Methods in Enzymology Volume 356 LASER CAPTURE MICROSCOPY AND MICRODISSECTION
METHODS IN ENZYMOLOGY EDITORS-IN-CHIEF
John N. Abelson
Melvin I. Simon
DIVISION OF BIOLOGY CALIFORNIA INSTITUTE OF TECHNOLOGY PASADENA, CALIFORNIA
FOUNDING EDITORS
Sidney P. Colowick and Nathan O. Kaplan
Methods in Enzymology Volume 356
Laser Capture Microscopy and Microdissection EDITED BY
P. Michael Conn OREGON HEALTH AND SCIENCE UNIVERSITY PORTLAND, OREGON, AND OREGON NATIONAL PRIMATE RESEARCH CENTER BEAVERTON, OREGON
Amsterdam Boston London New York Oxford Paris San Diego San Francisco Singapore Sydney Tokyo
∞ This book is printed on acid-free paper. C 2002, Elsevier Science (USA). Copyright
All Rights Reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the Publisher. The appearance of the code at the bottom of the first page of a chapter in this book indicates the Publisher’s consent that copies of the chapter may be made for personal or internal use of specific clients. This consent is given on the condition, however, that the copier pay the stated per copy fee through the Copyright Clearance Center, Inc. (222 Rosewood Drive, Danvers, Massachusetts 01923), for copying beyond that permitted by Sections 107 or 108 of the U.S. Copyright Law. This consent does not extend to other kinds of copying, such as copying for general distribution, for advertising or promotional purposes, for creating new collective works, or for resale. Copy fees for pre-2002 chapters are as shown on the title pages. If no fee code appears on the title page, the copy fee is the same as for current chapters. 0076-6879/2002 $35.00 Explicit permission from Academic Press is not required to reproduce a maximum of two figures or tables from an Academic Press chapter in another scientific or research publication provided that the material has not been credited to another source and that full credit to the Academic Press chapter is given.
Academic Press An imprint of Elsevier Science. 525 B Street, Suite 1900, San Diego, California 92101-4495, USA http://www.academicpress.com
Academic Press 84 Theobalds Road, London WC1X 8RR, UK http://www.academicpress.com International Standard Book Number: 0-12-182259-1 PRINTED IN THE UNITED STATES OF AMERICA 02 03 04 05 06 07 SB 9 8 7 6 5 4
3
2
1
Table of Contents
CONTRIBUTORS TO VOLUME 356 . . . . . . . . . . . . . . . .
ix
PREFACE . . . . . . . . . . . . . . . . . . . . . . .
xiii
. . . . . . . . . . . . . . . . . . .
xv
VOLUMES IN SERIES
Section I. Basic Principles 1. Comparison of Current Equipment
ANDA CORNEA AND ALISON MUNGENAST
2. Laser Capture Microdissection and Its Applica- JAMES L. WITTLIFF AND tions in Genomics and Proteomics MARK G. ERLANDER 3. Going in Vivo with Laser Microdissection
ANETTE MAYER, MONIKA STICH, DIETER BROCKSCH, ¨ KARIN SCHUTZE , AND GEORGIA LAHR
3 12
25
4. Use of Laser Capture Microdissection to Selec- RACHEL A. CRAVEN AND tively Obtain Distinct Populations of Cells for ROSAMONDE E. BANKS Proteomic Analysis
33
5. Optimized Tissue Processing and Staining for LORA E. HUANG, Laser Capture Microdissection and Nucleic VERONICA LUZZI, Acid Retrieval TORSTEN EHRIG, VICTORIA HOLTSCHLAG, AND MARK A. WATSON
49
6. Fluorescence in Situ Hybridization of LCM- DOUGLAS J. DEMETRICK, Isolated Nuclei from Paraffin Sections SABITA K. MURTHY, AND LISA M. DIFRANCESCO
63
7. Immunoblotting of Single Cell Types Isolated LIVIA CASCIOLA-ROSEN AND from Frozen Sections by Laser Microdissection KANNEBOYINA NAGARAJU
70
8. Noncontact Laser Catapulting: A Basic Procedure GABRIELA WESTPHAL, for Functional Genomics and Proteomics RENATE BURGEMEISTER, GABRIELE FRIEDEMANN, AXEL WELLMANN, NICOLAS WERNERT, VOLKER WOLLSCHEID, BERND BECKER, THOMAS VOGT, ¨ RUTH KNUCHEL , WILHELM STOLZ, ¨ AND KARIN SCHUTZE 80 v
vi
TABLE OF CONTENTS
9. Internal Standards for Laser Microdissection
LUDGER FINK AND RAINER MARIA BOHLE
10. Methacarn: A Fixation Tool for Multipurpose MAKOTO SHIBUTANI AND Genetic Analysis from Paraffin-Embedded CHIKAKO UNEYAMA Tissues
99
114
Section II. Specialized Uses 11. Use of Laser Capture Microdissection for Clonal VALE´ RIE PARADIS Analysis AND PIERRE BEDOSSA
129
12. Laser Capture Microdissection in Carcinoma YEN-LI LO AND Analysis CHEN-YANG SHEN
137
´ 13. Laser Capture Microdissection to Assess Devel- CARLOS A. SUAREZ -QUIAN, opment OSCAR M. TIRADO, FRANCINA MUNELL, ´ AND JAUME REVENTOS
145
14. Application of Laser Capture Microdissection to K. K. JAIN Proteomics
157
15. Laser Capture Microdissection of Mouse Intestine: Characterizing mRNA and Protein Expression, and Profiling Intermediary Metabolism in Specified Cell Populations
THADDEUS S. STAPPENBECK, LORA V. HOOPER, JILL K. MANCHESTER, MELISSA H. WONG, AND JEFFREY I. GORDON 168
16. Laser Capture Microdissection in Pathology
FALKO FEND, KATJA SPECHT, MARCUS KREMER, AND LETICIA QUINTANILLA-MART´INEZ 197
17. Use of Laser Capture Microscopy in the Analysis MERAL J. ARIN AND of Mouse Models of Human Diseases DENNIS R. ROOP
208
18. Use of Laser Microdissection in Complex Tissue HOLGER S. WILLENBERG, RHODRI WALTERS, AND STEFAN R. BORNSTEIN
217
19. Assessment of Clonal Relationships in Malignant KOJO S. J. Lymphomas ELENITOBA-JOHNSON
225
20. Comparison of Normal and Tumor Cells by Laser JAUME MORA, Capture Microdissection MUZAFFAR AKRAM, AND WILLIAM L. GERALD
241
21. Analysis of Folliculostellate Cells by Laser RICARDO V. LLOYD, Capture Microdissection and Reverse Trans- LONG JIN, cription–Polymerase Chain Reaction (LCM- KATHARINA H. RUEBEL, RT/PCR) AND JILL M. BAYLISS
249
TABLE OF CONTENTS
vii
Section III. Genetic Applications 22. Analysis of Gene Expression 23. Analysis of Specific Gene Expression
JANETTE K. BURGESS AND BRENT E. MCPARLAND
259
GEORGIA LAHR, ANNA STARZINSKI-POWITZ, AND ANETTE MAYER
271
24. Gene Discovery with Laser Capture Microscopy MAURICIO NEIRA AND EDWIN AZEN
282
25. DNA Fingerprinting from Cells Captured by Laser YONGYUT SIRIVATANAUKSORN, Microdissection VORAPAN SIRIVATANAUKSORN, AND NICHOLAS R. LEMOINE 289 26. Single Cell PCR in Laser Capture Microscopy
SINUHE HAHN, XIAO YAN ZHONG, AND WOLFGANG HOLZGREVE
295
27. Assessment of Genetic Heterogeneity in Tumors DAVE S. B. HOON, Using Laser Capture Microdissection AKIHIDE FUJIMOTO, SHERRY SHU, AND BRET TABACK
302
28. Gene Mutations: Analysis in Proliferative Pro- HITOSHI TAKAYAMA, static Diseases Using Laser Capture Microdis- NORIO NONOMURA, AND section KATSUYUKI AOZASA
309
29. Use of Laser Capture Microdissection-Generated HIROE OHYAMA, Targets for Hybridization of High-Density Oli- MAMATHA MAHADEVAPPA, gonucleotide Arrays HEIKKI LUUKKAA, RANDY TODD, JANET A. WARRINGTON, AND DAVID T. W. WONG
323
˚ SA PERSSON, 30. Single Cell Gene Mutation Analysis Using Laser- A ¨ Assisted Microdissection of Tissue Sections HELENA BACKVALL , FREDRIK PONTE´ N, MATHIAS UHLE´ N, AND JOAKIM LUNDEBERG
334
31. Methylation in Gene Promoters: Assessment after ARTHUR R. BROTHMAN Laser Capture Microdissection AND JIANG CUI
343
AUTHOR INDEX . . . . . . . . . . . . . . . . . . . . .
353
SUBJECT INDEX . . . . . . . . . . . . . . . . . . . . .
375
This Page Intentionally Left Blank
Contributors to Volume 356 Article numbers are in parentheses following the names of contributors. Affiliations listed are current.
ARTHUR R. BROTHMAN (31), Departments of Pediatrics and Human Genetics, University of Utah Health Sciences Center, Salt Lake City, Utah 84132
MUZAFFAR AKRAM (20), Department of Pathology, Memorial Sloan-Kettering Cancer Center, New York, New York 10021 KATSUYUKI AOZASA (28), Department of Pathology, Osaka University Medical School, Suita, Osaka 565-0871, Japan MERAL J. ARIN (17), Department of Dermatology, University of Cologne, 50924 Cologne, Germany EDWIN AZEN (24), Department of Medicine, University of Wisconsin, Madison, Wisconsin 53706 ¨ (30), Department of HELENA BACKVALL Genetics and Pathology, University Hospital, S-751 85 Uppsala, Sweden ROSAMONDE E. BANKS (4), Cancer Research UK Clinical Centre, St. James University Hospital, Leeds LS9 7TF, United Kingdom JILL M. BAYLISS (21), Department of Laboratory Medicine and Pathology, Mayo Clinic, Rochester, Minnesota 55905 BERND BECKER (8), Department of Dermatology, Institute of Pathology, University of Regensburg, 93053 Regensburg, Germany PIERRE BEDOSSA (11), Faculty of Pharmacy, CNRS ESA 8067, 75005 Paris, France RAINER MARIA BOHLE (9), Institute of Pathology, Justus-Liebig-Universit¨at Giessen, 35392 Giessen, Germany STEFAN R. BORNSTEIN (18), Department of Endocrinology, University of D¨usseldorf, D-40225 D¨usseldorf, Germany DIETER BROCKSCH (3), Servicebereich Corporate Communications, Carl Zeiss, D73447 Oberkochen, Germany
RENATE BURGEMEISTER (8), P.A.L.M. Microlaser Technologies AG, 82347 Bernried, Germany JANETTE K. BURGESS (22), Respiratory Research Group, Department of Pharmacology, University of Sydney, Sydney, New South Wales, 2006 Australia LIVIA CASCIOLA-ROSEN (7), Department of Dermatology, Johns Hopkins University School of Medicine, Baltimore, Maryland 21205 ANDA CORNEA (1), Oregon National Primate Research Center, Beaverton, Oregon 97006 RACHEL A. CRAVEN (4), Cancer Research UK Clinical Centre, St. James University Hospital, Leeds LS9 7TF, United Kingdom JIANG CUI (31), Departments of Pediatrics and Human Genetics, University of Utah Health Sciences Center, Salt Lake City, Utah 84132 DOUGLAS J. DEMETRICK (6), Calgary Laboratory Services, Departments of Pathology, Oncology, and Medical Biochemistry, University of Calgary, Calgary, Alberta T2N 1N4, Canada LISA M. DIFRANCESCO (6), Calgary Laboratory Services, Departments of Pathology, Oncology, and Medical Biochemistry, University of Calgary, Calgary, Alberta T2N 1N4, Canada ix
x
CONTRIBUTORS TO VOLUME
TORSTEN EHRIG (5), Department of Dermatology, University of Texas Southwestern Medical Center, Dallas, Texas 75235 KOJO S. J. ELENITOBA-JOHNSON (19), Division of Anatomic Pathology, University of Utah Health Sciences Center, Salt Lake City, Utah 84132 MARK G. ERLANDER (2), Arcturus Applied Genomics, West Carlsbad, California 92008 FALKO FEND (16), Institute of Pathology, Technical University Munich, D-81675 Munich, Germany
356
LORA V. HOOPER (15), Department of Molecular Biology and Pharmacology, Washington University School of Medicine, St. Louis, Missouri 63110 LORA E. HUANG (5), Department of Internal Medicine, University of Iowa, Iowa City, Iowa 52246 K. K. JAIN (14), Jain PharmaBiotech, CH4057 Basel, Switzerland LONG JIN (21), Department of Laboratory Medicine and Pathology, Mayo Clinic, Rochester, Minnesota 55905
LUDGER FINK (9), Institute of Pathology, Justus-Liebig-Universit¨at Giessen, 35392 Giessen, Germany
¨ RUTH KNUCHEL (8), Department of Dermatology, Institute of Pathology, University of Regensburg, 93053 Regensburg, Germany
GABRIELE FRIEDEMANN (8), P.A.L.M. Microlaser Technologies AG, 82347 Bernried, Germany
MARCUS KREMER (16), Institute of Pathology, Technical University Munich, D-81675 Munich, Germany
AKIHIDE FUJIMOTO (27), Department of Molecular Oncology, John Wayne Cancer Institute, Santa Monica, California 90404
GEORGIA LAHR (3, 23), Laser Laboratory and Department of Molecular Biology, Staedtisches Krankenhaus M¨unchenHarlaching, D-81545 Munich, Germany
WILLIAM L. GERALD (20), Department of Pathology, Memorial Sloan-Kettering Cancer Center, New York, New York 10021 JEFFREY I. GORDON (15), Department of Molecular Biology and Pharmacology, Washington University School of Medicine, St. Louis, Missouri 63110 SINUHE HAHN (26), Laboratory for Prenatal Medicine, Department of Obstetrics and Gynecology, University of Basel, CH-4031 Basel, Switzerland VICTORIA HOLTSCHLAG (5), The Siteman Cancer Center at Barnes-Jewish Hospital and Washington University School of Medicine, St. Louis, Missouri 63110 WOLFGANG HOLZGREVE (26), Laboratory for Prenatal Medicine, Department of Obstetrics and Gynecology, University of Basel, CH-4031 Basel, Switzerland DAVE S. B. HOON (27), Department of Molecular Oncology, John Wayne Cancer Institute, Santa Monica, California 90404
NICHOLAS R. LEMOINE (25), Cancer Research UK Molecular Oncology Unit, Department of Cancer Medicine, Imperial College of Science, Technology, and Medicine, London W12 ONN, United Kingdom RICARDO V. LLOYD (21), Department of Laboratory Medicine and Pathology, Mayo Clinic, Rochester, Minnesota 55905 YEN-LI LO (12), Institute of Biomedical Sciences, Academia Sinica, Taipei 11529, Taiwan JOAKIM LUNDEBERG (30), Department of Biotechnology, Royal Institute of Technology (KTH), SCFAB, S-106 91 Stockholm, Sweden HEIKKI LUUKKAA (29), Division of Oral Pathology, Department of Oral Medicine and Diagnostic Sciences, Harvard School of Dental Medicine, Boston, Massachusetts 02115
CONTRIBUTORS TO VOLUME
VERONICA LUZZI (5), Department of Pathology and Immunology, Washington University School of Medicine, St. Louis, Missouri 63110 MAMATHA MAHADEVAPPA (29), Affymetrix Inc., Santa Clara, California 95051 JILL K. MANCHESTER (15), Department of Molecular Biology and Pharmacology, Washington University School of Medicine, St. Louis, Missouri 63110 ANETTE MAYER (3, 23), Humangenetik f¨ur Biologen, Universit¨at Frankfurt, D60054 Frankfurt/Main, Germany BRENT E. MCPARLAND (22), Department of Pathology, University of Sydney, Sydney, New South Wales, 2006 Australia JAUME MORA (20), Department of Hematology and Oncology, Hospital Sant Joan de Deu de Barcelona, Barcelona, Spain FRANCINA MUNELL (13), Unitat de Recerca Biom`edica, Hospital MaternoInfantil, Vall d’Hebron Hospital, 08035 Barcelona, Spain ALISON MUNGENAST (1), Oregon Health and Science University, Portland, Oregon 97201, and Oregon National Primate Research Center, Beaverton, Oregon 97006 SABITA K. MURTHY (6), Departments of Pathology, Oncology, and Medical Biochemistry, University of Calgary, Calgary, Alberta T2N 1N4, Canada KANNEBOYINA NAGARAJU (7), Division of Rheumatology, Department of Medicine, Johns Hopkins University School of Medicine, Baltimore, Maryland 21205 MAURICIO NEIRA (24), Department of Zoology, University of British Columbia, Vancouver, British Columbia V6T 1Z4, Canada NORIO NONOMURA (28), Department of Urology, Osaka University Medical School, Suita, Osaka 565-0871, Japan HIROE OHYAMA (29), Division of Oral Pathology, Department of Oral Medicine and Diagnostic Sciences, Harvard School of Dental Medicine, Boston, Massachusetts 02115
356
xi
VALE´ RIE PARADIS (11), Department of Pathology, Beaujon Hospital, 92110 Clichy, France ˚ SA PERSSON (30), Department of BiotechA nology, Royal Institute of Technology (KTH), SCFAB, S-106 91 Stockholm, Sweden FREDRIK PONTE´ N (30), Department of Genetics and Pathology, University Hospital, S-751 85 Uppsala, Sweden LETICIA QUINTANILLA-MART´INEZ (16), Department of Pathology, GSF Research Center for Environment and Health, D-85758 Oberschleissheim, Germany ´ (13), Unitat de ReJAUME REVENTOS cerca Biom`edica, Hospital MaternoInfantil, Vall d’Hebron Hospital, 08035 Barcelona, Spain DENNIS R. ROOP (17), Departments of Molecular and Cellular Biology and Dermatology, Baylor College of Medicine, Houston, Texas 77030 KATHARINA H. RUEBEL (21), Department of Laboratory Medicine and Pathology, Mayo Clinic, Rochester, Minnesota 55905 ¨ KARIN SCHUTZE (3, 8), P.A.L.M. Microlaser Technologies AG, 82347 Bernried, Germany CHEN-YANG SHEN (12), Institute of Biomedical Sciences, Academia Sinica, Taipei 11529, Taiwan MAKOTO SHIBUTANI (10), Division of Pathology, National Institute of Health Sciences, Setagaya-ku, Tokyo 158-8501, Japan SHERRY SHU (27), Department of Molecular Oncology, John Wayne Cancer Institute, Santa Monica, California 90404 VORAPAN SIRIVATANAUKSORN (25), Faculty of Medicine, Mahidol University, Bangkok 10700, Thailand YONGYUT SIRIVATANAUKSORN (25), Faculty of Medicine, Mahidol University, Bangkok 10700, Thailand
xii
CONTRIBUTORS TO VOLUME
KATJA SPECHT (16), Institute of Pathology, Technical University Munich, D-81675 Munich, Germany THADDEUS S. STAPPENBECK (15), Department of Molecular Biology and Pharmacology, Washington University School of Medicine, St. Louis, Missouri 63110 ANNA STARZINSKI-POWITZ (23), Humangenetik f¨ur Biologen, Universit¨at Frankfurt, D-60054 Frankfurt/Main, Germany MONIKA STICH (3), Laser Laboratory and Department of Molecular Biology, Staedtisches Krankenhaus M¨unchenHarlaching, D-81545 Munich, Germany WILHELM STOLZ (8), Department of Dermatology, Institute of Pathology, University of Regensburg, 93053 Regensburg, Germany ´ CARLOS A. SUAREZ -QUIAN (13), Department of Cell Biology, Georgetown University Medical School, Washington, D.C. 20007 BRET TABACK (27), Department of Molecular Oncology, John Wayne Cancer Institute, Santa Monica, California 90404 HITOSHI TAKAYAMA (28), Department of Pathology, Osaka University Medical School, Suita, Osaka 565-0871, Japan OSCAR M. TIRADO (13), Unitat de Recerca Biom`edica, Hospital MaternoInfantil, Vall d’Hebron Hospital, 08035 Barcelona, Spain RANDY TODD (29), Division of Oral Pathology, Department of Oral Medicine and Diagnostic Sciences, Harvard School of Dental Medicine, Boston, Massachusetts 02115 MATHIAS UHLE´ N (30), Department of Biotechnology, Royal Institute of Technology (KTH), SCAFB, S-106 91 Stockholm, Sweden CHIKAKO UNEYAMA (10), Division of Pathology, National Institute of Health Sciences, Setagaya-ku, Tokyo 158-8501, Japan
356
THOMAS VOGT (8), Department of Dermatology, Institute of Pathology, University of Regensburg, 93053 Regensburg, Germany RHODRI WALTERS (18), Department of Endocrinology, University of D¨usseldorf, D40225 D¨usseldorf, Germany JANET A. WARRINGTON (29), Affymetrix Inc., Santa Clara, California 95051 MARK A. WATSON (5), Department of Pathology and Immunology, Washington University School of Medicine, St. Louis, Missouri 63110 AXEL WELLMANN (8), Institute of Pathology, University of Bonn, 53011 Bonn, Germany NICOLAS WERNERT (8), Institute of Pathology, University of Bonn, 53011 Bonn, Germany GABRIELA WESTPHAL (8), P.A.L.M. Microlaser Technologies AG, 82347 Bernried, Germany HOLGER S. WILLENBERG (18), Department of Endocrinology, University of D¨usseldorf, D-40225 D¨usseldorf, Germany JAMES L. WITTLIFF (2), Hormone Receptor Laboratory, James Graham Brown Cancer Center, University of Louisville, Louisville, Kentucky 40202 VOLKER WOLLSCHEID (8), Ciphergen Biosystems Ltd., Surrey Technology Centre, Guildford, Surrey GU2 7YG, United Kingdom DAVID T. W. WONG (29), University of California School of Dentistry, Dental Research Institute, Los Angeles, California 90095 MELISSA H. WONG (15), Department of Dermatology, Cell and Development Biology, Oregon Health and Science University, Portland, Oregon 97201 XIAO YAN ZHONG (26), Laboratory for Prenatal Medicine, Department of Obstetrics and Gynecology, University of Basel, CH-4031 Basel, Switzerland
Preface
Five years ago few people had heard of “laser microdissection” or “laser capture microscopy.” Now most major institutions have it as a core facility. This volume documents many diverse uses for this technique in disciplines that broadly span biology. The methods presented include shortcuts and conveniences not included in the sources from which they were taken. To the degree possible, we have included information needed to select equipment, prepare samples, and analyze data. The techniques are described in a context that allows comparisons to other related methodologies. The authors were encouraged to do this in the belief that such comparisons are valuable to readers who must adapt extant procedures to new systems. Also, so far as possible, methodologies are presented in a manner that stresses their general applicability and potential limitations. Although for various reasons some topics are not covered, the volume provides a substantial and current overview of the extant methodology in the field and a view of its rapid development. Particular thanks go to the authors for their attention to meeting deadlines and for maintaining high standards of quality, to the series editors for their encouragement, and to the staff of Academic Press for their help and timely publication of this volume. P. MICHAEL CONN
xiii
This Page Intentionally Left Blank
METHODS IN ENZYMOLOGY VOLUME I. Preparation and Assay of Enzymes Edited by SIDNEY P. COLOWICK AND NATHAN O. KAPLAN VOLUME II. Preparation and Assay of Enzymes Edited by SIDNEY P. COLOWICK AND NATHAN O. KAPLAN VOLUME III. Preparation and Assay of Substrates Edited by SIDNEY P. COLOWICK AND NATHAN O. KAPLAN VOLUME IV. Special Techniques for the Enzymologist Edited by SIDNEY P. COLOWICK AND NATHAN O. KAPLAN VOLUME V. Preparation and Assay of Enzymes Edited by SIDNEY P. COLOWICK AND NATHAN O. KAPLAN VOLUME VI. Preparation and Assay of Enzymes (Continued ) Preparation and Assay of Substrates Special Techniques Edited by SIDNEY P. COLOWICK AND NATHAN O. KAPLAN VOLUME VII. Cumulative Subject Index Edited by SIDNEY P. COLOWICK AND NATHAN O. KAPLAN VOLUME VIII. Complex Carbohydrates Edited by ELIZABETH F. NEUFELD AND VICTOR GINSBURG VOLUME IX. Carbohydrate Metabolism Edited by WILLIS A. WOOD VOLUME X. Oxidation and Phosphorylation Edited by RONALD W. ESTABROOK AND MAYNARD E. PULLMAN VOLUME XI. Enzyme Structure Edited by C. H. W. HIRS VOLUME XII. Nucleic Acids (Parts A and B) Edited by LAWRENCE GROSSMAN AND KIVIE MOLDAVE VOLUME XIII. Citric Acid Cycle Edited by J. M. LOWENSTEIN VOLUME XIV. Lipids Edited by J. M. LOWENSTEIN VOLUME XV. Steroids and Terpenoids Edited by RAYMOND B. CLAYTON VOLUME XVI. Fast Reactions Edited by KENNETH KUSTIN xv
xvi
METHODS IN ENZYMOLOZY
VOLUME XVII. Metabolism of Amino Acids and Amines (Parts A and B) Edited by HERBERT TABOR AND CELIA WHITE TABOR VOLUME XVIII. Vitamins and Coenzymes (Parts A, B, and C) Edited by DONALD B. MCCORMICK AND LEMUEL D. WRIGHT VOLUME XIX. Proteolytic Enzymes Edited by GERTRUDE E. PERLMANN AND LASZLO LORAND VOLUME XX. Nucleic Acids and Protein Synthesis (Part C) Edited by KIVIE MOLDAVE AND LAWRENCE GROSSMAN VOLUME XXI. Nucleic Acids (Part D) Edited by LAWRENCE GROSSMAN AND KIVIE MOLDAVE VOLUME XXII. Enzyme Purification and Related Techniques Edited by WILLIAM B. JAKOBY VOLUME XXIII. Photosynthesis (Part A) Edited by ANTHONY SAN PIETRO VOLUME XXIV. Photosynthesis and Nitrogen Fixation (Part B) Edited by ANTHONY SAN PIETRO VOLUME XXV. Enzyme Structure (Part B) Edited by C. H. W. HIRS AND SERGE N. TIMASHEFF VOLUME XXVI. Enzyme Structure (Part C) Edited by C. H. W. HIRS AND SERGE N. TIMASHEFF VOLUME XXVII. Enzyme Structure (Part D) Edited by C. H. W. HIRS AND SERGE N. TIMASHEFF VOLUME XXVIII. Complex Carbohydrates (Part B) Edited by VICTOR GINSBURG VOLUME XXIX. Nucleic Acids and Protein Synthesis (Part E) Edited by LAWRENCE GROSSMAN AND KIVIE MOLDAVE VOLUME XXX. Nucleic Acids and Protein Synthesis (Part F) Edited by KIVIE MOLDAVE AND LAWRENCE GROSSMAN VOLUME XXXI. Biomembranes (Part A) Edited by SIDNEY FLEISCHER AND LESTER PACKER VOLUME XXXII. Biomembranes (Part B) Edited by SIDNEY FLEISCHER AND LESTER PACKER VOLUME XXXIII. Cumulative Subject Index Volumes I-XXX Edited by MARTHA G. DENNIS AND EDWARD A. DENNIS VOLUME XXXIV. Affinity Techniques (Enzyme Purification: Part B) Edited by WILLIAM B. JAKOBY AND MEIR WILCHEK VOLUME XXXV. Lipids (Part B) Edited by JOHN M. LOWENSTEIN
METHODS IN ENZYMOLOZY
xvii
VOLUME XXXVI. Hormone Action (Part A: Steroid Hormones) Edited by BERT W. O’MALLEY AND JOEL G. HARDMAN VOLUME XXXVII. Hormone Action (Part B: Peptide Hormones) Edited by BERT W. O’MALLEY AND JOEL G. HARDMAN VOLUME XXXVIII. Hormone Action (Part C: Cyclic Nucleotides) Edited by JOEL G. HARDMAN AND BERT W. O’MALLEY VOLUME XXXIX. Hormone Action (Part D: Isolated Cells, Tissues, and Organ Systems) Edited by JOEL G. HARDMAN AND BERT W. O’MALLEY VOLUME XL. Hormone Action (Part E: Nuclear Structure and Function) Edited by BERT W. O’MALLEY AND JOEL G. HARDMAN VOLUME XLI. Carbohydrate Metabolism (Part B) Edited by W. A. WOOD VOLUME XLII. Carbohydrate Metabolism (Part C) Edited by W. A. WOOD VOLUME XLIII. Antibiotics Edited by JOHN H. HASH VOLUME XLIV. Immobilized Enzymes Edited by KLAUS MOSBACH VOLUME XLV. Proteolytic Enzymes (Part B) Edited by LASZLO LORAND VOLUME XLVI. Affinity Labeling Edited by WILLIAM B. JAKOBY AND MEIR WILCHEK VOLUME XLVII. Enzyme Structure (Part E) Edited by C. H. W. HIRS AND SERGE N. TIMASHEFF VOLUME XLVIII. Enzyme Structure (Part F) Edited by C. H. W. HIRS AND SERGE N. TIMASHEFF VOLUME XLIX. Enzyme Structure (Part G) Edited by C. H. W. HIRS AND SERGE N. TIMASHEFF VOLUME L. Complex Carbohydrates (Part C) Edited by VICTOR GINSBURG VOLUME LI. Purine and Pyrimidine Nucleotide Metabolism Edited by PATRICIA A. HOFFEE AND MARY ELLEN JONES VOLUME LII. Biomembranes (Part C: Biological Oxidations) Edited by SIDNEY FLEISCHER AND LESTER PACKER VOLUME LIII. Biomembranes (Part D: Biological Oxidations) Edited by SIDNEY FLEISCHER AND LESTER PACKER VOLUME LIV. Biomembranes (Part E: Biological Oxidations) Edited by SIDNEY FLEISCHER AND LESTER PACKER
xviii
METHODS IN ENZYMOLOZY
VOLUME LV. Biomembranes (Part F: Bioenergetics) Edited by SIDNEY FLEISCHER AND LESTER PACKER VOLUME LVI. Biomembranes (Part G: Bioenergetics) Edited by SIDNEY FLEISCHER AND LESTER PACKER VOLUME LVII. Bioluminescence and Chemiluminescence Edited by MARLENE A. DELUCA VOLUME LVIII. Cell Culture Edited by WILLIAM B. JAKOBY AND IRA PASTAN VOLUME LIX. Nucleic Acids and Protein Synthesis (Part G) Edited by KIVIE MOLDAVE AND LAWRENCE GROSSMAN VOLUME LX. Nucleic Acids and Protein Synthesis (Part H) Edited by KIVIE MOLDAVE AND LAWRENCE GROSSMAN VOLUME 61. Enzyme Structure (Part H) Edited by C. H. W. HIRS AND SERGE N. TIMASHEFF VOLUME 62. Vitamins and Coenzymes (Part D) Edited by DONALD B. MCCORMICK AND LEMUEL D. WRIGHT VOLUME 63. Enzyme Kinetics and Mechanism (Part A: Initial Rate and Inhibitor Methods) Edited by DANIEL L. PURICH VOLUME 64. Enzyme Kinetics and Mechanism (Part B: Isotopic Probes and Complex Enzyme Systems) Edited by DANIEL L. PURICH VOLUME 65. Nucleic Acids (Part I) Edited by LAWRENCE GROSSMAN AND KIVIE MOLDAVE VOLUME 66. Vitamins and Coenzymes (Part E) Edited by DONALD B. MCCORMICK AND LEMUEL D. WRIGHT VOLUME 67. Vitamins and Coenzymes (Part F) Edited by DONALD B. MCCORMICK AND LEMUEL D. WRIGHT VOLUME 68. Recombinant DNA Edited by RAY WU VOLUME 69. Photosynthesis and Nitrogen Fixation (Part C) Edited by ANTHONY SAN PIETRO VOLUME 70. Immunochemical Techniques (Part A) Edited by HELEN VAN VUNAKIS AND JOHN J. LANGONE VOLUME 71. Lipids (Part C) Edited by JOHN M. LOWENSTEIN VOLUME 72. Lipids (Part D) Edited by JOHN M. LOWENSTEIN
METHODS IN ENZYMOLOZY
xix
VOLUME 73. Immunochemical Techniques (Part B) Edited by JOHN J. LANGONE AND HELEN VAN VUNAKIS VOLUME 74. Immunochemical Techniques (Part C) Edited by JOHN J. LANGONE AND HELEN VAN VUNAKIS VOLUME 75. Cumulative Subject Index Volumes XXXI, XXXII, XXXIV–LX Edited by EDWARD A. DENNIS AND MARTHA G. DENNIS VOLUME 76. Hemoglobins Edited by ERALDO ANTONINI, LUIGI ROSSI-BERNARDI, AND EMILIA CHIANCONE VOLUME 77. Detoxication and Drug Metabolism Edited by WILLIAM B. JAKOBY VOLUME 78. Interferons (Part A) Edited by SIDNEY PESTKA VOLUME 79. Interferons (Part B) Edited by SIDNEY PESTKA VOLUME 80. Proteolytic Enzymes (Part C) Edited by LASZLO LORAND VOLUME 81. Biomembranes (Part H: Visual Pigments and Purple Membranes, I) Edited by LESTER PACKER VOLUME 82. Structural and Contractile Proteins (Part A: Extracellular Matrix) Edited by LEON W. CUNNINGHAM AND DIXIE W. FREDERIKSEN VOLUME 83. Complex Carbohydrates (Part D) Edited by VICTOR GINSBURG VOLUME 84. Immunochemical Techniques (Part D: Selected Immunoassays) Edited by JOHN J. LANGONE AND HELEN VAN VUNAKIS VOLUME 85. Structural and Contractile Proteins (Part B: The Contractile Apparatus and the Cytoskeleton) Edited by DIXIE W. FREDERIKSEN AND LEON W. CUNNINGHAM VOLUME 86. Prostaglandins and Arachidonate Metabolites Edited by WILLIAM E. M. LANDS AND WILLIAM L. SMITH VOLUME 87. Enzyme Kinetics and Mechanism (Part C: Intermediates, Stereochemistry, and Rate Studies) Edited by DANIEL L. PURICH VOLUME 88. Biomembranes (Part I: Visual Pigments and Purple Membranes, II) Edited by LESTER PACKER VOLUME 89. Carbohydrate Metabolism (Part D) Edited by WILLIS A. WOOD VOLUME 90. Carbohydrate Metabolism (Part E) Edited by WILLIS A. WOOD
xx
METHODS IN ENZYMOLOZY
VOLUME 91. Enzyme Structure (Part I) Edited by C. H. W. HIRS AND SERGE N. TIMASHEFF VOLUME 92. Immunochemical Techniques (Part E: Monoclonal Antibodies and General Immunoassay Methods) Edited by JOHN J. LANGONE AND HELEN VAN VUNAKIS VOLUME 93. Immunochemical Techniques (Part F: Conventional Antibodies, Fc Receptors, and Cytotoxicity) Edited by JOHN J. LANGONE AND HELEN VAN VUNAKIS VOLUME 94. Polyamines Edited by HERBERT TABOR AND CELIA WHITE TABOR VOLUME 95. Cumulative Subject Index Volumes 61–74, 76–80 Edited by EDWARD A. DENNIS AND MARTHA G. DENNIS VOLUME 96. Biomembranes [Part J: Membrane Biogenesis: Assembly and Targeting (General Methods; Eukaryotes)] Edited by SIDNEY FLEISCHER AND BECCA FLEISCHER VOLUME 97. Biomembranes [Part K: Membrane Biogenesis: Assembly and Targeting (Prokaryotes, Mitochondria, and Chloroplasts)] Edited by SIDNEY FLEISCHER AND BECCA FLEISCHER VOLUME 98. Biomembranes (Part L: Membrane Biogenesis: Processing and Recycling) Edited by SIDNEY FLEISCHER AND BECCA FLEISCHER VOLUME 99. Hormone Action (Part F: Protein Kinases) Edited by JACKIE D. CORBIN AND JOEL G. HARDMAN VOLUME 100. Recombinant DNA (Part B) Edited by RAY WU, LAWRENCE GROSSMAN, AND KIVIE MOLDAVE VOLUME 101. Recombinant DNA (Part C) Edited by RAY WU, LAWRENCE GROSSMAN, AND KIVIE MOLDAVE VOLUME 102. Hormone Action (Part G: Calmodulin and Calcium-Binding Proteins) Edited by ANTHONY R. MEANS AND BERT W. O’MALLEY VOLUME 103. Hormone Action (Part H: Neuroendocrine Peptides) Edited by P. MICHAEL CONN VOLUME 104. Enzyme Purification and Related Techniques (Part C) Edited by WILLIAM B. JAKOBY VOLUME 105. Oxygen Radicals in Biological Systems Edited by LESTER PACKER VOLUME 106. Posttranslational Modifications (Part A) Edited by FINN WOLD AND KIVIE MOLDAVE VOLUME 107. Posttranslational Modifications (Part B) Edited by FINN WOLD AND KIVIE MOLDAVE
METHODS IN ENZYMOLOZY
xxi
VOLUME 108. Immunochemical Techniques (Part G: Separation and Characterization of Lymphoid Cells) Edited by GIOVANNI DI SABATO, JOHN J. LANGONE, AND HELEN VAN VUNAKIS VOLUME 109. Hormone Action (Part I: Peptide Hormones) Edited by LUTZ BIRNBAUMER AND BERT W. O’MALLEY VOLUME 110. Steroids and Isoprenoids (Part A) Edited by JOHN H. LAW AND HANS C. RILLING VOLUME 111. Steroids and Isoprenoids (Part B) Edited by JOHN H. LAW AND HANS C. RILLING VOLUME 112. Drug and Enzyme Targeting (Part A) Edited by KENNETH J. WIDDER AND RALPH GREEN VOLUME 113. Glutamate, Glutamine, Glutathione, and Related Compounds Edited by ALTON MEISTER VOLUME 114. Diffraction Methods for Biological Macromolecules (Part A) Edited by HAROLD W. WYCKOFF, C. H. W. HIRS, AND SERGE N. TIMASHEFF VOLUME 115. Diffraction Methods for Biological Macromolecules (Part B) Edited by HAROLD W. WYCKOFF, C. H. W. HIRS, AND SERGE N. TIMASHEFF VOLUME 116. Immunochemical Techniques (Part H: Effectors and Mediators of Lymphoid Cell Functions) Edited by GIOVANNI DI SABATO, JOHN J. LANGONE, AND HELEN VAN VUNAKIS VOLUME 117. Enzyme Structure (Part J) Edited by C. H. W. HIRS AND SERGE N. TIMASHEFF VOLUME 118. Plant Molecular Biology Edited by ARTHUR WEISSBACH AND HERBERT WEISSBACH VOLUME 119. Interferons (Part C) Edited by SIDNEY PESTKA VOLUME 120. Cumulative Subject Index Volumes 81–94, 96–101 VOLUME 121. Immunochemical Techniques (Part I: Hybridoma Technology and Monoclonal Antibodies) Edited by JOHN J. LANGONE AND HELEN VAN VUNAKIS VOLUME 122. Vitamins and Coenzymes (Part G) Edited by FRANK CHYTIL AND DONALD B. MCCORMICK VOLUME 123. Vitamins and Coenzymes (Part H) Edited by FRANK CHYTIL AND DONALD B. MCCORMICK VOLUME 124. Hormone Action (Part J: Neuroendocrine Peptides) Edited by P. MICHAEL CONN VOLUME 125. Biomembranes (Part M: Transport in Bacteria, Mitochondria, and Chloroplasts: General Approaches and Transport Systems) Edited by SIDNEY FLEISCHER AND BECCA FLEISCHER
xxii
METHODS IN ENZYMOLOZY
VOLUME 126. Biomembranes (Part N: Transport in Bacteria, Mitochondria, and Chloroplasts: Protonmotive Force) Edited by SIDNEY FLEISCHER AND BECCA FLEISCHER VOLUME 127. Biomembranes (Part O: Protons and Water: Structure and Translocation) Edited by LESTER PACKER VOLUME 128. Plasma Lipoproteins (Part A: Preparation, Structure, and Molecular Biology) Edited by JERE P. SEGREST AND JOHN J. ALBERS VOLUME 129. Plasma Lipoproteins (Part B: Characterization, Cell Biology, and Metabolism) Edited by JOHN J. ALBERS AND JERE P. SEGREST VOLUME 130. Enzyme Structure (Part K) Edited by C. H. W. HIRS AND SERGE N. TIMASHEFF VOLUME 131. Enzyme Structure (Part L) Edited by C. H. W. HIRS AND SERGE N. TIMASHEFF VOLUME 132. Immunochemical Techniques (Part J: Phagocytosis and CellMediated Cytotoxicity) Edited by GIOVANNI DI SABATO AND JOHANNES EVERSE VOLUME 133. Bioluminescence and Chemiluminescence (Part B) Edited by MARLENE DELUCA AND WILLIAM D. MCELROY VOLUME 134. Structural and Contractile Proteins (Part C: The Contractile Apparatus and the Cytoskeleton) Edited by RICHARD B. VALLEE VOLUME 135. Immobilized Enzymes and Cells (Part B) Edited by KLAUS MOSBACH VOLUME 136. Immobilized Enzymes and Cells (Part C) Edited by KLAUS MOSBACH VOLUME 137. Immobilized Enzymes and Cells (Part D) Edited by KLAUS MOSBACH VOLUME 138. Complex Carbohydrates (Part E) Edited by VICTOR GINSBURG VOLUME 139. Cellular Regulators (Part A: Calcium- and Calmodulin-Binding Proteins) Edited by ANTHONY R. MEANS AND P. MICHAEL CONN VOLUME 140. Cumulative Subject Index Volumes 102–119, 121–134 VOLUME 141. Cellular Regulators (Part B: Calcium and Lipids) Edited by P. MICHAEL CONN AND ANTHONY R. MEANS VOLUME 142. Metabolism of Aromatic Amino Acids and Amines Edited by SEYMOUR KAUFMAN
METHODS IN ENZYMOLOZY
xxiii
VOLUME 143. Sulfur and Sulfur Amino Acids Edited by WILLIAM B. JAKOBY AND OWEN GRIFFITH VOLUME 144. Structural and Contractile Proteins (Part D: Extracellular Matrix) Edited by LEON W. CUNNINGHAM VOLUME 145. Structural and Contractile Proteins (Part E: Extracellular Matrix) Edited by LEON W. CUNNINGHAM VOLUME 146. Peptide Growth Factors (Part A) Edited by DAVID BARNES AND DAVID A. SIRBASKU VOLUME 147. Peptide Growth Factors (Part B) Edited by DAVID BARNES AND DAVID A. SIRBASKU VOLUME 148. Plant Cell Membranes Edited by LESTER PACKER AND ROLAND DOUCE VOLUME 149. Drug and Enzyme Targeting (Part B) Edited by RALPH GREEN AND KENNETH J. WIDDER VOLUME 150. Immunochemical Techniques (Part K: In Vitro Models of B and T Cell Functions and Lymphoid Cell Receptors) Edited by GIOVANNI DI SABATO VOLUME 151. Molecular Genetics of Mammalian Cells Edited by MICHAEL M. GOTTESMAN VOLUME 152. Guide to Molecular Cloning Techniques Edited by SHELBY L. BERGER AND ALAN R. KIMMEL VOLUME 153. Recombinant DNA (Part D) Edited by RAY WU AND LAWRENCE GROSSMAN VOLUME 154. Recombinant DNA (Part E) Edited by RAY WU AND LAWRENCE GROSSMAN VOLUME 155. Recombinant DNA (Part F) Edited by RAY WU VOLUME 156. Biomembranes (Part P: ATP-Driven Pumps and Related Transport: The Na, K-Pump) Edited by SIDNEY FLEISCHER AND BECCA FLEISCHER VOLUME 157. Biomembranes (Part Q: ATP-Driven Pumps and Related Transport: Calcium, Proton, and Potassium Pumps) Edited by SIDNEY FLEISCHER AND BECCA FLEISCHER VOLUME 158. Metalloproteins (Part A) Edited by JAMES F. RIORDAN AND BERT L. VALLEE VOLUME 159. Initiation and Termination of Cyclic Nucleotide Action Edited by JACKIE D. CORBIN AND ROGER A. JOHNSON VOLUME 160. Biomass (Part A: Cellulose and Hemicellulose) Edited by WILLIS A. WOOD AND SCOTT T. KELLOGG
xxiv
METHODS IN ENZYMOLOZY
VOLUME 161. Biomass (Part B: Lignin, Pectin, and Chitin) Edited by WILLIS A. WOOD AND SCOTT T. KELLOGG VOLUME 162. Immunochemical Techniques (Part L: Chemotaxis and Inflammation) Edited by GIOVANNI DI SABATO VOLUME 163. Immunochemical Techniques (Part M: Chemotaxis and Inflammation) Edited by GIOVANNI DI SABATO VOLUME 164. Ribosomes Edited by HARRY F. NOLLER, JR., AND KIVIE MOLDAVE VOLUME 165. Microbial Toxins: Tools for Enzymology Edited by SIDNEY HARSHMAN VOLUME 166. Branched-Chain Amino Acids Edited by ROBERT HARRIS AND JOHN R. SOKATCH VOLUME 167. Cyanobacteria Edited by LESTER PACKER AND ALEXANDER N. GLAZER VOLUME 168. Hormone Action (Part K: Neuroendocrine Peptides) Edited by P. MICHAEL CONN VOLUME 169. Platelets: Receptors, Adhesion, Secretion (Part A) Edited by JACEK HAWIGER VOLUME 170. Nucleosomes Edited by PAUL M. WASSARMAN AND ROGER D. KORNBERG VOLUME 171. Biomembranes (Part R: Transport Theory: Cells and Model Membranes) Edited by SIDNEY FLEISCHER AND BECCA FLEISCHER VOLUME 172. Biomembranes (Part S: Transport: Membrane Isolation and Characterization) Edited by SIDNEY FLEISCHER AND BECCA FLEISCHER VOLUME 173. Biomembranes [Part T: Cellular and Subcellular Transport: Eukaryotic (Nonepithelial) Cells] Edited by SIDNEY FLEISCHER AND BECCA FLEISCHER VOLUME 174. Biomembranes [Part U: Cellular and Subcellular Transport: Eukaryotic (Nonepithelial) Cells] Edited by SIDNEY FLEISCHER AND BECCA FLEISCHER VOLUME 175. Cumulative Subject Index Volumes 135–139, 141–167 VOLUME 176. Nuclear Magnetic Resonance (Part A: Spectral Techniques and Dynamics) Edited by NORMAN J. OPPENHEIMER AND THOMAS L. JAMES VOLUME 177. Nuclear Magnetic Resonance (Part B: Structure and Mechanism) Edited by NORMAN J. OPPENHEIMER AND THOMAS L. JAMES
METHODS IN ENZYMOLOZY
xxv
VOLUME 178. Antibodies, Antigens, and Molecular Mimicry Edited by JOHN J. LANGONE VOLUME 179. Complex Carbohydrates (Part F) Edited by VICTOR GINSBURG VOLUME 180. RNA Processing (Part A: General Methods) Edited by JAMES E. DAHLBERG AND JOHN N. ABELSON VOLUME 181. RNA Processing (Part B: Specific Methods) Edited by JAMES E. DAHLBERG AND JOHN N. ABELSON VOLUME 182. Guide to Protein Purification Edited by MURRAY P. DEUTSCHER VOLUME 183. Molecular Evolution: Computer Analysis of Protein and Nucleic Acid Sequences Edited by RUSSELL F. DOOLITTLE VOLUME 184. Avidin-Biotin Technology Edited by MEIR WILCHEK AND EDWARD A. BAYER VOLUME 185. Gene Expression Technology Edited by DAVID V. GOEDDEL VOLUME 186. Oxygen Radicals in Biological Systems (Part B: Oxygen Radicals and Antioxidants) Edited by LESTER PACKER AND ALEXANDER N. GLAZER VOLUME 187. Arachidonate Related Lipid Mediators Edited by ROBERT C. MURPHY AND FRANK A. FITZPATRICK VOLUME 188. Hydrocarbons and Methylotrophy Edited by MARY E. LIDSTROM VOLUME 189. Retinoids (Part A: Molecular and Metabolic Aspects) Edited by LESTER PACKER VOLUME 190. Retinoids (Part B: Cell Differentiation and Clinical Applications) Edited by LESTER PACKER VOLUME 191. Biomembranes (Part V: Cellular and Subcellular Transport: Epithelial Cells) Edited by SIDNEY FLEISCHER AND BECCA FLEISCHER VOLUME 192. Biomembranes (Part W: Cellular and Subcellular Transport: Epithelial Cells) Edited by SIDNEY FLEISCHER AND BECCA FLEISCHER VOLUME 193. Mass Spectrometry Edited by JAMES A. MCCLOSKEY VOLUME 194. Guide to Yeast Genetics and Molecular Biology Edited by CHRISTINE GUTHRIE AND GERALD R. FINK VOLUME 195. Adenylyl Cyclase, G Proteins, and Guanylyl Cyclase Edited by ROGER A. JOHNSON AND JACKIE D. CORBIN
xxvi
METHODS IN ENZYMOLOZY
VOLUME 196. Molecular Motors and the Cytoskeleton Edited by RICHARD B. VALLEE VOLUME 197. Phospholipases Edited by EDWARD A. DENNIS VOLUME 198. Peptide Growth Factors (Part C) Edited by DAVID BARNES, J. P. MATHER, AND GORDON H. SATO VOLUME 199. Cumulative Subject Index Volumes 168–174, 176–194 VOLUME 200. Protein Phosphorylation (Part A: Protein Kinases: Assays, Purification, Antibodies, Functional Analysis, Cloning, and Expression) Edited by TONY HUNTER AND BARTHOLOMEW M. SEFTON VOLUME 201. Protein Phosphorylation (Part B: Analysis of Protein Phosphorylation, Protein Kinase Inhibitors, and Protein Phosphatases) Edited by TONY HUNTER AND BARTHOLOMEW M. SEFTON VOLUME 202. Molecular Design and Modeling: Concepts and Applications (Part A: Proteins, Peptides, and Enzymes) Edited by JOHN J. LANGONE VOLUME 203. Molecular Design and Modeling: Concepts and Applications (Part B: Antibodies and Antigens, Nucleic Acids, Polysaccharides, and Drugs) Edited by JOHN J. LANGONE VOLUME 204. Bacterial Genetic Systems Edited by JEFFREY H. MILLER VOLUME 205. Metallobiochemistry (Part B: Metallothionein and Related Molecules) Edited by JAMES F. RIORDAN AND BERT L. VALLEE VOLUME 206. Cytochrome P450 Edited by MICHAEL R. WATERMAN AND ERIC F. JOHNSON VOLUME 207. Ion Channels Edited by BERNARDO RUDY AND LINDA E. IVERSON VOLUME 208. Protein–DNA Interactions Edited by ROBERT T. SAUER VOLUME 209. Phospholipid Biosynthesis Edited by EDWARD A. DENNIS AND DENNIS E. VANCE VOLUME 210. Numerical Computer Methods Edited by LUDWIG BRAND AND MICHAEL L. JOHNSON VOLUME 211. DNA Structures (Part A: Synthesis and Physical Analysis of DNA) Edited by DAVID M. J. LILLEY AND JAMES E. DAHLBERG VOLUME 212. DNA Structures (Part B: Chemical and Electrophoretic Analysis of DNA) Edited by DAVID M. J. LILLEY AND JAMES E. DAHLBERG
METHODS IN ENZYMOLOZY
xxvii
VOLUME 213. Carotenoids (Part A: Chemistry, Separation, Quantitation, and Antioxidation) Edited by LESTER PACKER VOLUME 214. Carotenoids (Part B: Metabolism, Genetics, and Biosynthesis) Edited by LESTER PACKER VOLUME 215. Platelets: Receptors, Adhesion, Secretion (Part B) Edited by JACEK J. HAWIGER VOLUME 216. Recombinant DNA (Part G) Edited by RAY WU VOLUME 217. Recombinant DNA (Part H) Edited by RAY WU VOLUME 218. Recombinant DNA (Part I) Edited by RAY WU VOLUME 219. Reconstitution of Intracellular Transport Edited by JAMES E. ROTHMAN VOLUME 220. Membrane Fusion Techniques (Part A) ¨ ¨ UNES Edited by NEJAT DUZGU VOLUME 221. Membrane Fusion Techniques (Part B) ¨ ¨ UNES Edited by NEJAT DUZG VOLUME 222. Proteolytic Enzymes in Coagulation, Fibrinolysis, and Complement Activation (Part A: Mammalian Blood Coagulation Factors and Inhibitors) Edited by LASZLO LORAND AND KENNETH G. MANN VOLUME 223. Proteolytic Enzymes in Coagulation, Fibrinolysis, and Complement Activation (Part B: Complement Activation, Fibrinolysis, and Nonmammalian Blood Coagulation Factors) Edited by LASZLO LORAND AND KENNETH G. MANN VOLUME 224. Molecular Evolution: Producing the Biochemical Data Edited by ELIZABETH ANNE ZIMMER, THOMAS J. WHITE, REBECCA L. CANN, AND ALLAN C. WILSON VOLUME 225. Guide to Techniques in Mouse Development Edited by PAUL M. WASSARMAN AND MELVIN L. DEPAMPHILIS VOLUME 226. Metallobiochemistry (Part C: Spectroscopic and Physical Methods for Probing Metal Ion Environments in Metalloenzymes and Metalloproteins) Edited by JAMES F. RIORDAN AND BERT L. VALLEE VOLUME 227. Metallobiochemistry (Part D: Physical and Spectroscopic Methods for Probing Metal Ion Environments in Metalloproteins) Edited by JAMES F. RIORDAN AND BERT L. VALLEE VOLUME 228. Aqueous Two-Phase Systems ¨ JOHANSSON Edited by HARRY WALTER AND GOTE
xxviii
METHODS IN ENZYMOLOZY
VOLUME 229. Cumulative Subject Index Volumes 195–198, 200–227 VOLUME 230. Guide to Techniques in Glycobiology Edited by WILLIAM J. LENNARZ AND GERALD W. HART VOLUME 231. Hemoglobins (Part B: Biochemical and Analytical Methods) Edited by JOHANNES EVERSE, KIM D. VANDEGRIFF, AND ROBERT M. WINSLOW VOLUME 232. Hemoglobins (Part C: Biophysical Methods) Edited by JOHANNES EVERSE, KIM D. VANDEGRIFF, AND ROBERT M. WINSLOW VOLUME 233. Oxygen Radicals in Biological Systems (Part C) Edited by LESTER PACKER VOLUME 234. Oxygen Radicals in Biological Systems (Part D) Edited by LESTER PACKER VOLUME 235. Bacterial Pathogenesis (Part A: Identification and Regulation of Virulence Factors) Edited by VIRGINIA L. CLARK AND PATRIK M. BAVOIL VOLUME 236. Bacterial Pathogenesis (Part B: Integration of Pathogenic Bacteria with Host Cells) Edited by VIRGINIA L. CLARK AND PATRIK M. BAVOIL VOLUME 237. Heterotrimeric G Proteins Edited by RAVI IYENGAR VOLUME 238. Heterotrimeric G-Protein Effectors Edited by RAVI IYENGAR VOLUME 239. Nuclear Magnetic Resonance (Part C) Edited by THOMAS L. JAMES AND NORMAN J. OPPENHEIMER VOLUME 240. Numerical Computer Methods (Part B) Edited by MICHAEL L. JOHNSON AND LUDWIG BRAND VOLUME 241. Retroviral Proteases Edited by LAWRENCE C. KUO AND JULES A. SHAFER VOLUME 242. Neoglycoconjugates (Part A) Edited by Y. C. LEE AND REIKO T. LEE VOLUME 243. Inorganic Microbial Sulfur Metabolism Edited by HARRY D. PECK, JR., AND JEAN LEGALL VOLUME 244. Proteolytic Enzymes: Serine and Cysteine Peptidases Edited by ALAN J. BARRETT VOLUME 245. Extracellular Matrix Components Edited by E. RUOSLAHTI AND E. ENGVALL VOLUME 246. Biochemical Spectroscopy Edited by KENNETH SAUER VOLUME 247. Neoglycoconjugates (Part B: Biomedical Applications) Edited by Y. C. LEE AND REIKO T. LEE
METHODS IN ENZYMOLOZY
xxix
VOLUME 248. Proteolytic Enzymes: Aspartic and Metallo Peptidases Edited by ALAN J. BARRETT VOLUME 249. Enzyme Kinetics and Mechanism (Part D: Developments in Enzyme Dynamics) Edited by DANIEL L. PURICH VOLUME 250. Lipid Modifications of Proteins Edited by PATRICK J. CASEY AND JANICE E. BUSS VOLUME 251. Biothiols (Part A: Monothiols and Dithiols, Protein Thiols, and Thiyl Radicals) Edited by LESTER PACKER VOLUME 252. Biothiols (Part B: Glutathione and Thioredoxin; Thiols in Signal Transduction and Gene Regulation) Edited by LESTER PACKER VOLUME 253. Adhesion of Microbial Pathogens Edited by RON J. DOYLE AND ITZHAK OFEK VOLUME 254. Oncogene Techniques Edited by PETER K. VOGT AND INDER M. VERMA VOLUME 255. Small GTPases and Their Regulators (Part A: Ras Family) Edited by W. E. BALCH, CHANNING J. DER, AND ALAN HALL VOLUME 256. Small GTPases and Their Regulators (Part B: Rho Family) Edited by W. E. BALCH, CHANNING J. DER, AND ALAN HALL VOLUME 257. Small GTPases and Their Regulators (Part C: Proteins Involved in Transport) Edited by W. E. BALCH, CHANNING J. DER, AND ALAN HALL VOLUME 258. Redox-Active Amino Acids in Biology Edited by JUDITH P. KLINMAN VOLUME 259. Energetics of Biological Macromolecules Edited by MICHAEL L. JOHNSON AND GARY K. ACKERS VOLUME 260. Mitochondrial Biogenesis and Genetics (Part A) Edited by GIUSEPPE M. ATTARDI AND ANNE CHOMYN VOLUME 261. Nuclear Magnetic Resonance and Nucleic Acids Edited by THOMAS L. JAMES VOLUME 262. DNA Replication Edited by JUDITH L. CAMPBELL VOLUME 263. Plasma Lipoproteins (Part C: Quantitation) Edited by WILLIAM A. BRADLEY, SANDRA H. GIANTURCO, AND JERE P. SEGREST VOLUME 264. Mitochondrial Biogenesis and Genetics (Part B) Edited by GIUSEPPE M. ATTARDI AND ANNE CHOMYN VOLUME 265. Cumulative Subject Index Volumes 228, 230–262
xxx
METHODS IN ENZYMOLOZY
VOLUME 266. Computer Methods for Macromolecular Sequence Analysis Edited by RUSSELL F. DOOLITTLE VOLUME 267. Combinatorial Chemistry Edited by JOHN N. ABELSON VOLUME 268. Nitric Oxide (Part A: Sources and Detection of NO; NO Synthase) Edited by LESTER PACKER VOLUME 269. Nitric Oxide (Part B: Physiological and Pathological Processes) Edited by LESTER PACKER VOLUME 270. High Resolution Separation and Analysis of Biological Macromolecules (Part A: Fundamentals) Edited by BARRY L. KARGER AND WILLIAM S. HANCOCK VOLUME 271. High Resolution Separation and Analysis of Biological Macromolecules (Part B: Applications) Edited by BARRY L. KARGER AND WILLIAM S. HANCOCK VOLUME 272. Cytochrome P450 (Part B) Edited by ERIC F. JOHNSON AND MICHAEL R. WATERMAN VOLUME 273. RNA Polymerase and Associated Factors (Part A) Edited by SANKAR ADHYA VOLUME 274. RNA Polymerase and Associated Factors (Part B) Edited by SANKAR ADHYA VOLUME 275. Viral Polymerases and Related Proteins Edited by LAWRENCE C. KUO, DAVID B. OLSEN, AND STEVEN S. CARROLL VOLUME 276. Macromolecular Crystallography (Part A) Edited by CHARLES W. CARTER, JR., AND ROBERT M. SWEET VOLUME 277. Macromolecular Crystallography (Part B) Edited by CHARLES W. CARTER, JR., AND ROBERT M. SWEET VOLUME 278. Fluorescence Spectroscopy Edited by LUDWIG BRAND AND MICHAEL L. JOHNSON VOLUME 279. Vitamins and Coenzymes (Part I) Edited by DONALD B. MCCORMICK, JOHN W. SUTTIE, AND CONRAD WAGNER VOLUME 280. Vitamins and Coenzymes (Part J) Edited by DONALD B. MCCORMICK, JOHN W. SUTTIE, AND CONRAD WAGNER VOLUME 281. Vitamins and Coenzymes (Part K) Edited by DONALD B. MCCORMICK, JOHN W. SUTTIE, AND CONRAD WAGNER VOLUME 282. Vitamins and Coenzymes (Part L) Edited by DONALD B. MCCORMICK, JOHN W. SUTTIE, AND CONRAD WAGNER VOLUME 283. Cell Cycle Control Edited by WILLIAM G. DUNPHY VOLUME 284. Lipases (Part A: Biotechnology) Edited by BYRON RUBIN AND EDWARD A. DENNIS
METHODS IN ENZYMOLOZY
xxxi
VOLUME 285. Cumulative Subject Index Volumes 263, 264, 266–284, 286–289 VOLUME 286. Lipases (Part B: Enzyme Characterization and Utilization) Edited by BYRON RUBIN AND EDWARD A. DENNIS VOLUME 287. Chemokines Edited by RICHARD HORUK VOLUME 288. Chemokine Receptors Edited by RICHARD HORUK VOLUME 289. Solid Phase Peptide Synthesis Edited by GREGG B. FIELDS VOLUME 290. Molecular Chaperones Edited by GEORGE H. LORIMER AND THOMAS BALDWIN VOLUME 291. Caged Compounds Edited by GERARD MARRIOTT VOLUME 292. ABC Transporters: Biochemical, Cellular, and Molecular Aspects Edited by SURESH V. AMBUDKAR AND MICHAEL M. GOTTESMAN VOLUME 293. Ion Channels (Part B) Edited by P. MICHAEL CONN VOLUME 294. Ion Channels (Part C) Edited by P. MICHAEL CONN VOLUME 295. Energetics of Biological Macromolecules (Part B) Edited by GARY K. ACKERS AND MICHAEL L. JOHNSON VOLUME 296. Neurotransmitter Transporters Edited by SUSAN G. AMARA VOLUME 297. Photosynthesis: Molecular Biology of Energy Capture Edited by LEE MCINTOSH VOLUME 298. Molecular Motors and the Cytoskeleton (Part B) Edited by RICHARD B. VALLEE VOLUME 299. Oxidants and Antioxidants (Part A) Edited by LESTER PACKER VOLUME 300. Oxidants and Antioxidants (Part B) Edited by LESTER PACKER VOLUME 301. Nitric Oxide: Biological and Antioxidant Activities (Part C) Edited by LESTER PACKER VOLUME 302. Green Fluorescent Protein Edited by P. MICHAEL CONN VOLUME 303. cDNA Preparation and Display Edited by SHERMAN M. WEISSMAN VOLUME 304. Chromatin Edited by PAUL M. WASSARMAN AND ALAN P. WOLFFE
xxxii
METHODS IN ENZYMOLOZY
VOLUME 305. Bioluminescence and Chemiluminescence (Part C) Edited by THOMAS O. BALDWIN AND MIRIAM M. ZIEGLER VOLUME 306. Expression of Recombinant Genes in Eukaryotic Systems Edited by JOSEPH C. GLORIOSO AND MARTIN C. SCHMIDT VOLUME 307. Confocal Microscopy Edited by P. MICHAEL CONN VOLUME 308. Enzyme Kinetics and Mechanism (Part E: Energetics of Enzyme Catalysis) Edited by DANIEL L. PURICH AND VERN L. SCHRAMM VOLUME 309. Amyloid, Prions, and Other Protein Aggregates Edited by RONALD WETZEL VOLUME 310. Biofilms Edited by RON J. DOYLE VOLUME 311. Sphingolipid Metabolism and Cell Signaling (Part A) Edited by ALFRED H. MERRILL, JR., AND YUSUF A. HANNUN VOLUME 312. Sphingolipid Metabolism and Cell Signaling (Part B) Edited by ALFRED H. MERRILL, JR., AND YUSUF A. HANNUN VOLUME 313. Antisense Technology (Part A: General Methods, Methods of Delivery, and RNA Studies) Edited by M. IAN PHILLIPS VOLUME 314. Antisense Technology (Part B: Applications) Edited by M. IAN PHILLIPS VOLUME 315. Vertebrate Phototransduction and the Visual Cycle (Part A) Edited by KRZYSZTOF PALCZEWSKI VOLUME 316. Vertebrate Phototransduction and the Visual Cycle (Part B) Edited by KRZYSZTOF PALCZEWSKI VOLUME 317. RNA–Ligand Interactions (Part A: Structural Biology Methods) Edited by DANIEL W. CELANDER AND JOHN N. ABELSON VOLUME 318. RNA–Ligand Interactions (Part B: Molecular Biology Methods) Edited by DANIEL W. CELANDER AND JOHN N. ABELSON VOLUME 319. Singlet Oxygen, UV-A, and Ozone Edited by LESTER PACKER AND HELMUT SIES VOLUME 320. Cumulative Subject Index Volumes 290–319 VOLUME 321. Numerical Computer Methods (Part C) Edited by MICHAEL L. JOHNSON AND LUDWIG BRAND VOLUME 322. Apoptosis Edited by JOHN C. REED VOLUME 323. Energetics of Biological Macromolecules (Part C) Edited by MICHAEL L. JOHNSON AND GARY K. ACKERS
METHODS IN ENZYMOLOZY
xxxiii
VOLUME 324. Branched-Chain Amino Acids (Part B) Edited by ROBERT A. HARRIS AND JOHN R. SOKATCH VOLUME 325. Regulators and Effectors of Small GTPases (Part D: Rho Family) Edited by W. E. BALCH, CHANNING J. DER, AND ALAN HALL VOLUME 326. Applications of Chimeric Genes and Hybrid Proteins (Part A: Gene Expression and Protein Purification) Edited by JEREMY THORNER, SCOTT D. EMR, AND JOHN N. ABELSON VOLUME 327. Applications of Chimeric Genes and Hybrid Proteins (Part B: Cell Biology and Physiology) Edited by JEREMY THORNER, SCOTT D. EMR, AND JOHN N. ABELSON VOLUME 328. Applications of Chimeric Genes and Hybrid Proteins (Part C: ProteinProtein Interactions and Genomics) Edited by JEREMY THORNER, SCOTT D. EMR, AND JOHN N. ABELSON VOLUME 329. Regulators and Effectors of Small GTPases (Part E: GTPases Involved in Vesicular Traffic) Edited by W. E. BALCH, CHANNING J. DER, AND ALAN HALL VOLUME 330. Hyperthermophilic Enzymes (Part A) Edited by MICHAEL W. W. ADAMS AND ROBERT M. KELLY VOLUME 331. Hyperthermophilic Enzymes (Part B) Edited by MICHAEL W. W. ADAMS AND ROBERT M. KELLY VOLUME 332. Regulators and Effectors of Small GTPases (Part F: Ras Family I) Edited by W. E. BALCH, CHANNING J. DER, AND ALAN HALL VOLUME 333. Regulators and Effectors of Small GTPases (Part G: Ras Family II) Edited by W. E. BALCH, CHANNING J. DER, AND ALAN HALL VOLUME 334. Hyperthermophilic Enzymes (Part C) Edited by MICHAEL W. W. ADAMS AND ROBERT M. KELLY VOLUME 335. Flavonoids and Other Polyphenols Edited by LESTER PACKER VOLUME 336. Microbial Growth in Biofilms (Part A: Developmental and Molecular Biological Aspects) Edited by RON J. DOYLE VOLUME 337. Microbial Growth in Biofilms (Part B: Special Environments and Physicochemical Aspects) Edited by RON J. DOYLE VOLUME 338. Nuclear Magnetic Resonance of Biological Macromolecules (Part A) ¨ Edited by THOMAS L. JAMES, VOLKER DOTSCH , AND ULI SCHMITZ VOLUME 339. Nuclear Magnetic Resonance of Biological Macromolecules (Part B) ¨ Edited by THOMAS L. JAMES, VOLKER DOTSCH , AND ULI SCHMITZ
xxxiv
METHODS IN ENZYMOLOZY
VOLUME 340. Drug–Nucleic Acid Interactions Edited by JONATHAN B. CHAIRES AND MICHAEL J. WARING VOLUME 341. Ribonucleases (Part A) Edited by ALLEN W. NICHOLSON VOLUME 342. Ribonucleases (Part B) Edited by ALLEN W. NICHOLSON VOLUME 343. G Protein Pathways (Part A: Receptors) Edited by RAVI IYENGAR AND JOHN D. HILDEBRANDT VOLUME 344. G Protein Pathways (Part B: G Proteins and Their Regulators) Edited by RAVI IYENGAR AND JOHN D. HILDEBRANDT VOLUME 345. G Protein Pathways (Part C: Effector Mechanisms) Edited by RAVI IYENGAR AND JOHN D. HILDEBRANDT VOLUME 346. Gene Therapy Methods Edited by M. IAN PHILLIPS VOLUME 347. Protein Sensors and Reactive Oxygen Species (Part A: Selenoproteins and Thioredoxin) Edited by HELMUT SIES AND LESTER PACKER VOLUME 348. Protein Sensors and Reactive Oxygen Species (Part B: Thiol Enzymes and Proteins) Edited by HELMUT SIES AND LESTER PACKER VOLUME 349. Superoxide Dismutase Edited by LESTER PACKER VOLUME 350. Guide to Yeast Genetics and Molecular and Cell Biology (Part B) Edited by CHRISTINE GUTHRIE AND GERALD R. FINK VOLUME 351. Guide to Yeast Genetics and Molecular and Cell Biology (Part C) Edited by CHRISTINE GUTHRIE AND GERALD R. FINK VOLUME 352. Redox Cell Biology and Genetics (Part A) Edited by CHANDAN K. SEN AND LESTER PACKER VOLUME 353. Redox Cell Biology and Genetics (Part B) Edited by CHANDAN K. SEN AND LESTER PACKER VOLUME 354. Enzyme Kinetics and Mechanisms (Part F: Detection and Characterization of Enzyme Reaction Intermediates) Edited by DANIEL L. PURICH VOLUME 355. Cumulative Subject Index Volumes 321–354 (in preparation) VOLUME 356. Laser Capture Microscopy and Microdissection Edited by P. MICHAEL CONN VOLUME 357. Cytochrome P450 (Part C) (in preparation) Edited by ERIC F. JOHNSON AND MICHAEL R. WATERMAN
METHODS IN ENZYMOLOZY
xxxv
VOLUME 358. Bacterial Pathogenesis (Part C: Identification, Regulation, and Function of Virulence Factors) (in preparation) Edited by VIRGINIA L. CLARK AND PATRIK M. BAVOIL VOLUME 359. Nitric Oxide (Part D) (in preparation) Edited by ENRIQUE CADENAS AND LESTER PACKER VOLUME 360. Biophotonics (Part A) (in preparation) Edited by GERARD MARRIOTT AND IAN PARKER VOLUME 361. Biophotonics (Part B) (in preparation) Edited by GERARD MARRIOTT AND IAN PARKER
This Page Intentionally Left Blank
Section I Basic Principles
This Page Intentionally Left Blank
[1]
COMPARISON OF CURRENT EQUIPMENT
3
[1] Comparison of Current Equipment By ANDA CORNEA and ALISON MUNGENAST Introduction Some of the most exciting new developments in biomedical research, such as DNA microarrays and proteomics, depend on the isolation of single cells or pure populations of cells with specific phenotypes. Several microscopic techniques are currently used for microdissection: suction of a cell content through micron-size glass pipettes, dissection using a piezo-activated metal knife and suction through a glass pipette (PPMD by Brinkmann), and dissection using lasers. Because of the high energy concentrated into a small area, the easy control of the beam position, and the lack of direct contact with the material to be dissected, lasers provide the best option for easy-to-use, large-scale microdissections. There are currently three commercially available systems designed specifically for laser capture microdissection: PixCell by Arcturus (Mountain View, CA), PALM by P.A.L.M. Mikrolaser Technologie (Wolfratshousen, Germany), and the Leica AS LMD by Leica (Heidelberg, Germany). The PixCell originated in a Cooperative Research and Development Agreement between the National Institutes of Health, the National Cancer Institute, and the National Institute for Child and Human Development and is now manufactured and marketed by Arcturus. The system uses an IR laser focused through a microscope objective to heat a plastic film placed above a section of tissue. The plastic melts temporarily in the small area irradiated and penetrates the tissue. When the laser beam is turned off, the plastic solidifies and forms bonds with the tissue it has penetrated. When the plastic sheet is removed, the tissue bonded to the plastic is removed as well and thereby isolated or dissected from the rest (Fig. 1). The dissected material may then be processed for the isolation of RNA, DNA, or proteins (Emmert-Buck et al.1 ). The PALM system uses an N2 laser with 336 nm wavelength. The laser, also focused through an inverted microscope objective, has enough energy to ablate tissue or cells that are in focus. Ablation, which destroys chemical bonds within a tissue by a mechanism not fully understood, may remove undesirable cells or groups of cells and isolate a region of interest. When the laser is slightly defocused, its energy may be used to catapult the dissected material up to where it may be collected into a cap and stored or immediately used for the isolation of RNA, DNA, 1
M. R. Emmert-Buck, R. F. Bonner, P. D. Smith, R. F. Chuaqui, Z. Zuang, S. R. Goldstein, R. A. Weiss, and L. A. Liotta, Science 274, 998 (1996).
METHODS IN ENZYMOLOGY, VOL. 356
Copyright 2002, Elsevier Science (USA). All rights reserved. 0076-6879/02 $35.00
4
[1]
BASIC PRINCIPLES
Laser beam Plastic film Tissue section Glass slide
A
Dissected cell
B
Plastic film
Tissue section Glass slide
FIG. 1. Principle of laser capture microscopy used by Arcturus. (A) A thin plastic film is lowered over the specimen to be dissected. An IR laser beam focus through a microscope objective illuminates a small area on the plastic causing it to melt locally and penetrate the tissue. (B) When the laser is interrupted, the plastic solidifies forming bonds with the underlying tissue. When the plastic is removed, the attached tissue is isolated from the rest of the specimen.
or proteins (Schutze and Lahr2 ). This process, termed laser pressure catapulting, is patented by P.A.L.M. (Fig. 2). The Leica AS LMD, more recently introduced as the third generation of laser microdissection systems, uses a pulsed UV laser similar to the PALM on an upright microscope. The laser beam may be moved with a software-controlled mirror system to select cells to be ablated or to isolate the area to be dissected. The dissected material is allowed to fall by gravity into a cap and may thereafter be used for isolating proteins or genetic material (Fig. 3). There are several parameters that may be used to characterize a laser microdissection system. Among them the most important are the resolution and the specificity of dissection and the integrity of the dissected material. Very important also are the ease of use, the reliability of the instrument, and the quality of service and support. Resolution The smallest area that can be isolated from the rest of a tissue by laser microdissection is related to the size of the laser beam and therefore depends on the numerical aperture (NA) of the objective used for dissection and on the wavelength of light used. The NA of an objective is defined as the sine of the collection 2
K. Schute and G. Lahr, Nat. Biotechnol. 16, 737 (1998).
[1]
COMPARISON OF CURRENT EQUIPMENT
5
Tissue section Plastic film Glass slide
A Laser beam (ablating)
B Laser beam (catapulting)
C Laser beam (catapulting) FIG. 2. Principle of the PALM laser microdissection. (A) A UV laser beam focused by the objective of an inverted microscope cuts a contour around the area to be dissected. (B) The laser is defocused and positioned within the selected area. (C) The laser pressure is used to lift the dissected sample into a collecting cap. This process, named laser pressure catapulting, is patented by P.A.L.M.
angle multiplied by the refractive index of the immersion medium. Most common configurations for laser microdissection use dry objectives, which somewhat limits the NA. The most stringent limitation comes, however, from the maximum collection angle that can be used. For limited lens diameters, a large angle imposes a short working distance. In most cases, specimens to be dissected are mounted on standard microscope glass slides with approximately 1 mm thickness. For the PALM and the Leica AS LMD systems, the light exiting the objective needs to cross this distance in order to be focused on the tissue. This requires long working distance objectives that necessarily have a lower NA and consequently a wider laser beam waist. For cases when the thinnest possible cuts are required, higher NA objectives may be used if the specimen is mounted on a thin glass coverslip, making possible the use of short working distance objectives with high NA. The main inconvenience in this case is the fragility of these coverslips, as all steps prior to dissection, including collection of sections, fixing, staining, transport, and mounting on the microscope stage, have to be done with extra care. In the PixCell, light leaving the objective passes through the collecting cap and that also limits the minimum usable working distance and consequently the NA of
6
[1]
BASIC PRINCIPLES
Laser beam (ablating) Glass slide Plastic film Tissue section
A
Laser beam (ablating)
B
C Dissected cell FIG. 3. Principle of the Leica AS LMD microdissection. (A) A laser beam similar to the one used by PALM is concentrated by the objective of an upright microscope. The specimen, mounted on a PEN membrane, is mounted upside down. (B) A contour is cut through the membrane and tissue around the area to be dissected by moving the laser beam and not the stage. (C) The dissected area, isolated from the rest of the specimen, falls into a collecting cap positioned under the specimen.
the objective. In this case, the user does not have the option to mount specimens on coverslips; the only improvement could come from redesigning the collection caps. The size of the beam waist depends on the wavelength of light used. The infrared light used by PixCell will give a larger beam waist than the UV light used by the PALM and Leica AS LMD for the same objective used. The demands for the smallest possible beam waist, however, are different for the two classes of microdissection instruments, as the dissection technique is different. For the PixCell, the beam waist gives the minimum dissected area. The minimum value cited by the manufacturer is 7.5 µm. This is about the size of a cell, or smaller than many cells. It can be argued that a smaller size is hardly necessary, as in most cases laser microdissection is used to isolate single cells or larger numbers of identical cells and, in this case, the intent is to collect as much material per cell as possible. The PALM and Leica AS LMD use the laser to ablate a contour line around a single cell and therefore the width of the cut is expected to be much smaller. The minimum cut sizes quoted for the PALM and Leica AS LMD were less than 1 µm and 2.5 µm, respectively.
[1]
COMPARISON OF CURRENT EQUIPMENT
7
Specificity and Integrity In all systems the material removed may include more than the specific area marked for dissection. In the case of the PixCell, the bond between the plastic and the tissue must be stronger than the bond between tissue and slide in order to allow removal of the cell or tissue part marked for dissection. Uncharged and unsubbed slides work well because the tissue adheres more loosely to the slides. This presents a problem, however, when the user desires to process the tissue beyond simple staining, such as using immunohistochemistry, as sections may be lost during the procedure. The bonds within the tissue, dependent on tissue type and fixation, may also overcome in some cases the attachment to the slide, and extra material will be removed contaminating the purity of the dissected sample (Fig. 4). For some of the collecting caps, the most frequently used ones, the plastic film touches the section of tissue to be dissected and may remove material that randomly adhered to it. This is, however, overcome in the more newly designed “CapSure” caps in which only the melted film touches the tissue. Sticky “prep strips” are also provided which can reduce contamination.
FIG. 4. Positive selection of cells to be dissected with PixCell laser capture microscope. (A) Brain section in which cells selected for dissection appear in a lighter color after the plastic transfer film was attached by melting. (B) Cells removed with the film. Arrows indicate cells for which nuclei failed to transfer. Arrowheads indicate dissection that removed extra material. Note the clarity of quality control.
8
BASIC PRINCIPLES
[1]
FIG. 5. Negative selection of area to be selected by a PALM laser microdissection microscope. (A) A freehand line is drawn on the screen over the image. (B) The laser cuts the PEN membrane and ablates the tissue along that line. (C) The tissue enclosed by the cut is catapulted using laser pressure. (D) Sample remaining after positive selection of single cells in a different tissue section mounted directly on a glass slide.
The laser ablation used in the PALM and Leica LDM is aimed at circumventing this contamination problem by destroying the tissue around the cell or the region of interest that can then be collected free of neighboring contamination (Fig. 5). The process of ablation itself is not well understood and the destruction of chemical bonds may not be complete. Particles of the material targeted for ablation may be seen, while cutting, landing on the cell or region to be dissected suggesting a potential for contamination. Cells adjacent to the cell of interest cannot be collected as well with this process, as they are ablated by the laser beam. With the Arcturus, however, it is possible to collect material from adjacent cells. The integrity of the material dissected may be an issue as in all cases the sample is irradiated with laser light that has the potential to alter chemical bonds by either destroying them or causing cross-linking. This is particularly an issue for the UV lasers even though the wavelength is slightly larger than the main absorption peak for proteins and nucleic acids. The high temperature created by the IR laser in the PixCell system, necessary for the melting of the film, may also degrade the dissected material. There is at this time a large body of literature suggesting that adequate RNA may be quantitatively isolated by both technologies and not much published evidence to the contrary. Ease of Use There are many steps involved in laser capture microdissection that may affect the ease of use. The type of slides required, demands on sectioning and processing of the tissue, visualization and selection of the region to be selected, and dissection
[1]
COMPARISON OF CURRENT EQUIPMENT
9
and collection of the dissected material are all variables that may make the technique easy and straightforward or complicated and awkward. Regular microscope slides may be used in all systems, albeit with suboptimal results. For the PALM and the Leica DML it is recommended that a thin PEN (polyethylene naphthalate) membrane be mounted between the slide and the tissue. The membrane may be cut by the ablation laser around the area targeted for dissection and then catapulted in the PALM or let drop in the Leica, preserving the integrity of the cells attached to it. Dissected pieces of tissue may be visualized after capture in the collecting cap only if the underlying membrane kept the structure intact. If cells are catapulted directly from a slide, the material will be pulverized and it will be impossible to assess the efficiency of capture by visualization. PEN coated slides may be purchased but are relatively expensive. Preparing them, however, may be very time consuming. We have also found that the membranes tear easily during immunohistochemical processing. For the PixCell, the tissue may be mounted directly on a glass slide. The dissected cells are then collected on a plastic film attached to the collecting caps and can be easily visualized. For all systems, the efficiency of dissection and collection is critically dependent on tissue preparation as well as on environmental factors.
Sample Preparation Arcturus provides a number of protocols for sample preparation that must be strictly followed in order to obtain a good dissection. The thickness of the section, method of fixation of the tissue, and staining are restricted. Specimens need to be perfectly dry to adhere to the transfer film. The PALM and the Leica systems rely largely on the ablation of the PEN foil for dissection and therefore offer more freedom for the preparation of the specimen. Tissue sections may be much thicker and various fixations may be used. The PALM may even isolate live cells. Not everything works, however, and the greater freedom in sample preparation comes with the necessity for users to optimize their own protocols.
Environment Humidity in the specimen, due either to incomplete drying or to absorption of air vapors, makes dissection in the PixCell impossible. Samples and instrument need to be kept in a dry environment. Humidity between the slide and the membrane prevents proper membrane cutting in the PALM. Lack of humidity, however, contributes to increased electrostatic forces that compete with gravity and affect collection in the Leica AS LMD system. In this case, the instrument should be kept in a high-humidity room free of air drafts.
10
BASIC PRINCIPLES
[1]
Control of Stage Movement The control of the stage and selection of areas for dissection may strongly affect the ease of use. The PixCell uses a joystick manual control of the stage movement. The stage is moved so that each intended target is brought into the middle of the field of view, under the laser beam, and dissected, before moving to the next target and so on. If areas larger than can be covered by a single beam are desired, multiple laser shots need to be fired as the stage is moved. This requires skill and concentration, as any misfire can contaminate the whole sample. The PALM uses computer-controlled stage motors that can move the stage along a predrawn path. This is particularly convenient when large areas of irregular shapes are dissected. A freehand drawing tool allows the operator to outline the target area, and then the stage moves as the laser fires and cuts the underlying membrane along the chosen path. The whole area is then catapulted into the collecting cap. When several isolated cells are collected into the same cap, they can each be outlined first, then the system isolates them one by one, automatically. The Leica AS LMD uses a computer-controlled mirror that moves the laser beam along a path also preselected by a freehand drawing tool. Cutting may be in this case much faster than for the PALM, as the stage is immobile and the laser beam is moved faster.
Service A major factor in the satisfaction with any of the instruments chosen is the reliability and speed and quality of service. Arcturus has the longest history of operation and, at least in the United States, has an excellent record of reliability. The IR laser it uses is designed for a longer lifetime than the UV laser used by the PALM and Leica estimated at 2 years or 2,000,000 pulses. The PALM, now marketed and supported by Zeiss, and Leica are expanding their operations in the United States, relying for applications and service on the preexisting networks of the two respective companies. We have tested all three systems in an attempt to find the one that best fits most of our needs. Our experience showed that for all three systems, sample preparation and environmental factors are critical for good dissection. Specimens freshly prepared according to the Arcturus-suggested protocols could be very easily dissected. Samples insufficiently dried or stored without desiccation could not be dissected. Specimens prepared for demonstration by the PALM could be easily dissected and catapulted. With a specimen that we prepared with membrane coated slides provided by PALM which maintained a little humidity, the membrane could not be cut reliably even with the highest setting for the laser energy. When the tissue was mounted directly on a glass slide without a membrane, individual cells could be easily catapulted into the collecting cap.
[1]
COMPARISON OF CURRENT EQUIPMENT
11
Dissection using the Leica AS LMD system was less sensitive to sample preparation. Tissue sections were mounted in this case on a thin PEN membrane held by a plastic frame the size of a microscope slide that was easy to dry. Dissected samples, however, were subjected to strong electrostatic attraction from the many charged plastic and metal surfaces close by, allowing only few of them to be collected in the caps. Air humidity in this case is helpful in reducing the electrostatic charges around the membrane. Shooting the slide with an electrostatic gun also helped to some degree. The ability to visually inspect the dissected material was excellent for the Arcturus system. The images of the remaining and dissected material, acquired with the same objective, match each other perfectly and may be added to reconstitute the image before the dissection. In the case of the PALM, larger areas dissected could be visually inspected with relative ease, albeit with a lower magnification objective. Cells mounted directly on the glass slide, dissected by catapulting, are practically pulverized and therefore cannot be seen. Dissected cells could not be seen in the Leica AS LMD. For each instrument we tested the presence and integrity of RNA in the collecting caps. Groups of cells or single cells were captured with each microscope. Cells were snap-frozen immediately after capture. RNA was extracted using the Arcturus PicoPure RNA Isolation kit following the Arcturus protocol. Extracted RNA was dried down to 1.6 µl in a vacuum centrifuge. The following components were added to each sample: 1.7 µl Invitrogen 1st Strand Buffer, 0.5 µl RNasin (Promega), 1.7 µl 0.1 M DTT (Invitrogen), 1 µl dNTPs (Promega), 100 ng Random Hexamers (Invitrogen), and 1 µl Superscript II Reverse Transcriptase (Invitrogen) to a final volume of 8.5 µl. The samples were incubated at 42◦ for 90 min.
FIG. 6. All captured samples contain intact mRNA. Laser-captured samples were subjected to RNA extraction with the Arcturus PicoPure kit and reverse transcription with the GIBCO-BRL Superscript II enzyme. Two rounds of PCR were performed with nested primers recognizing cyclophilin, a housekeeping gene. Lanes 1 and 2 contain a sample isolated with the PALM equipment, lanes 3 through 8 contain samples isolated with the Leica AS LMD, and lanes 9 and 10 contain a sample isolated with the Arcturus PixCell system. Lanes 11 and 12 contain PCR products from control rat hypothalamic mRNA. Lanes 1, 3, 5, 7, 9, and 11 represent the first round of nested PCR using outer primers. Lanes 2, 4, 6, 8, 10, and 12 represent the final PCR product after a second round of PCR with inner primers.
12
[2]
BASIC PRINCIPLES TABLE I COMPARISON OF CURRENT EQUIPMENT PixCell II
PALM
Leica LMD
+++ ++ + ++ +++ +++ ++ +++ ++ +++ ++
+ +++ +++ +++ ++ ++ +++ + ++ +++ +
+ +++ ++ +++ ++ ++ +++
Laser lifetime Resolution Versatility Sample preparation Sample preparation protocols Ease of use—single cells Ease of use—larger areas Visualization of dissected sample Sample recovery RNA integrity Price
+ +++ ++
After the reverse transcription reaction, 1 µl from each sample was subjected to two 35-cycle rounds of PCR with nested primers recognizing cyclophilin, a housekeeping gene. The PCR products from both rounds were electrophoresed on a 2% agarose gel (Fig. 6). Our experience with the instruments tested is summarized in Table I. Addendum: Useful Sites http://www.arctur.com/ http://www.palm.spacenet.de/ http://www.leica-microsystems.com/
[2] Laser Capture Microdissection and Its Applications in Genomics and Proteomics By JAMES L. WITTLIFF and MARK G. ERLANDER Background Human tissue collection, handling, and analyses present specific problems for clinically reliable genomic and proteomic testing unlike studies with animal tissues or homogeneous cell lines grown in culture. For example, determinations of levels of clinically relevant analytes in tissue biopsies, used as markers for detection, diagnosis, prognosis, or therapeutic response of a cancer patient, are performed
METHODS IN ENZYMOLOGY, VOL. 356
Copyright 2002, Elsevier Science (USA). All rights reserved. 0076-6879/02 $35.00
[2]
LASER CAPTURE MICRODISSECTION
13
either biochemically or by immunohistochemistry currently (e.g., Wittliff et al.1 ). If the analyte (e.g., estrogen receptor, HER-2/neu oncoprotein) is measured biochemically, a tissue specimen consisting of a heterogeneous cell population is homogenized and the final concentration of the analyte extracted from the cancer cells is “diluted” by the contribution of other proteins released from noncancerous cells (e.g., epithelium, histiocytes, macrophages, and connective tissue cells). Therefore, an underestimate of the analyte concentration is likely to be determined compromising the appropriate cutoff value between disease and normal states. While certain tumor markers in tissue biopsies have well served the clinical management of cancer patients (e.g., estrogen receptors in the selection of tamoxifen-responsive breast cancer1 ), many questions of analyte expression in normal and neoplastic cells remain. Likewise, immunohistochemistry is used to measure certain proteins in cancer tissue sections for clinical application in spite of reports indicating the results are often highly operator and antibody dependent and, at best, semiquantitative (e.g., Igarashi et al.2 ). As Wittliff1 and Cole et al.3 have noted, collection and processing of human tissue biopsies have focused on their clinical purpose (e.g., diagnosis, staging, prognosis, therapy selection) with little emphasis on sampling and cryopreservation for sophisticated genomic (e.g., microarrays) and proteomic analyses (e.g., protein chips). The obvious problem of cellular heterogeneity in the tissue section, which may result in misleading or confusing molecular findings,3 complicates these issues. Therefore, a reproducible method for obtaining homogeneous cell populations from normal tissue or from cancer biopsies was required in order to obtain accurate information from molecular analyses. Laser capture microdissection (LCM) was initially conceived by a team of investigators at the National Institutes of Health, led by Lance Liotta, Robert Bonner, and Michael Emmert-Buck, to address this need.4,5 LCM provides a rapid and direct method for procuring homogeneous subpopulations of cells or complex structures for biochemical and molecular biological analyses. Arcturus Engineering, Inc. (Mountain View, CA) developed the first commercial LCM instrument, made available in 1997, in collaboration with the NIH group as part of a Cooperative Research and Development Agreement (CRADA). 1
J. L. Wittliff, R. Pasic, and K. I. Bland, in “The Breast: Comprehensive Management of Benign and Malignant Diseases” (K. I. Bland and E. M. Copeland III, eds.), p. 458. W. B. Saunders Co., Philadelphia, 1998. 2 H. Igarashi, H. Sugimura, K. Maruyama, Y. Kitayama, I. Ohta, M. Suzuki, M. Tanaka, Y. Dobashi, and I. Kino, APMIS 102, 295 (1994). 3 K. A. Cole, D. B. Krizman, and M. R. Emmert-Buck, Nat. Genet. 21, 38 (1999). 4 M. R. Emmert-Buck, R. F. Bonner, P. D. Smith, R. F. Chuaqui, Z. Zhuang, S. R. Goldstein, R. A. Weiss, and L. A. Liotta, Science 274, 998 (1996). 5 R. F. Bonner, M. R. Emmert-Buck, K. Cole, T. Pohida, R. Chuaqui, S. Goldstein, and L. A. Liotta, Science 278, 1481 (1997).
14
BASIC PRINCIPLES
[2]
FIG. 1. Components of a laser capture microdissection instrument. The LCM station integrates a research grade inverted microscope, a low-power infrared laser, a joystick-controlled stage, and a custom CapSure LCM cap handling mechanism (cassette module and placement arm). The LCM employs a video camera connected to an image archiving unit (not shown) for annotation, storage, and review of the microdissection process.
Laser Capture Microdissection Instrumentation LCM represents a major advancement in nondestructive cell sampling technology that can be applied to genomic and proteomic studies. Studies conducted in our laboratories utilize the PixCell II LCM System (Arcturus Engineering, Inc.) composed of the LCM instrument with fluorescence microscopy, the CapSure Transfer Film Carrier, and the PixCell II Image Archiving Workstation (Fig. 1). Briefly, the LCM station integrates a research-grade inverted microscope, a lowpower infrared laser, a joystick-controlled stage, and a custom CapSure LCM cap handling mechanism with a video monitor and controller. Protocol for Processing Human Tissue Specimens To evaluate differences between normal and diseased cells, one must first isolate the cells or structures by LCM (Fig. 2) and extract them independently for DNA, RNA, or protein analyses.6,7 Proper tissue procurement, specimen handling, and cryopreservation are essential for the collection of quality information from these analyses.1 Briefly, biopsy specimens should be excised expeditiously and without trauma during the surgical procedure. Specimens must be chilled on ice, 6
N. L. Simone, R. F. Bonner, J. W. Gillespie, M. R. Emmert-Buck, and L. A. Liotta, Trends Genet. 14, 272 (1998). 7 J. L. Wittliff, S. T. Kunitake, S. S. Chu, and J. C. Travis, J. Clin. Ligand Assay 23, 66 (2000).
[2]
LASER CAPTURE MICRODISSECTION
15
FIG. 2. Sequence for LCM procurement of cells from a complex tissue section. After tissue collection, fixation, staining, and dehydration as described in the text, cells of interest are located and the CapSure optically transparent cap is placed on the tissue. A laser pulse releases the cell from surrounding structures transferring it to the thermoplastic film. The intact cell bound to the CapSure device is lifted and placed onto a standard 500 µl microcentrifuge tube for subsequent extraction and analysis.
and then well trimmed of necrotic tissue, leaving normal tissue present with the lesion in question. The tissue specimen should either be frozen on dry ice in the pathology suite within 20–30 min of collection or rapidly transported chilled in a petri dish or plastic bag immersed in ice prior to cryopreservation and frozen section preparation in the LCM laboratory to retain the biological integrity of macromolecules. Any procedure avoiding RNase contact is desirable. It is preferable to freeze the specimen immediately on dry ice after collection at the time of frozen section diagnosis if studies requiring RNA are to be conducted. With the advent of LCM and sensitive technologies of genomics and proteomics requiring nondestructive isolation of pure cell populations, new surgical pathology
16
BASIC PRINCIPLES
[2]
approaches and methods must be developed as recommended by Cole et al.3 and Wittliff et al.7 Specimens are processed according to accepted biohazard policies in clean rooms prepared to reduce RNase contamination. Specimens are frozen in an optimum cutting temperature compound (TissueTek OCT medium, VWR Scientific Products Corp.) and stored at −86◦ until LCM is performed. At that time, frozen sections are collected on sterile microscope slides without a coverslip and retained frozen by being placed on a flat surface of dry ice to preserve labile macromolecules. We recommend that glass slides without coatings (uncharged) be used to enhance LCM of selected cells. Frozen tissue biopsies or tissue sections collected on slides and stored in sterile plastic slide holders may be shipped to a distant laboratory for LCM analyses if the specimens are retained on dry ice during transfer. Preservation of the biological integrity of the biopsy tissue prior to arrival in the LCM laboratory is the shared responsibility of the pathologist and the surgeon, if proteomic and genomic analyses are to become routine clinical tests. In addition, sections of the tissue procured must be representative of the lesion. Fixation, Staining, and Dehydration Frozen sections mounted on uncoated glass slides are handled according to the following procedures depending on the type of staining reagent used. The intercalating dye, TO-PRO-3 (Molecular Probes, Inc., Eugene, OR), which binds tightly to double-stranded nucleic acids and exhibits a peak fluorescence at 661 nm, has been used to assess the integrity of DNA in LCM procured cells and structures.7 TO-PRO-3 Staining Protocol (1) Place frozen section in 70% ethanol for 1 min, (2) transfer to PBS for 30 sec, (3) place slide in tray and stain with 10 µM TO-PRO-3 for 2 min, (4) transfer to PBS for 2 min, (5) transfer to deionized water for 30 sec, (6) transfer to 70% ethanol for 30 sec, (7) transfer to 95% ethanol for 30 sec, (8) transfer to 100% ethanol for 30 sec, (9) transfer to xylene for 5 min, and (10) air dry for 20 min; store desiccated. Phosphate-buffered saline (PBS) and deionized water are used in all wash steps. All of the steps utilizing ethanol employed ethyl alcohol UPS (Aaper Alcohol and Chemical Co., Shelbyville, KY). H&E Staining Protocol (1) Place frozen section in 70% ethanol for 1 min, (2) transfer to hematoxylin Gill No. 3 (Sigma Diagnostics, St. Louis, MO) for 30 sec, (3) transfer to RNasefree water for 30 sec, (4) transfer to bluing agent (ThermoShandon, Pittsburgh, PA) for 30 sec, (5) transfer to 70% ethanol for 30 sec, (6) transfer to 70% alcohol-based eosin Y alcoholic (ThermoShandon, Pittsburgh, PA) for 30 sec, (7) transfer to 70%
[2]
LASER CAPTURE MICRODISSECTION
17
FIG. 3. Representative images of human breast carcinomas stained with H & E. (A) Typical human breast carcinoma showing infiltration of cancer cells into stromal elements; (B) breast carcinoma adjacent to a large population of inflammatory cells; (C) tissue specimen of DCIS with protein secretions; (D) breast carcinoma biopsy exhibiting freezing artifact.
ethanol for 30 sec, (8) transfer to 95% ethanol for 30 sec, (9) transfer to 100% ethanol for 30 sec, (10) transfer to xylene GR/ACS (EM Science, Gibbstown, NJ) for 5 min, and (11) air dry for 20 min; store desiccated. Desiccate only if slide will not be used for RNA extraction. Tissue slides to be used for total RNA extraction and gene expression profiling must be used within 1–2 hr for LCM procurement of cells. Prior to LCM, we evaluate the structural status of the frozen tissue biopsy after sectioning and H & E staining (Fig. 3). As illustrated in Fig. 3A, the section indicates the biopsy is acceptable for proceeding with LCM and gene expression profiles. The section shown in Fig. 3B also indicates an acceptable specimen but considerable caution must be exercised to avoid removing unwanted inflammatory cells with carcinoma cells. The tissue section shown in Fig. 3C illustrates that the specimen received in the laboratory contained considerable areas of DCIS although
18
BASIC PRINCIPLES
[2]
the diagnosis indicated invasive ductal carcinoma. This illustrates the value of pathology confirmation on the portion of the tissue biopsy received in the LCM laboratory. Finally, the tissue section shown in Fig. 3D depicts significant freezing artifact indicating the biopsy was unsatisfactory for LCM and gene expression profiling. Prior to LCM, other types of tissue preparations were utilized. These include either formalin-fixed or alcohol-fixed sections that are paraffin-embedded, as well as cytospin preparations of cells from blood or ascites fluids. Fixation conditions are dictated by the nature of the antigen of interest. Precipitative reagents such as acetone and methanol are used for intracellular antigens while cross-linking fixatives such as glutaraldehyde or paraformaldehyde are used for cell-surface antigens.8 Immunohistochemistry of protein analytes has been performed to guide cell selections by LCM (e.g., Fend et al.9 ). We routinely perform immunohistochemistry of clinically relevant analytes such as estrogen and progestin receptors, EGF receptors, and HER-2/neu oncoprotein to direct the procurement of cells expressing particular tumor markers.1 Steps in Laser Capture Microdissection The sequence of tissue collection, cell procurement by LCM, and macromolecular extraction is depicted in Fig. 4. Avoid the presence of moisture (e.g., frost, fingertips, breath, room humidity) during all steps prior to RNA extraction. Because our LCM laboratories are used for proteomic and gene expression studies, all procedures are conducted under RNase-free conditions, including cleaning of the stage and related areas of the LCM instrument and surrounding bench with RNase AWAY (Molecular BioProducts, San Diego, CA). Gloves and lab coats are worn at all times. Prior to performing LCM, the joystick should be positioned perpendicular to the bench and the CapSure LCM cap should be placed over the tissue under examination. The operator locates the cell or structure to be microdissected from the tissue section by viewing the histology on the monitor of the PixCell II LCM System.7 After test firing the IR laser in an area devoid of cells and observing the features of the melted plastic ring, the settings for power and duration are adjusted to obtain the desired spot size. Typically if one is using CapSure HS LCM caps, the following adjustments are suggested: spot size of 7.5 µm (power setting = 65–75 mW; duration setting = 650–750 µs), spot size of 15 µm (power setting = 35–45 mW; duration setting = 2.5–3.0 ms), spot size of 30 µm (power 8
S.-R. Shi, J. Gu, and C. R. Taylor, “Antigen Retrieval Techniques: Immunohistochemistry and Molecular Morphology.” Eaton Publishing, Natick, MA, 2000. 9 F. Fend, M. R. Emmert-Buck, R. Chuaqui, K. Cole, J. Lee, L. A. Liotta, and M. Raffeld, Am. J. Pathol. 154, 61 (1999).
[2]
LASER CAPTURE MICRODISSECTION
19
FIG. 4. Simple four-step process to capture cells and recover macromolecules. After location of the cells of interest, a CapSure or CapSure HS LCM cap is placed over the target area. Pulsing the laser through the cap activates the thermoplastic film to form a thin protrusion that bridges the gap between the cap and tissue and adheres to the target cell. Lifting of the cap removes the target cell(s) now attached to the cap. Macromolecules may be extracted from 200–1000 cells using the ExtracSure Sample Extraction Device which accommodates small volumes.
setting = 45–55 mW; duration setting = 6.0–7.0 ms). If CapSure LCM caps are employed, the following adjustments are suggested: spot size of 7.5 µm (power setting = 40–50 mW; duration setting = 550–650 µs), spot size of 15 µm (power setting = 30–40 mW; duration setting = 1.5–2.0 ms), spot size of 30 µm (power setting = 25–35 mW; duration setting = 5.0–6.0 ms). In order to correctly return to the initial area of cell capture for making photo records of
20
BASIC PRINCIPLES
[2]
“After” and “Cap” images (Fig. 5), the authors “mark” the region by placing several spots on areas devoid of cells adjacent to the cells of interest. This is particularly helpful when collecting multiple caps of cells from distant regions in the same tissue section for later comparison of proteomic and genomic analyses. Using the image archiving unit, characteristics of the tissue section are recorded before and after LCM, as well as those of the cells procured on each cap (Fig. 5). The cells or structures are microdissected after firing the IR laser and lifting the CapSure LCM cap with the intact cells collected on the transfer film. The CapSure and CapSure HS consist of a proprietary thermoplastic polymer film hermetically sealed to the bottom of a precision optical grade plastic cap. In certain experiments requiring extraction of small numbers of cells (200–1000) in low microliter volumes, we utilize the ExtracSure Sample Extraction Device (Arcturus) and the CapSure HS LCM caps for efficient removal of total RNA. The CapSure LCM caps containing the cells fit directly onto standard reagent tubes (500 µl Eppendorf) in preparation for cell extraction. Typically, 1–6 ng of total RNA may be extracted in this manner using Buffer RLT (Qiagen, Valencia, CA). The transfer process does not damage the captured cells or the surrounding cells remaining on the slide containing the original tissue preparation (Fig. 5). Usually there is no undesirable cellular contamination since the IR laser beam may be focused between 7.5 and 30 µm providing accurate selection. If necessary, we employ the CapSure pads to remove debris (e.g., stromal elements) from the CapSure LCM caps prior to extraction. Forces involved in an efficient LCM manipulation include (a) those between tissue and slide, (b) those between tissue and activated film, (c) tissue–tissue interactive forces, and (d) the force between tissue and inactivated film. The dynamics of the IR focusing and the melting properties of the thermoplastic transfer film on the CapSure LCM caps are optimized with those of cells in 5- to 10-µm tissue sections. After collection of cells on the CapSure LCM cap, macromolecules are extracted using a variety of procedures depending on whether the analyses are focused on DNA, RNA, or protein, as described in other chapters of this volume. Gene expression as measured by analyses of mRNA provides an understanding of the manner in which normal cells respond to endocrine changes, malignant transformation, and environmental insults.6,7,10–13 Determination of the level of gene 10
A. Glasow, A. Haidan, U. Hilbers, M. Breidert, J. Gillespie, W. A. Scherbaum, G. P. Chrousos, and S. R. Bornstein, J. Clin. Endocrin. Metab. 83, 4459 (1998). 11 L. Luo, R. C. Salunga, H. Guo, A. Bhittner, K. C. Joy, J. E. Galindo, H. Xiao, K. E. Rogers, J. S. Wan, M. R. Jackson, and M. G. Erlander, Nat. Med. 5, 117 (1999). 12 D. C. Sgroi, S. Teng, G. Robinson, R. LeVangie, J. R. Hudson, Jr., and A. G. Elkahloun, Cancer Res. 59, 5656 (1999). 13 S. M. Goldsworthy, P. S. Stockton, C. S. Trempus, J. F. Foley, and R. R. Maronpot, Mol. Carcinog. 25, 86 (1999).
[2]
LASER CAPTURE MICRODISSECTION
21
FIG. 5. Representative collection of human breast cancer cells by LCM. Three different regions of the same breast carcinoma biopsy are shown in the top three panels marked Before. The regions of the tissue section where carcinoma cells were removed by LCM are shown in the images marked After, and the isolated cells adhering to the CapSure LCM caps are shown in the images marked Capture. Each cap, containing 200–300 carcinoma cells, is extracted for RNA that is quantified and amplified before microarray analyses.
expression as well as the size and structure of RNA molecules requires retention of biological integrity. Because of the lability of mRNA, several workers have studied the effects of tissue fixation on RNA extraction and amplification after LCM,9,13 providing some insight into the stability of these labile molecules using current procedures. Advantages of LCM Manual microdissection techniques, which require tedious manipulation, significant manual dexterity, and a lengthy training program, are slow and the variability in tissue collection is significant. However, LCM, which uses standardized
22
BASIC PRINCIPLES
[2]
technology, allows rapid sample procurement of tissue structures with awkward geometry and efficient isolation of different cell types in close proximity or adjacent to each other.2–4,7 Furthermore, the transfer process is nondestructive and cell morphology is retained.7 Of particular importance in molecular diagnostics and gene discovery, there is a record of the original location of cells in the tissue and visual verification of cell capture. Some investigators have reported successful DNA analyses using 300–500 cells (e.g., Simone et al.,6 ; Sirivatanaukorn et al.14 ), while 500–1000 cells have been used to isolate RNA (e.g., Glasow et al.10 ; Luo et al.11 ; Goldsworthy et al.13 ). Examinations of proteins using a single technology have employed 1000– 5000 cells isolated by LCM (e.g., Banks et al.15 ; Emmert-Buck et al.16 ). Extraction and 2D PAGE of proteins from representative samples requires capturing 20,000–30,000 cells although new nanotechnology approaches are being developed.15,16 RNA Isolation, Characterization, and Amplification for Microarray In our laboratories, total RNA is isolated using the PicoPure (Arcturus) kits, which are optimized for extracting RNA from cells procured by LCM. Routinely 1–6 ng of total RNA may be isolated from 200–300 human breast cancer cells procured by LCM, using these reagents. The intactness of RNA in tissue sections is evaluated prior to proceeding with LCM by a variety of procedures including electrophoresis incorporating a series of markers of different base-pair lengths. For investigations of gene expression profiles of human tissues, we procure cells of interest (e.g., normal vs neoplastic) from at least three different regions of a single tissue section (Fig. 5). Note that the carcinoma cells were removed from each of the three regions of interest and procured cleanly and retained on the CapSure LCM caps (Fig. 5, lower images). Each cell capture (usually containing 200–1000 cells) is treated as an independent evaluation in that the RNA is extracted, purified, and amplified, then subjected to microarray (Fig. 6). RNA isolated from cells procured by LCM is amplified efficiently with the RiboAMP kits (Arcturus) enabling the production of microgram amounts of RNA from nanogram quantities isolated from breast carcinoma and normal cells. Amplification requires preparation of double-stranded cDNA from the mRNA fraction of total RNA followed by transcription in vitro. The use of exogenous primers maximizes reliability in the synthesis of cDNA template while reducing reaction 14
Y. Sirivatanaukorn, V. Sirivatanauksorn, S. Bhattacharya, B. R. Davidson, A. P. Dhillon, A. K. Kakkar, R. C. N. Williamson, and N. R. Lemoine, J. Pathol. 189, 344 (1999). 15 R. E. Banks, M. J. Dunn, M. A. Forbes, A. Stanley, D. Pappin, T. Naven, M. Gough, P. Harnden, and P. J. Selby, Electrophoresis 20, 689 (1999). 16 M. R. Emmert-Buck, J. W. Gillespie, C. P. Paweletz, D. K. Ornstein, V. Basrur, E. Appella, Q. H. Wang, J. Huang, N. Hu, P. Taylor, and E. F. Petricoin III, Mol. Carcinog. 27, 158 (2000).
[2]
LASER CAPTURE MICRODISSECTION
23
FIG. 6. A representative Eisen Color Map of the gene expression profiles of various human breast carcinomas. Although the microarray performed contained more than 12,000 genes, only a portion of the gene expression profile of each breast cancer is shown using the GeneMaths program (Applied Maths, Austin, TX). Note that without preconceived selection of criteria, gene clustering was observed. Through preliminary bioinformatic analyses, molecular signatures are being identified for several types of human breast cancers, such as those expressing estrogen receptors (ER+) compared with carcinomas lacking the receptor (ER−), which is a marker of anti-estrogen responsiveness. Principal component analysis (diagram on right) was performed using the data matrix shown on the left, and the collective results of the breast specimens are projected onto the three-dimensional space diagram using the first three components.
times. The aRNA prepared by this protocol is ready for labeling and hybridization necessary for microarray analyses. Preliminary studies of microarray analyses of independent amplifications from the same RNA preparation indicate an excellent correlation. RT-PCR was used to detect low-, medium-, and high-abundance genes within the amplified RNA population. Our laboratory has demonstrated that
24
BASIC PRINCIPLES
[2]
amplification of mRNA in all abundance classes ensures that differential gene expression patterns will be identified. Currently, we are using a microarray containing approximately 12,000 genes of which 10% were included through KnowledgeBased Selection based on reported alterations in cancer. From studies of more than 100 human breast cancers, we have demonstrated that the RNA isolated from LCM procured cells is intact for use in amplification of mRNA and subsequent microarray (Fig. 6). We are employing this approach to derive molecular signatures (gene expression profiles) to advance the classification of breast cancer and assessment of patient prognosis and therapeutic response. Additional Applications of Laser Capture Microdissection LCM is rapidly becoming the method of choice for selecting diseased cells from normal cells of the same tissue specimen for genomic and proteomic analyses. 7,10–16 Some of the applications of LCM in these areas related to molecular diagnostics and prognostics of human cancer are shown below. Genomics: Differential gene profiling Loss of heterozygosity Micro-satellite instability Gene quantification Mutation/clonal analysis Proteomics: Two-dimensional PAGE Western blots Immunoquantitation of proteins MALDI-TOF mass spectrometry The ability to procure homogeneous cell subpopulations of normal, premalignant, and malignant cell types and to accumulate data from each cell type advances our understanding of the underlying causes of tumor formation and permits the tracking, at the molecular level, of cell progression into a metastatic phenotype. Efforts are well underway to produce cDNA libraries that catalog genes differentially expressed during tumor progression (e.g., Peterson et al.17 ). The Cancer Genome Anatomy Project (CGAP) has utilized LCM to obtain normal and premalignant samples from human prostate, breast, ovary, colon, kidney, and 17
L. A. Peterson, M. R. Brown, A. J. Carlisle, E. C. Kohn, L. A. Liotta, M. R. Emmert-Buck, and D. B. Krizman, Cancer Res. 58, 5326 (1998).
[3]
LIVE CELL LASER MICRODISSECTION
25
lung tissue, to name a few. Information from CGAP is publicly available through the CGAP-NIH Web site.18 Laser capture microdissection has proved to be a powerful tool for research into the cellular basis of disease and is increasingly being employed in drug discovery and clinical diagnostics. Physiological changes occurring during development and progression of normal cells to neoplastic lesions may be explored easily with LCM and proteomics and gene expression profiling. For clinical diagnosis, the ability to sample specific types of cells creates a new analytical paradigm which will allow patients to be diagnosed based on qualitative and quantitative gene expression as well as on levels of cell-specific proteins. As Wittliff suggested previously,1 a new generation of laboratory tests is rapidly evolving in which analyses will be performed directly on human tissue biopsies. It is envisioned that tissue banks such as the Biorepository at the Hormone Receptor Laboratory will be developed for long-term preservation of human tumor samples. This will allow assessment of genetic and biochemical properties of the stored tumor tissues as new clinical, chemical, and molecular biological probes are developed for cancer management, and as technologies such as laser capture microdissection are utilized to separate normal from tumor cells. 18
www.ncbi.nlm.nih.gov/CGAP
[3] Going in Vivo with Laser Microdissection ¨ By ANETTE MAYER, MONIKA STICH, DIETER BROCKSCH, KARIN SCHUTZE , and GEORGIA LAHR Introduction Tissue microdissection and single-cell isolation is one of the most advanced techniques in modern gene analysis and is especially useful for studying expression of genes in isolated tumor cells. Till now, microdissection methods have been limited to cells from fixed or frozen tissues.1–9 An old dream of cell biologists 1
W. Meier-Ruge, W. Bielser, E. Remy, F. Hillenkamp, R. Nitsche, and R. Uns¨old, Histochem. J. 8, 387 (1976). 2 M. Schindler, M. L. Allen, M. R. Olinger, and J. F. Holland, Cytometry 6, 368 (1985). 3 Y. Kubo, F. Klimek, Y. Kikuchi, P. Bannasch, and O. Hino, Cancer Res. 55, 989 (1995). 4 M. R. Emmert-Buck, R. F. Bonner, P. D. Smith, R. F. Chuaqui, Z. Zhuang, S. R. Goldstein, R. A. Weiss, and L. A. Liotta, Science 274, 998 (1996). 5 R. F. Bonner, M. Emmert-Buck, K. Cole, T. Pohida, R. Chuaqui, S. Goldstein, and L. A. Liotta, Science 278, 1481 (1997).
METHODS IN ENZYMOLOGY, VOL. 356
Copyright 2002, Elsevier Science (USA). All rights reserved. 0076-6879/02 $35.00
26
BASIC PRINCIPLES
[3]
is to isolate living cells from tissue culture or unfixed and unfrozen sections of living tissues.10 But to date, laser-based microdissection of living cells resulted in the destruction of the isolated cells.2 Now, a modification of the laser microbeam microdissection (LMM) method in combination with the laser pressure catapulting (LPC) technique6 and a newly developed cell culture protocol allows microdissection and “ejection” of living single cells or cell clusters with ongoing cultivation for potential treatment and analysis. We established a unique technique in which cultured cells were microdissected and afterward catapulted by LPC into the cap of a microfuge tube. The viability of the catapulted cells is not affected as they enter the cell cycle and proliferate. Applying this protocol—select, microdissect, eject, and clone living cells—to biopsy slices will come true in the near future. As this, “going in vivo” opens up a broad spectrum of applications. Step I: Cell Culture Preparation A prerequisite for isolation of single living cells from cell cultures by laserassisted cell picking is the growth of the cells on a supporting membrane. The membrane is mounted in a specific cell culture chamber, the ROC chamber. For microdissection the membrane around the cell or cell clusters of interest is cut by the focused laser beam in a sufficient distance from the cell. Then the cell-membrane stack is catapulted by the laser beam into a conventional cap of a microfuge tube centered directly above the selected area (Fig. 1, A and B). Buffers, Reagents, and Equipment ROC chamber, Round Open Closed (PeCon and LaCon, Erbach-Bach, Germany) Polyethylene–naphthalene membrane, 1.35 µm (PEN membrane; P.A.L.M. Microlaser Technologies AG, Bernried, Germany) EJ28 cells, a bladder carcinoma cell line TPC-1 cells, a thyroid carcinoma cell line Dulbecco’s modified Eagle’s medium [Invitrogen GmbH (GIBCO-BRL), Karlsruhe, Germany] Dulbecco’s modified Eagle’s medium Nutrient Mixture F12-Ham (DME/F12 Hams) (Sigma-Aldrich GmbH, Deisenhofen, Germany) 200 mM L-glutamine (Sigma-Aldrich GmbH) 6
K. Sch¨utze and G. Lahr, Nat. Biotechnol. 16, 737 (1998). K. Sch¨utze, H. P¨osl, and G. Lahr, Mol. Cell Biol. 44, 735 (1998). 8 G. Lahr, Lab. Invest. 80, 1 (2000). 9 G. Lahr, M. Stich, K. Sch¨ utze, P. Bl¨umel, H. P¨osl, and W. B. J. Nathrath, Pathobiology 68, 218 (2000). 10 M. Schindler, Nat. Biotechnol. 16, 719 (1998). 7
[3]
LIVE CELL LASER MICRODISSECTION
27
FIG. 1. Schematic drawings of a cell culture grown on a PEN membrane in an ROC chamber. The chamber is attached to the microscope stage and the microfuge cap is centered above the line of laser fire directly in the ROC chamber (A). The selected cell-membrane stacks are microdissected by the laser beam (LMM). The cell-membrane stacks are catapulted by LPC directly into the cap of the sample tube supplied with a droplet of Hanks’ solution (B). The captured cells are covered with 25 µl of Hanks’ solution (C). The cap is topped with the remaining tube and the assembled tube is centrifuged to collect captured cells at the bottom of the microfuge tube (D).
