were deposited at http://www.ncbi.nlm.nih.gov/geo/info/MIAME.html in National Center for Biotechnology Information´s gene expression omnibus, GEO accession number GSE21419) in adventitia (A) versus media (M).

420

M. Beer et al.

Fig. 2. Gene frequency histograms of relative expression of LCM-derived aorta media microarrays of four individual wild-type mice. Mbp is exclusively expressed in the adventitia as revealed by immunohistochemistry (not shown) and serves as an internal control gene of individual aorta segments as several probe sets are available on the corresponding array. (a–d) Data of four individual aorta segments are shown as frequency histograms (divided in 50 sectors) of relative media expression. For these individual aorta segments, the array data were separately filtered to remove genes with low expression in adventitia and media. Note large number of filtered genes expressed in both media and adventitia. In each pair of aorta adventitia and media microarrays, four mbp probe set log signal intensities show significant changes by paired two-sided t-Test P £ 0.01. For each of four mbp probe sets (k), the factor FMbp(k) was determined (defined as the factor that describes the degree of media contamination by adventitia across the paired arrays per mouse shown as relative media expression): FMbp(k) = EM(Mbp(k))/(EA(Mbp(k)) + EM(Mbp(k))); EM(Mbp(k)) is the expression value of Mbp for mouse k measured in the media (M); EA(Mbp(k)) is the adventitia (A) value. Means of the resulting four FMbp(k) values yielded the FMbp correction factor ± SD (shown in the upper right quadrant of each panel). Correspondingly, F values for other transcripts were determined, i.e., plp probe set log signal intensities show significant changes by paired two-sided t-Test P £ 0.01 and four transcripts (selected from the original gene expression maps; see GEO GSE21419) preferentially expressed in the media are indicated at right: Acta2 (actin, aortic smooth muscle, aorta); hspb7 (heat shock protein family, member 7, cardiovascular); sgcg (sarcoglycan, gamma, dystrophin-associated glycoprotein); dsp, (desmoplakin), and many more (not shown).

2. Materials 2.1. LCM Procedure

1. O.C.T.® compound (Sakura Finetek). 2. Membrane slides (MembraneSlide 1.0 PEN, Carl Zeiss MicroImaging Gmbh). 3. Trizol (Invitrogen, Carlsbad, CA).

35  Laser-Capture Microdissection of Hyperlipidemic/ApoE−/− Mouse…

2.2. Preparation of RNA and Quality Control

421

1. RNeasy® Micro Kit (Qiagen, Valencia, CA). 2. NanoDrop ND spectrophotometer device (Thermo Scientific, Pittsburgh, PA). 3. Agilent RNA 6000 Pico Kit (Agilent, Santa Clara, CA). 4. Agilent Bioanalyzer 2100 (Agilent Technologies).

2.3. Preparation of Microarrays from Small Amounts of RNA

1. Superscript II, E.coli ligase, RNAse H, T4 DNA polymerase (Invitrogen, Carlsbad, CA). 2. p(A) RNA (component of the RNeasy® Micro Kit, Qiagen). 3. ENZOTM High Yield In Vitro Transcription Kit(ENZO Life Sciences, Lausen Switzerland). 4. Concentrated T7 RNA polymerase (Ambion, Austin, TX). 5. RNeasy® Mini kit Qiagen). 6. GeneChip® Hybridization, Wash, and Stain Kit, (Affymetrix, Santa Clara, CA). 7. GeneChip® mouse genome 430A 2.0 arrays, (Affymetrix).

2.4. Mice

1. Mice on the C57BL/6J background were purchased from The Jackson Laboratory (Bar Harbor). 2. Mice were housed in a pathogen-free environment on a 12-h light–dark cycle and fed a standard rodent chow (Altromin). 3. Animal procedures were approved by the animal Care and Use Committee of Thuringia. 4. Mice were killed at 78 weeks of age. The aorta was perfused in situ to remove blood using EDTA/PBS for 10 min, and care was taken to preserve the adventitia. Adipose tissue, lymph nodes, and large paraaortic ganglia were removed using forceps with the help of a Zeiss microscope.

3. Methods Normal and diseased arteries consist of distinct tissue compartments, referred to as laminae intima, media, and adventitia (Fig. 1). A major goal of cardiovascular research is to understand the functional impact and mechanisms of interaction of arterial wall cells (SMC, endothelial cells) with leukocyte subsets (macrophages/foam cells, T lymphocytes, and other immune cells) during atherosclerosis at a molecular level (1–6). While intima plaques and media SMC have been studied intensively in the past, few investigators took notice of the lamina adventitia. Indeed, the adventitia has somewhat escaped attention possibly because it has largely been viewed as a bystander tissue whose major function is to supply arteries with oxygen and nutrients through vasa vasora.

422

M. Beer et al.

Though morphological studies performed almost 50 years ago on human coronary arteries revealed that advanced atherosclerosis in the intima is associated with a marked inflammatory response (small round cell infiltrates probably representing T cells; see ref. 9 for further details) in adjacent adventitia, its systematic study represented considerable experimental hurdles. Recently, the adventitia regained attention as evidence has been obtained to suggest that adaptive (auto)immune responses against arterial wall (auto)antigens may be carried out in the adventitia (7–10). Ultimately, it is important to identify molecular mechanisms of cell–cell interactions in each of the three arterial wall laminae during physiological tissue homeostasis and disease progression. This can be accomplished in a variety of ways including confocal immunofluorescence microscopy, in situ hybridization, Western blotting, and other methods to visualize mRNAs and proteins. However, cell–cell interactions in atherosclerosis appear to be remarkably complex, while visualizing gene expression by traditional methods inherently remains both descriptive and strongly biased. Thus, methods to overcome these limitations are needed. Since arterial wall inflammation and the impact of immune responses in atherosclerosis are only beginning to be understood at the molecular level, we sought to generate large scale mRNA expression maps of LCM-derived (11–14) aorta tissue compartments. Here, we report on studies to apply LCM to the mouse aorta (9, 15, 16). 3.1. LCM Procedure (see Note 4)

1. Heat membrane slides to 180°C for 4 h to remove RNase. 2. Store RNase-free membrane slides at −20°C. 3. Prepare fresh-frozen cryotome sections (10 mm) from aorta embedded in Tissue Tek (maintained at −80°C); use every tenth section for Oil-red O/hematoxylin staining to identify ATLOs or other arterial wall structures. 4. Transfer five sections into Eppendorf cup to examine RNA integrity using Bioanalyzer electrophoresis (first RNA integrity control to test RNA quality before LCM). 5. Place 5 × 20 mm sections onto cooled membrane slides. 6. Save 3 × 10 mm sections for immunofluorescence analyses and store at −20°C. 7. Fix membrane slides for 25  s in ice-cold acetone within cryotome chamber. 8. Dry section on warm plate for 5–10 min at 37°C under argon. Dry sections were placed in an exsiccator under argon until LCM to avoid moisture uptake. Prepare approximately 200 sections on an appropriate number of membrane slides to get a yield of approximately 30–120 ng RNA for each of the arterial wall laminae.

35  Laser-Capture Microdissection of Hyperlipidemic/ApoE−/− Mouse…

423

9. Use PALM MicroBeam system, Carl Zeiss MicroImaging GmbH. 10. Separate manually isolated LCM material using forceps to transfer to Eppendorf cup containing 900 ml Trizol. 11. Subsequent to a complete LCM procedure of an individual mouse aorta, transfer five sections into separate Eppendorf cup to examine RNA integrity using Bioanalyzer electrophoresis (second RNA integrity control to test RNA degradation during LCM). 3.2. Preparation of RNA

1. Collect all LCM samples plus both samples for RNA quality control in each 900 ml Trizol (if necessary store them for up to 48 h at 4°C to assemble a complete set). 2. Add 180 ml chloroform and shake. 3. Centrifuge for 15 min at 15,000 × g 12°C. 4. Transfer clear supernatant into Eppendorf tube, add 280 ml ethanol. 5. Transfer to RNeasy® MinElute Column. 6. Centrifuge 1 min 10,000 × g RT. 7. Discard flowthrough and wash column with 500  ml RW1 buffer. 8. Centrifuge 1 min 10,000 × g RT. 9. Change tube and wash column with 500 ml RPE buffer. 10. Centrifuge 1 min 10,000 × g RT. 11. Discard flowthrough and wash the columns with 500 ml 80% ethanol. 12. Centrifuge 1 min 10,000 × g RT. 13. Discard throughput and open cap. 14. Centrifuge for 2 min 13,000 × g RT. 15. Place 1.5 ml nonstick tube under column; elute with 12 ml RNase-free H2O. 16. Centrifuge 2 min 13,000 × g RT. 17. Apply a 1:5 dilution to an Agilent RNA6000 Pico Chip to determine RNA quality and estimate concentration. 18. Store RNA at −80°C.

3.3. Preparation of Microarrays from Small Amounts of LCM Material (see Note 4)

1. Reverse-transcribe 30–120 ng RNA using Superscript II. T7 promoter is introduced by tailed p(dT) primer. 2. Synthesize second strand with the help of DNA-polymerase 1, RNAse H, and E. Coli ligase. To obtain blunt ends, T4 DNA Polymerase is used at the end of the reaction.

424

M. Beer et al.

3. Purify cDNA by phenol–chloroform extraction followed by ammonium acetate precipitation. To enhance recovery add 5 mg of p(A) RNA as a carrier. 4. Perform in  vitro transcription using ENZO™ High Yield in vitro Transcription Kit. To obtain optimal linear amplification from the low starting LCM material run in  vitro transcription for 68  h: After 16, 24, 40, 48, 64  h, freshly add concentrated T7 polymerase and nucleotides to the reaction mixture (see Note 5). 5. Purify cRNA using the RNeasy® Mini Kit. Measure yield with NanoDrop ND-1000 spectrophotometer. Estimate cRNA yield by subtracting 4 mg for the carrier added in Step 3. 6. Fragment the probe (0.5–12  mg cRNA) in the presence of Mg2+ and K+ for 30 min at 94°C. Make sure that the amount of cRNA does not differ more than twofold for different probes within one experiment. 7. Hybridize arrays for 20 h in hybridization buffer containing control oligoB2, Hering sperm DNA, acetylated BSA, and Tween-20 at 45°C under vigorous agitation. With cRNA yields less than 2 mg, hybridization time is extended to 24 h and temperature is reduced to 43°C. 8. Stain in the fluidic station (400 or 450) according to Affymetrix protocols using the Affymetrix staining kit. 9. Scan the Arrays immediately after staining. Scale raw data to 500 and export for further processing. 3.4. Microarray Data Analysis

1. Calculate signal intensities from the raw data and scale to an array trimmed mean of 500. Use all further steps with R and Bioconductor (17, 18). 2. Normalize logarithmized signals across arrays using quantile normalization (19). 3. Filter data prior to statistical analyses to remove genes with low expression or without significant changes between control group and analysing group. For upregulation, include probe sets if a minimum of 2 arrays (of three) or 3 arrays (of four) is present (detection p value P £ 0.05) and a minimum of 2 arrays show a log signal ³ log2(200). For downregulation, include probe sets if a minimum of 2 or 3 arrays was present in the control group (detection p value P £ 0.05) and a minimum of 2 or 3 arrays showed a log signal ³ log2(200). Recorded genes are required to be upregulated or downregulated with a fold change of at least log2 (2.0). 4. Use data from two groups and subject the resulting up-list or down-list to t-Test. Use data after applying filters with more

35  Laser-Capture Microdissection of Hyperlipidemic/ApoE−/− Mouse…

425

than two groups and subject the resulting up-list or down-list to one-factor variance analysis (ANOVA). Tests with p values P £ 0.01 are performed with Benjamini and Hochberg correction for multiple testing (20). 3.5. Model to Improve the Predictive Power of Microarray Expression Data from LCM-ISOLATED Aorta (see Note 6)

1. Analyze cross RNA contamination on a genome-wide scale by determining relative abundance of genes on the microarrays of adventitiae and their corresponding mediae in four individual mice (Fig. 2). 2. Use probe sets of mbp as marker for adventitia-selective genes. 3. Determine cross contamination as ratio of media expression to sum of media expression and adventitia expression (relative media expression). 4. Generate means of four pairs and determine SD (Fig. 2). 5. Devise a correction algorithm of RNA carryover between LCM-separated tissues (see Note 7).

4. Notes 1. Microarray data were deposited at http://www.ncbi.nlm.nih. gov/geo/info/MIAME.html in National Center for Biotechnology Information´s gene expression omnibus, GEO accession number GSE21419. Some of these data sets are complementary to those published in the same data bank as reported in ref. 9 (GSE10000). 2. To apply cross-contamination correction algorithms it is required to identify genes (preferably present on the array as multiple probe sets) that are exclusively or largely expressed in one of two adjacent tissues to be separated by LCM (see for adventitia versus media genes Figs.  1 and 2). For our model, we chose mbp as the model gene to correct contamination of RNA in LCM-derived material in each individual mouse as our immunohistochemical analyses revealed exclusive expression of neuronal genes in the adventitia. However, our microarray analyses reveal many more candidates. It is noteworthy that to further improve the preciseness of the correction algorithm, we applied our model on individual mice rather than on pools of RNA of different mice (see Fig. 2). Data derived from cross-contamination of the media by adventitial tissue, here mbp, have also been applied to estimate cross-contamination of media by plaque RNA because visualizing the media–plaque border are similar on phase-

426

M. Beer et al.

contrast optics and because plaques contain SMCs derived from the media refs. 1–5. 3. The genotype markedly affected the total amount of RNA obtained from each tissue section: Wild-type adventitiae show scarce cellularity necessitating considerably higher numbers of sections to obtain the amount of RNA needed for a microarray when compared to apoE−/− aortae (9). 4. This approach avoids immunohistochemical staining found to compromise RNA integrity even when several precautions were taken such as working under argon and rapid drying (12). However, our approach to improve RNA quality using phase contrast optics comes with the caveat that separation by LCM may be not as precise as with prior immunohistochemical analyses. 5. This modified protocol allows for obtaining a considerably larger amount of target amplification from small amounts of total RNA as starting material when compared with the standard Affymetrix protocol. 6. Presumptive media-specific genes acta2 (smooth muscle actin a2), smtn (smoothelin), and others believed to be largely expressed by SMCs were also found to be significantly expressed by cultured mouse aorta endothelial cells. Moreover, SMCs migrate from the media into the intima and proliferate there excluding presumptive media genes to be used for correcting media RNA contamination by LCM-derived plaque tissue. Finally, presumptive SMC-specific genes may also be expressed significantly by myofibroblasts in the adventitia and by adventitial arterioles but we do not know this for certain. Therefore, we chose neuronal genes as internal controls. 7. The crossover algorithm can be described as follows: Fig. 3 shows a schematic presentation of the real (measured) expression values in LCM-separated tissues versus those after correction in a single tissue section. EA(i) and EM(i) are the values for expression values of the adventitia for a given gene i (EA(i)) or the corresponding expression value for the media of this gene i (EM(i)) of the identical mouse as determined by the array. EA(i)* and EM(i)* are the corrected values (Fig.  3). CA(i) and CM(i) are the errors of gene expression due to cross-contamination of RNA pools for a given gene i. F is the factor that describes the degree of contamination across the paired arrays per mouse for adventitia and media. In the experiment described here, the factor F was calculated by the relative media expression of adventitia marker mbp: FMbp = EM(Mbp)/(EA(Mbp) + EM(Mbp)).

35  Laser-Capture Microdissection of Hyperlipidemic/ApoE−/− Mouse…

427

Fig.  3. Correction algorithm of microarray expression data in adjacent aorta laminae isolated by LCM. EA(i), EM(i) denote expression values for adventitia (A) or media (M) for gene i as determined in the respective microarrays before applying the correction algorithm; EA(i)* and EM(i)* are expression values for their corrected counterparts; CA(i) is the “contamination value” of adventitia expression by media for gene i; CM(i) is the value of media expression by adventitia for gene i; F is the array-wide correction factor for adventitia and media. Solid lines on top indicate the path of the laser beam; broken arrows at bottom indicate the “theoretical laser beam” after applying the correction algorithm.

References 1. Lusis, A.J. (2000) Atherosclerosis. Nature 407, 233–241. 2. Glass, C.K., and Witztum, J.L. (2001) Atherosclerosis. The road ahead. Cell 104, 503–516. 3. Libby, P. (2002) Inflammation in atherosclerosis. Nature 420, 868–874. 4. Witztum, J.L. (2002) Splenic immunity and atherosclerosis: a glimpse into a novel paradigm ? J. Clin. Invest. 109, 721–724. 5. Hansson, G.K. (2005) Inflammation, atherosclerosis, and coronary heart disease. N. Engl. J. Med. 352, 1685–1695. 6. Zhao, L., Moos, M.P., Grabner, R., Pedrono, F., Fan, J., Kaiser, B., John, N., Schmidt, S., Spanbroek, R., Lotzer, K., Huang, L., Cui, J., Rader, D.J., Evans, J.F., Habenicht. A.J., Funk, C.D. (2004) The 5-lipoxygenase pathway promotes pathogenesis of hyperlipidemiadependent aortic aneurysm. Nat. Med. 9, 966–973. 7. Moos, M.P.W., John, N., Grabner, R., Nossmann, S., Gunther, B., Vollandt, R., Funk, C.D., Kaiser, B., Habenicht, A.J.R. (2005) The lamina adventitia is the major site of immune cell accumulation in standard chow-fed

apolipoprotein E-deficient mice. Arterioscler. Thromb. Vasc. Biol. 25, 2386–2391. 8. Galkina, E., Kadl, A., Sanders, J., Varughese, D., Sarembock, I.J., and Ley, K. (2006) Lymphocyte recruitment into the aortic wall before and during development of atherosclerosis is partially L-selectin-dependent. J. Exp. Med. 203, 1273–1282. 9. Grabner, R., Lotzer, K., Dopping, S., Hildner, M., Radke, D., Beer, M., Spanbroek, R., Lippert, B., Reardon, C.A., Getz, G.S., Fu, Y.-X., Hehlgans, T., Mebius, R.E., van der Wall, M., Kruspe, D., Englert, C., Lovas, A., Hu, D., Randolph, G.J., Weih, F., Habenicht, A.J.R. (2009) Lymphotoxin b receptor signaling promotes tertiary lymphoid organogenesis in the aorta adventitia of aged apoE−/− mice. J. Exp. Med. 206, 233–248. 10. Lotzer, K., Dopping, S., Connert, S., Grabner, R., Lemser, B., Beer, M., Hildner, M., Hehlgans, T., van der Wall, M., Mebius, R.E., Lovas, A., Randolph, G.J., Weih, F., Habenicht, A.J.R. (2010) Mouse aorta smooth muscle cells differentiate into lymphoid tissue organizer-like cells upon combined TNFR1/LTbR NF-kB signaling. Arterioscler. Thromb. Vasc. Biol. 30, 395–402.

