Chromosome Structural Analysis: A Practical Approach

Chromosome Structural Analysis

The Practical Approach Series SERIES EDITOR B. D. HAMES Department of Biochemistry and Molecular Biology University of Leeds, Leeds LS2 9JT, UK

See also the Practical Approach web site at http://www.oup.co.uk/PAS * indicates new and forthcoming titles

Affinity Chromatography Affinity Separations Anaerobic Microbiology Animal Cell Culture (2nd edition) Animal Virus Pathogenesis Antibodies I and II Antibody Engineering Antisense Technology Applied Microbial Physiology Basic Cell Culture Behavioural Neuroscience Bioenergetics Biological Data Analysis Biomechanics - Materials Biomechanics - Structures and Systems Biosensors Carbohydrate Analysis (2nd edition) Cell-Cell Interactions The Cell Cycle Cell Growth and Apoptosis

if Cell Separation Cellular Calcium Cellular Interactions in Development Cellular Neurobiology * Chromatin if Chromosome Structural Analysis Clinical Immunology Complement if Crystallization of Nucleic Acids and Proteins (2nd edition) Cytokines (2nd edition) The Cytoskeleton Diagnostic Molecular Pathology I and II DNA and Protein Sequence Analysis DNA Cloning 1: Core Techniques (2nd edition) DNA Cloning 2: Expression Systems (2nd edition) DNA Cloning 3: Complex Genomes (2nd edition)

*

if

if if

*

DNA Cloning 4: Mammalian Systems (2nd edition) Drosophila (2nd edition) Electron Microscopy in Biology Electron Microscopy in Molecular Biology Electrophysiology Enzyme Assays Epithelial Cell Culture Essential Developmental Biology Essential Molecular Biology I and I Eukaryotic DNA Replication Experimental Neuroanatomy Extracellular Matrix Flow Cytometry (2nd edition) Free Radicals Gas Chromatography Gel Electrophoresis of Nucleic Acids (2nd edition) Gel Electrophoresis of Proteins (3rd edition) Gene Probes 1 and 2 Gene Targeting Gene Transcription Genome Mapping Glycobiology Growth Factors and Receptors Haemopoiesis Histocompatibility Testing HIV Volumes 1 and 2 HPLC of Macromolecules (2nd edition) Human Cytogenetics I and II (2nd edition)

Human Genetic Disease Analysis * Immobilized Biomolecules in Analysis Immunochemistry 1 Immunochemistry 2 Immunocytochemistry * In Situ Hybridization (2nd edition) lodinated Density Gradient Media Ion Channels * Light Microscopy (2nd edition) Lipid Modification of Proteins Lipoprotein Analysis Liposomes Mammalian Cell Biotechnology Medical Parasitology Medical Virology MHC Volumes 1 and 2 if Molecular Genetic Analysis of Populations (2nd edition) Molecular Genetics of Yeast Molecular Imaging in Neuroscience Molecular Neurobiology Molecular Plant Pathology I and II Molecular Virology Monitoring Neuronal Activity Mutagenicity Testing * Mutation Detection Neural Cell Culture Neural Transplantation Neurochemistry (2nd edition)

Neuronal Cell Lines NMR of Biological Macromolecules Non-isotopic Methods in Molecular Biology Nucleic Acid Hybridisation Oligonucleotides and Analogues Oligonucleotide Synthesis PCR1 PCR2 *PCR3:PCRInSitu Hybridization Peptide Antigens Photosynthesis: Energy Transduction Plant Cell Biology Plant Cell Culture (2nd edition) Plant Molecular Biology Plasmids (2nd edition) Platelets Postimplantation Mammalian Embryos Preparative Centrifugation Protein Blotting

Protein Expression Vol 1 * Protein Expression Vol 2 Protein Engineering Protein Function (2nd edition) Protein Phosphorylation Protein Purification Applications Protein Purification Methods Protein Sequencing Protein Structure (2nd edition) Protein Structure Prediction Protein Targeting Proteolytic Enzymes Pulsed Field Gel Electrophoresis RNA Processing I and II RNA-Protein Interactions Signalling by Inositides Subcellular Fractionation Signal Transduction * Transcription Factors (2nd edition) Tumour Immunobiology

Chromosome Structural Analysis A Practical Approach Edited by

WENDY A. BICKMORE Cell Genetics Section, MRC Human Genetics Unit, Western General Hospital, Edinburgh

OXFORD UNIVERSITY PRESS

1999

OXFORD UNIVERSITY PRESS

Great Clarendon Street, Oxford OX2 6DP Oxford University Press is a department of the University of Oxford and furthers the University's aim of excellence in research, scholarship, and education by publishing worldwide in Oxford New York Athens Auckland Bangkok Bogota Buenos Aires Calcutta Cape Town Chennai Dar es Salaam Delhi Florence Hong Kong Istanbul Karachi Kuala Lumpur Madrid Melbourne Mexico City Mumbai Nairobi Paris Sao Paulo Singapore Taipei Tokyo Toronto Warsaw and associated companies in Berlin Ibadan Oxford is a registered trade mark of Oxford University Press Published in the United States by Oxford University Press Inc., New York © Oxford University Press 1999 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Oxford University Press. Within the UK, exceptions are allowed in respect of any fair dealing for the purpose of research or private study, or criticism or review, as permitted under the Copyright, Designs and Patents Act, 1988, or in the case of reprographic reproduction in accordance with the terms of licenses issued by the Copyright Licensing Agency. Enquiries concerning reproduction outside those terms and in other countries should be sent to the Rights Department, Oxford University Press, at the address above. This book is sold subject to the condition that it shall not, by way of trade or otherwise, be lent, re-sold, hired out, or otherwise circulated without the publisher's prior consent in any form of binding or cover other than that in which it is published and without a similar condition including this condition being imposed on the subsequent purchaser Users of books in the Practical Approach Series are advised that prudent laboratory safety procedures should be followed at all times. Oxford University Press makes no representation, express or implied, in respect of the accuracy of the material set forth in books in this series and cannot accept any legal responsibility or liability for any errors or omissions that may be made. A catalogue record for this book is available from the British Library Library of Congress Cataloging in Publication Data (Data available) ISBN 0-19-963699-0 (Hbk) 0-19-963698-2 (Pbk) Typeset by Footnote Graphics, Warminster, Wilts Printed in Great Britain by Information Press, Ltd, Eynsham, Oxon.

Preface The transcription of genes and the duplication and segregation of the eukaryotic genome occurs within the context of chromatin and chromosomes. Chromosomes are macromolecular complexes of the primary DNA sequence complexed with protein, and perhaps also with RNA. The formation of chromatin occurs through a series of hierarchical interactions with proteins: starting with the interaction with the histone octamer to form nucleosomes and finishing with the chromosome in its fully condensed form ready for mitosis. By this stage a 10000-fold linear compaction of the DNA has been achieved. It is this very hierarchy of packaging that has made the analysis of chromatin and chromosome structure so difficult because of the problems in dissecting out any one particular layer of packaging without destroying or disrupting preceding levels. The history of research into the regulation of transcription well illustrates the progression of thinking about how DNA/protein interactions within the chromosome occur and what their roles are. Prokaryotic paradigms for transcriptional regulators are often single proteins, or multimers of a single protein, binding to a specific recognition sequence and influencing the activity of RNA polymerase directly. The study of single transcription factors binding to specific recognition motifs also has a long track record in studies of gene regulation in eukaryotes. Nucleosomes then became seen as a mere nuisance and hindrance to transcription factor-access to the DNA sequence. Chromatin became a bit more exciting when it was realized that modifications of histones, for example by acetylation, were important in regulating transcriptional potential and in propagating chromatin memory from cell to cell. The enzymes responsible for these modifications have now been found to be transcriptional regulators and to be part of large complexes of proteins that include other transcriptional activators or repressers. The complexes seem to be getting bigger and bigger. Large megadalton complexes are now known to be responsible for remodelling chromatin in the cell. In the most repressed parts of the genome (heterochromatin) it was soon apparent that both modifications of histones and the presence of large multiprotein complexes were at the heart of the genetic inertness of these regions of the genome. This extreme form of chromatin structure may also have an important role to play in maintaining the integrity of the chromosome itself. Advances in visual techniques for analysing the genome have added another layer of complexity to the problem. It is clear that in the interphase nucleus, chromosomes do not decondense into the tangled mess of spaghetti depicted in some textbook diagrams. Rather each chromosome maintains a distinctive identity (territory). Moreover, the silencing of some genes appears to be intimately linked with their spatial sequestration into discrete compartments

Preface of the nuclear volume—often those parts of the nucleus occupied by visible heterochromatin. To reflect this complexity of structure and control this Practical Approach volume seeks to bring together techniques from a variety of eukaryotes that all aim to study complex levels of chromatin and/or chromosome structure in vitro, ex vivo, and in vivo. The opening chapter from Donald MacLeod describes procedures that can be used to define specific sites of protein interaction (such as transcription factor binding sites or the position of nucleosomes) on DNA sequences within chromatin, both in vitro and in vivo, taking advantage of ligation-mediated PCR. Renato Paro and his colleagues and Janet Partridge and Karl Ekwall, have also explored ways of looking at specific interactions between chromatin proteins and DNA in vivo, by formaldehyde cross-linking of living cells (from Drosophila embryos or from fission yeast cells) and followed by chromatin immunoprecipitation, respectively. The chapter from Partridge and Ekwall also introduces us to visual analysis of chromosome structure using fluorescence in situ hybridization (FISH)—in this case in a yeast. Even in a eukaryote as small as Schizosaccharomyces pombe FISH has had a profound impact on our understanding of chromosome structure and biology. The chapter from Jeff Craig describes FISH in its more familiar form—as a tool to study the large and distinctive metaphase chromosomes of vertebrates. Special emphasis is placed here on the ways in which FISH can explore questions of chromosome structure, both by the nature of the probes that can be used and by the biochemical manipulation of the chromosomes to be used as hybridization substrates. Beth Sullivan develops this theme by showing how the combination of FISH to detect DNA sequence, immunofluorescence to detect specific proteins, and drugs to interfere with chromosome segregation and cytokinesis assist in our understanding of a key step in chromosome biology—the accurate segregation of the genetic material to daughter cells by mitosis. Analysis of mitotic chromosomes has been possible for so long because of each chromosome's compact structure and distinctive shape. Individual chromosomes in interphase cells cannot be distinguished by simple DNA stains alone. Hence to study interphase chromosome structure, individual chromosomes must be delineated with specific paints and probes. Joanna Bridger and Peter Lichter describe how FISH in combination with immunofluorescence can be used to analyse chromatin and chromosome structure in the nuclei of vertebrate cultured cells that have been treated to preserve as much as possible of their three-dimensional structure. Biochemical manipulation of the interphase nucleus prior to visual analysis is also described. Abby Dernburg takes us one step further to the real animal by describing ways in which FISH can be performed in whole-mount tissues of nonvertebrate animals. A delicate balance must be maintained here between the viii

Preface preservation of structure, the permeability of the cells/nuclei to probe, and the necessary denaturation of target chromosomal DNA. These last five chapters also all describe the benefits and limitations of the image acquisition and analysis systems that are necessary to do justice to the data produced from these fluorescence-based approaches. The preservation of subcellular and chromosomal architecture is a problem demanding compromises in FISH and immunofluorescence studies in whole cells and tissues. There are additional problems in studying complex underlying nuclear structures using classical biochemical approaches. Dean Jackson details the principal methods through which attempts have been made to study nuclear substructures using a variety of biochemical extraction procedures. Continuing the biochemical theme, Jason Swedlow presents techniques for generating and studying mitotic chromosomes in an in vitro system developed from Xenopus eggs. This system is proving its worth in identifying components important for higher order chromosome structure. Methods for the immunodepletion of specific factors from these extracts and also the immunolocalization of specific proteins on the resulting chromosomes are presented. The final chapter, from Christine Farr, describes a truly top-down approach to studying chromosome structure through the genetic manipulation and fragmentation of existing mammalian chromosomes. It is my belief, and I am sure that of the authors as well, that the methods presented here will be of continuing and increasing importance for studies of chromosome structure and function and gene regulation in eukaryotes, and that the combined forces of genetics, biochemistry, and cytology will augment and complement each other in this field to an increasing extent. Edinburgh May 1998

W.A.B.

iX

This page intentionally left blank

Contents List of contributors Abbreviations 1. Mapping protein/DNA interactions in vivo using ligation-mediated polymerase chain reaction Donald Macleod 1. Introduction Methods used in determining protein/DNA interactions in vitro Determining protein/DNA interactions in vivo

2. The ligation-mediated polymerase chain reaction (LMPCR) Applications of LMPCR

3. Mapping protein factor binding sites in vivo with LMPCR and DMS DNA modification by DMS in vitro DMS modification of DNA in vivo Amplification of DMS/piperidine-cleaved DNA by LMPCR Analysis of LMPCR reactions

4. Mapping nucleosomes using micrococcal nuclease and LMPCR Isolation of nuclei from cultured cells Preparation of DNA from nuclei treated with MNase Cleavage of genomic DNA with MNase

Acknowledgements References

2. Mapping DNA target sites of chromatinassociated proteins by formaldehyde cross-linking in Drosophila embryos Giacomo Cavalli, Valeria Orlando, and Renato Paro 1. Introduction 2. Outline of the method

1 1 1 2

3 4

5 5 1 7 9

13 13 14 16

18 18

21 21 22

Contents 3. Formaldehyde cross-linking in staged Drosophila embryos Preparation of fly cages and collection of staged embryos Optimizing cross-linking conditions

4. Immunoprecipitation of cross-linked embryonic chromatin and PCR amplification of the immunoprecipitated DNA 5. Analysing the enrichment of putative target sequences in the PCR-amplified DNA Slot-blot analysis of the enrichment of putative PC target sequences Mapping DNA target sites for Polycomb and GAGA factor in the Drosophila bithorax complex

6. Concluding remarks References

23 23 24

28 31 31 31

36 37

3. Fission yeast chromosome analysis: fluorescence in situ hybridization (FISH) and chromatin immunoprecipitation (CHIP) Karl Ekwall and Janet F. Partridge 1. Introduction 2. Fluorescence in situ hybridization (FISH) analysis of fission yeast Preparation of probes Cell fixation and cell-wall digestion

3. Chromatin immunoprecipitation from fission yeast Fixation of yeast cells to maintain protein localization Preparation of chromatin extract Immunoprecipitation of chromatin Analysis of immunoprecipitated DNA sequences

Acknowledgements References

39

39 40 40 43

48 49 50 51 55

56 56

4. Isolation of vertebrate metaphase chromosomes and their analysis by FISH

59

Jeff Craig 1. Introduction

59

2. General equipment required for FISH

59

xii

Contents 3. Production of metaphase chromosomes as substrates for FISH Production of fixed metaphase chromosome spreads Production of long prometaphase chromosomes Isolation of suspensions of unfixed metaphase chromosomes

60 60 62 63

4. Spreading fixed chromosomes

65

5. Pretreatments of slides Pretreatment of mitotic chromosome spreads Salt extraction of isolated metaphase chromosomes

66 66 67

6. Labelling DNA probes Choice of label Nick translation Random priming Labelling by PCR Quantifying label incorporation

69 69 69 70 70 71

7. Hybridization Preparation of probes and slides Hybridization

72 72 73

8. Detecting hybridized probe

75

9. Counterstaining and mounting Simple counterstaining Chromosome banding

77 77 77

Acknowledgements

78

References

78

5. Studying progression of vertebrate chromosomes through mitosis by immunofluorescence and FISH

81

Beth A. Sullivan and Peter E. Warburton 1. Introduction

81

2. Fundamental aspects of mitosis The mitotic spindle Chromosomes

81 81 82

3. Detecting centromere/kinetochore proteins on metaphase chromosomes

83

4. In situ hybridization following immunofluorescence

87

5. The use of anti-mitotic drugs Colcemid/colchicine Nocodazole

90 90 91

xiii

Contents Vinblastine and other drugs Cytochalasin/dihydrocytochalasin B (DCB)

91 91

6. Immunofluorescence on anaphase and telophase cells/chromosomes C-anaphase: sister chromatid separation and anaphase in the presence of Colcemid Anaphase chromosomes visualized on the mitotic spindle Anaphase studies on cytokinesis-blocked cells References

92 92 94 96 100

6. Analysis of mammalian interphase chromosomes by FISH and imimmofluorescence

103

Joanna M. Bridger and Peter Lichter 1. Introduction

103

2. Preparation of sample material Adherent cells Suspension cells DNA halo preparations Fixation and permeabilization Improving probe penetration

105 105 106 106 107 109

3. Probes

111

Chromosomal painting probes Probe labelling

111 112

4. Fluorescence in situ hybridization

113

Denaturation Hybridization Washing Detection of reporter molecules

113 115 115 116

5. Immunofluorescence in combination with FISH Primary and secondary antibody incubations after FISH Primary antibody incubation predenaturation and secondary antibody incubation postdenaturation Primary and secondary antibody incubations prior to FISH denaturation Reporter-conjugated primary antibody incubated predenaturation

116 118 119 119 119

6. Mounting the slides

119

7. Analysis

120

References

121

xiv

Contents

7. Fluorescence in situ hybridization in whole-mount tissues Abby F. Dernburg 1. Introduction 2. Probe synthesis and labelling

3. 4. 5. 6. 7.

125 125 127

General considerations Why use fluorescence-based detection? Choice of labelling and detection reagents Probe synthesis

127 127 128 131

Fixation methods for whole-mount FISH Hybridization methods Troubleshooting Microscopy and image analysis Future directions References Acknowledgement

133 136 141 143 144 144 144

8. Analysing the substructure of mammalian nuclei, in vitro Dean A. Jackson 1. Introduction 2. The nuclear matrix and nucleoskeleton 3. Methods used to analyse nuclear organization The nuclear matrix Nucleoids The nuclear scaffold The 'low-salt' nuclear matrix The nucleoskeleton

147 147 147 148 148 149 150 151 151

4. Studying the chromatin loops of different nuclear derivatives 153 Chromatin loops after hypertonic or hypotonic treatment Chromatin loops under 'physiological' conditions Technical tips on cutting and electroeluting chromatin The frequency and nature of attachment sites in different nuclear derivatives

5. The morphology of different nuclear derivatives 6. Assaying nuclear function and nuclear proteins in permeabilized cells Labelling sites of replication and transcription in vitro Technical tips on labelling sites of replication and transcription xv

153 154 156 157

159 160 160 161

Contents Studying protein distribution relative to sites of transcription or replication in permeabilized cells. A typical example

162 164

7. Conclusions

164

References

165

9. Chromosome assembly in vitro using Xenopus egg extracts

167

Jason R. Swedlow 1. Introduction

167

2. Chromosome structure and biochemistry

167

3. Preparation of Xenopus egg extracts for chromatin and chromosome assembly in vitro Xenopus egg maturation Xenopus egg extracts Chromosome assembly extracts—technical tips

168 168 168 169

4. Chromatin and chromosome assembly in vitro Assembly and isolation of chromatin and chromosomes—technical tips

175

5. Immunofluorescence of in vitro assembled chromosomes

178

6. Functional analysis of the role of specific proteins in chromatin and chromosome structure by irnmunodepletion

180

Acknowledgements

181

References

182

10. Chromosome fragmentation in vertebrate cell lines

175

183

Christine J. Farr 1. Introduction

183

2. Telomere-associated chromosome fragmentation

183

3. Experimentally induced de novo telomere formation Design of the telomere-seeding construct Transfection of the telomere-seeding construct Screening stable transfectants for de novo telomere formation

185 185 186 188

4. Targeted de novo telomere formation

191

5. Targeted truncation events in the recombination-proficient avian cell line DT40

192

xvi

Contents 6. The characterization of chromosomes modified by de novo telomere formation and fragmentation Estimation of minichromosome size Assays for mitotic stability of minichromosomes

193 194 196

7. Concluding remarks

197

References

197

A1. List of suppliers

199

Index

207

xvii

This page intentionally left blank

Contributors JOANNA M. BRIDGER

MRC Human Genetics Unit, Crewe Road, Edinburgh EH4 2XU, UK. GIACOMO CAVALLI

Zentrum fur Molekulare Biologie der Universita't Heidelberg, Im Neuenheimer Feld 282, D-69120 Heidelberg, Germany. JEFF CRAIG

The Murdoch Institute, Royal Children's Hospital, Flemington Road, Parkville, Victoria 3052, Australia. ABBY F. DERNBURG

Department of Developmental Biology, Stanford University School of Medicine, Stanford, CA 94305, USA. KARL EKWALL

Karolinska Institute, Department of Biosciences at Novum, SE-141 57 Huddinge, Stockholm, Sweden. CHRISTINE J. FARR

Department of Genetics, University of Cambridge, Downing Street, Cambridge CB2 3EH, UK. DEAN A. JACKSON

CRC Structure and Function Research Group, Sir William Dunn School of Pathology, University of Oxford, Oxford OX1 3RE. UK. PETER LICHTER

Abteilung Organisation Komplexer Genome, DKFZ, In Neuenheimer Feld 280, D-69120 Heidelberg, Germany. DONALD MACLEOD

Institute of Cell and Molecular Biology, University of Edinburgh, King's Buildings, West Mains Road, Edinburgh EH9 3JR, UK. VALERIO ORLANDO

DIBIT hSR Scientific Park, Via Olgettina 58, 20132 Milano, Italy. RENATO PARO

Zentrum fur Molekulare Biologie der Universita't Heidelberg, Im Neuenheimer Feld 282, D-69120 Heidelberg, Germany. JANET F. PARTRIDGE

MRC Human Genetics Unit, Western General Hospital, Crewe Road, Edinburgh EH4 2XU, UK.

Contributors BETH A. SULLIVAN

MRC Human Genetics Unit, Western General Hospital, Crewe Road, Edinburgh EH4 2XU, UK. JASON R. SWEDLOW

Department of Biochemistry, MSI/WTB Complex, University of Dundee, Dow Street, Dundee DD1 5EH, UK. PETER E. WARBURTON

Department of Human Genetics, Mount Sinai School of Medicine, New York, USA.

xx

Abbreviations Ab ActD AMCA ATCC BAC BCIP bp BrdU BSA CCD CENPs CHIP CHO CISS CLSM CREST

antibody actinomycin D 7-amino-4-methylcoumarin-3-acetic acid American Tissue Culture Company bacterial artificial chromosome 5-bromo-4-chloro-3-indolyl phosphate base pair 5-bromo-2'-deoxyuridine bovine serum albumin charged-coupled device centromeric proteins chromatin immunoprecipitation Chinese hamster ovary chromosomal in situ suppression confocal laser scanning microscopy calcinosis, Raynaud's phenomenon, (o)gsophogeal dysmotility, scleroderma, and telangectasia syndrome CSF cytostatic factor CytB cytochalasin B DABCO l,4-diazabicyclo-[2.2.2]-octane (triethylenediamine) DAPI 4, 6-diamidino-2-phenylindole DCB dihydrocytochalasin B dH2O double-distilled water dCTP deoxycytidine triphosphate DHS DNase hypersensitive site DIG digoxigenin DMS dimethylsulfate DMSO dimethylsulfoxide DNase deoxyribonuclease DNP dinitrophenol dNTP deoxyribonucleotide triphosphate(s) DTT dithiothreitol dTTP deoxythymidine triphosphate dUTP deoxyuridine triphosphate EDTA ethylenediaminetetraacetic acid, disodium salt EGS ethylene glycol-bis(succinimidylsuccinate) EGTA ethylene glycol-bis(b-aminoethyl ether-N,N,N',N''-tetraacetic acid EtBr ethidium bromide EtOH ethanol F farad (SI unit of capacitance)

Abbreviations FCS fetal calf serum FISH fluorescence in situ hybridization FITC fluorescein isothiocyanate GFP green fluorescent protein hCG human chorionic gonadotropin HPLC high-performance liquid chromatography HPRI human placental ribonuclease inhibitor HPRT hypoxanthine-guanine phosphoribosyl transferase IP immunoprecipitation IRS interspersed repetitive sequences kb kilobase LAS loop attachment sequences LCR locus control region LIS lithium 3,5-diiodosalicylic acid LMPCR ligation-mediated polymerase chain reaction LPS lipopolysaccharide mAb monoclonal antibody MARs matrix-associated regions MeOH methanol MI mitotic index MMCT microcell-mediated chromosome transfer MNase micrococcal nuclease MT microtubule MTX methotrexate NaOAc sodium acetate NBT nitroblue tetrazolium chloride NCS newborn calf serum NPG n-propyl gallate PAGE polyacrylamide gel electrophoresis PBL peripheral blood lymphocytes PBS phosphate-buffered saline PC Polycomb (chromatin-associated proteins in Drosophila) PCNA proliferating cell nuclear antigen PCR polymerase chain reaction pFA paraformaldehyde PFGE pulsed-field gel electrophoresis PHA phytohaemagglutinin PI propidium iodide PMSF phenylmethlysulfonyl fluoride PMSG pregnant mare serum gonadotropin PNA protein nucleic acid PPD p-phenylenediamine RBC red blood cells RNase ribonuclease xxii

Abbreviations SAAP SARs SDS SIR SSC TAE TBS TCA TdT TE TRF TRITC YAC

streptavidin-alkaline phosphatase conjugate scaffold-associated regions sodium dodecyl sulfate Silent Information Regulator standard saline citrate Tris-acetate, EDTA Tris-buffered saline trichloroacetic acid terminal deoxynucleotidyl transferase Tris-HCl, EDTA terminal restriction fragment tetramethylrhodamine isothiocyanate yeast artificial chromosome

xxiii

This page intentionally left blank

1

Mapping protein/DNA interactions in vivo using ligation-mediated polymerase chain reaction DONALD MACLEOD

1. Introduction This chapter describes the procedures currently used to identify sites of protein interaction on specific DNA sequences. One of these methods, ligationmediated PCR (LMPCR), can be used for in vivo footprinting analysis and is discussed in detail. Using the mouse aprt gene as an example, I show how this method can be used to build up a profile of gene structure in vivo, by mapping transcription factor binding sites and the position of nucleosomes.

1.1 Methods used in determining protein /DNA interactions in vitro There are a variety of methods used to examine protein interactions in vitro which are now standard protocols. Two methods are briefly described here: (a) Bandshift. This method uses polyacrylamide or agarose gel electrophoresis to determine whether a labelled DNA fragment contains binding sites for proteins in an in vitro assay. The labelled DNA probe is incubated with either a crude protein extract, usually prepared from nuclei, or with purified proteins. Binding of the protein to the DNA, in the presence of competitor, results in a 'bandshift' as the mobility of the probe is impaired during electrophoresis (1). (b) DNase I footprinting. This is in an extension of the previous procedure. Once bandshift conditions have been determined which optimize DNA/ protein binding, the specific sites of interaction on the DNA can be mapped using DNase I. This nuclease introduces random nicks into an end-labelled DNA probe to produce a ladder of bands resolvable by polyacrylamide gel electrophoresis (PAGE). The pattern of bands produced from DNase I cleavage of naked DNA is compared to that obtained when the probe has been preincubated with protein extract. Missing bands

Donald Macleod (footprints) indicate positions on the DNA which are not cleaved by DNase I due to masking by bound proteins. Some sites, generally at the edge of footprints, are often hypersensitive to DNase I compared to naked DNA (2).

1.2 Determining protein/DNA interactions in vivo In vitro analysis may not give a true picture of the DNA/protein interactions within the nucleus. Sites which bind protein factors in vitro may be unoccupied in vivo (3), and a different combination of factors may interact with a promoter to regulate the same gene in different cell types (4). Until the development of LMPCR it had been difficult to identify specific sites of protein binding in vivo due to the complexity of the genome. Most analyses have used DNase I and micrococcal nuclease (MNase) which are able to cleave exposed sites in DNA within nuclear chromatin. Cleavage sites in a specific DNA sequence can be mapped by indirect end-labelling (5): (a) DNase I hypersensitive sites. DNase I cleaves at preferred sites within chromatin. When these DNase I hypersensitive sites (DHS) are mapped, they are often found at gene promoters where polymerases and transcription factors interact with the DNA. However, they are also found at enhancers, silencers, replication origins, recombination sites, telomeres, and locus control regions (LCRs) (6). DHS are generally 50-100 bp stretches of DNA but 'domains' of sensitivity have also been identified which extend over much larger regions (7). DHS can be transient or persistent and are thought to indicate a localized decondensation of chromatin which allows access of proteins such as transcription factors and polymerases to the DNA. This open conformation can exist in the absence of factor binding (8). CpG islands, are CpG-rich, non-methylated regions of ~1 kb which are found at the 5' ends of many genes (9). Fractionation of CpG island chromatin has revealed a nucleosome-free fraction which is hypersensitive to nucleases such as DNase I (10). (b) MNase cleavage. MNase is a relatively non-specific nuclease, although it does prefer to cleave AT-rich DNA, and with prolonged digestion will eventually degrade DNA to small fragments and mononucleotides. When isolated nuclei are incubated with low concentrations of MNase there is preferential cleavage of linker DNA between nucleosomes. MNase will also preferentially cleave DNA at DHS. Limited digestion of nuclei with MNase in the presence of Ca2+ produces a nucleosomal ladder when fractionated by electrophoresis (11). The problem with in vivo analysis is in obtaining footprints at the same high level of resolution as can be achieved in vitro. This has been carried out by transferring fractionated genomic DNA from polyacrylamide gels to nylon membranes and probing with single-stranded probes labelled radioactively to

1: Mapping protein/DNA interactions high specific activity (12,13). Indirect end-labelling (5) and primer extension (14,15) have also been used. However, these methods are technically difficult due to the large amounts of DNA and long autoradiographic exposure times required, and the high signal to noise ratio (16). However, some analysis has been successful, especially when multiple copies of the target sequence are present in the genome (17).

2. The ligation-mediated polymerase chain reaction (LMPCR) LMPCR (16, 18) circumvents many of the technical problems of in-vivo analysis and is diagrammed in Figure 1. Partially cleaved DNA is denatured and a gene-specific primer is annealed and extended in a primer extension reaction to obtain blunt-ended fragments. These products are then ligated to

Figure 1. A flow diagram showing the LMPCR method. Step 1: partial cleavage of genomic DNA. Step 2: denaturation and annealing of gene-specific primer AP1. The primer will anneal to multiple fragments of different sizes—only one of which is shown in the diagram. Step 3: primer extension to produce blunt-ended fragments. Steps 4 and 5: ligation of linkers. Step 6: denaturation and annealing of second (nested) linker AP2. Step 7: primer extension in the first step of PCR which copies the lower strand (including the L1 sequence). Step 8: PCR amplification of molecules with AP2 and L1 primers.

Donald Macleod a blunt-ended, unidirectional linker molecule which consists of a 25mer (L1) and an llmer (L2) which have been preannealed together. The linker oligomers do not contain 5' phosphates so will not self-ligate. This will yield a mixture of fragments of different sizes each of which contain a linker molecule at one end which can then be amplified using a second (nested) gene-specific primer on the upper strand and the L1 primer on the lower strand.

2.1 Applications of LMPCR 2.1.1 Genomic sequencing Genomic sequencing is used to identify methylated cytosines in DNA. 5Methylcytosine is present in vertebrate genomic DNA where it is associated with transcriptional repression, and is also found in differentially methylated (imprinted) regions (19). Transcriptional repression may be directly caused by the methylated base excluding transcription factor interaction with its recognition sequence on the DNA, or, indirectly, by proteins which specifically bind to the methylated DNA (19). It is therefore of interest to accurately map the position of the methylated bases in the DNA sequence under study. However, as methylation is lost upon cloning into Escherichia coli, the sequence analysis has to be carried out directly on genomic DNA by Maxam and Gilbert chemical cleavage (20). Partial modification of the DNA by hydrazine or dimethylsulfate (DMS) and subsequent cleavage of the modified bases by piperidine is used to produce sequence ladders when the DNA is fractionated by PAGE. Methylated cytosine is not cleaved in the sequencing reaction and is identified as a 'missing' base in the sequence ladder. LMPCR considerably improves this analysis and has been used to amplify genomic DNA sequence from the 5' regions of the human PGK-1 and HPRT genes (16,21). 2.1.2 In vivo footprinting The same technology used in genomic sequencing is used for in vivo footprinting, and LMPCR is now widely used for this purpose (18). Cultured cells or isolated nuclei are treated with reagents which either directly cleave the DNA (e.g. DNase I, MNase) or modify specific DNA bases (e.g. DMS) which are subsequently cleaved after the DNA is isolated. (a) DNase I footprinting in vivo. DNase I will cleave exposed sites in DNA within the nucleus of cells permeabilized with lysolecithin (3,17). Cleaved DNA fragments from the gene of interest are specifically amplified by LMPCR, and the ladder of bands produced from in vitro modified DNA are compared to the in vivo LMPCR reactions after PAGE. As with in vitro DNase I footprinting, missing bands or 'footprints' of proteins which bind to a DNA sequence in vivo are revealed. In addition, hypersensitive cleavage sites can also be found, particularly at the edge of foot-

1: Mapping protein/DNA interactions prints, which are thought to be due to conformational changes in chromatin that makes these sites more accessible. (b) DM5 footprinting in vivo. DMS can be used to modify DNA at exposed sites in vivo which are then cleaved with piperidine when the DNA is isolated and amplified by LMPCR (18). This procedure is detailed in Section 3. 2.1.3 In vivo mapping of nucleosomes LMPCR has been used to map nucleosome-like structures on chromosomes in vivo using DNase I (3). In an analysis of the human PGK-1 gene on the inactive X chromosome two regions were cleaved by DNase I at 10 bp intervals, suggesting that the DNA is wrapped around nucleosomes. This periodicity is similar to that created by DNase I at exposed regions of the DNA helix, in isolated nucleosome core particles (22). MNase which preferentially cleaves nucleosome-free regions or linker DNA in vivo has also been used, but the results were not informative at the nucleotide level (3). A lowresolution method has been used to map positioned nucleosomes over the aprt gene using LMPCR and MNase cleavage (23) and this method is detailed in Section 4.

3. Mapping protein factor binding sites in vivo with LMPCR and DMS 3.1 DNA modification by DMS in vitro Conditions are determined which result in the incomplete cleavage of genomic DNA, whether in vitro or in vivo, to give a range of fragment sizes after cleavage by piperidine. Using DMS, a random selection of guanines will be methylated at the N7 position and subsequently cleaved by piperidine at the site of modification. Each of the fragments generated by this cleavage will terminate at a G in the sequence, and the DMS conditions must, therefore, be monitored to obtain a complete sequencing ladder where every G is represented. Fragments between 50 and 500 bp are optimal (16), and to achieve this, samples of DNA are incubated for various times with DMS, cleaved by piperidine, and subjected to denaturing gel electrophoresis to determine the average fragment size (Protocol 1). Protocol 1. DMS/piperidine treatment of naked DNA Caution: DMS and piperidine are hazardous chemicals. DMS waste should be deactivated by disposing into a waste bottle containing 5 M NaOH. The lyophilizer should be vented into a fume-hood to remove hazardous piperidine fumes. The use of the screw-capped microcentrifuge tube

Donald Macleod Protocol1. Continued provides an efficient seal when heating the samples at 90°C, an alternative is to seal the tube with Teflon tape, close the cap, and place a weight on top. Equipment and reagents • DMS (BDH) • TE buffer: 10 mM Tris-HCI pH 7.5, 1 mM EDTA . Chloroform/isoamyl alcohol (24:1) • 1:1 phenol (buffered with TE pH 8.0): chloroform/isoamyl alcohol (24:1) • 3 M sodium acetate (NaOAc), pH 5.0

. EtOH (absolute ethanol, stored at -20°C) • 10% piperidine (Sigma) • Alkaline gel electrophoresis buffer: 50 mM NaOH, 1 mM EDTA pH 14.0 . 5 M NaOH . Lyophilizer (Speedvac, Savant)

Method 1. Mix several samples of genomic DNA (10-50 mg) in 100 ml TE buffer with 11xmlDMS in a 1.5 ml microcentrifuge tube and incubate for various times (e.g. 0.5, 1, 1.5, or 2 min) at room temperature. 2. Stop the reactions by adding 100 ml 1.5 M NaOAc and 500 (ml chilled EtOH. Mix and chill the samples at -70°C for at least 1 h and centrifuge for 30 min in the cold at 15800 g in a microcentrifuge. Carefully remove the supernatant and gently wash the inside of the tube with 70% EtOH twice, then air-dry the DNA pellet. 3. To reduce the viscosity of the DNA sample, resuspend the DNA in 20 ml of an appropriate restriction enzyme buffer and incubate with a restriction enzyme whose site lies outside of the region of interest. 4. Extract the DNA with an equal volume of phenol:chloroform isoamyl alcohol, centrifuge for 5 min at 15800 g at room temperature, and remove the aqueous phase to a fresh tube. Repeat the extraction with an equal volume of chloroform:isoamyl alcohol and remove the aqueous phase to a screw-capped microcentrifuge tube. Add 1/10 volume 3 M NaOAc and 2.5 volumes of EtOH, then pellet the DNA by centrifugation as in step 2. Wash the pellet with 70% EtOH and air-dry. 5. Resuspend the samples in 100 ml 10% piperidine. Close the cap tightly and heat at 90°C for 30 min. Cool at room temperature for 5 min and spin down. Add 20 ul 3 M NaOAc and 500 ml EtOH. Chill at -70°C for at least 1 h and centrifuge for 30 min as in step 2. Gently wash the inside of the tube with 70% EtOH twice, and air-dry. 6. Resuspend in 50 ul dH2O (i.e. double-distilled), freeze on dry-ice, and lyophilize to remove traces of piperidine. Resuspend in 50 ml dH2O and repeat the lyophilization. Resuspend in 30 ml TE buffer. 7. Check a small amount of each sample on a 1.5% agarose, alkaline denaturing gel and run at 8V/cm. Stain the gel with ethidium bromide (EtBr) and determine the DNA size under UV light. Note: when the desired size range is achieved, the DNA can be used for LMPCR (Protocol 3).

1: Mapping protein/DNA interactions

3.2 DMS modification of DNA in vivo Since cell membranes are permeable to DMS, modification of DNA within the nucleus can be achieved by incubating cells for a short time in a culture media containing DMS. The DMS concentration, or the time that the cells are exposed, should be determined empirically. Protocol 2. Reagents

DMS treatment of culture cells

• . . .

Phosphate-buffered saline pH 7.3 (PBS) • Proteinase K buffer: 0.3 M NaCI, 10 mM Cells and culture medium EDTA, 10 mM Tris-HCI pH 7.5, 100 m-l/ml DMS (see Protocol 1) proteinase K 1 x trypsin-EDTA diluted with PBS from a • 20% sodium dodecyl sulfate (SDS) w/v in dH 10 x solution (Gibco/BRL) 2° • RNase A

Method 1. Grow several dishes (75 cm2) of culture cells until 60–80% confluent (approx. 1 x 107 cells). Remove the culture medium and replace with fresh medium containing 1% DMS (add just before use and use a control dish with no added DMS). Treat each dish for different time intervals (e.g. 0.5, 1, 1.5, and 2 min one dish for each); remove the medium/DMS and wash the cells twice with PBS. 2. Detach the cells from the dishes using 2 ml 1 x trypsin-EDTA for 2-5 min and centrifuge them at 219 g in 15 ml tubes. Resuspend the pellets in 10 ml aliquots of ice-cold PBS. Repeat the centrifugation and PBS wash to remove traces of DMS. 3. Extract DNA from the cells either by preparing nuclei (Section 4.1) or by resuspending the cells in 10 ml proteinase K buffer. Add SDS to 1% and incubate at 37°C for 3 h, then add 10 m.g RNase A/ml and continue the incubation at 37°C for a further 1 h. 4. Extract the DNA (Protocol 1, step 4), resuspend in 50 ml TE buffer, and assay DNA concentration by spectrophotometry. Note: The in-vivo modified DNA is then cleaved with piperidine to produce a range of fragment sizes that are assayed by denaturing gel electrophoresis (Protocol 1}. When the desired range of fragments is obtained, the DNA samples can be used in LMPCR reactions (Protocol 3).

3.3 Amplification of DMS-/piperidine-cleaved DNA by LMPCR DNA sequence ladders can be obtained by amplifying the partially cleaved molecules using gene-specific primers and linker molecules.

Donald Macleod Protocol 3. Ligation-mediated PCR amplification of cleaved DNA Equipment and reagents • . . • . . .

Linkers and gene specific primers • T4 DNA ligase (NBL) 250 mM Tris-HCI, pH 7.7 . DMS ligation mix: 50 mM Tris pH 7.5, 2 mM 0.1 M MgCI2 ATP 0.1 M dithiothreitol (DTT) • MNase ligation mix: 84 mM Tris pH 7.5, 2 mMATP Thermocycler (e.g. Techne) Sequenase T7 DNA polymerase (USB) * 3 M NaOAc 5 X Seq. buffer: 250 mM NaCI, 200 mM •tRNA Tris-HCI pH 7.7 • Dimethylsulfoxide (DMSO) (Sigma) . 10 mM dNTP (10 mM each of dATP, dGTP, • Red Hot Taq polymerase and supplied dCTP, and dTTP) buffer IV (Advanced Biotech.) . TE buffer (Protocol 7)

Method 1. Prepare annealed linkers. I have used the HPLC-purified oligomers described in ref. 18— LI: 5'-GCG GTG ACC CGG GAG ATC TGA ATT C3' and L2: 5'-GAA TTC AGA TC-3'. Anneal oligomers together at 20 pM/ml of each in 250 mM Tris-HCI pH 7.7. Denature at 95°C for 5 min. Place in a water-bath at 70°C and then cool to room temperature. Store at -20°C and thaw on ice before use. 2. Anneal 0.5 pM of a gene-specific primer to 1-3 mg of cleaved genomic DNA in a 15 ml reaction mix containing 1 x Seq buffer. Denature at 95°C for 2 min and anneal primer at 55°C for 30 min. Chill on ice. 3. Primer extension is carried out using Sequenase. Add 3 ml 5 x Seq. buffer, 3 ml 0.1 M MgCI2, 3 ml H20,1ml10 mM dNTP, 3 ml of 0.1 M DTT and mix. Then add 1.5 ml of Sequenase (diluted 1:8 in ice-cold TE immediately before use). Incubate at 45°C for 15 min and inactivate at 68°C for 15 min. Chill on ice and centrifuge briefly. 4. Ligate the linkers to the DNA. As the linkers are in 250 mM Tris-HCI pH 7.7, it is important that the ligation buffer is adjusted so that the final concentration of Tris is optimal for the ligase. For DMS reactions, add 30 ml ice-cold DMS ligation mix and 5 ml linkers (100 pM). For MNasetreated DNA samples add 30 ml MNase ligation mix and 1 ml linkers (40 pM). In each case, add 200 units (u) T4 DNA ligase and incubate at 18°C overnight. 5. Precipitate DNA by adding 1/10 volume 3 M NaOAc, 10 mg tRNA, and 2 vol. EtOH (prechilled at -20°C). Centrifuge for 30 min at 15800 g in the cold. Wash pellet carefully with 70% EtOH and air-dry. 6. Amplify the ligated DNA by PCR using the L1 primer and a second gene-specific primer that is nested with respect to that used in the annealing reaction (step 1). Resuspend the DNA pellet in 100 ml of a standard PCR reaction mix. The type of thermostable polymerase used

8

1: Mapping protein/DNA interactions is largely a matter of choice, currently I use Red Hot Taq polymerase with the supplied buffer. The final reaction mix contains 20 mM (NH4)3SO4, 75 mM Tris-HCI pH 9.0, 0.01% Tween-20, 1.5 mM MgCI2, 250 mM dNTP, 10 pM of gene-specific and L1 primers, and 1 U Taq polymerase. For GC-rich sequences include 10% DMSO in the reaction mixture. Denature samples at 94°C for 5 min, cycle 20 times (94°C x 30 sec, 55°C X 30 sec, 72°C x 2 min) and finally extend at 72°C for 10 min.