10% Fetal calf serum (FCS; Sigma-Aldrich GmbH) 100× Antibiotic–antimycotic solution (Sigma-Aldrich GmbH) Hanks’ solution (Sigma-Aldrich GmbH) Trypsin–EDTA solution (Sigma-Aldrich GmbH) Conventional culture dish plates for cell culture Gassed incubator for cell culture Laser microscope, Robot-MicroBeam (P.A.L.M. Microlaser Technologies AG) Inverted microscope Axiovert 135 (Carl Zeiss, G¨ottingen, Germany) Microfuge tubes (P.A.L.M. Microlaser Technologies AG) Procedure Assembly of the ROC Chamber 1. Cover the glass bottom of the ROC chamber with the polyethylene– naphthalene membrane (PEN membrane) by using a droplet of 100% ethanol for mounting it onto the glass. 2. Expose the opened chamber with the membrane to UV light for 20 min to change the hydrophobic nature of the membrane into a more hydrophilic one. 3. Assemble the ROC chamber totally and autoclave it at 121◦ for 20 min.
28
BASIC PRINCIPLES
[3]
Cell Culture. The EJ28 bladder carcinoma cell line and the papillary thyroid tumor cell line TPC-1 (a generous gift of Dr. B. Mayr, Med. Hochschule Hannover, Germany) were used for the experiments. EJ28 cells were grown in Dulbecco’s modified Eagle’s medium and TPC-1 cells were grown in Dulbecco’s modified Eagle’s medium Nutrient Mixture F12-Ham (DME/F12 Hams), both supplemented with 5 mM L-glutamine, 10% fetal calf serum (FCS), and 1× antibiotic– antimycotic solution. Seed the cell culture cells at the desired density onto the membrane-covered ROC chamber in their appropriate medium. 4. Incubate the cells in the ROC chamber at 37◦ in a gassed incubator. After 1–2 days in culture the cells are ready for microdissection. Laser Microdissection and Catapulting 5. Remove the medium completely from the ROC chamber before laser microdissection (Figs. 1–4A). 6. Microdissect the desired cell-membrane sample. The parameters concerning laser energy and laser focus during microdissection (LMM) are dependent on the laser microscope system used and have to be optimized before use (Figs. 1A, 2B, 3B, 4C, 5B). 7. Apply a 10-µl droplet of Hanks’ solution on top of the selected cells to facilitate LPC. Be careful not to wash away the microdissected specimen.
FIG. 2. Images using LMM and LPC to capture 40 EJ28 cells. Cells before microdissection (A), after microdissection (B), cells remaining after LPC (C), catapulted membrane with the cells (D). 11 hr after plating (E), after 1 day (F), after 5 days (G), after 8 days (H), and after 12 days (I). Black arrow: cell filopodium. White arrow: mitotic cell. Dotted line: area to be microdissected. Bar equals 100 µm in A–D, F, and H; 50 µm in E, G, and I. Objective lenses in A–D and F: 20× ; E and I: 40× ; G: 5×; and H: 10×.
[3]
LIVE CELL LASER MICRODISSECTION
29
FIG. 3. LMM and LPC images of a small TPC-1 cell cluster (10 cells). The sequence shows cells before LMM (A), after LMM (B), and remaining cell culture after LPC (C). (D) Aggregated cells within the “hanging droplet” 9 hr after collection. 1 day in culture the catapulted and aggregated cells begin to adhere to the bottom of the culture dish (E). Proliferating cells shown after 12 days in culture (F). Dotted line: area to be microdissected. Bar equals 50 µm in A–C; 100 µm in D–F. Objective lenses in A–C: 40×; D–F: 20×.
8. Pipette a 5-µl droplet of Hanks’ solution into the center of the cap of a microfuge tube and place the cap directly above the selected cells into the ROC chamber (Fig. 1A). 9. Catapult the cell-membrane stack with one single laser shot positioned at the border of the circumscribed membrane. For the catapulting the laser is focused below the microdissected target specimen. 10. Energy settings should be sufficiently high to catapult the microdissected cells with the membrane into a cap (Fig. 1, A and B). Even large cell-membrane stacks (for example 385× 248 µm) can be catapulted (Fig. 2C).
FIG. 4. Images of an experiment to destroy an “undesired cell” before LMM and LPC of about 18 living EJ28 cells. The sequence shows the cells before microdissection(A), after destruction of one specific cell (B), and after microdissection (C), the remaining cells after LPC (D), and the catapulted membrane with the cells (E). (F) Cells after 12 days in culture. Black arrow: cell destroyed by a precise laser shot. White arrow: cells on the membrane after catapulting. Dotted line: area to be microdissected. Bar equals 100 µm in A–E; 50 µm in F. Objective lenses in A–E: 20×; F: 10×.
30
BASIC PRINCIPLES
[3]
FIG. 5. Images of 5 pooled single EJ28 cells. The sequence shows the cells before LMM (A), after LMM of cell group 1 (B), after LMM of cell groups 1 and 2 (C), after LMM of cell groups 1 to 3 (D). The remaining cells after LPC of cell group 1 (E), 1 and 2 (F), and after LPC of cell group 3 (G). (H) 3 catapulted membranes with cells. Aggregated cells within the “hanging droplet” (I). Dotted lines: areas to be microdissected. Bar equals 100 µm in A–H; 50 µm in I. Objective lenses in A–H: 20×; I: 40×.
11. After LPC remove the ROC chamber from the microscope stage, take the cap, and inspect the catapulted cells in the cap now fixed within the manipulator (Figs. 2D, 4E, 5H). Notes to Step I a. To reduce the chance of contamination wear gloves during the whole cell culture procedure. Do not keep the cell cultures outside the incubator longer than necessary. In case of several experiments allow cells in the ROC chamber to recover from dryness by adding medium back to the cells. This medium has to be removed totally, otherwise during the cutting process the laser energy will be absorbed by the aqueous solution. This results in local heating of the medium and in visible steam bubbles, which destroy the viable cells. In addition, be aware that after several microdissection events the medium is entering the micro space between the membrane and the glass bottom of the ROC chamber. This makes further microdissection and catapulting more and more difficult and finally impossible. b. With the focused laser beam single cells or cell clusters are precisely separated together with the membrane from their surrounding (Figs 2B, 3B, 4C,
[3]
LIVE CELL LASER MICRODISSECTION
31
5B–5D) using 20× (Figs. 2, 4, 5) or 40× (Fig. 3) objective lenses. Laser circumscription of the cell-membrane stack results in a gap, free of any other biological material, separating the target from its surroundings.6 The width of the gap is about 3–5 µm, depending on the objective lens and the absorption behavior of the specimen. Where the selected specimen area contains an undesired cell, this cell can be eliminated by a direct laser shot (Fig. 4B). c. Increased laser energy catapults the target specimen into the cap of a microfuge tube. Even large cell-membrane stacks (for example 290× 369 µm) can be catapulted (Figs. 2D, 4E, 5H). This results in empty patches within the cell culture (Figs. 2C, 3C, 4D, 5G). The laser-catapulted cell-membrane stacks are well preserved and allow direct correlation with their templates in terms of shape, size, and original position (Figs. 2D, 4E, 5H). Microdissection and catapulting of cell clusters in these large sizes takes less than 2 min. The manipulation of single living cells is done within seconds. Laser-assisted isolation is performed with cell clusters of about 10 cells (Fig. 3) and several tens (Figs. 2 and 4), as well as single cells (Fig. 5).
Step II: Collection of Catapulted Cells Procedure 1. Cover the catapulted cells in the cap with 25 µl of Hanks’ solution. 2. Close the cap with the remaining tube and store for up to 30 min at room temperature. 3. Centrifuge the tube for 1 min at 8000g and discard the supernatant. 4. Resuspend the cells in 20 µl trypsin–EDTA solution and incubate for 10 min at room temperature to detach the cells from the membrane. 5. Centrifuge for 1 min at 8000g. 6. After centrifugation the trypsinized cells (pellet) are resuspended in 15 µl (<10 cells) or 20 µl of supplemented medium (4 parts medium: 1 part conditioned medium). Note: If there are fewer than 10 catapulted cells in the tube follow protocol step 7; if there are 10 or more cells in the suspension proceed with step III. 7. With fewer than 10 cells in the suspension form a so-called “hanging droplet”: place the 15 µl drops inside the lid of a culture dish. 8. Turn the lid and mount it on the remaining culture dish. (This results in “hanging droplets” where single cells come in close contact and start aggregation; Figs. 3D and 5I.) Incubate the cells within the “hanging droplet” overnight at 37◦ in a gassed incubator. 9. The next day flip the lid around again and add 5 µl of supplemented medium to the droplet (total: 20 µl) and proceed with step III.
32
BASIC PRINCIPLES
[3]
Step III: Recultivation of Catapulted Cells Procedure 1. Place the 20-µl droplets of the cell suspension with a sterile micropipette onto the bottom of a culture dish, either alone or with sufficient space from other droplets. 2. Incubate overnight at 37◦ . 3. 24 hr later mark the droplets on the bottom of the culture dish with a permanent pen. This facilitates tracing of the cells. 4. Remove the medium and fill the culture plate with 4–10 ml of fresh supplemented medium, depending on the diameter of the culture dish. 5. The growing of the cells can be observed during next days (Fig. 2F) and weeks (Figs. 3F and 4F). A regular change of cell culture medium is recommended. Notes to Step III a. Because of the round shape of a droplet the aggregated (Figs. 3D, 5I) or the nonaggregated cells (Fig. 2) are forced to move in the tension-free center of the droplet (Figs. 3E, 4F, 2F, respectively). After about 3 hr these cells begin to spread out on the culture dish now showing a more flattened cell shape with extending filopodia (Fig. 2E). After 1 day in culture the cells show the typical flat shape of epithelial cells (Figs. 2F, 3E, 4F). b. Depending on the density of the plated cells, after 3 days in culture most cells enter the cell cycle (more than 30 cells) and start proliferation (see Figs. 2G–2I). Fewer than 10 cells need more time before they proliferate (Fig. 3F). In our hands this took more than 3 weeks. Proliferating EJ28 cells show typically round-shaped cells sitting on top of the flat cell layer (Figs. 2H, 2I). We were unable to get a “clone” from one single isolated cell, even with supplemented medium, but succeeded with single pooled cells. The cells were grown further and observed for up to 4 weeks.
Future Applications The new approach of the described microdissection protocol for live cells depends on the special procedure of preparation and recultivation of cells. Now the laser microdissection includes the processing steps “select-microdissect-ejectand-cultivate.” This opens the way for “going in vivo” with a broad spectrum of applications, for example, the establishment of homogeneous cell populations out of heterogeneous cells [i.e., after transfection experiments with green fluorescent protein (GFP) chimeras]. In addition, specific cells within a mixed cell population can be identified by their morphology, isolated, and cultivated separately. This protocol fulfills an old dream of cell biologists to isolate and recultivate living cells.
[4]
LASER CAPTURE MICRODISSECTION AND PROTEOMICS
33
The just-described method has been applied to cell cultures but can be extended in the future to tissue cultures and to unfixed unfrozen sections of living tissues, e.g., slice cultures from biopsies. Such isolated live cells can be examined, manipulated, and potentially cloned in an effort to provide more biological material for analysis and studies of normal and aberrant biological processes. Acknowledgments The authors thank H. P¨osl, R. Sch¨utze, and A. Starzinski-Powitz for technical assistance and instrumental and conceptional support. A. Mayer was supported by a grant from the Boehringer Ingelheim Stiftung given to A. Starzinski-Powitz, Goethe Universitaet Frankfurt am Main.
[4] Use of Laser Capture Microdissection to Selectively Obtain Distinct Populations of Cells for Proteomic Analysis By RACHEL A. CRAVEN and ROSAMONDE E. BANKS Introduction The proteome is the complement of proteins expressed by a tissue or cell type at a particular point in time.1,2 Technological advances have now permitted the realistic study of gene expression at the protein level leading to a rapid growth in proteomics-based research.3 Although proteomics cannot routinely compete with the throughput of extensive array-based mRNA profiling it does have significant advantages over studies at the mRNA level. First, it gives a more accurate representation of the level of gene product in a cell as it takes account of posttranscriptional controls of gene expression that result in a frequent lack of correlation between mRNA and protein levels. Second, it allows the study of posttranslational modifications, which are a fundamental aspect of protein function and often altered during disease progression. Identification of disease biomarkers that have diagnostic or prognostic value or molecules important in disease pathogenesis is often complicated by the heterogeneity of tissue samples. Development of techniques to allow the selective 1
V. C. Wasinger, S. J. Cordwell, A. Cerpa-Poljak, J. X. Yan, A. A. Gooley, M. R. Wilkins, M. W. Duncan, R. Harris, K. L. Williams, and I. Humphery-Smith, Electrophoresis 16, 1090 (1995). 2 M. R. Wilkins, J. C. Sanchez, A. A. Gooley, R. D. Appel, I. Humphery-Smith, D. F. Hochstrasser, and K. L. Williams, Biotechnol. Genet. Eng. Rev. 13, 19 (1996). 3 R. E. Banks, M. J. Dunn, D. F. Hochstrasser, J. C. Sanchez, W. Blackstock, D. J. Pappin, and P. J. Selby, Lancet 356, 1749 (2000).
METHODS IN ENZYMOLOGY, VOL. 356
Copyright 2002, Elsevier Science (USA). All rights reserved. 0076-6879/02 $35.00
34
BASIC PRINCIPLES
[4]
enrichment of relevant cell types and thereby enhance the information available from the direct use of biological samples in the study of disease pathogenesis has received a great deal of attention. Laser capture microdissection (LCM), which was developed by Lance Liotta and co-workers at NIH,4 and is now made commercially available by Arcturus Engineering, Inc. (Mountain View, CA), is one of several laser-assisted dissection strategies that allow direct selection of cell types from tissue without the need for enzymatic processing or growth in culture.5 In this approach stained tissue sections are visualized using an inverted microscope. A cap coated with a thermolabile film is placed in contact with the tissue section and local melting over selected areas of tissue is then induced by a laser. A beam of 7.5, 15, or 30 µm can be chosen, depending on the fineness of the particular dissection. Material fused to the cap is then specifically removed when the cap is lifted. Several thousand “shots” can be captured on a single cap thereby concentrating the desired cell types. Material can then be solubilized in an appropriate buffer for downstream analysis. The study of proteins in LCM-procured samples has been accomplished using a number of techniques that include two-dimensional polyacrylamide gel electrophoresis (2D PAGE) and surface enhanced laser desorption ionization mass spectrometry (SELDI) for global protein expression profiling and immunoassays for the study of specific proteins.6 A number of factors must be considered in the context of protein analysis of LCM procured samples. The aim of the approach is to obtain a more representative protein profile of the relevant cell type(s). This assumes that removal of contaminating cell type(s) will give selective enrichment detectable in the dissected sample relative to the starting material. This gain of information compared with analysis of whole tissue must warrant the time taken to generate the dissected sample. LCM relies on staining of tissue sections to allow visualization of tissue morphology and is therefore potentially subject to in vitro artifacts that result from tissue section processing. These potential alterations and more general aspects of sample stability and integrity must therefore be considered. Furthermore, in the absence of an amplification step for proteins, there are limitations resulting from the amount of sample available for analysis. The usefulness of LCM is therefore dependent on the tissue type being studied and the particular sample being used as well as on the molecule(s) being studied and the method of downstream analysis. Here we review the current status of LCM and proteomics as a research strategy, concentrating on the methods we have adopted for collection and analysis of laser capture microdissected samples, and discuss the limitations of this type of approach. 4
M. R. Emmert-Buck, R. F. Bonner, P. D. Smith, R. F. Chuaqui, Z. Zhuang, S. R. Goldstein, R. A. Weiss, and L. A. Liotta, Science 274, 998 (1996). 5 Y. Sirivatanauksorn, R. Drury, T. Crnogorac-Jurcevic, V. Sirivatanauksorn, and N. R. Lemoine, J. Pathol. 189, 150 (1999). 6 R. A. Craven and R. E. Banks, Proteomics 1, 1200 (2001).
[4]
LASER CAPTURE MICRODISSECTION AND PROTEOMICS
35
Sample Processing General Considerations Classically tissue samples are routinely fixed with formalin, an aqueous solution of formaldehyde, that functions by cross-linking proteins, thus precluding the subsequent analysis of proteins by electrophoresis and other separation techniques. Tissue must therefore be banked by an alternative method to be useful for protein analysis, other than at the immunohistochemical level. The protocol we use involves the cryopreservation of tissue, with sections being fixed with ethanol prior to staining. The morphology of these sections is inferior to that of paraffinembedded formalin fixed material, particularly with the inverted optics of the LCM system, but it is adequate for selection of tissue areas for microdissection and fully compatible with the downstream analysis of proteins. The choice of slides is also an important factor. For many applications involving relatively complex downstream processing of tissue such as antigen retrieval procedures, the adhesiveness of slides for the tissue sections is of paramount importance. Consequently positively charged slides or slides coated with poly-L-lysine or APES-formalin are often used. For LCM, there is a need to balance the retention of the tissue on the slide with the obvious need to allow “lift” of tissue areas following microdissection. For this reason we use ethanol-dipped plain glass slides. Sections are cut on the day of microdissection and kept on dry ice prior to use but are not stored longer term. Sections are defrosted and stained immediately prior to use. The thickness of the sections is another factor that affects lift of tissue, and we generally use 8-µm sections. Although 10-µm sections can be used, often they do not microdissect as cleanly. Another consideration in this context is the threedimensional nature of the tissue; depending on the complexity of the tissue being dissected thinner sections may need to be used to allow microdissection without cutting through the area of interest and collecting contaminating surrounding material. Staining protocols employed for LCM are generally more rapid than conventional methods and incorporate the use of protease inhibitors to minimize protein degradation. Hematoxylin and eosin is the most commonly used histological stain, and certainly for the complex structure of the kidney we find it the most useful. Alternative stains such as methyl green, methylene blue, and toluidine blue are also compatible with protein analysis.7,8 Rapid immunostaining protocols modified to minimize the chance of proteolysis can also be adopted. Conventional enzyme-based reactions run the potential risk of modifications to 7
L. C. Lawrie, S. Curran, H. L. McLeod, J. E. Fothergill, and G. I. Murray, Mol. Pathol. 54, 253 (2001). 8 R. A. Craven, N. Totty, R. P. Harnden, P. J. Selby, and R. E. Banks, Am. J. Path. 160, 815 (2002).
36
BASIC PRINCIPLES
[4]
proteins and so we have chosen to develop a rapid silver-enhanced gold immunolabeling procedure.8 Fluorescent-based methods such as those developed for use with subsequent RNA extraction9 provide an attractive alternative if fluorescence optics are available. Obviously the concentration of each antibody needs to be optimized individually, but, in general, we have found that because of the short incubations, antibodies have tended to be used at significantly higher concentrations than in more conventional protocols. A further consideration is the fixative used, as many antibodies do not work with ethanol fixation but rather require the use of acetone. This can cause problems with regard to protein recovery. It is essential, whatever staining protocol is adopted, that it be optimized for the tissue in question and that the effects on subsequent downstream analysis steps be minimal. Materials RPMI medium is from Life Technologies (Paisley, UK), Complete Mini Protease Inhibitor Cocktail tablets from Roche (Lewes, UK), PBS tablets from Oxoid (Basingstoke, UK), Tris from ICN (Basingstoke, UK), sodium iodate from Sigma-Aldrich (Poole, UK), and gold-conjugated antibodies for light microscopy and silver enhancing kit from British BioCell International (Cardiff, UK). All other chemicals are supplied by BDH (Poole, UK). Milli-Q H2O is used throughout. Tissue Collection The procedure described below has been optimized for kidney specimens rich in proteases and has also been successfully used for bladder, ovary, and cervix samples. It is likely that it will be applicable to the majority of tissues, but trial runs are advised. Immediately following removal from the patient, samples of tissue are cut by a pathologist using disposable scalpels and placed in 10 ml ice-cold RPMI medium containing a Complete Mini Protease Inhibitor Cocktail tablet for immediate transport to the laboratory. Samples are then rinsed briefly in ice-cold PBS, followed by isotonic sucrose (0.25 M) to remove excess salts, lightly touched against filter paper to remove excess liquid, and then embedded in OCT, wrapped in foil, and snap-frozen in liquid nitrogen. Samples are then stored in liquid nitrogen or at −80◦ . Fixing and Staining The methods we use for hematoxylin and eosin staining and immunostaining of tissue sections are described below. 9
H. Murakami, L. Liotta, and R. A. Star, Kidney Int. 58, 1346 (2000).
[4]
LASER CAPTURE MICRODISSECTION AND PROTEOMICS
37
Hematoxylin and Eosin Staining 1. Cut 8-µm sections onto ethanol-dipped plain glass slides and place on dry ice. 2. Allow the required number of sections to just thaw out and fix in 70% v/v ethanol for 1 min. 3. Stain with Mayer’s hematoxylin (containing 1 Complete Mini Protease Inhibitor Cocktail tablet/Coplin jar) for 30 sec. 4. Briefly dip into water followed by Scott’s tap water (2% w/v magnesium sulfate-7-hydrate, 0.35% w/v sodium bicarbonate) for 10 sec and a further brief dip in water. 5. Stain with eosin (containing 1 Complete Mini Protease Inhibitor Cocktail tablet/Coplin jar) for 5–20 sec. 6. Briefly wash in water and dehydrate by sequentially placing in 70% v/v ethanol for 30 sec, 100% ethanol for 1 min, and xylene for 2 × 5 min. 7. Allow sections to dry and use immediately for LCM. Mayer’s hematoxylin is made as follows: Solution 1: Dissolve 3 g hematoxylin in 20 ml ethanol Solution 2: Dissolve 0.3 g sodium iodate, 1 g citric acid, 50 g chloral hydrate, and 50 g aluminum potassium sulfate sequentially in 850 ml H2O Add solution 1 to solution 2, mix, and add 120 ml glycerol. Mix and store in the dark at room temperature. The stain improves with aging. Eosin is a 1% aqueous solution (as supplied by BDH). Filter stains prior to use. Silver-Enhanced Gold Immunolabeling 1. Cut 8-µm sections onto ethanol-dipped plain glass slides and place on dry ice. 2. Allow the required number of sections to just thaw out and fix in 70% v/v ethanol or acetone for 1 min. 3. Incubate with primary antibody in TBS (containing 1× Complete Mini Protease Inhibitor Cocktail) for 5 min. 4. Briefly wash with TBS. 5. Incubate with gold-labeled secondary antibody in TBS (containing 1× Complete Mini Protease Inhibitor Cocktail) for 5 min. 6. Briefly wash in water and treat with the silver enhancing kit as per the manufacturer’s instructions. 7. Briefly wash in water and counterstain with Mayer’s hematoxylin (containing 1 Complete Mini Protease Inhibitor Cocktail tablet/Coplin jar) for 30 sec.
38
BASIC PRINCIPLES
[4]
8. Briefly dip into water followed by Scott’s tap water for 10 sec and a further brief dip in water. 9. Dehydrate by sequentially placing in 70% v/v ethanol for 30 sec, 100% ethanol for 1 min, and xylene for 2 × 5 min. 10. Allow sections to dry and use immediately for LCM. LCM We use a PixCell II laser capture microdissection system and CapSure LCM caps (Arcturus Engineering, Inc.). LCM with frozen material tends to be more problematic than with paraffinembedded formalin-fixed material. This is partly due to the inferior morphology and partly due to more difficulties in ensuring consistent “lifting” or capture of material. For the majority of our work, we use the 7.5- and 15-µm diameter laser beams and adjust the power settings as required to ensure good capture and also to optimize the laser diameter. For 7.5-µm dissections, for example, we typically have the power settings at 50–70 mW. The most common problem we encounter with LCM is incomplete capture of material on lifting the cap from the section surface. The cap should always be checked after dissection to examine the material present. The reasons for incomplete capture are not always clear but it is often due to incomplete removal of water from the sections. Such problems can sometimes be cured by regularly changing the ethanol and xylene solutions, by not shortening the incubation times in these solutions, and by staining of sections immediately prior to use. Decreasing the length of time of thawing of sections before fixing in ethanol can also be effective. Ambient temperature may also be a factor, with dissection not being optimal in warm environments (logically this may also increase proteolysis). If necessary, cutting thinner sections may also aid the dissection. Often loose contaminating tissue can be seen on the cap surface after dissection; we recommend using the adhesive part of a “Post-It” to remove such material prior to inserting the cap into extraction buffers. The use of PrepStrip tissue preparation strips to remove loose tissue debris from the slide and CapSure HS LCM caps with a 12-µm rail that prevents the transfer film from touching the tissue surface can also ensure a cleaner dissection. Sample Preparation A general outline of sample preparation is given below. Details of the times of extraction and the extraction buffers we use are given in the subsequent sections. Once microdissection is complete, a “Post-It” is used to remove extraneous material from the cap surface prior to protein extraction. If very small extraction volumes (<10 µl) are being used the buffer is added directly to the cap surface and the cap is then covered with an inverted Eppendorf Safe-Lock microtube. If
[4]
LASER CAPTURE MICRODISSECTION AND PROTEOMICS
39
larger volumes (30–50 µl) are being used the cap is inserted into an Eppendorf Safe-Lock microtube containing the buffer and the tube is then inverted taking care to ensure the cap surface is covered with buffer—vortex mixing at intervals can be used to assist extraction. At the end of the extraction period microtubes are spun at 13,000 rpm in a benchtop microfuge for 5 min to ensure all the extract is collected in the tube. The cap is then examined microscopically to ensure complete sample solubilization. Extracts are stored at −80◦ . Samples are not subjected to freeze–thaw; instead extracts from different days’ dissections are pooled prior to analysis. Some groups recommend stripping the cap polymer and inserting it into the buffer as an alternative strategy.7 Techniques for Protein Analysis We analyze laser capture microdissected samples by 2D PAGE, SELDI, and Western blotting and have optimized protocols for sample solubilization and downstream analysis which are dealt with individually in the following sections. General Considerations Buffers for sample solubilization are chosen to minimize proteolytic degradation and maximize sample recovery while being compatible with the subsequent analytical technique to be employed. The use of protease inhibitor(s) and the denaturing nature of the buffers used often eradicates proteolysis but this must be considered when developing new extraction protocols. Extraction buffer volumes are kept to a minimum to maximize protein concentration and multiple caps can be extracted sequentially in each microfuge tube. In studies using LCM the protein load is often described as that procured by a specific number of laser shots or the number of cells that have been captured. This can be used as a relative measure of protein load within studies but does not account for tissue or sample variability and is not readily transferable when taking into account factors such as different thicknesses of section or different laser diameters. Quantification of protein is difficult because of the lack of sensitivity of standard protein assays and the interference of some buffer components with many protein assays. The methods for protein estimation we use vary depending on the analytical technique being used (this reflects both the amount of sample being collected and the buffer used); this is an area for future development. Materials Thiourea, DTT, iodoacetamide and Tween 20 are from Sigma-Aldrich (Poole, UK), Tris and urea from ICN (Basingstoke, UK), glycine from Genomic Solutions (Cambridge, UK), CHAPS from Calbiochem (Nottingham, UK), bromphenol blue Pharmalyte pH 3–10, Immobiline DryStrip Gels, DryStrip Cover Fluid, and ECL
40
BASIC PRINCIPLES
[4]
Western blotting detection reagents from Amersham Pharmacia (Little Chalfont, UK), Protogel acrylamide (30% acrylamide : 0.8% bisacrylamide) from Flowgen (Sittingbourne, UK), LMP agarose from Life Technologies (Paisley, UK), SYPRO Ruby protein blot stain from Molecular Probes (Leiden, The Netherlands), TFA from Perbio (Tattenhall, UK), and Nonidet P-40 from Roche (Lewes, UK). All other chemicals are from BDH (Poole, UK). Milli-Q H2O is used throughout. Protein Profiling Using 2D PAGE 2D PAGE is the central protein separation tool in proteomic studies. Proteins are separated based on two independent characteristics—charge and size— allowing up to 2000 protein species to be resolved in a single gel. The use of sample fractionation and narrow-range pH strips10–12 permits the study of many thousands of proteins from a single biological sample. Identification of proteins from 2D gels is routinely carried out by tryptic digestion and mass spectrometry. Following 2D PAGE we detect proteins by silver staining which can detect nanogram(s) of protein. This is still the most common method used, although the fluorescent dyes such as SYPRO Ruby,13 which have a wider linear dynamic range, are increasing in popularity. A novel method, fluorescence two-dimensional differential gel electrophoresis (2D DIGE),14 which allows samples labeled using different fluorophores to be run on a single gel and imaged separately is an appealing option that can simplify gel analysis, but is not routinely available. When analyzing LCM procured material the limitation of the approach is its sensitivity, which means that with a realistic dissection time only the most abundant proteins can be studied; thus samples are generally analyzed using broad range gradients (pH 3–10NL) to maximize the information obtained. Nonetheless in our hands samples are sufficient to enable large format gels to be used, but minigels can also be successfully employed.7 There are a large number of different protocols used for running 2D PAGE gels, with fundamental variations in the choice of buffers and equipment.15–17 The protocols listed below are those we routinely use for analyzing LCM samples. 10
S. J. Cordwell, A. S. Nouwens, N. M. Verrills, D. J. Basseal, and B. J. Walsh, Electrophoresis 21, 1094 (2000). 11 R. Wildgruber, A. Harder, C. Obermaier, G. Boguth, W. Weiss, S. J. Fey, P. M. Larsen, and A. G¨ org, Electrophoresis 21, 2610 (2000). 12 S. Hoving, H. Voshol, and J. van Oostrum, Electrophoresis 21, 2617 (2000). 13 M. F. Lopez, K. Berggren, E. Chernokalskaya, A. Lazarev, M. Robinson, and W. F. Patton, Electrophoresis 21, 3673 (2000). 14 R. Tonge, J. Shaw, B. Middleton, R. Rowlinson, S. Rayner, J. Young, F. Pognan, E. Hawkins, I. Currie, and M. Davison, Proteomics 1, 377 (2001). 15 A. J. Link, “2-D Proteome Analysis Protocols.” Humana Press, Totowa, NJ, 1999. 16 S. R. Pennington and M. J. Dunn, “Proteomics: From Protein Sequence to Function.” BIOS Scientific Publishers Limited, Trowbridge, UK, 2001. 17 T. Rabilloud, “Proteome Research: Two-Dimensional Gel Electrophoresis and Identification Methods.” Springer Verlag, Berlin, 2000.
[4]
LASER CAPTURE MICRODISSECTION AND PROTEOMICS
41
Sample Solubilization and Protein Estimation. Material from successive caps is extracted into 30 µl of lysis buffer (7 M urea, 2 M thiourea, 4% w/v CHAPS 1% w/v DTT, 0.8% Pharmalyte pH 3–10, 1 mg/ml Pefabloc) for ∼30 min. Samples are stable for several hours at room temperature, readily allowing the successive solubilization of caps in the same aliquot of lysis buffer. Protein estimation is sometimes possible using a modified Bradford protein assay (BioRad, Hemel Hempstead, UK). Alternatively samples can be normalized by carrying out dot-blots and staining with Coomassie blue or probing with antibodies to a housekeeping gene product.18 Some researchers have used parallel dissections extracted into an assay compatible buffer,7 which can give only a rough estimation given the very different yield of protein in different extraction buffers. Isoelectric Focusing. Isoelectric focusing is carried out on 18-cm immobilized pH gradient (IPG) strips (Immobiline DryStrip Gels; pH 3–10NL) using the IPGphor system (Amersham Pharmacia). 1. Dilute samples to a final volume of 450 µl in reswell buffer (7 M urea, 2 M thiourea, 4% w/v CHAPS, 0.46% w/v DTT, 0.2% Pharmalyte pH 3–10 with a trace of bromphenol blue). 2. Pipette the sample into a strip holder, place the IPG strip over the sample, and cover with DryStrip Cover Fluid. 3. Apply the sample using in-gel rehydration (30 V, 13 hr). 4. Focus for a total of 65 kVh (200 V 1 hr, 500 V 1 hr, 1000 V 1 hr, 1000–8000 V 1 hr, 8000 V to end). 5. Remove the strips and drain off the cover fluid. Store between plastic sheets in a film cassette at −80◦ prior to SDS–PAGE. SDS–PAGE. For SDS–PAGE we use the ISO-DALT system (Amersham Pharmacia). Gels (generally 10%T resolving gels with a 4%T stack) are poured “in-house” using standard electrophoresis protocols. 1. Equilibrate IPG strips in equilibration buffer (6 M urea, 30% v/v glycerol, 2% w/v SDS, 0.05 M Tris-HCl pH 6.8) containing 1% w/v DTT for 15 min followed by equilibration buffer containing 4% w/v iodoacetamide for 10 min. 2. Rinse strips with gel running buffer and place on top of the second dimension gels. Place a 20-µl bead of 1% LMP agarose made up in Tris–glycine gel running buffer (24 mM Tris, 0.2 M glycine, 0.1% w/v SDS) containing molecular weight markers adjacent to the strip. 3. Overlay the strip with 1% LMP agarose in Tris–glycine gel running buffer and allow to set. 4. Carry out electrophoresis overnight at 12.5◦ using 15 to 20 mA per gel. 18
M. R. Emmert-Buck, J. W. Gillespie, C. P. Paweletz, D. K. Ornstein, V. Basrur, E. Appella, Q. H. Wang, J. Huang, N. Hu, P. Taylor, and E. F. Petricoin III, Mol. Carcinog. 27, 158 (2000).
42
BASIC PRINCIPLES
[4]
Silver Staining. There are a number of commercially available silver staining kits and we have elected to use the OWL silver staining kit (OWL Separation Systems, Portsmouth, NH). 1. Fix gels overnight in 50% v/v methanol, 10% v/v acetic acid. 2. Silver stain using the manufacturer’s instructions but using 300-ml volumes per gel and with the following incubations: fixing solution 2 : 30 min; pretreatment solution: 40 min; water wash: 2 × 10 min; silver staining solution: 40 min; water washes: 3 × 2 min; developer: 5–10 min. We scan gels using a Personal Densitometer (Molecular Dynamics, Chesham, UK). Gel analysis is carried out using Melanie 3 software (GeneBio, Geneva, Switzerland). Figure 1 shows the protein profile of 30,000 15-µm laser shots of cervical epithelium. It is difficult to generalize about the time taken to laser capture sufficient sample for analysis and the enrichment seen relative to the starting material. The abundance of the desired cell type(s) and the complexity of the tissue being microdissected, which dictates the size of laser that can be used, are perhaps the most significant factors. In the case of cervix, 7500 15-µm laser shots of epithelial tissue, which took 3 hr to collect, were sufficient to allow 590 protein spots to be
FIG. 1. 2D gel of cervical epithelium. 30,000 15-µm laser shots of cervical epithelium were procured by laser capture microdissection. Protein extracts were prepared and separated by 2D PAGE. 930 protein species were resolved.
[4]
LASER CAPTURE MICRODISSECTION AND PROTEOMICS
43
resolved, with significant enrichment in the protein profile being evident relative to total tissue. In contrast the dissection of proximal tubules from human kidney cortex required the much more time-consuming collection of 72,000 7.5-µm laser shots to resolve 660 protein species, and the protein profile was largely unchanged compared to the starting material.8 Only a small number of studies describing the use of LCM in combination with 2D PAGE have been published,7,8,18,19,20 but taken together they provide a very encouraging endorsement of the approach. However, our results do indicate that there are limitations with respect to sample and tissue type. Despite only screening the tip of the iceberg as far as the proteome is concerned, comparative analyses of tumor and normal samples have been successful in identifying putative disease markers for esophageal and prostate cancers.18,20 Study of less abundant proteins is currently beyond the limits of the technology. A problem that may be encountered in subsequent analysis of proteins implicated by a LCM and 2D PAGE strategy is in determination of protein identity. If this is achieved by selection of the spot of interest from a more heavily loaded preparative-grade gel of whole tissue lysate, it may not always be straightforward if contaminating proteins removed by the dissection obscure the profile of the dissected material. Alternatively if spots of interest are selected from a gel of dissected sample, there may not be sufficient protein for mass spectrometry. Protein Profiling Using SELDI The SELDI system developed by Ciphergen Biosystems (Fremont, CA) uses ProteinChip technology to selectively capture proteins from complex mixtures using standard chromatographic chemistries followed by time-of-flight mass spectrometry to generate a spectral profile of the sample under investigation.21 This technology provides an alternative highly sensitive tool for protein profiling, allowing the rapid collection of highly reproducible spectral profiles of biological samples. By using a combination of ProteinChip surfaces and extraction buffers it is possible to profile hundreds of proteins from a single sample. SELDI is most suited to the study of low molecular weight proteins (<20,000 Da) and is therefore complementary to 2D PAGE. With the assistance of neural networks and heuristic clustering, clinically valuable information relating to diagnosis, prognosis, or response to therapy can be directly obtained from the information within the spectra. The subsequent 19
R. E. Banks, M. J. Dunn, M. A. Forbes, A. Stanley, D. Pappin, T. Naven, M. Gough, P. Harnden, and P. J. Selby, Electrophoresis 20, 689 (1999). 20 D. K. Ornstein, J. W. Gillespie, C. P. Paweletz, P. H. Duray, J. Herring, C. D. Vocke, S. L. Topalian, D. G. Bostwick, W. M. Linehan, E. F. Petricoin III, and M. R. Emmert-Buck, Electrophoresis 21, 2235 (2000). 21 M. Merchant and S. R. Weinberger, Electrophoresis 21, 1164 (2000).
44
BASIC PRINCIPLES
[4]
determination of protein identity can represent a more difficult challenge. The use of antibodies has been successful in a number of cases where the size of the protein alone can be used to predict its likely identity.22,23 In cases where this is not possible a strategy of (partial) purification must be adopted which has been illustrated very elegantly24 but is potentially more problematic. The sensitivity of SELDI is in the femtomole to attomole range, which makes it ideally suited as a tool for analyzing samples generated by LCM. Two approaches are possible for analysis of sample: “dry down” experiments that profile all readily ionized proteins in a total sample and experiments making use of individual ProteinChip surface chemistries. Preparation of Whole Cell Extracts. Material from the cap is extracted into 5 µl 10 mM Tris-HCl, pH 8, 1% v/v NP-40 by adding buffer directly to the cap surface, mixing with a pipette tip, and incubating for 15 min at 4◦. This extraction buffer is fully compatible with SELDI analysis but is not optimal for total protein solubilization. Other groups have used buffers for solubilizing microdissected samples that incorporate a range of detergents and reducing agents including PBS, 1% w/v Triton X-100, 1% w/v MEGA 10, 1% w/v octyl-βglucopyranoside, 0.1% SDS25 and 10 mM Tris pH 7.5, 1 mM MgCl2, 1 mM EGTA, 0.1 mM PMSF, 5 mM 2-mercaptoethanol, 0.5% CHAPS, and 10% glycerol.26 The addition of urea can markedly improve protein extraction and is compatible with some ProteinChip surfaces. SELDI Analysis with Different ProteinChip Surfaces. Methods for each ProteinChip type we have used are listed below. We generally use sinapinic acid as the matrix and add two 0.35-µl volumes. Carry out all incubations in a humidity chamber to prevent drying. NP2 (normal phase) 1. Mix sample 1 : 1 with 20% v/v acetonitrile, 0.2% v/v TFA and apply 5 µl to the ProteinChip surface. 2. Allow to dry. 3. Wash with H2O (3 × 1 min). 4. Add matrix. 22
G. L. Wright, L. H. Cazares, S.-M. Leung, S. Nasim, B.-L. Adam, T.-T. Yip, P. F. Schellhammer, L. Gong, and A. Vlahou, Prostate Cancer Prostatic Dis. 2, 264 (1999). 23 A. Vlahou, P. F. Schellhammer, S. Mendrinos, K. Patel, F. I. Kondylis, L. Gong, S. Nasim, and G. L. Wright, Jr., Am. J. Pathol. 158, 1491 (2001). 24 V. Thulasiraman, S. L. McCutchen-Maloney, V. L. Motin, and E. Garcia, Biotechniques 30, 428 (2001). 25 C. P. Paweletz, J. W. Gillespie, D. K. Ornstein, N. L. Simone, M. R. Brown, K. A. Cole, Q. H. Wang, J. Huang, N. Hu, T.-T. Yip, W. E. Rich, E. C. Kohn, W. M. Linehan, T. Weber, P. Taylor, M. Emmert-Buck, L. A. Liotta, and E. F. Petricoin, Drug Dev. Res. 49, 34 (2000). 26 F. von Eggeling, H. Davies, L. Lomas, W. Fiedler, K. Junker, U. Claussen, and G. Ernst, Biotechniques 29, 1066 (2000).
[4]
LASER CAPTURE MICRODISSECTION AND PROTEOMICS
45
SAX2 (strong anion exchange) 1. Wash ProteinChip with 5 µl 10 mM Tris pH 8, 1% v/v NP-40 (2 × 5 min). 2. Mix sample 1 : 1 with 10 mM Tris, 1% v/v NP-40 and apply 5 µl to the ProteinChip surface. 3. Incubate for 30 min. 4. Wash with 10 mM Tris, 1% v/v NP-40 (5 × 1 min) then with H2O (2 × 1 min). 5. Add matrix. H4 (hydrophobic) 1. Pretreat ProteinChip with 2 µl 50% v/v acetonitrile. 2. Mix sample 1 : 1 with 20% v/v acetonitrile, 0.2% v/v TFA and apply 5 µl to the ProteinChip surface. 3. Incubate for 20 min. 4. Wash with 10% v/v acetonitrile 0.1% v/v TFA (1 × 1 min, 2 × 5 min). 5. Air dry. 6. Add matrix. Alternatively for dry down experiments follow steps 1 to 3, allow the sample to air dry, and add matrix directly. WCX2 (weak cation exchange) 1. 2. 3. 4. 5. 6. 7. 8. 9.
Add 5 µl 10 mM HCl to ProteinChip surface and incubate for 10 min. Wash with H2O (3 × 1 min). Wash with 100 mM ammonium acetate pH 6.5 (1 × 5 min). Mix sample 1 : 1 with 100 mM ammonium acetate pH 6.5. Apply 5 µl to the ProteinChip surface. Incubate for 30 min. Wash with 100 mM ammonium acetate pH 6.5 (3 × 1 min). Wash with H2O (2 × 1 min). Add matrix.
IMAC3 (immobilized metal affinity) 1. Wash ProteinChip with 5 µl 50 mM nickel sulfate (2 × 5 min). 2. Rinse with PBS NaCl (PBS, 0.5 M NaCl, 0.1% v/v NP-40; 2 × 5 min). 3. Mix sample 1 : 1 with 2× PBS NaCl and apply 5 µl to ProteinChip surface. 4. Incubate for 1 hr. 5. Wash with PBS NaCl (5 × 1 min) then with H2O (2 × 1 min). 6. Dry. 7. Add matrix.
46
BASIC PRINCIPLES
[4]
FIG. 2. SELDI spectra of cervical epithelium and stromal fractions. 2500 15-µm laser shots of cervical epithelium were microdissected and analyzed by SELDI using an NP2 ProteinChip. The profile of the stromal tissue is also shown.
Protein Estimation. In a SELDI-based approach the need to normalize protein load is a serious consideration if spectral profiles are to be compared. A protein assay with a detection limit in the nanogram range is required. To achieve this we use a membrane bound assay based on the fluorescent dye SYPRO Ruby. 1. Dilute samples 10-fold in 10 mM Tris pH 8.0 and prepare standards in 10 mM Tris pH 8.0, 0.1% v/v NP-40. 2. Pipette 5 µl into the well of a sterile 96-well plate with mixed cellulose ester membrane (MultiScreen-HA sterile plate, Millipore, Watford, UK) and incubate for 10 min at room temperature. 3. Wash wells four times for 1 min with Milli-Q H2O. 4. Incubate with SYPRO Ruby (diluted 1 : 24 in Milli-Q H2O) for 15 min in the dark with shaking. 5. Repeat step 3. 6. Read signals using a fluorescent plate reader. SELDI-generated spectral profiles of laser capture microdissected cervical epithelium and the stromal tissue remaining after LCM are shown in Fig. 2. SELDI requires comparatively little sample; indeed, profiles of a limited nature can be generated from fewer than 100 cells,25,26 making it realistic to profile a large number of samples. Tissue-specific profiles and novel candidate disease markers have been readily generated using SELDI.22,25–27 It will be very interesting to see whether the data obtained hold up to more rigorous testing and whether protein identities can be determined. 27
F. von Eggeling, K. Junker, W. Fiedle, V. Wollscheid, M. D¨urst, U. Claussen, and G. Ernst, Electrophoresis 22, 2898 (2001).
[4]
LASER CAPTURE MICRODISSECTION AND PROTEOMICS
47
Study of Individual Proteins Using Antibodies The study of individual proteins by immunoassays is likely to be the most widely used analytical tool in protein studies employing LCM. Assays using immobilized antibodies to bind proteins of interest with detection by second antibodies (as in sandwich ELISAs) or by SELDI have been successful in analyzing LCM procured samples. Western blotting allows different forms of proteins to be resolved, so it is perhaps the most informative of this type of analytical technique. Although tissue arrays lose this additional information they represent a more sensitive and higher throughput development of this technology.28 The sensitivity of Western blotting obviously depends on a number of factors including the copy number of the protein being analyzed and the antibody being used, with a detection limit in the low femtogram range being achievable. Preparation of Protein Extracts and Protein Estimation. Material is solubilized from the cap into 5–20 µl SDS–PAGE sample buffer (0.0625 M Tris, pH 6.8, 10% v/v glycerol, 2% w/v SDS, 10% v/v 2-mercaptoethanol, 0.0025% w/v bromphenol blue) for 15 min at room temperature. Samples are loaded based on the number of laser shots and subsequently samples are more accurately normalized by probing for a housekeeping gene product (such as β-actin). Western Blotting. We make use of a standard minigel format to run SDS–PAGE gels and use a Trans-blot SD semidry transfer cell (BioRad, Hemel Hempstead, UK). 1. Resolve samples by SDS–PAGE and transfer protein to Immobilon-P membrane (Millipore) using 48 mM Tris, 39 mM glycine, 20% methanol as the transfer buffer. 2. Block for 1 hr in TBST (Tris-buffered saline, 0.1% v/v Tween 20) containing 10% nonfat milk. 3. Incubate with primary antibody diluted in TBST containing 1% nonfat milk for 1 hr. 4. Wash in TBST (3 × 5 min). 5. Incubate with biotinylated secondary antibody diluted in TBST containing 1% nonfat milk for 1 hr. 6. Repeat step 4. 7. Incubate with streptavidin-HRP diluted in TBST containing 1% nonfat milk for 1 hr. 8. Wash in TBST (4 × 5 min). 9. Detect signal by enhanced chemiluminescence. 28
C. P. Paweletz, L. Charboneau, V. E. Bichsel, N. L. Simone, T. Chen, J. W. Gillespie, M. R. Emmert-Buck, M. J. Roth, E. F. Petricoin III, and L. A. Liotta, Oncogene 20, 1981 (2001).
48
BASIC PRINCIPLES
[4]
FIG. 3. Western blot analysis of laser captured renal cell carcinoma. 1250 (lane 1), 500 (lane 2), and 250 (lane 3) 7.5-µm laser shots of renal cell carcinoma were analyzed by immunoblotting with antibodies to Major vault protein (upper panel) and β-actin (lower panel).
For tissues such as kidney and liver that contain high levels of endogenous biotin-containing proteins, we use the Envision system as an alternative highsensitivity detection system (Banks et al., submitted). Western blotting can be carried out using very small amounts of sample. We have detected major vault protein and β-actin in 250 7.5-µm laser shots of renal cell carcinoma (Fig. 3). In other studies a number of proteins, including annexin I, CD34, and PSA, have been successfully studied in a variety of human tumors.29–31 Investigation of phosphorylated signaling molecules involved in cell survival in prostate cancer progression clearly illustrates the scope of the approach.28 Conclusions As the use of LCM becomes increasingly widespread it is important to revisit several fundamental points. First is the question of whether the dissected sample is representative of the desired cell population. Second is the question of whether the use of LCM, which involves an investment of both time and sample, has added to an investigation. 2D PAGE has been used as an analytical tool to assess artifacts introduced into protein profiles by tissue section processing and promising preliminary results have been obtained.8 Similarly using SELDI analysis, we have found that hematoxylin and eosin staining introduces reproducible changes into the protein profile. These results are encouraging but such in vitro changes require further investigation. LCM can be introduced into studies of disease at the stage of marker discovery and marker validation. In the case of marker discovery, LCM may allow more 29
Y. Natkunam, R. V. Rouse, S. Zhu, C. Fisher, and R. M. van De, Am. J. Pathol. 156, 21 (2000). D. K. Ornstein, C. Englert, J. W. Gillespie, C. P. Paweletz, W. M. Linehan, M. R. Emmert-Buck, and E. F. Petricoin, Clin. Cancer Res. 6, 353 (2000). 31 C. P. Paweletz, D. K. Ornstein, M. J. Roth, V. E. Bichsel, J. W. Gillespie, V. S. Calvert, C. D. Vocke, S. M. Hewitt, P. H. Duray, J. Herring, Q. H. Wang, N. Hu, W. M. Linehan, P. R. Taylor, L. A. Liotta, M. R. Emmert-Buck, and E. F. Petricoin, Cancer Res. 60, 6293 (2000). 30
[5]
LCM TISSUE PROCESSING
49
meaningful comparative analyses to be carried out by enriching for the pathologically relevant cell types, which are often a minor component of whole tissue. Although the information obtained is restricted by sample availability, such that study of only the most abundant proteins is likely to be feasible, this approach has been successful in identifying putative biomarkers. In some studies, such as those using samples that contain cells of different pathological stages, microdissection is obviously required. In other cases it is not clear whether putative biomarkers would not have been identified by the study of whole tissue. In the case of marker validation, LCM and antibody-based approaches such as immunoassays and Western blotting provide an alternative to immunohistochemistry, giving information about the relative level of protein expression in a particular cell type. Such analyses are highly sensitive, allowing the study of low-abundance proteins. In general the data are more quantitative than those provided by immunohistochemistry and additional information regarding isoforms or differentially modified forms of a particular protein can be obtained. Strategies combining LCM and protein arrays, which are described elsewhere, are likely to be used increasingly in high-throughput screening.
[5] Optimized Tissue Processing and Staining for Laser Capture Microdissection and Nucleic Acid Retrieval By LORA E. HUANG, VERONICA LUZZI, TORSTEN EHRIG, VICTORIA HOLTSCHLAG, and MARK A. WATSON Introduction Laser capture microdissection (LCM) is a powerful technique for the isolation and subsequent molecular analysis of discrete cell populations present in histologically heterogeneous tissue sections. However, to optimize recovery of nucleic acids (both DNA and RNA) from dissected specimens, particular attention to initial tissue processing is required. In this chapter we outline the considerations necessary for tissue processing and staining for LCM and nucleic acid retrieval, and we provide detailed protocols for tissue processing that have worked well in our shared resource facility for the past several years. Critical parameters for tissue processing include initial specimen procurement, preservation of the tissue specimen, and histological staining of the tissue to identify target cell populations.1 The exact methodological approach will depend on 1
N. Tanji, M. D. Ross, A. Cara, G. S. Markowitz, P. E. Klotman, and V. D. D’Agati, Exp. Nephrol. 9, 229 (2001).
METHODS IN ENZYMOLOGY, VOL. 356
Copyright 2002, Elsevier Science (USA). All rights reserved. 0076-6879/02 $35.00
50
BASIC PRINCIPLES
[5]
the available material (archival surgical pathology paraffin blocks vs prospectively procured animal tissues), the target nucleic acid to be isolated (DNA vs more labile mRNA), the quality of the nucleic acid required, and the relative need for high histological detail (e.g., to distinguish atypical hyperplastic cells from simple hyperplastic lesions). Ideally, tissue harvesting and processing should occur as rapidly as possible to preserve the molecular integrity of the specimen. Whereas this may be easily accomplished when working with animal tissues, it may not be as logistically feasible when human tissue specimens are to be obtained from the hospital operating room. Minor delays in specimen processing will have a minimal effect on isolation of more stable genomic DNA. Surprisingly, as shown in Fig. 1, we have also found that even modest delays associated with obtaining human tissue specimens from surgical areas do not qualitatively affect the isolation of total cellular RNA and also seem to have negligible quantitative effect on the specimen’s cellular mRNA population, as assayed by gene expression profiling using high-density oligonucleotide microarrays.2 In contrast, at least one other published report has concluded that there are significant differences in gene expression patterns from specimens that are processed with as little as a 40 min delay.3 Although rapid specimen processing is desired, in our experience, the “warm ischemia time” experienced by the specimen prior to processing is not a major contributing cause of poor nucleic acid yield from LCM tissue. Collected specimens may be snap frozen or fixed using several different formulations. Snap-frozen tissue will provide the highest molecular sample quality for both RNA and DNA retrieval. However, snap freezing tissue by direct immersion in liquid nitrogen (LN2) frequently generates ice crystals within the tissue that destroy histological detail (“freeze artifact”). Embedding small pieces of tissue in a polymer medium and freezing at a more controlled rate can maintain molecularly integrity of the sample and provide much improved histological detail that, depending on the tissue type, may be close to that obtained from standard fixation and paraffin wax embedding. As an alternative to snap freezing, specimens may be processed in a number of different precipitating fixatives such as ethanol, methanol/acetic acid,4 and acetone5 followed by paraffin wax embedding. These fixative are a more “molecularly friendly” alternative to standard formalin fixation. Tissues so processed yield good quality DNA. However, as shown in Fig. 2, RNA derived from these specimens is significantly degraded compared to snap-frozen material. This material is 2
M. Watson, manuscript in preparation. J. Huang, R. Qi, J. Quackenbush, E. Dauway, E. Lazaridis, and T. Yeatman, J. Surg. Res. 99, 222 (2001). 4 M. Shibutani, C. Uneyama, K. Miyazaki, K. Toyoda, and M. Hirose, Lab. Invest. 80, 199 (2000). 5 Y. Sato, K. Mukai, S. Furuya, and Y. Shimosato, J. Pathol. 163, 81 (1991). 3
[5]
LCM TISSUE PROCESSING
51
FIG. 1. Effect of warm ischemia time on RNA recovery from human surgical tissue specimens. (A) Human tissue specimens from surgical resections of colon (1–3), kidney (4–6), or breast (7–9) were dissected into 1 cm3 pieces and either frozen immediately upon receipt in the surgical pathology suite (1, 4, 7) or placed in 4◦ phosphate-buffered saline (PBS) for 6 hr (2, 5, 8) or 24 hr (3, 6, 9) prior to freezing. RNA was isolated from frozen tissues using Trizol reagent (In vitrogen) following the manufacturer’s protocol and subsequently analyzed by formaldehyde gel electrophoresis. Note that even at 6 hr after procurement, isolated RNA appears qualitatively comparable to that obtained from immediately snap freezing tissue. At 24 hr postprocurement, there is noticeable RNA degradation as evidenced by a low molecular weight smear below the 18S ribosomal band. (B) RNA isolated from spleen (1–3) and colon (4, 5) tissues as described in (A) either after immediate freezing (1, 4), or after 30 min (2) or 6 hr (3, 5) incubations in 4◦ PBS. (C) Spleen RNAs shown in (B) from immediate freezing or 6 hr delay specimens were utilized for biotinylated cRNA target synthesis and hybridization to Affymetrix Human Cancer Gene HCG110 Arrays as previously described [V. Luzzi, V. Holtschlag, and M. A. Watson, Am. J. Pathol. 158, 2005 (2001)]. Gene expression levels are plotted for immediately frozen tissue RNA (x axis) versus the 6 hr delayed tissue RNA (y axis) on a log scale. Each point represents one gene with darker points representing genes scored detected (‘P’) in one or both samples and lighter points representing genes scored not detected (‘A’) in both samples by the Affymetrix software. Lines indicate limits of twofold change difference. Of 2033 gene transcripts represented on the array, only 35 genes demonstrated a greater than twofold change in expression between the two tissue samples. Differential expression of these genes was not observed in a replicate set of data, suggesting that the difference in expression of even these 35 genes may be due to technical variability rather than consistent differences between tissue handled under different conditions. All human tissue specimens were collected using an Institutional Review Board (IRB)-approved protocol.
still suitable for reverse transcriptase polymerase chain reaction (RT-PCR)-based assays4,5 and, because of the superior histology of sections processed by these methods, may be a reasonable compromise for many investigators. Cross-linking fixatives, such as routine formalin fixation used in surgical pathology departments for clinical specimens, yield very poor quality nucleic acid
52
BASIC PRINCIPLES
[5]
FIG. 2. RNA isolation from ethanol fixed and paraffin embedded (EFPE) tissue vs snap-frozen tissue. Human foreskin samples were collected under an IRB-approved protocol and bisected. Onehalf of the tissue specimen was snap frozen in liquid nitrogen while the other half was subjected to 70% ethanol fixation and low-melting paraffin embedding as described in this chapter. Fifty µm frozen sections or paraffin embedded sections were cut and RNA was isolated using Trizol reagent following the manufacturer’s protocol. For paraffin embedded tissues, tissue sections were washed twice in xylene and twice in 100% ethanol followed by Trizol extraction. One µl of each resulting RNA preparation was run on an RNA LabChip using a BioAnalyzer2100 (Agilent Technologies). Lane L is a molecular weight ladder, lanes 1, 3, and 5 are RNAs derived from three independent snap-frozen specimens, and lanes 2, 4, and 6 are RNAs derived from the corresponding ethanol fixed, paraffin embedded tissue. Note the severe degree of RNA degradation in ethanol fixed paraffin embedded tissue as compared to snap-frozen material.
that is highly fragmented. For studies involving archived human specimens, there may be little choice but to work with this material. Many investigators have successfully used LCM material from formalin fixed tissues for DNA-based studies6,7 and even quantitative gene expression studies where target amplicons involve small stretches of RNA sequence.8,9 When possible, however, formalin or other crosslinking fixatives should be avoided except, perhaps, when there is a desire to preserve the tissue specimen for many years at ambient storage conditions. After tissue processing and subsequent sectioning, tissue sections can be stained with a number of different histological dyes. For routine histological staining, the objective is to visualize cell nuclei and cytoplasmic features with sufficient detail to allow for the selection of the target cell population. Because dyes bind cellular components such as DNA and RNA, the intensity of cell staining can often be directly proportional to the dye’s interference with nucleic acid retrieval. Also, because dye binding to cellular components may occur at relatively extreme pH 6
Z. P. Ren, J. Sallstrom, C. Sundstrom, M. Nister, and Y. Olsson, Pathobiology 68, 215 (2000). U. Lehmann, O. Bock, S. Glockner, and H. Kreipe, Pathobiology 68, 202 (2000). 8 T. E. Godfrey, S. H. Kim, M. Chavira, D. W. Ruff, R. S. Warren, J. W. Gray, and R. H. Jensen, J. Mol. Diagn. 2, 84 (2000). 9 K. Specht, T. Richter, U. Muller, A. Walch, M. Werner, and H. Hofler, Am. J. Pathol. 158, 419 (2001). 7
[5]
LCM TISSUE PROCESSING
53
conditions, processing steps needed to visualize stains may simultaneous chemically degrade nucleic acid, particularly more chemically labile RNA. We have examined the effect of a number of nuclear dyes on the retrieval and subsequent quantitative amplification of genomic DNA obtained from both fresh frozen and formalin fixed LCM material.10 Whereas previous studies have noted interference in DNA retrieval and amplification from manually dissected, hematoxylin stained tissue,11 we observed no such effect from LCM tissue. In fact, the use of hematoxylin, methyl green, toluidine blue O, and azure B gave comparable results for both histological nuclear detail and DNA recovery. The most probable explanation for this result is that the amount of stained LCM tissue per volume of sample lysis buffer is considerably smaller than that obtained from more grossly, manually dissected tissue sections and, therefore, is not at a sufficient concentration to cause inhibition of downstream applications. Unlike DNA, recovery of cellular RNA appears to be very sensitive to the histological staining process. As shown in Fig. 3, RNA retrieval from LCM tissue (as judged by a qualitative RT-PCR assay) is dependent on the histological stains employed. For this reason, we have adopted a staining process previously described12 that results in a less intense histological stain, but that allows recovery of high quality RNA that can be used for sophisticated applications such as molecular amplification and gene expression profiling with high density oligonucleotide arrays.13 Methods Frozen Tissue Processing On receipt, the tissue specimen should be rinsed in cold buffered saline or other physiologic buffer to remove excess and clotted blood. Grossly evident fat, necrosis, or cauterized tissue should be trimmed as much as possible. If the tissue specimen is large, it should be dissected into individual segments no larger than 1 cm × 1 cm × 1 cm. Fill an appropriately sized cryomold (Sakura Finetek, TissueTek II Cryomold) with OCT embedding compound (Sakura Finetek, Tissue-Tek OCT Compound). Place the tissue segment(s) in the mold so that it is completely submerged in the embedding compound and so that the desired cross section of the tissue is lying flat, face down in the mold. The tissue should be centered in the mold with a border of embedding compound surrounding the tissue. This orientation 10
T. Ehrig, S. A. Abdulkadir, S. M. Dintzis, J. Milbrandt, and M. A. Watson, J. Mol. Diagn. 3, 22 (2001). 11 T. Murase, H. Inagaki, and T. Eimoto, Mod. Pathol. 13, 147 (2000). 12 S. M. Goldsworthy, P. S. Stockton, C. S. Trempus, J. F. Foley, and R. R. Maronpot, Mol. Carcinog. 25, 86 (1999). 13 V. Luzzi, V. Holtschlag, and M. A. Watson, Am. J. Pathol. 158, 2005 (2001).
54
BASIC PRINCIPLES
[5]
FIG. 3. Effect of histological stains on RT-PCR analysis of RNA derived from LCM tissue. Fourteen identical 6-µm sections from frozen breast tumor tissue were cut and mounted on plain glass slides. Subsequently, duplicate tissue sections were scraped from the slides into 200 µl of lysis buffer without further treatment (lane 1), after staining with Harris’ hematoxylin and eosin Y (lane 2), after staining with a modified Harris’ hematoxylin and eosin Y (lanes 3), after staining with Mayer’s hematoxylin and eosin Y (lanes 4), after staining with Mayer’s hematoxylin and eosin Y, substituting “bluing reagent” with 1× automation buffer as described in this chapter (lanes 5), after staining with Eosin Y only (lane 6), or after dehydration in ethanols and xylene, omitting all staining solutions (lane 7). All 14 RNAs were isolated using the Stratagene MicroRNA isolation kit. Subsequently, 10% of the recovered RNA from each tissue sample was subjected to qualitative RT-PCR analysis using exon-spanning glyceraldehyde-6-phosphate dehydrogenase primers. Lane L is a 100-bp ladder, N is a negative control (no RT enzyme added to sample), and P is a positive control (purified cell line RNA). Note significantly attenuated amplification signal from RNA derived from routinely stained hematoxylin tissue.
will facilitate cutting of frozen sections. Without disturbing the tissue orientation, quickly transfer the mold into a cryobath (Shandon Lipshaw, Histobath) containing 2-methylbutane (Sigma) that has been equilibrated to approximately −50◦ . Such “histobaths” are commonly employed in frozen section rooms of pathology departments. If such an instrument is not available, a dry ice and 2-methylbutane slurry bath can be used to freeze the mold. Alternatively, a pair of long-handled forceps may be used to lower the mold into the vapor phase of a Dewar flask (Nalgene) containing liquid nitrogen. Maintain the mold in the freezing chamber for 10 to 15 min to ensure that the entire block is frozen. After freezing, remove the mold, label the mold with a cryomarker, wrap it tightly in aluminum foil, and store it at −80◦ until ready for sectioning. Ethanol Fixation As an alternative to freezing, tissue specimens may be fixed in 70% ethanol and embedded in low-melting paraffin. Clean, trim, and dissect the tissue specimen into 1 cm3 segments as described above. Place the tissue in a 50-ml conical tube containing 30–40 ml of 70% ethanol and fix for 12–16 hr at 4◦ . In the absence of an automated tissue processor, tissue may be further processed by hand by filling 50-ml conical tubes with 30–40 ml of the solutions indicated below. The tissue is then gently transferred from one tube to the next and incubated under the specified conditions. Note that xylene is a volatile liver toxin and should be used under an appropriate chemical fume hood. In the absence of such a fume hood, a tabletop
[5]
LCM TISSUE PROCESSING
55
charcoal filter hood (Labconco) may be used when working with small numbers of tissues and small volumes of xylene: Step 1: (After 16 hr 70% ethanol fixation) 80% ethanol at 4◦ for 15 min Step 2: 95% ethanol at 4◦ for 15 min Step 3: 100% ethanol at 4◦ for 20 min Step 4: 100% ethanol at 4◦ for 30 min Step 5: Xylene at room temperature for 20 min Step 6: Xylene at room temperature for 20 min Step 7: Melt and equilibrate low-melting paraffin (Paraplast X-tra) to 56◦ by incubation in a heated oven. Infiltrate tissue in warmed paraffin at 56◦ for 20 min Step 8: Infiltrate tissue in fresh, warmed paraffin again at 56◦ for 20 min Step 9: Infiltrate tissue in fresh, warmed paraffin again at 56◦ for 20 min Step 10: Infiltrate tissue in fresh, warmed paraffin again at 56◦ for 20 min After the final infiltration with paraffin, select an appropriately sized paraffin mold and fill it partially with warm paraffin. The tissue segment(s) should be removed with warmed forceps and oriented in the paraffin mold such that the tissue is centered in the mold with the cross section of interest lying flat, face down. Place the mold on a cold plate to partially solidify the paraffin and hold the tissue in the correct orientation. Place the bottom of a standard histology cassette on top of the mold and fill the combined mold and cassette with paraffin. Cool the block by immediately placing it on a cold plate and allow the paraffin to solidify for 15 min. Detach the cassette from the mold, label the cassette, and store at 4◦ until ready for sectioning. Paraffin Embedded Tissue Sectioning Trim excess paraffin away from the block, mount the block on the microtome, and cut into the block. Manually advance the blade while cutting to expose the embedded tissue surface and create a flat, cross-sectional plane that is parallel with the microtome knife. Paraffin sections for LCM are routinely cut at 6 µm. Cut a ribbon of paraffin sections and place on the surface of a 39◦ flotation water bath (Fisher Scientific). Although we have not found it necessary to utilize DEPCtreated water in the bath, the water should be changed between specimens and replaced with fresh, Milli-Q water. Separate the sections and place each flattened section onto a clean, uncharged slide (Fisher Premium Superfrost slides) that has been prelabeled with the specimen identifier. Slides are placed vertically for 5 min to drain excess water. Labeled slides are then stored in a vacuum desiccator until staining.
56
BASIC PRINCIPLES
[5]
Frozen Tissue Sectioning The cryostat is equilibrated to −22◦ . The facility of cutting tissue with varying fat content may be affected by cryostat temperature, but we have found this to be a generally acceptable temperature for most solid human tissues. Frozen, embedded tissue is removed from the −80◦ freezer, unwrapped, and allowed to equilibrate to the cryostat temperature for 10–15 min. Using a small amount of OCT compound, the frozen tissue block is attached to the tissue disk and frozen. The block is mounted on the cryostat and rough sections are cut to create a cross-sectional tissue face that is parallel with the cryostat blade. For LCM and nucleic acid retrieval, 6-µm sections are easy to cut, provide good histological detail, and are efficiently transferred to the LCM film. Although thicker sections may be cut (e.g., 20 µm), they are often more difficult to mount, provide less histological detail, and often do not transfer efficiently. The cut section should be immediately transferred to plain, untreated glass slides (Fisher Premium Superfrost slides). It is not necessary to pre treat, clean, or chill the slides before mounting. Immediately after the section is mounted to the slide, it should be transferred to a slide holder (see below) containing 70% ethanol at 4◦ . We have documented that immediate transfer of the mounted tissue section to the 70% ethanol fixative is critically important to preserve histological detail (i.e., to avoid “air-dry” artifact) and to prevent irreversible bonding of the tissue to the slide which impedes efficient LCM transfer. The mounted tissue should be maintained in the 70% ethanol for at least 1 min prior to subsequent staining, but can be left in the 70% ethanol for several hours without significant effect on histology or nucleic acid retrieval. This can allow a convenient means for cutting multiple tissue sections at once and then staining and dissecting one slide at a time (see below). Although some investigators prefer to store frozen section slides at −80◦ until needed for staining and dissection, we have found that slide storage results in decreased quality and quantity of isolated cellular RNA and also results in decreased histological detail (frequent “air-dry” artifact). For this reason, we recommend cutting, staining, and dissection all in the course of a 3- to 4-hour period. This approach is facilitated by having a cryostat, staining rack, and LCM instrument all in the same laboratory area (Fig. 4). Paraffin Embedded Tissue Staining Tissue staining is usually conducted with one or two slides at one time. To conserve reagents and to allow for frequent replenishing of potentially contaminated reagents, staining solutions are dispensed in disposable cytology slide mailers (Fisher), which can easily hold up to 4 slides at one time and require only 20 ml of solution. Slide staining (which involves xylene) must be performed under an appropriate chemical fume hood. If no such hood is available, it is often convenient to position a tabletop charcoal filter hood immediately adjacent to the
[5]
LCM TISSUE PROCESSING
57
FIG. 4. Schematic layout of LCM specimen preparation facility. Suggested arrangement of equipment needed for frozen tissue sectioning, tissue staining, and LCM to optimize processing efficiency and minimize time between sectioning and microdissection. Exploded view of miniature staining rack utilizing 50 ml disposable cytology mailers as staining vessels. Numbers refer to processing steps for paraffin embedded and frozen tissues, respectively, as detailed in the text. Note that the entire staining set can fit easily under a tabletop charcoal filter staining hood.
LCM instrument. For paraffin embedded tissue sections, slides are processed as follows: Step 1: Incubate in xylene for 5 min Step 2: Incubate in xylene for 5 min Step 3: Incubate in 100% ethanol for 30 sec Step 4: Incubate in 100% ethanol for 30 sec Step 5: Incubate in 95% ethanol for 30 sec Step 6: Incubate in 70% ethanol for 60 sec Proceed to “step 1” of frozen tissue staining Frozen Tissue Staining See comments above concerning the setup of staining solutions. After incubating frozen tissue sections in 70% ethanol for at least 1 min or after deparaffinizing paraffin sections as described above, proceed as follows: Step 1: Dip slide 5 times in distilled, Milli-Q water to rehydrate tissue and (for frozen tissue sections) remove embedding compound. We have not found a need for using autoclaved or DEPC-treated water, but frequently replenish with fresh water.
58
BASIC PRINCIPLES
[5]
Step 2: Aqueous nuclear stain. Depending on the tissue type and investigator preference, the tissue is stained as follows: Step 2.1: Mayer’s hematoxylin (Sigma). For optimal RNA recovery it is extremely important to use Mayer’s formulation of hematoxylin (as opposed to Harris’ or other formulations); see Fig. 3. Stain for 30 sec (frozen tissue and RNA recovery) to 2 min (formalin fixed tissue and DNA recovery). Intermediate staining times may produce more optimal results depending on conditions used to process the tissue and the degree of staining required for histological identification of cell populations. Step 2.2: Methyl green (Sigma). This stain works well for DNA retrieval, but has not been tested for RNA isolation. A 0.5% solution of methyl green is made in Milli-Q water and filtered prior to use. Stain both frozen tissue and fixed tissue for 30 min. Step 2.3: Toluidine blue O (Sigma). This stain works well for DNA retrieval, but has not been tested for RNA isolation. A 0.1% solution of toluidine blue is made in Milli-Q water and filtered prior to use. Stain for 30 sec (frozen tissue) to 2 min (formalin fixed tissue). Step 3: Dip slide 10 times in distilled, Milli-Q water to remove excessive stain. We have not found a need for using autoclaved or DEPC-treated water, but replenish this solution frequently. Step 4: If Mayer’s hematoxylin stain was used, place slide in 1× Automation buffer (Biomeda) for 30 sec to develop stain. Automation buffer is a previously described12 alternative to traditionally utilized “bluing reagent” (Richard-Allen Scientific) and, although it does not provide as an intense histological staining, it may provide a higher yield of intact nucleic acid (particularly RNA; see Fig. 3). Step 5: Dip slide 10 times in distilled, Milli-Q water Step 6: Dip slide 20 times in 70% ethanol Step 7: Dip slide 20 times in 95% ethanol Step 8: Stain tissue in eosin Y (Sigma) for 15 sec. Although this alcoholic, extranuclear stain does not appear to have a detrimental effect on nucleic acid recovery (e.g., see Fig. 3), the dye may interfere with downstream fluorogenic assays such as quantitative PCR.14 Therefore, if LCM tissue lysate is to be added directly to such a reaction without prior purification, this stain may be omitted. Step 9: Dip slide 10 times in 95% ethanol Step 10: Dip slide 20 times in 95% ethanol Step 11: Dip slide 20 times in 100% ethanol Step 12: Dip slide 20 times in 100% ethanol 14
J. Serth, M. A. Kuczyk, U. Paeslack, R. Lichtinghagen, and U. Jonas, Am. J. Pathol. 156, 1189 (2000).
[5]
LCM TISSUE PROCESSING
59
Step 13: Dip slide 20 times in xylene Step 14: Incubate slide for 5 min in xylene Step 15: Incubate slide for 5 min in xylene Remove the slide from the final xylene wash and allow the slide to dry under the fume hood for 5 min or until xylene is completely evaporated. After staining and dehydration, slides should be dissected immediately. If this is not possible, slides may be stored in a vacuum desiccator for 1–2 hr. Depending on the relative humidity, we have noted markedly decreased tissue transfer efficiency and RNA recovery from tissue sections that have been maintained for greater than 1 hr at ambient laboratory conditions. Although hematoxylin and eosin staining is the most widely used approach for both RNA and DNA isolation from LCM tissue, several other methods, such as Nissel stain, can be adapted and used for successful staining and retrieval of RNA and DNA.15 At this time Arcturus Engineering has announced the availability of a commercial staining kit (Histogene) specifically designed for LCM tissue staining and retrieval of RNA and DNA. Initial evaluation of this staining kit in our facility demonstrated performance comparable to the protocols described above, at least for tissue staining and isolation of cellular RNA. When evaluating any new staining protocol, we generally run several controls to assess its compatibility with nucleic acid isolation (see Troubleshooting below). Nucleic Acid Purification Detailed protocols for isolation of DNA and RNA from LCM tissue specimens are beyond the scope of this chapter. However, successful isolation of nucleic acid from tissue prepared as described in this chapter has been accomplished using a variety of methods. For isolation of DNA, tissue is routinely lysed in proteinase K buffer, heat inactivated, and then used directly in the amplification reaction.10 RNA has been isolated using both organic extraction micropreps (Stratagene, MicroRNA Isolation Kit) and affinity spin-column based methods (Arcturus Engineering, PicoPure RNA Isolation Kit). Depending on the number of cells captured, 5–10 ng of purified total cellular RNA may be directly visualized using capillary microchip electrophoresis (Agilent Technologies, Agilent BioAnalyzer 2100) to quality control RNA samples prior to downstream applications (Fig. 5). Troubleshooting Although the protocols described above work routinely for downstream DNA and RNA analyses, there can be numerous causes of failure. The following are some common problems encountered and possible ways to resolve them. 15
T. Betsuyaku, G. L. Griffin, M. A. Watson, and R. M. Senior, Am. J. Respir. Cell. Mol. Biol. 25, 278 (2001).
60
BASIC PRINCIPLES
[5]
FIG. 5. Example of LCM, RNA isolation, and downstream gene expression microarray analysis. (A) A lesion of human breast lobular carcinoma in situ (LCIS) stained following the Mayer’s hematoxylin and eosin protocol detailed in this chapter, both before and after LCM. (B) Total RNA isolated from 3000–5000 LCM 30-µm firings of breast carcinoma in situ lesions. RNA was isolated using the Stratagene MicroRNA Isolation kit and 10% (1 µl) of the resulting RNA was analyzed using an RNA LabChip and BioAnalyzer 2100 (Agilent Technologies). Lane L is an RNA ladder standard; lanes 1 and 2 are two different RNA isolations from LCM tissue. On average, 1000 firings of a 30 µm beam (approximately 5000 cells) yielded 12 ng of total RNA [V. Luzzi, V. Holtschlag, and M. A. Watson, Am. J. Pathol. 158, 2005 (2001)]. (C) After two rounds of linear transcript amplification and biotin labeling, labeled aRNA was applied to Affymetrix human Hu6800SubD GeneChips for expression profile analysis. The scatter plot shows gene expression in one RNA sample from LCM-derived carcinoma in situ RNA plotted on a log scale against gene expression in another RNA sample from adjacent LCM-derived normal epithelial cells. Lines indicate limit of 2-fold and 5-fold changes in gene expression. At least two genes (denoted by arrowheads) were shown to be differentially expressed between populations of LCM-derived normal breast epithelial cells and carcinoma in situ lesions from two independent patients. For additional experimental details, see V. Luzzi, V. Holtschlag, and M. A. Watson, Am. J. Pathol. 158, 2005 (2001).
Poor Histological Detail Tissue may appear vacuated without cytoplasmic or even extracellular features. This can be seen frequently with frozen tissue sections and is usually due to “freeze artifact” or “air-dry artifact.” When freezing tissue, do not immerse directly in liquid nitrogen. Also, do not attempt to freeze tissue by placing it inside the
[5]
LCM TISSUE PROCESSING
61
cryostat or −80◦ freezer. Tissue should be frozen in a −50◦ refrigerated bath or 2-methylbutane/dry-ice bath. Make certain tissue is dissected into pieces no larger than 1 cm3. After the frozen section is mounted on the slide, it should be placed immediately in cold 70% ethanol. Do not store frozen section slides at −80◦ . Rather, stain and dissect slides immediately after cutting. Slides may be stored at 4◦ in 70% ethanol for up to 72 hr prior to staining, but for best histological detail and RNA retrieval, slides should not be left in 70% ethanol for more than 3–4 hr. Poor Tissue Staining When working with paraffin embedded sections, make certain that sections have been completely deparaffinized. Increase the number or the time of initial xylene steps. Replace staining solutions frequently. For retrieval of DNA, increase the staining time (for RNA isolation, this is not recommended as increased staining time will increase the likelihood of RNA degradation). If working with hematoxylin, replace the automation buffer solution or increase the incubation time in automation buffer. If using hematoxylin staining for DNA isolation, replace the automation buffer with standard bluing reagent. Although we have not observed much difference in histological detail using the stains mentioned above, depending on the tissue, some stains may provide better detail than others. Poor Tissue Transfer to LCM Film This is usually caused by incomplete dehydration of the tissue section. If a “wave” or “ripple” effect is seen each time the laser is fired on the tissue, then incomplete dehydration is almost certainly the problem. Increase the number or time of final xylene washes. Once the tissue has been stained and dehydrated, work quickly to procure the cells of interest by LCM. If there is a delay between slide staining and dissection, store dehydrated slides in a vacuum desiccator. Ensure that when tissue sections are cut and mounted, they are completely flat on the slide. Wrinkles in the tissue will not allow the transfer film to sit properly over the tissue section and will impede transfer. If thicker sections are being cut (e.g., 20 µm), try using thinner sections. Ensure that sections are mounted on noncharged, plain glass slides. If frozen section slides have been stored at −80◦ , try dissection of slides that have been immediately cut and stained. Be certain that once cut and mounted, frozen tissue sections are immediately placed in 70% ethanol fixative. Do not allow tissue to dry between staining steps. No Nucleic Acid Recovery or Poor Downstream Assay Performance When isolated nucleic acid does not function in the downstream assay, perform a stepwise analysis to locate the problem source. First, perform duplicate dissections. In one of the dissected tissue lysates, spike in pure nucleic acid (DNA
62
BASIC PRINCIPLES
[5]
or RNA) in an amount that would be anticipated from the number of dissected cells (1 cell ∼10 pg total RNA and ∼6 pg genomic DNA). Perform the downstream application. If the spiked sample works and the unspiked sample does not, this indicates a problem with the LCM procedure. If neither sample works, this suggests a problem or inhibition of the downstream assay. Next, cut and stain one tissue section. After dehydration, use a scalpel or razor blade to scrape the entire tissue section into an appropriate amount of lysis buffer. Perform the downstream application. If the assay works, the staining process has been performed correctly; consider increasing the number of captured cells by LCM or monitoring the efficiency of cell capture and elution from the LCM transfer film. If the assay does not work, this indicates a problem with the tissue or staining procedure. Next, cut a single section. If working with frozen tissue, place the unprocessed section immediately into the appropriate volume of lysis buffer. If working with paraffin embedded tissue, mount the tissue section to a slide, deparaffinize the section (see paraffin section staining, above), and then immediately scrape the tissue section into the appropriate volume of lysis buffer. Perform the downstream application on the unstained tissue section. If the assay fails, this suggests a primary problem with the tissue (improper preservation, extensive tissue necrosis, acellular tissue). Examination of a routinely stained hematoxylin and eosin stained section can often identify this problem. If the assay works with the unstained tissue but not with the stained tissue, attempt the following. Replace all staining solutions. If performing RNA isolation and hematoxylin staining, ensure that Mayer’s formulation is being used. If using bluing reagent, replace this with automation buffer. Use a different histological stain (e.g., attempt staining with alcoholic eosin only) or shorten the staining times (particularly those incubations in aqueous solutions). Conclusions LCM is a powerful tool for dissecting molecular and genetic events that occur within discrete cell populations of complex, heterogeneous tissues. Using the protocols detailed in this chapter, many investigators have been able to independently carry out successful experiments involving genome and transcriptome analysis (see Fig. 5). Although initial optimization of tissue preservation and processing can be a tedious process, in the long run it is a critical step needed for successful and complex downstream analyses such as large-scale genome sequencing and global gene expression profile analysis.
[6]
LCM-FISH OF PARAFFIN SECTIONS
63
[6] Fluorescence in Situ Hybridization of LCM-Isolated Nuclei from Paraffin Sections By DOUGLAS J. DEMETRICK, SABITA K. MURTHY, and LISA M. DIFRANCESCO Introduction Fluorescent in situ hybridization (FISH) was first utilized to determine the cytogenetic map position of genes.1 This technique has also proved to be very useful in assessing gene copy number from clinical specimens, and such information may be used, for example, to predict the response of a cancer to treatment.2,3 The method usually works best with fresh or frozen tissue and offers localization of a genomic abnormality to a specific cell, very valuable in cancer specimens that are commonly contaminated with normal tissue elements such as inflammatory cells or fibroblasts. Heterogeneity of gene copy number may be a useful additional feature to assess in cancer specimens,4 as evidence suggests that metastases likely occur via specific dominant clones in the primary tumor.5,6 FISH of paraffin sections has been described, and commercial kits are currently available to perform such analyses. If the FISH target is highly amplified, one can often directly identify amplification within the cancer cells of paraffin sections. Tissue or fixative autofluorescence, however, can seriously compromise interpretation of low-level amplification, or of deletion.7,8 High-throughput methods of paraffin FISH,9,10 using tissue microarrays,11 have also been described but are subject to the same 1
D. Pinkel, T. Straume, and J. W. Gray, Proc. Natl. Acad. Sci. U.S.A. 83, 2934 (1986). J. D. Bitran, B. Samuels, L. Klein, S. Hanauer, L. Johnson, J. Martinec, E. Harris, J. Kempler, and W. White, Bone Marrow Transplant. 17, 157 (1996). 3 S. J. Houston, T. A. Plunkett, D. M. Barnes, P. Smith, R. D. Rubens, and D. W. Miles, Br. J. Cancer 79, 1220 (1999). 4 G. Sauter, H. Moch, D. Moore, P. Carroll, R. Kerschmann, K. Chew, M. J. Mihatsch, F. Gudat, and F. Waldman, Cancer Res. 53, 2199 (1993). 5 D. Theodorescu, I. Cornil, C. Sheehan, S. Man, and R. S. Kerbel, Int. J.Cancer 47, 118 (1991). 6 J. F. Simpson, D. E. Quan, J. P. Ho, and M. L. Slovak, Am. J. Pathol. 149, 751 (1996). 7 J. A. McKay, G. I. Murray, W. N. Keith, and H. L. McLeod, Mol. Pathol. 50, 322 (1997). 8 J. Szollosi, S. J. Lockett, M. Balazs, and F. M. Waldman, Cytometry 20, 356 (1995). 9 P. Schraml, J. Kononen, L. Bubendorf, H. Moch, H. Bissig, A. Nocito, M. J. Mihatsch, O. P. Kallioniemi, and G. Sauter, Clin. Cancer Res. 5, 1966 (1999). 10 J. Richter, U. Wagner, J. Kononen, A. Fijan, J. Bruderer, U. Schmid, D. Ackermann, R. Maurer, G. Alund, H. Knonagel, M. Rist, K. Wilber, M. Anabitarte, F. Hering, T. Hardmeier, A. Schonenberger, R. Flury, P. Jager, J. Luc Fehr, P. Schraml, H. Moch, M. J. Mihatsch, T. Gasser, O. P. Kallioniemi, and G. Sauter, Am. J. Pathol. 157, 787 (2000). 11 J. Kononen, L. Bubendorf, A. Kallioniemi, M. Barlund, P. Schraml, S. Leighton, J. Torhorst, M. J. Mihatsch, G. Sauter, and O. P. Kallioniemi, Nat. Med. 4, 844 (1998). 2
METHODS IN ENZYMOLOGY, VOL. 356
Copyright 2002, Elsevier Science (USA). All rights reserved. 0076-6879/02 $35.00
64
BASIC PRINCIPLES
[6]
caveats as above. Purification of nuclei from paraffin sections for FISH could decrease the effects of tissue autofluorescence; however, the advantage of in situ verification of the source of the cells is lost. The use of laser capture microdissection to prepare nuclei from breast carcinoma cells for both FISH and flow cytometric analysis has been described.12 This technique, LCM/FISH, allows detection of normal interphase copy numbers, and thus detection of low-level amplification or even single copy deletions, which would be very difficult in paraffin section FISH. Some minor refinements to the method have also improved the yield and reproducibility of the technique. Materials and Methods Laser Capture Microdissection Procedure Specimens of breast carcinoma were fixed in 10% neutral buffered formalin and embedded in Surgiplast matrix. Thirty-µm sections were cut from the paraffin blocks, stained with Harris’ hematoxylin, dehydrated through a graded ethanol series to xylene (three xylene washes), and then allowed to air dry. Staining with eosin yielded wide-spectrum background fluorescence that interfered with subsequent fluorescent analysis and is strongly discouraged. Dehydrated sections were subjected to laser capture microdissection using a PixCell II instrument (Arcturus Engineering, Inc.) using the typical conditions: 30-µm beam: power 70–80 mW; pulse duration 6–7 ms 15-µm beam: power 40–50 mW; pulse duration 3–5 ms The 7-µm beam is very difficult to use with thick sections due to contamination of the captured cells with adjacent uncaptured tissue or poor adhesion of the captured cells. The spotting laser is usually very difficult to see because of the thickness of the tissue section, so the position of the beam is marked on the video monitor with a dry-erase marker. Cells and cell nests are fixed to the polymer film on CapSure LCM caps and placed tightly into 0.5-ml microfuge tubes. Some experimentation is usually needed for the tissue at hand and multiple bursts to allow optimal adhesion to the polymer may also be needed to firmly fix the cells. The LCM caps in their tubes can be stored up to 1 year at room temperature prior to nucleus extraction. Nucleus Isolation Procedure The following procedure details the dissolution of the CapSure LCM cap membrane matrix with sequential organic solvents, followed by rehydration of the tissue and proteinase K-mediated release of the nuclei. All organic extractions should be 12
L. M. DiFrancesco, S. K. Murthy, J. Luider, and D. J. Demetrick, Mod. Pathol. 13, 705 (2000).
[6]
LCM-FISH OF PARAFFIN SECTIONS
65
performed in an appropriate fume hood to minimize occupational solvent exposure and care must be taken to avoid ignition of flammable materials. 1. 100 µl of fresh, high-quality chloroform is pipetted into an Eppendorf 0.5-ml microfuge tube and capped with a CapSure LCM cap containing a microdissected specimen. The tube is inverted for 10 sec, then microcentrifuged at 3000g for 30 sec to release the tissue specimen from the “capture” polymer of the LCM cap. 2. The LCM cap is removed and 200 µl of anhydrous ethyl ether is added and mixed with the chloroform by inversion. This step is necessary to lower the density of the solvent to allow successful centrifugation of the tissue fragments while ensuring that the polymer “capture” medium is still soluble. 3. Following microfuge centrifugation, the supernatant is removed by pipette suction with a narrow pipette tip and the pellet is washed 2 times as above with 300 µl of xylene to remove dissolved LCM cap polymers, taking care to not disturb the fragile tissue pellet. 4. The sample is rehydrated by washing once with 400 µl of absolute ethanol to remove xylene, followed by one wash with 95% ethanol, one wash of 70% ethanol, and two washes with TE (10 mM Tris, 1 mM EDTA, pH 8.0) for rehydration, always leaving approximately 50 µl of residual solution to minimize accidental removal of tissue. 5. The sample is finally washed once with 400 µl of proteinase buffer (50 mM Tris, 10 mM NaCl, 10 mM EDTA, pH 8.0) and approximately 100 µl is left in the tube. 6. Proteinase K is added (50 µl at 0.015% to a final concentration of 0.005%). The digest is then incubated for 30–60 min at 37◦ in a water bath, with gentle finger vortexing approximately every 10 min. For FISH analysis, the sample is diluted with 350 µl of TE, then centrifuged at 10,000g for 2–3 min followed by careful removal of as much supernatant as possible without disturbing the pellet. 7. The pellet is then gently resuspended in approximately 20 µl of 10 mM Tris pH 8.0 by pipetting with a wide-bore pipette tip, and the suspension is pipetted onto clean microscope slides (1–20 µl depending on the concentration of nuclei). Circling the site of sample application with a diamond pencil on the underside of the slide is helpful to later visualization. The slides are allowed to thoroughly air dry, then fixed in fresh methanol : glacial acetic acid (3 : 1) for 5 min, air dried, and baked at 37◦ for at least 4 hr prior to hybridization or storage. Slides are stored at −20◦ in slide boxes sealed within hybridization bags containing Drierite dessicant. It is important to bring the slides to room temperature prior to opening the storage bag. FISH is performed on the LCM and touch preparations according to our previously published methodology.12,13 Fluorescent images are captured with a 13
D. Demetrick, Mod. Pathol. 9, 133 (1996).
66
BASIC PRINCIPLES
[6]
FIG. 1. LCM preparation of “naked nuclei.” Cells were harvested from 30-µm sections of hematoxylin-stained breast carcinoma by LCM. The nuclei were purified using organic solvents, rehydration, and proteinase K digestion as described in the Nuclear Isolation Procedure section, dropped onto clean glass slides, and fixed. The specimens were photographed under phase contrast with the “naked nuclei” indicated by arrows. Inset shows a higher power view of the nuclei. Scale bars are as indicated.
PXL1400 cooled CCD camera (Photometrics, Phoenix, AZ) using Electronic Photograpy software (BioDx, Pittsburgh, PA). Bright-field and phase contrast images are captured with a Spot 1 digital camera (Diagnostic Instruments, Inc.). Discussion Evaluation of the preparation following isolation of the nuclei using phase contrast microscopy is highly recommended (Fig. 1). A loss of up to half the nuclei during the FISH procedure is not uncommon and is likely due to detachment from the glass slide during DNA denaturation. The size of the nuclei significantly affects the optimal thickness of the specimen. If the specimens are too thin, nuclear slicing occurs and the yield of whole nuclei following LCM is very low. FISH evaluation allows good visualization of normal gene copy numbers in interphase nuclei with very low background or autofluorescence (Fig. 2). Gene amplification is obvious (Fig. 3). We have used this technique on either formalin-fixed paraffinembedded normal breast or carcinoma sections, and it has proved to be reproducible among different investigators in our laboratory. We have also utilized paraffin tissue microarrays for FISH. While the paraffin microarrays save considerable time and effort with an optimal gene target, in our experience detection of low-level gene
[6]
LCM-FISH OF PARAFFIN SECTIONS
67
FIG. 2. LCM/FISH of breast carcinoma cells with normal p53 copy number. 30-µm paraffin sections of hematoxylin-stained breast carcinoma were subjected to laser capture of carcinoma cells. The nuclei were prepared and subjected to FISH using a PAC genomic DNA probe containing the p53 gene (red signals). DAPI-stained (blue) nuclei show two p53 signals per interphase nucleus.
amplification or assessment of gene deletions was very difficult (data not shown). As well, the use of a weak probe will also make interpretation of tissue microarray FISH difficult. Although a more laborious technique than tissue microarray FISH, LCM/FISH interpretation is much more obvious for low copy number assessment. The choice of tissue will dictate the thickness of the paraffin sections required for LCM/FISH, as large nuclei will necessitate thicker sections to preserve yield. We have been able to use 15- to 20-µm sections with some specimens, which offers superior morphology for dissections, and still had an acceptable yield, but 30-µm sections give a good yield with almost all carcinoma specimens that we have
68
BASIC PRINCIPLES
[6]
FIG. 3. LCM/FISH of infiltrating duct carcinoma cells. 30-µm paraffin sections from a case of infiltrating duct carcinoma were subjected to laser capture of cells or cell groups from formalin-fixed breast cancer tissue. The nuclei were prepared and subjected to FISH using c-erbB-2 (red signal) and an α-satellite centromeric probe for chromosome 17 (green signals). Normal copy numbers would be 2–4 individual small signals per interphase nucleus. Although both probes show an amplified signal, the α-satellite sequences are normally present in many copies, while the c-erbB-2 gene is abnormally amplified.
tried. When establishing this technique in one’s laboratory one should validate the technique first on thicker sections and then, if desired, on thinner sections (15–25 µm). Bear in mind that some microtomes may be significantly inaccurate when specifying specimen thickness during cutting of the paraffin blocks, and section thickness errors will lead to decreased nuclear yields.
[6]
LCM-FISH OF PARAFFIN SECTIONS
69
Troubleshooting Tips 1. Follow the Arcturus protocols for deparaffinizing and hematoxylin staining the sections, then dehydrate back through xylene. Dehydration and delipidation seem to be important in ensuring good adhesion to the LCM polymer film, so use fresh xylene in at least the last deparaffinization bath. 2. Visualization of the morphology of thick sections can be difficult. Although the use of the PixCell diffuser improves visualization during dissection, one risks scratching the tissue section with the cap holder, or wasting a cap to evaluate appropriate areas to capture before starting the dissection. A thin sheet of good quality photocopier paper or a thin business card laid on top of the slide will allow good visualization of the section without requiring use of the diffuser (personal communication, D. Billings, Arcturus, 2001). 3. The proteinase K digestion allows a great deal of latitude. If the yield of nuclei is poor, try digesting the tissue longer or with more enzyme. 4. One must try to avoid capturing large amounts of collagen during the LCM. The thickness of the section ensures that extraneous material will be removed with the LCM cap if a strong band of collagen is captured. Although adhesion to adhesive tape may help remove this extraneous material from the LCM cap, avoiding the capture of nonessential collagen is the best strategy. 5. Ensure that centrifugation speed is enough to pellet the nuclei but not so much as to distort or destroy the nuclei. 6. Store slides for several days prior to FISH as this ensures both better retention of nuclei to the glass and better quality (higher signal-to-noise) FISH signals. 7. Use very dilute DAPI (e.g., 40 pg/ml in McIlwaine’s buffer) to counterstain the FISH slides, as overly bright nuclei will interfere with FISH interpretation. Interphase nuclei may stain differently from these published conditions because of both the preparation and the particular source of DAPI, so be prepared to optimize the staining experimentally by varying DAPI concentrations. Acknowledgments This work was funded by the Canadian Breast Cancer Research Initiative and the Alberta Cancer Board. D.J.D. thanks the CIHR (Clinician–Scientist Program) for salary support and also gratefully acknowledges the Canadian Foundation for Innovation, the Alberta Science and Research Authority, and the Terry Fox Foundation for equipment funding. The authors are also grateful to Calgary Laboratory Services for salary support to D.J.D. and L.D., as well as for access to the clinical specimens necessary to develop this technology.
70
[7]
BASIC PRINCIPLES
[7] Immunoblotting of Single Cell Types Isolated from Frozen Sections by Laser Microdissection By LIVIA CASCIOLA-ROSEN and KANNEBOYINA NAGARAJU Introduction Immunohistochemistry is the commonly used method of choice to detect proteins in tissue specimens. This technique has been successfully used to investigate the protein expression in individual cell types in various tissues under a variety of physiological and pathological conditions. One of the major disadvantages of this method is that it is at best semiquantitative. Another frequently used approach is to dissociate various cell types from tissues by enzymatic (e.g., trypsin or collagenase), immunological (antibody coated magnetic beads), and mechanical methods, followed by culture of these cells and detection of specific proteins. Studies have clearly shown that primary mammalian cells experience both extrinsic and intrinsic stresses that alter gene expression, growth, and senescence when the cells are explanted into in vitro culture conditions.1 Hence the protein expression profiles observed in cultured cells may differ from those of cells in intact tissues. The optimal approach is therefore to isolate the individual cell types directly from tissues and to quantitatively analyze the protein expression patterns in these tissues. Analysis of proteins expressed in tissue samples is complex partly because of the heterogeneous nature of the cell types present. This complexity is more pronounced in diseased tissues because several abnormal cell types coexist with normal cells, and protein expression may differ in normal and abnormal cells. Studying proteins expressed in individual cell types in diseased tissues may yield important insights into the perturbed state in abnormal cells, and, ultimately, lead to the development of new diagnostic, prognostic, and therapeutic markers for diseases. The laser capture microdissection (LCM) technique has been successfully used to isolate single cells to investigate gene expression using reverse transcriptase– polymerase chain reaction (RT-PCR), a technique able to amplify a few transcripts several thousandfold.2 In contrast, it remains difficult to investigate the protein profiles in single cells because techniques to amplify single target protein sequences do not yet exist. Cells dissected using LCM have been used for proteomic analysis, including quantitative measurements of protein antigens isolated from pure populations of stained tissue cells.3–6 1
C. J. Sherr and R. A. DePinho, Cell 102, 407 (2000). K. Schutze and G. Lahr, Nat. Biotechnol. 16, 737 (1998). 3 R. E. Banks, M. J. Dunn, M. A. Forbes, A. Stanley, D. Pappin, T. Naven, M. Gough, P. Harnden, and P. J. Selby, Electrophoresis 20, 689 (1999). 2
METHODS IN ENZYMOLOGY, VOL. 356
Copyright 2002, Elsevier Science (USA). All rights reserved. 0076-6879/02 $35.00
[7]
IMMUNOBLOTTING OF SINGLE CELL TYPES
71
In this chapter, the isolation and analysis of skeletal muscle fibers and infiltrating lymphocytes from biopsies of patients with autoimmune inflammatory myopathies are described. Isolation of muscle cells is particularly well suited to LCM because (i) the cells are relatively big, (ii) they have clear boundaries and are regularly shaped, and (iii) the same muscle fibers can be easily tracked and dissected from serial muscle sections. Pooling of microdissected single cells of any one type from a single section is necessary to generate enough material to be detected by immunoblotting. Serial muscle sections have the additional advantage that the same cells can be microdissected from sequential sections, enabling the number of cells in a pooled sample consisting of a single cell type to be significantly increased. Tissue Preparation and Laser Microdissection of Individual Cell Types from Frozen Muscle Sections Detailed below are guidelines for processing skeletal muscle biopsies for LCM and analysis of proteins in pooled cells of a single type. Although the principles of the approach are applicable to other tissues and cell types, the details may require optimizing in different systems. Protein recovery from microdissected tissues depends on the initial processing of the tissues. Formalin-fixed paraffin-embedded material is not suitable for protein analysis studies. We have successfully isolated proteins from frozen tissues, which were properly snap frozen and stored at −80◦ up to 12 months. If it is not known how the tissue was handled during initial processing, starting with freshly frozen tissues is recommended. Reagents Phosphate buffered saline (PBS) without Ca2+ or Mg2+ (pH 7.4), isopentane (2-methylbutane) (Sigma), liquid nitrogen or dry ice and acetone, aluminum foil, OCT compound (Sakura Finetek, USA Inc., Torrance, CA), cryostat, uncoated microscope glass slides, cryomolds, Mayer’s hematoxylin (Dako, Carpinteria, CA), eosin Y, bluing agent (NH4OH, 0.8% in deionized water), ethanol, xylene, PixCell laser capture microdissection system (Arcturus Engineering, Mountain View, CA), caps (CapSure TM TF-100 transfer film, 5 mm-diameter optical-grade transparent plastic), self-stick note pads (The Blind and Visually Impaired, Inc., Rochester, NY). 4
L. C. Lawrie, S. Curran, H. L. McLeod, J. E. Fothergill, and G. I. Murray, Mol. Pathol. 54, 253 (2001). 5 N. L. Simone, C. P. Paweletz, L. Charboneau, E. F. Petricoin III, and L. A. Liotta, Mol. Diagn. 5, 301 (2000). 6 N. L. Simone, A. T. Remaley, L. Charboneau, E. F. Petricoin III, J. W. Glickman, M. R. EmmertBuck, T. A. Fleisher, and L. A. Liotta, Am. J. Pathol. 156, 445 (2000).
72
BASIC PRINCIPLES
[7]
Method 1. Immediately after biopsy, rinse the tissue in cold PBS and blot dry with paper towel. 2. Using forceps, place the tissue in aluminum foil and plunge into a beaker containing isopentane chilled in liquid nitrogen. At ∼−150◦ , isopentane is a highly viscous transparent liquid. Keep the tissue under isopentane for ∼20 sec to allow complete, rapid freezing. Do not directly immerse into liquid nitrogen as this may cause nitrogen to boil and the resulting gas bubbles may insulate the specimen and slow its quenching. Alternatively, tissues can be frozen in slurry of dry ice and acetone at −70◦ . Note that slow freezing over minutes or hours allows the formation of ice crystal artifact. Ice crystal growth occurs relatively slowly below −40◦ and does not occur at liquid nitrogen temperatures. 3. At this stage, frozen specimens can be stored at −80◦ in a freezer until transferred to the cryostat for the preparation of frozen sections. 4. Chill the empty cryomold on dry ice for a minute. Embed the stored tissue, properly oriented, in OCT compound. Alternatively, tissues can be directly embedded after step 1 on chilled empty cryomold in OCT compound, then snap frozen in liquid nitrogen and stored wrapped in aluminum foil at −80◦ . 5. Cut 8-µm-thick serial frozen sections with a cryostat, and mount these at the center of clean, uncoated plain glass slides. Avoid charged (e.g., poly-L-lysine) or coated (gelatin or chrome alum) glass slides as these may interfere with the LCM transfer of tissues. The number of serial sections needed depends on the cell types to be isolated and the abundance of the protein(s) to be immunoblotted in these cells. Generally, 10–15 serial sections give a sufficient number of cells for detection of specific proteins by immunoblotting from a typical muscle biopsy (a 0.6 by 0.6 cm specimen). 6. After cutting, sections are immediately placed and kept on dry ice until stained. It is preferable to stain and dissect the frozen sections individually to avoid degradation of proteins. If the target cells are readily identifiable and quick to isolate (e.g., skeletal muscle cells), it is technically feasible to stain two sections at once. The second section can be kept at room temperature for 2–3 min while the first section is being dissected. 7. Briefly (5 sec) air dry the section(s) and follow the staining procedure outlined in Fig. 1. Some investigators have used complete protease inhibitor cocktails in hematoxylin and eosin solutions to prevent degradation of proteins during staining.3 Rapid toluidine staining is another staining method of choice for LCM specimens.4 8. Allow the xylene to evaporate completely. Enter the sample and cap information into the image archiving workstation of the PixCell laser capture microdissection system. Place the stained slide on the microscope. The H&E stained sections generally appear darker because of the absence of refractive index
[7]
IMMUNOBLOTTING OF SINGLE CELL TYPES
73
FIG. 1. Flow chart of rapid H&E staining procedure for LCM.
matching provided by resin and coverslip. When isolating cells from serial sections, identify the cell types and areas in the tissue before starting laser transfer to avoid unnecessary time delay in acquiring cells. 9. Adjust the position of the tissue under a fixed laser beam using the joystick. The diameter of the laser beam of the LCM is between 7.5 and 30 µm. Select the spot size depending on the type of the cell to be dissected and capture the cells or areas of interest by activating the transfer film by laser. The following settings work well for skeletal muscle cells: spot diameter 30 µm, pulse duration 50 msecs and pulse power 50 mW. Multiple cells of the same type from the same section can be collected onto a single cap; a separate cap is used to collect the same cells dissected from serial sections. Separate caps are also used to isolate different cell types in each tissue section. 10. Immediately after isolation, fix each cap into an Eppendorf microcentrifuge tube and store these on dry ice until the samples are lysed as described below. Contamination by other adjacent cell types during LCM transfer may be a problem in certain tissues. Visual evaluation of the cells transferred onto the cap is advised as it gives a rough estimate of contaminating cell types and the extent of contamination. For example, a typical LCM transfer of abnormal skeletal muscle
74
BASIC PRINCIPLES
[7]
FIG. 2. Visual evaluation of LCM transfer of abnormal muscle cells from an H&E stained skeletal muscle biopsy from a myositis patient. Section before LCM (A), section after LCM (B), and transferred abnormal cells on the cap (C). The muscle cells marked with asterisks (∗) were selectively transferred for analysis (copyright 2000, The American Association of Immunologists) [K. Nagaraju, L. CasciolaRosen, A. Rosen, C. Thompson, L. Loeffler, T. Parker, C. Danning, P. J. Rochon, J. Gillespie, and P. Plotz, J. Immunol. 164, 5459 (2000)]. Note the contaminating lymphocytes adhered to the periphery of the dissected muscle cells on the cap (arrows in panel C).
cells from a myositis muscle biopsy is shown in Fig. 2. The abnormal muscle cells to be dissected from the biopsy (cells marked with asterisks) are densely surrounded by packed infiltrating mononuclear cells. After LCM transfer, the cells on the cap were inspected and a small amount of contamination with surrounding lymphocytes was noted (Fig. 2C, cells marked with arrows). The number of these loosely attached lymphocytes can be reduced by gently touching the cap to the sticky surface of self-stick note pads. Such contamination is not a major problem in areas of the biopsy with less lymphocyte infiltration. Another method of quantitative evaluation of contamination is to immunoblot for a marker that is specifically expressed by the contaminating cell type (for example, CD3 for lymphocytes). Techniques are currently being developed to reduce the nonspecific adherence and contamination by changing the contact force and contact area of the transfer film on the specimen.7 To date, there are no systematic studies addressing whether the H&E staining procedure and/or heat generated during laser activation cause protein damage or affect protein recovery and detection. We have addressed this using serial frozen muscle tissue sections as follows: (i) sections, either stained or unstained (but not microdissected), were lysed and immunoblotted and (ii) stained sections, subjected to LCM or not, were lysed and immunoblotted. In the tissue tested (muscle) and for the proteins immunoblotted [FLICE inhibitory protein (FLIP), vinculin, Mi-2, poly(ADP-ribose) polymerase (PARP)], neither staining nor the process of LCM altered protein detection.8 7
C. A. Suarez-Quian, S. R. Goldstein, T. Pohida, P. D. Smith, J. I. Peterson, E. Wellner, M. Ghany, and R. F. Bonner, Biotechniques 26, 328 (1999). 8 K. Nagaraju, L. Casciola-Rosen, A. Rosen, C. Thompson, L. Loeffler, T. Parker, C. Danning, P. J. Rochon, J. Gillespie, and P. Plotz, J. Immunol. 164, 5459 (2000).
[7]
IMMUNOBLOTTING OF SINGLE CELL TYPES
75
FIG. 3. Infiltrating lymphocytes, normal muscle cells, and abnormal muscle cells were microdissected from five serial frozen sections made from a muscle biopsy obtained from a patient with myositis. The same cells were dissected from each section. The pooled samples were electrophoresed on a 10% SDS–PAGE minigel, transferred to nitrocellulose, and immunoblotted with a mouse monoclonal antibody against vinculin (Sigma) or a rabbit anti-FLIP polyclonal antibody (Alexis Biochemicals, San Diego, CA). After exposure of the autorad for 10 min, both FLIP and vinculin were readily detected in the infiltrating lymphocytes (not shown).
Generation of Lysate from Microdissected Samples Since the amount of protein obtained from a single microdissected frozen section is too small to be detected by immunoblotting, cells that are microdissected from serial sections are pooled into a single sample for immunoblotting. In the case of frozen sections made from muscle biopsies (see Fig. 3), abnormal-looking muscle cells were laser microdissected from five serial sections, with ∼10 cells from each section being pooled onto a single cap (the same cells were dissected from the five serial sections). The contents of each of the five caps therefore represent the abnormal muscle cells dissected from one section. Normal muscle cells and infiltrating lymphocytes are similarly dissected onto a series of caps. Reagents Lysis buffer: 1% Nonidet P-40, 20 mM Tris (pH 7.4), 150 mM NaCl, 1 mM EDTA 5× Gel application buffer: 10% sodium dodecyl sulfate (SDS), 20% glycerol, 200 mM Tris (pH 6.8), 3 mg bromphenol blue. Immediately before use on lysates, 5% 2-mercaptoethanol (v/v) is added to this buffer.
76
BASIC PRINCIPLES
[7]
Method 1. Unless otherwise specified, all steps are performed on ice. 2. Remove the samples from −80◦ freezer storage and thaw them on ice. Because the cells are on the caps of each microfuge tube, add 20 µl of lysis buffer directly onto the cap of the first microfuge tube containing abnormal muscle cells. Aspirate well but gently using a P-20 pipette, then transfer this lysate onto the cap of the second microfuge tube containing abnormal muscle cells. Continue in this way until the contents of all five caps containing abnormal muscle cells have been lysed in the 20 µl of lysis buffer used on the first set of cells. Similarly, normal muscle cells and infiltrating lymphocytes (from the same five serial sections) are also each lysed in 20 µl of lysis buffer. 3. Add 5 µl of 5× gel application buffer to each sample and boil for 3 min. The samples can be used immediately for SDS–polyacrylamide gel electrophoresis (SDS–PAGE) or can be stored at −20◦ until use. SDS–PAGE Gel samples prepared from microdissected sections are electrophoresed on minigels, as the amount of protein available is small. Minigel apparatuses are commercially available (e.g., from Bio-Rad, Hercules, CA) and the gels can be purchased precast or can be made in the laboratory. The percentage gel used depends on the molecular weight of the protein being probed. In the examples shown in Fig. 3, both FLIP (55,000) and vinculin (115,000) are electrophoresed well on 10% SDS–PAGE. However, smaller proteins are better resolved on higher percentage gels and vice versa. Reagents for 10% SDS–PAGE Resolving Gel with a 3% Stacking Gel Resolving gel: 10% SDS (102 µl), 1.875 M Tris pH 8.8 (2.1 ml), 30% (w/v) acrylamide: 0.8% (w/v) bisacrylamide solution (3.5 ml), H2O (4.8 ml). Immediately before pouring the gel, 10% ammonium persulfate (70 µl) and TEMED (6.5 µl) are added to polymerize the mixture. Stacking gel: 10% SDS (40 µl), 1 M Tris pH 6.8 (502 µl), 30% (w/v) acrylamide: 0.8% (w/v) bisacrylamide solution (400 µl), H2O (3.06 ml). Immediately before pouring, 10% ammonium persulfate (32 µl) and TEMED (4 µl) are added to polymerize the mixture. Electrode buffer: 25 mM Tris, 192 mM glycine, 0.1% SDS Suitable protein molecular weight standard: e.g., broad-range SDS–PAGE molecular weight standards (Bio-Rad, Hercules, CA). Method 1. Clean and assemble plates per manufacturer’s directions. Pour in resolving gel and gently overlay the gel with water.
[7]
IMMUNOBLOTTING OF SINGLE CELL TYPES
77
2. When the resolving gel has polymerized, remove the water, insert a gel comb, and pour in the stacking gel. 3. When the stacking gel has polymerized, remove the comb and mount the gel in the minigel apparatus. 4. Add electrode buffer to the upper and lower reservoirs. 5. Load the gel samples into each lane, including a protein molecular weight standard. Run the gel using conditions recommended by the minigel apparatus manufacturer (e.g., 200 V) until the dye front is just above the bottom of the gel. Transfer of Electrophoresed Samples onto Nitrocellulose Several different membranes (e.g., nitrocellulose and immobilon) can be used to transfer proteins for subsequent use in immunoblotting. For the vinculin and FLIP blots shown in Fig. 3, 0.2 µm nitrocellulose (Schleicher and Schuell, Keene, NH) was used. Minigel transfer apparatuses are commercially available (e.g., from Bio-Rad). Reagents Transfer buffer: 20% methanol, 200 mM glycine, 25 mM Tris Ponceau stain: 0.3% Ponceau (Sigma) (w/v) in 3% trichloroacetic acid Method 1. Assemble the gel to be transferred in the apparatus, add transfer buffer, and transfer using the conditions recommended by the manufacturer (e.g., 100 V for 1 hr). 2. Transferred proteins on the membrane can be visualized by staining with a water-soluble stain such as Ponceau. Immerse the membrane in Ponceau stain (diluted 1 : 1 with water before use) for a few minutes, and then rinse the Ponceau off with water. Ponceau staining is very useful as a guide for cutting membranes containing transferred proteins. For example, in the vinculin and FLIP blots shown in Fig. 3, the nitrocellulose membrane was cut along the 66,000 molecular weight standard marker (markers were electrophoresed on either side of the experimental samples so that they could be used as a guide for cutting after transfer). The upper and lower portions of the transfer were used to blot vinculin and FLIP, respectively.
Detection of Proteins by Immunoblotting Several different protocols can be used to immunoblot. The method described below, which was used to detect vinculin and FLIP (see Fig. 3), involves blocking with BSA, the use of wash buffer containing NP-40 detergent, and detection
78
BASIC PRINCIPLES
[7]
of the immunoblotted bands by enhanced chemiluminescence (ECL). Other frequently used blotting reagents include blotto (dried milk powder) for blocking, and Triton X-100 or polyoxythylene–sorbitan monolaurate (Tween 20) as detergents in the wash buffer. Because not all antibodies immunoblot equally well using all protocols, blotting conditions should be optimized for each different antibody used. Detection using ECL is recommended when performing immunoblots on laser microdissected samples as it is extremely sensitive (capable of detecting picogram amounts). Ultrasensitive ECL kits are also commercially available (e.g., Supersignal West Femto Chemiluminescent Substrate, Pierce, Rockford, IL). We have not found the use of the ultrasensitive ECL detection system to be necessary for detection of the relatively abundant antigens vinculin and FLIP, but for proteins expressed in cells at low levels it will likely be useful. Reagents and Solutions Wash buffer: 10 mM Tris (pH 7.6), 0.5% NP-40, 150 mM NaCl Blocking buffer: 3% BSA (Sigma) dissolved in wash buffer, 0.02% sodium azide Primary antibodies and appropriate horseradish peroxidase (HRP)-conjugated secondary antibodies ECL detection system (e.g., Supersignal West Pico Chemiluminescent Substrate from Pierce) Method 1. All steps are performed at room temperature, with agitation on an orbital shaker. 2. Block transfer by incubating the membrane with blocking buffer for 30– 60 min. 3. Remove blocking buffer and add primary antibody diluted in wash buffer. Incubate for 60 min. Sodium azide (0.02%) is added to the primary antibody dilutions so that they can be stored at 4◦ and reused several times. Blocking buffer can be reused once. 4. Wash blot with wash buffer (1 quick wash, and three washes of 5 min per wash). 5. Incubate with HRP-conjugated secondary antibody diluted in wash buffer for 30 min. 6. Wash blot with wash buffer (one quick wash, one 15-min wash, two 5-min washes) 7. Perform ECL reaction according to the manufacturer’s directions. 8. Expose X-ray film for the length of time that gives best visualization of the blotted bands.