428

M. Beer et al.

11. Espina, V., Wulfkuhle, J.D., Calvert, V.S., VanMeter, A., Zhou, W., Coukos, G., Geho, D.H., Petricoin, E.F., Liotta, L.A. (2006) Laser-capture microdissection. Nature Protocols 1, 586–603. 12. Fend, F., Emmert-Buck, M.R., Chuaqui, R., Cole, K., Lee, J., Liotta, L.A., and Raffeld, M. (1999) Immuno-LCM: Laser capture microdissection of immunostained frozen sections for mRNA analysis. Am. J. Pathol. 154, 61–66. 13. Clément-Ziza, M., Munnich, A., Lyonnet, S., Jaubert, F., Besmond, C. (2008) Stabilization of RNA during laser capture microdissection by performing experiments under argon athmosphere or using ethanol as a solvent in staining solutions. RNA 14, 2698–2704. 14. Kuhn, D.E., Roy, S., Radtke, J., Gupta, S., and Sen, C.K. (2006) Laser microdissection and pressure-catapulting technique to study gene expression in the reoxygenated myocardium. Am. J. Physiol. Heart Circ. Physiol. 290:H2625–2632. 15. Uzonyi, B., Lotzer, K., Jahn, S., Kramer, C., Hildner, M., Bretschneider, E., Radke, D., Beer, M., Vollandt, R., Evans, J.F., Funk, C.D., Habenicht, A.J. (2006) Cysteinyl leukotriene 2 receptor and protease-activated receptor 1 activate strongly correlated early genes in human endothelial cells. Proc. Natl. Acad. Sci. USA 103, 6326–6331.

16. Ingersoll, M.A., Spanbroek, R., Lottaz, C., Gautier, E.L., Frankenberger, M., Hoffmann, R., Lang, R., Haniffa, M., Collin, M., Tacke, F., Habenicht, A.J., Ziegler-Heitbrock, L., Randolph, G.J. (2009) Comparison of gene expression profiles between human and mouse monocyte subsets. Blood 115, e10–9. 17. R Development Core Team. (2009) R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. ISBN 3-900051-07-0, URL http://www.R-project.org. 18. Gentleman, R.C., Carey, V.J., Bates, D.M., Bolstad, B., Dettling, M., Dudoit, S., Ellis, B., Gautier, L., Ge, Y., Gentry, J., Hornik, K., Hothorn, T., Huber, W., Iacus, S., Irizarry, R., Leisch, F., Li, C., Maechler, M., Rossini, A.J., Sawitzki, G., Smith, C., Smyth, G., Tierney, L., Yang, J.Y., and Zhang, J. (2004) Bioconductor: Open software development for computational biology and bioinformatics. Genome Biol 5:R80. URL http://genomebiology. com/2004/5/10/R80. 19. Bolstad, B.M., Irizarry R.A., Astrand M., and Speed T.P. (2003) A comparison of normalization methods for high density oligonucleotide array data based on variance and bias. Bioinformatics 19, 185–193. 20. Benjamini, Y., and Hochberg, Y. (1995) Controlling the false discovery rate: A practical and powerful approach to multiple testing. J. R. Stat. Soc. B 57, 289–300.

Chapter 36 Gene Expression Profiling in Laser-Microdissected Bone Marrow Megakaryocytes Kais Hussein Abstract Gene expression analysis of the megakaryocytic lineage requires isolation of megakaryocytes from their bone marrow microenvironment. Laser microdissection of megakaryocytes from diagnostic bone marrow biopsies allows analysis of standardised formalin-fixed samples that reflect the in situ grown status quo of a physiological or pathological condition. Taking into account that in neoplastic proliferation, e.g. myeloproliferative neoplasms, non-neoplastic haematopoietic clones proliferate in parallel, this morphologybased isolation enables selective analysis of the aberrant megakaryocytic population. Two different laser microdissection devices are presented, and the details of RNA extraction and sub-sequent real-time qPCR gene expression analysis of mRNA and microRNA are provided. Key words: Laser microdissection, Bone marrow, Megakaryocyte, Gene expression, Real-time qPCR, RNA, microRNA

1. Introduction Laser microdissection is a relatively new method and is now accepted as a helpful tool in molecular analysis (1, 2). Analyses of particular cell fractions from solid organs/ neoplasms require techniques for isolation of single cells or anatomical compartments. Macrodissection might not be sufficient for all experimental settings, and in many pathological conditions it is not useful. Particularly, the microscopic small anatomical organisation of stromal and parenchymatous cells cannot be separated. Antibody-based cell sorting (such as fluorescence or magnetic activated cell sorting (FACS or MACS)) is not applicable in all solid organ tissues because compact cell masses must be separated into a suspension of single cells. This might be

Graeme I. Murray (ed.), Laser Capture Microdissection: Methods and Protocols, Methods in Molecular Biology, vol. 755, DOI 10.1007/978-1-61779-163-5_36, © Springer Science+Business Media, LLC 2011

429

430

K. Hussein

technically complicated or insufficient; for example, suspension might be too viscous and require high dilution and, therefore, a longer time for FACS/MACS work flow. Many tissue banks comprise frozen and formalin-fixed and paraffin-embedded (FFPE) samples. Cell sorting in such tissues is the domain of laser microdissection. It is an elegant method that allows a precise microscopic evaluation of tissue sections and subsequent cutting of the tissue with a laser. Single cells or tissue areas can be isolated, collected, and analysed on the genomic and transcript level (1, 2). The advantage of frozen or otherwise fixed tissue is that the molecular analyses reflect the status quo in a given physiological or pathological condition in situ. Besides molecular research, laser microdissection is used in routine diagnostics in pathology and legal medicine for purification and/or specific isolation of particular cells. Furthermore, in human medicine, tissue banks allow the comparison of follow-up samples. Collection, processing, registration/sample-coding, and archiving of tissues/related patient data must be standardised for all samples. As a matter of course, research analysis of human samples always requires consideration of ethical and legal questions that must be addressed by the local ethics committee (3, 4) (see Note 1). 1.1. The Organisation of the Laboratory is Essential for Safe Use of Laser Microdissection in Molecular Analysis

In order to guarantee high standards, the molecular laboratory should be strictly separated into pre-amplification areas for sample preparation and post-amplification areas for analysis of PCR reaction products. Primary sample processing with registration, labelling, decalcification of bone marrow samples, and paraffin embedding is separate from the molecular laboratory. In our molecular laboratory, four different types of areas are used: (1) pre-amplification rooms for tissue cutting/laser microdissection and DNA/RNA isolation, as well as PCR preparation areas where PCR-mastermix solutions and DNA/cDNA solutions are mixed (2) a pre-amplification room for cell culture experiments, (3) a pre-amplification laboratory for preparation and storage of primers, Taq polymerase, and master mix solutions, (4) post-amplification rooms with conventional PCR/real-time qPCR cyclers and other devices for further analysis of amplified DNA/cDNA fragments. All post-amplification chemicals are kept strictly separate from the pre-amplification areas. Furthermore, DNAse-free/RNAse-free tips with aerosol protection are used. Different lab coats and new gloves are worn in each laboratory (lab coats are left in the laboratory). Laboratory benches are used, which can be decontaminated with integrated UV light and surfaces are cleaned regularly to avoid dust and aerosol deposition. These regulations are applicable for everyone, including trainees and researchers from other

36  Gene Expression Profiling in Laser-Microdissected Bone Marrow Megakaryocytes

431

institutes. This strict organisation is necessary to avoid contamination of master mix solutions and pre-amplified samples with highly enriched PCR products. 1.2. Advantage of Using Laser Microdissected Bone Marrow Megakaryocytes as a Source for Molecular Analysis

This chapter focus on gene expression profiling of laser microdissected bone marrow megakaryocytes. You will also find other chapters written by researchers from our institute who present molecular methods and specifications for analysing other tissues and pathological conditions. Peripheral blood mononuclear cells (granulocytes, monocytes, and lymphocytes but also progenitors/blasts) are easily accessible for molecular analysis. By contrast, analysis of the megakaryocytic and erythroid lineages mainly depends on bone marrow samples. Megakaryocytes are localised in the bone marrow and generate cytoplasmic pseudopodia (proplatelet formation), which are released into the bone marrow sinus as platelets (thrombocytes) (5). These platelets then circulate in the peripheral blood. Platelets contain no nuclear DNA, but some amounts of RNA. Therefore, isolation of platelets from peripheral blood samples may allow analysis of RNA/cDNA. However, purification of platelets is the critical point (6);for example, platelet-enriched plasma might not be pure enough. The RNA content of peripheral blood leukocytes is much higher than in platelets, and thus, contamination might result in co-analysis of leukocyte-derived transcripts. In addition, somatic mutations of the megakaryocytic lineage and megakaryocytic gene expression are only analysable indirectly in peripheral blood platelets. Fresh megakaryocytes from bone marrow aspirate can be isolated by FACS or MACS. However, the limitation of these methods might be the low number of megakaryocytes in an aspirate sample compared to other bone marrow haematopoietic cells. It is possible to incubate unsorted living bone marrow cells with cytokines, which generate megakaryocytic burst-forming units or colony-forming units (MK-BFU, MK-CFU). This is an artificial model that reflects considerable parts of megakaryocytic cell homeostasis, except the influence of the complex bone marrow microenvironment, e.g. endocrine and paracrine regulation, cell–cell and cell–extracellular matrix interaction (5). Analysis of laser microdissected megakaryocytes from bone marrow trephines has three major advantages: (1) in the same fashion, from each patient, cells are rapidly conserved by formalin fixation immediately after the biopsy has been taken (autolytic processes are unlikely); (2) these cells reflect the in situ growing conditions against the background of the bone marrow microenvironment; (3) megakaryocytes can be selected according to their aberrant phenotype, e.g. micromegakaryocytes, enlarged cells,

432

K. Hussein

clustered cells. Taking into account that, in neoplastic proliferation, non-neoplastic haematopoietic clones proliferate in parallel, this morphology-based isolation allows the analysis of the ill-defined megakaryocytic population. Therefore, molecular findings can be correlated to aberrant morphological features (see Note 2). In our institute, we focus on myeloproliferative neoplasms (MPN). The aberrant stem cell generates fully matured myeloid lineages with typical aberrant phenotypes of the neoplastic megakaryocytes (7). Depending on the MPN entity and the disease stage, megakaryocytes can increase the production of platelets, resulting in thrombocytosis. Despite essential thrombocytosis, marked thrombocytosis can also be present in polycythaemia vera, primary myelofibrosis (particularly in the pre-fibrotic stage), and chronic myelogenous leukaemia (7). Furthermore, MPN have a higher risk of generating a progressive myelofibrosis (7), and it is thought that megakaryocytes are central activators of bone marrow fibroblasts (8). The molecular basis is mainly unknown. Therefore, laser microdissected bone marrow megakaryocytes allow for the analysis of the haematopoietic lineage of interest. 1.3. Requirements for Analysis of DNA and RNA from Fixed Bone Marrow Samples

In principle, DNA and RNA are extractable from FFPE bone marrow and other organ specimens (2, 9–11). The critical point is not extraction, but fixation and fragmentation of DNA/ RNA. First, it is important to note that particularly rapid acid decalcification of FFPE bone marrow samples massively destroys the integrity of DNA and RNA strands. These samples are mainly lost for further diagnostic or research analysis of underlying mutations or aberrant gene expression. Some institutes have problems with extraction of sufficient amounts of DNA/RNA from FFPE samples. Sometimes, FFPE blocks from other institutes are sent to us that are not suitable for further mutation analysis due to other fixation protocols and/or rapid decalcification. Without changing the fixation and particularly the decalcification protocol, no significant progress will be made. Second, FPPE tissue-derived DNA/RNA is always fragmented, but however, as long as careful processing has been performed, fragments of a few hundred base pairs can be amplified by PCR. This must be taken into account when selecting primers for mutation analysis (100–500 nucleotide-long DNA-derived PCR-fragments) or evaluation of gene expression (60–100 nucleotide-long RNA/cDNA-derived fragments for real-time qPCR analysis; see Notes 3 and 4). Owing to the low number of laser microdissected FFPE cells and the known fragmentation of RNA strands, RNA quality control (e.g. absorbance ratio), as recommended for samples from cell lines or fresh/frozen tissue, is not applicable (11).

36  Gene Expression Profiling in Laser-Microdissected Bone Marrow Megakaryocytes

433

2. Materials Several companies have invented and developed different laser microdissection systems. In our laboratory, we use two systems (P.A.L.M. Microlaser Technologies, Bernried, Germany, and SmartCut Plus System, Olympus, Hamburg, Germany) but, in principle, any laser microdissection device is suitable. The main relevant difference between all systems is the mode of cell collection after the cells have been isolated with the laser beam (Fig. 1). The cells can be catapulted into the lid of a PCR tube (P.A.L.M. System) or stuck directly on the lid (SmartCut Plus System). In our laboratory, the former system allows a faster collection of megakaryocytes (approximately 40–60 min per case), while the latter device provides excellent histomorphology (12, 13) and also allows the use of fluorescence microscopy. Owing to the increased number of megakaryocytes in MPN, one slide with 1–3 bone marrow sections is sufficient per case. In order to collect the same number of cells in non-neoplastic control cases (see Note 5), two or more slides should be used. 1. Bone marrow fixative: 64% methanol (v/v), 4.48 M formaldehyde, 1.6 mM sodium hydrogen phosphate buffer pH 7.4, 7.4 mM glucose. 2. Decalcification solution: 270 mM Tris–HCL, 270 mM EDTA pH 7.4, Adjustment of pH with 60  mM sodium hydrogen phosphate pH 7.0–7.2. 3. Histological staining, xylene substitute, less toxic than xylene (Tissue-Tek® Tissue-Clear®, Sakura, Zoeterwoude, The Netherlands). 4. Ethanol (100, 96, 70%). 5. Sterile water. 6. Methylene blue (Loeffler’s methylene blue, MERCK, Darmstadt, Germany). 7. RNA digestion solution, 50 ml stock solution: 100 g guanidinium isothiocyanate, 6  ml Tris–HCl pH 7.6, 4.2  ml 97% sarcosyl, fill up with up to 200 ml 0.2% DEPC-H2O (Merck, Darmstadt, Germany; AppliChem, Darmstadt, Germany), 50  ml Proteinase K stock solution: 20  mg/ml in water, aliquots stored at −20°C (Merck, Darmstadt, Germany), 0.1 ml beta-mercaptoethanol (Carl Roth, Karlsruhe, Germany). 8. RNA precipitation, 3  M sodium acetate pH 7.0 100 mg/ml dextran T500 (SIGMA, Taufkirchen, pH 5.2, Roti-Aqua-Phenol (Roth, Karlsruhe, Chloroform (Merck, Darmstadt, Germany), (Roche, Basel, Switzerland).

containing Germany), Germany), Glycogen

Fig.  1. Laser microdissection of bone marrow megakaryocytes. In these two laser microdissection systems (P.A.L.M. Microlaser Technologies, Bernried, Germany, and SmartCut Plus System, Olympus, Hamburg, Germany), the laser and the microscope are fixed, while the tissue section is moved on a motor table. The position of the table is controlled by a computer and via movement of the cursor on the video camera image of the tissue. Morphological identification of megakaryocytes (a and d) is followed by contact-free laser microdissection. In the P.A.L.M. System, the lid of the reaction tube is above the tissue and has no contact with it (b). Note that the tissue is on top of the laser microdissection membrane. After the cell is completely cut from the surrounding tissue, it is subsequently catapulted into the adhesive lid (c). In the SmartCut Plus System, the lid of the tube is pressed on the membrane and has no contact with the tissue, because the tissue is on the opposite surface (e). The cut cell sticks to the adhesive inlay of the lid and is removed by elevation of the tube via a robot arm (f). The advantage of this system is the relatively good histological quality. In both systems, the specific cell isolation is demonstrated in the images from the adhesive surface of the lids.

36  Gene Expression Profiling in Laser-Microdissected Bone Marrow Megakaryocytes

435

3. Methods 3.1. Fixation and Decalcification of Bone Marrow Samples

1. The bone marrow biopsy is incubated in buffered formalin (volume of fixation solution to specimen >10:1) for 24 h at 4°C in the dark. Variations in duration or room temperature are tolerable. 2. The decalcification in pH 7.4 EDTA buffer can take several days and can be shortened considerably by using an ultrasonic bath. EDTA is stirred constantly and changed every day.

3.2. Preparation of Sections for Laser Microdissection

1. FFPE bone marrow biopsies are cut with a microtome (3–5 mm). Sections are mounted in a water bath (37°C) on foil-covered slides (provided by the microdissection device company) and are dried for several hours or overnight at 45°C. 2. Deparaffinisation, rehydration, and staining are performed as follows: 2 × 10 min xylene substitute, 2 × 5 min 100% ethanol, 2 × 5 min 96% ethanol, 1 × 5 min 70% ethanol, 1 × 5 min sterile water, 1 × 10s methylene blue, rinsed with distilled water. 3. Sections can now be used for laser microdissection or can be dried overnight. For all steps, shorter durations are tolerable.

3.3. Isolation and Collection of Megakaryocytes

1. Use a ×400 magnification for morphological identification of characteristically large megakaryocytic cells (Fig. 1). 2. Immunohistochemical labelling can be applied for identification and isolation of non-megakaryocytic cells (e.g. haemoglobin, CD15, mast cell tryptase, CD34, CD20, CD3) but is not necessary for highlighting megakaryocytes. The laser energy, which is capable of cutting the tissue, varies according to the thickness of the tissue section, the type of tissue (haematopoietic areas are easily dissectible) and the laser microdissection device. 3. When using the P.A.L.M. System, use high laser energy (>70%) because after cutting around the cells, they are subsequently catapulted into the lid. Special 500-ml tubes with adhesive lid inlays are provided by the manufactures. Cut accurately between the megakaryocytic cell membrane and the surrounding cells (Fig.  1) and collect 200–300 megakaryocytes.