3.4 Analysis of LMPCR reactions The products of LMPCR reactions can be analysed by agarose gel electrophoresis, but if high resolution is required, as in footprinting of protein factors, PAGE is used. The most convenient method is to include another PCR step using an end-labelled third nested gene-specific primer. This allows the products to be directly visualized by autoradiography after PAGE (Protocol 4). 3.4.1 LMPCR analysis using an end-labelled primer Protocol 4. LMPCR analysis by PCR with an end-labelled primer Equipment and reagents • [-y-32P]ATP 10 mCi/ml 5000 Ci/mmol (Amer- • 5 M NH4OAc (ammonium acetate), pH 7.0 sharn) • Red Hot Taq polymerase (Protocol 3) • T4 polynucleotide kinase (NBL) . Thermocycler (Protocol 3) • Tris-HCI pH 7.6 . 6% acrylamide sequencing gel and equip• Labelling mixture: 70 mM Tris-HCI pH 7.6, ment 10 mM MgCI2, 5 mM DTT, 50 mCi [y- . 10 x TBE buffer: 890 mM Tris, 890 mM 32 P]ATP, 50 pM primer, 10 U T4 polynuboric acid, 20 mM EDTA pH 8.0 cleotide kinase . Gel loading mix: 80% formamide, 20 mM . Labelled primer mix: 20 mM (NH4)3SO4, 75 EDTA, 0.05% Bromophenol blue, 0.05% mM Tris-HCI pH 9.0, 0.01% Tween-20, 1.5 xylene cyanol blue mMMgcl2,1mMdNTP' 2 • TE buffer (Protocol • tRNA

PM

of labelled • 3MM chromato9raPhY PaPer (Whatman) • XAR-5 autoradiographic film (Kodak) 1) ,_ . • Vacuum gel drier (Savant)

Method 1. End-label a third primer, nested to that used in the LMPCR reaction, using 10ml labelling mixture. Incubate the reaction mix at 37°C for 1 h and heat inactivate the enzyme at 65°C for 15 min. Precipitate the labelled primer by adding 40 ml TE, 40 ml 5 M NH4OAc, 1 ml tRNA, and 270 ml ice-cold EtOH. Mix and centrifuge for 10 min at 15800 g in the cold. Wash the pellet with 70% EtOH and air-dry. Resuspend in 50 ml TE buffer. 2. To the PCR reactions from Protocol 3, add 5 ml labelled primer mix and a further 1 U Red Hot Taq polymerase/reaction. Cycle twice (94°C x 30

9

Donald Macleod Protocol 4.

Continued

sec, 55°C x 30 sec, 72°C X 2 min) and then extend at 72°C for 10 min. Chill the samples and add 10 ml 3 M NaOAc, 1 ml tRNA, and 250 |xl EtOH; precipitate DNA by centrifugation as in step 1. Wash the pellet in 70% EtOH and air-dry. 3. Resuspend the samples in 10 ml gel loading mix. Denature 2-5 ml of the sample (the rest can be stored at -20°C to repeat the electrophoresis) at 90°C for 2 min and quench on ice. 4. Load the samples on a 6% acrylamide sequencing gel and electrophorese in 1 x TBE buffer. Include a set of DNA sequence reactions as a marker. A DNA clone which contains the sequence analysed can be used with dideoxy-terminator sequencing using established methods (e.g. Sequenase, USB). Preferably the sequencing reactions should be carried out using the same primer as in step 1 since this can be directly compared with the G ladder produced by the LMPCR reactions. 5. After electrophoresis, transfer the gel to 3MM paper, cover with Saran wrap and place on a vacuum drier. Dry down at 80°C, under vacuum, and expose to autoradiographic film. An alternative to this procedure is to transfer the DNA from the gel to a nylon membrane by electroblotting (Protocol 5) and then to hybridize the membrane to a gene-specific probe.

For low-resolution analysis, such as in nucleosome mapping, agarose gel electrophoresis can be used. DNA fragments from the agarose gels can be

Figure 2. Electroblotting apparatus. A drawing of a workshop-made apparatus as described in ref. 16. It consists of a Perspex box containing the blotting apparatus. The electrodes are two stainless-steel plates connected to the power supply through holes in the box lid. A safety mechanism should be incorporated to interrupt the power supply when the lid is removed. The lower plate sits on top of Perspex blocks and on top of this are 10 pieces of Whatman filter paper (cross-hatched) soaked in buffer and rollered to remove air bubbles. On top of this is the gel, then the nylon membrane, another stack of filter paper soaked in buffer, the top electrode plate, and finally a 2 kg weight. The buffer level is adjusted to halfway up the bottom filter paper stack.

10

2: Mapping protein/DNA interactions transferred to nylon membranes by conventional Southern blotting and hybridized to gene-specific probes. 3.4.2 Analysis of LMPCR reactions by electroblotting from acrylamide gels DNA can be transferred from the acrylamide gel by electroblotting using the method described in ref. 24. Equipment is commercially available, but I have used an apparatus made in our workshops as described in ref. 16 and shown in Figure 2. Protocol 5. Electroblotting from acrylamide gels Equipment and reagents • • • •

3 M NaOAc Loading dye mix (Protocol 4) PAGE equipment and reagents Genescreen Plus nylon membrane (NEN)

. TBE buffer (Protocol 4} • Electroblotting apparatus • Whatman 17 filter paper

Method 1. Chill LMPCR samples from Protocol 4, add 1/10 volume 3 M NaOAc and 2.5 volumes of EtOH precipitate the DNA by centrifugation for 30 min at 15800 g in the cold, wash the pellet with 70% EtOH and air-dry. 2. Resuspend in 10 ul loading dye mix and fractionate the DNA by PAGE. 3. After electrophoresis, transfer the gel from the glass plate to Saran wrap. Overlay with 10 layers of filter paper, presoaked in TBE, and squeezed out to remove trapped air bubbles. Invert, so the gel is to the top, remove Saran wrap after rolling out any bubbles, and transfer to the buffer chamber on top of the lower electrode. Lay the prewetted nylon membrane on top (again take care not to trap air bubbles) followed by a second layer of ten presoaked filter papers, the top electrode, and the 2 kg weight. Electroblot onto the membrane at 30V (1.6 A) for 1-2 h. Remove the membrane, air-dry for 15–30 min, and bake in a vacuum oven at 80°C. UV cross-link the DNA. The membranes can then be hybridized using a gene-specific probe.

After hybridization, LMPCR samples showing the most complete sequence ladders from the naked DNA or in vivo modified DNA reactions can be rerun alongside each other on a new gel so that the intensities of the bands can be directly compared. Figure 3 shows a region of the aprt gene promoter which interacts with the transcription factor Spl. This analysis was done with primers which amplify either the coding or the non-coding strand. In each case a footprint is identified at the three Sp1 binding sites by weak or missing G bands in the in vivo samples. Hyper-reactive G bands (relative to the naked

11

Donald Macleod

Figure 3. In vivo footprints at the aprt gene promoter. Autoradiographs showing footprints on lower strand [tracks B and C) and upper strand (tracks E and F) of the mouse aprt DNA sequence from ref. 23. Tracks A and B are LMPCR reactions produced from DMS-modified naked DNA and tracks B, C, D and E from in vivo modified DNA. Samples B and C were treated for 2 min with DMS and samples in D and E for 5 min. Each of the bands in the tracks terminate at a G residue in the aprt sequence. Weak or missing bands in the in vivo samples are indicated by open circles, hyper-reactive bands by filled circles. Numbered brackets refer to the positions of the Sp1 sites and the consensus sequence motifs are shown at the bottom of the figure.

12

1: Mapping protein/DNA interactions DNA sample) are also seen in the in vivo samples. The DNA sequence motifs have been shown to interact with Sp1 in vitro by DNase I footprinting (25).

4. Mapping nucleosomes using micrococcal nuclease and LMPCR MNase-cleaved DNA obtained from nuclei, or an in vitro reaction, can also be amplified using LMPCR.

4.1 Isolation of nuclei from cultured cells This procedure is based on the method described in ref. 26. Cells are treated with weak detergent that ruptures the cell membrane but not the nuclear membrane. The nuclei are purified from cell debris by pelleting through a sucrose cushion, resuspended in a buffer containing glycerol, and stored at -70 °C until use. Protocol 6.

Preparation of nuclei from tissue culture cells

Equipment and reagents • PBS pH 7.3 • Nuclei storage buffer: 50 mM Tris-HCI pH . Buffer R: 10 mM Tris-HCI pH 7.5, 10 mM 7.5, 5 mM MgCI2, 0.1 mM EDTA NaCI, 3 mM MgCI2, 0.1 mM PMSF, 0.25 M . Glycerol sucrose . Centrifuge with swing-out rotor (e.g. Beck• NP-40 man J2-21, JS13 rotor) • Sucrose • 30 ml Corex tubes . 1% SDS

Method 1. Culture cells (approx. 2 x 107) and detach from flasks as in Protocol 2. 2. Wash the cell pellet twice with PBS and resuspend in 10 ml ice-cold buffer R. Add 0.2 ml 10% NP-40 in buffer R and agitate cells by repeated pipetting back and forth using a 10 ml glass pipette (alternatively, use a glass Dounce homogenizer) to lyse cell walls and release nuclei. Check the degree of lysis by removing a drop of the mixture and examining under a microscope using phase contrast. 3. Prepare an ice-cold 5 ml cushion of 1.1 M sucrose in buffer R in a 30 ml glass Corex tube and overlay with 10 ml lysed cells. Centrifuge at 3900 g in a prechilled, swing-out rotor for 3 min at 0°C to pellet the nuclei. 4. Remove the supernatant and gently resuspend the nuclear pellet in 5 ml buffer R and repeat the centrifugation through a sucrose cushion (Step 3). 5. Resuspend the nuclei in 1 ml nuclei storage buffer and examine a drop under the microscope to check that the nuclei are intact and that there is no contaminating cell debris. If contamination is a problem, repeat procedure perhaps with an additional step 3. 13

Donald Macleod Protocol 6.

Continued

6. Remove a small aliquot (10 ml) into a fresh tube and add 1 ml of 1% SDS. Mix to lyse the nuclei and check the DNA concentration using a spectrophotometer. 7. Add glycerol to the nuclei suspension to 40%, aliquot into microcentrifuge tubes, and store at -70 °C.

The amount of NP–40 solution used can be varied depending on the cell type. However, it is important to treat the nuclei gently so that they do not lyse or break open. Checking the DNA concentration before freezing determines how much of the nuclei suspension to aliquot into each tube (generally this is equivalent to 1 mg of DNA/tube). 4.2 Preparation of DNA from nuclei treated with MNase MNase preferentially cleaves the chromatin between nucleosomes to give a nucleosomal ladder when the DNA is purified and fractionated by electrophoresis. For LMPCR, a range of DNA samples should be produced which are only partially digested with the nuclease, as overdigestion will result in preferential amplification of the smaller fragments. Protocol 7. Micrococcal nuclease (MNase) digestion of nuclei Reagents • Buffer M: 50 mM Tris-HCI pH 7.4, 60 mM KCI, 3 mM CaCI2, 0.34 M sucrose • MNase (Worthington) resuspended at 17 U/ml in 10 mM Tris-HCI pH 7.5, 0.5 mM EDTA, 0.5 mM DTT, and 50% glycerol • Proteinase K stop mix: 0.6 M NaCI, 20 mM EDTA, 20 mM Tris-HCI pH 7.5, 1% SDS, 200 m,g/ml proteinase K . 10 mg/ml RNase A (Sigma); heat treat before use (27)

Phenol:chloroform (1:1) (Protocol 1) Chloroform:isoamyl alcohol (24:1) (Protocol 7) 3 M NaOAc (Protocol 7) TE buffer (Protocol 7) T4 polynucleotide kinase buffer: 70 mM Tris-HCI pH 7.6 10 mM Mgcl2, 5 mM DTT' 2 mM ATP •4 polynucleotide kinase (NBL)

Method 1. Thaw a frozen aliquot of nuclei on ice and centrifuge at 720 g in a microcentrifuge for 3 min. Carefully remove the supernatant and resuspend the nuclear pellet in 600 ml ice-cold buffer M. 2. Remove 100 ml aliquots of nuclei into fresh microcentrifuge tubes containing variable amounts of MNase. The amount used must be determined empirically as enzyme activity varies from batch to batch. As an example, 1-15mlof a 1/500 dilution of the stock solution in buffer M was used to produce the range of digestion products shown in Figure 2. 3. Incubate the samples for 4 min at 37°C then stop the reaction with an

14

1: Mapping protein/DNA interactions

4.

5. 6.

7.

equal volume of proteinase K stop mix. Mix and continue the incubation at 37°C for 3 h. Add 10 mg RNase A and incubate at 37°C for a further 1 h, Extract DNA with an equal volume of 1:1 phenol:chloroform and again with chlorofom:isoamyl alcohol (24:1) at room temp. Precipitate the DNA by adding an equal volume of isopropanol and pellet by centrifugation for 30 min at 15800 g at room temp. Wash the DNA pellet with 70% EtOH and air-dry, Resuspend in 100 ml TE buffer. Determine the DNA concentration by UV spectrophotometry and analyse a small sample (e.g. 5 ug) by agarose gel electrophoresis. Resuspend 5 mg MNase-digested DNA in 20 ml T4 polynucleotide kinase buffer and add 5 U T4 polynucleotide kinase. Incubate at 37°C for 1 h. Inactivate the enzyme at 65°C for 15 min, then add 2 ml 3 M NaOAc, 50 ml ice-cold EtOH and precipitate the DNA by centrifugation for 30 min at 15800 g in the cold. Wash the pellet in 70% EtOH and air-dry. Resuspend in 10 ml TE buffer and use for LMPCR (Protocol 5).

8 As MNase cleavage does not produce a 5' phosphate for the ligation step of the LMPCR the DNA is phosphorylated by T4 poiynucleotide kinase.

Figure 4 shows EtBr-stained DNA isolated from nuclei that had been digested with increasing amounts (0, 1, 2, 5, 10, and 15 ml) of a 1/500 dilution

Figure 4. Agarose gel fractionation of micrococcal digested DNA. (A) shows samples of naked DNA incubated with increasing amounts of micrococcal nuclease. Track 0: no enzyme added; tracks 1, 2, 5, and 10 correspond to 1, 2, 5, and 10 ml.1 of a 1/1000 dilution of stock enzyme, (B) shows samples of DNA extracted from nuclei digested with MNase. As above, the numbering of the tracks refers to the amount of enzyme added in each sample (0, incubated without enzyme). DNA markers are 1 kb (m1) and 123 bp (m2) ladders IGibco/BRL).

15

Donald Macleod of MNase stock and run on a 1.5% agarose gel. As the amount of MNase is increased a ladder of nucleosomal bands is detectable.

4.3 Cleavage of genomic DNA with MNase As a control, a set of naked DNA samples cleaved with MNase is also prepared to run alongside the in vivo samples. This will allow preferentially cleaved sites to be mapped on the naked DNA which can be directly compared with sites cleaved in vivo. Protocol 8.

Micrococcal nuclease cleavage of DNA

Equipment and reagents • see Protocol 7

Method 1. Prepare several tubes of genomic DNA resuspended in buffer M and add increasing amounts of MNase. This needs to be determined empirically (e.g. 0, 1, 2, 5, 10, and 15 |xl of a 1/1000 dilution of MNase stock). 2. Incubate for 4 min at 37°C then stop with an equal volume of phenokchloroform mix and extract the DNA (Protocol 1, step 4). 3. Resuspend the DNA in 50 ml TE buffer and analyse fragment size by agarose gel electrophoresis. 4. Phosphorylate DNA ends (Protocol 7, step 6) and use in the LMPCR method (Protocol 3).

Figure 4 also shows an analysis of naked DNA cleaved by MNase. As with in vivo cleaved DNA a range of fragments is produced, but a smear of DNA is evident rather than a nucleosomal ladder. Figure 5 shows the results of in vitro and in vivo treated mouse F9 cell DNA samples which were subjected to LMPCR using primers from the aprt gene, fractionated on a 1.6% agarose gel, and hybridized to a DNA probe which lies downstream of the primer sequences used for LMPCR. Figure 5 clearly shows that MNase cleaves the naked DNA and nuclear DNA at hypersensitive sites. Some are cleaved in naked DNA but not nuclei (e.g. sites D1 and D2), and others are hypersensitive in both samples (e.g. site 1). Site 5 is weakly cleaved in nuclei but becomes more prominent with higher concentrations of MNase. Hypersensitive sites 2, 3, and 4 map to the promoter region where Sp1 interacts with the DNA. The other in vivo sites map at —200 bp intervals on either side and were mapped to the DNA sequence (see Figure 6). A suggested array of positioned nucleosomes over the gene is indicated in 16

1:Mapping protein/DNA interactions

Figure 5. Micrococcal cleavage of aprt DNA amplified by LMPCR. Autoradiographs of Southern blots of agarose gel-fractionated LMPCR samples probed with the mouse aprt gene Isee Figure 6), The LMPCR samples are produced from the same DNA samples shown in Figure 3. Each track refers to the amount of MNase used to cleave the DNA [see Figure 4) (0: incubated without enzyme). The DNA marker is a 1 kb ladder (Gibco/BRL) end-labelled with [32P)dNTP (track M). The arrows indicate prominent MNase cleavage sites in vivo (1-8) and sites which are hypersensitive only in vitro (D1 and D2) and are referred to in Figure 6.

Figures. Sites of in vivo protein/DNA interactions in the 5' region of the aprt gene. Numbers refer to base pairs, and short vertical lines show the positions of the CpG nucleotides in the gene sequence (the bracket indicates the boundaries of the CpG island). The position of the first two exons (open rectangles) and three Sp1 motifs are shown under the CpG plot. Below this, the mapped sites for MNase cleavage in vivo are indicated by arrows (see Figure 5) and the inferred positions of nucleosomes as hatched rectangles. The positions of the Ap2 primer and probe are also shown.

17

Donald Macleod Figure 6. The in vivo cleavage sites were confirmed using other primer sets (23); outside this region nucleosomes could not be mapped, suggesting that they may be randomly positioned in these regions or that the linker DNA is protected from MNase cleavage (23). A similar analysis, using a different method, has been undertaken on another CpG island gene, O6-methylguanine DNA methyltransferase, and shows a very similar in vivo organization (28).

Acknowledgements I would like to thank G.P. Pfeifer et al. (Molecular Biology Section, Beckman Research Institute of the City of Hope, Department of Biology, Duarte, California 91010) and P.R. Mueller and B. Wold (Division of Biology, California Institute of Technology, 156–29 Pasadena, CA 91125) for their detailed protocols on the LMPCR method. I thank Cold Spring Harbor Laboratory Press and the editors of Genes and Development for permission to include data from published work (23). I would also like to thank Adrian Bird for his helpful discussions and The Wellcome Trust, the Imperial Cancer Research Fund, and the Howard Hughes Medical Institute for their support.

References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22.

Strauss, E. and Varshavsky, A. (1984). Cell, 37,889. Galas, D. and Schmitz, A. (1978). Nucl. Acids Res., 5, 3157. Pfeifer, G.P. and Riggs, A.D. (1991). Genes Dev., 5,1102. Somma, M.P., Gambino, I., and Lavia, P. (1991). Nucl. Acids Res., 19, 4451. Wu, C., Wong, Y-C., and Elgin, S.C.R. (1979). Cell, 16, 807. Gross, M. and Garrard, W.T. (1988). Annu. Rev. Biochem., 57, 159. Bonifer, C., Vidal, M., Grosveld, F, and Sippel, A.E. (1990). EMBO J., 9, 2843. Weintraub, H. (1985). Cell, 42, 705. Bird, A.P. (1986). Nature, 321, 209. Tazi, J. and Bird, A. (1990). Cell, 60, 909. Kornberg, R. (1977). Annu. Rev. Biochem., 46, 931. Church, G.M. and Gilbert, W. (1984). Proc. NatlAcad. Sci. USA, 81,1991. Becker, P.B., Ruppert, S., and Schutz, G. (1987). Cell, 51, 435. Saluz, H.P. and Jost, J.P. (1989). Proc. NatlAcad. Sci. USA, 86, 2602. Mueller, P.R., Salser, S.J., and Wold, B. (1988). Genes Dev., 2, 412. Pfeifer, G.P., Steigerwald, S.D., Mueller, P.R., Wold, B., and Riggs, A.D. (1989). Science, 246, 810. Zhang, L. and Gralla, J.D. (1989). Genes Dev., 3,1814. Mueller, P.R. and Wold, B. (1989). Science, 246, 780. Tate, P. and Bird, A.P. (1993). Curr. Opin. Genet. Dev., 3, 226. Maxam, A.M. and Gilbert, W. (1980). In Methods in enzymology, Vol. 65 (ed. L. Grossman and K. Moldave) p. 499. Academic Press, New York. Hornstra, I.K and Yang, T.F. (1994). Mol. Cell. Biol., 14,1419. Noll, M. (1974). Nature, 251, 249. 18

1: Mapping protein/DNA interactions 23. Macleod, D., Charlton, J., Mullins, J., and Bird, A.P. (1994). Genes Dev., 8, 2282. 24. Saluz, H.P. and Jost, J.P. (1987). A laboratory guide to genomic sequencing. Birkhauser, Boston. 25. Dush, M.K., Briggs, M.R., Royce, M.E., Schaff, D.A., Khan, S.A., Tischfield, J.A., and Sambrook, P.J. (1988). Nucl. Acids Res., 16, 8509. 26. Shimada, T., Inokuchi, K., and Nienhuis, A.W. (1986). J. Biol. Chem., 261,1445. 27. Sambrook, J., Fritsch, E.F., and Maniatis, T. (1989). Molecular cloning. A laboratory manual. Cold Spring Harbor Press, NY. 28. Patel, S.A., Graunke, D.M., and Pieper, R.O. (1997). Mol. Cell. Biol., 17, 5813.

19

This page intentionally left blank

2

Mapping DNA target sites of chromatin-associated proteins by formaldehyde cross-linking in Drosophila embryos GIACOMO CAVALLI, VALERIC ORLANDO, and RENATO PARO

1. Introduction The method described here allows the mapping of protein-DNA interactions through the ability of formaldehyde to cross-link proteins and nucleic acids in living cells. Formaldehyde is a very reactive dipolar compound which reacts with the amino groups of proteins and amino acids (1, 2). It shows no reactivity, however, towards free double-stranded DNA, and thus does not cause the extensive DNA damage seen after prolonged exposure to other crosslinking reagents such as UV. Each formaldehyde molecule has the capacity to interact with two amino groups. Therefore DNA-protein, protein-protein, and RNA-protein cross-links are rapidly formed after formaldehyde treatment, creating a stable structure which prevents the redistribution of cellular components. Furthermore, a simple heat treatment is sufficient to reverse the reaction equilibrium, and to allow the isolation of pure DNA for further analysis (3). Formaldehyde cross-linking combined with chromatin immunoprecipitation (IP) is a way of mapping the in vivo distribution of chromatinassociated proteins. As such, this technique is of great value in the analysis of protein-DNA interactions, even more so when studying proteins which do not show specific DNA binding activities in vitro. This has recently been demonstrated for several chromatin-associated proteins, such as Polycomb (PC) in Drosophila (4) and the Silent Information Regulators (SIR) proteins in budding yeast (5). Additionally, analysis of the cross-linking pattern can not only allow mapping of sites of protein-DNA interaction, but can also give an estimation of the relative binding affinity to different sequences across a large genomic region. The method presented here is based on a previously described method of

Giacomo Cavalli, Valeria Orlando, and Renato Paro formaldehyde cross-linking in Drosophlla cultured Schneider SL-2 cells (4). This method was recently improved with a modification of the PCR amplification step, allowing a more accurate quantification of relative binding affinities for adjacent sequences with a resolution in the order of 500 bp (6). Here we present the adaptation of this methodology to the analysis of chromatin from Drosophila embryos. Although SL-2 cells have been successfully used as a model system for the study of various cellular processes, the analysis of embryonic chromatin permits additional functional studies. For example, competitive protein binding to overlapping target sites during development might be investigated at endogenous sites or in transgenic constructs. Moreover, the wealth of genetic mutants of Drosophila allows DNA binding of multimeric protein complexes to be studied in the absence of single components. In this chapter, we will discuss the different steps involved in the mapping of in vivo DNA binding sites by formaldehyde cross-linking, focusing in particular on the steps of preparation and cross-linking of the embryos. As an example of the application of the method, we will show the analysis of the binding profile of two proteins. The first is PC, a protein involved in the maintenance of the repressed state of homeotic genes from mid-embryogenesis to the adult state (7). The second is GAGA factor, which counteracts PCmediated silencing, maintaining the spatially restricted activation pattern of homeotic genes (8).

2. Outline of the method The analysis of in vivo DNA-protein interactions by formaldehyde crosslinking involves the following steps: (a) preparation of embryos for cross-linking; (b) formaldehyde cross-linking and purification of soluble cross-linked chromatin; (c) IP of purified cross-linked chromatin, reversal of the cross-links, and DNA purification; (d) PCR-amplification of the immunoprecipitated DNA; (e) analysis of the immunoprecipitated DNA, which can be done in two ways: (i) using probes to analyse the enrichment of specific DNA fragments in slot-blots, these fragments may represent putative target sites of the protein under study; or (ii) using PCR-amplified DNA as a radioactively labelled probe to hybridize a Southern blot of DNA from chromosome walk of a genomic region of interest. 22

2: Mapping DNA target sites of chromatin-associated proteins

3. Formaldehyde cross-linking in staged Drosophila embryos 3.1 Preparation of fly cages and collection of staged embryos For analysing cross-linked chromatin of 11-16-h-old (after egg lay) embryos, Drosophila embryo collection was performed at 25 °C in cylindrical fly cages of 30 cm diameter and 35 cm depth containing 7500 to 15 000 flies per cage. Flies were 5-12 days old. We used two fly cages for wild-type (Oregon R) flies, while four cages were necessary for the transgenic line 5F24 25,2, which lays significantly less eggs then the wild-type flies. For collection, six apple-juice agar plates, 14 cm in diameter, distributed on two levels in the cage were used. Under these conditions about 1 g wet weight of 11-16-h-old embryos could be collected (in our experience, a minimum of 0.5 g is required to obtain enough DNA for PCR amplification). For staging, a prelay collection of 2 h with plates streaked with 45% acetic acid and fresh yeast paste was first made. Collection was made from 5 p.m. to 10 p.m. and the plates were incubated overnight at 25 °C. At 9 a.m., embryos were collected in embryo wash buffer and further processed as described in Protocol 1. Essentially, embryos were dechorionated, washed, and cross-linked. Cross-linked embryos were extensively washed and then sonicated to produce soluble chromatin, which was used for further analysis (see below). Many chromatin proteins may change their target size distribution at different developmental stages, therefore a careful staging is essential in these experiments. In order to monitor staging, nuclei from a small aliquot of the cross-linked embryos were stained with the DNA staining dye Hoechst 33258. Stained embryos were analysed by fluorescence- and light microscopy to assess the average stage of development. We found that preparations containing a large proportion of embryos older then 16 h resulted in a high fraction of uncross-linked chromatin, most likely due to the poor permeability of these embryos to formaldehyde. On the other hand, cross-linking of embryos during early development can be easily performed, but it must be noted that the same weight equivalent contains a smaller number of nuclei, i.e. of chromatin. This is particularly important when studying early stages, such as preblastoderm embryos, where up to 5-10 g wet weight might be required to obtain DNA amounts sufficient to represent the whole genome complexity after IP. With this set-up, up to four embryo collections could be obtained from wild-type fly cages on alternate days. On the days when no cross-linking was performed, flies were fed in the morning and in the evening with three applejuice agar plates/cage, streaked with about 20 g of fly medium and 2 g of yeast paste. When transgenic flies were used, only two to three embryo collections per cage could be performed. 23

Giacomo Cavalli, Valeria Orlando, and Renata Paro

3.2 Optimizing cross-linking conditions Permcabili/alion of the embryos was obtained by adding three volumes of n-heptane directly to the cross-linking solution (Protocol 1), and cross-linking with vigorous shaking. This was sufficient to permeabilLze embryos until the late developmental stages, but it seems to be insufficient for cross-linking

Figure 1. Analysis of cross-linking efficiency with different concentrations of formaldehyde. Cross-linking of embryonic chromatin was performed with 1.8% (A, C), or 1% (B, D) formaldehyde. (A) and (B): Analysis of the DNA from each fraction collected from a CsCI gradient of 1.8% and 1% formaldehyde cross-linked chromatin, respectively. Fraction numbers are indicated at the top of each gel. The size of the DNA (Mr) markers (Boehringer) is shown in bp on the right of each panel. In the 1.8% sample, the majority of the DNA elutes with a density of 1.36 to 1.42 g/cm3, corresponding to protein-DNA complexes; while in the 1% sample, a large amount of the DNA elutes in the bottom of the gradient. This has a density of about 1.60 g/cm3, corresponding to free, uncrosslinked DNA. (C) and (D): The gels from (A) and (B), respectively, were photographed, and the intensity of the DNA signal was estimated by scanning each lane using the NIH Image software package. The relative intensity of the signal (after background subtraction, calculated from an empty lane) is plotted against the fraction number. The density of fractions which were pooled for chromatin analysis of the 1.8% sample is indicated.

24

2: Mapping DNA target sites of chromatin-associated proteins embryos at the very end of embryogenesis or upon hatching, as judged by the high loss of material observed from preparations containing a high fraction of 16–20-h–old embryos. These embryos were also refractory to Hoechst staining. Thus, more severe methods of fixation should be tested for these late stages. The concentration of formaldehyde required for optimal cross-linking has been tested extensively using different concentrations of formaldehyde. The optimal range was found to be between 1.8% and 3.7%. The effect of undercross-linking is shown in Figure 1. Using 1% formaldehyde, a large fraction of the embryonic chromatin is not cross-linked, resulting in the fractionation of a large amount of free DNA towards the bottom of the gradient (Figures 1B and ID). Sometimes, this was also observed using 1.4% formaldehyde. On the other hand, a concentration of 1.8% resulted in a reproducible cross-linking of about 80% of the material, eluting in a fraction with the density characteristic of protein-DNA complexes ( = 1.39, see Figures 1A and 1C). Concentrations of 2.8% and 3.7% gave a pattern indistinguishable from that observed with 1.8% formaldehyde, while higher concentrations resulted in the loss of a large fraction of the material and to DNA of a high molecular weight (Mr), probably since highly cross-linked material is refractory to shearing by sonication.

Protocol 1. Formaldehyde cross-linking of Drosophila embryos Equipment and reagents • Sonifier apparatus (Branson Ultrasonics Corporation, Sonifier Model 250) equipped with a Microtip 3/16 inch (cat. no. 101–148–069) . 0.1 mm diameter glass beads •Wash . 3% NaOCl . Cross-linking solution: 1.8% formaldehyde, 50 mM Hepes, 1 mM EDTA, 0.5 mM EGTA, 100 mM NaCI, pH 8.0. Add formaldehyde immediately before use from a 37% stock solution stabilized with 10% methanol. . n-heptane • Glycine • Phosphate buffered saline pH 7.4 (PBS) • Triton X-100

• 100% glycerol . Staining dye: Hoechst 33258 (Sigma) . Embryo wash buffer (EWB): 0.03% Triton X100, 0.4%NaCI solution A: 10 mM Hepes pH 7.6, 10 mM EDTA pH 8.0, 0.5 mM EGTA pH 8.0, 0.25% Triton X-100 . Wash solution B: 10 mM Hepes pH 7.6, 200 mM NaCI, 1 mM EDTA pH 8.0, 0.5 mM EGTA pH 8.0, 0.01% Triton X-100 . Sonication buffer: 10 mM Hepes pH 7.6, 1 mM EDTA pH 8.0, 0.5 mM EGTA, pH 8.0 • 10% N-lauroylsarcosine • CsCI (optical grade) • 15 ml and 50 ml Falcon tubes

Method 1. Dechorionate approximately 1 g embryos in 3% NaOCI in EWB for 2-3 min at room temperature (r.t). Wash extensively with EWB. Transfer the embryos to a 50 ml Falcon tube and wash once in 0.01% Triton X100 made up in PBS.

25

Giacomo Cavalli, Valeria Orlando, and Renato Paro Protocol 1. Continued 2. Cross-link with 10 ml of cross-linking solution for 15 min in the presence of 30 ml n-heptane. Shake vigorously. 3. Spin embryos for 1 min in a tabletop centrifuge at maximum speed. Stop cross-linking by washing with 50 ml of PBS, 0.125 M glycine, 0.01% Triton X-100. Let embryos sediment without centrifuging. 4. DNA stain a small aliquot of embryos to check embryo staging as follows: (a) Wash embryos with 0.5 ml PBS. (b) Add 250 ml 1 mg/ml Hoechst 33258 in PBS, stain for 4 min in the dark, allow embryos to sediment. (c) Wash twice with 0.4 ml PBS. (d) Resuspend embryos in 25 ml PBS (at this stage the embryos can be stored overnight at 4°C in the dark). (e) Add 25 ml of 100% glycerol and mix well. Mount on a slide and coverslip. (f)

Analyse on a fluorescence microscope with a suitable filter and score 100-200 embryos to control staging.

5. Resuspend embryos in 10 ml of wash solution A. Transfer them to a 15 ml Falcon tube and wash them for 10 min on a roller. Repeat with wash solution B. Resuspend in 5.5 ml (final volume) sonication buffer. Up to this stage, embryos should still be intact. This can be checked under a microscope. 6. Sonicate with the Branson Model 250 sonifier equipped with a microtip. Adjust sonication empirically and keep the sample on ice throughout. We recommend that the tip be immersed about 2 cm deep in the solution. Perform four sonication cycles of 30 sec each at constant power in the presence of 0.1 mm diameter glass beads. Gradually increase the power up the maximum level possible, but avoid foaming. It should be possible to reach level 6 to 8 in the output control scale of the Branson Model 250 sonifier. Pause for 90 sec between each cycle. At the end, inspect a 10 ml aliquot under phase-contrast microscopy to ensure that all nuclei are lysed (no large particles should be left). 7. Adjust to 0.5% N-lauroylsarcosine and rotate for 10 min. Spin the debris at high speed (in microcentrifuge tubes for 5 min). The supernatant can be further processed by CsCI-gradient purification (Protocol 2) immediately, or frozen in liquid N2 and stored for several days before further purification. This is particularly useful when working with fly lines that lay poorly, so that several cross-linking preparations can be pooled together.

26

2: Mapping DNA target sites of chromatin-associated proteins Protocol 2. Purification of soluble cross-linked chromatin Equipment and reagents • Supernatant from Protocol 7 • 1 mg/ml RNase A (DNase-free) . 10% W-lauroylsarcosine . 10 mg/ml proteinase K (in TE buffer, stored • Sonication buffer (Protocol 1) at -20°C) •Peristaltic pump with 0.114 cm (0.045 . 3 M sodium acetate (NaOAc) pH 5.2 inches) internal diameter tubing .Dialysis buffer: 4% glycerol, 10 mM •20% SDS/v Tris-HCI pH 8.0, 1 mM EDTA pH 8.0, 0.5 • 100 % ethanol (on ice) mM EGTA pH 8.0 . Phenol/chloroform 1:1 v/v . Dialysis bags: microcollodion bags (Sarto- •Gel loading buffer rius, cat. no. 13202) 1% agarose gel and electrophoresis equip. TE buffer: 10 mM Tris-HCI 1 mM EDTA, pH 8.0

Method 1. To the supernatant from Protocol 1 add sonication buffer (Protocol 1) containing 0.5% N-Mauroylsarcosine to a volume of 8 ml. Add 5.68 g CsCI (i.e. adjust to a density of 1.42g/ml) and make up to 10 ml final volume with sonication buffer containing 0.5% N-lauroylsarcosine. 2. Divide the sample between two 5 ml tubes to avoid overloading the gradient (after centrifugation a broad sarkosyl/lipid/protein aggregate is present at the top of the gradient, and overloading can result in poor resolution of the gradient, and thus poor quality chromatin). Spin the CsCI gradient at 195000 g (40000 r.p.m. in a Beckman rotor SW55Ti) for 72 hat 20 °C. 3. Elute 12 x 400 ml fractions per gradient with a peristaltic pump, fitted with 0.114 cm (0.045 inches) internal diameter tubing, at a speed of about 1 ml/min. 4. Check the density profile of the fractions (refraction index should be in the range of 1.390 to 1.362). 5. Dialyse fractions at 4°C in microcollodion bags against 2 litre dialysis buffer. After 2 h change the buffer and continue the dialysis overnight. 6. After the buffer change in step 5, remove about 1/10 vol. (40 ml) to a microcentrifuge tube to check the size and quantity of DNA in each gradient fraction (see Figure 1) as follows: (a) Add RNase A to 50 mg/ml and incubate for 30 min at 37°C. (b) Add proteinase K to 500 ng/ml and SDS to 1%. (c) Incubate at 56°C for 1 h to partially reverse the cross-links. (d) Purify DNA once by phenol/chloroform extraction, then ethanol precipitate and resuspend in 15 ml 1 x loading buffer for agarose gel electrophoresis. (e) Check DNA samples on a 1% agarose gel.

27

Giacomo Cavalli, Valeria Orlando, and Renato Paro Protocol 2. Continued 7. Pool appropriate dialysed chromatin fractions together (usually good fractions have densities of between 1.350 and 1.450 g/ml and, depending on the efficiency of the sonication, DNA is from 0.2 kb to 20 kb in size with an average size of 1.0 kb). Freeze 500 ml aliquots in liquid N2 and store at -80°C. These aliquots should contain 30–60 mg DNA. To analyse DNA of the pooled fractions, take 50 ml of the final pool and repeat the RNase treatment, proteinase K/SDS, and DNA purification of step 6. Estimate the DNA concentration by measuring the optical density at 260 nm and run an agarose gel to check DNA size-distribution. Frozen chromatin aliquots are suitable for IP and should be stable for several months.

4. Immunoprecipitation of cross-linked embryonic chromatin and PCR amplification of the immunoprecipitated DNA IP of the cross-linked purified chromatin, reversal of the cross-links, and DNA purification are performed as previously described (4, 9) and detailed in Protocol 3. Recently, an improvement of the PCR amplification step has been introduced in order to allow a more reliable quantification of the relative enrichment of neighbouring DNA fragments (6). The purified DNA is directly ligated to a blunt-ended linker and PCR-amplified. This assumes that: (a) chromatin DNA fragments which are not blunt-ended are not ligated to the linker; and (b) among the ligated molecules, short fragments, in the range of 200 to 500 bp, are amplified more efficiently than longer ones. Thus, a fraction of the cross-linked DNA molecules is lost during the procedure. Since this loss is random throughout the genome, it does not compromise the quantification as long as the whole genomic complexity is represented in the samples before PCR amplification. Amplification artefacts may be observed under non-optimal sonication conditions, since too large a proportion of very long fragments results in a correspondingly high loss of DNA molecules during PCR. This might result in the loss of entire regions of genomic DNA in the PCR-amplified samples. Such artefacts can be identified by analysing the hybridization profile of a PCR-amplified, radioactively labelled, sample from a mock IP to a restriction digest of a large genomic walk. Since no antibody is used in the mock IP, the probe obtained from this sample should hybridize uniformly to all fragments of the digest in the absence of artefactual loss of genomic complexity (see Protocol 4 and also Figures 3A, B, and C, mock IP samples). Finally, it should be noted that the preferential amplification of shorter DNA fragments sets the resolution of the technique to about 500 bp. 28

2: Mapping DNA target sites of chromatin-associated proteins Protocol 3. Chromatin immunoprecipitation Reagents • RIPA buffer: 140 mM NaCI, 10 mM Tris-HCI pH 8.0, 1 mM EDTA, 1% Triton X-100, 0.1% SDS, 0.1% sodium deoxycholate, 1 mM PMSF (on ice); add PMSF immediately before use from a 100 mM stock in isopropanol • Protein A Sepharose CL4B (PAS, Sigma), equilibrate in RIPA buffer by mixing at 4°C for 30–60 min. (100 mg PAS equilibrated in 1 ml RIPA buffer results in a 50% v/v suspension. After equilibration, PAS is stable for up to 1 week at 4°C.) . TE buffer: 10 mM Tris-HCI pH 8.0, 1 mM EDTA • RNase A • Proteinase K • 100 mM PMSF in isopropanol

Stock solutions for adjusting chromatin samples to RIPA conditions: 10% Triton X100, 1% SDS, 1% sodium deoxycholate, 1.4 M NaCI . LiCI buffer: 250 mM LiCI, 10 mM Tris-HCI pH 8.0, 1 mM EDTA, 0.5% NP-40, 0.5% sodium deoxycholate (on ice) > 1:1 v/v phenol:chloroform/isoamyl alcohol (24:1) . 50 mM Tris-HCI pH 8.0 > 20 mg/ml glycogen > 3 M sodium acetate pH 5.2 > 5 mg/ml RNase A (DNase-free) 10% SDS • 10 mg/ml proteinase K in TE buffer, store at -20°C

Method 1. Thaw a 500 ml aliquot of chromatin and adjust to RIPA buffer conditions by the sequential addition of 100 ml 10% Triton X-100, 100 ml 1% sodium deoxycholate, 100 ml 1% SDS, and 100 ml 1.4 M NaCI. Allow 2 min gentle mixing between additions for equilibration of the chromatin into the new conditions. Finally add 10 ml 100 mM PMSF. Always include one additional sample to be mock-treated as a negative control. 2. Add 30-40 ml of the 50% (v/v) PAS suspension to the chromatin sample. Incubate for 1 h at 4°C, then remove the PAS by centrifuging in a microcentrifuge at top speed for 30 sec. This acts as a preclearing step to reduce non-specific binding to the protein A Sepharose. 3. Remove the chromatin sample to a new tube, and add 2-5 mg of appropriate antibody. Incubate overnight at 4°C, with gentle mixing. The optimal amount of antibody may need to be determined empirically, and a control (mock) IP without antibody should be carried out in parallel. Mock IP isolates DNA non-specifically, but specific antibodybound DNA fragments should be several-fold enriched in the antibody IP. 4. Purify immunocomplexes by adding 30-40 ul 50% (v/v) PAS, and incubating for 3 h at 4°C, with gentle mixing. 5. Wash PAS-antibody-chromatin complexes 5 times for 10 min each in RIPA buffer, once in LiCI buffer, and twice in TE buffer. Carry out all wash steps at 4°C using 1 ml wash buffer, and between washes centrifuge at full speed for 20 sec to pellet the PAS before removing the supernatant. 29

Giacomo Cavalli, Valeria Orlando, and Renato Paro Protocol 3. Continued 6. Resuspend the PAS complexes in 100 ml TE buffer, add RNase A to 50 mg/ml, and incubate for 30 min at 37°C. 7. Adjust the samples to 0.5% SDS, 0.5 mg/ml proteinase K and incubate overnight at 37°C, followed by 6 h at 65°C. 8. Phenol-chloroform extract the sample, and back-extract the lower phenol phase by adding an equal volume of 50 mM Tris-HCI pH 8.0, mixing, and centrifuging. Combine the aqueous phases from the phenol extraction and the back-extraction, and extract the combined phases with chloroform/isoamyl alcohol. 9. Precipitate by adding 1 ml 20 mg/ml glycogen (as carrier), 1/10 volume 3 M sodium acetate pH 5.2, and 2 volumes ice-cold 100% ethanol. Store on ice for 30 min, before centrifuging at 4°C for a further 30 min. Wash the DNA pellet in 70% ethanol, air-dry and resuspend in 20 ul sterile dH20. Store at–20°C. To minimize the risk of contamination during the PCR, we would recommend the use of aerosol-free pipette tips, and the storage of nucleotides, linkers, primers, etc. in small aliquots to prevent contamination of valuable reagents (in addition to preventing frequent freeze-thawing which may cause the destabilization/inactivation of buffers). Protocol 4.

PCR amplification of immunoprecipitated DNA

Equipment and reagents . • Sample from Protocol 3 • 1 uM linker DNA: two oligonucleotides • annealed: (a) a 24mer of sequence 5'-AGA • AGC TTG AAT TCG AGC AGT CAG (phosphorylated at the 5'-end); (b) a 20mer of • sequence 5'-CTG CTC GAA TTC AAG CTT CT. Store in small aliquots at–20°C. . 10 x ligation buffer: 0.5 M Tris-HCI pH 7.6, 125 mM MgCI2, 250 mM DTT, 12.5 mM • ATP. Store at–20°C in small aliquots. . 10 mM dGTP, dATP, dCTP, dTTP: dilute • from commercially available stocks in 10 • mM Tris-HCI pH 8.0, and store in small aliquots at–20°C

T4 DNA ligase, 4 U/ml Taq polymerase and buffer (Boehringer) PCR primer: linker oligonucleotide (b) at 100 MM, store in small aliquots at–20°C Phenol–chloroform (Protocol 3), chloroform/isoamyl alcohol (24:1), 3 M NaOAc pH 5.2, 100% ethanol (20 mg/ml glycogen (Boehringer), 70% ethanol Hindlll restriction endonuclease and corresponding buffer PCR purification columns (e.g. Qiagen) 1% agarose gel, and electrophoresis equipment and reagents

Method 1. Ligate the linker to 7 ul of the sample from Protocol 3 by adding 1 ul 10 x ligation buffer, 1 ul 1 uM linker, and 1 ul (4 U) T4 DNA ligase. Incubate overnight at 4°C. 2. Carry out the amplification directly, without purifying the ligated DNA. Make the volume of the sample up to 78.5 ul with dH2O, and add 10 ml 10 x Taq polymerase buffer, 2.5 ul each 10 mM dNTP, 1 ml 100 mM 30

2: Mapping DNA target sites of chromatin-associated proteins primer and 0.5 ml Taq polymerase. Use the following amplification scheme: 1 cycle of 94°C for 2 min; 35 cycles of 94°C for 1 min, 55°C for 1 min, 72°C for 3 min; 1 cycle of 94°C for 1 min, 55°C for 1 min, 72°C for 10 min. 3. Check 5 ul of the amplified samples on a 1% agarose gel; the product should be a smear ranging between 200 and 500 bp. Obvious bands in the smear may be a result of contamination. 4. Purify the amplified DNA by phenol-chloroform extraction and ethanol precipitation. Remove linker DNA sequences by digesting with 10 U HinDIII. Purify the amplification products from linker DNA using Qiagen PCR purification columns, according to the manufacturer's conditions. The expected yield from the PCR reaction is approximately 5 ug, and amplification of DNA from the mock IP should be as efficient as that of the antibody IP.