[7]
IMMUNOBLOTTING OF SINGLE CELL TYPES
79
Results In the examples shown in Fig. 3, approximately equal numbers of abnormal and normal muscle cells were loaded in each gel lane. The precise number of infiltrating lymphocytes (which are much smaller than muscle cells) dissected from the same sections was impossible to determine accurately, but likely was ∼5–10 times greater than the number of muscle cells. An alternative approach is to load equal amounts of lysate protein in each gel lane; however, performing a protein assay is not technically feasible with laser microdissected samples as the amount of protein is so limited. Since different cell types contain different amounts of protein, one way to quantitate the relative levels of the protein of interest (e.g., FLIP) in the various microdissected cell populations is to express these relative to the level of a “housekeeping” protein (e.g., vinculin). To this end, optimal exposures of autorads are scanned and the optical density readings of FLIP/vinculin are calculated for each sample. Concluding Remarks There are several practical and technical limitations (e.g., size and shape of cells in sectioned tissues, suboptimal optical resolution of uncovered tissues, nonspecific adherence of neighboring cells, and the inability of current protein detection methods to detect minute amounts of lysed proteins) on achieving quantitative determination of proteins from laser dissected single cells. For studying diseased tissues, the pathogenetic process can be well probed at the protein expression level by immunoblotting proteins from pooled, rather than single, abnormal and normal cells isolated by LCM. Comprehensive analysis of events at the protein level in a diseased tissue may help in diagnosis, understanding the pathogenesis, and designing better therapeutic measures. Several studies have clearly demonstrated the potential use of LCM in proteomic studies.3,5,8–11 Acknowledgments These studies were supported by the National Institutes of Health Grant AR 44684 (to L. CasciolaRosen), and by an Arthritis Investigator Award from the National Arthritis Foundation (to K. Nagaraju). Both investigators are recipients of research grants from the Arthritis Foundation, Maryland Chapter, MARRC Program. 9
D. K. Ornstein, C. Englert, J. W. Gillespie, C. P. Paweletz, W. M. Linehan, M. R. Emmert-Buck, and E. F. Petricoin III, Clin. Cancer Res. 6, 353 (2000). 10 F. Fend, M. R. Emmert-Buck, R. Chuaqui, K. Cole, J. Lee, L. A. Liotta, and M. Raffeld, Am. J. Pathol. 154, 61 (1999). 11 D. K. Ornstein, J. W. Gillespie, C. P. Pawelewtz, P. H. Duray, J. Herring, C. D. Vocke, S. L. Topalian, D. G. Bostwick, W. M. Linehan, E. F. Petricoin III, and M. R. Emmert-Buck, Electrophoresis 21, 2235 (2000).
80
[8]
BASIC PRINCIPLES
[8] Noncontact Laser Catapulting: A Basic Procedure for Functional Genomics and Proteomics By GABRIELA WESTPHAL, RENATE BURGEMEISTER, GABRIELE FRIEDEMANN, AXEL WELLMANN, NICOLAS WERNERT, ¨ , VOLKER WOLLSCHEID, BERND BECKER, THOMAS VOGT, RUTH KNUCHEL ¨ WILHELM STOLZ, and KARIN SCHUTZE Introduction Pure sample preparation is one of the most challenging tasks in modern molecular biology and medicine. Laser microdissection and laser pressure catapulting (LMPC) with a focused nitrogen laser enable the transfer of selected cells without any mechanical contact. This amazing laser technique uses only the force of an extremely focused, nanosecond laser pulse to eject a selected specimen from the object slide or from a living cell layer (see [3], this volume) and to directly catapult it into the cap of a routine collection tube. Depending on the nature of the sample, laser microdissection may be performed prior to catapulting to isolate the selected specimen from its surroundings, thus warranting pure sample preparation. Noncontact laser microdissection and sampling of a specimen as small as filaments or chromosomes and also the capture of single cells or entire tissue areas are frequently performed in numerous research institutes or industrial laboratories around the world. There is no heat involved in this laser technique, and the applied laser wavelength neither affects the viability of the living specimen nor destroys the biological information of fixed cells or tissue. Thus, subsequent molecular genetic examinations can be carried out immediately after transfer of the catapulted material into an appropriate reaction vial. In addition, the same laser system can be used to microinject drugs or genetic material into living cells without harming their viability. This enables genetic engineering without mechanical tools or viral vectors. With this unique combination of microdissection and catapulting the noncontact extraction of selected living cells or homogeneous tissue samples has become one of the most interesting techniques for functional genomics and proteomics, as the catapulted samples are surely derived from a morphologically defined origin. History of Laser Micromanipulation and Catapulting Soon after the advent of lasers,1 a great deal of work was done in cell and developmental biology to demonstrate the feasibility of laser micromanipulation as well 1
M. Bessis, F. Gires, G. Mayer, and G. Nomarski, CR Acad. Sci. 255, 1010 (1962).
METHODS IN ENZYMOLOGY, VOL. 356
Copyright 2002, Elsevier Science (USA). All rights reserved. 0076-6879/02 $35.00
[8]
NONCONTACT LASER CATAPULTING
81
as the harmlessness of the applied laser wavelength to living objects or to the molecular biological information within fixed cells.2–5 The first application describing laser microdissection of tissue slices was published in 1976 by Meier-Ruge et al.6 As an extension of this work microdissection was used to isolate single renal cells from freeze-dried sections for subsequent genetic analysis.7 Laser microdissection of human chromosomes was reported in 1986 by Monajembashi et al.8 However, those laser systems required extensive hardware and intensive maintenance from skilled personnel. With the development of small and compact microlaser systems an increasing number of biologists and medical researchers gained access to laser micromanipulation and microdissection. The combination of laser microdissection and laser trapping using a UV-A laser MicroBeam and NIR (near infrared) Optical Tweezers enabled interesting experiments in the field of cell and developmental biology such as blastomer fusion and sperm manipulation.9–12 In 1996 the first UV-A laser-assisted capture of small samples from routinely formalin fixed, archival tissue for subsequent genetic analysis was reported.13,14 The real breakthrough in laser-based microdissection and sample recovery came with laser pressure catapulting (LPC), a unique and fascinating technology that enables the transfer of the selected cells without any mechanical contact, but solely by the force of focused laser light.15 Since then, much research has been
2
K.-O. Greulich and G. Weber, J. Microsc. 167, 127 (1991). M. W. Berns, J. Aist, J. Edwards, K. Strahs, J. Girton, P. McNeill, J. B. Rattner, M. Kitzes, M. Hammer-Wilson, L. H. Liaw, A. Siemens, M. Koonce, S. Peterson, S. Brenner, J. Burt, R. Walter, P. J. Bryant, D. van Dyk, J. Coulombe, T. Cahill, and G. S. Berns, Science 213, 505 (1981). 4 A. deWith and K.-O. Greulich, J. Photochem. Photobiol. B 30, 71 (1995). 5 K.-O. Greulich, U. Bauder, S. Monajembashi, N. Ponelies, S. Seeger, and J. Wolfrum, Labor 2000, 36 (1989). 6 W. Meier-Ruge, W. Bielser, E. Remy, F. Hillenkamp, R. Nitsche, and R. Unsold, Histochem. J. 8, 387 (1976). 7 Y. Kubo, F. Klimek, Y. Kikuchi, P. Bannasch, and O. Hino, Cancer Res. 55, 989 (1995). 8 S. Monajembashi, C. Cremer, T. Cremer, T. Wolfrum, and K.-O. Greulich, Exp. Cell Res. 167, 262 (1986). 9 K.-O. Greulich, “Micromanipulation by Light in Biology and Medicine.” Birkh¨ auser Verlag, Basel, 1999. 10 K. Sch¨ utze and A. Clement-Sengewald, Nature 368, 667 (1994). 11 A. Clement-Sengewald, K. Sch¨ utze, A. Ashkin, G. A. Palma, G. Kerlen, and G. Brehm, J. Assist. Reprod. Gen. 13, 259 (1996). 12 A. Clement-Sengewald, K. Sch¨ utze, S. Sandow, C. Nevinny, and H. P¨osl, in “Photomedicine in Gynecology and Reproduction” (P. Wyss, Y. Tadir, B. J. Tromberg, and U. Haller, eds.), p. 340. Karger, Basel, 2000. 13 I. Becker, K.-F. Becker, M. H. R¨ ohrl, G. Minkus, K. Sch¨utze, and H. H¨ofler, Lab. Invest. 75, 6 (1996). 14 K. Sch¨ utze, I. Becker, K.-F. Becker, S. Thalhammer, R. Stark, W. M. Heckl, M. B¨ohm, and H. P¨osl, Gen. Anal. BME, 1 (1997). 15 K. Sch¨ utze and G. Lahr, Nat. Biotechnol. 16, 737 (1998). 3
82
BASIC PRINCIPLES
[8]
done using either direct catapulting or the combination of microdissection and catapulting.16–19 Technical Setup and Basic Principles The PALM MicroBeam A pulsed nitrogen laser is coupled via the epifluorescence path to an inverted Zeiss Axiovert 200 microscope and focused through the objective. Within the narrow laser focal spot, forces are generated that enable ablation, i.e., cutting of material from animate or inanimate origin without any mechanical contact. This socalled PALM MicroBeam is equipped with a customized, extremely precise motorized microscope stage. The combination of an especially designed “RoboSoftware” and the unique PALM CapMover allows the largely automated and highly reliable capture of thousands of cells within a very short time (minutes only), thus enabling high throughput specimen sampling, which is especially important for array techniques or proteomic studies (Fig. 1). If desired, simultaneous fluorescence illumination is possible as well. The PALM MicroBeam allows laser microdissection with several objective magnifications from 5× to 100×. As the laser focus is dependent on the numerical aperture of the objective, the smallest cutting size ever measured was 300 nm using a 100× objective20 (Robert Stark, personal communication, 2001). As the 337-nm nitrogen laser works within the UV-A range of 320–400 nm, which is outside the range of absorption by biological matter, the laser catapulted specimen as well as the adjacent tissue show no detrimental effect on DNA, RNA, or protein recovery. Laser Microdissection The principle of cutting with a pulsed UV-A laser is supposed to be a photochemical or “cold ablation” process without heat formation. Within the narrow laser beam focus a power density of more than 1 megawatt/cm2 is achieved. At the site of laser cutting the unwanted material is photofragmented into atoms and small molecules, which are blown away at supersonic velocities. But, as the process is photochemical and rapid (submicrosecond), there is no lateral damage to the adjacent material. Thus laser microdissection is used to isolate a selected specimen from its surroundings or to selectively destroy unwanted objects to obtain pure and homogeneous sample preparations. 16
K. Sch¨utze and G. Lahr, Am. Biotechnol. Lab., March, 24 (1999). L. Fink, W. Seeger, L. Ermert, J. H¨anze, U. Stahl, F. Grimminger, W. Kummer, and R. M. Bohle, Nat. Med. 4, 1329 (1998). 18 G. Lahr, Lab. Invest. 80, 1477 (2000). 19 U. Lehmann, S. Gl¨ ockner, W. Kleeberger, H. Feist, R. von Wasielewski, and H. Kreipe, Am. J. Pathol. 156, 1855 (2000). 20 S. Thalhammer, R. W. Stark, and K. Sch¨ utze, J. Biomed. Optics 2, 115 (1997). 17
[8]
NONCONTACT LASER CATAPULTING
83
FIG. 1. PALM MicroBeam system for noncontact laser micromanipulation and catapulting (stage and PALM CapMover with Multi-CapHolder).
Laser Pressure Catapulting Laser pressure catapulting (LPC) appears to result from a gas pressure cloud that develops beneath the specimen due to laser-based ablation of material. LPC was discovered by accident. During the ablation procedure that cleared the area around the selected specimens the cells sometimes lost contact with their surroundings and suddenly flew away with a subsequent laser shot. Thus, it was obvious that the power of focused laser light seemed to be sufficient to dislodge selected specimens from the object plane and transport them evenly over centimeter distances along the direction of the incident laser light. This phenomenon is now used to prepare samples in an entirely noncontact, fast, and elegant manner and was named laser pressure catapulting (LPC). With a single laser shot the selected specimens are ejected from the object plane and catapulted directly into an appropriate collection vial (e.g., the cap of a standard microfuge tube) mounted within the laser beam path (Fig. 2). Pieces of
84
BASIC PRINCIPLES
[8]
FIG. 2. Principle of laser pressure catapulting (LPC): example of one single cell catapulted from a cytocentrifuged specimen (a) and visualized within the capture cap (b,c). LPC of an entire tissue area from a membrane-mount tissue section (d) arriving in the capture cap with its morphology beautifully preserved (e).
chromosomes and not only single cells but also large cell areas having a diameter of more than a millimeter have been catapulted using different objective magnifications, and even living cells or entire small organisms [e.g., Caenorhabditis elegans (Fig. 5a, p. 92)] have survived this unique catapulting procedure. Sample Features and Specimen Preparation Any routine biological sample is suitable for noncontact laser microinjection, microdissection, or laser capture. So far, no restrictions have been reported regarding the origin of the selected specimen or the applied preparation and staining procedures. Routinely fixed and paraffin-embedded and also frozen sections have been processed as well as cells cultured within a custom-made petri dish. Up to now, samples of any shape and size from 1 µm up to 1 mm in diameter have been captured. (Note: With a cryopreserved specimen the freezing medium OCT has to be removed prior to laser application, as this substance interferes with laser cutting.) Improved Visualization Due to the fact that tissue sections for laser microdissection or catapulting cannot routinely be embedded and coverslipped, their morphology sometimes appears
[8]
NONCONTACT LASER CATAPULTING
85
quite different when compared with embedded tissue sections. Several approaches to improve specimen visualization have been tried using various embedding materials. Important, however, is the convenience of substance application as well as its suitability for subsequent genetic or proteomic analysis. Several embedding materials have been tested and a few are recommended for improved visualization. Combination of Laser Microdissection and Laser Pressure Catapulting (LMPC) Depending on the nature of the sample, catapulting may be performed directly. Cytocentrifuged specimens (Fig. 3a), for example, are highly suitable for direct catapulting, but so are single cells or entire tissue areas from homogeneous or less
FIG. 3. These photos give an overview of LPC of a cytocentrifuged specimen (column a), glass-mounted specimen (column b), membrane-mounted tissue slice (column c), and chromosome metaphases spread on membrane (column d). First image of each column shows the marked specimen prior to laser microdissection. Different steps after cutting and/or catapulting are given in the second image of columns: one single shot is enough to catapult single cells (a); for a glass-mounted tissue area several laser shots are necessary (b). Membrane-mounted specimen can be cut out and be catapulted with one laser shot (c,d). Third images of columns represent the recovery of catapulted specimen in the cap of a microfuge tube.
86
BASIC PRINCIPLES
[8]
critical tissue areas. However, tissue preparations are usually inhomogeneous and consist of a mixture of different cells. To avoid contamination with unselected material it might be advisable to perform laser microdissection prior to catapulting in order to obtain pure sample capture. The laser cuts a clear gap between selected and nonselected regions. Within the narrow laser cut all biological matter is destroyed, so that nothing will contaminate the captured specimen. Within the immediate surroundings of the cut, the cells remain intact and are perfectly suitable for subsequent genetic or proteomic analysis. In addition, if there are unwanted structures or cells within a selected area they may selectively be destroyed by laser ablation prior to catapulting. Glass-Mounted Specimen The laser can only grab a small portion of material with each shot. When larger areas from a routinely glass-mounted specimen have to be collected with multiple laser shots, numerous tissue flakes within the collection cap result (Figs. 3a and 3b). However, the collected material is pure as it is solely catapulted from the selected area and thus derived from a morphologically defined origin. LPC has no impact on subsequent molecular analysis. Direct catapulting from glass is recommended for cytocentrifuged specimens, small cell clusters, and pooled single cells. Membrane-Mounted Specimen To preserve morphology, the specimen should be mounted on a thin membrane covering a routine glass slide.21 The LPC membrane serves as a backbone, which holds together the selected tissue area and facilitates the capture of large tissue areas or preparations such as cell smears, but also of fragile samples such as very small cells, cell nuclei, or chromosome preparations. The laser first cuts around the selected area following a predefined line. One single shot is now sufficient to catapult the entire selected area out of the object plane and directly into the collection vial, keeping its morphology intact (Figs. 3c and 3d). In this manner, larger areas of up to 1 mm in diameter and also single cells or fragments of chromosomes can be collected independent of their shape and nature. Unwanted material within a larger area could selectively be destroyed prior to catapulting to obtain pure samples. Comparison of samples catapulted from membrane-mounted versus glass-mounted tissue has shown that there is no difference in subsequent molecular analysis. Membrane-mounted slides can be autoclaved and used for cell cultivating. Cells can grow on the membrane and can be fixed prior to catapulting. It is also possible to catapult membrane-bound living cells. It has been shown that cells 21
M. B¨ohm, I. Wieland, K. Sch¨utze, and H. R¨ubben, Am. J. Pathol. 15, 63 (1997).
[8]
NONCONTACT LASER CATAPULTING
87
can survive the microdissection and catapulting procedure and are capable of subsequent cultivation and cloning. In this case a special double-layer petriperm dish with a special LPC-membrane layer is recommended for use. There are ready-to-use LPC-membrane slides available which have two different kinds of membranes attached that differ in their chemical compounds and laser cutting behavior and are used for different applications: the PEN (polyethylenenaphthalate) membrane is recommended for use with tissue sections, critical single cells, cell smears, blood smears, and mucus smears; the POL (polyester) membrane is recommended for laser ablation prior to cutting to avoid contamination with nonselected material: chromosomes, chromosomal parts, filaments, cell organelles, etc. Special preparations require covering the specimen with a sheet of the LPC membrane serving as a cover slip. This so-called “sandwich-prep” is recommended for either immobilization of a specimen that barely adheres to the membrane or for conserving a minimum of humidity on the specimen so it will not dry out. The procedure of catapulting can be performed in the manner described.
Specimen Collection and Recovery The catapulted specimens are simply collected within a microfuge cap or any other appropriate vial placed above the objective. The cap should be filled with a small amount of liquid to achieve reliable attachment of the catapulted sample, thus avoiding accidental specimen loss. We recommend filling the collection vial with a fine film of routine lightweight PCR mineral oil, with a small droplet of buffer, or, for example, an RNA-protective solution, which reliably preserves the RNA. After catapulting, the samples are suspended in appropriate buffers and spun down into the tube for further processing. If mineral oil is used the specimen will remain within the vial until suspended with buffer solution. This way the specimen can be kept for several days before further processing. If buffer is used, the specimen should be spun down immediately to avoid evaporation and resulting specimen loss. PALM RoboSoftware for Automated Microdissection and Catapulting The RoboSoftware controls the motion of the motorized microscope stage as well as the motor-driven laser energy and laser focus settings. Saving of the selected elements with respect to a reference position in personal files allows the relocation of the stored elements on each individual slide. Computer graphic tools (freehand lines, straight lines, ellipses or circles, rectangles or quadrats) provide the possibility of outlining the specimen of interest. The outlined areas
88 BASIC PRINCIPLES
[8]
[8]
NONCONTACT LASER CATAPULTING
89
may be numbered and/or named with a text note, and the circumscribed area can be calculated in µm2 (Fig. 4a). Different laser functions are available to process the preselected and outlined specimen. Cut: The laser cuts precisely along the predefined line yielding a clear-cut gap between the selected and non-selected material. Within the cutting line all biological material is destroyed due to ablation. Thus pure sample preparation is possible without danger of contamination. LPC (laser pressure catapulting): Only LPC dot-marked specimens are catapulted. This function is of special benefit for cytocentrifuged specimens, but is also used for individual isolated cells within a histological preparation or to individually catapult membrane-mounted specimens. AutoLPC enables the automatic catapulting of larger areas from glassmounted preparations. With glass-mounted preparations only a small amount of cellular material can be catapulted with each single shot. Therefore larger areas have to be catapulted with multiple shots. The distance between the single shots depends on the tissue characteristics which can be preselected in the Setup menu (see white dots in Fig. 3b). With the use of CloseCut&AutoLPC critical preparations can be isolated prior to AutoLPC to avoid contamination with neighboring tissue. Thus pure sample preparations can be obtained. RoboLPC means cut and catapulting in a single step! And it is only possible for use with membrane-mounted specimens. The marked line is entirely closed leaving a small connecting piece from which the entire area is immediately catapulted with one single shot. The size of the connecting piece can be preselected from the laser Setup menu and displayed together with the RoboLPC dot (Figs. 3c,d). The list of elements is the main tool for the sorting of the outlined areas and for laser activation (Fig. 4b). Choosing from the colored chart, the computer will microdissect and/or catapult only elements showing the particular color chosen. In this manner sorting is facilitated and laser capture becomes quick and easy, is therefore important for high throughput analysis and is necessary for proteomics and DNA array technology.
FIG. 4. With computer graphic tools the interesting areas of the specimen are marked and numbered. Presorting is enabled by using different colors as, for example, to discriminate between benign and malignant cell types (a). The outlined areas are displayed in the “list of elements” including area measurements (b). The list of elements is a main switchboard for sorting and laser activation. By clicking on to the colored group (red), the computer will microdissect and/or catapult only elements showing this particular color. This way sorting is facilitated, and laser capture becomes quick and easy (c). Specimens are selected this way, which are subsequently cut and catapulted automatically into the cap of a microcentrifuge tube (d).
90
BASIC PRINCIPLES
[8]
Downstream Applications With the MicroBeam many downstream applications have been established and described.22–24 In general, any common PCR and RT-PCR method can be applied to laser microdissected and catapulted specimens. The applications represent the importance of laser based noncontact and pure sample capture for functional genomic and proteomic analysis. Expressed genes from archival material: The feasibility of laser microdissection and catapulting of single cells from formalin fixed tissue and the subsequent extraction of RNA from those preserved cells in combination with the real time RT-PCR method has been demonstrated in several publications. The ability to study preserved tissues at the molecular level makes possible retrospective studies on large numbers of tissue samples and may permit tracking, over long periods of time, of genetic changes or infectious agents that are associated with the development of diseases.18,19 Working with cell smears: Mucus smears or other cell smears should be spread over the LPC membrane. The laser cuts the membrane around the cell or nucleus of interest, yielding a specimen-membrane island, which is catapulted into the cap. The membrane facilitates catapulting of those otherwise tightly bound to the glass surface specimans and allows analysis on the single cell level. Fetal cell isolation: The isolation of fetal cells from a transcervical mucus smear of a pregnant woman and the subsequent testing for Rhesus genotyping and individual genotyping with microsatellites is possible.25 Chromosome metaphase spreads: For capture of chromosomes, the metaphase spread should be brought onto the 1.0-µm thin polyester (POL) membrane. The laser first destroys surrounding chromosomal material to avoid contamination (ablation). With slightly higher laser energy the membrane is cut around the target chromosome and, subsequently, the chromosome-membrane island is catapulted into the cap.20,26 “Donut” preparation: The cells are cultivated on slides mounted with LPCmembrane and are thereafter fixed and stained. Using a 100× magnification 22
M. R. Bernsen, H. B. P. M. Dijkman, E. de Vries, C. G. Figdor, D. J. Ruiter, J. Adema, and G. N. P. van Muijen, Lab. Invest. 78, 1267 (1998). 23 S. Hahn, X. Y. Zhong, C. Troeger, R. Burgemeister, K. Gloning, and W. Holzgreve, CMLS Cell. Mol. Life Sci. 57, 96 (2000). 24 G. Vona, A. Sabile, M. Louha, V. Sitruk, S. Romana, K. Sch¨ utze, F. Capron, D. Franco, M. Pazzagli, M. Vekemans, B. Lacour, C. Br´echot, and P. Paterlini-Br´echot, Am. J. Pathol. 156, 57 (2000). 25 R. Burgemeister, K. Sch¨ utze, S. Minderer, and K.-Ph. Gloning, BIOforum Int. 3, 119 (1999). 26 L. Schermelleh, S. Thalhammer, T. Cremer, H. P¨ osl, W. M. Heckl, K. Sch¨utze, and M. Cremer, BioTechniques Int. 27, 362 (1999).
[8]
NONCONTACT LASER CATAPULTING
91
the cell nucleus is located and catapulted into vial number 1. In a second step the cytoplasm is cut and catapulted into a second vial. This way the nuclei are separated from the rest of the cell. This experiment might be useful for studying the kinetics of material transfer or viral infection.26a Polar body extraction: For prefertilization diagnosis the polar body is of interest to learn about the genetic pattern of an oocyte prior to fertilization. Polar body extraction can be performed in an entirely “noncontact” manner by laser zona drilling combined with laser trapping, but also in combination with mechanical tools. In this case, the laser drills a hole into the zona pellucida which is a fast and very gentle procedure followed by sucking the polar body out with a blunt pipette or using optical tweezers.27 Catapulting of living organisms: Recent experiments provided evidence for the possibility of even catapulting living cells for subsequent culturing or cloning (see Mayer et al., [3], this volume). Also endosymbionts from the cytoplasm of Amoeba proteus, phytoplankton, or even entire organisms, e.g., Caenorhabditis elegans, have successfully been catapulted alive (Fig. 5). In some cases a special trick has to be applied to laser-process the vigorously moving specimen, i.e., placing it in between two LPC membranes, thus forming a duplex-membrane sandwich slide. For proteomics and DNA array technology the following applications have been developed. Analysis of Microdissected Prostate Tissue with ProteinChip Arrays—for New Insights into Carcinogenesis and Diagnostic Tools Prostate cancer is one of the most common cancers in Western countries.28 It has two characteristic features: it develops within a complex epithelial–mesenchymal tissue with well coordinated intercellular cross talks29 and up to 70% of prostate carcinomas arise in the prostate proper,30 whereas only about 25% develop in the transitional zone. Since there are no obvious histomorphological differences between these zones, important molecular peculiarities of the prostate proper must be postulated which favor development of carcinoma. Early diagnosis of prostate carcinoma is of great importance. Therefore identification of novel biomarkers 26a
S. Thalhammer, A. K¨olzer, G. Fr¨osner, and W. M. Heckl, Eur. Biophys. J. 29, 12D-5 (2000). A. Clement-Sengewald, T. Bucholz, and K. Sch¨utze, Pathobiology 68, 232 (2000). 28 W. G. Nelson and T. L. DeWeese, Urology 57, 39 (2001). 29 G. R. Cunha, Cancer 74, 1030 (1994). 30 J. E. McNeal, Am. J. Surg. Pathol. 12, 619 (1989). 27
92
BASIC PRINCIPLES
[8]
FIG. 5. Catapulting of Caenorhabditis elegans: (a) 10× magnification, and marine phytoplankton; (b,c) 100× magnification. C. elegans is shown after catapulting and is creeping away from the catapulted membrane. Phytoplankton circumscribed within the sandwich stack; (b) the catapulted sandwich stack within the cap (c).
has great potential for improving tumor management. A classical way of discovering new markers on the protein level is with 2D gel electrophoresis.31,32 However, although 2D gel PAGE is able to resolve thousands of proteins, it requires large amounts of starting material. In contrast, better understanding of prostate carcinoma development requires separation of stromal and epithelial cells by laser-assisted microdissection.15 The development of surface enhanced laser desorption/ionization time of flight mass spectrometry (SELDI-TOF-MS) has overcome many limitations of matrix-assisted laser desorption/ionization time of flight (MALDI-TOF) and 2D gels.33 This technology reverses the conventional MALDI 31
P. H. O’ Farrel, J. Biol. Chem. 250, 4007 (1975). S. D. Patterson, Anal. Biochem. 221, 1 (1994). 33 H. Kuwata, T. T. Yip, C. L. Yip, M. Tomita, and T. W. Hutchens, Biochem. Biophys. Res. Commun. 245, 764 (1998). 32
[8]
NONCONTACT LASER CATAPULTING
93
sample preparation by using a ProteinChip array of addressable protein binding sites on a solid substrate. Captured individual proteins from complex mixtures are subsequently resolved by mass spectrometry. This report describes our initial studies using laser assisted microdissection in conjunction with SELDI and a new software for elaborate analysis to detect potential new diagnostic markers for prostate carcinomas in an effort to understand additional aspects of carcinogenesis. Methods Radical prostatectomy specimens were obtained from 4 patients with carcinoma (Gleason scores between 5 and 9) and 1 control. Samples from the prostate proper, transitional zone, and prostate carcinoma were immediately frozen in −180◦ liquid nitrogen. Populations of epithelial and mesenchymal cells from the different zones and tumor were isolated from frozen tissue sections using a PALM System (PALM AG, Bernried, Germany; Fig. 6). For SELDI analyses of cell lysates a total of 500 microdissected cells were lysed in 15 µl of 50 mM HEPES with 1% TRITON X-100 (pH 7.4). The lysates were homogenized by vortexing, resuspended and centrifuged for 5 min at maximum speed with a benchtop centrifuge. The spots of a SAX-2 (strong anionic exchange surface using quaternary ammonium groups) ProteinChip array were preincubated with PBS (pH 7.5). PBS was replaced after 5 min by 5 µl of the resulting supernatant after centrifugation and then incubated for 30 min at room temperature in a humidity chamber. After incubation the lysate was withdrawn and the spots washed 2 times with PBS + 0.05% TRITON, then washed twice with aqua bidest. Mass analysis was performed with the Protein Biology System II using 2 applications of saturated sinapinic acid (SPA) dissolved in 50% acetonitrile containing 0.5% trifluoroacetic acid (TFA). Twice 0.7 µl of TFA was applied to each spot after drying. ProteinChip arrays were analyzed in a ProteinChip reader according to an automated data collection protocol. The instrument was operated with a source and detector voltage of 20 and 1.8 kV, respectively. Laser intensity was set to 250, detector sensitivity to 10 with an optimization range of the time lag focusing feature between 1000 Da and 25,000 Da with a focus on the center of the optimization range at 13,000 Da. Data interpretation was augmented using the ProteinChip software version 2.1b. Results To determine sensitivity limitations of SELDI-TOF-MS we performed a pilot study and tested a panel of extracts obtained from less than 500 cells. These extracts contained enough protein lysate to generate sufficient reliable protein profiles as we observed spectra similar to those obtained from more cells. Further data were generated using the equivalent amount of protein of approximately 500 cells for binding on the spots of the SAX-2 protein chips (per extract). Processing on a
94
BASIC PRINCIPLES
[8]
FIG. 6. Laser-assisted microdissection of a normal and a carcinoma gland, and of adjacent periglandular and peritumoral stroma. Normal gland (a–d) and carcinoma gland (e–h ) and their adjacent stroma before (a,e) and after (b–d, f–h) microdissection. Hematoxylin–eosin (H&E)-stained frozen sections. Bar = 100 µm.
[8]
NONCONTACT LASER CATAPULTING
95
FIG. 7. Protein-chip SELDI-TOF mass spectrometry from 500 laser-microdissected cells (58-year old patient, strong anionic exchange surface, SAX-2, 1-5 kDa)
strong anionic exchange surface (SAX-2) resolved up to 100 peaks in the mass range between 1 and 30 kDa. Figure 7 demonstrates representative protein mass spectra (from 1 to 5 kDa) obtained with samples from a 58-year-old patient. A number of differences are evident between the epithelial and the stromal compartments of the normal prostate proper, transitional zones, and prostate carcinomas. For data analysis features of the ProteinChip software version 2.1b were used. The spectra obtained under similar experimental conditions from all 28 subpopulations analyzed on SAX-2 chips were imported to one experiment file and the peaks automatically labeled (automatic peak detection) with highest detection sensitivity. The samples were grouped related to the different subpopulations and,
96
BASIC PRINCIPLES
[8]
subsequently, combination maps were generated by the software combining all spectra of one subpopulation into one combination map. The resulting combination maps were then compared groupwise to produce comparison maps. We were able to detect a number of differentially regulated proteins in the mass range between 1.5 and 30 kDa. Discussion The search for molecular differences between specific subpopulations of cells within the prostate and prostate carcinoma will not only help to understand carcinogenesis but identify biomarkers that could support tumor screening. Laserassisted microdissection combined to the SELDI technology offers a very sensitive tool for analyzing complex protein mixtures from just a few cells from epithelial and stromal compartments of different zones of the prostate and prostate cancer. Transcriptome Profiling of Melanocytic Cells from Human Malignant Melanoma After the completion of the human genome project, molecular profiling of biopsy material has become the main challenge in molecular medicine. Using cDNA array technology a genome wide screening of gene expression profiles is now feasible. The analysis of biopsies is considered to be superior to the comparison of different cultured cell types because tissue culture may not reflect the physiologic phenotype of a given cell type in its natural microenvironment. The laser pressure catapulting (LPC) microscope allows the direct isolation of cell populations or even of single cells of interest without contaminating cells directly from the histologically HE-stained sections (Fig. 8). The yields of RNA prepared from laser microdissected tissues are limited and need to be amplified in order to generate a probe for the hybridization of cDNA-arrays.
FIG. 8. Steps in laser microdissection and LPC: (a) before, (b) after microdissection, and (c) after LPC. (a) shows a HE-stained section from a cryopreserved biopsy of a benign melanocytic nevus. The melanocytic cells are separated from the surrounding tissue by laser microdissection (b) and can be catapulted out of the section by one shot of LPC (c).
[8]
NONCONTACT LASER CATAPULTING
97
LMPC of Stained Sections from Cryo-Fixed Biopsies The material from which sections for laser microdissection and catapulting are prepared have to be snap frozen in liquid nitrogen and stored at −80◦ . The sections must be handled under RNase-free conditions. After the preparation of each section by cryostat at −25◦ , the section is mounted on the PEN foil covered slide and each slide is immediately fixed with a drop of RNA protecting solution until the next section is done. If LMPC is discontinued after staining, slides can be stored in dry ice or at −80◦ , but not for longer than 18–20 hr. RNA Isolation Best results of RNA isolation were obtained using the Purescript RNA isolation kit (# R5500-A, Gentra systems, Minneapolis, MN). The procedure should be performed as quickly as possible without any interruptions. For the final precipitation step of the RNA, glycogen is added as carrier molecule. Probe Preparation and Hybridization For the generation of probes for array hybridization an amplification of the whole transcriptome of the cells is necessary due to the small RNA amounts gained from microdissected samples. We use the SMART system from Clontech (# K1052-1, BD Clontech, Heidelberg, Germany). This method uses PCR amplification protocols (see Ref. 34 for examples), which are much easier and faster than T7-based protocols for linear amplification35,36 and therefore useful for larger screening studies in clinical samples. Amplification by PCR might produce artifacts due to normalization effects, which may appear when the PCR reaction reaches the end of the logarithmic reaction phase. We were able to demonstrate the reliability and speed of this method for the generation of reproducible probes from small amounts of RNA when the process during the PCR reaction is controlled carefully using a real time PCR machine like the Rotor-Gene (Corbett Research)37 (Fig. 9a). When the signal intensity of the PCR fragments produced reaches the turning point (the middle of the logarithmic phase of the amplification reaction), the PCR reaction is stopped by removing the sample from the PCR machine (marks in Fig. 9a). For demonstration, PCR reactions of two different samples were performed twice, i.e., one for demonstrating the performance of the PCR reaction in 34
B. Becker, T. Vogt, M. Landthaler, and W. Stolz, J. Invest. Dermatol. 116, 983 (2001). H. Ohyama, X. Zhang, Y. Kohno, I. Alevizos, M. Posner, D. T. Wong, and R. Todd, Biotechniques 29, 530 (2000). 36 O. Kitahara, Y. Furukawa, T. Tanaka, C. Kihara, K. Ono, R. Yanagawa, M. E. Nita, T. Takagi, Y. Nakamura, and T. Tsunoda, Cancer Res. 61, 3544 (2001). 37 B. Becker, T. Vogt, M. Landthaler, and W. Stolz, AACR San Francisco, 43, 231 (2002). 35
98
BASIC PRINCIPLES
[8]
FIG. 9. (a) A typical protocol during SMART cDNA amplification PCR with two different templates recorded with the Rotor-Gene real-time PCR machine. The marks A and B label the number of cycles used for equal amplification of the templates: 22 and 25 cycles, respectively. -RT represents the control reaction without reverse transcriptase during the first-strand synthesis step in the SMART protocol. The inset shows an agarose gel with the cDNA isolated from template A after 22 cycles and template B after 25 cycles; on the lane labeled -RT the negative control is shown. Equal volumes from the cDNAs and the negative control were loaded on each lane. M: DNA size marker (sizes given in kilobases). (b) Reproducibility of array hybridization with SMART-amplified RNA from LPC purified melanocytic cells from one patient biopsy (#2068). The tumor cells were prepared independently from two slides by LPC of the same biopsy. The two probes were hybridized onto two different cDNA arrays. In the corresponding sections of the array autoradiographs are shown.
[9]
INTERNAL STANDARDS FOR LASER MICRODISSECTION
99
the Rotor-Gene and the other one for agarose gel electrophoresis. This reaction demonstrates that the cDNA amounts generated from different samples starting with different amounts of template are comparable (inset in Fig. 9a; the examples represent a twofold difference in template amount used for single-strand synthesis in SMART). The labeling of the cDNA for hybridization is performed according to standard protocols.37 The hybridization follows already published conventional protocols.37 Figure 9b shows an example of the reproducibility of array hybridization with SMART amplified RNA, which was isolated from two microdissections from the same tumor biopsy.34 Array Analysis After hybridization the array is exposed to a phosphoimager screen. The autoradiography is analyzed by suitable array analysis software like the AIDA software package (Raytest, Berlin, Germany) Comments The protocols for probe labeling and hybridization recommended herein constitute one possible method of array hybridization. The method of choice depends strongly on the type of array used and may therefore vary. After cDNA synthesis, the cDNA can be used for any labeling protocol recommended by the supplier of the array or also for other protocols, which might enhance the number of hybridizations feasible (e.g., Strip-EZ DNA; #1470, Ambion, Inc., Austin, TX).
[9] Internal Standards for Laser Microdissection By LUDGER FINK and RAINER MARIA BOHLE Introduction Morphological study and the identification of cell differentiation within tissues have been the basis for assessment of physiological and pathological processes since the nineteenth century. Applying molecular techniques, the analysis of commonly used tissue homogenates bears the risk of masking genetic deviations or expression changes of an individual cell type by the bulk of surrounding cells. Neoplasms, for example, are composed of tumor cells as well as stromal components, vessels, inflammatory cells, and others. Cell cultures, on the other hand, are homogeneous populations but cannot reflect the complex interactions
METHODS IN ENZYMOLOGY, VOL. 356
Copyright 2002, Elsevier Science (USA). All rights reserved. 0076-6879/02 $35.00
100
BASIC PRINCIPLES
[9]
and cross-talk of cells in tissue settings and in contact with stromal elements. Thus, to overcome the problem of tissue heterogeneity, cells have to be harvested selectively for further analysis. Even in nonneoplastic processes such as inflammatory or degenerative diseases, separation of different cell types is desirable. Therefore, microdissection and micromanipulation techniques were developed. Using a scalpel blade or pulled glass micropipettes, several thousands down to single cells have been isolated for further PCR/ RT-PCR analysis.1–3 In particular, the laser-based microdissection systems make retrieval of target cells simple, rapid, and precise. Accordingly, the use of these techniques has increased exponentially within the last few years. The first two laser systems were introduced in 1996. Using an inverted microscope, a pulsed low-energy infrared laser was used to fuse a thermoplastic film to the selected underlying cell area. The cells become adherent to the film and both could be removed together. This type of microdissection was developed in Bethesda at NIH and is called laser capture microdissection (LCM; now distributed by Arcturus Engineering, Mountain View, CA).4 Alternatively, a pulsed ultraviolet laser is used to microdissect tissue directly (laser microbeam microdissection, LMM; distributed by P.A.L.M. Microlaser Technology, Bernried, Germany).5 The dissected cell islet is either harvested by micromanipulation or is catapulted by the laser beam into the cap of a reaction tube positioned above. This is called laser pressure catapulting (LPC) and allows the transfer of dissected cells without any further contact.6 A short survey of the different techniques is given by Walch et al.7 This paper highlights technical procedures, from tissue fixation to several subsequent analytical methods with a special focus on RNA investigation. Standard protocols for DNA or RNA preparation from frozen or formalin-fixed tissue sections are presented. Specific protocols are provided for few cell approaches as well as for higher cell scales. Most procedures can be used with any microdissection system. Additionally, special features of each system are mentioned. Standard Precautions The small amounts of nucleic acids that can be obtained from a few cells require high amplification rates by PCR. This, of course, increases the risk of contamination and, therefore, false positive results. To avoid this potential danger, 1
R. Kuppers, M. Zhao, M. L. Hansmann, and K. Rajewsky, EMBO J. 12, 4955 (1993). C. A. Moskaluk and S. E. Kern, Am. J. Pathol. 150, 1547 (1997). 3 T. Hiller, L. Snell, and P. H. Watson, Biotechniques 21, 38 (1996). 4 M. R. Emmert-Buck, R. F. Bonner, P. D. Smith, R. F. Chuaqui, Z. Zhuang, S. R. Goldstein, R. A. Weiss, and L. A. Liotta, Science 274, 998 (1996). 5 K. Schutze, I. Becker, K. F. Becker, S. Thalhammer, R. Stark, W. M. Heckl, M. Bohm, and H. Posl, Genet. Anal. 14, 1 (1997). 6 K. Schutze and G. Lahr, Nat. Biotechnol. 16, 737 (1998). 7 A. Walch, K. Specht, J. Smida, M. Aubele, H. Zitzelsberger, H. Hofler, and M. Werner, Histochem. Cell Biol. 115, 269 (2001). 2
[9]
INTERNAL STANDARDS FOR LASER MICRODISSECTION
101
tissue preparation, microdissection, extractions, and especially RT/PCR preparations should be carried out in different rooms and separated from the amplification and postamplification settings. When working with RNA, RNase-free solutions must be used (i.e., DEPC-treated H2O). Every work bench needs to be equipped with a set of tools to prevent transmission of contamination. Positive air displacement pipettes with sterile filter tips help prevent liquid and aerosol contamination.8,9 Plasticware must be kept sterile. Disposable gloves should be worn throughout the procedure and replaced when leaving the work site. Glassware and buffers should be autoclaved. All procedures should comply with standard molecular biology rules.10 Tissue Preparation Strategies for the isolation of nucleic acids involve a number of steps, which can lead to fragmentation and degradation (especially of RNA), resulting in a low recovery of an already limited amount of input material. This has to be taken into account when planning the study from the time of tissue harvesting to the analytic steps. The type of tissue preparation, storage, and fixation are not only major factors for analytical success, but also affect the workup of the microdissection studies. First, the time between onset of ischemia (i.e., during surgical maneuver for tumor excision or death of the individual) and start of fixation may vary between minutes and hours or even days. During this time at body or room temperature, enzymes such as endogenous RNases are able to degrade their targets. In our experience, this time has a greater impact on the RNA integrity than the time of tissue fixation. Thus, the time span should be kept as short as possible to obtain high quality specimens. Next, the fixation of fresh tissues requires several hours, depending on the type of fixative, rate of penetration into the tissue, and size of the tissue sample. This time also influences the molecular quality of the tissue (i.e., due to extensive cross-linking and strand scissions of DNA, RNA, and proteins by formaldehyde). Finally, the embedding procedure may have an important effect, i.e., high temperature exposure during submersion in melted paraffin. Snap freezing of unfixed tissue in liquid nitrogen-cooled isopentane circumvents these problems and results in the best available product for further molecular applications. This advantage, however, is offset by a poorer quality in morphology with possible split artifacts and higher costs for tissues storage at −80◦ . Several promising trials have been performed in an effort to find optimal strategies for tissue processing, i.e., fixation of small tissue specimens in 70% ethanol 8
G. D. Cimino, K. Metchette, S. T. Isaacs, and Y. S. Zhu, Nature 345, 773 (1990). S. Kwok and R. Higuchi, Nature 339, 237 (1989). 10 J. Sambrook, E. Fritsch, and T. Maniatis, “Molecular Cloning: A Laboratory Manual,” 2nd Ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, 1989. 9
102
BASIC PRINCIPLES
[9]
followed by embedding in paraffin or low-melting polyester, suggested by the CGAP group at NIH for the treatment of prostate cancer specimens.11 Alternatively, methacarn fixation (60% methanol, 30% chloroform, 10% glacial acetic acid) and paraffin embedding were also shown to yield reasonable quantity and quality of mRNA and protein, significantly superior to that obtained from crosslinking fixatives.12 Low temperature embedding in plastic resin was suggested to obtain a higher degree of integrity of nucleic acids,13 but low-melting polyester also has the disadvantage of drying out and cracking after storage. Finally, data are lacking on the integrity of cellular morphology and nucleic acids after long-term storage of tissues processed in this manner. In another approach, Goldsworthy et al.14 tested the application of several fixatives to frozen tissue sections. They confirmed the preservative effect of alcoholic fixatives (ethanol, acetone) for RNA and found a remarkably lower yield after cross-linking fixation. In consequence, there is no optimal preparation strategy to date. If microdissection is intended, especially in combination with full length cDNA requirement, tissue specimens should be frozen airproof at −80◦ or in liquid nitrogen. If fixation is desired (because of easier storage and handling), a non-cross-linking fixative should be combined with low-temperature embedding. Only when immunolabeling has to be combined with microdissection, can a mild neutral buffered-formalin fixation be considered, accepting a higher degradation rate of nucleic acids. In this case, 4.5% neutral buffered formalin (e.g., Roti-Histofix, Roth, Karlsruhe, Germany) is appropriate. Fixation time of small specimens should not exceed 12 hours to limit the cross-linking and shearing of nucleic acids. Otherwise the necessary proteinase K digestion has to be prolonged to set nucleic acids free. Apart from the new preparation of specimens, archival tissues are, most often, formalin-fixed and paraffin embedded (FFPE). Due to the ease of storage and good long-term preservation of morphology, such tissue specimens have accumulated over recent decades and are often desirable subject tissue. To make these tissues available for molecular approaches, procedures had to be found to handle this type of material. Several reports have adapted existing protocols to the fixation inherent problems dealing with DNA or RNA. Due to the cross-linking of all tissue components, proteinase K digestion was shown to be indispensable to release nucleic acids from the cells. Fragmentation of the nucleic acids can be overcome with the use of primers that amplify very short PCR products (see below). However, the use of RNA derived from routine FFPE tissue remains problematic for hybridization to cDNA arrays, and especially oligonucleotide arrays. Moreover employing this 11
http://cgap-mf.nih.gov/ M. Shibutani, Lab. Invest. 80, 199 (2000). 13 S. D. Finkelstein, R. Dhir, M. Rabinovitz, M. Bischeglia, P. A. Swalsky, P. DeFlavia, J. Woods, A. Bakker, and M. Becich, J. Mol. Diagn. 1, 17 (1999). 14 S. M. Goldsworthy, P. S. Stockton, C. S. Trempus, J. F. Foley, and R. R. Maronpot, Mol. Carcinog. 25, 86 (1999). 12
[9]
INTERNAL STANDARDS FOR LASER MICRODISSECTION
103
RNA for cDNA library construction and RNA preamplification, techniques have to be considered with reservation due to the remarkable rate of degradation and fragmentation. Sectioning, Staining, and Immunolabeling For slide preparation, disposable knives are used (i.e., cryotome CM 3000, Leica, Bensheim, Germany, with Feather microtome blade, via PFM, Cologne, Germany). Contact areas are cleaned with 70% ethanol followed by 0.1 M NaOH. The cryo-sections (5–10 µm) or FFPE sections (3–10 µm) are mounted on glass slides. Sections are made at maximal thickness, but still allow precise microscopic cell recognition. While nonadhesive glass slides are sufficient for short staining protocols, immunolabeling procedures require slides precoated with poly-L-lysine and/or 3-aminopropyltriethoxysilane (APES) to ensure attachment.15,16 In the case of LPC, sections are mounted onto a 1.35–1.5 µm polyethylene membrane treated with 1% poly-L-lysine or 8% APES solution in acetone. Prior to this procedure, the membrane has to be fixed wrinkle-free to the glass slides.17 When using frozen tissue, sectioning should be performed immediately before staining. When applying FFPE-tissue sections, the time span from sectioning to microdissection should not exceed a few days, since a decrease in the yield of nucleic acids could also be observed. Staining procedures for frozen and FFPE sections are presented in Table I. After staining, the sections are stored in 100% ethanol to keep the tissue soft for LMM and micromanipulation. In contrast, using membrane and LPC, the ethanol is allowed to evaporate. For LCM in particular, the sections are finally dehydrated in xylene for 2 min twice before air drying.14,18 To investigate RNA from frozen sections, the time span between staining and microdissection must be kept as short as possible. Due to feasible staining, we have not seen a disadvantageous effect of routine hemalaun staining on RNA recovery, but several reports indicate that hematoxylin may have an inhibitory effect, especially on DNA PCR. Therefore, methyl green, nuclear fast red, and eosin stains are recommended as alternatives.19,20 15
F. d’Amore, J. A. Stribley, T. Ohno, G. Wu, R. S. Wickert, J. Delabie, S. H. Hinrichs, and W. C. Chan, Lab. Invest. 76, 219 (1997). 16 L. Fink, T. Kinfe, W. Seeger, L. Ermert, W. Kummer, and R. M. Bohle, Am. J. Pathol. 157, 1459 (2000). 17 L. M. Gjerdrum, I. Lielpetere, L. M. Rasmussen, K. Bendix, and S. Hamilton-Dutoit, Mol. Diagn. 3, 105 (2001). 18 F. Fend, M. R. Emmert-Buck, R. Chuaqui, K. Cole, J. Lee, L. A. Liotta, and M. Raffeld, Am. J. Pathol. 154, 61 (1999). 19 M. P. Burton, B. G. Schneider, R. Brown, N. Escamilla-Ponce, and M. L. Gulley, Biotechniques 24, 86 (1998). 20 T. Murase, H. Inagaki, and T. Eimoto, Mod. Pathol. 13, 147 (2000).
104
[9]
BASIC PRINCIPLES TABLE I ROUTINE STAINING PROCEDURE FOR FFPE AND FROZEN TISSUE SECTIONS
Dewax with xylene 100% Ethanol 95% Ethanol 70% Ethanol 0.1% Mayer’s hematoxylin, Harris’ hematoxylin, hemalaun Rinse in DEPC water Eosin (optional) Rinse in DEPC water 70% Ethanol 95% Ethanol 100% Ethanol
FFPE tissue
Frozen tissue
2 × 5 min 30 sec 30 sec 30 sec 30–60 sec
— Optional 30 sec — — 30–60 sec
2 × 5 sec 10–30 sec 2 × 5 sec 30 sec 30 sec 30 sec
2 × 5 sec 10–30 sec 2 × 5 sec 15 sec 15 sec 15 sec
Precise characterization of defined cell types within complex tissues often requires immunolabeling. Several articles reported on factors influencing the quality of nucleic acids following this process. While reduction in total staining time seems to be unnecessary for FFPE tissue sections, it is crucial in unfixed and alcoholfixed tissues. Thus, it should be kept as short as possible. Moreover, the number of incubation steps, as well as the type of fixation and staining complex/enzymatic reaction, may influence the results.20,21 In consequence, for immunohistochemical staining the total incubation time should not exceed 30 min in unfixed sections. This may be obtained by reduction of the antibody incubation steps. Therefore, moderate increases in the antibody working concentrations are often necessary. After acetone or 100% ethanol treatment for 1 to 5 min sections are rinsed in Tris-buffered saline (TBS; pH 7.5) briefly. The antibody incubations last about 1 to 5 min, and the enzymatic staining reaction takes about 5 to 15 min.18,21 In particular, for analysis of RNA, the duration of exposure to the aqueous phase is crucial for its recovery.18,21,22 Murakami et al.23 clearly showed for frozen sections that RNA is degraded rapidly so that labeling has to be accomplished within 2 or 3 min. To omit the conversion of the staining complex, which requires time and influences negatively RNA quality, direct or indirect immunofluorescence labeling is advantageous. Applying a mild formalin fixation (short perfusion or ∼2 hr immersion) allowed us to increase the labeling time to 10–15 min.16 Primary and secondary antibody can be incubated for 1 to 5 min followed by a short rinse 21
L. Fink, T. Kinfe, M. M. Stein, L. Ermert, J. Hanze, W. Kummer, W. Seeger, and R. M. Bohle, Lab. Invest. 80, 327 (2000). 22 Y. Kohda, H. Murakami, O. W. Moe, and R. A. Star, Kidney Int. 57, 321 (2000). 23 H. Murakami, L. Liotta, and R. A. Star, Kidney Int. 58, 1346 (2000).
[9]
INTERNAL STANDARDS FOR LASER MICRODISSECTION
105
in TBS. Optionally, RNase inhibitor (RNasin, 400 U/ml, Promega, Mannheim, Germany) can be added to the antibody solutions. Using micromanipulation, the labeled sections have to be kept under ethanol continuously until isolation, as drying leads to marked unspecific background signals. Even a (longer lasting) routine immunohistochemistry staining was combined successfully with DNA analysis for a few or single cells microdissected from archival FFPE tissue.15,24,25 However, the staining complex has to be resistant to alcohol and must not interfere with further reactions. For DNA analysis, DAB and new fuchsin, for example, showed no deleterious effect.20 High temperature antigen-retrieval treatments should be omitted as they can adversely affect recovery of nucleic acids. This includes a microwave oven and pressure cooker to unmask epitopes. Recently, it was shown for DNA that this problem can be overcome using a high pH buffer and low temperature retrieval at 60◦ .17 Microdissection The laser beam is either used to dissect the tissue (i.e., PALM) or to focally melt a thermoplastic membrane to form a composite with the tissue (Arcturus). In the latter case, the membrane is mounted on a small cap. After cooling, the tight adherence of the membrane and tissue allows lifting of the cap with the membrane and results in removal of the cells of interest. Fitting on a reaction tube, the caps with adherent cells are transferred for further procedures. Low beam energy is almost completely absorbed by the membrane, leaving the biomolecules intact (reviewed by Fend and Raffeld26 ). In the case of LMM, a pulsed nitrogen laser with a wavelength of 337 nm is applied. The small beam focus of the UV laser allows very accurate photoablation. Due to its short wavelength, this type of nonthermic ablation avoids energy dispension into adjacent cells.27 These characteristics confer preservation of cytoplasm, including RNA integrity in target cells. Depending on the objective, the focus accuracy can be refined to less than 1 µm, making even the dissection of single chromosomes possible.28 Additionally, the laser beam is advantageous for isolating cells from fiber-rich tissues. This is nearly impossible due to the tight adhesion of the fiber network if pure micromanipulation is used. After microdissection, the tissue islet or cells can be procured traditionally with a sterile needle 24
H. Kanzler, R. K¨uppers, M. L. Hansmann, and K. Rajewsky, J. Exp. Med. 184, 1495 (1996). C. A. Suarez-Quian, S. R. Goldstein, T. Pohida, P. D. Smith, J. I. Peterson, E. Wellner, M. Ghany, and R. F. Bonner, Biotechniques 26, 328 (1999). 26 F. Fend and M. Raffeld, J. Clin. Pathol. 53, 666 (2000). 27 R. Srinivasan, Science 234, 559 (1986). 28 L. Schermelleh, S. Thalhammer, W. Heckl, H. Posl, T. Cremer, K. Schutze, and M. Cremer, Biotechniques 27, 362 (1999). 25
106
BASIC PRINCIPLES
[9]
mounted on a motorized micromanipulator.29 One drop of 100% ethanol onto the uncovered section can improve the morphology remarkably. Moreover, the tissue and cells remain flexible and adhere tightly to the needle, which can be easily lifted after rapid evaporation of the alcohol. Alternatively, the tissue section can be mounted on a polyethylene membrane. Cells or tissue islets can be cut and catapulted by the laser beam pressure into a tube cap positioned above (LPC). Recently, another system also equipped with an ultraviolet laser for microdissection became available. Using an upright microscope, the dissected area falls into a cap that is positioned below (distributed by Leica Microsystems, Wetzlar, Germany).30 Both LCM and LPC are rapid, reliable, and precise techniques. They can readily be learned and applied to various fields of investigation. This is also valid for the recently presented Leica system. LCM may be advantageous to procure islets of hundreds or thousands of cells in a few minutes. Limitations are the isolation of smaller single cells and because of the given spot form (7.5 to 30 µm diameter) the removal of long but narrow structures like endothelial cells. Sometimes, the large contact area of the membrane causes contamination by adhesion of loose tissue fragments. On the other hand, strong tissue adhesion to the glass slide after longer periods of drying can cause problems in removing the cells. LPC and microdissection by the Leica system are real “no-contact” techniques, and, due to the high accuracy of the laser beam, especially suitable for dissecting single cells of all forms and sizes. Problems may arise when liquid remains between membrane and glass slide or electrostatic forces divert the catapulted tissue, thus failing the cap. Use of a micromanipulator after LMM allows a certain flexibility in isolating single cells as well as large tissue islets and is independent of membranes and devices. However, the isolation requires more time and practice than the other techniques. Thus, the preference for employing a special system depends on individual requirements, such as the cell numbers, precision of single isolations, frequency of use, and the number of users. Liberation of Nucleic Acids: Extraction and Digestion The requirements for microdissection may differ markedly. For some approaches, single cells have to be isolated, whereas other studies require a cell islet consisting of a hundred or even thousands of cells. Usually, nucleic acids have to be extracted for further application. However, isolation of a few cells allows proceeding without an extraction step. The limit is 50 to 100 cells.29,31 Up 29
L. Fink, U. Stahl, L. Ermert, W. Kummer, W. Seeger, and R. M. Bohle, Biotechniques 26, 510 (1999). K. Kolble, J. Mol. Med. 78, B24-5 (2000). 31 M. D. To, S. J. Done, M. Redston, and I. L. Andrulis, Am. J. Pathol. 153, 47 (1998). 30
[9]
107
INTERNAL STANDARDS FOR LASER MICRODISSECTION
TABLE II RECOMMENDED TREATMENT FOR THE INVESTIGATION OF DNA AND RNA FROM MICRODISSECTED CELLS HARVESTED FROM FFPE AND FROZEN TISSUE SECTIONS FFPE tissue
Frozen tissue
DNA
RNA
DNA
RNA
+
+
+
+
(a) Few-cell approach, ≤50 cells Resuspension in reaction buffer, 10 µl Proteinase K digestion; 0.5 µg/µl final conc. Denaturation Extraction DNase digestion; 2U Denaturation
Indispensible; 6–10 hr, 58◦ 7 min, 95◦ − − −
− Optional; ≤30 min, 37◦ 7 min, 95◦
Optional; 30 min, 58◦ 7 min, 95◦ − − −
− Optional; ≤30 min, 37◦ 7 min, 95◦
(b) ≥50 cells Resuspension in
Proteinase K digestion; 0.5 µg/µl final conc. Denaturation Extraction DNase digestion; 2–10 U Denaturation
Reaction buffer, 1 M GTC buffer, 10–100 µl or Tris/EDTA (depending on buffer 200 µl cell amount) Indispensable; 12–16 hr, 58◦ ◦ 7 min, 95 − − Phenol/chloroform − Optional; ≤30 min, 37◦ − 7 min, 95◦
Reaction buffer, 10–100 µl (depending on cell amount) Optional; ≤1 hr, 58◦ 7 min, 95◦ − − −
4 M GTC-buffer, 200 µl − − Phenol/chloroform Optional; ≤30 min, 37◦ 7 min, 95◦
to this number, cells can be transferred to a reaction buffer suitable for necessary digestions, reverse transcription, and PCR (52 mM Tris-HCl, pH 8.3, 78 mM KCl, 3.1 mM MgCl2). Alternatively, a PCR buffer can be used. In case of LPC, the buffer can be prefilled into the cap of a reaction tube to collect the catapulted cells. In case of the Leica system, the cells fall into the buffer-filled cap. Thus the buffer sometimes reduces electrostatic diversion. For RNA analysis, RNase inhibitor can be added (4% v/v). To disrupt the cell membrane of intact cells (i.e., from cytospins), addition of a nonionic detergent, e.g., Igepal CA-630 or Tween 20, can be advantageous. Its concentration should not exceed 1% v/v, otherwise further enzymatic reactions can be inhibited. We therefore snap freeze the samples, followed by thawing three times and centrifuge the samples at 10,000g for 1 min to destroy the cell structure. Afterward, the samples are kept in liquid nitrogen. Table II shows the recommended procedures for RNA and DNA, including digestion steps from frozen and FFPE tissues, respectively.
108
BASIC PRINCIPLES
[9]
DNA For DNA analysis, the microdissected cells are resuspended in 10 to 100 µl of the reaction buffer regardless of the type of tissue fixation. Depending on the number of cells the buffer volume is adapted accordingly (i.e., 50 cells in 10 µl, 5000 cells in 100 µl). Alternatively, to obtain very pure DNA, cells can be transferred to a buffer containing 50 mM Tris-HCl (pH 8.1), 1 mM ethylenediaminetetraacetic acid (EDTA), and 0.5% Tween 20. After proteinase K digestion, organic extraction and alcoholic precipitation are performed.32 RNA While less than 50 cells can be processed without extraction, this step is necessary for a larger number of cells. After microdissection of frozen tissue, the cells are resuspended in 200 µl of lysis buffer containing 4 M guanidine thiocyanate (GTC), 25 mM sodium citrate, 0.5% sarcosyl, 0.72% 2-mercaptoethanol, 20 mM Tris-HCl, pH 7.5. After incubation for 10 min at room temperature, 20 µl 2 M sodium acetate, 220 µl phenol (pH 4.3), and 60 µl chloroform/isoamyl alcohol (24 : 1) are added. The samples are vortexed and centrifuged for 15 min at 4◦ . The aqueous layer is collected, 1 µl glycogen (1 mg/ml) added, and then precipitated with 200 µl isopropanol. Samples are frozen for 1 hour at −20◦ and centrifuged for 15 min at 12,000g. The pellets are washed with 75% ethanol, air dried, and finally resuspended in 10 µl H2O. Microdissected cells from FFPE tissue must be digested before extraction. They are resuspended in 200 µl of a lysis buffer containing 10 mM Tris-HCl (pH 8.0), 0.1 mM EDTA (pH 8.0), 2% sodium dodecyl sulfate (pH 7.3) and are then digested with proteinase K (0.5 µg/µl end concentration) at 58◦ for 12–16 hr. Then RNA is extracted by the above-described phenol/chloroform procedure.33 Alternatively, the microdissected cells are suspended in 200 µl 1 M GTC, 0.5% sarcosyl, 0.72% 2-mercaptoethanol, 20 mM Tris-HCl, pH 7.5.34 Adding 0.5 µg/µl proteinase K to the buffer, the samples are digested for 12–16 hr (58◦ ) and phenol/chloroform extraction follows. While the first technique is more sensitive, the latter one results in lower DNA contamination. Use of silica columns for DNA and RNA extraction is an interesting alternative, but, so far, the total elution requires a considerably higher volume (at least 30–40 µl). This again has to be reduced by speed vacuum centrifugation. 32
U. Lehmann, S. Glockner, W. Kleeberger, H. F. von Wasielewski, and H. Kreipe, Am. J. Pathol. 156, 1855 (2000). 33 K. Specht, T. Richter, U. Muller, A. Walch, M. Werner, and H. Hofler, Am. J. Pathol. 158, 419 (2001). 34 G. Stanta and C. Schneider, Biotechniques 11, 304 (1991).
[9]
INTERNAL STANDARDS FOR LASER MICRODISSECTION
109
RNA extraction from FFPE tissue with supermagnetic beads results in a somewhat lower recovery; probably because of RNA fragmentation only 3 ends are caught by this technique. Digestion As already mentioned, proteinase K digestion is indispensable for the release of nucleic acids from FFPE material. Optimal conditions are a 53◦ to 60◦ incubation temperature and a final concentration of 0.1 to 0.5 µg/µl16,33,35 for 12 to 16 hr. If the buffer is not clear to the eye, digestion should be repeated.36 For frozen tissues, the proteinase K digestion is optional and may improve the results when fiber-rich tissues are investigated or proteins are expected to be bound to DNA and RNA. The concentration and temperature of digestion are the same (≤0.5 µg/µl, 53–60◦ ) and time should not exceed 30–60 min. For denaturation of the enzyme, another 7 min at 95◦ are necessary. Employing RNase-free DNase digestion (i.e., Ambion, Austin, TX; 1–10 U, ≤30 min, 37◦ ) depends on subsequent analysis, and should be limited to those cases for which intron-spanning primers are useless, such as presence of pseudogenes37 or investigation of an intron-free gene. The use of RNA for cDNA libraries or array hybridization also requires this kind of digestion. A subsequent precipitation to eliminate DNase is optional and depends on the volume of the sample and the planned analysis. Its presence may impair further enzymatic reactions. cDNA Synthesis Reverse transcription (RT) should be performed shortly after microdissection. Especially for few-cell approaches, microdissection, digestion, and cDNA synthesis should be processed immediately and not stored until RT has been completed. Tubes with microdissected cells diluted in the reaction buffer as well as H2O diluted RNA are heated to 70◦ for 10 min and then cooled on ice for 5 min. With a preceding digestion step, denaturation of the enzyme simultaneously serves to denaturate the RNA, thus the 70◦ step can be omitted. Ingredients are from PE Applied Biosystems (Foster City, CA): cDNA synthesis from 10 µl H2O-diluted RNA is performed with 4 µl MgCl2 (25 mM), 2 µl 10× buffer II (100 mM TrisHCl, pH 8.3, 500 mM KCl), 1 µl dNTP (10 mM each), 1 µl random hexamers (50 µM ), 0.5 µl RNase inhibitor (10 U), and 1 µl MMLV reverse transcriptase in a total volume of 19.5 µl. Due to the presence of MgCl2 in the cell buffer, only 2 µl MgCl2 are added resulting in a volume of 17.5 µl for few-cell approaches 35
I. Becker, K. F. Becker, M. H. Rohrl, and H. Hofler, Histochem. Cell Biol. 108, 447 (1997). Z. P. Ren, J. Sallstrom, C. Sundstrom, M. Nister, and Y. Olsson, Pathobiology 68, 215 (2000). 37 J. Finke, R. Fritzen, P. Ternes, W. Lange, and G. Dolken, Biotechniques 14, 448 (1993). 36
110
BASIC PRINCIPLES
[9]
without RNA extraction. Samples are incubated at 20◦ for 10 min followed by 42◦ for 60 min. The reaction is stopped by heating to 95◦ for 5 min and then cooled to 4◦ . Compositions and ingredients are only recommendations and may vary. Depending on the requirements, RT-PCR can be performed using enzyme mixtures (reverse transcriptase and DNA polymerase). For few-cell analysis, the application of recombinant Thermus thermophilus enzyme is not recommended because both reverse transcriptase and polymerase activities are often lower than those of the respective single enzymes. Random hexamers are preferable, especially when working with FFPE-derived RNA. Due to degradation, all fragments will be reverse-transcribed. In the case of oligo(dt) priming, only the poly A tailcarrying mRNA fragments are transcribed. cDNA from extracted RNA can be applied to several PCRs. In the case of few-cell analyses samples are split into two identical volumes for further PCR reactions (i.e., for target gene and standard gene analysis). Division into three or more PCR samples leads to a decrease in respective recoveries.29 PCR for Qualitative Analysis To detect the presence of specific nucleic acid sequences, qualitative PCR is sufficient. In the case of mRNA, further questions may concern defined splicing variants,38,39 the neoexpression in a pathological state,40 or the interaction of proteins and RNA. To distinguish DNA and RNA, intron-spanning primers should be constructed. Sole detection of mRNA can be performed with optimal efficiency using very short PCR products (≤100 bp). This is particularly important for the investigation of archival FFPE tissue. Due to degradation and fragmentation of nucleic acids, increasing the length of the amplicon results in decreased recovery.14,33,41 Not more than 20% of the PCR volume should consist of the RT product. Because of an inhibitory effect of the RT product on the polymerase activity, up to 10 µl of the cDNA are transferred to the PCR reaction. In a final volume of 50 µl, a standard PCR master mix for one sample consists of 5 µl PCR 10× buffer, 4 µl MgCl2, forward and reverse primer in a final concentration of 300 nM, 1 µl dNTP with 10 mM each, 0.5 µl AmpliTaq Gold (PE Applied Biosystems), and cDNA. As a three-step PCR often yields a somewhat higher recovery, the following PCR conditions are applied: 95◦ for 6 min followed by 45 to 50 cycles with 95◦ for 20 sec, annealing temperature for 30 sec, and 73◦ for 30 sec. 38
W. Kummer, L. Fink, M. Dvorakova, R. Haberberger, and R. M. Bohle, Neuroreport 9, 2209 (1998). K. Pauls, L. Fink, and F. E. Franke, Lab. Invest. 79, 1425 (1999). 40 H. Holschermann, R. M. Bohle, H. Zeller, H. Schmidt, U. Stahl, L. Fink, H. Grimm, H. Tillmanns, and W. Haberbosch, Am. J. Pathol. 154, 211 (1999). 41 M. H. Roehrl, K. F. Becker, I. Becker, and H. Hofler, Diagn. Mol. Pathol. 6, 292 (1997). 39
[9]
INTERNAL STANDARDS FOR LASER MICRODISSECTION
111
Detection of DNA and mRNA from low-copy genes of few-cell or even singlecell approaches may fail if a single PCR is used. In these cases, PCR product can be applied to a second “nested” PCR.6 PCR for Quantitative Analysis Real-time PCR turned out to be a reliable approach for combination with microdissection.42 To investigate genomic amplification or regulation of gene expression, quantitative assays are necessary. Determination of hypermethylation within promoters was also shown to be applicable to microdissected tissue, providing new insights concerning gene inactivation in tumors.43 Moreover, interaction of mRNA and binding proteins can be assessed by this technique.44 The method is based on the 5 nuclease activity of Taq polymerase to hydrolize a duallabeled fluorogenic and sequence-specific hybridization probe. Valid measurement over a large range of initial starting quantities is obtained with known sensitivity of PCR. Absolute quantitation by comparing the target to the constant dilution series requires remarkable effort. In most instances, relative quantitation (CT) is sufficient for measuring gene regulation. Based on Eq. (1), the target sequence is normalized to an internal standard or reference (i.e., nonamplified genomic sequence, almost unregulated “standard” gene mRNA). To = K(1 + E)(CT,R-CT,T) Ro
(1)
where To is initial number of target copies; Ro, initial number of standard copies; E, efficiency of amplification; CT,T, threshold cycle of target gene; CT,R, threshold cycle of standard gene; and K, constant. As amplification efficiency may vary between 0 ( = 0%) and 1 ( = 100%), in pilot experiments it must be shown that the efficiency of the target and the reference primer/probe sets are approximately equal. This has to be determined by calculating the slope of the dilution series. K is assumed to be equal within a definite fluorogenic-labeled primer/probe system and thus does not influence the comparison of the ratios. Calculating the ratio of the threshold cycles for both PCR products allows comparison of the different states of regulation and variation.32,42–46 In the case of 42
L. Fink, W. Seeger, L. Ermert, J. Hanze, U. Stahl, F. Grimminger, W. Kummer, and R. M. Bohle, Nat. Med. 4, 1329 (1998). 43 U. Lehmann, B. Hasemeier, R. Lilischkis, and H. Kreipe, Lab. Invest. 81, 635 (2001). 44 K. Steger, L. Fink, T. Klonisch, R. M. Bohle, and M. Bergmann, Histochem. Cell Biol. 117, 227 (2002). 45 R. M. Bohle, E. Hartmann, T. Kinfe, L. Ermert, W. Seeger, and L. Fink, Pathobiology 68, 191 (2000). 46 Y. Nagasawa, M. Takenaka, Y. Matsuoka, E. Imai, and M. Hori, Kidney Int. 57, 717 (2000).
112
BASIC PRINCIPLES
[9]
relative mRNA quantitation, apart from the aspects mentioned above (short PCR products), special requirements mainly concern the construction of the primer/ probe system. The primers are selected to span a long intron to prevent amplification of DNA by suitable PCR conditions. Additionally, the probe is placed centrally onto the exon–exon transition. This assures complete detachment of the probe from the nucleic acid in case of DNA binding. For exclusive DNA amplification, the probe is positioned in an intron. To investigate few-cell approaches, SYBR-Green is not suitable since primer–dimer aggregations also result in a singal that occurs in a similar threshold cycle as the real PCR product. We therefore recommend SYBRGreen only when more than a hundred cells are investigated or after proving that the primer–dimer signals are detected considerably later than the target signals. Internal Controls Apart from the above-mentioned precautions, several internal controls should accompany the procedure. Many preparation steps may lead to mistakes resulting in negative results, on the one hand; highest sensitivity and many amplification cycles open the risk of contamination, on the other hand. Thus, positive as well as negative controls are routinely introduced into the procedure, and it is strongly recommended that the results in relation to the controls be assessed. Per 10 samples, one or two samples should contain only the reaction buffer. In the case of micromanipulation, a needle is transferred to the buffer without contact to any tissue sample to insure the absence of buffer and needle contamination. To distinguish DNA and RNA when intron-spanning primers cannot be applied, samples have to be split after proteinase digestion/extraction to one with and one without reverse transcription.38,47 The RT-master mix should differ only in the enzyme that is replaced in −RT (mock) samples by H2O; all other ingredients and incubation steps remain identical. Subsequent quantitative PCR allows determination of the portion of RNA in relation to DNA by calculation of the ratio of the threshold cycles. cDNA synthesis can be checked by an additional sample containing RNA in a comparable concentration and already tested to be positive. For PCR analysis, one or two aliquots of the PCR master mix per 10 samples are used for buffer controls. H2O is added to the final volume and these buffer controls are processed in parallel to the others to assess contamination of the PCR ingredients. To check PCR amplification, a positive control should be added. Therefore, an unambiguous positive sample (i.e., PCR product) is applied at a concentration comparable to the unknown sample. 47
S. Busche, S. Gallinat, R. M. Bohle, A. Reinecke, J. Seebeck, F. Franke, L. Fink, M. Zhu, C. Sumners, and T. Unger, Am. J. Pathol. 157, 605 (2000).
[9]
INTERNAL STANDARDS FOR LASER MICRODISSECTION
113
Use of uracil-N-glycosylase digestion is helpful but also occasionally fails. Thus, it can be applied routinely, but cannot replace other controls. Further Applications Several further applications are employed and adapted to microdissection approaches, such as whole genome amplification using degenerate oligonucleotide primer (DOP) PCR or primer extension preamplification (PEP) PCR,48 e.g., for microsatellite and loss of heterozygosity analyses,49 comparative genomic hybridization microarrays,7 or cDNA arrays.50,51 RNA preamplification techniques, such as linear antisense RNA amplification52 as well as PCR-based techniques,53 were shown to augment the introduced original RNA pool and preserve the initial expression profile. Therefore, minimally degraded and fragmented RNA is required. In analogy to cDNA library construction, where full length RNA copies are required, (fresh) frozen tissues are preferred. Future studies are needed to show to what extent routine cross-linking-fixed tissues are reliably applicable. Apart from nucleic acids, microdissected material is increasingly applied to protein investigation. 2D PAGE followed by matrix-assisted laser desorption and ionization time-of-flight mass spectrometry (MALDI-TOF MS) was combined with microdissection of several thousand cells.54,55 Alternatively, proteins can be immobilized to special affinity surfaces of protein chips eliminating the need of preseparation techniques, and TOF-MS can follow directly.56,57 These approaches are promising and standards have to be defined.
48
W. Dietmaier, A. Hartmann, S. Wallinger, E. Heinmoller, T. Kerner, E. Endl, K. W. Jauch, F. Hofstadter, and J. Ruschoff, Am. J. Pathol. 154, 83 (1999). 49 P. Wild, R. Knuechel, W. Dietmaier, F. Hofstaedter, and A. Hartmann, Pathobiology 68, 180 (2000). 50 L. Luo, R. C. Salunga, H. Guo, A. Bittner, K. C. Joy, J. E. Galindo, H. Xiao, K. E. Rogers, J. S. Wan, M. R. Jackson, and M. G. Erlander, Nat. Med. 5, 117 (1999). 51 L. Fink, S. Kohlhoff, M. M. Stein, J. Hanze, N. Weissmann, F. Rose, E. Akkayagil, D. Manz, F. Grimminger, W. Seeger, and R. M. Bohle, Am. J. Pathol. 160, 81 (2002). 52 J. E. Kacharmina, P. B. Crino, and J. Eberwine, Methods Enzymol. 303, 3 (1999). 53 A. Chenchik, Y. Y. Zhu, L. Diachenko, R. Li, J. Hill, and P. D. Siebert, “Gene Cloning and Analysis by RT-PCR,” p. 305. BioTechniques Books, Natick, MA, 1998. 54 D. K. Ornstein, J. W. Gillespie, C. P. Paweletz, P. H. Duray, J. Herring, C. D. Vocke, S. L. Topalian, D. G. Bostwick, W. M. Linehan, E. F. Petricoin III, and M. R. Emmert-Buck, Electrophoresis 21, 2235 (2000). 55 L. C. Lawrie, S. Curran, H. L. McLeod, J. E. Fothergill, and G. I. Murray, Mol. Pathol. 54, 253 (2001). 56 F. von Eggeling, H. Davies, L. Lomas, W. Fiedler, K. Junker, U. Claussen, and G. Ernst, Biotechniques 29, 1066 (2000). 57 K. K. Jain, Pharmacogenomics 1, 385 (2000).
114
[10]
BASIC PRINCIPLES
[10] Methacarn: A Fixation Tool for Multipurpose Genetic Analysis from Paraffin-Embedded Tissues By MAKOTO SHIBUTANI and CHIKAKO UNEYAMA Introduction Recent advances in microfield dissection have facilitated biochemical and molecular biological analysis of specific cells of interest within tissue specimens.1,2 Quantitative measurements of gene expression in pathologically altered cell populations can provide valuable information regarding the mechanism underlying biological phenomena, such as cell growth, differentiation, and apoptotic cell death. In addition, mutation analysis of single cells is now essential for the investigation of molecular events during carcinogenic processes. For histological assessment, tissue fixation and subsequent paraffin embedding are routinely employed because of the ease of handling tissues and subsequent staining, as well as the good preservation of morphology. Until recently, formaldehyde-based fixatives, such as buffered formalin, have been used for this purpose. However, with such cross-linking agents, the efficiency of extraction and quality of extracted RNA, protein, and genomic DNA is limited,3–11 with consequent difficulty in the analysis of microdissected, histologically defined tissue areas. Therefore, unfixed-frozen tissue preparation has now become the gold standard for the analysis of microdissected cells. However, preparation of cryosections from unfixed frozen tissue for the purpose of microdissection may not be applicable for routine samples because of the inconvenience in terms of tissue storage and the skill required for cryosection preparation and subsequent microdissection. As compared to unfixed frozen tissue, paraffin-embedded tissue (PET) permits the
1
M. R. Emmert-Buck, R. F. Bonner, P. D. Smith, R. F. Chuaqui, Z. Zhuang, and S. R. Goldstein, Science 274, 998 (1996). 2 K. Sch¨ utze and G. Lahr, Nat. Biotechnol. 16, 737 (1998). 3 M. Shibutani, C. Uneyama, K. Miyazaki, K. Toyoda, and M. Hirose, Lab. Invest. 80, 199 (2000). 4 D. H. Berger, H. Chang, M. Wood, L. Huang, C. W. Heath, T. Lehman, and B. A. Ruggeri, Cancer 85, 326 (1999). 5 B. Blomeke, W. P. Bennett, C. C. Harris, and P. G. Shields, Carcinogenesis 18, 1271 (1997). 6 N. J. Coombs, A. C. Gough, and J. N. Primrose, Nucleic Acids Res. 27, e12 (1999). 7 S. J. Diaz-Cano and S. P. Brady, Diagn. Mol. Pathol. 6, 342 (1997). 8 J. R. Howe, D. S. Klimstra, and C. Cordon-Cardo, Histol. Histopathol. 12, 595 (1997). 9 S. Merkelbach, J. Gehlen, S. Handt, and L. Fuzesi, Am. J. Pathol. 150, 1537 (1997). 10 M. E. Ortiz-Pallardo, Y. Ko, A. Sachinidis, H. Vetter, H. P. Fischer, and H. Zhou, J. Hepatol. 32, 406 (2000). 11 J. Poncin, J. Mulkens, J. W. Arends, and A. De Goeij, Diagn. Mol. Pathol. 8, 11 (1999).
METHODS IN ENZYMOLOGY, VOL. 356
Copyright 2002, Elsevier Science (USA). All rights reserved. 0076-6879/02 $35.00
[10]
METHACARN FOR MULTIPURPOSE GENETIC ANALYSIS
115
handling of tissue from many samples and accessibility to histologically defined cells such as anatomically defined neuronal cell populations. As the main focus of this chapter, we present tissue fixation procedures, followed by quantitative gene expression analysis and genomic DNA analysis applicable for microdissected small cell areas in PET sections. Principle of Genetic Analysis in Small Tissue Specimens Utilizing Methacarn-Fixed PETs Methacarn is a non-cross-linking protein-precipitating fixative, which has been developed as a methanol-based Carnoy’s fluid to reduce tissue shrinkage by ethanol.12 If an antigen is properly immobilized, protein-precipitating fixatives usually give immunohistochemical results superior to those from aldehyde-based cross-linking fixatives13–15 because antigenicity is usually maintained intact.16 For molecular analysis of microdissected cells, extraction efficiency and quality of molecules are critical. We have found that methacarn meets these critical criteria for analysis of RNAs, proteins, and genomic DNAs in defined areas of PET sections.3,17 By application of a simple extraction method for RNA or protein to methacarn-fixed PET, mRNA-specific PCR amplification and immunoblot detection of proteins appropriate for quantitative analysis can be performed with small tissue specimens. In addition, methacarn fixation extends its availability to genomic DNA analysis in terms of target fragment size and number of microdissected cells required. Methods and Comments Fixation and Paraffin Embedding Like the protocols for other fixatives, the one for methacarn is very simple. Methacarn solution consisting of 60% (v/v) absolute methanol, 30% chloroform, and 10% glacial acetic acid is freshly prepared and stored at 4◦ before fixation. Organs/tissues of interest can be trimmed to 3–4 mm in thickness to facilitate fixation. Exposure of tissues to saline prior to fixation may cause severe shrinkage artifacts of histological components.12 Tissues are fixed with methacarn for 2 hr at 4◦ . The ratio of fixative to tissue should be 20 : 1–30 : 1. For embedding, tissue samples are dehydrated 3 times for 1 hr in fresh 100% ethanol at 4◦ , immersed in 12
H. Puchtler, F. S. Waldrop, S. N. Meloan, M. S. Terry, and H. M. Conner, Histochemie 21, 97 (1970). P. M. Banks, J. Histochem. Cytochem. 27, 1192 (1979). 14 T. O. Rognum, P. Brandtzaeg, H. Orjasaeter, and O. Fausa, Histochemistry 67, 7 (1980). 15 T. B. Orstavik, P. Brandtzaeg, K. Nustad, and J. V. Pierce, J. Histochem. Cytochem. 29, 985 (1981). 16 D. Mitchell, S. Ibrahim, and B. A. Gusterson, J. Histochem. Cytochem. 33, 491 (1985). 17 C. Uneyama, M. Shibutani, K. Nakagawa, N. Masutomi, and M. Hirose, Curr. Topics Biochem. Res. 3, 237 (2000). 13
116
BASIC PRINCIPLES
[10]
xylene once for 1 hr and then 3 times for 30 min at room temperature, and finally immersed in hot paraffin (60◦ ) 3 times for 1 hr, for a total of 3 hr. Comment 1. Glassware for the preparation of methacarn should be autoclaved before use. Do not use disposable plasticware, which can be deteriorated by chloroform, for the preparation and/or storage of methacarn solution. If necessary, tissue processing can be stopped at the ethanol dehydration step, and tissue blocks can be kept at 4◦ overnight after fixation. Microdissection In our laboratory, we use the laser microbeam microdissection (LMM) technique with laser pressure catapulting by PALM Robot-Microbeam equipment (Carl Zeiss Co., Ltd., Tokyo, Japan) because whole targeted cell areas of any size can be recovered with this system. Briefly, thin sections from PET, 4–30 µm in thickness, are prepared and mounted onto 1.35 µm thin polyethylene film, deparaffinized with xylene 3 times for 10 min, and placed in 100% ethanol for washing 3 times for 10 min. If necessary, tissue sections can briefly be stained with hematoxylin or cresyl violet. Then the membrane with the attached specimen is mounted in reverse (membrane side up) onto a new coverslip. The targeted tissue area is then subjected to microbeam microdissection and removed either by a needle tip or laser pressure catapulting, which allows contact-free collection of tissue fragments in the cap of a microcentrifuge tube. Comment 2. There are two major techniques for microdissection utilizing the precision of lasers. One technique is LMM as described above. This system is based on a pulsed UV laser with a small beam focus to cut out areas or cells of interest by photoablation of adjacent tissue. Another technique is laser capture microdissection (LCM), which uses a low energy infrared laser pulse to capture the targeted cells by focal melting of the thermoplastic membrane through laser activation. Advantages and disadvantages of these systems are described elsewhere.18 If expression analysis of RNA is desired, ultrapure water should be used for the preparation of tissue section samples after treatment with diethyl pyrocarbonate (DEPC). To immerse the tissue section in aqueous solution for a long time may increase the risk of degradation of molecules in tissues, such as RNA degradation by contaminating RNases. If nuclear staining is desired, the staining solution should be treated with DEPC or prepared with DEPC-treated water and autoclaved or filtrated if possible. As compared to formalin-fixed tissue sections, methacarnfixed tissue sections can easily be stained with dyes used for histological stainings, and therefore, the tissue-staining procedure can be minimized.
18
F. Fend and M. Raffeld, J. Clin. Pathol. 53, 666 (2000).
[10]
METHACARN FOR MULTIPURPOSE GENETIC ANALYSIS
117
RNA, DNA, and Protein Extraction Microdissected tissue fragments are lysed with RNA STAT-60 (Tel-Test “B,” Inc., Friendswood, TX), and total cytoplasmic RNA is isolated with isopropanol precipitation in the presence of glycogen at 2 µg/ml as a carrier. Isolated RNA is labeled with a RiboGreen RNA Quantitation kit (Molecular Probes, Eugene, OR), and concentrations are estimated with a fluorescence spectrophotometer. For protein extraction, the tissue fragment is solubilized and sonicated for 10 min in 2× sodium dodecyl sulfate (SDS) gel loading buffer. If the mobility shift of the molecule of interest appears in the SDS–polyacrylamide gel electrophoresis (PAGE) by Western blotting, protein precipitation with 10% trichloroacetic acid (TCA) in saline should be performed before solubilization in 2× SDS buffer. Protein concentrations are estimated with a NanoOrange Protein Quantitation kit (Molecular Probes). For DNA extraction, microdissected cells are trapped in PCR tube caps and lysed with 4 µl of DEXPAT (Takara Shuzo Co. Ltd., Kyoto, Japan) at 95◦ for 10 min, and the entire lysate is used as a template for PCR by adding it directly to the master mix for a total of 50 µl. For a large tissue area (0.5–1 mm2), specimens can be lysed with 40 µl of DEXPAT for DNA extraction. If necessary, DNA concentration in the final preparation can be measured using Hoechst 33258 (Molecular Probes). Comment 3. In our laboratory, we found that methacarn-fixed rat liver PET allows the extraction of total RNA in a yield of 52 ± 15 ng/mm2, sufficient for quantitative RT-PCR of many genes, from a deparaffinized 10-µm-thick section by a simple, single-step extraction method with commercially available solution.3 When deparaffinized small tissue blocks were examined for extraction, methacarnfixed rat liver PET showed a 46% RNA yield from unfixed frozen tissue, which was superior to any of the cross-linking fixatives examined (Table I).3 In addition, RiboGreen fluorescent dye is very sensitive for measuring RNA concentrations, with a detection range of 1 ng–1 µg/ml. In terms of protein extraction, tissue sections from methacarn-fixed PET can easily be solubilized in 2× SDS gel loading buffer. We found that the extraction of protein yielded 4.9 ± 2.1 µg/mm2 from a 10-µm-thick rat liver section, allowing quantitative expression analysis of protein by Western blotting. On the other hand, methacarn-fixed rat liver PET showed a 77% protein yield from unfixed frozen tissue (Table II).3 Although the so-called “diffusion artifact” of such an alcohol-based fixative may be responsible for the slight loss, signal intensities of membrane-bound proteins, such as epidermal growth factor receptor (EGFR) and cytochrome P450 (CYP) 2E1, examined by Western blotting did not differ from those of methacarn-fixed PETs and unfixed frozen samples, suggesting the loss in the protein yield could be negligible in analysis by Western blotting (Fig. 1).3 In addition, NanoOrange fluorescent dye is very sensitive for estimating protein concentrations, with a detection range of 10 ng–10 µg/ml.
118
[10]
BASIC PRINCIPLES TABLE I RNA YIELDS AND CONCENTRATIONS OF CONTAMINATING DNA OF THE RNA PREPARATION EXTRACTED FROM RAT LIVER PETSa
Fixative
No. of samples
Unfixed frozen Methacarn Acetone Paraformaldehyde Buffered formalin Ufix Bouin’s solution
6 5 5 5 5 5 5
RNA yield (µg/mg wet tissue) 1.55 0.74 0.82 0.16 0.30 0.05 0.14
± ± ± ± ± ± ±
DNA contamination (ng/mg wet tissue)
0.38b 0.11c 0.13c 0.17b,c 0.21b,c 0.10b,c 0.14b,c
5.85 0.78 5.55 1.12 0.75 0.32 1.88
± ± ± ± ± ± ±
1.31 0.20c 2.29b 0.55c 0.51c 0.45c 0.91c
a
From M. Shibutani, C. Uneyama, K. Miyazaki, K. Toyoda, and M. Hirose, Lab. Invest. 80, 199 (2000), with permission of Lippincott Williams & Wilkins. b Significantly different from methacarn-fixed samples (p < 0.005). c Significantly different from unfixed frozen samples (p < 0.0001).
In terms of genomic DNA analysis, methacarn-fixed PET showed an approximately 8 times higher DNA recovery from a unit area of 10-µm-thick rat liver section when compared to that of buffered formalin-fixed PET.17 Reverse Transcription–Polymerase Chain Reaction (RT-PCR) Single-strand DNA is generated from 15–1000 ng of total RNA with random hexamers and the SUPERSCRIPT Preamplification System (Life Technologies, TABLE II PROTEIN YIELDS FROM RAT LIVER PETSa
Fixative
No. of samples
Unfixed frozen Methacarn Acetone Paraformaldehyde Buffered formalin Ufix Bouin’s solution
4 5 5 5 5 5 5
a
Yield of protein (µg/mg wet tissue) 179 137 175 19 23 21 111
± ± ± ± ± ± ±
21 26 41 4b,c 5b,c 5b,c 75
From M. Shibutani, C. Uneyama, K. Miyazaki, K. Toyoda, and M. Hirose, Lab. Invest. 80, 199 (2000), with permission of Lippincott Williams & Wilkins. b Significantly different from unfixed frozen sample (p < 0.0001). c Significantly different from methacarn-fixed sample (p < 0.0001).
[10]
A
METHACARN FOR MULTIPURPOSE GENETIC ANALYSIS
119
B
FIG. 1. Comparison of protein signals detected in methacarn-fixed PETs and unfixed frozen tissues by Western blotting. (A) EGFR expression in rat liver tissue. Each protein lysate (10 µg) was subjected to 5% SDS–PAGE. Relative band intensity of the protein derived from methacarn-fixed samples to unfixed frozen samples was 101.1 ± 30.6% (n = 4). (B) CYP2E1 expression in rat liver tissue. Ten µg of protein lysate was subjected to 7.5% SDS–PAGE. Protein extracts were prepared either with or without TCA precipitation of deparaffinized tissue section. Relative band intensity of the protein from methacarn-fixed samples to unfixed sample was 101.9 ± 21.9% (n = 5). Purified 2E1: Liver microsomal fraction isolated from the rat treated with acetone. From M. Shibutani, C. Uneyama, K. Miyazaki, K. Toyoda, and M. Hirose, Lab. Invest. 80, 199 (2000), with permission of Lippincott Williams & Wilkins.
Inc., Rockville, MD). Hot start PCR is performed with PLATINUM Taq DNA polymerase (Life Technologies, Inc.) in a total 20 µl volume. The cycle parameters for the PCR of target fragments smaller than 1 kilobase (kb) are typically 94◦ for 1 min, 55◦ for 1 min, and 72◦ for 1 min. If information on exon–intron boundaries is available for the gene of interest, primers for such genes should be designed to overlap two or more exons to distinguish PCR products derived from genomic DNA and mRNA. Under prescribed PCR conditions, only the shorter mRNA-derived cDNA is successfully amplified at the expense of any potential genomic DNA contamination.19 The PCR conditions for each gene are estimated by optimizing the amount of total RNA for the RT, followed by the cycle number for PCR. The amount of total RNA for RT is determined by negativity for PCR amplification of the template derived from contaminating genomic DNA by reverse transcriptase (−) mock RT-PCR at 35 cycles. The optimum cycle number for each gene is determined within the range of the exponential phase of PCR amplification using RT products. Comment 4. For analysis of RNA expression, purity and quality of extracted RNA must be ensured. We have already demonstrated that the integrity of total 19
R. D. Foss, N. Guha-Thakurta, R. M. Conran, and P. Gutman, Diagn. Mol. Pathol. 3, 148 (1994).
120
A
BASIC PRINCIPLES
[10]
B
FIG. 2. RT-PCR results with methacarn-fixed paraffin-embedded rat liver section. (A) RT-PCR results of CYP1A1, CYP2B1, CYP2B2, and CYP4A1, whose steady-state expression levels in the rat liver or hepatocytes are known to be extremely low unless xenobiotically stimulated [T. A. Kocarek, J. M. Kraniak, and A. B. Reddy, Mol. Pharmacol. 54, 474 (1998); C. J. Omiecinski, C. A. Redlich, and P. Costa, Cancer Res. 50, 4315 (1990); R. A. Prough, S. J. Webb, H. Q. Wu, D. P. Lapenson, and D. J. Waxman, Cancer Res. 54, 2878 (1994)]. The RT was performed using 500 ng of total RNA, followed by 35 cycles of PCR with cycle parameters being 94◦ for 1 min, 55◦ for 1 min, and 72◦ for 1 min. A primer set for CYP2B1 also amplifies CYP2B2 RNA-derived cDNA template (arrowhead). PCR of 35 cycles is also performed using reverse transcriptase (−) mock RT product from 500 ng total RNA. Positive control for RT-PCR was performed using in vitro-transcribed mRNA of chloramphenicol acetyltransferase (CAT-RNA RT-PCR). Positive PCR control was performed using primers for glyceraldehyde-3-phosphate dehydrogenase (GAPDH) with the GAPDH cDNA template. (B) RT-PCR results of EGFR RNA fragments of 693 bp and 1.9 kb. PCR of 35 cycles was performed using RT product from 500 ng total RNA with cycle parameters being 94◦ for 1 min, 58◦ for 1 min, and 72◦ for 2 min 30 sec. RT-product was generated by a gene-specific primer (5 -AAGGCCTGGC CCAGCACATC-3 ). Lanes 1 and 2, unfixed frozen liver tissue; lanes 3–6, methacarn-fixed liver section; lane 7, CAT-RNA RT-PCR; lane 8, GAPDH cDNA PCR. Reverse transcriptase (−) mock RT-PCR was performed for each 693 bp (lane 4) and 1.9 kb (lane 6) fragments. From M. Shibutani, C. Uneyama, K. Miyazaki, K. Toyoda, and M. Hirose, Lab. Invest. 80, 199 (2000), with permission from Lippincott Williams & Wilkins.
RNA extracted from methacarn-fixed paraffin-embedded rat PC12 cells was well preserved and similar to that from unfixed frozen cells, judging from the resolution of 18S and 28S ribosomal RNAs.3 Furthermore, the total RNA extracted from methacarn-fixed PET showed low concentration of contaminating genomic DNA as compared to those from acetone-fixed PET or unfixed frozen tissue (Table I). PCR amplification of long mRNA sequences (∼1.9 kb) and any mRNA species, even those expressing low copy numbers,20–22 may prove the quality of extracted RNA from methacarn-fixed PET sections (Fig. 2).3 20
T. A. Kocarek, J. M. Kraniak, and A. B. Reddy, Mol. Pharmacol. 54, 474 (1998). C. J. Omiecinski, C. A. Redlich, and P. Costa, Cancer Res. 50, 4315 (1990). 22 R. A. Prough, S. J. Webb, H. Q. Wu, D. P. Lapenson, and D. J. Waxman, Cancer Res. 54, 2878 (1994). 21
[10]
METHACARN FOR MULTIPURPOSE GENETIC ANALYSIS
121
Where genomic DNA sequence information is not available for the gene of interest, we employ the following validation experiments for each gene. (1) The maximum amount of total RNA for the RT is determined by performing a mock RT followed by PCR with the maximum 35 cycles. (2) The optimum cycle number is determined within the range of exponential amplification of RT products. With this two-step validation, mRNA-specific template amplification can be detected even when genomic sequence information is lacking. Competitive RT-PCR and Plate Hybridization Copy numbers of the target gene transcript in microdissected cells can be estimated by competitive RT-PCR following plate hybridization. Isolated total RNA from microdissected cells is reverse-transcribed to cDNA. PCR is performed with 5 -biotinylated upstream primer and unlabeled downstream primer. The resulting amplicons are detected with streptavidin–horseradish peroxidase (HRP) conjugate and 3,3 ,5,5 -tetramethylbenzidine (TMB) as a substrate. The competitor fragment is designed based on the DNA sequence nonhomologous with any mammalian gene. In our laboratory, we utilized firefly luciferase or lambda phage DNA as a competitor template, which has been amplified with “composite primers” consisting of the target gene sequence followed by nucleotides hybridized to the competitor fragment. Six tubes of a twofold serial dilution of PCR competitor are prepared for each competitive PCR sample. For plate hybridization, capture wells of a 96-well microtiter plate (AquaBind; M&E Corp., Copenhagen, Denmark) are precoated with oligonucleotide (approximately 30 nucleotides) specific to each target and competitor fragments. AquaBind plates are used for covalent linkage of thiol or amino groups. Procedure for oligonucleotide-coating to the capture well is as follows: 1. Prepare 5 -aminated oligonucleotide probe and dilute the probe solution to 10 pmol/0.1 ml with oligonucleotide binding buffer (0.1 M sodium carbonate buffer containing 2.8 M sodium thiosulfate and 0.5% SDS, pH 9.6). 2. Add 0.1 ml diluted probe to each well of the AquaBind plate and incubate at 37◦ for 2 hr. 3. Discard oligonucleotide binding buffer and wash with phosphate-buffered saline (pH 7.2). Tap dry. Following competitive RT-PCR, the PCR products (target and competitor amplicons) are denatured with 0.4 N NaOH at room temperature for 10 min. The sample is then neutralized with the addition of 10–1000 volumes of hybridization buffer (5× SSC, 0.02% SDS, 0.1% sodium N-laurylsarcosine). Then 100 µl of the neutralized sample is hybridized at 50◦ for 2 hr in oligonucleotide capture wells for each amplicon. In order to aquire quantitative results over a broad range, the denatured amplicons can be analyzed by serial dilution in hybridization buffer.
122
BASIC PRINCIPLES
[10]
All dilutions should be promptly done with addition to the capture wells occurring within 30 min of neutralization to prevent inaccuracies due to reannealing of the amplicons. Following hybridization, unbound material in the well is removed with washing buffer (10 mM Tris-buffered saline, pH 7.4, 1 mM EDTA, 0.05% Tween 20). Dilute streptavidin-HRP conjugate (BioSource International, Camarillo, CA) 1 : 100–1 : 1000, add 100 µl to each well, and incubate for 30 min at 37◦ . The wells are again washed to remove unbound conjugate and then 100 µl of substrate (BioSource International) is added. The coloration reaction contains methanol, hydrogen peroxide, and TMB. The reaction is stopped by the addition of 100 µl of 1.8 N H2SO4 and the optical density at 450 nm is measured. The optical density in each well is proportional to the amount of amplicon present, which can be related to the copy number of either the competitor or target gene in the original PCR reaction. The copy number of the target gene in each PCR reaction is calculated from the ratio of the total optical density for the target gene well to the total optical density for the competitor well and the input copy number of the competitor. Comment 5. For competitive RT-PCR, a preliminary validation study should be performed with the practical RNA samples on the concentration of competitor fragments and cycle parameters including cycle numbers. Similarly, the concentration of amplicons should be set in excess of the background binding level due to the cross-hybridization of the target amplicons (or competitor amplicons) to the competitor capture wells (or target wells). With 150–250 ng of total RNA isolated from microdissected cells, competitive RT-PCR can be performed. For example, we investigated the sex difference in the expression of GABA transporter (GAT)-1 RNA in the sexually dimorphic nucleus of the preoptic area (SDN-POA) in neonatal rats (Fig. 3). Immunoblotting Analysis Protein samples are subjected to SDS–PAGE and then transferred to a polyvinylidine difluoride membrane (Millipore, Bedford, MA). After blocking with 0.2% casein (Merck, Whitehouse Station, NJ), blots are incubated with the primary antibody against the molecules of interest. Bound antibodies are detected with the secondary antibodies conjugated with HRP (Amersham Pharmacia Biotech AB, Uppsala, Sweden) and analyzed either with the Supersignal Chemiluminescent Substrate (Pierce Technology Corp., Inc., New York, NY) or the Bio-Rad HRP color reagent (Hercules, CA). Comment 6. For protein analysis, non-cross-linking fixatives such as acetone, ethanol, or modified Carnoy’s solution have proved to give clear protein bands for cytokeratin molecules.23 We have also demonstrated that the polypeptide band of 23
C. J. Conti, F. Larcher, J. Chesner, and C. M. Aldaz, J. Histochem. Cytochem. 36, 547 (1988).
[10]
METHACARN FOR MULTIPURPOSE GENETIC ANALYSIS
123
FIG. 3. Sex difference in the expression of GAT-1 RNA in SDN-POA of neonatal rats. Neonatal rats at postnatal day 9 were sacrificed under deep anesthesia and the removed brains were immediately fixed with methacarn for 2 hr at 4◦ . After dehydration, the coronal brain slice including the hypothalamus, was embedded in paraffin, and serial sections, one 3 µm thick section between every two 21 µm sections, were prepared. Localization of SDN-POA was determined by microscopic observation of the 3 µm sections stained with hematoxylin and eosin, and the SDN-POA area was microdissected from the adjacent unstained 21 µm sections. Following isolation of total RNA, competitive RT-PCR and plate hybridization were performed. For competitive RT-PCR, firefly luciferase sequence from pGL3-basic vector (GenBank/EMBL Data Bank, Accession No. U47295) was utilized as a competitor template. The target fragment (nucleotides 2542-2940) derived from GAT-1 (Accession No. M59742) was amplified in the presence of competitor fragments in a total reaction volume of 20 µl using an upstream primer, 5 -AACAGCAAAC AGCCTATCCA G-3 , and a downstream primer, 5 -ACAGAATCCA AGGCCAGAAG-3 . The sizes of the PCR products of the GAT-1 target and the luciferase competitor were 399 and 519 bp, respectively. The cycle parameters for PCR were 94◦ for 1 min, 57◦ for 1 min, and 72◦ for 1 min, and the cycle numbers for males and females were 32 and 33, respectively. Capture sequences for plate hybridization of GAT-1 (target) and firefly luciferase (competitor) genes were 5 -TCCTCCGTTA GTGGGTGTGT ACATCTGAAA-3 (nucleotides 2815–2844) and 5 -CTTTCGGTAC TTCGTCCACA AACACAACTC-3 (nucleotides 1631–1660).
all examined proteins, in which two glycoproteins (cathepsin D and EGFR) were included, migrated to the positions of correct molecular size (17,000–170,000) in our methacarn-fixed PET. However, it should be noted that some unknown modification appeared on certain populations of polypeptides affecting the mobility in SDS–polyacrylamide gel without TCA precipitation of the tissue samples, such as the case of the CYP2E1 signal (Fig. 1). Therefore, TCA precipitation of microdissected cells before solubilization with 2× SDS buffer is recommended. Genomic DNA Amplification by PCR Hot start PCR of the genomic sequence of the gene of interest is performed with platinum Taq DNA polymerase in a total reaction volume of 50 µl. If nested
124
BASIC PRINCIPLES
[10]
FIG. 4. Single-step PCR-amplification of α2u -globulin genomic sequence (Accession No. M24108 in GenBank/EMBL Data Bank) extracted from a methacarn-fixed rat liver PET section. Genomic DNA was extracted from a deparaffinized, 10 µm thick tissue section using Wizard Genomic DNA Purification Kit (Promega, Madison, WI). Genomic sequence of 1 kb sized fragment was amplified with a primer set of 5 -TAATTAAGTG AGGTGTTTGC-3 (upstream) and 5 -TAAAGGAGGT GTCTACTGC-3 (downstream). Four-kb fragments were amplified with primers of 5 -TAATTAAGTG AGGTGTTTGC-3 (upstream) and 5 -CTGTCAAGGG GTGGTAAATC-3 (downstream) in combination. PCR was performed in a reaction volume of 50 µl with cycle parameters of 95◦ for 5 min, 35 cycles of 95◦ for 1 min, 50◦ for 1 min, 72◦ for 3.5 min. From C. Uneyama, M. Shibutani, K. Nakagawa, N. Masutomi, and M. Hirose, Curr. Topics Biochem. Res. 3, 237 (2000), with permission of Research Trends.
PCR is desired, 1 µl of the first PCR product is used as a template in a total volume of 20 µl. Comment 7. In the preliminary study, we have shown that extensive portions of the genomic DNA sequence, up to 4 kb, could be amplified by single-step PCR with extracted DNA from the rat liver PET section as a template (Fig. 4).17 We have investigated the ability of nested PCR to amplify genomic DNA and found that a DNA fragment of 522 base pairs (bp) from single cell could be amplified in 20% of microdissected Purkinje cells from a 10 µm rat cerebellar PET section. The minimum number of cells required for practical PCR analysis as estimated using rat hippocampal neurons was of the order of 10–20.24 Conclusion We have described protocols of molecular analysis for RNA, protein, and genomic DNA in small tissue samples using methacarn-fixed PET sections in conjunction with a microdissection technique. Using sensitive quantitation methods, such as those utilizing fluorescent dyes specific for RNA or protein, molecules of small quantity can be normalized between samples, and, therefore, quantitative expression analysis for RNA or protein can be applied to microdissected small tissue specimens. 24
U. Uneyama, M. Shibutani, N. Masutomi, H. Takagi, and M. Hirose, J. Histochem. Cytochem. (in press).
[10]
METHACARN FOR MULTIPURPOSE GENETIC ANALYSIS
125
Benefits of paraffin embedding in handling tissues extend the availability of methacarn fixation to genetic analysis in large scale experiments. In addition, considering its advantages for immunohistochemistry,16 tissue embedding after methacarn fixation should be recommended as a valuable approach for routine application, possibly in combination with immuno-LCM,25,26 a recently developed method which allows targeted RNA or DNA analysis of immunophenotypically defined cell populations. The question of how long molecules are retained intact in methacarn-fixed PET should now be addressed. Although we do not have data on archival tissues stored for several years/decades, mRNA, as well as protein and DNA, could be analyzed with methacarn-fixed PET that had been prepared a year ago. Acknowledgments This work was supported by a Grant-in-Aid from the Ministry of Health and Welfare of Japan (Grant H11-Seikatsu-20). We thank Ms. Naoko Abe and Mr. Hitoshi Tainaka from Asahi Techno Glass Corporation (Chiba, Japan) for the support to establish a plate hybridization technique in quantitative RNA expression analysis.
25
F. Fend, M. R. Emmert-Buck, R. Chuaqui, K. Cole, J. Lee, L. A. Liotta, and M. Raffeld, Am. J. Pathol. 154, 61 (1999). 26 F. Fend, M. Kremer, and L. Quintanilla-Martinez, Pathobiology 68, 209 (2000).
This Page Intentionally Left Blank
Section II Specialized Uses
This Page Intentionally Left Blank
[11]
USE OF
LCM FOR CLONAL ANALYSIS
129
[11] Use of Laser Capture Microdissection for Clonal Analysis By VALE´ RIE PARADIS and PIERRE BEDOSSA Clonal Analysis: A Diagnostic Tool in Human Proliferation Clonal study is a useful approach to the analysis of tissue proliferation.1 Indeed, clonality helps define the real nature of lesions, resulting in a better understanding of their behavior and appropriate management. Such an assay allows the differentiation between polyclonal and monoclonal lesions. In fact, monoclonal patterns would support neoplastic nature, whether benign or malignant, whereas polyclonality would be consistent with a regenerative or inflammatory process. Thus, clonal studies have been applied mainly to the study of neoplastic transformation and tumor progression.2–4 The presence of a common and nonrandom genetic alteration in a group of cells confirms their clonal origin only when these particular alterations are acquired early in the neoplastic process and then transmitted to daughter cells. Clonal analysis relies on various tests, including microsatellite analysis (linked or not to the X chromosome) and specific gene rearrangements.5–7 The choice of test is mainly determined by the nature of the pathologic process of interest. Whereas LOH analysis is a useful tool for the detection of inactivation of tumor suppressor genes, specific gene rearrangements are shown to be useful in the diagnosis of various tumors, especially malignant lymphomas.7,8 However, some of these genetic modifications may occur late in the course of a neoplastic process, involving only a few subclones. This is avoided in X-linked clonal analysis since the clonal marker investigated in such an approach is constitutively present in each somatic cell independent of the genetic or epigenetic events occuring during neoplastic transformation or clonal expansion.9 These methods, applied only to informed female patients, based on X chromosome inactivation include the ability to 1
P. J. Fialkow, Biochem. Biophys. Acta 458, 283 (1976). P. C. Nowell, Science 194, 23 (1976). 3 B. Vogelstein, E. R. Fearon, S. E. Kern, A. C. Preisinger, M. Leppert, Y. Nakamura, R. White, A. M. Smits, and J. L. Bos, N. Engl. J. Med. 319, 525 (1988). 4 M. Sternlicht, C. Mirell, S. Safarians, and S. Barsky, Biochem. Biophys. Res. Commun. 199, 511 (1994). 5 T. A. Brentnall, Am. J. Pathol. 147, 561 (1995). 6 J. Koreth, J. J. O’Leary, and J. O. D. McGee, J. Pathol. 178, 239 (1996). 7 K. J. Trainor, M. J. Brisco, J. H. Wan, S. Neoh, S. Grist, and A. A. Morley, Blood 78, 192 (1991). 8 A. G. Knudson, Proc. Natl. Acad. Sci. U.S.A. 90, 10914 (1993). 9 B. Vogelstein, E. R. Fearon, S. R. Hamilton, and A. P. Feinberg, Science 227, 642 (1985). 2
METHODS IN ENZYMOLOGY, VOL. 356
Copyright 2002, Elsevier Science (USA). All rights reserved. 0076-6879/02 $35.00
130
SPECIALIZED USES
[11]
[11]
USE OF
LCM FOR CLONAL ANALYSIS
131
determine the paternally derived X chromosome from the maternally derived one. According to Lyon’s hypothesis, genes on either of the two X chromosomes are randomly inactivated by methylation of cytosine residues within promoter regions during early embryogenesis.10 Once methylated, such CpG islands are functionally and heritably inactive. It is widely believed that this inactivation is stable, even during the neoplastic change. Therefore, in the context of clonal proliferation, any transformed cell will transmit to its progeny its own X chromosome pattern of inactivation. Consequently, reactive polyclonal processes will have an admixture of cells with inactivated maternal and paternal X chromosomes, whereas clonal processes will have a dominant population of cells with only one inactivated allele. Many genes on the X chromosome are polymorphic and allow the distinction between maternally and paternally inherited X chromosomes. This information is related directly to the frequency of their polymorphism in a population, from 29% heterozygosity for hypoxanthine phosphoribosyltransferase to more than 90% for the human androgen receptor gene (HUMARA).11 Therefore, most of the studies investigating clonal status examine the methylation status of the HUMARA gene. This gene contains several restriction sites of the methylation-sensitive HhaI and HpaII endonucleases adjacent to a polymorphic [(CAG)n] repeat region in exon 1. Enzymatic digestion with HhaI or HpaII, followed by PCR targeted to the HUMARA gene region, using primers whose product spans both the HpaII sites and the [(CAG)n] polymorphism, will result in amplification of the methylated undigested alleles but not the unmethylated alleles (Fig. 1). Variations in the lengths of the [(CAG)n] repeats on the paternal and maternal X chromosomes will yield HUMARA alleles of different lengths which can be resolved by gel electrophoresis. LCM and Clonal Analysis Clonal analysis appears to be an accurate tool in the understanding and management of various human disorders. However, reliable interpretation of the results of clonality tests depends mainly on correlation with tissue morphology. In that 10 11
M. F. Lyon, Biol. Rev. 47, 1 (1972). D. G. Gilliland, K. L. Blanchard, J. Levy, S. Perrin, and H. F. Bunn, Proc. Natl. Acad. Sci. U.S.A. 88, 6848 (1991).
FIG. 1. Top: Structure of the androgen receptor gene (exon 1). A highly polymorphic CAG repeat sequence located downstream of a potentially methylated CG island allows a reliable yield of both components in a single chain reaction amplification. Both alleles, easily differentiated by size, can be amplified only from the inactive X chromosome (Xi) in the hpaII-digested sample. Bottom: Allelic patterns of androgen receptor gene in monoclonal and polyclonal proliferations. The presence of both alleles in hpaII-digested samples (D) characterizes polyclonal proliferation (left), whereas the presence of a single allele in hpaII-digested samples (D) [two alleles in undigested samples (U)] defines monoclonal proliferations (right).
132
SPECIALIZED USES
[11]
context, the use of LCM provides major benefits in the clonal evaluation since it allows accurate selection of the cell population of interest. As a matter of fact, both normal and lesional tissues are composed of a mixture of cell types, such as epithelial, mesenchymal, endothelial, and inflammatory cells. This is obvious, especially in neoplasia where the neoplastic cells are intimately associated with the nonneoplastic stromal component. Such cell heterogeneity could lead to false polyclonal results in the analysis of the clonal status of a lesion. It is known that contaminating stromal and inflammatory cells within tumors should not exceed 20% in LOH studies, and that an increase in sensitivity of more than 50% in allelic imbalance analysis is obtained by using microdissected cell populations compared with crushed frozen tumor samples.12 It appears, then, that isolation and enrichment of defined cell populations from biological samples are necessary steps for improving the sensitivity of the clonal analysis. The search for such higher sensitivity prompted the development of new methods of tissue microdissection. The early assisted-microdissection techniques, involving manual or micromanipulator guidance of a needle to scrape off the cells of interest under a microscope, were in fact not accurate enough to isolate cells of interest in a cellular mixture.13,14 They are now replaced by technologies based on laser capture. Two main procedures are currently available: one using a pulsed ultraviolet laser with a small beam focus to cut areas of cells of interest by photoablation of adjacent tissue; the other based on the selective adherence of visually targeted cells to a thermoplastic membrane activated by a low energy infrared laser pulse.15–18 This latter microdissection technique is attractive since it does not destroy adjacent tissues of the area of interest and allows sequential samplings of the same slide. Whichever system is used, collections of pure cell populations in a simple and quick manner are now easily accomplished in a one-step transfer. Such collections are also allowed by the good preservation of morphology of both captured cells and the residual tissue. Applications of LCM and Clonal Analysis in Human Disorders Use of LCM in clonal analysis gave rise to significant advances in the understanding of the pathogenesis of various human diseases, especially those of 12
H. E. Giercksky, L. Thorstensen, H. Qvist, J. M. Nesland, and R. A. Lothe, Diagn. Mol. Pathol. 6, 318 (1997). 13 M. Emmert-Buck, M. J. Roth, Z. Zhuang, E. Campo, J. Rozhin, B. F. Sloane, L. A. Liotta, and W. G. Stetler-Stevenson, Am. J. Pathol. 145, 1285 (1994). 14 S. Noguchi, H. Motomura, H. Inaji, S. Imaoka, and H. Koyama, Cancer Res. 54, 1849 (1994). 15 I. Becker, K. F Becker, M. H. R¨ ohrl, G. Minkus, K. Sch¨utze, and H. H¨ofler, Lab. Invest. 75, 801 (1997). 16 M. B¨ ohm, I. Wieland, and K. Sch¨utze, Am. J. Pathol. 151, 63 (1997). 17 M. R. Emmert-Buck, R. F. Bonner, P. D. Smith, R. F. Chuaqui, Z. Zhuang, S. R. Goldstein, R. A. Weiss, and L. A. Liotta, Science 274, 998 (1996). 18 R. F. Bonner, M. Emmert-Buck, K. Cole, T. Pohida, R. Chuaqui, S. Goldstein, and L. A. Liotta, Science 278, 1481 (1997).