3.4. Extraction of Total RNA for Analysis of Megakaryocytic Gene Expression

1. Bone marrow cells are digested overnight at 55°C in 50 ml 20 mg/ml proteinase K solution plus 50 ml RNA digestion solution. Digestion can be performed in the laser microdissection tube or after transfer of cells to another PCR tube (see Note 6).

436

K. Hussein

2. All subsequent steps are performed on ice-cooled racks. Between the RNA precipitation steps, we recommend vigorous shaking of the chemicals by hand and not with a vortex device to prevent mechanical fragmentation of RNA molecules. 3. For RNA precipitation, 10 ml 3 M sodium acetate pH 5.2 is added (mix and short centrifugation) and then 63  ml RotiAqua-Phenol (mix and short centrifugation) and finally 27 ml chloroform (mix and short centrifugation) are added. Following phase separation through incubation on ice (30  min) and centrifugation (30  min at 4°C), the aqueous phase (95  ml) was carefully removed and added to 100  ml 2-propanol and 1 ml 20 mg/ml glycogen as precipitation carrier. RNA is stored at −20°C. 4. After centrifugation, the RNA pellet is washed in 70% ethanol and dissolved in 10  ml DEPC-treated water (see Note 7). Subsequently, 2 ml (1 ng) of the RNA solution are each reverse transcribed as described in the relevant kit manual. 3.5. Gene Expression Profile Analysis

1. Analyses of protein-coding mRNA (14–16) and non-proteincoding regulatory microRNA (17, 18) can be performed from laser microdissected bone marrow megakaryocytes. 2. We recommend quantitative real-time reverse transcription PCR-based gene expression analysis (RT-qPCR) because it is a highly specific, sensitive, and reproducible method (suggested reference genes: glucuronidase, beta/GUSB, GeneID 2990; glyceraldehyde-3-phosphate dehydrogenase/GABDH, GeneID 2597; polymerase (RNA) II (DNA directed) polypeptide A, 220kDa/POLR2, GeneID 5430; actin, beta/ACTB, GeneID 60). In principle, any commercially available kit and RT-qPCR cycler system should be suitable. We do not recommend the use of fluorescence-labeled cDNA glass slide microarrays. We tried this method for microRNA but found that samples from lasermicrodissected megakaryocytes do not generate reliable results. With the appropriate array, the RT-qPCR method for the analysis of single genes (14, 16) can be extended to the analysis of up to several hundred genes (16, 18). Per case, 1 ng of FFPEderived megakaryocytic RNA should be sufficient. It is not always necessary, but the megakaryocytic cDNA may be pre-amplified before further analysis (15, 16).

4. Notes 1. Guideline for the use of archived samples for research analysis: the diagnosis must have been established before research use, the remaining tissue should not be required for further

36  Gene Expression Profiling in Laser-Microdissected Bone Marrow Megakaryocytes

437

therapy and the samples must be anonymised. Non-diagnostic tissue samples that are donated for research use need written approval by the donor. 2. Other methods, such as FACS/MACS, can isolate the neoplastic cell population only if the immunophenotype is different from non-neoplastic cells. However, mature neoplastic megakaryocytes, granulocytes, monocytes, and erythroid cells exhibit the same marker profile as their non-neoplastic counterparts. Mutation analysis, after cell sorting, does not exclude contamination by non-neoplastic cells. 3. For mutation analysis with restriction enzymes, multiplex PCR, melting curve analysis, Pyrosequencer assay, or direct sequencing, 100–200 nucleotide-long DNA/cDNA fragments should be sufficient. Screening for unknown mutations is difficult in these small fragments. Real-time qPCR expression analysis should be performed in 60–100 nucleotide-long cDNA-derived fragments. 4. It is not advisable to test new primers only in DNA or cDNA from leukaemic cell lines or design primers that generate >500 nucleotide-long fragments. Use FFPE-derived DNA/ cDNA from complied tissue sections, rather than laser microdissected cells for testing. 5. Non-neoplastic control cases: a bone marrow biopsy had been taken in a diagnostic setting for evaluation of bone marrow status, e.g. clinically suspected haematological disorder. Aberrant peripheral blood parameters, e.g. thrombocytosis/ thrombocytopenia, are tolerable as long as bone marrow histomorphology shows normal maturation of haematopoietic cells. Post-radio/chemotherapy and post-bone marrow transplantation samples should not be included. 6. When using tubes that are provided by the laser microdissection company, make sure that the tubes do not leak, particularly when the tubes are incubated at 55°C on top of the lid. The proteinase K solution might evaporate and the probe will be lost. In this case, transfer the cells from the laser microdissection tube to a conventional PCR tube by washing the cells with digestion solution and a pipette from the lid. Never use the same laser microdissection tube for different probes, even after cleaning. 7. Commercially available membrane filter-based kits are suitable for DNA as well as RNA extraction from laser microdissected cells. The critical point is the number of cells collected and the relatively small RNA fragments. Owing to their short length, microRNA might be too small for the filter and could be lost during the washing processes. Furthermore, the amount of RNA from a few hundred cells might be diluted during the

438

K. Hussein

last washing-out step from the membrane filter. This might result in low RNA concentration. We, therefore, suggest conventional phenol–chloroform extraction of RNA from laser microdissected cells, such as megakaryocytes. Accordingly, DNA analysis can be performed after proteinase K digestion from crude lysate (after denaturation of proteinase K, 95°C for 10 min) or after conventional DNA precipitation.

Acknowledgements The author would like to thank his mentors Professor Dr. med. Hans H. Kreipe and Professor Dr. med. Oliver Bock (Institute of Pathology, Hannover Medical School) as well as Dr. phil. Khadra Hussein, Dr. med. Ulrich Thorns and Sir Jan Off. References 1. Emmert-Buck MR, Bonner RF, Smith PD et al (1996) Laser capture microdissection. Science 274:998–1001 2. Lehmann U, Glockner S, Kleeberger W et  al (2000) Detection of gene amplification in archival breast cancer specimens by laser-assisted microdissection and quantitative realtime polymerase chain reaction. Am J Pathol 156:1855–1864 3. Bauer K, Taub S, Parsi K (2004) Ethical issues in tissue banking for research: a brief review of existing organizational policies. Theor Med Bioeth 25:113–142 4. Caulfield T (2004) Tissue banking, patient rights, and confidentiality: tensions in law and policy. Med Law 23:39–49 5. Bluteau D, Lordier L, Di Stefano et al (2009) Regulation of megakaryocyte maturation and platelet formation. J Thromb Haemost 7:227–234 6. Birschmann I, Mietner S, Dittrich M et  al (2008) Use of functional highly purified human platelets for the identification of new proteins of the IPP signaling pathway. Thromb Res 122:59–68 7. Swerdlow SH, Campo C, Harris NL, Jaffe ES, Pileri SA, Stein H, Thiele J, Vardiman, JW (2008) WHO Classification of Tumours of Haematopoietic and Lymphoid Tissues. IARC, Lyon 8. Bock O, Hussein K, Kreipe H (2007) Stem cell defects in Philadelphia chromosome negative chronic myeloproliferative disorders: a phenotypic and molecular puzzle? Curr Stem Cell Res Ther 2:253–263

9. Bock O, Kreipe H, Lehmann U (2001) One-step extraction of RNA from archival biopsies. Anal Biochem 295:116–117 10. Li J, Smyth P, Flavin R et al (2007) Comparison of miRNA expression patterns using total RNA extracted from matched samples of formalinfixed paraffin-embedded (FFPE) cells and snap frozen cells. BMC Biotechnol 7:36 11. Bustin SA, Benes V, Garson JA et  al (2009) The MIQE guidelines: minimum information for publication of quantitative real-time PCR experiments. Clin Chem 55:611–622 12. Hussein K, Bock O, Theophile K et  al (2009) Biclonal expansion and heterogeneous lineage involvement in a case of chronic myeloproliferative disease with concurrent MPLW515L/-JAK2V617F mutation. Blood 113:1391–1392 13. Hussein K, Theophile K, Buhr T et al (2009) Different lineage involvement in myelodysplastic/myeloproliferative disease with combined MPL and JAK2 mutation. Br J Haematol 146:510–520 14. Bock O, Schlué J, Lehmann U et  al (2002) Megakaryocytes from myeloproliferative disorders show enhanced nuclear bFGF expression. Blood 100:2274–2275 15. Theophile K, Jonigk D, Kreipe H et al (2008) Amplification of mRNA from laser-microdissected single or clustered cells in formalin-fixed and paraffin-embedded tissues for application in quantitative real-time PCR. Diagn Mol Pathol 17:101–106 16. Theophile K, Hussein K, Kreipe H et al (2008) Expression profiling of apoptosis-related genes

36  Gene Expression Profiling in Laser-Microdissected Bone Marrow Megakaryocytes in megakaryocytes: BNIP3 is downregulated in primary myelofibrosis. Exp Hematol 36:1728–1738 17. Hussein K, Dralle W, Theophile K et al (2009) Megakaryocytic expression of miRNA 10a, 17–5p, 20a and 126 in Philadelphia chromo-

439

some-negative myeloproliferative neoplasm. Ann Hematol 88:325–332 18. Hussein K, Theophile K, Dralle W et al (2009) MicroRNA expression profiling of megakaryocytes in primary myelofibrosis and essential thrombocythemia. Platelets 20:391–400

wwwwwww

Chapter 37 Specific RNA Collection from the Rat Endolymphatic Sac by Laser-Capture Microdissection (LCM): LCM of a Very Small Organ Surrounded by Bony Tissues Kosuke Akiyama, Takenori Miyashita, Ai Matsubara, and Nozomu Mori Abstract Laser-capture microdissection (LCM) is an excellent tool to selectively obtain target tissue or cells. The endolymphatic sac (ES) is part of the inner ear, and a large part of the ES is located in the temporal bone. The rat ES is conventionally harvested using stereomicroscopy. In this method, contamination is unavoidable because of its size and location; therefore, additional checks, such as in situ hybridization, are necessary to confirm the cellular localization, and quantitative analysis is difficult in the ES. We have shown a selective epithelial tissue method using LCM to obtain RNA without contamination from ES epithelial tissue. Key words: Endolymphatic sac, Laser-capture microdissection, Temporal bone, Inner ear, Tissue contamination, RT-PCR

1. Introduction The endolymphatic sac (ES) is part of the inner ear, and a large part of the ES is located inside the temporal bone (1, 2). It is generally accepted that ES plays important roles in the homeostasis of the inner ear and endolymph (3–5). Rats are most commonly used as an experimental animal for ES molecular analyses, and ES isolation is conventionally performed under stereomicroscopy (6, 7). Using this conventional method, contamination is unavoidable because ES is quite a small organ surrounded by bony tissues (1). Thus, other methods that can selectively isolate ES epithelial cells are required.

Graeme I. Murray (ed.), Laser Capture Microdissection: Methods and Protocols, Methods in Molecular Biology, vol. 755, DOI 10.1007/978-1-61779-163-5_37, © Springer Science+Business Media, LLC 2011

441

442

K. Akiyama et al.

Laser capture microdissection (LCM) is superior to selectively obtain target tissue or cells (8). We have demonstrated a more selective collection of rat ES epithelial tissue using LCM to obtain pure RNA of ES epithelia with no contamination, compared with that using the conventional method (9). Our recent study has shown that LCM is useful to isolate a very small organ surrounded by bony tissues.

2. Materials 2.1. Animals

1. Four-week-old Sprague-Dawley rats (Charles River Japan, Inc. Yokohama, Japan) (see Note 1).

2.2. Tissue Preparation

1. Ketamine hydrochloride (50  mg/kg i.m. Daiichi-Sankyo, Tokyo, Japan) and diethyl ether. 2. Ethanol is dissolved in diethyl pyrocarbonate (DEPC)-treated water at 70% before use (see Note 2). 3. 0.12 M Ethylenediamine tetraacetic acid (EDTA) (0pH 6.5) including RNAlater (1:50) (Invitrogen, CA, USA) is stored at 4°C. 4. OCT tissue compound and noncoated glass slides.

2.3. LCM

1. LCM is performed using Arcturus Pixcell Il (MDS Analytical Technologies, CA, USA). 2. 70, 90, and 100% ethanol/DEPC-treated water and xylene are adjusted shortly before use. 3. CapSure HS LCM Caps (Arcturus).

2.4. RNA Extraction and Reverse TranscriptionPolymerase Chain Reaction

1. PicoPure RNA isolation Kit (Arcturus). 2. 200  U/ml; Molony murine leukemia virus reverse transcriptase (Invitrogen), 100 pM/ml; Random primers, 5× RT buffer, 10  mM dNTP3 (Takara Bio, Otsu, Japan), 0.1  M DTT, and 10 U/ml RNase inhibitor (Invitrogen). 3. TaKaRa PCR thermal cycler MP (Takara Bio). 4. 5 unit/ml; TaKaRa Ex Taq, 10× Ex Taq buffer, each 2.5 mM; dNTP mixture (Takara Bio). 5. Specific primer pairs, osteocalcin (GenBank M23637) (10), calponin H1 (GenBank NM031747) (11), NKCC-2 (GenBank U10096) (12), and G3PDH (Toyobo Co. Ltd, Osaka, Japan) are designed from the GenBank database. Table 1 shows each primer sequence (see Note 3). 6. 1.0% agarose gels and ethidium bromide.

37  Specific RNA Collection from the Rat Endolymphatic Sac…

443

Table 1 Primer sequences used to identify the expression of osteocalcin, calponin H1, NKCC2, and G3PDH

Osteocalcin

Primer sequence

Length of PCR product (bp)

S 5¢-GGTGCAAAGCCCAGCGACTCT-3¢

199

AS 5¢-GGAAGCCAATGTGGTCCGCTA-3¢ Calponin H1

S 5¢-GGGGGTCAAATATGCAGAGA-3¢

310

AS 5¢-GACGTTGAGCGTGTCACAGT-3¢ NKCC2

S 5¢-CCGTCGCCTACATAGGTGTT-3¢

343

AS 5¢-CCCTTTGCGAAGAACTGAAG-3¢ G3PDH

S 5¢- TGAAGGTCGGTGTCAACGGATTTGGC -3¢

983

AS 5¢- CATGTAGGCCATGAGGTCCACCAC -3¢ The expected lengths of PCR products are indicated (reproduced from ref. 9 with permission from Elsevier Science)

3. Methods 3.1. Tissue Preparation

1. Rats are deeply anesthetized using ketamine hydrochloride and diethyl ether. 2. The rats are decapitated after exsanguination via the left ventricle with 70% ethanol/DEPC-treated water. Exsan­ guination is continued for about 5 min (see Note 4). 3. Temporal bones on bilateral sides are removed and ES is separated carefully using a stereomicroscope (see Note 5). 4. ES samples are fixed in 70% ethanol/DEPC-treated water for 4 h at 4°C. 5. ES are decalcified in 0.12 M EDTA, including RNAlater, for about 5–7 days at 4°C (see Note 6). 6. After decalcification, ES is embedded in OCT tissue compound, frozen in liquid nitrogen immediately, and stored at −80°C until use. 7. The entire ES is cut into slices using a cryostat and placed on slides. The thickness of cryostat sections is set up at 10–20 mm. These procedures are performed at −20°C, and the slides are stored at −80°C until use. An example tissue sample is shown in Fig. 1.

3.2. LCM

1. Prepare the slides to perform LCM. The slides are refixed in 70% ethanol/DEPC-treated water for a short time, washed in DEPC-treated water in a short time, dehydrated in steps of 70, 90, and 100% ethanol/DEPC-treated water each for

444

K. Akiyama et al.

Fig. 1. Temporal bone tissues around the endolymphatic sac (ES) stained with hematoxylin. (a) Lower power field. (b) Higher power field, and the lower left area shows ES epithelium. The lumen of the ES is covered with monolayer epithelial cells. SC semicircular canal, O operculum, V vein of vestibular aqueduct, BM bone marrow, B brain. Scale bar in A = 2 mm and in B = 200 mm (reproduced from ref. 9 with permission from Elsevier Science).

1 min and twice in xylene for a few minutes. They are well air-dried in an air chamber at room temperature. 2. LCM is performed using Arcturus Pixcell Il with CapSure HS LCM Caps. Spot size is adjusted to 15  mm, laser power is adjusted to 40 mW, and duration is adjusted to 2.0 ms. An example tissue sample before and after LCM is shown in Fig. 2. 3. After LCM, LCM cap is stored at −80°C immediately. 3.3. RNA Extraction

RNA extraction from LCM samples is performed using a Micro scale RNA isolation kit in accordance with the manufacturer’s protocol. 1. Cover the captured cells with 10  ml extraction buffer onto the LCM cap and incubate at 42°C for 30 min. 2. Centrifuge at 800 × g for 2  min to collect cell extract into another microcentrifuge tube. 3. Add 10 ml of 70% ethanol into the cell extract and mix with preconditioning purification column (purification column is

37  Specific RNA Collection from the Rat Endolymphatic Sac…

445

Fig. 2. Collection of endolymphatic sac (ES) epithelium by laser-capture microdissection. (a and c) show dehydrated ES before dissection. (b and d) show dehydrated ES after dissection. (a and b) are shown in lower power field and the scale bar indicates 500 mm. (c and d) are in higher power field and the scale bar indicates 100 mm. Arrows and arrowheads indicate ES epithelium and the vein of the vestibular aqueduct, respectively. Only the ES epithelium was dissected selectively. Vein or other tissues around the ES remained. BM bone marrow (reproduced from ref. 9 with permission from Elsevier Science).

incubated with 250  ml conditioning buffer for 5  min and centrifuged at 16,000 × g for 1 min beforehand). 4. Centrifuge for 2 min at 100 × g immediately followed by centrifugation at 16,000 × g for 30 s. 5. Wash three times with washing buffer and centrifuge. 6. The purification column is transferred to a new microcentrifuge tube and 15 ml elution buffer is added to the column. 7. Centrifuge the column for 1 min at 1,000 × g and then spin at 16,000 × g for 1 min to elute RNA. 3.4. RT-PCR

1. Extracted RNA is reverse-transcribed in cDNA. Incubation is performed using a TaKaRa PCR thermal cycler MP. 15  ml extracted RNA is incubated with 7  ml random primers and 13.5  ml DEPC-treated water at 65°C for 5  min. Reaction buffer is added and then incubated: 25°C for 10 min, 37°C for 60  min, and 95°C for 5  min. The reaction buffer is composed of 5× RT buffer 10 ml, DTT 5 ml, dNTP3 2.5 ml, RNase inhibitor 1  ml, and Molony murine leukemia virus reverse transcriptase 1 ml. 2. PCR mixture: TaKaRa Ex Taq 0.1 ml, 10× Ex Taq buffer 2 ml, dNTP mixture 1.6  ml, reverse and forward specific primer each 1 ml, synthesized cDNA 3 ml, and made up to 20 ml with sterilized distilled water. 3. PCR is performed using a TaKaRa PCR thermal cycler MP. PCR cycle is as follows: 35 cycles, denature at 94°C for 30 s, extension at 60°C for 30 s, and annealing at 72°C for 1 min.