5. Analysing the enrichment of putative target sequences in the PCR-amplified DNA 5.1 Slot-blot analysis of the enrichment of putative PC target sequences When putative target sequences of the chromatin protein under study are known, it is possible to test whether they are enriched in the complex DNA sample (obtained by IP from the cross-linked chromatin) using slot-blot analysis: 100 ng of mock IP and antibody IP DNAs are immobilized on the membrane and hybridized with a radioactive probe from the DNA fragment to be tested (Protocol 5A). Relative enrichments are calculated as a ratio between antibody and mock IPs. Figure 2A shows the relative enrichment for a known PC target site, a 6.0 kb EcoRI fragment from the bithorax complex (BX-C), named MCP (10,11). A strong enrichment of this sequence in the PC IP compared to the mock IP was seen in two independent Drosophila lines, as previously reported for Drosophila cultured SL-2 cells (4,6). Only specific PC target sites are enriched by this procedure (Figure 25). Heat-shock sequences, which are known not to be targets of PC-binding, are not enriched. It should be noted that, although this method is very valuable for quantitative measurements of relative enrichments, it is inconvenient for scanning binding to extended chromosomal domains.

5.2 Mapping DNA target sites for Polycomb and GAGA factor in the Drosophila bithorax complex In order to scan binding to a large genomic region, or to compare relative binding affinity to neighbouring DNA fragments from regions with a size of several kb, the PCR-amplified DNA can be used as a complex radioactive 31

B Figure 2. The relative enrichment of the MCP sequence by PC immunoprecipitation in wild-type and transgenic fly lines. Cross-linked Chromatin from two Drosophita lines was analysed: wild type, OregonR flies (wt) and the 5F24 25,2 transgentc line, which contains a P element-derived transgene inserted in the X chromosome (11), The enrichment factor of PC IP versus Mock IP Chromatin, calculated by Phosphorimager quantitation of the hybridization signals, is reported to the right of each panel. (A): Slot-blot enrichment of the 6.0 kb Ecor MCP element from the BX-C was assayed by hybridization to 100 ng of crosslinked, PCR-amplified DNA obtained from mock or PC IP. A substantial enrichment factor is observed in both lines. (B) The same DNAs were hybridized to a 10 kb FcoRI DMA fragment from the heat-shock 87 °C genomic locus (4). In this case, no significant enrichment is detected. Figure 3. Hybridization of control, PC, and GAGA immune precipitated DNA to putative target sites in the BX-C. From the previously published analysis of PC and GAGA factor binding to the BX-C, three fragments sutacloned into Bfuescript KS+ (6) were analysed in order to characterize the binding of PC and GAGA at higher resolution. The three fragments are (i) a 3384 bp EcoRI fragment, from coordinate 218 241 to 221 625 of the published BX-C sequence (15). This fragment contains 3 PC and GAGA factor target site, named the BXD PRE (12). (ii) a 7652 bp FcoRI fragment (coordinates 123 772-131 424), a PC and GAGA target site named Peak C (6). (iii) a 5989 bp EcoRI fragment (coordinates 109 688-115 677) containing the MCP element, a known PC target (6, 10). (A-C): DNAs were digested, run on a 1% agarose gel, blotted and hybridized to PCR-amplified mock IP (panels A, B, C, left), GAGA IP (panel A, right), or PC IP (panels B and C, right). The size of the DNA fragments (in bp) most strongly bound by PC and GAGA factor is shown on the right side of each panel. Panels A and B, lanes 1 and 4; BXD PRE digested by Kpnl (K) and Ps (P). Panels A and B, lanes 2 and 5: Peak C digested by Pstl, BamHI (B), and Xhol (Xh), Panels A and B, lanes 3 and 6: MCP digested by Acc\ (A) and Pstl. In order to analyse PC binding to the two largest MCP fragments of 2208 bp and 1556 bp, the 2208 bp fragment was gel-elmed and digested with Rsall (R) and Xmnl (X) (C, lanes 8, 10), while the gel-eluted 1556 bp fragment was digested by Pvull (Pv) (C, lanes 7, 9). (D-F): Calculated relative enrichments for each DNA fragment. For BXD PRE and Peak C (D, F, respectively!, enrichments for PC and GAGA factor are shown, whilst only PC enrichments are shown for MCP (E) since no GAGA factor binding could be detected to any fragments of the MCP region (panel A, compare lane 3 with 6). The plots show the DNA fragments of each DNA element to scale in the proximal-distal orientation along the BX-C.

32

2: Mapping DNA target sites of chromatin-assaciated proteins probe to hybridize a membrane obtained by Southern blotting an agarose gel containing a restriction digest of DNA from the corresponding genomic region. When large regions are analysed, genomic walks of severa) lambda phage or P1 clones are digested and loaded side by side on the gel. The restriction digests can be set to obtain DNA fragments as small as 500 bp, the maximal resolution of the methodology. Figure 3 shows an analysis of PC- and

33

Giacomo Cavalli, Valeria Orlando, and Renato Paro GAGA factor-binding to three regions from the BX-C. The first is the 6.0 kb MCP fragment mentioned previously. The second is another PC target, a 3.4 kb EcoRl fragment containing the so-called BXD Polycomb response element (PRE) (12). The third sequence is a 7.6 kb fragment, named Peak C, which was identified during the analysis of the whole 340 kb BX-C in SL-2 cultured cells as a specific target for both PC and GAGA binding (6). Specific enrichments were observed for the binding of both proteins to the same DNA subfragments in Drosophila embryos as previously described for SL-2 cells. These enrichments can be plotted to display relative binding affinities to neighbouring regions better (see Figure 3D to F). Importantly, the binding profile of the two proteins is qualitatively different. PC, a chromatin-associated protein which associates to large regions of its target genes, shows rather broad peaks. On the other hand GAGA factor, a DNA binding protein which recognizes GA repeats on the DNA both in vitro and in vivo, binds only to DNA fragments containing its target sites. Therefore, the methodology presented here is suited to mapping the binding of both broadly distributed proteins as well as sequence-specific binding proteins.

Protocol 5. Analysis of the enrichment of specific DNA sequences in PCR-amplified DNA Equipment and reagents • Standard materials for agarose gel electro• Standard materials for random primed DNA phoresis and Southern blotting onto posilabelling with (a-32P)dATP (Specific activity tively charged nylon membrane (e.g. Gene10 mCi/ml 3000 Ci/mmol, Amersham) screen Plus, NEN) . Hybridization buffer: 0.5 M NaHP04 pH 7.2, •Slot-blot apparatus minifoldll (Schleicher 7%SDS, 1 mM EDTA pH 8.0, 1% BSA (a 1 M and Schuell, SRC 072/0) NaHP04 pH 7.2 stock is 0.5 M Na2HP04 con. Phosphorimager apparatus taining 4 ml orthophosphoric acid per litre) . Denaturation buffer: 0.5 M NaOH, 1 M NaCI • Wash buffer 1: 40 mM NaHPO4 pH 7.2, 5% . Dilution buffer: 0.1 x SSC solution, 0.125 M SDS, 1 mM EDTA PH8.0.0.5%BSA NaOH • Wash buffer 2: 40 mM NaHPO4 pH 7.2, 1% SDS 1 mM EDTA . Neutralization buffer: 0.5 M Tris-HCI PH 7.5, ' PH8.0,0.5%BSA 0.5 M NaCI

A. Analysis by slot-blot of enrichment of specific DNA fragments 1. Denature 100 ng DNA of the mock IP and antibody IP samples made up to 15 ml with dH2O by adding 15 ml of the denaturation buffer and incubating for 10 min at room temperature. 2. Add 270 ml of dilution buffer and transfer on ice. Load samples into the slots of the minifold apparatus containing the hybridization membrane and let the liquid flow through the membrane for 30 min. Connect the minifold to a vacuum pump until no liquid is left in the slot. Uncast the slot-blot and neutralize the membrane for 1-2 min in neutralization buffer. Fix the DNA to the membrane by baking at 80°C for 1 h.

34

2: Mapping DNA target sites of chromatin-associated proteins 3. Prehybridize for 2-4 h at 65°C in hybridization buffer (hybridization is carried out essentially as described in ref. 13). 4. Label 25 ng of the specific DNA fragment to be tested for specific enrichment by random priming with 50 ml (a-32P)dATP. 5. Denature the probe by boiling it for 5 min. Cool in ice, then add to 5-10 ml hybridization buffer and incubate overnight at 65°C. 6. Wash the filter twice for 10 min at 65°C in wash solution A, and 3-4 times for 5 min each at 65°C in wash solution B. Expose the filter overnight to a Phosphorimager screen. Quantify the signals of the mock- and the antibody-IP samples to calculate the specific enrichment due to IP of the cross-linked chromatin for the sequence used as probe. 7. After exposure, strip the probe from the membrane according to manufacturer's instructions, and reprobe with 25 ng of labelled Drosophila genomic DNA. This gives a precise estimate of the amount of DNA loaded onto the membrane, to use as a correction factor for standardizing the calculated specific enrichments. B. Searching target DNA binding sites by analysis of the Southern hybridization profile to large genomic walks or subcloned DNA fragments 1. Prepare a Southern filter of a genomic walk which is a potential target of the immunoprecipitated protein, or with the subcloned putative target DNA fragments digested by appropriate restriction enzymes. Prehybridize for 2-4 h at 65°C in hybridization buffer. 2. Random prime label 50–100 ng of amplified DNA with (a-32P)dATP. 3. Hybridize as described above (see part A, steps 3-6), and expose the filter (overnight or several days) to a Phosphorimager screen. 4. Quantify the signal intensity of each band using the software package of the Phosphorimager, and calculate the enrichment of each fragment taking into account the following points: (a) Intensity is proportional to Mr and the resulting values should be normalized with respect to M, if the relative enrichments of different DNA fragments are being compared. The amount of signal per kb of DNA in each fragment may be calculated and plotted against a map of the genomic region (as shown for PC and GAGA factor in Figures 3D to F). (b) The mock IP probe should hybridize approximately uniformly to all fragments (dependent on Mr). As the method is only semiquantitative some sequence-specific differences in amplification may occur, but amplification of different fragments generally varies by no more than 50% from the mean. This degree of error

35

Giacomo Cavalli, Valerio Orlando, and Renato Paro Protocol 5. Continued must therefore be assumed for all experiments. If only a few random restriction fragments of a genomic walk hybridize to the mock IP probe, it is likely that too little input DNA was added to the ligation reaction, or that ligation occurred at low efficiency. (c) Repetitive elements are always strongly enriched in immunoprecipitations and therefore hybridize strongly to all IP DNA probes. These elements can be identified by their strong hybridization to genomic DNA. (d) When analysing large genomic walks, antibody IP probes will hybridize to all fragments of a genomic region to some extent (although enriched fragments hybridize much more strongly). Slot-blot analysis may be used to determine the background level accurately. Typically 100-200 ng of DNA from mock and antibody IPs are immobilized on nylon membrane by slot-blot and hybridized to a number of probes derived from the target DNA of interest. The resulting signals are quantitated and the actual enrichment accurately determined. Comparison between a number of fragments allows the setting of a 'background' level, and only hybridization signals above this level are considered to be enriched.

6. Concluding remarks The formaldehyde cross-linking method presented here allows in-vivo mapping of the DNA target sites of chromatin-associated proteins and site-specific DNA binding proteins in Drosophila embryos. This has several applications in the molecular dissection of the regulation of gene expression for developmentally relevant genes. For example, whenever the potential target sites of a given regulatory factor are known, it can be determined whether or not they are actually bound; in addition, the developmental timing of DNA binding can also be analysed. Moreover, competition between regulatory proteins within the nucleus can be analysed, either at natural target sites or in transgenic constructs. In several regulatory sequences of developmental genes, it has been found that DNA binding sites for regulatory transcription factors are overlapping. With cross-linking analysis, it might be possible to investigate whether binding of a given regulator physically displaces a second from the DNA. Many chromatin-associated proteins show no site-specific DNA binding in vitro. In this case, formaldehyde cross-linking appears to be the method of choice for the analysis of binding to the regulatory regions of potential target genes. Moreover, this method is also applicable to the analysis of histone 36

2: Mapping DNA target sites of chromatin-associated proteins modifications in the chromatin template in vivo. Formaldehyde cross-linking has already been successfully applied in the study of changes in histone acetylation in the mating type loci and telomeres of yeast chromatin (14). Additionally other chromosomal processes, e.g. DNA replication, recombination, and repair, can now be subjected to an in-vivo analysis of participating components.

References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15.

McGhee, J.D. and von Hippel, P.H. (1975). Biochemistry, 14, 1281. McGhee, J.D. and von Hippel, P.H. (1975). Biochemistry, 14,1297. Solomon, M.J. and Varshavsky, A. (1985). Proc. Natl Acad. Sci. USA, 82, 6470. Orlando, V. and Paro, R. (1993). Cell, 75,1187. Hecht, A., Strahl-Bolsinger, S., and Grunstein, M. (1996). Nature, 383, 92. Strutt, H, Cavalli, G., and Paro, R. (1997). EMBO J., 16, 3621. Paro, R. (1995). Trends Genet., 11, 295. Farkas, G., et al. (1994). Nature, 371, 806. Orlando, V., Strutt, H., and Paro, R. (1997). Methods. A companion to methods in enzymology, Vol. 11 (ed. K. Zaret), pp. 205-14. Academic Press, New York. Busturia, A. and Bienz, M. (1993). EMBO J., 12, 1415. Zink, D. and Paro, R. (1995). EMBO J., 14, 5660. Chan, C.S., Rastelli, L., and Pirrotta, V. (1994). EMBO J., 13, 2553. Church, G.M. and Gilbert, W. (1984). Proc. Natl Acad. Sci. USA, 81,1991. Braunstein, M., et al. (1993). Genes Dev., 7, 592. Martin, C.H., et al. (1995). Proc. Natl Acad. Sci. USA, 92, 8398.

37

This page intentionally left blank

3

Fission yeast chromosome analysis: fluorescence in situ hybridization (FISH) and chromatin immunoprecipitation (CHIP) KARL EKWALL and JANET F. PARTRIDGE

1. Introduction The fission yeast (Schizosaccharomyces pombe/S. pombe) is a unicellular ascomycete, that is a well-studied model organism for cell biology and molecular genetics. Some of the main advantages of using fission yeast as a model eukaryote are: • well-defined classical and molecular genetics; • distinct growth properties for cell-cycle analysis; • approximately 50% of the 15 Mb genome has been sequenced; (http://www.sanger.ac.uk/); • immunofluorescence- and electron-microscopy techniques are well developed. Here we describe two recently developed methods for chromosome analysis in fission yeast: fluorescence in situ hybridization (FISH) and chromatin immunoprecipitation (CHIP). FISH allows the detection of any chromosomal DNA sequences (>30 kb) in the nuclei of fixed yeast cells, using fluorescence microscopy after hybridization to DNA probes labelled either directly or indirectly with fluorochromes. In addition, FISH in combination with immunofluorescence, can be used to demonstrate the co-localization of specific proteins to their associated DNA sequences during all stages of the cell cycle (for example co-localization of proteins to centromere DNA). CHIP makes it possible to isolate and precisely map the DNA sequences associated with chromosomal proteins (see Chapter 2). This is achieved by the chemical cross-linking (fixation) of cells and the preparation of soluble chromatin extracts from which specific chromatin fragments are then precipitated using

Karl Ekwall and Janet F. Partridge antibodies to chromatin proteins. Both FISH and CHIP have already contributed significantly to our understanding of chromosomal proteins in many areas of biology. It is our belief that these two methods will be of increasing importance for studies of fission yeast chromosomal proteins and their associated DNA sequences, augmenting and complementing conventional biochemical and genetic approaches. Here we describe detailed protocols for FISH and CHIP in fission yeast, with the emphasis on problems that are peculiar to studies in yeast and that are not encountered in other eukaryotes.

2. Fluorescence in situ hybridization (FISH) analysis of fission yeast The FISH method for fission yeast described in this chapter was originally developed by Uzawa and Yanagida (1). The same laboratory has also reported some modifications to the original protocol (2). Fission yeast FISH involves five steps: (a) (b) (c) (d) (e)

preparation of probes; fixation of cells; denaturation of cellular DNA by alkali treatment; in situ hybridization of the probe to cellular DNA; and fluorescence microscopy.

The main improvement to the FISH protocol described in this chapter and outlined below is the optimization of probe preparation (3). The cell fixation conditions used in the FISH protocol are based on the aldehyde fixation method described by Hagan and Hyams (4). Recently two-colour FISH has been reported (5), enabling the simultaneous detection and co-localization of different DNA sequences, for example centromeres and telomeres, by direct labelling probe DNAs with FITC-dUTP (green) and Cy3-dUTP (red), respectively. The protocols developed for FISH analysis of budding yeast (S. cerevisiae) are quite different from the procedures outlined below for fission yeast. For guidelines on the budding yeast procedure see Guacci et al. (6).

2.1 Preparation of probes Two parameters have to be considered: • the size of the target sequence; and • the length of the probe fragments. As with all FISH experiments the larger the target sequence, the stronger the resulting hybridization signal. In practice this means that, for example, detection of ribosomal DNA is relatively easy, because the target is large and 40

3: Fission yeast chromosome analysis repetitive. The smallest low copy number targets for FISH reported in fission yeast are the mating type loci, probed using three plasmids with a combined target size of ~20 kb (7). Probe cos212 utilized in this chapter contains a 30-kb insert of telomere adjacent sequence and detects telomere regions of chromosome / and // (2). 2.1.1 Probe labelling It is very important to use hybridization probes of denned fragment length. Probe fragments that are too large are unable to penetrate yeast cells and stick to the cell wall, causing high background. Probe fragments that are too short tend to hybridize inefficiently (Figure 1). At least three methods of probe labelling for FISH in fission yeast have been reported: random priming (1); nick translation (3); and direct labelling using terminal deoxynucleotidyl transferase (5). Here we describe the method for nick translation (Protocol 1), as the use of calibrated amounts of DNase I gives control over the length of probe fragments produced. The amount of DNase I has to be titrated and the resulting probe fragment lengths checked (Protocol 2, Figure 1A). Once calibrated, the same batch of DNase I can be used for years (enzyme stock is stable in glycerol at -20 °C). Protocol 1. Labelling of cosmidorplasmid DNA probes by nick translation Equipment and reagents • Cosmid or plasmid DNA of good quality (e.g. supercoiled CsCI preparation) . Water-bath at 15°C .• 0.5 DTT, 500 mg/ml BSA, 100 mM MgS04, 500 alkali stable' (DIGdUTP) (Boehringer, cat. no. 1093088) . dNTPs •7'5

• 10U/mlDNase I (Gibco BRL) . •10 u/ml DNA polymerase 1 (Gibco BRL) MEDTA pH 8.0 • Sephadex 17004201)

G-50 (Pharmacia, cat.

no.

MNH Ac

'

Method 1. To 1.0 M-g DNA add 4 ml each of: 10 XNTS buffer, 2 mM dATP, 2 mM dCTP, 2 mM dGTP, and 1 mM DIG-dUTP; 2 ml of 0.5 mM dTTP, 1 ml (10 Units) DNA polymerase 1 and 1 ml DNase I freshly diluted,b dependent on the batch from 1:15 to 1:500 in sterile dH20. Bring the volume up to 40 ml with dH2O. 2. Incubate for 90 min at 15°C. Stop the reaction by adding 1.2 ml 0.5 M EDTA and place on ice. 3. Purify the labelled DNA by either step (a) or (b): (a) Purify the labelled DNA by filtering, using a G-50 Sephadex spin column as described in ref. 8. 41

Karl Ekwall and Janet F. Partridge Protocol 1. Continued (i)

Plug an empty 1 ml syringe barrel with glass-wool. Fill with Sephadex G-50 swollen in TE buffer pH 8.0 and pack the column to 0.9 ml by centrifuging at 3000 fir for 5 min at room temp. (ii) Add 40 ml of TE buffer pH 8.0 and check that the same volume is recovered by centrifuging at 3000 g for 5 min at room temp. (iii) Add 40 ml of the nick translation reaction from step 2 to the column and elute the purified probe by centrifuging at 3000 g for 5 min at room temp. (b) Purify the labelled fragments by removing proteins by precipitation (step (i)) and then precipitate DNA from free dNTPs (steps (ii) and (iii)). (i)

Add 20 ml 7.5 M NH4Ac to the nick translation reaction, and centrifuge for 5 min at 15000 g at room temp, (ii) Transfer the supernatant to afresh tube. Add 150 ml EtOH, mix and centrifuge for 15 min at 15000 g at room temp. Wash the DNA pellet with 150 ml 80% EtOH. (iii) Centrifuge for 5 min as in (ii). Dry the pellet for 10 min under vacuum then resuspend the probe in 40 ml TE buffer 8.0. 'It is important to use alkali-stable rather than labile DIG-UTP, because NaOH is used during the denaturation step (Protocol 4). b See Section 2.1.1 and Figure 1A for calibration of DNase 1. We recommend a titration procedure for each batch.

2.1.2 Assessing probe fragment length Probe fragments are only able to diffuse through the permeabilized yeast cell wall if they are of sufficiently small size. However, if probe fragments are too short they will not hybridize efficiently to the target sequence. The desired mean probe fragment size is 0.2 kb (range between 0.1 and 0.5 kb) (see Figure 1). The nick translation reaction described in Protocol 1 uses the single-strand nicking activity of DNase I in combination with the polymerization activity of DNA polymerase 1 to incorporate DIG-dUTP into double-stranded DNA. The length of the resulting fragments is measured using a strand-separating gel system such as alkaline agarose gel electrophoresis (8). The efficiency of DIG incorporation into probe fragments can be assessed by Southern blotting of the alkaline gel, followed by detection of the labelled fragments using an anti-DIG alkaline phosphatase assay (Protocol 2). Protocol 2. Checking probe fragment length and efficiency of DIG labelling This protocol is based on ref. 9 and the manufacturer's instructions for the Vectastain kit (Vector)

42

3: Fission yeast chromosome analysis Equipment and reagents . DIG-labelled molecular weight marker VI (Boehringer, cat. no. 1 218 611) . 5 M NaOH . 0.5 M EDTA (pH 8.0) . 6 x alkaline loading buffer (ALB): 300 mM NaOH, 6 mM EDTA, 18% Ficoll (w/v), 0.15% Bromocresol green (wM) » Alkaline gel running buffer: 30 mM NaOH, 1 mM EDTA pH 8.0 • • Midigel (for example Pharmacia GNA 100 apparatus, 10 cm)

. 1MTris-HCI pH 7.6, 1.5 M NaCI . Buffer 1: 0.1 M Tris-HCI pH 7.5, 1.5 M NaCI . 20 X SSC buffer: 0.3 M sodium citrate pH 7.0, 3 M NaCI • BSA fraction V .• 0.1 M Tris-HCI pH 9.5 . Anti-DIG-alkaline phosphatase (anti-DIGAP, Boehringer) > Vectastain kit IV BCIP/NBT (Vector SK5400) or equivalent

Method 1. Make up fresh gel running buffer and place in a cold room. 2. Melt 1% agarose in H2O and cool to 60 °C. 3. To the molten agarose add 5 M NaOH to 50 mM and 0.5 M EDTA (pH8.0) to1 mM. 4. Pour a midigel (for example Pharmacia GNA 100 apparatus 10 cm) and place in the cold room. 5. To 10–15 ml probe sample, add 0.5 M EDTA (pH 8.0) to 10 mM and 0.2 volumes of 6 x ALB. 6. Load the gel dry and layer agarose on top of the wells. 7. Electrophorese in alkaline gel running buffer in the cold room at 75 V for 3-4 h (should give <0.1 A). 8. Neutralize the gel by soaking for 30 min in 1 M Tris-HCI pH 7.6, 1.5 M NaCI 9. Southern blot overnight in 20 x SSC buffer to a nylon filter, then UVcross-link DNA to the filter. 10. Develop the blot using the Vectastain kit IV at room temperature: (a) (b) (c) (d)

Wash for 5 min in Buffer 1. Incubate for 30 min in Buffer 1 + 3% BSA. Incubate for 30 min in 20 ml Buffer 1 + 20 (ml anti-DIG-AP. Wash twice for 15 min in 100 ml Buffer 1 and once for 5 min in 100 ml 0.1 M Tris-HCI pH 9.5. (e) Incubate the blot in the dark with 10 ml 0.1 M Tris-HCI pH 9.5 and 2 drops each of solutions A, B, and C from the Vectastain kit. Once the blue colour has developed, stop the reaction by washing in 100 ml H2O for 5 min.

2.2 Cell fixation and cell-wall digestion The aldehyde fixation method is used (3). To carry out combined antibody labelling and FISH, it is preferable to first label the cells with primary and 43

0

DIG Size Marker

1/500_

1/250

1/125

1/60

1/30

1/15

Control Probe

3: Fission yeast chromosome analysis Figure 1. The effect of DNase I concentration on cos212 probe length and FISH hybridization. Alkaline agarose gel of cos212 DNAs nick translated with digoxigenin in the presence of decreasing amounts of DNase I. The dilution of DNase I from the stock (10U/ml) (Gibco BRL) is shown above the lanes. The gel was transferred to a membrane and developed with anti-DIG alkaline phosphatase (Protocol 2). FISH with probe cos212, labelled with digoxigenin under the conditions in (A), on fission yeast. Digoxigenin-labelled DNA detected with FITC (green). Cells were counterstained with DAPI (blue). For this batch of DNase I the 1:60 dilution gave the strongest specific hybridization (bottom left). Higher dilution and thus longer fragments gave higher background on the cell surface (top left and right). Lower dilution gave a specific but weaker signal (bottom right).

secondary antibodies and then fix the stained cells again for FISH. The choice of fixation method is very dependent on the antigen used. Therefore, it is important to first determine the optimal conditions for each protein (see also Section 3.1). We frequently add sorbitol to 1.2 M to the culture prior to the first fixation to stabilize some antigens osmotically, but this can alter the appearance of nuclear DNA. For combined staining of microtubules in combination with FISH, an initial fixation with 3.8% paraformaldehyde (pFA) and subsequent refixation using a combination of pFA and glutaraldehyde gave good results (4, 7). After cell fixation, the cell walls are digested by Zymolyase 100T and then gently permeabilized using Triton X-100 (Protocol 3). Protocol 3. Fixation and cell wall digestion of fission yeast cells Equipment and reagents • • • •

Shaking water-bath at 32°C Water-baths at 37°C and 65°C Paraformaldehyde (pFA) (Sigma P6148) PEM buffer: 100 mM Pipes pH 6.9, 1 mM EDTA, 1 mM MgSO4, . PEMS buffer: PEM with 1.2 M sorbitol

10 M NaOH 2.4 M sorbitol dissolved in yeast medium Sodium borohydride Glutaraldehyde (BDH, cat. no. 28682) Zymolyase 100 T (ICN, cat. no. 320931) Triton X-100

Method 1. Inoculate fission yeast cells in rich medium (0.5% yeast extract, 3% glucose), or if necessary minimal medium, to obtain 5 x 106 cells/ml in the morning of the next day. 2. Prepare 10 x fixative by adding pFA powder to 38% (w/v) in PEMS buffer, and 10 M stock of NaOH to a final concentration of 0.25 M NaOH. Incubate for 10 min (mix after 5 min) at 65°C to dissolve the pFA. The solution should become transparent. Let the fixative cool to

32 °C. 3. Check the cell density (should be mid-log phase, i.e. 5 x 106 cells/ml). Add 1 volume 2.4 M sorbitol dissolved in yeast medium. Wait for 5 min and then add 1/10 volume of 10 x pFA fixative from step 2

45

Karl Ekwall and Janet F. Partridge Protocol 3.

Continued

followed by glutaraldehyde to a final concentration of 0.25% (v/v). Incubate in a 32°C shaking water-bath for 30 min. 4. Harvest the cells and then wash with 10 ml 1 x PEMS buffer at room temperature. Spin down the cells3 and resuspend in 1 x PEMS with 1.0 mg/ml Zymolyase 100T. Incubate for approximately 90 min at 37°C.b 5. Lyse the cells with 1% Triton X-100 in 0.5 ml 1 x PEMS and leave for 5 min at room temperature. 6. If starting from prestained cells combining immunofluorescence with FISH, then fix the stained cells in 3.8% pFA and 0.25% glutaraldehyde for 45 min at room temperature and then proceed to the PEM washes. 7. Wash three times a with 1 ml of 1 x PEM buffer in a microcentrifuge tube. 8. Weigh out 10 mg of sodium borohydridec into three tubes. 9. Pellet the cells and add 10 ml of 1 x PEM to the first tube containing sodium borohydride, and immediately add 1.0 ml of this solution to the cell pellet. Resuspend and incubate for 5 min. 10. Repeat step 9 twice. 11. Wash three times3 with 1 ml of 1 X PEM buffer in a microcentrifuge tube. a All microcentrifugations of fixed cells throughout the protocol are carried out as double-spin procedures: first spin at 15000 g for 10 sec at room temperature to pellet the cells, then rotate the microcentrifuge tube 180° and spin for another 10 sec. This will ensure that the cells are pelleted efficiently and not lost during the repeated washing steps. b There may be batch-to-batch variation in Zymolyase activity. Therefore, it is a good idea to try out different digestion times with Zymolyase 100T, i.e. 30, 60, 90, 120 min. c Sodium borohydride is a quenching agent to reduce non-specific background for antibody staining after glutaraldehyde fixation.

Protocol 4.

FISH to fission yeast cells

Equipment and reagents • Water-baths at 37°C and 65°C, and a boil. PBS–BAG buffer: PBS, 1% BSA, 0.1% ing water-bath (100°C) sodium azide, 0.5% cold-water fish skin . Rotary wheel gelatin (Sigma, cat. no. G7765) . PEMBAL buffer: 1 x PEM (Protocol 3) with • 5 x Denhardts solution: 0.5 g Ficoll 400, 0.5 1% BSA, 0.1% sodium azide, 100 mM lysine 9 polyvmylpyrrdidone, 0.5 g BSA made up HCI to 500 ml with H2O RNasp A * Hybridization mix: 10% dextran sulfate . nrassa M (Pharmacia, cat. no. 17034002), 50% deion. 0.1 M NaOH ized formamide (BDH, cat. no. 10326), 2 x « 2 x SSC: 0.3 M NaCI, 0.03 M trisodium SSC, 5 x Denhardfs solution, 0.5 mg/ml citrate pH 7.0 denatured salmon sperm DNA

46

3: Fission yeast chromosome analysis < 2 x SSC, 0.1% sodium azide • Anti-DIG-FITC (Boehringer, cat. no. 1207741) > Vectashield mounting medium (Vector H1000)

• DAPI • Polylysine (Sigma, cat. no. P1274) • No.1 22 X 22 mm coverslips (Chance Propper Ltd. )

Method 1. Resuspend fixed cells from Protocol 3 in 0.5 ml 1 x PEMBAL buffer with 0.2 mg/ml RNase A, and incubate for 2 h at 37°C. 2. Boil the DIG-labelled probe (Protocol 1) for 5 min and then add 50–100 ng (2-4 ml) to 100 ml of the Hybridization mix prewarmed to 65°C. Incubate at 65°C for 15 min. 3. Spin down the cells (see footnote a in Protocol 3) and add 100ml,l0.1 M NaOH to each cell pellet containing 0.5-1 x 107 cells, resuspend and wait for 2 min (including the spin time), before removing the NaOH by centrifugation and adding 100 ml of the probe solution. Resuspend the cells by pipetting and incubate at 37°C overnight. 4. Next day, wash out the non-specifically bound probe by adding 200 ml 2 X SSC + 0.1% sodium azide. Spin down the cells as in Protocol 3. 5. Wash the cells three times for 30 min each time at 37°C with 2 x SSC + 0.1% sodium azide. 6. Pellet the washed cells, resuspend in 100 ml PBS-BAG, and incubate for 30 min at room temperature. 7. Pellet the cells and resuspend in 150 ml PBS-BAG containing a 1/100 dilution of FITC-conjugated anti-DIG antibodies (raised in sheep). 8. Incubate, covered in aluminium foil to exclude light, for at least 12 h on a rotary wheel at room temperature. 9. Pellet the cells and wash out secondary antibodies once with 150 ml PBS-BAG and twice with PBS + 0.1% sodium azide. 10. Resuspend the cells in 50 ml of PBS + 0.1% sodium azide with DAPI (0.5 (mg/ml). 11. Mount 5 ml of the cell suspension on polylysine-coated clean coverslips3 and allow to dry in the dark. 12. Mount the coverslip onto a slide using one drop of Vectashield, gently press using Kleenex tissue to remove excess Vectashield. a Coverslips are usually greasy and need to be washed extensively using hot water and detergent, then rinsed several times with water and finally with acetone. Coating with polylysine is carried out by spreading 1 mg/ml solution (w/v) on the clean coverslips, wiping off excess with filter paper, and allowing to dry.

47

Karl Ekwall and Janet F. Partridge

3. Chromatin immunoprecipitation from fission yeast The technique of chromatin immunoprecipitation (CHIP), as used to study histone occupation of transcribed promoters (10), and localization of regulatory chromatin factors in Drosophila (11,12 and Chapter 2) and S. cerevisiae (13, 14), has recently been modified to analyse chromatin components in fission yeast (15,16). The technique relies upon the rapid fixation of large multimeric complexes within whole cells, prior to chromatin extraction and analysis of recognition sites for particular proteins by immunoprecipitation (IP). We describe methods developed for the analysis of histone isoform distribution, and for the isolation of DNA sequences associated with the Swi6 chromodomain protein (Figure 2). Formaldehyde Fix cells

i i Sonicate chromatin

Prepare chromatin extracts

to 0.5 - 1 kb

.

Immunoprecipitate chromatin protein Wash immunoprecipitates Recover DNA Quantitate DNA by competitive PCR

Figure 2. Schematic diagram of chromatin immunoprecipitation procedure. Cells are fixed (Protocol 5) and soluble chromatin extracts, with chromatin sheared to a few nucleosomes in size, are prepared (Protocol &. Protein-DNA complexes of interest are immunoprecipitated by adding antibodies and Sepharose beads (Protocol 7). DNA is recovered by reversing the cross-linking (Protocol 8), and analysed by PCR (Section 3.4).

48

3: Fission yeast chromosome analysis The CHIP protocol consists of four steps: (a) fixation of cells to maintain localization of proteins of interest; (b) preparation of chromatin extract, with DNA of 500–1000 bp target length; (c) IP of chromatin extract, to specifically isolate target sequences; and (d) evaluation of the IP by PCR or slot-blot analysis.

3.1 Fixation of yeast cells to maintain protein localization For histones and other abundant components of chromatin, a good starting point for fixation is to use the conditions outlined in Protocol 5A. For less abundant antigens, base your trial experiments on the conditions you have determined for immunofluorescence microscopy. We rapidly screened a wide range of pFA fixation parameters using fluorescence microscopy of cells bearing a functional green fluorescent protein (GFP) Swi6 fusion, which is expressed at levels similar to the endogenous Swi6p (A.L. Pidoux unpublished observation). Our assay was the detection of bright punctate spots of fluorescence in the nucleus. Conditions tested were: fixing cells in the absence/presence of 1.2 M sorbitol; varying pFA concentration from 0.5 to 4%; and using either a 15- or 30-min fixation time. In parallel, we performed a more limited series of experiments on wild-type yeast, performing standard immunocytochemistry to visualize endogenous Swi6p staining. These parameters must be evaluated in conjunction with the preparation of a chromatin extract (Section 3.2). For instance, we found that 3% pFA fixation for 15 min in the presence of sorbitol generated chromatin that was too highly cross-linked to be efficiently extracted, and resulted in no detectable signal for our target DNA in the crude lysate, even before IP. In the absence of sorbitol, the fixation time could be increased to 30 min with 3% pFA to yield good cytology and extractable chromatin. Protocol 5.

Fixation of cells for CHIP

Equipment and reagents • Shaking incubator(s) at 32°C and 18°C, 65°C water-bath in fume hood . 2 ml screw-top Sarstedt tubes for the mini beadbeater (ref. no. 72693005) • 2.5 M glycine • 50 ml centrifuge tubes

• Paraformaldehyde (pFA) : Fisons AR 37% stock (F/1500/PB17); or for part B, paraformaldehyde powder (Sigma P6148) . Tris-buffered saline (TBS): 8 g NaCI, 0.2 g KCI, 3 g Tris base in 1 litre, pH to 7.4 with HCI

A. For histones and other abundant chromatin proteins 1. Grow 20 ml of yeast culture overnight to 5 X 106/ml. 2. Add pFA to 1%, from a 37% stock (Fisons).

49

Karl Ekwall and Janet F. Partridge Protocol 5.

Continued

3. Fix cells for 30 min at room temperature with shaking. 4. Stop the fixation by adding 2.5 M glycine to 0.125 M, with shaking for 5 min at room temperature. 5. Transfer the cells to a 50 ml centrifuge tube, and spin at 3000 g for 5 min at 4°C, dispose of the supernatant (containing pFA) in a fumehood sink. 6. Gently resuspend the pelleted cells in 20 ml ice-cold TBS. 7. Repeat step 5. 8. Resuspend the cells in 1 ml TBS, and transfer to a 2 ml screw-cap Sarstedt tube. Microcentrifuge the tube at 15000 g for 30 sec and remove the supernatant. Cells can be stored on ice at this point for up to2h. B. For less abundant chromatin proteins 1. Grow 50 ml of yeast overnight to 5 x 106 /ml.alf sorbitol is to be used, add 50 ml of 2.4 M sorbitol in growth medium to the culture and incubate for 5 min prior to fixation. 2. Prepare 50% pFA (w/v) in growth media. To dissolve the pFA, add 5 M NaOH to 0.25 M, and heat at 65°C. Allow the pFA to cool to room temperature, and add to the yeast culture while swirling to the desired percentage. 3. Fix the cells for 15-30 min at room temperature. 4. Follow Steps 4 to 8, part A. * If antigen localization is improved at low temperature, incubate the culture at 18°C for 2 h prior to fixation.

3.2 Preparation of chromatin extract Efficient cell lysis can be achieved by bead-beating cells in the presence of glass beads. As heat is generated, this lysis should be performed in pulses at 4°C. The chromatin is then sheared by sonication to generate fragments of a few nucleosomes in length (~600 bp). It is important to determine the parameters required for good fixation in parallel with the conditions for the generation of sheared chromatin. Material that is overfixed can not be efficiently sheared, and will be pelleted during the clarification steps at the end of Protocol 6. Time courses of sonication should be performed, and the size of the recovered DNA (Protocol 8) evaluated by agarose gel electrophoresis. 50

3: Fission yeast chromosome analysis Protocol 6.

Preparation of chromatin extract

Equipment and reagents • Mini beadbeater (Biospec) • Lysis buffer: 50 mM Hapes-KOH pH 7.5, 140 . Soniprep 150 MSE sonicator or equivalent MM NaCI, 1 mM EDTA, 1% (v/v) Triton Xthat can be used to sonicate material in a 100. 0-1% (w/v) sodium deoxycholate 1.5 ml tube • Protease inhibitors: 'complete' protease . Acid-washed glass beads (425-600 inhibitor cocktail (Boehringer, cat. no. microns; Sigma, cat. no. G8772) 16997498), 1 mM PMSF (Sigma, cat. no. . 15 ml round-bottomed Falcon tubes (cat. P7626) no. 2006) • 19-gauge needle • 1.5 ml and 15 ml centrifuge tubes

Method 1. Resuspend the fixed cell pellet from Protocol 5 in 400 ml ice-cold lysis buffer with protease inhibitors added. 2. Add approximately 400 ml of cold glass beads to the resuspended cells. 3. Lyse the cells with 4 x 40-sec pulses on the bead beater at 4°C, set at maximum power. 4. Check for >95% cell lysis under a microscope. 5. Cut the lid off a 1.5 ml centrifuge tube, and place it inside a 15ml centrifuge tube on ice. 6. Puncture the base of the 2 ml screw-cap Sarstedt tube with a 19gauge needle, and place it on top of the 1.5 ml tube. 7. Centrifuge the 15 ml tube containing both tubes at 3000 g for 5 min, at 4°C, to spin the cell lysate, but not the glass beads, into the rnicrocentrifuge tube. 8. Place the lysate on ice, and resuspend it gently with a 1 ml Gilson pipette blue tip. 9. Sonicate the lysate on ice to shear the chromatin to —600 bp, e.g. 1 x 10-sec pulse at 22 m amplitude.9 10. Spin out the cell debris at 15000 g. 4°C, for 5 min. 11. Transfer the supernatant to a fresh tube, and spin at 15000 g for a further 15 min, at 4°C. 12. Transfer the supernatant (crude lysate) to a fresh 1.5 ml tube on ice. "The sonication time required will vary dependent on the fixation conditions, and the sonicator used.

3.3 Immunoprecipitation of chromatin The parameters used for IP depend to a large extent on the abundance of the antigen. For histones, follow the procedure outlined in Protocol 7A, and for less abundant antigens, Protocol 7B. The crude lysate generated in Protocol 6 51

Karl Ekwall and Janet F. Partridge will suffice for six reactions for abundant antigens, and for one IP of a nonabundant chromatin protein. In general, the cleanest results will be generated by using the most specific reagents possible. For IP of centromeric heterochromatin to which Swi6p is bound, we used affinity-purified polyclonal antibodies (Ab) either crosslinked to protein A-Sepharose (Protocol 7), or we added protein ASepharose to cell lysates after incubating with Ab to collect the antigen-bound chromatin. We also found that it was necessary to preclear extracts with protein A-Sepharose beads before adding antibodies, to reduce non-specific interactions with the beads (Protocol 7B). In addition, high backgrounds resulted from overnight incubations, and these were reduced by shortening the incubation times of Ab with lysate to a maximum of 3 h (Protocol 7B). For histone antibodies, crude sera coupled to protein A-Sepharose gave good results, preclearing was not necessary, and the time of IP was not critical (Protocol 7A and ref. 16). Protein A-Sepharose can be used to bind polyclonal Abs and monoclonal mouse Abs of classes IgG2a, IgG2p, or IgG3. For epitopes for which there are only mouse monoclonal antibodies (mAb) of class IgG1 available, protein G-Sepharose should be substituted. Before use, protein A-Sepharose beads should be swollen in TBS and thoroughly washed twice by resuspending them in 10 volumes of TBS, and centrifuging at 3000 g for 5 min. The beads should finally be resuspended to 50% (v/v) in TBS, sodium azide added to 0.01%, and stored at 4°C. Protein G-Sepharose is sold as a slurry in ethanol; again these beads need to be thoroughly washed and resuspended to 50% (v/v) in TBS before use. Note that there is some batch-to-batch variation in the efficiency of coupling of protein A in the protein A-Sepharose available from suppliers. It is well worth spending the time to properly wash away residual non-covalently coupled protein A from the beads before use, as the efficiency of cross-linking or immunoprecipitating Ab onto beads will be greatly reduced by competition with free protein A. Protocol 7 gives details on CHIP for antibodies which can be bound by protein A. For mouse IgG1, protein A-Sepharose should be substituted by protein G-Sepharose. Protocol 7. Immunoprecipitation of yeast chromatin Equipment and reagents • Rocking platform at 4°C • Antibody-conjugated Sepharose beads (see Protocol 9) or protein A-Sepharose (Sigma, cat. no. P3391)

• Refrigerated microcentrifuge • Duck-bill pipette tips (Sorenson Bioscience Ltd, cat. no. 13760)

A. Immunoprecipitation of abundant chromatin antigens 1. Store 50 ml of the crude lysate from Protocol 6 as the control non-IP chromatin.

52

3: Fission yeast chromosome analysis 2. Add 25 ml of antibody-conjugated Sepharose beads (50% v/v, Protocol 5) to 50 ml of the crude lysate. 3. Incubate the bead/lysate mixture with rocking at 4°C for at least 2 h to overnight. B. Immunoprecipitation of non-abundant chromatin antigens 1. Preclear the crude lysate from Protocol 6, by adding 60 (ml 50% (v/v) protein A-Sepharose beads to 400 ml of the crude lysate. Incubate the bead/lysate mixture with rocking for 1 h at 4°C. 2. Centrifuge at 15000 g for 5 min at 4°C. Carefully remove the supernatant to a clean tube using a duck-bill tip, taking care to leave the beads behind. Save 10% of this material at -20°C as the control nonimmunoprecipitated chromatin sample. 3. To the remainder of the sample, add 60 ml of antibody-conjugated beads (Protocol 9), or antibody.8 4. Bind the antibody-conjugated beads with the lysate for 3 hb with gentle rocking at 4°C. 'Antibody-conjugated beads can be replaced at this stage by the addition of Ab alone. We use 1010mlof affinity-purified polyclonal sera/350 ml precleared lysate; this sera is used at 1/600 for Westerns, and at 1/30 for immunofluorescence. b Antibody incubation with lysate is performed for 1 h prior to the addition of 60 ml (50% (v/v)) protein A-Sepharose beads, and a further 2 h incubation at 4°C.