[11]
USE OF
LCM FOR CLONAL ANALYSIS
133
uncertain behavior. It gives further insights into human diseases the morphologic patterns of which do not allow per se the distinction between a reactive and neoplastic lesion. For instance, whereas initial studies demonstrated the neoplastic condition of Langerhans’ cell histiocytosis, recent results showed that pulmonary Langerhans’ cell histiocytosis nodules are polyclonal, suggesting that LCH involving specific sites, such as lungs, might behave rather as regenerative disorders despite some systemic forms of the disease which are clonal.19,20 Such an approach also clarifies the nature of several entities usually considered neoplastic. For instance, the analysis of the epithelial component of Warthin’s tumor, collected by microdissection, showed patterns of polyclonal proliferation in the HUMARA assay, suggestive of a nonneoplastic tumor-like condition.21 In addition, clonal analysis of morphologically distinct components within a tumor, successfully performed with the use of LCM, provided further highlights on the cellular origin in tumors and the relationships between all components. This has been illustrated in cases of pleomorphic adenomas of salivary glands which displayed a monoclonal pattern, showing that the two morphologically different areas in these tumors, the stromal and the epithelial components, arise from the same clone.22 Such an approach has also been very useful in the study of liver cirrhosis which constitutes the main preneoplastic change in the liver. Indeed, the detection of the minute potential premalignant precancerous monoclonal nodules is critically important, helping in the management of patients with chronic liver diseases and cirrhosis. Before the development of LCM, clonal analysis was performed on large nodules, also called macronodules. Such analysis demonstrated that aproximately 50% of them were monoclonal in origin, suggestive of neoplasia.23,24 Interestingly, clonal status did not correlate with the morphological classification of the macronodules, indicating the poor prognostic value of pathological classifications.24 These results prompted us to extend this study to the analysis of smaller cirrhotic micronodules.25 To address this issue, liver micronodules of a mean size of 1 mm were microdissected from one 16-µm serial paraffin section of cirrhotic tissue by LCM with a PixCell instrument (Arcturus Engineering, Mountain View, CA). Parameters of the LCM used included a laser diameter of 30 µm, a laser power of 50 mW, and a pulse length of 10 ms. Each micronodule was microdissected separately after being identified in the adjacent hematoxylin and 19
C. L. Willman, L. Busque, B. B. Griffith, B. E. Favara, K. L. McClain, M. H. Duncan, and D. G. Gilliland, N. Engl. J. Med. 331, 154 (1994). 20 S. A. Yousem, T. V. Colby, Y.-Y. Chen, W.-G. Chen, and L. M. Weiss, Am. J. Surg. Pathol. 25, 630 (2001). 21 K. Honda, K. Kashima, T. Daa, S. Yokoyama, and I. Nakayama, Hum. Pathol. 31, 1377 (2000). 22 P.-S. Lee, M. Sabbath-Solitare, T. C. Redondo, and E. H. Ongcapin, Hum. Pathol. 31, 498 (2000). 23 T. Aihara, S. Noguchi, Y. Sasaki, H. Nakano, and S. Imaoka, Gastroenterology 107, 1805 (1994). 24 V. Paradis, I. Laurendeau, M. Vidaud, and P. Bedossa, Hepatology 28, 953 (1998). 25 V. Paradis, D. Darg` ere, F. Bonvoust, L. Rubbia-Brandt, N. Ba, P. Bioulac-Sage, M. Vidaud, and P. Bedossa, Lab. Invest. 80, 1553 (2000).
[11]
USE OF
LCM FOR CLONAL ANALYSIS
135
in about 7% of melanoma cases.26 Intratumoral molecular heterogeneity was also described in aflatoxin-induced murine lung tumors based on p53 status.27 Some true neoplastic disorders have been found to be oligoclonal. Such findings have been demonstrated by the use of careful microdissection as illustrated by Cabras et al.28 in a study investigating the clonality of morphogically distinct areas of gastric lymphomas. This study clearly showed that some gastric lymphomas exhibited biclonal IgH genes, consistent with the presence of two malignant clones arising from different progenitor cells. At last, the combination of clonal analysis and LCM could provide further evidence for the nature of field cancerization as a discontinous, multifocal, and polyclonal process, as has been proposed in some tissues.29,30 These data confirm the role of LCM in the study of defined cell components within the lesion, especially since studies are carried out on very few cells. In this respect, several papers reported that clonality can be assessed when clonally derived cells comprise 20% or more of the population.19,31,32 Pitfalls and Drawbacks The application of laser-assisted microdissection techniques can be widespread, to DNA, RNA, and protein analysis, since it has been shown that these procedures do not significantly alter additional molecular analysis. However, integrity and quality of the biological material provided should be checked, since these techniques can be applied on both frozen and archival tissue specimens Indeed, the main parameter influencing clonal analysis is specimen fixation, which critically determines the conservation and integrity of the nucleic acids. For instance, fixation leads to some degree of fragmentation of DNA and RNA and induces several chemical modifications, including cross-linking between amino groups on DNA bases, resulting in potentially poor PCR amplification. Appropriate controls, able to check the quality of the material and the efficiency of the molecular procedures, should be simultaneously run. Finally, clonal analysis can be quite easily applied to archival tissue specimens when analysis is performed on samples fixed without 26
T. Nakayama, B. Taback, R. Turner, D. L. Morton, and D. S. B. Hoon, Am. J. Pathol. 158, 1371 (2001). 27 A. S. Tam, J. F. Foley, T. R. Devereux, R. R. Maronpot, and T. E. Massey, Cancer Res. 59, 3634 (1999). 28 A. D. Cabras, S. Candidus, F. Fend, M. Kremer, S. Schulz, C. Bordi, G. Weirich, H. H¨ ofler, and M. Werner, Lab. Invest. 81, 961 (2001). 29 H. K. Yang, R. I. Linnola, N. K. Conrad, M. J. Krasna, S. C. Aisner, B. E. Jonson, and M. J. Kelley, Int. J. Cancer 64, 229 (1995). 30 R. D. Mashal, M. L. S. Fezjo, A. J. Friedman, N. Mitchner, R. A. Nowak, M. S. Rein, C. C. Morton, and J. Sklar, Genes Chromosomes Cancer 11, 11 (1994). 31 T. Enomoto, T. Haba, M. Fujita, T. Hamada, K. Yoshino, R. Nakashima, H. Wada, H. Kurachi, K. Wakasa, M. Sakurai, Y. Murata, and K. R. Shroyer, Int. J. Cancer 73, 339 (1997). 32 K. Krohn, D. Fuhrer, H. P. Holzappel, and R. Paschke, J. Clin. Endocrinol. Metab. 83, 130 (1998).
136
SPECIALIZED USES
[11]
fixatives containing picric acid.20,21,24–26 The possibility of performing such studies on archival tissue has led to further applications in the field of oncology where clonal analysis of microscopic premalignant lesions and multifocal small cancers is inevitably discussed. Numerous situations have been investigated, such as cervical intraepithelial neoplasia31,33 or colonic neoplasia.34 This type of analysis, usually with very tiny samples, introduces the concept of patch size.35 A patch is usually regarded as a group of cells which share a common genotype and have the same X chromosome inactivation pattern. Regarding this concept, the sample size becomes a critical factor which should be taken into account in studies investigating clonality. Therefore, this notion of patch size can impair clonal analysis in tissue samples that are too small. To date, there have been few data concerning patch size in human tissues. However, a large discrepancy in patch size can be observed in tissues from various origins, from a few mm2 in the myometrium or the liver to cm2 in the bladder or arteries.36–39 It is then assumed that any tumor arising within a patch will be, by definition, monoclonal with an X-linked marker. Consequently, X chromosome inactivation cannot be relied on to assess clonality because of the very large patch size of tissue in some cases, such as arteries. An additional potential drawback concerns skewed-X chromosome expression. This phenomenon of extreme lyonization can also mimic clonal derivation of a group of cells. This occurs when there is skewing in the random inactivation of the X chromosome during embryogenesis which may vary from tissue to tissue. Such skewed profiles are transmitted to all somatic cells of normal tissue, resulting in a false monoclonal pattern. To optimally control for lyonization, it is necessary to study control normal tissues, expected polyclonal, that are closely related to the tissue of interest. In conclusion, molecular techniques such as PCR-based methods using X-linked genes as clonal markers have been increasingly used to demonstrate the clonal nature of many tumoral processes. Finally, the development of LCM importantly contributed to a broader application in various human disorders allowing a better knowledge of their biological behavior.
33
H.-S. Park, R. A. Goodlad, and N. A. Wright, Am. J. Pathol. 147, 1416 (1995). M. R. Novelli, J. A. Williamson, I. P. M. Tomlinson, G. Elia, S. V. Hodgson, I. C. Talbot, W. F. Bodmer, and N. A. Wright, Science 272, 1187 (1996). 35 G. H. Schmidt and R. Mead, BioEssays 12, 37 (1990). 36 G. Linder and S. M. Gartler, Science 150, 67 (1965). 37 T. Ochiai, Y. Urata, T. Yamano, H. Yamagishi, and T. Ashihara, Hepatology 31, 615 (2000). 38 Y. C. Tsai, A. R. Simoneau, C. H. Spruck, P. W. Nichols, K. Steven, J. D. Buckley, and P. A. Jones, J. Urol. 153, 1697 (1995). 39 I.-M. Chung, S. M. Schwartz, and C. E. Murry, Am. J. Pathol. 152, 913 (1998). 34
[12]
LASER CAPTURE MICRODISSECTION
137
[12] Laser Capture Microdissection in Carcinoma Analysis By YEN-LI LO and CHEN-YANG SHEN Development of gene sequencing and amplification techniques has now allowed cancer investigators to pinpoint molecular aberrations based on DNA and RNA extracted from cancer tissue biopsies or cytological smears. Successful performance of these sophisticated genetic testing methods, however, depends on the purity and precision of the tumor cell population being analyzed. The conventional method of homogenizing the biopsy sample will always result in an impure combination of cancerous tissue and surrounding normal tissue. To overcome this critical limitation, laser capture microdissection (LCM) has been invented,1 which aims at procuring pure cell populations for subsequent molecular analysis. LCM integrates a standard laboratory microscope, a low-energy laser, and a transfer film in a convenient, one-step, aim-and-shoot method. Based on standard protocols mentioned below, LCM is able to transfer a single cell or large groups of cells from paraffin-embedded and frozen solid-tissue samples, which are usually on stained and immunolabeled slides or, alternatively, on cytological smears. More importantly, LCM preserves the exact morphologies of both the captured cells and the surrounding tissues, very helpful in evaluating capture efficiency. LCM has been widely applied to cancer research of various forms of tumors, including brain, breast, colorectal, gastric, liver, lung, prostate, and uterus. The captured premalignant lessions and primary and metastatic tumors have been used for nucleic acid analysis, such as loss of heterozygosity (LOH) and microsatellite instability. Furthermore, combined with other techniques, such as DOP-PCR, archival formalin-fixed paraffin-embedded tumor samples can be analyzed by comparative genome hybridization (CGH). LCM has been combined with RNA amplification and cDNA microarrays to study differential gene expression profiles. All relevant publications are now available on web sites.2 The basic procedures for LCM will be described as follows. 1. Histological Preparation for Laser Capture Microdissection Histological slides are ideal substrates for LCM. They allow microscopic visualization and two-dimensional manipulation of a single cell layer obtained from a tissue. Microdissections can be carried out on frozen sections as well as 1
M. R. Emmert-Buck, R. F. Bonner, P. D. Smith, R. F. Chuaqui, Z. Zhuang, S. R. Goldstein, R. A. Weiss, and L. A. Liotta, Science 274, 998 (1996). 2 http://www.arctur.com/
METHODS IN ENZYMOLOGY, VOL. 356
Copyright 2002, Elsevier Science (USA). All rights reserved. 0076-6879/02 $35.00
138
SPECIALIZED USES
[12]
paraffin-embedded sections (surgical and autopsy). Each method has its advantages and drawbacks. Paraffin-embedded tissues preserve morphology best and are consistent in quality and reproducibility, but they have comparatively poor molecular preservation and recovery. In contrast, frozen sections have poorer histology and are inconsistent in quality, but they are more easily processed and demonstrate excellent recovery of DNA and RNA for analysis. If the isolation of DNA for PCR is the end use, sample preparation requirements are usually less stringent than if the samples are to be used for isolating RNA. In preparing histology samples for RNA isolation, considerable effort may be required to minimize exposure to aqueous environments in which RNase will be active and to keep the slides desiccated prior to microdissection. Tissue thickness during microtomy is another issue. Good histology can be at 5 µm, but more DNA and RNA are present in a thicker section. Ten µm is presumably better for DNA or RNA isolation than 5 µm, but morphology will be less than optimal. 2. Preparation of Paraffin-Embedded Tissue Sections for LCM3,4 Paraffin embedding is a process in which the tissue specimen is fixed to preserve its cellular structure, and then blocked out and embedded in paraffin to stabilize it for long-term storage and easy sectioning. RNA is a more labile species, and the paraffin-embedding process has been shown to greatly harm it. For each sample, a total of 7 LCM slides are prepared. The first 3 slides are cut at 5 µm and for harvesting the sample. They will be stained, but not coverslipped. Slide 4 will be cut at 5 µm, stained with H&E, and coverslipped. This is the morphology reference slide. Slides 5 through 7 will be cut at 5 µm and left as unstained paraffin sections. These will serve as duplicate samples should there be insufficient material from the first 3 slides to allow collection of duplicate samples. 2.1. Fixation of Tissue for LCM Prior to paraffin embedding, the tissue must be “fixed” to preserve cellular morphology and prevent autolysis. For any fixative used, a fixing period of 16– 24 hr is recommended. 1. 10% Neutral buffered formalin fixative is appropriate for isolation of tissues for DNA extraction. The fixation of tumor tissues in formalin results in extensive cross-link of nuclear proteins and formation of tight complexes between proteins 3 4
http://dir.nichd.nih.gov/lcm/lcm.htm http://dir.niehs.nih.gov/dirlep/lcm.html
[12]
139
LASER CAPTURE MICRODISSECTION
and DNA, thus making the molecules rigid and susceptible to mechanical shearing and fragmentation. It is therefore not surprising that it is difficult to extract DNA from these fixed nuclei with compact chromatin. The poor yield and degradation of DNA are clearly evident with increasing formalin fixation time. When using formalin-fixed tissue, 2.5–4 hr for fixation is recommended. 2. 70% Ethanol is currently being recommended as the best fixative for the maximum isolation of RNA and improved integrity and recovery of nucleic acids. 3. Acidic fixative and methanol should be avoided. 2.2. Processing, Embedding, Sectioning, and Mounting After fixation, paraffin processing can be performed via a routine overnight or accelerated cycle in an automated tissue-processing machine (see Tables I and II). After processing, the specimen is embedded in paraffin and blocked. Sectioning and Mounting 1. Block the wax and specimen out in a mold. 2. Cut sections on a microtome with a clean blade (5–10 µm thick). 3. Float paraffin ribbons on 43–44◦ deionized water (no adhesives). 4. Mount the tissue on the center of plain glass slides (i.e., not charged or coated). 5. Dry slides in a 37◦ oven overnight, or at 60◦ for 10–15 min. TABLE I ROUTINE OVERNIGHT PROCESSING Step
Solution
Concentration (%)
Time (min)
Temp.
1
NEF (skip steps if formalin not used as the fixative) NBF Ethanol Ethanol Ethanol Ethanol Ethanol Ethanol Xylene Xylene Paraffi Paraffi Paraffi Paraffi
10
120 : 00
40◦
10 70 80 95 95 100 100 100 100
120 : 00 0 : 30 0 : 30 0 : 45 0 : 45 0 : 45 0 : 45 0 : 45 0 : 45
40◦ 40◦ 40◦ 40◦ 40◦ 40◦ 40◦ 40◦ 40◦
2 3 4 5 6 7 8 9 10 11 12 13 14
◦ ◦ ◦ ◦
140
[12]
SPECIALIZED USES TABLE II ACCELERATED PROCESSING Step
Solution
Concentration (%)
Time (min)
Temp.
1 2 3 4 5 6 7 8 11 12 13 14
Ethanol Ethanol Ethanol Ethanol Ethanol Xylene Xylene Xylene Paraffin Paraffin Paraffin Paraffin
70 80 95 100 100 100 100 100
0 : 10 0 : 10 0 : 15 0 : 20 0 : 30 0 : 30 0 : 30 0 : 30 0 : 20 0 : 20 0 : 30 0 : 20
40◦ 40◦ 40◦ 40◦ 40◦ 40◦ 40◦ 40◦ ◦ ◦ ◦ ◦
2.3. Staining: Hematoxylin and Eosin 1. Xylene (prewarm at 40◦ ) to deparaffinize the slides (5 min 2×). 2. 100% Ethanol (xylene wash) 30 sec. 3. 100% Ethanol (rehydration steps 3–6) 30 sec. 4. 95% Ethanol wash for 30 sec. 5. 70% Ethanol wash for 30 sec. 6. Purified water wash for 30 sec. 7. Mayer’s hematoxylin (store in the dark) for 30 sec. (The length of the hematoxylin and eosin-Y staining depends on the degree of contrast in the histology desired and is tissue-dependent.) 8. Purified water rinse. 9. Bluing reagent, 30 sec or until differentiation. 10. 70% Ethanol wash (60 sec). 11. 95% Ethanol wash (60 sec). 12. Eosin Y (30 sec). (Hematoxylin may bind to nucleic acids and adversely affect downstream research processes. Two possible binding sites for the hematoxylin–aluminum complex: the phosphate groups of DNA and the histones that represent protein bound by the phosphate groups. Tissues stained with hematoxylin seem resistant to complete digestion with proteinase K solution, which may simply make the DNA less available for enzymatic replication and might contribute to the poor outcome following PCR.) 13. 95% Ethanol wash (60 sec 2×). 14. 100% Ethanol wash (dehydration) 60 sec. 15. Xylene wash (ensures dehydration) 10–30 min.
[12]
LASER CAPTURE MICRODISSECTION
141
16. Air dry (allow the xylene to evaporate completely). 17. The slides should not be coverslipped. 3. LCM Transfer Protocol For identifying areas to microdissect (1) double-scope the potential sample on a morphology reference slide (a pathologist should identify the areas for microdissection). (2) An enlarged photocopy or scanned image should be made with the areas of interest marked for reference at the time of laser capture microdissection. Detailed instructions on the microscope’s assembly and use can be found in the LCM microscope instruction manual.2 For the optimal transfer of frozen tissue sections, it is best to keep sections <10 µm thick. Thicker sections are more difficult to visualize. If there are folds in the tissue it is best not to place the cap over that area. The cap may not make direct contact with the entire surface at that area. Therefore, it is advisable to inspect the tissue before placing the cap down. The tissue section must be dry and not coverslipped for effective LCM transfer. The staining appears darker and more granular because of light scattered from the irregular air–tissue interface. The tissue where the polymer melts and bonds after laser activation appears lighter and resembles a coverslipped slide because of the replacement of the air in the tissue with the polymer. This phenomenon is called index matching or polymer wetting. Poor transfers may result if the slide is not fully dehydrated (i.e., the 100% ethanol becomes hydrated after repeated use). The final xylene rinse facilitates the efficiency of transfer with LCM. If a tissue section does not transfer well, a longer xylene rinse may help. Although other staining protocols can be used, the slides should be dehydrated in a final xylene step. Any water in the ethanol will prevent complete dehydration and result in poor transfer. One way to check water presence in the ethanol is to put a small amount into the xylene. The xylene will cloud up if water is present in the ethanol. 4. DNA, RNA Recovery and Analysis3,4 Most DNA and RNA extraction protocols are easily adapted to LCM. 4.1. Proteinase K Protocol for DNA Extraction with LCM 1. After microdissection, the cap is inserted into a 0.5-ml Eppendorf tube containing digestion buffer [50–100 µl: 10 mM Tris (pH 8.0), 1 mM EDTA, 1% Tween 20, and 0.04% proteinase K; final pH of buffer is 8.0]. There should be about a 1 mm space between the cap top and tube edge.
142
SPECIALIZED USES
[12]
2. Shake Eppendorf tube once inverted to make sure digestion buffer covers all of the caps. 3. The tube is placed upside-down and incubated overnight in a 37–42◦ oven. 4. Centrifuge the tube at 12,000 rpm for 5 min to pull the buffer down, and remove the cap. 5. Heat the buffer at 95◦ for 10 min to inactivate the proteinase K. It can then be used directly as template for PCR. 4.2. RNA Isolation (Modified Stratagene RNA Micro-isolation Kit) RNA Recovery. RNA recovery is best performed on fresh-frozen (or frozen), ethanol-fixed tissue. All solutions must be RNase free, and the tissue must be handled quickly to prevent any endogenous nucleases in the tissue from degrading the RNA. 1. Place the cap in an Eppendorf 0.5-ml tube containing 200 µl RNA denaturing buffer (guanidinium isothiocyanate, GITC) and 1.6 µl 2-mercaptoethanol. 2. Invert several times over the course of 2 min to digest the tissue off the cap. 3. Spin the tube briefly to bring down any buffer clinging to the film and remove the cap. Multiple caps can be processed this way in the same buffer to increase the starting material used. 4. Add 20 µl (0.1× volume) 2 M sodium acetate (pH 4.0). 5. Add 220 µl (1× volume) water-saturated phenol (bottom layer). 6. Add 60 µl (0.3× volume) chloroform–isoamyl alcohol. 7. Vortex vigorously and put on wet ice for 15 min. 8. Centrifuge at 12,000 rpm for 30 min (at 4◦ ) to separate the aqueous and organic phases. 9. Transfer the upper aqueous layer to a new tube. 10. Add 2 µl glycogen (10 µg/µl). 11. Add 220 µl cold isopropanol. 12. Precipitate at −80◦ freezer for at least 30 min. 13. Centrifuge at 12,000 rpm for 30 min at 4◦ . 14. Remove the supernatant. 15. Wash the glycogen pellet with 400 µl 70% ethanol. 16. Vacuum or air-dry the pellet to remove any residual ethanol. 17. Store the pellet at −80◦ until use. DNase Treatment 1. Add 15 µl DEPC water to the RNA pellet. 2. Add 1 µl 20 U/µl RNase inhibitor, 2 µl 10× DNase buffer.
[12] 3. 4. 5. 6.
LASER CAPTURE MICRODISSECTION
143
Heat at 50◦ for 3 min. Add 2 µl 10 U/µl DNase I (20 units total). Incubate at 37◦ for 2 hr. Then boil at 95◦ for 5 min to deactivate the DNase.
Reextraction of RNA 1. 2. 3. 4. 5. 6. 7.
Add 2 µl sodium acetate. Add 22 µl phenol. Add 6 µl chloroform–isoamyl alcohol. Place on ice for 15 min. Centrifuge for 10 min at 4◦ . Transfer the upper layer to a new tube. Repeat RNA extraction steps 10–17.
Reverse Transcription (Final Volume 20 µl) 1. Add 24 µl water and 1 µl 20 U/µl RNase inhibitor to RNA pellet. 2. Aliquot 12 µl into 2 tubes for the (+) and (−) RT. 3. Add 4 µl 5× RT buffer. 4. Add 2 µl dNTP (250 µM ). 5. Add 1 µl 10 µM random hexamer primers or 2 µl 10 µM oligo(dT) primers. 6. Incubate at 65◦ for 5 min. 7. Incubate at 37◦ for 10 min for oligo(dT) (10 min at 25◦ if using hexamer primers). 8. Add 1 µl 100 units/µl MMLV reverse transcriptase to the (+) RT tube only, add 1 µl water to the (−) RT tubes (if random priming, incubate another 10 min at 25◦ ). 9. Heat 40 min at 37◦ (40 min at 37◦ if random priming). 10. Boil at 95◦ for 5 min. 11. The cDNA may be stored at −20◦ until use.
5. Preparation and RNA/DNA Extraction of Frozen Tissue Sections3 Frozen embedding is another way to preserve specimens and stabilize them for long-term storage. This method is recommended for recovery of RNA and has faster processing and excellent molecular recovery. Up to 800 base pairs for both RNA and DNA have been recovered from OCT-embedded tissue, and up to 2 kilobases in cDNA library smears. However, histology is poor compared to paraffin-embedded tissue and sometimes inconsistent.
144
SPECIALIZED USES
[12]
5.1. Embedding 1. Place an empty, labeled cryomold on dry ice for 1 min. 2. Cover the bottom of the cryomold with embedding medium (i.e., OCT). 3. Place the frozen tissue against the bottom. 4. Fill the cryomold with the embedding medium, cover the dry ice container, and allow the OCT to harden (it will turn white when frozen). 5. Wrap the block in foil and store at −80◦ until cutting. 5.2. Cutting 1. Remove the block from the cryomold and attach it to the chuck in the cryostat with OCT. (The cutting surface should be as parallel as possible.) 2. Allow the block to equilibrate to the cryostat temperature (−20◦ ) for about 15 min. If the block is too cold during cutting this time may need to be extended. 3. Cut 10 µm (or thinner) sections onto plain uncoated glass slides. 4. Slides should be kept in the cryostat or on dry ice if LCM is to be performed that day. Alternatively, they may be stored in paper slide boxes at −80◦ until needed. 5.3. Staining 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12.
70% Ethanol wash (30 sec). Purified water wash (30 sec). Mayer’s hematoxylin (store in the dark) for 30 sec. Purified water rinse. Bluing reagent, 30 sec or until differentiation. 70% Ethanol wash (60 sec). 95% Ethanol wash (60 sec). Eosin Y (30 sec). 95% Ethanol wash (60 sec 2×). 100% Ethanol wash (dehydration) 60 sec. Xylene wash (ensures dehydration) 10 min. Air dry (allow the xylene to evaporate completely).
For detailed protocols of DNA and RNA preparations from microdissected samples, see Refs. 2, 3, and 4.
[13]
LCM TO ASSESS DEVELOPMENT
145
[13] Laser Capture Microdissection to Assess Development ´ By CARLOS A. SUAREZ -QUIAN, OSCAR M. TIRADO, FRANCINA MUNELL, ´ and JAUME REVENTOS Introduction Complex processes in developing tissues are particularly difficult to understand in terms of biochemical phenomena. For the most part, not only are the tissues in small supply from which to harvest material, but, as well, the events to study are transient, often occurring for no more than a few hours at a time. Thus, even when a discrete biological phenomenon is observed, it is often nearly impossible to isolate cells or groups of cells to homogeneity to then define the biological process in terms of its molecular fingerprint. Although modern genomic and proteomic approaches promise to decipher developmental processes in molecular terms, not until the limitations of tissue and cellular harvest from the developing embryos are overcome will these powerful new tools yield a fruitful outcome. Laser capture microdissection (LCM) represents that bridge to genomics and proteomics. LCM allows for the efficient and precise capture of cells or groups of cells from developing tissues in sufficient quantities and within the context of time and space to permit the subsequent molecular fingerprinting and characterization of the targeted tissue. A detailed description of how LCM works and how cells and/or groups of cells are harvested from tissue sections will not be presented here, since LCM has been the subject of reviews and the harvesting of cells is now a routine process.1,2 Instead, discussion will focus on the features of LCM that are of particular interest to cell and developmental biologists. Below, we provide successful applications of LCM to tissues of the male reproductive tract, a standard model system to study differentiation of a complex epithelium that occurs in both time and space. LCM in Developmental Studies: Specific Considerations Disadvantages The steps involved in acquiring cells or groups of cells from tissues by LCM are as follows: tissue harvest from animal, tissue fixation, sectioning, and finally targeting and capture. For purposes of definition, capture is defined as initially
1
R. F. Bonner, M. Emmert-Buck, K. Cole, T. Pohida, R. Chuaqui, S. Goldstein, and L. A. Liotta, Science 278, 1481 (1997). 2 C. A. Suarez-Quian, S. R. Goldstein, and R. F. Bonner, J. Androl. 21, 601 (2000).
METHODS IN ENZYMOLOGY, VOL. 356
Copyright 2002, Elsevier Science (USA). All rights reserved. 0076-6879/02 $35.00
146
SPECIALIZED USES
[13]
described using the Arcturus PixCell instrument. That is, a low energy laser transiently melts a thermoplastic film bonded onto a flat support and the film then infiltrates the interstices of the tissue section, not unlike paraffin entering tissue during routine embedding of histological blocks. The distance away from an epicenter that the film is capable of infiltrating is a function of the duration and power of the laser and is regulated by the investigator. Inactivation of the laser leads to the almost instantaneous hardening of the thermoplastic film and capturing the area of the targeted tissue to which the film infiltrated. It is important to note that within the area of capture the forces between the targeted tissue and the thermoplastic film are greater than between the targeted tissue and the rest of the tissue section, or between the targeted tissue and the glass slide. On removal of the flat support, the targeted tissue area is now attached to the support (captured). Thus, true capture entails a firm association of targeted tissue to the thermoplastic film and will require the extraction of the captured tissue from the thermoplastic film prior to molecular analysis; mere nonspecific bonding of nontargeted tissue to the flat support does not constitute capture. This latter nonspecific tissue bonding is easily removed with the aid of a sticky tape now commercially available from Arcturus, Inc. The mechanism of how cellular capture is accomplished with LCM and described above is now routine in many laboratories, in particular when capture is performed from homogeneous tissues or cellular aggregates larger than a few millimeters, e.g., tumors within normal tissues. In the specific application for developmental studies, however, LCM remains an investigative tool whose full potential has yet to be realized. Two factors inherent in the design of LCM are responsible for these limitations. First, capture of individual cells from tissue sections requires that they not be mounted with a coverslip. Thus, imaging of discrete cellular outlines within a complex epithelium is nearly impossible, especially if the targeted cells are highly irregular in shape, or fusiform in design. The Sertoli cell of the seminiferous epithelium is such an example, and within a 5- to 6-µm thick section it is impossible to image the entirety of its arboreal cytoplasm interspersing within the developing germ cells using the current PixCell II LCM apparatus. Neurons also present a problem of this kind, especially the targeting of the dendrites and axons surrounding the cell bodies. The difficulty in imaging the desired cells for capture is due to the increased diffraction of light as it passes trough the tissue section and the absence of an embedding medium filling in the interstices of the section. Under these conditions tissue histology appears more akin to an etching than to images that are routinely observed by a conventional microscopist, and cell boundaries are impossible to distinguish with certainty. While a trained histologist or an investigator extremely familiar with a particular tissue may be able to discern in part some cell boundaries and target distinct cell types from a developing tissue, the capture of irregularly shaped cells from
[13]
LCM TO ASSESS DEVELOPMENT
147
a complex epithelium still remains an unattainable goal given current technology. Second, the diameter of capture can only go as small as 7–8 µm. Thus, cells smaller than this diameter will not be possible to capture without also capturing nondesired, adjacent cells. However (see below), if desired cells are lining a tubule lumen, then it may be possible to capture cellular areas smaller than 7–8 µm in diameter. Advantages In spite of these two limitations, LCM remains the gold standard by which investigators can best target, isolate, and harvest discrete cells from heterogeneous tissues and still be able to perform a molecular analysis of their in situ conditions. Several recent advances in technology make this possible. First, while the appearance of tissues lacking a coverslip will never be identical to conventional tissues observed with routine light microscopy, nevertheless currently enhanced hematoxylin and eosin staining of tissues provides a very good approximation of usual tissue histology. Arcturus, Inc., now sells a hematoxylin and eosin staining kit that, when instructions are followed precisely, including duration of the stains and rinses, provides tissue histology that closely approximates conventional light microscopy images. For those individuals who do not routinely perform histology, or do not have ready access to a trained histologist well versed in paraffin and cryosectioning and staining, the convenience of initiating a research project with the use of kits is well worth the cost. Further, given that the ultimate goal of cellular capture is to perform molecular analysis of protein, RNA, or DNA, quality control of the commercial reagents may provide an advantage to ensuring the best possible macromolecular extraction from the captured cells. Once stained, it is now customary to prepare a “road map” of the tissue to be targeted before the LCM process. This road map is prepared by briefly applying a drop of xylene to the tissue section and capturing an image of the area to be dissected. The xylene works similar to an embedding medium and coverslip combined, yet is volatile and will evaporate, leaving interstices once again within the tissue section that can be infiltrated by the thermoplastic film during the capture event. (Indeed, this one LCM “trick” is learned if a novice investigator visits a functioning LCM core facility.) The road map image can be continuously examined before making targeting decisions. Thus, even though tissue histology is not ideal in real time, a highly accurate map of the section is. The precision of the instrument is such that the laser pulse, and hence the diffusion of the thermoplastic film, precisely corresponds to the targeted area designated on the road map.2 Second, if antibody probes are available that specifically immunostain the desired cell or groups of cells to be acquired by LCM, then prior immunostaining will significantly enhance visualization, targeting, and capture of discrete cell types. Initial immunostaining protocols employed DAB as
148
SPECIALIZED USES
[13]
the reporter molecule and RNA and DNA could be specifically isolated and assayed by RT-PCR.3,4 The current PixCell II apparatus from Arcturus, Inc., is also designed to image sections immunostained using fluorescently tagged antibodies. Thus, if an antibody is available in sufficient quantities that it can be fluorescently labeled and will specifically immunostain the desired cell type to be captured, direct immunofluorescence can be performed, a much quicker procedure than the biotin–immunoperoxidase method initially used. The fluorescent attachment of the PixCell II can also be exploited in imaging tissues before capture if these have been stained with hematoxylin and eosin. Eosin is weakly fluorescent and the sections can be visualized with fluorescence microscopy. Importantly, cellular boundaries in some tissues can be more readily distinguished under these conditions than with the use of conventional light microscopy. Assaying Cells Acquired by LCM: Western Analysis and RNA Extraction Embryonic cells, as discussed above, are often in limited supply for study and may at times be impossible to isolate to homogeneity by conventional means. In principle, it is possible to continue harvesting specific cell types by LCM until sufficient material is acquired from which it will be possible to perform Western analysis and extract enough RNA for DNA chip analysis. Acquisition of thousands of embryonic cells can be a laborious process, however, since it may include the harvesting of embryos, sectioning, and then capturing by LCM. Further, even if an investigator is willing to invest the required amount of time, the cost of generating sufficient transgenic mice with a desirable gene insertion or deletion from which to harvest discrete cell types may be prohibitively expensive, especially if the transgenics are embryologically lethal. For Western analysis, advances in chemiluminescent detection systems have significantly enhanced detection of minimal amounts of protein. The Pierce Supersignal West Pico Chemiluminescent Substrate permits detection of as low as 0.003 ng of protein in a single band by Western analysis. Although an initial report demonstrating the feasibility of Western analysis of LCM cells detected a signal in approximately 50,000 cells,5 it is likely that an abundant protein could be detected in as few as 500 cells using the Supersignal substrate (personal observation). Initial quantities of cellular captures to assay by Western analysis may be based on preliminary immunohistochemistry results in combination with available Western analysis data. In this regard, it is 3
F. Fend, M. R. Emmert-Buck, R. Chuaqui, K. Cole, J. Lee, L. A. Liotta, and M. Raffeld, Am. J. Pathol. 154, 61 (1999). 4 C. A. Suarez-Quian, S. R. Goldstein, T. Pohida, P. D. Smith, J. I. Peterson, E. Wellner, M. Ghany, and R. F. Bonner, BioTechniques 26, 328 (1999). 5 D. K. Ornstein, C. Englert, J. W. Gillespie, C. P. Paweletz, W. M. Linehan, M. R. Emmert-Buck, and E. F. Petricoin, Clin. Cancer Res. 6, 353 (2000).
[13]
LCM TO ASSESS DEVELOPMENT
149
important to note that Western analysis of cells acquired by LCM is likely to be a less definitive method to determine protein distribution in discrete cells of heterogeneous tissues than immunohistochemistry. In contrast, Western analysis of LCM procured cells may be a more reliable quantitative method to assay specific cellular protein concentrations in response to drug treatments or environmental toxins than conventional immunohistochemistry, a method that is often fraught with investigator bias and at times not reproducible between different laboratories, even when using identical antibody reagents. Indeed, it is proposed that Western analysis of LCM cells may ultimately be the method of choice in which specific protein concentrations in normal or pathological tissues will be determined. A word of caution, however, is warranted. Capture of cells from a section by LCM does not represent the acquisition of complete cells; rather, a 5- to 6-µm section of a 12- to 15-µm diameter cell will represent as little as 30–50% capture of part of a cell. Thus, careful bookkeeping of cell equivalents will be necessary if Western analysis is to be used for quantitative purposes. Also, when trying to perform quantitative Western analysis, investigators are urged to err on the side of capture of large numbers of cells rather than just a few. While it may be temping to minimize the number of captured cell for each time point, significance of results is likely to be greater when Western analysis is performed on thousands of cells rather than hundreds, even when a signal can be detected in the smaller protein quantities. RNA Isolation Ever since the development of microarray technology, gene expression studies have become ever more present in the characterization of developmental processes. Serial analysis of gene expression (SAGE) is also rapidly gaining favor as a means to quantitatively determine gene expression profiles in normal and pathological conditions.6 The value of these analyses, however, principally relies on the quality of the starting material, i.e., the RNA. In addition, both of these gene expression analyses require relatively large amounts of starting RNA to ultimately use as probes. Thus, as with Western analysis, availability of sufficient cellular material from developmental processes is a key consideration before initiating a gene expression analysis of a specific cell type. At present, RNA isolation of good quality is best achieved working with frozen sections fixed in precipitating fixatives such as 70–95% ethanol.7 Although commercial kits are available indicating that RNA can be isolated from paraffin embedded tissues fixed in aldehydes, to date there are no reports in the literature certifying the veracity of such claims for LCM tissues. Indeed, it would seem 6 7
V. F. Velculescu, L. Zhang, B. Vogelstein, and K. W. Kinzler, Science 270, 484 (1995). S. M. Goldsworthy, P. S. Stockton, C. S. Trempus, J. F. Foley, and R. R. Maronpot, Mol. Carcinog. 25, 86 (1999).
150
SPECIALIZED USES
[13]
that such an approach is counterintuitive and unlikely to succeed, since the crosslinking properties of aldehydes may make RNA extraction from tissues less than ideal. In contrast, if tissues were to be fixed in precipitating fixatives such that RNA cross-linking would not occur and then embedded in paraffin, then it is quite likely that tissue histology would be significantly compromised. Nevertheless, given the amount of archival tissue material embedded in paraffin, especially human tissue, it is hoped that conditions are ultimately developed that permit extraction of good quality RNA from these collections. Fastidiousness is perhaps the ideal quality for an investigator to possess before embarking on RNA isolation. Working with LCM material requires double meticulousness, given that RNA degradation may occur at any of the steps taken by an investigator to capture cells to then extract the RNA. In the usual procedure to prepare tissue for LCM, animals are euthanized, and desired tissues quickly dissected free and then frozen by immersing in liquid nitrogen. After sectioning, sections are fixed in 70% ethanol and then stained with hematoxylin and eosin. During these steps the cells are subjected to various aqueous environments, e.g., freeze–thawing is required for the sections to attach to the glass slides and the hematoxylin step requires water rinses, with the possibility of RNases degrading the RNA. In addition, the tissue will come in contact with surfaces that may contain RNases, e.g., blade of knife, glass slides. To preempt these problems, it may be possible to perfuse animals with the 70% ethanol as a means to reduce aqueous exposure before freezing and the LCM steps. Unfortunately, for reasons unknown, some tissue histology is significantly impaired when processed in this manner, making it impossible to discern cellular integrity in tissue sections. Testes, for example, cannot be processed in this fashion, whereas liver, epididymis, and prostate from mice fixed by perfusing them with 70% ethanol are indistinguishable from untreated tissues, yet RNA recovery is greater and the quality of the RNA better. Investigators will have to check histology of their tissue of choice first if they wish to take this approach. Even when the problems of RNA quality extracted from LCM harvested cells are overcome an investigator must still have a means to extract enough RNA to then perform gene profiling studies. For example, current estimates are that microgram quantities of total RNA starting material are needed to perform microarray analysis and one report in the literature, in which the development of a microSAGE procedure was presented, indicated that at a minimum 0.2 µg of total RNA was required.8 While gaining access to large numbers of cells from a cell culture to isolate RNA is now a relatively trivial process, the same is not true for LCM. Indeed, the harvesting of 50,000 cell equivalents by LCM could easily take a week, including the time that it takes to prepare, section, and capture tissue. If the 8
N. A. Datson, J. van der Perk-deJong, M. P. van den Berg, E. R. de Kloet, and E. Vreugdenhil, Nucleic Acids Res. 27, 1300 (1999).
[13]
LCM TO ASSESS DEVELOPMENT
151
desired cells are rare, the time to harvest may be increased by one order of magnitude, if not longer. In the summer of 2001, Arcturus, Inc., introduced a new RNA amplification kit (RiboAmp) that will yield 10–50 µg of amplified RNA from as little as 250 cells. Independent verification of RiboAmp kit efficacy in amplifying this amount of RNA from a large number of cells and not introducing any bias into the RNA population is not yet present in the literature. However, the results presented by Arcturus, Inc., amplifying RNA using these kits and then using the amplified RNA in microarray analysis to distinguish the differential expression of specific genes in normal vs pathological breast tissues were dramatic. Clearly, if these kits deliver as promised, this new technology will open up new frontiers in the field of developmental biology. New Caps Attention must also be paid to the possibility of diluting the concentration of captured cells by the extraction solution used to remove cells from the thermoplastic film. Most micro RNA and protein isolation procedures, for example, are usually designed to work at 200 to 400 µl volumes, ideal volumes for cell lines and tissue isolations procedures but overwhelmingly large for limiting cell equivalents harvested LCM cells. Arcturus, Inc., has designed the CapSure cap in which captured cells are extracted with volumes as small as 10 µl. These new caps have an offset ridge approximately 10 µm high that creates a shallow well within the capture zone. In combination with an extraction device that fits over the flat support, captured cells are easily extracted from the thermoplastic film and pooled in 10 µl volumes. Another advantage of these newly designed caps is that the offset ridge eliminates any contact between the thermoplastic film and the tissue section, except at points where the film is melted and diffused into the targeted area. Thus, the probability of nontargeted cells attaching to the thermoplastic film is further reduced. Although in general this is not a consideration when capturing nonlimiting tissues (see discussion above), any nonspecific capture may significantly compromise the quality of the starting material when availability is limiting, as is likely to be the case with developing tissue. The CapSure caps were specifically designed for this purpose. Targeting Decisions The current PixCell II LCM instrument permits the investigator to target and capture three different-sized spheres of tissues from tissue sections, 60 µm, 30 µm, and 7.5 µm in diameter, respectively. Once the tissue sections are prepared using a cryostat, the use of the PixCell instrument to capture groups of cells greater than 30 µm in diameter becomes a trivial event. The PixCell II is easy to use and entails robust technology, and excellent results are often obtained by first-time
152
SPECIALIZED USES
[13]
users. Thus, a detailed description of how to use the instrument to capture cells is not presented here. Instead, a theoretical basis for how to target and capture discrete cells from heterogeneous tissues is presented. Single Cells The small spot size of 7.5 µm is ideally suited to capture most individual cells larger than 7.5 µm in diameter. The strategy for targeting and capture is no different from when larger diameter cell aggregates are desired. In principle, the geometric center of the cell is aligned with the epicenter of the laser diode that is then activated. If the targeted cell is larger than 7.5 µm in diameter, then two or more laser shots are fired to diffuse the thermoplastic film into the targeted area. One feature of the capture process that is yet to be understood, however, is how a true spherical laser spot melts the thermoplastic film to capture the entirety of a cell that is not a true sphere (see Fig. 1). The assumption is made that the diffusion of the thermoplastic film encounters resistance to flow across cell boundaries, and at the periphery of the targeted areas, it tends to flow along the path of least resistance, i.e., to the periphery of the cell. Regardless of how thermoplastic film diffusion is altered by tissue architecture, this feature of the capture process is routinely observed and can be exploited in the targeting and capture of discrete cells. While the capture of a mostly spherical cell appears to be a relatively simple process, the same cannot be said for a highly irregular shaped cell, in particular if the area to be captured falls below the 7.5 µm threshold and these are interspersed within nondesired cells. Although a solution to this problem is yet to be achieved for all cases likely to be encountered, two strategies may be employed to overcome in part the limitations of the 7.5 µm diameter capture zone. First, if the goal of the research project is to capture a discrete cell type to the highest degree of purity, then investigators must realize that it is not necessary to capture all of
FIG. 1. LCM harvest of a single liver cell of diameter 10 µm. Targeted cell in section (A) and void left in section (B) are indicated by asterisks. Captured cell on cap is presented in C. Although the laser shoots a spherical beam, the thermoplastic film diffuses into the single cells so as to approximate its cell boundary.
[13]
LCM TO ASSESS DEVELOPMENT
153
the cytoplasm of the desired cell type present within a particular tissue section. The capture zone may be extended solely to the cytoplasm of the perinuclear area, likely to be within the 7.5 µm capture zone. In addition, the capture zone does not need to fully extend to the cell’s perimeter; thus, capture yield could be sacrificed to gain in cell purity. As discussed above, the more sensitive chemiluminescent detection kits for Western analysis and the RNA amplification kits are likely to make this strategy worthwhile as these latter technologies gain wider acceptance. Second, paradoxically, histologists strive for maintenance of cell–cell contacts in preparing tissues for fixation, embedding, sectioning, and imaging. The history of tissue processing for imaging is a long one, and investigators are biased toward methods that tend to leave few if any spaces between cells. Indeed, it is the axiom of morphologists that there are no true spaces in biology. Unfortunately, this investigator bias may work against designing tissue processing strategies that leave cellular architecture nearly intact (so that desired cells can be targeted in tissue sections), but generates natural “cleavage points” between cells of complex tissues. One such approach that may prove useful is to perfuse animals with hyperosmolar solutions of saline (600 mosmol) before harvesting tissues. The necessary time to subject different tissues to a hyperosmolar condition may vary from tissue to tissue and will have to be developed empirically by each investigator; however, preliminary studies using this strategy to try and separate cells of the seminiferous epithelium have proven surprisingly successful. Whether or not the quality of the RNA is drastically altered remains to be determined. Epithelia The capture of discrete cell types from a columnar epithelium is presented in Fig. 2. As with spherical cells, specific targeting decisions must be employed to capture desired cell types or discreet areas of cells. The capture zone may be
FIG. 2. LCM of cells from a columnar epithelium. A single capture event is demonstrated in A–C. Arrow points to epithelium before capture (A), void left in tissue (B), and void at higher magnification in (C). Note that basally-located nucleus is left in section, and only apical cytoplasm was harvested.
154
SPECIALIZED USES
[13]
FIG. 3. LCM capture of epithelial cells. Using the larger spot size (30 µm), three different areas of epithelium were targeted and captured. By continuing to link the different capture zones, it is possible to harvest the complete epithelium, yet leave the myoepithelial layer behind.
initiated from a luminal direction, as indicated in the figure, or the interstitium surrounding the tubules may be approached first. The precision and reliability of the capture zone is such that the investigator has several options to choose from in targeting and capturing cells forming an epithelium. First, the apical cytoplasm can be targeted and captured. Next, the basally located nuclear zone can be individually removed, followed by the myoepithelial layers. Finally, the surrounding interstitium can be targeted and captured. Each of the discrete capture events can then be pooled and used as starting material for subsequent analysis. Using this approach, large numbers of cells that may be less than the 7.5 µm diameter of the capture zone can be harvested from columnar epithelia. In Fig. 3, the beginning of this process is presented. The capture zones are eventually joined together and the captured material pooled in one cap to minimize dilution of the harvested cells. Tissue Preparation In preparing tissue for LCM, all general conditions necessary to maintain an RNase-free environment are absolutely required. Instruments that will come in contact with fresh tissues must be baked, and all glassware, knives to be used in sectioning, etc., should be treated with RNase AWAY. Fresh gloves should be worn at all times when handling tissues, sectioning, and transferring sections to LCM apparatus. Gloves should be changed between entering body cavity and removing desired tissue, and instruments should also be changed between those that were used to open body wall and those used to remove tissue from euthanized animal. Freezing of specimens is performed in liquid nitrogen. Standard protocol for embryos is to first place embryos in sterile 30% sucrose solution until embryos sink to the bottom of a scintillation vial (usually takes several hours). Not only does this cryoprotect the embryos, but subsequent RNA extraction is improved.
[13]
LCM TO ASSESS DEVELOPMENT
155
Similar treatments with male reproductive tissues yielded excellent results. Next, remove tissue from sucrose, place into cryomold filled with OCT compound, and place in liquid nitrogen. Embryos treated in this fashion may be stored in a −80◦ freezer until sectioning. Sectioning is performed in a conventional cryostat whose knife edge and any area that may come in contact with the tissue is cleaned with RNase AWAY. (Various reports in the literature suggest that the glass slides should not be coated in any way if they are to be used for LCM. My own experience is that Fisher Plus slides do not compromise in any way LCM of cells.) After sectioning, tissue needs to be fixed and stained with hematoxylin and eosin. Because of RNA susceptibility, each slide should be processed individually, and extreme care should be taken that tissue contact with water is minimized. Ideal times for fixation and staining are available at the Arcturus, Inc. web site (
[email protected]). A detailed and comprehensive set of instructions is presented at this site and need not be reproduced here. At this point, the sections are ready for LCM. In general, batches of 4–5 slides are prepared and used for LCM before generating more sections. In this way the investigator can ensure that the majority of the sections are subjected to identical fixation, staining, and environmental times. For RNA Extraction As described above, Arcturus, Inc. now sells a kit to extract and amplify the RNA from LCM cells. The convenience and quality control of using such a kit to prepare larger numbers of RNA aliquots from different LCM sample pools cannot be overstated. Nevertheless, successful RNA extraction from LCM samples was achieved by several investigators using a variety of micro RNA isolation kits. (Our laboratory uses the Stratagene micro RNA isolation kits and precisely follows the instructions provided in the instructions manual.) The equivalent of 50,000 captured cells rendered enough total RNA to perform five reverse transcriptions and PCR. Western Analysis There are so many variables possible in rendering a robust signal by Western analysis that it is not recommended to put in place an assay working with LCM cells and using an uncharacterized antibody probe. If, however, an investigator is familiar with the particular characteristics of an antibody probe, then Western analysis of LCM is feasible. The least number of cells harvested by LCM that will render a robust signal will have to be determined empirically. As a minimum starting material, 5000 LCM cells should be assayed and the numbers adjusted accordingly. Extraction of proteins for subsequent PAGE and transfer should be performed with previously established conditions that render positive results for the particular protein of interest. It is convenient to pool captured cells from different
156
SPECIALIZED USES
[13]
caps into one single Eppendorf tube as follows: Add 100 µl of extraction medium to tube and attach cap, vortex for 1 min, centrifuge to collect solution, and transfer to a new cap. Repeat steps until desired number of captured cells is extracted. At this point continue with PAGE, transfer, and blot as per usual. (We have had significant success using T-PER Tissue Protein Extraction Reagent from Pierce, Inc., to extract captured cells on caps. An advantage of using these reagents is their compatibility with standard protein assays.) Concluding Remarks In this paper general considerations and strategies were presented to help guide investigators embarking on LCM. The technical event of the capture is a straightforward process requiring a PixCell instrument. The PixCell II instrument is capable of targeting and capturing tissue spot sizes approximately 7.5 µm and greater in diameter. Thus, creative strategies must be exerted in designing experiments to harvest discrete cell types from heterogeneous tissues. While round cells are still the cells of choice to harvest by LCM, it is hoped that irregularly shaped cells, or at least parts of these cells, may become amenable to harvesting. The technology to assay LCM acquired cells has improved rapidly since the initial report of EmmertBuck et al. in 1996.9 More recently, robust methods to extract and amplify RNA so as to make gene profiling experiments feasible appeared in the market. Commercial reagents are also available to facilitate Western analysis. Development of these novel technologies promises to bridge LCM and gene profiling assays to better define the temporal, macromolecular content of discrete cell types. As this information accumulates, windows into normal developmental processes will emerge, as well as the result of catastrophic outcomes resulting from toxic insults. Addendum A new version of the PixCell apparatus, PicCell IIe, is now available from Arcturus, Inc. It allows the capture of 5-µm-diameter tissue spots, significantly facilitating the harvest of single cells. Acknowledgments The international collaboration of the work presented here was made possible by a Travel Grant to C.A.S.-Q. from the Office of the Provost, Georgetown University, to facilitate academic exchanges between non-U.S. universities and Georgetown.
9
M. R. Emmert-Buck, R. F. Bonner, P. D. Smith, R. F. Chuaqui, Z. Shuang, S. R. Goldstein, R. A. Weiss, and L. A. Liotta, Science 274, 998 (1996).
[14]
APPLICATION OF
LCM TO PROTEOMICS
157
[14] Application of Laser Capture Microdissection to Proteomics By K. K. JAIN Introduction Laser capture microdissection (LCM) provides an ideal method for extraction of cells from specimens in which the exact morphologies of both the captured cells and the surrounding tissue are preserved. LCM methods (for an example, see Ref. 1) enable successful extraction of DNA, mRNA, and proteins from tissue fragments down to the single cell level. Selected cells are retrieved by activation of a transfer film placed in contact with a tissue section and use of a laser beam focused on a selected area of tissue through an inverted microscope. The laser bonds the film to the tissue beneath it and these cells are then lifted free of surrounding tissue. LCM is compatible with subsequent nucleic acid analysis of tissues and is important for comprehensive molecular characterization of normal, precancerous, and malignant cells by means of DNA-array technology. However, discovering the genetic sequence encoding a protein is not sufficient for predicting the size or biological nature of a protein. Studies at the messenger RNA (mRNA) level can assess the expression profiles of transcripts, but these analyses only measure the relative amount of an mRNA encoding a protein and not the actual amount of protein in a tissue. This is particularly important in cancer research where posttranslational modifications of a protein can specifically lead to the disease. To address this area, several protein-based analysis technologies have been developed. Proteomics, is a term that indicates study of the “proteome” (PROTEins expressed by a genOME) and is applied to the systematic analysis of protein profiles of tissues.2 LCM has been shown to be a valuable adjunct to proteomic technologies, particularly in the study of cancer.3 Immunohistochemical Staining Techniques Prior to LCM Microdissection of routinely stained or unstained frozen sections has been used successfully to obtain purified cell populations for the analysis of cell-specific gene expression patterns in primary tissues with a complex mixture of cell types. However, it may be difficult to identify different cell types and structures by morphology 1
Janette Burgess and Brent E. McParland, Methods Enzymol. 356, [22], 2002 (this volume). K. K. Jain, “Proteomics: Technologies, Companies and Markets.” Jain Pharmabiotech Publications, Basel, Switzerland, 2002. 3 K. K. Jain, Pharmacogenomics 1, 385 (2000). 2
METHODS IN ENZYMOLOGY, VOL. 356
Copyright 2002, Elsevier Science (USA). All rights reserved. 0076-6879/02 $35.00
158
SPECIALIZED USES
[14]
alone, especially in tissues which lack easily discernible architectural features. Therefore, several groups have adapted immunohistochemical staining techniques for tissues to be examined by LCM. In addition to providing high contrast targets for microdissection, immunostaining allows selection of cells according to not only morphological, but also phenotypical and functional criteria. In order to allow reliable tissue transfer and preserve the integrity of the target of analysis such as DNA, RNA, and proteins, immunostaining protocols have to be modified when combined with LCM. If the problem of RNA recovery from immunostained sections can be resolved satisfactorily, gene expression can be correlated with phenotypic and functional properties of the examined cell population, such as proliferation, maturation stage, and oncoprotein expression.4 Microdissection Techniques Used for Proteomics The PixCell II Laser Capture Microdissection System (LCM) was initially conceived at the U.S. National Institutes of Health and commercially developed by Arcturus Engineering, Inc. (Mountain View, CA). It can locate a single cell or a group of cells in minutes and extracts them for molecular analysis using a simple aim-and-shoot method. It uses a low power infrared excitation to eliminate sample photodegradation. A related but different technology is laser beam microdissection in which a pulsed ultraviolet laser cuts out cells of interest by photoablation of adjacent tissue. One example of this type is UV CUT (SL Microtest, Jena, Germany). It enables cell-specific extraction from heterogeneous tissue for the analysis of DNA, RNA, and proteins. The cold ablation of the UV laser avoids any heating of the tissue. In particular, the material to be extracted is never directly exposed to the laser but only circumscribed by it. Another technique of laser microdissection for in vitro incision and excision of biological specimens and other microscopic particles is by a LaserScissors module and computer control provided by the Microscope Workstation (Cell Robotics International Inc., Albuquerque, NM). It can make laser microscopic incisions using standard microscope optics and can biopsy at the single cell level. It enables the microdissection of pathology samples for DNA, RNA, and protein analysis as a diagnostic instrument. MicroBeam Laser Pressure Catapulting (P.A.L.M. Microlaser Technologies, Bernried, Germany) dissects selected specimens from various sources with a pulsed nitrogen laser and allows precise cutting or microdissection.5 The specimens are ejected directly into a standard microfuge tube for subsequent proteomic analysis.
4 5
F. Fend, M. Kremer, and L. Quintanilla-Martinez, Pathobiology 68, 209 (2000). K. Schutze and G. Lahr, Nat. Biotechnol. 16, 737 (1998).
[14]
APPLICATION OF
LCM TO PROTEOMICS
159
TABLE I PROTEOMIC TECHNOLOGIES USED IN COMBINATION WITH LCM Protein separation Two-dimensional gel electrophoresis: 2D polyacrylamide gel electrophoresis (PAGE) Protein detection Silver staining Western blot Fluorescence detection Protein identification and characterization Mass spectrometry (MS): matrix-assisted laser desorption mass spectrometry High performance liquid chromatography ProteinChip: surface-enhanced laser desorption/ionization
Uses of LCM in Combination with Proteomic Technologies Proteomics technologies that can be combined with LCM are shown in Table I. The application of proteomics to microdissected tissues has opened a new bridge to “molecular morphology.” The feasibility of 2D gel electrophoresis, immunoblotting, and immunoassays performed on cells obtained by LCM has been demonstrated. 2D Gel Electrophoresis Proteins can be recovered from laser capture microdissected tissue in a form suitable for 2D gel electrophoresis. In one study on colon cancer tissue, the solubilized proteins retained their expected electrophoretic mobility in 2D gels as compared with bulk samples, and mass spectrometric analysis was also unaffected.6 LCM has been used to supply purified normal human and renal cortical tissues for proteomic analyses.7 The most widely used proteomics tool is 2D PAGE, which can display protein expression patterns to a high degree of resolution. However, 2D PAGE can be time consuming; the analysis is complicated and, compared with DNA techniques, is not very sensitive. Although some of these problems can be alleviated by using high-quality homogeneous samples, such as those generated using microdissection techniques, the quantity of sample is often limited and may take several days to generate sufficient material for a single 2D PAGE analysis. Currently, it is possible to microdissect approximately 50,000 cells, which reveal approximately 675 distinct proteins as visualized by 2D PAGE stained with silver.
6 7
L. C. Lawrie, S. Curran, and H. L. McLeod, Mol. Pathol. 54, 253 (2001). R. E. Banks, M. J. Dunn, and M. A. Forbes, Electrophoresis 20, 689 (1999).
160
SPECIALIZED USES
[14]
ProteinChip ProteinChip (Ciphergen Biosystems, Fremont, CA) uses patented SELDI (surface-enhanced laser desorption/ionization) to rapidly perform the separation, detection, and analysis of proteins at the femtomole level directly from biological samples. The ProteinChip system can replace and complement a wide range of traditional analytical methods, which not only are more time consuming but require specialized scientific expertise. As an alternative to 2D PAGE, SELDI allows the retention of proteins on a solid-phase chromatographic surface or ProteinChip Array with direct detection of retained proteins by time-of-flight mass spectrometry. Known proteins are analyzed using on-chip functional assays. For example, chip surfaces can contain enzymes, receptor proteins, or antibodies, enabling onchip protein–protein interaction studies, ligand binding studies, or immunoassays. Using this system, tumor and normal tissue from head and neck cancer and microdissected melanoma have been analyzed to determine differentially expressed proteins.8 Comparisons of the protein expression patterns from microdissected normal and tumor tissues indicate several differences, highlighting the importance of extremely defined tissue lysates for protein profiling. By applying this fast and powerful ProteinChip array technology it becomes possible to investigate complex changes at the protein level in cancer associated with tumor development and progression. Applications of LCM in Proteomics Applications of LCM in proteomics are listed in Table II. LCM and Oncoproteomics The most important use of LCM is in cancer proteomics, also referred to as oncoproteomics. LCM is being used in the Cancer Genome Anatomy Program for systematic identification and cataloging of known and novel genes expressed during tumor development. It can be applied to any disease process accessible through tissue sampling, such as premalignant lesions. LCM has played an important role in basic investigations of oncoproteomics. CD34, a heavily glycosylated transmembrane protein of ∼110 kDa is found in several neoplasms and has immunohistological reactivity with anti-CD34 antibodies. Laser capture microdissection of CD34 immunohistologically reactive epithelioid sarcoma and nonreactive epidermal cells illustrates that this reactivity for anti-CD34 antibodies in apparently unrelated tumors is specific to tumor cells not to shared epitopes on unrelated proteins.9 Analysis on a microdissected 8 9
F. von Eggeling, H. Davies, and L. Lomas, Biotechniques 29, 1066 (2000). Y. Natkunam, R. A. Warnke, and B. Haghighi, J. Cutan. Pathol. 27, 392 (2000).
[14]
APPLICATION OF
LCM TO PROTEOMICS
161
TABLE II APPLICATIONS OF LCM IN PROTEOMICS Oncoproteomics Establishing the genetic fingerprints of malignant neoplasms Differential protein expression of cancer versus normal cells Cancer Genome Anatomy Project Diagnosis of premalignant lesions Anticancer drug development Proteomic diagnostics Neuroproteomics Tissue proteomics Drug discovery and development of personalized medicines
epithelioid sarcoma shows that the ∼110-kDa band is present in the sample containing tumor cells, which excludes the possibility that adjacent nonneoplastic cells are responsible for the immunoblot result. High-density protein arrays, antibody arrays, and small molecular arrays, coupled with LCM, could have a substantial impact on proteomic profiling of human malignancies.10 In combination with techniques such as expression library construction, cDNA array hybridization, and differential display, LCM enables the establishment of “genetic fingerprints” of specific pathological lesions, especially malignant neoplasms.11 This approach could help in establishing individualized treatment tailored to the molecular profile of a tumor. LCM facilitates studies of new anticancer drugs. By analyzing the protein composition of tumors, the beneficial and toxic effects of treatments can be determined in the laboratory before the treatments are used in patients, thus optimizing chemotherapy. Examples of the use of LCM in the study of cancers of some organs are given in the following paragraphs. Head and Neck Squamous Cell Carcinoma. Cancers of the oral cavity, salivary glands, larynx, and pharynx, collectively referred to as squamous cell carcinomas of the head and neck (HNSCC), are the sixth most common cancer among men in the developed world. The prognosis of HNSCC patients is still poor, which reflects the fact that although the risk factors for HNSCC are well recognized, very little is known about the molecular mechanisms responsible for this malignancy. LCM technologies will be helpful in molecular studies aimed at revealing the mechanisms involved in squamous cell carcinogenesis. They are also expected to provide a molecular blueprint for HNSCC, thus helping to identify suitable markers for the early detection of preneoplastic lesions, as well as novel targets for pharmacological intervention in this disease.12 10
V. E. Bichsel, L. A. Liotta, and E. F. Petricoin, Cancer J. 7, 69 (2001). F. Fend and M. Raffeld, J. Clin. Pathol. 53, 666 (2000). 12 V. Patel, C. Leethanakul, and J. S. Gutkind, Crit. Rev. Oral Biol. Med. 12, 55 (2001). 11
162
SPECIALIZED USES
[14]
Prostate Cancer. The proportion of unbound serum prostate-specific antigen (PSA; percent free PSA) is reported to be lower in men with prostate cancer, compared to men with benign prostates. The majority of immunoreactive PSA in serum is complexed to α 1-antichymotrypsin (ACT). LCM has been used to assess the bound versus free form of intracellular PSA in both benign and malignant epithelium procured from prostate tissue.13 Western blotting analysis of one-dimensional PAGE gels revealed that in the vast majority of intracellular tumors, normal PSA exists within cells in the “free” form. Binding studies showed that PSA recovered from LCM-procured cells retained full ability to bind ACT, and 2D PAGE Western analysis demonstrated that the PSA/ACT complex was stable under strong reducing conditions. Intracellular PSA, therefore, exists in the “free” form and binding to ACT occurs exclusively outside the cell. Prostate-specific antigen or histological examination of bulk tissue may not accurately reflect molecular events that take place in the actual ductal epithelium that change as a consequence of the malignant process in the prostate gland. Alternative proteomic-based approaches including LCM enable the identification of protein markers in the actual premalignant and frankly malignant epithelium.14 The phenotype of a given cell type is ultimately determined by the composition and activation status of its proteins. Therefore, quantitative and qualitative proteomic measurement of normal and neoplastic prostate cells is an important experimental approach that will complement genomic DNA and gene expression analyses. The National Cancer Institute Prostate Group has been studying protein profiles of prostate cancer using tissue microdissection and two protein analysis methods: 2D PAGE and SELDI. Specific populations of normal and malignant epithelium from radical prostatectomy tissue specimens, procured by LCM, were analyzed by 2D PAGE. Comparison of 2D PAGE profiles of microdissected cells with matched in vitro cell lines from the same patient and metastatic prostate cancer cell lines showed striking differences between prostate cells in vivo and in vitro with less than 20% shared proteins.15 The data demonstrate that 2D PAGE analysis of LCM-derived cells can reliably detect alterations in protein expression associated with prostate cancer, and that these differentially expressed proteins are produced in levels high enough to allow for their clinical utility as new targets for therapeutic intervention, serum markers, and/or imaging markers. Quantitation of the number of PSA molecules/cells was conducted on human prostate tissue cells procured by LCM from fixed and stained frozen sections.16 Calibration of the chemiluminescent assay was conducted by developing a standard 13
D. K. Ornstein, C. Englert, and J. W. Gillespie, Clin. Cancer Res. 6, 353 (2000). C. P. Paweletz, L. A. Liotta, and E. F. Petricoin, Urology 57(4 Suppl 1), 160 (2001). 15 D. K. Ornstein, J. W. Gillespie, and C. P. Paweletz, Electrophoresis 21, 2235 (2000). 16 N. L. Simone, A. T. Remaley, and L. Charboneau, Am. J. Pathol. 156, 445 (2000). 14
[14]
APPLICATION OF
LCM TO PROTEOMICS
163
curve using known concentrations of PSA. After the sensitivity, precision, and linearity of the chemiluminescent assay were verified for known numbers of solubilized microdissected tissue cells, it was then possible to calculate the number of PSA molecules per microdissected tissue cell for case samples. Immunohistochemical staining of human prostate for PSA was compared with the results of the soluble immunoassay for the same prostate tissue section. Independent qualitative scoring of anti-PSA immunohistochemical staining intensity paralleled the LCM quantitative immunoassay for each tissue subpopulation and verified the heterogeneity of PSA content between tissue subpopulations in the same case. Extraction buffers were successfully adapted for both secreted and membrane-bound proteins. This technology has broad applicability for the quantitation of protein molecules in pure populations of tissue cells. SELDI ProteinChip MS technology has been used for the rapid, reproducible, and simultaneous identification of four well-characterized prostate cancerassociated biomarkers: PSA (free and complexed forms), prostate-specific peptide, prostate acid phophatase, and prostate-specific membrane antigen in cell lysates, serum, and seminal plasma.17 Proteins corresponding to the mass of these biomarkers could readily be captured and detected using either chemically defined or antibody coated ProteinChip arrays. Several other proteins were found up-regulated in cell lysates of pure populations of prostate cancer-associated cells procured by LCM when compared with mass spectra of normal cell lysates. Coupling LCM with SELDI provides tremendous opportunities to discover and identify the signature proteins associated with each stage of tumor development. Esophageal Cancer. 2D PAGE analysis of cells procured by LCM from human esophageal cancer specimens and comparison of the microdissected protein profiles showed a high degree of similarity (98% identical) between the matched normal-tumor samples.18 Protein profiling of LCM-captured cells is shown in Fig. 1. In this study, 17 proteins showed tumor-specific alterations, including 10 that were uniquely present in the tumors and seven that were observed only in the normal epithelium. Two of the altered proteins were characterized by mass spectrometry and immunoblot analysis and were identified as cytokeratin 1 and annexin I. Acquisition of 2D PAGE protein profiles, visualization of disregulated proteins, and subsequent determination of the identity of selected proteins through high-sensitivity MS-MS microsequencing are possible from microdissected cell populations. These separation and analytical techniques are uniquely capable of detecting tumor-specific alterations. Annexin I protein expression was evaluated in patient-matched longitudinal study sets of LCM normal, premalignant, and invasive epithelium from human
17 18
G. L. Wright, L. H. Cazares, and S. M. Leung, Prostate Cancer Prostatic Dis. 2, 264 (2000). M. R. Emmert-Buck, J. W. Gillespie, and C. P. Paweletz, Mol. Carcinogenesis 27, 158 (2000).
164
SPECIALIZED USES
Normal epithelium
[14]
Tumor epithelium
FIG. 1. 2D PAGE (polyacrylamide gel electrophoresis) comparison of microdissected normal squamous epithelium and tumor cells. Circled proteins in the normal epithelium panel indicate proteins that are down-regulated. Up-regulated proteins are 3 blobs in an oval on the tumor epithelial panel on the right. Reproduced from M. R. Emmert-Buck, J. W. Gillespie, and C. P. Paweletz, Mol. Carcinogenesis 27, 158 (2000) by permission of John Wiley & Sons.
esophageal squamous cell cancer.19 Western blot and immunohistochemistry show either complete loss or a dramatic reduction in the level of annexin I protein expression in premalignant lesions compared with patient-matched normal epithelium. Correlation of this finding with immunohistochemical analysis suggests that annexin I may be an essential component for the maintenance of the normal epithelial phenotype and that functional consequences of annexin I protein loss in tumor cells need further investigation. Use of LCM in Proteomic Diagnostics The new proteomic-based approaches to molecular diagnostics are aimed at the identification and investigation of protein markers in the actual histologically defined cell populations immersed in heterogeneous diseased tissue. It is envisioned that these investigations will eventually lead to novel diagnostic, prognostic, or therapeutic markers that can be applied to monitor therapeutic toxicity or efficacy.20 Coupling LCM with sensitive, quantitative chemiluminescent immunoassays has broad applicability to normal, diseased, or genetically modified tissue in the field of proteomics. The fluctuations of expressed genes or alterations in the cellular DNA that correlate with a particular disease stage can ultimately be compared within or between individual patients. Such a fingerprint of gene expression patterns can provide 19 20
C. P. Paweletz, D. K. Ornstein, and M. J. Roth, Cancer Res. 60, 6293 (2000). N. L. Simone, C. P. Paweletz, and L. Charboneau, Mol. Diagn. 5, 301 (2000).
[14]
APPLICATION OF
LCM TO PROTEOMICS
165
crucial clues for the etiology and might ultimately contribute to diagnostic decision and therapies tailored to individual patients. Molecules found to be associated with a defined pathological lesion might serve as imaging or therapeutic agents. LCM and Neuroproteomics Neuroproteomics is the term used for the application of proteomics to the study of the nervous system and its disorders with the aim of developing diagnostics and therapeutics. Several abnormalities of proteins in neurological disorders can be detected by examination of body fluids such as blood and cerebrospinal fluid. These findings can be strengthened through subsequent proteomic analysis of specific brain areas implicated in the pathology. New isolation strategies of clinically relevant cellular material such as LCM, protein enrichment procedures, and proteomic approaches to neuropeptide and neurotransmitter analysis give us the opportunity to map out complex cellular interaction at an unprecedented level of detail. In neurological disorders multiple underlying pathogenic mechanisms as well as acute and chronic CNS disease components may require a selective repertoire of molecular targets and biomarkers rather than an individual protein to better define a complex disease. The resulting proteome database bypasses many ambiguities of experimental models and may facilitate preclinical and clinical development of more specific disease markers and new selective fast-acting therapeutics.21 LCM has been used with cDNA microarrays to profile gene expression of adjacent neuronal subtypes.22 Expression profiles generated with this integration are useful for screening cDNAs as well as for producing databases of cell-type specific gene expression. Coordinate gene expression may indicate functional coupling between the encoded proteins and facilitates determination of function of most cDNAs now cloned. A detailed analysis of individual cell types such as hippocampal glial cells is necessary for investigating a novel putative therapeutic target. This may be achieved by LCM. MicroBeam laser pressure catapulting can be used for selective catapulting of single neurons to study degenerative neurological disorders such as Alzheimer’s and Parkinson’s diseases. It is possible to capture single plaques or amyloid deposits for proteomic analysis. There is also the potential to investigate single prion deposits in patients with Creutzfeldt–Jakob disease. Use of LCM has been considered for use on postmortem brain tissue, which can yield good quality mRNA and intact protein antigens enabling the successful application of molecular profiling techniques. The combination of LCM with high throughput profiling techniques offers opportunities to obtain precise genetic fingerprints of individual neurons allowing comparisons of normal and pathological states.23 21
C. Rohlff, Electrophoresis 21, 1227 (2000). L. Luo, R. C. Salunga, and H. Guo, Nat. Med. 5, 117 (1999). 23 S. Bahn, S. J. Augood, and M. Ryan, J. Chem. Neuroanat. 22, 79 (2001). 22
166
SPECIALIZED USES
[14]
Tissue Proteomics Tissue Proteomics uses LCM as a launching point for proteomics. The project, which began 3 years ago, is a joint effort between the FDA and NCI. Study of proteins in microdissected tissue will help early cancer detection, cancer prevention, and therapy, as well as the development of clinical trials. This program uses existing technologies, such as 2D gel electrophoresis and SELDI, and developing new technologies to generate protein “fingerprints.” More than 100 proteins have been identified to date and are currently being validated. The Tissue Proteomics initiative is to ultimately have a strong impact on the drug development process. The joint initiative was designed to address pharmaceutical drugs rejected by the FDA, usually for two reasons: unforeseen toxicity or inaccurate drug targets. With the help of LCM, the joint initiative is generating protein fingerprints at each stage of cancer development. This information will identify the critical proteins involved in cancer development and progression, will provide new therapeutic targets, and will potentially match tumor stage with therapeutic strategies to improve treatment regimens. In addition, the FDA can test new drugs on the market for efficacy by examining changes in protein profiles before and after drug treatment. The initiative will also generate early toxicity fingerprints, such as those occurring during vascular damage, and identify specific toxicity markers. With this collection of information, they can develop algorithms and test new drugs for early toxicity based on their defined protein fingerprints. Drug Discovery and Development of Personalized Medicine Arrays using LCM-procured cancer epithelial cells can test the functional status of the pathways of interest and may be used for rapid identification of targets for pharmacologic intervention, as well as for assessment of the therapy in correcting the deranged pathways. The impact of proteomics on human cancer and diseases will not be limited to the identification of new biomarkers for early detection and new targets. These tools will be used to design rational drugs tailored according to the molecular profile of the protein circuitry of the diseased cell. Proteomics will provide one approach to personalized medicine. Advantages of LCM LCM is fast, high-throughput, reproducible, and precise based on high resolution. LCM requires little manual dexterity and provides opportunity for documentation. Unlike traditional cell-isolation methods, LCM enables the isolation of normal, precancerous, and tumor cells from a patient biopsy while maintaining the protein expression patterns of the cells.
[15]
LCM OF MOUSE INTESTINE
167
After LCM, the tissue remaining on the slide is fully accessible for comparative molecular analysis of adjacent cells. LCM is a convenient alternative to cell culture needed for biopsy tissue from patients during clinical trials when only a few hundred cells are available to work with.
Limitations of LCM One problem with LCM is that the sample to be studied is subjected to the heat and radiation of the laser, possibly damaging the cellular nucleic acids. This can be avoided by using the laser to circumscribe the cells to be excised in a process called “laser ablation.” As the sample is never actually exposed to the laser, there is no possibility of damaging it. Tissue sections without coverslip, necessary for LCM, become dehydrated leading to a significant decrease in optical resolution. Poor quality of nucleic acids and proteins from the archival samples is a limitation of LCM application. A drawback of LCM is working with frozen samples. To overcome these problems, a rapid simple method for the enrichment of normal and tumor cells has been developed called epithelial aggregate separation and isolation. Although LCM is capable of procuring single cells, some nonspecific tissue retained on the capture membrane can contaminate a sample containing only a few cells.
[15] Laser Capture Microdissection of Mouse Intestine: Characterizing mRNA and Protein Expression, and Profiling Intermediary Metabolism in Specified Cell Populations By THADDEUS S. STAPPENBECK, LORA V. HOOPER, JILL K. MANCHESTER, MELISSA H. WONG, and JEFFREY I. GORDON The adult mouse intestine has several features that make it an attractive system for using laser capture microdissection (LCM). It is composed of repeating discrete anatomic units. Epithelial cell proliferation, differentiation, migration, and cell death occur continuously within well-defined regions of each of these units. Moreover, the stem cell hierarchy of the intestine makes it ideally suited for performing genetic mosaic analyses with LCM. In such analyses, mice are
METHODS IN ENZYMOLOGY, VOL. 356
Copyright 2002, Elsevier Science (USA). All rights reserved. 0076-6879/02 $35.00
[15]
LCM OF MOUSE INTESTINE
169
created whose intestines contain cohorts of genetically engineered cells juxtaposed to clearly demarcated cohorts of normal cells. These normal cells serve as internal controls for defining the consequences of genetic manipulation of the “experimental” population. This chapter reviews the structural and functional organization of the mouse intestine and describes methods for LCM of targeted cell populations. Protocols are presented for real-time quantitative RT-PCR and DNA microarray-based profiling of mRNA levels in captured cells. In addition, we describe procedures for analyzing cellular proteins and for performing microanalytic biochemical assays to characterize intermediary metabolism in retrieved cells. Organization of Mouse Intestine Small Intestine The adult mouse small intestine is composed of crypt-villus units (Fig. 1A). Epithelial proliferation occurs in flask-shaped mucosal invaginations known as crypts of Lieberk¨uhn. The small intestine contains ∼106 crypts, with each crypt containing ∼250 epithelial cells. Long-lived active multipotential stem cells are positioned at or near the base of each crypt.1–9 These stem cells produce committed daughters10 that undergo several rounds of cell division, forming a transit amplifying (TA) population of ∼150 cells positioned in the middle of each crypt. The small intestine contains four epithelial lineages: absorptive enterocytes plus three types of secretory cells: goblet, enteroendocrine, and Paneth (Fig. 1B). Paneth cells complete their differentiation at the crypt base. Each crypt contains 30–50 mature Paneth cells that have an average lifespan of 18–23 days.5 These cells secrete antimicrobial peptides, digestive enzymes, and growth factors. Enterocytes 1
H. Cheng and C. P. Leblond, Am. J. Anat. 141, 461 (1974). H. Cheng and C. P. Leblond, Am. J. Anat. 141, 503 (1974). 3 H. Cheng and C. P. Leblond, Am. J. Anat. 141, 537 (1974). 4 H. Cheng, Am. J. Anat. 141, 481 (1974). 5 H. Cheng, Am. J. Anat. 141, 521 (1974). 6 M. Bjerknes and H. Cheng, Am. J. Anat. 160, 51 (1981). 2
FIG. 1. Crypt-villus units and their component cell populations. (A) Hematoxylin- and eosin-stained section of crypt-villus units located in the middle third (jejunum) of the adult mouse small intestine. The dashed line indicates the general boundary between the upper portion of crypts and the base of villi. The inset is a high power view of the boxed portion of the villus. The arrow points to an intraepithelial lymphocyte. (B) Current view of the origins of small intestinal epithelial lineages [M. Bjerknes and H. Cheng, Gastroenterology 116, 7 (1999); G. Rindi, C. Ratineau, A. Ronco, M. E. Candusso, M. Tsai, and A. B. Leiter, Development 126, 4149 (1999)]. (C) High power view of a crypt stained with hematoxylin and eosin. Paneth cells contain prominent apical secretory granules (open arrow). The arrowhead points to a pericryptal fibroblast. A submucosal plexus neuron is highlighted with an arrow. Bars in A and C = 25 µm.
170
SPECIALIZED USES
[15]
and goblet and enteroendocrine cells differentiate during a rapid, highly organized migration from the crypt up adjacent finger-shaped villi. Each villus is supplied by several crypts: the average number of crypts surrounding the base of each villus varies along the length of the intestine.11 The journey from the crypt to the villus tip takes 2–5 days and terminates with loss of cells by exfoliation and/or apoptosis. The steady-state epithelial cell population of a villus ranges from 2000 to 7000 cells, depending on where the villus is positioned along the cephalocaudal axis of the gut (average cellular census in proximal villi is greater than in distal villi).11 Colon The colon lacks Paneth cells12 and villi. The descendants of the multipotent colonic crypt stem cell migrate up onto a hexagonal surface epithelial cuff that surrounds the orifice of each crypt.13 Mesenchymal Cell Populations Crosstalk between mesenchymal and epithelial cells is essential for normal intestinal development.14–18 The adult intestinal mesenchyme contains a heterogenous collection of functionally important cellular elements. Crypts are surrounded by a layer of fibroblasts (Fig. 1C). These cells continuously divide and migrate upward from the crypt base,19 placing them in a strategic position to communicate with neighboring crypt epithelial lineage progenitors. The gut-associated lymphoid tissue (GALT) is composed of a complex repertoire of immune cells located within and underneath the epithelium20 (Fig. 1A). Genetically engineered T-cell receptor (TCR) δ subunit-deficient mice that lack crypt γ δ TCR+ intraepithelial lymphocytes (IELs) have reduced epithelial proliferation,21 indicating that there is 7
M. Bjerknes and H. Cheng, Am. J. Anat. 160, 77 (1981). M. Bjerknes and H. Cheng, Am. J. Anat. 166, 76 (1982). 9 M. H. Wong, J. R. Saam, T. S. Stappenbeck, C. H. Rexer, and J. I. Gordon, Proc. Natl. Acad. Sci. U.S.A. 97, 12601 (2000). 10 M. Bjerknes and H. Cheng, Gastroenterology 116, 7 (1999). 11 N. A. Wright and M. Irwin, Cell Tissue Kinet. 15, 595 (1982). 12 W. W. L. Chang and C. P. Leblond, Am. J. Anat. 131, 73 (1971). 13 G. H. Schmidt, M. M. Wilkinson, and B. A. J. Ponder, Cell 40, 425 (1985). 14 L. Karlsson, P. Lindahl, J. K. Heath, and C. Betsholtz, Development 127, 3457 (2000). 15 K. H. Kaestner, D. G. Silberg, P. G. Traber, and G. Schutz, Genes Dev. 11, 1583 (1997). 16 O. Pabst, R. Zeigerdt, and H.-H. Arnold, Development 126, 2215 (1999). 17 F. Beck, F. Tata, and K. Chawengsaksophak, BioEssays 22, 431 (2000). 18 M. Ramalho-Santos, D. A. Melton, and A. McMahon, Development 127, 2763 (2000). 19 M. N. Marsh and J. S. Trier, Gastroenterology 67, 636 (1974). 20 V. J. McCracken and R. G. Lorenz, Cell. Microbiol. 3, 1 (2001). 21 H. Komano, Y. Fujiura, M. Kawaguchi, S. Matsumoto, Y. Hashimoto, S. Obana, P. Mombaerts, S. Tonegawa, H. Yamamoto, S. Itohara, M. Nanno, and H. Ishikawa, Proc. Natl. Acad. Sci. U.S.A. 92, 6147 (1995). 8
[15]
LCM OF MOUSE INTESTINE
171
direct or indirect crosstalk between these immune cells and epithelial progenitors. The enteric nervous system (ENS) consists of a complex network of neurons and supporting glial cells that control motility, mucosal secretion, and blood flow. All of the different types of neurotransmitters found in the central nervous system are expressed in the gut. ENS neuronal cell bodies are located in two distinct anatomic regions: the myenteric and submucosal plexus (Fig. 1C). Those in the myenteric plexus send most of their axonal projections to the muscle layers of the gut to control motility. Those in the submucosal plexus send the majority of their projections to villi. Recent studies suggest that ENS may form part of the stem cell niche. Glucagon-like peptide 2 (GLP-2) produced by a subset of enteroendocrine cells is recognized by GLP-2 receptors expressed in ENS neurons. This interaction produces signals that regulate the proliferative activity of a long-lived committed stem cell daughter that gives rise to enterocytes.22 The intestinal mesenchyme also contains a microvasculature that surrounds the crypt and permeates the villus core. There are pronounced variations in epithelial differentiation, immune function, and microbial ecology along the length of the intestine.23 This remarkably complex spatial diversification makes LCM an ideal tool for recovering specified cell populations from a given region of the gut. Retrieval from frozen sections allows the influence of surrounding cells and the luminal environment on gene expression to be preserved. Genetic Mosaic Mouse Models and LCM The stem cell hierarchy of the intestine combined with the orderly nature of cell migration and the distinctive anatomic organization of crypt-villus units also make this tissue ideally suited for LCM-based genetic mosaic analyses. All active stem cells in each adult intestinal crypt appear to be derived from a single progenitor cell that occupies the highest position in the established stem cell hierarchy.9,24 Because of this clonal organization, an adult chimeric mouse, produced by introducing normal or genetically manipulated 129/Sv embryonic stem (ES) cells into C57Bl/6ROSA26 (B6-ROSA26) blastocysts,25 will contain discrete patches of crypt-villus units that are entirely 129/Sv or entirely B6-ROSA26. These patches can be distinguished because all B6-ROSA26 cells express Escherichia coli β-galactosidase (LacZ) which can be readily visualized by staining whole mount preparations or sections of intestine with X-Gal. If a villus is supplied with both monoclonal 129/Sv and monoclonal B6-ROSA26 crypts, it will appear striped. There will be coherent vertical columns 22
M. Bjerknes and H. Cheng, Proc. Natl. Acad. Sci. U.S.A. 98, 12497 (2001). P. G. Falk, L. V. Hooper, T. Midtvedt, and J. I. Gordon, Microbiol. Mol. Biol. Rev. 62, 1157 (1998). 24 G. H. Schmidt, D. J. Winton, and B. A. J. Ponder, Development 103, 785 (1988). 25 G. Friedrich and P. Soriano, Genes Dev. 5, 1513 (1991). 23
FIG. 2. The stem cell hierarchy and anatomy of the intestine make it well suited for genetic mosaic analysis. (A) Idealized three-dimensional drawing of the small intestine from an adult B6-ROSA26↔129/Sv chimeric mouse, produced by injecting 129/Sv ES cells into B6-ROSA26 blastocysts. The relationship between individual crypts and their adjacent villi is shown. In adult chimeras, crypts are composed of either B6-ROSA26 or 129/Sv epithelial cells, but not a mixture of both. Some villi are polyclonal, supplied by monoclonal 129/Sv crypts and monoclonal B6-ROSA26 crypts. (B) X-Gal and Nuclear Fast Red stained section of a jejunal crypt-villus unit showing a polyclonal villus. (C) Whole mount preparation of a segment of small intestine from an adult B6-ROSA26↔129/Sv chimeric mouse. The mid-portion of the intestine (jejunum) has been opened, fixed, and stained with X-Gal. The white epithelium is 129/Sv. The blue epithelium is composed of LacZ-expressing B6-ROSA26 cells. The arrow points to a polyclonal villus. Note that each column is composed of either blue or white cells but not a mixture of both, and that the borders between blue and white cellular columns are clearly demarcated. (D) Whole mount preparation of colon. The orifice of each colonic crypt is surrounded by a surface epithelial cuff representing the villus homolog. Each cuff receives the output from a single crypt. (E) X-Gal and Nuclear Fast Red stained section from the preparation shown in panel D. As in the small intestine, each adult colonic crypt is monoclonal. Bars in B and E = 25 µm.
[15]
LCM OF MOUSE INTESTINE
173
of 129/Sv epithelium (white after X-Gal staining) positioned next to coherent vertical columns of B6-ROSA26 cells (blue after X-Gal staining; Figs. 2A–2C). The borders between cellular columns are very distinct, reflecting the remarkable orderliness of epithelial cell migration from the crypt to the villus tip (Fig. 2C). If 129/Sv ES cells are genetically manipulated prior to injection into B6-ROSA26 blastocysts, the effects of the genetic manipulation can be defined in a single villus by comparing epithelial cells harvested by LCM from adjacent 129/Sv and B6 columns. Normal B6-ROSA26↔129/Sv mouse chimeras generated using nonmanipulated ES cells and B6-ROSA26 blastocysts can be used as external controls to identify phenotypic differences ascribable to the C57Bl/6 vs 129/Sv strain backgrounds, rather than to an engineered genetic manipulation. In the colon, the phenotypic comparison has to be made between cells harvested from adjacent 129/Sv and B6-ROSA-26 crypts, since the villus homolog (surface epithelial cuff) only receives the cellular output of a single crypt (Figs. 2D, 2E). These types of chimeric mouse models increase the likelihood of identifying subtle phenotypes produced from mutant alleles because there is a reference control population of normal cells present in the same microenvironment as the experimental cohort. Importantly, they allow molecular interactions to be defined at the interface between normal and genetically manipulated cellular cohorts, thereby allowing exploration of the pathogenesis of diseases having focal origins. Genetic mosaic models can also be produced using other strategies. For example, other markers such as green fluorescent protein (or variants) can be used to genotype cells prior to LCM.26,27 We have taken advantage of transgenic mice that express Cre recombinase in some but not all of their crypts to create genetic mosaic models. Remarkably, Cre expression in these mice is either totally off or fully on in each crypt.9 This type of Cre-based genetic mosaic mouse model eliminates the potentially confounding problem of strain background effects encountered in aggregation or injection chimeras, since all cells in the intestine will have an identical genotype, other than at the Cre-recombined floxed allele. Preparing Mouse Intestine for LCM General Comments One key to successful LCM of the mouse intestine is careful attention to the low-tech methods employed for handling and preparing tissue. Dissection must be swift and gentle, and the harvested material must be quickly frozen in an unfixed state to preserve intact RNA and protein. Individual cryo-sections are subsequently fixed by immersion in 70% ethanol. We have found that recently developed 26 27
W. C. Kisseberth, N. T. Brettingen, J. K. Lohse, and E. P. Sandgren, Dev. Biol. 214, 128 (1999). S. Srinivas, T. Watanabe, C. S. Lin, C. M. William, Y. Tanabe, T. M. Jessell, and F. Costantini, BMC Dev. Biol. 1, 4 (2001).
174
SPECIALIZED USES
[15]
methods for recovering RNA from some fixed, paraffin-embedded, extraintestinal tissues (e.g., Ref. 28) do not yield intact mRNA (or protein) from the mouse gut. Typically, we produce frozen blocks of tissue within 3 min of the start of the dissection. Because of the length of the adult mouse gut (the small intestine is 35–45 cm long; the cecum and colon have an aggregate length of 12–15 cm), we do not recommend processing the entire gut from the same mouse, unless multiple people are involved. When working alone, the best results are obtained if an 8- to 10-cm-long specimen is recovered from a specified region of the intestine. Since the gut exhibits such profound regional differences in function and cellular differentiation, when multiple animals are being compared it is very important that the targeted segment be harvested from the same relative position along the cephalocaudal axis of their intestines. Harvesting and Freezing Intestine 1. Open the abdomen with a pair of scissors and cut the junction of the stomach and proximal small intestine. Grab the proximal end of the small intestine with serrated tissue forceps and gently pull upward. Simultaneously, with the opposite hand, dissect the mesentery attached to the proximal 2–4 cm of the small intestine using blunt tip 10.5-cm-long scissors (Miltex). This dissection will disrupt the biliary and pancreatic duct connections. To separate the remainder of the small intestine from its mesentery, simply continue to gently pull the proximal end upward (over a distance of ∼35–50 cm). Cut the junction of the small intestine and cecum and carefully place the harvested intact small intestine on a polyethylene dissecting board with imprinted ruler (Fisher). Do not stretch the intestine to make the measurement. Excise an 8- to 10-cm-long specimen with a fresh razor blade. To recover the colon, the junction of the distal small intestine and the cecum is cut as above, and the mesentery and associated soft tissue freed with blunt-tipped scissors. Identify the ano-rectal junction by opening the pelvic bone with a pair of heavy scissors: remove the bladder and reproductive organs, and dissect the distal rectum away from surrounding skin. The intact colon is then removed from the mouse and measured as above. 2. Flush the harvested 8- to 10-cm specimen of small intestine, or the intact colon, with ice-cold phosphate buffered saline, pH 7.4 (PBS), using a 12-ml syringe containing an attached 1.5-inch-long, 18-gauge blunt needle. Carefully hold the proximal end of intestine with forceps during the flush. 3. Flush with Tissue-Tek OCT (VWR) from the proximal end of the specimen. Use a 60-ml syringe fitted with a 200-µl pipette tip clipped at its distal end so that it can fit into the opening at the proximal end of the intestinal segment. Gently distend 28
M. Shiutani, C. Uneyama, K. Miyazaki, K. Toyoda, and M. Hirose, Lab. Invest. 80, 199 (2000).
[15]
LCM OF MOUSE INTESTINE
175
the intestine with OCT. Continue to perfuse for the next 5–10 sec to remove all of the PBS used in the initial flush. If this step is performed incorrectly, cellular preservation will be poor. (Note: The intestine will not expand properly if the 18-gauge needle employed for the PBS flush is used.) 4. Excise and discard the very proximal portion of the intestine that had been grasped by forceps. Use a razor blade and the cutting board to divide the remaining OCT-perfused specimen into three segments (each segment is typically 2–3 cm long). 5. Place the three adjacent segments lengthwise in the base of a Tissue-Tek cryomold (VWR) and carefully overlay with additional OCT. It is important that the three intestinal segments remain flat in the bottom of the cryomold to maximize the area of tissue that can be sectioned. If the OCT is not gently overlaid, the segments will begin to float. 6. Freeze the tissue as follows [Note: Perform steps (a–c) just before the dissection is started.]: (a) Place a 50-ml side arm flask and the bottom portion of a 10-cm glass petri dish in a styrofoam cooler filled with crushed dry ice. (b) Liquify Stephens Scientific Cytocool II (VWR) by spraying directly into the bottom of the side arm flask. While spraying the Cytocool, hold the side arm with an insulated glove and point the opening of the flask away from you. (c) Pour ∼20 ml of liquid Cytocool into the petri dish. (d) Submerge the cryomold containing the intestinal segments and overlaid OCT completely under liquid Cytocool, until bubbling ceases. To ensure optimal sections, make sure the base of the cryomold remains flat by compressing it against the bottom of the petri dish during the freezing process until the material is completely frozen (usually 5–10 sec). For compression, we use a 3.5-cm-diameter aluminum canister (e.g., the type employed for packaging restriction enzymes from Roche) filled with dry ice. If the compression is not performed, the base of the cryomold tends to bow outward, resulting in uneven sectioning of the three intestinal segments. The sample can be left in the cooler surrounded by dry ice until all blocks from all animals are processed. 7. Store individual frozen blocks, wrapped in aluminum foil, at −80◦ until sectioning begins. Cutting and Staining Frozen Sections for LCM Cutting Frozen Sections for LCM 1. Equilibrate the temperature of the tissue block in a cryostat (set at −15◦ ) for 20–30 min. When blocks are too cold, they will not cut evenly. 2. Mount a temperature-equilibrated tissue block on a specimen disc with OCT. Orient the mounted block so that its three intestinal segments can be cut along their cephalocaudal axis. Cutting sections perpendicular to this axis will cause tissue to be chipped out of the block in large chunks.
176
SPECIALIZED USES
[15]
3. The optimal thickness of sections for LCM is 7–8 µm. Thicker sections are more difficult to visualize and dissect. To obtain sections with well-oriented crypt-villus units, you need to cut into the block to the middle of the segment. 4. When cutting sections, take care that they are flat, free of folds, and have consistent thickness. A major source of failure during LCM stems from cutting thick and thin sections: i.e., alternating sections that are thicker followed by sections that are much thinner than the set thickness. This problem usually occurs when one or more of the moveable levers that control the position of the knife holder and tissue chuck are not tight, or when the knife holder, tissue, or chuck has not been equilibrated to the temperature of the cryostat. 5. Minimize degradation of RNA and protein from exogenous sources. We use a cryostat where gloves are employed at all times to handle all cutting surfaces and implements. We also use disposable blades. All surfaces of the cryostat are routinely cleaned with 70% ethanol. 6. Minimize degradation of RNA and protein from endogenous sources. Sections mounted onto slides should not sit at room temperature. Sections should either be fixed and stained or placed in a slide holder on crushed dry ice. Sections prepared in this manner can be stored at −80◦ , wrapped in foil, for up to 1 week. 7. Glass slides are a very important reagent. The tissue must stick to the slide during staining but must allow LCM. We have found that Superfrost/Plus coated slides from Fisher work well, though there is some lot-to-lot variation. Staining Frozen Sections of Intestine for LCM 1. For slides stored at −80◦ , thaw one slide at a time for 1 min at room temperature. Otherwise fix and stain slides directly after cutting a single section. 2. Fix in 70% ethanol for 15 sec at room temperature.29 After fixation, dip the slide 10 times in nuclease-free water (Ambion). This is a critical step. Any OCT remaining on the slide after staining will inhibit laser capture. 3. Stain the section for 5–10 sec. If RNA is to be retrieved, hematoxylin (Vector Laboratories), eosin Y (Richard Allan Scientific), methyl green (Vector Laboratories), or Nuclear Fast Red (NFR, Vector Laboratories) are appropriate stains. If protein is to be retrieved, avoid eosin since it inhibits recovery. 4. After staining is completed, dehydrate the section with two successive washes in 95% ethanol, two washes in 100% ethanol (AAPER Alcohol and Chemical Co.), and three washes in histology grade xylene (Fisher). Complete dehydration is very important for maintaining the stability of cellular RNA and protein. To minimize exogenous RNases and proteases, and to ensure proper dehydration, use fresh 50-ml Falcon tubes for staining and dehydration, plus a set of dedicated glass Coplin staining jars (Wheaton Scientific) for xylenes. Store fixed 29
S. M. Goldsworthy, P. S. Stockton, C. S. Trempus, J. F. Foley, and R. R. Maronpot, Mol. Carcinog. 25, 86 (1999).
[15]
LCM OF MOUSE INTESTINE
177
and stained slides in a slide holder in an airtight container with DriRite. Sections should be used promptly, especially for protein extraction. Summary of Staining Protocol Method 1 for RNA retrieval. (All solutions are room temperature.) 1. Fix in 70% ethanol 15 sec 2. Nuclease free water Dip 10× 3. Methyl green, hematoxylin, or NFR Dip 1× (10 sec) 4. Nuclease free water Dip 10× 5. 70% Ethanol Dip 10× 6. 95% Ethanol Dip 10× 7. Eosin Y (if cytoplasmic features need to be well visualized) Dip once (2–3 sec) 8. 95% Ethanol Dip 10× 9. 95% Ethanol 1 min 10. 100% Ethanol 1 min 11. 100% Ethanol 1 min 12. Xylenes ×3 5 min total 13. Air dry 2 min Method 2 for protein retrieval 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12.
Fix in 70% ethanol Nuclease free water Nuclear Fast Red Nuclease free water 70% Ethanol 95% Ethanol 95% Ethanol 95% Ethanol 100% Ethanol 100% Ethanol Xylenes ×3 Air dry
15 sec Dip 10× 5 sec Dip 10× Dip 10× Dip 10× Dip 10× 1 min 1 min 1 min 5 min total 2 min
There are reports that cells in some tissues can be marked for LCM using rapid immunostaining procedures and still yield intact RNA.30,31 Unfortunately, we have found that just 1 min in an aqueous solution degrades ∼99% of the RNA in intestinal epithelial cells.9 An alternative to direct immunostaining is to prepare two adjacent 30
F. Fend, M. R. Emmert-Buck, R. Chuaqui, K. Cole, J. Lee, L. A. Liotta, and M. Raffeld, Am. J. Pathol. 154, 61 (1999). 31 L. Jin, C. A. Thompson, X. Qiun, S. J. Kuecker, E. Kulig, and R. V. Lloyd, Lab. Invest. 79, 511 (1999).
178
SPECIALIZED USES
[15]
cryosections. One section is subjected to immunohistochemical staining, and then employed as an image template (guide) to direct dissection of the adjacent slide.9 Laser Capture Microdissection Before beginning LCM, stained slides from a given block need to be examined by light microscopy to verify that cellular preservation is satisfactory. Epithelial cells on villus tips are very sensitive to oxygen deprivation and therefore are good reporters of tissue integrity. Cells that are shrunken, vacuolated, or lost indicate poor preservation (Figs. 3A, 3B). A more subtle indication of damage is separation of epithelial cells from the underlying mesenchyme. Poor quality sections are typically the result of slow or forceful tissue handling during preparation of the frozen block. Occasionally, poor quality fixation and/or staining reagents can affect morphology. Sections maintained at −80◦ that have begun to dehydrate from prolonged or improper storage will also contain epithelial cells with aberrant morphology. Intestinal sections with well-oriented crypt-villus units should be used for dissection: i.e., sections prepared perpendicular and not tangential to the cryptvillus axis. The discrete nature of crypt-villus units and colonic crypts makes it relatively easy to recover specific epithelial populations (Figs. 3C–3G). In laser capture microdissection, targeted cells are transferred to an ethylene vinyl acetate transfer film attached to an optical quality cap placed over the section. One problem with the initial generation of commercially available LCM caps was that their entire surface contacted the tissue section. This presented a problem with the intestine since lymphocytes and the outer muscularis layer are highly susceptible to nonspecific sticking to these caps. A new generation of caps has been developed that contain four 12-µm-high rails to minimize nonspecific contact sticking (Arcturus, CapSure HS LCM Caps). These rail-containing caps should always be used for LCM of the intestine. One aspect of LCM often overlooked is the environment of the room housing the instrument. Both the quality of the dissection and the stability of cell-associated macromolecules depend on maintaining sections in a dehydrated state. For this reason, we perform LCM in a constant temperature room (20◦ ) with low humidity (<50%). In addition, we surround the dissection bench with lightweight fiberglass curtains that trap dust and eliminate static electricity. The procedure for LCM using the Arcturus PixCell II system is clearly outlined in the instruction manuals that accompany the instrument. The major problem that most new investigators encounter is lack of capture. Troubleshooting tips for overcoming this problem include the following. Make sure the laser is properly focused. An out-of-focus laser cannot deliver sufficient power to wet the transfer film. Verify the proper focus by test firing the laser in a blank area of the slide. You should see a dark black halo around an area of melted transfer film. If the cap
[15]
LCM OF MOUSE INTESTINE
179
FIG. 3. LCM of targeted cell populations in the small intestine. A segment from the middle portion of the small intestine was perfused with OCT, frozen, and sectioned. Sections were fixed in 70% ethanol and stained with eosin Y and methyl green. (A) High power view of a villus tip showing good cellular preservation. (B) Example of poor cellular preservation. (C–G) Examples of LCM of selected cell populations. Panel C, crypt-villus unit prior to LCM. Panel D, postdissection photo after crypt epithelium has been removed. Panel E, view of cap containing dissected villus epithelium. Panel F, cap containing dissected villus core mesenchyme. Panel G, cap with recovered crypt epithelium. Bars = 25 µm.
180
SPECIALIZED USES
[15]
does not wet properly when fired, increase the power and/or duration of the laser pulse. Finally, check the cap for damage. The cap will not function properly if it contains dust, scratches, or fingerprints. Analysis of mRNAs Recovered from LCM Cell Populations RNA Isolation On average, 2–5 pg of total RNA can be recovered from a captured, 7-µmthick intestinal epithelial cell. For example, we typically obtain 10 ng of total RNA from 5000 recovered cells (20–30 min of dissection time). Until recently, we used the protocol for RNA isolation described on the NIH web site (dir.nichd.nih.gov/ lcm/lcm.htm). This protocol has been incorporated into a commercially available kit (Stratagene). The method involves guanidine isothiocyanate denaturation, phenol–chloroform extraction, and ethanol precipitation. However, this protocol is quite time consuming. Silica-based columns have been adapted for isolation of nanogram quantities of RNA. One of these systems, Arcturus Genomics PicoPure RNA Isolation Kit (www.arctur.com), permits rapid isolation (15 min as opposed to several hours) and allows an optional DNase step to be performed on the column. This oncolumn DNase treatment can be performed after binding RNA to the column and completing the first wash step in the PicoPure kit protocol. The DNase treatment consists of (a) adding 10 µl of a 10 U/µl stock of RNase-free DNase (Qiagen) to 70 µl of Qiagen’s RDD buffer supplemented with 10 mM MgCl2, 1 mM DTT); (b) incubating the column in the presence of the DNase solution for 15 min at 25◦ ; (c) centrifuging the column for 1 min at 8000g (microfuge); (d) adding 50 µl of PicoPure kit Wash Buffer 1; and (e) centrifuging the column for 1 min at 8000g. The treated RNA is eluted from the column in a total volume of 11 µl, per the kit protocol. The DNase step is usually performed when cells are harvested from intestinal neoplasms, from areas with dense inflammatory cell infiltrates, or from submucosal lymphoid aggregates that form part of the gut-associated lymphoid tissue (GALT). The amount of total cellular RNA recovered is determined using the RiboGreen RNA quantitation kit (Molecular Probes). Real-Time Quantitative RT-PCR Because of the limited amounts of RNA obtained from LCM procured material, quantitating levels of specific mRNAs is best done using real-time quantitative reverse-transcriptase polymerase chain reaction (qRT-PCR). The following method requires small amounts of total RNA (400 pg per reaction, representing ∼200 captured cell equivalents). Principles of SYBR Green Based Real-Time Quantitative RT-PCR. In qRTPCR, the amount of amplicon (PCR product) generated after each cycle of PCR is
[15]
LCM OF MOUSE INTESTINE
181
monitored in real time. Increasing amounts of double-stranded DNA product are reflected by increasing fluorescence emitted from a dye that binds to the DNA, or from a fluorogenic probe that anneals to the middle of the amplicon (see below). A threshold cycle (CT) for each amplicon is defined as the PCR cycle at which the fluorescence intensity crosses a user-established threshold (Fig. 3A). CT is inversely proportional to the log of the copy number of the mRNA.32 There are two basic approaches for performing qRT-PCR. One employs a labeled fluorogenic probe (TaqMan probe) that anneals to sequences within the amplicon. The TaqMan probe has a 5 -conjugated fluorescent tag that serves as the reporter and a 3 -conjugated tag that quenches reporter fluorescence. The intact probe does not fluoresce. Amplicon is detected when the 5 nuclease activity of Taq polymerase cleaves the TaqMan probe during the extension phase of PCR, releasing the reporter dye and resulting in a fluorescence emission (monitored in real time). A second approach is based on the use of SYBR Green I (Molecular Probes) as a fluorescence label when bound to double-stranded DNA. Applied Biosystems (home.appliedbiosystems.com) has developed conditions compatible with their 5700 and 7700 Sequence Detection Systems that allow qRT-PCR to be performed with SYBR Green I. The fluorogenic TaqMan probe allows detection of multiple transcripts within a single sample and is highly specific for a given sequence. Nonetheless, SYBR Green offers a number of advantages. The cost is lower. The method is more versatile since the dye can be used to detect amplicon from any transcript. Binding of multiple SYBR Green molecules to each amplicon increases the sensitivity of detection. The principal drawback to the SYBR Green method is its lack of specificity, since any contaminating double-stranded DNA species (such as primer– dimer) will contribute to the fluorescence signal. However, this can be overcome by establishing a melting temperature for the amplicon that exceeds that of any primer–dimer complexes (see below). Fluorescence emission is then measured at this melting temperature after the extension phase of each PCR cycle. cDNA Synthesis. As with any RT-PCR application, the first step in qRT-PCR is cDNA synthesis. If you begin with 10 ng of total RNA, purified from ∼5000 captured cells, cDNA synthesis will produce enough product for assaying 8 different mRNA species in triplicate. Protocol 1. Add 1.0 µl oligo(dT) primers (0.5 µg) or 1.0 µl random hexanucleotide primers (3 µg) to 10 ng purified RNA (in 11 µl of elution buffer from the PicoPure column protocol). Incubate the mixture at 70◦ for 5 min, then place on ice for another 5 min. 32
C. A. Heid, J. Stevens, K. J. Livak, and P. M. Williams, Genome Res. 6, 986 (1996).
182
SPECIALIZED USES
[15]
2. Add the following reagents to the mixture in (1): (a) 1 µl Sensiscript reverse transcriptase (Qiagen); (b) 2 µl 10× Sensiscript reverse transcriptase buffer; (c) 2 µl dNTPs (5 mM stock); (d) 0.2 µl of RNase Inhibitor (Roche); and (e) nuclease-free water (Ambion) so that the final volume of the reaction is 20 µl. 3. Incubate at 37◦ for 1 hr. The reaction mixture can be stored at −20◦ until further use. Primer Design. Selecting a primer pair that generates a single, specific amplicon is a critical step. Primers that comply with the following requirements yield the best results: Tm = 58–60◦ ; GC content = 20–80%; length ∼20 nucleotides; and no more than two G or C residues among the five bases at the 3 end. In addition, the amplicon should be 50–150 bp to minimize PCR cycle time, should have a melting temperature of 78–84◦ , and should ideally span an intron–exon junction to allow differentiation between the genomic DNA and cDNA product. Prior to qRT-PCR, all primer pairs are tested by conventional PCR using cDNA prepared from RNA isolated from the intact intestine (or intestinal segments), genomic DNA, and a water blank to establish that a single amplicon is produced in the cDNA sample and that minimal primer–dimer bands are produced in the absence of template. To ensure that the fluorescence measurement after each cycle is only due to the specific amplicon, a melting temperature must be identified that results in a maximal difference between amount of amplicon and primer–dimer. Therefore, qRT-PCR amplifications are performed (95◦ for 15 sec; anneal at 55◦ for 45 sec; extend at 72◦ for 30 sec for a total of 40 cycles). The product obtained after 40 cycles of PCR is then subjected to a melt curve analysis (Fig. 4). To generate this melt curve, fluorescence is measured at 0.2◦ intervals between 60◦ and 92◦ . We exploit the PE Biosystems 7700 Sequence Detection System software to collect and analyze the 160 separate incremental readings as follows. The machine is programmed to collect increments of four 0.2◦ fluorescence readings. The first four readings span the temperature range of 60.0◦ to 60.6◦ . This collection of four readings is treated as one round of data collection. The second round collects four fluorescence measurements at 0.2◦ increments between 60.8◦ and 61.4◦ . This iterative process is continued for a total of 40 rounds (160 data points) in order to span 60◦ to 92◦ . We then plot the fluorescence measurement from the first or third data point from each round to generate a melt curve such as that shown in Fig. 4. Comparison of the amplicon melt curve, generated with cDNA as template, vs the primer–dimer melt curve, generated with no template, provides a temperature range at which the majority of the fluorescence signal is derived from the amplicon. Note that when plotting this data, cycle number is converted to temperature using the following equation: (cycle number − 1)(0.8◦ ) + 60◦ Once this temperature has been determined from the melt curve analysis, subsequent experimental qRT-PCR assays are performed by programming the Sequence
⌬Rn
A Amplification
Cycle
Rn
B Melt Curve
Cycle FIG. 4. SYBR Green based quantitative real time RT-PCR. (A) Triplicate qRT-PCR assays comparing the relative expression of keratin 8 mRNA in LCM populations of (1) villus epithelial cells, (2) crypt epithelial cells, and (3) villus mesenchymal cells. The amount of amplicon present after each cycle of PCR is monitored in real time. Increasing amounts of amplicon are reflected by increasing fluorescence from the SYBR Green dye that binds to double-stranded DNA. The threshold cycle (CT) is defined as the cycle at which fluorescence intensity crosses a user-established threshold. CT is inversely proportional to the log of the starting mRNA copy number. In the example shown, the amount of keratin 8 mRNA is greatest in villus epithelial cells (population 1) and least abundant in population 3 (villus mesenchymal cells). (B) Melt curve for keratin 8 mRNA. Contaminating primer–dimer will contribute to the fluorescence signal in SYBR Green based qRT-PCR. Therefore, an amplicon melting temperature exceeding that of any primer–dimer complexes must be established. See text for detailed descriptions of the assay and its interpretation.
184
SPECIALIZED USES
[15]
Detection System to ramp to the melt temperature following the extension phase of each cycle so that fluorescence data can be collected. Assembling qRT-PCR Reaction Mixture. The goal is to perform each assay using cDNA representing ∼200 cells (for low abundance mRNA transcripts, more cell equivalents may be required). Therefore, the 20 µl cDNA synthesis reaction mixture described above (representing ∼5000 cells) is diluted with nuclease-free water to a final volume of 50 µl (i.e., 100 cell equivalents/µl). Two µl of this diluted mixture is then added to a solution containing the following components (final volume after all additions = 25 µl): (a) 12.5 µl 2× SYBR Green Mix (PE Biosystems). This mixture contains 8% glycerol, 2.5 mM MgCl2, 2× PCR buffer (40 mM Tris-HCl, pH 8.4, 100 mM KCl), 200 µM dATP, 200 µM dGTP, 200 µM dCTP, 400 µM dUDP, 0.32× SYBR Green I dye in DMSO (Molecular Probes), 225 nM ROX standard I (6-mer with 5 ROX dye conjugated and 3 phosphorylated, Synthegen, Houston, TX), and 0.25 U Platinum Taq Polymerase (Invitrogen). The 2× SYBR Green stock can be stored at 4◦ for at least 2 weeks. (b) 300 nM of each primer (c) 0.25 U uracil DNA glycosylase (UDG; Invitrogen). Adding glycosylase eliminates problems with amplicon contamination from previous runs, since it recognizes and cleaves double-stranded nucleic acid sequences containing dUDP. If UDG is included, the PCR reaction mixture must be incubated for 2 min at 50◦ prior to initiating thermocycling. Quantitation of Gene Expression. Comparison of mRNA levels between samples can be done using either the standard curve method or the comparative CT method. Each of these methods is detailed in the Applied Biosystems ABI Prism 7700 Sequence Detection System User Bulletin #2 (available in PDF format from home.appliedbiosystems.com). In both methods, an endogenous reference control transcript is used to correct for variations in the efficiency of cDNA synthesis between samples. We routinely use glyceraldehyde-3-phosphate dehydrogenase (Gapdh) mRNA (forward primer = 5 TGGCAAAGTGGAGATTGTTGCC, reverse primer = 5 AAGATGGTGATGGGCTTCCCG), or 18S rRNA (forward primer = 5 CATTCGAACGTCTGCCCTATC, reverse primer = 5 CCTGCTGC CTTCCTTGGA) as reference controls. Using 18S rRNA requires cDNA synthesis to be primed with random hexamers. In the standard curve method, amplicon quantity is determined by comparison to a standard curve generated using serial dilutions of a stock solution of any cDNA preparation that contains a high concentration of the sequence of interest. (Note: A stock solution of recombinant plasmid DNA containing the sequence can also be used, but there are caveats to this approach if you want to calculate the absolute amount of the sequence; see below.) A separate standard curve is prepared for both
[15]
LCM OF MOUSE INTESTINE
185
the target sequence and reference control (Gapdh or 18S rRNA). Within a sample, the target value is divided by the reference value to obtain a normalized value. To compare the relative levels of target mRNA in two or more samples, one sample is designated the calibrator and the normalized target values in the experimental samples are divided by the normalized calibrator value. Thus, quantities in the experimental samples are expressed as n-fold differences relative to the calibrator. It is also possible to use the standard curve method to quantify absolute amounts of a transcript in a sample, but this requires that the absolute quantities of the standard be established by an independent method (such as A260). Plasmid DNA cannot be used as a standard to quantify absolute transcript levels from cDNA since there is no control for the efficiency of in vitro transcription. In the comparative CT method, a mathematical formula is used to determine a fold change in normalized target levels between an experimental and calibrator sample. A standard curve is not necessary, provided the target and reference control sequences are amplified with equal efficiency (this is established in a separate validation experiment). We favor this approach since it permits direct comparison of multiple samples and allows more samples to be assayed in a single plate due to elimination of the standard curve. First, the difference in CT between target and reference sequences in each sample is calculated (CT). Second, the difference between the CT of the experimental and that of the calibrator sample is determined (CT). Because the fluorescence data collected from the qRT-PCR assay is plotted in a log-linear fashion, CT can be converted to a fold difference with the formula 2−CT (see Applied Biosystems ABI Prism 7700 Sequence Detection System User Bulletin #2 for further details). As controls for the fidelity of the microdissection, genes known to be expressed in a particular cellular compartment or cell type are also quantified. For example keratin 8 (forward primer = 5 GCTGAAGTTCGTGCCCAGTAC, reverse primer = 5 CTTTGTGCGGCGCAGAT) is expressed in epithelial cells, vimentin (forward primer = 5 TGCTTCTCTGGCACGTCTTG, reverse primer = 5 GGACATGCTGTTCCTGAATCTG) is expressed in mesenchymal cells, and phospholipase A2 (forward primer = 5 CCAAATCACCTGTTCTGCAAAC, reverse primer = 5 CATTCAGCGGCGGCTTTA) is expressed in Paneth cells. DNA Microarrays RNA prepared from a population of laser capture microdissected cells can be used for functional genomics studies with commercially available high density, oligonucleotide-based microarrays (GeneChips, www.affymetrix.com). Each GeneChip contains up to 12,000 different genes, each gene represented by at least one set of ∼20 different “probe pairs.” A probe pair consists of a 25-bp oligonucleotide perfect match “probe” and a 25-bp mismatch probe where the thirteenth position does not match the target sequence. (Note the nomenclature:
186
SPECIALIZED USES
[15]
the oligonucleotide probe sequences are on the chip while the target is the cRNA generated from cellular mRNA.) Information across all 20 paired probes (the probe set) is integrated by proprietary GeneChip software. The software compares mRNA levels in two RNA preparations by analyzing probe-set signals from two GeneChips: one hybridized with cRNA made from the first RNA preparation, the other hybridized with cRNA generated from the second RNA preparation. From a practical standpoint, the maximum number of cells that can be dissected for an experiment is generally of the order of tens of thousands (corresponding to 10–100 ng of total RNA). GeneChip hybridizations require microgram quantities of labeled cRNA target. Methods have been developed for amplification of cRNA from the small quantities of mRNA isolated from laser captured cells (Fig. 5). The key to obtaining linear amplification is to optimize the amount of the oligo(dT)-T7 primer (5 -GGCCAGTGAATTGTAATACGACTCACTATAGGGA GGCGG(T)24 -3 ) used for cDNA synthesis. If excess amounts of primer are used, the primer is carried over to the in vitro transcription reaction (step 1, Fig. 5). This produces deleterious side reactions that affect the fidelity of amplification.33 To verify that amplification is linear, internal controls are included in the reaction mixture prior to initiation of first strand cDNA synthesis. A commonly used internal reference control consists of a mixture of Bacillus subtilis gene transcripts detectable by probe sets present on Affymetrix GeneChips. Each reference bacterial RNA is added in different amounts, so that the efficiency of amplification can be monitored over a range of starting transcript concentrations: i.e., 1000 copies of trp, 5000 copies of thr, 20,000 copies of phe, and 100,0000 copies of lys RNA are added per 10 ng of total (intestinal) RNA. The protocol for amplification of cRNA targets described below is a modification of a procedure developed by Luo et al.34 All reagents are from the Superscript kit (Invitrogen) unless otherwise indicated. Amplification Round 1. (See Fig. 5.) Step 1: First strand cDNA synthesis (a) Incubate purified total RNA (minimum 10 ng) in 10 µl of elution buffer from the PicoPure column with 1 µl of oligo(dT)-T7 primer (10 ng/µl) at 70◦ for 10 min. (b) Place mixture on ice and add 4 µl of 5× first strand synthesis buffer (from kit), 2 µl 0.1 M DTT, 1 µl 10 mM dNTPs. (c) Incubate at 42◦ for 2 min. (d) Add 2 µl Superscript II reverse transcriptase (200 U/µl); incubate at 42◦ for 1 hr. 33 34
L. R. Baugh, A. A. Hill, E. L. Brown, and C. P. Hunter, Nucleic Acids Res. 29, e29 (2001). L. Luo, R. C. Salunga, H. Guo, A. Bittner, K. C. Joy, J. E. Galindo, H. Xiao, K. E. Rogers, J. S. Wan, M. R. Jackson, and M. G. Erlander, Nat. Med. 5, 117 (1999).