446

K. Akiyama et al.

4. 15 ml of each PCR product is mixed with 3 ml of 6× loading buffer, separated by electrophoresis on 1.0% agarose gels, and visualized by ethidium bromide staining. An example result is shown in Fig. 3.

Fig. 3. PCR amplification with osteocalcin, calponin H1, and Na-K-2Cl cotransporter-2 (NKCC2) specific primer pairs. The method using LCM was compared with the conventional method. In the conventional method, ES was removed under stereomicroscopy and RNA was extracted by Trizol. One microgram of total RNA was reverse-transcribed into cDNA, as shown in Methods. (a) Osteocalcin expression is observed in temporal bone and conventional ES, whereas it is not observed in LCM ES. (b) Calponin H1 expression is observed in the thoracic aorta and conventional ES, whereas no band is seen in LCM ES. (c) NKCC-2 expression is observed in the kidney and in both conventional ES and LCM ES. (d) G3PDH expressions in each tissue. cDNAs of conventional ES and LCM ES were derived from three individual rats, respectively. These results indicate that the LCM method has higher specificity than the conventional method. LCM laser-capture microdissection, ES endolymphatic sac, B temporal bone, A thoracic aorta, K kidney (reproduced from ref. 9 with permission from Elsevier Science).

37  Specific RNA Collection from the Rat Endolymphatic Sac…

447

4. Notes 1. Young adult rats are the most suitable because their temporal bone is not so osteoid and soft. Therefore, 5–7 days decalcification is enough for the following treatment. When using older rats, a longer decalcification period is needed. 2. DEPC-treated water is prepared the day before use. DEPC is added to distilled water at 1:1,000 and allowed to stand at room temperature overnight. It is then autoclaved and stored at 4°C. 3. Osteocalcin is used as a bone-specific marker, and calponin H1 is used as an arteriole-specific marker, which are indications of ES surrounding tissue contamination. NKCC-2, which is expressed only in ES, except in kidneys, is used as an ES epithelium marker. Namely, NKCC2 can be mirrored by ES epithelium. 4. The thoracic cavity is opened and the heart is visualized. A 21G needle connected to a roller pump pierces the left ventricle cavity, and then the right auricle is cut to eliminate blood. Fixative solution is poured and the stiffened whole body is an indication that fixation has been achieved. 5. ES is located under the operculum. Temporal bone tissue, including the operculum, is cut by a surgical knife to take the entire ES with some margin. 6. Decalcification of approximately 7 days is sufficient to use the cryostat. Shorter decalcification makes it a little difficult to prepare cryostat sections because the specimens are solid.

Acknowledgments This work was supported in part by Grants-in-Aid for Scientific Research (#13671783 and #16591711 to N.M., #16791003 and #791208 to T.M.). An outline of this work was published in ref. 9 and is reproduced with permission from Elsevier Science.

448

K. Akiyama et al.

References 1. Dahlmann A., von During M. (1995) The endolymphatic duct and sac of the rat: A histological, ultrastructural, and immunocytochemical investigation, Cell Tissue Res. 282: 277–289. 2. Lo W. W., Daniels D. L., Chakeres D. W et al (1997) The endolymphatic duct and sac, AJNR Am. J. Neuroradiol. 18: 881–887. 3. R.S. Kimura. (1967) Experimental blockage of the endolymphatic duct and sac and its effect on the inner ear of the guinea pig. A study on endolymphatic hydrops, Ann. Otol. Rhinol. Laryngol. 76: 664–687. 4. R.S. Kimura. (1982) Animal models of endolymphatic hydrops, Am. J. Otolaryngol. 3: 447–451. 5. Miyashita T., Tatsumi H., Hayakawa K. et  al (2007) Large Na(+) influx and high Na(+), K (+)-ATPase activity in mitochondria-rich epithelial cells of the inner ear endolymphatic sac. Pflugers Arch. 453: 905–13. 6. Furuta H., Sato C., Kawaguchi Y et al (1999) Expression of mRNAs encoding hormone receptors in the endolymphatic sac of the rat, Acta Otolaryngol. 119: 53–57.

7. Sawada S., Takeda T., Kitano H et  al (2002) Aquaporin-2 regulation by vasopressin in the rat inner ear, Neuroreport 13: 1127–1129. 8. Murray G. I. (2007) An overview of laser microdissection technologies, acta histochmica. 109: 171–176. 9. Akiyama K., Miyashita T., Mori T et al (2008) A new approach for selective rat endolymphatic sac epithelium collection to obtain pure specific RNA, Biochem. Biophys. Res. Commun. 376: 611–614. 10. Yoon K. G., Rutledge S. J., Buenaga R. F et al (1988) Characterization of the rat osteocalcin gene: Stimulation of promoter activity by 1,25-dihydroxyvitamin D3, Biochemistry 27: 8521–8526. 11. Nishida W., Kitami Y., Hiwada H (1993) cDNA cloning and mRNA expression of calponin and SM22 in rat aorta smooth muscle cells, Gene 130: 297–302. 12. Gamba G., Miyanoshita A., Lombardi M et al (1994) Molecular cloning, primary structure, and characterization of two members of the mammalian electroneutral sodium-(potassium)chloride cotransporter family expressed in kidney, J. Biol. Chem. 269: 17713–17722.

Chapter 38 The Use of Laser Capture Microdissection on Adult Human Articular Cartilage for Gene Expression Analysis Naoshi Fukui, Yasuko Ikeda, and Nobuho Tanaka Abstract The integrity of articular cartilage is maintained by chondrocytes, the sole type of cell that resides within the tissue. The noncalcified region of articular cartilage can be divided into three zones based on histological features, in which the chondrocyte metabolism is known to differ obviously among the zones. In pathological cartilage, the chondrocyte metabolism may change dramatically, which could play a pivotal role in the progression of the disease. Since such change in metabolism differs obviously from site to site within cartilage, it is crucial to determine the chondrocyte metabolism in respective regions. To this end, we have employed laser-capture microdissection (LCM) to analyze chondrocyte metabolism in various regions of pathological and control cartilage. In this report, we describe our protocol for LCM on adult human cartilage tissue. With this protocol, a specific site of cartilage tissue was successfully obtained by LCM for gene expression analysis. Key words: Laser-capture microdissection, Articular cartilage, mRNA, Osteoarthritis, Chondrocyte, Metabolic change, Regional difference

1. Introduction Articular cartilage is a thin layer of connective tissue that covers bone surfaces within synovial joints. It is a unique tissue with the capacity to withstand repetitive loads, allowing frictionless motion of the joint. Articular cartilage is primarily composed of water and extracellular matrix, with a small volume of cells. The matrix of articular cartilage is composed of several types of collagen and proteoglycans (1). Chondrocytes are the only kind of cell that resides within cartilage, and since no other types of cells are seen in cartilage, chondrocytes are considered to be entirely responsible for the integrity and turnover of cartilage matrix.

Graeme I. Murray (ed.), Laser Capture Microdissection: Methods and Protocols, Methods in Molecular Biology, vol. 755, DOI 10.1007/978-1-61779-163-5_38, © Springer Science+Business Media, LLC 2011

449

450

N. Fukui et al.

Based on histological features, articular cartilage can be divided into four layers of superficial (tangential) zone, middle (transitional) zone, deep (radial) zone, and calcified cartilage underneath (1–3) (Fig.  1). Among them, calcified cartilage is unique in that the matrix contains an abundance of calcium apatite crystals and is clearly distinguished from the above three zones by a tide mark. In the three noncalcified zones, the metabolic activity of chondrocytes is known to differ considerably among the zones (1, 2, 4–6). Osteoarthritis (OA) is a disease that primarily affects articular cartilage. In OA, the cartilage matrix is lost gradually over years, eventually devastating functional joints. OA is an age-related disease. With the aging of society, OA has become the most prevalent joint disease in developed countries (7). OA of a knee joint is particularly an issue, in view of its prevalence, severity of symptoms, and association with disability (8, 9). Focal loss of articular cartilage is a unique feature in the pathology of OA. In OA joints, cartilage is most severely degenerated in weight-bearing areas, but may remain almost intact in nonweight-bearing areas. Reflecting such regional differences, the chondrocyte metabolism differs considerably from site to site within cartilage. Since chondrocytes become catabolic and promote cartilage degeneration in OA, it is critically important to determine chondrocyte metabolism at respective sites within OA cartilage. Conventionally, such regional differences can be known only by histological methods, such as immunohistochemistry or in situ hybridization.

Fig. 1. Histology of normal adult human articular cartilage. The section is prepared in a plane vertical to the surface of cartilage. Superficial (S), middle (M), and deep (D) zones are shown together with the tide mark (TM), calcified cartilage (CC), and subchondral bone (SCB). Safranin O/fast green staining. Scale bar, 0.5 mm.

38  The Use of Laser Capture Microdissection on Adult Human Articular Cartilage…

451

Laser capture microdissection (LCM) is an innovative technology that enables acquisition of specific sites of the tissues based on histological features. We have employed this method and successfully analyzed gene expression levels of chondrocytes in respective regions of OA and control cartilage. In this section, we describe in detail our LCM protocol for adult human articular cartilage tissue. The use of LCM on cartilage tissue could clarify many novel aspects of cartilage pathology, which cannot be known by any other methods.

2. Materials 2.1. Preparation of Cartilage Tissue Blocks

1. DMEM/F-12 (Life Technologies, Carlsbad, CA). 2. Scalpel blade (No. 15; Feather, Osaka, Japan). 3. Scalpel holder (No. 3; Feather). 4. Optimum Cutting Temperature (OCT) compound (Sakura Finetek Japan, Tokyo, Japan). 5. Metal mold for tissue blocks. 6. Liquid nitrogen. 7. Deep freezer (−80°C).

2.2. Sectioning

1. Cryostat (CM1900; Leica Microsystems, Houston, TX). 2. Disposable microtome blades (818; Leica Microsystems). 3. Precleaned glass slide (S2444; Matsunami, Osaka, Japan). 4. RNAse-free water (Nippon Gene, Tokyo, Japan). 5. 0.5 M Ethylenediaminetetraacetic acid disodium salt (EDTA) solution, (Sigma, St. Louis, MO). 6. Ethanol (Wako Pure Chemical, Osaka, Japan). 7. Xylene (Sigma). 8. RNase-free pipette tips and pipettes (1,000, 200, 20 ml). 9. Lint-free paper towels. 10. Fume hood.

2.3. LCM

1. Laser-Capture Microdissection System (Arcturus PixCell IIe; Molecular Devices, Sunnyvale, CA). 2. LCM caps (CapSure LCM caps; Molecular Devices). 3. Fine-pointed tweezers. 4. (Optional) Dissection microscope.

2.4. RNA Extraction

1. RNeasy Micro kit (Qiagen, Hilden, Germany). 2. 2-Mercaptoethanol (Sigma).

452

N. Fukui et al.

3. 10  mM ribosomal RNA from Escherichia coli (Roche Diagnostics, Bazel, Switzerland). Dissolve in RNase-free water. 4. Tabletop centrifuge. 5. Clean, autoclaved 1.5-ml centrifuge tubes. 2.5. cDNA Synthesis

1. Sensiscript RT kit (Qiagen). 2. Oligo dT primer (Life Technologies). 3. RNAse inhibitor (Roche Diagnostics). 4. Heat block for 1.5-ml centrifuge tubes (37°C).

3. Methods Due to the ubiquitous presence of RNases, RNA is extremely unstable. To minimize RNA degradation, all reagents and instruments used for the experiment should be RNase-free. Disposable experimental gloves should be worn throughout the procedure, and the experiment should be performed as quickly as possible (see Note 1). 3.1. Preparation of Cartilage Tissues

1. Cartilage tissue obtained at surgery or dissection is transferred to the laboratory without delay in chilled DMEM/F-12 medium (see Note 2). 2. Cartilage tissue should be processed immediately after arrival to minimize RNA degradation. 3. Cartilage tissues are separated from the subchondral bone in full thickness with a sharp scalpel (Fig. 2 a, b). We prefer to use a small crescent-shaped blade to do this. 4. The cartilage tissue is cut into an appropriate size (Fig. 2 c, d). 5. OCT compound is poured in a mold, and the tissue is placed within it with the cross section of the tissue parallel to the surface (see Notes 3 and 4). The mold is immediately dipped in liquid nitrogen to solidify the compound (Fig. 2e). 6. Remove the compound block from the mold (Fig. 2f ), put it in an appropriate container, and store it at −80°C until use. In this condition, RNA in the tissue can be preserved for at least 3 years, as long as it is maintained strictly at −80°C or below.

3.2. Preparation of Cryosections

1. Cryostat is prepared for sectioning. Set a new microtome blade and cool down the chamber and specimen disc holder. We usually set the temperatures of the chamber and the holder at −20 and −30°C, respectively, but the optimum temperature may differ among cartilage tissues and may need further optimization. A lower temperature may be better for control cartilage.

38  The Use of Laser Capture Microdissection on Adult Human Articular Cartilage…

453

Fig.  2. Preparation of cartilage tissue. (a) Cartilage tissue obtained at surgery. (b) Macroscopically intact area of the cartilage tissue (arrow ) is separated from the underlying subchondral bone in full thickness. (c) Cartilage tissue is cut into a strip (arrow ) which has dimensions of 20 mm × 6 mm. (d) Cross section of the cartilage strip. Arrowheads indicate surface of cartilage. (e) Cartilage tissue is embedded in OCT compound (black arrow ), which is solidified in the mold (white arrow ) in liquid nitrogen. (f ) Close view of the solidified OCT compound containing the cartilage strip. Arrowheads indicate the cross section of the embedded cartilage tissue.

2. A tissue block is mounted on a specimen disc and attached to the cryostat. Adjust the orientation of the specimen discs so that the sections will be cut parallel to the cross section of the cartilage tissue (Fig. 3a). 3. After waiting for 5–10 min to equilibrate the block temperature, cut cryosections into 20–30 mm thicknesses and mount on a glass slide (see Note 5). Place three to five sections on one slide in view of possible loss during processing. 4. These sections are immediately processed for LCM (see Note 6).

454

N. Fukui et al.

Fig. 3. Preparation of cryosections for LCM. (a) Cartilage tissue in solidified OCT compound is mounted on a cryostat. Arrowheads indicate the cross section of the cartilage tissue. (b) Cryosection is prepared and processed for LCM (Arrow ). After processing, the section may curl and detach from the glass slide. (c) LCM cap may be placed directly onto the cryosection. The cap may be pressed down to flatten the section. (d) Photomicrograph of the cryosection obtained through the LCM device. S, M, and D indicate superficial, middle, and deep zones, respectively.

3.3. Processing of Tissue Sections for LCM

1. To complete the processing and LCM as quickly as possible, we usually treat one glass slide at a time (see Note 7). 2. The sections on the glass slide are washed twice briefly with 1–2 ml of ice-cold RNAse-free water to remove the OCT compound. RNAse-free water is placed on a glass slide with a pipette until the entire sections are covered. The water is pipetted up and down several times and then removed by a pipette. 3. Treat the sections with 1–2 ml of 0.5 M EDTA solution for 3 min at room temperature. 4. After removing the EDTA solution, the sections are rinsed once briefly with ice-cold RNAse-free water (see Note 8). 5. The sections are then treated with 75, 95, and 100% ethanol sequentially, each for 30 s. 6. Treat with 100% ethanol again for 30 s (see Note 9). 7. Finally, treat the sections with xylene for 5 min. Then, remove xylene with a pipette and a paper towel and allow the sections to dry up in a fume hood. When completely dry, proceed to LCM immediately (Fig. 3b).

38  The Use of Laser Capture Microdissection on Adult Human Articular Cartilage…

3.4. Acquisition of Cartilage Tissues by LCM

455

1. Place a glass slide and cryosections on the stage of an LCM device (see Note 10). 2. Identify cartilage zones through the LCM device. We usually use a 2× or 4× objective for this purpose (see Note 11). 3. The slide is fitted to the stage by a vacuum chuck equipped in the device. 4. Place an LCM cap on the section following the manufacturer’s instruction. 5. Alternatively, the cap may be placed manually on the section before setting the glass slide on the stage (Fig. 3c). 6. Confirm that the area of interest is seen through the cap (Fig. 3d). 7. Pulse the laser in the area outside the section as a trial to adjust the focus. We prefer to set the spot size of the laser at 15 or 30 mm in diameter. 8. The power and duration of the laser are set at 60–100 mW and 1–3 ms, respectively. Since we use thicker sections, these parameters are greater than those for regular LCM. 9. Pulse the laser repeatedly until the area of interest is entirely attached to the plastic film of the LCM cap. 10. Obtain the area of interest of the tissue (see Notes 12 and 13).

3.5. Extraction of RNA from Tissues Obtained by LCM

RNA is extracted from the harvested tissue by the following protocol. This protocol is based on the manufacturer’s instruction for the RNeasy micro kit, but we made some modifications (see Note 14). 1. Place 400 ml of the RNA lysis buffer contained in the RNeasy micro kit (buffer RLT with 2-mercaptoethanol) into a clean 1.5-ml centrifuge tube. Detach the plastic film and harvested cartilage tissue from the LCM cap, and immerse them in the lysis buffer in the tube (see Note 15). 2. Add 20 ng of bacterial ribosomal RNA in the lysis buffer as carrier RNA, and vortex for 30  s at the maximum speed (see Note 16). 3. Transfer 350 ml of the lysis buffer to a new clean centrifuge tube. 4. Add 350 ml of 70% ethanol to the lysis buffer and mix well by pipetting. 5. Transfer the mixture to a spin column contained in the kit (RNeasy MinElute spin column), and centrifuge for 15 s at ³8,000 × g. 6. Add 350 ml of the wash buffer (buffer RW1) to the column and centrifuge for 15 s at ³8,000 × g.