Protocol 8 gives methods for washing the IPs and reversal of cross-linking for the recovery of DNA. See Protocol 9 for methods for the covalent linking of antibodies to Sepharose beads.

Protocol 8. Washing immunoprecipitations, and reversal of cross-linking Equipment and reagents • • • •

Duck-bill tips {Protocol 7) . Wash buffer: 10 mM Tris-HCI pH 8, 0.25 M LiCI, 0.5% NP–40, 0.5% sodium deoxyRotating wheel cholate, 1 mM EDTA 65 °C water-bath . • TE buffer: 10 mM Tris-HCI, 1 mM EDTA pH 2 ml centrifuge tubes (e.g. Anachem, cat. 8.0 no. HT-20-NG/1000) » 10 mg/ml proteinase K (Boehringer • Lysis buffer (see Protocol & Mannheim, cat. no. 1000144) « Lysis buffer, 500 mM NaCI: replace 140 mM • Phenol:chloroform 1:1 v/v NaCI in lysis buffer (Protocol 6) with 500 . 3 M NaOAc mM NaCI • Glycogen (Sigma, cat. no. G1508) . TES: 50 mM Tris-HCI pH 8.0, 10 mM EDTA, • RNase A (DNase free) 1% SDS

53

Karl Ekwall and Janet F. Partridge Protocol 8.

Continued

Method 1. Collect the beads from Protocol 7 by centrifuging at room temperature for 2 min at 15000 g. Use a duck-bill tip to discard the supernatant, being careful to retain all beads in the tube. 2. Wash the beads successively with 1 ml of the following buffers. Wash the beads for 5 min at room temperature on a rotating wheel, then collect the beads by centrifuging for 2 min at 15000 g, and discard the supernatant: • 1 ml lysis buffer twice • 1 ml lysis buffer, 0.5 M NaCI • 1 ml wash buffer • 1 ml TE buffer 3. Carefully remove all the supernatant. Add 50 ml TES to the beads and incubate at 65°C for 10 min. 4. Centrifuge the beads at 15000 g for 1 min at room temperature, and transfer the supernatant to a fresh 1.5 ml tube. 5. Wash the beads with 200 ml TES, centrifuge as in step 4, and pool the supernatants. 6. Add 200 ml TES to the control non-immunoprecipitated chromatin from Protocol 7A, step 1, or Protocol 76, step 2. 7. Incubate the control and IP chromatin samples at 65°C for at least 6 h to reverse the cross-linking. 8. Cool the samples to room temperature. Add 25 ml 10 mg/ml proteinase K and 250 ml TE buffer to each tube. Incubate at 37°C for 2 h. 9. Phenol:chloroform-extract the DNA, and transfer the supernatant to a 2 ml microcentrifuge tube. 10. Ethanol-precipitate the DNA by adding 1/10 volume of 3 M NaOAc, 2.5 volumes of 96% ethanol, and 2 ml 10 mg/ml glycogen. Freeze the mixture on dry ice for 30 min, and collect the DNA by centrifuging at 15000 g for 30 min at 4°C. 11. Vacuum-dry the pellet, resuspend in 40 ml TE + 10 ml RNase A and incubate for 30 min at 37°C.

54

3: Fission yeast chromosome analysis Protocol 9. Covalent coupling of antibodies to protein Sepharose Equipment and reagents • Protein A-Sepharose or Protein G-Sepharose"'Sigma, cat. no. P3391 or P4691) • 0.2 M borax pH 9.0 (sodium tetraborate; Sigma, cat. no. B9876) « TBS (Protocol 75), 0.01% sodium azide

• Dimethyl pimelimidate (Sigma, cat. no. D8388) • 0.2 M ethanolamine pH 8.0 (Sigma, cat. no. E9508)

Method (based on ref. 17) 1. Incubate Ab with protein-Sepharose for 1 h with rocking at room temperature. We use a 1:10 ratio of affinity-purified antisera to 50% (v/v) protein-Sepharose. This sera is used at 1/600 for Western blotting. Perform a control preparation of placebo-coupled beads for the preclearing steps (Protocol 76). 2. Wash the beads twice with 10 vol. of 0.2 M sodium tetraborate pH 9.0 by spinning at 3000 g for 5 min.b 3. Dissolve dimethylpimelimidate in 0.2 M sodium tetraborate pH 9.0 to a final concentration of 20 mM, and incubate the Sepharose beads in 10 vol. of this solution for no longer than 30 min, including step 4. 4. Rapidly remove the coupling mixture by centrifuging at 3000 g for 5 min. 5. Wash the beads twice with 10 vol. of 0.2 M ethanolamine pH 8.0. 6. Resuspend the beads in 10 vol. of 0.2 M ethanolamine, and incubate with rocking for 2 h at room temperature.6 7. Wash the beads by centrifuging with 10 vol. TBS three times. 8. Resuspend the beads to 50% (v/v) in TBS, 0.01% sodium azide, and store at 4°C. ' Protein-Sepharose of appropriate type for the Ab should be chosen (Section 3.3), and should be washed and resuspended to 50% (v/v) in TBS before use. b Coupling efficiency of Ab to beads can be estimated by taking samples of beads at these points, boiling them in Laemmli buffer (17), and comparing the amount of heavy chain (55 kDa) stained by Coomassie blue after SDS-PAGE of these samples before and after coupling.

3.4 Analysis of immunoprecipitated DNA sequences Several methods have been described in the literature for the analysis of immunoprecipitated chromatin sequences. Analyses by slot blotting, Southern blotting, and by PCR are described in Chapter 2. Here we describe a strategy (competitive PCR) that takes advantage of the ease of genetic manipulation in fission yeast and that gives an accurate measurement of the relative enrichment of a particular locus compared with an internal control within the 55

Karl Ekwall and Janet F. Partridge same genome (Figure 2, ref. 16). We use 5. pombe strains bearing a ura4+ marker gene inserted at particular sites within the telomeres, centromeres, or mating type loci. In addition, these strains carry a ura4 minigene, called DS/E, with a 268-bp deletion, at the endogenous ura4 locus (18). Primers were designed which span the region deleted in the minigene, and give rise to products of 426 and 694 bp from the minigene and ura4+, respectively. These primers can be used in competitive PCR reactions (19) to determine accurately the ratio of ura4+ to minigene in the input DNA sample, and the ratio of ura4+ to minigene in the IP sample. The fold enrichment of the IP is calculated by dividing the IP DNA ratio by the non-IP ratio for each strain. A further control can be built into these calculations, by determining the ratios for a strain with a non-heterochromatic ura4+ locus at the same time. The potential disadvantages of this method are that strains need to be constructed, and that the technique relies on the spreading of the protein of interest across the ura4+ chromatin. For these experiments, as only a small proportion of the total input DNA is immunoprecipitated, we use 3 ml of a 1/100 dilution of the input material, and 3 ml of the IP material in a 40-ml PCR reaction. For quantification of the relative ratios of ura4+ and minigene, we include 5 ml [a-32P]dCTP (10mCi/ml) (3000 Ci/mmol) in each ml of PCR mixture. PCR-amplified products are resolved on 4% polyacrylamide gels, dried onto Whatman 3MM paper, exposed to a phosphorimage screen, and quantified using Molecular Dynamics software. IP can be fickle, it is therefore essential to reproduce results several times, and to build appropriate controls (e.g. IP with placebo beads, or unrelated antibodies) into the experimental design.

Acknowledgements FISH technology was transferred to the Allshire laboratory with the generous support of Prof. Mitsuhiro Yanagida. We would like to thank the following for advice, help, and encouragement: 'Chromosome Biology', 'Photography', W. Bickmore, I. Hagan, J.P. Javerzat, A. Pidoux, S. Strahl-Bolsinger, and B. Turner. Special thanks to Robin Allshire for his continued support and enthusiasm. KE is supported by the MFR (K97-13P-1182), and our work is supported by a core grant to R. Allshire from the Medical Research Council

References 1. Uzawa, S. and Yanagida, M. (1992). J. Cell Sci., 101, 267. 2. Funabiki, H., Hagan, I., Uzawa, S., and Yanagida, M. (1993). J. Cell Biol., 121, 961. 3. Ekwall, K., Nimmo, E. R., Javerzat, J-P., Borgstrom, B., Egel, R., Cranston, G., and Allshire, R. (1996). J. Cell Sci., 109,2637. 56

3: Fission yeast chromosome analysis 4. Hagan, I. M. and Hyams, J. S. (1988). J. Cell Set., 89,343. 5. Chikashige, Y., Ding, D-Q., Imai, Y., Yamamoto, M., Haraguchi, T., and Hiraoka, Y. (1997). EMBOJ., 16, 193. 6. Guacci, V., Hogan, E., and Koshland, D. (1994) J. Cell Biol., 125, 517. 7. Ekwall, K., Javerzat, J-P, Lorentz, A., Schmidt, H., and Allshire, R. (1995). Science, 269,1429. 8. Maniatis, T., Fritsch, E. F., and Sambrook, J. (1989). In Molecular cloning—A laboratory manual (1st edn). Cold Spring Harbor Laboratory Press, New York. 9. Sambrook, J., Fritsch, E.F., and Maniatis, T. (1989) In Molecular cloning—A laboratory manual (2nd edn). Cold Spring Harbor Laboratory Press, New York. 10. Solomon, M.J., Larsen, P.L., and Varshavsky, A. (1988). Cell, 53, 937. 11. Orlando, V. and Paro, R. (1993). Cell, 75,1187. 12. Strutt, H., Cavalli, G., and Paro, R. (1997). EMBO J., 16, 3621. 13. Hecht, A., Strahl-Bolsinger, S., and Grunstein, M. (1996). Nature, 383, 92. 14. Strahl-Bolsinger, S., Hecht, A., Luo, K., and Grunstein, M. (1997). Genes Dev., 11, 83. 15. Saitoh, S., Takahashi, K., and Yanagida, M. (1997). Cell, 90,131. 16. Ekwall, K., Olsson, T., Turner, B.M., Cranston, G., and Allshire, R.C. (1997). Cell, 91,1021. 17. Harlow, E. and Lane, D. (1988). In Antibodies a laboratory manual. Cold Spring Harbor Laboratory Press, New York. 18. Allshire, R.C., Nimmo, E.R., Ekwall, K., Javerzat, J-P., and Cranston, G. (1995). Genes Dev., 9,218. 19. Gilliland, G., Perrin, S., and Bunn, H.F. (1990). In PCR protocols (ed. M.A. Innis, D.H. Gelfand, J.J. Sninsky, and, T.J. White), p. 60. Academic Press, San Diego, California.

57

This page intentionally left blank

4

Isolation of vertebrate metaphase chromosomes and their analysis by FISH JEFF CRAIG

1. Introduction This chapter concentrates on the in situ hybridization of DNA probes to metaphase chromosomes isolated from vertebrate cells. Fluorescence in situ hybridization (FISH) has a number of applications within the field of chromosome structure, ranging from the simple mapping of specific genes or markers (1-3), mapping transgene integration sites (4-7), through to analysis of the distribution of particular classes of DNA sequence across the genome, e.g. repeat sequence families (8-10), CpG islands (11), fractions of defined base composition (12), and scaffold or matrix-associated sites (13). Whole chromosomes can also be highlighted by chromosome painting probes (14, 15, and Chapter 6), and interspecies hybridization can highlight regions of conserved synteny (16, 17). There are also other novel applications of FISH in studies of chromosome structure, these either analyse karyotype changes or explore the higher order organization of chromosomes (18-24). The protocols detailed in this chapter have been developed from those used by two different research groups (25, 26), after various workshop discussions—such as those set up by Oncor Appligene, and from the research of this author and colleagues.

2. General equipment required for FISH Some basic equipment is needed for all the different FISH techniques. This includes: • two or three water-baths; • a hot-block capable of reaching 75 °C; • frosted slides;

Jeff Craig • coverslips of various sizes (the most commonly used being 22 X 22 mm, 22 X 32 mm, and 22 X 50 mm and between 0.1 and 0.2 mm thick); • rubber cement; • a phase-contrast microscope; • a small lightproof slide incubation box; • slide forceps; and • a rotating platform. Also essential are 50 ml glass Coplin jars; these can hold up to 10 slides backto-back or 9 slides in a zigzag pattern, the latter being better for buffer circulation. If more slides are needed this can be done in 200 ml slide troughs (20 slides). Equipment for FISH can be bought from many large suppliers. Optional, but useful, equipment includes a shaking hot-block and a rotating vacuum dryer. Many of the methods in this chapter contain slide washing steps. Solutions and Coplin jars are prewarmed in water-baths; washing is performed by placing the Coplin jars on a rotating platform, preferably within a small Perspex incubation hood.

3. Production of metaphase chromosomes as substrates for FISH Described below are methods for obtaining fixed metaphase and prometaphase chromosome spreads, as well as techniques for obtaining suspensions of isolated, unfixed metaphase chromosomes.

3.1 Production of fixed metaphase chromosome spreads The technique outlined below describes the production of chromosome spreads from peripheral blood lymphocyte (PBL) cultures and other cell cultures. Subconfluent cell cultures contain a proportion of cells undergoing cell division at any given time. Human PBLs are stimulated to divide with the plant lectin phytohaemagglutinin (PHA), resulting in a wave of cell division after 2-3 days. Similarly, cells isolated from mouse spleens can be stimulated to divide in culture by the addition of lipopolysaccharide (LPS). Mitotic cells are accumulated using a spindle inhibitor, most commonly colchicine, or its derivative demecolcine (Colcemid). A longer incubation time with these agents not only increases the percentage of mitotic cells within a population (the mitotic index, MI), but also increases the degree of chromosome condensation. If absolute numbers of mitotic cells are important, as in Protocol 3, cells can be incubated with Colcemid for many hours and they will have high MIs (50–100%) but extremely condensed chromosomes. On the other hand, cultures released from other cell-cycle blocks and harvested without spindle 60

4: Isolation of vertebrate metaphase chromosomes inhibitors can produce a low, but workable, MI (0.5-5%) of long prometaphase chromosome spreads (Protocol 2). Protocol 1 uses a 40-min incubation with Colcemid. Cells are then harvested and swollen in a hypotonic solution before being fixed with 3:1 methanol:acetic acid. It is important, especially when working with PBLs, to take care with these two steps—especially the fixation—as lysed red blood cells (RBCs) can precipitate if the 'fix' is added too quickly. The number of RBCs can be reduced significantly by culturing only the 'buffy coat' (the whitish interface between the RBCs and plasma after centrifugation of whole blood) that is enriched in white blood cells. The fixative should be freshly made with absolute methanol (HPLC grade if possible) and glacial (>99% pure) acetic acid just before it is needed, since it absorbs water from the atmosphere. Protocol 1. Production of metaphase chromosomes from human peripheral blood Caution: Treat all blood samples with extreme care as 'Biological hazard' material, and take all recommended precautions. Equipment and reagents Venipuncture equipment Sodium heparin (Sigma) RPMI 1640 medium, 200 mM L-glutamine (Gibco BRL) Fetal calf serum (PCS, Sigma) Penicillin-streptomycin (Sigma P0781) PHA (PHA-L, Sigma) 75 cm3 tissue culture flasks

• 10 mg/ml Colcemid (Sigma); store at–20°C in 10 ml aliquots • 5% CO2 incubator . Hypotonic 0.075 M KCI • 37 °C water-bath • 3:1 fixative (HPLC grade methanol:glacial acetic acid) (Sigma)

Method 1. Collect peripheral blooda and transfer immediately to tubes containing preservative-free sodium-heparin (e.g. 50 ul of 5000 U/ml Naheparin for 10 ml blood). Centrifuge in 15 ml tubes at 200 g for 7 min at room temp. 2. Remove 2 ml of the buffy coat.b Alternatively, use heparinized whole blood. 3. Add 2 ml of the buffy coat or 3.5 ml whole blood to 50 ml prewarmed RPMI 1640 medium containing 15% FCS, 100 U/ml penicillin, 100 mg/ ml streptomycin, 2 mM L-glutamine, and 1% PHA. 4. Mix well, transfer to a 75 cm2 tissue culture flask and incubate at 37°C with 5% CO2 for 69-75 h. 5. Add 500 ul 10 mg/ml Colcemid for 40 min. Alternatively, synchronize the cultures by arresting the cells in S-phase with methotrexate (Protocol 2). 61

Jeff Craig Protocol 1.

Continued

6. Centrifuge at 200 g for 7 min at room temp, remove the supernatant down to 5 ml and mix with the pellet using a vortex mixer on a low setting. 7. While mixing as in step 6, add 5 ml prewarmed (to 37°C) hypotonic KCI dropwise, then top up to 50 ml. 8. Incubate in a 37°C water-bath for 15-20 min, then add 5 ml freshly made fixative at room temperature. 9. Repeat step 6. 10. Add 45 ml fixative, adding the first 10 ml dropwise while vortexing on a low setting. 11. Repeat steps 9 and 10. 12. Repeat step 6. (The pellet should now be white). Add 10 ml fix, transfer to a 10-15 ml tube, centrifuge again at 200 g for 7 min, and remove the supernatant. 13. Repeat step 12 twice more, and finally resuspend the pellet in 5 ml fix, mixing well. The chromosome preparation is now ready to spread on slidesc or it can be stored at -20°C. * It is advisable to test the peripheral blood from a number of individuals, as there can be some variability in the quantity and quality of chromosome preparations produced. b The buffy coat is not always obvious; aim to remove approximately equal amounts of plasma and RBCs. This step increases the proportion of white blood cells, making the culture easier to harvest and fix. c It is advisable to spread as soon after fixation as possible, as preparations deteriorate slowly during storage.

Protocol 1 is also applicable for cultured cell lines. After the required length of exposure to Colcemid, cells are harvested, washed in PBS, and processed as in Protocol 1 (from step 7). The concentration of cells in hypotonic KCI should be <2 X 107/ml. For rodent chromosomes, and for some human fibroblast cell lines, we have found that better chromosome morphology is obtained using an alternative hypotonic (0.034 M KCI, 0.017 M trisodium citrate).

3.2 Production of long prometaphase chromosomes Longer and less condensed chromosomes than those obtained from Protocol 1 can be produced by cell synchronization before mitosis. Methotrexate (MTX) is a folic acid analogue that inhibits dihydrofolate reductase, so blocking thymidine synthesis. MTX-treated cells are blocked at either the G1/S boundary or in mid S-phase. Cells can be released from the block by the thymidine analogue 5-bromo-2'-deoxyuridine (BrdU). BrdU mildly inhibits chromosome condensation and is immunodetectable, so that chromosomes can be banded post-FISH with fluorochrome-conjugated anti-BrdU. For a more pronounced inhibition of chromosome condensation, the DNA intercalating agent actinomycin D (ActD) can be added 1 h prior to harvesting (27). 62

4: Isolation of vertebrate metaphase chromosomes Protocol 2. Methotrexate synchronization to produce prometaphase chromosomes Equipment and reagents • Equipment and reagents as Protocol 7 • 4.5 ml/ml methotrexate (MTX/amethopterin, Sigma); store at -20°C

•

• 3 mg/ml BrdU (Sigma); store at–20°C 670 Mg/ml actinomycin D (ActD, Sigma); store at–20°C

Method 1. Set up a 50 ml PBL culture as detailed in steps 1-5 of Protocol 1. After approximately 48 h, add 500 ul 4.5 mg/ml MTX. 2. After approximately 17 h, centrifuge at 200 g for 7 min and remove the supernatant. 3. Adding 500 ml prewarmed medium (as in Protocol 1) and 30 mg/ml BrdU. 4. (Optional) Add 500 ul 670mg/mlActD after 3.5 h. 5. After 4.5 h add 500 ul 10 mg/ml Colcemid, mix, and incubate for a further 10 min. 6. Harvest as for Protocol 1, from step 6.

3.3 Isolation of suspensions of unfixed metaphase chromosomes Protocol 3, adapted from refs 20 and 28, takes advantage of the nuclear membrane breakdown that accompanies mitosis in vertebrates and the selectivity of the detergent digitonin for the cell membrane, to isolate a suspension of unfixed metaphase chromosomes. The structure of these chromosomes can then be manipulated, e.g. by the extraction of soluble chromosomal proteins (20), before they are fixed and used as substrates for FISH. Overnight Colcemid treatment of exponentially growing cells produces a high MI, and the condensed nature of the chromosomes so produced is maintained with a polyamine buffer. Chromosomes may be further purified over a sucrose step gradient. Protocol 3. Isolation of metaphase chromosome suspensions Equipment and reagents • Polyamine (PA) buffer: 15 mM Tris-HCI pH • Haemocytometer 7.2, 0.2 mM spermine (free base), 0.5 mM « Refrigerated variable speed rnicrocenspermidine (free base), 0.5 mM EGTA, 2 mM trifuge EDTA adjusted to pH 7.2 with 1 M KOH, 80 . Hypotonic 0.075 M KCI mM KCI, 20 mM NaCI. Prepare 2 X stock of . Swelling buffer (PME): 5 mM Pipes pH 7.2, fresh buffer weekly, and store at 4°C 5 mM NaCI, 5 mM MgCI2, 1 mM EGTA

63

Jeff Craig Protocol 3.

Continued

• Digitonin (Sigma); add 20 mg digitonin to 20 ml PA buffer, heat to 37°C for 15–20 min, filter (0.22 mm) to remove any undissolved digitonin then transfer to ice. Prepare fresh. . 10 m/ml Colcemid (Sigma) . PBS pH 7.3

• 30%, 40%, 50%, 60% sucrose in swelling buffer . •Glycerol . •6 ml polyallomer tubes .•3:1fixative(Protocol 1)

A. Crude chromosome preparation 1. Add Colcemid to 0.1 mg/ml to exponentially growing cells and leave overnight. 2. Harvest the cells: for monolayer cultures rounded-up mitotic cells can be detached by mitotic shake-off. Disperse any celt clumps by gently resuspending the cells in 10 ml PBS, then spin at 400 g for 5 min at room temp. 3. Resuspend the cells in 10 ml PBS and count using a haemocytometer. Centrifuge as in step 2 then resuspend the cells in hypotonic KCI solution at 2 x 106cells/ml. 4. Leave for 10 min at 37°C, centrifuge at 275 g for 5 min at room temp. 5. Resuspend the pellet in ice-cold PA buffer at 8 x 106 cells/ml.6 6. Withdraw a 0.5 ml aliquot to fix in 3:1 and spread onto slides to assess the Ml. 7. Spin the rest of the cells at 200 g at 4°C for 5 min. 8. Resuspend the pellet at 107 cells/ml in cold PA buffer containing 1 mg/ml digitonin.3 Vortex for 2 X 15 sec to burst the cell membranes and release the mitotic chromosomes. 9. Spin at 200 g for 10 min at 4°C to pellet the nuclei. Reserve the supernatant containing the chromosomes. 10. Add another 1 ml of PA buffer with digitonin to the nuclear pellet and respin, as many chromosomes get trapped in amongst the nuclei. 11. Pool the two supernatants. If further purification is required, proceed to part 6. 12. Transfer the chromosome suspension (the supernatant) to 1.5 ml microcentrifuge tubes and spin at 4000 g for 5 min at 4°C. 13. Resuspend the visible pellet of chromosomes in PA buffer containing 40% glycerol to 108–109 chromosomes/ml and store at -70°C. B. Purification of chromosome suspension using a sucrose step gradient (This method was modified from that described at http://skyemed.harvard.edu/pages/chrl.html) 1. Prepare a sucrose step gradient in 16 ml polyallomer tubes as follows: • 3 ml 60% sucrose in PME • 2 ml 40% sucrose in PME • 2 ml 50% sucrose in PME • 2 ml 30% sucrose in PME 64

4: Isolation of vertebrate metaphase chromosomes 2. Layer the supernatants from part A, step 11 onto the gradient and centrifuge at 4000 g in a swing-out rotor for 15 min at 4°C with no brake. 3. Aspirate until the middle of the 40% step, and collect the flocculent white material at the 40–50% and 50–60% interfaces using a Pasteur pipette. 4. Mix well to resuspend the chromosomes and store 10 ul aliquots at -70 °C. • Digitonin is toxic and should be handled with care. b CuSO4 may be added to 10-4 M to the PA buffer at this stage—it may help to stabilize the chromosome scaffold.

4. Spreading fixed chromosomes When cells fixed as in Protocol 1 are dropped onto glass slides, the cell membrane bursts and mitotic chromosomes are dispersed onto the slide. Environmental factors, particularly relative humidity, play a role in producing good spreads (29). In Protocol 4, a local humid environment is created with moist, warm towels. Clean slides are also imperative; most slides come 'precleaned' but need further cleaning. An additional short wash with ethanol/ether will suffice. The quality of spreading can be monitored by the extent and the evenness to which a drop spreads out over a slide, and by examining the slides by phase-contrast microscopy. A well-spread preparation will expand evenly to fill almost the whole slide; fix will evaporate evenly, and the chromosomes will be free of refractive cytoplasm. Protocol 4.

Making mitotic chromosome spreads

Equipment and reagents • Hot-block or slide warmer • Phase-contrast microscope with 10 x and 20 x objective lenses • Slide storage box, with desiccant . Absolute ethanol

• Ether (Sigma) • 3:1 methanol:acetic acid fixative (Protocol 7) • Ethanol series (Coplin jars of 70%, 90%, and 100% ethanol)

Method 1. Set a hot-block or slide warmer to 37°C. Cover with a cloth or paper towels and saturate with d(H2O. 2. In a fume hood, preclean the slides by dipping them briefly into a beaker containing 1:1 absolute ethanoliether and wiping them with a lint-free cloth. Muslin is best for this, but laboratory towels will suffice. 65

Jeff Craig Protocol 4.

Continued

Place a number of precleaned slides on the moist warm towels to warm up. 3. Resuspend chromosomes in the appropriate volume (approx. 5 ml for a 50 ml culture) of fresh fixative (the suspension should be slightly milky) and mix well. 4. Take up a small volume of fixed cells and drop onto a slide from a height of —30 cm. 5. Monitor spreading by eye, leave to dry, then examine under a phasecontrast microscope. 6. Label the slides with a pencil and mark, on the back of each slide, the extent of the spread chromosomes; leave to dry completely for 3-17 h. 7. Wash the slides through an ethanol series (2 min each) and dry before storage. This is thought to lessen depurination by acetic acid. 8. Freeze in a slide box containing desiccant. Alternatively, if slides are to be used within 3 days, leave covered at room temperature until needed.

5. Pretreatments of slides Slides may need to be pretreated before hybridization. Before hybridization to chromosomes fixed in 3:1 methanol:acetic acid, excess RNA can be removed from the slide (Protocol 5). Isolated and unfixed suspensions of metaphase chromosomes (Protocol 3) can be manipulated by extracting soluble chromosomal proteins under high-salt conditions (Protocol 6).

5.1 Pretreatment of mitotic chromosome spreads RNase pretreatment of slides is necessary if hybridization probes contain substantial amounts of coding DNA, and it also helps minimize the background signal from small single-copy probes. It is best to pretreat slides in batches before storing them at -20 °C. Protocol 5.

RNase treatment of slides

Equipment and reagents • 20 mg/ml RNase A (Sigma) in 10 mM • Tris-HCI pH 7.5, 15 mM NaCI. Boil for 10 min to destroy any DNase activity and store • in aliquots at –20°C. • 50 ml glass Coplin jars (or 200 ml glass troughs)

66

2 x SSC: 300 mM NaCI, 30 mM trisodium citrate pH 7.0 Ethanol series: Coplin jars containing 70%, 90%, and 100% ethanol

4: Isolation of vertebrate metaphase chromosomes Method 1. If the slides are stored at -20°C, remove and leave them to thaw at room temperature in their container for 30 min. 2. Fill a 50 ml Coplin jar with 2 x SSC and place in a 37°C water-bath. Add 250 ml 20 mg/ml RNase to the prewarmed 2 x SSC. Incubate the slides in the RNase solution at 37°C for 1 h. 3. Wash the slides for 3 X 5 min in 2 x SSC at room temperature. 4. Take the slides through an ethanol series (2 min each in 70%, 90%, and 100% ethanol) and air-dry.

5.2 Salt extraction of isolated metaphase chromosomes After extracting proteins under high-salt concentrations, vertebrate mitotic chromosomes retain a characteristic shape when examined by electron or light microscopy (20, 30, 31). A central chromosome scaffold running the length of the chromosome is surrounded by a halo of DNA. FISH has be used to examine the spatial relationship between denned DNA sequences and the axial chromosome scaffold produced after salt extraction of metaphase chromosomes (20). This allows visual analysis of the topology of genomic DNA within the mitotic chromosome, and is analogous to similar extractions of interphase nuclei that have been used to examine the association of specific sequences with the nuclear matrix (ref 32 and Chapter 6). Salt extraction is performed by lowering the slides, in a horizontal position, into a buffer containing varying concentrations of NaCl. Agitation of the chromosomes prior to fixation must be minimized to prevent streaking of chromatin fibres. It was found that the best way to achieve this was to construct a platform from a metal mesh (see Figure 1). Protocol 6. Salt extraction of isolated metaphase chromosomes Equipment and reagents • Plastic trays in which to treat slides • Platform on which to transfer the slides between buffers (e.g. made from a folded piece of metal mesh, see Figure 7) . Various concentrations of NaCl in CIBa

• Chromosome isolation buffer (CIB): 10 mM Tris-HCI pH 8.0, 10 mM EDTA pH 8.0, 0.1% Nonidet P-40, 0.1 mM CuSO4 20 mg/ml PMSF • 3:1 methanol:acetic acid (Protocol 1)

Method 1. Smear 10–15 ml of the unfixed chromosome suspension from Protocol 3 across the slide using the side of a yellow pipette tip. Cover and allow the chromosomes to settle onto the slide overnight. 2. Place the slides, sample uppermost, onto the wire mesh and gently lower into 150 ml CIB solution (Figure 1). Leave for 5 min.

67

Jeff Craig Protocol 6. Continued 3. Transfer the slides carefully to a CIB solution containing 0.5 M NaCI and leave for 5 min. 4. Repeat step 3, increasing the NaCI concentration until the desired concentration is reached.3 5. Transfer the slides carefully into fresh 3:1 fixative. Leave for 10 min. 6. Repeat step 5 with fresh fixative. 7. Remove the slides from the fixative and air-dry. • In practice, it was found that 1.8 M NaCI is the maximum concentration that still gives a reasonable FISH signal.

After the extraction of metaphase chromosomes there is no need to pretreat the slides with RNase prior to hybridization. During the hybridization of such preparations it is important that the slide denaturation time is kept to a minimum and that the pH of the slide denaturant is checked, to preserve the delicate chromosome morphology (Protocol 9). Extractions can also be performed with other reagents, e.g. 25 mM lithium 3,5-diiodosalicylic acid (LIS) or NH4SO4, but the resulting decondensed chromosomes are too fragile to retain their morphology after FISH. Extracted chromosomes can

Slides

Figure 1. Equipment set-up for salt extraction of isolated metaphase chromosomes. Slides are lowered into extraction buffer (Protocol 3) on a folded metal mesh to minimize displacement of extracted chromatin.

68

4: Isolation of vertebrate metaphase chromosomes also be fixed in other ways, e.g. with 4% paraformaldehyde and are suitable substrates for immunofluorescence (Chapter 5).

6. Labelling DNA probes DNA probes for FISH can range in size and complexity from short probes of <1 kb, through whole plasmids, cosmids, BACs, and YACs, up to complex probe sets representing chromosome subregions, whole chromosomes, and entire genomes. After labelling, probe size should be <600 bp. Probes longer than 1 kb before labelling are usually labelled by nick translation (Chapter 3), those <1 kb can be labelled by random priming. Probes may also be labelled by PCR provided that the amplification products remain in the 400–1000 bp size range (Section 6.4)

6.1 Choice of label Labels are dNTP analogues joined covalently via a hydrocarbon linker arm to a molecule that can act as a hapten. In direct labelling, this hapten is a fluorochrome (e.g. FITC, Cy3). In indirect labelling, the hapten is a nonfluorescent molecule (e.g. biotin, digoxigenin) which can be detected using a fluorochrome attached to a molecule with a high affinity to the indirect label (e.g. avidin for biotin or anti-digoxigenin antibody). For most probes, especially repetitive or complex ones such as chromosome paints, direct labelling is sufficient. Probes that are smaller benefit from the signal amplification afforded by indirect labelling. Of the green fluorochromes, FITC is the most widely used and generally considered the best. Of the red fluorochromes, TRITC and rhodamine have traditionally been used, though Cy3, which is brighter, is now the label of choice. Texas Red, another red fluorochrome has significantly different spectral properties to TRITC, rhodamine, and Cy3 and cannot usually be detected with the same filter sets. The best and most commonly used dNTPs are dUTPs with a long linker length, e.g. fluorescein–12-dUTP, biotin-16dUTP, and digoxigenin-11-dUTP.

6.2 Nick translation The basic protocol for labelling by nick translation is described in Chapter 3 (Protocol /). Longer probes, e.g. BACs and cosmids require slightly more DNase (up to 2-3fold) than shorter ones. In addition, digoxigenin- and FITCconjugated dNTPs need slightly more (1-1.5 X) DNase than do reactions using biotin as a label, Cy3 needs slightly less (0.5-1 x). These dilutions are worked out empirically by checking the probe length after labelling (Chapter 3, Protocol 2). Unincorporated label should be removed by column chromatography through Sephadex G-50 (Chapter 3, Protocol 1). There are also a number of 69

Jeff Craig premade columns, the most convenient of which is the Bio-Rad Micro BioSpin 6 Tris column. These columns consist of acrylamide beads preswollen in 10 mM Tris-HCl, pH 7.4, and require a microcentrifuge capable of 1000 g. After draining the excess buffer, the labelled probe is spun through the columns in 4 min.

6.3 Random priming Random priming is used for probes too small to be labelled by nick translation (200–600 bp). The constituents of the random prime reaction are available as a kit (Boehringer Mannheim). Protocol 7.

Probe labelling by random priming

Equipment and reagents • TE buffer: 10 mM Tris-HCl, 1 mM EDTA pH • Random priming kit (Boehringer Mann8.0 heim). This contains tubes of each dNTP at • Boiling water-bath or 96°C hot-block/ther0.5 mM, 'reaction mixture' (hexanucleotide mal cycler mixture in 10 x reaction buffer), 2 U/ul .1 mM biotin–16-dUTP, digoxigenin-11Klenow DNA polymerase. Mix the dNTPs in the dUTP, fluorescein-12-dUTP (Boehringer ratio of,,1 dGTP:1 dCTP:1 dATP:°.8 dTTP. Store Mannheim), or Cy3-dUTP (Amersham) all reagents at -20°C. • Stop mix: 0.1% (w/v) Bromophenol Blue, 0.5% (w/v) Dextran Blue, 0.1 M NaCI, 20 mM EDTA pH 8.0

Method 1. Denature 1 mg of probe in 12 ul TE by heating in a microcentrifuge tube to 96°C for 5 min. Cool rapidly on ice.

2. Add: • 2 ml 'reaction mixture' • 3.8mldNTP mix • 0.8 ml of labelled dNTP, and • 1 ml 2 U/ml Klenow polymerase. Vortex, spin briefly, and incubate at 37°C for 1 h. 3. Add 80 ul of stop mix and remove unincorporated nucleotides (as in Chapter 3, Protocol 1). The labelled probe can be stored at -20°C.

6.4 Labelling by PCR Probes can be labelled directly during PCR by the inclusion of labelled dNTP into the reaction mix. The best candidates for labelling this way are those whose amplification products are the right size for FISH (<1 kb for chromosome paints and genomic DNA, <600 bp for other probes). If the labelled amplified DNA is too long, it can be reduced in size using DNase I. Biotinand digoxigenin-labelled dNTPs incorporate well in PCR reactions. Direct 70

4: Isolation of vertebrate metaphase chromosomes labels can give variable results, probably due to the impaired ability of Tag polymerases to incorporate nucleotide analogues with bulky and hydrophobic adducts. The ratio of the labelled:unlabelled dNTP is important. In a standard PCR reaction in which dATP, dCTP, and dGTP are all present at a final concentration of 200 mm, final concentrations of 80 (mm dTTP and 40 mm labelled dUTP give good incorporation.

6.5 Quantifying label incorporation To check labelling, probes are diluted to the same concentration as control labelled DNA and spotted onto nitrocellulose filters. Filters are then incubated with an alkaline phosphatase-conjugated label-specific molecule and detected with the substrates BCIP (5-bromo-4-chloro-3-indoyl-phosphate) and NBT (4-Nitro blue tetrazolium chloride). The intensity of the blue colour on the filter is proportional to incorporation of the label. Protocol 8. Measuring label incorporation Equipment and reagents • Standard UV transilluminator or UV crosslinker • Gridded nitrocellulose membrane filters (e.g. 85 mm diameter, circular Protran filters, Schleicher and Schuell) • Small, light-tight plastic container to fit the filters • Control labelled DNA: biotin-labelled control DNA (Gibco) or digoxigenin-labelled control DNA (Boehringer Mannheim). Dilute to 1 ng/ml and 0.1 ng/ml and store at 4°C. •TE buffer (Protocol 7) • NBT and BCIP (Gibco BRL)

. Buffer AP1: 0.1 M Tris-HCI, pH 7.5, 0.1 M NaCI, 2 mM MgCI2, 0.05% Triton X-100 « Buffer AP2: 3% BSA fraction V in buffer AP1 + 0.04% sodium azide • Buffer AP3: 0.1 M Tris-HCI pH 9.5, 0.1 M NaCI, 50 mM MgCI2 • 0.75 U/ml streptavidin-alkaline phosphatase conjugate (SAAP, Boehringer Mannheim) • 0.75 U/ml anti-digoxigenin-AP Fab fragments (Boehringer Mannheim) . 0.75 U/ml anti-FITC-AP Fab fragments (Boehringer Mannheim)

Method 1. Label the grid in ballpoint pen to form a table where each row represents a different labelled probe (or control DNA) and each column (labelled '1'and '0.1') one of the two serial probe dilutions. 2. Dilute each probe to approximately 1 ng/ml and 0.1 ng/ul as follows. Label a strip of Parafilm in the same way as the gridded filter. Pipette onto the Parafilm, two 9 ul spots of TE buffer under the columns marked '1' and '0.1'; add 1 ul probe to the first 9 ul, mix by repeated pipetting, transfer 1 ul of this to the 2nd spot of TE buffer and mix in the same way. 3. Pipette 1 ul of each dilution to form a spot within the appropriate square on the gridded nitrocellulose filter. In the same way, spot 1 ul of 1 ng/ml and 0.1 ng/ml control labelled DNA onto the filter. Dry at room temperature for 10-15 min. 4. Cross-link with UV light for 2 min. 71

Jeff Craig Protocol 8.

Continued

5. Incubate the filter in 10 ml AP1 with shaking. 6. Discard the AP1 and block the filter by incubating in AP2 for 10 min. 7. Dilute the appropriate alkaline phosphatase conjugate in 10 ml AP1. For biotin-labelled probes use 1.5 ul SAAP, for detecting digoxigenin use 2 ml anti-digoxigenin-AP; for FITC-labelled probes use 2 ul antiFITC-AP. 8. Discard the AP2 buffer and incubate with AP conjugate for 10 min. 9. Wash twice for 2 min each time in AP1 then in APS3 10. Make up the colour development solution: to 5 ml AP3, add 16.6 ul BCIP and 22 ul NBT and mix well. Discard the last wash solution, add the colour development solution and incubate with shaking. A colour reaction will take between 5 min and 1 h to develop. Stop the reaction by washing briefly with distilled water. Air-dry and compare the intensity of probe spots with each other and with the control spots.a a If the labelled probes were originally 10 ng/ml, their spots should have the same intensity as those of the control.

7. Hybridization 7.1 Preparation of probes and slides DNA probes are precipitated with ethanol and resuspended in hybridization buffer (typically 50% formamide and 10% dextran sulfate in 1 X SSC), the dyes present in the labelling stop mixes act as a visible marker of precipitation and resuspension. Carrier DNA, such as sonicated salmon sperm DNA, and unlabelled DNA enriched in repetitive sequences are added, if necessary, to the probe prior to precipitation (see Section 7.1.3). After resuspending the probe mix, the DNA is denatured at 75 °C and repetitive DNA left to reanneal if necessary, before adding the probe to slides (Protocol 9). 7.1.1 Slides Slides are denatured in 70% formamide in 2 X SSC for 1.5 min at 72°C; recently prepared slides may require less time than this, very old slides more. If in doubt, denature a test slide and check for chromosome morphology. If the chromosomes look 'puffy', the denaturation time should be reduced. 7.1.2 Probe DNA Labelled DNA probes are generally used at a concentration of 25 ng/m1 in hybridization buffer (250 ng probe in 10 ul hybridization buffer). Probes labelled in different ways can be mixed together, as long as the concentration of each probe remains at 25 ng/ml. Some commercially available probes are sold already dissolved in hybridization buffer and ready to denature. 72

4: Isolation of vertebrate metaphase chromosomes 7.1.3 Suppressing hybridization from repetitive DNA Many FISH probes contain repetitive DNA. If only the single-copy component of the probe is to be visualized, hybridization between repetitive elements in the probe and target chromosomes must be suppressed. This is achieved using an excess of unlabelled DNA enriched in repeats. Cot-1 DNA is widely used for this purpose and is available commercially (Gibco BRL) (33-35). Cot-1 is usually used in 50–100fold excess over probe DNA, i.e. 12.5-25 mg for 250 ng of the labelled probe, but the precise amount depends on the number of repeats within a probe. Protocol 9 recommends reannealing the probe and Cot-1 DNA for 30 min; this can be increased to 2 h if the required degree of suppression is not achieved initially. Some commercially available probes come preannealed with Cot-1 or similar DNA.

7.2 Hybridization Kinetics of hybridization relate to the ratio of the probe to the target and the degree of homology between them. Repetitive DNAs will hybridize within 0.5–4 h. Chromosome paints will hybridize within 4 h, but the hybridization signal improves after overnight hybridization. For single-copy probes, overnight hybridization is usually sufficient. However, for applications where optimal hybridization is important, hybridizations are left for up to 3 days. Interspecies hybridization usually requires 3-5 days' hybridization due to the low degree of homology between the target and the probe (16). Protocol 9.

Hybridization of probes to slides

Equipment and reagents • Water-baths at 37°C and 75°C • Rotary vacuum dryer (e.g. Spin-Vac, Savant) . Shaking hot-block (e.g. Eppendorf Thermomixer) . Coverslips (22 x 22 mm or smaller) . Ethanol series (Coplin jars containing 70%, 90% and 100% ethanol) • Absolute ethanol, prechilled to –20°C • 10 mg/ml sonicated salmon sperm DNA (Sigma)

• Slide denaturant: 70% formamide in 2 x SSC (Protocol 5); adjust pH to 7.0 with 1 M HCI and store at -20 °C • 1mg/mlCot-1 DNA (Gibco BRL) . Hybridization mastermix: 20% dextran sulfate in 2 x SSC pH 7.0; filter (45mm)and store at–20°C in 1 ml aliquots • Deionized formamide; store at -20°C in 1 ml aliquots • Rubber cement, e.g. TipTop, Fixogum, Pang

Method 1. Add together: probe(s), 50 x excess Cot-1 DNA if needed, and 5ml 10 mg/mlsalmon sperm DNA. Add 2.5 volumes of cold (–20°C) absolute ethanol. 2. Precipitate DNA mixture at –80°C for 15 min or -20°C for 30 min. 3. If slides are stored at -20°C, remove and leave to thaw at room temperature in their container for 30 min. 73

Jeff Craig Protocol 9. Continued 4. Transfer 50 ml of the slide denaturant to a glass Coplin jar and place in a small water bath, Turn the water-bath to 75°C.a Place the 70% ethanol on ice. Thaw out aliquots of the deionized formamide and the hybridization mastermix and keep at room temperature. 5. Microcentrifuge the precipitated DNA mix at maximum speed for 15 min at room temperature. 6. As soon as the centrifuge has stopped, remove up to six tubes, quickly pour off the supernatant and leave the tubes inverted on absorbent paper for 2-3 min. If there are more than six tubes, bring the centrifuge up to full speed, stop, and take the next six tubes out. 7. Dry the DNA pellets, using a rotary vacuum dryer or a hot block at 45°C.b 8. Add 5.5 ml deionized formamidec to the pellet and dissolve the DNA at 45°C using a hot-block, preferably a rotary dryer. Vortex briefly every 5 min. The probes will dissolve in 15-20 min. 9. After a brief spin, add 5.5 ml of the hybridization mastermix. Mix with a pipette tip and by vortexing; spin briefly again. 10. Denature the probes at 75°C for 5 min. 11. If the hybridization mixture contains Cot-1 DNA, preanneal the probe by incubating at 37°C for 30 min. 12. Preheat the slides in batches of 2-4 for 1 min using a 72°C hot-block or a small oven.a 13. Check the temperature of the denaturation solution, adjust to 72°C if necessary, and denature the slides in batches of two to four for 1–1.5 min.a 14. Take the slides through an ethanol series (including the prechilled 70% ethanol) and air-dry. 15. On a piece of absorbent paper, lay down the appropriate number of 22 x 22 mm coverslips and pipette the probe onto these. 16. With each slide in turn, pick up the coverslip and probe. Gently squeeze out any air bubbles and seal with rubber cement. 17. Place the slides into a metal tray with a non-airtight lid, in a 37°C waterbath, or, alternatively, in a 37°C incubator overnight (Section 7.2). a The desired temperature for slide denaturation is 72°C and it is critical that this temperature is accurate. Always measure the temperature of the formamide solution directly within the Coplin jar, which will always be less than the water-bath temperature. Preheating the slides to 72°C circumvents the problem of cold slides cooling the temperature of the formamide. It is best to titrate the denaturation time for each batch of slides, as fresher slides may need a shorter denaturation time. Overdenaturatuion results in 'puffy' chromosomes. b lt is important not to over-dry the DNA pellet at this stage. Check the tubes after intervals of 2 min and remove when dry. Drying using a 45 °C hot-block takes longer (10-20 min). c The DNA will dissolve quicker if the formamide is added first. Using 5.5 ul of formamide and mastermix, loss due to the viscosity of the hybridization solution, usually results in a total of total.