[15]
LCM OF MOUSE INTESTINE
FIG. 5. Amplification rounds 1 and 2.
187
188
SPECIALIZED USES
[15]
Step 2: Second strand synthesis (a) To first strand synthesis reaction add 91 µl nuclease-free H2O, 30 µl 5× second strand buffer (from kit), 3 µl 10 mM dNTPs, 1 µl of a 10U/µl stock of E. coli DNA ligase, 4 µl of a 10 U/µl stock of E. coli DNA polymerase, 1 µl of a 2 U/µl stock of E. coli RNase H. (b) Incubate at 16◦ for 2 hr. (c) Add 2 µl of 5 U/µl T4 DNA polymerase. (d) Incubate at 16◦ for 15 min. (e) Stop reaction by adding 10 µl 0.5 M EDTA, pH 8.0. Step 3: Cleanup and precipitation of double stranded cDNA (a) Pellet phase lock gel (Eppendorf) in microfuge (12,000g for 30 sec). (b) Add 162 µl phenol/chloroform/isoamyl alcohol (25 : 24 : 1; Ambion) to the in vitro transcription mixture and vortex. (c) Transfer the entire contents of the tube (organic and aqueous phases) to the tube containing the pelleted phase lock gel (do not vortex). Spin for 2 min at 12,000g. (d) Transfer aqueous phase to fresh 1.5-ml Eppendorf tube. (e) Ethanol precipitate double-stranded cDNA by adding 1 µl Glycoblue (Ambion), 0.5 volume of 7.5 M ammonium acetate, and 2.5 volumes 100% ethanol. Vortex and centrifuge for 20 min at room temp (16,000g). Wash resulting pellet twice in 80% ethanol, dry pellet in a Speed Vac, and resuspend in 8 µl H2O. Step 4: T7-directed in vitro transcription using double-stranded cDNA template. This step is performed using reagents provided in the AmpliScribe T7 kit (Epicentre Technologies, www.epicentre.com). All 8 µl of the dsDNA from step 3 is mixed with 2 µl of 10× reaction buffer, 1.5 µl of each ribonucleotide (A,C,G,U), 2 µl 10× DTT, 2 µl T7 RNA polymerase in a final reaction volume of 20 µl. The reaction mixture is incubated at 42◦ for 3 hr (tap tube every hour). Step 5: Cleanup of in vitro transcripts. This step uses reagents in the RNeasy kit from Qiagen. (a) Add 80 µl nuclease-free H2O to the reaction mixture in step 4. (b) Add 350 µl RLT buffer plus 3.5 µl of 2-mercaptoethanol. (c) Add 250 µl 100% ethanol, gently mix and apply all 700 µl of the solution to a RNeasy spin column. (d) Centrifuge 15 sec at 12,000g. (e) Wash column twice with 500 µl RPE buffer. (f) Elute cRNA with 40 µl distilled H2O. Let eluant stand for 1 min at room temperature, centrifuge for 1 min at 12,000g, dry down to 10 µl in Speed Vac.
[15]
LCM OF MOUSE INTESTINE
189
Amplification Round 2. (See Fig. 5.) Step 1: First strand DNA synthesis (a) Mix 10 µl of cRNA from step 5 of round 1 with 2 µl of a 50 ng/µl solution of random hexamers (Superscript kit, Invitrogen). Incubate at 70◦ for 10 min and place on ice. (b) Add 4 µl 5× first strand buffer (Superscript kit), 2 µl 0.1 M DTT, 1 µl 10 mM dNTPs, 1 µl RNase inhibitor (Roche), and 1 µl of Superscript II reverse transcriptase. Incubate at 42◦ for 1 hr. Step 2: Second strand DNA synthesis (a) Add 3 µl oligo(dT)-T7primer (10 ng/µl); incubate at 70◦ for 2 min, followed by 42◦ for 2 min. (b) Add 87 µl nuclease-free H2O, 30 µl 5× second strand buffer (Superscript kit), 3 µl of 10 mM dNTPs, 4 µl of 10 U/µl E. coli DNA polymerase, 1 µl of 10 U/µl E. coli DNA ligase, and 1 µl of 2 U/µl E. coli RNase H. (c) Incubate at 16◦ for 2 hr. (d) Add 2 µl 5 U/µl T4 DNA polymerase and incubate at 16◦ for 15 min. Terminate reaction with 10 µl 0.5 M EDTA, pH 8.0. Step 3: Clean up and precipitate ds cDNA as described in the first round amplification. Step 4: Production of biotinylated cRNA targets. This step uses reagents from the BioArray RNA transcription labeling kit manufactured by Enzo Diagnostics (www.enzo.com). (a) Mix 22 µl dsDNA with 4 µl 10× HY reaction buffer, 4 µl 10× biotinlabeled ribonucleotides, 4 µl 10× DTT, 4 µl 10× RNase inhibitor, and 2 µl 20× T7 RNA polymerase. (b) Adjust final volume to 40 µl with nuclease-free water and incubate at 37◦ for 5 hr (tapping tube every hour). Step 5: Perform cleanup of cRNA with RNeasy columns as described in first round amplification. Notes: For each RNA preparation, we generate two amplified cRNA targets in separate reaction mixtures. Each of these independently amplified cRNA targets is then hybridized to a GeneChip. Thus, if two RNA preparations are being compared (one from a control population of cells, and the other from an experiment population), a total of 4 cRNA targets are used for duplicate GeneChip hybridizations (cRNA1control, cRNA2control are each hybridized to separate GeneChips, each of which will be designated the baseline chip and compared to the two experimental chips hybridized with cRNA1experimental and cRNA2experimental. GeneChip software is used to compare the signal intensity of a given probe set in experimental chip 1
190
SPECIALIZED USES
[15]
to the intensity of the corresponding probe set in baseline chip 1. Similarly, the signal intensity of that probe set in experimental chip 2 will be compared to the corresponding probe set in baseline chip 2. Only those probe sets that show differences of ≥2-fold in the same direction (increased or decreased) in the duplicate comparisons are culled and put into a dataset. However, using an arbitrary threshold for fold change (e.g., defining an increase or decrease of ≥2-fold as significant) means that potentially important and reproducible biological changes will be masked. In addition, fold change is a ratio: probe set intensities only reflect expression differences linearly within a limited range; if either probe set has hybridization intensities outside this range (high or low), the ratio will be skewed. Because of these concerns, we have designed an alternative system for filtering false positives. The system is based on the results of our analysis of signal intensities produced by genes whose expression level was called “changed” in chip-to-chip comparisons of an identical RNA population.35 All called changes in expression in these same-same comparisons were defined as false positives. The distribution of signal intensities of the false positives in baseline and partner GeneChips was compared to the distribution of signal intensities produced when biologically distinct RNAs were analyzed. The results were used to create a series of look-up tables (LUTs). These LUTs can be used to score transcripts whose expression level is called changed by GeneChip software in comparisons of biologically distinct RNAs. LUT-derived scores range from 0 to 6, with 0 most likely and 6 least likely to represent noise. Eliminating transcripts with LUT scores of <4 reduces false positives by 90%35 and increases experiment-to-experiment reproducibility to above that obtained by imposing the arbitrary ≥2-fold change threshold. LUTs and software needed to score GeneChip datasets are available (gordonlab.wustl.edu/mills). Analyzing Proteins in LCM Cell Populations A central challenge to doing protein analysis is efficient extraction of material from LCM cells on the cap. Efficient extraction of 50,000 intestinal epithelial cells yields ∼10 µg of total protein. A comparison of proteins present in two populations of intestinal cells harvested by LCM can entail two-dimensional gel electrophoresis (2D GE) followed by immunoblotting with antibodies directed against a protein of interest. Alternatively, gels can be stained with a sensitive dye, and the relative abundance of a protein or proteins compared in the two cellular populations. Protocol for Protein Isolation and 2-Dimensional Gel Electrophoresis 1. Lyse captured cells on a single cap by adding 20 µl of a solution containing 7 M urea, 2 M thiourea, 4% CHAPS (Sigma Chemical Co.), 40 mM 35
J. Mills and J. I. Gordon, Nucleic Acids Res. 29, e13 (2001).
[15]
LCM OF MOUSE INTESTINE
191
Tris-base, 50 mM DTT, 1% decanoyl-N-methlyglucamide (Sigma), 1% n-octylβ-D-glucopyranoside (Sigma), 2 mM tributylphosphine (Bio-Rad), and 0.5% ampholytes (Bio-Rad).36 This buffer extracts total cellular proteins. The extracted proteins can be directly analyzed with 2D GE. 2. First dimension isoelectric focusing is carried out using 7-cm-long Ready Strips IPG Strips with a pH 3–10 gradient (Bio-Rad). The strips are equilibrated in two buffers (2 washes of 10 min each): (a) buffer 1 (6 M urea, 2% SDS, 0.375 M Tris-HCl, pH 8.8, 20% glycerol, 130 mM DTT) followed by (b) buffer 2 [6 M urea, 2% SDS, 0.375 M Tris-HCl, pH 8.8, 20% glycerol, 135 mM iodoacetamide (Sigma)]. 3. For second dimension separation, equilibrated strips are applied to a 4–12% NuPage precast Bis-Tris gradient gel (Invitrogen). Electrophoresis is performed in MES buffer (50 mM MES, 50 mM Tris-base, pH 7.3, 3.5 mM SDS, 1 mM EDTA) at 80 volts for 15 min followed by 150–200 volts for 1 hr. The gels can be either electrophoretically transferred to PVDF membranes (Millipore) for immunoblotting, or treated with SYPRO Ruby IEF gel stain (Molecular Probes). SYPRO Ruby can detect as little as 1 ng of protein and has a dynamic range of 1–1000 ng.37 An approach developed by Cravatt and co-workers may be particularly useful for analysis of the proteomes of laser capture microdissected cells. It is based on changes in protein activity rather than abundance and uses labeled chemical probes directed at active sites (e.g., serine hydrolases).38 They have also generated combinatorial libraries of electrophilic probes to proteomes: these probes are incubated with cellular proteins prior to electrophoretic separation. A control incubation is performed at a nonphysiological temperature so that heat-sensitive interactions can be distinguished from heat-insensitive (nonspecific) interactions. Reactive protein species exhibiting heat-sensitive interactions are then excised for mass spectrometry-based identification. The approach yields workable sized signatures of enzyme/protein activities.39 Assaying Enzymes and Metabolites Involved in Intermediary Metabolism We have developed a method for performing LCM under conditions that preserve cellular metabolites and enzyme activity for subsequent quantitative biochemical assays. This requires an embedding medium that does not interfere with 36
D. K. Ornstein, J. W. Gillespie, C. P. Paweletz, P. H. Duray, J. Herring, C. D. Vocke, S. L. Topalian, D. G. Bostwick, W. M. Linehan, E. F. Petricoin, and M. R. Emmert-Buck, Electrophoresis 21, 2235 (2000). 37 W. F. Patton, Electrophoresis 21, 1123 (2000). 38 Y. Liu, M. P. Patricelli, and B. F. Cravatt, Proc. Natl. Acad. Sci. U.S.A. 96, 14694 (1999). 39 G. C. Adam, B. F. Cravatt, and E. J. Sorensen, Chem. Biol. 8, 81 (2001).
192
SPECIALIZED USES
[15]
enzyme/metabolite measurements or with laser capture microdissection of unstained, unfixed tissue sections. In our method, OCT is replaced with a rabbit brain paste and the sectioned tissue is affixed to the glass slide by freeze-drying. Tissue Preparation and Sectioning Frozen, stripped rabbit brains (Pel-Freez) are thawed and then disrupted, without added water or buffer, using a high shear tissue homogenizer (Tekmar) set at full power for 3 min. The resulting brain paste should be smooth and free of lumps. An 8- to 10-cm-long specimen of intestine is rapidly harvested as described above, flushed with PBS, and perfused with brain paste. The perfused specimen is subdivided into three ∼2.5 cm segments which are then placed at the base of a TissueTek cryomold. The segments are overlaid with additional brain paste and the cryomold immediately submerged in liquefied Cryocool. Sections (7 µm) are cut in a cryostat cooled to −10◦ . Superfrost/Plus glass slides are precooled on dry ice before the tissue section is applied. Slides containing the tissue section are then immediately dropped into liquid nitrogen and freeze-dried at −35◦ under vacuum (0.01 mm Hg) for 2 days. Slides can be stored at −80◦ under vacuum for at least 1 month. Preparation of Cell Lysates for Enzyme Assays For descriptions of the equipment used to obtain nanoliter-sized aliquots and for other aspects of the cycling assays, see Passoneau and Lowry (1993).40 Approximately 3000 epithelial cells (50–70 ng soluble protein), or equivalent amounts of mesenchyme or muscle, are collected per cap. Lysates are prepared by adding 1.6 µl of extraction buffer to the surface of the cap [extraction buffer contains 20 mM phosphate, pH 7.4, 5 mM 2-mercaptoethanol, 25% glycerol, 0.5% Triton X-100, 1 mM Pefabloc (Roche) plus 1 tablet of “complete mini-EDTA-free protease inhibitors” (Roche) dissolved in 5 ml of the buffer]. Following a 1 min incubation at 20◦ , lysates are removed from the cap surface and transferred to an oil well (a Teflon block containing drilled wells filled with a mixture of mineral oil and hexadecane to prevent evaporation and to facilitate long-term storage). The Teflon block containing the lysates can be stored at −80◦ under vacuum indefinitely. Aliquots of the lysates are removed for assays of enzyme activities as well as determination of soluble protein. Levels of enzymes and metabolites are expressed per microgram of soluble cellular protein. Soluble protein concentration in lysates is defined using the colloidal gold method (Diversified Biotech). A 0.2-µl aliquot of lysate is added to 0.5 ml colloidal gold reagent. Following a 40
J. V. Passoneau, and O. H. Lowry, “Enzymatic Analysis: A Practical Guide.” Humana Press, Totowa, NJ, 1993.
[15]
LCM OF MOUSE INTESTINE NAD+
Malate Malic dehydrogenase
193
Ethanol Alcohol dehydrogenase
Oxaloacetate
NADH
Acetylaldehyde
SCHEME I. NAD cycle.
Glutamate
NADP+
Glutamate dehydrogenase α-Ketoglutarate + NH4+
NADPH
G-6-phosphate Glucose 6-phosphate dehydrogenase 6-Phosphogluconate
SCHEME II. NADP cycle.
60 min incubation at 37◦ , absorbance is determined at 595 nm. Typically, BSA standards are surveyed in the range of 1–10 ng. Enzyme Cycling: General Principles This sensitive, versatile, and well-established analytic method allows the levels of enzymes, metabolites, and nucleotides to be measured in laser captured cell populations through reactions that generate reduced or oxidized forms of NAD or NADP. The usefulness of using NAD and NADP for analytic purposes has been described by Passoneau and Lowry.40 The low levels of some enzyme activities in LCM cell lysates require that NAD or NADP generated in the primary analytic reaction be amplified through a series of cycling steps. Each cycling step involves one of the following coupled reactions (see Schemes I and II). After an appropriate number of cycles are performed (see Ref. 41 for details about how to determine the number of cycles), the reaction is terminated by heating at 100◦ for 5 min. Samples are cooled to room temperature, and 1 ml of indicator reagent is added to convert the cycled product (malate in the case of NAD cycling, 6-phosphogluconate in the case of the NADP cycling) to NADH or NADPH, respectively. The NAD cycling indicator step (see Scheme III) involves addition of 1 ml of malate reagent [50 mM aminomethylpropanol, pH 9.9, 5 mM L-glutamate, pH 9.9, 0.2 mM NAD+, 5 µg/ml malic dehydrogenase (3000 U/mg protein, Sigma Chemical Co.), and 2 µg/ml glutamate oxaloacetate transamidase (200 U/mg, Roche)]. The NADP cycling indicator step (see Scheme IV) is begun by adding 1 ml of 6-phosphogluconate reagent [50 mM imidazole-HCl, pH 7.0, 25 mM acetic acid, 1 mM EDTA, 30 mM ammonium acetate, 5 mM MgCl2, 0.1 mM NADP+, and 2.5 µg/ml 6-phosphogluconate dehydrogenase (20 U/mg, Sigma)]. These 41
S. S. Lin, J. K. Manchester, and J. I. Gordon, J. Biol. Chem. 276, 36000 (2001).
194
Glutamate
Malate
[15]
SPECIALIZED USES
Malate dehydrogenase
Glutamate oxaloacetate α-Ketoglutarate transamidase
Oxaloacetate
NAD +
Aspartate
NADH SCHEME III. Indicator step for NAD cycle.
6-Phosphogluconate
6-Phosphogluconate Ribulose 5-phosphate dehydrogenase
NADP+
NADPH
SCHEME IV. Indicator step for NADP cycle.
F-1,6bP
F-6P Pi
F-6P
Phosphoglucoisomerase
G-6P NADP+
G6PdH
NADPH 6-PG
SCHEME V. Scheme for fructose-1,6 bisphosphatase assay.
indicator reactions are incubated for 10 min at 25◦ . NADH generated from malate or NADPH produced from 6-phosphogluconate are measured fluorimetrically (excitation monitored at 365 nm, emission at 460 nm). Care is taken to ensure that for each enzyme, metabolite, or nucleotide determination, product formation is linear with respect to the range of cell extract used. Standards consisting of the metabolite of interest, or the product of the enzyme being assayed, are run in parallel with cell extracts. NAD+ or NADP+ standards are added at the cycling step in a minus extract control to coincide with levels produced by the experimental reactions. A known amount of malate or 6-phosphogluconate, corresponding to the predicted concentration of pyridine nucleotide produced from the cycling reaction, is included as a third reference control. See Ref. 41 for further details about calculating the amount of standards to include in these reactions. Example: Fructose-1,6-bisphosphatase An assay for measuring fructose-1,6-bisphosphatase is presented in Scheme V as an example of how pyridine nucleotide-based enzyme cycling can be applied to
[15]
LCM OF MOUSE INTESTINE
195
LCM cell populations. Lysate (0.1 µl) is added to 0.5 µl of a solution containing 50 mM imidizole-HCl (pH 7.0), 1 mM EDTA, 0.05% BSA, 2 mM MgCl2, 0.2 mM NADP+, 0.2 mM fructose 1,6-bisphosphate, 0.25 U/ml yeast phosphoglucoisomerase (specific activity 350 U/mg; Roche), and 0.18 U/ml yeast glucose-6phosphate dehydrogenase (G6PdH). Also included at this step are 0.1-µl aliquots of 5 and 10 µM fructose 6-phosphate which serve as internal standards. Following a 60 min incubation at 20◦ , 0.5 µl of 0.15 M NaOH is added, and the samples are incubated at 80◦ for 20 min. A 0.5-µl aliquot is removed and added to 0.1 ml of NADP cycling reagent [100 mM imidizole-HCl (pH 7.0), 7.5 mM α-ketoglutarate, 5 mM glucose 6-phosphate (G-6P), 25 mM ammonium acetate, 0.02% BSA, 0.1 mM ADP, 1.5 U/ml glutamate dehydrogenase (GDH, specific activity = 120 U/mg), and 1.5 U/ml G6PdH] (see Scheme II). After a 60 min incubation at 38◦ , samples are incubated for an additional 5 min at 100◦ . Once cooled to 25◦ , 1 ml of 6-phosphogluconate indicator reagent is added (see Scheme IV). The subsequent production of NADPH, corresponding to the amount of 6-PG produced in the cycling step, is determined fluorimetrically. Preparation of Extracts for Assay of Cellular Metabolites The same number of cells is needed for analysis of metabolites as for enzymes. Different metabolites have different stabilities at different pH values. Therefore, an alkali as well as an acid extract is made from a single cap of captured cells. First, 1.2 µl of ice-cold 0.05 M NaOH containing 1 mM EDTA is added to a cap containing freeze-dried cells. Two 0.6-µl aliquots are transferred to an “oil well.” The first aliquot is designated as the alkali extract. The second aliquot is added to 0.6 µl 0.1 M HCl and is designated as the acid extract. Both extracts are heated to 80◦ for 20 min. The alkali extract is neutralized by adding 0.6 µl of 100 mM Tris-HCl (pH 8.1) and 0.05 M HCl. The acid extract is neutralized by adding 0.3 µl of 0.4M Tris-base. Extracts are stored at −80◦ under vacuum until needed. The soluble protein content of the extract is determined using the colloidal gold method. A 0.2-µl aliquot of the alkali extract is added to 0.5 ml colloidal gold reagent. Following a 60 min incubation at 37◦ , absorbance is determined at 595 nm. Metabolite levels are determined using “oil wells” and the schemes described above with minor modifications. Acid or alkali extracts are employed depending on the metabolite.41 Example: Glucose (Scheme VI) A 0.1-µl aliquot from the acid extract is added to 0.1 µl of a solution containing 50 mM Tris-HCl, pH 8.0, 0.04% BSA, 0.4 mM DTT, 0.1 mM NADP+, 4 mM MgCl2, 0.5 mM ATP, 0.08 U/ml G6PdH, and 0.5 U/ml yeast hexokinase (specific activity 450 U/mg). The reaction mixture is incubated for 20 min at 20◦ followed by 0.1 µl 0.1 M NaOH. The mixture is incubated for 20 min at 80◦ . Five µl of NADP cycling reagent (composition described above with the exception that GDH
196
[16]
SPECIALIZED USES
ATP Glucose
HK
ADP G-6P
G6PdH
NADP+
6-PG NADPH
SCHEME VI. Glucose assay.
and G6PdH concentrations are doubled) is added to each sample and the solution incubated for 60 min at 38◦ . Cycling is terminated with 0.5 µl 1 M NaOH and heating to 80◦ for 20 min. A 5-µl aliquot is transferred to 1 ml 6-PG indicator reagent and NADPH fluorescence is determined.
[16] Laser Capture Microdissection in Pathology By FALKO FEND, KATJA SPECHT, MARCUS KREMER, and LETICIA QUINTANILLA-MARTI´NEZ Introduction The molecular genetic analysis of pathologically altered tissues has greatly increased our understanding of the etiologies and pathogenesis of human disease processes. The identification of recurrent genetic alterations has a major impact on the pathologic diagnosis of cancer, and conventional morphological tumor classification will rapidly be replaced by defining disease entities based on the integration of clinical, morphological, phenotypical, and genetic information. Moreover, the establishment of individual molecular profiles of tumors may help to identify targets for specific therapeutic intervention. However, primary tissues are a complex mixture of various cell types, and tumors contain an abundance of reactive stromal and inflammatory cells, which frequently outnumber the neoplastic population. This inherent complexity can crucially influence the results of molecular genetic examinations of primary tissues, since many alterations such as loss of heterozygosity or point mutations in tumor suppressor genes or oncogenes can go undetected by standard detection methods if the percentage of “contaminating” stromal cells reaches a certain threshold. In expression profiling of bulk tissue, admixed cell populations can potentially obscure tumor-specific signatures and can make message assignment to specific cell types impossible. Furthermore, early pathologic lesions, such as dysplasia or carcinoma in situ, are frequently inaccessible for conventional molecular analysis.
METHODS IN ENZYMOLOGY, VOL. 356
Copyright 2002, Elsevier Science (USA). All rights reserved. 0076-6879/02 $35.00
[16]
LCM IN PATHOLOGY
197
To circumvent these problems and to obtain more homogeneous cell populations for molecular analysis, manual and micromanipulator-based microdissection techniques have been developed during the past decade.1–5 Despite the unquestionable progress achieved with these approaches, such as the unveiling of the lineage and clonality of the malignant cells in Hodgkin’s disease,6 the time-consuming nature of microdissection and the significant manual dexterity required for it have until recently prevented its broad application in pathology. The development of easy-to-handle, laser-assisted technologies such as laser capture microdissection (LCM) or laser microbeam microdissection (LMM) allows rapid and highly precise procurement of purified cell populations suitable for a variety of downstream analyses.7–11 For many diagnostic applications, the use of microdissected cells as template source requires few, if any, modifications of standard protocols, and LCM can easily be integrated into the routines of a molecular pathology laboratory. This chapter gives an overview of applications for laser-assisted microdissection in pathology, focused on, but not restricted to, LCM. We describe various protocols for tissue preparation, microdissection, and DNA and RNA analysis from microdissected tissues. Tissue Preparation Both fresh frozen tissues and fixed, paraffin-embedded specimens, as well as cytological preparations, can be used for LCM. The major difference between routinely fixed and frozen specimens lies in the amount and quality of nucleic acids and protein which can be isolated from these sources. However, pre-LCM tissue preparation and microdissection itself are also critically influenced by the type of starting material. 1
F. d’Amore, J. A. Stribley, T. Ohno, G. Wu, R. S. Wickert, J. Delabie, S. H. Hinrichs, and W. C. Chan, Lab. Invest. 76, 219 (1997). 2 G. Deng, Y. Lu, G. Zlotnikov, A. D. Thor, and H. S. Smith, Science 274, 2057 (1996). 3 R. K¨ uppers, M. Zhao, M. L. Hansmann, and K. Rajewsky, EMBO J. 12, 4955 (1993). 4 L. Whetsell, G. Maw, N. Nadon, P. D. Ringer, and F. V. Schaefer, Oncogene 7, 2355 (1992). 5 J. J. Going and R. F. Lamb, J. Pathol. 179, 121 (1996). 6 R. K¨ uppers, K. Rajewsky, M. Zhao, G. Simons, R. Laumann, R. Fischer, and M. L. Hansmann, Proc. Natl. Acad. Sci. U.S.A. 91, 10962 (1994). 7 M. B¨ ohm, I. Wieland, K. Sch¨utze, and H. R¨ubben, Am. J. Pathol. 151, 63 (1997). 8 R. F. Bonner, M. Emmert-Buck, K. Cole, T. Pohida, R. Chuaqui, S. Goldstein, and L. A. Liotta, Science 278, 1481 (1997). 9 M. R. Emmert-Buck, R. F. Bonner, P. D. Smith, R. Chuaqui, Z. Zhuang, S. R. Goldstein, R. A. Weiss, and L. A. Liotta, Science 274, 998 (1996). 10 L. Fink, W. Seeger, L. Ermert, J. H¨ anze, U. Stahl, F. Grimminger, W. Kummer, and R. M. Bohle, Nat. Med. 4, 1329 (1998). 11 K. Sch¨ utze and G. Lahr, Nat. Biotechnol. 16, 737 (1998).
198
SPECIALIZED USES
[16]
Paraffin-Embedded Tissues Paraffin embedding, usually preceded by fixation in neutral, buffered 10% formalin, is still the most widely used method for the processing and conservation of diagnostic pathological specimens. However, cross-linking fixatives, such as formalin, lead to significant fragmentation of nucleic acids and cannot be used for techniques such as Southern or Northern blot analysis, which require large amounts of high molecular weight DNA and RNA, respectively. Nevertheless, microdissected paraffin-embedded tissues can serve as template sources for PCR-based analyses of both DNA and RNA, as long as relatively small amplicon sizes are used. Several groups have made efforts to replace formalin with precipitating fixatives such as ethanol, followed by paraffin embedding for improved preservation of nucleic acids.12,13 Although this represents a promising step to ensure both excellent morphology and superior quality of biologic macromolecules, formalin-fixed specimens still represent the biggest source of archival material for molecular studies. The preparation of paraffin sections for LCM and subsequent DNA or RNA extraction requires little deviation from standard laboratory procedures. Paraffin sections are cut under precautions against cross contamination between different tissue samples, then mounted on standard or plus-charged glass slides, depending on the subsequent staining procedure. After drying at 60◦, usually overnight, the slides are dewaxed in xylene 2× for 5 min, rehydrated through graded alcohols, and finally immersed in distilled water. If the slides are used for RNA extraction, nuclease-free water (DEPC-treated water) should be employed. Hematoxylin–Eosin Staining. The slides are stained in Mayer’ hematoxylin solution (Sigma-Aldrich, Deisenhofen, Germany) for 30 sec to 1 min, followed by blueing solution, 70% ethanol, eosin (5–20 sec), and dehydration through graded alcohols and 2 changes of xylene. The use of fresh, 100% ethanol as the final step before xylene is of great importance, because residual humidity can severely interfere with tissue transfer during LCM. Stained, dehydrated slides should be used for LCM as soon as possible, because prolonged storage may lead to reduced tissue transfer and may also influence the quality of nucleic acids. Stained slides should be stored in the presence of desiccants. Several other staining techniques such as hematoxylin or hemalum only, nuclear fast red, and others have been tested and may give equivalent or superior morphology, depending on the type of tissue.14 Immunohistochemical Staining for Paraffin Sections. A drawback of laserassisted microdissection is the requirement of dehydrated tissue sections or cell 12
S. M. Goldsworthy, P. S. Stockton, C. S. Trempus, J. F. Foley, and R. R. Maronpot, Mol. Carcinog. 25, 86 (1999). 13 M. Shibutani, C. Uneyama, K. Miyazaki, K. Toyoda, and M. Hirose, Lab. Invest. 80, 199 (2000). 14 T. Ehrig, S. A. Abdulkadir, S. M. Dintzis, J. Milbrandt, and M. A. Watson, J. Mol. Diagn. 3, 22 (2001).
[16]
LCM IN PATHOLOGY
199
preparations without coverslip, which leads to a significant decrease in optical resolution and loss of cytological detail. Although routinely stained sections (e.g., hematoxylin–eosin) are frequently sufficient for target recognition in wellstructured tissues, precise microdissection from tissues lacking easily identifiable architectural features such as lymphoid tissues, inflammatory infiltrates, or diffusely infiltrating neoplasms can be virtually impossible. In practice, the precision of laser-assisted microdissection is more frequently limited by difficulties in recognizing the target cells rather than by the technical specifications of the dissection tool. Immunohistochemical or cytochemical staining techniques can improve precision of LCM by rendering high-contrast targets and further allow separation of morphologically homogeneous cell populations according to phenotypical or functional criteria. Standard immunohistochemical procedures for paraffin-embedded tissues do not interfere with LCM and do not seem to have a major influence on subsequent DNA recovery.1,15 Our laboratory uses for most part routine staining protocols for diagnostic immunohistochemistry, including appropriate heat-induced antigen retrieval, as determined by the primary antibody used. To prevent detachment of slides during the staining procedure, plus-charged (Superfrost Plus, Fisher Scientific, Pittsburgh, PA) or coated (e.g., with poly-L-lysine) slides should be used, which do not interfere with tissue transfer during LCM if handled properly. For the detection of most primary antibodies, the ABC (avidin–biotin complex) technique is employed, with a biotinylated secondary antibody and horseradish peroxidase-labeled avidin as third step. Diaminobenzidine (DAB) as chromogen is well suited for LCM, since the stained sections can be counterstained with hemalum and dehydrated as above, and the stain does not interfere with PCR. In fact, the color precipitate remains on the membrane after digestion or elution of captured cells and can serve as visual control and documentation of dissection specificity (Fig. 1). Since strong staining results may be beneficial for the identification of the targeted cells, standard protocols may be optimized accordingly, including overnight incubation with the primary antibody or increased antibody concentrations. A short immunostaining protocol for frozen sections designed to reduce the exposure to aqueous media is described below. LCM of Paraffin Sections. LCM of paraffin sections, whether routinely stained or immunostained, is usually straightforward, and even archival, stained sections can be used successfully after removal of the coverslip with xylene. However, some problems may be encountered during LCM. 1. Poor visualization of targeted cell population. Depending on tissue architecture and type of staining, poor visualization can severely compromise dissection 15
F. Fend, L. Quintanilla-Martinez, S. Kumar, M. W. Beaty, L. Blum, L. Sorbara, E. S. Jaffe, and M. Raffeld, Am. J. Pathol. 154, 1857 (1999).
200
SPECIALIZED USES
[16]
A
B
FIG. 1. (A) LCM of a paraffin section of a composite non-Hodgkin’s lymphoma immunostained for CD5. The neoplastic follicles are CD5 negative (asterisks), whereas the interfollicular neoplastic cells express CD5, in addition to reactive T-cells. The holes left behind by the procedure are clearly visible. (Three-step immunoperoxidase technique, ×100.) Amplification of rearranged immunoglobulin heavy chain genes from the microdissected CD5+ and CD5− cell populations repeatedly rendered two products of different size and sequence (not shown), confirming the presence of two different clones. [F. Fend, L. Quintanilla-Martinez, S. Kumar, M. W. Beaty, L. Blum, L. Sorbara, E. S. Jaffe, and M. Raffeld, Am. J. Pathol. 154, 1857 (1999)]. (B) Cap surface after proteinase K digestion. Although the cellular material has already been removed, the outlines of the captured cells immunostained for CD5 are clearly visible due to the residual DAB precipitate which remains on the thermoplastic membrane and allows control of dissection specificity.
[16]
LCM IN PATHOLOGY
201
specificity. Here are some remedies: Use the diffuser or add a drop of xylene to the slide, which works like a coverslip, and create a roadmap image. After evaporation of the xylene, use this image as guidance. Reduce the time in hematoxylin or change to another stain. For certain tissues, the use of immunostaining may be mandatory. Mounted and stained parallel sections can also aid to “navigate” on the slide used for LCM, but this is only appropriate for tissues with larger, predictable anatomical structures which can be followed through step sections.16 2. The captured tissue remains on the slide. This is probably the most vexing problem. Frequently, insufficient dehydration is the reason, and reimmersion of the section in absolute ethanol followed by xylene may be a remedy. Another strategy is to incubate the slide in water with 3% glycerol before dehydration, which can be of help for various cell or tissue preparations.17 Uneven section surfaces, slightly tilted placement of the cap, or repositioning of the cap after dissection with adherent cells on the lower surface can compromise tissue contact and dissection efficiency. DNA Extraction. After control of dissection specificity and, if necessary, removal of nonspecifically adherent cells with a light adhesive, put the cap on a 500-µl tube which contains 50–100 µl of TE buffer with 400 µg/ml proteinase K. The optimal buffer volume depends on the amount of captured cells. For small numbers of cells, the area containing the captured cells can be cut from the cap surface under the microscope with a sterile blade and directly immersed into 10–20 µl of proteinase K-containing buffer. Alternatively, specially designed caps for small amounts of cells can be used in conjunction with the micro-extraction chamber suitable for small fluid volumes (Arcturus Engineering, Santa Clara, CA). Unless a membrane fragment is directly immersed in the buffer, the tube carrying the cap is inverted and incubated at 55◦ for 4–8 hr or overnight. Despite the small amount of cells, complete digestion of cells from paraffin-embedded sources requires at least several hours. It is advisable to control for complete digestion by restaining the cap to detect any remaining cell fragments. Because of the limited amount of cells, proteinase K digestion without subsequent purification steps (e.g., organic extraction) is usually sufficient for standard PCR. After heat inactivation of proteinase K and spinning down of any particulate debris, the supernatant can be used for PCR directly. Determination of B-Cell Clonality by PCR Using Consensus Primers against Framework Three Region (FR3) of Immunoglobulin Heavy Chain Genes (IgH). The determination of B- or T-cell clonality of lymphoid proliferations is one of the 16
M. H. Wong, J. R. Saam, T. S. Stappenbeck, C. H. Rexer, and J. I. Gordon, Proc. Natl. Acad. Sci. U.S.A. 97, 12601 (2001). 17 L. Jin, C. A. Thompson, X. Qian, S. J. Kuecker, E. Kulig, and R. V. Lloyd, Lab. Invest. 79, 511 (1999).
202
SPECIALIZED USES
[16]
most frequently used diagnostic molecular assays in pathology. Since the percentage of clonal cells must reach a certain threshold of at least 2–3%—in reality often closer to 10% of the total cell population—to be reliably detected, microdissection can be used to enrich the cell population in question. Besides Hodgkin’s disease, many other lesions such as nodular lymphoid infiltrates in the bone marrow or extranodal locations or composite lymphoma lend themselves to microdissection (Fig. 1). However, the use of microdissected tissue fragments for clonality determination requires rigorous control and careful interpretation of results, since the small amounts of lymphoid cells serving as template for amplification can result in “pseudoclonal” amplification products, which should not be equated with malignancy. The effects of formalin fixation with decrease in template quality and subsequent preferential amplification of some rearrangements over others, due to the use of consensus primers with varying binding affinity, can further aggravate this problem. Whenever possible, a single-step PCR should be used to reduce the possibility of clonal amplification products of uncertain relevance. If larger amounts (hundreds to thousands) of cells can be collected, conventional single-step PCR is sufficient to generate enough PCR product for fragment length analysis, direct sequencing, or cloning. Perform all reactions in duplicate, preferably using cells from different microdissections. If clonal bands are not identical in all reactions, they are likely the result of preferential amplification of rare B cells. The following protocol is currently used in our laboratory15,18 : The reaction volume of 25 µl contains 0.4 µM/liter of each primer (FR3a and LJH19 ), 0.2 mM/liter dNTPs, 2 mM/liter MgCl2, 1.25 U of Taq polymerase (Amplitaq Gold, PerkinElmer, Weiterstadt, Germany) and 1–5 µl of template DNA. After initial denaturation at 94◦ for 4 min, 40 cycles of amplification are performed at 94◦ for 1 min, 56◦ for 30 sec, and 72◦ for 30 sec, followed by a final extension step at 72◦ for 10 min. The PCR products can be run on a 3% Metaphor gel (FMC Bioproducts, Rockland, ME) or on a 16% polyacrylamide gel. For computer-assisted fragment length analysis, one of the primers is end-labeled with fluorescein, and amplification is performed under identical conditions. Fluorescein-labeled PCR products are analyzed on a high-resolution polyacrylamide gel using an ABI Prism 377 automated sequencer and Genescan software (PerkinElmer). If only very small numbers of cells are available, such as groups of Hodgkin and Reed–Sternberg cells, a seminested protocol using an internal primer directed against homologous sequences of the IgH joining region genes (VLJH) is necessary.20 After first-round amplification as described above, another 25 cycles under identical conditions are performed, replacing the LJH primer with the VLJH 18
M. Kremer, A. D. Cabras, F. Fend, S. Schulz, K. Schwarz, H. Hoefler, and M. Werner, Hum. Pathol. 31, 847 (2000). 19 G. H. Segal, T. Jorgensen, A. S. Masih, and R. C. Braylan, Hum. Pathol. 25, 1269 (1994). 20 M. Kremer, M. Sandherr, B. Geist, A. D. Cabras, H. H¨ ofler, and F. Fend, Mod. Pathol. 14, 91 (2001).
[16]
LCM IN PATHOLOGY
203
primer and using 3–5 µl of the first-round product, both undiluted and 1 : 10 diluted, as template for the second round. The isolation of groups of 30 to 50 RS cells per cap and their joint analysis greatly reduce the work needed for clonality analysis of single cells and allow repeat amplification or investigation of other genes. Since contamination by occasional nonneoplastic lymphocytes is likely with this approach, cloning and sequencing of PCR products obtained from several distinct groups of RS cells originating from different microdissections is necessary to confirm the tumor cell origin of the amplification product. Amplified bands are purified from agarose or polyacrylamide gels with appropriate techniques, ligated into the PCR2.1 vector (TA cloning kit, Invitrogen, Carlsbad, CA), and cloned into INVαF’ bacteria.15,20 Sequencing of several inserts obtained from each PCR product will confirm or disprove the clonal identity of the isolated cells, and a minority of contaminating clones will not interfere with the interpretation. RNA Extraction and Real-Time TaqMan RT-PCR in Formalin-Fixed, ParaffinEmbedded Tissues. RNA extracted from formalin-fixed, paraffin-embedded (FFPE) tissues is generally of poor quality because degradation of the RNA can occur before completion of the formalin fixation process. Moreover, as mentioned above, formalin causes cross-linkage of nucleic acids and proteins and covalently modifies RNA by the addition of monomethylol groups to the bases, making RNA extraction, cDNA synthesis, and quantitation analysis problematic.21 When performing gene expression analysis in FFPE tissues, it is therefore extremely important (a) to choose an RNA extraction procedure that provides only minimally cross-linked RNA and (b) to select very small target sequences in a range of 60–100 bp for real-time RT-PCR, enabling the detection of fragmented and degraded RNA.22,23 Generally, there is a huge number of methods available for RNA extraction; however, in our experience, the most successful method in terms of yield of extractable RNA and suitability of the RNA for real-time RT-PCR analysis involves proteinase K digestion.22 Briefly, RNA from a small number of microdissected cells (20–10,000 cells) is extracted using a modification of the method described by Rupp and Locker.24 Microdissected cells are transferred to a 1.5-ml microcentrifuge tube and lysed in 200 µl lysis buffer, containing 10 mmol/liter Tris-HCl (pH 8.0), 0.1 mmol/L EDTA (pH 8.0), 2% sodium dodecyl sulfate (pH 7.3), and freshly added 500 µg/ml proteinase K. Cells are digested for 12 hr at 60◦ until the tissue is completely solubilized. After heat inactivation of the proteinase K for 5 min at 95◦ , the RNA is purified by phenol/chloroform extraction: 1/10 volumes 2 M sodium acetate (pH 4.0 ), 1 volume water-saturated acidic 21
N. Masuda, T. Ohnishi, S. Kawamoto, M. Monden, and K. Okubo, Nucleic Acids Res. 27, 4436 (1999). 22 K. Specht, T. Richter, U. M¨ uller, A. Walch, M. Werner, and H. H¨ofler, Am. J. Pathol. 158, 419 (2001). 23 A. E. Krafft, B. W. Duncan, K. E. Bijwaard, J. K. Taubenberger, and J. H. Lichy, Mol. Diagn. 2, 217 (1997). 24 G. M. Rupp and J. Locker, BioTechniques 6, 56 (1988).
204
SPECIALIZED USES
[16]
phenol, and 1/5 volume chloroform are added to the reaction. After vortexing, the samples are put on ice for 15 min and then centrifuged at 14,000 rpm for 20 min at 4◦ . The upper, aqueous phase containing the RNA is transferred to a new microcentrifuge tube and the RNA is precipitated with 2 µl of 10 mg/ml carrier glycogen and 1 volume isopropanol. After incubation for 2 hr at −20◦ , the RNA is pelleted by centrifugation at 14,000 rpm for 20 min. The pellet is washed once with 70% ethanol, dried, and resuspended in 20 µl RNase-free H2O. If the primers and probes used for subsequent real-time RT-PCR do not span an intron, the removal of genomic DNA by DNase digestion is necessary at this point. cDNA synthesis reaction is carried out with Superscript II Reverse Transcriptase in a final reaction volume of 20 µl as described in the instruction manual (Superscript Choice system) provided by Life Technologies. One-half of the isolated RNA (10 µl) is used for reverse transcription, while a no RT control reaction should be performed in parallel with the other half of the RNA. RNA is annealed with 250 ng random primers at 25◦ for 10 min and then reverse transcribed with 200 U (1 µl) Superscript reverse transcriptase in 4 µl of 5× first-strand buffer [250 mM Tris-HCL (pH 8.3), 375 mM KCL, 15 mM MgCl2], 10 mM dithiothreitol (DTT), 1 µl of 0.5 mM of each dNTP, and 1 µl of RNase inhibitor (40 U)] for 60 min at 42◦ . Typically, 1/10–1/20 of the cDNA reaction is then used for subsequent real-time TaqMan PCR. Frozen Tissues Although frozen tissue represents a superior source for intact biomolecules compared to FFPE tissues, the morphology of frozen sections frequently is poor and further compromises the precision of microdissection. As mentioned above, IHC is an excellent tool for improving the visualization of the target populations. However, standard IHC protocols require several hours of incubation in aqueous media, which results in a significant loss of RNA through the action of ubiquitous RNases. Therefore, several groups have developed immunostaining protocols suitable for LCM and subsequent gene expression analysis.17,25,26 The following rapid immunostaining protocol can be performed with many different primary antibodies and standard reagents and renders mRNA of good quality, although a loss of mRNA in comparison to rapid routine staining (e.g., hematoxylin–eosin) does still occur.25 Immuno-LCM of Frozen Sections. The Quick Staining kit (DAKO Corp., Carpinteria, CA) used in the initial publication is no longer commercially available.25 Alternatively, the procedure can be performed with analogous reagents suitable for quick immunostaining, such as a three-step (strept-)avidin–biotin 25
F. Fend, M. R. Emmert-Buck, R. Chuaqui, K. Cole, J. Lee, L. A. Liotta, and M. Raffeld, Am. J. Pathol. 154, 61 (1999). 26 H. Murakami, L. Liotta, and R. A. Star, Kidney Int. 58, 1346 (2000).
[16]
LCM IN PATHOLOGY
205
FIG. 2. LCM of a frozen section of Hodgkin’s disease immunostained for the CD30 antigen. The holes left behind are clearly visible, as well as the membrane staining of other tumor cells. The arrow denotes granulocytes, which show positivity due to endogenous peroxidase activity. (Rapid three-step immunoperoxidase technique, × 600.)
system with a biotinylated secondary antibody and horseradish peroxidase (HRP) (strept-)avidin complex optimized for high sensitivity (Fig. 2). A useful method is the employment of secondary antibodies coupled to a polymer backbone carrying multiple HRP molecules (EnVision, DAKO), which abolish the necessity for a third incubation step before color development and show increased sensitivity. Furthermore, the system is biotin-free and therefore lacks background staining as a result of endogenous biotin.27 Staining procedure Frozen sections are mounted on charged slides (Superfrost Plus) and immediately refrozen on dry ice. Sections can be stored at −80◦ . The sections are briefly thawed at room temperature and immediately immersed in cold acetone for 1–2 min. Drying of the sections before fixation will severely compromise subsequent tissue capture. Do not stain more than 1–3 sections at one time, because multiple slides will lead to a significantly prolonged incubation time. After evaporation of acetone, the slides are rinsed briefly in buffered PBS or Tris pH 7.4 and incubated with 70–100 µl primary antibody for 1.5–3 min. 27
U. K¨ammerer, M. Kapp, A. M. Gassel, T. Richter, C. Tank, J. Dietl, and P. Ruck, J. Histochem. Cytochem. 49, 623 (2001).
206
SPECIALIZED USES
[16]
The optimal dilution and staining time has to be determined individually. The slide is rinsed briefly with buffer and incubated with the secondary reagent for 2–3 min. If a three-step avidin–biotin system with a biotinylated secondary antibody is used, it is followed by incubation with the streptavidin–HRP complex for 2–3 min. Subsequently, the slide is covered with 100 µl of freshly prepared 3,3 diaminobenzidine solution for 2–5 min. Then the slide is rinsed in water and briefly counterstained with hematoxylin (20–30 sec) if desired. The sections are dehydrated through graded alcohols (15–30 sec each, including twice 100% ethanol) and xylene (twice for 2 min each). For all aqueous solutions, the use of pure, RNase treated water is recommended. During the incubation steps, placental RNase inhibitor (PerkinElmer) may be added in a concentration of 200–400 U/ml. RNA Extraction. After LCM, the tubes carying the caps with the captured cells are put on ice. RNA extraction is performed using lysis in guanidinium isothiocyanate followed by extraction with water-saturated phenol and subsequent precipitation with cold isopropanol and glycogen added as carrier (Micro RNA isolation kit, Stratagene, La Jolla, CA). If the presence of contaminating DNA is critical, DNase digestion is mandatory. RNA is dissolved in pure water containing 1 µl of RNase inhibitor and incubated for 2 hr at 37◦ with 10–20 U of DNase I (Genhunter Corp., Nashville, TN). After reextraction of RNA following the same protocol as above, the RNA is dissolved in pure water, and 1 µl of RNase inhibitor is added. Reverse transcription is performed with 2.5 µmol/liter of random hexamers, 250 µmol/liter of each dNTP, and 100 U of MMLV reverse transcriptase (GenHunter) in a final volume of 20 µl. A mock reaction without RT should be performed in parallel.9,15 Depending on the amount of isolated cells, 20 µl of cDNA is usually sufficient for 10–20 or more single-step PCR reactions, and fragments larger than 500 bp can be amplified successfully. Conclusions The above protocols represent a small selection of fairly simple, easy-to-use techniques which can be readily performed in most molecular pathology laboratories. However, methodical improvements and further technical developments are realized at a rapid pace in all relevant areas, including tissue conservation and pre-LCM preparation, identification, and precise isolation of targeted cells, as well as more sensitive downstream analytical techniques adapted to small amounts of tissue, including high throughput screening methods such as cDNA microarrays or proteomics. Our increasing ability to correlate molecular findings with morphology and phenotype on the microscopic level will have a profound impact on all aspects of pathology.
[17]
LCM ANALYSIS OF MOUSE MODELS
207
[17] Use of Laser Capture Microscopy in the Analysis of Mouse Models of Human Diseases By MERAL J. ARIN and DENNIS R. ROOP Introduction The use of laser capture microdissection (LCM) has been a major step toward a better understanding of the molecular mechanisms that play a role in various disease processes.1 The ability to readily obtain and isolate cells of interest from complex tissues has made LCM an attractive technology in the fields of genomics and proteomics. To study genetic changes that occur in a particular cell type, tiny subpopulations of cells need to be isolated to reduce the risk of contamination. Before the development of LCM, this process of isolating specific cells of interest was very tedious, irreproducible, and inefficient.2 The skin is an attractive organ for the application of LCM since it is composed of a complex mixture of different cell types. Keratinocytes represent the major cell type in the epidermis, the outer layer of the skin. They contain highly polymeric molecules, the keratins, that confer mechanical stability on these cells. Beneath the epidermis lies the dermis which is an elastic support structure and contains fibroblasts as the predominant cell type. Dermal fibroblasts synthesize the structural components of the dermis which include collagen and elastic fibers as well as ground substance. Molecular defects in structural components of the epidermis and dermis have been identified in several hereditary blistering disorders of the skin.3 These disorders are characterized by mutations in various genes encoding intra- and extracellular molecules leading to blister formation in distinct layers of the skin. We describe the use of LCM in the analysis of animal models of hereditary blistering disorders. Animal Models of Hereditary Blistering Disorders Animal models are valuable tools to study pathways involved in disease processes. Several animal models have been generated and are currently being used to study the molecular and cellular basis of genetic skin disorders.4 For the analysis of two hereditary blistering disorders, epidermolytic hyperkeratosis (EHK) and 1
M. R. Emmert-Buck, R. F. Bonner, P. D. Smith, R. F. Chuaqui, Z. Zhuang, S. R. Goldstein, R. A. Weiss, and L. A. Liotta, Science 274, 998 (1996). 2 J. J. Going and R. F. Lamb, J. Pathol. 179, 121 (1996). 3 B. P. Korge and T. Krieg, J. Mol. Med. 74, 59 (1996). 4 M. J. Arin and D. R. Roop, Trends Mol. Med. 7, 422 (2001).
METHODS IN ENZYMOLOGY, VOL. 356
Copyright 2002, Elsevier Science (USA). All rights reserved. 0076-6879/02 $35.00
208
SPECIALIZED USES
[17]
epidermolysis bullosa simplex (EBS), we generated mouse models that reproduce the human disease at both the phenotypic and genetic levels. To understand the role of stem cells in these disorders, we generated mouse models that allow focal induction of the EBS and EHK phenotypes in a circumscribed area of the skin.5,6 In these mouse models, we used LCM to isolate epidermal keratinocytes from stained tissue sections obtained from areas of interest (see below). Tissue Microdissection Several methods have been developed to isolate subpopulations of cells from tissue sections. These include ablation of unwanted regions and subsequent collection of the remaining cells and manual separation of the cells of interest by fine needle, pipette, or blade, as well as irradiation of manually ink-stained sections to destroy unwanted genetic material.7,8 These conventional techniques are very time consuming and require a high degree of manual dexterity, which limits their practical use. Compared to LCM, which uses a laser pulse to precisely target the cells of interest, these techniques have proved ineffective and nonefficient. Using LCM, it has been possible to reliably isolate pure populations of cells from tissue sections under microscopic visualization. To perform LCM, a histological section containing the tissue of interest is placed under the specifically designed microscope and the image is transferred to a computer screen. A transparent thermoplastic membrane is mounted on an optically clear cap which fits on a standard 0.5-ml microcentrifuge tube for further processing. The cap is positioned to cover the area of interest. Cells can then be melted onto the ethylene vinyl acetate (EVA) membrane by a near infrared laser beam, which builds a strong focal bond between the cells and the film.1 The strong adherence of the tissue to the activated membrane allows selective removal of the cells of interest. Large numbers of cells can be isolated in a short time by repeating multiple laser impulses across the whole cap surface. The size of the laser beam spot can be selected at 7.5, 15, and 30 µm. We used a spot size of 30 µm in our experiments to capture 1–2 keratinocytes per laser pulse. The cells of interest can then be easily pulled away from the slide by simply lifting the membrane from the tissue slide. The cells are then ready for molecular analysis. This technique can be used on formalin-fixed, paraffin-embedded tissue sections, as well as on frozen tissues. The quantity of products amplified by PCR (see below) is higher from frozen tissue than from paraffin-embedded tissue.9 5
M. J. Arin, M. A. Longley, X. J. Wang, and D. R. Roop, J. Cell. Biol. 152, 645 (2001). T. Cao, M. A. Longley, X. J. Wang, and D. R. Roop, J. Cell. Biol. 152, 651 (2001). 7 L. Whetsell, G. Maw, N. Nadon, D. P. Ringer, and F. V. Schaefer, Oncogene 7, 2355 (1992). 8 D. Shibata, D. Hawes, Z. H. Li, A. M. Hernandez, C. H. Spruck, and P. W. Nichols, Am. J. Pathol. 141, 539 (1992). 9 S. M. Goldsworthy, P. S. Stockton, C. S. Trempus, J. F. Foley, and R. R. Maronpot, Mol. Carcinog. 25, 86 (1999). 6
[17]
LCM ANALYSIS OF MOUSE MODELS
209
The most important advantage of LCM is its speed combined with its precision. Thousands of cells can be captured within a few minutes. The exact morphology of microdissected cells is retained during this procedure and can be visualized directly after microdissection. This allows control and documentation of captured cells throughout the procedure. Some limitations exist, however, including the considerable amount of time required when a large number of tissue samples is involved. In addition, the optical resolution obtained is limited because of scattering of transmitted light that passes through the air spaces within dehydrated tissue (and absence of a coverslip), which may limit the identification of cells of interest.10,11 In certain tissues, it is difficult to identify different cell types by morphology alone. Immuno-LCM allows a rapid and precise isolation of specific subpopulations of cells that express distinct proteins. The advantages of immunohistochemically stained frozen sections are good optical resolution and faster and more precise identification of stained cell populations. In addition, frozen sections have been shown to yield a much higher quality of RNA for various applications such as generation of expression libraries and screening of cDNA arrays.12 We used LCM to analyze keratinocytes from our disease models. In these models, topical application of RU486 to a focal area of the skin results in activation of the mutant keratin allele giving rise to focal areas of affected skin (see below). Keratinocytes from affected and adjacent nonaffected skin were used for LCM analysis. LCM in Analysis of Mouse Models for Mosaic Skin Disorders Mosaic skin disorders are characterized by the presence of at least two genetically distinct cell populations from the same differentiation lineage. The molecular and cellular mechanisms that lead to clinical mosaicism are poorly understood. It remains unclear why mosaicism exists for certain disorders and not for others. Two candidate disorders to analyze the mechanisms that lead to clinical mosaicism are the keratin disorders, epidermolytic hyperkeratosis (EHK) and epidermolysis bullosa simplex (EBS). EHK is caused by point mutations in the genes encoding keratins 1 and 10, leading to blister formation in the suprabasal layers of the epidermis. In EBS, blistering takes place in the basal compartment and is caused by point mutations in keratins 5 and 14. Whereas mosaic forms exist for EHK with linear lesions of affected and unaffected skin, this has never been described for EBS.13 To investigate the underlying cause of this phenomenon, we generated mouse models for both disorders. In these 10
S. Curran, J. A. McKay, H. L. McLeod, and G. I. Murray, Mol. Pathol. 53, 64 (2000). F. Fend and M. Raffeld, J. Clin. Pathol. 53, 666 (2000). 12 F. Fend, M. R. Emmert-Buck, R. Chuaqui, K. Cole, J. Lee, L. A. Liotta, and M. Raffeld, Am. J. Pathol. 154, 61 (1999). 13 A. S. Paller, A. J. Syder, Y. M. Chan, Q. C. Yu, E. Hutton, G. Tadini, and E. Fuchs, N. Engl. J. Med. 331, 1408 (1994). 11
210
SPECIALIZED USES
[17]
FIG. 1. Schematic representation of LCM analysis of keratinocytes that were captured from the epidermis of an inducible mouse model that allows focal induction of the phenotype in a circumscribed area of the skin. In the example shown, bigenic mice were generated that contain a keratin 10-point mutation (∗), a neomycin resistance gene cassette (neo) flanked by loxP sites on one allele (mutK10neo), and an inducible Cre transgene (CrePR1) under the transcriptional control of the keratin 14 promoter. These mice lack a phenotype due to suppression of the mutant allele by the presence of neo cassette in intron 1. Topical application of the inducer (RU486) to the skin of these mice results in excision of the neo cassette and restoration of expression of the mutant allele (mutK10loxP) with the expected phenotype at the site of induction. Skin biopsies were taken from previously treated, phenotypic areas and keratinocytes were isolated from the epidermis by LCM for PCR analysis to confirm that the neo cassette was excised in lesional areas.
mouse models expression of mutated keratin proteins can be focally induced in restricted areas of the skin using the ligand-inducible CrePR1 system.14 Topical application of RU486 on these mice results in activation of CrePR1 (an inducible form of Cre recombinase, which recognizes very specific sequences termed loxP sites). In its activated form, CrePR1 excises the neomycin (neo) cassette, which is flanked by loxP sites and inhibits expression of the mutant keratin allele when inserted into the first intron. Excision of the neo cassette activates expression of the mutant keratin allele, resulting in a phenotype characteristic of either EHK or EBS (Fig. 1). We used LCM to capture keratinocytes from the epidermis from treated 14
C. Kellendonk, F. Tronche, A. P. Monaghan, P. O. Angrand, F. Stewart, and G. Sch¨utz, Nucleic Acids Res. 24, 1404 (1996).
[17]
LCM ANALYSIS OF MOUSE MODELS
loxP
211
loxP neo
mtK10neo R156C
K14.CREPR1 + RU486 loxP mtK10loxP FIG. 2. Induction and characterization of blisters in K14-CrePR1/mtK10neo mice. (A) RU486 was applied once a day for 3 to 5 consecutive days to the upper trunk and paws of newborn bigenic pups (+/mutK10neo.CrePR1). A 6-day-old pup is shown, where blisters formed at the site of induction. After rupture of the blisters, scaling developed around the site of the previous blister. (B) Paw of the same mouse 5 months after induction shows the persisting phenotype with thick hyperkeratoses, as seen in older EHK patients. (C) H&E staining of a biopsy taken from a blistered paw shown in (A). The areas of cytolysis in the suprabasal layers of this section were subjected to LCM. (D) Schematic representation of the Cre-mediated excision of the neo cassette in skin areas treated with the inducer. [Modification of Fig. 2 in M. J. Arin, M. A. Longley, X. J. Wang, and D. R. Roop, J. Cell. Biol. 152, 645 (2001) by copyright permission of The Rockefeller University Press.]
and untreated areas. Biopsies were processed as described below and sections were visualized using the PixCell LCM system (Arcturus Engineering, Mountain View, CA). DNA was extracted and subjected to PCR analysis to determine the presence or absence of the neo cassette. In the EHK model, we found permanent excision of the neo cassette, and thus activation of the mutant allele, confirming that Cre-mediated recombination occurred in epidermal stem cells (Fig. 2).5 Our results indicate that in EHK, mutant and wild-type stem cells can coexist in the
212
SPECIALIZED USES
[17]
basal compartment, leading to a persistent phenotype.5 In the EBS model, blisters formed after activation of the mutant allele; however, they healed after 10 days and never reappeared. We used LCM to document that the neo cassette was initially excised in the blistered area, but cells in these lesions were replaced by phenotypically normal epidermal stem cells (i.e., those surrounding the treated area that retained the neo cassette) (Fig. 3).6 The schematic shown in Fig. 4 summarizes the results obtained with our inducible mouse models. Selective pressure against stem cells containing an activated EBS allele occurs because these stem cells express the mutant K14 protein. In the EHK model, the EHK allele is also activated in stem cells; however, it is the differentiated progeny of these cells, not the stem cells themselves, that express the mutant K10 protein. These observations could explain why mosaic forms have been reported for EHK, but not EBS. Protocol for LCM in Our Analysis of Disease Models 1. Skin biopsies are fixed in 10% formalin for a few hours to overnight; 2.5 to 4 hr seem to work best since DNA yield decreases with prolonged fixation times. After dehydration through graded ethanol, the sections are embedded in paraffin. 2. Tissue sections are mounted on plain, untreated, and uncharged glass slides. The use of charged slides increases the strength of adhesion of the section to the slide, which interferes with tissue capture. We routinely use 5-µm sections of paraffin-embedded tissue for capturing keratinocytes, but sections of between 5 and 20 µm in thickness can be used depending on the diameter of the nucleus of the cell of interest. Air dry sections overnight at room temperature. If sections do not adhere to the slide sufficiently they can be baked at 42◦ for up to 8 hr. 3. Deparaffinize 5-µm sections in xylene two times for 5 min and rinse in a graded alcohol series (100%, 95%, 70% ethanol for 30 sec each) with a final rinse in distilled water for 30 sec. The slides are now ready for staining, e.g., with Nuclear Fast Red (Vector Laboratories, Burlingame, CA). 4. Fix the sections in 75% ethanol for 30 sec or in acetone for 4 min at 4◦ . Transfer to distilled water for 30 sec. Stain with Nuclear Fast Red for 30–60 sec. Rinse in distilled water. Dehydrate 70% ethanol for 30 sec, 95% ethanol for 30 sec, 100% ethanol for 30 sec. Dip in xylene two times for 5 min. Air dry for at least 20 min in a hood. One reference slide is stained with H&E for identification of areas and cells of interest. 5. Sections are then visualized using the PixCell LCM system (Arcturus Engineering, Mountain View, CA). A thermoplastic polymer coating (ethylene vinyl acetate, CapSure) attached to a rigid support is placed in contact with the tissue section. The transfer film is activated by a near infrared laser pulse which melts the film onto the targeted cells, thus forming a strong focal bond. The desired amount of cells from the area of interest is captured onto one cap and is now ready for analysis. Conventional LCM typically uses 20–100 cells for one PCR reaction or
B
A
C
D
E
F loxP mtK14 neo
loxP neo
K14.CREPR1 + RU486 loxP mtK14 loxP FIG. 3. Induction and characterization of blisters in K14-CrePR1/mtK14neo mice. (A) Gross phenotype of an induced blister after treatment of the right paw with inducer. No blisters developed on the untreated leg (left paw). (B) The right front paw of a K14-CrePR1/mtK14neo pup 10 days after blister formation upon treatment. (C) The left front paw and leg of a K14-CrePR1/mtK14neo mouse 6 months after blister formation and cessation of inducer treatment. The blistered area, including the palm and leg, healed without scarring. No additional blisters formed without further inducer treatment. (D and E) H&E staining (D) and immunofluorescence microscopy with an anti-K14 antibody (Texas Red, E) of an induced blister edge. Blistering occurred in the basal cell layer (arrowheads) of the paw. The asterisk denotes cytolysis. (F) Schematic representation of the Cre-mediated excision of the neo cassette in skin areas treated with the inducer. LCM analysis was performed on nontreated skin, on the roof of the blister in panel A, and on the blistered area after healing. [Modification of Fig. 4 in T. Cao, M. A. Longley, X. J. Wang, and D. R. Roop, J. Cell. Biol. 152, 651 (2001) by copyright permission of The Rockefeller University Press.]
EBS-Model
EHK-Model
A
D
Activator
Activator
B
E
C
F
Phenotypically Normal Stem Cell Stem Cell with Activated K10 Mutation TA Cells and Differenting Progeny with Activated K10 Mutation FIG. 4. Schematic representation of the differences in behavior of epidermal stem cells in our mouse models for EBS and EHK. (A, D) Stem cells in the basal cell layer express low levels of the mutant K14 allele (mtK14neo). The mutant K10 allele (mtK10neo) is expressed at a low level in suprabasal cells. The expression of both mutant keratin alleles is insufficient to cause cell fragility, i.e., resulting in a subclinical phenotype. (B, E) On topical application of an inducer to the skin of K14-CrePR1/mtK14neo and K14-CrePR1/mtK10neo mice, respectively, the mutant alleles are activated by excision of the neo cassettes, thereby generating the dominant disease alleles (mtK14loxP, mtK10loxP). Blisters are formed as a consequence of the activator treatment. Within a few days, lesions are reepithelializing from the surrounding epidermis (arrows). (C) Nonphenotypic stem cell (mtK14neo) colonizes the wound area in the EBS model. As a consequence, blisters heal without scarring. (F) Although the neo cassette is excised from the mutant K10 allele, the gene is not expressed in stem cells, only in its differentiating progeny. Consequently, there is no selective pressure against stem cells containing the mtK10loxP allele. These stem cells persist and give rise to islands of mutant cells that result in persistent lesions throughout life.
[17]
LCM ANALYSIS OF MOUSE MODELS
215
5000 cells to generate a cDNA library.15 We routinely capture about 1000 cells per cap to perform several PCR reactions. LCM works best with stained and dehydrated sections. Any moisture, even a small amount, appears to inhibit the transfer of cells to the transfer film. In our hands, it has helped to rerinse the sections in xylene and let them air dry briefly before continuing the capturing process. 6. For DNA analysis, DNA was extracted in a buffer containing 1 mg/ml proteinase K, 0.5% Nonidet P-40, 0.25% Tween 20, 0.2 mM EDTA pH 8.0, and 10 mM Tris-HCl pH 8.0. We usually use a volume of 50 µl. The cap is placed onto a 0.5-ml tube and incubated upside down at 37◦ overnight with or without shaking. It is critical that the digestion buffer contact the tissue on the cap. 7. The next morning, the tube is centrifuged for 5 min and the cap is removed. The reaction is heated to 95◦ for 8 min to inactivate the proteinase K. An aliquot can be used directly for PCR amplification. In our experience, 40 cycles work well to amplify the sequence of interest. Conclusions Laser capture microdissection was a valuable tool in the analysis of these inducible mouse models for inherited skin blistering diseases. It allowed us to isolate contaminant-free keratinocytes from lesional and nonlesional areas of the skin and confirm at the molecular level whether Cre-mediated excision of the neo cassette had occurred. The information obtained by LCM provided new insight into the molecular and cellular basis of mosaic skin disorders, suggesting that a lack of selective pressure against stem cells containing certain mutations could explain the existence of mosaic forms of some diseases, but not others. Acknowledgments We thank C. Allred for use of the LCM equipment, P. Koch and B. Eckes for help with artwork, and M. Koster for comments on the manuscript. This work was in part supported by the Deutsche Forschungsgemeinschaft (Ar 291/1-1 and Ar 291/3-1) and the Koeln Fortune Program, Faculty of Medicine, University of Cologne (M.J.A.) and NIH Grants HD25479, AR62228, and AR47898 to D.R.R.
15
C. A. Suarez-Quian, S. R. Goldstein, T. Pohida, P. D. Smith, J. I. Peterson, E. Wellner, M. Ghany, and R. F. Bonner, BioTechniques 26, 328 (1999).
216
[18]
SPECIALIZED USES
[18] Use of Laser Microdissection in Complex Tissue By HOLGER S. WILLENBERG, RHODRI WALTERS, and STEFAN R. BORNSTEIN Introduction Light microscopy from a classical point of view was designed primarily to observe objects rather than to perform manipulations of cells or tissues. However, the knowledge that macroscopic phenomena originate from regulated ultrastructural and molecular alterations led to the desire to specifically investigate single cells, or even subcellular particles and structural elements. The characteristics of light permit particles the size of micrometers to be trapped within an electromagnetic field without being mechanically or chemically altered when the light is not absorbed. On the other hand, when a parallel laser light beam is strongly focused and energy absorption at the site of interest is very high, it is possible to cut thin lesions within the tissue. At such intensities, light is, however, able to damage ultrastructures by heat absorption causing tissue degeneration, development of plasma, or mechanical deformation. Therefore, the pulse rate and wavelength of the laser is adjustable within a certain range. Instruments designed to perform laser manipulation under visual control are now commercially available and playing an ever more prominent role in modern routine research of complex tissue. Brief History of Laser Microdissection The development of laser microdissection shares a common history with the development of both molecular biology and microelectronics. In the late 1960s and early 1970s, the first steps were taken when chromosomal lesions were achieved using an argon laser.1 Scientists later succeeded in dissecting tissue with precision in the range of micrometers employing a UV laser.2,3 Unfortunately, this approach required bulky equipment and was susceptible to difficulties. Besides, it was a time-consuming procedure and, therefore, did not find its way into routine biomedical investigation. Micropipettes and needles were easier to handle and were thus used manually to scrape off tissue fragments as small as 2 square millimeters.4,5 1
M. W. Berns, R. S. Olson, and D. E. Rounds, Nature 221, 74 (1969). G. Isenberg, W. Bielser, W. Meier-Ruge, and E. Remy, J. Microsc. 107, 19 (1976). 3 W. Meier-Ruge, W. Bielser, E. Remy, F. Hillenkamp, R. Nitsche, and R. Unsold, Histochem. J. 8, 387 (1976). 4 M. Palkovits, Int. Rev. Cytol. 56, 315 (1979). 5 M. Palkovits, Methods Enzymol. 103, 368 (1983). 2
METHODS IN ENZYMOLOGY, VOL. 356
Copyright 2002, Elsevier Science (USA). All rights reserved. 0076-6879/02 $35.00
[18]
LASER MICRODISSECTION IN COMPLEX TISSUE
217
When employing a micromanipulator, however, the number of cells selected could be reduced from many hundreds to fewer than 100 cells,6 a method which could equally well have been applied to isolate cells from pancreatic islets, acinar glands, and other structures from within paraffin-embedded tissue. However, this did not yet allow the study of single cells within complex tissues, let alone elements from single cells. In addition, this method was very limited in its reproducibility. Selective ultraviolet radiation fractionation (SURF) was described that same year. Chinese ink serves to protect DNA within the cell groups of interest from stray UV photons when irradiation of the whole tissue is performed to ablate nonprotected cells and tissue regions.7 This technique improved reproducibility considerably. However, it did not improve biological resolution. Microdissection methods were modified and technically improved,8–10 but remained tedious and very time-consuming. A revival of laser microdissection began with the work of Y. Kubo et al.11 and M. Emmert-Buck et al.12 when tissue areas in the micrometer range became readily dissectable. Meanwhile laser technology became more refined and powerful, and computers became available which could run sophisticated software for controlling the laser, video processing, image acquisition, analysis, and documentation. Today some countries have funded programs for developing microdissection devices (e.g., United States, Australia), and several companies from different countries offer commercial laser microdissection systems which are user friendly. Principles UV Laser Microdissection Laser beams with a wavelength within the ultraviolet (UV) spectrum of light are very powerful. Proteins, however (280 nm), and nucleic acids (230–260 nm) are the main energy absorbers resulting in tissue damage. Therefore this technique is used effectively as an optical scissors or knife. Laser spot widths within the range of nanometers are generated and are only used to create linear cuts. Inverted (SL Microtest, Germany; P.A.L.M. Microlaser Technologies AG, Germany; and others) and standard microscopes (Leica, Germany) are equipped with a robotized 6
L. Whetsell, G. Maw, N. Nadon, D. P. Ringer, and F. V. Schaefer, Oncogene 7, 2355 (1992). D. Shibata, D. Hawes, Z. H. Li, A. M. Hernandez, C. H. Spruck, and P. W. Nichols, Am. J. Pathol. 141, 539 (1992). 8 R. K¨ uppers, M. Zhao, M. L. Hansmann, and K. Rajewsky, EMBO J. 12, 4955 (1993). 9 Z. Zhuang, P. Bertheau, M. R. Emmert-Buck, L. A. Liotta, J. Gnarra, W. M. Linehan, and I. A. Lubensky, Am. J. Pathol. 146, 620 (1995). 10 J. J. Going and R. F. Lamb, J. Pathol. 179, 121 (1996). 11 Y. Kubo, F. Klimek, Y. Kikuchi, P. Bannasch, and O. Hino, Cancer Res. 55, 989 (1995). 12 M. R. Emmert-Buck, R. F. Bonner, P. D. Smith, R. F. Chuaqui, Z. Zhuang, S. R. Goldstein, R. A. Weiss, and L. A. Liotta, Science 274, 998 (1996). 7
218
SPECIALIZED USES
[18]
FIG. 1. (A) Illustrates the steps involved in laser capture microdissection (LCM). After pretreatment of the tissue on a slide, it is viewed under an inverted microscope. Step 1—A film-coated cap is placed onto the tissue. Step 2—A light spot appears which serves to aim at cells of interest. Then a laser is pulsed at the demarcated region. Step 3—This procedure is subsequently followed by the capture of the cells which stick to the cap, which is then removed and brought in contact with prepared solutions. (B) Illustrates the principle of UV laser microdissection (UV µD). Specimens are prepared on a membrane holder which is supported by a slide. On the video screen, one can draw projected laser lines with a computer to program the laser and the robotized microscope stage. Subsequently, the dissected structures can be removed mechanically with a sticky cap or they fall into microcentrifuge tubes. (C) Gives an example of tissue (adrenal) before and after laser microdissection.
stage and controlled by a computer. An image is displayed on a video screen. Specimens are placed on holders or slides which are covered with a specialized membrane and can be subjected to pretreatment in the same manner as glass slides, with applications such as histochemistry, paraffin removal, or immunhistological procedures. The cells or areas to be excised are marked on the video screen using a mouse in the same way as modern graphic programs are used to draw lines, circles, etc. Then the robotized microscope-operated laser and stage are used to cut out the marked objects as per prior computer generated circumscription (Fig. 1). The dissected tissue fragments are collected automatically in microcentrifuge vials, or else mechanically through microcentrifuge lids coated with an adhesive film that becomes activated by the laser energy. Since the samples are protected by a membrane, they can be transferred without being touched.
[18]
LASER MICRODISSECTION IN COMPLEX TISSUE
219
UV lasers are also employed as “zona knives,” for example, to dissect the zona pellucida of mammalian oocytes, or as “nano knives,” for instance, to cut the tails of sperm cells. UV laser microdissection (UV µD) can even be used to perform incisions into subcellular structures such as membranes. Thus, for example, in vitro fertilization is facilitated using UV laser microdissection techniques. Laser Capture Microdissection This method, developed at the National Cancer Institute (NCI), NIH, Bethesda, MD, with the company Arcturus, Mountain View, CA, makes use of a low energy infrared (IR) laser whose energy is not absorbed by proteins and is too low to heat water over a point at which tissue destruction could result.12 A high precision inverted microscope provides visual control over the tissue to be dissected. A sterile film-coated cap is applied to the top of the tissue, and an area of interest is chosen. A laser penetrating the cap is triggered under optical control. The film is an inert membrane and melts under heat, becoming glue-like and sticking to the tissue area on which the laser impinges (Fig. 1). Because of the low energy and the short pulse duration of the laser beam, almost no damage is done to the structures to be investigated. Nevertheless, the cells are subjected to heat as well as photons from the laser itself which may result in partial destruction of nucleic acids. After successful dissection, the cap can be removed and transferred to a 500-µl microcentrifuge tube or a special container which allows a minimal volume of appropriate lytic solution to be added to extract the material from the microdissected cells. The laser spots are so small that even single cells can be isolated from slides carrying quite complex tissue structures. This procedure is very fast and has therefore been integrated into the Cancer Genome Anatomy Project (CGAP) of the National Cancer Institute at NIH. Applications Visually controlled microdissection techniques combine the advances of anatomical and physiological approaches. The optical contrast achieved by histologic pretreatment of the specimen and dissection of the tissue under fine visual control, in conjunction with the added resolution of molecular studies with emphasis on genetics and proteomics, result in a deeply impactive augmentation of biofunctional resolution. Thus, laser microdissection affords a wide range of applications. Paraffin-embedded samples can be processed as well as frozen tissue, cell smears, and even cell culture samples. New histological procedures improve optical resolution, particularly when special membranes are employed to select and collect samples directly from cryo-microtomes. Histochemistry and immunohistochemistry can be performed prior to microdissection to mark or highlight objects of interest. Thus, the contrast between benign and malignant cells, or the
220
SPECIALIZED USES
[18]
contrast between specialized organ-specific cells and cells belonging to other tissues, e.g., nerve cells, endothelial cells, infiltrating immune cells, or cells from within complex tissues, can be enhanced. The clinical function of a whole organ results from a complex interaction between various cell types and the development of the ultrastructure within which these cells are embedded. Dependent on their age, their exposure to changes within the micromilieu, or their localization within the tissue, cells of the same type and origin may fulfill differing functions or undergo further differentiation. Adrenal chromaffin cells, for instance, may develop into a sympathetic neuron-like cell type or else into a neuroendocrine type which produces mainly catecholamines13 under the influence of adrenocortical glucocorticoids.14,15 Chromaffin cells also regulate the function of adrenocortical cells.16,17 When grown in coculture with medullary cells, adrenocortical cells increase their cortisol production as much as tenfold as compared to adrenocortical cells maintained in monoculture.18 Using microdissection systems it is now much easier to achieve molecular access to understanding these paracrine interactions. Structures can now be separated via laser knives and be subjected to downstream analytical methods without fear of contamination by surrounding tissue. This is even more important in organs built up by various types of cells which intermingle to such a great degree as in the adrenal gland. Thus, for instance, it is possible to examine the differential expression of the receptor for prolactin and the receptor for leptin within the three zones of the adrenal cortex, in the adrenal medulla, and also in tumorous adrenal tissue.19,20 In a clinical case of Cushing’s syndrome, which arose due to ectopic adrenal ACTH secretion, an adrenocortical–pituitary hybrid tumor was identified using LCM followed by combined molecular and ultrastructural analysis.21 Studies concentrating on spatial resolution can be performed as well as studies which aim at improving temporal resolution, focusing on development, regeneration, and remodeling of tissues, on cell proliferation or necrosis and apoptosis. 13
D. J. Anderson and R. Axel, Cell 47, 1079 (1986). R. J. Wurtman and J. Axelrod, Science 150, 1464 (1965). 15 D. P. Merke, G. P. Chrousos, G. Eisenhofer, M. Weise, M. F. Keil, A. D. Rogol, J. J. Van Wyk, and S. R. Bornstein, N. Engl. J. Med. 343, 1362 (2000). 16 M. Ehrhart-Bornstein, J. P. Hinson, S. R. Bornstein, W. A. Scherbaum, and G. P. Vinson, Endocr. Rev. 19, 101 (1998). 17 S. R. Bornstein, H. Tian, A. Haidan, A. Bottner, N. Hiroi, G. Eisenhofer, S. M. McCann, G. P. Chrousos, and S. Roffler-Tarlov, Proc. Natl. Acad. Sci. U.S.A. 97, 14742 (2000). 18 A. Haidan, S. R. Bornstein, A. Glasow, K. Uhlmann, C. Lubke, and M. Ehrhart-Bornstein, Endocrinology 139, 772 (1998). 19 A. Glasow, S. R. Bornstein, G. P. Chrousos, J. W. Brown, and W. A. Scherbaum, Horm. Metab. Res. 31, 247 (1999). 20 A. Glasow, A. Haidan, U. Hilbers, M. Breidert, J. Gillespie, W. A. Scherbaum, G. P. Chrousos, and S. R. Bornstein, J. Clin. Endocrinol. Metab. 83, 4459 (1998). 21 N. Hiroi, G. P. Chrousos, B. Kohn, A. Lafferty, M. Abu-Asab, S. Bonat, A. White, and S. R. Bornstein, J. Clin. Endocrinol. Metab. 86, 2631 (2001). 14
[18]
LASER MICRODISSECTION IN COMPLEX TISSUE
221
This remains one of the key foci of the CGAP project at NIH. Genetic fingerprints of various tumors at different stages of progression and with different malignant characteristics are compiled within a huge gene library and then catalogued (cgap.nci.nih.gov/).22,23 Thus, it was possible to identify the mutations and the gene responsible for the development of multiple endocrine neoplasia (MEN) type 1.24 To examine somatic mutations there is a special need to separate with high precision stromal cells from neighboring cells which derive from the endothelium, from nervous tissue, or from fibroblasts. In using preparations derived from single cells, it is possible to detect a single point mutation in the codon 12 of the c-Ki ras 2 mRNA.25 The study of identified proteins or mRNAs expressed within malignant tissue and a direct comparison of the tissue structure to the entries within the gene library resulted in the discovery of genes which are excusively expressed in prostate cancer.26 New concepts which study the aberrant expression or overexpression of proteins in context of tumor development27–29 led to the identification of ectopic receptors for cytokines30 or other trophic factors such as hormones and neuropeptides.31 Tissue infiltration by immune or metastasized cells may lead to other metabolic alterations or consequences for neighboring cells and the establishment of alternative feedback systems. It was in 1889 when Stephan Paget created the hypothesis of “seed and soil” and recognized the role of tissue that embedded the tumor in oncogenesis.32,33 A hundred years later it is now possible to have a closer look at early observed phenomena by piercing into the underlying molecular mechanisms with laser microdissection technology. 22
R. L. Strausberg, J. Pathol. 195, 31 (2001). J. W. Gillespie, M. Ahram, C. J. Best, J. I. Swalwell, D. B. Krizman, E. F. Petricoin, L. A. Liotta, and M. R. Emmert-Buck, Cancer J. 7, 32 (2001). 24 S. C. Chandrasekharappa, S. C. Guru, P. Manickam, S. E. Olufemi, F. S. Collins, M. R. EmmertBuck, L. V. Debelenko, Z. Zhuang, I. A. Lubensky, L. A. Liotta, J. S. Crabtree, Y. Wang, B. A. Roe, J. Weisemann, M. S. Boguski, S. K. Agarwal, M. B. Kester, Y. S. Kim, C. Heppner, Q. Dong, A. M. Spiegel, A. L. Burns, and S. J. Marx, Science 276, 404 (1997). 25 K. Sch¨ utze and G. Lahr, Nat. Biotechnol. 16, 737 (1998). 26 G. Vasmatzis, M. Essand, U. Brinkmann, B. Lee, and I. Pastan, Proc. Natl. Acad. Sci. U.S.A. 95, 300 (1998). 27 M. S. Katz, T. M. Kelly, E. M. Dax, M. A. Pineyro, J. S. Partilla, and R. I. Gregerman, J. Clin. Endocrinol. Metab. 60, 900 (1985). 28 A. Lacroix, E. Bolte, J. Tremblay, J. Dupre, P. Poitras, H. Fournier, J. Garon, D. Garrel, F. Bayard, R. Taillefer, R. J. Flanagan, and P. Hamet, N. Engl. J. Med. 327, 974 (1992). 29 Y. Reznik, V. Allali-Zerah, J. A. Chayvialle, R. Leroyer, P. Leymarie, G. Travert, M. C. Lebrethon, I. Budi, A. M. Balliere, and J. Mahoudeau, N. Engl. J. Med. 327, 981 (1992). 30 H. S. Willenberg, C. A. Stratakis, C. Marx, M. Ehrhart-Bornstein, G. P. Chrousos, and S. R. Bornstein, N. Engl. J. Med. 339, 27 (1998). 31 N. N’Diaye, J. Tremblay, P. Hamet, and A. Lacroix, Horm. Metab. Res. 30, 440 (1998). 32 S. Paget, Lancet 1, 571 (1889). 33 G. Poste and L. Paruch, Cancer Metast. Rev. 8, 93 (1989). 23
222
SPECIALIZED USES
[18]
FIG. 2. Schematic flow of the combination of methods possible employing laser (capture) microdissection systems.
Laser microdissection is thus not only a tool for “digging into unknown ground” and identifying factors or receptors implicated in tumorigenesis, but also capable of excluding peptides which were believed to play a major role in these processes. For example, authors were able to discount the hypothesis that mutations in the gene coding for the prolactin receptor were a pathophysiologically relevant step in the development of human breast carcinomas.34 Perspectives The classical experimental sequence—isolation of a “specific” cell, amplification of its genome, characterization of its mRNA expression profile, identification of characteristics which lead to its specialized role—is no longer a scientist’s dream today (Fig. 2). Purified preparations of nucleic acids can now be hybridized to one of thousands of gene probes on a microchip array.35 Proteins can be denatured or maintained on a slide and analyzed by applying the protein on antibody arrays that recognize selective epitopes, or else be analyzed within a matrix-assisted laser desorption device or by ionization-time-of-flight mass spectrometry devices, with many more technologies in development within the new wave of proteomics. 34
A. Glasow, L. C. Horn, S. E. Taymans, C. A. Stratakis, P. A. Kelly, U. Kohler, J. Gillespie, B. K. Vonderhaar, and S. R. Bornstein, J. Clin. Endocrinol. Metab. 86, 3826 (2001). 35 M. A. Rubin, J. Pathol. 195, 80 (2001).
[18]
LASER MICRODISSECTION IN COMPLEX TISSUE
223
Aspiration of biological material, laser microdissection of diseased structures, and subsequent analysis using modern techniques of molecular medicine represent a combination of methods which will be in routine use in the years ahead in diagnostics, development of therapeutical agents, and monitoring treatment. Summary Concomitant with the rapid development in biomedical knowledge, including the methods of molecular biology and proteomics, and the manufacture of ever more precise optical instruments, powerful lasers, and sophisticated microcomputing hardware and software, laser microdissection systems have emerged which are now entering the field of routine research. Today, several devices are commercially available, congresses devoted to the latest advances in laser microdissection are now held on regular occasions, and the number of publications based on the use of these techniques has risen to over 250. With laser microdissection, histological treatment, such as chemical or immunological fixation and staining, can readily be combined with methods suitable for molecular biology or proteomics. As the optical, technical, and methodological resolution of polymerase chain reaction (PCR) and microdissection increases, genetic and phenotypic studies of biological material are possible even at the level of single cells and subcellular elements. Moreover, questions such as the paracrine interaction of cells within complex tissues, the development of cancer, and the role of single cells in tissue remodeling or development on the microscopic and molecular level can now be addressed precisely at the molecular level. This chapter reviewed the development of laser microdissection platforms, its potential impact on the future of research, and how, in particular, these technologies can be successfully integrated into modern research and routine histopathological studies of complex tissue. Acknowledgment This work was supported by a DFG grant (441/1-1 to S.R.B). We would like to express our thanks to Dr. J. Gillespie and the NIH LCM Lab for technical support and advice.
224
[19]
SPECIALIZED USES
[19] Assessment of Clonal Relationships in Malignant Lymphomas By KOJO S. J. ELENITOBA-JOHNSON Introduction The recent development of new technologies that provide the opportunity for direct molecular examination of pathologic cells and tissues has dramatically transformed investigative studies in research and diagnostic medicine. These techniques permit exploration and analysis of molecular and cellular aberrations as well as of mechanisms of development of disease processes. One technology that has found widespread utilization for the isolation of specific cell populations is laser capture microdissection (LCM).1 LCM is a recently developed technology that involves microscopic visualization and isolation of defined cell populations from tissue sections. The histologic area of interest on the tissue section is visualized and directly overlaid by a specialized cap on which a transparent ethylene vinyl acetate thermoplastic film is coated. The thermoplastic polymer film contains infrared (IR) absorbing dyes that permit utilization with near-IR gallium arsenide laser diodes, which have been incorporated into routine microscopes. Operatordependent pulsed laser activation effects transient melting of the thermoplastic polymer in the vicinity of the targeted cells. The cells are captured within the film and can be retrieved in aqueous solutions for subsequent experimental manipulation. The commercially available versions of the LCM instruments permit modulation of the spot size and laser beam intensity. The transient temperature transitions encountered in the process are innocuous to nucleic acids in the tissues, and DNA and RNA can be easily recovered for utilization in experiments. The unique property of lymphoid cells to undergo specific rearrangements of their antigen receptor genes (ARG) within a virtually infinite repertoire of combinatorial diversity provides a convenient modality of assessment of clonality. The assessment of clonal relationships between two or more putative neoplastic populations is a useful strategy in the determination of the origin or the relationships between anatomically or temporally separate populations. The advent of LCM has dramatically improved the ability to assess clonal relationships, by permitting tumor cell purification or enrichment, and in some cases, isolation of single cells at the histological level. Once tumor cell enrichment or isolation of pure tumor populations is accomplished, it is feasible to evaluate clonal relationships by examining for genetic markers characteristic of the particular neoplasm of interest. 1
M. R. Emmert-Buck, R. F. Bonner, P. D. Smith, R. F. Chuaqui, Z. Zhuang, S.R. Goldstein, R. A. Weiss, and L. A. Liotta, Science 274, 998 (1996).
METHODS IN ENZYMOLOGY, VOL. 356
Copyright 2002, Elsevier Science (USA). All rights reserved. 0076-6879/02 $35.00
[19]
CLONAL RELATIONSHIPS IN MALIGNANT LYMPHOMAS
225
The ability to directly assess lymphoid populations in this fashion has had important implications for the primary distinction of malignant from benign processes, molecular staging, and monitoring of minimal residual disease following therapy. In this chapter, a number of strategies and protocols that have been successfully implemented in our laboratory for the establishment of clonal relationships between two microdissected lymphoid populations will be described. Amplification of the ARG in benign lymphoid populations yields a polyclonal smear or ladder pattern on gel electrophoresis indicating the presence of recombined antigen receptor genes of different lengths. On the other hand, a malignant population arising by clonal expansion of a neoplastically transformed lymphoid cell exhibits only one or two specific gene rearrangements. This antigen receptor rearrangement is unique to the founder neoplastic cell and its progeny and thus polymerase chain reaction (PCR) analysis of ARG in a monoclonal population yields either a single or two prominent bands. This unique rearrangement serves as a marker for this particular neoplasm and can be used as a parameter for the establishment of a clonal relationship between this and another lymphoid population. In this regard, the reader will note that all of the examples illustrated in this section relate to B-lymphoid neoplasms. Similar principles apply to T-cell processes in which the T-cell receptor γ or β genes may be amplified for the establishment of clonal relationships between two or more microdissected T-cell populations. While the strategies and protocols discussed here are by no means exhaustive, they provide a framework for the design of similar assays based on parallel fundamental concepts. Protocols Preparation of Fixed Paraffin-Embedded Tissue Sections for Laser Capture Microdissection Unstained 5 µm thick sections are deparaffinized by dipping into xylene, rinsed twice with 95% ethanol, briefly stained with eosin, and air-dried. Preparation of OCT-Embedded Frozen Tissue Sections for Laser Capture Microdissection Five-micron sections are cut from OCT embedded snap-frozen tissue blocks using a cryostat. The sections are mounted on poly-L-lysine precoated glass slides (C to C Laboratory Supplies, Chicago, IL). The frozen sections are thawed at room temperature for 60 sec and submerged in either cold acetone (5 min), methanol (5 min), 4% paraformaldehyde (5 min), 70% ethanol (30 sec), ethanol/formalin (3 : 1, 1 min), and ethanol 70% (15 sec) followed by acetone (5 min). The slides are then briefly rinsed in phosphate buffered saline (PBS), pH 7.4, and allowed to dry.
226
SPECIALIZED USES
[19]
Immunohistochemical Staining for Laser Capture Microdissection Immunohistochemical staining is performed using a three-step streptavidin– biotin technique with monoclonal or polyclonal (rabbit anti-human) antibodies against the appropriate antigen that is specific for the intended cell population, and the DAKO Quick Staining kit (DAKO Corp., Carpinteria, CA). The slides are incubated at room temperature with the primary and secondary antibodies, and the tertiary reagent for 90 sec each, and rinsed with 1 × PBS between each step. Color development is performed using diaminobenzidine (DAB) (3 min) as the chromogen. The sections are counterstained with hematoxylin for 15 sec (optional), serially dehydrated through graded alcohols (3 dips in each alcohol solution), and finally immersed in xylene before air-drying at room temperature. After drying, the immunohistochemically stained section is ready to be utilized for LCM. Laser Capture Microdissection Laser capture microdissection is performed using a PixCell laser capture microscope (Arcturus Engineering, Santa Clara, CA). Briefly, 5 µm hematoxylin and eosin stained tissue sections obtained from fixed paraffin-embedded or frozen OCT-embedded tissues are placed under the microscope and the area of interest is focused between the optical cross hairs visible in an eyepiece lens. The optically clear plastic caps bearing the thermoplastic membranes (which effect the capture of cells from the glass slides) are loaded onto the transporting arm attached to the microscope. The transporting arm with the loaded plastic cap is swiveled across to overlay the plastic cap on the area of interest on the glass slide (Arcturus Engineering, Santa Clara, CA). Laser activation of the film leads to focal melting at the precise spot of activation with attendant capture of the underlying visually selected cells (Fig. 1). The cell-laden cap may now be utilized as the cover for the Eppendorf tube that contains a previously aliquoted volume (∼20 µl) of the digestion buffer. The tube is inverted gently and repeatedly to immerse the microdissected cells within the digestion buffer. Brief pulses of vortexing may be performed to facilitate dissolution of the cells into the digestion buffer. DNA Extraction Approximately 500 cells from each biopsy specimen are procured by microdissection and immediately transferred into a 20 µl digestion buffer solution containing 0.05 mol/liter Tris-HCL, 0.001 mol/liter EDTA, 1% Tween 20, and 0.1 mg/ml proteinase K (pH 8.0) and incubated overnight at 37◦ . Alternatively, microdissectates may be incubated at 55◦ for 3–5 hr. In either case, the mixture is heated to 95◦ for 10 min to inactivate proteinase K, and 1.0 µl of this solution is used as the template for PCR. Further purification of the DNA can be achieved
[19]
CLONAL RELATIONSHIPS IN MALIGNANT LYMPHOMAS
227
FIG. 1. Illustration of microdissection of two histologically distinct lymphoid populations. (A) 100 × (original) magnification of a hematoxylin and eosin stained section showing follicular lymphoma with attendant monocytoid B-cell proliferation before microdissection. (B) After microdissection of both follicular and monocytoid B-cell components.
in some cases by pursuing the additional steps of phenol–chloroform extraction and ethanol precipitation. In our hands, we have found that the crude proteinase K digests yield adequate DNA material for PCR. The integrity of the DNA from the microdissected sample may be established by successful PCR of a segment of a housekeeping gene (e.g., human β-globin). During the configuration of the housekeeping gene PCR, the product size should be comparable to or greater than that of the intended gene of interest. We have successfully verified the integrity of isolated DNA from frozen and paraffin-embedded tissue material using primers from the human hemoglobin beta chain gene.2 These primers yield 123 bp and 268 bp PCR products and can be readily visualized by ultraviolet transillumination of ethidium bromide-stained (0.2 µg/ml) 1.5% NuSieve agarose gels (Nusieve 3 : 1 Agarose blend, FMC BioProducts, Rockland, ME). On occasion, the 123 bp may be present, but the 268 bp fragment may absent. This is interpreted as evidence that the extracted DNA is adequate for amplifying only smaller DNA fragments. For instance, such a sample would be considered inadequate for immunoglobulin 2
G. Wu, T. C. Greiner, and W. C. Chang, Diagn. Mol. Pathol. 6, 147 (1997).
228
SPECIALIZED USES
[19]
heavy chain (IgH) PCR using framework II primers which typically yield products of approximately 250 base pairs. Assessment of Clonal Relationships by Polymerase Chain Reaction Analysis of Immunoglobulin Heavy Chain Gene Polymerase chain reaction assays to detect IgH gene rearrangements exploit conserved sequences within the variable and joining regions of the IgH gene that are located several kilobases apart in the germ-line configuration. In B cells that have undergone IgH gene rearrangement, these variable and joining regions are brought into close proximity. No product is amplified in nonlymphoid cells, whereas B cells yield short PCR products whose length and sequence composition is determined by the nature of their recombined variable-diversity-joining (V-D-J) regions and by their unique N (nontemplated nucleotide) region sequences. Importantly, because of the tremendous repertoire for diversity present at the antigen receptor loci, simultaneously tested lymphoid samples with IgH PCR products of identical sequence may be presumed to be clonally identical. Although the unique IgH complementarity determining region (CDR) sequence of a monoclonal lymphoid population is the gold standard for assessing clonal relatedness, it is also reasonable to surmise that samples demonstrating IgH PCR products of identical size are clonally related (Fig. 2). A similar interpretation holds for T-cell receptor γ chain gene PCR assays for the establishment clonal relationships in T-cell populations. IgH PCR For IgH heavy chain PCR, crude extract (1 or 2 µl) from microdissected tissue and 1 µl each of 50 ng/µl positive and negative controls are used as the template for amplification. PCR is performed in a PerkinElmer 2400 thermal cycler (PerkinElmer, Norwalk, CT). Target DNA is amplified in a 20 µl reaction containing GenAmp PCR buffer, 1.5 mM MgCl2, four deoxynucleotide triphosphates at 200 µM, primers at 0.5 µM, and 1 unit of AmpliTaq DNA Polymerase (PerkinElmer). Labeled reactions include a 32P dCTP, and the concentration of cold nucleotides is reduced (50 µM dATP, dGTP, and dTTP; 10 µM dCTP). A hot start technique is employed using TaqStart Antibody (Clontech, Palo Alto, CA), and immunoglobulin heavy chain variable region (VH-FRIII); 5 -ACA CGG C[C/T] T[G/A]T ATT ACT GT-3 ) and joining region (JH; 5 -ACC TGA GGA GAC GGT GAC C-3 ) consensus primers. The amplification protocol consists of 35 cycles of denaturation at 96◦ for 1 min, annealing at 56◦ for 1 min, and extension at 74◦ for 1 min. Amplification products may be analyzed on ethidium bromide stained standard polyacrylamide or sequencing gels, with molecular weight standards. Radiolabeled products are best analyzed on 8% sequencing gels, dried and autoradiographed. One or two dominant bands indicate the presence of a monoclonal
[19]
CLONAL RELATIONSHIPS IN MALIGNANT LYMPHOMAS
229
FIG. 2. Immunoglobulin heavy chain PCR from two microdissected lymphoid populations analyzed on a 10% polyacrylamide gel. Lane 1 corresponds to the H2O control. Lane 2 corresponds to the positive control (Raji cell line). Lanes 3–4 represent duplicate IgH PCR amplifications from one microdissected population, and lanes 5–6 contain products from the other component. There is comigration of monoclonal bands indicating that the two different lymphoid populations are clonally identical.
population, whereas a polyclonal population results in either a smear (standard acrylamide gel) or a ladder of bands of nearly equal intensity (sequencing gel). Replicates of all PCR-based gene rearrangement analyses of microdissected samples should be run parallel in order ensure reproducibility and to avoid erroneous scoring of monoclonality as a result of low lymphocyte numbers.3 Isolation of Monoclonal Bands for Direct DNA Sequencing Amplicons resulting from IgH PCR can be subjected to electrophoresis using ethidium bromide-stained 1.5% agarose gels, and the band of expected size excised, dissolved in distilled H2O, purified, and concentrated using Amicon DNA Extraction columns (Millipore, Bedford, MA), according to manufacturer’s instructions. A denaturation gel-based approach has been suggested by Wu et al.,2 which facilitates ready distinction of monoclonal bands. This method requires the utilization of GC-clamped IgH V region primers. The GC clamp portions of the GC-clamped amplicons derived from the IgH PCR may be used as priming sites for direct DNA sequencing. 3
K. S. Elenitoba-Johnson, S. D. Bohling, R. S. Mitchell, M. S. Brown, and R. S. Robetorye, J. Mol. Diagn. 2, 92 (2000).
230
SPECIALIZED USES
[19]
Direct Purification of PCR Amplicons for Cycle Sequencing The generation of optimal DNA sequence requires elimination of excess primers and dNTPs that remain following PCR amplification. We have utilized the QIAquick PCR purification kit (Qiagen, Valencia, CA) with good success. The purification is based on column technology that utilizes the selective binding properties of silica-gel membranes. The silica membrane adsorbs DNA in the presence of high concentrations of chaotropic salts and optimal pH. Adsorption is typically 95% at pH ≤ 7.5. The purified DNA is eluted with distilled water or Tris buffer. We have observed that lower salt concentrations and basic conditions (pH 7.5–8.5) generally enhance elution efficiency. Ten replicates of the sample of interest are amplified and pooled. Five volumes of Buffer PB are added to 1 volume of the pooled PCR product. The samples are the applied to the wells of the QIAquick strips and the vacuum source is switched on. After all the liquid has been suctioned through the column, the vacuum is switched off and the QIAquick 6S is vented by releasing the tubing connecting the unit to the vacuum. The wells of the QIAquick strips are washed by adding 1 ml of the Buffer PE to each well and then the vacuum switched on. For optimum washing, the last two steps are repeated with 2 washes of Buffer PE. After Buffer PE has been drawn through the columns, the membrane is dried by applying the maximal vacuum for an additional 5 min. After this step, the vacuum is switched off and the QIAquick 6S is ventilated. The top plate is disengaged from the base and the nozzles of the QIAquick strips are blotted with clean Kimwipes. The DNA is eluted by adding 75–100 µl of Buffer EB (10 mM Tris-HCl, pH 8.5) directly to the center of each membrane in the QIAquick strips and allowed to stand for 1 min. The vacuum is switched on for 2 min to effect suction through the QIAvac 6S. Samples in which sequencing is to be performed in the short term (≤6 hours) should be stored at 4◦ until needed. Samples in which sequencing is anticipated in the long term should be stored at −20◦ . Our protocol for cycle sequencing requires that the purified DNA sample to be analyzed be constituted at 30–90 ng (in a maximum volume of 8 µl) for each cycle sequencing reaction. The primers to be used for the sequencing reaction are diluted to 0.8 pmol/ml with distilled H2O. DNA Sequencing of IgH Chain Gene PCR Products As has been stated previously, the CDRIII of the IgH is a highly variable region that can be used as a specific marker for a B-lymphoid neoplasm.4 Improvements and automation of DNA sequencing procedures now permit ready sequencing of shorter DNA fragments as may be encountered in IgH framework III PCR assays 4
M. Yamada, S. Hudson, O. Tournay, S. Bittenbender, S. S. Shane, B. Lange, Y. Tsujimoto, A. J. Caton, and G. Rovera, Proc. Natl. Acad. Sci. U.S.A. 86, 5123 (1989).
[19]
CLONAL RELATIONSHIPS IN MALIGNANT LYMPHOMAS
231
where the product size may range from 80 to 120 bp. Following purification and dilution of the PCR product to the appropriate concentration, the internal J-region primer (5 CTG TCG ACA CGG CCG TGT ATT ACT G 3 ) can be used for direct sequencing. An alternative approach involves the usage of nested IgH VH- and JH region primers to which a segment of the M13 vector sequence have been ligated.2 The procedure entails utilization of an appropriate dilution of 1–5 µl of the original IgH PCR amplicon (after purification) as the template in the subsequent nested PCR. The IgH PCR products from this reaction may then be gel-purified or directly purified in solution and subjected to direct sequencing using primers complementary to M13 by a cycle sequencing procedure with fluorescently labeled dideoxynucleotide terminators (Applied Biosystems, Foster City, CA).
Establishment of Clonal Identity Using Clone-Specific Primers or Internal Probes As previously discussed, the establishment of clonal identity between two lymphoproliferative processes indicates that they have arisen from the same founder clone. The marked diversity of the V-D-J combinatorial repertoire and the uniqueness of the junctional “N-DH-N” sequences permit the design of clone-specific oligonucleotide primers or probes as specific markers of a B-cell clone and its progeny. In this section, we describe the utilization of clone-specific PCR to establish a clonal relationship between two histologically distinct samples. This can be done using conventional or real-time PCR. In both assays, establishment of clonality and sequencing of the monoclonal IgH PCR product are necessary in an initial specimen, to determine the unique IgH sequence of the neoplastic population. Once established, unique portions of this sequence may be utilized for the design of primers or internal oligonucleotide probes for PCR amplification and assessment of clonal identity with other (microdissected) populations. For the conventional PCR and gel electrophoresis, the PCR products resulting from amplification reactions using primers specific to the unique CDRIII sequences can be separated using polyacrylamide, sequencing, or even metaphor agarose gels. Monoclonal bands of appropriate size comigrating in the samples tested are indications that they are clonally related (Fig. 3). The real-time PCR assay for establishment of clonal identity may be configured using any one of several fluorescence-based formats including linear hybridization,5 exonuclease (TaqMan),6 or hairpin (Molecular Beacon)7 probes specific 5
K. S. Elenitoba-Johnson, S. D. Bohling, C. T. Wittwer, and T. C. King, Nat. Med. 7, 249 (2001). P. M. Holland, R. D. Abramson, R. Watson, and D. H. Gelfand, Proc. Natl. Acad. Sci. U.S.A. 88, 7276 (1991). 7 S. Tyagi and F. R. Kramer, Nat. Biotechnol. 14, 303 (1996). 6
232
SPECIALIZED USES
[19]
FIG. 3. Immunoglobulin heavy chain PCR using clone-specific primers and analyzed on a 10% polyacrylamide gel. IgH PCR was performed on one of the lymphoid populations and sequenced. Following sequence determination, PCR primers complementary to the specific IgH V and J regions are utilized for amplification of both microdissected populations. Lane 1 contains the H2O control. Lane 2 containing DNA from the first microdissected sample shows a monoclonal band of approximately 180 bp in size. Lane 3 containing DNA from the other microdissected population shows a monoclonal band of identical size with the initial lymphoid population. This indicates that both lymphoid populations are clonally related. Lane 4 contains the size marker.
to the junctional or specific CDRIII sequences present in the initial specimen. The internal probe-based approach utilizes fluorescently labeled probes that are perfectly complementary to the junctional CDRIII sequence of the monoclonal population of interest. The demonstration of fluorescence resonance energy transfer (FRET), and hence amplification is indicative of the presence of the unique CDRIII sequence identified in the original population. We have had the most experience with the dual linear hybridization probe format in which one oligonucleotide probe is labeled with a donor dye (e.g., fluorescein) and the other is labeled with an acceptor fluorophore (e.g., LC Red 640).5,8 Hybridization of the probes adjacent to one another on the target strand leads to the generation of FRET and an exponential increase in fluorescence that is detected by the fluorimeter when amplification has occurred (Fig. 4). The demonstration of amplification evidenced by exponential increase in fluorescence in earlier PCR cycles is indicative of sequence identity and hence clonal identity. Conversely, the absence of an exponential phase in fluorescence in the fluorescence (F ) vs time (T ) graphs indicates that the junction specific oligonucleotide probes do not hybridize adjacent to one another on the IgH sequence in question. This represents sufficient evidence for the absence of clonal identity in the lymphoid populations being evaluated. 8
C. T. Wittwer, M. G. Herrmann, A. A. Moss, and R. P. Rasmussen, BioTechniques 22, 130 (1997).
[19]
CLONAL RELATIONSHIPS IN MALIGNANT LYMPHOMAS
233
FIG. 4. (A) Schematic representation of CDRIII specific hybridization probe-based fluorescence PCR for determination of clonality. The CDRIII sequence of interest is a 41 base pair sequence which includes the unique N-regions flanking the DH region sequence. The FITC-labeled probe is complementary to the DH sequence and the LCRed 640-labeled probe is complementary to the N-region sequence. The external primers (arrows) are complementary to the specific V and J regions identified by sequencing the initial specimen. (B) Amplification curves show exponential increase in fluorescence at approximately 25 cycles for microdissected sample 1 (positive 1) and microdissected sample 2 (positive #2). This finding indicates that the two sequences share the same unique CDRIII sequence and hence are clonally identical. By contrast, the unrelated sample (negative) does not show amplification and yields a profile similar to that of H2O which is flat.
Assessment of Clonal Relationships by Polymerase Chain Reaction Analysis of Chromosomal Translocations Nonrandom chromosomal translocations represent rearrangement processes that occur in abnormal cells. During the translocation process, a segment of one chromosome is juxtaposed to another, with resulting deregulation of one or more
234
SPECIALIZED USES
[19]
[19]
CLONAL RELATIONSHIPS IN MALIGNANT LYMPHOMAS
235
genes located within the translocated segment. Mature lymphoid neoplasms often contain balanced reciprocal translocations in which there is no net loss of genetic material. The DNA fusions arising from such aberrations can be detected using molecular techniques such as Southern blot hybridization analysis or PCR. Because the DNA fragment size in the translocations within each individual tumor sample is typically unique, the presence or absence of the translocation can be used to infer clonal relatedness within samples, and furthermore, translocation fragments that generate PCR products of identical size within different samples have been considered identical (Fig. 5A).9 The melting temperature (Tm) of a fragment of DNA also provides another modality for the assessment of clonal relationships between two populations carrying the same chromosomal translocation. The Tm of a dsDNA fragment is defined as the temperature at which 50% of the strands polynucleotide duplex are dissociated.10 Since the Tm of a particular segment of dsDNA is related to its specific DNA sequence, GC content, and length, deductions about similarities between DNA sequences of identical length can be made based on their Tms.9 We have employed this strategy to establish clonal relationships between two microdissected populations, based on the melting temperature of MBR/JH translocation products that were identified in two histologically distinct but molecularly identical populations in an single biopsy specimen.9 DNA sequence analysis does provide a veritable means to assess sequence identity in equivocal cases. In this chapter, only the PCR-based methodologies for detection of a specific chromosomal translocation and the establishment of clonal relationships between two separate lymphoid 9
R. S. Robetorye, S. D. Bohling, L. J. Medeiros, and K. S. Elenitoba-Johnson, Lab. Invest. 80, 1593 (2000). 10 J. Wetmur, in “Molecular Biology and Biotechnology: A Comprehensive Desk Reference” (R. Myers, ed.), p. 605. VCH Publishers, New York, 1995.
FIG. 5. Determination of clonal relationship using bcl-2/JH translocation (major breakpoint cluster region) PCR. (A) 2% agarose gel electrophoresis of MBR/JH PCR products from the microdissected follicular lymphoma and the monocytoid B-cell cell proliferation in Fig. 1. Lane 1 shows the positive control (SUDHL-6 cell line). Lane 2 shows MBR/JH product from the follicular lymphoma and lane 3 shows a PCR product of identical size from the monocytoid B-cell population. This result indicates that the follicular lymphoma and monocytoid B-cell proliferation in this case are clonally related. Lane 4 shows a band of different size in an unrelated sample from another patient included as an additional control. Lane 5 shows the H2O control with no amplification product. (B) Fluorescence PCR for the bcl-2 MBR/JH translocation product using SYBR Green I. The sample showing a melting peak at Tm 86.8◦ is the positive control SUDHL-6 known to contain the t(14;18) translocation. PCR products from the samples described in panel A show melting profiles with an identical Tm (87.7◦ ). MBR/JH products with identical molecular weight size and identical Tm are considered clonally identical. (C) The MBR/JH product from the unrelated sample described in (A), in addition to being of different length, also shows a different Tm (88.3◦ ) confirming the lack of clonal relationship.
236
SPECIALIZED USES
[19]
processes with or without the translocation are described. Although the examples described in this text involve the detection of chromosomal translocation products that are characteristic of a specific form of non-Hodgkin’s lymphoma (i.e., follicular lymphoma), the principle is applicable to a variety of molecular diagnostic scenarios in which the clonal relationship between two or more cell populations is sought to be established. Gel Electrophoretic Determination of Translocation Product Size Agarose gel electrophoresis is often adequate for the demonstration of translocation products of identical size. Greater resolution may be achieved using gel electrophoretic systems such as polyacrylamide, single-strand polymorphism, and denaturation gradient gels which are capable of higher resolution and more accurate results. Regardless of the gel electrophoretic methodology employed, it is recommended that the products to be compared be run side by side in adjacent lanes to facilitate size comparison, and equal volumes of PCR product from each microdissected sample be utilized (Fig. 5A). Establishment of Clonal Relationships by Fluorescence Melting Curve Analysis of PCR Products for Chromosomal Translocations In this section, the utilization of the t(14;18) is described as a marker of clonal relatedness. The t(14;18) is characteristic of a form of non-Hodgkin’s lymphoma known as follicular lymphoma.11 Polymerase chain reaction amplification for detection of the t(14;18) which juxtaposes the bcl-2 gene on band 18q21 with the immunoglobulin heavy chain (JH) locus at 14q32 yields varying sized bands depending on the precise site of the breakpoint at the bcl-2 locus. The majority of the breakpoints on the bcl-2 locus are located in the major breakpoint region (MBR).12 Rapid cycle PCR for the MBR/JH is performed with oligonucleotide primers specific for the major breakpoint region of the bcl-2 gene (5 GAG TTG CTT TAC GTG GCC TG 3 ) and the JHa region of the immunoglobulin heavy chain joining region (5 ACC TGA GGA GAC GGT GAC C3 ) in a thermal cycler integrated with a fluorimeter (LightCycler, Roche Molecular Biochemicals, Indianapolis, IN).13 Fifty to 100 ng of DNA is amplified in a 10- to 20-µl reaction in glass capillary tubes containing 50 mmol/l Tris (pH 8.5), 3.0 mmol/l MgCl2, four deoxynucleotide triphosphates at 200 µmol/l each, primers at 0.5 µmol/l, and 0.4 U of Taq DNA 11
Y. Tsujimoto, M. M. Bashir, I. Givol, J. Cossman, E. Jaffe, and C. M. Croce, Proc. Natl. Acad. Sci. U.S.A. 84, 1329 (1987). 12 M. Crescenzi, M. Seto, G. P. Herzig, P. D. Weiss, R. C. Griffith, and S. J. Korsmeyer, Proc. Natl. Acad. Sci. U.S.A. 85, 4869 (1988). 13 C. T. Wittwer, K. M. Ririe, R. V. Andrew, D. A. David, R. A. Gundry, and U. J. Balis, BioTechniques 22, 176 (1997).
[19]
CLONAL RELATIONSHIPS IN MALIGNANT LYMPHOMAS
237
polymerase (Promega, Madison, WI) with 11 ng/µl TaqStart antibody (ClonTech, Palo Alto, CA) per 10-µl sample. The PCR reaction mixture includes the doublestranded DNA binding dye SYBR Green I at a 1 : 30,000 dilution (Molecular Probes, Eugene, OR). A hot start technique is employed with most applications, with an initial denaturation at 94◦ for 20 sec. The thermal cycling protocol entails 45 cycles of denaturation (94◦ for 10 sec), annealing (68◦ for 0 sec), and extension (74◦ for 20 sec). Postamplification, the samples of interest are cooled to 45◦ and then heated slowly at 0.2◦ /sec until 95◦ . In the LightCycler, data acquisition is achieved using Labview graphical programming language (National Instruments, Austin, TX). Fluorescence signals are obtained once per cycle by sequential monitoring of each tube for 70 msec at the end of extension. In addition to the test samples of interest, all experimental runs should include a negative DNA control sample (e.g., placental DNA, microdissected hyperplastic tonsil), a positive control [t(14;18) MBR/JH positive cell line DNA, e.g., SUDHL-6], and a control in which a DNA template is omitted (H2O). Fluorescence signals are collected continuously from 72◦ to 95◦ to monitor the fluorescence signals derived from the dsDNA-specific dye SYBR Green I. Fluorescence vs cycle number graphs provide a graphical depiction of the occurrence (or lack thereof ) of amplification in real time. As PCR progresses, product synthesis and accumulation is observed as an increase in fluorescence with progressive cycles of amplification. The postamplification melting protocol gradually converts dsDNA into single-stranded species. When this occurs, the interaction between the dsDNA binding dye (SYBR Green I) and the DNA is lost at the melting temperature (Tm) of the DNA fragment. Fluorescence melting curves are generated from fluorescence versus temperature (F vs T ) plots. In the F vs T representation of the melting profile of a DNA fragment, a sharp decline in fluorescence corresponding to the Tm of the dsDNA fragment is identified at a specific temperature for that fragment. For easier visualization, the melting profiles are best represented as fluorescence melting peaks which are derived from the F vs T graphs by plotting the negative derivative of fluorescence over temperature vs temperature (−dF/dT vs T ).14 Using this methodology, we have previously inferred clonal identity based on the demonstration of MBR/JH translocation products with identical size and Tms in two separate microdissected lymphoid populations (follicular lymphoma and monocytoid B-cell proliferation) within the same biopsy specimen9 (Figs. 5B,C). Establishment of Clonal Relatedness by Identification of Identical Gene Mutations in Microdissected Samples DNA sequencing remains the most veritable method for establishing the presence of genetic mutations. In tumor samples in which mutations of specific 14
S. D. Bohling, T. C. King, C. T. Wittwer, and K. S. Elenitoba-Johnson, Am. J. Pathol. 154, 97 (1999).
238
SPECIALIZED USES
[19]
FIG. 6. Single-strand conformational polymorphism (SSCP) analysis. DNA isolated from a microdissected population was amplified using primers specific for exon 5 of the p53 gene. Lane 1 shows the wild-type pattern. Lane 3 represents the H2O control. Lanes 2 and 4 contain p53 exon 5 amplicon from one microdissected sample. Lanes 5 and 6 contain p53 amplicon from another microdissected population. The reactions were run in duplicate to verify reproducibility. The arrows indicate that the anomalous conformer (mutant allele) comigrates in both tumor populations, providing presumptive evidence that they contain the same p53 mutation and hence are clonally related.
oncogenes or tumor suppressor genes have been previously demonstrated, identification of the same mutation in a synchronous or metachronous sample (microdissected) within the same individual establishes that both tumors are clonally related. Indirect evidence of such a relationship may be obtained by the demonstration of identical migration of non-germ-line bands using length-independent gel-based methods for mutation screening. These include single-strand conformation polymorphism (SSCP) analysis15 and denaturation gradient gel electrophoresis.16 Here we briefly describe a PCR-SSCP protocol for detection of mutations in the p53 tumor suppressor gene that we have successfully implemented in our laboratory for the establishment of a clonal relationship between microdissected lymphoma samples (Fig. 6). Immunohistochemical studies to detect the presence of p53 overexpression may be helpful as an initial rapid screening test for the presence of p53 mutations, before molecular analysis is performed.17
15
M. Orita, H. Iwahana, H. Kanazawa, K. Hayashi, and T. Sekiya, Proc. Natl. Acad. Sci. U.S.A. 86, 2766 (1989). 16 S. G. Fischer and L. S. Lerman, Proc. Natl. Acad. Sci. U.S.A. 80, 1579 (1983). 17 K. S. Elenitoba-Johnson, L. J. Medeiros, J. Khorsand, and T. C. King, Am. J. Clin. Pathol. 106, 728 (1996).