456

N. Fukui et al.

7. Add 10 ml of DNase I solution prepared as indicated by the manufacturer onto the membrane of the column and incubate for 15 min at RT. 8. Add 350  ml of the wash buffer and centrifuge for 15  s at ³8,000 × g. 9. Add 500 ml of another wash buffer (buffer RPE) to the spin column, and centrifuge for 2 min at ³8,000 × g. 10. Spin the column again for 5 min at full speed to dry the membrane completely. 11. Add 16 ml of RNase-free water onto the membrane, wait for 1 min, and then spin the column for 1 min at full speed to elute the RNA (see Note 17). 3.6. Synthesis of cDNA from Extracted RNA

1. Transfer 12.5 ml of the above obtained eluate to a new clean tube. Since some RNase-free water is lost during the elution, this amount of eluate contains almost all the obtained RNA. 2. Add 10× buffer RT (2 ml), dNTP mix (5 mM each dNTP; 2  ml), oligo dT primer (10  mM; 2  ml), RNase inhibitor (40  U/ml; 0.5  ml) and reverse transcriptase (1  ml), and mix them. Buffer RT, dNTP mix, and reverse transcriptase are contained in the Sensiscrpt kit (see Note 18). 3. Incubate the mixture for 60 min at 37°C. 4. Proceed to qPCR or store at −20°C until use (see Note 19).

4. Notes 1. This protocol is based on our experience with the Arcturus PixCell IIe LCM system. Although we do not have experience with other LCM models, this protocol may work well with them as long as the system uses an infrared laser and thermoplastic films. We do not have experience with microdissection by laser cutting, so another protocol should be prepared for this type of microdissection. 2. To prepare fresh cartilage samples is the first and most important step for successful RNA acquisition by LCM. For OA cartilage from an operation room, we ask the surgeons to put cartilage tissues into the cooled culture media as soon as possible during the surgery. The tissues are kept at a low temperature (4–10°C) and brought to the laboratory immediately, where it is processed upon arrival. For control cartilage from a dissection room, the condition of the sample is more critical. We ask the pathologist to harvest cartilage from donors within 24  h after death. However, even obtained

38  The Use of Laser Capture Microdissection on Adult Human Articular Cartilage…

457

within this time period, RNA in cartilage tissues could have significantly degenerated, depending on the condition of the donor before death. 3. In our experience, OCT compound from Sakura Finetek is most suitable for the preparation of cryosections for LCM for gene expression analysis. Compounds of other brands may inhibit successful RNA acquisition, although they work fine for preparing cryosections. 4. To make tissue blocks, we use a custom-designed metal mold which measures 20 mm in length, 15 mm in width, and 20 mm in depth. It is deeper than regular molds so that the tissue blocks would have a few millimeters of compound layer at the bottom. This compound layer will minimize elevation of tissue temperature when a block is mounted on a cryostat. 5. Although the manufacturer of the LCM device recommends that sections of 5–8 mm in thickness be prepared, the amount of RNA obtained from cartilage sections of this thickness might not suffice for subsequent analyses. For this reason, we usually prepare cartilage sections 20–30 mm thick. 6. To ensure you succeed in obtaining RNA, make it a rule to complete the entire process of LCM (from the preparation of cryosections till the immersion of harvested tissue in the lysis buffer) within 30 min. 7. We do not perform tissue staining for LCM to separate cartilage zones, because such zones are easily identified by histological features without staining. Staining may be performed when LCM is performed for other purposes. 8. EDTA treatment of cartilage sections is essential to perform LCM. Without it, tissues may not attach firmly to the plastic film on an LCM cap, and may detach during tissue separation. EDTA treatment significantly improves the strength of tissue adhesion. From our experience, treatment with 0.5 M EDTA solution for 3 min should be adequate to assure attachment. 9. Sections must be completely dehydrated for LCM. To ensure this, we routinely treat them with 100% ethanol twice before xylene. 10. After dehydration, cartilage sections may curl up and detach from the slide (Fig. 3b). Such sections must be spread out for LCM. Fine-pointed tweezers or pipette tips may be used to do this. 11. Cartilage tissues separated from the subchondral bone by a scalpel may contain calcified cartilage. This zone is easily recognized through a microscope and can be removed by LCM. 12. Tissue separation by LCM may be difficult for some cartilage samples. This is more often the case with control cartilages

458

N. Fukui et al.

from nonarthritic joints because the tissues are tougher than those from OA joints. During tissue separation, the tissue once fixed to the thermoplastic film may detach from the film, or the film may pull away from the LCM cap with the tissue. In such cases, you may have to cut out the attached area of the section using a scalpel and tweezers under a dissection microscope. 13. Even when the area of interest is separated by LCM, the separation may be imperfect and unnecessary parts of the section may come together. We routinely examine the harvested tissue under an LCM device or a dissection microscope to remove any unnecessary parts by a scalpel and tweezers. 14. We use the RNeasy micro kit from Qiagen, but similar kits are now available from other manufacturers, and they may work equally well. 15. Since cell density in cartilage tissues is very low, we often cut and process 3–10 cryosections from a single block and put them together in one tube for RNA extraction. 16. For RNA extraction, we do not use the poly-A RNA in the RNeasy micro kit as carrier RNA, since it might interfere with reverse transcription, given we use oligo-dT primers for the reaction. 17. Although we tried several different conditions, to date, we have not yet succeeded in obtaining RNA of sufficient quality and quantity for cDNA microarray analysis by LCM. The RNA obtained by LCM might be used for microarray analysis if it is amplified, although we have not attempted it. 18. We compared several reverse transcriptases and finally chose Sensiscript from Qiagen. In our experience, this enzyme works consistently well with small amounts of RNA or RNA of relatively low purity. 19. Owing to a limited amount of RNA, for some genes, gene expression may not be evaluated reliably by qPCR. For such genes, it may be worth trying to extract RNA again using an increased number of cryosections. Alternatively, more general methods may be attempted such as altering PCR conditions, changing primer designs, and using another Taq DNA polymerase. For certain genes, it is effective to use random primers instead of, or together with, the oligo dT primer. Occasionally, denaturation of RNA before reverse transcription may improve the result of qPCR.

Acknowledgments This work was sponsored in part by Grants-in-Aids from the Japan Society for the Promotion of Science (Nos. 15390467 and 18390424).

38  The Use of Laser Capture Microdissection on Adult Human Articular Cartilage…

459

References 1. Sandell, L. J., Heinegard, D., Hering, T. M. (2007) Cell biology, biochemistry, and molecular biology of articular cartilage in osteoarthritis, In: Moskowitz, R. W., Altman, R. D., Hochberg, M. C., Buckwalter, J. A., and Goldberg, V. M. (eds) Osteoarthritis. Diagnosis and medical/surgical management, 4th edn. pp 73–106, Lippincott Williams & Wilkins, Philadelphia. 2. Buckwalter, J. A., Mankin, H. J., Grodzinsky, A. J. (2005) Articular cartilage and osteoarthritis. AAOS Instructional Course Lectures, 54, 465–80. 3. Poole, R. A., Guilak, F., Abramson, S. B. (2007) Etiopathogenesis of osteoarthritis, In: Moskowitz, R. W., Altman, R. D., Hochberg, M. C., Buckwalter, J. A., and Goldberg, V. M. (eds) Osteoarthritis. Diagnosis and medical/ surgical management, 4 edn. pp 27–49, Lippincott Williams & Wilkins, Philadelphia. 4. Fukui, N., Ikeda, Y., Ohnuki, T., Tanaka, N., Hikita, A., Mitomi, H., et al. (2008) Regional differences in chondrocyte metabolism in osteoarthritis: a detailed analysis by laser capture microdissection. Arthritis Rheum, 58, 154–63. 5. Fukui, N., Miyamoto, Y., Nakajima, M., Ikeda, Y., Hikita, A., Furukawa, H., et  al. (2008) Zonal gene expression of chondrocytes in osteoarthritic cartilage. Arthritis Rheum, 58, 3843–53.

6. Poole, R. A. (2004) Cartilage in Health and Disease, In: Koopman, W. J., Moreland, Larry W. (eds) Arthritis and Allied Conditions. A Textbook of Rheumatology, 15 edn. pp 223–69, Lippincott Williams & Wilkins, Philadelphia. 7. Kotlarz, H., Gunnarsson, C. L., Fang, H., Rizzo, J. A. (2009) Insurer and out-of-pocket costs of osteoarthritis in the US: evidence from national survey data. Arthritis Rheum, 60, 3546–53. 8. Altman, R., Asch, E., Bloch, D., Bole, G., Borenstein, D., Brandt, K., et  al. (1986) Development of criteria for the classification and reporting of osteoarthritis. Classification of osteoarthritis of the knee. Diagnostic and Therapeutic Criteria Committee of the American Rheumatism Association. Arthritis Rheum, 29, 1039–49. 9. Jinks, C., Jordan, K., Croft, P. (2007) Osteoarthritis as a public health problem: the impact of developing knee pain on physical function in adults living in the community: (KNEST 3). Rheumatology (Oxford), 46, 877–81. 10. Fassler, R., Schnegelsberg, P. N., Dausman, J., Shinya, T., Muragaki, Y., McCarthy, M. T., et al. (1994) Mice lacking alpha 1 (IX) collagen develop noninflammatory degenerative joint disease. Proc Natl Acad Sci USA, 91, 5070–4.

wwwwwww

Chapter 39 Laser-Capture Microdissection of Developing Barley Seeds and cDNA Array Analysis of Selected Tissues Johannes Thiel, Diana Weier, and Winfriede Weschke Abstract Laser microdissection provides a useful method for isolating specific cell types from complex biological samples for downstream applications. In contrast to the texture of mammalian cells, most plant tissues exhibit a cell organization with hard, cellulose-containing cell walls, large vacuoles, and air spaces, thus complicating tissue preparation and extraction of macromolecules such as DNA and RNA. Especially, barley seeds show cell types with enormous differences in osmolarity (degenerating and differentiating tissues) and contain high amounts of the main storage product starch, thus requiring specific procedures for morphological preservation and RNA extraction. In this study, we report about methods allowing tissue-specific gene expression profiling of developing barley seeds. Details on aspects of tissue preparation, including fixation and embedding procedures, laser-capture microdissection, RNA isolation, and linear mRNA amplification to produce high-quality labelled probes for large-scale expression analysis are provided. Particular emphasis is placed on the fidelity of transcript data obtained by the developed methods in relation to the in vivo transcriptome. Key words: Laser-capture microdissection, mRNA amplification, Gene expression analysis, cDNA array, Barley seeds

1. Introduction Laser-assisted microdissection systems and combinations with large-scale gene expression analysis have been developed and frequently used in mammalian systems (1, 2). In plants, these applications have been used to monitor gene expression in a variety of different tissue types, such as epidermal cells, vascular tissues, roots, shoot apical meristems, embryo cells, and siliques of different species (3, 4). Cryosections from frozen tissues provide

Graeme I. Murray (ed.), Laser Capture Microdissection: Methods and Protocols, Methods in Molecular Biology, vol. 755, DOI 10.1007/978-1-61779-163-5_39, © Springer Science+Business Media, LLC 2011

461

462

J. Thiel et al.

the highest yield and the best quality of extracted RNA from microdissected tissues and form the standard method for animal and human tissues. In plant tissues, chemical fixation is the predominant method used for tissue preparation in microdissection, since it guarantees a high morphological integrity, especially when fine structures and small tissue types have to be isolated. Owing to the enormous differences in structure of plant tissues, no standard protocol for tissue preparation exists. Existing protocols vary in the choice of the chemical fixative, the duration of fixation, and the subsequent embedding procedure, and it can be congruently summarized that protocols have to be adapted and optimized for every tissue type. Seed development proceeds in a gradual manner following a defined timescale of tissue-specific differentiation. Nucellar projection and endosperm transfer cells represent the link between maternal and filial seed tissues in barley and mediate nutrient transfer from the maternal seed part into the developing endosperm (5), thereby determining endosperm development. As little is known about specific gene expression patterns in these important tissues, we were interested in tissue-specific transcriptional networks influencing cellular differentiation and assigned functions in nutrient transfer. Recent transcript profiling experiments to study barley grain development using manual separation of maternal and filial tissue fractions has remained critical due to problems in precise allocation of distinct tissue types and the inaccessibility of specific cell layers inside the seed. In this chapter, we describe feasible methods allowing tissuespecific transcriptome analysis of barley seeds. Fixation in ethanol– acetic acid (EAA, Farmer’s fixative) and combination with an alternative embedding procedure using low-melting Steedman’s wax preserved morphological integrity and provided RNA of sufficient quality for subsequent amplification steps. Because of low amounts of extracted RNA, two rounds of T7-based amplification were performed with a labelling step during the second round of in vitro transcription to generate radioactively labelled antisense RNA (aRNA) for hybridization to the barley 12K macroarray. One focal point during methodical development was to get information about the influence of mRNA amplification and fixation/embedding procedures on RNA quality and gene expression levels compared to a conventional probe preparation from untreated frozen grains. Using the developed methods, we were able to identify distinct groups of genes differentially expressed between nucellar projection and endosperm transfer cells. These findings provide the basis to unravel transcriptional networks determining differentiation and function of these tissues in the context of endosperm development.

463

39  Laser-Capture Microdissection of Developing Barley Seeds…

2. Materials 1. Farmer’s fixative (75% ethanol, 25% glacial acetic acid). 2. Ethanol, 100, 75, and 90% in DEPC-treated water. 3. Steedman’s wax, assembled as follows (6): embedding medium was prepared in advance by melting 90 g polyethylene glycol 400 distearate at 60°C and adding 10 g of 1-hexadecanol; the mixture was stirred for several hours and stored at room temperature. 4. Steedman’s wax in ethanol, 25, 50, 75%. 5. Silicon molds (Plano, Wetzlar, Germany). 6. Nuclease-free 1.0-mm PEN-membrane slides (PALM). 7. Rotary microtome (Leica RM 2165). 8. RNase AWAY® (Roth, Karlsruhe, Germany). 9. PALM Laser Microbeam (Palm, Martinsried). 10. 0.5-mL adhesive caps (PALM). 11. RNA isolation reagents, Absolutely RNA Nanoprep Kit (Stratagene). 12. Nuclease-free 0.5, 1.5 and 2.0-mL microcentrifuge tubes. 13. Linear amplification kit, MessageAmpTM Amplification Kit (Ambion).

II

aRNA

14. Programmable thermocycler and regulated hybridization oven. 15. Phosphorylated UTP ([a-33P]-UTP, 3000 Ci/mmol, GE healthcare). 16. 12K barley seed array, designed and spotted according to (7). 17. Phosphor scanner (Fuji FLA 5100).

3. Methods In the following sections, methods are described for (1) tissue preparation, (2) laser-capture microdissection, (3) extraction and analysis of total RNA, (4) T7-based amplification of mRNA and preparation of labelled aRNA probes for 12K barley macroarray hybridization, and (5) quality assessment of macroarray data . 3.1. Tissue Preparation

The protocol in this chapter describes the usage of ethanol–acetic acid (EAA) fixative combined with Steedman’s wax in the embedding procedure (see Note 1). Whereas protocols for EAA fixation are well known in plant science, the embedding in low-melting-point

464

J. Thiel et al.

Fig. 1. Laser microdissection and laser pressure catapulting (LMPC) of the nucellar projection from 20-mm cross sections of EAA-fixed barley grains mounted on PEN-membrane slides. (a) Barley grain at 8 days after flowering (DAF). Black area indicates the position of cross section. (b) Median cross section. (c) Magnification of the region around NP, cutting of NP with a laser beam. (d) Section after pressure catapulting with a defocused laser beam. (e) Collection of complete sections of NP in the lid of an adhesive cap. NP nucellar projection, P pericarp, SE starchy endosperm, TC endosperm transfer cells. Bars = 100 mm.

polyester (Steedman’s wax) at 40°C provided a more gentle treatment of sample material avoiding high temperatures as necessary for paraffin embedding and resulted in a significant improved preservation of morphology (Fig. 1). 1. Caryopses were divided into three parts using a razor blade, placed in Farmer’s fixative, and fixed overnight at 4°C. After fixation, samples were washed in 75% ethanol, dehydrated in a graded series of ethanol, and infiltrated with increasing concentration of Steedman’s wax at 40°C. The details of the procedure are presented in Table 1. Objects were then put into embedding molds and left to polymerize overnight at room temperature. 2. Prior to sectioning, PALM membrane slides were irradiated with UV light at 254 nm for 30 min. Owing to irradiation,

39  Laser-Capture Microdissection of Developing Barley Seeds…

465

Table 1 Fixation and embedding procedure Solution

Time/exposure

Temperature (°C)

Farmer’s fixative (75% ethanol, 25% Glacial acetic acid)

Overnight

75% ethanol in DEPC-treated water

Four times for 15 min

22

90% ethanol in DEPC-treated water

1 h

22

100% ethanol

1 h

22

100% ethanol

1 h

40

25% Steedman’s wax in ethanol

3 h

40

50% Steedman’s wax in ethanol

Overnight

40

75% Steedman’s wax in ethanol

3 h

40

Steedman’s wax

3 h

40

Steedman’s wax

Overnight

40

Steedman’s wax

3 h

40

4

membranes show a higher hydrophilicity, facilitating the adhesion of sections (see Note 2). For removal of RNase activity, the slides were dipped into pure RNase-AWAY® solution for some seconds, followed by two washing steps in DEPC-treated water and drying at 37°C. 3. Ribbons of 20-mm cross sections were prepared with a rotary microtome, mounted onto membrane slides, and stretched by addition of a small drop of water to one end of the ribbon. The slides were allowed to dry overnight at room temperature. 4. Prior to microdissection, Steedman’s wax was removed by incubating the slides in ethanol (two times for 10 min) and dried for 30 min at 37°C. Otherwise, embedded sections can be stored at 4°C in sealed slide boxes in the presence of a desiccant (see Note 3). 3.2. Laser-Capture Microdissection

1. The PALM Laser Microbeam instrument was used to isolate distinct tissues from dewaxed, dried cross sections of barley grains. 2. The power of the laser beam (diameter <1.0 mm) was adjusted to 45–60 mW for cutting and to 70–90 mW for laser pressure catapulting. PALM RoboSoftware was used as graphic tool for targeting of cells. The sections from nucellar projection and transfer cells were catapulted into the lid of 0.5-mL PALM adhesive caps.