74

4: Isolation of vertebrate metaphase chromosomes

8. Detecting hybridized probe After hybridization, slides are washed in 2 X SSC to loosen the coverslips and wash off the hybridization mixture. The slides are then washed at a higher stringency to remove non-specifically bound probe. Sufficient amounts of washing buffers are brought up to the relevant temperature in preheated water-baths. Biotin is detected with avidin-fluorochrome conjugates and amplified with anti-avidin antibodies. Digoxigenin is detected with anti-digoxigenin antibodies and amplified with anti-idiotype antibodies. One, two, or three layers of reagents may be used to increase signal amplification (Table 1), but multiple layers may result in a higher level of background. Probes such as chromosome paints usually need only one layer of detection, short single-copy probes may need three. Table 1 lists the different ways of detecting biotin and digoxigenin recommended in Protocol 10 and also gives the recommended reagent dilutions. Biotin and digoxigenin can be detected simultaneously if care is taken to use antibodies that do not cross-react. Digoxigenin usually gives a slightly stronger signal than biotin with the same fluorochrome. Detection reagents may be obtained from a number of companies including Amersham, Boehringer Mannheim, Jackson ImmunoResearch, Sigma, and Vector laboratories. Traditionally, glass coverslips have been used for incubations. However, plastic coverslips may be used instead; they are thought to adsorb less antibody, float to the surface during washes, and can be made from Parafilm or autoclave bags. Slides should not be allowed to dry at any time during detection otherwise background will result. Table 1. Basic detection strategies for biotin and digoxigenin Label

First layer*

Second layer

Third layer

Biotin

Avidin-Cy3b Avidin-FITCc

— Goat anti-avidin-FITCc Biotinylated goat anti-avidinc —

— —

Sheep anti-mouseCy3b

—

Avidin-FITCc Digoxigenin

Sheep anti-digoxigenin– FITCC Mouse anti-digoxigeninb

Avidin-FITCc —

" For combined detection of biotin and digoxigenin, combine the reagents for the one-layer protocols, the two-layer protocols, or combine the biotin three-layer protocol with the digoxigenin two-layer protocol. bDilute 1:500 in 4 x SSC/Tween. "Dilute 1:200 in 4 x SSC/Tween.

75

Jeff Craig Protocol 10. Washing and detection Equipment and reagents • Water-baths • Lightproof incubation box, lined with moist tissue paper and big enough to house up to nine slides • Perspex incubation hood big enough to fit a small rotating platform and the incubation box, or a cell-culture incubator . 50 ml Coplin jars . 50 x 22 mm coverslips (glass or plastic)

• 1.5 ml lightproof (amber) microcentrifuge tubes • Wash buffers: 2 x SSC and 0.2 x SSC both at pH 7.0 . 4 x SSC/0.1% Tween-20 pH 7.0 » BSA fraction V (Sigma) .Blockingsolution. 3% (w/v) BSA in 4 x SSC/0.1% Tween . Detection reagentsa (Table 7); store at 4°C

Method 1. Prewarm 250 ml of 2 x SSC to 37°C, 0.2 x SSC to 60°C, and 600 ml of 4 x SSC/0.1% Tween-20 to 45°C,b in 50 ml Coplin jars (2 jars of 2 x SSC at 37°C). Prewarm 50 ml of the blocking solution to 37 °C. 2. With each slide in turn, carefully peel off the rubber solution with forceps. Place the slides, with their coverslips still on, into a Coplin jar of 2 x SSC at 37 °C. 3. Wash slides for 5 min.c This loosens the coverslips and minimizes damage to the specimen. 4. Transfer each slide in turn into a fresh Coplin jar containing 2 X SSC at 37 °C. If coverslips have not dropped off, remove carefully. 5. Wash for 3 X 5 min.0 6. Directly labelled probes need no further washing. For indirectly labelled probes, transfer the slides to the Coplin jar containing 0.2 x SSC at 60°C and wash for 3 x 5 min. 7. Wash the slides briefly in 4 x SSC/0.1% Tween-20.d 8. Add the prewarmed blocking solution and incubate at 37°C for 15 min. 9. During blocking, dilute the relevant detection reagents (Table 1) in 1.5 ml lightproof microcentrifuge tubes; using 200 ul per slide, estimate the total amount required. Spin at maximum speed for 3 min and transfer to ice. 10. On a piece of absorbent paper, lay down the appropriate number of 50 x 22 mm coverslips and pipette onto these 150 ul of the first layer of detection reagents. 11. With each slide in turn, remove from the blocking solution, shake off excess buffer, pick up the coverslip and antibody and transfer to the lightproof incubation box. Incubate at 37°C for 30 min. 12. Carefully remove the coverslips and transfer the slides to 4 x SSC/0.1% Tween-20 at 37°C. Wash 3x5 min. 76

4: Isolation of vertebrate metaphase chromosomes 13. Repeat steps 10–12 for each layer of detection. The slides are now ready for counterstaining and mounting. ' Use in combination and dilute as in Table 1. b Protect slides from light as much as possible when washing slides with directly labelled probes, and washing between antibodies. This can be done by placing an upturned lightproof container over the Coplin jars. c All wash steps are carried out on the shaker platform, if possible within a Perspex incubator. d Slides can be left for up to 2 h in 4 x SSC/0.1% Tween-20 if needed.

9. Counterstaining and mounting After FISH, chromosomes are stained with a DNA-specific fluorescent dye, usually DAPI (4, 6-diamidino-2-phenylindole), Hoechst stain, or propidium iodide (PI). The choice of counterstain depends on the spectral overlap with the fluorochromes used as labels, the capability of the fluorescence microscope, and whether chromosome banding is required.

9.1 Simple counterstaining Slides can be mounted in a solution of the counterstain in a mountant, 1-2 mg/ml PI or 2-10 mg/ml DAPI. Low concentrations of counterstain produce the best chromosome banding. Because of the decondensed nature of chromosomes that have been extracted with high-salt (Protocol 6}, high concentrations (10 mg/ml) of DAPI are needed to see the halo of DNA loops that surround the chromosome core. A better method of DAPI staining is to incubate the slides for 2-3 min in DAPI diluted to 25 ng/ml in 2 X SCC. Slides can then be mounted after briefly washing with dH2O twice then drying for 3-5 min.

9.2 Chromosome banding DAPI itself produces a weak G-banding pattern, which can be enhanced with the non-fluorescent dye ActD (36) (Protocol 11). A similar technique uses a mixture of 1 mg/ml DAPI and 1.2 mg/ml PI in antifade. Chromosomes are normally stained uniformly with PI, but a mixture of PI and DAPI viewed with a red filter produces R bands, due to the competition between binding sites in G bands. After the incorporation of BrdU in early S-phase (R-bands, ref. 37) or late S-phase (G-bands, Protocol 2) cells, anti-BrdU antibodies can be used to highlight these bands. Incubate the slides with FITC-conjugated anti-BrdU (Boehringer) diluted to 10 mg/ml during the probe detection steps in Protocol 10, ensuring that there is no cross-reactivity to other antibodies being used. BrdU can also detected with other fluorochromes by consecutive incubations with anti-BrdU and fluorochrome-conjugated anti-idiotype antibody. 77

Jeff Craig Protocol 11. DAPI/actinomycin D banding Equipment and reagents • 2 x 50 ml Coplin jars . Lightproof incubation box, lined with moist tissue paper and big enough to house up to nine slides • 50 mg/ml DAPI (Sigma)

• Mcllvaine's buffer: add 0.15 M citric acid to 500 ml 0.15 M Na2HP04 until the pH is 7.0 • 10 mg/ml actinomycin D (Sigma), in methanol

Method 1. Dilute 25 ul ActD to 0.25 mg/ml with 975 M1 Mcllvaine's buffer, and dilute 2 ul DAPI stock solution to 0.1 mg/ml with 1 ml Mcllvaine's buffer. 2. Incubate the slides in 50 ml Mcllvaine's buffer in a Coplin jar for 5 min. 3. On a piece of absorbent paper, lay down the appropriate number of 50 x 22 mm coverslips and pipette onto these 80 ul 0.25 mg/ml ActD. 4. With each slide in turn, pick up the coverslip and transfer to the lightproof incubation box. Incubate at room temperature for 15 min. 5. Carefully remove the coverslips and wash briefly in Coplin jars of distilled water, then Mcllvaine's buffer. 6. Incubate the slides with 0.1 ug/ml DAPI, stain for 5 min, and wash as above. Mount.

Fluorochromes fade, depending on the length of exposure to light and the mounting medium. It is therefore important to mount in an antifade. A simple but effective antifade is AFT10 (Citifluor) To make this up, dissolve 1 tablet in 2.5 ml PBS, add 2.5 ml non-fluorescent glycerol (Merck), mix well, and store in 1 ml aliquots at 4°C. Ready-made antifades are also available commercially, and one of the best is Vectashield (Vector Laboratories).

Acknowledgements I thank Irina Solovei, Michael Speicher, Jurgen Kraus, Anna Jauch, Wendy Bickmore, and Judy Fantes for help in formulating the protocols.

References 1. Inazawa, J., Saito, H., Ariyama, T., Abe, T., and Nakamura, Y. (1993). Genomics, 17,153. 2. Pappas, G.J., Polymeropoulos, M.H., Boyle J.M., and Trent, J.M. (1995). Genomics, 25,124.

78

4: Isolation of vertebrate metaphase chromosomes 3. Bray-Ward, P., Menninger, J., Lieman, J., Desai, T., Mokady, N., Banks, A., and Ward, D.C. (1996). Genomics, 32,1. 4. Taruscio, D. and Manuelidis, L. (1991). Chromosoma, 101, 141. 5. Murnane, J.P. and Yu, L.-C. (1993). Mol. Cell Biol., 13, 977. 6. Heng, H.H. Q., Tsui, L.-C., and Moens, P.B. (1994). Chromosoma, 103, 401. 7. Warburton, P.E., and Cooke, H.J. (1997). Chromosoma, 106,149. 8. Korenberg, J.R. and Rykowski, M.C. (1988). Cell, 53,391. 9. Boyle, A.L., Ballard, S.G., and Ward, D.C. (1990). Proc. NatlAcad. Sci. USA, 89, 4913. 10. Haaf, T., Sirigu, G., Kidd, K.K., and Ward, D.C. (1996). Nature Genet., 12, 183. 11. Craig, J.M., and Bickmore, W.A. (1994). Nature Genet., 7,376. 12. Saccone, S., De Sario, A., Wiegant, J., Raap, A.K., Delia Valle, G., and Bernardi, G. (1993). Proc. NatlAcad. Sci. USA, 90,11929. 13. Craig, J.M., Boyle, S., Perry, P., and Bickmore, W.A. (1997). J. Cell Sci., 110, 2673. 14. Schrock, E., du Manoir, S., Veldman, T., Schoell, B., Wienberg, J., FergusonSmith, M.A., Ning, Y., Ledbetter, D.H., Bar-Am, I., Soenksen, D., Garini, Y., and Ried, T. (1996). Science, 273, 494. 15. Speicher, M.R., Ballard, S.G., and Ward, D.C. (1996). Nature Genet., 12, 368. 16. Wienberg, J. and Stanyon, R. (1995). Curr. Opin. Genet Dev., 5, 792. 17. Scherthan, H., Cremer, T., Arnason, U., Weier, H-U., Lima-de-Faria, A., and Fronicke, L. (1994). Nature Genet., 6, 342. 18. Baumgartner, M., Dutrillaux, B., Lemieux, N., Lilienbaum, A., Paulin, D., and Viegas-Pequignot, E. (1991). Cell, 64, 761. 19. Lemieux, N., Malfoy, B., Fetni, R., Muleris, M., Vogt, N., Richer, C.L., and Dutrillaux, B. (1994). Cytogenet. Cell Genet., 66, 107. 20. Bickmore, W.A. and Oghene, K. (1997). Cell, 84, 95. 21. Kallioniemi, A., Kallioniemi, O.P., Sudar, D., Rutovitz, D., and Gray, J.W. (1992). Science, 258, 818. 22. Kalliomiemi, O.-P. (1997). Nature Genet., 15, 5. 23. Ledbetter, D.H. (1992). Am. J. Hum. Genet., 51, 451. 24. Meltzer, P.S., Guan, X.-Y., and Trent, J.M. (1993). Nature Genet., 4, 252. 25. Fantes, J.A., Bickmore, W.A., Fletcher, J.M., Ballesta, F., Hanson, I.M., and van Heyningen, V. (1992). Am. J. Hum. Genet., 51,1286. 26. Cremer, T., Popp, S., Emmerich, P., Lichter, P., and Cremer, C. (1990). Cytometry, 11,110. 27. Yunis, J.J. (1981). Hum. Genet., 56, 293. 28. Blumenthal, A.B., Dieden, J.D, Kapp, L.N. and Sedat, J.W. (1979). J. Cell Biol., 81, 255 29. Hlisics, R., Muhlig, P., and Claussen, U. (1997). Cytogenet. Cell Genet., 76, 167. 30. Paulson, J.R. and Laemmli, U.K. (1977). Cell, 12, 817. 31. Jeppesen, P. and Morten, H. (1985). J. Cell Sci., 73, 245. 32. Gerdes, M.G., Carter, K.C., Moen, P.T., and Lawrence, J.B. (1994) J. Cell Biol. 126, 289. 33. Landegent, J.E., Jansen in de Wal, N., Dirks, R.W., Baas, F., and van der Ploeg, M. (1987). Hum. Genet., 77, 366. 34. Lichter, P., Cremer, T., Borden, J., Manuelidis, L., and Ward, D.C. (1988). Hum. Genet, 80, 224. 79

Jeff Craig 35. Pinkel, D., Landegent, J., Collins, C., Fuscoe, J., Segraves, R., Lucas, J., and Gray, J.W. (1988). Proc. NatlAcad. Sci. USA, 85, 9138. 36. Schweizer, D. (1976). Chromosoma, 58, 307. 37. Rooney, D.E. and Czepulkowski, B.H. (ed.) (1986). Human cytogenetics. A practical approach, IRL press, Oxford.

80

5

Studying the progression of vertebrate chromosomes through mitosis by immunofluorescence and FISH BETH A. SULLIVAN and PETER E. WARBURTON

1. Introduction Mitosis is a dynamic process in which replicated chromosomes, that have aligned at the metaphase plate, move along spindle microtubules (MTs) and become equally distributed into daughter cells. This chapter describes several methods used to study chromosome structure and segregation in mitosis, particularly during anaphase and telophase. Incorporating FISH and immunofluorescence, the protocols include the treatment of cells with drugs that reversibly affect spindle dynamics and cell cleavage, as well as immunofluorescence techniques on unblocked cells or unfixed cells. Advantages to each approach are discussed with respect to cell type and growth characteristics. The techniques are generally suitable for basic fluorescence microscopy, however confocal microscopy and imaging using charge-coupled devices (CCDs) have also been used to analyse cells prepared according to the protocols in this chapter.

2. Fundamental aspects of mitosis 2.1 The mitotic spindle At metaphase, the mitotic spindle is a predominating structure in the cell. The spindle is composed of several kinds of MTs—long filaments containing linear, polarized tubulin subunits, which function in various structural and functional capacities. The spindle provides the framework onto which the chromosomes will attach, and is associated with some of the motors which move the chromosomes (1). There are kinetochore MTs and non-kinetochore MTs. The kinetochore, a trilaminar proteinaceous structure located on either

Beth A. Sullivan and Peter E. Warburton side of the centromere, secures the chromosome to the spindle and is necessary for chromosome movement along the spindle. Kinetochore MTs are attached at one end to a kinetochore, while the other end can be attached at the spindle pole or just short of the pole. Instead of associating with the spindle pole, the other end may establish lateral contact with non-kinetochore MTs (1). Non-kinetochore MTs can be polar (one end at a pole, the other end free), astral (one end at a pole, the other outside the spindle body proper), interdigitating (one end at a pole, the other end laterally associated with MTs from the other pole), continuous (one end at each pole), or free (neither end associated with a pole nor a chromosome). These many types of MTs contribute to form a structure that is dynamic, characterized by constant polymerization and depolymerization (addition and subtraction of tubulin subunits), and repeated capturing and release by kinetochores. The mitotic spindle first appears in prometaphase when MT organizing centres, or spindle poles, move away from each other to occupy sites at opposite sides of the cell. The nuclear envelope breaks down, and MTs that have been assembled from each pole are captured by kinetochores. Successful kinetochore attachment is crucial to mitotic progression, since it creates tension which is an important indicator that chromosomes have properly attached to the spindle and are prepared for segregation (2). Cell then enter anaphase, during which time MTs depolymerize, or shorten, to permit chromosome migration. Anaphase consists of two phases: anaphase A when chromosomes move to the poles; and anaphase B during which the spindle elongates (1). Both phases involve intricate and carefully orchestrated mechanisms that generate poleward forces and MT growth, sliding, and depolymerization. By telophase, sister chrornatids have reached the spindle poles, the spindle disassembles, and cytokinesis begins, thus dividing the mother cell into two daughters, each containing identical chromosome complements.

2.2 Chromosomes Proper segregation of replicated chromosomes is controlled, in large part, by the centromere. Cytologically defined on each chromosome as the primary constriction, it is the site of sister chromatid cohesion until their separation during anaphase and is the point of chromosome attachment to the mitotic spindle (3). Several proteins termed CENPs (centromeric proteins), reacting with antibodies from the sera of patients with autoimmune diseases such as CREST syndrome (calcinosis, Raynaud's phenomenon, (o)esophageal dysmotility, scleroderma and telangectasia) preferentially localize to, and contribute to, structural and functional aspects of the centromere and/or kinetochore (4). CENP-A is a DNA-binding protein with homology to histone H3 (5). CENP-B is a DNA-binding protein, recognizing a 17-bp motif, the CENP-B box, in alpha satellite DNA (6). CENP-C is a structural protein of the inner kinetochore plate with DNA-binding potential (7), CENP-E is a 82

5: Studying the progression of vertebrate chromosomes protein located in the outer plate and fibrous corona and is transiently associated with the kinetochore (8, 9). It contains a kinesin-like motor domain and is involved in chromosome congression during metaphase and chromosome movement in anaphase, presumably due to a minus-end (poleward) motor activity (10). Injection of antibodies to CENP-E into cells both delays the onset of anaphase and inhibits poleward chromosome migration, indicating that CENP-E plays an important role in directing chromosome segregation during anaphase (8,11). CENP-F, another protein located in the outer kinetochore, accumulates in G2 and may be involved in early kinetochore assembly (12). The centromere is also involved in the regulation of mitosis. Attachment of chromosomes to the spindle is monitored by a tension-sensing mechanism at the centromere (2). An epitope at the centromere remains phosphorylated until tension is established at all kinetochores by attachment to the spindle (13). Once the kinetochores are firmly associated with spindle MTs, the epitope is dephosphorylated and anaphase commences. CENP-A, -B, and -C remain associated with centromeres throughout anaphase and telophase. However, CENP-E and -F dissociate from the chromosomes in anaphase B (spindle elongation) and relocate to the interzonal MTs at the spindle midzone, suggesting that they function in kinetochore-specific functions in metaphase and anaphase (12,14).

3. Detecting centromere/kinetochore proteins on metaphase chromosomes Immunofluorescence is commonly used to locate chromosomal proteins on metaphase chromosomes and to track the fate of these proteins during mitosis. It has been beneficial in studies which have investigated aspects of chromosomal structure, such as the assembly of functional centromeres and kinetochores at both normal and structurally abnormal chromosomes (15,16) and modification of chromosomal proteins (17). It has been used to study chromosome structure in human fibroblasts, lymphoblasts, and amniocytes, as well as fibroblasts from other primates, mice, hamsters, and muntjacs. One important consideration is that many antigens are destroyed or are made inaccessible to antibodies with harsh fixation, such as 3:1 methanol:acetic acid and even formaldehyde, therefore many investigators choose to prepare unfixed chromosomes for immunofluorescence. The best technique for obtaining unfixed metaphase spreads is centrifugation of hypotonically swollen cells onto clean microscope slides. The Cytospin 3 (Shandon Inc.) has a fixed rotor and 12 slots for microscope slides. An important feature of this particular model is the choice of acceleration speed. Excellent unfixed preparations can be obtained using high acceleration, presumably because the cells quickly come in contact with the slide surface and spread flat. Disposable single- and 83

Beth A. Sullivan and Peter E. Warburton double-chambered cytofunnels, already fitted with filter paper to absorb excess liquid from the sample area, are available as well as reusable singlechambered funnels. The reusable funnels must be fitted with filter-paper wicks at each use. Normal centrifuges, with swinging bucket rotors that have been equipped to hold microscope slides, have also been successfully used for the cytocentrifugation of cells. Even under the best conditions, unfixed chromosomes lack the distinct morphology and two-dimensionality of methanolracetic acid chromosomes (Chapter 4). Cell concentration is crucial to chromosome morphology. Colcemid and hypotonic conditions can be adjusted for each cell line to optimize mitotic index and chromosome spreading. Although optimal concentrations for cytospinning may vary among cell lines, Table 1 lists general guidelines for various types of cells. As the cells are unfixed, they often depend on mechanical adhesion to each other to remain intact on the slide. While obtaining the proper cell concentration will facilitate this adhesion, Shandon manufactures a Superfrost Plus slide (cat. no. 67761214) which has a permanent positive charge and noticeably improves specimen adhesion. These perform better than polylysine-coated or silane-prep slides (Sigma) on which many cell preparations clump and chromosomes fail to spread sufficiently. After the cells are spun onto slides, KCM, a buffer containing potassium and a detergent, is used to lyse the cell membrane, thus providing better access of antibodies to the chromosomes. Slides may be stored in KCM or PBS overnight or until antibody detection. Detection of the antigen with primary antibodies should be performed within 24 hours, as cell morphology decreases over time and the unfixed cells will eventually detach from the slides. Primary antibodies are subsequently detected with fluorochrome-conjugated secondary antibodies. Secondary antibodies are commonly conjugated to FITC, Texas Red, or rhodamine and are available from several companies e.g. Jackson ImmunoResearch (USA) or Vector Laboratories (Burlingame, CA). Protocol 1 outlines the procedure for obtaining unfixed metaphase chromosome spreads using a Shandon Cytospin 3 cytocentrifuge; these are then suitable for immunofluorescence (Protocol 2) and FISH as described in Protocol 3. This method is also useful for obtaining anaphase and telophase cells, and will be discussed in detail for these applications later in this chapter. Protocol 1. Preparing unfixed metaphase chromosomes for immunofluorescence Equipment and reagents • Colcemid • T-25 flask

• Cytofunnels, single or double chamber (Life Sciences International: single chambers, cat. no. 5991025;" double chambers, cat. • Cytocentrifuge (e.g. Shandon Cytospin 3 or no 5991039a) Ames Cyto-Tek) • Clean microscope slides (Superfrost Plus, • Trypsin-EDTA (1 X) Shandon are recommended)

84

5: Studying the progression of vertebrate chromosomes • PBS: 136 mM NaCI, 2 mM KCI, 10.6 mM Na2HP04, 1.5 mM KH2PO4, pH 7.4 . KCM: 120 mM KCI, 20 mM NaCI, 10 mM Tris-HCI pH 7.7,0.1% Triton X-100

• Hypotonic solution: 75 mM KCI for suspension cells and fibroblasts or 8 mM trisodium citrate, 66 mM KCI for human, mouse, hamster fibroblasts

A. Monolayer cultures 1. Add 0.1 mg/ml Colcemid to actively dividing cells in a T-25 flask (106-107 cells. Incubate for 1-4 h at 37°C.b 2. Reserve the culture medium in a centrifuge tube, since many dividing cells often detach and float in the media. Wash the cells once with 2 ml PBS. Add PBS wash to the centrifuge tube. 3. Add 1-2 ml trypsin-EDTA to the flask and the incubate cells at 37°C until the cells are detached. Inactivate the trypsin-EDTA with the reserved medium from the centrifuge tube. 4. Pellet the cells at 400 g for 5 min at room temp. 5. Resuspend in 10 ml PBS and determine the cell concentration using a haemocytometer. Proceed to part C. B. Suspension cells 1. Add 0.1 mg/ml Colcemid to actively dividing cells, usually 24 h after having split the culture. Incubate for 30–60 min at 37°C. 2. Pellet the cells by centrifuging at 400 g for 5 min at room temp. 3. Resuspend in 10 ml of PBS and determine the cell concentration using a haemocytometer. Proceed to part C. C. Cytocentrifugation 1. Resuspend the cells to 8 X 104 cells/ml in hypotonic solution.c Add the hypotonic solution dropwise and ensure that there are no cell clumps. Incubate the cells at room temperature for 15 min. 2. Prepare the centrifuge rotor.d Load clean microscope slides into the metal (or plastic) clips. Place a cytofunnel fitted with filter paper against the slide and secure the metal clip. 3. Gently mix using a pipette tip to ensure that cells are evenly dispersed throughout the hypotonic solution. 4. Load 500 ml of cells in hypotonic solution into the cytofunnel. If you are using double cytofunnels, pipette 250 ml into each side of the funnel.d 5. Spin with high acceleration at 1900–2000 r.p.m. (-750 g) for 10 min at room temp. 6. Gently remove the funnels from the slides, taking care to not scrape off cells. Allow excess liquid to dry from the slides.

85

Beth A. Sullivan and Peter E. Warburton Protocol 1.

Continued

7. Incubate the slides in KCM at room temperature for at least 10 min before adding the antibodies (Protocol 2). a Catalogue numbers listed are for orders within the United States. International order numbers are: single chambers 59910025 and double chambers 59910039) b Incubation time will vary according to the cell line. Somatic cell hybrid lines, mouse and hamster cells which tend to grow rapidly usually require 30–60 min in Colcemid. Primary human fibroblast cells often require longer. cThe ideal cell concentration for cytospinning will depend on the cell type and the type of cytofunnels used. See Table 7 for general cell-concentration guidelines. d Remember that several cytocentrifuge models require that the rotor is loaded and unloaded outside the centrifuge. Familiarize yourself with the manufacturer's instructions before attempting this protocol. "Shandon provides lids for cytofunnels. While their use is completely optional, they are useful for minimizing debris on the sample area which often results from loose filter-paper particles circulating in the closed rotor.

Protocol 2. Antibody detection on cytospun preparations Equipment and reagents • Humidified chamber (Pyrex dish or other container covered with Saran wrap or a fitted lid; use wet paper towels or filter paper to create humidity within the dish) • Parafilm—cut into squares just larger than specimen area • Appropriate primary and secondary antibodies diluted in KCM • 4% paraformaldehyde (pFA) in KCM or PBS (Protocol 7) (optional)

. KCM (Protocol 7) DAPI (4'-6-diamidino-phenylindole): 50 mg/ ml stock diluted to 1-2 mg/ml in antifadent > Antifadent (Vectashield, Vector Laboratories) or 100 mg phenylenediamine dihydrochloride, 10 ml PBS, pH 8.0, with 0.5 M bicarbonate/carbonate buffer (0.42 g NaHCO3 in 10 ml dH20, adjust to pH 9.0 with NaOH). Add antifade to 90 ml glycerol and filter through a 0.45 mm filter

Method 1. Remove slides in Protocol 1 from the KCM. Take care that the slides do not dry out at any point during antibody detection. 2. Apply 20-30 ml of the primary antibody diluted in KCM. Do not pipette directly onto the sample area as this can dislodge the cells. Carefully cover the sample area with a square of Parafilm. 3. Incubate at 37°C for 30–90 min in a humidified chamber. 4. Wash the slides 3 x in KCM at room temperature for 3 min each time. Do not agitate the slides as many cells may be dislodged. 5. Add 20-30 ml of the secondary antibody diluted in KCM. Incubate and wash the slides as in steps 3 and 4. 6. Briefly drain any liquid from the slides and mount with DAPI diluted in antifadent. Do not apply pressure to the coverslip to remove air 86

5: Studying the progression of vertebrate chromosomes bubbles as this can often destroy chromosome morphology.a If the cells are to be used for FISH following immunofluorescence, proceed to Protocol 3.b "To prevent the destruction of chromosome morphology during mounting, slides can be fixed in 4% paraformaldehyde (pFA) in KCM or PBS for 10 min at room temperature, followed by two washes of 1 min each in dH20. After drying, slides can be mounted in DAPI in antifade. b If the slides will be processed for FISH following immunofluorescence, they can be mounted for quick analysis under the microscope first. Excess immersion oil on coverslips should be gently wiped away and coverslips should always been floated off in KCM or PBS before proceeding with FISH.

Table 1. Suggested cell concentrations for cytocentifugation of metaphase chromosomes Cell type Lymphoblasts Human fibroblasts Somatic cell hybrids Mouse fibroblasts Hamster fibroblasts

Single chamber (cells/ml) 5

Double chamber (cells/ml) 1-1.2 x 105 5–8 X 104 6–9 x 104 7-8 x 104 7-8 X 104

1.8 X10 3–5 X 104 4-6 X 10* 3X 10 4 3X 10 4

4. In situ hybridization following immunofluorescence (Figure 1} It is often necessary to perform FISH analysis subsequent to the indirect immunofluorescent technique described in Protocol 2. To maintain the immunofluorescence signal and to preserve chromosome morphology during the harsh denaturing conditions of FISH, chromosomes should be fixed with either formaldehyde/formalin alone or formaldehyde followed by 3:1 methanol:acetic acid. Chromosomes retain their shapes following these fixation steps, although they are no longer so easily accessible to DNA probes. Therefore, several steps can be taken to ensure successful in situ hybridization, including longer denaturation times or higher denaturation temperatures (for other suggestions for increasing accessibility of formaldehyde fixed chromatin to DNA probes, see Chapter 6). Denaturing and washing solutions should be at pH 7.0 to minimize the damage that can occur to the chromosome morphology during FISH. Generally, 25-50 ng of probe DNA per slide (24 X 60 mm) is used for repetitive probes; 100 ng-1 mg DNA for single-copy probes. For cytospin preparations in which each specimen area is approximately 6 mm, 10–20 ng of repetitive probes and 50–500 ng of singlecopy probes per specimen area are adequate. However, it may be necessary to optimize certain probes for use with this protocol. 87

Beth A. Sullivan and Peter E. Warburton Protocol 3. In-situ hybridization following immunofluorescence Equipment and reagents • . • • • • . • • • •

Speedvac (Savant) Parafilm (cut to fit the sample area) KCM or PBS (Protocol 7) 4% formaldehyde (formalin: Sigma cat no. F1268) in KCM or PBS 3:1 (v/v) methanol:acetic acid 20 x SSC: 3 m NaCI, 0.3 M trisodium citrate pH 7.4 70, 90, and100%ethanol Denaturing solution: 70% formamide/2 x SSC pH 7.0 3 M Na acetate, pH 5.5 Salmon sperm DNA Hybridization buffer/mix: 50–65% formamide, 2 x SSC, 20% dextran sulfate, 0.1% Tween-20 (v/v) (see also Chapter 4); make at least 100 ml total, bringing the volume to 100mlwith dH2O

• Cot-1 DNA (or commercial probe) for single-copy probes • 12-16 mm round coverslips • 24 x 50 mm or 24 x 60 mm glass coverslips (0.13–0.16 mm thick) • Rubber cement and vulcanizing solutions • Post-hybridization washes: 50–55% formamide/2 x SSC pH 7.0 . Blocking buffer (5% skimmed milk in 4 x SSC) • Antibodies for detecting biotin or digoxigenin(dig)-conjugated probes (see Chapter 4, Table 7): for biotin-labelled probes use FITC or Texas Red-conjugated avidin, for dig-labelled probes use rhodamine- or FITC-conjugated anti-dig • Wash buffer: 4 x SSC/0.1% Tween-20 (v/v) . DAPI in antifade (Protocol 2)

A. Fixation of cells prior to FISH 1. Fix cells in 4% formaldehyde in KCM or PBS for 10 min at room temperature. 2. Wash the slides twice briefly in dH2O (about 30 sec). 3. Allow the slides to air-dry. Slides can be processed for FISH directly or can undergo further fixation (step 4). 4. Fix the cells in 3:1 methanohacetic acid for 15 min at room temperature. Allow to dry overnight before proceeding with in situ hybridization (part B). B. 1. 2. 3. 4.

Preparation of slides for hybridization Treat slides in 2 X SSC at 37°C for 30-60 min. Dehydrate through 70/90/100% ethanol series for 2 min each. Air-dry on a slide warmer set to 37 °C. Denature the slides in 70% formamide/2 x SSC, pH 7.0 for 8 min at 70°C.a 5. Repeat steps 2 and 3. C. Preparation of DNA probes 1. To the required amount of probe, in a 1.5 ml microcentrifuge tube, add 1/10th volume of 3 M Na acetate, pH 5.5, 500 ng salmon sperm DNA, and 2 volumes of cold 100% ethanol. For single-copy probes, 0.2 ng2 mg of Cot-1 DNA should be added.b Precipitate at -80°C for 2 h, or overnight at–20°C. 2. Centrifuge at 14000 g for 20 min at 4°C.

88

5: Studying the progression of vertebrate chromosomes 3. Aspirate the supernatant, taking care not to disturb the pellet. Wash the pellet with 100 ml 70% ethanol. 4. Centrifuge at 14000 g for 8 min at 4°C and aspirate 30 ml of the supernatant. 5. Dry the pellet for 10 min (or until dry) in a Speedvac. 6. Resuspend the pellet in prepared 8 ml of hybridization buffer. Vortex and spin briefly. Denature probe at 70°C for 5 min and reanneal if necessary (Chapter 4) 7. Apply the probe to the denatured slide from part B. Apply a coverslip and seal with rubber cement/vulcanizing solution; 12-16 mm round coverslips are adequate for covering the small cell area produced by cytocentrifugation. 8. Incubate overnight in a humidified chamber at 37 °C. D. Post-hybridization washes and signal detection 1. Wash the slides 3 x for 5 min each time in 50% formamide, 2 x SSC, pH7.0at 45°C.c 2. Wash the slides 3 x for 3 min each time in 2 x SSC at 45°C. 3. Soak the slides in wash buffer for at least 5 min. 4. Apply 15 ml of the blocking buffer to the cell area.d Incubate at room temperature for 5 min. 5. Drain excess blocking buffer from the slides. Add 15-30 ml of fluorochrome-conjugated avidin or anti-dig antibody. 6. Coverslip with Parafilm cut just larger than the sample area. Incubate at 37°C for 30–60 min. 7. Wash the slides 3 x for 2 min each time in wash buffer. 8. If necessary, amplify the signal0 (see Chapter 4 for standard amplification procedures). 9. Counterstain the DNA with DAPI in antifade. Use a 24 X 50 mm or 24 X 60 mm glass coverslip (0.13-0.16 mm thick) to cover both specimen areas and seal around the edges with a vulcanized sealing solution. 'Alternatively, slides may be denatured at 70°C, first for 3 min in 70% formamide/2 x SSC pH 7.0, followed by 2 min in 50% formamide/2 x SSC pH 7.0. Apply the probe immediately after the 2nd denaturation step, apply a coverslip and seal with rubber cement. b If using commercial probes, follow the manufacturer's instructions for hybridization. c Less stringent post-hybridization washes are often necessary in this procedure, particularly for a-satellite DNA probes, as the efficiency of hybridization is often decreased by formaldehyde fixation compared to that for conventionally fixed metaphase chromosomes. Washes containing 50–60% formamide/2 x SSC are sufficient to produce a reasonably strong probe signal. d'Blocking buffer, as well as antibodies diluted in blocking buffer, should be centrifuged at 4°C for 5 min before applying to slides. b It is sometimes unnecessary to amplify a-satellite signals. Slides may be counterstained and analysed at this point. If further amplification is needed, the coverslip can be floated off in wash buffer and the slides can be amplified until a signal is clearly visible.

89

Beth A. Sullivan and Peter E. Warburton

Figure 1. Determining centromere function using unfixed chromosome preparations. Immunofluorescence and sequential FISH on fibroblast cells containing a dicentric X chromosome. White dots bordering each primary constriction represent staining with anti-CENP-E antibodies, indicating a functional kinetochore. The arrow points to the functional centromere of the dicentric chromosome, which is also identified by the localization of an X-specific a-satellite probe. The DNA probe is located between anti-CENP-E antibodies. The arrowhead denotes the inactive centromere of each dicentric which is detected only by the DNA probe.

5. The use of anti-mitotic drugs Colccmid, nocodazole, and vinblastine are potent inhibitors of cell division, although each varies in its effect on mitosis. These drugs are commonly used to study stages of the cell cycle, MT dynamics, and spindle integrity, and to analyse chromosome structure and segregation.

5.1 Colcemid/colchicine Colcemid, and its more potent derivative colchicine, specifically binds to the tubulin dimer and inhibits the polymerization of MTs. Spindle formation is 90

5: Studying the progression of vertebrate chromosomes disrupted by the conformational change caused by the Colcemid-tubulin complex, and either a bipolar structure does not form properly or chromosomes are not oriented correctly upon it. Mitotic progression is significantly delayed at prometaphase, although some cells will eventually progress into anaphase in the absence of a spindle. Colcemid was previously thought to permanently arrest cells in metaphase, but more recent evidence indicates that it delays mitosis in a manner similar to other anti-mitotic drugs like nocodazole and vinblastine (18). Exposure to Colcemid causes rapid accumulation of cells in prometaphase but does not interfere with chromosome condensation. This has made Colcemid a widely used drug for the preparation of highly condensed metaphase chromosomes for cytogenetic banding or FISH.

5.2 Nocodazole Nocodazole, a benzimidazole derivative, has an immediate effect on MT dynamics by binding to tubulin dimers and preventing MT assembly. Overall, nocodazole causes a rapid depolymerization of MTs, efficiently arresting cells in metaphase. It competitively inhibits colchicine-binding to tubulin, indicating that nocodazole acts in a similar fashion. However, it associates with and dissociates from tubulin at a much faster rate than Colcemid (19) so that cells eventually progress into anaphase. This unique effect on mitosis makes it an attractive alternative to Colcemid and colchicine for use in anaphase studies. Large numbers of cultured cells can be arrested in metaphase and, within a short period of time after removal of nocodazole from the culture medium, they will synchronously enter anaphase. The deleterious effects on MT recovery after exposure to low concentrations of nocodazole are minimal, although at micromolar concentrations astral MT disorganization, reduced numbers of interzonal MTs, and slow spindle pole separation have been observed (20).

5.3 Vinblastine and other drugs Vinblastine, vincristine, and podophyllotoxin are anti-mitotic drugs which, at high concentrations, bind to and depolymerize MTs. The effects of low concentrations of vinblastine on mitotic cells is very similar to those seen after nocodazole treatment. Rates of MT growth and shortening are suppressed, and the spindle, while stabilized, is no longer capable of dynamic MT turnover (21). In contrast, low concentrations of podophyllotoxin have a more severe mitotic effect by completely depolymerizing MTs and leading to distorted spindles. Furthermore, cells treated with podophyllotoxin are not as efficiently blocked in metaphase as is seen after treatment with nocodazole or vinblastine (21, 22).

5.4 Cytochalasin/dihydrocytochalasin B (DCB) Cytochalasins affect various cellular aspects, including cell division, motility, shape, and sugar transport (23,24). Their competition with sugars for binding 91

Beth A. Sullivan and Peter E. Warburton sites in the plasma membrane has no bearing on their cytological effects on MT dynamics and cell cleavage. Cytochalasins disrupt actin formation and microfilament growth and stability. Dihydrocytochalasin B (DCB), a derivative of cytochalasin B (CytB), only disrupts actin assembly and in a more specific manner than most other cytochalasins. However, it is less frequently associated with structurally abnormal microfilaments, indicating that, overall, it is considerably less potent than CytB (23). Typically, in late anaphase, a structure called the telophase disc is formed from proteins that have dissociated from the chromosomes during anaphase (25, 26). The telophase disc expands across the width of the cell and at the position of what will become the cleavage furrow (25). After sister chromatids have successfully reached opposite poles in late telophase, actin and myosin filaments become concentrated between the poles, interacting to form a contractile ring (invagination of the plasma membrane). DCB prevents cytokinesis by binding to the growing ends of actin, preventing elongation of filaments and formation of the contractile ring (23, 25). DCB and other cytochalasins appear to have no effects on spindle integrity, chromosome segregation, or mitotic progression, since cells treated with DCB maintain a normal mitotic spindle and progress through mitosis into a Gr-like state in which binucleated cells are formed but remain uncleaved (27). This delay in cell cleavage is reversible—upon release from a DCB block, cells that have been in G1 for a short time will undergo cleavage.

6. Immunofhiorescence on anaphase and telophase cells/chromosomes Chromosomes in the later stages of mitosis can be studied using various approaches. Treating cells with the anti-mitotic drugs described in Section 5 can delay cells in late metaphase, anaphase, or even telophase. The point of delay can be easily identified using immunofluorescence with antibodies to tubulin. Other antibodies can be used simultaneously to study various chromosomal and cellular characteristics, such as histone acetylation, kinetochore and centromere structure and assembly, telomere integrity, and sister chromatid cohesion (17, 27, 28).

6.1 C-anaphase: sister chromatid separation and anaphase in the presence of Colcemid Usually a 1-3-h treatment of cells with 0.1 (mg/ml Colcemid yields metaphase chromosomes in their highest state of condensation. However, upon prolonged exposure to Colcemid (>16 h) or at lower concentrations of Colcemid (<0.05 (mg/ml), mitosis proceeds in the absence of a spindle. Sister chromatids separate as the cells enter anaphase, and chromosomes decon92

5: Studying the progression of vertebrate chromosomes

Figure 2. C-anaphase preparations from Chinese hamster ovary (CHO) cells containing ectopic a-satellite DNA arrays from human chromosome 17, to study sister chromatid cohesion. The extended incubation time with Colcemid produces chromosomes in which sister chromatids are largely separated. Note that some centromeres are still joined on several chromosomes. A chromosome a -satellite probe (white) was used to identify the CHO chromosome containing the integrated human DNA and to demonstrate that while the endogenous hamster centromere chromatids are separated, the ectopic human a-sateltite DNA is still joined.

dense into a telophase-like stage. These chromosome structures, termed 'Canaphases', can be easily prepared with 3:1 methanol:acetic acid fixation and dropped onto microscope slides for analysis by FISH (Protocol 4). Various stages of disjunction are observed, from complete pairing along the length of the chromosome to total separation of chromatids, even at the centromere (28). C-anaphases are useful for studying aneuploidy, non-disjunction, and centromere separation, for instance, of structurally abnormal chromosomes or cells demonstrating premature centromere separation (Figure 2). Protocol 4.

Preparation of 'C-anaphasesF

Reagents • 10 mg/ml Colcemid stock solution (Gibco-

• 3:1 methanol:acetic acid {Chapter 4, Protocol 1) • 1 x trypsin-EDTA

BRL)

• Hypotonic solution: 75 mM KCI

93

Beth A. Sullivan and Peter E. Warburton Protocol 4. Continued Method 1. Incubate actively dividing cells (106-107 cells) in 0.1 mg/ml Colcemid for 16-36 h. 2. Trypsinize the cells as in Protocol 1 and spin for 10 min at 400 g at room temp. 3. Gently resuspend the cell pellet in 8 ml of 75 mM KCI. Incubate for 10 min at room temperature or at 37°C. 4. Spin as in step 2. 5. Aspirate the supernatant and carefully resuspend the pellet in hypotonic solution. Add 5 ml 3:1 methanol:acetic acid dropwise, gently mixing the cell suspension after every 10-20 drops. 6. Spin as in step 2. 7. Repeat the fixation with 3:1 methanol:acetic acid at least three more times. 8. Prepare slides as described in Chapter 4. Slides may be stored at -20°C or under vacuum at room temperature until analysis by in-situ hybridization.