[19]
CLONAL RELATIONSHIPS IN MALIGNANT LYMPHOMAS
239
Polymerase Chain Reaction-Single Strand Conformation and Polymorphism Analysis for Mutations of p53 Gene We have used oligonucleotide primers as previously described for amplification of exons 5–8 of the p53 tumor suppressor gene: p53 exon 5 sense primer-5 TTC CTC TTC CTG CAG TAC TC-3 ; p53 exon 5 antisense primer 5 - ACC CTG GGC AAC CAG CCC TGT-3 ; p53 exon 6 sense primer 5 - ACA GGG CTG GTT GCC CAG GGT-3 ; p53 exon 6 anti-sense primer 5 -AGT TGC AAA CCA GAA CCT CAG-3 ; p53 exon 7 sense primer 5 -GTG TTG TCT CCT AGG TTG GC-3 ; p53 exon 6 antisense primer 5 -GTC AGA GGC AAG CAG AGG CT-3 ; p53 exon 8 sense primer 5 - TAT CCT GAG TAG TGG TAA TC-3 ; exon 8 antisense primer 5 -AAG TGA ATC TGA GGC ATA AC -3 .18 Ten µl of amplicons from p53 PCR are mixed with 45 µl of 95% formamide, 20 mM EDTA, 0.05% bromphenol blue, and 0.05% xylene cyanol. The entire mixture is then heated at 94◦ for 3 min and rapidly chilled in an ice-water bath, and 5–10 µl is immediately loaded onto an SSCP gel [0.25× MDE (AT Biochem, Malvern, PA) cast in a Bio-Rad (Richmond, CA) sequencing gel apparatus]. Electrophoresis is performed in 0.7× TBE, at 8–10 watts constant power, for 5–16 hr at room temperature. Note that run times and conditions may vary for optimal resolution of different segments of the same gene.17 The gels are transferred onto filter paper (3MM Whatman, Maidstone, UK), dried, and autoradiographed at −70◦ for 12–24 hr using Kodak (Rochester, NY) XAR or XP X-ray film. The samples in question are run simultaneously in adjacent lanes, and the migratory characteristics of the wild-type conformers are examined in comparison with the “mutant” conformers. Depending on how the SSCP conditions are optimized, it may be possible to distinguish all conformers from each strand of each allele in the amplified fragment of DNA of the sample with the wild-type sequence. Anomalous mobility shifts indicative of the presence of mutations are identified by being absent in the wild-type sample. Comigration of such anomalous bands in samples run in parallel establishes the presence of a mutation in the microdissected samples. Summary DNA sequencing of the antigen receptor genes remains the gold standard for the establishment of clonal relationships between samples. However, a variety of strategies may be employed as surrogates for the determination of the actual sequence of the clonally rearranged antigen receptor genes. The methods described in this chapter provide a framework for the rapid determination of clonal relationships between (microdissected) lymphoid populations. All of the methods described are PCR-based because of its versatility and ability to utilize very small 18
A. Y. Matsushima, E. Cesarman, A. Chadburn, and D. M. Knowles, Am. J. Pathol. 144, 573 (1994).
240
[20]
SPECIALIZED USES
amounts of DNA. For illustration purposes, the descriptions have been confined to B-cell populations. Although not described here, Igκ or Igλ PCR may also be utilized for determination of B-cell clonality and clonal relationships in the same manner. Similarly, the principles utilized may be extended to T-cell populations and T-cell receptor chain genes. Regardless of the methodology or targets involved, it is strongly recommended that all assays on microdissected material be run on parallel replicates of each sample to ensure reproducibility of results. The information about clonal relationships obtained by LCM has more than an academic significance and has utility in routine diagnostics for the establishment of minimal residual disease and the determination of microscopic disease recurrence vs the development of a secondary malignancy. Acknowledgment This work was supported by CA83984 (NIH) and the ARUP Institute for Clinical and Experimental Pathology.
[20] Comparison of Normal and Tumor Cells by Laser Capture Microdissection By JAUME MORA, MUZAFFAR AKRAM, and WILLIAM L. GERALD Introduction Most tissues including neoplasms are composed of a complex, heterogeneous mixture of cells and extracellular material. This creates inherent difficulties in analysis of pure populations of cells. In the study of neoplastic disease it is critically important to accurately identify and purify individual cell components within a tumor. For example, a single region of tumor tissue may contain several types of normal epithelial and stromal cells, preinvasive and invasive neoplastic cells, and reactive stromal and inflammatory cells. Laser capture microdissection (LCM) has the advantage of relying on standard histologic or phenotypic criteria to guide precise separation and procurement of the individual cellular components providing very pure populations from heterogeneous tissues while preserving macromolecules.1 This technique incorporates microscopic visualization and targeting of sample tissue with a user friendly method for cell capture and has been used extensively 1
M. R. Emmert-Buck, R. F. Bonner, P. D. Smith, R. F. Chuaqui, Z. Zhuang, S. R. Goldstein, R. A. Weiss, and L. A. Liotta, Science 274, 998 (1996).
METHODS IN ENZYMOLOGY, VOL. 356
Copyright 2002, Elsevier Science (USA). All rights reserved. 0076-6879/02 $35.00
[20]
LASER CAPTURE MICRODISSECTION
241
for investigation of neoplastic disease. This chapter briefly describes useful methods for LCM in the analysis of tumors and nonneoplastic cells. Preparation of Samples for Laser Capture Microdissection Paraffin-Embedded Tissues Fixation. Sample preparation for LCM is largely based on standard histologic techniques for sectioning tissues for microscopic examination.2 Most commonly, formalin-fixed (10% neutral-buffered formalin), paraffin-embedded tissues, or OCT embedded frozen tissues are used. Formalin fixation cross-links nucleic acids and proteins making macromolecules more susceptible to shearing. Duration of the formalin fixation is important for recovery of macromolecules. For tissue blocks less than 3 mm in thickness, 2–4 hr of fixation is adequate. Longer periods of fixation may reduce yields of macromolecules. RNA is much more labile and recovery is less predictable. Ethanol-based fixatives are best for nucleic acid preservation but are not routinely used in clinical practice. Mercury-based fixatives are not recommended for analysis of macromolecules. After fixation the tissues can be processed for paraffin infiltration to stabilize tissue for sectioning and for long term storage using standard automated procedures. Preparation of Sections. After processing and embedding, tissues are sectioned at the desired thickness using a microtome. Thickness of section is a compromise between microscopic resolution and amount of cellular material that can be captured. For detailed dissection of larger cell types, such as those of many tumors, sections of 10 to 20 µm are reasonable. Thicker sections make visualization of morphology difficult and adherence of the tissue to the activated thermoplastic is less efficient. Thinner sections may be needed for good resolution of smaller cell types. Sections are placed in a water flotation bath set at 42–44◦ to be smoothed and eliminate folds. We do not routinely use adhesive additives and sections are placed on uncharged, uncoated slides to reduce adherence and allow efficient transfer. The tissue sections should be placed in the center of the slides to make it easier to align the area of interest. The quality of sections is critical. Wrinkles or scratches can become a limiting factor with LCM due to the need for uniform contact between the thermoplastic and the tissue. Staining. The slides are dried at 42◦ overnight. Deparaffinization, staining, and dehydration are performed as follows using coplin jars in a fume hood: 1. Xylene (histology grade) for 5 min (×2) 2. 100% Ethanol for 30 sec (×2) 3. 95% Ethanol for 30 sec 2
D. Sheehan and B. Hrapchak, “Theory and Practice of Histotechnology,” 2nd Ed. Battelle Press, Columbus, OH, 1987.
242
SPECIALIZED USES
[20]
4. 70% Ethanol for 30 sec 5. One-minute wash under running distilled water 6. Mayer’s hematoxylin for 30 sec 7. Wash under running distilled water for a few seconds 8. Bluing reagent, one dip or until differentiation 9. Wash under running distilled water for a few seconds 10. Eosin Y for 10–30 sec (optional) 11. 95% Ethanol for 30 sec (×3) 12. 100% Ethanol washes for 30 sec (×2) 13. Xylene for 5 min (×2) 14. The samples are air dried for at least 20 min before microdissection. Any moisture left in the tissue section will hinder microdissection. Samples are now ready for LCM (see below). The dehydration steps are of prime importance for effective LCM and for preservation of labile molecules such as RNA. Dehydration of the tissue section inhibits RNase and the dry sections are efficiently captured by the thermoplastic. The tissue sections prepared for LCM are not coverslipped. The length of time in hematoxylin and eosin depends on the degree of desired contrast and the thickness of the tissue section. We tend to use a light hematoxylin stain alone to limit exposure to inhibitory factors and contaminants that might affect subsequent procedures. Other stains (methyl green and toluidine blue) and even immunohistochemistry (see below) can be used to define tissue and cell architecture; however, exposure to aqueous solutions and reagents that might bind to macromolecules should be limited as much as possible. Frozen Tissues Frozen tissues yield better quality macromolecules; however, morphology is not as well preserved as in paraffin-embedded tissues. Frozen tissues in OCT (Tissue-Tek) embedding medium can be stored at −80◦ or in liquid nitrogen for years with good preservation of tissue morphology and macromolecules. A critical step in dealing with frozen specimens is fixation after sectioning. Method of fixation will depend on use of the material and molecules to be isolated. Preparation of Sections and Staining 1. The frozen tissue block should be attached to the cryostat chuck with OCT and then allowed to come to temperature in the cryostat at −20◦ for about 15– 30 min. Section the frozen block at the desired thickness (10 to 20 µm). 2. Pick up tissue with a room temperature, uncoated, uncharged slide and
[20]
LASER CAPTURE MICRODISSECTION
243
immediately fix in the desired medium. Fixation with 70 or 95% ethanol for 2 min is useful for isolation of nucleic acids and most proteins. 3. Wash under running distilled water for 3 min. The slides can then be stained (start at step 5 above for hematoxylin and eosin staining). It is best to process slides as close as possible to the time of use for LCM although frozen tissue sections may be stored at −80◦ . Completely processed and stained sections may be stored dry in a desiccator. Immunohistochemistry for Laser Capture Microdissection The lengthy regular immunostaining procedure may interfere with recovery of macromolecules, especially RNA. It is necessary to minimize the amount of time samples are incubated in an aqueous medium. Immunostaining can be modified to help reduce the overall time and rapid methods for immunochemistry have been described.3 For example, the incubation time of the primary and secondary antibodies can sometimes be reduced sharply by increasing the concentration of the antibody dilution. Protocol for Immunostaining Procedure 1. Tissue sections may need to be mounted on charged or coated slides for proper adherence throughout the procedure; however, this may interfere with LCM. Deparaffinize the tissue section with xylene and rehydrate the slides with a descending series of alcohols diluted in RNase-free deionized water (95%/70%/50%). For frozen sections they should be fixed in 70% ethanol for 1–2 min or acetone for 4 min followed by two washes of deionized RNase-free water. 2. 1× PBS or 1× TBS wash for a few seconds. 3. Block with normal serum (1 : 10) from the same animal species as secondary antibody for 1 min and blot off serum by tapping the edge of the slide on an absorbent towel. 4. Incubate with primary antibody for 10–30 min (cover tissue section with reagent). 5. Wash in 1× PBS or 1× TBS three times. 6. Secondary enzyme-conjugated [horseradish peroxidase (HRP) or alkaline phosphatase (AP)] antibody for 10–20 min (may need to increase usual concentration). 7. Wash with 1× PBS or 1× TBS three times. 8. Develop the signal with appropriate substrate under microscopic visualization. 3
F. Fend, M. R. Emmert-Buck, R. Chuaqui, K. Cole, J. Lee, L. A. Liotta, and M. Raffeld, Am. J. Pathol. 154, 61 (1999).
244
SPECIALIZED USES
[20]
9. Wash for 30 sec in running distilled water and dehydrate with a series of ascending ethanols (70%, 95%, 100%) diluted in RNase-free deionized water for 30 sec each and xylene for 5 min twice. 10. Air-dry for 10 min. Samples are ready for LCM. Principles and Technical Basis of Laser Capture Microdissection LCM was developed at the National Cancer Institute and commercialized as the PixCell system by Arcturus Engineering, Inc. (Mountain View, CA).1 The system is designed to allow the user to visually target cells through an optical-grade plastic microcentrifuge tube cap that has a layer of ethylene vinyl acetate thermoplastic on the bottom. The bottom of the cap is in contact with the surface of the tissue section (Fig. 1). A low energy laser pulse melts the thermoplastic and targeted cells are trapped during the milliseconds of melting and resolidifying. The sample is then transferred from the bottom of the cap to a microcentrifuge tube containing the appropriate buffer for macromolecule extraction. The laser amplitude and pulse width may be adjusted to create the appropriate area of thermoplastic activation from 7.5- to 60-µm diameter spot size (depending on instrument model). The laser can be activated as many times as necessary to capture other regions of interest in the same sample. The LCM microscope is connected to a personal computer for additional laser control and image archiving. Image archiving is essential in correlating morphology and molecular analyses. Detailed protocols for use of the instrument are provided by the manufacturer. LCM has the advantages of accuracy, speed, and versatility. LCM does not require manual dexterity because it is a no-touch system. The preserved morphology of both the residual cells and the captured cells on the cap allows validation of collected tissue and the system provides a mechanism for digital image documentation and archiving (Fig. 2). Since the cells are strongly polymerized in the EVA plastic there is minimal danger of tissue loss. The laser is safe and does not affect the quality of the DNA, RNA, and protein. It does not destroy the adjacent
FIG. 1. Diagrammatic representation of laser capture microdissection. (a) Laser pulse activation of thermoplastic on the bottom of microcentrifuge cap. (b) Adherence of targeted tissue to bottom of cap.
[20]
LASER CAPTURE MICRODISSECTION
245
FIG. 2. Laser capture microdissection of breast lobule. (a) Photomicrograph of breast lobule prior to dissection (hematoxylin and eosin). (b) Same microscopic field after removal of captured tissue. (c) Lobule adherent to microcentrifuge cap after capture.
tissue; therefore normal and tumor cells can be picked up from the same slide by changing caps. The versatility of the system allows use on archival tissue sections, smears, and touch preps. There are a few limitations of LCM. Since the tissue sections are not coverslipped, the optical resolution is poor. This may make distinction of various cell types difficult. Using a light diffuser provided with the system can help with this problem. It is important that the user be a very competent histopathologist, familiar with both the tissues and diseases under study and the artifacts of tissue processing and microscopy. Although LCM is quite precise, a potential problem is that the activated thermoplastic may pick up neighboring tissue as selected cells are torn away from the tissue section. Using a piece of Scotch tape to remove loosely adherent contaminates, or the new “no-touch” cap (Arcturus) with a minimal contact force, minimizes nonspecific transfer of cells and tissue. Other limitations are associated with the preparation of the material rather than the LCM system. For example, lack of adherence of the cells to the membrane is usually due to poor preparation of sections; limited amount of sample results in small yields of macromolecules; and degradation of the DNA, RNA, and proteins is usually due to poor tissue handling. The protocols outlined here can help eliminate some of these limitations. Although the yields of the technique are largely dependent on the type of sample, we attempt to collect 500 to 5000 cells from a 10-µm section for DNA or RNA isolation. This typically yields enough material for several PCR experiments even from paraffin-embedded tissues. Molecular Extraction from Laser Capture Microdissection Samples Most standard methods of macromolecule isolation can be used with cells and tissues procured by LCM. Many companies now provide procedures in mini-kit form that are appropriate for small samples such as those achieved with LCM. The following protocols are simple procedures that we have successfully used.
246
SPECIALIZED USES
[20]
DNA Extraction 1. After LCM, insert the cap in a 0.5-ml microcentrifuge tube (sterile) containing 50 µl of digestion buffer (10 mM Tris-HCL, pH 8.0, 1 mM EDTA, 1% Tween 20 with 0.08% proteinase K). Invert the tube and place it in an incubator or on a heating block at 37◦ overnight. 2. Next morning centrifuge the lysate containing digested cells at 10,000g for 5 min. Add a fresh aliquot of proteinase K and incubate the lysate at 50◦ for 20 min, then heat to 95◦ for 10–15 min to inactivate the proteinase K. The lysate can now be used for PCR template. 3. The DNA can be further purified by conventional phenol–chloroform extraction followed by ethanol precipitation. We have found that this step is helpful if samples are to be stored for long periods of time prior to use. In this case a carrier such as glycogen or tRNA should be used for efficient recovery. RNA Extraction 1. Insert the cap into a 0.5-ml Eppendorf microfuge tube containing 200 µl of denaturing buffer [4 M guanidinium thiocyanate, 25 mM sodium citrate-2H2O, 0.5% (w/v) sodium lauryl sarcosinate] and 1.6 µl 2-mercaptoethanol. Gently shake the tube for 30 sec, then incubate at room temperature for 5 min with occasional or continuous agitation to digest all the tissue from the cap. 2. Briefly centrifuge the tube to bring down any buffer clinging to the cap or sides and remove the cap. Multiple caps can be processed in this way to increase the starting material used. 3. Transfer the solution to a 1.5-ml centrifuge tube. The buffer can be kept on ice or stored at −80◦ until further extraction. 4. Add 20 µl of 2 M sodium acetate (pH 4), 220 µl of water-saturated phenol, and 60 µl of chloroform–isoamyl alcohol (49 : 1). 5. Vortex and put on ice for 15 min. 6. Centrifuge at 10,000g for 10 min at 4◦ and transfer the upper (aqueous) layer to a new tube. 7. Mix in 1 µl of carrier glycogen (GenHunter 10 mg/ml) and precipitate by adding 200 µl of cold isopropanol (mix completely) at −80◦ for at least 30 min. It may be left for longer periods if convenient. 8. Centrifuge at 10,000g, 4◦ for 30 min. 9. Carefully remove the supernatant to minimize disruption of the pellet. 10. Wash the glycogen pellet with 400 µl of cold 70% ethanol. Centrifuge for 5 min to pellet the precipitate. 11. Carefully remove the supernatant. 12. Vacuum or air-dry the pellet to remove any residual ethanol. 13. Store at −80◦ until use.
[20]
LASER CAPTURE MICRODISSECTION
247
DNase Treatment For some applications it may be critical to remove residual genomic DNA. 1. Resuspend the pellet with 15 µl of RNase-free DEPC water. 2. Add 1 µl of RNase inhibitor (PerkinElmer) and 2 µl of 10× DNase I buffer (GenHunter) and 2 µl of DNase I (GenHunter). 3. Incubate at 37◦ for 30 min to 2 hr. 4. Reextract the RNA following DNAse treatment with the phenol–chloroform method as described above. Finally, resuspend the pellet with RNase-free water. Analysis of Laser Capture Microdissection Samples The prepared nucleic acid can be successfully used in most standard procedures that employ PCR including microsatellite analysis for loss of heterozygosity and clonality, sequencing, sequence detection, and cDNA synthesis.4–9 Components and protocols will depend largely on the template quantity and quality. In general template is limiting and care should be taken to avoid contamination and increase yield of PCR product. This may include multiple rounds of amplification with nested primers and using sensitive methods of detection of final product.
4
C. J. M. Best and M. R. Emmert-Buck, J. Mol. Diagn. 1, 53 (2001). R. G. Weber, M. Scheer, I. A. Born, S. Joos, J. M. Cobbers, C. Hofele, G. Reifenberger, J. E. Zoller, and P. Lichter, Am. J. Pathol. 153, 295 (1998). 6 J. Mora, N. K. V. Cheung, G. Juan, P. Illei, I. Cheung, S. Chi, M. Ladanyi, C. Cordon-Cardo, and W. L. Gerald, Can. Res. 61, 6892 (2001). 7 M. R. Bernsen, H. B. Dijkman, E. de Vries, C. G. Figdor, D. J. Ruiter, G. J. Adema, and G. N. van Muijen, Lab. Invest. 78, 1267 (1998). 8 L. Luo, R. C. Salunga, H. Guo, A. Bittner, K. C. Joy, J. E. Galindo, H. Xiao, K. E. Rogers, J. S. Wans, M. R. Jackson, and M. G. Erlander, Nat. Med. 5, 117 (1999). 9 V. Luzzi, V. Holtschlag, and M. A. Watson, Am. J. Pathol. 158, 2005 (2001). 5
248
[21]
SPECIALIZED USES
[21] Analysis of Folliculostellate Cells by Laser Capture Microdissection and Reverse Transcription–Polymerase Chain Reaction (LCM-RT/PCR) By RICARDO V. LLOYD, LONG JIN, KATHARINA H. RUEBEL, and JILL M. BAYLISS Introduction Folliculostellate (FS) cells were first described in the rat anterior pituitary gland by Farquhar.1 These cells have many functions including phagocytosis, serving as supporting or sustentacular cells and in the paracrine regulation of the secretory anterior pituitary cells.2–6 They produce many substances that influence anterior pituitary function, including fibroblast growth factor,7 vascular endothelial growth factor,8 leukemia inhibitory factor,9 interleukin-6 (IL-6),10 follistatin,11 nitric oxide synthase,12,13 and leptin.14 Previous attempts to obtain homogeneous populations of FS cells have been unsuccessful.15 Biophysical separation techniques have led to enriched populations of FS cells,15–17 but not homogenous populations. Cell lines of FS cells have been 1
M. G. Farquhar, [Abstract] Anat. Rec. 127, 291 (1957). T. Nakajima, H. Yamaguchi, and K. Takahashi, Brain Res. 191, 523 (1980). 3 N. Shirasawa, S. Yamaguchi, and F. Yoshimura, Cell Tissue Res. 237, 7 (1984). 4 R. V. Lloyd and J. Mailloux, Am. J. Pathol. 133, 338 (1988). 5 M. Baes, W. Allaerts, and C. Denef, Endocrinology 120, 685 (1987). 6 W. Allaerts, P. Carmeliet, and C. Denef, Mol. Cell. Endocrinol. 71, 73 (1990). 7 N. Ferrara, L. Schweigerer, G. Neufeld, R. Mitchell, and D. Gospodarowicz, Proc. Natl. Acad. Sci. U.S.A. 84, 5773 (1987). 8 N. Ferrara and W. Henzel, Biochem. Biophys. Res. Commun. 161, 851 (1989). 9 N. Ferrara, J. Wener, and W. Henzel, Proc. Natl. Acad. Sci. U.S.A. 89, 698 (1992). 10 H. Vankelecom, P. Matthys, J. Van Damme, H. Heremans, A. Billiau, and C. Denef, J. Histochem. Cytochem. 41, 151 (1993). 11 U. B. Kaiser, B. L. Lee, R. S. Carroll, G. Unabia, W. W. Chin, and G. V. Childs, Endocrinology 130, 3048 (1992). 12 S. Ceccatelli, A. L. Hulting, X. Zhang, L. Gustafson, M. Villar, and T. Hokfelt, Proc. Natl. Acad. Sci. U.S.A. 90, 11292 (1993). 13 R. V. Lloyd, L. Jin, X. Qian, S. Zhang, and B. W. Scheithauer, Am. J. Pathol. 146, 86 (1995). 14 L. Jin, B. G. Burguera, M. E. Couce, B. W. Scheithauer, J. Lamsan, N. L. Eberhardt, E. Kulig, and R. V. Lloyd, J. Clin. Endocrinol. Metab. 84, 2903 (1999). 15 C. Denef, E. Hautekeete, A. De Wolf, and B. Vanderschueren, Endocrinology 103, 724 (1978). 16 C. Denef, P. Maertens, W. Allaerts, A. Mignon, W. Robberecht, L. Swennen, and P. Carmeliet, Methods Enzymol. 168, 47 (1989). 17 X. Qian, L. Jin, and R. V. Lloyd, Endocr. Pathol. 9, 339 (1998). 2
METHODS IN ENZYMOLOGY, VOL. 356
Copyright 2002, Elsevier Science (USA). All rights reserved. 0076-6879/02 $35.00
[21]
LCM-RT/PCR ANALYSIS OF FS CELLS
249
used to study FS cells in vitro,18–20 but these cell lines may be different from primary FS cells in biological and biophysical properties. Laser capture microdissection (LCM) methods have been used to obtain homogenous cell populations for cellular and molecular analyses.21–25 Several laboratories have combined immunophenotypic characterization of specific cell types with LCM (immuno-LCM) which have resulted in the preparation of highly homogeneous cell populations.23–25 This approach of immuno-LCM in combination with RT-PCR allows the study of gene expression in immunophenotypically characterized homogeneous cells. This chapter describes the procedures for the use of immuno-LCM and RTPCR to prepare homogeneous populations of FS cells. The properties of primary populations of FS cells are compared to those of cultured populations of FS cells.26 Materials and Methods Anterior pituitaries are from 60- to 90-day-old female Wistar-Furth rats (Harlan Sprague Dawley, Inc., Indianapolis, IN). All experiments using animals were conducted in accordance with the principles and procedures outlined in the NIH Guidelines for the Care and Use of Experimental Animals. The mouse FS cell line (TtT/GF) was a gift from Dr. K. Inoue (Gumna University, Maebashi, Japan). Dulbecco’s minimum essential medium (DMEM) with 15% horse serum, 2.5% fetal calf serum, 1 µg/ml insulin, and 1% antibiotics (penicillin and streptomycin) were from Life Technologies, Inc. (Grand Island, NY). TGFβ1 was obtained from R&D Systems, Inc. (Minneapolis, MN). Pituitary adenylase cyclase-activating polypeptide-38 (PACAP-38) was from Peninsula Laboratories, Inc. (Belmont, CA). The PixCell II Laser Capture Microdissection System was purchased from Arcturus Engineering, Inc. (Mountain View, CA). Anti-S100 protein polyclonal antiserum is from DAKO Corp. (Carpinteria, CA). 18
H. Ishikawa, H. Nogami, and N. Shirasawa, Nature 303, 711 (1983). K. Inoue, H. Matsumoto, C. Koyama, K. Shibata, Y. Nakazato, and A. Ito, Endocrinology 131, 3110 (1992). 20 T. Yamasaki, H. Fujita, K. Inoue, T. Fujita, and N. Yamashita, Endocrinology 138, 4346 (1997). 21 M. R. Emmert-Buck, R. F. Bonner, P. D. Smith, R. F. Chuaqui, Z. Zhuang, S. R. Goldstein, R. A. Weiss, and L. A. Liotta, Science 274, 999 (1996). 22 R. F. Bonner, M. R. Emmert-Buck, K. Cole, T. Pohida, R. Chuaqui, S. Goldstein, and L. A. Liotta, Science 278, 1481 (1997). 23 L. Jin, C. A. Thompson, X. Qian, S. J. Kuecker, E. Kulig, and R. V. Lloyd, Lab. Invest. 79, 511 (1999). 24 S. J. Kuecker, L. Jin, E. Kulig, G. L. Oudraogo, P. C. Roche, and R. V. Lloyd, Appl. Immunohistochem. Mol. Morphol. 7, 192 (1999). 25 F. Fend, M. R. Emmert-Buck, R. Chuaqui, K. Cole, J. Lee, L. A. Liotta, and M. Raffeld, Am. J. Pathol. 154, 61 (1998). 26 L. Jin, I. Tsumanuma, K. H. Ruebel, J. M. Bayliss, and R. V. Lloyd, Endocrinology 142, 1703 (2001). 19
250
SPECIALIZED USES
[21]
Immuno-LCM Normal rat pituitaries are freshly dissociated with 0.25% trypsin and placed onto glass slides placed by cytocentrifugation using 1 × 104 cells/slide for immunoLCM analysis as previously reported to prepare homogeneous populations of pituitary FS cells.23 The pituitary cells are rehydrated with PBS buffer, and immunocytochemistry is performed using an anti-S100 antibody (from DAKO Corp.; diluted 1 : 1000) to characterize FS cells in the normal pituitaries. Immunostaining is performed within 2.5 hr using the avidin–biotin peroxidase complex method, as previously described.23 The slides are then lightly counterstained with hematoxylin and dehydrated with 95% and 100% ethanol, incubated in xylene for 6 min, and air-dried before LCM. All reactions are performed in ribonuclease-free solution to prevent RNA degradation. The PixCell II Laser Capture Microdissection instrument is used for LCM analysis. LCM parameters included a laser power of 90 milliwatts, laser pulse duration of 1.2 ms, and laser spot size of 7.5 µm diameter. The infrared laser is pulsed over cells of interest, and this melts the film directly onto the targeted cells, embedding the captured cells. Approximately 400 S100-positive cells from each sample are captured using 2–3 slides. After LCM, total RNA extraction from the captured cells is performed using the TRIzol reagent kit (Life Technologies, Inc.). The caps with LCM cells are immediately placed into sterile 0.5-ml microcentrifuge tubes (PGC Scientifics, Frederick, MD) containing 200 µl TRIzol reagent and inverted at room temperature for 1 hr before storing at −70◦ overnight. On the following day, the RNA extraction is performed according to the manufacturer’s instructions. After ethanol precipitation, the RNA pellet is resuspended in 10 µl diethylpyrocarbonate–H2O and used for the RT-PCR reactions. Cell Culture The effects of TGFβ1 and PACAP-38 on FS cells are analyzed by growing cells in DMEM with complete serum in a 37◦ , 5% CO2 atmosphere, as previously reported.27 Aliquots of TtT/GF cells are treated with TGFβ1 (10−9M) or PACAP-38 (250 nM) for 4 days and then harvested and used for RNA extraction. Dissociated rat pituitary cells are incubated in DMEM with 2% FCS, and aliquots of cells are treated with TGFβ1 (10−9M) or PACAP-38 (250 nM). After 4 days of treatment, the pituitary cells are harvested and attached to slides by cytocentrifugation using 1 × 104 cells/slide for immuno-LCM analysis. At this cell density it is possible to capture individual cells without contamination from neighboring cells. The slides are fixed in 100% ethanol for 5 min, air-dried, and kept at −70◦ until they are used for LCM. 27
X. Qian, L. Jin, J. P. Grande, and R. V. Lloyd, Endocrinology 137, 3051 (1996).
[21]
LCM-RT/PCR ANALYSIS OF FS CELLS
251
RT-PCR First-strand DNA is prepared from total RNA by using a First Strand Synthesis Kit (Stratagene, La Jolla, CA), according to the manufacturer’s instruction. The RT reaction is performed in a final volume of 50 µl with 10 µl total RNA from LCM transfer cells or 5 µg total RNA from the FS cell line. Total RNA (5 µg) from normal rat pituitary tissues without LCM is used as a positive control. The sequence of primers for PCR and internal probes for Southern hybridization are as follows: rat S100b (GenBank accession no. X01090; product size, 211 bp), 5 -GTTGCCCTCATTGATGTCTTC (sense), 5 -AGACGAAGGCCATAAACTC CT (antisense), and 5 -CCATCCCCATCTTCGTCCAGCGTCTCCATC (probe); mouse glial fibrillary acidic protein (GFAP; X02801; 391 bp), 5 -GCTGAACT GAACCAGCTTCGA (sense), 5 -CTTGGCCACATCCATCTCCAC (antisense), and 5 -AGAACTGGATCTCCTCCTCCAGCGATTCAA (probe); rat PACAP (M63006; 215 bp), 5 -CATCTTCACAGACAGCTATAG (sense), 5 -GTTTGG AAAGAACACATGAGT (antisense), and 5 -CCCTAGCACGGCCGCCAAGTA TTTCTTGAC (probe); rat PACAP-RI (303-bp) (40), 5 -CTTGTACAGAAGCTC CAGTCC (sense), 5 -CCGGTGCTTGAAGTCATAGT (antisense), and 5 -GAT GAGCAGTAGGGTGGAGCGGGCCAGCCG (probe); and mouse hypoxanthine phosphoribosyl transferase (HPRT; J0042; 478 bp), 5 -TTCCTCCTCAGACCG CTTTTT (sense), 5 -GTTTGCATTGTTTTACCAGTG (antisense), and 5 -AGCA CACAGAGGGCCACAATGTGATGGCCT (probe). The other primers and probes used in this study have been published in previous reports, including those for rat TGFβ1 (161 bp),27 rat TGFβ-RII (304 bp),28 rat leptin (244 bp), and leptin receptor (OB-Rb; 302 bp).29 Rat GH (V01237; 376 bp),23 rat PRL (344 bp),23 and rat POMC (K01877; 318 bp)23 primers are also used as controls to check the homogeneity of LCM-captured FS cells. The specificity of the primers and probes is verified by GenBank searches. Most primers were designed to match both rat and mouse sequences. PCR amplification is performed in a 50-µl final reaction volume containing 16 µl RT reaction product from 400 LCM captured cells as template DNA. For the FS cell line, a 100-µl final volume containing 10 µl RT reaction product is used. PCR amplification is performed for 40 cycles for LCM samples and 30 cycles for the FS cell line. The annealing temperatures range from 55◦ to 60◦ and are obtained with the Oligo-5 software program (Molecular Biology Insights, Inc., Cascade, CO). After the final cycle, the elongation step is extended by 10 min at 72◦ . The housekeeping gene, mouse HPRT, is amplified from the same RT products and used as internal control. In some experiments the Immuno-LCMcaptured FS cells are analyzed by RT-PCR for GH, PRL, and POMC to determine 28 29
X. Qian, L. Jin, and R. V. Lloyd, Endocr. Pathol. 7, 77 (1996). L. Jin, S. Zhang, B. G. Burguera, M. E. Couce, R. Y. Osamura, E. Kulig, and R. V. Lloyd, Endocrinology 141, 333 (2000).
252
SPECIALIZED USES
[21]
the homogeneity of the cell population. Omission of reverse transcriptase during the RT reaction is used as a negative control. A 20-µl aliquot of the PCR product is routinely analyzed by electrophoresis on a 2% agarose gel with ethidium bromide staining. Titration studies with different amounts of cDNA are performed to verify that each amplification is in the linear range. The PCR products are transferred to nylon membrane filters, and Southern hybridization is performed with 33P-labeled internal probes at 42◦ for 18 hr. After washing in 6× SSC (standard saline citrate)/0.1% SDS at 23◦ for 20 min and at 42◦ for 10–20 min, autoradiography is performed with Kodak X-Omat-AR film (Eastman Kodak Co., Rochester, NY). The amounts of leptin and IL-6 mRNA are quantitated by densitometry, and the mRNA levels are normalized relative to HPRT. Comments After immunostaining for S100 protein, the FS cells can be readily identified by cytoplasmic and nuclear positivity and represent 5–10% of the total cells/slide. S100-positive cells are collected by immuno-LCM and used for RNA extraction (Fig. 1).26 RT-PCR analysis shows that rat FS cells express mRNA for S100 protein,
FIG. 1. LCM of cells from pituitary after immunostaining for S100 protein. (A) Normal FS cells stained positively for S100 protein. The arrow shows the cell to be captured. (B) The cell indicated by the arrow is captured and transferred to the cap. (C) The captured cell is transferred from cap to the TRIzol reagent and used for RNA extraction and RT-PCR. Reproduced from L. Jin, I. Tsumanuma, K. H. Ruebel, J. M. Bayliss, and R. V. Lloyd, Endocrinology 142, 1703 (2001) with permission from the Endocrine Society.
[21]
A GFAP
S100 2
1
2
3
4
5
6
1
2
5
6
1
2
4
3
4
Leptin Receptor
Leptin
4
5
6
1
2
5
6
1
2
TGFβRII
TGFβI 3
3
3
4
5
6
1
2
5
6
1
2
PACAP 3
4
3
4
5
IL-6 6
1
2
6
1
2
4
4
5
6
5
6
HPRT
PACAP RI 3
3
5
3
4
B PRL 1
2
3
4
GH 5
6
1
2
3
POMC 4
5
6
1
2
3
4
5
6
253
FIG. 2. RT-PCR analysis of FS cells. Approximately 400 S100-positive cells are collected and analyzed by RT-PCR. (A) Analysis of various mRNA in FS cells. Lane 1, Rat pituitary FS cell; lane 3, TtT/GF cells; lane 5, normal rat pituitary without LCM used as positive control. Lanes 2, 4, and 6, Negative control lanes without RT. Top panel: The ethidium bromide-stained gel; bottom panel: Southern hybridization with the internal probes described in Materials and Methods. (B) Analysis of PRL, GH, and POMC expression. The lanes are the same as in A. Only the normal pituitary control expressed these hormone mRNAs. Reproduced from L. Jin, I. Tsumanuma, K. H. Ruebel, J. M. Bayliss, and R. V. Lloyd, Endocrinology 142, 1703 (2001) with permission from the Endocrine Society.
LCM-RT/PCR ANALYSIS OF FS CELLS
1
254
[21]
SPECIALIZED USES
GFAP, leptin, the long form of the leptin receptor, IL-6, TGFβ1, TGFβII, PACAP38, and PACAP receptor I (Fig. 2A).26 The cells are negative for PRL, GH, and POMC mRNA, supporting the specificity of the procedure for capturing FS cells (Fig. 2B).26 Aliquots of total RNA from the TtT/GF cell line and normal pituitary are analyzed with the same primers. Only PACAP-38 mRNA is not detected in
A
Leptin 1 2 3 4 5
IL - 6 1 2 3 4 5
HPRT 1 2 3 4 5
FIG. 3. (A) RT-PCR analysis of the effects of TFGβ1 and PACAP-38 on leptin and IL-6 mRNA expression in pituitary FS cells. Cultured cells are collected by immuno-LCM, analyzed by RT-PCR, and normalized with HPRT. Lane 1, Control pituitary FS cells; lane 2, TGFβ1-treated FS cells; lane 3, PACAP-38-treated FS cells; lane 4, normal rat pituitary cells without LCM, used as a positive control; lane 5, normal rat pituitary cells without RT, used as a negative control. The top panel in A shows the ethidium bromide-stained gel; the bottom panel of A shows Southern hybridization with the internal probe described in Materials and Methods. (B) Densitometric analysis show a 1.8-fold increase in leptin mRNA by TGFβ1 treatment. Data are the mean ± SEM from three experiments. Reproduced from L. Jin, I. Tsumanuma, K. H. Ruebel, J. M. Bayliss, and R. V. Lloyd, Endocrinology 142, 1703 (2001) with permission from the Endocrine Society.
[21]
LCM-RT/PCR ANALYSIS OF FS CELLS
255
the TtT/GF cell line, which may be related to the rat primers used. PRL, GH, and POMC mRNA were detected in the normal pituitary, but not in the TtT/GF cells. TGFβI Regulation of Leptin mRNA Expression After treatment of dissociated pituitary cells with TGFβ1 or PACAP-38 followed by immuno-LCM, RT-PCR, and Southern hybridization, there is increase in leptin, but not IL-6, mRNA expression (Figs. 3A and 3B).26 This effect is specific for TGFβ1, as PACAP-38 did not influence leptin or IL-6 mRNA expression.
This Page Intentionally Left Blank
Section III Genetic Applications
This Page Intentionally Left Blank
[22]
ANALYSIS OF GENE EXPRESSION
259
[22] Analysis of Gene Expression By JANETTE K. BURGESS and BRENT E. MCPARLAND Introduction The recent advances of the collaborative sequencing programs that are decoding the complete genomes of many organisms are providing opportunities for new approaches to gene analysis. With these developments have come greater needs for attention to the source material to ensure that the gene expression profiles reflect the gene activity in the cells of interest rather than an experimental artifact. The advent of techniques that allow the manipulation of genetic material has significantly improved the understanding of cellular events, but the study of crude tissue extracts is complicated by the heterogeneous nature of their cellular components. The development of laser capture microdissection (LCM) has enabled the isolation of populations of defined single cell types which can then be analyzed for their DNA, RNA, or protein content (discussed in [14], this volume).1 The ability to isolate and analyze specific individual cells from a complex multicellular sample reduces the opportunity for nonspecific contribution from colocalized cell types and will enhance our understanding of the role played by each cell type. LCM-based analysis is applicable to any disease process for which histopathological lesions are accessible through tissue sampling or other sources, e.g., cytospins of blood. Examples include mapping the field of genetic changes associated with the progression of premalignant and malignant cancer lesions; analysis of gene expression patterns in atherosclerosis, inflammation, Alzheimer’s disease plaques, multiple sclerosis, and infectious microorganism diagnosis; and analysis of genetic abnormalities in utero from selected rare fetal cells in maternal fluids.2 The ability to isolate defined cell populations from small tissue sections enables the utilization of biopsy samples for analysis of gene expression. The identification of gene expression patterns may provide vital information for the understanding of the disease process and may contribute to diagnostic decisions and therapies tailored to the individual patient. Molecules found to be associated with defined pathological lesions may provide opportunities for new therapeutic targets in the future. Sample Preparation Particular attention must be paid to the pretreatment of samples from which RNA, and to a lesser degree DNA, will be isolated downstream. RNA of reasonable 1 2
K. K. Jain, Methods Enzymol. 356, [14], 2002 (this volume). R. F. Bonner, M. Emmert-Buck, K. Cole, T. Pohida, R. Chuaqui, S. Goldstein, and L. A. Liotta, Science 278, 1481 (1997).
METHODS IN ENZYMOLOGY, VOL. 356
Copyright 2002, Elsevier Science (USA). All rights reserved. 0076-6879/02 $35.00
260
GENETIC APPLICATIONS
[22]
integrity has not been successfully isolated from paraffin embedded tissues. The treatment of the tissue between tissue resection and the freezing of fresh frozen specimens is important. Tissue pH can play a critical role in the integrity of the RNA.3,4 Successful isolation of intact mRNA in the brain is associated with a tissue pH between 6.1 and 7.0. Tissues with a pH below 6.0 result in fragmented or absent RNA. After mounting in a protective medium such as OCT (Miles, Elkhart, IN) and snap freezing in liquid nitrogen cooled isopentane or hexane, tissues should be stored at −80◦ . The freshly frozen tissues should be brought to −20◦ and immediately cut on the cryostat. The optimal thickness of the sections varies for different tissue types but is generally between 4 and 10 µm. The sections are placed on a plain glass slide at room temperature and immediately fixed. Pretreatment of the glass slides to remove any RNases present either by prebaking at 230◦ for 4 hr5 or by treating with RNase Away (Molecular BioProducts, San Diego, CA) helps to reduce further RNA degradation. The blade of the cryostat should be changed and the blade holder cleaned with acetone between each different tissue sample. Once sectioning of the tissue commences, it is important to work rapidly through to completion of the staining protocol to avoid RNA degradation. The integrity of the RNA within a tissue section can be checked by staining one section with 10 µg/ml acridine orange in 0.2 M dibasic sodium phosphate/0.1 M citric acid (pH 4.0) using the method described by Ginsberg and colleagues.3,6 Fixation and Staining Protocol All solutions used throughout the fixation and staining procedures should be prepared in diethylpyrocarbonate (DEPC)-treated water and refreshed between staining sections from a different tissue sample. Sections are fixed in 70% ethanol at room temperature for 2 to 4 min and rinsed rapidly in DEPC–water. Mayer’s hematoxylin is applied to the slide surface (200 µl per section) for 1 to 2 min followed by a rapid rinse in DEPC–water. Thirty seconds in Scott’s blueing solution enhances the staining of the nuclei followed by 30 sec in DEPC–water and then 30 sec in 70% ethanol. Eosin stain (500 mg/dl, alcohol soluble) is then applied to the slide surface (200 µl per section) for 1 min. The sections are then dehydrated through 95% ethanol for 30 sec, followed by two changes of molecular sieve dried 100% ethanol for 1 min each before a final dehydration step in two changes of 3
S. Bahn, S. J. Augood, M. Ryan, D. G. Standaert, M. Starkey, and P. C. Emson, J. Chem. Neuroanat. 22, 79 (2001). 4 A. E. Kingsbury, O. J. Foster, A. P. Nisbet, N. Cairns, L. Bray, D. J. Eve, A. J. Lees, and C. D. Marsden, Brain Res. Mol. Brain Res. 28, 311 (1995). 5 Y. Kohda, H. Murakami, O. W. Moe, and R. A. Star, Kidney Int. 57, 321 (2000). 6 S. D. Ginsberg, P. B. Crino, V. M. Lee, J. H. Eberwine, and J. Q. Trojanowski, Ann. Neurol. 41, 200 (1997).
[22]
ANALYSIS OF GENE EXPRESSION
261
xylene for 2 min each. Care should be taken to ensure that the tissue is not left in the xylene for too long as the tissue becomes so dry that nonspecific cell adhesion to the caps occurs. The molecular sieves should be pretreated by washing in DEPC– water for three changes of 30 min each followed by baking at 160◦ for 5 days to remove RNases. Once stained, the sections can be wrapped in aluminum foil and placed in an airtight container in the presence of silica gel. Storage time for stained sections prior to cell capture varies for different tissue types as prolonged storage can lead to irreversible adherence to the slide surface. In our experience airway smooth muscle cells from human lung sections bind to the slide surface in an irreversible manner after 72 hr, whereas the surrounding cell types are easily captured after weeks in storage. In contrast, mouse brain sections can be stored for prolonged periods and all cell types can still be captured. Alternative Fixatives Fixation methods that alter the three-dimensional structure of the proteins within the tissues (such as ethanol, methanol, or acetone) allow the isolation of RNA from captured cells. The stronger fixation agents (such as formalin and paraformaldehyde) that cross-link the proteins within the tissues inhibit the dissociation of the tissues in the RNA lysis buffer and degrade the RNA.7 A method of fixing tissues using methacarn prior to paraffin embedding has been described that yielded half the quantity of RNA that could be isolated from fresh frozen tissue but the total RNA integrity was well preserved.8 The DNA contamination in this preparation was also reduced compared to fresh frozen tissue. This method of fixation may allow for subsequent paraffin embedding of tissues while still providing the option of RNA retrieval at a later stage. One other potential problem with RNA retrieval from paraffin embedded tissue is that large RNA molecules may fail to be retrieved, but further studies are needed to confirm this effect.7,8 Alternative Stains Many stains work only on tissues that have been prepared in a specific manner. One step often used is mordanting of tissue prior to staining. A commonly used fixative with mordanting properties is Bouin’s solution which requires treatment for 1 hr at 60◦ . This step is crucial for trichrome stains; therefore they cannot be used on fresh frozen tissue that has been fixed in 70% ethanol as required for the isolation of RNA. Periodic acid Schiff (PAS) staining has been observed to lead to significant RNA degradation.5
7
S. M. Goldsworthy, P. S. Stockton, C. S. Trempus, J. F. Foley, and R. R. Maronpot, Mol. Carcinog. 25, 86 (1999). 8 M. Shibutani, C. Uneyama, K. Miyazaki, K. Toyoda, and M. Hirose, Lab. Invest. 80, 199 (2000).
262
GENETIC APPLICATIONS
[22]
Stains that have been reported to differentiate cell types that facilitated the isolation of quality RNA include toluidine blue,9 cresyl violet,8 and methyl green.3 Identification of Cells of Interest Familiarity with the morphology of the tissue being examined is essential, as LCM requires identification of the cell types of interest in the tissue sections without the aid of a coverslip. In some cases it is not possible to distinguish between cell types without the use of cell surface markers, so rapid immunostaining techniques have been developed which enable the identification of cells while preserving the integrity of the RNA in the cells of interest. Fend and colleagues10 have described a method of immunostaining that requires between 12 and 25 min for the complete staining protocol, depending on the nature of the primary antibody. The slides are incubated with the primary, secondary, and tertiary antibodies for 90 to 120 sec each at room temperature with 1× PBS wash between each step. Diaminobenzidine is used for 3–5 min for color development before counterstaining with hematoxylin for 15 to 30 sec and dehydration in graded alcohols (15 sec each) and xylene (twice for 2 min). The final step is to air-dry the slides before LCM is commenced. The antibodies in the Dako Quick Staining kit (Dako, Carpinteria, CA) enabled the shortest incubation times but other primary antibodies which required prolonged incubation times up to 10 min also allowed identification of cells and successful collection of RNA. RNase inhibitor (200 to 400 U/ml) was added to the primary antibody and the color development steps. An alternative method for the identification of cells within a tissue section is to stain one section using a rapid staining protocol. The stained cells of interest in this section are then used as a guide to determine which cells should be captured from an adjacent section which has been stained using a rapid general stain (Fig. 1). This approach reduces the number of treatment steps before the RNA isolation which also reduces the risk of RNA degradation. Cell Capture and Storage This review focuses on the techniques involved in capturing cells using the Arcturus Engineering PixCell II system (Mountain View, CA) but the methods outlined are equally applicable to the preparation of specimens and the isolation of RNA using other laser capture systems. Immediately prior to capturing the tissue sections on the slides are flattened using an electrostatic charged plastic film (PrepStrip Tissue preparation strips 9
J. Pan, E. J. Kunkel, U. Gosslar, N. Lazarus, P. Langdon, K. Broadwell, M. A. Vierra, M. C. Genovese, E. C. Butcher, and D. Soler, J. Immunol. 165, 2943 (2000). 10 F. Fend, M. R. Emmert-Buck, R. Chuaqui, K. Cole, J. Lee, L. A. Liotta, and M. Raffeld, Am. J. Pathol. 154, 61 (1999).
[22]
ANALYSIS OF GENE EXPRESSION
a
c
263
b
d
e
FIG. 1. Identification and capturing of airway smooth muscle cells. Five µm serial sections of human airway tissue were cut and stained with (a) mouse anti-human alpha smooth muscle actin and the Dako LSAB2 new fuchsin detection system and counterstained with Mayer’s hematoxylin for the identification of the different cell types or (b) rapid Mayer’s hematoxylin and eosin and viewed through a visualizer on the LCM microscope (roadmap image). For capturing, the H&E stained section was viewed without the visualizer demonstrating the (c) before image, (d) after image showing where the captured cells were removed, and (e) cap image of the captured cells.
Arcturus) which also serves to remove cells only loosely attached to the slide surface. Cells of interest are captured onto either the standard LCM caps or the CapSure High Sensitivity LCM caps (HS). The standard caps bind more nonspecific material than the HS caps when certain tissue types are studied (for example, human lung and mouse brain). This nonspecific material must be removed using either the adhesive strip on the rear of a Post-It note (3M, St. Paul, MN) or the CapSure cleanup pads from Arcturus, although the adhesive force of the CapSure strips can remove some of the specific material if extreme care is not taken. Once the cells of interest have been captured, the caps are placed on the 0.5 ml tubes in the presence of 100 µl lysis buffer for standard caps or 10–20 µl for HS caps. Arcturus now provides ExtracSure sample extraction devices with the HS caps which enable the cell lysis buffer to be added to a much smaller area of the cap surface thereby reducing the volume required and enhancing the effective yield of RNA. In our experience Eppendorf (Hamburg, Germany) 0.5 ml micro
264
GENETIC APPLICATIONS
[22]
test tubes fit the caps snugly but alternative brands may be appropriate. The tubes can be stored at −80◦ for at least 1 month without appreciable RNA degradation but should be kept inverted to ensure that the lysis buffer is in contact with the cells. RNA Isolation RNA isolation from laser captured cells requires the same procedure as a standard RNA isolation but on a small scale. Many manufacturers are now producing kits for RNA isolation from small numbers of cells and, in our hands, the kits tested all gave reasonable yields for the number of cells processed. The column-based isolation procedures generally provide a greater yield than the methods that rely on multiple phenol–chloroform washes to clean the RNA, as a percentage of the RNA is lost at each wash step. The columns that have a smaller surface area, many of which are now associated with the RNA isolation kits for small cell numbers, enable the RNA to be eluted in a smaller volume, thereby reducing the need for concentrating the RNA by precipitation or vacuum centrifugation. Although it is still possible to develop a method for RNA isolation using individual reagents made up in the laboratory, the nature of the samples available and the time taken for LCM have resulted in many groups turning to kit-based isolations to avoid the loss of precious samples as the method is developed. RNA Isolation Kits One of the earliest kits to be marketed for use with LCM samples was the Stratagene (La Jolla, CA) Micro RNA isolation kit. This kit has been reported by many users of LCM to be a successful method for RNA isolation from LCM captured cells.10–13 Some users have adapted this kit to adjust the phenol–chloroform wash steps.7 More recently Stratagene has released the Total RNA Microprep kit which has been further optimized for isolation of RNA from LCM captured cells. In our hands this kit works well. This newer kit is a column-based extraction which does not contain the phenol–chloroform steps. The Qiagen (Valencia, CA) RNeasy mini kit is another column-based kit that has been reported to yield good quality RNA from LCM samples.14 The widely used RNA isolation reagent from Life Technologies (Invitrogen, Carlsbad, CA), Trizol, reacted adversely with the thermoplastic film on the LCM caps and consequently is not appropriate for the isolation of RNA from LCM. 11
L. Luo, R. C. Salunga, H. Guo, A. Bittner, K. Joy, J. E. Galindo, H. Xiao, K. E. Rogers, J. S. Wan, M. R. Jackson, and M. G. Erlander, Nat. Med. 5, 117 (1999). 12 T. N. Darling, C. Yee, J. W. Bauer, H. Hintner, and K. B. Yancey, J. Clin. Invest. 103, 1371 (1999). 13 R. B. Nagle, J. Histochem. Cytochem. 49, 1063 (2001). 14 M. Neira, V. Danilova, G. Hellekant, and E. A. Azen, Mamm. Genome 12, 60 (2001).
[22]
ANALYSIS OF GENE EXPRESSION
265
Arcturus has released its own version of the RNA isolation kit for use with the LCM caps. This kit is reported to have been optimized for use with the very small numbers of cells isolated during capturing, but to date there are no independent data evaluating the efficiency of this kit. Following RNA isolation, or as part of the isolation procedure, the sample needs to be treated with RNase-free DNase I to remove any contaminating DNA. In the Stratagene Total RNA Microprep kit the DNase treatment is included during the RNA isolation procedure. For other RNA isolation protocols, the DNase treatment is usually carried out after the RNA has been redissolved in DEPC– water by incubating with 20 U of DNase I for 1 to 2 hr at 37◦ . The DNase I is removed with a phenol–chloroform wash followed by RNA precipitation in isopropanol in the presence of sodium acetate and glycogen carrier (10 µg/µl) or by purifying the RNA by binding it to a column and washing twice before eluting the RNA from the column. Appropriate columns include those used for RNA isolation or Microcon-100 columns (Millipore, Bedford, MA). Once again, the columns with the smaller surface area allow the RNA to be eluted in a smaller volume. Alternatively, if the downstream application is reverse transcriptase–polymerase chain reaction (RT-PCR) the primers can be designed to cross intron–exon boundaries to prevent the amplification of any contaminating genomic DNA.
RNA Analysis Reverse Transcriptase–Polymerase Chain Reaction RT-PCR is often used to measure the gene expression for particular genes of interest in RNA isolated from LCM samples. In some cases the RNA lysis buffer is taken straight from the caps to the RT-PCR, but it is more usual to isolate the RNA completely before commencing downstream analysis. The RNA can be reverse transcribed to cDNA using random hexamers or oligo(dT)primers. Gene-specific primers can then be used for the PCR. This method will allow the examination of multiple genes from one reverse transcription reaction. Alternatively, the RT-PCR can be performed as a one-step reaction using the gene-specific primers for the reverse transcriptase reaction and the amplification reaction. Real-time PCR is being used where available for the detection of gene expression from LCM RNA samples, as the sensitivity of this technique requires less starting material than a standard PCR run to produce a reliable result. The method described here uses the Applied Biosystems equipment and reagents, but any of the real-time PCR systems can be used for this application. A real-time RT-PCR is prepared using the TaqMan One-Step RT-PCR Master Mix Reagents Kit (PE Applied Biosystems). For precise quantitative analysis of gene expression primers and probe for a control gene, for example the
266
GENETIC APPLICATIONS
[22]
Pre-developed TaqMan Assay Reagents [Endogenous Control Ribosomal RNA Control (18s rRNA)] (PE Applied Biosystems), can be included in the RT-PCR reactions. Ten microliters of total RNA isolated from LCM captured cells is analyzed in a 25 µl reaction containing 1× Master Mix, 1× MultiScribe and RNase Inhibitor Mix, 50–900 nM gene-specific forward primer, 50–900 nM gene-specific reverse primer, 50–250 nM gene-specific probe, and 1× control Primer and Probe Mix. The concentration of the gene-specific primers and probe needs to be optimized for each reaction. The multiplexing of the control gene and the gene-specific primers and probe also needs to be individually assessed. RT-PCR reactions are performed in the ABI Prism 7700 Sequence Detection System (PE Applied Biosystems). The thermal cycle conditions consist of reverse transcription at 48◦ for 30 min, denaturation at 95◦ for 10 min, followed by 40 cycles of 95◦ for 15 sec and 60◦ for 1 min. Data from the reaction are collected and analyzed by the complementary computer software. Reverse transcription can also be performed with random hexamers or oligo(dT)primers as described previously and the real-time PCR performed using the TaqMan Universal PCR Master Mix (PE Applied Biosystems). SYBR Green detection methods using the appropriate master mix can also be used for the detection of the amplified product with a melting curve analysis ensuring the specificity of the amplification. The sensitivity of this detection technique allows the reliable detection of sequences from as few as 100 cells (Fig. 2). The number of cells required for detection varies with cell type as different cell types contain different amounts of RNA. Where available, cells grown in culture can be used to estimate the number of cells required for the detection of specific genes. Adherent cells can be grown directly on the microscope slides before fixation and staining for LCM. Precise cell numbers captured are easier to calculate using this method but caution should be exercised in equating cell numbers captured from tissue culture cells to cell numbers captured from tissue sections, as the cells cultured on the slides are not subjected to sectioning and are therefore whole cells. The cells isolated from sections are not all whole, so consequently the RNA yield is lower. In our experience, there is a difference of about 100-fold in the RNA yield for smooth muscle cells. cDNA Microarrays RNA isolated from LCM captured cells can also be applied to cDNA or oligonucleotide microarrays. Before labeling for array analysis, the mRNA is amplified linearly11,15 to increase the amount of sample but still reflect the proportions of message for each gene in the original sample. 15
J. Eberwine, H. Yeh, K. Miyashiro, Y. Cao, S. Nair, R. Finnell, M. Zettel, and P. Coleman, Proc. Natl. Acad. Sci. U.S.A. 89, 3010 (1992).
[22]
ANALYSIS OF GENE EXPRESSION
267
FIG. 2. Real-time RT-PCR following LCM. Human airway smooth muscle cells were grown on microscope slides before 70% ethanol fixation and rapid H&E staining. Ten, 100, or 1000 cells were captured on HS caps (n = 6) and the RNA isolated using the Stratagene Total RNA Microprep kit. Real-time RT-PCR was used to quantify the expression of a control and target gene using a multiplex RT-PCR with two dyes. One hundred nanograms of RNA isolated from human cultured airway smooth muscle cells was used as a positive control and the absence of RNA template (NTC) as the negative control.
Linear Amplification. The RNA is isolated as described above and, after DNase I treatment, resuspended in 11 µl DEPC–water. One microliter of RNA is kept as the negative control for real-time RT-PCR. Reverse Transcription. The remaining 10 µl is mixed with 1 µl 0.5 mg/ml T7oligo(dT) primer (5 TCTAGTCGACGGCCAGTGAATTGTAATACGACTCAC TATAGGGCGT21 –3 )11 and heated to 70◦ for 10 min. The sample is immediately placed on ice while 4 µl first strand reaction buffer, 2 µl 0.1 M DTT, 1 µl RNasin and 1 µl 10 mM dNTPs [Life Technologies (Invitrogen), Carlsbad, CA] are added. Incubation at 42◦ for 5 min is carried out before the addition of 1 µl Superscript II (Life Technologies) and the continuation of the incubation at 42◦ for 1 hr. To the sample is added a cocktail containing 92 µl water, 30 µl 5× secondstrand synthesis buffer, 3 µl 10 mM dNTPs, 4 µl DNA polymerase I, 1 µl RNase H, and 1 µl Escherichia coli DNA ligase and this mix is incubated at 16◦ for 2 hr. Two microliters of T4 DNA polymerase is then added and the sample incubated at 16◦ for a further 15 min. The enzymes are then inactivated by heating to 70◦ for 10 min and the cDNA extracted by the addition of 150 µl phenol–chloroform and spinning at 14,000 rpm for 5 min. The aqueous phase is washed 3 times with 500 µl water in a Microcon-100 column (Millipore) and eluted in 8 µl of water. In Vitro Transcription. Using the Ampliscribe T7 transcription kit (Epicentre Technologies, Madison, WI) a cocktail of 2 µl 10× Ampliscribe T7 buffer, 1.5 µl
268
GENETIC APPLICATIONS
[22]
each of 100 mM ATP, UTP, CTP, and GTP, 2 µl 0.1 M DTT, and 2 µl T7 RNA polymerase is added to the cDNA. All reagents except the enzymes should be at room temperature before the cocktail is mixed. The mixture is incubated at 42◦ for 3 hr before the addition of 1 µl DNase I and a further 20 min at 37◦ . The amplified RNA (aRNA) is then washed three times using a Microcon-100 column and eluted in 11 µl water. One microliter is kept for analysis by real-time RT-PCR to check the efficiency of the amplification and the remaining 10 µl is carried on for a second round of amplification. Second Round Amplification. To the 10 µl aRNA is added 1 µl random hexamers (1 µg/µl) and the sample is incubated at 70◦ for 10 min before being chilled on ice and equilibrated to room temperature. To this is added 4 µl 5× first-strand buffer, 2 µl 0.1 M DTT, 1 µl 10 mM dNTPs, and 1 µl RNasin. After 5 min at room temperature 1 µl Superscript II is added. After a further 5 min at room temperature the mixture is incubated at 37◦ for 2 hr. One microliter of RNase H is then added and the incubation continued at 37◦ for a further 20 min before inactivating the enzyme at 95◦ for 2 min and placing on ice. One microliter of 0.5 mg/ml T7 oligo(dT)primer is added and the mixture incubated at 70◦ for 5 min followed by 42◦ for 10 min. To this is added 90 µl water, 30 µl 5× second-strand synthesis buffer, 3 µl 10 mM dNTPs, 4 µl DNA polymerase I, and 1 µl RNase H and the mixture is incubated at 16◦ for 2 hr. Two microliters T4 DNA polymerase is added and the incubation is continued at 16◦ for a further 10 min. The sample is then heated to 65◦ for 10 min before being extracted with 150 µl phenol–chloroform and purification using the Microcon-100 columns. The cDNA is then put through a second round of T7 in vitro transcription as described above (Fig. 3) before a final purification ready for labeling using a standard microarray labeling protocol. After the final round of amplification, 1 µl of the aRNA is taken for real-time RT-PCR to check the efficiency of the amplification reaction. Following the linear amplification of the mRNA, the aRNA can be labeled using a standard microarray labeling reaction before hybridization with the microarray using a standard protocol for a microarray experiment. It has been reported that between 20013 and 100011 cells are enough for RNA linear amplification and subsequent microarray analysis. DNA Analysis DNA Isolation DNA can be isolated from fresh frozen tissue samples but can also be isolated from paraffin embedded samples providing a much wider set of source material for studies of genes using DNA-based techniques. Ten micron tissue sections are cut on the cryostat and mounted on clean, flat microscope slides. Sections are stained appropriately for the identification of the cells types being studied. Hematoxylin and eosin are commonly used. Cells are captured onto the standard or HS LCM caps and the caps stored in a manner similar to that described for RNA isolation. Once again DNA isolation is carried out using scaled-down standard isolation
[22]
ANALYSIS OF GENE EXPRESSION
269
FIG. 3. Linear amplification of RNA. Schematic of the amplification method used for increasing the amount of mRNA available for downstream analysis. aRNA, Amplified RNA.
methods. Twenty to 50 µl of extraction buffer containing 10 mM Tris pH 8.0, 2 mM EDTA, 0.2% Tween 20, and 200 µg/ml proteinase K is incubated with the cells on the cap (in an inverted 0.5-ml Eppendorf tube) at 37◦ overnight. The following day the mixture is heated to 100◦ for 10 min to inactivate the proteinase K and 3 to 5% of the resultant solution can be used directly as PCR template. The QIAamp tissue kit (Qiagen, Valencia, CA) has been reported to produce quality DNA. Loss of Heterozygosity and Mutation Analysis Standard PCR conditions using primers designed to flank mutation sites can be used to detect mutations in LCM isolated samples. One or two rounds of PCR
270
GENETIC APPLICATIONS
[22]
amplification consisting of 35 to 40 cycles followed by direct sequencing has been used to identify FAS gene mutations in prostatic intraepithelial neoplasia and concurrent carcinoma,16 androgen-regulated homeobox gene NKX3.1 mutations in benign and malignant prostate epithelium,17 and Ki-ras and/or p53 mutations in tumor genotyping.18 Takayama et al. also used DNA amplification to look for loss of heterozygosity at four known polymorphism sites in prostatic intraepithelial neoplasia.16 Methylation Analysis Patel et al.19 have described a method for the analysis of the methylation state of a gene (p16Ink4a) DNA isolated from LCM captured cells. Briefly, the DNA is isolated from the cells by the addition of 50 µl lysis buffer (0.5% Tween 20, 1 mM EDTA pH 8.0, 50 µM Tris pH 8.5, and 0.5 µg/µl proteinase K) and incubation overnight, inverted, at 37◦ . The samples are then incubated at 95◦ for 8 min before precipitation of the DNA by the addition of 1.8 ml 100% ethanol in the presence of 2 µl glycogen (20 mg/ml). The DNA is washed in 70% ethanol twice and resuspended in 10 µl water. The samples are digested overnight at 37◦ with EcoRI in a total volume of 20 µl before denaturation at 75◦ with 2 µl 3 M NaOH. Bisulfite modification is then carried out by the addition of 250 µl 4.8 M sodium bisulfite and 14 µl 20 mM hydroquinine before overlaying the samples with light mineral oil and incubating at 55◦ for 5 hr. The DNA is purified using Centricon-30 columns (Millipore) and eluted in 100 µl water. The samples are then desulfonated with 4.5 µl 3 M NaOH and neutralized with 28 µl 5 M ammonium acetate. The DNA is precipitated with 3 volumes of 100% ethanol, in the presence of 1 µg glycogen, overnight at −20◦ and then washed with 70% ethanol and resuspended in 20 µl water. Two rounds of nested PCR can then be performed using primers that have been designed with all of the cytosines replaced with thymines to enable the amplification of the bisulfite-treated DNA. The unmethylated cytosines are converted to uracils during the modification with sodium bisulfite. During the subsequent PCR the uracils are converted to thymine residues. This method allows site-specific and region-specific methylation to be studied.19 16
H. Takayama, T. Takakuwa, Z. Dong, N. Nonomura, A. Okuyama, S. Nagata, and K. Aozasa, Lab. Invest. 81, 283 (2001). 17 D. K. Ornstein, M. Cinquanta, S. Weiler, P. H. Duray, M. R. Emmert-Buck, C. D. Vocke, W. M. Linehan, and J. A. Ferretti, J. Urol. 165, 1329 (2001). 18 D. Dillon, K. Zheng, and J. Costa, Exp. Mol. Pathol. 70, 195 (2001). 19 A. C. Patel, C. H. Anna, J. F. Foley, P. S. Stockton, F. L. Tyson, J. C. Barrett, and T. R. Devereux, Carcinogenesis 21, 1691 (2000).
[23]
ANALYSIS OF SPECIFIC GENE EXPRESSION
271
[23] Analysis of Specific Gene Expression By GEORGIA LAHR, ANNA STARZINSKI-POWITZ, and ANETTE MAYER Introduction One of the key points for understanding the molecular basis of development or carcinogenesis is the analysis of gene expression in specific cell populations. To isolate these cell populations laser-assisted microdissection is commonly used. Here, the P.A.L.M. Robot-MicroBeam system allows contamination-free isolation of single cells or defined cell clusters from frozen or archival tissue sections.1 After laser microbeam microdissection (LMM) cells of interest are catapulted by the force of the laser directly into the cap of a common microfuge tube. This catapulting process is called “laser pressure catapulting” (LPC).1,2 Microdissected cells are now suitable for subsequent experiments. Many different methods for RNA isolation and RT-PCR analysis of cells from cryopreserved and archival tissues have been published in the past few years.3–7 The decision to use paraffin-embedded or frozen tissue depends on subsequent experiments and tissue supply. Paraffin-embedded tissues may be used for retrospective studies and may permit tracking over long periods of time. However, cross-linking of proteins with nucleic acids by formalin, which is the most common fixative, leads to RNA or DNA fragmentation after nucleic acid isolation procedures, making them useless for some downstream methods. Because of these strand breaks the length of an amplifiable RT-PCR product from archival tissue should not exceed 380 bp. For isolation of longer fragments and of nearly undegraded RNA (e.g., for library construction or array technique) cryopreserved tissue should be chosen since the quality of the isolated RNA is much higher. Cryopreserved tissue sections require some changes in protocol (see below) since mild fixatives, which keep RNA and DNA almost undegraded, are not able to completely inhibit endogenous RNase activity. Starting with archival or cryopreserved laser-microdissected cells, we established improved protocols for the isolation of RNA and downstream RT-PCR analysis, including Real-time RT-PCR. The combined LMM and LPC provide 1
K. Sch¨utze and G. Lahr, Nat. Biotechnol. 16, 737 (1998). K. Sch¨utze, H. P¨osl, and G. Lahr, Mol. Cell Biol. 44, 735 (1998). 3 M. R. Bernsen, H. B. Dijkman, E. de Vries, C. G. Vigdor, D. J. Ruiter, G. J. Adema, and G. N. P. Muijen, Lab. Invest. 78, 1267 (1998). 4 L. Jin, C. A. Thompson, X. Qian, S. J. Kuecker, E. Kulig, and R. V. Lloyd, Lab. Invest. 79, 511 (1999). 5 G. Lahr, Lab. Invest. 80, 1477 (2000). 6 G. Lahr, M. Stich, K. Sch¨ utze, P. Bl¨umel, H. P¨osl, and W. B. J. Nathrath, Pathobiology 68, 218 (2000). 7 Y. Imamichi, G. Lahr, and D. Wedlich, Dev. Genes Evol. 211, 361 (2001). 2
METHODS IN ENZYMOLOGY, VOL. 356
Copyright 2002, Elsevier Science (USA). All rights reserved. 0076-6879/02 $35.00
272
GENETIC APPLICATIONS
[23]
an extremely useful tool for selecting and pooling morphologically similar cells, which opens the field for the analysis of specific gene expression or differential gene expression screens between microarrays. Step I: Cell/Specimen Preparation Buffers, Reagents, and Equipment 1. For tissue preparation: 1.35 µm thin polyethylene–naphthalene membrane (PEN; P.A.L.M. Microlaser Technologies AG, Bernried, Germany) 0.1% poly-L-lysine, Mayer’s hematoxylin, eosin, and mineral oil (all Sigma, Deisenhofen, Germany) 2. For laser microbeam microdissection (LMM) and laser pressure catapulting (LPC): Robot-MicroBeam (P.A.L.M. Microlaser Technologies AG, Bernried, Germany) coupled onto an inverted Axiovert 135 microscope (Carl Zeiss, G¨ottingen, Germany) Procedure Specimen Preparation Formalin-preserved tissues. We commonly use 5 µm serial sections of a routinely formalin-fixed and paraffin-embedded tissue block obtained from the pathology department. 1. Mount the tissue sections either directly onto a common glass slide or onto a membrane-mounted glass slide. 2. For membrane mounting: use a droplet of 100% ethanol to attach the 1.35 µm thin PEN membrane to the glass slide. 3. Fix the edges of the membrane with nail polish. 4. Cover the nail polish after drying with an autoclave tape to avoid its dissolving during deparaffinization in xylene. 5. Cover the membrane with 0.1% poly-L-lysine prior to tissue mounting and air dry the coated membrane. 6. Mount the tissue section on the membrane, deparaffinize, and stain the sections with hematoxylin–eosin as usual. Cryopreserved tissues. We prepare 8 µm serial sections of snap-frozen tissue derived from surgery on a cryostat. 1. Mount the PEN membrane on a glass slide (see points 2–4 above). 2. UV irradiation of the membrane for 20 min renders the hydrophobic membrane more hydrophilic.
[23]
ANALYSIS OF SPECIFIC GENE EXPRESSION
3. 4. 5. 6.
273
Mount the frozen tissue section on the membrane. Fix the sections with 70% ethanol for 30 sec and rinse briefly twice in PBS. Stain sections with hematoxylin for 10 sec and rinse again twice in PBS. Dry sections by incubation for 15 sec each in 70%, 90%, and 100% ethanol.
Laser Microbeam Microdissection and Laser Pressure Catapulting 7. Insert the object slide onto the microscope stage (Figs. 1A, 2A, 2E, 3A). 8. Moisten the cap of a microfuge tube with a 2 µl droplet of mineral oil by using a micropipette tip, or apply a 5 µl droplet of lysis buffer (included in PureScript isolation kit) to the cap.
A
D
B
E
C
FIG. 1. Isolation and RT-PCR analysis of microdissected archival cells. Microscopic illustration of one specific experiment using LMM and LPC to capture pooled single cells from membrane-mounted archival hematoxylin–eosin stained tissue slices of a differentiated colon adenocarcinoma. The view shows specimen before LMM (A), the remaining tissue after LMM (B), the remaining tissue or cells after LPC (C), and the catapulted and captured cells in the cap of a conventional microfuge tube (D). The experiment was performed by using a 40× objective. Panel E shows the MspI digests of Ki-ras2 RT-PCR-amplified codon 12 products (217 bp RT-PCR fragments) which were separated on a 1.5% ethidium bromide-stained agarose gel. The uncut RT-PCR fragment of 7 pooled tumor cells (shown in D) is indicated by (1). The MspI digest of this sample is shown in 2. An RT-PCR sample of about 100 isolated SW480 cells digested with MspI was loaded in lane S. A 100 bp ladder size marker (M) was run in parallel. The sizes of the 100 bp ladder are indicated on the right side as are the uncut 217 bp RT-PCR fragments (white arrow).
274
[23]
GENETIC APPLICATIONS
A
B
C
D
E
F
G
H
I
J
FIG. 2. Isolation and RT-PCR analysis of microdissected cryopreserved cells. Microscopic illustration of two specific experiments using LMM and LPC to capture epithelial (A–D) or stromal (E–H) cells from membrane-mounted cryopreserved hematoxylin-stained tissue sections of an endometriotic lesion. The view shows a specimen before LMM (A and E), the tissue after LMM (B and F), the remaining tissue after LPC (C and G), and the catapulted cells captured in the cap of a conventional microfuge tube (D and H; all 40× objective). (I) GP130 RT-PCR analysis of microdissected epithelial (1) or stromal (2) cells derived from an endometriotic lesion. Second PCR products (220 bp) were loaded on 1.5% ethidium bromide-agarose gels. C1 is a H2O control of 1st PCR used as a template in 2nd PCR; C2 is control containing H2O instead of DNA. Long β-actin fragments derived from about 2000 dissected endometrial cells are shown in (J), where (3) is from from 1st and (4) is from 2nd (nested ) PCR. The sizes of the 100 bp ladder are indicated on the right side. The arrowheads indicate the faint signal of the 1st and the arrows correspond to the 2nd PCR products. M, 100 bp ladder size marker.
[23]
A
ANALYSIS OF SPECIFIC GENE EXPRESSION
B
275
C
D
FIG. 3. Isolation, RT, and real-time PCR analysis of microdissected archival cells. In this example LMM and LPC were used to capture a cell cluster of about 100 cells from an archival hematoxylinstained tissue slice of a follicular variant of a papillary thyroid carcinoma. The view shows specimen before (A) and after microdissection and LPC (B), and the catapulted cells (arrowhead) in the cap (C; all 40× objective). The melting curve analysis after 50 cycles in a real-time PCR reaction from RET/PTC1 LightCycler SYBR Green I assay is shown in (D). Here, the first negative derivative of the fluorescence (-dF/dT ) is plotted as a function of temperature (◦ C; Tm). The RET/PTC1 melting curve analysis of the dissectates composed of about 100 laser-microdissected cells is shown in black (1). The dotted line (T) indicates the translocation-positive control, derived from the tumor cell line TPC-1.
9. Insert the “prepared” cap into the LPC collector. 10. Focus the laser microbeam through a 20× or 40× dry objective lens. Energy settings are dependent on the absorption behavior of the specimen and on the transmission rate of the objective lens. 11. For microdissection adjust the laser energy either to solely cut the tissue (glass-mounted; Fig. 2B) or to cut the entire membrane-tissue stack (membranemounted; Figs. 1B, 3B, 3F). 12. For LPC adjust the energy settings. They should be sufficiently high to catapult the specimen into the microfuge cap, which is now centered above the line of laser fire by a special LPC-collector device. 13. Inspect the catapulted cells within the cap by using the “checkpoint” position of the system (Figs. 1D, 2D, 2H, 3C). 14. Remove the cap including the captured cells from the collector device.
276
GENETIC APPLICATIONS
[23]
15. Cover the catapulted cells by a total volume of 30 µl of RNA lysis buffer. 16. Top the caps with the remaining tube and process as described below. Notes to Step I Mounting of the PEN membrane takes less than 1 min per slide. The membranecovered slides can be stored at room temperature until needed. Step II: RNA Isolation Buffers, Reagents, and Equipment Enzymes and buffers are derived from the PUREscript RNA Isolation Kit. PUREscript RNA Isolation Kit (BIOzym Diagnostik, Hess. Oldendorf, Germany) Glycogen (MB grade; 20 µg/µl) (Roche Diagnostics GmbH, Mannheim, Germany) Procedure Preparation of Total RNA from Microdissected Cell Samples Cell lysis 1. Add 30 µl Cell Lysis Solution to the catapulted cells in the cap and pipette up and down 15–30 times to lyse the cells. 2. Centrifuge at 13,000–16,000g for 1 min to pellet the lysis solution including the cells. Protein–DNA precipitation 3. Add 10 µl Protein–DNA Precipitation Solution to the cell lysate. 4. Invert tube gently 10 times and place tube into an ice bath or cryo pack for 5 min. 5. Centrifuge at 13,000–16,000g for 3 minutes. The precipitated proteins and DNA will form a tight white pellet. RNA precipitation 6. Pour the supernatant containing the RNA (leaving behind the precipitated protein–DNA pellet) into a clean microfuge tube containing 30 µl 100% 2-propanol. 7. Add 1.5 µl of glycogen (1 : 10 diluted with H2O) as a carrier. 8. Mix the sample by inverting gently 50 times. Optional: incubate at 4◦ for 20 min. 9. Centrifuge at 13,000–16,000g for 5 min; the RNA will be visible as a small, translucent pellet.
[23]
ANALYSIS OF SPECIFIC GENE EXPRESSION
277
10. Pour off the supernatant and drain tube briefly on clean absorbent paper. Add 30 µl 70% ethanol and invert the tube several times to wash the RNA pellet. 11. Centrifuge at 13,000–16,000g for 3 min. Carefully pour off the ethanol. 12. Invert and drain the tube on clean absorbent paper and allow sample to air dry 10–15 min or for a maximum of 5 min in a Speed Vac. RNA hydration 13. Add 5 µl RNA Hydration Solution. 14. Allow RNA to rehydrate for at least 30 min on ice. Alternatively, store RNA sample at −70◦ to −80◦ until use. 15. Before use vortex sample vigorously for 5 sec and pulse spin. 16. Store purified RNA sample at −70◦ to −80◦ or proceed with cDNA synthesis. Notes to Step II Longer incubation and centrifugation steps may increase total RNA yield. Chemicals and staining dyes, such as hematoxylin and eosin, and dyes from an optional immunostaining procedure may inhibit the downstream RT-reaction, but this effect is negated by purification of the “cell extracts” by RNA isolation.5 Step III: Reverse Transcription Buffers, Reagents, and Equipment ExpeRT-PCR Kit (Hybaid-AGS, Heidelberg, Germany) dNTP solution containing all four dNTPs (10 mM each) RNase inhibitor (40 units/µl) (Hybaid-AGS, Heidelberg, Germany) Hexanucleotide Mix, 10× concentration (random hexamers; Roche Diagnostics GmbH, Mannheim, Germany) Procedure Preparation of cDNA 1. RNA must be denatured for 5 min at 80◦ and immediately transferred to ice before being added to the reaction mix. 2. Set up the RT reaction mix in a total volume of 12.5 µl: 10× ExpeRT-PCR-Buffer, complete (including 1.75 mM MgCl2 final concentration) DMSO (2% final concentration) Template: total RNA 10× Random hexamers
1.25 µl 0.25 µl 5 µl 0.75 µl
278
[23]
GENETIC APPLICATIONS
10 mM dNTP-Mix 40 units/µl RNase inhibitor H2O, double distilled, DEPC treated 5 units/µl AMV Reverse Transcriptase
0.25 µl 0.06 µl 4.7 µl 0.25 µl
3. Incubate 60 min at 42◦ . 4. Inactivate AMV Reverse Transcriptase (10 minutes at 80◦ ). Notes to Step III a. To reduce the chance of contamination with exogenous nucleic acids, prepare and use a special set of reagents and solutions for RNA isolation, RT and PCR only. b. Temperature for RT: any template denaturation to overcome secondary structures in the RNA should be performed in the absence of the AMV-Reverse Transcriptase, as the enzyme is denatured at elevated temperatures. AMV-Reverse Transcriptase has optimum activity at 42◦ [using oligo(dT) primers and random hexamers] but can be used at temperatures of up to 60◦ (using gene-specific primers) to minimize problems with secondary structures. c. RNase inhibitor: the use of RNase inhibitor is optional.
Step IV: PCR Analysis Buffers, Reagents, and Equipment 1. For PCR in the Thermal Block Cycler: ExpeRT-PCR Kit (Hybaid-AGS, Heidelberg, Germany) Thermal block cycler instrument 9600 (PerkinElmer Applied Biosystems GmbH, Weiterstadt, Germany) 2. For real-time PCR in the LightCycler: LightCycler instrument (Roche Diagnostics GmbH, Mannheim, Germany) LightCycler capillaries (Roche Diagnostics GmbH) LightCycler FastStart DNA Master SYBR Green I (Roche Diagnostics GmbH) 3. Special equipment: Barrier tips for micropipettes Microfuge tubes (0.5 ml, thin-walled for PCR amplification) Thermal block cycler programmed with desired amplification protocol Water baths or heating blocks, preset to 80◦ and to 42◦ Ice-water bath or cryopack 1.5% Agarose gels in TAE buffer containing 0.25 µg/ml ethidium bromide TAE buffer: 40 mM sodium acetate (pH 8.3), 40 mM Tris base, 2 mM EDTA
[23]
279
ANALYSIS OF SPECIFIC GENE EXPRESSION
Procedure A. PCR Reactions in Block Cycler 1. Pipette the following PCR reaction mix (1st or 2nd PCR) in a total volume of 25 µl: 10× ExpeRT-PCR-Buffer (complete) (including 1.75 mM MgCl2 final concentration) Template: aliquot of RT- or 1st PCR reaction 10 pmol/µl 5 -Primer 10 pmol/µl 3 -Primer 10 mM dNTP-Mix 2.5 units/µl Proof Sprinter Taq/Pwo Mix DMSO (2% final concentration); note: 2% DMSO is already included in the RT- and 1st PCR mix MgCl2 (1.75–5 mM MgCl2 final concentration) H2O double distilled (adjust to a total volume of 25 µl)
2.5 µl 3.1–10 µl 1 µl 1 µl 0.5 µl 0.2 µl x µl x µl x µl
2. Subject the samples to 1 cycle at 94◦ for 2 min, 35–45 cycles at 93◦ for 30 sec, at 56–62◦ for 30 sec (this is dependent on the primer composition and on the amount of template DNA), at 72◦ for 1 min; at 72◦ for 10 min, and storage at 4◦ . 3. For detection of PCR products analyze the sizes of the amplified products of the 1st and 2nd PCR reaction, using a 15–20 µl aliquot of each reaction, by agarose gel electrophoresis. Notes to Step IV, A a. Sense and antisense primers (10 µM each) in H2O. Each primer should be 20–25 nucleotides in length and have a GC content of approximately 50–60%. b. Oligonucleotide primers synthesized on an automated DNA synthesizer can generally be used in RT-PCR without further purification. c. Magnesium is required by both the AMV and the Taq/Pwo mix and should be optimized for each PCR reaction. d. Times and temperatures may need to be adapted to suit other types of equipment and reaction volumes. e. If the thermal cycler is not equipped with a heated lid, use either mineral oil or paraffin wax to prevent evaporation of liquid from the reaction mixture during RT and PCR. f. Controls containing H2O instead of DNA (negative) and containing cDNA or DNA (positive), e.g., from a cell line, are always run in parallel. The results of a specific RT-PCR analysis using archival microdissected cells from a colon adenocarcinoma are shown in Fig. 1E. Related single and oligo cell
280
[23]
GENETIC APPLICATIONS
analyses have been published.1,5 Briefly, after microdissection and RT-PCR analysis one aliquot of the 2nd Ki-ras2 PCR reaction, example shown for the seven pooled tumor cells in Figs. 1A–1D and for 100 pooled immunostained SW480 cells, was digested with the restriction enzyme MspI. The SW480 tumor cell line differs from the wild-type Ki-ras2 at codon 12.8,9 MspI cuts within the 217 PCR fragment only if codon 12 is wild type.10 The products of the nested PCR and the MspI digests were fractionated on 1.5% agarose–ethidium bromide gels to yield the “uncut” DNA fragments of the appropriate length (Fig. 1E). This indicates that both the microdissected tumor cells and the SW480 cells carry a mutation in codon 12 of the Ki-ras2 gene. The results of a specific RT-PCR analysis using microdissected cells from cryosections of endometriotic lesions are shown in Figs. 2I–2J. Tissue-specific expression of GP130 mRNA was investigated by first microdissecting either epithelial (Figs. 2A–2D) or stromal (Figs. 2E–2H) cells. RT-PCR analysis of 100– 300 cells detects GP130 mRNA in both cell populations; aliquots of the second GP130 PCR reaction are examples shown in Fig. 2I. The use of cryosections allows the isolation of large RNA fragments; this is demonstrated by the amplification of a 1.7 kb (first PCR) and a 755 bp (second PCR) β-actin fragment (Fig. 2J). B. Real-Time PCR Used for Archival Tissue 1. Pipette 60 µl from the LightCycler-FastStart Reaction Mix SYBR Green I into the LightCycler-FastStart Enzyme (“HotStart” reaction mix), mix gently, and protect from light. 2. Pipette the LightCycler-Sybr Green I PCR reaction mix (1st or 2nd PCR) in a total volume of 20 µl. “HotStart” reaction mix Template: RT- or 1st PCR reaction product 10 pmol/µl 5 -Primer 10 pmol/µl 3 - Primer 25 mM MgCl2 (3 mM final concentration) H2O double distilled (adjust at a total volume of 25 µl)
2 µl 6.5 µl or 3 µl 1 µl 1 µl 1.6 µl x µl
3. The amplification is performed in the LightCycler running 50 cycles for 15 sec at 95◦ , 5 sec at 56–60◦ (depending on the oligonucleotides), and 10 sec at 72◦ , starting with a 10 min denaturation at 95◦ . 8
W. Jiang, S. M. Kahn, J. G. Guillen, S.-H. Lu, and B. Weinstein, Oncogene 4, 923 (1989). M. Verlann-de Vries, M. E. Bogaard, H. van den Elst, J. H. van Boom, A. J. van der Eb, and J. L. Bos, Gene 50, 313 (1986). 10 A. Haliassos, J. C. Chomel, S. Grandjouan, J. Kruh, J. C. Kaplan, and A. Kitzis, Nucleic Acids Res. 17, 8093 (1989). 9
[23]
ANALYSIS OF SPECIFIC GENE EXPRESSION
281
4. For detection of PCR products assess the specificity of the amplified PCR products by performing melting curve analysis with the LightCycler software. Melting analysis was performed for 0 sec denaturation at 95◦ , hybridization for 15 sec at 65◦ (depending on the target sequence), and continuously increasing the temperature (0.1◦ /sec) from 65◦ (depending on the target sequence) to 95◦ . 5. If desired, analyze the sizes of the amplified products of the PCR reaction in a 15–20 µl aliquot of each of the reactions by agarose gel electrophoresis. Notes to Step IV, B a. SYBR Green I is a dye, specific for double-stranded DNA. It can be used for the amplification on every DNA or cDNA target. Each protocol needs adaptation to the appropriate reaction conditions. b. The amplicon size should not exceed 1 kb in length. For optimal results, select a product length of 700 bp. c. Use primers at a final concentration of 0.3–1 µM each. A recommended starting concentration is 0.5 µM each. d. For specific and efficient amplification using the LightCycler instrument, it is essential to optimize the target-specific MgCI2 concentration that may vary from 1 to 5 mM. Real-time RT-PCR analysis has become an indispensable tool for rapid testing of various biological specimens, such as blood and biopsies, for the presence of gene mutations, translocations, and microorganisms. The results of a real-time RT-PCR to analyze a specific translocation using archival microdissected cells from a follicular variant of a papillary thyroid carcinoma are shown in Fig. 3D. There, the tumor tissue section was mounted directly onto a conventional glass slide. After microdissection and RT one aliquot was applied to real-time PCR. The melting curve analysis after 50 cycles from the RET/PTC1 assay (see reference 6) is shown in Fig. 3D, where the first negative derivative of the fluorescence (−dF/dT ) was plotted as a function of temperature (◦ ; Tm). Analysis of about 100 lasermicrodissected cells resulted in a melting peak at about 87◦ , which corresponds to the expected melting peak derived from the tumor cell line TPC-1 (T) used as a RET/PTC1 translocation-positive control. Acknowledgments The authors thank M. Stich and K. Sch¨utze for technical assistance and instrumental support. This work was supported by a grant of the Boehringer Ingelheim Stiftung to A.S.-P.
282
[24]
GENETIC APPLICATIONS
[24] Gene Discovery with Laser Capture Microscopy By MAURICIO NEIRA and EDWIN AZEN Introduction The isolation of pure cell populations, suitable for nucleic acid extraction, from complex tissues by laser capture microdissection (LCM) is the basis for the straightforward method of gene discovery presented in this article. We describe, in detail, a fast method for the construction of cDNAs from material obtained by LCM and an example leading to the finding of new genes specifically expressed in taste cells as opposed to the surrounding epithelium. Taste buds are onion-shaped specialized neuroepithelial cell structures of 50–100 cells, 50 µm in size and commonly found embedded in tongue taste papillae. These characteristics make taste buds suitable for microdissection using a 30- to 50-µm laser beam. A detailed procedure will be given for the method we used to make cDNAs from material obtained by laser capture microdissection, and for all other procedures involved in the gene discovery process we will point out the relevant stages and will refer the reader to the specific kit or source describing the detailed procedure. General Considerations Laser capture microdissection performed on frozen sections of rhesus monkey circumvallate taste papillae (easily visible in the back of the tongue) was used to obtain two separate populations of cells: taste buds and epithelial cells immediately surrounding the taste buds.1 The differential screening strategy for the discovery of genes, specifically expressed in taste buds, is as follows: 1. Construction of a λ ZAP II cDNA library from the microdissected taste buds 2. Plating of a portion of the taste bud library in duplicate filters 3. Hybridization with radioactively labeled complex cDNA probes obtained from the microdissected taste buds and microdissected surrounding epithelium 4. Selection of clones giving a positive signal with the taste bud cDNA probe but not with the control probe 5. Selection of clones not showing signals with either probe, as they may represent low-abundance genes missed out in the differential screening due to sensitivity 1
M. Neira, V. Danilova, G. Hellekant, and E. Azen, Mammalian Genome 12, 60 (2001).
METHODS IN ENZYMOLOGY, VOL. 356
Copyright 2002, Elsevier Science (USA). All rights reserved. 0076-6879/02 $35.00
[24]
GENE DISCOVERY WITH
LCM
283
issues involved in using complex cDNAs as probes. In order to determine to what cell population they are specific to, a number of clones from this set were chosen randomly, sequenced, PCR primers designed, and PCR analysis with specific primers performed on separate cDNAs from taste buds and from surrounding epithelium. Clones showing a PCR signal in the taste bud cDNA population but not in the cDNAs from surrounding epithelium are selected as potentially taste bud specific. 6. Final selection of clones as taste cell specific is made by RNA in situ hybridization. Reagents The following reagents and buffers are necessary for the cDNA synthesis and are purchased from different companies or prepared as follows: 1. Dynabeads mRNA purification kit from Dynal: contains oligo(dT)25 beads, 2× binding buffer, washing buffer, and magnetic stand 2. Superscript RT II from GIBCO: contains reverse transcriptase, 0.1 M DTT, 5× first-strand buffer 3. RNase inhibitor from GIBCO 4. T4 DNA polymerase and 10× T4 buffer from Epicentre 5. RNase H and RNase H buffer from Boehringer Mannheim 6. Terminal deoxynucleotidyltransferase (TdT) and 10× One-Phor All plus buffer, from Pharmacia 7. Taq polymerase, dATP, and dNTPs from Epicentre 8. 1 M TMAC from Sigma (a chemical that causes an oligonucleotide to hybridize based on length and not GC content which results in an improvement in the quality of the PCR product2 ) Preparation of different reaction mixtures is as follows: 1. 2× first-strand buffer: from 5× first-strand buffer with RNase-free water 2. Reverse transcriptase mix: 5× First strand buffer RNase-free water 0.1 M DTT 10 mM Each dNTPs mix RNase inhibitor (10 U/µl) 2
4 µl 11 µl 2 µl 1 µl 1 µl
K. N. Lambert and V. M. Williamson, Methods Mol. Biol. 69, 1 (1997).
284
GENETIC APPLICATIONS
[24]
3. T4 DNA polymerase reaction mix: Double distilled H2 O 10× T4 buffer 10 mM Each dNTPs Mix T4 polymerase (1 U/µl)
41.5 µl 5.0 µl 2.5 µl 1.0 µl
4. RNase H buffer: 20 mM Tris-HCl pH 8.0, 50 mM KCl, 10 mM MgCl2, 1 mM DTT (prepare fresh) 5. RNase H reaction mix: RNase H buffer RNase H (1 U/µl)
20 µl 0.5 µl
6. Terminal transferase mix: Double distilled H2 O 10× One-Phor All plus buffer 1.5 mM dATP TdT (22 U/µl)
14 µl 2 µl 3 µl 1 µl
7. 5× PCR buffer: 100 mM Tris-HCl pH 8.3, 12.5 mM MgCl2, 125 mM KCl, 0.25% Tween 20. 8. Primers: Chosen from Ref. 3. AL1T : 5 -ATT GGA TCC AGG CCG CTC TGG ACA AAA TAT GAA TTC (T)24-3 AL1 : 5 -ATT GGA TCC AGG CCG CTC TGG ACA AAA TAT GAA TTC-3 Note: The AL1 primer contains an EcoRI site at the 3 end and will be subsequently used for subcloning into λZAP II phage vector. 9. Taq polymerase reaction mix 1: Double distilled H2 O 5× PCR buffer 1 M TMAC AL1 T primer (20 pmol/µl) 10 mM Each dNTPs mix Taq polymerase (5U/µl) 3
C. Dulac and R. Axel, Cell 83, 195 (1995).
36.65 µl 10 µl 0.4 µl 1.45 µl 1 µl 0.5 µl
[24]
GENE DISCOVERY WITH
LCM
285
10. Taq polymerase reaction mix 2: Double distilled H2 O 5× PCR buffer 1 M TMAC AL1 T primer (1 pmol/µl) AL1 T primer (20 pmol/µl) 10 mM Each dNTPs mix Taq polymerase (5U/µl)
34.6 µl 10 µl 0.4 µl 1 µl 2.5 µl 1 µl 0.5 µl
11. Taq polymerase reaction mix 3: Double distilled H2 O 5× PCR buffer 1 M TMAC AL1 primer (20 pmol/µl) 10 mM Each dNTPs mix Taq polymerase (5U/µl)
35.6 µl 10 µl 0.4 µl 2.5 µl 1 µl 0.5 µl
Methods Laser Capture Microdissection For tissue procurement and sectioning we refer the reader to Ref. 1, and for laser capture microdissection to other chapters in this volume. One hundred to 300 laser beam shots of approximately 30–50 µm were made with the Laser Capture Microdissection instrument to transfer selected cells to the polymer film (in one cap). RNA Isolation RNA extraction was performed using the RNeasy Mini Kit (Qiagen Inc., Valencia, CA). A 50 µl final volume of denaturing solution without addition of 2-mercaptoethanol was used per cap and stored in tubes at −70◦ until ready for RNA extraction. Just before the extraction, following the procedures of the RNeasy Mini Kit, 2-mercaptoethanol was added. cDNA Synthesis The method outlined here was adapted from Ref. 2. Shown in square brackets are the names of the programs (detailed below) in the PTC-100 programmable thermal controller (MJ Research, Inc., Waltham, MA) used to perform the procedures in the protocol. The actual writing of the programs is only a suggestion
286
GENETIC APPLICATIONS
[24]
as they may change depending on the thermocycler and not for every step is it absolutely necessary to use the thermocycler. 1. Resuspend RNA in 25 µl RNase-free water. 2. Heat at 65◦ for 2 min [LCM65] then cool on ice. 3. Put 20 µl (100 µg) of Dynabeads in 0.5-ml centrifuge tube and place it in the magnetic stand. 4. Remove supernatant. 5. Remove from stand and resuspend in 25 µl of 2× binding buffer. 6. Remove buffer using magnetic stand. 7. Add 25 µl of 2× binding buffer to beads. 8. Add 25 µl of RNA from step 2 to beads (total volume now 50 µl) and incubate for 15 min at 22◦ [LCM22]. 9. Remove supernatant using magnetic stand. 10. Wash beads twice with 50 µl of washing buffer and remove supernatant. 11. Wash beads in 50 µl of 2× first-strand buffer and remove supernatant. 12. Add 19 µl of reverse transcriptase mix and heat at 37◦ for 2 min [LCMRT]. 13. Add 1 µl of reverse transcriptase and mix, continue with [LCMRT]. 14. Continue incubation at 37◦ for 15 min, then increase temperature to 42◦ for 45 min (mixing tube every 15 min) and heat at 65◦ for 10 min to inactivate the reverse transcriptase [LCMRT]. 15. Remove buffer and add 20 µl of T4 DNA polymerase reaction mix. Incubate at 16◦ for 1 hr and inactivate by heating at 74◦ for 10 min [LCMTRIM]. 16. Remove buffer and add 20 µl of RNaseH reaction mix. Incubate at 37◦ for 1 hr [LCM37H], remove buffer, then add 50 µl of 1 mM EDTA and heat at 75◦ for 5 min [LCM75IH]. Remove EDTA solution. 17. Add 20 µl of Terminal Transferase mix. Incubate at 37◦ for 15 min, then stop by adding 2 µl of 0.5 M EDTA and remove buffer [LCMTT]. 18. Add 50 µl of Taq polymerase reaction mix 1 (containing 29 pmol of AL1T primer). Carry out annealing and extension at 30◦ for 3 min, 40◦ for 3 min, and 72◦ for 5 min [LCM2ST]. Remove supernatant. 19. Add 50 µl of Taq polymerase reaction mix 2 (containing 50 pmol AL1 primer and 1 pmol AL1T primer) and heat at 95◦ for 2 min [LCM95], then transfer supernatant to another vial and save beads for future use (suggested 70% ethanol at 4◦ ). 20. Add oil to the new vial if necessary, incubate at 30◦ for 15 min, 40◦ for 15 min, 72◦ for 15 min, then heat at 95◦ for 2 min and amplify 15 cycles 95◦ for 1 min, 72◦ for 5 min with final extension at 72◦ for 30 min [LCMAMP]. (These conditions may change depending on the primers and polymerase being used.) 21. For gel electrophoresis analysis, if sample from final amplification is not seen, then reamplify by adding 5 µl of the reaction in Taq polymerase reaction
[24]
GENE DISCOVERY WITH
LCM
287
mix 3 (95◦ for 1 min, 72◦ for 5 min 15 times, and final extension at 72◦ for 7 min) [LCMRE]. 22. Proceed with subcloning into an appropriate vector or use the ds cDNA for labeling to be used as a probe. Programs LCM65 1. T = 65◦ , t 2. T = 22◦ , t LCM22 1. T = 22◦ , t LCMRT 1. T = 37◦ , t 2. T = 37◦ , t 3. T = 42◦ , t 4. T = 70◦ , t 5. T = 16◦ , t LCMTRIM 1. T = 16◦ , t 2. T = 74◦ , t 3. T = 37◦ , t LCM37H 1. T = 37◦ , t LCM75IH 1. T = 75◦ , t 2. T = 37◦ , t LCMTT 1. T = 37◦ , t 2. T = 30◦ , t LCM2ST 1. T = 30◦ , t 2. T = 40◦ , t 3. T = 72◦ , t LCM95 1. T = 95◦ , t 2. T = 30◦ , t LCMAMP 1. T = 30◦ , t 2. T = 40◦ , t 3. T = 72◦ , t 4. T = 95◦ , t
= 2 min =∞ = 15 min = 2 min (add reverse transcriptase after this incubation) = 15 min = 45 min (mix every 15 min) = 10 min =∞ = 1 hr = 10 min =∞ = 1 hr = 5 min =∞ = 15 min =∞ = 3 min = 3 min = 5 min = 2 min =∞ = 15 min = 15 min = 15 min = 2 min
288
GENETIC APPLICATIONS
[24]
5. T = 95◦ , t = 1 min 6. T = 72◦ , t = 5 min 7. Go to 5 14 times (or more as needed) 8. T = 72◦ , t = 30 min 9. T = 4◦ , t = ∞ LCMRE 1. T = 95◦ , t = 2 min (hot start) 2. T = 95◦ , t = 1 min 3. T = 72◦ , t = 5 min 4. Go to 2 14 times 5. T = 72◦ , t = 7 min 6. T = 4◦ , t = ∞ Phage λZAP II Library Construction The ds cDNAs obtained with the former protocol are flanked with the AL1 primer which can be cut out of the library by restriction with EcoRI. The EcoRI restricted ds cDNAs are further purified with QIAquick PCR purification kit (Qiagen Inc.) and subcloned into λZAP II vector (Stratagene, La Jolla, CA) following the protocol from the company. Differential Screening Approximately 12,000 clones from the taste bud library were transferred to duplicate 150-mm round nylon filters (∼2000 clones per filter), which were separately prehybridized at 68◦ for 1 hr in 6× SSC + 1× Denhardt’s (Sigma, St. Louis, MO) + 40 mg/ml poly A + (Boehringer Mannheim Corp.) and then hybridized at 68◦ overnight with either the 32P-labeled (with Random Primed DNA Labeling Kit, Boehringer Mannheim Corp.) total taste bud or control epithelial cell cDNAs (1.3–2 × 106 cpm/ml) in 6× SSC + 1× Denhardt’s + 500 mg/ml poly A+. Washing was performed three times, for 30 min each, at 68◦ in 0.5× SSC + 0.5% SDS and then exposed for 1–3 days to Xar-5 films (Eastman Kodak Company, Rochester, NY) with intensifying screens at −70◦ . Plaques picked on the first screen were screened a second time in the same fashion. Phage from plaques giving a signal with the taste bud-specific cDNA probe, but not with the probe from control epithelium, were processed by plasmid in vivo excision into pBluescript SK(−) phagemid vector (as recommended in the λZAP II phage vector kit), and the plasmids were analyzed by automatic fluorescence sequence analysis. Similarly, a random set of the clones not showing hybridization with either probe can be analyzed by sequencing and chosen for primer design for subsequent PCR analysis. Note: Poly A+ is added to suppress any unwanted hybridization due to poly(dT) stretches present in the cDNAs probes.
[25]
DNA FINGERPRINTING FROM LCM-CAPTURED CELLS
289
Concluding Remarks The use of oligo(dT) magnetic beads in the cDNA synthesis allows the storage of a cell-specific ss cDNA population that can be used later for amplification. In our hands it was possible to use the beads 2–3 times more. It is important to be aware of the possible cross-cell contamination when performing laser capture microdissection. The quality of staining of the frozen sections may be crucial for the recognition of the cell population of interest and this will influence the purity of the cDNAs to be obtained. Minute cross-cell contamination may lead to contradictory results when a gene specific for a cell population shows a PCR signal in the cDNAs from this cell population and in the cDNAs from surrounding cells at the same time when differential hybridization and RNA in situ hybridization shows expression only in the specific cell population of interest. The same caution applies when analyzing clones that do not show a signal with either probe on differential hybridization and are further analyzed by PCR.
Acknowledgments We are especially grateful to Dr. Robert F. Bonner from NIH for the scientific advice in using LCM and giving us access to the NIH LCM Core Lab.
[25] DNA Fingerprinting from Cells Captured by Laser Microdissection By YONGYUT SIRIVATANAUKSORN, VORAPAN SIRIVATANAUKSORN, and NICHOLAS R. LEMOINE Introduction There have been dramatic advances in our knowledge of the molecular processes involved in human diseases, but it is certain that other molecular and genetic lesions remain to be identified, and there is a pressing need to integrate such information with structural and architectural data derived from conventional morphological approaches. It is obviously an advantage to use microdissected cell samples in molecular analysis because the confounding effect of contaminating cells is eliminated. Laser-assisted microdissection is one of the most advanced techniques and has been rapidly developed to procure precisely the cells of interest from complex normal and diseased tissues. Precise microdissection of phenotypically
METHODS IN ENZYMOLOGY, VOL. 356
Copyright 2002, Elsevier Science (USA). All rights reserved. 0076-6879/02 $35.00
290
GENETIC APPLICATIONS
[25]
similar tissue samples revealed genetic heterogeneity.1 An increase in sensitivity of more than 50% in allelic imbalance analysis was obtained by using microdissected cell populations compared with crushed frozen tumor samples.2 The fundamental advantage of this technique is the possibility of exploiting capture on a single-cell basis and isolating high quality DNA and mRNA for analysis of sequence and quantitation of expression. In cancer models, laser-assisted microdissection provides the capacity for isolating cells from specific stages of tumorigenesis, including normal, precancerous, malignant, and metastatic cells. This will allow us to define the genetic changes associated with functional state, malignant transformation, tumor progression, tumor heterogeneity, and clonal progression. Polymerase chain reaction (PCR)-based methods for nucleic acid detection and fingerprinting have become vital to modern molecular genetics, whether for the analysis of populations of organisms to determine population structure of an ecosystem, sampling a set of DNA sequences to infer evolutionary history, sampling genetic loci to build a map, or sampling differentially expressed genes to identify phenotypic markers. PCR can be used to generate high resolution genetic maps of human and comparative genomes. The classic approach to DNA fingerprinting utilizes variable number tandem repeat (VNTR) polymorphism in which alleles differ by a variable number of tandem repeats. Although the term VNTR could encompass a wide range of repeat lengths, it is usually reserved for moderately large arrays of a repeat unit that is typically in the 5- to 64-bp region. If the VNTR locus is a member of a repeated DNA family, the use of a VNTR probe will produce a complex polymorphic band pattern on hybridization. The hybridizing bands appear on the filter as a ladder of bands, referred to as the DNA fingerprint, which visually resembles the bar codes used by stores to identify and price merchandise. Arbitrarily Primed PCR The arbitrary primer-based DNA amplification technique has been proposed as an alternative targeting tool for genetic typing and mapping. This strategy uses randomly generated primers to initiate amplification of discrete but arbitrary portions of the genome. Arbitrarily primed PCR (AP-PCR) is one of the fingerprinting techniques described by Welsh and McClelland.3 This technique is a modification of PCR, a method that is widely used to copy sections of DNA for identifying gene structure or matching tissue specimens. PCR uses two primers whose complementary sequences flank the desired sequence to amplify a region of DNA. 1
C. A. Macintosh, M. Stower, N. Reid, and N. J. Maitland, Cancer Res. 58, 23 (1998). H. E. Giercksky, L. Thorstensen, H. Qvist, J. M. Nesland, and R. A. Lothe, Diagn. Mol. Pathol. 6, 318 (1997). 3 J. Welsh and M. McClelland, Nucleic Acids Res. 18, 7213 (1990). 2
[25]
DNA FINGERPRINTING FROM LCM-CAPTURED CELLS
291
The primers usually have specific nucleotide sequences that bind to previously identified segments of DNA. They bind to specific sites on opposing strands of the double-stranded DNA and are extended by a thermostable DNA polymerase to make millions of copies of the intervening stretch of DNA. Normally, the primers are annealed to the template DNA at relatively high stringency. High stringency during the primer-annealing step ensures that the primers do not interact with the template DNA at positions where they do not match. By contrast, AP-PCR allows the detection of polymorphisms without prior knowledge of nucleotide sequence. It is based on the selective amplification of genomic sequences that, by chance, are flanked by adequate matches to an arbitrarily chosen primer. The method utilizes short primers of arbitrary nucleotide sequence (10 to 20 bases) that are annealed in the first few cycles of PCR at low stringency. The low stringency of the early cycles ensures the generation of products by allowing priming with fortuitous matches or near-matches between primers and template. This approach results in a high number of products having the original primer sequence at both ends. After a few low-stringency cycles, the annealing temperature is raised and the reaction is allowed to continue under standard, high-stringency PCR conditions. This step will amplify a discrete number of sequences among those initially targeted and permits the unbiased analysis of the cell genome. Alternatively, an intermediate stringency primer-annealing step may be used throughout the PCR to achieve a similar outcome. AP-PCR products are resolved on polyacrylamide gels and are detected by autoradiography. If two template genomic DNA sequences are different, their AP-PCR products display different banding patterns. Such differences can be exploited in ways largely analogous to the uses of restriction fragment length polymorphisms, including genetic mapping, taxonomy, phylogenetics, and the detection of mutations. AP-PCR permits the rapid and cost-effective detection of polymorphisms and genetic markers in a variety of experiments.4 Moreover, it is dramatically easier and faster than established methods of genetic mapping. The reproducible and semiquantitative amplification of multiple sequences provides a powerful tool for studying somatic genetic alterations in tumorigenesis. Peinado and colleagues showed the ability of AP-PCR to detect both qualitatively and quantitatively and to isolate, in a single step, DNA sequences representing two of the genetic alterations that underlie the aneuploidy of colorectal cancer cell, i.e., losses of heterozygosity and chromosomal gains.5 Moreover, they confirmed that AP-PCR could yield information on the overall chromosomal composition of the cell. The intensities of the bands derived from single-copy sequences were 4
J. G. Williams, A. R. Kubelik, K. J. Livak, J. A. Rafalski, and S. V. Tingey, Nucleic Acids Res. 18, 6531 (1990). 5 M. A. Peinado, S. Malkhosyan, A. Velazquez, and M. Perucho, Proc. Natl. Acad. Sci. U.S.A. 89, 10065 (1992).
292
GENETIC APPLICATIONS
[25]
proportional to the concentration of the target sequences. The outstanding result using AP-PCR fingerprinting in the field of cancer research was the discovery of the microsatellite mutator phenotype mechanism for carcinogenesis in colonic and lung cancers.6,7 AP-PCR is also useful for the detection and isolation of DNA sequences to levels well below the minimum levels required by other available methods,8 and the products can be used to clone or hybridize back to digested genomic DNA.9 Unbiased DNA fingerprinting by AP-PCR is a powerful molecular approach for the cytogenetic analysis of solid tumors. It detects both gains of chromosomal regions, reflecting the presence in these regions of cancer genes, and losses of genes with negative role in cell growth. Nevertheless, the finding that DNA sequences have undergone heterozygous deletions or gains of extra copies in tumor relative to normal tissue does not ensure that these sequences are linked to genes playing an active role in oncogenesis because of the high level of random genetic damage in the genome of solid tumors. Moreover, AP-PCR is an uncomplicated and effective method for scanning the genomes of tumor samples to show the genomic heterogeneity and the evolution of differences.10,11 Laser Capture Microdissection The laser capture microdissection (LCM) system has been developed by Emmert-Buck et al.12 and comprises a novel membrane-based microdissection technique. The system has been subsequently commercialized and used in many laboratories. A thermoplastic ethylene vinyl acetate transfer film containing a near-infrared absorbing dye attached to a 6-mm diameter rigid, flat cap is placed in contact with a routinely prepared, hematoxylin and eosin stained tissue section. The isolation of cells from immunohistochemical or molecule-specific, fluorescently labeled sections improves sample imaging and can help in obtaining specific cell populations more precisely.13 The film over the cells of interest is precisely 6
Y. Innov, M. A. Peinado, S. Malkhosyan, D. Shibata, and M. Perucho, Nature 363, 558 (1993). Y. Anami, T. Takeuchi, K. Mase, J. Yasuda, S. Hirohashi, M. Perucho, and M. Noguchi, Int. J. Cancer 89, 19 (2000). 8 I. B. Roninson, J. E. Chin, K. G. Choi, P. Gros, D. E. Housman, A. Fojo, D. W. Shen, M. W. Gottesman, and I. Pastan, Proc. Natl. Acad. Sci. U.S.A. 83, 4538 (1986). 9 C. S. Wesley, M. Ben, M. Kreitman, N. Haga, and W. F. Eanes, Nucleic Acids Res. 18, 599 (1990). 10 Y. Sirivatanauksorn, V. Sirivatanauksorn, S. Bhattacharya, B. R. Davidson, A. P. Dhillon, A. K. Kakkar, R. C. N. Williamson, and N. R. Lemoine, Gut 45, 761 (1999). 11 Y. Sirivatanauksorn, V. Sirivatanauksorn, S. Bhattacharya, B. R. Davidson, A. P. Dhillon, A. K. Kakkar, R. C. N. Williamson, and N. R. Lemoine, J. Pathol. 189, 344 (1999). 12 M. R. Emmert-Buck, R. F. Bonner, P. D. Smith, R. F. Chuaqui, Z. Zhuang, S. R. Goldstein, R. A. Weiss, and L. A. Liotta, Science 274, 998 (1996). 13 F. Fend, M. R. Emmert-Buck, R. Chuaqui, K. Cole, J. Lee, L. A. Liotta, and M. Raffeld, Am. J. Pathol. 154, 61 (1999). 7
[25]
DNA FINGERPRINTING FROM LCM-CAPTURED CELLS
293
activated by a near-infrared laser pulse and bonds strongly to the selected cells. Removal of the cap from the tissue section effectively procures the targeted cells. The identity of the transferred cells attached to the film can then be recorded by image capture. The cap is lifted off the tissue and placed directly onto a 0.5-ml microfuge tube containing 50 µl of proteinase K buffer. The tube is inverted and incubated overnight at 37◦ . After the incubation period, the tube is centrifuged at 10,000g for 5 min and the cap is removed. Then the buffer is inactivated at 95◦ for 10 min and the sample is ready to use as a template for PCR. To examine the quality of the DNA samples from the microdissection technique, 4 µl of DNA solution is used for amplification with GAPDH primers in a total volume of 50 µl solution containing 50 mM KCl, 10 mM Tris-HCl pH 8.3, 1.5 mM MgCl2, 100 pM primers, and 5 units Taq polymerase. Templates are denatured for 5 min at 95◦ and are subject to 35 cycles at 94◦ , 1 min; 55◦ , 1 min; and 72◦ , 2 min. The PCR products are run on 1.5% agarose gel staining with ethidium bromide. Then the DNA fingerprint from the cells of interest is amplified by the AP-PCR technique. Arbitrarily Primed PCR Materials 1. Stocks of all four dNTPs (5 mM) 2. Stock of arbitrary primer (100 µM) 3. Radioisotope ([α-32P] or [γ -33P]) dATP (>2500 Ci/mmol) 4. Taq polymerase (5 U/µl) 5. Formamide dye solution: 96% formamide, 0.1% bromphenol blue, 0.1% xylene cyanol, 10 mM EDTA 6. 10× Tris–borate–EDTA (TBE) buffer: 90 mM Tris–borate, 20 mM Na2EDTA, pH 8.3 7. Acrylamide stock solution [40% acrylamide : bisacrylamide (29 : 1)] 8. Thermocycler (e.g., Peltier Thermal Cycler, model PTC-100) 9. Sequencing gel electrophoresis apparatus (e.g., 40 cm long, 30 cm wide, 0.4 mm thick) 10. Gel dryer 11. X-ray film and exposure cassette Methods A. Polynucleotide Kinase Reaction. This procedure is useful for radioactive labeling of the 5 end of oligonucleotide primers. The kinase enzyme requires that the 5 end of the oligonucleotide has been previously dephosphorylated with
294
GENETIC APPLICATIONS
[25]
alkaline phosphatase. The forward kinase reaction catalyses the exchange of the terminal γ -phosphate, which is labeled with 33P from the ATP to the terminal phosphate on the oligonucleotide. 1. Mix 100 pM of oligonucleotide with 50 mM Tris-Cl pH 7.5, 10 mM MgCl2, 5 mM DTT, 0.5 mM spermidine, 50 Ci of [γ -33P]dATP, and 10 U of T4 polynucleotide kinase enzyme. 2. Incubate the reaction for 1 hr at 37◦ . B. Arbitrarily Primed PCR 1. Mix template DNA (100 ng) with reaction mixture for a 25-µl final reaction containing [γ -33P]ATP-labeled and kinased arbitrary primer, 0.2 mM each dNTP (Bioline), 10 mM Tris-Cl pH 9.2, 3.5 mM MgCl2, 75 mM KCl, and 0.5 U of Taq DNA polymerase. 2. Perform thermocycling using 5 low-stringency steps, followed by 35 highstringency steps as follows: 5 low-stringency cycles (94◦ for 1 min; 50◦ for 5 min; 72◦ for 5 min) then 35 high-stringency cycles (94◦ for 1 min; 60◦ for 1 min; 72◦ for 2 min) and A final chase cycle of 72◦ for 5 min to allow complete elongation of all products (After 5 low-stringency cycles, exact copies of the primer sequence flank a handful of anonymous sequences. Thus, the annealing temperature can be raised after a few cycles and the reaction allowed to continue under standard, high-stringency PCR conditions. The two-step low–high stringency protocol was designed to avoid internal priming within a larger amplifying product.) 3. Mix amplification product with 5 µl of formamide dye solution, denature at 95◦ for 3 min, and load 5 µl of mixture onto an 8% acrylamide gel matrix prepared in 1× TBE buffer. 4. Perform electrophoresis using a Sequencing Gel Electrophoresis Apparatus at 12 W, constant voltage overnight. 5. After electrophoresis, transfer the gel to Whatman 3MM paper, and dry under vacuum at 80◦ for 2 hr. 6. Autoradiograph the dried gel using an X-ray film at room temperature overnight or for 2–3 days as required.