466

J. Thiel et al.

3. Typically, between 12 and 25 sections were processed per cap. One advantage of using PEN membrane slides is that specimens are cut out and catapulted completely so that efficiency of tissue transfer to the lid of reaction tubes can easily be controlled by microscope. Sections from two tubes were pooled prior to RNA isolation. 3.3. RNA Isolation

There are a number of commercially reagent systems available for isolating micro-amounts of RNA. Most of them are column-based systems avoiding organic extraction, phase separation and precipitation steps and, thus, probably minimizing loss of RNA yield during these steps. The RNA purification described here is done with the Absolutely RNA nanoprep Kit (Stratagene) using columns with a silica-based fiber matrix for RNA binding. The procedure has been conducted after manufacturer’s instructions with slight modifications, which are described in detail. Quality assessment of total RNA extracted from EAA-fixed and Steedman’s wax-embedded sections (see Note 9) revealed RNA of sufficient quality (Fig. 2). 1. Prepare lysis buffer by adding 0.7 mL ß-mercaptoethanol to 100 mL lysis buffer for each sample. Lysis buffer is preheated to 60°C in a heating block for 10 min. 2. Centrifuge PALM adhesive caps containing microdissected tissues for 2  min (12,000 × g) to spin down sections. Add 100 mL of lysis buffer and vortex tubes for 2 min (also upsidedown for elution of possible leavings in the tube cap). After a short spin in a microcentrifuge, vortexing was repeated again for 2  min. The sections in lysis buffer were incubated for 15 min at 60°C (see Note 4). 3. Add an equal volume of 70% ethanol (usually 100 mL) to the cell lysate and vortex thoroughly for some seconds. Transfer the mixture immediately to an RNA-binding nano-spin column seated in a 2-mL collection tube and spin for 1  min at ³12,000 × g. 4. Discard filtrate and place the column in the same tube. Add 300  mL of 1× low-salt wash buffer onto the column and spin the sample in a centrifuge for 1  min at ³12,000 × g. Retain tube, discard filtrate, and spin the column again for 2 min (³12,000 × g) to dry the filter matrix. It is important that the filter matrix is completely dry; otherwise, remaining ethanol residues would inhibit on-column DNase digestion. Place the column in a new 2.0-mL tube. 5. Add 24 mL of DNase solution (4 mL DNase I reconstituted in 20 mL DNase digestion buffer) directly on the fiber matrix of the column. Incubate at 37°C for 15 min. 6. Add 300 mL of 1× high-salt wash buffer and spin in a centrifuge for 1 min at ³12,000 × g. Retain spin cap and discard filtrate.

39  Laser-Capture Microdissection of Developing Barley Seeds…

467

Fig. 2. Representative examples of RNA quality from EAA-fixed probes. Total RNA was extracted from EAA-fixed and Steedman’s wax-embedded sections of barley grains. 50 ng of total RNA was amplified by T7-based in vitro transcription for generating labelled probes. Microcapillary electrophoretic analysis using an Experion RNA HighSens Analysis Kit (BioRad Laboratories) of 5 ng total RNA [upper electropherogram in (b), middle lane in gel-like image in (a)] revealed relative intact total RNA with distinct fluorescent peaks and electrophoretic bands of 18S and 28S ribosomal subunits. A slight degradation of RNA is visible by nearly equal amounts of 28S and 18S peaks (ratio 28S:18S ~0.9). Antisense RNA (aRNA) synthesized from 50 ng total RNA displayed a distribution of sizes from 300 to 2,000 nucleotides after the first round of amplification, with a maximum between 500 and 1,000 nucleotides [lower electropherogram in (b), right lane in gel-like image in (a)] implicating that marginal decline in RNA quality does not affect size distribution of amplified aRNA.

Perform two additional washing steps with 300 mL of 1× low-salt wash buffer as described above. 7. Discard filtrate and transfer the column to a dry 2.0-mL tube. Spin the column again for 3  min (³12,000 × g) to dry the fiber matrix. 8. Transfer the column to a new 2.0-mL collection tube and add 13 mL of preheated elution buffer directly on the fiber matrix. The elution buffer has to be warmed to 60°C for 10 min to increase the yield of RNA. The sample is incubated for 5 min at room temperature. 9. Spin the sample in centrifuge for 5 min at ³12,000 × g. The purified RNA (~12 mL) is in the eluate in the collection tube. One microliter can be used for estimation of RNA amount and/or quality, while the remaining ~11 mL is used for subsequent amplification steps.

468

J. Thiel et al.

3.4. Linear Amplification of mRNA and Preparation of Labelled Targets for Macroarray Analysis

Up to 30  ng of total RNA could be extracted from 25 to 50 microdissected specimens (8, see Note 5). Subsequently, two successive rounds of in vitro transcription are necessary to generate sufficient amounts of labelled probes for cDNA array analysis (see Note 6). In our protocol, we used the MessageAmpTM II aRNA Amplification Kit (Ambion) based on the RNA amplification protocol developed from Eberwine and Van Gelder (9). Using this reagent system, each probe yielded around 400 ng of aRNA after the first round of amplification and around 1.6  mg of labelled aRNA after the second round, sufficient amounts to perform transcriptome analysis with the 12K barley macroarray or with other platforms for high-throughput expression analyses (10). Amplified RNA from EAA-fixed and Steedman’s wax-embedded sections (see Note 9) displayed a size distribution similar to amplified RNA from frozen samples implying few degradation and disruption during the probe preparation procedure (Fig. 2). The procedure has been conducted after manufacturer’s instructions, but throughout the protocol some points should be remembered: 1. Consistency in work flow is very important for amplification experiments. Standardize reaction incubation times and prepare master mixes for every step of the amplification procedure to minimize effects of pipetting errors. 2. We recommend using a calibrated hybridization oven for prolonged 37°C or 42°C incubations and a calibrated thermal cycler for the 16°C second strand synthesis reaction incubation. Because this step is very sensitive to temperature and temperatures above 16°C compromise the yield of doublestranded cDNA, it is very important to use either thermal cycler with a regulated lid temperature that matches the block temperature or not to close the lid if you use a static system. In general, condensation in the reaction tubes during any of the incubations leading to changes in the composition of reaction mixtures should be strictly avoided.

3.4.1. Reverse Transcription to Synthesize First Strand cDNA

1. Add 1 mL T7 oligo dT primer to the eluted RNA (11 mL) in a fresh 0.5-mL tube. Mix the reagents, incubate at 70°C for 10 min, spin briefly, and place on ice. 2. Add 8 mL of reverse transcription master mix (2 mL 10× first strand buffer, 4 mL dNTP mix, 1 mL RNase inhibitor, 1 mL Array script) to each tube and incubate the final 20 mL reaction at 42°C for 2 h. 3. Briefly spin the reactions and incubate on ice for 5 min.

3.4.2. Second Strand cDNA Synthesis

1. On ice, prepare second strand master mix (63 mL nucleasefree water, 10 mL 10× second strand buffer, 4 mL dNTP mix, 2 mL DNA polymerase, and 1 mL RNase H) and add immediately to each sample.

39  Laser-Capture Microdissection of Developing Barley Seeds…

469

2. Mix thoroughly by pipetting up and down and flicking the tube 3–4 times and spin briefly to collect the reaction in the bottom of the tube. Incubate the final 100 mL reaction in a thermal cycler at 16°C for 2 h. 3.4.3. cDNA Purification

1. Add 250 mL cDNA binding buffer to each sample and mix thoroughly as described above. After a short spin, proceed quickly to the next step. 2. Pass the mixture through a cDNA filter cartridge and centrifuge for 1 min at 10,000 × g. All centrifugations in the purification procedure should be done at 10,000 × g at room temperature. Higher RCFs could cause mechanical damage or may deposit glass filter fiber in the eluate. 3. Apply 500 mL wash buffer to the cartridge and centrifuge for 1 min. Discard filtrate and centrifuge the filter cartridge for an additional minute to remove trace amounts of wash buffer. 4. Transfer column to a new 2.0-mL tube and apply 18  mL nuclease-free water (preheated to 55°C for at least 10 min) to the center of the filter in the cDNA cartridge. 5. Leave at room temperature for 5 min and centrifuge for 2 min at 10,000 × g. The double-stranded cDNA will now be in the eluate (~16 mL).

3.4.4. In Vitro Transcription to Synthesize aRNA

1. Add 24  mL IVT master mix (4  mL ATP, CTP, GTP, UTP (each 75 mM) and 4 mL of 10× reaction buffer and T7 enzyme mix) to each cDNA and mix thoroughly by pipetting and flicking the tube. For samples that will undergo two rounds of amplification only unmodified UTP can be used in the IVT reaction! 2. Incubate for 14–16 h (usually overnight) at 37°C in a calibrated hybridization oven. We generally recommend long incubation times up to 16 h when working with small tissue amounts. 3. Add 60 mL nuclease-free water to each sample to bring the volume to 100 mL and to stop the reaction. Incubate samples on ice and proceed to the aRNA purification step.

3.4.5. aRNA Purification

1. It is important for aRNA yield that the IVT reaction is exactly brought to 100 mL. Add 350 mL aRNA binding buffer and immediately 250 mL of 100% ethanol to each sample and mix by pipetting the mixture up and down three times. Do not vortex to mix and do not centrifuge! Pipet the mixture (700 mL) immediately to the center of the filter in the aRNA cartridge and centrifuge for 1 min. Any delay in this procedure could result in a loss of aRNA, since aRNA is in a semiprecipitated state when ethanol is added.

470

J. Thiel et al.

2. Wash with 650  mL wash buffer, centrifuge, and discard the filtrate. Spin the column again for 1 min to remove traces of wash buffer and transfer the cartridge to a new 2.0-mL tube (provided RNA collection tube). 3. Add 100 mL of nuclease-free water (preheated to 60°C!) to the center of the filter. Leave at room temperature for 5 min and centrifuge for 2 min. Purified aRNA will now be eluted in 100 mL water. 3.4.6. Second Round Amplification to Synthesize Radioactively Labelled Probes

The procedure is similar to the first round of amplification resulting in equivalent reaction products, but the reaction setup is different for some specific steps. Random primers (provided as “second round primers”) are used for the synthesis of first strand cDNA. After first strand synthesis, an aRNA digestion step using RNase H is inserted to degrade aRNA and to leave only cDNA as template for second strand synthesis. Second strand cDNA synthesis is primed by T7 oligo dT primer. 1. Because we work with very low RNA concentrations, purified aRNA from the first round of amplification has to be concentrated to a volume of £10 mL prior to first strand cDNA synthesis by using a vacuum centrifuge. It is important that the heating of the vacuum centrifuge is turned off and to avoid to dry the aRNA completely. Both, heat and overdrying, compromise aRNA quality and, thus, impede reverse transcription. Using the volumes mentioned above, careful concentration of samples requires 1.5–2  h with checking the progress of drying every 5 min after 1.5 h. 2. The probes were labelled during the in  vitro transcription reaction to synthesize [a-33P]-UTP labeled aRNA. The IVT master mix was prepared with 4  mL [33P]-UTP instead of unmodified UTP corresponding to a concentration of 13.3 pmol UTP in the master mix. We decided to use solely radioactively labeled UTP for the reaction, because T7 RNA polymerase activity is not hampered by 33P and thus, guarantees the highest incorporation rates. During aRNA purification procedure, incorporation rates of [33P]-UTP could be determined by using a Geiger counter.

3.5. Target Hybridization, Data Analysis, and Fidelity Confirmation of Expression Data

One main obstacle for interpreting data generated by such multistep procedures is how reliable and reproducible these data are. As it is well known that every round of amplification introduces a certain bias in global transcript abundance and that every chemical treatment has an impact on RNA quality, we evaluated the influence of mRNA amplification (see Note 8) and tissue preparation procedure (see Note 9) on gene expression levels separately by comparison to a nonamplified and untreated reference probe (see Note 7).

39  Laser-Capture Microdissection of Developing Barley Seeds…

471

Data from independent two-round amplifications from frozen and from fixed and embedded material revealed a high correlation to the reference probe representing the in vivo trancriptome (Fig. 3a, c). This confirms that probe preparation procedure described in this protocol is feasible to conduct tissue-specific gene expression profiling and allows the identification of differentially expressed genes between microdissected tissues from barley seeds (Fig. 3e). 1. 12K barley macroarray was prehybridized for at least 1 h at 65°C in 12 ml of Church buffer (0.5 M sodium phosphate pH 7.2, 7% SDS, 1% BSA, 1 mM EDTA) containing sheared salmon sperm DNA. Purified labelled probes (100 mL) were denatured (5 min, 70°C) and cooled on ice (5 min) before hybridization. Probes were added together with fresh Church buffer (12 ml, preheated to 65°C) and hybridized at 65°C for at least 12 h. 2. Macroarrays were washed three times with 40  mM sodium phosphate pH 7.2, 1% SDS, 2mM EDTA for 15 min at 65°C and shrink-wrapped in a plastic bag. Macroarrays were exposed to an imaging plate for 20 h and scanned at 100 mm resolution using a Fuji FLA 5100 phosphor scanner (Tokyo, Japan). 3. Images of hybridized nylon membranes were subjected to automatic spot detection using a suite of customized MATLAB programs. The total number of 11,786 genes per array is covered by 23,572 double spots, enabling one technical replicate per gene for quality control. Quantile normalization was carried out on the data sets according to Bolstad (2003). For data comparison, normalized expression values were log2­ transformed and plotted against each other using MATLAB software.

4. Notes 1. Despite the fact that it would be useful to work with cryosections from frozen tissues as it is established in mammalian systems, the different cell organization of plant tissues frequently results in nonacceptable histological integrity after freezing. If small tissue types or cell layers have to be isolated, tissue sections of high quality are needed, as well as resolution, ensuring a precise identification of target cells. We recommend using a precipitative fixative (e.g., EAA) which provides a compromise between RNA quality and histological integrity. When more crude tissue types or fragments will be isolated and the histological quality is acceptable, we would

472

J. Thiel et al.

Fig. 3. Fidelity and reproducibility of expression signals obtained after hybridization of amplified and nonamplified probes and comparison of expression signals from microdissected tissues. Representative scatter plots of log2-transformed signal intensities display correlation between the different probes. Signal intensities differing less than twofold are bracketed. Numbers within each plot represent the pearson correlation (r). (a) Comparison of a two-round amplification with 50 ng input RNA to a non-amplified probe with 35 mg RNA (reference) from a common source indicates high concordance of gene expression levels. (b) Reproducibility of independent two-round amplifications with 50 ng input amount from a common source. (c) Comparison of a probe from EAA-fixed sections of a completely sectioned caryopsis after two rounds of amplification with 50 ng input RNA to the nonfixed and nonamplified reference probe showed marginal additional effects of fixation and embedding on fidelity of expression data. The correlation coefficient of 0.76 (and 0.78 for the second biological replicate) corresponds to overlapping expression of 72–74% of the genes, demonstrating that the developed methods generate high-quality probes reproducing the original mRNA profile to a large degree. (d) Reproducibility of expression values from two independent biological replicates of EAA-fixed sections of completely sectioned caryopses. (e) Comparison of signal intensities from one biological replicate of endosperm transfer cells (TC1) and of nucellar projection (NP1) revealed high scatter, indicating a large number of differentially expressed genes between the tissue types. (f) Reproducibility of expression values from two independent biological replicates of endosperm transfer cells (TC1, TC2) generated by the described methods.

39  Laser-Capture Microdissection of Developing Barley Seeds…

473

suggest using cryosections, guaranteeing the best quality and highest yield of extracted RNA. 2. Mounting of Steedman’s wax-embedded sections to PENmembrane slides is a critical point, because Steedman’s wax displays different adhesive properties than other commonly used waxes (e.g., paraffin). Therefore, it is a prerequisite to irradiate slides with UV light before usage. 3. It is crucial for the laser-assisted cutting and catapulting procedure that sections are totally dried. Any traces of moisture will inhibit transfer of microdissected cells and bear the risk of RNA degradation. If it is necessary to store embedded sections or microdissected cells, use always sealed slide boxes in the presence of a dessicant (e.g., silica gel) for storage at 4°C (embedded sections) and −80°C (microdissected tissues), respectively. In general, prolonged storage should be avoided, and sample material should be used as fast as possible. 4. Complete lysis of sample material is mandatory for getting optimal yields and quality of extracted RNA. We recommend to preheat the lysis buffer and to incubate the microdissected cells in the lysis buffer for 15 min at 60°C, as our own experiments showed that the yield of total RNA is significantly increased upon heat incubation. The lysis buffer contains guanidine thiocyanate which prevents RNase activity. 5. A precise estimation of RNA yields from these minimum tissue amounts is rather difficult. Based on our experimental data from a fiber-optic spectrophotometer and microcapillary electrophoresis the yield of total RNA from tissue amounts we used generally ranges between 3 and 30 ng RNA. More reliable results in terms of RNA quantities can be achieved after the first round of amplification. 6. If in some cases two rounds of amplification will not yield enough labelled aRNA for high-throughput expression analysis, a third round of amplification can be performed to increase the total amount of aRNA. However, our experimental data revealed that hybridization of these probes to barley 12K array results in an unfavorable signal-to-noise ratio. Owing to the higher background, a reduced number of transcripts were detected compared to two-round amplifications. We recommend collection of sample material sufficient for the two-round amplification procedure. 7. As one focus of methodical establishment was to check the fidelity of transcriptome data obtained by our probe processing procedure we had to define a reference transcriptome. Reference probes were prepared from purified total RNA (35 mg) from shock-frozen barley grains of the same developmental stage, which was used to synthesize [33P]-CTP-labelled

474

J. Thiel et al.

cDNA probes. This conventional probe preparation procedure was defined as base for the different comparisons and was designated as in vivo transcriptome. 8. To assess the effects of a two-round amplification on global transcript abundance we prepared labelled probes using the described methods starting with 50 ng total RNA from the same shock-frozen material used for the reference probe. After extraction and a first round of amplification, probes were labelled with [33P]-UTP during the second round, hybridized to the barley 12K array and compared with the reference probe. 9. Beside the influence of mRNA amplification, we wanted to get information about the influence of EAA fixation and Steedman’s wax embedding on RNA quality after extraction and amplification and on gene expression levels. Therefore, complete, fixed barley grains were sectioned, dewaxed in ethanol, air-dried, and homogenized with a mortar and pestle. A small proportion of the mixture was used for RNA extraction and subsequent analyses. The whole procedure of probe processing is included, and the comparison to the reference probe allows a precise estimation of aspects influencing fidelity of expression data important in the way that this represents the framework for the analyses of microdissected tissues.