6.2 Anaphase chromosomes visualized on the mitotic spindle An effective way to prepare anaphase cells for immunofluorescence is by cytocentrifugation, a similar procedure to that outlined in Protocol 1 for obtaining unfixed metaphase chromosomes. This technique has been used to demonstrate segregation abnormalities of dicentric chromosomes and of hamster chromosomes containing amplified ectopic arrays of a-satellite DNA (Figure 3) (B.A. Sullivan, unpublished data, and ref. 28). Preparation of cells in this manner may be done in the absence of anti-mitotic drugs and by isolating mitotic cells by 'shake-off. The most successful mitotic shake-offs are done on cultured cell lines with high mitotic indices, although if cell lines grow at a slow rate it is possible to do shake-offs after treating cells with nocodazole which will not destroy the mitotic spindle. After shake-off, cells are incubated for several hours or overnight to accumulate cells in anaphase and are then centrifuged onto glass microscope slides. Cells can be processed for immunofluorescence immediately or after a brief (and optional) fixation in cold methanol. The mitotic spindle can be stained using antibodies to tubulin, the major component of MTs. Monoclonal antibodies to the two subunits of microtubules, a-tubulin and 3-tubulin, are commercially available (Sigma). Protocols 5 and 6 have successfully been used with antibodies to b-tubulin and acetylated atubulin. Acetylation is a post-translational modification of tubulin and has 94

5: Studying the progression of vertebrate chromosomes been correlated with increased MT stability and longer-lived (slow turnover) MTs (29, 30). Therefore, while b-tubulin antibodies will stain virtually all MTs, antibodies to acetylated a-tubulin will stain a subset of them, presumably the most stable ones. For instance, astral MTs are not stained with antibodies to acetylated a-tubulin.

Protocol 5. Mitotic spindle preparation using cytocentrifugation of attached Cells Equipment and reagents • Haemocytometer • Cytocentrifuge (Shandon Cytospin; Ames Cyto-Tek, Bayer Diagnostics) . Clean microscope slides (Superfrost Plus, Life Sciences International) • PBS (Protocol 1) • 100% methanol at -20°C

• KCM buffer (Protocol 1) • Anti-tubulin antibody: b-tubulin (Sigma, cat. no. T4026) or anti-acetylated tubulin (Sigma, cat. no. T6793)a • Anti-mouse secondary antibody (FITC-, rhodamine-, or Texas Red-conjugated) • DAPI/antifadent (Protocol 2}

A. Isolation of unblocked mitotic cells and attachment to microscope slides 1. To obtain actively dividing cells by mitotic 'shake-off': aspirate the medium and add 3 ml PBS. Alternatively, treat slower growing cells overnight with 30-40 mg/ml ( 1 0 0 nM) nocodazole to arrest cells in metaphase. Tighten the lid and knock the flask against a rubber stopper 10-20 times. This is sufficient to dislodge most of the mitotic cells. 2. Transfer the cells to a 15 ml centrifuge tube. 3. Incubate the tube at 37°C for 1-2 h to allow as many cells as possible to enter anaphase. Incubation time may vary according to the cell type or cell line. 4. Determine the cell concentration using a haemocytometer. 5. Pellet the cells by centrifuging at 400 g for 5 min at room temp. 6. Dilute the cells in 75 mM KCI to the appropriate cell concentration (see below). Incubate at room temperature for 10 min. Recommended cell concentrations for obtaining anaphase cells are 5–7 x 104 cells/ml for double-chamber cytofunnels and 1-4 x 10* cells/ml for singlechamber cytofunnels.b 7. Load 500 ml of the cell suspension into cytofunnels (250 ml per side for double cytofunnels). 8. Spin at 100–200 g for 5 min at room temp. 9. Place slides into 100% methanol at -20°C for 10 min to fix the cells. 10. Allow slides to air-dry briefly. 95

Beth A. Sullivan and Peter E. Warburton Protocol 5.

Continued

B. Immunofluorescence with anti-tubulin antibodies 1. Incubate the slides in KCM buffer for at least 10 min at room temperature. 2. Drain the excess liquid from the area on the slide which contains the cells. 3. Add 30 ml of anti-tubulin antibody (diluted 1:100-1:1000 in KCM).C Carefully place a plastic or Parafilm coverslip over the cell area. 4. Incubate at 37°C for 30-60 min. 5. Wash the slides three times in KCM for 3 min each wash. 6. Add 30 ml of anti-mouse secondary antibody. Incubate and wash the slides as in steps 4 and 5. 7. Drain off the excess liquid and mount the slides in DAPI/antifadent (Protocol 2). 8. If the slides will be used for FISH they should be fixed in 4-10% formaldehyde in KCM for 10 min, washed briefly twice in distilled water and air-dried. a The MT epitope recognized by acetylated anti-tubulin may often be inaccessible or even absent in certain cell types. In our experience, the anti-acetylated tubulin antibody does recognize epitopes in human and mouse fibroblasts and human lymphoblastoid cells. b Note that the cell concentrations recommended here for fibroblast cells are slightly less than those shown in Table 1. Because the cells will be spun at a slower speed, it is necessary to use less concentrated cell suspensions. Lower speeds will protect cell and spindle morphology, while the more dilute cell concentration will ensure that cells will spread sufficiently flat. c Double immunofluorescence may also be performed. Other primary antibodies can be applied at the same time, as long as both antibodies were not raised in the same type of animal. Alternatively, labelled secondary antibodies (e.g. one Texas Red-labelled and one FITC-labelled) can be added simultaneously in part B, step 6.

6.3 Anaphase studies on cytokinesis-blocked cells Segregation of chromosomes can be conveniently studied using cytochalasins which block cytokinesis (Section 5.4). As chromosome segregation and decondensation is unaffected by the delay in cell cleavage, chromosomes can be visualized during and after segregation to daughter cells using standard fixation and in situ hybridization techniques. Most importantly, artefactual chromosomal aneuploidy is minimal, and this technique can be applied to several cell types, such as fibroblasts, somatic cell hybrids, and lymphocytes (27,31). 6.3.1 Preservation of the mitotic spindle To visualize cytokinesis-blocked chromosomes on the mitotic spindle, MTs must be preserved through mild fixation. In Protocol 6, 2-4% paraformalde96

5: Studying the progression of verlebrate chromosomes

Figure 3. Anti-tubulin staining to detect the mitotic spindle, (a) At metaphase, chromosomes have congressed to the metaphase plate. Anti-acetytated tubulin antibodies have been detected with FITC (green! and chromosomes were counterstained with DAPI (blue), (b) Analysis of chromosome segregation in African green monkey (AGM) cells containing integrated human a satellite DNA. AGM cells were treated with nocodazole for 16 h, released from the mitotic block, and treated with DCB for 45 min before fixation with pFA. Double immunfkiorescence was performed with anti tubulin antibodies detected with Texas Red and anti-centromere antibodies detected with FITC (green). This late anaphase cell shows that most chromosomes have migrated to the spindle poles (yellow signals located at opposite ends of spindle), however, several yellow dots in the middle of the spindle indicate that several chromosomes lag in anaphaso. hyde (pFA) is used to fix cells to slides whilemaintaining;spindle and chromosome morphology. Permeobilizing cells with 0.5% Triton X-100 in PBS is suflieient to provide anlibody access to the chromosomes and spindle, If spindle staining is unnecessary, alternative cell fixation can be performed (Protocol 7).

Protocol 6. Analysis of chromosome segregation in cytokinesisblocked cells Equipment and reagents • 100 1000 •• nocodazole (Sigma) in sterie DMSO • 100 /. dihydrocytochalasin B DCB. Sigma) in sterile DMSO

Glass Coplin jat 100 mm tissue culture dishes or quadriPERM plus dishes (Heraeus. cat. no. /60//310I Cytokinesis-blocked cei.s and culture medium • 80% ethanol GaS ':oucrt' flamer

• PBS (Protocol 1)

• 2 4% paraformaidehyde IpFA) in PBS . Extraction buffer: 0.6% Tritnn X-100 in PBS

97

Beth A. Sullivan and Peter E. Warburton Protocol 6.

Continued

Method 1. Ethanol-flame glass microscope slides and place one slide into each tissue culture dish or quadriPERM chamber. 2. Place ~10* cells onto each slide. Leave the cells for 3–5 min to allow initial attachment to slides. Add 10-12 ml of the culture medium to each dish (5–6 ml for quadriPERM plates), taking care not to add the medium directly onto the slides. 3. Grow the cells on the slides at 37°C until 50-70% confluent (24-48 h). 4. Add 40 ng/ml (100 nM) nocodazole to the medium and incubate at 37°C for 12-16 h. 5. Wash the cells three times with fresh medium. Pipette the medium against the side of the dish as mitotic cells can be easily washed off the slide. 6. Add 10 mM (DCB) to the medium. Gently swirl to disperse the DCB. Incubate at 37°C for 45-120 min.a 7. Carefully aspirate the medium and add 10 ml 2-4% pFA (prewarmed to 37°C). Fix cells at 37°C for 20-30 min. 8. Aspirate the pFA and immerse the slides in a Coplin jar filled with extraction buffer for 5 min at room temperature. 9. Soak the slides in PBS until imrnunofluorescence is performed. Tubulin and/or other proteins (chromosomal, centromere, etc.) should be detected by following Protocol 5B. a It is necessary to optimize the incubation time in DCB for each cell line, since each cell line/cell type often differs in its recovery from the nocodazole block and its progression through anaphase and telophase. For example, human fibroblasts need to be incubated in DCB for at least 90 min to accumulate most cells in telophase. Somatic cell hybrids, which have a faster division time than primary cell lines, only require 30-45 min in DCB.

6.3.2 Fixation of anaphase cells using 3:1 methanol:acetic acid If spindle staining is unnecessary, we recommend fixing the cells with 3:1 methanol: acetic acid. Consistent with acetic acid fixation, the chromosomes will appear more two-dimensional than those fixed with formaldehyde, and they are suitable for FISH analysis of chromatid separation and segregation (Figure 4).

98

5: Studying the progression of vertebrate chromosomes Protocol 7.

Acid fixation of anaphase cells and FISH

Equipment and reagents • Hypotonic solution: 75 mM KCI (prewarmed to 37 °C) . 3:1 (v/v) methanol: acetic acid, ice-cold . 2 x SSC (Protocol 3) . 70,90, and 100% ethanol

• 24 x 60 mm coverslips . Denaturant: 70% formamide, 2 x SSC pH 7.0 . Slide-warmer/hotplate set at 37°C (optional) . Rubber cement • DAPI in antifadent (Protocol 2)

A. Acid fixation of cytokinesis-blocked cells 1. Follow Protocol 6 to block cells in metaphase and at cytokinesis using nocodazole and DCB, respectively. 2. Add 10 ml prewarmed (37°C) 75 mM KCI to each dish/slide chamber. Incubate at room temperature for 15 min. 3. Quickly pipette an equal volume of ice-cold fixative (3:1 methanol: acetic acid) into the hypotonic solution. Incubate at room temperature for 10 min. 4. Aspirate the hypotonic-fix solution and add 10 ml ice-cold fixative to the cells. 5. Repeat steps 3 and 4 three more times. 6. Air-dry the slides3 and allow them to sit overnight at room temperature or on a 37°C slide-warmer/hotplate. B. In situ hybridization on acid-fixed anaphase cells 1. Incubate the slides in 2 x SSC at 37°C for 30-60 min. 2. Dehydrate the slides through a 70/90/100% ethanol series at room temperature.6 3. Denature the slides in 70% formamide/2 x SSC, pH 7.0, at 70°C for 2 min. 4. Dehydrate the slides as in step 2. 5. Place the slides on a 37 °C hotplate/slide-warmer. 6. Prepare the DNA probe in hybridization mix (Protocol 3).c Denature the probe at 70°C for 5–10 min. 7. Apply the probe to each slide and cover with 24 x 60 mm coverslips. Seal with rubber cement and incubate at 37°C for 4-24 h. 8. Remove the rubber cement from around the coverslips and wash the slides as in Protocol 3. 9. Mount the slides with DAPI in antifadent (Protocol 2). ' Slides can be dried naturally at room temperature. However, depending on weather conditions and the lab. environment, it may be necessary to dry the slides with humidity (over a steaming water-bath) or, in conditions of high humidity, with forced air. 'Alternatively, dehydration steps may be performed in cold ethanol. "For general analysis of chromosome segregation, a-satellite probes are the most efficient and easily visualized. However, single-copy probes, such as cosmids have been used successfully on cytokinesis-blocked cells.

99

Beth A. Sullivan and Peter E. Wurburton

Figure 4. Acid fixation of cytokinesis-blocked cells for studying chromosome segregation in anaphase and telophase. (a) Cells containing a dicentric X chromosome were blocked in cytokinesis and fixed with methanol:acetic acid. Hybridization with an X-specific satellite probe, detected with FITC (green), indicated that the entire dicentric chromosome lagged in anaphosc, remaining at the metaphase plate, (b) In telophase. a single chromatid of the dicentric X chromosome failed to segregate with the remaining chromosomes (blue) into daughter cells. The chromosome X n-satellite probe was detected with FITC (green) and chromosome S n-satellite, detected with rhodamine anti-digoxigenin (red), was used as a control for normal segregation.

References 1. [lyams, J.S. and B n n k l o y , B.R. ( e d . ) (1989). Mitosis Academic.Press, San Diepo. CA. 2. Li. X. and Nk'klas. R.B. (1997).J. Cell Set.. 410.537

100

molecules

andmechanism.

5: Studying the progression of vertebrate chromosomes 3. Pluta, A.F., Mackay, A.M., Ainsztein, A.M., Goldberg, I.G., and Earnshaw, W.C. (1995). Science, 270,1591. 4. Earnshaw, W.C. and Rothfleld, N. (1985). Chromosoma, 91, 313. 5. Palmer, D.K., O'Day, H.L., Trong, H., Charbonneau, H., Margolis, R.L. (1991).. Proc. Natl. Acad. Sci. USA, 88, 3734. 6. Masumoto, H., Masukata, H., Muro, Y., Nozaki, N., Okazaki, T. (1989)..J. Cell Biol, 109,1963. 7. Saitoh, H., Tomkiel, J., Cooke, C.A., Ratrie, H., Maurer, M., Rothfield, N., and Earnshaw, W.C. (1992). Cell, 70,115. 8. Yen T.J., Compton, D.A., Wise, D., Zinkowski, R.P., Brinkley, B.R., Earnshaw, W.C., and Cleveland, D.W. (1991). EMBO J., 10,1245. 9. Cooke, C.A., Schaar B., Yen T.J., and Earnshaw, W.C. (1997). Chromosoma, 106, 446. 10. Thrower, D.A., Jordan, M.A., Schaar, B., Yen, T.J., and Wilson, L. (1995). EMBO J., 14, 918. 11. Lombillo, V.A., Nioslow, C., Yen, T.J., Gelfand, V.I., and Mclntosh, J.R. (1995). J. Cell Biol., 128, 107. 12. Liao, H., Winkfein, R.J., Mack, G., Rattner, J.B., and Yen, T.J. (1995). J. Cell Biol., 130, 507. 13. Gorbsky, G.J. and Ricketts, W.A. (1993). J. Cell Biol., 122,1311. 14. Brown, K.D., Wood, K.W., and Cleveland, D.W. (1997). J. Cell Sci., 109, 961. 15. Sullivan, B.A. and Schwartz, S. (1995). Hum. Mol. Genet., 4, 2189. 16. Depinet, T.W., Zackowski, J.L., Earnshaw, W.C., Kaffe, S., Sekhon, G.S., Stallard, R., Sullivan, B.A., Vance, G.H., Van Dyke, D.L., Willard, H.F., Zinn, A.B., and Schwartz, S. (1997). Hum. Mol. Genet., 6,1195. 17. Jeppesen, P. and Turner, B.M. (1993). Cell 74,281. 18. Rieder, C.L. and Palazzo, R.E. (1992). J. Cell Sci., 102, 387. 19. Jha, M.N., Bamburg, J.R., and Bedford, J.S. (1994). Cancer Res., 54,5011. 20. Vasquez, R.J., Howell, B., Yvon, A.-M.C, Wadsworth, P., and Cassimeris, L. (1997). Mol. Biol. Cell, 8, 973. 21. Jordan, M.A., Thrower, D., and Wilson, L. (1992). J. Cell Sci., 102, 401. 22. Snyder, J.A., Vogt, S.I., and McLelland, S.L. (1983). Cell Motility, 3, 79. 23. Aubin, J.E., Osborn, M., and Weber, K. (1981). Exp. Cell Res., 136, 63. 24. Yahara, K., Harada, F., Setsuko, S., Yoshihira, K., and Natori, S. (1982). J. Cell Biol., 92, 69. 25. Andreassen, P.R., Palmer, D.K., Wener, M.H., and Margolis, R.L. (1991). J. Cell Sci., 99,523. 26. Eackley, D.M., Ainsztein, A.M., Mackay, A.M., G.oldberg, I.G., and Earnshaw, W.C. (1997). J. Cell Biol., 136,1169. 27. Martineau, S.N., Andreassen, P.R., and Margolis, R.L. (1995). J. Cell Biol., 131, 191. 28. Warburton, P.E. and Cooke, H.J. (1997). Chromosoma, 106,149. 29. Piperno, G., LeDizet, M., and Chang, X.-J. (1987). /. Cell Biol., 104, 289. 30. Webster, D.R. and Borisy, G.G. (1989). /. Cell Sci., 92, 57. 31. Zijno, A., Leopardi, P., Marcon, F., and Crebelli, R. (1996). Mut. Res., 372, 211.

101

This page intentionally left blank

6

Analysis of mammalian interphase chromosomes by FISH and immunofluorescence JOANNA M. BRIDGER and PETER LICHTER

1. Introduction The analysis of chromosome structure and organization in mammals has traditionally been performed on metaphase chromosomes. This is especially true for the localization of specific DNA sequences and chromosomalassociated proteins longitudinally along the metaphase chromosomes (1). It is not yet clear to what extent the structure and organization of metaphase chromosomes is different from, or related to, the structure and organization of the same chromosomes in interphase nuclei. Furthermore, different proteins may contribute to the maintenance of chromosome structure and organization during mitosis and interphase. Analysis of metaphase chromosomes has been possible for many years because of the chromosomes' inherent compact structure during mitosis, they are easily separated from each other and can be recognized due to their distinctive banding patterns when stained with certain dyes (2). Individual chromosomes in interphase mammalian cells cannot be distinguished by staining DNA or chromatin alone. Hence, for any analysis of interphase chromosome structure the individual chromosomes must be delineated in other ways. For a long time it was not clear how specific chromosomes were spatially organized in interphase, but with the application of fluorescence in situ hybridization (FISH), using the first whole chromosome paints, it became obvious that chromosomes in interphase were territorially organized (3-5; Figure 1). It appears that there is no extensive intermingling of chromatin from juxtaposed chromosome territories (6). There are a number of methodologies which can be employed to obtain specific chromosome painting probes and these will be discussed in this chapter. The organization of interphase chromosomes could be dynamic, changing as cells traverse interphase (7, 8). To study the differences between chromo-

Joanna M. Bridger and

Figure 1, interphase chromosome territories. Digital image of a human dermal fibroblast nucleus, acquired using a cooled CCD camera and displaying two chromosome territories delineated using a biotinylated painting probe (green). The probe was PCRamplified from a catch-linkered DNA sequence, incorporating biotin-16-dUTP during the PCR amplification. The biotin hapten was detected using avidin conjugated to FITC as the first detection layer, the second layer consisted of biotinylated anti-aviriin which was detected via a third layer of avidin-FITC (green). Note the cloud-like morphology of the interphase chromosomes. DNA is stained with DAPI (blue). Scale bar = 5 mm.

somes at the various stages of the cell cycle, cells can be synchronized. Alternatively, the cell-cycle phase of individual cells in an asynchronous culture can be assessed using specific cell-cycle markers detected with specific anlibodies. Such studies necessitate the use of FISH in combination with immunofluorescence, and this is discussed in Section 5. In some cases it is also possible to determine the cell-cycle stage by morphology or by incorporation of 'dNTPs in S-phase. With the rapid maturation of fluorescence microscopy techniques and the number of different reporter molecules available there are increasing possibilities tor more structures to be simultaneously analysed within the same sample. Another limiting factor has been the analysis of interphase cells in 104

6: Analysis of mammalian interphase chromosomes three dimensions, critical when performing spatial analyses, since the interphase nucleus is obviously a 3-dimensional (3-D) structure and the chromosomes should be thought of as 'clouds'. Hence cells need to be fixed so that their 3-dimensionality is preserved as much as possible. Methods for fixation, permeabilization, and preservation of the 3-D structure will be discussed in the following section.

2. Preparation of sample material The most readily available source of biological starting material for the study of interphase chromosomes are mammalian tissue-culture cells or maintained primary cells in culture. There are a variety of human chromosome painting probes available, and hence the majority of the studies performed on mammalian interphase chromosome structure use human cell types. However, it is possible to use certain primate cells in combination with human probes since the level of homology between the two genomes can be very high. Although for a given application this must be determined empirically. For example, the DNA of COS-7 African green monkey cells is highly homologous with human DNA and so specific human chromosome paints can be used (Bridger, Kraeusslich, and Lichter, unpublished data). Chromosome specific paints for mouse and Chinese hamster chromosomes are also commercially available (Oncor Appligene, Biovation, Cambio, Cytocell, Growing Point, Vysis).

2.1 Adherent cells For performing FISH, especially when analysing cells fixed to preserve their 3-dimensionality, cells are grown on glass microscope slides which survive extremes of temperatures, such as Superfrost™ slides (e.g. Merck). Glass coverslips can be used, but they are much more difficult to manipulate. Furthermore, cells fixed in paraformaldehyde can be stored at -80 °C (see Section 2.5) and this freezing step is considerably easier if microscope slides are employed. There are customized tissue culture dishes (e.g. Quadriperm™, Heraeus) which conveniently take four microscope slides in separate chambers. Each chamber takes between 3–4 ml of medium and cells should be plated at 1-5 X 103/cm2, depending on individual culturing needs. Normal tissue culture dishes can obviously be used if such dishes prove to be too expensive. Slides should be clean and sterile before use in tissue culture. It is not important to have sterile slides for suspension cells, since the cells are only in contact with the slides for a relatively short time. We clean slides by boiling them in detergent for several minutes in a Pyrex beaker. This step can be performed in a microwave or over a Bunsen burner. After washing in copious amounts of tap water to remove detergent, the slides are rinsed in distilled water and placed in methanol. At this stage slides can either be sterilized by allowing them to dry and packaging them in aluminium foil for autoclaving, or 105

Joanna M. Bridger and Peter Lichter they can be kept in methanol, flamed, and placed immediately into the tissue culture dish.

2.2 Suspension cells There are at least two possible ways of performing FISH on suspension cells. Cells can be attached to a substratum by coating glass microscope slides with 40 mg/ml poly-L-lysine. This can be made from powder or purchased as a liquid. Alternatively, ready-prepared poly-L-lysine slides are also commercially available. When preparing your own slides, place 500 m1 poly-L-lysine solution on the slide and, using a second slide, encourage the solution to coat the slide. Allow the slide to dry and then repeat the procedure to build up several layers. Suspension cells, in the appropriate medium, should be applied to the central area of the slide at a density of approximately 5 X 104/cm2. The slides can then be incubated at the appropriate temperature and CO2 concentration for 1-3 h, by which time the cells should be attached to the substratum. This procedure of securing suspension cells to glass slides is the method of choice when analysing interphase chromosome structure, as cells settle on the substratum and become secured by the cells own interaction with the substratum and do not endure physical forces which may have detrimental effects on nuclear and/or chromosome architecture. There may be subtle differences in chromosome architecture in suspension cells that remain in suspension compared to suspension cells that become attached, but these are thought to be minimal. Cytospinning can also be used to secure suspension cells to microscope slides. The cell suspension is forced onto a glass microscope slide by centrifugation (Chapter 5). This method is not thought suitable for preserving the 3-D structure of cells because of the centrifugal forces involved. If this method is used then it would be advisable to fix cells (preferably in paraformaldehyde, see Section 2.4) prior to spinning.

2.3 DNA halo preparations Cell nuclei contain a structural component, termed the nuclear matrix, and when certain extraction procedures are performed on permeabilized cells in the preparation of samples, regions of DNA containing matrix-associated sequences (MARs) (at the base of DNA loops) remain attached to the inextractable nuclear matrix structure. The rest of the unattached looped DNA is released from the confines of the nucleus and is distributed as a halo surrounding the extracted nucleus. The permeabilization and extraction procedures remove soluble proteins within the cells, these include the histones. It is possible to perform FISH on such preparations since the DNA is not digested. Such analyses are informative as to which chromosomal sub-regions are associated with the nuclear matrix and which are not. Most DNA halo preparations are made by extracting with high molarities of sodium chloride (9-12), 106

6: Analysis of mammalian interphase chromosomes but it is also possible to use LIS (lithium 3,5-diiodosalicylic acid) buffer (see ref. 10). These preparations can also be used to perform high-resolution FISH, since highly extended loops of DNA are produced which can be used for mapping specific sequences. Nuclei can either be grown on slides, cytospun onto slides, or allowed to settle on slides. There are various different approaches used to prepare nuclear matrices/DNA halo structures, but they all have aspects in common; we will only present one methodology here, however one should refer to the aforementioned references for variations which may suit the individual. Protocol 1.

Preparation of DNA halos (10)

Equipment and reagents • Coplin jars . CSK buffer: 10 mM Pipes pH 7.8, 100 mM NaCI, 0.3 M sucrose, 3 mM MgCI2, 0.5% Triton-X 100 • Phosphate-buffered saline, 10 x, 5 x, 2 x, and 1 x (PBS 1 x: 140 mM NaCI, 27 mM KCI, 110 mM Na2HP04, 15 mM KH2PO4)

• Extraction buffer: 2 M NaCI, 10 mM Pipes PH 6.8, 10 mM EDTA, 0.1% digitonin, 0.05 mM spermine, 0.125 mM spermidine . Ethanol series: 10%, 30%, 70%, and 95%

Method 1. If using suspension or harvested cells, place 1.5 x 106 cells in CSK buffer and incubate on ice for 15 minutes. 2. Cytospin (see Chapter 5) the cells onto glass microscope slides at 500g. 3. Rinse the slides twice in 1 x PBS. 4. Gently place the slides into a Coplin jar containing extraction buffer for 4 min. 5. Rinse the slides consecutively in 10 x, 5 x, 2 x, and 1 x PBS for 1 min each. 6. Take the slides through an ethanol series, comprising 10%, 30%, 70%, and 95% ethanol. 7. Air-dry the slides and bake for 2 h at 70°C. The slides are now ready for denaturation (ProtocolsSand 6).

2.4 Fixation and permeabilization Flattened and swollen interphase nuclei are produced when cells are prepared according to the standard method for cytogenetic analyses on metaphase chromosome spreads (7,13 and Chapters 4 and 5). Cells prepared in this way should be avoided for interphase chromosome structural analyses, since the chromatin and morphology of the chromosome territory is modified by the hypotonic treatment and dropping of the cells onto the slide and may not give 107

Joanna M. Bridger and Peter Lichter a very accurate view of the in-vivo situation. Furthermore, since the hypotonic treatment is performed prior to fixation, immunofluorescence analysis of any antigen of interest may not be reliable. If 3-D preservation of the sample cells is not imperative and/or antigen detection precludes the use of paraformaldehyde, then organic fixatives (such as methanol, ethanol, acetone) can be used, or hypotonic treatment followed by fixation in methanolracetic acid may also be used (Protocol 2A). Good FISH signals are routinely easy to achieve when using these types of fixation. However, such fixations do not preserve cellular architecture, many soluble proteins are immediately lost as the nuclear envelope is punctured. Fixation is based on the aggregation of, in particular, proteinaceous elements of cells and hence cells are flattened, extracted to a certain extent, dehydrated, and lose components which may be necessary for the structural integrity of both the nucleus and chromosomes. Furthermore, if the spatial relationship between a specific nuclear entity and chromosome territories is being investigated then fixations, causing protein aggregations, potentially produce artefactual colocalizations. Cross-linking agents such as paraformaldehyde should be the fixative of choice when preservation of the 3-D structure of cells and nuclei is of prime importance. It has also been shown in cells fixed with paraformaldehyde, permeabilized, and processed for FISH, that centromeres remain in the same spatial position before and after the FISH procedure (14,15). We fix cells in paraformaldehyde and permeabilize with different procedures so that FISH protocols can be applied. The problem with fixing cells so that their 3-D structure is preserved with paraformaldehyde is that crosslinking is produced which hinders the penetration of the probe sequences. However, the fixation protocol in Protocol 2B in combination with the additional procedures to improve probe penetration (Protocols 3 and 4) allows good-quality FISH signals to be attained, in addition to preservation of the 3-D structure of the cell (15,16). Protocol 2. Fixation of cells on slides with organic fixatives or with paraformaldehyde Equipment and reagents • Glass Coplin jars or glass staining troughs . Organic fixative—either methanol, ethanol, acetone—or combinations of fixatives such as methanohacetone (1:1), ethanol:glacial acetic acid (19:1)

• PBS (Protocol 7) • 4% paraformaldehyde in PBS (w/v), freshly made • 0.5% saponin (Sigma) (w/v), 0.5% Triton-X 100 (v/v) in PBS

Method 1. Wash cells attached to glass microscope slides three times with PBS in a glass Coplin jar or staining trough to remove the culture medium. Then follow either A or B. 108

6: Analysis of mammalian interphase chromosomes A. 2. Incubate with the organic fixative of choice, at 4°C for methanol: acetone (1:1) or at -20°C for other organic fixatives. Fixation will also occur if the incubation is performed at room temperature or at 4°C. 3. Wash three times in PBS and store in PBS at 4°C until ready to perform the FISH denaturation step.

B. 2. Incubate cells in freshly prepared 4% paraformaldehyde for 10 min at room temperature.3 3. Rinse in PBS three times. 4. Place in freshly prepared 0.5% saponin and 0.5% Triton-X 100 made up in PBS and incubate for 10 min at room temperature to permeabilize the cells. 5. Rinse in PBS again. a Some laboratories use 4% formaldehyde in PBS (v/v) to fix cells (17).

Of course nothing can beat performing analyses in vivo on live unfixed cells. A number of nuclear proteins have been analysed in vivo by expressing specific proteins tagged with the green fluorescent protein (GFP). It has even been possible to visualize areas of chromosomes by targeting GFP-labelled proteins to specially engineered regions containing exogenous DNA sequences (18). In vivo, in situ hybridization is possible with certain RNA sequences (19). However, the delineation of a specific single species of mammalian chromosome, by hybridization with specific probes, has not yet been achieved in vivo.

2.5 Improving probe penetration After fixation and permeabilization are completed (Protocol 2), there are a number of optional steps which can be used to improve penetration of the probe deep into the nuclei of cells which have been fixed with paraformaldehyde. They include a freeze-fracture procedure (Protocol 3) and depurination of DNA using 0.1 M HC1. Cells are incubated in 20% glycerol/PBS and snapfrozen in liquid N2, these steps can be repeated several times to improve probe penetration. This kind of procedure is a double-edged sword, since the more freeze-thaw steps that are performed the better the probe penetration but at the expense of cells. A happy medium should be reached where there are sufficient freeze and thaw steps to achieve a good signal but no more. We routinely perform between one and five cycles of freeze and thaw, depending on the cell type and length of fixation. One advantage of this regime is that once cells have been frozen in liquid N2 they can then be stored at –80 °C for long periods. 109

Joanna M, Bridger and Peter Lichter Protocol 3. Freeze and thaw procedure to aid the penetration of probe in paraformaldehyde-fixed cells Equipment and reagents • Glass Coplin jar or staining trough • PBS (Protocol 7) « 20% glycerol (v/v) in PBS

Small Dewar of liquid N2 and safety equipment > Plastic freezing containers for slides

Method 1. Place temperature-stable glass microscope slides with attached fixed and permeabilized cells in a Coplin jar or staining trough containing 20% glycerol/PBS (v/v). Leave for 30 min at room temperature (longer incubations are preferable). Slides can be kept in the glycerol solution overnight at 4°C. 2. Take hold of a slide with long forceps and suspend the slide in liquid N2 for at least 30 sec. A good indication of when to remove the slide is when the slide stops making cracking and fizzing sounds. The slide should be covered in a thick layer of white frozen glycerol solution.3 3. Place the slide, right-side up, on the bench and allow the slide to thaw slowly at room temperature. 4. The freeze and thaw procedure can be repeated up to 10 times. When the slide has thawed on the bench return it immediately to the glycerol solution and incubate for a short time, >1 min, in glycerol/PBS before repeating step 2. If you need to perform more than five rounds of freeze and thaw leave the slide in the glycerol solution for 10–30 min. 5. Wash the slides in several changes of PBS to remove the glycerol a At this stage cells can be stored for long periods at –80°C. Put the frozen slide immediately into a plastic slide container and place in a -80°C freezer. Do not forget to either label the slide with a pencil or the slide holder with a permanent marker.

Depurination with HC1 can be performed before or after freeze and thaw, or before or after storage. Place the slides in a Coplin jar or staining trough containing 0.1 M HC1 and incubate for 5-10 min, then wash away the acid with three changes of PBS. One should be aware that this acid treatment may also have an impact on protein structure. It is also possible to perform an RNase treatment which will greatly aid the movement of the probe through nucleoplasm and chromosome territories, by incubating slides with 200 ml of 100 mg/ml RNase A, made up in PBS or 2 X SSC, under a glass coverslip in a 37 °C humidified chamber for 20 min. However, since it is as yet unclear to what extent nuclear RNA is involved in both nuclear and chromosomal structure (20), removal of RNA should only be performed after fixation. 110

6: Analysis of mammalian interphase chromosomes After fixation in paraformaldehyde it is also possible to place cells in organic fixatives such as methanol or methanol:acetone for 5-10 min to improve penetration of the probe. However, in our hands such treatments are unnecessary, and the resulting painted chromosome territories are more diffuse than those seen with paraformaldehyde fixation alone, implying that there are detrimental effects on chromatin structure using organic fixations. If immunofluorescence is not being performed then it is possible to employ a pepsin digestion (see refs 17, 21) to remove proteinaceous components of the nucleus, so permitting better penetration of the probe sequences (Protocol 4). Protocol 4. Pepsin digestion of interphase nuclei Equipment and reagents • Glass Coplin jar « Pepsin solution 50 mg/ml (w/v) in 0.01 M HCI pH 2.0

• 4% paraformaldehyde • 1 M NaSCN (optional) . PBS (Protocol 1)

Method 1. Pretreat fixed and permeabilized slides in 1 M NaSCN overnight at 4°C.a 2. Wash the slides in PBS. 3. Place the fixed and permeabilized slides in the pepsin solution for 10-20 min at 37°C. 4. Wash in PBS. "This pretreatment is optional but it does not effect the 3-D structure of the nuclei.

3. Probes 3.1 Chromosomal painting probes Chromosomal painting probes are a collection of sequences generated from a specific chromosome which, when used in FISH, will paint that chromosome sufficiently so that the whole chromosomal area is delineated. Before interphase chromosomes are painted with such probes, the probe should be hybridized to control metaphase chromosomes (Chapter 4). Chromosome painting probes can be generated by various means, and the different types of probe have all been used successfully to paint interphase chromosomes. Generally, the individual chromosomal sequences are obtained from flow-sorted or microdissected chromosomes, or by using human chromosome sequences in somatic cell hybrids as the source material. Either chromosome-specific DNA libraries are created or PCR protocols are

111

Joanna M. Bridger and Peter Lichter Table 1. Different types of chromosomal painting probes Lambda phage and cosmid chromosomal painting libraries (available from the American Tissue Culture Company, ATCC) Cloned human chromosome painting libraries made from flow-sorted chromosomes (27) Degenerate oligonucleotide-primed PCR products (DOP-PCR) from flow-sorted chromosomes (28-30) Interspersed repetitive sequences (IRS) amplified by PCR from flow-sorted chromosomes (31) IRS-PCR on DNA from somatic-cell hybrid lines containing a single human chromosome (32, 33) Linker adapter PCR from flow-sorted chromosomes (34, 35) Chromosome arm painting probes from microdissected metaphase chromosomes (22)

used to amplify sequences from the chromosome in question. There are also a number of commercially available whole chromosome and chromosome-arm specific painting probes (22) which can be combined to paint entire chromosomes or used separately. Although, most studies analysing interphase chromosomes have employed human specific sequences, the technology for flow-sorting has improved allowing other mammalian chromosomes to be sorted, so that individual chromosomes can now be painted in many other mammalian species (23-26). Table 1 summarizes the different human chromosomal painting probes currently available.

3.2 Probe labelling Probes can be labelled in several ways depending on the size of the original DNA fragments (Chapter 4). For good painting of chromosomes in interphase nuclei the probe length should be 200–500 base pairs. If probe sequences are smaller than this, they can be labelled by incorporating nucleotides conjugated to reporter molecules such as biotin, digoxigenin, or dinitrophenol (DNP) in a PCR mix or in a subsequent amplification round of PCR (13, 28, 36, and Chapter 4). Nick translation is commonly used to incorporate hapten-conjugated nucleotides into larger probe sequences (13, 37, 38, and Chapter 3). When using FISH probes that contain tandem or interspersed repetitive sequences, a process termed chromosomal in situ suppression (CISS, 4, 24) is required (Chapter 4). The C0t1 fraction of DNA is generally used as competitor—human and mouse C0t1 DNA is commercially available from, for example, Boehringer Mannheim and Gibco/BRL. The probe mix (10 |xl) for one slide should contain: 100-500 ng labelled chromosome paint, 3-7mgC0t1i DNA (for IRS-PCR amplified probe use >30 mg), and 1-3 mg salmon sperm DNA. The probe mix can either be lyophilized in a SpeedVac or ethanol112

6: Analysis of mammalian interphase chromosomes precipitated followed by two washes in 70% ethanol and then drying. The dried probe is resuspended in hybridization mix (50% deionized formamide, 2 X SSC, 10% dextran sulfate, 50 mM sodium phosphate and 1% Tween-20) (see also Chapter 4). Obviously, if analysing repetitive sequences themselves do not includeC0t1DNA in the probe mix.

4. Fluorescence in situ hybridization 4.1 Denaturation In most cases both the probe and the sample need to be denatured so that both single-stranded probe and target sequences are available for hybridization to each other. If single-stranded probe sequences are used then denaturation of the probe is unnecessary, but it can be used to destroy any secondary structures. The probe should be denatured at 75 °C for 5 min and, when competitor DNA is included, an annealing step is required for 10–60 min at 37 °C before applying the probe to the denatured sample. There are really only two methods used for denaturation of the sample, these either use a combination of formamide, salt, and high temperature or an alkaline solution of NaOH to denature the DNA. Most laboratories use formamide denaturation (Protocol 5), but NaOH denaturation (Protocol 6) has also been used successfully (see, for example, ref. 39). We have tested both methods and find that formamide denaturation gives better signals and is more consistently successful (Kurz and Lichter, unpublished observations). Protocol 5.

Denaturation of sample by formamide

Equipment and reagents Two glass Coplin jars with lids Water-baths heated to at least 75°C and 70°C 37 °C water-bath or incubator Moist incubation chamber, prewarmed Prewarmed hot plate at 37-40°C 20 x SSC: 3.0 M NaCI, 0.3 M sodium citrate

• Glass coverslips at 37-40°C (18 x 18 mm or 22 x 22 mm) 70% deionized formamide, 2 x SSC; adjust to pH 7.0 with HCI • 50% deionized formamide, 2 x SSC pH 7.0 • Rubber cement • Mercury thermometera

A. Denaturation of cells fixed in paraformaldehyde 1. Place 70-100 ml of 70% deionized formamide, 2 x SSC pH 7.0, and 70-100 ml 50% deionized formamide, 2 x SSC pH 7.0 in glass Coplin jars. Place in a water-bath set to 75°C and switch on. The optimal temperature for denaturation of samples fixed in pFA is 73-75°C. The temperature of the water-bath should be raised 1°C for each slide being denatured, i.e. for 3 slides use 76°C to achieve a denaturation temperature of 73°C.b 113

Joanna M. Bridger and Peter Lichter Protocol 5.

Continued

2. After the slides have been permeabilized keep them in PBS until denaturation. 3. Wipe the excess buffer from the reverse side of the slide and from both ends of the front of the slide. 4. Place the slides into the 70% deionized formamide, 2 x SSC as quickly as possible. Incubate for 3 min. Place a mercury thermometer into the Coplin jar and monitor the temperature throughout the incubation.a If the temperature is too high (>75–76°C) take the jar out of the waterbath and/or take the lid off. 5. Move the slides carefully, but with speed, to the 50% deionized formamide, 2 x SSC. Incubate at 73-75°C for 1 min. Continue to monitor the temperature. 6. While the slides are in 50% formamide, 2 x SSC, place the denatured probe mix onto prewarmed coverslips on a hot plate. Use 18 x 18 mm or 22 x 22 mm glass coverslips for 10 ml of the probe mix. 7. Take the slides out of the 50% formamide 2 x SSC, wiping the excess fluid from the reverse of the slide and the ends on the front of the slide. Flick off the excess fluid with a quick wrist movement and present the slide to the probe solution on the coverslip. It is important at this stage to work as quickly as possible, and without trapping any air bubbles between the slide and coverslip. Flip the slides over so that the coverslip is on the top.c 8. Seal around the edge of the coverslip with rubber cement. B. Denaturation of cells fixed in organic fixatives 1. Preheat 70% formamide, 2 x SSC pH 7.0 to 70°C. 2. Place the slides in the preheated 70% formamide, 2 x SSC for 2 min. 3. Perform steps 7 and 8 of part A. 8Use a mercury thermometer as the temperature is critical. * Never put more than 4 slides into a Coplin jar as this will reduce the temperature too much. c When hypotonically treated and dropped nuclei are being prepared for denaturation, slides are dehydrated and dried pre- and post-denaturation by putting them through a series of icecold 70%, 90%, and 100% ethanol. The probe is then added to the dry slide.

Protocol 6.

Denaturation using NaOH

Equipment and reagents • Glass Coplin jars or staining dishes . 0.2 M NaOH in 70% ethanol

• PBS (Protocol 7)

114

6: Analysis of mammalian interphase chromosomes Method 1. Place the fixed slides in the glass jar and add the NaOH denaturing solution. 2. Incubate the slides for 5 min at room temperature. 3. Wash in PBS and add denatured probe to the chosen area on the slide and seal with rubber cement (Protocol 5A).

4.2 Hybridization Hybridization is generally performed at 37 °C for approximately 18 h, but longer incubations can be used and can even be beneficial to signal intensity if hybridizing to low-abundance sequences. Longer incubations can only be used if the slides are kept humid. This can be achieved by either placing slides in a floating enamel or plastic tray in a water-bath or in a humid chamber in a 37 °C water-bath. For longer hybridization times it is advisable to use a humid chamber in an incubator.

4.3 Washing Washing the hybridized slides removes non-specifically bound probe and so reduces background staining. It is possible to change the stringency of FISH experiments by adjusting post-hybridization washes. Protocol 7 works well in combination with the denaturation regimes presented here, and is a standard washing procedure in many laboratories (Chapter 4). Protocol 7. Washing hybridized slides Equipment and reagents • Two glass Coplin jars or staining troughs • Two heated water-baths, one at 42-45°C and one at 60°C . 0.1 x SSC, pH 7.0

• 50% formamide, 2 x SSC, pH 7.0 (see Protocol 5 for composition of 20 x SSC, pH 7.0 • 4 x SSC

Method 1. Prewarm the glass Coplin jars in a 42°C water-bath. 2. Place 50% formamide, 2 X SSC, pH 7.0 at 42°C and 0.1 X SSC, pH 7.0 at 60°C. 3. Remove the rubber cement from around the coverslip and place slides into the Coplin jar containing 50% formamide, 2 x SSC, pH 7.0 at 42°C. Wash the slides with gentle shaking for 3x5 min. 4. Replace 50% formamide, 2 x SSC, pH 7.0 with 0.1 x SSC, pH 7.0 at 60°C and wash in the 42°C water-bath. 5. Wash the slides for 3 x 5 min.

115

Joanna M. Bridger and Peter Lichter Protocol 7.

Continued

6. Place the slides into the 4 x SSC at room temperature. Either continue with probe detection (below) or store slides at 4°C in 4 x SSC for few days.