[26]
SINGLE CELL
PCR IN LCM
295
[26] Single Cell PCR in Laser Capture Microscopy By SINUHE HAHN, XIAO YAN ZHONG, and WOLFGANG HOLZGREVE Introduction The ability to analyze single cells by PCR has opened up many new avenues for diagnosis and research. One of the most prominent diagnostic applications of this technology currently is preimplantation genetic diagnosis (PGD).1,2 In this examination one or two embryonic cells are biopsied from human preimplantation embryos. An advantage of this procedure is that it is fairly simple to obtain single embryonic cells free from any potential paternal (spermatozoa) or maternal (cumulus cells) contaminants. Another potential diagnostic application of single cell PCR is the examination of single fetal erythroblasts retrieved from maternal circulation, where, by a combination of enrichment and micromanipulation, it is possible to isolate the desired single fetal cell from a relatively large host of coenriched maternal cells.3 This is in stark contrast to the isolation of single cells from tissue sections, where conventional micromanipulation techniques, of the type applied for PGD or isolation of fetal cells, are wholly unsatisfactory. To circumvent these limitations several laser-based tools have been devised for the reliable and effective retrieval of single cells from complex preparations. Laser Capture Microdissection The laser capture microdissection (LCM) system was developed by Liotta and colleagues4 at NIH (the National Institutes of Health, Bethesda, MD) and is marketed by a spin-off company, Arcturus Engineering (www.arctur.com), under the name PixCell. The LCM system employs a 980 nm IR diode laser mounted above the microscope stage, which is pulsed through a special optical quality cap the size of a standard PCR reaction vessel. The base of this cap is lined with a 5-µm synthetic transfer membrane which is brought to rest on the microscope slide with the tissue preparation. The application of a pulse of laser energy, which is focused on the desired target cell, melts the thermoplastic membrane, thereby covalently bonding the target cell to the PCR reaction vessel cap. This process can then be repeated to permit the collection of numerous singly isolated cells, or the cap can 1
A. H. Handyside, Prenat. Diagn. 18, 1345 (1998). S. Hahn, X. Y. Zhong, C. Troeger, R. Burgemeister, K. Gloning, and W. Holzgreve, Cell. Mol. Life Sci. 57, 96 (2000). 3 W. Holzgreve and S. Hahn, Clin. Perinatol. 28, 353, ix (2001). 4 M. R. Emmert-Buck, R. F. Bonner, P. D. Smith, R. F. Chuaqui, Z. Zhuang, S. R. Goldstein, R. A. Weiss, and L. A. Liotta, Science 274, 998 (1996). 2
METHODS IN ENZYMOLOGY, VOL. 356
Copyright 2002, Elsevier Science (USA). All rights reserved. 0076-6879/02 $35.00
296
GENETIC APPLICATIONS
[26]
be placed on a standard-sized PCR reaction vessel for further molecular biological analysis. Since its inception,4 this tool has gained rapid acceptance in many research and diagnostic circles,5,6 most prominently in those centers interested in tumor heterogeneity,7 cellular differentiation,8 and preimplantation genetic diagnosis,9 as well as cell-specific drug interactions.10 These studies have used PCR, RTPCR, real-time PCR, and gene expression profiling by microarray analysis. It is worth noting that although LCM in theory is capable of isolating single cells, almost all of these numerous studies have been performed using pools of cells which had been singly captured. This partly stems from the inability to focus the laser sufficiently finely to accommodate the size and shape of single cells.11 Furthermore, when a membrane-lined PCR cap is placed over the entire specimen, it is very likely that cells or cell debris can easily become attached to this membrane, thereby contaminating the captured single cell preparation. Although this can be overcome to some extent by blotting the membrane-lined cap on a piece of sterile adhesive tape to remove noncovalently bonded cells, this approach is less than satisfactory for work requiring high degrees of purity. The other reason may be that the reliability of single cells from PCR is still technically challenging.2 In order to facilitate the better capture of single cells, two approaches have been taken. In the first, a high-tech approach developed by NIH, a computer-controlled arm carefully positions a 40-µm wide strip of the thermoplastic membrane with a very light contact force on the specimen.11 A modified laser is then used to epi-irradiate individual cells with highly focused rapid laser pulses. This system uses computer-assisted rotation of a membrane-coated cylinder, whereby it is possible to capture multiple single cells individually. By the use of this approach, it is reasoned that highly selective transfer of single cells is possible. Unfortunately, this system is so complex that it is not yet commercially available. Arcturus, the company marketing the LCM system, has taken a rather low-key approach, developing a special cap (termed CapSure HS noncontact caps) used for the capture procedure whereby only a very small membrane area is in contact with the specimen. The usefulness of this in practice still needs to be verified in large-scale studies. 5
M. A. Rubin, J. Pathol. 195, 80 (2001). J. C. Mills, K. A. Roth, R. L. Cagan, and J. I. Gordon, Nat. Cell Biol. 3, E175 (2001). 7 F. Fend and M. Raffeld, J. Clin. Pathol. 53, 666 (2000). 8 H. Ohyama, X. Zhang, Y. Kohno, I. Alevizos, M. Posner, D. T. Wong, and R. Todd, BioTechniques 29, 530 (2000). 9 A. Clement-Sengewald, T. Buchholz, and K. Sch¨ utze, Pathobiology 68, 232 (2000). 10 T. Betsuyaku, G. L. Griffin, M. A. Watson, and R. M. Senior, Am. J. Respir. Cell. Mol. Biol. 25, 278 (2001). 11 C. A. Suarez-Quian, S. R. Goldstein, T. Pohida, P. D. Smith, J. I. Peterson, E. Wellner, M. Ghany, and R. F. Bonner, BioTechniques 26, 328 (1999). 6
[26]
SINGLE CELL
PCR IN LCM
297
Laser Pressure Catapulting Biological application of laser pressure catapulting (LPC) was pioneered by Sch¨utze and colleagues12,13 and developed at the same time as the LCM system (www.palm.microlaser.com). This system also uses an inverse microscope, but uses an epifluorescence laser system instead of an overhead laser. In LPC, a 337 nm high quality pulsed nitrogen laser is focused through the microscope optics onto the specimen and with the use of high numerical long-distance objectives a spot size of 2 µm can be attained. The LPC system differs from the LCM system in that the cell preparation is transferred onto a glass slide covered with a thin polyethylene membrane. Once the desired target cells have been localized, they are first excised by the high energy laser beam and then in a second step are catapulted by a focused pulse of laser energy directly into the cap of a PCR reaction vessel suspended above the microscope slide. The use of the synthetic membrane permits retention of the target cells during the microdissection and catapulting steps. Advantages of this system are the ability to finely focus the laser beam and the fact that the high energy laser can be used to ablate cells adjacent to the target cell. It therefore appears better suited to the isolation of single cells than the current LCM version. Comparative Analysis As our interest lies in the reliable retrieval of single fetal cells from maternal circulation,3 we have performed a comparative analysis in which we investigated the efficacy of LCM and LPC systems and classical manual micromanipulation. In this study, cells were transferred onto glass slides by cytocentrifugation and histochemically stained. Single lymphocytes were isolated by the three different microdissection methods and examined by a nested PCR procedure for the ubiquitous β-globin gene which had been optimized for the analysis of single cells.14,15 This control locus had previously been shown to be very effective in monitoring the efficacy of the PCR reaction and in ascertaining whether a single cell had indeed been transferred to the reaction vessel.16 In our study a total of 359 single cells were analyzed: 61 by LCM, 99 by LPC, and 199 by capillary-based micromanipulation. The single cell PCR analysis indicated that the β-globin gene could be readily detected in 52% of the cells isolated by LCM, 56% of the cells isolated 12
K. Sch¨utze, A. Clement-Sengewald, and A. Ashkin, Fertil. Steril. 61, 783 (1994). K. Sch¨utze, I. Becker, K. F. Becker, S. Thalhammer, R. Stark, W. M. Heckl, M. Bohm, and H. Posl, Genet. Anal. 14, 1 (1997). 14 C. Troeger, X. Y. Zhong, R. Burgemeister, S. Minderer, S. Tercanli, W. Holzgreve, and S. Hahn, Mol. Hum. Reprod. 5, 1162 (1999). 15 X. Y. Zhong, W. Holzgreve, J. C. Li, K. Aydinli, and S. Hahn, Prenat. Diagn. 20, 838 (2000). 16 S. Hahn, X. Y. Zhong, M. R. Burk, C. Troeger, and W. Holzgreve, Ann. N.Y. Acad. Sci. 906, 148 (2000). 13
298
GENETIC APPLICATIONS
[26]
by LPC, and 43% of those retrieved by manual micromanipulation. It, therefore, appears that the isolation of single cells by either of the laser-based systems is more effective than by traditional methods. A major factor to consider here is the reliable transfer of the isolated single cell into the PCR reaction vessel, which appears to be better safeguarded by the laser-mediated tools than by use of a finely drawn capillary, from which it may be more easily lost. Current Limitations of Individual Methods Although it may appear that the LPC system is better suited than current LCM tools for the retrieval of single cells, both systems have their individual advantages and disadvantages. As mentioned above, LCM of single cells is currently hampered by the use of a rather large membrane surface covering a significant portion of the target cell preparation. Since this membrane is in direct contact with the specimen, it is difficult to be certain that no extraneous genetic material has been introduced into the final single cell preparation. This large membrane surface also hinders the effective analysis of single cells, as it is very difficult to extract the DNA from the single cell attached to this membrane in a sufficiently small volume of extraction buffer which still permits an effective PCR analysis. Indeed, in our experience, it was not possible to reliably isolate the single cell DNA by moistening the membrane while attached to the cap; rather, the membrane had to be removed and immersed in a separate PCR reaction vessel. This additional step was not only tedious, but a significant potential source of contamination. It is to be hoped that the use of new noncontact caps will alleviate this problem, until the more complex cylinder rotation system becomes commercially viable. On the other hand, we have determined that the membrane used for the LPC system, although important for the maintenance of cell integrity, does compromise the use of this system. In this regard, we have experienced that while the membrane coated slides are very suitable for use with microtome-cut tissue sections, they are less than ideal with cytocentrifuged cell preparations. This is due to the lack of adequate adhesion of cell suspensions to this membrane, which results in significant cell loss during staining procedures. Although this is of little concern when dealing with an abundant cell type, it is a definite problem when investigating rare events, such as fetal cells enriched from the maternal circulation. Staining Procedures Including Fluorescence-Based Methods A further problem we have encountered is that LCM is adversely affected by the majority of histological staining procedures, such as May-Gruenwald Giemsa, which are commonly used for the initial distinction of various cells types.14,15 This is apparently because the preparation is not sufficiently dehydrated to permit effective transfer of the laser energy through the target cell to achieve its
[26]
SINGLE CELL
PCR IN LCM
299
thermobonding to the carrier membrane. The use of extra dehydration steps has the effect of leaching the desired dyes from the preparation, thereby rendering identification of the desired target cell impossible. Of the diverse histological and immunohistochemical staining procedures we have used for the putative identification of fetal erythroblasts, we determined that 3,3 -diaminobenzidine (DAB) based stains were optimal. It is unclear how this physical need for extremely dry preparations will affect the use of fluorescent immunohistochemical protocols. We have, however, determined that the use of FISH (fluorescence in situ hybridization) protocols does prevent the effective retrieval of target cells by LCM. Consequently, when considering cell recycling approaches, other strategies have to be chosen in which individual cells are first analyzed by FISH to determine chromosomal ploidy and subsequently by PCR to examine for the presence of a specific inherited single gene disorder.2,17 In the same manner, we have noted that the use of membrane-coated slides for LPC severely restricts the use of fluorescence-based approaches because of the very high level of autofluorescence of this membrane. Furthermore, as the laser of the LPC system uses the same pathway as that of the epifluorescence system, the fluorescence intensity in such modified microscopes is dramatically reduced, particularly for certain wavelengths, such as those needed to excite the DAPI channel. We have not been able to optimize this system to the extent that it can be used for cell recycling approaches. It is clear that it will also need to be optimized carefully when using fluorescent immunohistochemistry for the identification of potential target cells. RT-PCR and Microarray Analysis Even though several studies using laser microdissection have focused on the detection of genetic mutations, such as the presence of malignant cells in a given tissue, a considerable body of work has been performed examining gene expression using this technology.8,18–21 One aspect worth noting with regard to these studies is that these studies are almost exclusively performed on pools of cells. Several of these studies have employed cutting-edge technology, such as real-time PCR,22 or 17
A. Sekizawa, O. Samura, D. K. Zhen, V. Falco, and D. W. Bianchi, Am. J. Obstet. Gynecol. 181, 1237 (1999). 18 K. Schutze and G. Lahr, Nat. Biotechnol. 16, 737 (1998). 19 I. Alevizos, M. Mahadevappa, X. Zhang, H. Ohyama, Y. Kohno, M. Posner, G. T. Gallagher, M. Varvares, D. Cohen, D. Kim, R. Kent, R. B. Donoff, R. Todd, C. M. Yung, J. A. Warrington, and D. T. Wong, Oncogene 20, 6196 (2001). 20 M. Neira, V. Danilova, G. Hellekant, and E. A. Azen, Mamm. Genome 12, 60 (2001). 21 K. E. Dolter and J. C. Braman, BioTechniques 30, 1358 (2001). 22 L. Fink, W. Seeger, L. Ermert, J. Hanze, U. Stahl, F. Grimminger, W. Kummer, and R. M. Bohle, Nat. Med. 4, 1329 (1998).
300
GENETIC APPLICATIONS
[26]
high density gene microarrays.5,6,8,19 In this regard, Fink and colleagues examined TNF-α expression by real-time PCR in small numbers of individually isolated alveolar macrophages obtained by LPC from bronchiolar lavage specimens.22 LCM technology has also been used by Betsuyaku et al. to explore the action of drugs on distinct cell populations, such as bleomycin on bronchiolar epithelium by real-time PCR.10 Furthermore, the expression profiles of LCM dissected tumor and normal epithelial cells have been examined by microarray analysis.5,6 In one such study approximately 600 cancer-associated genes were identified.19 Because of the number of cells required for these analyses, between 100 and 1000, considerable progress still has to be made before this technology becomes available for single cell analysis. One way of overcoming this deficit may be to use whole genome gene amplification strategies similar to those used by numerous research groups for the PCR analysis of several genetic loci from single cells.2,23,24 Additionally, the mRNA can be linearly amplified using T7 RNA polymerase.8 This latter approach has been shown to be suitable for generating biotinylated cRNA species which can be readily detected using high density oligonucleotide arrays.8 Another major problem when dealing with mRNA expression studies is the preparation of the specimen. Studies have indicated that archival tissue can be readily used for DNA analyses, whereas those used for mRNA expression need to be specially prepared to prevent mRNA degradation.25–27 In this manner cryopreserved tissues yielded a greater abundance of mRNA species than paraffin embedded sections as measured by real-time PCR for standard housekeeping genes. A further major concern is that exposing the specimen to aqueous solutions can lead to the destruction of almost 99% of the RNA. To overcome this, special rapid immunofluorescence methods have been developed in which the specimen is labeled and fixed in a 1 min procedure.26,28–30 Results from other studies have indicated that brief counterstaining of cryosections with hematoxylin may leave cells sufficiently intact to permit an examination of several distinct mRNA species.27 23
V. G. Cheung and S. F. Nelson, Proc. Natl. Acad. Sci. U.S.A. 93, 14676 (1996). C. P. Beltinger, F. Klimek, and K. M. Debatin, Mol. Pathol. 50, 272 (1997). 25 S. M. Goldsworthy, P. S. Stockton, C. S. Trempus, J. F. Foley, and R. R. Maronpot, Mol. Carcinog. 25, 86 (1999). 26 F. Fend, M. Kremer, and L. Quintanilla-Martinez, Pathobiology 68, 209 (2000). 27 N. Tanji, M. D. Ross, A. Cara, G. S. Markowitz, P. E. Klotman, and V. D. D’Agati, Exp. Nephrol. 9, 229 (2001). 28 F. Fend, M. R. Emmert-Buck, R. Chuaqui, K. Cole, J. Lee, L. A. Liotta, and M. Raffeld, Am. J. Pathol. 154, 61 (1999). 29 L. Fink, T. Kinfe, M. M. Stein, L. Ermert, J. Hanze, W. Kummer, W. Seeger, and R. M. Bohle, Lab. Invest. 80, 327 (2000). 30 H. Murakami, L. Liotta, and R. A. Star, Kidney Int. 58, 1346 (2000). 24
[26]
SINGLE CELL
PCR IN LCM
301
The ability to obtain small numbers of a particular cell type has facilitated the generation of well-defined cDNA libraries which permit the identification of new tissue-specific genes.20 In this manner a taste bud specific transcript (rmSTG) was detected by differential screening from a small number of laser manipulated Rhesus monkey taste bud cells.20 It is to be expected that these techniques will be optimized to such an extent that it will become possible to examine gene expression differences in various cell types in a tissue section, thereby permitting a more detailed analysis of cell differentiation and the process of cancer formation. Conclusions In the few years since the development of laser-mediated microdissection technologies tremendous progress has been made in the analysis of cells captured by these means. It is clear that the analysis of single cells will always be hampered by the limiting amount of target template, be this DNA or mRNA, despite the use of strategies to amplify the template in an unbiased manner.2,31,32 This aspect may also become more apparent when examining different cell types, such as erythroblasts, which have higher allele dropout rates than other hemopoietic cells.31 Other issues which currently need to be addressed are reliable contamination-free excision of single cells from complex specimens using LCM and methods to obviate the need for membrane-coated slides in LPC, from which rare cells are prone to be lost during the preparation of the specimen. Furthermore, effective methods will have to be developed which enable the efficient manipulation of cells labeled by immunofluorescent means, especially those which have been previously analyzed by FISH for their chromosomal complement. There is, however, little doubt that the analysis of single or few cells obtained by laser mediated microdissection is going to gain in clinical importance in the near future and that this technology will also open new, exciting avenues for research, especially for those interested in cellular differentiation, be this normal or pathological.
31 32
A. M. Garvin, W. Holzgreve, and S. Hahn, Nucleic Acids Res. 26, 3468 (1998). S. Hahn, A. M. Garvin, E. Di Naro, and W. Holzgreve, Genet. Test. 2, 351 (1998).
302
GENETIC APPLICATIONS
[27]
[27] Assessment of Genetic Heterogeneity in Tumors Using Laser Capture Microdissection By DAVE S. B. HOON, AKIHIDE FUJIMOTO, SHERRY SHU, and BRET TABACK Introduction Assessment of genetic changes that contribute to the tumor heterogeneity within a tumor lesion is an important and technically complex problem. We have employed laser capture microdissection (LCM) to assess archived small metastatic melanoma lesions for tumor genetic changes. The focus of the study is to examine inter- and intratumor genetic heterogeneity within melanoma lesions. DNA microsatellite markers with loss of heterozygosity (LOH) are assessed in melanoma lesion sections microdissected with LCM. Both inter- and intratumor genetic heterogeneity are observed in these tumor lesions. There are now ample studies demonstrating that human solid tumors are heterogeneous in specific genetic markers or expression of individual genes. Tumor heterogeneity is the inherent problem that has significantly hampered cancer diagnosis and treatment. Tumor heterogeneity occurs at the “macro” level involving clonal cell types within a tumor lesion and at the “micro” level involving genetic differences and gene expression levels within cells of a tumor lesion. The heterogeneity of tumors at the molecular level has only recently been appreciated. This has come about through the development of improved and more sensitive molecular detection assays, and most importantly through new approaches to microdissection of a given tumor lesion and isolation of a specific group of cells. The latter have significantly evolved through the development of LCM. The LCM technique has allowed investigation of tumor lesions at the micro level, which has not previously been available to researchers. The approach has allowed more accurate, focused, and comparative analysis of defined regions or cells within a tumor lesion. The rapid evolution of molecular biology of tumors and LCM analysis has provided a very powerful analytic approach to assess tumor heterogeneity and understand tumor progression better at the molecular level. In our laboratory a major ongoing study is the assessment of genetic markers and their relevance in tumor progression such as in melanoma. The focus of our studies has been on specific multiple DNA microsatellite markers in which there is LOH. Studies from our laboratory and others have demonstrated that there are frequent LOH of specific microsatellite markers covering several major chromosome arms in melanoma.1–4 The detection of LOH at specific chromosome sites 1
T. Nakayama, B. Taback, R. Turner, D. L. Morton, and D. S. B. Hoon, Am. J. Pathol. 158, 1371 (2001).
METHODS IN ENZYMOLOGY, VOL. 356
Copyright 2002, Elsevier Science (USA). All rights reserved. 0076-6879/02 $35.00
[27]
TUMOR GENETIC CLONALITY ASSESSMENT BY
LCM
303
in various cancers has been instrumental in the identification of tumor suppressor genes or tumor-related genes. The detection of specific LOH markers in tumor lesions has also been shown to be a very valuable prognostic marker for disease outcome.2,5,6 Traditional methods have been to dissect out tumor lesions from paraffin-embedded sections under a micrcoscope using a scalpel or needle. This approach was basically the only practical method available until the development of LCM. The limitations of previous approaches of tumor dissection were contamination of normal cells, the size of tumor cell clumps, and the accuracy of particular cell types being dissected. It is now obvious that previous approaches may have underestimated LOH of multiple markers in a tumor lesion, and, also, the assessment of the particular site of the lesion may not be representative of the lesion as a whole. In assessment of tumor lesions for genetic markers one of the major limitations has been the size of the tumor lesion to be dissected. The majority of the studies of LOH have been on large tumor lesions that are easily dissected out, usually hematoxlyin and eosin positive-stained lesions greater than 5 mm, with a scalpel/needle with light microscopy. In recent years primary tumor lesions and metastasis have been detected at earlier stages; thus the lesions are being detected when they are smaller. Recently, there has been significant growing importance in the detection of “micrometastasis” in tumor-draining lymph nodes. The genetic changes in these early stages of metastatic tumor establishment are important in assessment. Micrometastasis is a loosely defined term that can refer to anything from a few occult tumor cells to a clump (10–100 cells) of tumor cells. The advent of LCM has allowed us to improve our dissection of these micrometastases, isolate a sufficient amount of nucleic acids, and perform molecular analyses. Obviously, the limitations are specimen preservation and the amount of nucleic acids recovered. Nevertheless, LCM has opened up a new area of the accurate assessment of the early stages of tumor metastasis. In this study we describe the utilization of LCM in the assessment of LOH markers in two types of metastatic melanoma diseases, in-transit melanomas and lymph node metastasis.
2
Y. Fujiwara, D. J. Chi, H. J. Wang, P. Kelemen, D. L. Morton, R. Turner, and D. S. B. Hoon, Cancer Res. 59, 1561 (1999). 3 T. Nakayama, B. Taback, D.-H. Nguyen, D. L. Morton, and D. S. B. Hoon, Ann N.Y. Acad. Sci. 906, 87 (2000). 4 R. Morita, A. Fujimoto, N. Hatta, K. Takehara, and M. Takata, J. Invest. Dermatol. 111, 919 (1998); E. Healy, C. Belgain, M. Takata, A. Vahlquist, I. Rehman, H. Rigby, and J. Rees, Cancer Res. 56, 589 (1996). 5 B. Taback, A. E. Giuliano, N. M. Hansen, and D. S. B. Hoon, Ann. N.Y. Acad. Sci. 945, 22 (2001). 6 B. Taback, Y. Fujiwara, H.-J. Wang, L. J. Foshag, D. L. Morton, and D. S. B. Hoon, Cancer Res. 61, 5723 (2001).
304
GENETIC APPLICATIONS
[27]
Materials and Methods Specimens Primary and metastatic melanoma paraffin-embedded tissue blocks were obtained from Saint John’s Health Center pathology department. All studies were under patients’ written consent through an approved Human Subject’s Committee IRB protocol. Control DNA for each melanoma patient was obtained from either their peripheral blood lymphocytes (PBL) or from spotted blood on FTA cards (GIBCO, NY) using the QIAamp blood kit. LCM and DNA Extraction Paraffin-embedded tumor tissue from melanoma patients was assessed for intratumor heterogeneity using LCM. Two or three random and discrete regions from each melanoma lesion were selected for LCM and DNA extraction. Amplitude, pulse duration, and number of hits were adjusted to capture approximately 4 × 106µm3 of tissue. DNA was isolated with 55 µl proteinase K (0.18 mg/ml proteinase K, 45 mM Tris-HCL pH 8.0, 0.9 mM EDTA, and 0.45% Tween 20) at 42◦ overnight, followed by heat-denature of proteinase K at 95◦ for 10 min. Microsatellite-PCR LOH Analysis Eight markers covering six different chromosomes were selected for PCR amplification. These markers were selected because they showed a high incidence of LOH in either primary melanoma tumors or advanced metastases. Studies were conducted with FAM-labeled microsatellite markers for vertical gel electrophoresis or Beckman dye fluorolabeled dye microsatellite markers for capillary array electrophoresis (CAE). The following FAM-labeled microsatellite markers were used in this study: D1S228 at 1p36; D1S314 at 1p36.3; D3S1293 at 3pter3p24.2; D6S264 at 6q25.2-q27; D9S157 at 9p23-p22; D9S04 at 9p21; D10S212 at 10q26; and D11S2000 at 11q22-q23. The following microsatellite markers were used for CAE: D1S228; DS1293; D9S157; and D11S200. These markers were labeled with WellRed phosphoramidite-linked dye or active ester-labeled dye. PCR primer sets for specific allele loci were obtained from Research Genetics, Inc. (Huntsville, AL). Genomic DNA (∼50 ng) extracted by LCM and matching PBL was amplified using PCR in a 10 µl reaction volume, containing 15 mM Tris-HCl (pH 8.0), 50 mM KCl, 1.5 mM MgCl2, 0.8 mM deoxynucleotide triphosphates, 0.25 µM forward primer, 0.25 µM reverse primer, and 0.5 U of Amplitaq Gold DNA polymerase (PerkinElmer, Norwalk, CT). PCR cycles consisted of 30 sec at 94◦ , 30 sec at 50–56◦ depending on the primer sets, and 30 sec at 72◦ for a total of 40 cycles. This was followed by a 5 min final extension at 72◦ . One µl of PCR product was mixed with 40 µl of loading buffer and 0.5 µl of CEQ DNA size standards (Beckman Coulter, Inc., Fullerton, CA) in a 96-well microtiter plate.
[27]
TUMOR GENETIC CLONALITY ASSESSMENT BY
LCM
305
Sample was then electrophoresed on the CEQ 2000XL DNA Analysis System (Beckman Coulter, Inc.). The CAE system contains 8 capillaries, each with a 33 cm length (Beckman Coulter, Inc.). Samples were denatured at 90◦ for 120 sec, introduced into the capillaries by electrokinetic injection for 30 sec at 2.0 kV, and electrophoresed at 6.0 kV for 35 min. The total cycle for each row of eight samples, which included denaturation, injection, separation, data analysis, and capillary replenishment, was approximately 45 min. Peak intensity and relative size were generated automatically by CEQ 2000 Fragment Analysis System Software (Beckman Coulter, Inc.). To estimate the degree of LOH, normalized ratios are calculated as (T1/T2)/(N1/N2) where T1 and N1 are the peak heights of the lighter alleles and T2 and N2 are the peak heights of the heavier alleles of tumor DNA (T) and PBL DNA (N). The tumor was scored as exhibiting LOH when the ratio was greater than 2.0 or lower than 0.5. For vertical gel electrophoresis the gel was scanned by a fluorescent/optical GenomyxSC scanner (Beckman Coulter, Inc.). Densitometry was performed on the gel images and analyzed using ClaritySC 3.0 software (Media Cybernetics, Silver Spring, MD). The tumor was scored as exhibiting LOH if there was a >50% reduction in signal intensity of one allele when compared to the respective allele in the corresponding normal DNA (lymphocytes) for both CAE and vertical gel electrophoresis analysis. Results and Discussion In-transit melanoma is a locoregional metastasis that often develops into an aggressive disease and eventual systemic metastasis. The disease is often characterized by rapid recurrence of local regional metastases after the primary or metastastic lesions have been removed. This melanoma is often cutaneous and recurrence is in the form of multiple nodules usually <1 cm. This form of melanoma is very difficult to treat and remains an enigma to clinicians. The disease is also very interesting in terms of its biological behavior. There have been no major studies assessing the genetic analysis of in-transit melanomas. However, there are questions as to whether the metastases are clonal in origin. We took on the approach of assessing clonality of these in-transit metastatic and respective primary lesions by assessing commonly found microsatellites with LOH in melanomas. The use of LCM provided a unique opportunity to assess intratumor and intertumor heterogeneity. In 19 of the 25 patients (informed patients) LOH was detected for at least one of the eight microsatellite markers. The frequency of individual markers was as follows in order of frequency: D9S157 (56%); D9S304 (47%); D11S2000 (39%); D10S212 (32%); D1S214 (32%); D6S264 (22%); D1S228 (11%); and D3S1293 (10%). In examining 79 lesions from 25 patients there were only six lesions showing different LOH profiles compared to the other respective lesions of the patient. In these six lesions no significant marker stood out. We examined intratumor
D11S2000
306
D9S157
Before LCM N
After LCM A B
GENETIC APPLICATIONS
A
B
[27]
FIG. 1. Assessment of tumor heterogeneity using LCM. Subcutaneous recurrent melanoma in lymph node. Assessment of two microsatellite loci (D9S157 and D11S2000). The tumor tissue sections were isolated separately from two regions by LCM. Photograph of before and after LCM shown (H&E stain, magnification 40×). The arrows indicate the two alleles being compared. N refers to DNA from respective patients lymphocytes. LOH for D9S157 was noted in A section, and for D11S2000 in both A and B sections.
D3S1293
D9S157
N
After LCM
B
C
A
B
307
FIG. 2. Assessment of tumor heterogeneity using LCM. Metastatic melanoma in lymph node. Assessment of three microsatellite loci (D1S228, D3S1293, and D9S157). The tumor tissue sections were isolated separately from three regions by LCM. Photograph of before and after LCM shown (H&E stain, magnification 40×). The arrows indicate the three alleles being compared. N refers to DNA from respective patients lymphocytes. LOH for D1S228 and D9S157 was noted in A, B, and C sections. Retention was noted for D3S1293 in A, B, and C sections.
LCM
C
TUMOR GENETIC CLONALITY ASSESSMENT BY
A
[27]
D1S228 Before LCM
308
[27]
GENETIC APPLICATIONS TABLE I LOH ANALYSIS: INTER- AND INTRATUMOR HETEROGENEITYa Patient A (inter) A (intra) B (inter) B (intra) C (inter) C (intra) D (inter) D (intra) a
Marker D1S228 D9S304 D1S214 D11S2000
Lesion 1 R RRR R RRR L LLL R RRR
Lesion 2
Lesion 3
Lesion 4
R RRR L LRL R LRR L RLL
R RRL L LLL R RRR R RRR
L LLL L LLL R RRR — —
Representative examples of inter- and intratumor heterogeneity of specific LOH markers from individual melanoma patients (A–D). Lesions 1 to 4 refer to separate melanoma lesions within a melanoma patient with in-transit metastases. L, LOH of an allele; R, retained allele.
heterogeneity in 26 lesions from six patients. Tumor analysis involved LCM of three areas within each tumor lesion. The LOH analysis demonstrated that six of 26 lesions (23%) showed intratumor heterogeneity for at least one microsatellite marker. Representative examples of inter- and intratumor heterogeneity of melanoma lesions from in-transit patients are shown in Table I. In assessment of primary lesions there was no LOH marker occurrence that was in the metastasis and not the primary tumor. We developed a more sensitive approach of assessing LOH of microsatellite markers using CAE. The CAE allows utilization of smaller amounts of DNA and is more sensitive than traditional approaches using slab gel analysis. In combination with LCM of small amounts of tissue the CAE provides a powerful approach of assessment of multiple markers. In our laboratory we have been validating the utilization of CAE in assessment of LCM tissues. Using a single section, 5 to 7 µm, we are able to dissect out small sections with the LCM and run several markers. Figures 1 and 2 are representative examples of analysis of melanoma lymph node metastasis. Tumor heterogeneity in LOH markers is shown in Fig. 1, whereas in Fig. 2 there is no heterogeneity observed for the markers assessed. Although there is significant improvement in microdissection of tissue with the LCM there is still a problem in nucleic acid isolation from specimens. This is particularly a major problem when using archived (formalin fixed) paraffin-embedded tissue blocks. Improvement in nucleic acid isolation from small numbers of cells as well as alternative approaches to fixation of tissue is needed. Studies on frozen sections are one alternative; however, the morphology is poor and LCM on frozen sections is a logistical problem. Analysis of LOH using CAE provides a very
[28]
GENE MUTATIONS IN PROLIFERATIVE PROSTATIC DISEASES
309
sensitive assessment of LOH. This approach is more accurate and quantitative compared to traditional radioisotope labeling and vertical gel electrophoresis. The CAE analysis allows one to use smaller amounts of DNA. This is particularly useful when employing LCM and examining a small number of cells. The combination of CAE and LCM provides a very powerful tool for assessment of genetic changes of clones within tumors. This approach will help identify genetic changes in clonal populations within tumor lesions. These studies in the future will help identify tumor clonal populations based on genetic markers and how they exist within lesions. The utilization of fine-tuned microdissection may one day provide information on genetic patterns of evolution within a tumor that will lead to specific tumor phenotypes. The studies presented also indicate that caution should be taken in assessment of tumor lesions in terms of sampling numbers and sites. This is particularly important in using genetic markers as “correlatives” of predicting disease outcome.
Acknowledgment Supported in part by the National Institutes of Health PO1, Project II Grant CA-29605.
[28] Gene Mutations: Analysis in Proliferative Prostatic Diseases Using Laser Capture Microdissection By HITOSHI TAKAYAMA, NORIO NONOMURA, and KATSUYUKI AOZASA Introduction Prostatic cancer (PCA) is one of the commonest forms of cancer. For prevention or early detection, it is important to understand the character of the precancerous lesions in the prostate. Prostatic intraepithelial neoplasia (PIN) is characterized by an intraluminar proliferation of epithelial cells in the ducts and acini. PIN frequently coexists with prostatic carcinoma (PCA)1–3 and is commonly found in the nontransition zone which is the dominant site for PCA.4,5 According to histologic and cytologic findings, PIN is divided into high-grade PIN (HGPIN) 1
D. G. Bostwick and M. K. Brawer, Cancer 59, 788 (1987). D. G. Bostwick, Cancer 75, 1823 (1995). 3 F. J. Skjørten, A. Berner, S. Harvei, T. E. Robsahm, and S. Tretli, Cancer 79, 1172 (1997). 4 M de la Torre, M. J. H¨ aggman, S. Br¨andstedt, and C. Busch, Br. J. Urol. 72, 207 (1993). 5 J. Qian, P. Wollan, and D. G. Bostwick, Hum. Pathol. 28, 143 (1997). 2
METHODS IN ENZYMOLOGY, VOL. 356
Copyright 2002, Elsevier Science (USA). All rights reserved. 0076-6879/02 $35.00
310
GENETIC APPLICATIONS
[28]
and low-grade PIN (LGPIN).6 Studies have revealed that HGPIN and PCA share common cytogenetic features. Allelic loss of chromosome 8p was frequent in both HGPIN and invasive PCA.7,8 Mutations of the H-ras gene were closely associated with the progression of HGPIN to invasive PCA in transgenic mice, although this was not confirmed in humans.9 These findings indicate that HGPIN, not LGPIN, is the most likely precursor lesion for PCA. For detailed characterization of molecular genetic changes in PIN lesions, in situ excision of the lesions is indispensable. However, it is difficult to excise exclusively the PIN lesions, which are fairly small and cannot be defined macroscopically. In addition, HGPIN frequently coexists with cancer or benign glands in haphazard fashion in the nontransition zone. Laser capture microdissection under direct microscopic visualization enables rapid one-step procurement of selected human cell populations from histologic sections. This method has made microdissection of selected cells much easier; thus extensive study on the objective lesions has become possible. Using this method on whole-mount samples, we selectively microdissected numerous lesions of PIN and PCA from the nontransition and transition zones. DNA extracted from each lesion was analyzed for p53 mutations by single-strand conformation polymorphism (SSCP) of polymerase chain reaction (PCR)-amplified DNA fragments, followed by direct sequencing. Genomic DNA extracted from each lesion was analyzed for Fas mutations. The p53 gene is a tumor suppressor gene on the short arm of chromosome 17, which consists of 11 exons and 10 introns and encodes 393 amino acids of p53 proteins.10,11 In a wide variety of human tumors, p53 gene mutations have been detected mainly in exons 5 through 8, which include highly conserved domains II–V.12 Fas antigen is a 45-kDa transmembrane protein of the tumor necrosis factor (TNF) receptor superfamily that can induce programmed cell death (apoptosis) through cross-link with the Fas ligand (FasL).13,14 Fas is situated on chromosome 10q24.1 and comprises 9 exons and 8 introns. The Fas gene encodes 325 amino acids, which are divided into extracellular, transmembrane, and intracytoplasmic domains. The 80-amino acid portion in the intracytoplasmic domain is essential 6
R. Montironi, D. G. Bostwick, H. Bonkhoff, A. T. K. Cockett, B. Helpap, P. Troncoso, and D. Waters, Cancer 78, 362 (1996). 7 M. R. Emmert-Buck, C. D. Vocke, R. O. Pozzatti, P. H. Duray, S. B. Jennings, C. D. Florence, Z. Zhuang, D. G. Bostwick, L. A. Liotta, and W. M. Linehan, Cancer Res. 55, 2959 (1995). 8 M. J. H¨ aggman, K. J. Wojno, C. P. Pearsall, and J. A. Macoska, Urology 50, 643 (1997). 9 M. Shibata, J. M. Ward, D. E. Devor, M. L. Liu, and J. E. Green, Cancer Res. 56, 4894 (1996). 10 S. J. Baker, E. R. Fearon, J. M. Nigro, S. R. Hamilton, A. C. Preisinger, J. M. Jessup, P. van Tuinen, D. H. Ledbetter, D. F. Baker, Y. Nakamura, R. White, and B. Vogelstein, Science 244, 217 (1989). 11 C. A. Finlay, P. W. Hins, and A. J. Levine, Cell 57, 1083 (1989). 12 M. Hollstein, D. Sidransky, B. Vogelstein, and S. R. Harris, Science 253, 49 (1991). 13 T. Suda, T. Takahashi, P. Goldstein, and S. Nagata, Cell 75, 1169 (1993). 14 S. Nagata, Cell 88, 355 (1997).
[28]
GENE MUTATIONS IN PROLIFERATIVE PROSTATIC DISEASES
311
for apoptotic signal transduction and thus is designated as death-signaling.15,16 Mutations of the Fas gene in the death domain lead to loss of its apoptosis function, a loss-of-function mutation, which may contribute to the pathogenesis of human malignancies. Indeed, Fas gene mutations have been reported in both lymphoid lineage and epithelial malignancies: approximately 10% of cases with multiple myeloma17 and sporadic non-Hodgkin’s lymphoma,18 7.7% with lung cancer,19 and 28% with urinary bladder cancer.20 Cases Twenty-seven PCA patients, who underwent prostatectomies, were selected for the study. They had been admitted to hospitals during the period 1996 to 1998. None of the 27 patients received preoperative chemotherapy or radiation therapy. Based on the American staging system (modified by Whitmore-Jewett),21 11 cases (41%) were determined to be in stage T2 and 16 (59%) in T3. Histologic specimens were fixed in 10% neutral buffered formalin and routinely processed for paraffin embedding. Serial 5-µm sections were cut and stained with hematoxylin and eosin and reviewed independently by three pathologists (Fig. 1). Mean number of sections examined was 9.3 per case. Diagnosis of HGPIN was made based on histologic and cytologic features, i.e., intraluminar proliferation of glandular epithelial cells with large nuclei and prominent nuclei. Basal layer is partially disrupted. Laser Capture Microdissection and DNA Extraction Microdissection of each lesion was performed using a PixCell laser capture microscope (Arcturus Engineering, Santa Clara, CA) according to the previously described methods with some modifications.22,23 Briefly, histologic sections are 15
N. Itoh, S. Yonehara, M. Ishii, S. Yonehara, M. Mizushima, A. Sameshima, A. Hase, Y. Seto, and S. Nagata, Cell 66, 233 (1991). 16 R. Watanabe-Fukunaga, C. Brannan, N. Itoh, S. Yonehara, N. G. Copeland, N. A. Jenkins, and S. Nagata, J. Immunol. 148, 1274 (1992). 17 T. H. Landowski, N. Qu, I. Buyuksal, S. Painter, and W. S. Dalton, Blood 90, 4266 (1997). 18 K. Grønbæck, P. T. Straten, E. Ralfkiaer, V. Ahrenkiel, M. K. Andersen, N. E. Hansen, J. Zeuthern, K. Hou-Jensen, and P. Guldberg, Blood 92, 3018 (1998). 19 S. H. Lee, M. S. Shin, W. S. Park, S. Y. Kim, H. S. Kim, J. Y. Han, G. Y. Park, S. M. Dong, J. H. Pi, C. H. Kim, S. H. Kim, J. Y. Lee, and N. J. Yoo, Oncogene 18, 1754 (1999). 20 S. H. Lee, M. S. Shin, W. S. Park, S. Y. Kim, H. S. Kim, J. Y. Han, G. Y. Park, S. M. Dong, J. H. Pi, C. H. Kim, S. H. Kim, J. Y. Lee, and N. J. Yoo, Cancer Res. 59, 3068 (1999). 21 D. G. Bostwick, R. P. Myers, and J. E. Oesterling, Semin. Surg. Oncol. 10, 60 (1994). 22 M. R. Emmert-Buck, R. F. Boner, P. D. Smith, R. Chuaqui, Z. Zhaung, S. R. Goldstein, R. A. Weiss, and L. A. Liotta, Science 274, 998 (1996). 23 R. F. Bonner, M. Emmert-Buck, K. Cole, T. Pohida, R. Chuaqui, S. Goldstein, and L. A. Liotta, Science 278, 1481 (1997).
312
GENETIC APPLICATIONS
[28]
FIG. 1. Representative cases of prostatic carcinoma (Gleason pattern 5 = 2 + 3, Grade II) (A), and high-grade prostatic intraepithelial neoplasia (B). Hematoxylin and eosin staining; original magnification 200×.
dehydrated, and then the histologic fields of interest are selected, are overlaid with a thermoplastic film mounted on a transparent cap, and are captured by the film through laser energy (Fig. 2). The dissected pieces are allowed to adhere to the transparent cap and collected in 0.6-ml Eppendorf tubes. The cells are subsequently resuspended in 20–50 µl of extraction buffer containing 10 mM Tris (pH 8.0), 2 mM EDTA, 0.2% Tween 20, and 200 µg/ml proteinase K, and are incubated overnight at 37◦ . The mixture is heated at 100◦ for 10 min to inactivate
[28]
GENE MUTATIONS IN PROLIFERATIVE PROSTATIC DISEASES
313
FIG. 2. HGPIN lesion (A) in representative case is successfully microdissected (B). Note successful resection of intraductal epithelial cells (Arcturus Engineering, Santa Clara, CA). Hematoxylin and eosin staining; original magnification 200×.
the proteinase K, and 3–5% of solution is used as a template for each PCR. The total number of microdissected lesions from the 27 cases was 193 : 111 lesions with HGPIN (75 lesions from nontransition and 36 from transition zone), 55 with PCA (30 lesions from nontransition and 25 from transition zone), and 27 with benign glands.
314
GENETIC APPLICATIONS
[28]
p53 Gene Mutations Mutations of the p53 gene from exon 5 to exon 8 were analyzed. One µl microdissected DNA template was subjected to PCR of 35 cycles with the oligonucleotide primers, denaturation for 30 sec at 95◦ , annealing for 30 sec at variable temperatures, and extension for 30 sec at 72◦ (Table I) in a 9700 Applied Biosystems Thermocycler (Foster City, CA). Nonradioactive SSCP was performed according to the previously reported method, with some modifications.24,25 Briefly, a mixture containing 7 µl of PCR product, 0.4 µl of 1 M methylmercury hydroxide (Wako), 3.0 µl of 15% (w/v) Ficoll loading buffer containing 0.25% bromphenol blue and 0.25% xylene cyanol, and 10 µl of 10× TBE buffer was prepared, heated at 85◦ for 4 min, and then put on ice. Twenty-µl aliquots of the mixture were subjected to electrophoresis in 18% polyacrylamide TBE gels at 500 volts with the temperature maintained at 35◦ for exons 5 and 6 and 25◦ for exons 7 and 8 in a circulating temperature control unit. Possible mutated bands detected by SSCP were extracted from gels and amplified by PCR under the same primer. PCR products were purified using the QIAquick PCR Purification Kit (Qiagen, CA, USA) and were sequenced by the dideoxy chain termination method using the DNA sequencing kit (Applied Biosystems). The samples were analyzed with the Genetic Analyzer (ABI PRISM 310, Applied Biosystems). As shown in Table II, 27 mutations of the p53 gene were detected in 24 lesions from 12 cases (Fig. 3). All were point mutations; 17 were missense, 7 silent, and 2 nonsense mutations. There were no mutational hot spots, although exon 5 was the commonest site. In cases 3, 19, and 20, there were double mutations with different types of nucleotide substitutions in the same exons. Mutations were deteced in 6 cases (22.2%) or 13 of 111 lesions (11.7%) with HGPIN and 8 cases (29.6%) or 11 of 55 lesions (20.0%) with PCA. Benign proliferative glands adjoining PIN and/or PCA had no mutations of the p53 gene. The PCA cases with mutations were in stage T2 (2 cases) and T3 (6 cases). In cases 19 and 20, two each of PCA and PIN lesions had different mutations. In cases 6, 16, 17, 18, and 21, each of the HGPIN and PCA lesions had mutations different from one another. Mutations at CpG sites were found in one case (case 21). Regarding patterns of p53 mutations, G-to-A transition was the commonest (6/27; 22.2%), followed by C-to-T transition (5 mutations) and A-to-C transversion (5 mutations). Frequency of p53 mutation of PCA in the nontransition zone (33.3%) was significantly higher than that in the transition zone (4%) (p < 0.05) (Table III). Frequency of p53 mutation in PCA with stage T3 (30.3%), 10 of 33 lesions, was significantly higher than that with stage T2 (4.5%), 1 of 22 lesions ( p < 0.05). 24 25
T. Hongyo, G. S. Buzard, R. J. Calvert, and C. M. Weghorst, Nucleic Acids Res. 21, 3337 (1993). T. Hongyo, G. S. Buzard, D. Palli, C. M. Weghorst, A. Amorosi, M. Galli, N. E. Caporaso, J. F. Fraumeni, and M. Rice, Cancer Res. 55, 2665 (1995).
[28]
TABLE I AMPLIFICATION PRIMERS FOR MUTATION ANALYSIS OF THE p53 AND Fas GENE
Gene/Exon p53 (Exon5) p53 (Exon6) p53 (Exon7) p53 (Exon8) Fas (promoter) Fas (promoter) Fas (Exon3) Fas (Exon7) Fas (Exon8) Fas (Expon9)
Primer
Sequence (5 → 3 )
5U 5D 6U 6D 7U 7D 8U 8D PA-F PA-R PB-F PB-R 3-F 3-R 7-F 7-R 8-F 8-R 9-F 9-R
5 -gtactcccctgccctcaaca-3 5 -ctcaccatcgctatctgagca-3 5 -ttgctcttaggtctggcccc-3 5 -cagacctcaggcggctcata-3 5 -taggttggctctgactgtacc-3 5 -tgacctggagtcttccagtgt-3 5 -agtggtaatctactgggacgg-3 5 -acctcgcttagtgctccctg-3 5 -gccctataccatcctccttat-3 5 -ctgtcactgcacttaccacc-3 5 -cctcttgaaaataaaaact-3 5 -tcactcagagaaagacttgcgg-3 5 -acttcccaccctgttacctg-3 5 -acttcccaccctgttacctg-3 5 -tcttagtgtgaaagtatgttctc-3 5 -caaatcactaatttctctatttt-3 5 -attaaggaaaaattagaagttcacat-3 5 -atcccataatatgtcactgaaa-3 5 -ggttttcactaatgggaatttca-3 5 -tatgttggctcttcagcgcta-3
Second PCR Size of PCR product (bp)
Annealing temperature
194
58◦
128
60◦
117
60◦
141
60◦
172
55◦
242
56◦
310
55◦
223
46◦
217
50◦
536
50◦
Sequence (5 → 3 )
5 -gccctataccatcctccttat-3 5 -gtaggtgttgataggcttgtct-3 5 -cctcttgaaaataaaaact-3 5 -gacttgcggggcatttgac-3 5 -acttcccaccctgttacctg-3 5 -tgtgtgtcaacatagcaccac-3 5 -ctacaaggctgagacctgagtt-3 5 -aggaagtaacaaaaagccaaatc-3 5 -attaaggaaaaattagaagttcacat-3 5 -atcccataatatgtcactgaaa-3 5 -ggttttcactaatgggaatttca-3 5 -ctaattgcatatactcaggaa-3
Size of PCR product (bp)
Annealing temperature
145
55◦
228
56◦
250
55◦
203
55◦
181
50◦
443
50◦
GENE MUTATIONS IN PROLIFERATIVE PROSTATIC DISEASES
First PCR
315
TABLE II p53 MUTATIONS IN PIN AND CONCURRENT PCAa
Case
Zone of lesion
3
N-T
PCA/Gleason 2, T3
6
19
N-T N-T N-T N-T N-T N-T N-T N-T N-T N-T
PCA/Gleason 2, T3 PCA/Gleason 3, T3 PCA/Gleason 2, T3 HGPIN HGPIN HGPIN HGPIN HGPIN HGPIN PCA/Gleason 3, T3
20
N-T N-T N-T N-T
9 16 17 18
N-T N-T N-T N-T T T T N-T N-T
21
24 25 26
Distance from PCA (mm)
1 1 2 1 1 1
0 1
1 0 0 13 13
Histology/ pathological stage
PCA/Gleason 3, T3 HGPIN HGPIN PCA/Gleason 5, T3 PCA/Gleason 5, T3 HGPIN HGPIN HGPIN HGPIN HGPIN PCA/Gleason 2, T2 PCA/Gleason 3, T3 PCA/Gleason 3, T3
Exon/codon
Mutation
Pattern
7/235 7/237 7/235 8/296 7/243 5/145 5/128 8/285 8/291 8/285 5/145 5/164 5/154 5/172 5/142 7/248 5/151 5/167 5/172 5/145 5/141 5/151 5/165 5/158b 8/285 8/279 5/151
Asp(AAC)→His(CAC) Met(ATG) →Arg(AGG) Asp(AAC)→Thr(ACC) His(CAC) →Asn(AAC) Cys(TGC) →Tyr(TAC) Gly(GGC)→Asp(GAC) Pro(CCT) →Leu(CTT) Glu(GAG)→Stop(TAG) Lys(AAG)→Thr(ACG) Glu(GAG)→Stop(TAG) Leu(CTG) →Leu(CTA) Lys(AAG →Lys(AAA) Gly(GGC) →Asp(GAC) Val(GTT) →Ala(GCT) Pro(CCT) →Leu(CTT) Arg(CGG)→Arg(AGG) Pro(CCC) →Pro(CCG) Gln(CAG) →Gln(CAA) Val(GTT) →Ala(GCT) Leu(CTG) →Leu(TTG) Leu(CTG) →Leu(TTG) Pro(CCC) →Pro(CCG) Gln(CAG) →Gln(CAA) Arg(CGC) →Asp(CAC) Glu(GAG) →Stop(TAG) Gly(GGG)→Gly(GGT) Pro(CCC) →Pro(CCG)
Tv missense Tv missense Tv missense Tv missense Tv missense Ts silent Ts missense Ts nonsense Tv missense Tv nonsense Ts silent Ts silent Ts missense Ts missense Ts missense Tv silent Tv silent Ts silent Ts missense Ts silent Ts missense Tv silent Ts silent Ts missense Tv nonsense Tv silent Tv silent
a
PIN, prostatic intraepithelial neoplasia; PCA, prostatic carcinoma; Ts, transition; Tv, transversion; T, transition zone; N-T; nontransition zone. b Mutation at CpG sites. From H. Takayama, M. Shin, N. Nonomura, A. Okuyama, and K. Aozasa, Jpn. J. Cancer Res. 91, 941 (2000).
FIG. 3. PCR-single-strand conformation polymorphism (SSCP) analysis of p53 mutation. Nonradioactive SSCP analysis of exon 7. Aberrant migration patterns (arrow) were seen in lane 1 (case 3, PCA), lane 2 (case 6, PCA), and lane 3 (case 19, PIN). Wild-type SSCP bands are shown in lane 4 (case 19, benign prostatic hypertrophy).
[28]
317
GENE MUTATIONS IN PROLIFERATIVE PROSTATIC DISEASES TABLE III SUMMARY OF p53 MUTATIONS IN PCA AND PIN Mutation frequency (%) Stage
PCA PIN
Zone
Distance from PCA
Case
Lesion
T2a
T3a
Transitiona
Non-transitiona
< =2mm
8/27 (29.6) 6/27 (22.2)
11/55 (20.0) 13/111 (11.7)
1/22 (4.5) 4/33 (12.1)
10/33 (30.3) 9/78 (11.5)
1/25 (4.0) 2/36 (5.6)
10/30 (33.3) 11/75 (14.7)
— 11/46a (24.0)
>2mm — 2/65a (3.0)
< 0.05. From H. Takayama, M. Shin, N. Nonomura, A. Okuyama, and K. Aozasa, Jpn. J. Cancer Res. 91, 941 (2000).
ap
Frequency of p53 mutation in PIN in the nontransition zone (14.7%) was higher than that in the transition zone (5.6%), although the difference was not significant. The frequency rate of p53 mutation in HGPIN close to PCA (< =2 mm of distance) was significantly higher (24%) than that in an HGPIN lesion with >2 mm of distance from PCA (3%) (p < 0.05). Information on the molecular genetic characteristics of PIN was quite limited until the development of the microdissection technique. Previous studies showed that allelic loss of chromosomes 8p, 10q, and 16q was frequent both in the HGPIN and in invasive PCA, suggesting the involvement of tumor suppressor genes or oncogenes located on these loci.7,8,26,27 In our studies, the frequency rate of p53 gene mutations in the PCA lesions, mainly from stages T2 and T3 of the disease (20%), was close to that in the cases reported with advanced PCA.28–30 Frequency of the p53 gene mutations in the HGPIN lesions was 12%. There is a difference in mutation frequency of PCA lesions in the different stages of the disease: one of 22 lesions (4.5%) in T2 and 10 of 33 lesions (30.3%) in T3 stage showed the mutations. This suggests that p53 gene mutations may be 26
I. C. Gray, S. M. A. Phillips, S. L. Lee, J. P. Neoptolemos, J. Weissenbach, and N. Spurr, Cancer Res. 55, 4800 (1995). 27 S. E. Strup, R. O. Pozzatti, C. D. Florence, M. R. Emmert-Buck, P. H. Duray, L. A. Liotta, D. G. Bostwick, W. M. Linehan, and C. D. Vocke, J. Urol. 162, 590 (1999). 28 R. Bookstein, D. MacGrogan, S. G. Hilsenbeck, F. Sharkey, and D. C. Allred, Cancer Res. 53, 3369 (1993). 29 M. Watanabe, T. Ushijima, H. Kakiuchi, T. Shiraishi, R. Yatani, J. Shimazaki, T. Kotake, T. Sugimura, and M. Nagao, Jpn. J. Cancer Res. 85, 904 (1994). 30 D. Mirchandani, Z. Zheng, G. J. Miller, A. K. Ghosh, D. K. Shibata, R. J. Cote, and P. Roy-Burman, Am. J. Pathol. 147, 92 (1995).
318
GENETIC APPLICATIONS
[28]
involved in the progression of PCA. Berner et al.31 reported that C-to-G transversion at codon 273 was frequently found in PCA. They suggested that this may be a mutational hot spot in the progression of PCA. Although exon 5 was the commonest site for mutations in our series, there were no mutational hot spots. Mutations at the CpG site were found in one case (case 21). G-to-A transition and C-to-G transversion substitution was most common in our series. In 8 of 12 cases with PCA and/or PIN lesions, p53 mutations had at least one mutation that changes an amino acid, which may provide selection pressure for expansion. PCA with various grades of histologic differentiation is common. Previous studies using the fluorescence in situ hybridization (FISH) technique showed allelic loss of 8p in PIN and PCA lesions32 and suggested a common “genetic history” for these proliferations. Also, FISH studies revealed overexpression of c-myc.33 Konishi et al.34 reported different patterns of p53 alterations among multifocal lesions of PCA. In cases 19 and 20 of our series, the direct sequencing of the PCRSSCP products from 4 independent foci, 2 PIN and 2 PCA, showed a different pattern of p53 mutations, indicating each focus to be derived from different cell clones. The presence of different clones in the same prostatic lesions was also shown in another five cases (cases 6, 16, 17, 18, and 21). Multiclonality of prostatic precancerous and cancerous lesions is not surprising in the light of multistep carcinogenesis. The presence of precancerous lesions on the verge of becoming cancerous should be taken into account when treating patients with PCA. Our study35 has shown that the HGPIN, like PCA, is sensitive to androgen deprivation therapy and is occasionally hard to recognize after hormone therapy, even on whole-mount prostatectomy specimens. The nontransition zone is known to be the dominant site for PCA and HGPIN.36,37 Coexistence of PCA and HGPIN lesions in the nontransition zone was found in approximately 75% of PCA cases,35 supporting the precancerous nature of HGPIN. Our study revealed that the frequency of p53 mutations in PCA lesions was significantly higher in the nontransition than in the transition zone. As for HGPIN, p53 mutations in the nontransition zone were significantly more 31
A. Berner, G. Geitvik, F. Karlsen, S. D. Fossa, J. M. Nesland, and A. L. Børresen, J. Pathol. 176, 299 (1995). 32 J. Qian, D. G. Bostwick, S. Takahashi, T. J. Borell, J. F. Herath, M. M. Lieber, and R. B. Jenkins, Cancer Res. 55, 5408 (1995). 33 R. B. Jenkins, J. Qian, M. M. Lieber, and D. G. Bostwick, Cancer Res. 57, 524 (1997). 34 N. Konishi, Y. Hiasa, H. Matsuda, M. Tao, T. Tsuzuki, I. Hayashi, Y. Kitahori, T. Shiraishi, R. Yatani, J. Shimazaki, and J. C. Lin, Am. J. Pathol. 147, 1112 (1995). 35 M. Shin, H. Takayama, N. Nonomura, A. Wakatsuki, A. Okuyama, and K. Aozasa, Prostate 42, 81 (2000). 36 J. E. McNeal, E. A. Redwine, F. S. Freiha, and T. A. Stamey, Am. J. Surg. Pathol. 12, 897 (1988). 37 D. R. Greene, T. M. Wheeler, S. Egawa, J. K. Dunn, and P. T. A. Scardino, J. Urol. 146, 1069 (1991).
[28]
GENE MUTATIONS IN PROLIFERATIVE PROSTATIC DISEASES
319
frequent than in the transition zone. These findings suggest that p53 gene mutations play a role in the development of precancerous and cancerous lesions in the nontransition zone, but not in the transition zone. The distance between the PCA and HGPIN was reported to be frequently within 2 mm.38 Bostwick and Brawer1 reported that the frequency of appearance of HGPIN increased in cases with PCA compared to those without PCA. In our previous study35 close association (distance within 2 mm) of HGPIN with PCA was more frequently found in the nontransition zone (63% of lesions) than in the transition zone (38% of lesions). These findings provide a basis for suggesting the precancerous nature of HGPIN, especially in the nontransition zone. Indeed, the frequency rate of p53 mutations in HGPIN lesions close to PCA (24% of lesions) was significantly higher than in those distant from PCA (>2 mm) (3% in total and none in the noncastrated cases). Our study using the laser capture microdissection method clearly showed the significant role of p53 gene mutations in the development of HGPIN and PCA in the nontransition zone with the sequential occurrence of HGPIN to PCA when these lesions were close to one another.
Fas Gene Mutations The death domain is necessary for the transduction of the apoptotic signal15,39,40; therefore we examined mutations in exons 7, 8, and 332 bp of exon 9. DNA was subjected to first-round PCR of 10 cycles with the oligonucleotide primers followed by second PCR of 35 cycles with use of 0.1% of first-round PCR products as the template, denaturation for 30 sec at 95◦ , annealing for 30 sec at variable temperatures, and extension for 30 sec at 72◦ (Table I) in a 9700 Applied Biosystems Thermocycler (Foster City, CA). PCR products were purified using the QIAquick PCR Purification Kit (Qiagen, CA) and were sequenced by the dideoxy chain termination method using the DNA sequencing kit (Applied Biosystems). The samples were analyzed with the Genetic Analyzer (ABI PRISM 310, Applied Biosystems). PCR products with suspicious mutations were cloned in the pCR 2.1-TOPO (Invitrogen), then sequenced to confirm whether the mutation exist. As shown in Table IV, 4 mutations of the Fas gene were detected in 4 HGPIN lesions from 4 cases. All mutations were point mutations; 3 missense and 1 nonsense mutation detected in exon 9, which encodes the death domain region of the Fas receptor.14 Substitutions at codon 261 of the Fas cDNA sequence (GenBank 38
J. Qian and D. G. Bostwick, Pathol. Res. Pract. 191, 860 (1995). N. Itoh and S. Nagata, J. Biol. Chem. 268, 10932 (1993). 40 J. Cheng, O. Liu, and J. D. Mountz, J. Immunol. 154, 1239 (1995). 39
320
TABLE IV MUTATIONS AND LOH OF THE Fas GENE IN HIGH-GRADE PROSTATIC INTRAEPITHELIAL NEOPLASIA AND PROSTATIC CANCERa Mutation Lesion
Castration
Stage
3 10 21
HGPIN HGPIN HGPIN PCA HGPIN1 HGPIN2 PCA1 PCA2 HGPIN1 HGPIN2 HGPIN3 HGPIN4 HGPIN5 PCA HGPIN1 HGPIN2 HGPIN3 HGPIN4 PCA HGPIN PCA
+ + − − + + + + − − − − − − − − − − − + +
pT2bNxMx pT2bNxMx pT3aNxMx pT3aNxMx pT3aNxMx pT3aNxMx pT3aNxMx pT3aNxMx pT2bNxMx pT2bNxMx pT2bNxMx pT2bNxMx pT2bNxMx pT2bNxMx pT3aNxMx pT3aNxMx pT3aNxMx pT3aNxMx pT3aNxMx pT3aNxMx pT3aNxMx
Positive Negative Negative Negative Positive Positive Positive Positive Positive Positive Positive Positive Positive Positive Positive Negative Negative Positive Positive Positive Positive
23
24
25
27
a
Sites
LOH analysis
Codon Position Nucleotide Amino acid −1377 −670 Exon 3 Exon 7
Exon 9 Exon 9 Exon 9
254 260 260
1002 1020 1021
ACA/GCA CAA/TAA CAA/CGA
Thr/Ala Gln/Stop Gln/Arg
Exon 9
285
1099
AAT/AGT
Asn/Ser
NI NI NI NI NI NI NI NI NI NI NI NI NI NI NI NI NI NI NI NI NI
NI NI HET LOH LOH LOH LOH LOH LOH LOH LOH LOH LOH LOH LOH LOH LOH LOH LOH LOH LOH
NI NI NI NI NI NI NI NI NI NI NI NI NI NI NI NI NI NI NI NI NI
NI NI NI NI HET LOH HET LOH NI NI NI NI NI NI NI NI NI NI NI NI NI
[28]
HGPIN, High-grade prostatic intraepithelial neoplasia; PCA, prostatic cancer; NI, not informative; HET, retention of heterozygosity; LOH, loss of heterozygosity. Reproduced with permission from Lab. Invest. 81, 283 (2001).
GENETIC APPLICATIONS
Case
Immunohistochemistry for Fas protein
[28]
GENE MUTATIONS IN PROLIFERATIVE PROSTATIC DISEASES
321
accession No. M67454), Gln to Stop (case 10) and Gln to Arg (case 21), respectively, were found in the HGPIN lesions. Regarding the mutational patterns, all were transitional: G to A, C to T, and A to G. G-to-C or C-to-A transversion was not found. All of the HGPIN lesions with mutations were present in the nontransition zone. Neither PCA nor benign proliferative glands adjoining HGPIN and/or PCA showed Fas mutations. Loss of heterozygosity (LOH) was examined at four sites of known polymorphisms, i.e., at positions −1377, −670 (promoter region), 416 (exon 3), and 836 (exon 7). DNA was amplified using primers flanking the four polymorphic sites (Table I). Polymorphisms at −1377, 416, and 836 were examined by direct sequencing, and that at position −670 by restriction fragment length polymorphism by digestion with MvaI enzyme (Fermentas, Vilnius, Lithuania).41 Sixteen of 27 cases (59.6%) were heterozygous for one or more sites of the known biallelic polymorphisms, i.e., at the positions −1377, −670, 416, and 836. Of these 16 cases, 5 (31.3%) with PCA and 4 (25%) with HGPIN showed LOH at promoter region (−670) and exon 7. A HGPIN lesion in case 24 had missense mutation at position 1100 and LOH at −670. In case 21, the HGPIN lesion had mutation at exon 9 but no LOH, and the PCA lesion had LOH but no mutation. Immunohistochemical studies on the paraffin sections were carried out using the avidin–biotin–peroxidase complex (ABC) method. For detection of Fas protein, mouse anti-human Fas antibody (4B4-B3) that recognizes the extracellular domain of Fas was prepared by Dr. S. Nagata (unpublished data). No relationship was found between immunoreactivity for anti-Fas antibody and mutation or LOH in HGPIN and PCA lesions. Fas protein was expressed in 11 of 15 HGPIN lesions (73.3%) and 5 of 6 PCA lesions (83.3%), respectively. Through construction of a detailed deletion map spanning 10q23–25, Gray et al.26 suggested the presence of prostate tumor suppressor genes near the 10q23–24 boundary, which is close to the location of the Fas gene, 10q24.1. With use of the laser capture microdissection method, we could analyze the Fas gene mutations in numerous HGPIN lesions. The Fas gene mutations were detected in 4 of 27 cases (14.8%) or 4 of 111 (3.6%) lesions with HGPIN, whereas none of the 55 lesions with PCA had Fas gene mutations, indicating that PCA develops among HGPIN without Fas gene mutations. Benign proliferative glands adjoining HGPIN and/or PCA never showed Fas gene mutations. Thus Fas gene mutation may not contribute to pathogenesis of PCA. As for mutational sites and patterns of the Fas gene, the point mutation at codon 253 was reported in two patients with multiple myeloma.17 In non-Hodgkin’s lymphomas, mutations at codons 248 and 251 were identified.18 Lee et al.20 reported that 8 of 12 mutations found in bladder cancer showed G-to-A transition 41
Q. I. Huang, D. Morris, and N. Manollos, Mol. Immunol. 34, 577 (1997).
322
GENETIC APPLICATIONS
[28]
at codon 251, thus suggesting that this might be a mutational hot spot. Two of the HGPIN lesions showed mutations at codon 261. All mutations in our series were transitions, suggesting that some “endogenous” mutagens act in the pathogenesis of HGPIN. Missense mutations in the death domain are suggested to affect receptor function in a dominant-negative fashion,42 i.e., mutant Fas protein derived from mutated Fas gene of one allele may bind with normal Fas protein derived from another normal allele to construct a structurally abnormal Fas trimer, which may have a defect in binding to adapter proteins. Among four HGPIN lesions with a mutated Fas gene, one case (case 24) had LOH at position −670. Because the distance of sites between LOH and mutation is approximately 10 kb, LOH may involve the mutation site in the same allele. Therefore it is reasonable to consider that LOH and the mutation found in the HGPIN lesion of case 24 occurred at different alleles, thus resulting in the production of a predominantly mutant Fas protein. Occurrence of LOH was unknown in the remaining three cases (cases 3, 10, 21) with Fas mutations. In these cases, Fas function may be lost or reduced in a dominant-negative effect of the mutant Fas protein in cases in which LOH is absent or predominant production of mutant Fas protein in cases with LOH. In any case, Fas-mediated apoptosis may be disrupted in these four HGPIN lesions. Normal DNA repair mechanisms are important in maintaining the integrity of the genome. Humans are frequently exposed to naturally occurring DNA-damaging agents; thus the combined occurrence of DNA damage and impaired DNA repair function results in the development of neoplasias. Indeed, replication error, as revealed by microsatellite instability (MSI), was found in cases with PCA.43,44 Occurrence of LOH also indicates the underlying genetic instability in the lesional proliferating cells. Rohrbach et al.44 reported that MSI and LOH were found in 35% and 16%, respectively, of their PCA cases.44 In our cases, LOH at the four sites of the Fas gene was found in 31.6% of PCA and 25% of HGPIN lesions. Our results show that genetic instability occurs during the early phase of prostatic carcinogenesis.
42
G. H. Fisher, F. J. Rosenberg, S. E. Straus, J. K. Dale, L. A. Middleton, A. Y. Lin, W. Strober, M. J. Lenardo, and J. M. Puck, Cell 81, 935 (1995). 43 R. Dahiya, C. Lee, J. McCarville, W. Hu, G. Kaur, and G. Deng, Int. J. Cancer 72, 762 (1997). 44 H. Rohrbach, C. J. Hass, G. B. Baretton, A. Hirschmann, J. Diebold, R. P. Behrendt, and U. L¨ ohrs, Prostate 40, 20 (1999).
[29]
LCM-GENERATED RNA FOR MICROARRAY ANALYSIS
323
[29] Use of Laser Capture Microdissection-Generated Targets for Hybridization of High-Density Oligonucleotide Arrays By HIROE OHYAMA, MAMATHA MAHADEVAPPA, HEIKKI LUUKKAA, RANDY TODD, JANET A. WARRINGTON, and DAVID T. W. WONG Introduction The light microscope (LM) is a powerful research tool for visualization of cellular and tissue architecture. Its simplicity in operation makes it an attractive choice for obtaining precisely defined homogenous cell populations. Until recently it was impractical to use LM to harvest selective cell types from histological sections for biochemical and genetic analysis. Laser capture microdissection (LCM) allows for the precise isolation of individual cells or pure cell populations from complex tissue architectures suitable for biochemical and molecular analysis. DNA, RNA, and protein have been successfully isolated from LCM-procured cells from virtually any anatomic site or disease type. LCM represents a major advance over previous, manual-based approaches using fine (30-gauge or smaller) needles to dislodge cells from a tissue section. LCM uses a laser to indirectly or directly isolate cells from tissues. Indirect LCM approaches use a laser to remove surrounding tissue away from the target tissue. By ablating the circumscribing tissue, the target tissue is not exposed to the heat and radiation of the laser. By far the more common application, direct LCM uses a laser to melt a thermoplastic polymer over the target tissue, thereby binding the tissue to an Eppendorf tube cap for nucleic acid or protein isolation. Introduced 4 years ago, LCM has been successfully used to understand both normal human physiology and pathophysiology, including infectious disease, endocrine disorders, and cancer. This chapter details a specific application of LCM technology: the use of LCMprocured cells to generate sufficient RNA for global gene expression analysis by high-density oligonucleotide arrays. Rationale for LCM in Solid Tumor Research To the tumor biologist, the ability to selectively procure homogenous tumor, premalignant, and normal cells from the same patient is an important research objective. Although there are many research studies that will require tumor and stroma interactions, the segregation of these cell types was practically impossible prior to the development of LCM. LCM was originally developed to isolate pure premalignant cells from surrounding tissues to study the molecular events leading to invasive cancer. In many cases, this target cell population represented
METHODS IN ENZYMOLOGY, VOL. 356
Copyright 2002, Elsevier Science (USA). All rights reserved. 0076-6879/02 $35.00
324
GENETIC APPLICATIONS
[29]
less than 5% of the entire tissue section. This is particularly important for solid tumors where tumor cells form cords and nests as they infiltrate into the underlying stroma, making it impossible to separate cell types of distinct morphology using conventional approaches such as microdissection. LCM has been used to study adrenal-gland,1 brain,2 breast,3 colon,4,5 lymphoma,6 oral,7–10 pancreas,11 prostate,12–15 and thyroid16 tumors. Figure 1 illustrates a case of oral cavity tumor from the lateral border of the tongue. The cells of interest were transferred to a polymer film activated by laser pulses (Figs. 1B, 1C, and 1D). Need for Amplification of Laser Capture Microdissection-Generated RNA for Genome-Wide Expression Profiling by DNA Microarrays An important aspect of our research goal is to obtain the genome-wide molecular profiles of gene expression in normal, premalignant, and tumor oral keratinocytes. Although this could be done using proteomic and/or functional genomic approaches, current technology makes RNA profiling an efficient and cost-effective first step. While there are a number of different approaches for gene expression 1
A. Glasow, A. Haidan, J. Gillespie, P. A. Kelly, G. P. Chrousos, and S. R. Bornstein, Endocr. Res. 24, 857 (1998). 2 J. Mora, M. Akram, N. K. Cheung, L. Chen, and W. L. Gerald, Med. Pediatr. Oncol. 35, 534 (2000). 3 D. C. Sgroi, S. Teng, G. Robinson, R. LeVangie, J. R. Hudson, Jr., and A. G. Elkahloun, Cancer Res. 59, 5656 (1999). 4 O. Kitahara, Y. Furukawa, T. Tanaka, C. Kihara, K. Ono, R. Yanagawa, M. E. Nita, T. Takagi, Y. Nakamura, and T. Tsunoda, Cancer Res. 61, 3544 (2001). 5 D. Dillon, K. Zheng, and J. Costa, Exp. Mol. Pathol. 70, 195 (2001). 6 S. Yegappan, B. Schnitzer, and E. D. Hsi, Mod. Pathol. 14, 191 (2001). 7 H. Ohyama, X. Zhang, Y. Kohno, I. Alevizos, M. Posner, D. T. Wong, and R. Todd, BioTechniques 29, 530 (2000). 8 I. Alevizos, M. Mahadevappa, H. Ohyama, X. Zhang, Y. Kohno, M. Posner, G. T. Gallagher, M. Varvares, D. Cohen, D. Kim, R. Kent, R. B. Donoff, R. Todd, J. A. Warrington, and D. T. W. Wong, Oncogene 20, 6196 (2001). 9 C. Leethanakul, V. Patel, J. Gillespie, M. Pallente, J. F. Ensley, S. Koontongkaew, L. A. Liotta, M. Emmert-Buck, and J. S. Gutkind, Oncogene 19, 3220 (2000). 10 C. Leethanakul, V. Patel, J. Gillespie, E. Shillitoe, R. M. Kellman, J. F. Ensley, V. Limwongse, M. R. Emmert-Buck, D. B. Krizman, and J. S. Gutkind, Oral Oncol. 36, 474 (2000). 11 M. C. Chang, Y. T. Chang, M. S. Wu, C. T. Shun, Y. W. Tien, and J. T. Lin, J. Formos. Med. Assoc. 100, 352 (2001). 12 J. Cui, L. R. Rohr, G. Swanson, V. O. Speights, T. Maxwell, and A. R. Brothman, Prostate 46, 249 (2001). 13 N. L. Simone, A. T. Remaley, L. Charboneau, E. F. Petricoin III, J. W. Glickman, M. R. EmmertBuck, T. A. Fleisher, and L. A. Liotta, Am. J. Pathol. 156, 445 (2000). 14 D. K. Ornstein, C. Englert, J. W. Gillespie, C. P. Paweletz, W. M. Linehan, M. R. Emmert-Buck, and E. F. Petricoin III, Clin. Cancer Res. 6, 353 (2000). 15 D. K. Ornstein, J. W. Gillespie, C. P. Paweletz, P. H. Duray, J. Herring, C. D. Vocke, S. L. Topalian, D. G. Bostwick, W. M. Linehan, E. F. Petricoin III, and M. R. Emmert-Buck, Electrophoresis 21, 2235 (2000). 16 J. W. Gillespie, A. Nasir, and H. E. Kaiser, In Vivo 14, 139 (2000).
[29]
LCM-GENERATED RNA FOR MICROARRAY ANALYSIS
325
FIG. 1. Use of laser capture microdissection (LCM) to selectively harvest an epithelial tumor island. This is a case of well-differentiated squamous cell carcinoma from the oral cavity. (A) 5-µm sections from the snap frozen specimen are counterstained using H&E before LCM to illustrate detailed histomorphology of the oral cancer. (B) LCM was performed on the periphery of the epithelial tumor island using 30-µm diameter laser pulses. No coverslip is used in the LCM process, so the image in B differs from A because of enhanced refraction of the light passing through post-LCM treated tissue. (C) The tissue void created by the LCM-dissected tumor island lifted from the surrounding connective tissue. (D) The LCM-captured tumor island on a cap is transferred to an Eppendorf tube where it can be processed for RNA and protein applications as described [from H. Ohyama, X. Zhang, Y. Kohno, I. Alevizos, M. Posner, D. T. Wong, and R. Todd, BioTechniques 29, 530 (2000), with permission].
profiling using microarrays, we chose to use the Affymetrix high-density oligonucleotide arrays (GeneChip) primarily because of published reports of reproducibility, sensitivity, and specificity.17–21 17
D. J. Lockhart, H. Dong, M. C. Byrne, M. T. Follettie, M. V. Gallo, M. S. Chee, M. Mittmann, C. Wang, M. Kobayashi, H. Horton, and E. L. Brown, Nat. Biotechnol. 14, 1675 (1996). 18 L. Wodicka, H. Dong, M. Mittmann, M. H. Ho, and D. J. Lockhart, Nat. Biotechnol. 15, 1359 (1997). 19 A. de Saizieu, U. Certa, J. Warrington, C. Gray, W. Keck, and J. Mous, Nat. Biotechnol. 16, 45 (1998). 20 C. H. Redfern, M. Y. Degtyarev, A. T. Kwa, N. Salomonis, N. Cotte, T. Nanevicz, N. Fidelman, K. Desai, K. Vranizan, E. K. Lee, P. Coward, N. Shah, J. A. Warrington, G. I. Fishman, D. Bernstein, A. J. Baker, and B. R. Conklin, Proc. Natl. Acad. Sci. U.S.A. 97, 4826 (2000). 21 R. Todd and D. T. Wong, submitted (2001).
326
GENETIC APPLICATIONS
[29]
Regardless of the type of microarray used, the investigator is often required to provide labeled target cellular RNA in the range of 5–10 µg per microarray experiment. It is unrealistic to expect to obtain this amount of RNA from a tissue sample. Using the Arcturus PixCell II LCM, we and others routinely harvest ∼100 ng of total cellular RNA from ∼100,000 cells in ∼ 4 hr (∼40,000 LCM pulses at 30 µm diameter).7,8 Simply scaling up the LCM collection to obtain more cells is impractical, costly, and, because of limited amounts of tissue, often not an option. In the given example scaling up to obtain 5–10 µg of cellular RNA would require 50 times the effort of a trained operator and an experienced pathologist translating into ∼200 hr of LCM time to procure ∼5,000,000 cells. This is certainly an impractical approach. Aside from labor and time, tissue availability is a limiting factor, as well as RNA integrity that is compromised with additional handling and time. In order to harness the power of LCM and be able to monitor global gene expression using microarrays, methods have emerged that either amplify the total RNA prehybridization (targets)22 or amplify the signal posthybridization.23 Although this field is rapidly evolving with no apparent “ideal” approach yet, the linear amplification of cellular RNA using multiple rounds of the bacteriophage T7 RNA polymerase is a frequently published approach to amplify sufficient target RNA for various applications including microarray-based expression analysis.7,8 Two commercial vendors have released preoptimized kits for this particular application (RiboAmp RNA Amplification Kit, Arcturus, Mountain View, CA; MessageAmp aRNA Kit, Ambion, Inc., Austin, TX). We published the first papers using LCM-generated RNA from oral cancer tissues, prepared and labeled via multiple rounds of T7 linear amplification and hybridized to the Affymetrix 6.8k HuGeneFL probe array.7,8 Our results are not only encouraging but also strongly supportive of this experimental approach to unravel the critical molecular alterations during a pathological process (such as cancer) or during normal development. The LCM/microarray approach revealed molecular changes that theretofore had not been identified. These results provide gene expression profile information regarding individual genes and clusters of genes associated with pathways of significant clinical interest. Laser Capture Microdissection, RNA Isolation, and Linear Amplification of RNA Since our research focus is human oral cancer, this solid tumor will serve as the illustrative tissue. Under RNase-free conditions, surgically excised human oral cancer tissues are immediately snap-frozen and embedded in OCT compound and 5 µm cryosections are prepared. Normal and malignant oral keratinocytes are then 22 23
J. E. Kacharmina, P. B. Crino, and J. Eberwine, Methods Enzymol. 303, 3 (1999). K. K. Wong, R. S. Cheng, and S. C. Mok, BioTechniques 30, 670 (2001).
[29]
LCM-GENERATED RNA FOR MICROARRAY ANALYSIS
327
procured using the PixCell II LCM System (Acturus Engineering, Mountain View, CA). About 120,000 keratinocytes (∼28,000 LCM pulses using a 30-µm beam diameter) are laser captured from each of the tissue samples. LCM-isolated cells are lysed and homogenized by vortexing using a denaturation GITC-based buffer containing 2-mercaptoethanol. Total RNA is extracted using the RNeasy Kit (Qiagen, Valencia, CA). A DNase I step is incorporated to remove residual genomic DNA before washing steps while the RNA is still bound to the silica-gel membrane. Using RT-PCR we routinely check the quality of isolated RNA by measuring the integrity of five cellular maintenance gene transcripts including glyceraldehyde-3-phosphate dehydrogenase (GAPDH); α-tubulin; β-actin; ribosomal protein S9; and ubiquitin. The quantity of isolated RNA is assessed using the RiboGreen RNA Quantitation Reagent and kit (Molecular Probes, Eugene, OR) using spectrofluorometry (Bio-Rad, Hercules, CA). Only those samples exhibiting good quality PCR products for all five cellular maintenance genes are used for subsequent analysis. Target Sample Preparation Double-stranded cDNA is synthesized from the LCM-derived isolated RNA using the Superscript Choice System (Life Technologies, Rockville, MD). Five µl (75–100 ng) of total RNA isolated from ∼120,000 human oral cancer cells is mixed with 1 µl of 20 µM T7-oligo(dT)24 primer [5 -GGCCAGTGAATTGTAATACGA CTCACTATAGGGAGGCGG-(dT)24-3 ] in a total volume of 11 µl to initiate firststrand synthesis. The primer and RNA are heat denatured at 70◦ for 10 min, followed by annealing at 42◦ for 2 min. Four µl of 5× first-strand reaction buffer, 2 µl of 0.1 M DTT, 1 µl of 10 mM dNTPs, 1 µl of RNase inhibitor (40 units/µl) (Promega, Madison, WI), and 1 µl of Superscript II (200 units/µl) are added and incubated for 1 hr at 42◦ for the first-strand cDNA synthesis. For secondstrand cDNA synthesis, 30 µl of 5× second-strand synthesis buffer, 3 µl of 10 mM dNTPs, 4 µl of DNA polymerase I (10 units/µl), 1 µl of Escherichia coli RNase H (2 units/µl), 1 µl of E. coli DNA ligase (1 unit/µl), and 91 µl of RNase-free water are added and the mixture is incubated at 16◦ for 2 hr. Two µl of T4 DNA polymerase (5 units/µl) is then added and incubated for an additional 5 min at 16◦ . Ten µl of 0.5 M EDTA is then added to stop the reaction. The resultant cDNA is extracted with phenol–chloroform and washed three times with 500 µl of RNase-free water in a Microcon-100 spin column (Millipore, Bedford, MA), each spun at 2500 rpm for 12 min. The cDNA is then collected and adjusted to a volume of 15 µl for linear amplification by T7 RNA polymerase. The cDNA synthesis and T7 RNA polymerase linear amplification are modified according to the protocols of Luo et al. and Kacharmina.22,24 The first two rounds 24
L. Luo, R. C. Salunga, H. Guo, A. Bittner, K. C. Joy, J. E. Galindo, H. Xiao, K. E. Rogers, J. S. Wan, M. R. Jackson, and M. G. Erlander, Nat. Med. 5, 117 (1999).
328
GENETIC APPLICATIONS
[29]
of linear amplification are carried out using the Ampliscribe T7 Transcription Kit (Epicentre Technologies, Madison, WI). Twelve µl of double-strand cDNA, 2 µl of 10× Ampliscribe T7 buffer, 0.5 µl each of 40 mM ATP, CTP, GTP, and UTP, and 2 µl of T7 RNA polymerase are incubated at 37◦ for 14 hr. The cRNA is then extracted and washed three times in a Microcon-100 column (each at 2500 rpm for 12 min), collected, and adjusted to a volume of 10 µl. Using the resultant cRNA as a template, double-stranded cDNA synthesis is performed. Two µl of random hexamers (50 ng/µl) is added to the cRNA and the mix is incubated at 70◦ for 10 min, chilled on ice, and further incubated at room temperature for 10 min. Four µl of 5× first-strand reaction buffer, 2 µl of 0.1 M DTT, 1 µl of 10 mM dNTPs, 1 µl of RNase inhibitor (40 units/µl), and 1 µl of Superscript II (200 units/µl) are added and incubated for 5 min at room temperature followed by another incubation for 1 hr at 37◦ for the first-strand cDNA synthesis. One µl of E. coli RNase H (2 units/µl) is added and incubated at 37◦ for 20 min and then heated at 95◦ for 2 min. A custom oligo T7(dT)24 primer [5 GGCCAGTGAATTGTAATACGACTCACTATAGGGAGGCGG-(dT)24-3 ] is added to initiate the second-strand cDNA synthesis by incubations at 70◦ for 5 min and 42◦ for 10 min. Thirty µl of 5× second-strand synthesis buffer, 3 µl of 10 mM dNTPs, 4 µl of DNA polymerase I (10 units/µl), 1 µl of E. coli RNase H (2 units/µl), and 90 µl of RNase-free water are added and the mixture is incubated at 16◦ for 2 hr. Two µl of T4 DNA polymerase (5 units/µl) is then added and incubated for an additional 10 min at 16◦ . The resultant cDNA is extracted as previously described. The double-stranded cDNA, 2 µl of 10× Ampliscribe T7 buffer, 1.5 µl each of 100 mM ATP, CTP, GTP, and UTP, 2 µl of 0.1 M DTT, and 2 µl of T7 RNA polymerase are incubated together in a volume of 20 µl at 37◦ for 14 hr. The resultant cRNA from the second T7 amplification is extracted and then converted to cDNA as previously described. The resultant double-stranded cDNA prepared by two rounds of T7 linear amplification is now ready for labeling by biotinylation. The third in vitro transcription (IVT) reaction is performed to produce biotinlabeled cRNA from the double-stranded cDNA. The BioArray High Yield RNA Transcript Labeling System (Enzo, Farmingdale, NY) is a preoptimized kit specifically designed for this application. The biotinylated cRNA (IVT product) is purified using the RNeasy kit (Qiagen, Valencia, CA). The quantity and purity of the biotinylated cRNA are determined by spectrophotometry and an aliquot of sample is further checked by gel electrophoresis. Hybridization of Biotinylated cRNA to Test-1 and HuGeneFL GeneChip Probe Arrays The cRNA is fragmented as described by Wodicka et al.18 All array washing, staining, and scanning are carried out as described in the Gene Expression
[29]
LCM-GENERATED RNA FOR MICROARRAY ANALYSIS
329
Manual (Affymetrix, Inc., 1999). The probe sets consist of oligonucleotides 25 bases in length. Probes are complementary to the published sequences (GenBank) as previously described.17 The sensitivity and reproducibility of the GeneChip probe arrays is such that RNAs present at a relative concentration of 1 : 100,000 are unambiguously detected, and detection is quantitative over more than three orders of magnitude.20,25 Array controls and performance with respect to specificity and sensitivity are the same as those previously described.17,18,26 Information regarding the genes represented on the arrays used in this study can be found at www.netaffx.com. Summary and Discussion Our collective experience tells us that the successful generation of a target sample from biopsy tissue depends on both the ability to isolate sufficient quantities of intact mRNA and the ability to amplify mRNA without distorting gene expression levels. Using human oral cancer tissues, about 70% of the cases collected (normal and cancer) contained RNA of sufficient quality. Failure to capture sufficient quantities of RNA from these specimens likely reflects a prolonged ischemic time between harvesting and freezing the material. Ischemic times greater than 30 sec allow considerable degradation of RNA and can distort results. Even if sufficient tissue exists, RNA degradation can reduce enough of the transcripts to impair quantitative results.27 Therefore, good tissue harvesting methods remain the basis for successful target sequence generation. In addition to minimizing tissue ischemic time, we found careful processing to be critical for RNA isolation. The advantages of the procedures reported here include elimination of phenol– chloroform extraction and alcohol precipitation steps, and performing DNase I treatment while the RNA is still bound to the silica-gel membrane. LCM is used to microdissect ∼120,000 cells, yielding ∼450 ng of total RNA. Reverse transcription is performed on 75–100 ng of isolated RNA, followed by two rounds of T7 RNA polymerase amplification, and produced ∼5 µg of doublestranded cDNA. Evaluation of the cDNA reveals that five out of five cellular maintenance transcripts are present. We started with ∼75 ng of LCM-generated RNA and assuming 2% are mRNA (1.5 ng), the yield of 5 µg of double-stranded cDNA after two rounds of linear amplification represents a ∼3000-fold amplification. Since theoretically a 106-fold amplification is achievable, our results represent 3.3% efficiency. The key consideration is the need to minimize the number of T7 linear amplifications in order to generate ∼1 µg of double-stranded cDNA for 25
J. A. Warrington, S. Dee, and M. Trulson, in “Microarray Biochip Technology” (M. Schena, ed.), p. 119. Eaton Publishing, MA, 2000. 26 M. Mahadevappa and J. A. Warrington, Nat. Biotechnol. 17, 1134 (1999). 27 N. Simone, R. Bonner, J. Gillespie, M. Emmert-Buck, and L. Liotta, Trends Genet. 14, 253 (1998).
330
GENETIC APPLICATIONS
[29]
probe generation by biotinylation. Using the Enzo BioArray High Yield RNA Transcript Labeling kit ∼50 µg of biotinylated cRNA was produced using 1 µg of input cDNA. The biotinylated cRNA was purified using the RNeasy kit and then quantified by spectrophotometry. From 1 µg of double-stranded cDNA ∼50 µg of biotinylated cRNA was generated, representing an efficiency of about ∼50%. We have determined the fidelity of the amplification procedure (two rounds of T7 RNA amplification) by comparing the hybridization profiles on the HuGeneFL GeneChip Array (∼7000 genes) of unamplified and amplified total RNA isolated from human endometrial adenocarcinoma cells (AN3 CA). The number of transcripts detected in unamplified RNA is 38% vs 30% of the three rounds of T7 amplification. Scatter plot analysis revealed a strong linear relationship (r 2 = 0.928). Our data are similar to those published by Wang et al.28 The biotinylated cRNA from the 10 samples (normal and cancer) was used to hybridize Test-1 probe arrays to determine cRNA quality and integrity. The arrays contain probes representing a handful of maintenance genes and a number of controls. Analysis of the arrays confirmed the RT-PCR findings above. cRNA linearly amplified from human oral cancer tissue produced no nonspecific or unusual hybridization patterns and the transcripts for the maintenance genes were detected. The 5 region of the RNA was degraded but enough 3 transcript was intact to proceed because probes on the microarrays are biased to the 3 region of the genes represented. The sample was subsequently hybridized to HuGeneFL probe arrays containing probes representing ∼7000 full-length genes, and 26.5–33.0% of the genes represented on the arrays were detected as expressed in the 10 samples examined. In general, about 80% of the transcripts detected were in low abundance, fewer than or equal to 5 copies per cell; ∼9% were in low-moderate abundance, 5 to 10 copies per cell; ∼8% were in moderate abundance, 10 to 50 copies per cell; ∼2% were in moderate–high abundance, 50 to 100 copies per cell; and 0% were detected as high-abundance transcripts, more than 100 copies per cell. Given the LCM handling of the tissue and the amount of sample hybridized, 10 µg, detection of ∼30% of the 7000 transcripts is consistent with previous findings. In other experiments, using nonLCM methods and hybridizing 20 µg of cRNA the average percent of transcripts detected is 2300 or ∼43% of the 7000 transcripts.26 In titration experiments comparing hybridization amounts, it was determined that reducing the hybridization amount to 10 µg reduced the detectable transcript number by 8%. We have hybridized the biotinylated cRNA from the 5-paired cases of oral cancer onto the new U95A GeneChip microarrays consisting of probes for 12,000 full-length human cDNAs. The data revealed that there is a great degree of consistency between the HuGeneFL probe array results and the U95A arrays. In addition, a comparable
28
E. Wang, L. D. Miller, G. A. Ohnmacht, E. T. Liu, and F. M. Marincola, Nat. Biotechnol. 18, 457 (2000).
[29]
LCM-GENERATED RNA FOR MICROARRAY ANALYSIS
331
number of transcripts are detected (36.1 to 46.1% of the 12,000 transcripts are detected). We have selectively validated the microarray data. As a start, three metastatic pathway genes whose expressions are consistently altered in the five paired cases of oral cancer were selected. Real-time quantitative PCR (RT-QPCR) in conjunction with the TaqMan specific probe system or SYBR Green system was used to validate the expression levels of interstitial collagenase (a member of the MMPs involved in metastasis), urokinase plasminogen activator (UPA, associated with metastasis), and cathepsin L (a member of the serine proteases).
B 1
2
3
4
5
M N T N T N T N T N T
bp 200 100
Collagenase (140-bp)
FIG. 2. Comparison and validation of microarray data by RT-QPCR. (A) Comparison of gene expression data (from GeneChip) and by RT-QPCR for collagenase. (B) Visualization of actual RT-QPCR products by agarose gel electrophoresis [from I. Alevizos et al., Oncogene 20, 6196 (2001) with permission].
332
GENETIC APPLICATIONS
[29]
Actual Microarray or GeneChip Data from 5-Paired Cases of Oral Cancers
β-Actin (Control)
EGFR (
)
MMP1 (
Nmu(
5/5)
5/5)
UPA ( 5/5)
ALD9 ( 5/5)
Cathepsin L( 5/5)
HER3 ( 5/5)
FIG. 3. Screen shots of expression level analysis of selected cellular genes in five paired cases of human oral cancer (GeneSpring, Silicon Genetics System, Redwood City, CA). Each panel contains five subpanels, each showing the relative expression level of a cellular gene in each of the five oral cancers. Normal level is on the left while the tumor expression level is on the right.
Comparison of the microarray and RT-QPCR data revealed that they approximate each other. The actual comparative data for collagenase are graphically shown in Fig. 2A and Fig. 2B. Similar data were obtained for UPA and cathepsin L. We have further validated a number of other high and low abundant transcripts including Neuromedin U, GST, cytochrome P450, ALDH-9, ALDH-10, and Wilm’s tumor-related protein (data not shown).26 A final note is that we have found this LCM/microarray approach to be informative and revealing of consistent molecular alterations in oral cancer development that otherwise would not have been identified. Using bioinformatics tools, we analyzed the data using multiple methods including self-organizing maps, principal component analysis, and cluster analysis. Figure 3 is a composite illustration of screen shots showing the relative expression levels of eight genes in each of the five pairs of oral cancer. The β-actin panel demonstrates that the level of this house-maintenance gene is relatively similar between the normal and tumor keratinocytes as well as among the five paired cases. Three of the known metastatic genes, MMP1, UPA, and cathepsin L, are consistently upregulated in the tumors. Our data also revealed differential expression of cellular
[29]
LCM-GENERATED RNA FOR MICROARRAY ANALYSIS
333
genes that are not yet functionally characterized or genes that have not been studied by classic methods in head and neck/oral carcinogenesis (Nmu, ALD9, and HER3). One such example is neuromedin U (Nmu), which is down-regulated in 5/5 tumors.29 Nmu is a poorly understood protein that manifests potent contractile activities on smooth muscle cells. Two G-protein coupled receptors (NMU1 and NMU2) have been identified as interacting with Nmu with nanomolar potency.30,31 Our data provide strong evidence that Nmu is relevant in the development of oral malignancy and suggest the need for further study of the role of Nmu (downregulated expression in tumor) in carcinogenesis. On the other hand, EGFR, one of the most frequently reported alterations in epithelial cancers including oral cancer, did not show a consistent pattern of expression in the five paired cases we examined, although this could be due to the samples selected for our analysis which may not be reflective of the majority of human oral cancers. Alternatively, the discrepancy could also be indicative that the reported overexpression of EGFR may not be due to the tumor keratinocytes but perhaps to the tumor stroma and/or inflammatory infiltrate. A final advantage of the LCM/microarray approach to solid tumor analysis by global gene expression is that the resultant databases serve as permanent expression registries. These databases can be used to examine the expression profiles of any gene or cluster of genes pertaining to any experimental or epidemiological question, all without the need to reperform the experiment, the data are accessible at any time via a personal computer. Acknowledgments The authors acknowledge the input of Drs. Xue Zhang, Ilias Alevizos, and Yohko Kohno on the initial phase of the experiments. The work is supported by the National Institute of Dental and Craniofacial Research (NIDCR) Grants P01 DE12467 (D.T.W.W.), PO1 DE2467-S1 (D.T.W.W.), and P30 DE11814 (D.T.W.W.).
29
P. G. Szekeres, A. I. Muir, L. D. Spinage, J. E. Miller, S. I. Butler, A. Smith, G. I. Rennie, P. R. Murdock, L. R. Fitzgerald, H. Wu, L. J. McMillan, S. Guerrera, L. Vawter, N. A. Elshourbagy, J. L. Mooney, D. J. Bergsma, S. Wilson, and J. K. Chambers, J. Biol. Chem. 275, 20247 (2000). 30 R. Fujii, M. Hosoya, S. Fukusumi, Y. Kawamata, Y. Habata, S. Hinuma, H. Onda, O. Nishimura, and M. Fujino, J. Biol. Chem. 275, 21068 (2000). 31 R. Raddatz, A. E. Wilson, R. Artymyshyn, J. A. Bonini, B. Borowsky, L. W. Boteju, S. Zhou, E. V. Kouranova, R. Nagorny, M. S. Guevarra, M. Dai, G. S. Lerman, P. J. Vaysse, T. A. Branchek, C. Gerald, C. Forray, and N. Adham, J. Biol. Chem. 275, 32452 (2000).
334
[30]
GENETIC APPLICATIONS
[30] Single Cell Gene Mutation Analysis Using Laser-Assisted Microdissection of Tissue Sections ˚ SA PERSSON, HELENA BACKVALL ¨ By A , FREDRIK PONTE´ N, MATHIAS UHLE´ N, and JOAKIM LUNDEBERG Introduction The ability to analyze very small amounts of material, as well as being able to select exactly defined cell populations, or even single cells, in heterogeneous material, has been a long-sought goal for researchers in fields such as genetic archeology, forensics, prenatal diagnostics, and tumor biology. In the past several years developments in the fields of microdissection and amplification have made this a reality. Today there are several commercial laser-assisted systems suitable for cell population and single cell dissection. The two main principles for laser-assisted microdissection are laser microbeam microdissection/laser pressure catapulting (LMM/LPC) developed by Sch¨utze1,2 and laser capture microdissection (LCM) developed by Liotta and co-workers.3 In LMM, a 337-nm pulsed nitrogen laser is used to ablate undesired biological material surrounding the cell or cells of interest. The high-quality laser beam allows for fine focused microdissection (focus point <2 µm) without damage to surrounding cells. The isolated cell (or cells) can be retrieved using a needle or glass capillary or through laser catapulting into the cap of a microfuge tube (LPC). Tissue sections mounted on membranes or ordinary glass slides can be used with this technique. LCM uses a heat-generating infrared laser to fuse a thin temperature-sensitive transparent film with selected parts of the tissue on an underlying histological section. The film is mounted on a rigid flat cap and on removal the selected parts of the tissue are detached. The techniques employing laser-assisted microdissection have provided a new field of possibilities for analysis of tissue components. Microscopically defined cell populations, including single cells, can be retrieved from heterogeneous samples, e.g., complex tumor and normal tissues.4 Our focus has been on skin and skin cancer. Chronically sun-exposed skin harbors scattered p53 immunoreactive cells of unclear significance. The aim of a recent study was to investigate whether such 1
K. Sch¨utze and A. Clement-Sengewald, Nature 368, 667 (1994). K. Sch¨utze and G. Lahr, Nat. Biotechnol. 16, 737 (1998). 3 M. R. Emmert-Buck, R. F. Bonner, P. D. Smith, R. F. Chuaqui, Z. Zhuang, S. R. Goldstein, R. A. Weiss, and L. A. Liotta, Science 274, 998 (1996). 4 F. Pont´ en, C. Williams, G. Ling, A. Ahmadian, M. Nist´er, J. Lundeberg, J. Pont´en, and M. Uhl´en, Mut. Res. 382, 45 (1997). 2
METHODS IN ENZYMOLOGY, VOL. 356
Copyright 2002, Elsevier Science (USA). All rights reserved. 0076-6879/02 $35.00
[30]
SINGLE CELL GENE MUTATION ANALYSIS
335
cells differ from their neighboring cells with respect to mutations in the p53 gene.5 The p53 tumor suppressor gene is implicated in cancer and is mutated in about 50% of all human cancers.6–9 A major risk factor for development of skin cancer is ultraviolet (UV) radiation.10–12 UVB (290–320 nm) is anticipated as the major carcinogen in UV mediated tumorigenesis. UVB induces DNA photoproducts that underlie the typical UV-signature mutations,13 which are often found in the p53 gene from skin tumors. Although to a lesser extent, UVA can also induce scattered p53 immunoreactive cells. The genetic background of such cells is unknown. We have used the PALM laser microscope system (P.A.L.M. GmbH, Bernried, Germany) to perform LMM on tissue sections from frozen biopsies. Based on p53 immunoreactivity, single keratinocytes were isolated and collected using a micromanipulator. Single cells were subsequently subjected to PCR and direct DNA sequencing to determine the p53 gene status (Fig. 1). Using the protocols described below we have successfully investigated the p53 gene in both scattered and clustered p53 immunoreactive single keratinocytes from normal sun-exposed and experimentally UVA exposed skin. Comments on Protocol Sample Material Despite the superior morphology of formalin-fixed tissue, frozen tissue is preferred for genetic analysis. Formalin fixation degrades genomic DNA and may also create artifactual mutations.14,15 The emergence of artifactual mutations is pronounced in sample sizes containing fewer than 300 cells. When the amount of starting material was decreased from 150 to 80 cells, the artifactual mutation rate increased from 1/4000 to 1/1000 bases. When only 20 cells were used as 5
G. Ling, A. Persson, B. Berne, M. Uhl´en, J. Lundeberg, and F. Ponten, Am. J. Pathol. 159, 1247 (2001). 6 P. Hainaut, T. Hernandez, A. Robinson, P. Rodriguez-Tome, T. Flores, M. Hollstein, C. C. Harris, and R. Montesano, Nucleic Acids Res. 26, 205 (1998). 7 M. S. Greenblatt, W. P. Bennett, M. Hollstein, and C. C. Harris, Cancer Res. 54, 4855 (1994). 8 D. P. Lane, Nature 358, 15 (1992). 9 A. J. Levine, J. Momand, and C. A. Finlay, Nature 351, 453 (1991). 10 S. Tornaletti and G. P. Pfeifer, Bioessays 18, 221 (1996). 11 M. R. Gailani, D. J. Leffell, A. Ziegler, E. G. Gross, D. E. Brash, and A. E. Bale, J. Natl. Cancer Inst. 88, 349 (1996). 12 S. Tornaletti and G. P. Pfeifer, Science 263, 1436 (1994). 13 D. E. Brash, J. A. Rudolph, J. A. Simon, A. Lin, G. J. McKenna, H. P. Baden, A. J. Halperin, and J. Pont´en, Proc. Natl. Acad. Sci. U.S.A. 88, 10124 (1991). 14 N. Yagi, K. Satonaka, M. Horio, H. Shimogaki, Y. Tokuda, and S. Maeda, Biotech. Histochem. 71, 123 (1996). 15 S. Paabo, D. M. Irwin, and A. C. Wilson, J. Biol. Chem. 265, 4718 (1990).
336
GENETIC APPLICATIONS
[30]
Laser microdissection on tissue section
Multiplex outer amplification Chromosomal DNA
Region specific inner PCR
Mt 4 5 6 7 8 9 10 11
Sequence analysis
FIG. 1. Schematic illustration of the different steps involved in genetic analysis of single cells from tissue sections.
[30]
SINGLE CELL GENE MUTATION ANALYSIS
337
template, the mutation rate was 1/500 bases.16 We have previously demonstrated that the presence of EDTA during sample preparation inhibits nuclease activity and increases the rate of successful amplification.17 PCR PCR has been performed in a multiplex/nested fashion to allow for amplification of exon 4-11 of the p53 gene as well as a mitochondrial control sequence. Mitochondrial DNA exists in approximately 1000 copies in each cell compared to two copies of cellular genes. In the case of amplification failure, mitochondrial DNA was used as a specific marker to distinguish between loss of the cell (no amplified fragments) and degradation of the template (amplification only of mitochondrial DNA). The first four cycles of the outer PCR have extended annealing and extension times to increase primer annealing, a strategy earlier shown to affect the success rate of single cell amplification.17 We have used Pfu Turbo DNA polymerase to minimize the risk of polymerase-mediated artifactual mutations, which is increased using low copy number samples.15,18 Pfu Turbo DNA polymerase has proofreading activity and an error rate of 1.6 × 10−6,19,20 which can be compared to the error frequency of Taq DNA polymerase, which yields approximately one misincorporation in 104–105 bases.21,22 Potential problems with single cell PCR include risk of contamination and random amplification failure of one allele (allele dropout, ADO). Preparation of PCR reagent mixture, addition of PCR reagent mixture to the single cell sample, and addition of the outer PCR product to the reagent mix for the inner PCR should be performed at separate locations to avoid contamination. Coats and gloves should be used at all times. All pipetting should be performed with filter tips and the bench area where the template is added should be decontaminated using UV light. Negative controls without DNA should be included in abundance together with at least one positive control containing a sufficient amount of genomic DNA. The random frequency of allele dropout (ADO) must be considered when interpreting data. The ADO rate using the outlined procedure with cells retrieved from frozen sections is approximately 50%, but a lower rate is achieved using 16
C. Williams, F. Pont´en, C. Moberg, P. S¨oderkvist, M. Uhl´en, J. Pont´en, G. Sitbon, and J. Lundeberg, Am. J. Pathol. 155, 1467 (1999). 17 A. E. Persson, G. Ling, C. Williams, H. Backvall, J. Pont´ en, F. Pont´en, and J. Lundeberg, Anal. Biochem. 287, 25 (2000). 18 J. Odeberg, A. Ahmadian, C. Williams, M. Uhl´ en, F. Pont´en, and J. Lundeberg, Gene 235, 103 (1999). 19 P. Andre, A. Kim, K. Khrapko, and W. G. Thilly, Genome Res. 7, 843 (1997). 20 K. S. Lundberg, D. D. Shoemaker, M. W. Adams, J. M. Short, J. A. Sorge, and E. J. Mathur, Gene 108, 1 (1991). 21 K. A. Eckert and T. A. Kunkel, PCR Methods Appl. 1, 17 (1991). 22 K. R. Tindall and T. A. Kunkel, Biochemistry 27, 6008 (1988).
338
GENETIC APPLICATIONS
[30]
cultured cells.17 The reported ADO rates, usually based on analysis of cultured cells, are between 10% and 80%23–27 and are known to be affected by fragment length and number of fragments amplified in parallel. Sequencing Confirmation of a detected mutation can only be done by performing a new PCR from the same starting material. Evidently this cannot be accomplished using single cells. However, a partial confirmation is achieved by performing a new inner PCR on the outer PCR product followed by resequencing of the sample. Protocol Sample Preparation Sectioning. Cut 16-µm-thick cryosections from the frozen tissue biopsy of interest and place on ordinary glass slides. Cover the sections immediately with 10–20 µl EDTA (10 mM) to inhibit nuclease activity. If sections are not to be used immediately they can be stored at −70◦ . Staining. The tissue sections can be stained using histochemical or immunohistochemical methods, depending on what criteria are used for selection of cells. Immunohistochemical staining of p53. In this method DO-7, recognizing both mutated and wild-type p53, is used as primary antibody. The p53 protein is visualized by avidin–biotin-coupled immunohistoperoxidase staining using DAB as chromogen. Note: All solutions used during the staining, including tap water, should contain 10 mM EDTA to inhibit nuclease activity. 1. Let the sections air dry for 30–60 min at room temperature. 2. Equilibrate slides in phosphate-buffered saline (PBS) for approximately 10 min at room temperature. 3. Blocking of endogenous peroxidase. Incubate with blocking solution (PBS 50 ml + 500 µl H2O2) for 30 min using a vibrax. 23
P. F. Ray and A. H. Handyside, Mol. Hum. Reprod. 2, 213 (1996). I. Findlay, P. Matthews, and P. Quirke, Prenat. Diagn. 18, 1413 (1998). 25 A. M. Garvin, W. Holzgreve, and S. Hahn, Nucleic Acids Res. 26, 3468 (1998). 26 S. Rechitsky, C. Strom, O. Verlinsky, T. Amet, V. Ivakhnenko, V. Kukharenko, A. Kuliev, and Y. Verlinsky, J. Assist. Reprod. Genet. 15, 253 (1998). 27 A. R. Thornhill, J. A. McGrath, R. A. Eady, P. R. Braude, and A. H. Handyside, Prenat. Diagn. 21, 490 (2001). 24
[30]
SINGLE CELL GENE MUTATION ANALYSIS
339
4. Prepare 1% bovine serum albumin (BSA) solution by diluting 10% BSA in PBS. 5. Rinse slides in PBS 3× for 5 min using a vibrax. 6. Preincubation with 1% BSA (prevents nonspecific protein binding). Dry area around section. Add 50–100 µl (depending on size of section) of 1% BSA on section. Incubate slides in a dark chamber for 30 min at room temperature. 7. Preparation of the primary and secondary antibodies and avidin/biotin complex. All the solutions are made in 1% BSA. Primary ab: human p53 (DO-7, DAKO Ltd) 1 : 200 Secondary ab: biotinylated rabbit anti-mouse antibody, 1 : 200 Developing ab: avidin/biotin complex, 1 : 200 8. Pour off the BSA. 9. Incubation with primary antibody. Add 50–100 µl primary ab (DO-7). Incubate slides in a dark chamber for 30 min at room temperature. 10. Rinse slides in PBS 3× for 5 min using a vibrax. 11. Incubation with secondary antibody. Add 50–100 µl secondary ab. Incubate slides in a dark chamber for 30 min at room temperature. 12. Rinse slides in PBS 3× for 5 min using a vibrax. 13. Incubation with avidin/biotin complex. Add 50–100 µl AB complex. Incubate slides in a dark chamber for 30 min at room temperature. 14. Rinse slides in PBS 3× for 5 min using a vibrax. 15. Developing. The immunoreaction is visualized by the AB-complex using hydrogen peroxide as substrate and diaminobenzidine (DAB) as chromogen. Incubate for 7 min in color solution (50 ml PBS + 1000 µl DAB + 10 µl H2O2) using a vibrax. 16. Rinse in tap water for 2–3 min. 17. Counterstaining. Counterstain with Mayer’s hematoxylin for 20–30 sec. 18. Rinse in tap water for 5–10 min. 19. Let the sections dry at room temperature. The immunohistochemically stained slides should be kept at −70◦ prior to microdissection. Laser-Assisted Microdissection of Single Cells 1. Prepare small (0.5 ml) Eppendorf tubes containing 10 µl of 1× PCR buffer (PE II).
340
[30]
GENETIC APPLICATIONS
2. Dissect a single cell by ablation of the neighboring cells using the PALM laser microscope system (P.A.L.M. GmbH, Bernried, Germany) or by using the system of one’s choice. 3. Mount a small glass capillary (Femtotips, Eppendorf ) to the manipulator. Break off the tip by lowering the manipulator toward the glass slide, so that the diameter becomes large enough to fit a single cell. 4. Immerse the tip in a drop of PCR buffer adjacent to the tissue and let the capillary forces fill the tip with liquid. 5. Use the manipulator with the mounted Femtotip to wet and pick up the cell. 6. Break off the remaining tip in one of the prepared PCR tubes after confirming under the microscope that the cell is attached to the tip. 7. Add 50 µl of mineral oil. Samples should be kept at −70◦ and amplified as quickly as possible after dissection. PCR Amplification of p53 Using Multiplex/Nested Approach Samples are amplified in a multiplex/nested fashion making it possible to analyze eight exons from the p53 gene and a mitochondrial control sequence from each single cell. The mitochondrial sequence is specific for each individual and as such serves as a control. Outer Multiplex Amplification 1. For the outer multiplex amplification prepare a reagent mix (on ice) as described below. Always prepare mix sufficient for a couple of additional samples, since a small amount is lost during multiple pipetting. Do not forget to include a positive control and some negative controls without DNA (at least three for every 10 samples is appropriate) in each run. Ingredient
Volume(µl)/sample
10× Pfu buffer dNTP (2 mM) Outer primer mix (18 primers) (5 pmol/µl)∗ Pfu Turbo Polymerase 2.5 U/µl (Stratagene) Millipore water
2 2 1 0.7 4.3
∗
The primer mix consists of 5 pmol/µl (in water) of each of the 18 outer primers listed in Table I.
2. Add 10 µl of the mix to each single cell sample, just letting the tip touch the liquid surface beneath the oil. Keep samples on ice.