Acknowledgments We are grateful to Uta Siebert for her excellent and expert assistance in tissue processing and operating of the PALM Laser Microbeam instrument. We also wish to thank Ursula Tiemann and Karin Lipfert for graphical artwork. This work was supported by the Deutsche Forschungsgemeinschaft (DFG, FKZ 39205123) and by the Federal Ministry of Education and Research (BMBF, FKZ 0313821A). References 1. Emmert-Buck MR, Bonner RF, Smith PD, Chuaqui RF, Zhuang Z, Goldstein SR, Weiss RA and Liotta LA (1996) Laser capture microdissection. Science 274:998–1001 2. Luo L, Salunga RC, Guo H, Bittner A, Joy KC, Galindo JE, Xiao H, Rogers KE, Wan JS, Jackson MR and Erlander MG (1999) Gene expression profiles of laser-captured adjacent neuronal subtypes. Nat Med 5:117–122 3. Nelson T, Tausta SL, Gandotra N and Liu T (2006) Laser microdissection of plant tissue:

what you see is what you get. Annu Rev Plant Biol 57:181–201 4. Day RC, McNoe LA and Macknight C (2007) Transcript analysis of laser microdissected plant cells. Physiol Plant 129:267–282 5. Weschke W, Panitz R, Sauer N, Wang Q, Neubohn B, Weber H and Wobus U (2000) Sucrose transport into developing barley seeds: Molecular characterization of two transporters and implications for seed development and starch accumulation. Plant J 21:455–467

39  Laser-Capture Microdissection of Developing Barley Seeds… 6. Vitha S, Baluska F, Mews M and Volkman D (1997) Immunofluorescence detection of F-actin on low melting point wax sections from plant tissues. J Histochem Cytochem 45:89–95 7. Sreenivasulu N, Radchuk V, Strickert M, Miersch O, Weschke W and Wobus U (2006) Gene expression patterns reveal tissue-specific signaling networks controlling programmed cell death and ABA-regulated maturation in developing barley seeds. Plant J 47:310–327 8. Thiel J, Weier D, Sreenivasulu N, Strickert M, Weichert N, Melzer M, Czauderna T, Wobus U, Weber H and Weschke W (2008) Different

475

hormonal regulation of cellular differentiation and function in nucellar projection and endosperm transfer cells – a microdissectionbased transcriptome study of young barley grains. Plant Physiol 148:1436–1452 9. Van Gelder RN, von Zastrow ME, Yool A, Dement WC, Barchas JD and Eberwine JH (1990) Amplified RNA synthesized from limited quantities of heterogenous cDNA. Proc Natl Acad Sci USA 87:1663–1667 10. Duggan DJ, Bittner M, Chen Y, Meltzer P and Trent JM (1999) Expression profiling using cDNA microarrays. Nat Genet 21:10–14

wwwwwww

Chapter 40 Quantitative RT-PCR Gene Expression Analysis of a Laser Microdissected Placenta: An Approach to Study Preeclampsia Yuditiya Purwosunu, Akihiko Sekizawa, Takashi Okai, and Tetsuhiko Tachikawa Abstract Preeclampsia is still one of the leading causes of maternal and neonatal mortality and morbidity. Despite intensive research, the cause of preeclampsia has not yet been established. However, for a variety of reasons, the numerous aspects of normal and pathological pregnancies, particularly with regard to preeclampsia, remain difficult to study and are, therefore, poorly understood. The development of a laser-based microdissection system has provided rapid morphologically and phenotypically distinct types of cells in the placenta for molecular analysis. Alterations in gene expression in the cytotrophoblast and syncytiotrophoblast or other specific placental cell types from patients who later develop preeclampsia have been reported. Laser microdissection is an attractive method to study each specific placental cell type to characterize the development of preeclampsia or to study the pathophysiology of preeclampsia. This method may contribute to a better understanding of the pathophysiology of preeclampsia. Key words: Preeclampsia, Laser microdissection, Placenta, RT-PCR

1. Introduction Despite intensive research, the cause of preeclampsia (PE) has not yet been established, and PE is still a leading cause of maternal and perinatal mortality and morbidity. PE is a multisystem disorder unique to human pregnancy, and its clinical features are well recognized, characteristically manifesting in the second to third trimester. The underlying pathology is associated with a failure of trophoblastic invasion of the maternal arteries during early gestation. Several theories regarding its etiology have been proposed

Graeme I. Murray (ed.), Laser Capture Microdissection: Methods and Protocols, Methods in Molecular Biology, vol. 755, DOI 10.1007/978-1-61779-163-5_40, © Springer Science+Business Media, LLC 2011

477

478

Y. Purwosunu et al.

over recent decades, including defective placental vascular modeling early in pregnancy, genetic polymorphism, immune tolerance, vascular endothelial cell activation, and exaggeration of a systemic inflammatory disease that might induce reactive oxygen species (ROS) (1–5). Because PE has a long preclinical phase before clinically manifesting during later gestation, clinical prediction is believed to be possible, and clinical prediction may offer an early opportunity for intervention. A variety of methods have been reported as potential early markers for PE. However, none of these alone have sufficient clinical discrimination for use in clinical practice. However, for a variety of reasons, numerous aspects of normal and pathological pregnancies, particularly with regard to PE, remain difficult to study and are, therefore, poorly understood. The use of all tissue in the placenta for the quantitative measurement by real-time PCR results in an average value that accounts for all types of tissue in the placenta, and which is present in an unknown proportion. For example, direct evidence of alterations of gene expressions in the cytotrophoblasts or syncytiotrophoblasts from patients destined to develop PE later has not yet been reported. This requires the preparation of pure samples of specific cell types. Among other methods, laser microdissection offers the high-resolution differentiation of sample composition by the selection of individual cells or a group of cells (6). Laser microdissection and pressure catapulting (LMPC) is an attractive method for examining the pathophysiology of PE by isolating specific cells in a noninvasive and contamination-free manner (7). Using LMPC and real-time PCR and RT-PCR, it is now feasible to study genetic alteration (8–10), gene expression features (11, 12), and protein expression in each placental cell type in control and preeclampsia patients. Therefore, in preeclampsia patients, each type of tissue pathophysiology can be studied using LMPC, which allows rapid sample procurement and the collection of a specific type of cells that can be used for investigating the pathophysiology of preeclampsia.

2. Materials 2.1. Samples Preparation

1. Surgical tissue knife and pin set for sampling and cutting the placenta into small segments. 2. Isopentane (Wako Pure Chemical, Japan) cooled in liquid nitrogen (N2O). 3. Liquid nitrogen jar (Fig. 1a), on which a stainless-steel cup (Fig.  1b) filled with isopentane (Fig.  1c) is cooled by the liquid nitrogen.

40  Quantitative RT-PCR Gene Expression Analysis of a Laser Microdissected…

479

Fig. 1. The instruments for obtaining frozen tissue sections. A liquid nitrogen jar (a), on which a stainless-steel cup (b) filled with isopentane (c) is cooled by liquid nitrogen. The Optimal Cutting Temperature (OCT) compound (Sakura Finetek, Corp., Tissue-Tek, Cat, no. 4583, (d)) is used for cryopreservation.

4. Optimal Cutting Temperature (OCT) compound (Sakura Finetek, Corp. Tissue-Tek, Fig. 1d) for cryopreservation. 5. Snap-freeze plastics template Cryomold (Sakura Finetek, Corp., Tissue-Tek, Cat, no. 4557, Fig.  1e) for embedding samples into OCT. 6. Pin set to hold the sample during snap-frozen process and −80°C refrigerator. 7. A refrigerated microtome HM 550 Cryostat (Carl Zeiss, Germany). 8. Special microdissection glass slides covered UV-absorbing membrane (Membrane Slide, Carl Zeiss MicroImaging). 9. Staining solutions: (a) Toluidine blue solution (Wako Pure Chemical, Japan). (b) Sodium tetraborate (Wako Pure Chemical, Japan). (c) 100% ethanol (Wako Pure Chemical, Japan). (d) 100% methanol (Wako Pure Chemical, Japan). (e) DEPC-treated water (Nippon Gene, Japan). 10. 50-ml conical tube (Becton Dickinson, USA). 11. Refrigerated centrifuge. 2.2. Laser Microdissection

1. PALM Laser Microdissection unit using a 337-nm nitrogen laser (PALM MBIII, Carl Zeiss MicroImaging, Munich, Germany). 2. Laser Microdissection Microtube (AdhesiveCaps opaque, Carl Zeiss MicroImaging).

480

Y. Purwosunu et al.

2.3. Quantitative RT-PCR

1. RNA extraction using RNeasy Plus Micro Kit (Qiagen, Germany). 2. Sensiscript Reverse Transcription Kit (Qiagen, Germany). 3. Oligo dT (Promega, USA). 4. RNase Inhibitor (Quiagen). 5. Master mix: 10× buffer RT, 2.0  ml, 1×; dNTP mix (5  mM each dNTP), 2.0 ml, 0.5 mM each dNTP, Oligo-dT primer (10 mM), 2.0 ml, 1 mM; RNase inhibitor (10 U/ml)†, 1.0 ml, 10 U (per reaction), Sensiscript Reverse Transcriptase, 1.0 ml; Template RNA, <50  ng(per reaction), RNase-free water to 20 ml. 6. ABI 7900 (Applied Biosystem, USA). 7. Qiagen QT mix with supplied H2O (Qiagen, Germany). 8. 20× Primer-Probe mix (Assay-on-Demand, Applied Biosystems).

3. Methods 3.1. Preparation of Fresh Placental Samples

1. Prepare fresh placenta (less than 15 min) from cesarean section or normal delivery. Store the placenta in 4°C during placental transport from the delivery room to the laboratory (see Notes 1–3). 2. Cut the placenta into 3 × 10 × 10  mm cubes with the full thickness of placenta maintained, to obtain all placental cell types. The placenta sampling method was the Salafia method, as previously described (13). 3. Place the minced placental samples into the snap-frozen plastic template Cryomold embedded into the OCT compound (Fig. 2). Ensure that the samples are fully embedded into the OCT compound. The samples must be placed in a position such that the full thickness of placenta can be sliced using a Cryostat. 4. Place the embedded samples into liquid nitrogen-cooled isopentane (Fig. 2b, c). 5. After the surface of OCT compound is frozen, the embedded samples must be immersed in the isopentane for several seconds (Fig. 2d). Thereafter, the samples can be stored at −80°C until further processing and the subsequent molecular analysis. 6. Cut tissue sections using a refrigerated microtome HM 550 Cryostat (Carl Zeiss, Germany, Fig. 3). It is recommended to cut OCT-embedded tissue blocks at 8–10mm. The temperature of the microtome should be at −20°C. This thickness allows for good histological visualization, and it is technically easier to laser-microdissect the tissue (see Notes 4–6).

40  Quantitative RT-PCR Gene Expression Analysis of a Laser Microdissected…

481

Fig. 2. Methods to make frozen tissue samples. The cut placenta samples were placed into a snap-frozen plastic template Cryomold and embedded into the OCT compound (a). The embedded samples were placed into isopentane, which was cooled by liquid nitrogen (b and c). After the surface of OCT compound is frozen, the embedded samples are immersed in the isopentane for several seconds (d). Thereafter, the samples could be stored in −80°C until further processing and molecular analyses.

Fig. 3. The tissue sections were cut using a refrigerated microtome HM 550 Cryostat.

7. Place the sliced samples onto a specialized UV-absorbing glass slide for laser microdissection. Ensure that the OCT along with the sample is attached to the slide by warming the slide using body heat or 37°C, and stain with toluidine blue thereafter.

482

Y. Purwosunu et al.

8. Prepare individual staining dishes with the following solutions and treat slides for the described duration. All solutions and staining are conducted at room temperature. After each incubation period, briefly drain the slide and move it to the next solution. (a) 90% ethanol, 30 s to 1 min. (b) 100% methanol, 30 s to 3 min. (c) Wash with DEPC-treated water, briefly drain. (d) Stain in 1% toluidine blue solution with 1% sodium tetraborate, approximately 15 s (tailor the time according to the tissue thickness or quality of samples histological structure under a light microscope in previous samples). (e) Wash completely with DEPC-treated water. (f ) Put two slides with opposing faces into 50-ml conical tube (Becton Dickinson, USA). Centrifuge the tube at 3,000 × g for 30–40 s at 4°C to clean the samples from the OCT substance, which will inhibit subsequent reverse transcription (RT) reactions. 9. The stained tissue section is ready for laser microdissection. Ensure that the slides are cooled (4°C or in dry ice) before laser microdissection to avoid RNA degradation. Do not store the slide; the slides must be processed immediately with the laser microdissection step. 3.2. Laser Microdissection of Placenta

1. Place the glass slide with the stained tissue section on the laser microscope stage. 2. Place the microtube with the laser microdissection cap facing down (Fig. 4). 3. Visualize the tissue section through the microscope. Take a low-magnification image.

Fig. 4. (a) The microtube was placed with the laser microdissection cap facing down. (b) The strategy of cell isolation. The laser beam originates from the lower section. The irradiated cells are moved to the cap.

40  Quantitative RT-PCR Gene Expression Analysis of a Laser Microdissected…

483

Fig. 5. Process of microdissection.

4. Before beginning the microdissection, set the appropriate laser parameter (focus, power, and pulse duration). Test the laser in a free space with no cap to evaluate the laser parameter (see Note 7). 5. Take the predissection image. 6. Excise the borderline of the tissue of the targeted cell type using the laser (Fig.  5a, b). After the microdissection of the complete periphery, fire the laser on the middle of the tissue specimen and guide the laser up the tissue to the microtube cap (Fig. 5c, d). In each sample, the number of microdissected cells is estimated to be approximately 2,500–3,000 cytotrophoblasts or syncytiotrophoblasts. The amount of cells can also be estimated on the cumulative surrounding 1 mm2 area. 7. Remove the cap of the slide. Acquire a postdissection image. 8. Photograph the cap image. The dissected cells can be visualized (Fig. 5c). 9. Take note of the efficiency of tissue lifting onto the cap and the presence of any contaminating cells. 10. Place lysis buffer into the cap of the microtube and centrifuge. 3.3. mRNA Extraction

1. Adjust the sample volume to 75 ml with buffer RLT. Note: If processing fewer than 500 cells, 20  ng carrier RNA (5 ml of a 4 ng/ml solution) may be added to the lysate prior to homogenization. Ensure that b-mercaptoethanol

484

Y. Purwosunu et al.

(b-ME) or DTT is added to buffer RLT before use. Prepare the carrier RNA as described in the protocols recommended by the manufacturer (see Notes 8 and 9). 2. Vortex the sample for 30  s. No further homogenization is necessary. 3. Add 1 volume of 70% ethanol to the homogenized lysate, and mix well by pipetting. Do not centrifuge. Proceed immediately to Step 4. The volume of lysate may be less than 75 ml due to loss during homogenization. Precipitates may be visible after the addition of ethanol. This does not affect the procedure. 4. Transfer the sample, including any precipitate that may have formed, to an RNeasy MinElute spin column placed in a 2 ml collection tube (supplied). Close the lid gently and centrifuge for 15 s at 8,000 × g. Discard the flowthrough. * Reuse the collection tube in Step 5. 5. Add 350  ml buffer RW1 to the RNeasy MinElute spin column. Close the lid gently and centrifuge for 15 s at 8,000 × g to wash the spin column membrane. Discard the flowthrough. Reuse the collection tube in Step 8. Optional: If an on-column DNase digestion is not desired, add 700 ml buffer RW1instead, centrifuge for 15 s at 8,000 × g, and discard the flowthrough and collection tube. Proceed to Step 9. 6. Add 10 ml DNase I stock solution to 70 ml buffer RDD. Mix by gently inverting the tube. DNase I is particularly sensitive to physical denaturation. Mixing should only be carried out by gently inverting the tube. Do not vortex. 7. Add the DNase I incubation mix (80  ml) directly to the RNeasy MinElute spin column membrane, and place on the benchtop (20–30°C) for 15 min. Be sure to add the DNase I incubation mix directly to the RNeasy MinElute spin column membrane. DNase digestion will be incomplete if any of the mix sticks to the walls or the O-ring of the spin column. 8. Add 350 ml buffer RW1 to the RNeasy MinElute spin column. Close the lid gently and centrifuge for 15  s at 8,000 × g to wash the spin column membrane. Discard the flowthrough and collection tube. 9. Place the RNeasy MinElute spin column in a new 2-ml collection tube (supplied). Add 500 ml buffer RPE to the spin column. Close the lid gently and centrifuge for 15 s at 8,000 × g to wash the spin column membrane. Discard the flowthrough. Reuse the collection tube in Step 10. buffer RPE is supplied as a concentrate. Ensure that ethanol is added to buffer RPE before use.

40  Quantitative RT-PCR Gene Expression Analysis of a Laser Microdissected…

485

10. Add 500  ml of 80% ethanol to the RNeasy MinElute spin column. Close the lid gently and centrifuge for 2  min at 8,000 × g to wash the spin column membrane. Discard the flowthrough and collection tube. 11. Prepare the 80% ethanol with ethanol (96–100%) and the RNase-free water supplied with the kit. After centrifugation, carefully remove the RNeasy MinElute spin column from the collection tube so that the column does not contact the flowthrough. Otherwise, ethanol carryover will occur. 12. Place the RNeasy MinElute spin column in a new 2-ml collection tube (supplied). Open the lid of the spin column and centrifuge at full speed for 5 min. Discard the flowthrough and collection tube. To avoid damage to the lids, place the spin columns into the centrifuge with at least one empty position between columns. Orient the lids such that they point toward a direction opposite to the rotation of the rotor (e.g., if the rotor rotates clockwise, orient the lids counterclockwise). It is important to dry the spin column membrane because residual ethanol may interfere with downstream reactions. Centrifugation with the lids open ensures that no ethanol is carried over during RNA elution. 13. Place the RNeasy MinElute spin column in a new 1.5-ml collection tube (supplied). Add 14  ml RNase-free water directly to the center of the spin column membrane. Close the lid gently and centrifuge for 1 min at full speed to elute the RNA. As little as 10 ml RNase-free water can be used for elution if a higher RNA concentration is required, but the yield will be reduced by approximately 20%. Do not elute with less than 10  ml RNase-free water, as the spin column membrane will not be sufficiently hydrated. The dead volume of the RNeasy MinElute spin column is 2  ml: elution with 14 ml RNase-free water results in a 12 ml eluate. 3.4. Two-Tube Reverse Transcript (Directly Processed) (see Note 9)

1. Thaw the 10× buffer RT, dNTP mix, and RNase-free water at room temperature. Store on ice immediately after thawing. Mix each solution by vortexing, and centrifuge briefly to collect residual liquid from the sides of the tubes. 2. Dilute RNase inhibitor to a final concentration of 10 U/ml in ice-cold 1× buffer RT (dilute an aliquot of the 10× buffer RT accordingly using the supplied RNase-free water). Mix carefully by vortexing for no more than 5 s and centrifuge briefly to collect the residual liquid from the sides of the tube. Commercially available RNase inhibitor is commonly supplied at 40  U/ml. Dilute to make it easier to pipet small amounts when preparing the master mix in Step 1.