4.4 Detection of reporter molecules Before detection of the reporter molecules can be performed a blocking step is required to prevent non-specific binding of the detection reagents. Either 4% bovine serum albumin (BSA) or 4% dried milk powder in 4 X SSC (w/v) can be used. Place 200 (ml onto the hybridized area of the slide and cover with a glass coverslip. Place in a humid chamber at room temperature for 5-10 min. Most reporter haptens (Chapter 4) are detected via a specific antibody conjugated to a fluorochrome, but biotin can be detected via its affinity for avidin/streptavidin which can also be conjugated to a fluorochrome. Many of the species-specific antibodies used to detect reporter molecules in FISH are generally raised in animals other than mouse, rabbit, or guinea-pig. This is convenient when performing immunofluorescence in combination with FISH (Protocol 8), since it is very important that the secondary antibody used in the indirect immunofluorescence should not bind to the antibodies used to detect the reporters. It is also possible to amplify hybridization signals by building up different layers of detection reagents (13, 40, Chapter 4). FISH detection reagents are made at an appropriate dilution in 1 % BSA or dried milk and incubated on the slide for 30 min at 37 °C, or for 1 h at room temperature. Slides are washed at 42 °C in 4 X SSC, 1% Tween-20 (v/v) for 15 min with three changes of buffer. Do not be tempted to perform longer secondary antibody or avidin\streptavidin incubations as this will lead to non-specific binding, thus increasing the background to signal ratio.

5. Immunofluorescence in combination with FISH In most cases, combining immunofluorescence with FISH does not cause problems with antigen detection. However, occasionally there may be problems with the distortion or destruction of the antigen by either the fixation or denaturation regimes. For example, proliferating cell nuclear antigen (PCNA), the auxiliary coenzyme for polymerase delta, used to visualize active replication clusters in S-phase in mammalian cells (41), cannot be recognized by certain antibodies if fixatives such as paraformaldehyde or formaldehyde have been used. In contrast, certain antibodies recognizing the cell proliferation marker pKi-67, which can be used to determine specific cell-cycle stages in human cells (42), cannot be used on cells which have been fixed with alcohol-based fixatives. So one must choose the fixation regime suitable for the antibody/antigen. It is also possible that the denaturation procedure for FISH may hinder antigen detection. In this situation the immunofluorescence reactions can be performed prior to denaturation. 116

6: Analysis of mamalian interphase chromosomes

figure 2 shows examples of interphase whole chromosome territories detected by FISH with indirect immunofluoresence for a nuclear antigen.

Figure 2. Indirect immunofluorescence in combination with FISH. Chromosome territories have been delineated in human cells (green). The nuclear structures displayed in blue and red have been revealed using indirect immunofluorescence after the FISH procedure. Panel A shows the nuclear periphery using a human autoimmune serum reacting with an antigen in the nuclear rim. The secondary antibody used was anti-human conjugated to the blue emitting dye 7-amino 4 methykoumarin-3-acetic acid (AMCA). The red structures are exogenous assembled vimentin filaments in a stable transfected cell line. These structures were revealed using a monoclonal anti vimentin antibody and then an anti-mouse antibody conjugated to TRITC, This image was reproduced from ref. 48. Panel B. Early G1 human dermal fibroblast, the nuclear periphery was revealed using a monoclonal anti-human lamin antibody with an anti-mouse AMCA as the secondary antibody. The antigen pKi-67 is shown in red and is revealed by a rabbit anti-Ki-67 antibody and a secondary antibody, anti-rabbit conjugated to Texas Red. All of the secondary antibodies were raised in goat. Both images were generated by CCD. Scale bars 55mm. 117

Joanna M. Bridger and Peter Lichter Protocol 8.

Indirect immunofluorescence of interphase cells

Equipment and reagents • PBS (Protocol 1) . Primary antibody appropriately diluted in PBS containing 1% (v/v) newborn or fetal calf serum (NCS, PCS) • Humid incubation chamber

• Fluorochrome-conjugated secondary antibody diluted to 10–20 (mg/ml in PBS/1% NCS • 22 x 50 mm coverslips . Glass Coplin jar

Method 1. After fixation rinse the slides in PBS. Place 100–200 ml of the primary antibody on the slide and cover with a 22 x 50 mm coverslip. 2. Place in a humid chamber and incubate either at 37 °C for 30 min, or at room temperature for 1 h, or at 4°C overnight.a 3. Wash three times for 15 min each time in PBS in a glass Coplin jar. 4. Place 100-200 ml of the secondary antibody (fluorochrome-conjugated) onto the slide and cover with a 22 X 50 mm glass coverslip. 5. Place in a humid chamber and incubate either at 37°C for 30 min, or at room temperature for 1 h, or at 4°C for 4 h. 6. Wash in PBS for 3 x 5 min.b a Note that some antigens cannot be revealed with shorter incubations at higher temperatures and so in some cases it is imperative to perform overnight incubations at 4°C (43). bDetergent such as Tween-20 at a concentration of 0.1% (v/v) can be added to the PBS if the fluorescence signal is too intense or the background staining is too high.

5.1 Primary and secondary antibody incubations after FISH This combination of FISH and immunofluorescence is the best for optimizing antigen detection, since denaturation procedures can cause decreased antigen discrimination when primary or primary and secondary antibodies are applied before FISH. The primary antibody can be included in the solutions used to detect the FISH reporter haptens, or it can be incubated with the sample separately after detection of the FISH signal. Most antibodies that have been tried in our hands are suitable for postdenaturation incubation, these include those reacting with nuclear structural proteins, chromatin associated proteins, RNA polymerases, nucleolar antigens, centromere associated proteins, and proteins of nuclear bodies. However, this combination of FISH and indirect immunofluorescence cannot be used if denaturation steps distort or destroy the antigen of interest. Another phenomenon which appears to occur when performing FISH in combination with immunofluorescence is that, with some antibodies, the distribution of the antigen may appear to change after a short time. More 118

6: Analysis of mammalian interphase chromosomes specifically, the antibody and/or antibodies appear to become associated with the nucleoli. This can be prevented by performing a postimmunofluorescence fixation in 4% paraformaldehyde for 10 min immediately after washing and prior to mounting.

5.2 Primary antibody incubation predenaturation and secondary antibody incubation postdenaturation The primary antibody should be incubated prior to denaturation and the secondary antibody postdenaturation when the antigen is deleteriously affected by the denaturing step. To establish if the antigen of choice is destroyed by this process perform a mock denaturation with no addition of FISH probe and then perform indirect immunofluorescence. If necessary the primary antibody can be fixed in place with paraformaldehyde.

5.3 Primary and secondary antibody incubations prior to FISH denaturation If the antigen of interest and the primary antibody are distorted or destroyed by the denaturation process then the antibody incubations can be performed after fixation and permeabilization and before denaturation. This will require a postimmunolocalization fixation in 4% paraformaldehyde, so that conjugated fluorochromes do not become detached from the antibodies during denaturation. This sequence of steps can lead to a dull immunohistochemical signal due to adverse effects of the denaturation solution on the labelled secondary antibody.

5.4 Reporter-conjugated primary antibody incubated predenaturation Another possibility is to use a primary antibody with a reporter molecule conjugated to it which can survive the denaturation conditions. Such reporters include the haptens used to label probes for FISH (biotin, digoxigenin, or dinitrophenol (DNP)). These molecules can then be detected by the same protocols used to detect FISH probes. To use this method demands an adequate amount of antibody, as labelling procedures need a sufficient concentration of protein to work and there is also a possibility that epitope recognition can be destroyed by the conjugation. Commercial kits are available for conjugation.

6. Mounting the slides To view under high-power oil-immersion objectives, the slides need to be mounted. This is usually performed in 90% glycerol with the addition of antifade compounds such as 1,4-diazabicyclo-[2.2.2]-octane (DABCO), 119

Joanna M. Bridger and Peter Lichter p-phenylenediamine-dihydrochloride, or isopropylgallate. If a mountant contains Mowiol, or if one uses Vectashield (Vecta Laboratories), then these possess a self-sealing property and do not require sealing. Place 10–50 ml of the mounting solution onto a glass coverslip. Place the slide over the top, then put the slide face down onto a clean tissue and allow the excess mountant to squeeze out from under the coverslip by applying very slight pressure.

7. Analysis For 2-D analyses, such as those performed on organically fixed flattened cells the charged-coupled device (CCD) camera digital acquisition system, without deconvolution, is the method of choice (Chapter 4). However, one of the main aims of interphase studies is to analyse cellular structures in three dimensions, so the out-of-focus signal from regions below and above the plane being imaged must be either removed or not collected. All 3-D data are constructed from a series of 2-D images. There are only two light microscopy technologies detecting fluorescence signals which are commonly used to view whole cells or nuclei. The first is wide-field optical microscopy using a photon-collecting device such as a cooled CCD camera or video, this is used in combination with deconvolution algorithms to remove the out-of-focus data (44, 45). The other is confocal laser scanning microscopy (CLSM) in which only the fluorescence from the plane of focus is collected, apart from a small region above and below. It is not within the realm of this chapter to discuss the details of these different technologies, but a short overview will be given concerning how and why they should be used. For further details see ref. 46, and references therein. Both systems have advantages and disadvantages, although both are used with great effect in laboratories studying nuclear structure. Most CLSM studies have used systems that permit the excitation of two fluorochromes in red and green (e.g. rhodamine or Texas Red and fluorescein). However, the number and wavelengths of the lasers can be varied so that more fluorochromes can be excited and detected (e.g. the blue fluorochromes DAPI or AMCA and fluorochromes emitting in the far red, such as cyanine 5). CLSM can be used to perform 3-D analyses on series of 2-D confocal images that are collected without having to perform 3-D reconstructions. However, basic 3-D viewing programmes are available as standard on CLSM such as stereo pairs, red-green stereo images (which need to be viewed using a pair of red-green stereo glasses) and a throughview capability. Furthermore, comparisons between the spatial relationship of structures can be performed immediately without further computation. Figure 3 shows a series of confocal images through a nucleus with painted chromosomes and a 3-D reconstruction of the series. Some groups prefer to use deconvolution algorithms after image acquisition because of the superior sensitivity of the fluorescence detectors such as 120

6. Analysisi of mammalianinterphasechromosomes

Figure 3. A 3-D reconstruction using a z-serios of conlbcal images. This image displays a 3 D reconstruction of a polyploid human cell nucleus in which a single chromosome paint has been used (green) and vimentin filaments have bean revealed by indirect imrminofluoresoence (red). This imago has been produced using a series of con focal sections taken at 0.5 mM intervals through a 3-D preserved nucleus using a volume rendering program. The image has been reproduced from ref. 48.

CCD camems. The increased resolution that can he achieved in in the -axis also xix more accurate image reconstruction. In addition, the number of chromes chromes which can he delected is easily increased, making morespatiallcomparisons between structures possible. Deconvolution involves the subtraction of the out-of-focus fluorescence signal by computation. Deconvolution software is now available (e.g. Applied precision Inc.. AutoQuant Imaging Inc.. Life Sciences Resources. Scanalytics Ine., Scientific Volume imaging B.V.. and VayTek) but requires computers capable of running such programs. For a much more detailed discussion of (he merits of CCD camera image acquisition and deconvolution see ret. 47.

References 1. Turner. B.M.. Birley. A.J.. and Lavender..1 ( 1 9 9 2 ) . (Cel. 69. 375.. 2. Craig J.M and Bickmore W.A. {1993}.Bioessays 15, 349. .1. I.ichler. P., Cremer T., Borden. .J.. Manuelidis. I,., and ward D.C. ( 1 9 9 8 ) . Genet80.224 4. Cremer. T. (1998 ) Hum.Genet 80 5. Pinkel. D. Landegent .1., Collins.

C.Fuseoe

, Seagraves R.,

Gray. J. W. (1998). Proc: Natl Acad. Sci, USA. 85, 9138. 121

Lueas

Hum

and

Joanna M. Bridger and Peter Lichter 6. Dietzel, S., Jauch, A., Kienle, D., Qu, G., Holtgreve-Grez, H., Eils, R., Munkel, C., Bittner, M., Meltzer, P.S., Trent, J.M., and Cremer, T. (1998). Chrom. Res., 6, 25. 7. Gosden, C.M., Davidson, C., and Robertson, M. (1992). In Human cytogenetics: a practical approach, Vol. I (ed. D.E. Rooney and B.H. Czepulkowski), pp. 31-54. IRL Press, Oxford. 8. Ferguson, M. and Ward, D.C. (1992). Chromosoma, 101, 557. 9. Wiegant, J., Kalle, W., Mullenders, L., Brookes, S., Hoovers, J.M.N., Dauwerse, J.G., van Ommen, G.J.B., and Raap, A.K. (1992). Hum. Mol. Genet., 1, 587. 10. Gerdes, M.G., Carter, K.C., Moen, P.T. Jr, and Lawrence, J.B. (1994). J. Cell Biol, 126, 289. 11. Tocharoentanaphol, C., Cremer, C., Schrock, E., Blonden, L., Kilian, K., Cremer, T., and Reid, T. (1994). Hum. Genet., 93, 229. 12. Haaf, T. (1996). Biotechniques, 21,1050. 13. Bridger, J.M., Lampel, S., and Lichter, P. (1997). In Cells, a laboratory manual (ed. D. Spector, R. Goldman, and L. Leinwand). Cold Spring Harbor Press, New York. 14. Cremer, T., Kurz, A., Zirbel R., Dietzel S., Rinke B., Schroeck E., Speicher M., Mathieu U., Jauch A., Emmerich P., Scherthan H., Ried T., Cremer C., and Lichter P. (1993). Cold Spring Harbor Symp. Quant. Biol., 58, 777. 15. Kurz, A., Lampel, S., Nickolenko, J.E, Bradl, J., Benner, A., Zirbel, R.M., and Lichter, P. (1996). J. Cell Biol, 135, 1195. 16. Zirbel, R.M., Mathieu, U.R., Kurz, A., Cremer, T., and Lichter, P. (1993). Chrom. Res., 1, 93. 17. Eils, R., Dietzel, S., Bertin, E., Schroeck, E., Speicher, M.R., Ried, T., RobertNicoud, M., Cremer, C., and Cremer, T. (1996). J. Cell Biol., 135, 1427. 18. Robinett, C.C., Straight, A., Li, G., Sudlow, G., Murray, A., and Belmont, A.S. (1996). J. Cell Biol, 135,1685. 19. Paillasson, S., van de Corput, M., Dirks, R.W., Tanke, H.J. Robert-Nicoud, M., and Ronot, X. (1997). Exp. Cell Res., 231, 226. 20. Singer, R. and Green, M. (1997). Cell, 91, 291. 21. Coonen, E., Dumoulin, J.C., Ramaekers, F.C., and Hopman, A.H. (1994). Hum. Reprod., 9, 533. 22. Guan, X-Y., Zhang, H., Bittner, M., Jiang, Y., Meltzer, P., and Trent, J. (1996). Nat. Genet., 12,10. 23. Burkin, D.J., O'Brien, P.C., Broad, T.E., Hill, D.F., Jones, C.A., Wienberg, J., and Ferguson-Smith, M.A. (1997). Chrom. Res., 5,102. 24. Langford, C.F, Telenius, H., Carter, N.P., Miller, N.G., and Tucker, E.M. (1993). Cytogenet. Cell Genet., 61, 221. 25. Hoebee, B., de Stoppelaar, J.M., Suijkerbuijk, R.F., and Monard, S. (1994). Cytogenet. Cell Genet., 66, 277. 26. Rabbitts, P., Impey, H., Heppel-Parton, A., Langford, C., Tease C., Lowe, N., Bailey, D., Ferguson-Smith, M., and Carter, N. (1995). Nat. Genet., 9,369. 27. Collins, C., Kuo, W.L., Seagraves, R., Fuscoe, J., Pinkel, D, and Gray, J. (1991). Genomics, 11, 997. 28. Telenius, H., Pelmear, A.H., Tunnacliffe, A., Carter, N.P., Behmel, A., FergusonSmith, M.A., Nordenskjold, M., Pfagner, R., and Ponder, B.A. (1992). Genes, Chrom., Cancer, 4, 257. 122

6: Analysis of mammalian interphase chromosomes 29. Carter, N.P., Ferguson-Smith, M., Perryman, M.T., Telenius, A.H., Pelmear, A.H., Leversha, M.A., Glancy, M.T., Wood, S.L., Cook, K., and Dyson, H.M. (1992). J. Med. Genet., 29,299. 30. Bailey, D.M., Carter, N.P., de Vos, D., Leversha, M.A., Perryman, M.T., and Ferguson-Smith, M.A. (1993). Nucleic Acids Res., 21,5117. 31. Suijkerbuijk, R.F., Matthopoulos, D., Kearney, L., Monard, S., Dhut, S., Cotter, F.E., Herbergs, J., van Kessel, A.G., and Young, B.D. (1992). Genomics, 13, 355. 32. Lichter, P., Ledbetter, S.A., Ledbetter, D.H., and Ward, D.C. (1990). Proc. Natl Acad. Sci. USA, 87, 6634. 33. Lengauer, C. Riethman, H., and Cremer, T. (1990). Hum. Genet., 86,1. 34. Chang, K.S., Vyas, R.C., Deaven, L.L., Trujillo, J.M., Stass, S.A., and Hittelman, W.N. (1992). Genomics, 12, 307. 35. Vooijs, M., Yu, L.C., Tkachuk, D., Pinkel, D., Johnson, D., and Gray, J.W. (1993). Am. J. Hum Genet., 52, 586. 36. Bohlander, S.K., Espinosa, M., Le Beau, M., Rowley, J.D., and Diaz, M.O. (1992). Genomics, 13,1322. 37. Langer, P.R., Waldrop, A.A., and Ward, D.C. (1981). Proc. Natl Acad Sci. USA, 7, 6633. 38. Lichter, P. and Cremer, T. (1992). In Human cytogenetics: a practical approach, Vol. I (ed. D.E. Rooney and B.H.Czepulkowski), p. 157. IRL Press, Oxford. 39. Xing, Y., Johnson, C.V., Moen, P.T. Jr, McNeil, J.A., and Lawrence, J.B. (1995). J. CellBiol., 131,1635. 40. Pinkel, D., Straume, T., and Gray, J.W. (1986). Proc. Natl Acad. Sci. USA, 83, 2934. 41. Kill, I.R., Bridger, J.M., Campbell, K.H.S., Maldonado-Codina, G., and Hutchison, C.J. (1991). J. Cell Sci., 100, 869. 42. Bridger, J.M., Kill, I.R., and Lichter, P. (1998). Chrom. Res., 6,13. 43. Bridger, J.M., Kill, I.R., O'Farrell, M., and Hutchison, C.J. (1993). /. Cell Sci., 104, 297. 44. Svedlow, J.R., Sedat, J.W., and Agard, D.A. (1996). Deconvolution in optical microscopy. In Deconvolution of images and spectra (ed. P.A. Jansson), p. 284. Academic Press, San Diego, CA. 45. Arndt-Jovin, D.J., Robert-Nicoud, M., Kaufman, S.J., and Jovin, T.M. (1985). Science, 230, 247. 46. Harvath, L. (1994). Overview of fluorescence analysis with the confocal microscope. In Methods of moleular biology, Vol. 34 (ed. L.C. Javois), pp. 337. Humana Press, Totowa, NJ. 47. Paddy, M.R. (1997). Determining nuclear structure using the fluorescence light microscope. In Methods in cell biology, Vol. 53 (ed. M. Berrios), pp. 49. Academic Press, San Diego, CA. 48. Bridger, J.M., Herrmann, H., Muenkel, C., and Lichter, P. (1998). J. Cell Sci., 111, 1241.

123

This page intentionally left blank

7

Fluorescence in situ hybridization in whole-mount tissues ABBY F. DERNBURG

1. Introduction This chapter outlines techniques for the investigation of chromosome organization using fluorescence in-situ hybridization (FISH) to DNA in wholemount tissues. Emphasis is placed on procedures that preserve global tissue morphology. Nuclear architecture is not static, but may vary with time, cell type, and stage of development or differentiation. Therefore, many questions demand the examination of chromosome organization in a relevant developmental, spatial, and temporal context. These techniques were designed to permit the investigation of a variety of such questions, including chromosome pairing (1, 2), interaction of chromosomal DNA with proteins or other nuclear structures (3), and the global arrangement of chromosomes within the nucleus (4). Whole-mount FISH also allows direct karyotyping of gametes or embryos ((5), Figure 1). To preserve the structure of a nucleus, cell, or tissue while performing FISH involves a trade-off. Hybridization requires the denaturation of target chromosomal DNA, and also that the tissue be permeable enough to allow probes to diffuse and bind to the chromosomal target. At the same time, chromosome and cell structure must be maintained well enough so that meaningful conclusions may be drawn. While the methods in this chapter have been developed to minimize damage to morphology, one consequence of this preservation is some sacrifice of signal and an increase in background compared to methods in which chromosomes are fixed with acid and flattened on a glass slide prior to hybridization. Nevertheless, with the reagents currently available and application of sensitive microscopy, these methods are quite robust. These procedures were developed primarily using Drosophila melanogaster as a model system, but they have been readily adapted to tissues from other organisms. The protocols given here are designed to be as general as possible, with some specific guidelines based on work with tissues from Drosophila, Caenorhabditis elegans, and budding yeast. In addition, an effort has been

Figure 1. Examples of whole-mount FISH applications in Cacnorhabditis elegans. A: Results of an experiment to karyotype embryos with respect to their sex chromosome constitution. Hybridization was carried out using a cosmid probe to a region of the X chromosome, shown in green. One of the two embryos (lower) is XX (hermaphrodite), while the other (upper) has a single X chromosome in each nucleus (male). Scale bar 10 mm. B: The green probe recognizes an unstable extrachromosomal DNA array, which is present in multiple copies in some of these germline nuclei, and absent in others; FISH is thus useful as a way of performing mosaic analysis. The red probe recognizes a region of chromosome V, the homologous chromosomes are paired at this stage of meiosis, as evidenced by the fact that only one red signal is detected in each nucleus. Scale bar 5 mm. C: Following the progression of chromosome pairing within the temporal and spatial context of the worm gonad, The tip of the gonad (upper left in this image) contains a zone of mitotic nuclear proliferation. As nuclei progress through meiosis, the homologous chromosomes pair, and later unpair prior to the first meiotic division. The oldest nuclei in this image (bottom left) are at the end of meiotic prophase, or diakinesis. Scale bar 20 um.

126

7: Fluorescence in situ hybridization in whole-mount tissues made to survey the available reagents for probe labelling and detection, and to evaluate their performance in whole-mount FISH. Successful application of FISH requires four separate experimental components: probe synthesis; tissue preparation; hybridization and detection; and analysis by microscopy. Each of these will be discussed separately below.

2. Probe synthesis and labelling 2.1 General considerations Either single-copy or repetitive DNA elements can be targeted in FISH, but the more abundant the target the greater the likelihood of a detectable signal. Nevertheless, using high-quality microscope equipment and some of the better fluorescent labelling reagents available, it has been possible to detect probes covering as little as a few kilobases of genomic DNA in Drosophila embryos, oocytes from C. elegans, and other tissues, and further improvements in labelling technology may push the limit further still. In principle, either DNA or RNA may be used to probe genomic DNA sequences, but in practice it is simpler to synthesize and use DNA probes due to their greater chemical stability. A variety of options are available for labelling DNA with haptens or fluorophores. Hybridization procedures in which the target nucleic acid is bound to a membrane or spread on a microscope slide typically employ probes labelled by: nick translation; random priming; PCR-based incorporation of labelled nucleotides; or 5'- or 3'-endlabelling methods (Chapters 3, 4). In hybridization to intact nuclei additional constraints are imposed: the probe fragments must be very small in order to diffuse through fixed tissue; and the labelling procedure itself is more critical since such experiments frequently involve working near a threshold imposed by background fluorescence where every little bit of signal helps. A 3'-endlabelling scheme has given excellent results when compared with several other methods, and it has the additional advantage that it can be used to label any DNA, whether single- or double-stranded, oligonucleotide or cosmids, or total genomic DNA. Unless synthetic oligonucleotides or very small PCR products (<150 bp) are being used, the DNA must be fragmented before labelling so that the probe molecules are sufficiently small. A potential disadvantage of 3'-end-labelling is that terminal deoxynucleotidyl transferase (TdT) may be more finicky than some other DNA polymerases in terms of its willingness to incorporate modified nucleotides. However, good results have been achieved with biotinylated and digoxigeninlabelled nucleotides, and with the fluorescent nucleotides included in Table 1.

2.2 Why use fluorescence-based detection? Fluorescent probes provide several major advantages for the in-situ localization of DNA sequences. In contrast to other histochemical detection schemes 127

Abby F. Dernburg that involve enzymatic precipitation of a coloured product, fluorescent probes remain physically associated with the site of hybridization, providing optimal spatial localization. The signal to background ratio provided is also superior, since the tissue itself does not usually fluoresce strongly. Fluorescent signals can be localized at high resolution within a three-dimensional (3-D) volume, especially using confocal laser scanning microscopy (CLSM) or deconvolution-based optical sectioning microscopy (see Section 6), and this is likely to be critical when questions regarding subnuclear localization are being investigated. Multiple fluorescent probes can be detected and discriminated in the same sample through the use of specific excitation and emission filters. Fluorescent reagents are stable for long periods if stored frozen and protected from light, and detection is simple and rapid. The only major disadvantage is that specimens deteriorate over time due to photobleaching and other degradative processes. However, samples can be archived through image acquisition, or by permanent mounting in organic medium (6).

2.3 Choice of labelling and detection reagents A number of companies are currently marketing reagents for FISH, including ready-made probes to a variety of sequences (mostly to human and mouse targets), as well as probe labelling and detection reagents. These can be expensive, and the abundance of options makes the choice confusing. In this section, criteria are outlined for selecting reagents and some specific recommendations are given. 2.3.1 Direct vs. indirect detection A choice must be made between labelling probes directly with fluorophores, usually by incorporating fluorescently-conjugated nucleotides, or using hapten-labelled probes, most often by incorporating biotinylated nucleotides or digoxigenin-dUTP, which are then detected indirectly. The chief advantage of direct labelling is convenience; after hybridization and washing, probes can be visualized without time-consuming secondary staining procedures. Signal to background levels achieved with direct-labelled probes are often as good as or better than those seen using secondary detection, although the signals themselves tend not to be as bright. There are enough different fluorescent nucleotides currently available to distinguish up to three probes in the same sample, plus a counterstain, while there are only two well-developed hap ten labels (biotin and digoxigenin). Thus to distinguish three probes in the same sample without using combinatorial labelling schemes, at least one of them should probably be directly fluorescent. In particular cases, fluorescent probes (especially short oligos) will penetrate tissues better than secondary detection reagents. Hapten-labelled probes coupled with secondary detection, on the other hand, are more economical, given the expense of some of the better fluorescent dNTPs. Labelling a probe with digoxigenin (Boehringer Mannheim) or 128

7: Fluorescence in situ hybridization in whole-mount tissues biotin (Life Technologies or other suppliers) provides versatility, since the same probe may be detected with different secondary reagents (Chapter 4). Another small advantage is that the fluorescent reagents are only added towards the end of a hybridization procedure, minimizing photodamage during sample handling. Perhaps most importantly, secondary detection allows for signal amplification, since an antibody or avidin molecule that binds to a single hapten molecule can itself carry multiple fluorophores (Chapter 4). This may be particularly helpful when laser illumination is used to excite the sample, as in confocal microscopy, or when the signal will be detected by eye. However, the use of multiple layers of antibodies and/or biotin-avidin reagents to further enhance the signal tends to produce high fluorescent background in whole-mount tissues. For this reason, a single application of a secondary detection reagent is usually optimal. 2.3.2 Choice of fluorophores FISH probes may be labelled and detected using a variety of fluorophores. The recently developed cyanine dyes Cy2 and Cy3 (Cy is an Amersham trademark) provide improved photostability and brightness over their spectral counterparts, fluorescein and rhodamine. Indodicarbocyanine, or Cy5, absorbs red light and emits in the infrared, providing an additional region of the spectrum for probe detection. With either laser or mercury arc lamp illumination, Cy3 is probably the single best fluorophore readily available for FISH procedures. It is not useful, however, when using propidium iodide (PI) as a counterstain. When detecting multiple probes in the same sample, some consideration should be given to the relative sensitivities of different reagents, fluors, and detection strategies, such that the least abundant genomic target is probed in the most easily detectable way. Fluorescent nucleotides are available from a number of commercial suppliers. Fluorescein- and other green-labelled dNTPs have not shown dramatic differences in performance in my hands. Cy2 represents an advance over fluorescein in terms of photostability, but is not yet available directly conjugated to nucleotides. Cy3 has proven to be significantly brighter and more photostable than other red-emitting fluorophores (7, 8), and is available conjugated to dUTP and dCTP (Amersham). Cy5 is currently the sole option for infrared emission, and is also available as a nucleotide conjugate (Amersham). At this time, there is no blue-emitting fluorophore that works reliably for FISH probes. Accordingly, this wavelength range is conveniently reserved for a counterstain such as DAPI or Hoechst. With laser illumination, PI is typically used as a counterstain. Digoxigenin-labelled probes may be detected using fluorescein- and rhodamine-conjugated anti-digoxigenin Fab fragments (Boehringer Mannheim), and I have found that commercially available antibodies made to digoxin (Jackson ImmunoResearch), although not marketed for FISH, can also detect digoxigenin-labelled probes and yield excellent signal/background ratios. The 129

Abby F. Dernburg latter are available conjugated with Cy2, Cy3, and Cy5, which may provide superior performance in terms of brightness, photostability, and versatility. Biotin can be detected using fluorescent avidin, modified avidins such as UltraAvidin (Leinco Technologies) or NeutraLite Avidin (Molecular Probes), streptavidin, or anti-biotin antibodies. Different secondary reagents may work better in different tissues. In my experience, the detection of digoxigenin-labelled probes provides somewhat lower levels of background than the detection of biotinylated probes with avidin, streptavidin, or anti-biotin antibodies in whole-mount tissues. This is primarily true of samples fixed with moderate to high concentrations of formaldehyde (3.7%). Table 1 lists some of the many choices available for labelling and detecting probes in the four most useful wavelength ranges. This is not meant to be a comprehensive list, but to point to some basic reagents and commercial suppliers. It is impossible to provide a recommendation of labelling and detection reagents that are optimal for every situation, since success depends on many parameters, including the labelling method used, the autofluorescence spectra of different samples, and the microscope illumination sources and fluorescence filters. Moreover, new reagents are constantly becoming available.

Table 1. A useful subset of available labelling and detection reagents Emission Infrared

Blue*

Green

Direct label

Cascade BluedUTPM

FITC-dUTPA,B, M, N Cy3-dCTP/dUTP*

Cy5-dCTP/dUTP"

Biotinylated probe B'G,N

AMCA-anti-biotinJ AMCA-streptavidinJ AMCA-NeutraLite avidinm

Cy3-anti-biotinJ Cy2-anti-biotinJ Cy2-streptavidin J,A Cy3-streptavidin-M FITC-UltraAvidin^

Cy5-Anti-biotinJ Cy5-StreptavidinJ, A

DigoxigeninAMCA-antilabelled probe8 digoxigeninB AMCA-anti-digoxinJ Counterstain

DAPIS Hoechst 33258s

Red

FITC-antidigoxigeninB Cy2-anti-digoxinJ

Rhod-antidigoxigeninB Cy3-anti-digoxinJ

Oli-GreenM YOYO-1M

Propidium iodides

Cy5-Anti-digoxi n J

*UV-excitable, blue-emitting dyes are less easily detected than others and are thus recommended only for detecting very abundant sequences in experiments involving multiple probes. ''Amersham; BBoehringer Mannheim; GGibco-BRU/Life Technologies; JJackson ImmunoResearch; LLeinco Technologies; MMolecular Probes; NNew England Nuclear; sSigma. Currently, there is not a well-characterized green fluorescent DNA counterstain that is as generally useful as DAPI, Hoechst, or PI. Oli-Green is marketed as a single-stranded nucleic-acid detection reagent, and has been useful as a green DNA counterstain in Drosophila tissues. It should be tested at a dilution of 1:1000-1:10000 from the solution provided. YOYO-1 is marketed as a green DNA counterstain, but may bind to other cellular components in whole-mount tissues.

130

7: Fluorescence in situ hybridization in whole-mount tissues

2.4 Probe synthesis Good probes for whole-mount FISH must satisfy two major requirements: the probe fragments must be very small, and they must be labelled to a high level. To control both of these parameters consistently, the following scheme has been developed. Probe DNA is first digested to reduce the average fragment size to approximately 100 bases. A mixture of labelled and unlabelled nucleotides are then added to the 3' ends of the fragments by TdT, a templateindependent DNA polymerase. This method can be used to label a variety of DNA templates, regardless of initial size (e.g. plasmid, cosmid, or P1 clones, PCR products, or total genomic DNA). Synthetic oligonucleotides may also be labelled this way, but require no digestion since their small size allows them to penetrate thick tissues readily. The presence of 1 mM Co2+ in the TdT reaction (Protocol 1) allows the tailing of any type of 3'-end. An additional advantage of labelling in this way is that all the label is incorporated into a tail that does not anneal to the target, and thus the label may interfere less with hybridization than it can when incorporated into the complementary sequence. The free 3'-tail may also make haptens more accessible to detection reagents. However, TdT may show variability in its ability to incorporate different modified nucleotides. Thus, this method is presented as a straightforward and reliable option, but others may be worth considering. Protocol 1. Digestion and labelling of probe DNA Caution: cacodylate is toxic Equipment and reagents • 5 x 4BC buffer: 50 mM Tris-HCI pH 7.5, 250 • 0.4 M Na cacodylate, pH 7.2 with HCI M NaCI, 40 mM MgCI2 . Labelled and unlabelled dNTP . 5% w/v BSA . coCI2 . 100 mM DTT . TdT (prOmega or other supplier) . Restriction enzymes: Alul, Haelll, Msel, Mspl, ,0 ma/ml glycogen Rsal, and Sau3AI (New England Biolabs) •NH, OAc . 3 M NaOAc pH 5.2 2% . Cold 70%, 75%, and absolute EtOH • agrose gel. electrophoresis equip.

buffer: 10 mM Tris-HCI, pH 7.5, 0, mM

A. Restriction enzyme fragmentation of probe DNA 1. In a 1.5 ml polypropylene tube add 50 ml 5 x 4BC buffer, 2.5 ml 5% BSA, and 2.5 ml 100 mM DTT to 25 mg of probe DNA and add water to approximately 225ml(depending on the volume of enzymes required). 2. Add 25 U of each restriction enzyme (total volume should be 250 ml), mix, and incubate at 37°C for 2-4 h or overnight.3 3. Precipitate the digested products by adding 1/10 volume 3 M NaOAc, then 2.5 volumes cold absolute ethanol. Chill the tube and centrifuge for 131

Abby F. Dernburg Protocol 1.

Continued

15 min at maximum speed to pellet the DNA. Wash the pellet with cold 70% EtOH, allow to dry briefly, then resuspend in TE buffer to 1 4. Analyse 0.5 mg of the products by electrophoresis through a 2% agarose gel. A smear of fragments should be detected with an average size of 100-150 bp. If the concentration looks markedly different than expected (by comparison with a known quantity of DNA molecular weight markers) measure the concentration accurately by determining the absorption at 260 nm (1 OD260 = 50 m9/ml). B. Labelling with terminal deoxynucleotidyl transferase This reaction is designed to label 10 mg of probe, which is generally sufficient for 20-500 hybridizations, depending on the length and complexity of the target sequence. It can easily be scaled down and carried out in a smaller volume. 1. In a 1.5 ml microcentrifuge tubeb on ice add 50 ml 0.4 M Na cacodylate to 10 mg DNA fragments that have been denatured by heating for 2 min in a 95°C water-bath and then rapidly chilled. 2. Add 13.5 nmoles unlabelled dNTP (whichever the labelled nucleotide analogue is derived from, i.e. use dTTP if incorporating a modified dUTP) and 6.75 nmoles labelled dNTP (usually 6.75 ml of a commercially supplied 1 mM stock solution). Add DTT to 200 mM and CoCI2 to 1 mM. Adjust the volume to 97 ml with water. 3. Add 50 U TdT and incubate for 1 h at 37 °C. 4. Add 2 ml 20 mg/ml glycogen as a carrierc and precipitate the probe by adding NH4 OAc to 2 M (NaOAc is not used here to minimize the coprecipitation of unincorporated nucleotides) followed by 2.5 volumes of absolute ethanol. Chill and pellet by spinning for 15 min at top speed in a microcentrifuge (fluors labelled with fluorescent nucleotides should give a visibly coloured pellet, although it may be difficult to see this through an amber tube). Remove the ethanol, wash the pellet with cold 75% EtOH, and dry briefly. 5. Resuspend in TE bufferd Store at -20°C. a As an alternative to restriction enzyme digestion, DNA may be fragmented with DNase in a buffer containing Mn2+ (not Mg2+) to generate double-strand breaks. This allows fragment sizes to be analysed by gel electrophoresis. This option is more economical but less convenient, since the DNase concentration and digestion time must be carefully controlled to produce an appropriate distribution of fragment sizes. This calibration must be performed for each new template. The reaction may be stopped by adding EDTA, but the products should also be immediately phenol-extracted to kill residual enzyme activity. b'Amber tubes are useful for labelling and storing fluorescent probes. c(optional) Spin the probe through a 1 ml G25-Sephadex spin column (or a commercial version) to remove unincorporated nucleotides (Chapters 3 and 4). This may help to reduce background staining (reports vary) but it will also result in the loss of some precious probe. d lt is convenient to store probes at a concentration of 50-200 ng/ml, such that a small volume is used per sample in the hybridization procedure.

132

7: Fluorescence in situ hybridization in whole-mount tissues

3. Fixation methods for whole-mount FISH Identifying a good fixation method is crucial for achieving efficient in-situ hybridization while preserving sample morphology (Chapter 6). This is the most variable, tissue-dependent aspect of the entire procedure. The tissue needs to be sufficiently permeable to allow the probe molecules access to chromatin, but also adequately fixed to withstand the procedure. The doublestranded chromosomal DNA also need to be denatured, or melted, so that the probe can anneal to its target. To accomplish this without destroying the chromosome organization is the challenge. In general, fixation of tissues with moderate concentrations (1.5-4%) of formaldehyde provides the best combination of structural preservation and accessibility to hybridization. Ideally, fixation is carried out in an aqueous medium, preferably in a buffer that mimics the intracellular environment reasonably well (e.g. Ringer's solution or other balanced salt solutions). However, the signal to background level can often be increased at the expense of morphology by fixing with a mixture of acetic acid and alcohol, or alcohol alone. In special cases, formaldehyde fixation may not be compatible with other experimental goals (e.g. preservation of cytoskeletal elements) and other fixatives may be substituted. Postfixation with formaldehyde may help to preserve a sample fixed initially with alcohol or acid:alcohol. Glutaraldehyde should not be used as a fixative as it results in autofluorescence and prohibits its hybridization. With certain tissues, additional steps will be required for permeability. Such treatments may include enzymatic digestion (e.g. spheroplasting to remove the cell wall of yeasts—Chapter 3), incubation with detergent (preferably non-ionic detergents such as Triton X-100 or Tween-20), or physical dissection to remove impermeable or opaque barriers. Fixation conditions that are compatible with immunodetection should be used as a starting point. They are often also suitable for in-situ hybridization and both procedures can be performed on the same sample. Combining immunofluorescence and FISH together provides a way of examining the interaction between chromosomal subregions and other components of nuclear architecture (Chapter 6), and staining of the nuclear lamina or nuclear pore proteins provides the most accurate way to delineate the nuclear periphery (3, 4,9). In general, it is simplest to perform immunostaining steps following FISH. Surprisingly, however, antibodies added prior to the hybridization can also be detected (Chapter 6), even without postfixation steps to cross-link the antibodies to their antigens. Most antibodies show good reactivity after hybridization; a monoclonal anti-histone antibody has been the sole exception. Some soluble proteins may be partially extracted by the hybridization procedure. Specific fixation procedures for FISH to Drosophila embryos, spermatids, and egg chambers have been provided in other sources (2, 4, 5,10). Protocol 2 should serve as a starting point for other tissues and is a general approach to 133

Figure 2. Assessment of morphological preservation following whole-mount FISH. Because denaluration of chromosomal DNA necessarily subjects the target tissue to extreme conditions, the tissue must first be fixed so as to withstand this treatment. It is difficult to assess the degree of preservation following FISH to interphase nuclei by staining the DNA with fluorescent dyes, since the chromatin normally appears diffuse and structural landmarks are largely absent. Instead, condensed mitotic or meiotic chromosomes, such as these pachvtenc nuclei from C. elegans, can provide a simple visual assay. The nuclei here have been hybridized with a probe to the 5S rDNA locus (red) to monitor the pairing of homologous chromosomes. DAPI fluorescence is shown in blue. A: When a tissue is underfixed (here, fixed in methanol without formaldehyde), hybridization will cause condensed chromosomes to appear fuzzy and poorly preserved, although hybridization may result in signals that are more intense than in well-preserved specimens. B: Even well-fixed nuclei can be damaged by excessive heat treatment. C and D: However, with optimized fixation and denaturation conditions, the appearance of hybridized nuclei will differ little from samples that have simply been fixed and stained with DAPI without hybridization, as shown in E. All images are projections through 1-2 mm microns of 3-D image data, processed by deconvolution of 3-D data stacks. Scale bars 2 mm.

134

7: Fluorescence in situ hybridization in whole-mount tissues fixing tissue or cells onto microscope slides in a manner compatible with whole-mount FISH analysis. It includes a simple formaldehyde fixation in buffer, followed by postfixation in cold ethanol. Postfixation of the tissue in methanol or ethanol can markedly improve permeability. In some instances, it may be preferable to fix onto coverslips, since microscope optics are usually designed to optimize imaging immediately adjacent to the coverslip. On the other hand, coverslips are more fragile, and thus harder to handle without damage during hybridization. To evaluate the success of a particular fixation procedure, the hybridized sample should be compared to identically fixed tissue that has not undergone FISH. In particular, the appearance of condensed mitotic or meiotic chromosomes with and without hybridization can provide a useful criterion for deciding whether preservation is adequate (Figure 2). Underfixation will result in deterioration of morphology; overfixation will lead to weak or undetectable FISH signals and a high fluorescent background. With some tissues it may be a challenge to find the middle ground. The ultimate goal is to maintain the organization found in the living state. However, this requirement can be relaxed somewhat if the question being asked does not address chromosome organization; e.g. if only the presence or absence of particular sequences is being assayed. As with any cytological procedure, consideration should be given to ensuring that results do not represent fixation artefacts. When different fixation procedures converge to yield the same answer, confidence in the experimental results is enhanced. Protocol 2. Formaldehyde fixation for whole-mount FISH Equipment and reagents • Dissecting tools, which may include: stereo • Liquid N2 in a Dewar flask or a flat aluminium dissecting microscope; fine forceps (e.g. block on dry ice to freeze slides Dumont et Fils #5 purchased through . 37% formaldehyde, made fresh from Electron Microscopy Sciences or Ted Pella); paraformaldehyde (pFA) (EM grade. Polyscalpel or single-edged razor blade; dissciences): in a screw-cap 100 ml Pyrex test secting surface, e.g. a siliconized multiwell tube mix 1.85 g pFA in 3.5 ml H2O and place slide, siliconized coverslip, or a surface in a boiling water-bath. After 30 sec, add made from Sylgard elastomer (a clear sili90 ml 1 M NaOH. Cap loosely and agitate for cone rubber distributed by K. R. Anderson 1 min. The solution should become nearly Company Inc.); haemostats, the tips covclear as the pFA is hydrolysed. Cool by ered with short pieces of silicone rubber shaking briefly under a stream of tap water, tubing, for handling slides Filter through a (non-sterile) 0.22 urn syringe • Siliconized glass 18 x 18 mm coverslipsa filter into an airtight vial. Use the same day. • Glass slidesb . Dissecting buffer, typically a buffered saline • Whatman #3 filter paper solution such as Ringer's, egg salts, or PBS. . 95% ethanol, chilled to-20°C in a Coplin jar Some tissues may require osmotic protecor other slide-staining container tion by the addition of sucrose, sorbitol, . 2 x SSCT: 0.3 M NaCI, 0.03 M Na citrate, etc. to the dissecting medium. 0.1% Tween-20

Method 1. On a dissecting surface or siliconized glass coverslip, dissect or otherwise isolate the desired tissue in a suitable buffer, such as Ringer's

135

Abby F. Dernburg Protocol 2.