[30]
341
SINGLE CELL GENE MUTATION ANALYSIS TABLE I PRIMERS USED FOR ANALYSIS OF P53 EXON 4-11 AND MITOCHONDRIAL SEQUENCE Primer sequence
Target Exon 4
Exon 5
Exon 6
Exon 7
Exon 8
Exon 9
Exon 10
Exon 11
Mt
Amplification
Forward
Reverse
Outer
5 -CTGGGACCTGGAGGGCTGGG
5 -AGAGGAATCCCAAAGTTCCA
Inner
5 -CTGAGGACCTGGTCCTCTGAC
5 -ATACGGCCAGGCATTGAAGT
Outer
5 -TGCTGCCGTGTTCCAGTTGC
5 -CAATCAGTGAGGAATCAGAGG
Inner
5 -TTCACTTGTGCCCTGACTT
5 -ACCAGCCCTGTCGTCTCTCC
Outer
5 -GGCTGGAGAGACGACAGGGC
5 -CGGAGGGCCACTGACAACCA
Inner
5 -TTGCCCAGGGTCCCCAGGCC
5 -CTTAACCCCTCCTCCCAGAG
Outer
5 -CCTCCCCTGCTTGCCACAGG
5 -GGAAGAAATCGGTAAGAGGTGG
Inner
5 -CGCACTGGCCTCATCTTGGG
5 -CAGCAGGCCAGTGTGCAGGG
Outer
5 -ACAGGTAGGACCTGATTTCC
5 -TGAATCTGAGGCATAACTGC
Inner
5 -GCCTCTTGCTTCTCTTTTCC
5 -CCCTTGGTCTCCTCCACCGC
Outer
5 -AGCAAGCAGGACAAGAAGCG
5 -GTTAGCTACAACCAGGAGCC
Inner
5 -GCCTCAGATTCACTTTTATCACC 5 -CTGGAAACTTTCCACTTGAT
Outer
5 -GATCCGTCATAAAGTCAAAC
Inner
5 -CTTGAACCATCTTTTAACTCAGG 5 -AATCCTATGGCTTTCCAACCTAGG
Outer
5 -CTTCAAAGCATTGGTCAGGG
5 -GGGTTCAAAGACCCAAAACC
Inner
5 -CACAGACCCTCTCACTCATG
5 -GCAGGGGAGGGAGAGATGGG
Outer
5 -CCTGAAGTAGGAACCAGATG
5 -ACACCAGTCTTGTAAACCGG
Inner
5 -CTCCACCATTAGCACCCAAAG
5 -TGATTTCACGGAGGATGGTGG
5 -TTGACCATGAAGGCAGGATG
3. Initiate the PCR by denaturation at 98◦ for 2 min. Then amplify the samples in two steps. First run 4 cycles of denaturation at 98◦ for 15 sec; annealing at 55◦ for 4 min; and extension at 72◦ for 30 min. This is followed by 26 cycles of denaturation at 98◦ for 15 sec; annealing at 55◦ for 30 sec, and extension at 72◦ for 1 min. End the program with an extension step at 72◦ for 10 min and a hold step at 4◦ . Inner Nested Amplification. Each sample from the outer PCR will yield nine samples in the inner amplification, corresponding to each of the eight exons and the mitochondrial sequence.
342
[30]
GENETIC APPLICATIONS
1. Prepare reagent mixes (on ice), one for each fragment, as described below. EXON 4, 6-11 AND Mt Ingredient
Volume(µl)/sample
10× PCR buffer II dNTP (2 mM) MgCl2 (25 mM) ∗ Inner primer (forward) (10 pmol/µl) ∗ Inner primer (reverse) (10 pmol/µl) AmpliTaq DNA pol. 5 U/µl (Stratagene) Millipore water
5 5 4 1 1 0.2 33.3
∗
See primer sequences in Table I. EXON 5 Ingredient
Volume(µl)/sample
10× Pfu buffer dNTP (2 mM) MgCl2 (25 mM) ∗ Inner primer (forward) (10 pmol/µl) ∗ Inner primer (reverse) (10 pmol/µl) Pfu Turbo Polymerase 2.5 U/µl (Stratagene) Millipore water
5 5 5 1 1 0.7 31.8
∗
See primer sequences in Table I.
2. For each sample dispense 49.5 µl of the reagent mix in an Eppendorf tube (0.5 ml) and cover with 30–50 µl mineral oil. 3. Add 0.5 µl of the outer reaction product as template. Make sure that the oil layer is penetrated when dispensing. 4. Initiate the PCR by denaturation at 94◦ for 5 min (Taq polymerase) or 98◦ for 2 min (Pfu polymerase). Amplify the samples for 30 cycles by denaturation at 94◦ for 30 sec or 98◦ for 15 sec, annealing at 63◦ for 30 sec, and extension at 72◦ for 1 min. The program is ended with an extension step at 72◦ for 10 min and a hold step at 4◦ . 5. The resulting PCR products are checked on a 1% agarose gel. p53 Sequence Analysis The DNA sequence is determined by direct sequencing, using the ABI 377 DNA sequencer (PerkinElmer Applied Biosystems, Inc., Foster City, CA). All sequences are determined in both directions and possible mutations are resequenced using the product of a new inner PCR reaction.
[31]
METHYLATION IN GENE PROMOTERS
343
Cycle sequencing. The sequencing reactions are carried out using the BigDye Terminator Cycle Sequencing kit (PerkinElmer Applied Biosystems), which contains the four dideoxy bases labeled with four different fluorophores. The sequencing products are obtained by linear amplification using the inner PCR primers as sequencing primers. 1. Mix in a microfuge tube: 2 µl Big Dye Terminator mix 6 µl 1× CS 8–10 µl Millipore water 1 µl primer (5 pmol/µl) 1–3 µl inner PCR product 2. Run in thermocycler for 30 cycles at 96◦ for 10 sec, 50◦ for 5 sec, and 60◦ for 4 min. 3. Precipitate the samples in ethanol or 2-propanol. 4. Dissolve in 2 µl loading solution and run on ABI. Conclusion The protocol described here for microdissection and genetic analysis of single cells has been successfully used in our laboratory for amplification and analysis of more than 500 single cell samples. The method is consistent and contamination can be avoided using the above-mentioned precautions. The inherent nature of single cell DNA analysis disables the possibility of confirming experiments. However, multiple sampling as well as partial confirmation by a new inner PCR followed by resequencing minimizes the uncertainty of data. The method provides a powerful tool in genetic analysis of microscopically defined cells, which would not be possible using crude tissue samples as template.
[31] Methylation in Gene Promoters: Assessment after Laser Capture Microdissection By ARTHUR R. BROTHMAN and JIANG CUI Introduction Laser capture microdissection (LCM) is a powerful tool for the assessment of pure populations of cells in heterogeneous tumors.1 Many cancers, particularly adenocarcinomas, contain not only an admixture of phenotypically distinct cells 1
M. R. Emmert-Buck, R. F. Bonner, P. D. Smith, R. F. Chuaqui, Z. Zhuang, S. R. Goldstein, R. A. Weiss, and L. A. Liotta, Science 274, 998 (1996).
METHODS IN ENZYMOLOGY, VOL. 356
Copyright 2002, Elsevier Science (USA). All rights reserved. 0076-6879/02 $35.00
344
GENETIC APPLICATIONS
[31]
(such as epithelial vs stromal cells), but also normal cells mixed in with genetically and epigenetically abnormal cells. Gross dissection of tumor-containing tissue therefore limits the types of studies and assays that can be performed. In contrast, the LCM procedure can precisely generate essentially pure populations of cells based on morphological characteristics. While LCM has been shown to be useful in the analyses of morphologic and genetic markers as discussed elsewhere in this volume, the focus of this article is on applications of this procedure to the study of epigenetic characteristics of cells, specifically DNA methylation. Methylation of cytosines in CpG islands of gene promoters has been shown to be strongly associated with histone acetylation changes and gene inactivation by transcriptional silencing.2 Multiple cancers have been shown to arise following deletion or mutation of alleles at the same locus on homologous chromosomes, most commonly termed tumor suppressor genes.3 The concept is that these genes suppress uncontrolled cell growth in specific tissues under normal circumstances. When one of these genes is mutated or deleted (either in somatic tissue or in the germ line), a second “hit” therefore completely knocks out the function of that gene and the suppression of growth is eliminated, resulting in a tumor.3 This traditional “two hit” hypothesis has been modified to include epigenetic inactivation as one of the “hits.” 4 Since DNA methylation is believed to render a gene inactive, it would thus function in the same way as a deletion or mutation. Studies of DNA methylation in association with various cancers are now shedding light on several mechanisms of tumorigenesis.5 Briefly, several techniques are now available to determine the methylation status of a particular DNA sequence. Historically, the use of restriction endonucleases, which differentially cut DNA based on the methylation of CpGs in a specific sequence, would yield DNA fragments of different sizes, depending on whether a particular sequence was methylated. This technique requires relatively large amounts of DNA and is thus not practical for the analysis of a heterogenous cancer. The introduction of several powerful PCR-based methods for detecting methylated cytosines has greatly enhanced the ability to detect changes in extremely small quantities of DNA.6,7 The in vitro conversion of unmethylated, but not methylated, cytosines by bisulfite has proved to be a powerful method to detect methylation differences in individual genes.6 This essential reaction is shown in Fig. 1. Whereas some techniques such as Ms-PCR7 or Ms-SNuPE 8 examine 2
P. L. Jones and A. P. Wolffe, Semin. Cancer Biol. 9, 339 (1999). A. G. Knudson, Jr., Cancer Detect. Prev. 7, 1 (1984). 4 P. A. Jones and P. W. Laird, Nat. Genet. 21, 163 (1999). 5 M. Esteller, P. G. Corn, S. B. Baylin, and J. G. Herman, Cancer Res. 61, 3225 (2001). 6 M. Frommer, L. E. McDonald, D. S. Millar, C. M. Collis, F. Watt, G. W. Grigg, P. L. Molloy, and C. L. Paul, Proc. Natl. Acad. Sci. U.S.A. 89, 1827 (1992). 7 J. G. Herman, J. R. Graff, S. Myohanen, B. D. Nelkin, and S. B. Baylin, Proc. Natl. Acad. Sci. U.S.A. 93, 9821 (1996). 8 M. L. Gonzalgo and P. A. Jones, Nucleic Acids Res. 25, 2529 (1997). 3
[31]
METHYLATION IN GENE PROMOTERS
345
FIG. 1. Treatment with bisulfite showing (a) deamination of unmethylated cytosine to uracil resulting in thymine following PCR, and (b) no conversion when cytosine is methylated.
base-specific sites to get an overall estimate of methylation status, we routinely use the bisulfite sequencing technique,6 which provides information on the methylation status of every cytosine within an amplified fragment of a particular gene promoter sequence. An example of the bisulfite sequence scheme for both methylated and unmethylated CpGs is shown in Fig. 2. In this article we describe the basic techniques of LMC followed by PCR of specific primers designed for the promoter region of the caveolin-1 gene in prostatic adenocarcinoma.9 The technique is applicable to virtually any primer design, gene, or tumor type of interest. Likewise, our analysis after PCR involves genomic sequencing, but this technique can easily be adapted to alternative assays such as Ms-PCR. Materials and Methods Slide Preparation It is essential that good quality, histological specimens be prepared for dissection. Several 5 µm sections are ideal and it is suggested that an adjacent section 9
J. Cui, L. R. Rohr, G. Swanson, V. O. Speights, T. Maxwell, and A. R. Brothman, Prostate 46, 249 (2001).
346
GENETIC APPLICATIONS
[31]
FIG. 2. Bisulfite sequencing of unmethylated (CG) and methylated (mCG) sites. DNA is first denatured, then treated with bisulfite, and then PCR amplified and sequenced.
to one which will be microdissected be previously stained with hematoxylin and eosin (H&E) to ensure that tumor boundaries and regions of high tumor content (preferably greater than 80%) can be marked. These should be outlined by an experienced histopathologist, and we routinely use different color ink to discern tumor and benign regions of a section. This H&E section can then be used as a template and aligned with the newly stained section from which cells will be microdissected. Sections should be cut with a clean, sterile microtome blade and mounted on plain glass microscope slides (no coating). Slides should be dried overnight in a 37◦ oven (not baked), then deparaffinized by incubation in a xylene solution (2× for 3 min), followed by rehydration through a series of 95, 70, and 50% ethanol solutions (2× for 3 min each). After rinsing in distilled H2O, the sections are then H&E stained and dehydrated in 50, 70, and 95% ethanol (2× for 1 min each). The slides can be air-dried and are then ready for LCM. Laser Capture Microdissection We have used the PixCell II LCM System (Arcturus Engineering, Mountain View, CA), and techniques described are modifications of those recommended by the manufacturer. The concept is essentially to “melt” a portion of a transparent thermoplastic film, on the cap of a microcentrifuge tube, onto cells with a carbon dioxide laser pulse, causing adhesion of the targeted cells onto the cap. This cap is then inverted into a buffer solution to begin DNA preparation. The complete
[31]
METHYLATION IN GENE PROMOTERS
347
LCM system, in addition to the laser and light sources, includes a movable arm for exact placement and removal of the tube cap plus computer interface with imaging capabilities for documentation of all experiments. Details of the mechanics of the system are provided in the PixCell User’s Manual (Arcturus Engineering). Variables which need to be determined for different specimen types appear to be directly related to adherence of the tissue section to the glass slide. Initial mounting and attachment of sections can play a role (specifically temperatures at which sections were dried). An easy test is placing a small piece of cellophane tape over a test section, gently pressing, and then peeling off. If the cells stick to the tape they will likely adhere to the cap after laser pulsing. The other controlling variables are the amplitude and duration of the laser pulse, which both can be adjusted to maximize cell removal (without “frying” cells). For prostate cells, we have succeeded in using 60 milliwatts with a duration of 5 msec pulses; for different cell types, this should be adjusted accordingly. Last, the approximate amount of cells microdissected should be determined to obtain sufficient DNA for a reliable PCR reaction; we have found that approximately 1000 epithelial cells from representative tumor areas yield an acceptable DNA amount. The cap turns during the LCM procedure, allowing maximal use of the thermoplastic surface. Once the desired number of cells is removed, the microcentrifuge tube is inverted onto the cap and cells are digested in 50 µl of proteinase K buffer (2 mg/ml proteinase K, 10 mM Tris-HCl, 5 mM EDTA, and 1% Tween 20) at 45◦ overnight prior to DNA isolation. An example of a prostate section before and after LCM, in addition to the same cells within the cap, is shown in Fig. 3. Genomic DNA Isolation and Bisulfite Treatment These techniques can be performed on LCM cells in virtually the same way as on larger quantities of cells isolated by other methods. We have used the Puregene DNA isolation kit (Gentra System, Minneapolis, MN) with the addition of 2 µl
FIG. 3. Images of a prostate section before (a) and after (b) LCM, and then the attached section on the cap (c) which is used for direct DNA isolation.
348
GENETIC APPLICATIONS
[31]
of glycogen (20 mg/ml) as carrier to assist in the precipitation of DNA. We then use the CpGenome DNA Modification Kit (Intergen, Purchase, NY) for bisulfite treatment, and elute DNA in 30 µl of TE (10 mM Tris-HCl pH 8.0, 0.1 mM EDTA) and store it at −20◦ . Primer Design Multiple programs can be used to design specific primers. Ideally one should pick a sequence within the 5 promotor region that can be easily sized and cut from an agarose gel (recommended between 100 and 500 bp). We have had success with Primer3 primer design software (http://www-genome.wi.mit.edu/cgibin/primer/primer3 www.cgi). It is critical that the primers be carefully designed with consideration of sequence modification after bisulfite treatment (all unmethylated Cs will be converted to Ts). No or few CpGs should be present in the primer sequences; if more than a few are present, mismatch to both methylated and unmethylated sequence should be considered in the design. Also, it should be apparent that a large number of CpGs are present in the product as these are the informative sequences. A detailed description of primer design concerns for the bisulfite sequencing technique is given elsewhere.10 Also, our experience indicates that some primers do not work, so it is wise to design multiple primers initially; this also maximizes information by increasing the sequence of the product(s). In some cases, nested PCR may be applied to increase the yield for sequencing. The example given in this article is for a primer set designed to the promoter region of the caveolin-1 gene, which we term CPM3. The sequence used was forward, 5 -GGA TAG GGT AGG ATT GTG GAT T-3 , reverse, 5 -CAC ATC CCC AAA ATT CTA ACA-3 . Control primers specific for unconverted (not bisulfite treated) DNA can be for any sequence routinely used in the laboratory. This control will help determine the overall efficiency of the bisulfite reaction and is recommended for any PCR-based methylation assay since no product should ever be seen when bisulfite-treated DNA is used. We used a portion of the androgen receptor (AR) gene, with a forward sequence, 5 -GTT TGG TGC CAT ACT CTG TCC AC, reverse, 5 -CTG ATG GCC ACG TTG CCT ATG AA. PCR and Sequencing PCR amplifications in our lab are generally performed in 50-µl reaction mixtures containing 5–10 µl of bisulfite-modified genomic DNA, 200 µM each of the four dNTPs, 1 µM each of respective primers, 1 mM MgCl2, 2 units AmpliTaq DNA polymerase (PerkinElmer, Foster City, CA), and reaction buffer consisting of 67 mM Tris-HCl, 16.6 mM ammonium sulfate, 1.7 mg/ml bovine serum albumin, and 10 mM 2-mercaptoethanol in TE (100 mM Tris-HCl pH 8.8, 0.1 mM EDTA) 10
S. J. Clark, J. Harrison, C. L. Paul, and M. Frommer, Nucleic Acids Res. 22, 2990 (1994).
[31]
METHYLATION IN GENE PROMOTERS
349
buffer. Reactions are carried out using specified conditions for each primer set. For the examples shown in this article we used a Genius Techne thermocycler (Techne, Princeton, NJ) with a hot start at 95◦ for 10 min, 35–45 cycles of amplification (40 sec at 95◦ , then 40 sec at 58◦ , and 1 min at 72◦ ). CPM3 generates a product of 275 bp and our AR generates a product of 413 bp. Each PCR product (50 µl) is then loaded onto a 1.5% agarose gel, with ethidium bromide and ultimately cut from the gel and extracted using a QIAquick gel extraction (Qiagen, Inc., Valencia, CA). For direct PCR sequencing, 100 ng of fresh PCR product is used as a template for the sequencing reaction, which is performed using the ABI Prism Big Dye sequencing method (PerkinElmer, Applied Biosystems Division, Foster City, CA). The same forward primers (3.2 nmol) used for the PCR amplification are used as sequencing primers. Results and Discussion After DNA isolation from LCM, we routinely run PCR reactions with appropriate controls to ensure both that the bisulfite conversion was successful and that the DNAs of interest are amplified. An example of such an experiment is shown in Fig. 4, which represents DNA isolated from a prostate tumor (as shown in Fig. 3) and amplification with the primer set CPM3, which is specific for bisulfite-modified DNA and in the 5 promoter region of the caveolin-1 gene (methylation specific); a second primer set for a region of the androgen receptor gene (AR) which is not specific for bisulfite-modified DNA was also used as a control. Experiments such as this are necessary to ensure that the bisulfite conversion is complete, as no
FIG. 4. PCR reactions after LCM using methylation-specific primers for a region of the caveolin-1 promoter (CPM3) and the androgen receptor gene (AR). Lanes: 1, 100-bp ladder; 2, CPM3 primers on unconverted DNA; 3, CPM3 primers on bisulfite-converted DNA; 4, AR primers on unconverted DNA; and 5, AR primers on bisulfite-converted DNA.
350
GENETIC APPLICATIONS
[31]
FIG. 5. Representative sequencing result for a portion of CPM3 following bisulfite treatment from a normal epithelial cell population (a), and from a prostate tumor population in which there was a high frequency of methylation (b).
product was obtained in the reaction using the AR primers after bisulfite (lane 5). This also ensures that the PCR reaction generates the appropriately sized product with the CPM3 primers after bisulfite treatment (lane 3). Analysis of data involves careful comparison of any sequences after bisulfite treatment with unmodified sequence. The simplest analysis involves highlighting
FIG. 6. Comparison of tumor vs normal methylation index (MI) of CPM3. Average MI for each of the 17 CpG sites within CPM3 from 25 prostate tumor samples (squares, solid line) and 23 normal samples (circles, dotted line). Bars represent standard error. The difference between tumor and normal specimens was statistically significant ( p < 0.05).
[31]
METHYLATION IN GENE PROMOTERS
351
all CpG sites and then determining whether they have remained CpG (indicating methylation) or changed to TpG (signifying it is not methylated). An example of sequence for a CPM3 experiment is given in Fig. 5. We use a “methylation index” (MI), which simply represents the frequency of observed methylated CpGs over total number of CpGs within a region. An advantage of detailed LCM is the ability to compare two different cell populations, and we show a difference in MI for tumor vs normal regions of sections from multiple prostate cancer patients in Fig. 6. Clearly there is a difference in methylation of the tumor sections, suggesting that the caveolin-1 gene is hypermethylated in a subset of prostate cancers. As noted above, data using LCM can be extremely useful in analyzing tumor specimens from heterogeneous adenocarcinomas. Since methylation is now thought to be a more widespread phenomenon in the development of multiple cancers, this technique can allow for rapid and accurate assessment of virtually any gene of interest. Acknowledgments This work was supported by a grant from the NCI (NIH), R01CA46269 (A.R.B.), and by core laboratories at the University of Utah (HCI Analytical Morphology Core and HCI Sequencing Core, 5P30CA42014).
134
SPECIALIZED USES
[11]
FIG. 2. Distribution of mono- and polyclonal micronodules microdissected from liver cirrhosis. Clonal analysis showed a random distribution of mono- (M) and polyclonal (P) micronodules in a case of cirrhosis without macronodule.
eosin stained section. The captured micronodular tissue was immediately placed in a 0.5-ml microfuge tube containing proteinase K. The HUMARA assay was performed after DNA extraction. In this study, 51% of the micronodules were monoclonal with a random distibution throughout the liver cirrhosis, as shown in Fig. 2. These results demonstrate the presence of neoplastic nodules in liver cirrhosis and suggest that monoclonal nodules could represent potential markers for the future development of a carcinoma.
Specific Advances through the Use of LCM Associated with Clonal Analysis Clonal analysis combined with LCM technology also provides further highlights on specific issues. For instance, before the development of assisted microdissection techniques, tumor heterogeneity was poorly demonstrated at the molecular level, although it is a common feature of human malignancies, well-recognized by histological analysis. Study of the adjacent foci of cells exhibiting various morphological patterns included inside a lesion is now feasible with the use of LCM. This has been illustrated in the analysis of the molecular clonality of melanomas. Using LOH microsatellite markers after LCM, authors assessed intratumor heterogeneity
This Page Intentionally Left Blank
Author Index
Numbers in parentheses are footnote reference numbers and indicate that an author’s work is referred to although the name is not cited in the text.
A A’Agati, V. D. D., 300 Abdulkadir, S. A., 53, 59(10), 198 Abramson, R. D., 231 Abu-Asab, M., 220 Ackermann, D., 63 Adam, B.-L., 44, 46(22) Adam, G. C., 191 Adams, M. W., 336 Adema, G. J., 247, 271 Adema, J., 90 Adham, N., 333 Agarwal, S. K., 221 Ahmadian, A., 334, 336 Ahram, M., 221 Ahrenkiel, V., 311, 321(19) Aihara, T., 133 Aist, J., 81 Akkayagil, E., 113 Akram, M., 240, 324 Aldaz, C. M., 122 Alevizos, I., 97, 296, 299, 299(8), 300(8), 324, 326(7; 8) Allaerts, W., 248 Allali-Zerah, V., 221 Allen, M. L., 25, 26(2) Allred, D. C., 317 Alund, G., 63 Amet, T., 337 Amorosi, A., 314 Anabirarte, M., 63 Anami, Y., 292 Andersen, M. K., 311, 321(19) Anderson, D. J., 220 Andre, P., 336 Andrew, R. V., 236 Andrulis, I. L., 106 Angrand, P. O., 210 Anna, C. H., 270
Aozasa, K., 270, 309, 318, 319(36) Appel, R. D., 33 Appella, E., 22, 24(16), 41, 43(18) Arends, J. W., 114 Arin, M. J., 207, 208, 211, 211(5), 212(5) Arnold, H.-H., 170 Artymyshyn, R., 333 Ashihara, T., 136 Ashkin, A., 81, 297 Aubele, M., 100, 113(7) Augood, S. J., 165, 260, 262(3) Axel, R., 220, 284 Axelrod, J., 220 Aydinli, K., 297 Azen, E., 264, 282, 283, 285(1), 299, 301(20)
B Ba, N., 133, 135(25) B¨ackvall, H., 334, 336, 337(17) Baden, H. P., 335 Baes, M., 248 Bahn, S., 165, 260, 262(3) Baker, A. J., 325, 329(20) Baker, D. F., 310 Baker, S. J., 310 Bakker, A., 102 Balazs, M., 63 Bale, A. E., 335 Balis, U. J., 236 Balliere, A. M., 221 Banks, P. M., 115 Banks, R. E., 22, 24(15), 33, 34, 36, 43, 43(8), 48(8), 70, 72(3), 79(3) Bannasch, P., 25, 81, 217 Baretton, G. B., 322 Barlund, M., 63 Barnes, D. M., 63 353
354
AUTHOR INDEX
Barrett, J. C., 270 Barsky, S., 129 Bashir, M. M., 236 Basrur, V., 22, 24(16), 41, 43(18) Basseal, D. J., 40 Bauder, U., 81 Bauer, J. W., 264 Baugh, L. R., 186 Bayard, F., 221 Baylin, S. B., 344 Bayliss, J. M., 248, 249, 252, 254, 254(26), 255(26) Beaty, M. W., 199, 200, 202(15), 203(15), 206(15), 207(15) Becich, M., 102 Beck, F., 170 Becker, B., 80, 97 Becker, I., 81, 100, 109, 110, 132, 297 Becker, K.-F., 81, 100, 109, 110, 132, 297 Bedossa, P., 129, 133, 135(24; 25) Behrendt, R. P., 322 Belgain, C., 303 Beltinger, C. P., 300 Ben, M., 292 Bendix, K., 103, 105(17) Bennett, W. P., 114, 335 Berger, D. H., 114 Berggren, K., 40 Bergmann, M., 111 Bergsma, D. J., 333 Berne, B., 335 Berner, A., 309, 318 Berns, G. S., 81 Berns, M. W., 81, 216 Bernsen, M. R., 90, 247, 271 Bernstein, D., 325, 329(20) Bertheau, P., 217 Bessis, M., 80 Best, C. J., 221, 247 Betsholtz, C., 170 Betsuyaku, T., 59, 296, 300(10) Bhattacharya, S., 22, 24(14), 292 Bhittner, A., 20, 22(11), 24(11) Bianchi, D. W., 299 Bichsel, V. E., 47, 48, 48(28), 161 Bielser, W., 25, 81, 216 Bijwaard, K. E., 203 Billiau, A., 248 Bioulac-Sage, P., 133, 135(25)
Bischeglia, M., 102 Bissig, H., 63 Bitran, J. D., 63 Bittenbender, S., 230 Bittner, A., 113, 186, 247, 264, 266(11), 327 Bjerknes, M., 169, 169(7; 8; 10), 170, 171 Blackstock, W., 33 Blanchard, K. L., 131 Bland, K. I., 13, 18(1), 25(1) Blomeke, B., 114 Blum, L., 199, 200, 202(15), 203(15), 206(15), 207(15) Bl¨umel, P., 25(9), 26, 271 Bock, O., 52 Bogaard, M. E., 280 Boguski, M. S., 221 Boguth, G., 40 Bohle, R. M., 82, 99, 103, 104, 106, 109(16), 110, 110(29), 111, 112, 112(38), 113, 197, 299, 300 Bohling, S. D., 229, 231, 232(5), 235, 237, 237(9) B¨ohm, M., 81, 85, 100, 132, 197, 297 Bolte, E., 221 Bonat, S., 220 Boner, R. F., 311 Bonini, J. A., 333 Bonkhoff, H., 310 Bonner, R. F., 3, 13, 14, 20(6), 22(4; 6), 25, 34, 74, 100, 105, 114, 132, 137, 145, 147(2), 148, 156, 197, 206(9), 207, 207(9), 215, 217, 219(12), 224, 240, 244(1), 249, 259, 292, 295, 296, 296(4), 311, 329, 334, 343 Bonvoust, F., 133, 135(25) Bookstein, R., 317 Bordi, C., 135 Borell, T. J., 318 Born, I. A., 247 Bornstein, S. R., 20, 22(10), 24(10), 216, 220, 221, 222, 324 Borowsky, B., 333 Børresen, A. L., 318 Bos, J. L., 280 Bostwick, D. G., 43, 79, 113, 191, 309, 310, 311, 317, 317(8), 318, 319, 324 Boteju, L. W., 333 Bottner, A., 220 Brady, S. P., 114
355
AUTHOR INDEX Braman, J. C., 299 Branchek, T. A., 333 Br¨andstedt, S., 309 Brandtzaeg, P., 115 Brannan, C., 311 Brash, D. E., 335 Braude, P. R., 337 Brawer, M. K., 309 Bray, L., 260 Braylan, R. C., 202 Br´echot, C., 90 Brehm, G., 81 Breidert, M., 20, 22(10), 24(10), 220 Brenner, S., 81 Brentnall, T. A., 129 Brettingen, N. T., 173 Brinkmann, U., 221 Brisco, M. J., 129 Broadwell, K., 262 Brocksch, D., 25 Brothman, A. R., 324, 343, 345 Brown, E. L., 186, 325, 329(17) Brown, J. W., 220 Brown, M. R., 16(17), 24, 44, 46(25) Brown, M. S., 229 Brown, R., 103 Bruderer, J., 63 Bryant, P. J., 81 Bubendorf, L., 63 Bucholz, T., 91, 296 Buckley, J. D., 136 Budi, I., 221 Bunn, H. F., 131 Burgemeister, R., 80, 295, 296(2), 297, 299(2), 300(2), 301(2) Burgess, J., 157, 259 Burguera, B. G., 248, 251 Burk, M. R., 297 Burns, A. L., 221 Burt, J., 81 Burton, M. P., 103 Busch, C., 309 Busche, S., 112 Busque, L., 133, 135(19) Butcher, E. C., 262 Butler, S. I., 333 Buyuksal, I., 311, 321(18) Buzard, G. S., 314 Byrne, M. C., 325, 329(17)
C Cabras, A. D., 135, 202 Cagan, R. L., 296, 300(6) Cahill, T., 81 Cairns, N., 260 Calvert, R. J., 314 Calvert, V. S., 48 Candidus, S., 135 Cao, T., 208, 212(6), 213 Cao, Y., 266 Caporaso, N. E., 314 Capron, F., 90 Cara, A., 49, 300 Carlisle, A. J., 16(17), 24 Carmeliet, P., 248 Carroll, P., 63 Carroll, R. S., 248 Casciola-Rosen, L., 70, 74, 79(8) Caton, A. J., 230 Cazares, L. H., 44, 46(22), 163 Ceccatelli, S., 248 Cerpa-Poljak, A., 33 Certa, U., 325 Cesarman, E., 239 Chadburn, A., 239 Chambers, J. K., 333 Chan, W. C., 103, 105(15) Chan, Y. M., 209 Chandrasekharappa, S. C., 221 Chang, H., 114 Chang, M. C., 324 Chang, W. C., 227, 231(2) Chang, W. W. L., 170 Chang, Y. T., 324 Charboneau, L., 47, 48(28), 70(5; 6), 71, 79(5), 162, 164, 324 Chavira, M., 52 Chawengsaksophak, K., 170 Chayvialle, J. A., 221 Chee, M. S., 325, 329(17) Chen, L., 324 Chen, T., 47, 48(28) Chen, W.-G., 133, 135(20) Chen, Y.-Y., 133, 135(20) Chenchik, A., 113 Cheng, H., 169, 169(7; 8; 10), 170, 171 Cheng, J., 319 Cheng, R. S., 326 Chernokalskaya, E., 40
356
AUTHOR INDEX
Chesner, J., 122 Cheung, I., 247 Cheung, N. K. V., 247, 324 Cheung, V. G., 300 Chew, K., 63 Chi, D. J., 302(2), 303 Chi, S., 247 Childs, G. V., 248 Chin, J. E., 292 Chin, W. W., 248 Choi, K. G., 292 Chomel, J. C., 280 Chrousos, G. P., 20, 22(10), 24(10), 220, 221, 324 Chu, S. S., 14, 16(7), 18(7), 20(7), 22(7), 24(7) Chuaqui, R. F., 3, 13, 18, 21(9), 22(4), 25, 34, 79, 100, 103, 104(18), 114, 125, 132, 137, 145, 148, 156, 177, 197, 204, 206(9), 207, 207(9), 209, 217, 219(12), 224, 240, 244, 244(1), 249, 259, 262, 264(10), 292, 293, 295, 296(4), 300, 311, 334, 343 Chung, I.-M., 136 Cimino, G. D., 101 Cinquanta, M., 270 Clark, S. J., 348 Claussen, U., 44, 46, 46(26), 113 Clement-Sengewald, A., 81, 91, 296, 297, 334 Cobbers, J. M., 247 Cockett, A. T. K., 310 Cohen, D., 299, 324, 326(8) Colby, T. V., 133, 135(20) Cole, K., 13, 16(3), 18, 21(9), 22(3), 25, 44, 46(25), 79, 103, 104(18), 125, 132, 145, 148, 177, 197, 204, 209, 243, 249, 259, 262, 264(10), 293, 300, 311 Coleman, P., 266 Collins, F. S., 221 Conklin, B. R., 325, 329(20) Conner, H. M., 115 Conrad, N. K., 135 Conran, R. M., 119 Conti, C. J., 122 Coombs, N. J., 114 Copeland, N. G., 311 Cordon-Cardo, C., 114, 247 Cordwell, S. J., 33, 40 Corn, P. G., 344 Cornea, A., 3 Cornil, I., 63
Cossman, J., 236 Costa, J., 270, 324 Costa, P., 120 Costantini, F., 173 Cote, R. J., 317 Cotte, N., 325, 329(20) Couce, M. E., 248, 251 Coulombe, J., 81 Coward, P., 325, 329(20) Crabtree, J. S., 221 Cravatt, B. F., 191 Craven, R. A., 33, 34, 36, 43(8), 48(8) Cremer, C., 81 Cremer, M., 90, 105 Cremer, T., 81, 90, 105 Crescenzi, M., 236 Crino, P. B., 113, 260, 326, 327(22) Crnogorac-Jurcevic, T., 34 Croce, C. M., 236 Cui, J., 324, 343, 345 Cunha, G. R., 91 Curran, S., 35, 39(7), 40(7), 41(7), 43(7), 70(4), 71, 72(4), 113, 209 Currie, I., 40
D Daa, T., 133, 135(21) D’Agati, V. D. D., 49 Dahiya, R., 322 Dai, M., 333 Dalton, W. S., 311, 321(18) d’Amore, F., 103, 105(15), 197, 199(1) Danilova, V., 264, 283, 285(1), 299, 301(20) Danning, C., 74, 79(8) Darg`ere, D., 133, 135(25) Darling, T. N., 264 Datson, N. A., 150 Dauway, E., 50 David, D. A., 236 Davidson, B. R., 22, 24(14), 292 Davies, H., 44, 46(26), 113, 160 Davison, M., 40 Dax, E. M., 221 Debatin, K. M., 300 Debelenko, L. V., 221 Dee, S., 329 DeFlavia, P., 102
357
AUTHOR INDEX De Goeij, A., 114 Degtyarev, M. Y., 325, 329(20) de Kloet, E. R., 150 Delabie, J., 103, 105(15) de la Torre, M., 309 Demetrick, D. J., 63, 64, 65, 65(12) Denef, C., 248 Deng, G., 197, 322 DePinho, R. A., 70 Desai, K., 325, 329(20) de Saizieu, A., 325 Devereux, T. R., 134, 270 Devor, D. E., 310 de Vries, E., 90, 247, 271 DeWeese, T. L., 91 deWith, A., 81 De Wolf, A., 248 Dhillon, A. P., 22, 24(14), 292 Dhir, R., 102 Diachenko, L., 113 Diaz-Cano, S. J., 114 Diebold, J., 322 Dietl, J., 205 Dietmaier, W., 113 diFrancesco, L. M., 63, 64, 65(12) Dijkman, H. B., 90, 247, 271 Dillon, D., 270, 324 Di Naro, E., 301 Dintzis, S. M., 53, 59(10), 198 Dolken, G., 109 Dolter, K. E., 299 Done, S. J., 106 Dong, H., 325, 328(18), 329(17; 18) Dong, Q., 221 Dong, S. M., 311, 321(21) Dong, Z., 270 Donoff, R. B., 299, 324, 326(8) Drury, R., 34 Dulac, C., 284 Duncan, B. W., 203 Duncan, M. H., 133, 135(19) Duncan, M. W., 33 Dunn, J. K., 318 Dunn, M. J., 22, 24(15), 33, 40, 43, 70, 72(3), 79(3), 159 Dupre, J., 221 Duray, P. H., 43, 48, 79, 113, 191, 270, 310, 317, 317(8), 324 D¨urst, M., 46 Dvorakova, M., 110, 112(38)
E Eady, R. A., 337 Eanes, W. F., 292 Eberhardt, N. L., 248 Eberwine, J. H., 113, 260, 266, 326, 327(22) Eckert, K. A., 336 Edwards, J., 81 Egawa, S., 318 Ehrhart-Bornstein, M., 220, 221 Ehrig, T., 49, 53, 59(10), 198 Eimoto, T., 53, 103, 104(20), 105(20) Eisenhofer, G., 220 Elenitoba-Johnson, K. S., 224, 229, 231, 232(5), 235, 237, 237(9), 238, 239(17) Elkahloun, A. G., 20, 24(12), 324 Elshourbagy, N. A., 333 Emmert-Buck, M. R., 3, 13, 14, 16(3; 17), 18, 20(6), 21(9), 22, 22(3; 4; 6), 24, 24(16), 25, 34, 41, 43, 43(18), 44, 46(25), 47, 48, 48(28), 70(6), 71, 79, 100, 103, 104(18), 113, 114, 125, 132, 137, 145, 148, 156, 163, 164(18), 177, 191, 197, 204, 206(9), 207, 207(9), 209, 217, 219(12), 221, 224, 240, 243, 244(1), 247, 249, 259, 262, 264(10), 270, 292, 293, 295, 296(4), 300, 310, 311, 317, 317(8), 324, 329, 334, 343 Emson, P. C., 260, 262(3) Endl, E., 113 Englert, C., 48, 79, 148, 162, 324 Enomoto, T., 135 Ensley, J. F., 324 Erlander, M. G., 12, 20, 22(11), 24(11), 113, 186, 247, 264, 266(11), 327 Ermert, L., 82, 103, 104, 106, 109(16), 110(29), 111, 197, 299, 300 Ernst, G., 44, 46, 46(26), 113 Escamilla-Ponce, N., 103 Essand, M., 221 Esteller, M., 344 Eve, D. J., 260
F Falco, V., 299 Falk, P. G., 171 Farqugar, M. G., 248 Fausa, O., 115 Favara, B. E., 133, 135(19)
358
AUTHOR INDEX
Fearon, E. R., 129, 310 Feinberg, A. P., 129 Feist, H., 82, 90(19) Fend, F., 18, 21(9), 79, 103, 104(18), 105, 116, 125, 135, 148, 158, 161, 177, 196, 199, 200, 202, 202(15), 203(15), 204, 206(15), 207(15), 209, 243, 249, 262, 264(10), 293, 296, 300 Ferrara, N., 248 Ferretti, J. A., 270 Fey, S. J., 40 Fezjo, M. L. S., 135 Fialkow, P. J., 129 Fidelman, N., 325, 329(20) Fiedler, W., 44, 46, 46(26), 113 Figdor, C. G., 90, 247 Fijan, A., 63 Findlay, I., 337 Fink, L., 82, 99, 103, 104, 106, 109(16), 110, 110(29), 111, 112, 112(38), 113, 197, 299, 300 Finke, J., 109 Finkelstein, S. D., 102 Finlay, C. A., 310, 335 Finnell, R., 266 Fischer, H. P., 114 Fischer, S. G., 238 Fisher, C., 48 Fisher, G. H., 322 Fishman, G. I., 325, 329(20) Fitzgerald, L. R., 333 Flanagan, R. J., 221 Fleisher, T. A., 70(6), 71, 324 Florence, C. D., 310, 317, 317(8) Flores, T., 335 Flury, R., 63 Fojo, A., 292 Foley, J. F., 20, 21(13), 22(13), 24(13), 53, 58(12), 102, 103(13), 110(14), 134, 149, 198, 208, 261, 264(7), 270, 300 Follettie, M. T., 325, 329(17) Forbes, M. A., 22, 24(15), 43, 70, 72(3), 79(3), 159 Forray, C., 333 Foshag, L. J., 303 Foss, R. D., 119 Fossa, S. D., 318 Foster, O. J., 260 Fothergill, J. E., 35, 39(7), 40(7), 41(7), 43(7), 70(4), 71, 72(4), 113
Fournier, H., 221 Franco, D., 90 Franke, F. E., 110, 112 Fraumeni, J. F., 314 Freiha, F. S., 318 Friedemann, G., 80 Friedman, A. J., 135 Friedrich, G., 171 Fritsch, E., 101 Fritzen, R., 109 Frommer, M., 344, 345(6), 348 Fuchs, E., 209 Fuhrer, D., 135 Fujii, R., 333 Fujimoto, A., 302, 302(4), 303 Fujino, M., 333 Fujita, H., 249 Fujita, M., 135 Fujita, T., 249 Fujiura, Y., 170 Fujiwara, Y., 302(2), 303 Fukusumi, S., 333 Furukawa, Y., 97, 324 Furuya, S., 50, 51(5) Fuzesi, L., 114
G Gailani, M. R., 335 Galindo, J. E., 20, 22(11), 24(11), 113, 186, 247, 264, 266(11), 327 Gallagher, G. T., 299, 324, 326(8) Galli, M., 314 Gallinat, S., 112 Gallo, M. V., 325, 329(17) Garcia, E., 44 Garon, J., 221 Garrel, D., 221 Gartler, S. M., 136 Garvin, A. M., 301, 337 Gassel, A. M., 205 Gasser, T., 63 Gehlen, J., 114 Geist, B., 202 Geitvik, G., 318 Gelfand, D. H., 231 Genovese, M. C., 262 Gerald, C., 333 Gerald, W. L., 240, 247, 324
359
AUTHOR INDEX Ghany, M., 74, 105, 148, 215, 296 Ghosh, A. K., 317 Giercksky, H. E., 132, 290 Gillespie, J. W., 14, 20, 20(6), 22, 22(6; 10), 24(10; 16), 41, 43, 43(18), 44, 46(25), 47, 48, 48(28), 74, 79, 79(8), 113, 148, 162, 163, 164(18), 191, 220, 221, 222, 324, 329 Gilliland, D. G., 131, 133, 135(19) Ginsberg, S. D., 260 Gires, F., 80 Girton, J., 81 Giuliano, A. E., 303 Givol, I., 236 Gjerdrum, L. M., 103, 105(17) Glasow, A., 20, 22(10), 24(10), 220, 222, 324 Glickman, J. W., 70(6), 71, 324 Gl¨ockner, S., 52, 82, 90(19) Gloning, K.-Ph., 90, 295, 296(2), 299(2), 300(2), 301(2) Gnarra, J., 217 Godfrey, T. E., 52 Going, J. J., 197, 207, 217 Goldstein, P., 310 Goldstein, S. R., 3, 13, 22(4), 25, 34, 74, 100, 105, 114, 132, 137, 145, 147(2), 148, 156, 197, 206(9), 207, 207(9), 215, 217, 219(12), 224, 240, 244(1), 249, 259, 292, 295, 296, 296(4), 311, 334, 343 Goldsworthy, S. M., 20, 21(13), 22(13), 24(13), 53, 58(12), 102, 103(13), 110(14), 149, 198, 208, 261, 264(7), 300 Gong, L., 44, 46(22) Gonzalgo, M. L., 344 Goodlad, R. A., 135 Gooley, A. A., 33 Gordon, J. I., 167, 169(9), 170, 171, 171(9), 173(9), 174(9), 177(9), 178(9), 190, 190(35), 193, 195(41), 201, 296, 300(6) G¨org, A., 40 Gospodarowicz, D., 248 Gosslar, U., 262 Gottesman, M. W., 292 Gough, A. C., 114 Gough, M., 22, 24(15), 43, 70, 72(3), 79(3) Graff, J. R., 344 Grande, J. P., 250, 251(27) Grandjouan, S., 280 Gray, C., 325 Gray, I. C., 317, 321(27) Gray, J. W., 52, 63
Green, J. E., 310 Greenblatt, M. S., 335 Greene, D. R., 318 Gregerman, R. I., 221 Greiner, T. C., 227, 231(2) Greulich, K.-O., 81 Griffin, G. L., 59, 296, 300(10) Griffith, B. B., 133, 135(19) Griffith, R. C., 236 Grimm, H., 110 Grimminger, F., 82, 111, 113, 197, 299 Grist, S., 129 Grønbæck, K., 311, 321(19) Gros, P., 292 Gross, E. G., 335 Gu, J., 18 Gudat, F., 63 Guerrera, S., 333 Guevarra, M. S., 333 Guha-Thakurta, N., 119 Guillen, J. G., 280 Guldberg, P., 311, 321(19) Gulley, M. L., 103 Gundry, R. A., 236 Guo, H., 20, 22(11), 24(11), 113, 165, 186, 247, 264, 266(11), 327 Guru, S. C., 221 Gustafson, L., 248 Gusterson, B. A., 115, 124(16) Gutkind, J. S., 161, 324 Gutman, P., 119
H Haba, T., 135 Habata, Y., 333 Haberberger, R., 110, 112(38) Haberbosch, W., 110 Haga, N., 292 H¨aggman, M. J., 309, 310, 317(9) Haghighi, B., 160 Hahn, S., 295, 296(2), 297, 299(2), 300(2), 301, 301(2), 337 Haidan, A., 20, 22(10), 24(10), 220, 324 Hainaut, P., 335 Haliassos, A., 280 Halperin, A. J., 335 Hamet, P., 221 Hamilton, S. R., 129, 310
360
AUTHOR INDEX
Hamilton-Dutoit, S., 103, 105(17) Hammer-Wilson, M., 81 Han, J. Y., 311, 321(21) Hanauer, S., 63 Handt, S., 114 Handyside, A. H., 295, 337 Hanh, S., 295 Hansen, N. E., 311, 321(19) Hansen, N. M., 303 Hansmann, M. L., 100, 105, 197, 217 Hanze, J., 82, 104, 111, 113, 197, 299, 300 Harder, A., 40 Hardmeier, T., 63 Harnden, P., 22, 24(15), 43, 70, 72(3), 79(3) Harnden, R. P., 36, 43(8), 48(8) Harris, C. C., 114, 335 Harris, E., 63 Harris, R., 33 Harris, S. R., 310 Harrison, J., 348 Hartmann, A., 113 Hartmann, E., 111 Harvei, S., 309 Hase, A., 311, 319(16) Hasemeier, B., 111 Hashimoto, Y., 170 Hass, C. J., 322 Hatta, N., 302(4), 303 Hautekeete, E., 248 Hawes, D., 208, 217 Hawkins, E., 40 Hayashi, I., 318 Hayashi, K., 238 Healy, E., 303 Heath, C. W., 114 Heath, J. K., 170 Heckl, W. M., 81, 90, 100, 105, 297 Heid, C. A., 181 Heinmoller, E., 113 Hellekant, G., 264, 283, 285(1), 299, 301(20) Helpap, B., 310 Henzel, W., 248 Heppner, C., 221 Herath, J. F., 318 Heremans, H., 248 Hering, F., 63 Herman, J. G., 344 Hernandez, A. M., 208, 217 Hernandez, T., 335 Herring, J., 43, 48, 79, 113, 191, 324
Herrmann, M. G., 232 Herzig, G. P., 236 Hewitt, S. M., 48 Hiasa, Y., 318 Higuchi, R., 101 Hilbers, U., 20, 22(10), 24(10), 220 Hill, A. A., 186 Hill, J., 113 Hillenkamp, F., 25, 81, 216 Hiller, T., 100 Hilsenbeck, S. G., 317 Hino, O., 25, 81, 217 Hinrichs, S. H., 103, 105(15) Hins, P. W., 310 Hinson, J. P., 220 Hintner, H., 264 Hinuma, S., 333 Hirohashi, S., 292 Hiroi, N., 220 Hirose, M., 50, 51(4), 114, 115, 115(3), 117, 117(3), 118, 118(3; 17), 119, 120, 120(3), 124, 174, 198, 261, 262(8) Hirschmann, A., 322 Ho, J. P., 63 Ho, M. H., 325, 328(18), 329(18) Hochstrasser, D. F., 33 Hofele, C., 247 H¨ofler, H., 52, 81, 100, 109, 110, 113(7), 135, 202, 203 Hofstadter, F., 113 Hokfelt, T., 248 Holland, J. F., 25, 26(2) Holland, P. M., 231 Hollstein, M., 310, 335 Holschermann, H., 110 Holtschlag, V., 49, 51, 60, 247 Holzappel, H. P., 135 Holzgreve, W., 295, 296(2), 297, 299(2), 300(2), 301, 301(2), 337 Honda, K., 133, 135(21) Hongyo, T., 314 Hoon, D. S. B., 134, 135(26), 302, 302(2; 3), 303 Hooper, L. V., 167, 171 Hori, M., 111 Horio, M., 335 Horn, L. C., 222 Horton, H., 325, 329(17) Hosoya, M., 333 Hotlschlag, V., 53 Hou-Jensen, K., 311, 321(19)
361
AUTHOR INDEX Housman, D. E., 292 Houston, S. J., 63 Hoving, S., 40 Howe, J. R., 114 Hrapchad, B., 241 Hsi, E. D., 324 Hu, N., 22, 24(16), 41, 43(18), 44, 46(25), 48 Hu, W., 322 Huang, J., 22, 24(16), 41, 43(18), 44, 46(25), 50 Huang, L., 114 Huang, L. E., 49 Huang, Q. I., 321 Hudson, J. R., Jr., 20, 24(12), 324 Hudson, S., 230 Hulting, A. L., 248 Humphery-Smith, I., 33 Hunter, C. P., 186 Hutchens, T. W., 92 Hutton, E., 209
I Ibrahim, S., 115, 124(16) Igarashi, H., 13, 22(2) Illei, P., 247 Imai, E., 111 Imamichi, Y., 271 Imaoka, S., 132, 133 Inagaki, H., 53, 103, 104(20), 105(20) Inaji, H., 132 Innov, Y., 292 Inoue, K., 249 Irwin, D. M., 335, 336(15) Irwin, M., 170 Isaacs, S. T., 101 Isenberg, G., 216 Ishii, M., 311, 319(16) Ishikawa, H., 170, 249 Ito, A., 249 Itoh, N., 311, 319, 319(16) Itohara, S., 170 Ivakhnenko, V., 337 Iwahana, H., 238
J Jackson, M. R., 20, 22(11), 24(11), 113, 186, 247, 264, 266(11), 327
Jaffe, E. S., 199, 200, 202(15), 203(15), 206(15), 207(15), 236 Jager, P., 63 Jain, K. K., 113, 157, 259 Jauch, K. W., 113 Jenkins, N. A., 311 Jenkins, R. B., 318 Jennings, S. B., 310, 317(8) Jensen, R. H., 52 Jessell, T. M., 173 Jessup, J. M., 310 Jiang, W., 280 Jin, L., 177, 201, 204(17), 248, 249, 250, 250(23), 251, 251(23; 27), 252, 254, 254(26), 255(26), 271 Johnson, L., 63 Jonas, U., 58 Jones, P. A., 136, 344 Jones, P. L., 344 Joos, S., 247 Jorgensen, T., 202 Joy, K. C., 20, 22(11), 24(11), 113, 186, 247, 264, 266(11), 327 Juan, G., 247 Junker, K., 44, 46, 46(26), 113
K Kacharmina, J. E., 113, 326, 327(22) Kaestner, K. H., 170 Kahn, S. M., 280 Kaiser, H. E., 324 Kaiser, U. B., 248 Kakkar, A. K., 22, 24(14), 292 Kakuichi, H., 317 Kallioniemi, A., 63 Kallioniemi, O. P., 63 K¨ammerer, U., 205 Kanazawa, H., 238 Kanzler, H., 105 Kaplan, J. C., 280 Kapp, M., 205 Karlsen, F., 318 Karlsson, L., 170 Kashima, K., 133, 135(21) Katz, M. S., 221 Kaur, G., 322 Kawaguchi, M., 170 Kawamata, Y., 333
362
AUTHOR INDEX
Kawamoto, S., 203 Keck, W., 325 Keil, M. F., 220 Keith, W. N., 63 Kelemen, P., 302(2), 303 Kellendonk, C., 210 Kellman, R. M., 324 Kelly, P. A., 222, 324 Kelly, T. M., 221 Kempler, J., 63 Kent, R., 299, 324, 326(8) Kerbel, R. S., 63 Kerlen, G., 81 Kern, S. E., 100, 129 Kerner, T., 113 Kerschmann, R., 63 Kester, M. B., 221 Khorsand, J., 238, 239(17) Khrapko, K., 336 Kihara, C., 97, 324 Kikuchi, Y., 25, 81, 217 Kim, A., 336 Kim, C. H., 311, 321(21) Kim, D., 299, 324, 326(8) Kim, H. S., 311, 321(21) Kim, S. H., 52, 311, 321(21) Kim, S. Y., 311, 321(21) Kim, Y. S., 221 Kinfe, T., 103, 104, 109(16), 111, 300 King, T. C., 231, 232(5), 237, 238, 239(17) Kingsbury, A. E., 260 Kinzler, K. W., 149 Kisseberth, W. C., 173 Kitahara, O., 97, 324 Kitahori, Y., 318 Kitzes, M., 81 Kitzis, A., 280 Kleeberger, W., 82, 90(19) Klein, L., 63 Klimek, F., 25, 81, 217, 300 Klimstra, D. S., 114 Klonisch, T., 111 Klotman, P. E., 49, 300 Knonagel, H., 63 Knowles, D. M., 239 Kn¨uchel, R., 80 Knudsen, A. G., 129 Knudson, A. G., Jr., 344 Knuechel, R., 113 Ko, Y., 114
Kobayashi, M., 325, 329(17) Kocarek, T. A., 120 Kohda, Y., 104, 260, 261(5) Kohler, U., 222 Kohlhoff, S., 113 Kohn, B., 220 Kohn, E. C., 16(17), 24, 44, 46(25) Kohno, Y., 97, 296, 299, 299(8), 300(8), 324, 326(7; 8) Kolble, K., 106 Komano, H., 170 Kondylis, F. I., 44 Konishi, N., 318 Kononen, J., 63 Koonce, M., 81 Koontongkaew, S., 324 Koreth, J., 129 Korge, B. P., 207 Korsmeyer, S. J., 236 Kotake, T., 317 Kouranova, E. V., 333 Koyama, C., 249 Koyama, H., 132 Krafft, A. E., 203 Kramer, F. R., 231 Kraniak, J. M., 120 Kreipe, H., 52, 82, 90(19), 111 Kreitman, M., 292 Kremer, M., 125, 135, 158, 196, 202, 300 Krieg, T., 207 Krizman, D. B., 13, 16(3; 17), 22(3), 24, 221, 324 Krohn, K., 135 Kruh, J., 280 Kubelik, A. R., 291 Kubo, Y., 25, 81, 217 Kuczyk, M. A., 58 Kuecker, S. J., 177, 201, 204(17), 249, 250(23), 251(23), 271 Kukharenko, V., 337 Kuliev, A., 337 Kulig, E., 177, 201, 204(17), 248, 249, 250(23), 251(23), 252, 271 Kumar, S., 199, 200, 202(15), 203(15), 206(15), 207(15) Kummer, W., 82, 103, 104, 106, 109(16), 110, 110(29), 111, 112(38), 197, 299, 300 Kunitake, S. T., 14, 16(7), 18(7), 20(7), 22(7), 24(7) Kunkel, E. J., 262
AUTHOR INDEX Kunkel, T. A., 336 K¨uppers, R., 100, 105, 197, 217 Kuwata, H., 92 Kwa, A. T., 325, 329(20) Kwok, S., 101
L Lacour, B., 90 Lacroix, A., 221 Ladanyi, M., 247 Lafferty, A., 220 Lahr, G., 4, 25, 25(6–9), 26, 31(6), 70, 81, 82, 90(18), 92(15), 100, 111(6), 114, 158, 197, 221, 271, 277(5), 280(1), 281(5), 299, 334 Laird, P. W., 344 Lamb, R. F., 197, 207, 217 Lambert, K. N., 283 Lamsan, J., 248 Landowski, T. H., 311, 321(18) Landthaler, M., 97 Lane, D. P., 335 Langdon, P., 262 Lange, B., 230 Lange, W., 109 Lapenson, D. P., 120 Larcher, F., 122 Larsen, P. M., 40 Laurendeau, I., 133, 135(24) Lawrie, L. C., 35, 39(7), 40(7), 41(7), 43(7), 70(4), 71, 72(4), 113, 159 Lazarev, A., 40 Lazaridis, E., 50 Lazarus, N., 262 Leblond, C. P., 169, 170 Lebrethon, M. C., 221 Ledbetter, D. H., 310 Lee, B., 221 Lee, B. L., 248 Lee, C., 322 Lee, E. K., 325, 329(20) Lee, J., 18, 21(9), 79, 103, 104(18), 125, 148, 177, 204, 209, 243, 249, 262, 264(10), 293, 300, 311, 321(21) Lee, P.-S., 133 Lee, S. H., 311, 321(21) Lee, S. L., 317, 321(27) Lee, V. M., 260 Lees, A. J., 260
363
Leethanakul, C., 161, 324 Leffell, D. J., 335 Lehman, T., 114 Lehmann, U., 52, 82, 90(19), 111 Leighton, S., 63 Lemoine, N. R., 22, 24(14), 34, 289, 292 Lenardo, M. J., 322 Lerman, G. S., 333 Lerman, L. S., 238 Leroyer, R., 221 Leung, S.-M., 44, 46(22), 163 LeVangie, R., 20, 24(12), 324 Levine, A. J., 310, 335 Levy, J., 131 Leymarie, P., 221 Li, J. C., 297 Li, R., 113 Li, Z. H., 208, 217 Liaw, L. H., 81 Lichtinghagen, R., 58 Lichy, J. H., 203 Lieber, M. M., 318 Lielpetere, I., 103, 105(17) Lilischkis, R., 111 Limwongse, V., 324 Lin, A. Y., 322, 335 Lin, C. S., 173 Lin, J. C., 318 Lin, J. T., 324 Lin, S. S., 193, 195(41) Lindahl, P., 170 Linder, G., 136 Linehan, W. M., 43, 44, 46(25), 48, 79, 113, 148, 191, 217, 270, 310, 317, 317(8), 324 Ling, G., 334, 335, 336, 337(17) Link, A. J., 40 Linnola, R. I., 135 Liotta, L. A., 3, 13, 14, 16(17), 18, 20(6), 21(9), 22(4; 6), 24, 25, 34, 36, 44, 46(25), 47, 48, 48(28), 70(5; 6), 71, 79, 79(5), 100, 103, 104, 104(18), 125, 132, 137, 145, 148, 156, 161, 162, 177, 197, 204, 206(9), 207, 207(9), 209, 217, 219(12), 221, 224, 240, 243, 244(1), 249, 259, 262, 264(10), 292, 293, 295, 296(4), 300, 310, 311, 317, 317(8), 324, 329, 334, 343 Liu, E. T., 330 Liu, M. L., 310 Liu, O., 319 Liu, Y., 191
364
AUTHOR INDEX
Livak, K. J., 181, 291 Lloyd, R. V., 177, 201, 204(17), 248, 249, 250, 250(23), 251, 251(23; 27), 252, 254, 254(26), 255(26), 271 Lo, Y.-L., 137 Locker, J., 203 Lockett, S. J., 63 Lockhart, D. J., 325, 328(18), 329(17; 18) Loeffler, L., 74, 79(8) L¨ohrs, U., 322 Lohse, J. K., 173 Lomas, L., 44, 46(26), 113, 160 Longley, M. A., 208, 211, 211(5), 212(5; 6), 213 Lopez, M. F., 40 Lorenz, R. G., 170 Lothe, R. A., 290 Louha, M., 90 Lowry, O. H., 192, 193(40) Lu, S.-H., 280 Lu, Y., 197 Lubensky, I. A., 217, 221 Lubke, C., 220 Luc Fehr, J., 63 Luider, J., 64, 65(12) Lundberg, K. S., 336 Lundeberg, J., 334, 335, 336, 337(17) Luo, L., 20, 22(11), 24(11), 113, 165, 186, 247, 264, 266(11), 327 Luukkaa, H., 323 Luzzi, V., 49, 51, 53, 60, 247 Lyon, M. F., 131
M MacGrogan, D., 317 Macintosh, C. A., 290 Macoska, J. A., 310, 317(9) Maeda, S., 335 Maertens, P., 248 Magao, M., 317 Mahadevappa, M., 299, 323, 324, 326(8), 329, 330(26), 332(26) Mahoudeau, J., 221 Mailloux, J., 248 Maitland, N. J., 290 Malkhosyan, S., 291, 292 Man, S., 63 Manchester, J. K., 167, 193, 195(41) Maniatis, T., 101
Manickam, P., 221 Manollos, N., 321 Manz, D., 113 Marincola, F. M., 330 Markowitz, G. S., 49, 300 Maronpot, R. R., 20, 21(13), 22(13), 24(13), 53, 58(12), 102, 103(13), 110(14), 149, 198, 208, 261, 264(7), 300 Marsden, C. D., 260 Marsh, M. N., 170 Martinec, J., 63 Maruyama, K., 13, 22(2) Marx, C., 221 Marx, S. J., 221 Mase, K., 292 Mashal, R. D., 135 Masih, A. S., 202 Masuda, N., 203 Masutomi, N., 115, 118(17), 124 Mathur, E. J., 336 Matsuda, H., 318 Matsumoto, H., 249 Matsumoto, S., 170 Matsuoka, Y., 111 Matsushima, A. Y., 239 Matthews, P., 337 Matthys, P., 248 Maurer, R., 63 Maw, G., 197, 208, 217 Maxwell, T., 324, 345 Mayer, A., 25, 271 Mayer, G., 80 McCann, S. M., 220 McCarville, J., 322 McClain, K. L., 133, 135(19) McClelland, M., 290 McCracken, V. J., 170 McCutchen-Maloney, S. L., 44 McGee, J. O. D., 129 McGrath, J. A., 337 McKay, J. A., 63, 209 McKenna, G. J., 335 McLeod, H. L., 35, 39(7), 40(7), 41(7), 43(7), 63, 70(4), 71, 72(4), 113, 209 McMahon, A., 170 McMillan, L. J., 333 McNeal, J. E., 91, 318 McNeil, P., 81 McParland, B. E., 157, 259 Mead, R., 136
365
AUTHOR INDEX Medeiros, L. J., 235, 237(9), 238, 239(17) Meier-Ruge, W., 25, 81, 216 Melton, D. A., 170 Mendrinos, S., 44 Merchant, M., 43 Merke, D. P., 220 Merkelbach, S., 114 Metchette, K., 101 Middleton, B., 40 Middleton, L. A., 322 Midtvedt, T., 171 Mignon, A., 248 Mihatsch, M. J., 63 Milbrandt, J., 53, 59(10), 198 Miles, D. W., 63 Miller, G. J., 317 Miller, J. E., 333 Miller, L. D., 330 Mills, J., 190, 190(35) Mills, J. C., 296, 300(6) Minderer, S., 90, 297 Minkus, G., 81 Mirchandani, D., 317 Mirell, C., 129 Mitchell, D., 115, 124(16) Mitchell, R. S., 229, 248 Mittmann, M., 325, 328(18), 329(17; 18) Miyashiro, K., 266 Miyazaki, K., 50, 51(4), 114, 115(3), 117, 117(3), 118, 118(3), 119, 120, 120(3), 174, 198, 261, 262(8) Mizushima, M., 311, 319(16) Moberg, C., 336 Moch, H., 63 Moe, O. W., 104, 260, 261(5) Mok, S. C., 326 Momand, J., 335 Mombaerts, P., 170 Monaghan, A. P., 210 Monajembashi, S., 81 Monden, M., 203 Montesano, R., 335 Montironi, R., 310 Mooney, J. L., 333 Moore, D., 63 Mora, J., 240, 247, 324 Morita, R., 302(4), 303 Morley, A. A., 129 Morris, D., 321 Morton, D. L., 134, 135(26), 302, 302(2; 3), 303
Moskaluk, C. A., 100 Moss, A. A., 232 Motin, V. L., 44 Motomura, H., 132 Mountz, J. D., 319 Mous, J., 325 Muijen, G. N. P., 271 Muir, A. I., 333 Mulkens, J., 114 Muller, U., 52, 203 Munell, F., 145 Mungenast, A., 3 Murakami, H., 36, 104, 204, 260, 261(5), 300 Murase, T., 53, 103, 104(20), 105(20) Murdock, P. R., 333 Murray, G. I., 35, 39(7), 40(7), 41(7), 43(7), 63, 70(4), 71, 72(4), 113, 209 Murry, C. E., 136 Murthy, S. K., 63, 64, 65(12) Myers, R. P., 311 Myohanen, S., 344
N Nadon, N., 197, 208, 217 Nagaraju, K., 70, 74, 79(8) Nagasawa, Y., 111 Nagata, S., 270, 310, 311, 319, 319(16) Nagle, R. B., 264 Nagorny, R., 333 Nair, S., 266 Nakagawa, K., 115, 118(17), 124 Nakajima, T., 248 Nakamura, Y., 97, 310, 324 Nakano, H., 133 Nakayama, I., 133, 135(21) Nakayama, T., 134, 135(26), 302, 302(3), 303 Nakazato, Y., 249 Nanevicz, T., 325, 329(20) Nanno, M., 170 Nasim, S., 44, 46(22) Nasir, A., 324 Nathrath, W. B. J., 25(9), 26, 271 Natkunam, Y., 48, 160 Naven, T., 22, 24(15), 43, 70, 72(3), 79(3) N’Diaye, N., 221 Neira, M., 264, 282, 283, 285(1), 299, 301(20) Nelkin, B. D., 344 Neloan, S. N., 115
366
AUTHOR INDEX
Nelson, S. F., 300 Nelson, W. G., 91 Neoh, S., 129 Neoptolemos, J. P., 317, 321(27) Nesland, J. M., 290, 318 Neufeld, G., 248 Nevinny, C., 81 Nguyen, D.-H., 302(3), 303 Nichols, P. W., 136, 208, 217 Nigro, J. M., 310 Nisbet, A. P., 260 Nishimura, O., 333 Nister, M., 52, 109, 334 Nita, M. E., 97, 324 Nitsche, R., 25, 81, 216 Nocito, A., 63 Nogami, H., 249 Noguchi, M., 292 Noguchi, S., 132, 133 Nomarski, G., 80 Nonomura, N., 270, 309, 318, 319(36) Nouwens, A. S., 40 Novelli, M. R., 135 Nowell, P. C., 129 Nustad, K., 115
O Obana, S., 170 Obermaier, C., 40 Ochiai, T., 136 Odeberg, J., 336 Oesterling, J. E., 311 O’Farrel, P. H., 92 Ohnishi, T., 203 Ohnmacht, G. A., 330 Ohno, T., 103, 105(15) Ohyama, H., 97, 296, 299, 299(8), 300(8), 323, 324, 326(7; 8) Okubo, K., 203 Okuyama, A., 270, 318, 319(36) O’Leary, J. J., 129 Olinger, M. R., 25, 26(2) Olson, R. S., 216 Olsson, Y., 52, 109 Olufemi, S. E., 221 Omiecinski, C. J., 120 Onda, H., 333 Ongcapin, E. H., 133
Onho, T., 197, 199(1) Ono, K., 97, 324 Orita, M., 238 Orjasaeter, H., 115 Ornstein, D. K., 22, 24(16), 41, 43, 43(18), 44, 46(25), 48, 79, 113, 148, 162, 164, 191, 270, 324 Orstavik, T. B., 115 Ortiz-Pallardo, M. E., 114 Osamura, R. Y., 251 Oudraogo, G. L., 249
P P¨aa¨ bo, S., 335, 336(15) Pabst, O., 170 Paeslack, U., 58 Paget, S., 221 Painter, S., 311, 321(18) Palkovits, M., 216 Pallente, M., 324 Paller, A. S., 209 Palli, D., 314 Palma, G. A., 81 Pan, J., 262 Pappin, D., 22, 24(15), 33, 43, 70, 72(3), 79(3) Paradis, V., 129, 133, 135(24; 25) Park, G. Y., 311, 321(21) Park, H.-S., 135 Park, W. S., 311, 321(21) Parker, T., 74, 79(8) Partilla, J. S., 221 Paruch, L., 221 Paschke, R., 135 Pasic, R., 13, 18(1), 25(1) Passoneau, J. V., 192, 193(40) Pastan, I., 221, 292 Patel, A. C., 270 Patel, K., 44 Patel, V., 161, 324 Paterlini-Br´echot, P., 90 Patricelli, M. P., 191 Patterson, S. D., 92 Patton, W. F., 40, 191 Paul, C. L., 348 Pauls, K., 110 Paweletz, C. P., 22, 24(16), 41, 43, 43(18), 44, 46(25), 47, 48, 48(28), 70(5), 71, 79, 79(5), 113, 148, 162, 163, 164, 164(18), 191, 324
367
AUTHOR INDEX Pazzagli, M., 90 Pearsall, C. P., 310, 317(9) Peinado, M. A., 291, 292 Pennington, S. R., 40 Perrin, S., 131 Persson, A., 334, 335, 336, 337(17) Perucho, M., 291, 292 Peterson, J. I., 74, 105, 148, 215, 296 Peterson, L. A., 16(17), 24 Peterson, S., 81 Petricoin, E. F., 44, 46(25), 48, 148, 161, 162, 191, 221, 324 Petricoin, E. F. III, 22, 24(16), 41, 43, 43(18), 47, 48(28), 70(5; 6), 71, 79, 79(5), 113, 324 Pfeifer, G. P., 335 Phillips, S. M. A., 317, 321(27) Pi, J. H., 311, 321(21) Pierce, J. V., 115 Pineyro, M. A., 221 Pinkel, D., 63 Plotz, P., 74, 79(8) Plunkett, T. A., 63 Pognan, F., 40 Pohida, T., 13, 25, 74, 105, 132, 145, 148, 197, 215, 259, 296, 311 Poitras, P., 221 Poncin, J., 114 Ponder, B. A. J., 170, 171 Ponelies, N., 81 Pont´en, F., 334, 335, 336, 337(17) Pont´en, J., 335, 336, 337(17) P¨osl, H., 25(7; 9), 26, 81, 90, 100, 105, 271, 297 Posner, M., 97, 296, 299, 299(8), 300(8), 324, 326(7; 8) Poste, G., 221 Pozzatti, R. O., 310, 317, 317(8) Preisinger, A. C., 310 Primrose, J. N., 114 Prough, R. A., 120 Puchtler, H., 115 Puck, J. M., 322
Q Qi, R., 50 Qian, J., 309, 318, 319 Qian, X., 201, 204(17), 248, 249, 250, 250(23), 251, 251(23; 27), 271 Qiun, X., 177
Qu, N., 311, 321(18) Quackenbush, J., 50 Quan, D. E., 63 Quintanilla-Martinez, L., 125, 158, 196, 199, 200, 202(15), 203(15), 206(15), 207(15), 300 Quirke, P., 337 Qvist, H., 132, 290
R Rabilloud, T., 40 Rabinovitz, M., 102 Raddatz, R., 333 Rafalski, J. A., 291 Raffeld, M., 18, 21(9), 79, 103, 104(18), 105, 116, 125, 148, 161, 177, 199, 200, 202(15), 203(15), 204, 206(15), 207(15), 209, 243, 249, 262, 264(10), 293, 296, 300 Rajewsky, K., 100, 105, 197, 217 Ralfiaer, E., 311, 321(19) Ramalho-Santos, M., 170 Rasmussen, L. M., 103, 105(17) Rasmussen, R. P., 232 Rattner, J. B., 81 Ray, P. F., 337 Rayner, S., 40 Rechitsky, S., 337 Reddy, A. B., 120 Redfern, C. H., 325, 329(20) Redlich, C. A., 120 Redondo, T. C., 133 Redston, M., 106 Redwine, E. A., 318 Rees, J., 303 Rehman, I., 303 Reid, N., 290 Reifenberger, G., 247 Reinecke, A., 112 Remaley, A. T., 70(6), 71, 162, 324 Remy, E., 25, 81, 216 Ren, Z. P., 52, 109 Rennie, G. I., 333 Revent´os, J., 145 Rexer, C. H., 169(9), 170, 171(9), 173(9), 174(9), 177(9), 178(9), 201 Reznik, Y., 221 Rice, M., 314 Rich, W. E., 44, 46(25)
368
AUTHOR INDEX
Richter, J., 63 Richter, T., 52, 203, 205 Rigby, H., 303 Ringer, D. P., 208, 217 Ringer, P. D., 197 Ririe, K. M., 236 Rist, M., 63 Robberecht, W., 248 Robetorye, R. S., 229, 235, 237(9) Robinson, A., 335 Robinson, G., 20, 24(12), 324 Robinson, M., 40 Robsahm, T. E., 309 Roche, P. C., 249 Rodriguez-Tome, P., 335 Roe, B. A., 221 Roerl, M. H., 110 Roffler-Tarlov, S., 220 Rogers, K. E., 20, 22(11), 24(11), 113, 186, 247, 264, 266(11), 327 Rognum, T. O., 115 Rogol, A. D., 220 Rohlff, C., 165 Rohon, P. J., 74, 79(8) Rohr, L. R., 324, 345 Rohrbach, H., 322 R¨ohrl, M. H., 81, 109, 132 Romana, S., 90 Roninson, I. B., 292 Roop, D. R., 207, 208, 211, 211(5), 212(5; 6), 213 Rose, F., 113 Rosen, A., 74, 79(8) Rosenberg, F. J., 322 Ross, M. D., 49, 300 Roth, K. A., 296, 300(6) Roth, M. J., 47, 48, 48(28), 164 Rounds, D. E., 216 Rouse, R. V., 48 Rovera, G., 230 Rowlinson, R., 40 Roy-Burman, P., 317 R¨ubben, H., 85, 197 Rubbia-Brandt, L., 133, 135(25) Rubens, R. D., 63 Rubin, M. A., 222, 296, 300(5) Ruck, P., 205 Rudolph, J. A., 335 Ruebel, K. H., 248, 249, 252, 254, 254(26), 255(26)
Ruff, D. W., 52 Ruggeri, B. A., 114 Ruiter, D. J., 90, 247, 271 Rupp, G. M., 203 Ruschoff, J., 113 Ryan, M., 165, 260, 262(3)
S Saam, J. R., 169(9), 170, 171(9), 173(9), 174(9), 177(9), 178(9), 201 Sabbath-Solitare, M., 133 Sabile, A., 90 Sachinidis, A., 114 Safarians, S., 129 Sallstrom, J., 52, 109 Salomonis, N., 325, 329(20) Salunga, R. C., 20, 22(11), 24(11), 113, 165, 186, 247, 264, 266(11), 327 Sambrook, J., 101 Sameshima, A., 311, 319(16) Samuels, B., 63 Samura, O., 299 Sanchez, J. C., 33 Sandgren, E. P., 173 Sandherr, M., 202 Sandow, S., 81 Sasaki, Y., 133 Sato, Y., 50, 51(5) Satonaka, K., 335 Sauter, G., 63 Scardino, P. T. A., 318 Schaefer, F. V., 197, 208, 217 Scheer, M., 247 Scheithauer, B. W., 248 Schelhammer, P. F., 44, 46(22) Scherbaum, W. A., 20, 22(10), 24(10), 220 Schermelleh, L., 90, 105 Schindler, M., 25, 26, 26(2) Schmid, U., 63 Schmidt, G. H., 136, 170, 171 Schmidt, H., 110 Schneider, B. G., 103 Schnitzer, B., 324 Schonenberger, A., 63 Schraml, P., 63 Schulz, S., 135, 202 Schute, K., 4 Schutz, G., 170, 210
AUTHOR INDEX Sch¨utze, K., 25, 25(6; 7; 9), 26, 31(6), 70, 80, 81, 82, 85, 90, 90(20), 91, 92(15), 100, 105, 111(6), 114, 132, 158, 197, 221, 271, 280(1), 296, 297, 299, 334 Schwartz, S. M., 136 Schwarz, K., 202 Schweigerer, L., 248 Seebeck, J., 112 Seeger, S., 81 Seeger, W., 82, 103, 104, 106, 109(16), 110(29), 111, 113, 197, 299, 300 Segal, G. H., 202 Sekiya, T., 238 Sekizawa, A., 299 Selby, P. J., 22, 24(15), 33, 36, 43, 43(8), 48(8), 70, 72(3), 79(3) Senior, R. M., 59, 296, 300(10) Serth, J., 58 Seto, M., 236 Seto, Y., 311, 319(16) Sgroi, D. C., 20, 24(12), 324 Shah, N., 325, 329(20) Shane, S. S., 230 Sharkey, F., 317 Shaw, J., 40 Sheehan, C., 63 Sheehan, D., 241 Shen, C.-Y., 137 Shen, D. W., 292 Sherr, C. J., 70 Shi, S.-R., 18 Shibata, D., 208, 217, 292, 317 Shibata, K., 249 Shibata, M., 310 Shibutani, M., 50, 51(4), 102, 114, 115, 115(3), 117, 117(3), 118, 118(3; 17), 119, 120, 120(3), 124, 198, 261, 262(8) Shields, P. G., 114 Shillitoe, E., 324 Shimazaki, J., 317, 318 Shimogaki, H., 335 Shimosato, Y., 50, 51(5) Shin, M. S., 311, 318, 319(36), 321(21) Shiraishi, T., 317, 318 Shirasawa, N., 248, 249 Shiutani, M., 174 Shoemaker, D. D., 336 Short, J. M., 336 Shuang, Z., 156 Shun, C. T., 324
369
Sidransky, D., 310 Siebert, P. D., 113 Siemens, A., 81 Silberg, D. G., 170 Simon, J. A., 335 Simone, N. L., 14, 20(6), 22(6), 44, 46(25), 47, 48(28), 70(5; 6), 71, 79(5), 162, 164, 324, 329 Simoneau, A. R., 136 Simpson, J. F., 63 Sirivatanauksorn, V., 22, 24(14), 34, 289, 292 Sirivatanauksorn, Y., 22, 24(14), 34, 289, 292 Sitbon, G., 336 Sitruk, V., 90 Skjørten, F. J., 309 Slovak, M. L., 63 Smida, J., 100, 113(7) Smith, A., 333 Smith, H. S., 197 Smith, P. D., 3, 13, 22(4), 25, 34, 63, 74, 100, 105, 114, 132, 137, 148, 156, 197, 206(9), 207, 207(9), 215, 217, 219(12), 224, 240, 244(1), 249, 292, 295, 296, 296(4), 311, 334, 343 Snell, L., 100 S¨oderkvist, P., 336 Soler, D., 262 Sorbara, L., 199, 200, 202(15), 203(15), 206(15), 207(15) Sorensen, E. J., 191 Sorge, J. A., 336 Soriano, P., 171 Specht, K., 52, 100, 113(7), 196, 203 Speigel, A. M., 221 Speights, V. O., 324, 345 Spinage, L. D., 333 Spruck, C. H., 136, 208, 217 Spurr, N., 317, 321(27) Srinivas, S., 173 Srinivasan, R., 105 Stahl, U., 82, 106, 110, 110(29), 111, 197, 299 Stamey, T. A., 318 Standaert, D. G., 260, 262(3) Stanley, A., 22, 24(15), 43, 70, 72(3), 79(3) Stappenbeck, T. S., 167, 169(9), 170, 171(9), 173(9), 174(9), 177(9), 178(9), 201 Star, R. A., 36, 104, 204, 260, 261(5), 300 Stark, R., 81, 82, 90(20), 100, 297 Starkey, M., 260, 262(3) Starzinski-Powitz, A., 271
370
AUTHOR INDEX
Steger, K., 111 Stein, M. M., 104, 113, 300 Sternlicht, M., 129 Steven, K., 136 Stevens, J., 181 Stewart, F., 210 Stich, M., 25, 25(9), 26, 271 Stockton, P. S., 20, 21(13), 22(13), 24(13), 53, 58(12), 102, 103(13), 110(14), 149, 198, 208, 261, 264(7), 270, 300 Stolz, W., 80, 97 Stower, M., 290 Strahs, K., 81 Stratakis, C. A., 221, 222 Straten, P. T., 311, 321(19) Straume, T., 63 Straus, S. E., 322 Strausberg, R. L., 221 Stribley, J. A., 103, 105(15), 197, 199(1) Strober, W., 322 Strom, C., 337 Strup, S. E., 317 Suarez-Quian, C. A., 74, 105, 145, 147(2), 148, 215, 296 Suda, T., 310 Sugimura, H., 13, 22(2) Sugimura, T., 317 Sumners, C., 112 Sundstrom, C., 52, 109 Swalsky, P. A., 102 Swalwell, J. I., 221 Swanson, G., 324, 345 Swennen, L., 248 Syder, A. J., 209 Szekeres, P. G., 333 Szollosi, J., 63
T Taback, B., 134, 135(26), 302, 302(3), 303 Tadini, G., 209 Taillefer, R., 221 Takagi, T., 97, 324 Takahashi, K., 248 Takahashi, S., 318 Takahashi, T., 310 Takakuwa, T., 270 Takata, M., 302(4), 303 Takayama, H., 270, 309, 318, 319(36)
Takehara, K., 302(4), 303 Takenaka, M., 111 Takeuchi, T., 292 Tam, A. S., 134 Tanabe, Y., 173 Tanaka, T., 97, 324 Tanji, N., 49, 300 Tank, C., 205 Tao, M., 318 Tata, F., 170 Taubenberger, J. K., 203 Taylor, C. R., 18 Taylor, P., 22, 24(16), 41, 43(18), 44, 46(25), 48 Taymans, S. E., 222 Teng, S., 20, 24(12), 324 Tercanli, S., 297 Ternes, P., 109 Terry, M. S., 115 Thalhammer, S., 81, 82, 90, 90(20), 100, 105, 297 Theodorescu, D., 63 Thilly, W. G., 336 Thompson, C. A., 74, 79(8), 177, 201, 204(17), 249, 250(23), 251(23), 271 Thor, A. D., 197 Thornhill, A. R., 337 Thorstensen, L., 132, 290 Thulasiraman, V., 44 Tian, H., 220 Tien, Y. W., 324 Tillmanns, H., 110 Tindall, K. R., 336 Tingey, S. V., 291 Tirado, O. M., 145 To, M. D., 106 Todd, R., 97, 296, 299, 299(8), 300(8), 323, 324, 325, 326(7; 8) Tokuda, Y., 335 Tomita, M., 92 Tomlinson, I. P. M., 135 Tonegawa, S., 170 Tonge, R., 40 Topalian, S. L., 43, 79, 113, 191, 324 Torhorst, J., 63 Tornaletti, S., 335 Totty, N., 36, 43(8), 48(8) Tournay, O., 230 Toyoda, K., 50, 51(4), 114, 115(3), 117, 117(3), 118, 118(3), 119, 120, 120(3), 174, 198, 261, 262(8)
371
AUTHOR INDEX Traber, P. G., 170 Trainor, K. J., 129 Travert, G., 221 Travis, J. C., 14, 16(7), 18(7), 20(7), 22(7), 24(7) Tremblay, J., 221 Trempus, C. S., 20, 21(13), 22(13), 24(13), 53, 58(12), 102, 103(13), 110(14), 149, 198, 208, 261, 264(7), 300 Tretli, S., 309 Trier, J. S., 170 Troeger, C., 295, 296(2), 297, 299(2), 300(2), 301(2) Trojanowski, J. Q., 260 Tronche, F., 210 Troncoso, P., 310 Trulson, M., 329 Tsai, Y. C., 136 Tsujimoto, Y., 230, 236 Tsumanuma, I., 249, 252, 254, 254(26), 255(26) Tsunoda, T., 97, 324 Tsuzuki, T., 318 Turner, R., 134, 135(26), 302, 302(2), 303 Tyagi, S., 231 Tyson, F. L., 270
U Uhl´en, M., 334, 335, 336 Uhlmann, K., 220 Unabia, G., 248 Uneyama, C., 50, 51(4), 114, 115, 115(3), 117, 117(3), 118, 118(3; 17), 119, 120, 120(3), 124, 174, 198, 261, 262(8) Unger, T., 112 Unsold, R., 25, 81, 216 Urata, Y., 136 Ushijima, T., 317
Vanderschueren, B., 248 van Dyk, D., 81 Vankelecom, H., 248 van Muijen, G. N. P., 90, 247 van Oostrum, J., 40 van Tuinen, P., 310 Van Wyk, J. J., 220 Varvares, M., 299, 324, 326(8) Vasmatzis, G., 221 Vawter, L., 333 Vaysse, P. J., 333 Vekemans, M., 90 Velazquez, A., 291 Velculescu, V. F., 149 Verlann-de Vries, M., 280 Verlinsky, O., 337 Verlinsky, Y., 337 Verrills, N. M., 40 Vetter, H., 114 Vidaud, M., 133, 135(24; 25) Vierra, M. A., 262 Vigdor, C. G., 271 Villar, M., 248 Vinson, G. P., 220 Vlahou, A., 44, 46(22) Vocke, C. D., 43, 48, 79, 113, 191, 270, 310, 317, 317(8), 324 Vogelstein, B., 129, 149, 310 Vogt, T., 80, 97 Vona, G., 90 Vonderhaar, B. K., 222 von Eggeling, F., 44, 46, 46(26), 113, 160 von Wasielewski, R., 82, 90(19) Voshol, H., 40 Vranizan, K., 325, 329(20) Vreugdenhil, E., 150
W V Vahlquist, A., 303 van Boom, J. H., 280 Van Damme, J., 248 van De, R. M., 48 van den Berg, M. P., 150 van den Elst, H., 280 van der Eb, A. J., 280 van der Perk-deJong, J., 150
Wagner, U., 63 Wakatsuki, A., 318, 319(36) Walch, A., 52, 100, 113(7), 203 Waldman, F. M., 63 Waldrop, F. S., 115 Wallinger, S., 113 Walsh, B. J., 40 Walter, R., 81 Walters, R., 216 Wan, J. H., 129
372
AUTHOR INDEX
Wan, J. S., 20, 22(11), 24(11), 113, 186, 264, 266(11), 327 Wang, C., 325, 329(17) Wang, E., 330 Wang, H.-J., 302(2), 303 Wang, Q. H., 22, 24(16), 41, 43(18), 44, 46(25), 48 Wang, X. J., 208, 211, 211(5), 212(5; 6), 213 Wang, Y., 221 Wans, J. S., 247 Ward, J. M., 310 Warnke, R. A., 160 Warren, R. S., 52 Warrington, J. A., 299, 323, 324, 325, 326(8), 329, 329(20), 330(26), 332(26) Wasinger, V. C., 33 Watanabe, M., 317 Watanabe, T., 173 Watanabe-Fukunaga, R., 311 Waters, D., 310 Watson, M. A., 49, 50, 51, 53, 59, 59(10), 60, 198, 247, 296, 300(10) Watson, P. H., 100 Watson, R., 231 Waxman, D. J., 120 Webb, S. J., 120 Weber, G., 81 Weber, R. G., 247 Weber, T., 44, 46(25) Wedlich, D., 271 Weghorst, C. M., 314 Weiler, S., 270 Weinberger, S. R., 43 Weinstein, B., 280 Weirich, G., 135 Weise, M., 220 Weisemann, J., 221 Weiss, L. M., 133, 135(20) Weiss, P. D., 236 Weiss, R. A., 3, 13, 22(4), 25, 34, 100, 137, 156, 197, 207, 207(9), 217, 219(12), 224, 240, 244(1), 249, 292, 295, 296(4), 311, 334, 343 Weiss, W., 40 Weissenbach, J., 317, 321(27) Weissmann, N., 113 Wellmann, A., 80 Wellner, E., 74, 105, 148, 215, 296 Welsh, J., 290 Wener, J., 248
Werner, M., 52, 100, 113(7), 135, 202, 203 Wernert, N., 80 Wesley, C. S., 292 Westphal, G., 80 Wetmur, J., 235 Wheeler, T. M., 318 Whetsell, L., 197, 208, 217 White, A., 220 White, R., 310 White, W., 63 Wickert, R. S., 103, 105(15) Wieland, I., 85, 132, 197 Wilber, K., 63 Wild, P., 113 Wildgruber, R., 40 Wilkins, M. R., 33 Wilkinson, M. M., 170 Willenberg, H. S., 216, 221 William, C. M., 173 Williams, C., 334, 336, 337(17) Williams, J. G., 291 Williams, K. L., 33 Williams, P. M., 181 Williamson, J. A., 135 Williamson, R. C. N., 22, 24(14), 292 Williamson, V. M., 283 Willman, C. L., 133, 135(19) Wilson, A. C., 335, 336(15) Wilson, A. E., 333 Wilson, S., 333 Winton, D. J., 171 Wittliff, J. L., 12, 13, 14, 16(7), 18(1; 7), 20(7), 22(7), 24(7), 25(1) Wittwer, C. T., 231, 232, 232(5), 236, 237 Wodicka, L., 325, 328(18), 329(18) Wojno, K. J., 310, 317(9) Wolffe, A. P., 344 Wolfrum, J., 81 Wolfrum, T., 81 Wollan, P., 309 Wollscheid, V., 46, 80 Wong, D. T., 97, 296, 299, 299(8), 300(8), 323, 324, 325, 326(7; 8) Wong, K. K., 326 Wong, M. H., 167, 169(9), 170, 171(9), 173(9), 174(9), 177(9), 178(9), 201 Wood, M., 114 Woods, J., 102 Wright, G. L., 44, 46(22), 163
373
AUTHOR INDEX Wright, G. L., Jr., 44 Wright, N. A., 135, 170 Wu, G., 103, 105(15), 227, 231(2) Wu, H., 333 Wu, H. Q., 120 Wu, M. S., 324 Wurtman, R. J., 220
Yoshimura, F., 248 Young, J., 40 Yousem, S. A., 133, 135(20) Yu, Q. C., 209 Yung, C. M., 299
Z X Xiao, H., 20, 22(11), 24(11), 113, 186, 247, 264, 266(11), 327
Y Yagi, N., 335 Yamada, M., 230 Yamagishi, H., 136 Yamaguchi, H., 248 Yamamoto, H., 170 Yamano, T., 136 Yamasaki, T., 249 Yamashita, N., 249 Yan, J. X., 33 Yanagawa, R., 97, 324 Yancey, K. B., 264 Yang, H. K., 135 Yatani, R., 317, 318 Yeatman, T., 50 Yee, C., 264 Yegappan, S., 324 Yeh, H., 266 Yip, T.-T., 44, 46(22), 46(25), 92 Yokoyama, S., 133, 135(21) Yonehara, S., 311, 319(16) Yoo, N. J., 311, 321(21)
Zeigerdt, R., 170 Zeller, H., 110 Zettel, M., 266 Zeuthern, J., 311, 321(19) Zhang, L., 149 Zhang, S., 248, 251 Zhang, X., 97, 248, 296, 299, 299(8), 300(8), 324, 326(7; 8) Zhao, M., 100, 197, 217 Zhaung, Z., 311 Zhen, D. K., 299 Zheng, K., 270, 324 Zheng, Z., 317 Zhong, X. Y., 295, 296(2), 297, 299(2), 300(2), 301(2) Zhou, H., 114 Zhou, S., 333 Zhu, M., 112 Zhu, S., 48 Zhu, Y. S., 101 Zhu, Y. Y., 113 Zhuang, Z., 13, 22(4), 25, 34, 100, 114, 137, 197, 206(9), 207, 207(9), 217, 219(12), 221, 224, 240, 244(1), 249, 292, 295, 296(4), 310, 317(8), 334, 343 Zitzelsberger, H., 100, 113(7) Zlotnikov, G., 197 Zoller, J. E., 247 Zuang, Z., 3
This Page Intentionally Left Blank
Subject Index
A
nucleic acid extraction, see also Nucleic acid isolation, laser capture microdissection samples DNA, 108, 141–142, 246 DNase treatment, 247 overview, 59, 106–107 proteinase K digestion, 109, 141–142, 246 RNA, 22–24, 108–109, 142–143, 246 oral cancer, see Oral cancer prospects, 222–223 prostate cancer, see Prostate cancer proteomics applications CD34 detection, 160–161 esophageal cancer, 163–164 gel electrophoresis, see Two-dimensional polyacrylamide gel electrophoresis; Western blot head and neck squamous cell carcinoma, 161 overview, 24–25, 160–161, 221 prostate cancer, 162–163 SELDI, see Surface enhanced laser desorption ionization mass spectrometry rationale, 323–324 skin cancer, see Melanoma; Skin cancer specimen processing cryopreservation, 14–15 dehydration, 242 fixation, 18, 138–139, 241 frozen tissue section embedding and staining, 143–144, 242–243 hematoxylin and eosin staining, 16–17, 140–141, 241–242 immunostaining, 243–244 overview, 14–15, 137–138 paraffin embedding, 139 sectioning, 16, 139, 241 TO-PRO-3 staining, 16 tumor markers and conventional detection, 12–13 Clonal analysis human androgen receptor gene, 131, 133
Adrenal chromaffin cell, laser capture microdissection, 220 Annexin I, esophageal cancer proteomics, 163–164 AP-PCR, see Arbitrarily primed polymerase chain reaction Arbitrarily primed polymerase chain reaction advantages in polymorphism detection, 291 laser capture microdissection for DNA fingerprinting amplification reaction, 294 materials, 293 microdissection, 292–293 polynucleotude kinase reaction, 294 principles, 292–293 rationale, 289–290 variable number tandem repeat polymorphism, 290 principles, 290–291 tumorigenesis applications, 291–292
B Beam waist, resolution in laser microdissection, 6
C Cancer biopsy, laser capture microdissection advantages, 21–22, 240–241 clonal analysis, see Clonal analysis fluorescence in situ hybridization of nuclei, see Fluorescence in situ hybridization genomics applications, 24–25, 137, 221–222, 247 instrumentation components, 14 operation, 18–21 transfer techniques, 141 lymphoma, see Lymphoma 375
376
SUBJECT INDEX
laser capture microdissection gastric lymphoma, 135 Langerhan’s cell histiocytosis, 132–133 limitations, 135–136 liver cirrhosis, 133–134 melanoma, 134–135 rationale, 131–132 salivary gland adenoma, 133 Warthin’s tumor, 133 lymphoma, see Lymphoma melanoma capillary array electrophoresis of amplification products, 304, 308–309 DNA extraction, 304 laser capture microdissection micrometastases, 303 rationale, 302 technique, 304 loss of heterozygosity, 302–303, 305, 308 microsatellite polymerase chain reaction, 304–305 specimen preparation, 304 tissue proliferation diagnostics, 129, 131 X chromosome inactivation, 129, 131, 136 Caveolin-1, see DNA methylation
D DNA fingerprinting, see Arbitrarily primed polymerase chain reaction DNA isolation using laser capture microdissection, see Nucleic acid isolation, laser capture microdissection DNA methylation caveolin-1 promoter analysis in prostatic adenocarcinoma bisulfite treatment, 345, 348 controls, 349 data analysis, 349–351 genomic DNA isolation, 347–348 laser capture microdissection, 346–347, 351 polymerase chain reaction amplification reaction, 348–349 primer design, 348 sequencing of products, 349 slide preparation and staining, 345–346 p16 analysis from microdissected cells, 270 polymerase chain reaction techniques for analysis, 344–345
restriction endonuclease analysis, 344 tumor promoters, 343–344 DNA microarray intestinal RNA analysis from mouse first round amplification, 186, 188 internal references, 186 primers, 186 RNA cleanup, 189–190 second round amplification, 189–190 throughput, 186 melanoma transcriptome profiling array analysis, 99 laser pressure catapulting, 96–97 probe preparation and hybridization, 97, 99 RNA isolation, 97 specimen preparation, 97 oral cancer gene expression profiling amplification fidelity, 330 complementary DNA synthesis, 327–328 data analysis, 332–333 GeneChip system, 325, 328–329 microdissection, 327 RNA extraction, 327 hybridization of biotinylated complementary RNA, 328–329 yield following laser capture microdissection, 326, 329 sensitivity, 330 specimen preparation, 326–327 validation with reverse transcriptaseÑpolymerase chain reaction, 331–332 in vitro transcription, 328 RNA laser capture microdissection samples isolation, 22–24 linear amplification, 267 reverse transcription, 267 second round amplification, 268 in vitro transcription, 267–268 single-cell analysis, 299–301 Drug discovery, laser capture microdissection applications, 166
E Epidermolysis bullosa simplex, see Skin, laser capture microdissection from mouse Epidermolytic hyperkeratosis, see Skin, laser capture microdissection from mouse
377
SUBJECT INDEX
F Fas apoptosis signaling, 311 gene structure, 310 prostate cancer gene mutation analysis biopsy cases, 311 comparison of lesions, 319–321 DNA extraction, 312–313 DNA repair, 322 immunoreactivity relationship with mutations, 321 laser capture microdissection, 311–312 loss of heterozygosity, 321 missense mutations, 322 point mutations, 321–322 polymerase chain reaction and primers, 314, 319 FISH, see Fluorescence in situ hybridization Fluorescence in situ hybridization, laser capture microdissection-isolated nuclei cancer applications, 63–64 hybridization analysis, 65–67 laser capture microdissection, 64 nucleus isolation, 64–66 section thickness, 67–68 specimen preparation, 64 troubleshooting, 69 Folliculostellate cell cell culture of microdissected samples, 250 functions, 248 laser capture microdissection materials, 249 microdissection technique, 250 rationale for isolation, 248–249 reverse transcriptaseÑpolymerase chain reaction of microdissected samples amplification reaction, 251 gel electrophoresis of products, 252 primers and targets, 251 RNA isolation, 250 transcript levels for specific proteins, 252–255 transforming growth factor-β regulation of leptin expression, 255 Fructose-1,6-bisphosphatase, laser capture microdissection assay of mouse intestine cycling assay principles, 193–194 incubation conditions, 195 lysate preparation, 192–193
sectioning, 192 tissue preparation, 192 FS cell, see Folliculostellate cell
G Glucose, laser capture microdissection assay of mouse intestine, 195–196
H Hodgkin’s disease, see Lymphoma
I Immunoglobulin heavy chain clonal analysis, see Lymphoma Immunohistochemical staining cancer biopsy specimens, 243–244 cell identification for nucleic acid extraction, 262 Hodgkin’s disease samples frozen sections, 204–206 paraffin sections, 198–199 laser capture microdissection combination, 157–158 lymphoma samples, 226 nucleic acid isolation, laser capture microdissection samples, 104–105 p53 in skin cancer samples, 337, 339 Intestine, laser capture microdissection from mouse advantages of system, 167, 169 anatomy colon, 170 small intestine, 169–170 genetic mouse mosaic models, 171, 173 glucose assay, 195–196 intermediary metabolism enzyme assays cycling assay principles, 193–194 fructose-1,6-bisphosphatase assay, 194–195 lysate preparation, 192–193 sectioning, 192 tissue preparation, 192 laser capture microdissection caps, 178, 180 capture troubleshooting, 178, 180 environmental factors, 178 sample quality, 178
378
SUBJECT INDEX
mesenchyme enteric nervous system, 171 epithelial cell crosstalk, 170 gut-associated lymphoid tissue, 170 RNA analysis DNA microarray first round amplification, 186, 188 internal references, 186 primers, 186 RNA cleanup, 189–190 second round amplification, 189–190 throughput, 186 extraction, 180 real-time quantitative reverse transcriptaseÑpolymerase chain reaction complementary DNA synthesis, 181–182, 184 controls, 185 primer design, 182, 184 principles, 180–181 quantification of gene expression, 184–185 sample preparation freezing, 174–175 harvesting, 174–175 overview, 173–174 sectioning, 175–176 staining, 176–178 two-dimensional polyacrylamide gel electrophoresis, protein isolation, 190–191
L Laser capture microdissection advantages, 21–22, 24, 244–245 cancer biopsy analysis, see Cancer biopsy; specific cancers clonal analysis, see Clonal analysis developmental studies advantages, 147–148 assays, 148–151 disadvantages, 145–147 prospects, 156 rationale, 145 targeting decisions, 151–154 tissue preparation, 154–156 DNA methylation analysis, see DNA methylation
fluorescence in situ hybridization of nuclei, see Fluorescence in situ hybridization historical perspective, 216–217 instrumentation, see Leica AS LMD; PALM; PixCell limitations, 245 live cells, see Live cell laser microdissection mouse models, see Intestine, laser capture microdissection from mouse; Skin, laser capture microdissection from mouse nucleic acid isolation, see Nucleic acid isolation, laser capture microdissection pituitary, see Folliculostellate cell principles, 82, 216–219, 224, 244, 334 proteomics analysis, see also specific techniques advantages and limitations, 166–167 approaches, 158 diagnostics, 164–165 drug discovery, 166 neuroproteomics, 165 tissue proteomics, 166 taste bud, see Taste bud, gene discovery with laser capture microdissection ultraviolet laser microdissection, 217–219 Laser microbeam microdissection principles, 334 reverse transcriptaseÑpolymerase chain reaction sample preparation laser microbeam microdissection, 273, 275 laser pressure catapulting, 275–276 overview, 271–272 polymerase chain reaction, 278–280 real-time polymerase chain reaction, 280–281 reverse transcription, 277–278 RNA isolation hydration, 277 lysate preparation, 276 materials, 276 proteinÑDNA precipitation, 276 RNA precipitation, 276–277 specimen preparation cryopreserved tissues, 272–273 formalin-preserved tissues, 272 materials, 272 Laser pressure catapulting applications, 80, 90–91 collection of specimens, 87 combination with microdissection
SUBJECT INDEX contamination prevention, 85 glass-mounted specimens, 85 membrane-mounted specimens, 85–87 RNA analysis, see Reverse transcriptaseÑpolymerase chain reaction historical perspective, 80–82 instrumentation, see PALM live cells, see Live cell laser microdissection melanoma transcriptome profiling array analysis, 99 laser pressure catapulting, 96–97 probe preparation and hybridization, 97, 99 RNA isolation, 97 specimen preparation, 97 principles, 3–4, 83–84, 100 single cells neurons, 165 polymerase chain reaction of samples, 297–298 specimen requirements, 84 visualization improvements, 84–85 LCM, see Laser capture microdissection Leica AS LMD cell selection specificity, 8 comparison with other laser capture microdissection systems, 10–12 ease of use, 8–9 humidity effects, 9 integrity of dissected material, 8 principle of laser capture microdissection, 4 resolution, 4–6 sample preparation, 9 service, 10 stage movement control, 10 Live cell laser microdissection catapulted cells collection, 31 recultivation, 32 cell culture preparation chamber assembly, 27 contamination prevention, 30 culture, 28 laser microdissection and catapulting, 28–31 materials, 26–27 time requirements, 31 overview, 25–26 prospects, 32–33
379
LMM, see Laser microbeam microdissection Loss of heterozygosity, see Clonal analysis LPC, see Laser pressure catapulting Lymphoma antigen receptor gene features, 224–225, 239 chromosomal translocation characterization in microdissected samples fluorescence melting curve analysis of amplification products, 236–237 gel electrophoresis for product size determination, 236 melting temperature differences, 235 polymerase chain reaction, 233, 235 Hodgkin’s disease clonality analysis overview, 196–197 polymerase chain reaction of immunoglobulin heavy chain genes, 201–203 frozen sections immunohistochemical staining, 204–206 RNA extraction, 206 paraffin-embedded tissues DNA extraction, 201–203 hematoxylin and eosin staining, 198 immunohistochemical staining, 198–199 microdissection troubleshooting, 199, 201 overview, 197 real-time TaqMan reverse transcriptaseÑpolymerase chain reaction, 203–204 RNA extraction, 203 immunoglobulin heavy chain clonal analysis in microdissected samples clone-specific primers, 231 DNA extraction, 226–228 gel electrophoresis of amplification products, 229 internal probes, 231–232, 235 polymerase chain reaction, 228–229 purification of amplification products, 230 sequencing, 230–231 microdissection of samples, 226 single-strand conformation polymorphism analysis of p53 in microdissected samples gel electrophoresis, 239 overview, 237–238 polymerase chain reaction, 239
380
SUBJECT INDEX
tissue preparation for laser capture microdissection frozen tissue sections embedded with OCT, 225 immunohistochemical staining, 226 paraffin-embedded tissues, 225
rationale, 114–115, 124–125 Western blot analysis, 122–123 Muscle biopsy proteomics, see Western blot, laser capture microdissection for sample collection
N M Mass spectrometry, see Surface enhanced laser desorption ionization mass spectrometry Melanoma clonal analysis capillary array electrophoresis of amplification products, 304, 308–309 DNA extraction, 304 laser capture microdissection micrometastases, 303 rationale, 302 technique, 304 loss of heterozygosity, 302–303, 305, 308 microsatellite polymerase chain reaction, 304–305 specimen preparation, 304 DNA microarray for transcriptome profiling array analysis, 99 laser pressure catapulting, 96–97 probe preparation and hybridization, 97, 99 RNA isolation, 97 specimen preparation, 97 recurrence and metastasis, 305 Metharcin, paraffin-embedded tissue fixation extraction and yields DNA, 117–118 protein, 117–118 RNA, 117–118 fixation conditions, 115–116 microdissection of fixed samples, 116 polymerase chain reaction genomic DNA amplification, 123–124 reverse transcriptionÑpolymerase chain reaction amplification reactions, 119 competitive reverse transcriptionÑpolymerase chain reaction and plate hybridization, 121–122 reverse transcription, 118–119 RNA quality control, 119–120 validation, 121
NA, see Numerical aperture Nucleic acid isolation, laser capture microdissection samples applications, 62, 99–100, 259 cell identification, 262 contamination prevention, 100–101, 260 critical parameters, 49–50, 101 DNA analysis, see specific techniques isolation, 108, 141–142, 268–269 proteinase K digestion, 109, 141–142 extraction overview, 59, 106–107 Hodgkin’s disease sections DNA extraction, 201–203 RNA extraction, 203–204, 206 intestine studies, see Intestine, laser capture microdissection from mouse melanoma transcriptome profiling array analysis, 99 laser pressure catapulting, 96–97 probe preparation and hybridization, 97, 99 RNA isolation, 97 specimen preparation, 97 metharcin fixation of paraffin-embedded tissues, see Metharcin microdissection and laser pressure catapulting, 105–106, 262–263 polymerase chain reaction applications, 113 complementary DNA synthesis via reverse transcription, 109–110 internal controls, 112–113 qualitative analysis, 110–111 quantitative analysis, 111–112 RNA analysis, see DNA microarray; Reverse transcriptionÑpolymerase chain reaction degradation prevention, 259–260 developmental studies applications, 148 CapSure caps, 151
381
SUBJECT INDEX epithelial samples, 154–154 prospects, 156 RNA isolation, 149–151 single cell targeting, 152–153 tissue preparation, 154–156 extraction, 108–109, 142–143, 276–277 integrity checking, 260 isolation from cancer biopsy, 22–24, 203–204, 206 isolation kits, 264–265 specimen processing ethanol fixation, 54–55 fixation, 50–52, 101–102, 260–261 formalin-fixed paraffin-embedded samples, 102–103 optimization, 101–102 rapidity importance, 50–51 sectioning frozen tissue, 56, 103, 260 paraffin-embedded tissue, 55, 103 snap freezing, 50, 53–54, 101, 260 staining effects on recovery, 52–53 frozen tissue, 57–59, 103–105, 260–261 hematoxylin and eosin staining, 260 immunohistochemistry, 104–105 paraffin-embedded tissue, 56–57, 103–105 types of stains, 261–262 storage of microdissected cells, 263–264 troubleshooting film transfer, 61 histological detail, 60–61 recovery, 61–62 staining, 61 Numerical aperture, resolution in laser microdissection, 4–5
O Oral cancer DNA microarray gene expression profiling amplification fidelity, 330 complementary DNA synthesis, 327–328 data analysis, 332–333 GeneChip system, 325, 328–329 microdissection, 327 RNA extraction, 327 hybridization of biotinylated complementary RNA, 328–329
yield following laser capture microdissection, 326, 329 sensitivity, 330 specimen preparation, 326–327 validation with reverse transcriptaseÑpolymerase chain reaction, 331–332 in vitro transcription, 328 laser capture microdissection rationale, 323–324
P p16, methylation analysis from microdissected cells, 270 p53 gene structure and mutations in cancer, 310 prostate cancer gene mutation analysis biopsy cases, 311 comparison of lesions, 316–317, 319 DNA extraction, 312–313 laser capture microdissection, 311–312 polymerase chain reaction primers, 314 single-strand conformation polymorphism analysis, 314 stage effects, 318 types of mutations, 314, 316–317 zone distribution, 317–319 single-strand conformation polymorphism analysis, lymphoma microdissected samples gel electrophoresis, 239 overview, 237–238 polymerase chain reaction, 239 skin cancer, laser microbeam microdissection for single-cell analysis microdissection, 339–340 polymerase chain reaction allele dropout rate, 336–337 inner nested amplification, 341–342 mitochondrial control sequence, 336 outer multiplex amplification, 340–341 sample preparation overview, 335–336 p53 immunostaining, 337, 339 sectioning, 337 sequencing of p53 amplification products, 337, 342–343
382
SUBJECT INDEX
PALM cell selection specificity, 8 comparison with other laser capture microdissection systems, 10–12 ease of use, 8–9 humidity effects, 9 integrity of dissected material, 8 laser, 82 laser pressure catapulting processing, 3–4 resolution, 4–6 RoboSoftware for automated microdissection and catapulting, 82, 87, 89 sample preparation, 9 service, 10 stage movement control, 10, 82 PCR, see Polymerase chain reaction Pituitary, see Folliculostellate cell PixCell cell selection specificity, 7 comparison with other laser capture microdissection systems, 10–12 components, 14 development, 13, 219 ease of use, 8–9 humidity effects, 9 integrity of dissected material, 8 operation, 18–21 principle of laser capture microscopy, 3 resolution, 4–6 sample preparation, 9 service, 10 stage movement control, 10 Polymerase chain reaction, laser capture microdissection samples applications, 113 capillary array electrophoresis of amplification products, 304, 308–309 DNA fingerprinting, see Arbitrarily primed polymerase chain reaction internal controls, 112–113 lymphoma clonality analysis, see Lymphoma melanoma microsatellite polymerase chain reaction, 304–305 metharcin-fixed, paraffin-embedded tissues and genomic DNA amplification, 123–124 methylation analysis of promoters amplification reaction, 348–349 primer design, 348 sequencing of products, 349
p53 mutation analysis in single skin cancer cells allele dropout rate, 336–337 inner nested amplification, 341–342 microdissection, 339–340 mitochondrial control sequence, 336 outer multiplex amplification, 340–341 sample preparation overview, 335–336 p53 immunostaining, 337, 339 sectioning, 337 sequencing of p53 amplification products, 337, 342–343 qualitative analysis, 110–111 quantitative analysis, 111–112 reverse transcriptionÑpolymerase chain reaction, see Reverse transcriptionÑpolymerase chain reaction, laser capture microdissection samples single-cell polymerase chain reaction laser pressure catapulting, 297–298 microdissection comparison of techniques, 297–298 limitations, 298 optimization, 295–296 micromanipulation comparison with microdissection techniques, 297–298 preimplantation genetic diagnosis, 295 staining considerations, 298–299 Promoter methylation, see DNA methylation Prostate cancer caveolin-1 promoter methylation analysis in adenocarcinoma bisulfite treatment, 345, 348 controls, 349 data analysis, 349–351 genomic DNA isolation, 347–348 laser capture microdissection, 346–347, 351 polymerase chain reaction amplification reaction, 348–349 primer design, 348 sequencing of products, 349 slide preparation and staining, 345–346 classification, 309–310 early diagnosis importance, 91 gene mutation analysis biopsy cases, 311 DNA extraction, 312–313 Fas
SUBJECT INDEX comparison of lesions, 319–321 DNA repair, 322 immunoreactivity relationship with mutations, 321 loss of heterozygosity, 321 missense mutations, 322 point mutations, 321–322 polymerase chain reaction and primers, 314, 319 laser capture microdissection, 311–312 p53 comparison of lesions, 316–317, 319 polymerase chain reaction primers, 314 single-strand conformation polymorphism analysis, 314 stage effects, 318 types of mutations, 314, 316–317 zone distribution, 317–319 loss of heterozygosity and mutation analysis, 269–270 precursor lesions, 309–310 surface enhanced laser desorption ionization mass spectrometry of microdissected samples biomarkers, 162–163 data analysis, 95–96 laser capture microdissection, 93 sensitivity, 93, 95 specimen preparation, 93 two-dimensional polyacrylamide gel electrophoresis analysis, 162 Western blot of prostate-specific antigen, 162 ProteinChip, see Surface enhanced laser desorption ionization mass spectrometry
R Reverse transcriptionÑpolymerase chain reaction, laser capture microdissection samples complementary DNA synthesis via reverse transcription, 109–110, 267–268 folliculostellate cell microdissected samples amplification reaction, 251 gel electrophoresis of products, 252 primers and targets, 251 RNA isolation, 250 transcript levels for specific proteins, 252–255
383
transforming growth factor-β regulation of leptin expression, 255 Hodgkin’s disease, real-time TaqMan reverse transcriptaseÑpolymerase chain reaction, 203–204 intestinal RNA from mouse, real-time quantitative reverse transcriptaseÑpolymerase chain reaction complementary DNA synthesis, 181–182, 184 controls, 186 primer design, 182, 184 principles, 180–181 quantification of gene expression, 184–185 laser microbeam microdissection with laser pressure catapulting for sample preparation laser microbeam microdissection, 273, 275 laser pressure catapulting, 275–276 overview, 271–272 polymerase chain reaction, 278–280 real-time polymerase chain reaction, 280–281 reverse transcription, 277–278 RNA isolation hydration, 277 lysate preparation, 276 materials, 276 proteinÑDNA precipitation, 276 RNA precipitation, 276–277 specimen preparation cryopreserved tissues, 272–273 formalin-preserved tissues, 272 materials, 272 metharcin-fixed, paraffin-embedded tissues amplification reactions, 119 competitive reverse transcriptionÑpolymerase chain reaction and plate hybridization, 121–122 reverse transcription, 118–119 RNA quality control, 119–120 validation, 121 real-time TaqMan system, 203–204, 265–266 sensitivity, 266 single-cell analysis, 299–301 taste bud gene discovery, 285–288 RNA isolation using laser capture microdissection, see Nucleic acid isolation, laser capture microdissection
384
SUBJECT INDEX
RTÑPCR, see Reverse transcriptionÑpolymerase chain reaction
S SELDI, see Surface enhanced laser desorption ionization mass spectrometry Single-strand conformation polymorphism, p53 analysis in microdissected samples lymphoma gel electrophoresis, 239 overview, 237–238 polymerase chain reaction, 239 prostate cancer, 314 Skin, laser capture microdissection from mouse epidermolysis bullosa simplex and epidermolytic hyperkeratosis models DNA extraction, 215 overview, 207–209 phenotypes, 211 RU486 induction of mutant keratin, 210 stem cell characterization, 211–212 microdissection advantages and limitations, 209 technique, 208, 212, 215 tissue preparation fixation, 212 paraffin embedding and sectioning, 212 rationale, 207 Skin cancer, see also Melanoma laser microbeam microdissection for single-cell analysis microdissection, 339–340 polymerase chain reaction allele dropout rate, 336–337 inner nested amplification, 341–342 mitochondrial control sequence, 336 outer multiplex amplification, 340–341 sample preparation overview, 335–336 p53 immunostaining, 337, 339 sectioning, 337 sequencing of p53 amplification products, 337, 342–343 p53 mutations, 334–335 SSCP, see Single-strand conformation polymorphism Surface enhanced laser desorption ionization mass spectrometry
advantages in proteomics analysis, 92–93, 160 extract preparation, 44 laser capture microdissection for sample collection applications, 33–34, 46–48 prostate cancer biopsy biomarkers, 162–163 data analysis, 95–96 early diagnosis importance, 91 laser capture microdissection, 93 sensitivity, 93, 95 specimen preparation, 93 sample preparation, 38–39 tissue sample processing collection, 36 considerations for staining, 35–36 hematoxylin and eosin staining, 37 laser capture microdissection, 38 materials, 36 sectioning, 35 silver-enhanced gold immunolabeling, 37–38 slides, 35 overview, 43–44 protein estimation, 46 ProteinChip types and sample preparation H4, 45 IMAC3, 45–46 NP2, 44–45 SAX2, 45 WCX2, 45
T Taste bud, gene discovery with laser capture microdissection differential screening strategy, 282–283, 288–289 library construction, 288 microdissection, 285 rationale, 282 reagents and solutions, 283–285 reverse transcriptaseÑpolymerase chain reaction, 285–288 RNA isolation, 285 Two-dimensional polyacrylamide gel electrophoresis denaturing gel electrophoresis, 41 esophageal cancer proteomics, 163 isoelectric focusing, 41
385
SUBJECT INDEX laser capture microdissection for sample collection applications, 33–34, 42–43, 48 intestinal proteins from mouse, 190–191 prostate cancer, 162 rationale, 159 sample preparation, 38–39 tissue sample processing collection, 36 considerations for staining, 35–36 hematoxylin and eosin staining, 37 laser capture microdissection, 38 materials, 36 sectioning, 35 silver-enhanced gold immunolabeling, 37–38 slides, 35 materials, 39–40 overview, 40 protein estimation, 41 silver staining, 42 solubilization of samples, 41
W Western blot, laser capture microdissection for sample collection applications, 33–34, 48–49, 79 developmental studies applications, 148–149 CapSure caps, 151 epithelial samples, 154–154 prospects, 156 single cell targeting, 152–153 tissue preparation, 154–155 esophageal cancer proteomics, 163–164
metharcin-fixed, paraffin-embedded tissues, 122–123 muscle cell sample processing abnormal cell analysis in myositis, 79 advantages, 71 blotting, 77 contamination prevention, 73–74 freezing, 71–72 gel electrophoresis, 76–77 immunoblotting, 77–78 laser capture microdissection, 73–74 lysate preparation, 75–76 reagents, 71 sectioning, 72 staining, 72–73 prostate-specific antigen, 162 rationale, 70 tumor tissue sample processing collection, 36 considerations for staining, 35–36 electrophoresis and blotting, 47–48 hematoxylin and eosin staining, 37 laser capture microdissection, 38 materials, 36 protein estimation, 47 sample preparation, 38–39, 47 sectioning, 35 silver-enhanced gold immunolabeling, 37–38 slides, 35
X X chromosome, inactivation and clonal analysis, 129, 131, 136
This Page Intentionally Left Blank