486

Y. Purwosunu et al.

3. Prepare a fresh dilution of RNase inhibitor. To minimize the amount of RNase inhibitor and buffer RT used, dilute no more than needed for your current series of reactions. 4. Prepare a fresh master mix on ice Mix thoroughly and carefully by vortexing for no more than 5 s. Centrifuge briefly to collect residual liquid from the walls of the tube, and store on ice. The master mix contains all the components required for first-strand synthesis except template RNA. If setting up more than one reaction, prepare a volume of master mix 10% greater than that required for the total number of reverse-transcription reactions to be performed. 5. If setting up more than one reverse-transcription reaction, distribute the appropriate volume of master mix into individual reaction tubes. Keep the tubes on ice. 6. Add template RNA to the individual tubes containing the master mix. Mix thoroughly and carefully by vortexing for no more than 5  s. Centrifuge briefly to collect residual liquid from the walls of the tube. The protocol is optimized for use with under 50  ng RNA. This amount corresponds to the entire amount of RNA present, including any rRNA, mRNA, viral RNA, and carrier RNA present, and is independent of the primers used or the cDNA analyzed. 7. Incubate for 60 min at 37°C. 8. Add an aliquot of the finished reverse-transcription reaction to the PCR mix. When performing real-time PCR, no more than 1/5 of the final PCR volume should be derived from the finished reverse-transcription reaction. For example, for a 50 ml PCR assay, use £10 ml of the finished reverse-transcription reaction. 3.5. Quantitative PCR (Directly Processed) (see Notes 10 and 11)

1. Thaw the 2× Qiagen QuantiTect RT-PCR Master Mix, template DNA, Primer-Probe solutions (Assay-on-Demand, Applied Biosystems), and the RNase-free water. Mix the individual solutions and place them on ice. QuantiTect Multiplex RT Mix should be taken from −20°C immediately prior to use, kept on ice, and returned to storage at −20°C immediately after use. Store on ice after thawing. Mix each solution by vortexing and centrifuge briefly to collect the residual liquid from the sides of the tubes. 2. Prepare a reaction mix; mix thoroughly and carefully by vortexing for no more than 5 s. Centrifuge briefly to collect residual liquid from the walls of the tube and store on ice. Keep the samples on ice while preparing the reaction mix. The master mix contains all components required except the template DNA. If setting up more than one reaction, prepare a volume of master mix 10% greater than that required for the total number of reverse-transcription reactions to be performed.

40  Quantitative RT-PCR Gene Expression Analysis of a Laser Microdissected…

487

3. Mix the reaction mix thoroughly, and dispense the appropriate volumes into the wells of a PCR plate. Keep tubes or plate on ice. Place the master mix into 96-well plates designated for the standard curve and the blank. 4. Add the template DNA, the standard curve DNA and the blank (H2O) to individual wells containing the master mix. Keep the tubes or plate on ice until the real-time cycler is programmed. 5. Program the real-time cycler. Other settings are set according to the manufacturer’s protocols. Check the real-time cycler’s user manual for correct instrument setup for multiplex analysis (e.g., setting up the detection of multiple dyes from the same well). Activate the detector for each reporter dye used. Depending on the instrument, it may also be necessary to perform a calibration procedure for each of the reporter dyes before they are used for the first time. 6. Place the PCR tubes or plate in the real-time cycler and start the cycling program. 7. Perform the data analysis. Before performing the analysis, select the analysis settings for each probe (i.e., baseline settings and threshold values). Optimized analysis settings are a prerequisite for accurate quantification data. If using the Applied Biosystems 7500, it is necessary to adjust the preset threshold value to a lower value. Use a value of 0.01 as a starting point.

4. Notes 1. The way from the delivery room or clinics to the laboratory for specimen preparation must be considered. Snap-freezing provides the highest quality of RNA for the subsequent analyses, but histological details are inferior to those of fixed samples. If the tissue is adequately fresh and preserved, it is feasible to recover sufficient RNA quantities (100–200 ng). In our experience, immediate processing and fresh samples are important factors for the recovery of sufficient amounts of RNA and subsequent successful quantitative PCR. 2. Do not place the samples directly into the liquid nitrogen, which will break the histologic structure of the samples, thus rendering them unrecognizable during laser microdissection. 3. Consider the orientation of the specimen (placenta) before cuttting into small pieces that are embedded into the OCT. Particular cells of interest in the placenta must be sufficiently represented in the slides.

488

Y. Purwosunu et al.

4. After the OCT-embedded snap-freeze, in our experience, samples cannot be stored for over 6 months. However, processing the samples within 1 week greatly improves RNA recovery. Slides with frozen sections are directly processed with subsequent laser microdissection. When cryostat sectioning, ensure that there is no contamination from other tissues of other blocks. Clean the cryostat, change the blade, and clean from contaminating RNA before cutting the new samples to avoid contamination. 5. To reduce tissue waste, snap-frozen tissue blocks are aligned with the blade. In our experience, once the tissue blocks are thawed, the block cannot be refrozen for later processing, so it is important to obtain enough sections for subsequent processing. Make sure that the tissue block is maintained at −20°C during cryostat sectioning. 6. Tissue samples might contain substances that inhibit the quantitative PCR analyses. OCT is a major factor in qPCR; therefore, ensure that OCT is removed by centrifugation prior to the quantitative PCR analyses. 7. Before beginning microdissection, RNA extraction, and quantitative PCR, follow the general protocol of RNA handling. No more than 30  min should elapse until the microdissected tissues are placed into lysis buffer. 8. There are several methods for RNA extraction or purification; generally, 5–20  pg of RNA are obtained from one cell. Because microdissection samples yield low quantities of RNA, NanoDrop (NanoDrop Technologies, USA) is a good choice to quantify the RNA. 9. Because of the low yield of RNA from microdissection samples, in our experience, it is better to use a two-step quantitative PCR protocol. With one sample, the reverse transcript reaction can provide up to ten quantitative PCR reactions. RNA preamplification is not required. 10. Our experience suggests that single-plexing of quantitative PCR reactions (separate reaction from probe of interest gene and internal control of reference gene) is better than dual- or triple-plexing. We chose to use a smaller reaction (20 ml), which yields reliable and conclusive results. However, larger reactions (50 ml) can be used for the quantitative PCR reactions. 11. Normalization strategies may be performed using any one of the following: (1) microdissection cell counting, (2) total RNA quantification, or (3) the use of endogenous housekeeping genes. In the present study, we used an endogenous housekeeping gene as a normalization control.

40  Quantitative RT-PCR Gene Expression Analysis of a Laser Microdissected…

489

References 1. Khong TY, De Wolf F, Robertson WB, Brosens I. (1986) Inadequate maternal vascular response to placentation in pregnancies complicated by pre-eclampsia and by small-for-gestational age infants. Br J Obstet Gynaecol 93:1049–59. 2. Kupferminc MJ, Fait G, Many A, Gordon D, Eldor A, Lessing JB. (2000) Severe preeclampsia and high frequency of genetic thrombophilic mutations. Obstet Gynecol 96:45–9. 3. Levine RJ, Maynard SE, Qian C, Lim KH, England LJ, Yu KF, et al (2004). Circulating angiogenic factors and the risk of preeclampsia. N Engl J Med 350:672–83. 4. Maynard SE, Min JY, Merchan J, Lim KH, Li J, Mondal S, et  al. (2003) Excess placental soluble fms-like tyrosine kinase 1 (sFlt1) may contribute to endothelial dysfunction, hypertension, and proteinuria in preeclampsia. J Clin Invest 111:649–58. 5. von Dadelszen P, Magee LA. (2002) Could an infectious trigger explain the differential maternal response to the shared placental pathology of preeclampsia and normotensive intrauterine growth restriction? Acta Obstet Gynecol Scand 81:642–8. 6. Murray GI. (2007) An overview of laser microdissection technologies. Acta Histochem 109:171–6. 7. Sekizawa A, Purwosunu Y, Yoshimura S, Nakamura M, Shimizu H, Okai T, et al. (2009) PP13 mRNA expression in trophoblasts from preeclamptic placentas. Reprod Sci 16:408–13.

8. Amemiya S, Sekizawa A, Otsuka J, Tachikawa T, Saito H, Okai T. (2004) Malignant transformation of endometriosis and genetic alterations of K-ras and microsatellite instability. Int J Gynaecol Obstet 86:371–6. 9. Otsuka J, Okuda T, Sekizawa A, Amemiya S, Saito H, Okai T, et al. (2004) K-ras mutation may promote carcinogenesis of endometriosis leading to ovarian clear cell carcinoma. Med Electron Microsc 37:188–92. 10. Sekizawa A, Amemiya S, Otsuka J, Saito H, Farina A, Okai T, et  al. (2004) Malignant transformation of endometriosis: application of laser microdissection for analysis of genetic alterations according to pathological changes. Med Electron Microsc 37:97–100. 11. Yamamoto G, Irie T, Aida T, Nagoshi Y, Tsuchiya R, Tachikawa T. (2006) Correlation of invasion and metastasis of cancer cells, and expression of the RAD21 gene in oral squamous cell carcinoma. Virchows Arch 448:435–41. 12. Zhang X, Wang Y, Yamamoto G, Tachikawa T. (2009) Expression of matrix metalloproteinases MMP-2, MMP-9 and their tissue inhibitors TIMP-1 and TIMP-2 in the epithelium and stroma of salivary gland pleomorphic adenomas. Histopathology 55:250–60. 13. Salafia CM. (1997) Placental pathology of fetal growth restriction. Clin Obstet Gynecol 40:740–9.

wwwwwww

Index A Aorta atherosclerosis.............................................. 417–427 Atherosclerosis................................................... 9, 417–427

B Barley seed.............................................................. 461–474 Barrett’s esophagus..................................165, 175, 181–186 Blood vessel aorta.................................................................. 417–427 atherosclerosis............................................... 9, 417–427 Budd–Chiari syndrome.................................... 405–415 endothelium.......................................279, 386, 405–415 Bone marrow megakaryocyte................................. 429–438 Brain..............................20, 37, 54, 315, 327–329, 331, 332, 345–361, 363–373, 375, 382, 386–390, 393, 444 Breast cancer........................... 9, 18, 96, 107–117, 119–141, 143–154, 203, 217 Budd–Chiari syndrome.......................................... 405–415

C Cartilage................................................................. 449–458 Central nervous system brain..........................................................386–388, 390 development of......................................................... 375 Parkinson’s disease.............................338, 340, 363–373 spinal cord................................................................ 375 Chemotherapy.........................................108, 155, 199, 437 Cirrhosis................................................................. 234, 235 Colon cancer.......................................................... 199, 234 Colorectal cancer................................................ 9, 203–220

D DNA copy number.......................... 223, 226, 229–231, 317 Drug resistance esophageal cancer............................................. 197–201 prostate cancer...............................9, 291–299, 301–304

E Ear endolymphatic sac............................................ 441–447 inner ear............................................................ 319, 441

Endolymphatic sac (ES)......................................... 441–447 Endothelium brain..................................................386–388, 390, 393 liver................................................................... 405–415 neoplasm................................................................... 406 Esophageal cancer.................................................. 197–201 Esophagus Barrett’s esophagus............................165, 175, 181–186 cancer................................................................ 197–201

F 5-Fluorouracil........................................................ 301–304 Formalin fixed breast cancer..................................................... 119–140 prostate cancer...................................291–299, 302, 303

G Gene expression microarray.............................................17–44, 165, 166, 204, 237, 241, 262, 286, 308, 311, 338, 345–347, 349, 350 oligonucleotide microarray........................203–220, 234 PCR.............................................. 18, 20, 27, 31, 32, 37, 38, 43, 47, 49, 78, 79, 82, 83, 86, 108, 166, 235, 236, 284, 304, 333, 351, 356–358, 430–432, 435–437, 458 quantitative PCR........................ 93, 200, 296–298, 361, 363, 366, 370, 372, 373, 437, 456, 458, 488 quantitative RT-PCR................... 10, 20, 21, 36–38, 40, 44, 49, 53, 69, 72, 77–79, 82–83, 241, 258, 259, 296, 297, 308, 329, 337–340, 346, 351, 363, 364, 366, 369, 373, 392–393, 430, 432, 436, 477–488

I Western blot..............................................10, 245–255, 422 Immunohistochemistry (IHC).......................10, 57–64, 73, 78, 88, 91, 93, 108, 199, 235, 237, 285, 327, 329–331, 333–335, 341, 342, 387, 389, 393, 408, 409, 420, 450 Inner ear endolymphatic sac (ES)............................................ 441

Graeme I. Murray (ed.), Laser Capture Microdissection: Methods and Protocols, Methods in Molecular Biology, vol. 755, DOI 10.1007/978-1-61779-163-5, © Springer Science+Business Media, LLC 2011

491

Laser Capture Microdissection 492  Index

  

Intestine small intestine........................................................... 228

K Kidney cancer................................................................ 279–284 proximal tubule..........................................267–276, 282

L Leydig cell...................................................................... 308 Liquid chromatography (LC).....................97, 98, 103–104, 144, 145, 149–151, 156, 158, 160, 161, 316, 322 Liver Budd–Chiari syndrome.................................... 405–415 cirrhosis............................................................ 233–243 hepatocellular carcinoma (HCC)...................... 233–243

M Mass spectrometry proteomics............................. 10, 96, 103, 144, 149–151 Megakaryocyte................................................. 86, 429–438 Microarray cDNA...................17, 203–220, 262–263, 355, 436, 458 expression..........................18, 26, 33, 36, 165, 166, 175, 203–220, 234, 241, 262, 286, 308, 311, 338, 345–347, 349, 350, 380–382, 419, 425, 427 gene...................................18, 26, 33, 36, 165, 166, 175, 203–220, 234, 241, 262, 286, 308, 311, 338, 345–347, 349, 350, 380–382, 419, 425, 427 PCR...............................21, 36, 168, 175, 207, 215, 217, 241, 308, 329, 338–340, 373, 436 MicroRNA (miRNAs) breast cancer..................................................... 119–140 Mitochondria DNA copy number.................... 223, 226, 229–231, 317 DNA deletion....................................224, 226, 315–326 neurogastrointestinal encephalomyopathy........ 223–232 renal toxicity..................................................... 267–276 spiral ganglion.................................................. 315–326 mRNA........................... 5, 6, 10, 53, 68, 70, 72, 85, 93, 116, 120, 121, 190, 194, 201, 204, 215, 219, 235, 239, 262, 298, 302–304, 327, 328, 330, 337, 338, 340, 345, 346, 351, 357, 363–373, 413, 418, 422, 436, 462, 463, 468–470, 472, 474, 483–486 Multiplex ligation-dependent probe amplification (MLPA)...................................................... 107–117

N Neuron.................................... 258, 327–342, 363–373, 418

O Ovarian cancer................................................... 9, 155–163 Ovary

cancer............................................................ 9, 155–163

P Pancreas cancer................................................................ 245–255 Pancreatic cancer.................................................... 245–255 Parkinson’s disease (PD)..........................338, 340, 363–373 PCR quantitative PCR (QPCR).........................93, 123, 131, 167, 170, 174–176, 178, 197–201, 215, 225, 296–298, 357, 361, 363, 366, 370, 372, 373, 456, 458, 486–488 quantitative RT PCR (qRT-PCR).................21, 36–38, 40, 53, 77, 79, 190–194, 199, 217, 226, 229–231, 241, 257–259, 264–265, 302–304, 329, 337–340, 366, 369–371, 373, 385–394, 477–488 Placenta pre-eclampsia.................................................... 477–488 Pre-eclampsia......................................................... 477–488 Prostate cancer....................................................9, 291–299, 301–304 Prostate gland cancer.............................................9, 291–299, 301–304 Proteomics breast cancer..................................................... 143–154 western blot................................................ 10, 245–255 liquid chromatography (LC), 103, 144, 149–150 mass spectrometry.......................... 10, 96, 98, 103, 144, 149–151 matrix-assisted laser desorption/ionization (MALDI)................................................... 149–150 surface enhanced laser desorption ionization-time of flight mass spectrometry (SELDI-TOF MS)..................................... 155–163 Proximal tubule...............................................267–276, 282

Q Quantitative PCR (QPCR)..............................93, 123, 131, 167, 170, 174–176, 178, 197–201, 215, 225, 296–298, 357, 361, 363, 366, 370, 372, 373, 456, 458, 486–488 Quantitative RT PCR (qRT-PCR)......................21, 36–38, 40, 53, 77, 79, 190–194, 199, 217, 226, 229–231, 241, 257–259, 264–265, 302–304, 329, 337–340, 366, 369–371, 373, 385–394, 477–488

R Radiotherapy...................................................197–201, 292 RNA micro RNA (miRNAs)...........................41, 85–94, 119, 125, 131, 132, 134, 136, 138, 140, 293, 338, 437 mRNA....................... 5, 10, 70, 121, 219, 239, 262, 298, 302–304, 330, 337, 338, 462, 463, 468, 474, 486

Laser Capture Microdissection 493 Index     



S

T

SELDI. See Surface enhanced laser desorption ionization-time of flight mass spectrometry Spinal cord..................................................................... 375 Surface enhanced laser desorption ionization-time of flight mass spectrometry (SELDI)......... 155–163

Testis germ cell................................................................... 308 Leydig cell................................................................ 308 Toxicity kidney........................................................267–276, 282