Continued

solution or PBS. Transfer the tissue to a siliconized coverslip in 20-30 ml of the dissecting buffer, and physically remove large contaminants. 2. Add an equal volume of 2 X pFA in dissecting buffer (typically 3-8% pFA, made by diluting 37% pFA in dissecting buffer). 3. Remove all but 10-20 ml of the fixative by careful pipetting. 4. Touch the centre of the glass slide to the solution and invert carefully to sandwich the tissue between the slide and coverslip.b 5. Wick away excess buffer with pieces of absorbent paper, e.g. Whatman #3 filter paper. The goal is to press the tissue gently against the slide, but not to flatten it. 6. Immerse the slide and coverslip into liquid N2, or place on a flat aluminium block on dry ice, to freeze the sample. Once frozen, crack the coverslip using a fresh single-edged razor blade, and quickly transfer the slide to a staining jar containing 95% EtOH at-20°C. Store in EtOH at-20°C if hybridization will not be performed immediately. 7. When sufficient samples have been prepared, allow the ethanol to rise to above 0°C (i.e. frost on the outside of the container melts). Transfer the samples to a container filled with 2 x SSCT. Proceed with hybridization (Protocols 3 and 4). a To prepare siliconized glass slides or coverslips, immerse briefly in a 1-2% solution of Surfasil (Pierce) in chloroform, rinse with chloroform, and allow to air-dry. 'With many tissues it is helpful to first treat slides with agents that improve adhesion. These include polylysine solutions, aminoalkylsilane, and gelatin 'subbing' solutions. Commercially available treated slides such as Superfrost Plus (Fisher) may also be useful.

4. Hybridization methods The key to success in these experiments is to synthesize probes of high quality and to optimize the fixation and permeabilization for the particular tissue. Once these are achieved, hybridization itself is remarkably straightforward. The general strategy is to equilibrate the tissue in buffered formamide, to add the probe(s), and to denature both the chromosomal DNA and probe together by heat treatment. The probe is then allowed to anneal for several hours at an appropriate temperature, unbound probe is washed away, and secondary detection (if required) is performed. The sample is then counterstained and mounted for microscopy. For experiments in which immunolocalization of other cellular components is desired, these staining steps can conveniently be performed after hybridization. The major decision that must be made is whether the tissue can be easily handled in small polypropylene tubes, in which solutions are changed by 136

7; Fluorescence in situ hybridization in whole-mount tissues allowing the tissue to settle to the bottom of the tube, aspirating the solution carefully, and replacing with 400-500 ml of the new solution. If so, this is the most convenient way to perform the hybridization and requires a minimum volume for prehybridization and washing steps. The denaturation step is also easy to perform, since the sample can be heated and annealed in a thermal cycler or conventional water-bath (Protocol 3). Protocol 3.

Hybridization to tissue in small polypropylene tubes

Equipment and reagents • Mounting medium. Typically a buffered 90% • Fixed tissue (Section 3) in standard or thinglycerol solution containing an antifade walled 0.5 ml polypropylene tubes such as DABCO (diazabicyclooctane), PPD . Labelled probe(s) (Section 2 and Protocol 1) (p-phenylenediamine), or NPG (n-propyl . Fluorescent detection reagents, such as flugallate). An inexpensive all-purpose medium orescently labelled streptavidin, anti-biotin, can be made by dissolving 4% NPG (w/v) ,n or anti-digoxiger antibodies (Section 2 high-quality glycerol; this requires agitation and chapter 4) over several hours, but the solution may be Thermal stored indefinitely at room temperature. The cycler°r.othermeans of heating pH should be adjusted by adding 70ml 2 M tubes to 91 °C and incubating at 37 °C Aspirator with fine tip and liquid trap Tris base (no HCI) and 30 ml water to 900 ml 20 X SSC: 3 M NaCI, 0.3 M Na citrate glycerol/NPG, followed by thorough mixing. This pH-adjusted mountant should be used Tween-20 within a day or two. Note that VectaShield Formamide (Fluka, cat. no. 47670) and other mounting solutions containing PPD should not be used with the cyanine Dextran 2 x SSCT (Protocol 2) dyes, particularly Cy2, as the antifade will 2 x SSCT containing 20%, 40%, and 50% chemically degrade these fluorophores over time (William Stegeman, Jackson Immunoformamide Research, personal communication). The Cy . (optional) 10 mg/ml boiled RNase A in 2 x dyes are quite photostable in the absence of SSCT antifading agents.

Method 1. Wash the tissue 3 x with 2 x SSCT. Incubate for at least 10 min per wash. 2. (optional) Treat the sample with 10 mg/ml boiled RNase A in 2 x SSCT for 30 min. This step is usually dispensable, but probes to highly tran. scribed sequences, e.g. rDNA, may also pick up RNA, so this step may serve as a useful control. This treatment is also useful if the nuclear DNA will be counterstained with PI or another dye that binds to RNA (step 15). 3. Step the tissue gradually into 2 x SSCT containing 50% formamide, by adding sequentially and incubating for 10 min each in: (a) 2 x SSCT containing 20% formamide; (b) 2 x SSCT containing 40% formamide; (c) 2 x SSCT containing 50% formamide. 4. Add fresh 2 X SSCT/50% formamide, place samples at 37 °C, and incubate for at least 30 min before adding probe.3

137

Abby F. Demburg Protocol 3.

Continued

5. Remove as much of the prehybridization solution as possible without aspirating the tissue. Add the probe solution: 25-500 ng each probeb in 3 x SSC, 50% formamide, 10% dextran sulfate.c Mix gently by flicking the tube. 6. Denature the probe and chromosomal DNA by heating the sample to 91°Cd for 2 min. 7. Reduce the sample temperature to 37°Ce and allow the probe to anneal for several hours, typically overnight. 8. Add 500 ml of 2 x SSCT/ 50% formamide to the sample; mix, and allow to settle. 9. Wash the sample 3 x with 2 x SSCT/50% formamide at the annealing temperature over at least 1 h in total. 10. Step the sample back out of formamide through: (a) 2 x SSCT containing 40% formamide; (b) 2 x SSCT containing 20% formamide. 11. Wash with 3 changes of 2 x SSCT without formamide. 12. If secondary detection or other immunostaining will be carried out, include the following steps. For avidin and many antibodies (including anti-digoxigenin Fab fragments from Boehringer Mannheim), staining may be carried out in 2 x SSCT, but if another buffer (e.g. PBS, TBS) is preferred, it is simple to switch at this point. Prior to staining with antibodies and/or avidin, the sample should be blocked with a 0.5% (w/v) protein solution; this may be BSA, normal serum proteins, etc. Incubate the sample in blocking solution for 30 min, then add diluted antibody and/or avidin (Chapter 4). 13. Incubate the sample with the antibody/avidin solution for 2 h, protected from light, with agitation on a nutator or similar device. 14. Wash the tissue with 3 changes of 2 X SSCT for at least 10 min each wash to remove unbound detection reagents. 15. Counterstain DNA with DAPI or another fluorescent counterstain in the second wash. For nucleic acid dyes that also bind to RNA (e.g. PI), an RNase step prior to counterstaining is crucial (step 2). 16. Store samples in buffer (made to 1 mM EDTA) at 4°C until shortly before imaging, as samples tend to deteriorate over time in glycerol. 17. Mount the samples by exchanging the tissue into mounting medium and then transferring to a slide or coverslip; or alternatively, by pipetting the tissue in buffer onto a slide or coverslip (polylysine-treated glass may facilitate this), gently removing most of the buffer, and overlaying with mounting medium. A minimal volume of mounting 138

7: Fluorescence in situ hybridization in whole-mount tissues medium should be used, so that the samples are as close to the coverslips as possible and are immobilized by sandwiching between the two layers of glass. 'This prehybridization incubation can have a marked impact on the permeability of some tissues, and should be increased to as long as several hours if inconsistent probe penetration or high background is observed. 'The probe concentration must be optimized empirically, but will usually be 0.5-10 ng/ml. Beyond an optimal level, increasing the concentration of double-stranded probe molecules may preferentially enhance their ability to anneal to themselves, since these kinetics follow the square of probe concentration while hybridization to the chromosomes should increase only linearly with probe concentration. c lt is convenient to make up a '90% hybridization' solution containing 1.5 ml 20 x SSC, 1 g dextran sulfate, 5 ml formamide, made to 9 ml total volume with water. This solution is stable for months at 4°C. To 36 ul of this solution 4 ml of probe(s) plus water is added and mixed, and then added to the sample. dThis temperature has been optimized by examining signal/noise and morphological preservation in formaldehyde-fixed samples. With other fixatives, 75-80°C may be sufficient. •This stringency will work with most probes, but for very AT-rich targets, if hybridization fails, try lowering the annealing temperature to 30-37°C.

If the tissue is not amenable to the type of handling described in Protocol 3 (e.g. it is too large, delicate, transparent, or buoyant), it can be fixed directly on a microscope slide or coverslip, which is then carried through the procedure in Coplin jars, larger staining jars, or coverslip staining jars (Thomas Scientific) (Protocol 4). Coplin jars require 50-60 ml of solution to cover the sample, larger staining jars require 200 ml, and coverslip staining jars hold about 7-10 ml. Staining steps after hybridization are performed by pipetting the solution directly onto the tissue, covering with a glass or Parafilm coverslip, and incubating in a humid chamber at the appropriate temperature. Protocol 4. Hybridization to tissue on slides or coverslips Equipment and reagents • Fixed tissue (as described in Section 3) on slides or coverslips • Haemostats, the tips covered with short pieces of silicone rubber tubing, for handling slides • Labelled probe(s) as described in Section 2 and Protocol 1 (see also footnote b, Protocol 3) • Probe solution (see footnote c. Protocol 3) • Fluorescent detection reagents, such as fluorescently labelled streptavidin, anti-biotin, or anti-digoxigenin antibodies (Section 2) • Humidified denaturation chamber. A simple home-made version may be constructed by setting an aluminium heat block to 94°C and covering this with a plastic lid containing a layer of Kimwipes wetted with 2 x SSC around the inner rim {Figure 3).

• Humidified incubation chamber, constructed of a watertight plastic container with a lid with a few layers of paper towels moistened with 2 x SSC and overlaid with a sheet of Parafilrn • Aspirator with fine tip and liquid trap . 2 x SSCT (Protocol 2) . 20 X SSC (Protocol 3) • Tween-20 • Formamide (Fluka, cat. no. 47670) • Dextran sulfate • (optional) 10 mg/ml boiled RNase A in 2 x SSCT • 2 x SSCT containing 25%, 50% formamide • Mounting medium (Protocol 3)

139

Abby F. Dernburg Protocol 4.

Continued

Method 1. Wash the tissue with 2 changes of 2 x SSCT. Incubate for at least 10 min per wash. 2. (optional) Treat the sample with RNase A (step 2, Protocol 3). RNase may be added to the buffer in a Coplin jar, or smaller volumes can be used by transferring the slide(s) to a humid chamber, pipetting the RNase solution directly onto the sample, and covering with a Parafilm coverslip. 3. Step the tissue gradually into 2 x SSCT containing 50% formamide, by transferring sequentially into and incubating for 10 min each in: (a) 2 x SSCT containing 25% formamide; (b) 2 x SSCT containing 50% formamide. 4. Transfer the samples to fresh 2 x SSCT/50% formamide, place the staining jar in a 37 °C water-bath, allow the temperature to equilibrate, and incubate for at least 30 min before probe addition. (See footnote d, Protocol 3.) 5. Carefully remove the slides/coverslips from the prehybridization solution and drain on paper towels. (Save the prehybridization solution for the first posthybridization wash.) Wearing gloves, use the tip of an aspirator and/or Kimwipes to remove as much of the prehybridization solution as possible without damaging the tissue. Wipe the back of the glass dry with Kimwipes. 6. Add the probe solution: 10-200 ng each probe (see footnote b, Protocol 3) in 12 ml 3 x SSC, 50% formamide, 10% dextran sulfate (see footnote c, Protocol 3). Pipette this solution onto a clean 22 x 22 mm coverslip or, if the sample is on a coverslip, pipette it onto a slide. Touch the drop of solution to the sample and invert the slide. 7. Denature probe and chromosomal DNA by placing the slide onto a hot block pre-equilibrated to 94°C. Cover with a humidifying lid (Figure 3), and incubate for 2 min.a 8. Transfer the slides to a humidified chamber at 37°C and incubate for several hours to overnight. 9. Transfer the samples to a Coplin or staining jar containing 2 x SSCT/50% formamide at the annealing temperature.b 10. Wash the sample with one further change of 2 x SSCT/50% formamide at the annealing temperature over at least 1 h in total. 11. Transfer the sample into 2 x SSCT/25% formamide at room temperature and incubate for at least 10 min. 12. Wash with 3 changes of 2 x SSCT without formamide. 140

7: Fluorescence in situ hybridization in whole-mount tissues 13. If secondary detection or other immunostaining will be carried out, include the following steps. Perform staining steps by transferring the slides into a humid chamber, pipette the blocking or staining solution onto the sample, and overlay with a piece of Parafilm cut to approximately 25 x 25 mm (or as needed to cover the specimen). For avidin and many antibodies (including anti-digoxigenin and anti-digoxin antibodies), carry out the staining in 2 x SSCT; if another buffer (e.g. PBS, TBS) is preferred, it is simple to switch at this point. Prior to staining with antibodies and/or avidin, block the sample with a 0.5% (w/v) protein solution (BSA, normal serum proteins, etc.). Incubate the sample in blocking solution for 30 min, then add diluted antibody and/or avidin: 14. Incubate the sample with antibody/avidin solution for 2 h protected from light. 15. Wash the tissue with at least 3 changes of 2 x SSCT for at least 10 min each wash to remove unbound detection reagents. 16. Counterstain with DAPI or another fluorescent dye in the second wash. 17. Mount the samples for microscopy (Protocol 3). To mount samples adhered to microscope slides, pipette 12-15 ml mounting medium onto a clean 22 x 22 mm coverslip. Transfer the slide briefly to a staining jar containing 50 mM Tris-HCI pH 7.5 (to remove most of the salt). Remove the slide and allow to drain briefly on paper towels. Use Kimwipes and/or an aspirator to remove as much buffer as possible from the slide without damaging the sample or allowing it to dry out. Touch the sample to the drop of mountant, invert the slide, and seal with clear nail polish. a As an alternative to constructing a humidified denaturation chamber, seal the coverslip to the slide with a gasket of rubber cement, most easily applied using a syringe with a wide-bored needle. Allow the rubber cement to dry, and then denature the slide directly on a heat block at 94°C. bIf the samples are on slides, allow the coverslips to fall off and discard them. If rubber cement was used, this carefully peel this off using forceps while the slide is immersed in washing solution. If the samples are on coverslips, retrieve them carefully using flat-tipped forceps (Millipore) and transfer to coverslip staining jars containing 2 x SSCT/50% formamide, at the annealing temperature.

5. Troubleshooting Whole-mount FISH experiments can be plagued by three general problems: no detection of probe above background; non-specific binding of probe or detection reagents; and unacceptable morphological preservation. Efforts should be made to isolate the source of the problem to either the probe or the tissue preparation. 141

Abby F. Dernburg

Figure 3. Denaturation of FISH samples on slides using a humidified aluminium block. This diagram illustrates a simple way to heat-denature samples on glass slides using a conventional aluminium heat block. The heat block is set and equilibrated at 94°C using a thermometer placed in one of the wells, and the aluminium block is then inverted so that its flat side faces up. A humidifying chamber can be assembled using a shallow plastic box; the lids from racks of P2 Pipetman tips are perfectly sized for this. Kimwipes are placed around the interior rim of the lid and moistened with 2 x SSC or water (salt will damage anodized aluminium surfaces). The lid is placed over the slides during the 2-min denaturation period.

Preparations of metaphase chromosome squashes or spreads from the organism being investigated are an ideal way to test probes for specificity and successful labelling. Interphase nuclei may be prepared from many cell types by squashing the tissue with thumb pressure under a siliconized coverslip in a fresh 3:1 mixture of ethanol:acetic acid. After squashing, the slide should be fro/en on an aluminium block on dry ice, the coverslip cracked, and the slide transferred to an ethanol bath and then air-dried. Although such preparations will destroy both the three-dimensional chromosome organization and the structural context, the chromosomes are more accessible to probes. To hybridize to acid-fixed and flattened specimens, incubate the slides in 2 x SSC at 70aC for 30 min, and dehydrate through an ethanol series (70%, 70%, 95%, 95% ethanol in water, 5 min each), then air-dry. Immerse the slides in 0,07 M NaOH for 3 min, then again dehydrate through the same ethanol series and allow to dry. Dilute the probe into the same hybridization solution used in Protocols 3 and 4 (3 X SSC, 50% formamide, 10% dextran sulfate) and heat in a 95 °C water-bath for 2 min to denature. Chill the denatured probe on 142

7: Fluorescence in situ hybridization in whole-mount tissues ice briefly, then pipette onto a coverslip, and pick up by touching the squash preparation on the slide to the drop of probe solution. Seal the edges of the coverslip with rubber cement. Anneal, wash, and stain the slides as in Protocol 4, starting from step #8. Antibody incubation times may be reduced to 1 h. For complete procedures, see refs 11 or 12. If spread preparations are not easily obtained, the next-best option is to fix the target tissue without squashing in a 3:1 mixture of ethanohacetic acid. Hybridization can be performed using the procedures described in Section 4, using a denaturation temperature of 80 °C. If the probe cannot be seen easily in flat or acid-fixed chromosome preparations, it is pointless to try to detect it in a formaldehyde-fixed, whole-mount sample. The probe synthesis must first be improved. To ensure that the labelling procedure is working, it is a good idea to make a probe to a very abundant sequence in the genome, ideally a simple-sequence repeat that can be targeted using a synthetic oligonucleotide or PCR amplicon. More often than not, disappointing or inconsistent results are due to overor under-fixation of the tissue. If the probe can be readily detected on flattened or acid-fixed preparations, then attention should be paid to improving the whole-mount fixation, using the morphological criteria outlined in Section 3 and Figure 2. Modifications that have proven useful with some tissues include: (a) warming the fixative solution to 30-37 °C before adding to the sample; (b) performing fixation at low temperature, typically 0-4 °C for 1 h or more; and (c) reducing the formaldehyde concentration to 1-2%. However, all these modifications may have deleterious effects on the chromosome preservation.

6. Microscopy and image analysis Analysis of FISH signals in whole-mount tissues requires the application of suitable fluorescence microscopy. The requirements on the microscope system will be imposed by the size of the nuclei being imaged, the size of the target sequence(s) and resulting intensity of the fluorescence signals, and by the nature of the question being addressed. Diploid nuclei in model organisms are typically about 2-20 mm in diameter, only 1-2 orders of magnitude greater than the resolution limit of most optical microscopes. To achieve high optical resolution it is helpful to use techniques that remove out-of-focus information from fluorescence images. Two popular methods are CLSM and wide-field deconvolution microscopy, in which the out-of-focus light is collected but later restored to its proper position within the object using computational algorithms (13). Each has advantages (Chapter 6). The majority of my work has employed wide-field deconvolution microscopy, which satisfies a number of requirements for the analysis of 3-D FISH samples. Mercury arc lamp illumination is used, so that fluors can be excited 143

AbbyF. Dernburg throughout the visible spectrum. The use of a charge-coupled device (CCD) camera gives excellent sensitivity, permits the detection of probes that emit in the infrared, and also facilitates quantitative analysis of probe intensities and volumes. For some whole-mount FISH applications, such as karyotyping, conventional fluorescence microscopy will be perfectly adequate.

7. Future directions Cytological investigation of chromosome structure is an area of active research and development, and new reagents are constantly being made available that may extend the utility of the general approach described in this chapter. For example, the development of probes utilizing new chemistry, such as peptide nucleic acid (PNA) oligos, can substantially increase both the intensity and the reliability of hybridization, since PNA-DNA hybrids are stable under low salt conditions where DNA fails to anneal to itself (14). New fluorophores, particularly the cyanine dyes (Cy3, etc.) have already reduced the size of the target that can be detected, and further innovations in fluorophore chemistry are being explored. Tissue fixation is something of a 'black art', in which the chemical effects of different fixatives on proteins and nucleic acids are poorly characterized, and advances in this area may also eventually improve FISH technology. The methods presented here should thus be treated as a starting point for further exploration.

Acknowledgement This work was supported in part by a postdoctoral fellowship (DRG: 1392) from the Cancer Research Fund of the Damon Runyon-Walter Winchell Foundation.

References 1. Hiraoka, Y., Dernburg, A.F., Parmelee, S.J., Rykowski, M.C., Agard, D.A., and Sedat, J.W. (1993). J. Cell Biol., 120, 591. 2. Dernburg, A.F., Sedat, J.W., and Hawley, R.S. (1996). Cell, 86,135. 3. Marshall, W.F., Dernburg, A.F., Harmon, B., Agard, D.A., and Sedat, J.W. (1996). Mol. Biol. Cell, 7, 825. 4. Dernburg, A.F., Broman, K.W., Fung, J.C., Marshall, W.F., Philips, J., Agard, D.A., and Sedat, J.W. (1996). Cell, 85,745. 5. Dernburg, A.F., Daily, D.R., Yook, K.J., Corbin, J.A., Sedat, J.W., and Sullivan, W. (1996). Genetics, 143,1629. 6. Brelje, T.C., Wessendorf, M.W., and Sorenson, R.L. (1993). In Methods in cell biology, Vol. 38 (ed. B. Matsumoto), p. 97. Academic Press, San Diego, CA. 7. Wessendorf, M.W. and Brelje, T.C. (1992). Histochemistry, 98, 81. 144

7: Fluorescence in situ hybridization in whole-mount tissues 8. Wiegant, J., Verwoerd, N., Mascheretti, S., Bolk, M., Tanke, H.J., and Raap, A.K. (1996). J. Histochem. Cytochem., 44, 525. 9. Gotta, M., Laroche, T., Formenton, A., Maillet, L., Scherthan, H., and Gasser, S.M. (1996). J. Cell Biol., 134, 1349. 10. Dernburg, A.F. and Sedat, J.W. (1998). In Methods in cell biology, Vol. 53 (ed. M. Berrios), p. 187. Academic Press, San Diego, CA. 11. Ashburner, M. (1989). Drosophila: a laboratory manual. Cold Spring Harbor Laboratory Press, New York. 12. Pardue, M.-L. (1986). In Drosophila: a practical approach (ed. D.B. Roberts), p. 111. IRL Press Limited. Oxford. 13. Swedlow, J.R., Sedat, J.W., and Agard, D.A. (1997). In Deconvolution of images and spectra (ed. P.A. Jansson), p. 284. Academic Press, San Diego, CA. 14. Martens, U.M., Zijlmans, J.M., Poon, S.S., Dragowska, W., Yui, J., Chavez, E.A., Ward, R.K., and Lansdorp, P.M. (1998). Nat. Genet., 18,76.

145

This page intentionally left blank

8

Analysing the substructure of mammalian nuclei, in vitro DEAN A. JACKSON

1. Introduction If we are to understand the importance and complexities of interactions within the nucleus, as they exist in vivo, it is vital to develop approaches that allow us to access nuclear contents in vitro in such a way that the multiple layers of their organization can be analysed. This can only be worthwhile, however, if every effort is made to maintain and protect the organization as it exists inside the cell. Chapters 6 and 7 have shown how these goals can be pursued whilst using FISH and immunofluorescence to study chromosome structure in whole cells and tissues. There are particular additional problems in studying complex underlying nuclear structures using classical biochemical approaches. This chapter details the principal methods through which attempts have been made to study nuclear substructures that are important for both nuclear structure and function.

2. The nuclear matrix and nucleoskeleton In the 1950s, Mirsky and Ris (1) showed that chromosomes retain their shape even if the major chromatin proteins are removed, and so the existence of a proteinaceous chromosome core was proposed. About a decade later, Georgiev and Chentsov (2) used electron microscopy (EM) to visualize a potential interphase counterpart of this chromosome core in nuclei digested with deoxyribonuclease (DNase) and extracted with 2 M NaCl. In 1974, Berezney and Coffey (3) coined the term 'nuclear matrix' to describe the remnants of nuclei extracted with 2 M NaCl, Triton, and nucleases, in turn. Nuclear matrices, prepared under defined experimental conditions, are presumed to derive from a 'nucleoskeleton' found in vivo, although definitive evidence in support of this is lacking. The nucleoskeleton is thought to provide a framework upon which vital nuclear functions are performed. If this is so, then understanding the role of this structure could provide important information about the control of

Dean A. Jackson different nuclear functions. Early experiments demonstrated that sites of replication and transcription remained 'structured' inside mammalian nuclei when most nuclear proteins and much of the DNA had been removed (4, 5). On the basis of these experiments, it was argued that a nuclear substructure played an important role in coordinating these activities. In later studies, roles in RNA processing, DNA repair, and chromatin modifications such as histone acetylation, have also been proposed (6). The nuclear matrix isolated from proliferating mammalian cells retains 10-20% of nuclear protein and has a protein composition that reflects these diverse roles (6,7). Despite strong evidence that the matrix influences different nuclear functions in vitro, two principal arguments suggested that the nucleoskeleton might be of minor importance in vivo. First, it is clear that the major nuclear functions can be reconstituted, from soluble components, in reactions that appear to perform all steps in the apparent absence of any higher levels of organization. This implies that any order seen in matrices is a consequence of the experimental procedures used, probably reflecting the high local concentrations of components in vivo. Second, while a nuclear network similar to the cytoplasmic intermediate filament network has been described, the molecular details of this structure have proved elusive, implying that it might represent some experimental artefact. Here, I describe different approaches that have been used to study the structure and function of the nuclear matrix, scaffold, and skeleton, and the advantages and problems associated with particular methods.

3. Methods used to analyse nuclear organization 3.1 The nuclear matrix The original protocol for preparing nuclear matrices (3) involved purifying rat liver nuclei followed by sequential extraction with buffers containing 2 M NaCl, 1% Triton X-100, and 200 mg/ml DNase and RNase. For most purposes, it is more convenient to prepare nuclear matrices from cells grown in culture, following Protocol 1. Protocol 1. Preparing the nuclear matrix by high-salt extraction Equipment and reagents • Dounce homogenizer or 22-gauge syringe needle . PBS: 0.01 M Na/KPO4, 0.137 M NaCl, 0.0027 M KCI pH 7.4 (Sigma P4417)

• Hypotonic TM buffer: 10 mM Tris-HCI pH 7.4, 2 mM MgCI2, 0.5 mM phenylmethlysulfonyl fluoride (PMSF) • DNase I

. Triton X-100 .

• 4 M NaCI • TM (0.2) buffer: as TM buffer but with 0.2 mM MgCI2 in place of 2 mM MgCI2.

MqCl2

Method 1. Prepare nuclei using a standard procedure. For example, collect cells grown in culture, wash 2 x in PBS and resuspend in hypotonic TM

148

8: Analysing the substructure of mammalian nuclei, in vitro buffer at ~107 cells/ml. Incubate the cells for 1 min at room temperature then for 10 min on ice. The cytoplasmic volume will increase by >3-fold though the nuclear volume is almost unaffected. 2. Add Triton X-100 to 0.5% and break the cells using either 10-20 strokes of a tight-fitting Dounce homogenizer or 3-10 passes through a 22-gauge syringe needle; nuclei prepared using a syringe generally have less cytoplasmic material (8). Use a light microscope to inspect the sample at appropriate intervals and stop when the cells are 95-100% broken, but before nuclear/chromatin fragments appear. 3. Collect nuclei by centrifuging at 300 g for 5 min at 4°C, wash the pellet 2 x in TM buffer (mix without pipetting at this stage to avoid damage that will cause the nuclei to aggregate) and resuspend at ~5 x 107 nuclei/ml. 4. Add MgCI2 to a final concentration of 2.5 mM and DNase I at ~30 U/ml, incubate for 30 min at 4°C. 5. Mix gently, while adding an equal volume of 4 M NaCI, then add TM (0.2) buffer to give ~107 nuclei/ml. Collect nuclear matrices by centrifuging at 750 g for 15 min at 4°C and resuspend the pellet in TM 0.2.

3.2 Nucleoids For some purposes it is necessary to use nuclear derivatives that retain all nuclear DNA. This can be achieved using Protocol 1 but omitting step 4. However, such preparations tend to be unstable and will usually aggregate on manipulation. If cells are lysed directly, as described in Protocol 2, the resulting 'nucleoids' are stabilized by cytoskeletal elements that collapse onto the nuclear lamina during lysis (9). Nucleoids have been used to investigate the organization of sites of DNA replication, transcription, and repair (4).

Protocol 2.

Nucleoid preparations

Equipment and reagents • PBS (Protocol 1) . 30% (w/v) sucrose in 10 mM Tris-HCI pH . Lysis mix: 2.7 mM Tris-HCI pH 8.0, 133 mM 8.0,1.95 M NaCI, 1 mM EDTA EDTA, 2.6 M NaCI, 0.67% Triton X-100 » 10 mM Tris pH 8.0 . 15% (w/v) sucrose in 10 mM Tris-HCI pH • Beckman SW 28 rotor (or similar) 8.0, 1.95 M NaCI, 1 mM EDTA • Siliconized, wide-bore Pasteur pipettes

Method 1. Harvest 108 tissue culture cells, wash once in PBS, and resuspend in 3 ml PBS. Lyse the cells by adding 9 ml of the lysis mix. 149

Dean A. Jackson Protocol 2.

Continued

2. Layer the mixture directly onto two 25 ml step gradients of 15% (w/v) sucrose floating on 5 ml 30% (w/v) sucrose. Incubate on ice for 10 min. 3. Centrifuge at 7500 gmax (5500 r.p.m. in the Beckman SW 28 rotor) for 25 min at 4°C.a 4. Use a siliconized, wide-bore Pasteur pipette to recover the fluffy white aggregate of nucleoids from the surface of the sucrose cushion. Dilute the nucleoids with 9 vol. 10 mM Tris, pH 8.0 to a working concentration of ~5 x 106 nucleoids/ml. a If nucleoids from a particular cell type do not sediment under these conditions, use trial and error to establish the centrifugation conditions that should be used.

3.3 The nuclear scaffold The possibility that structures isolated following hypertonic extraction might not reflect in-vivo organization, stimulated the development of procedures that gave stable nuclear derivatives without the need for extraction with buffers containing high concentrations of salt. The two 'mild' treatments used most often extract chromatin with either lithium 3,5-diiodosalicylic acid (LIS; (10)) or 0.25M (NH4)2SO4 (11). While nascent replication products maintain a tight association with the nuclear scaffold, the detergent treatment used destroys transcription sites so that nascent transcripts and active genes are not preferentially associated with this structure (10-12). Protocol 3.

Preparing nuclear scaffolds

Equipment and reagents • Dounce homogenizer . Isolation buffer: 3.75 mM Tris-HCI pH 7.4, 0.05 mM spermine, 0.125 mM spermidine, 0.5 mM EDTA/KOH pH 7.4, 1% thiodiglycol, 20 mM KCI, 0.2 mM PMSF • Low-salt extraction buffer: 5 mM Hepes/ NaOH pH 7.4, 0.25 mM spermidine, 2 mM EDTA/KOH pH 7.4, 2 mM KCI, 0.1% digitonin, 25 mM LIS (Sigma)a

• 0.5 mM CuS04 . Digestion buffer: 20 mM Tris-HCI pH 7.4, 0.05 mM spermine, 0.125 mM spermidine, 20 mM KCI, 70 mM NaCI, 5 mM MgCI2, 0.1% digitonin, 0.2 mM PMSF

Method 1. Prepare nuclei as described in Protocol 1 steps 1-3 using isolation buffer and 0.1% digitonin instead of Triton X-100. Release nuclei by Dounce homogenization, wash 2 x, and resuspend in isolation buffer without EDTA but with 0.1% digitonin at ~5 x 107 nuclei/ml. 2. To 'stabilize' nuclei, either incubate at 37°C for 1 h or add 0.5 mM CuSO 4 for 10 min at O°C.

150

8: Analysing the substructure of mammalian nuclei, in vitro 3. Slowly add 50 vol. of the low-salt extraction buffer. Incubate at room temperature for 10 min. 4. Recover the histone-depleted nuclei by centrifuging at 750 g for 20 min at room temperature, wash 3 x in 10 ml digestion buffer and resuspend in digestion buffer at ~5 x 106 nuclei/ml. s

ln some circumstances it may be better to USe a lower concentration, e.g. E mM.

3.4 The'low-salt'nuclear matrix A 'low-salt' nuclear matrix can be prepared using Protocol I, steps 1-5, by extracting with 0.4-0.5 M (NH4)2SO4 in place of 4 M NaCl at step 5. This extraction is commonly used for morphological studies (8,11).

3.5 The nucleoskeleton While the different preparations described so far suggest that eukaryotic nuclei are highly organized, their impact is compromised by the possibility that the extraction conditions used generate structures that do not always resemble those existing in vivo. To avoid such extreme treatments, nuclei and derivatives can be prepared under isotonic conditions, but they then tend to lyse and aggregate upon manipulation making analysis difficult. However, if cells are first encapsulated in agarose microbeads (13) as in Protocol 4, this physically isolates cells and protects the fragile structures that result when cells are treated under conditions resembling those existing in vivo. Protocol 4 describes a simple procedure for encapsulating living cells in microbeads (Figure /). Cells in PBS supplemented with molten agarose are

Figure 1. Cells encapsulated in agarose microbeads. HeLa cells were encapsulated in agarose as described in Protocol 4. Beads were incubated in medium and visualized by phase microscopy either before (A) or after (B) permeabilizing with Triton X-100 in a physiological buffer. Bar = 100 mm.

151

Dean A. Jackson homogenized with an immiscible phase of liquid paraffin to form an emulsion. When this is cooled, suspended agarose droplets containing cells gel to form microbeads. Subsequently, cells can be grown to give microcolonies or can be permeabilized in solutions containing a 'physiological' concentration of salts. Such cells are permeable to triphosphates and continue to replicate and transcribe the chromatin template at rates close to those found in vivo (14–16). Protocol 4 can be used for any cell culture but is most appropriate for use with cells that are generally grown in suspension. Protocol 4. Encapsulating cells in agarose microbeads Equipment and reagents • . . . •

100 ml round-bottomed flask Flask-shaker (BDH) PBS DEAE-cellulose (Whatman) 50 ml Falcon tubes

•Low-gelling temperature agarose (Sigma Type VII) • Liquid paraffin (Boots the chemist; BDH cat. cat. no. 7162) no.29436:Merck

Method 1. Dissolve 5 g low-gelling temperature agarose in 200 ml PBS and mix with 100 ml fined DEAE-cellulose (pre-equilibrated with PBS) at 50°C for 15 min.a 2. Pour the slurry into 50 ml Falcon tubes and pellet the DEAE-cellulose using a bench centrifuge at 1000 g for 5 min at 37 °C. Decant the molten agarose and store the purified agarose in 10 ml aliquots at -20 °C. 3. For most purposes, encapsulate mammalian cells at cell densities ranging from 106-107 cells/ml beads. Harvest the cells and resuspend to give 5 ml of the desired cell density in 4 ml PBS. Warm the cells to 37°C in a 100 ml round-bottomed flask. 4. Melt 2.5% agarose in PBS, cool to 37°C, add 1 ml to the cell suspension and mix thoroughly. 5. Pour 10 ml liquid paraffin at 37°C into the flask, cover with Parafilm and shake at room temperature at about 800 cycles/min in a flaskshaker until a creamy emulsion forms (—10 see).b Plunge the flask into ice-cold water and leave for ~10 min, until the agarose sets. 6. Recover the beads by mixing the cooled slurry with 35 ml ice-cold PBS, then transfer to a 50 ml Falcon tube and pellet the beads using a bench centrifuge at 750 g for 2.5 min. If some beads remain at the PBS/paraffin interface—remove most of the paraffin, mix and respin.

152

8: Analysing the substructure of mammalian nuclei, in vitro 7. Aspirate the PBS and any remaining paraffin, resuspend the encapsulated cells in fresh medium and incubate under growth conditions for at least 2 h.c a Different batches of agarose contain variable amounts of contaminants that inhibit restriction enzymes and ligases. Lower gelling types have fewer sulfate groups and, for work with DNA, tend to have fewer inhibitors. However, for nuclease digestions, it is advisable to further remove impurities by extracting agarose with DEAE-cellulose. b If a flask shaker is not available shake by hand, as fast as possible for 10 sec. Bead size is controlled by shaking speed, temperature, viscosity, and the interfacial tension between water and the immiscible liquid used. Step 5 gives spherical beads, 50-100 mm in diameter; beads prepared by hand are often non-spherical and more varied in size. Larger beads can be removed by filtering dilute solutions through monofilament nylon filters (R. Cadisch and Sons) in a Swinnex filter (Millipore). Filtration through a 150 mm mesh should remove few beads and leave a filtrate that passes freely through yellow pipette tips. c The cell cycle of encapsulated cells continues normally for one or two cycles, cells then slow and eventually stop growing.

Though almost any cell type can be encapsulated in this way some are more sensitive than others to the encapsulation process. For example, HeLa cells grown in suspension are robust and will survive much more vigorous shaking than freshly isolated lymphocytes, that are rather fragile. Optimal conditions for each cell type should be established by trial and error.

4. Studying the chromatin loops of different nuclear derivatives 4.1 Chromatin loops after hypertonic or hypotonic treatment DNA inside mammalian cells is constrained at intervals so that it forms topologically isolated DNA 'loops' when the histones are removed. These were first described using 'nucleoid' preparations that, unlike early nuclear matrix preparations, retained supercoiled, undamaged, DNA (9). Subsequently, a number of studies demonstrated that particular classes of sequence were associated preferentially with the nuclear cage (4), nuclear matrix (5, 17), and nuclear scaffold (10,12), in vitro. By convention, DNA sequences that remain bound to these structures after removing most chromatin/DNA by nuclease digestion are called nuclear cage-associated sequences, nuclear matrix-associated regions (MARs), and nuclear scaffold-associated regions (SARs), respectively. Nuclear derivatives extracted using either high-salt or LIS have few residual histones. The naked DNA remaining in these preparations provides an excellent substrate for restriction endonucleases that can be used to analyse the arrangement of sequences around the DNA loops, as described in Protocol 5. 153

Dean A. Jackson Protocol 5. Fractionating chromatin loops from histone-depleted nuclear derivatives Reagents • EDTA • Restriction endonucleases

• Gel electrophoresis equipment and reagents • Southern blotting equipment and reagents

Method 1. Prepare derivatives of choice using Protocols 1-3 and dilute them in a suitable digestion buffer3 at a density of ~5 x 106 nuclei/ml. Add the desired restriction endonuclease(s) at an appropriate concentration.b Incubate at 33°C for 30 min, mixing intermittently. 2. Add EDTA to 5 mM to stop the reaction and separate the soluble (released or loop) DNA from the insoluble, derivative-associated DNA by centrifuging at 5-10000 g for 15 min at 4°C. 3. Recover the supernatants (loop) and pellets, prepare DNA,C and analyse by gel electrophoresis and Southern blotting, using standard procedures (see Figure 2). a Most restriction endonucleases work over a wide salt range. Isolation buffers with -100 mM NaCI and 1-2 mM MgCI2 should be OK. Proprietary buffers can also be used. b The concentration required should be established by trial and error; usually in the range 100-1000 U/ml. c To quantitate DNA, it is better to prelabel DNA as described in Protocol 6 and establish the amounts in the pellet and supernatant fractions by scintillation counting. Crude estimates can be obtained from UV scans of the purified material.

4.2 Chromatin loops under 'physiological' conditions If cells are encapsulated in agarose microbeads as in Protocol 4, they can subsequently be lysed using detergents and the chromatin examined under an appropriate environment without difficulty (18, 19). The precise 'physiological' conditions inside mammalian nuclei are not known. Nevertheless, it is clear that a buffer system such as that described in Protocol 6 offers advantages for the preservation of nuclear structure and function. Protocol 6. Fractionation of chromatin loops from the nucleoskeleton under 'physiological' conditions Equipment and reagents • • . .

10 ml clear polycarbonate tubes Vertical gel electrophoresis apparatus Southern blotting apparatus and reagents Peristaltic pump

• Glass-fibre filters • Scintillation counter and Ecolite Scintillation cocktail (ICN) • 200 ™M KH2PO4

154

8: Analysing the substructure of mammalian nuclei, in vitro PBS (Protocol 1) . 40mMEGTA 10 ml clear polycarbonate tubes • |3H)thymidine (0.1 mCi/ml; -50 Ci/mmol, Amersham) 1% SDS • Physiological buffer (PB): 100 mM potas5% trichloroacetic acid (TCA) sium acetate, 30 mM KCI, 10 mM Na2HPO4, Diethylether 1 mM MgCI2, 1 mM Na2ATP (Sigma Type I), Sarkosyl 1 mM dithiothreitol, 0.2 mM PMSF in HPLC pure water RNase Proteinase K . Triton X-100 . TAE buffer: 40 mM Tris, 20 mM NaAl, 2 mMPhenol and phenol/chloroform (1:1 v/v) EDTA pH 8.3

Method 1. To allow quantitation of DNA, grow cells for 24 h in medium supplemented with [3H]thymidine (0.05-0.25 ml (Ci/ml). 2. Adjust the pH of the PB to 7.4 by adding <1% (v/v) 200 mM KH2PO4; the amount required varies slightly as the acidity of the ATP varies from batch to batch. The free Ca2+ levels of this buffer are below 0.3 m,M and can be clamped at 0.1 |xM using 40 mM EGTA. 3. Collect cells encapsulated at ~2 x 106 cells/ml in microbeads (Protocol 4) by centrifuging for 1-2 min in a bench centrifuge at 750 g. Aspirate the medium, add ice-cold PBS to give 1 ml bead slurry/10 ml, invert the tubes to resuspend, and collect the beads by centrifugation, as before. To reduce bead losses, use 10 ml clear polycarbonate tubes for this and subsequent steps. 4. Resuspend the bead pellets in ice-cold PB, collect the beads by centrifugation, resuspend the washed pellet in ice-cold PB containing 0.5% Triton X-100 to lyse the cells, and incubate the sample on ice for 10 min, mixing intermittently. 5. Collect the beads and repeat the incubation in PB containing 0.5% Triton X-100, as in step 4. 6. Wash the beads 3 X in ice-cold PB to remove the Triton X-100. 7. Add restriction endonucleases and incubate at 33°C for 15 min.a Treat a control sample, without added enzyme, in the same way. 8. Prepare a 6-mm thick 1% agarose gel in TAE buffer in a vertical gel box; use a comb to produce wells in the usual way. Prerun the gel for 1 h at 1-1.5 volts/cm using 2 litres of PB, at 4°C. To prevent pH drift, recirculate the buffer at —50 ml/min. 9. Wash the beads once in ice-cold PB, pellet the beads and load the bead slurry onto the gel, with —250 ml bead slurry/well.b Perform electrophoresis at 1-1.5 volts/cm for 15 h or 5 volts/cm for 4 h to remove detached chromatin fragments. Recirculate the buffer throughout to maintain the pH at 7.4 +/- 0.1.

155

Dean A. Jackson Protocol 6.

Continued

10. Recover the beadsc from the gel wells and count the 3H in equivalent samples of digested and control beadsd as follows: (a) add 100 ml beads to 300 ml 1% SDS for 1 h; (b) pipette 3 x 100 ml aliquots onto glass-fibre filters; (c) extract with 5% TCA; (d) dry the filters with ethanol then diethylether; (e) and count 3H in acid-insoluble material. 11. Recover DNA from ~1 ml beads in 5 ml PB by extracting sequentially with: 0.25% sarkosyl and 50 mg/ml RNase for 1 h, 50 mg/ml proteinase Kfor 1 h, then 0.25% SDS for 5 h. Melt the agarose at 70°C for 10 min and extract twice with an equal volume of phenol to remove agarose. Then extract twice with phenol/chloroform, then with chloroform, and finally with diethylether. Precipitate DNA with ethanol, resuspend and analyse DNA fragments by gel electrophoresis and Southern blotting, using standard procedures (Figure 2 and refs 16, 17). "To minimize the general deterioration of nuclear structure, digestions should not be performed at higher temperatures or for longer than 15 min. Use lower temperatures and shorter times, if possible. b Note that beads containing permeabilized cells tend to stick to pipette tips so it is advisable to siliconize the tips to prevent losses. c lf it is necessary to recover both attached and detached chromatin fragments prepare a gel with a single deep well and perform electroelution with the beads inside wide dialysis tubing, with the direction of electrophoresis reversed. Pellet the beads and prepare DNA from the beads and supernatant as described. d To estimate packed bead volumes, use a bench centrifuge to sediment the beads in 100 ml glass capillary tubes, plugged at one end with Plasticine, and measure the volumes of packed beads and supernatant.

As well as allowing the analysis of chromatin loops under isotonic conditions, encapsulated cells also provide a system for comparing the structure of loops in different preparations. Hence, encapsulated cells can be treated with the different buffers described in Protocols 1-3 and the resulting chromatin/DNA loops compared. The beads prevent damage to the fragile preparations. If necessary, nuclei can be prepared and encapsulated as described for cells in Protocol 4 (reduce the shaking speed to prevent damage).

4.3 Technical tips on cutting and electroeluting chromatin Inappropriate Mg2+ and Ca2+ concentrations alter chromatin structure and loop size. The concentrations of MgCl2 and ATP used in Protocol 6 appear to preserve chromatin structure and allow sufficient free triphosphate and Mg2+ both for cutting with restriction enzymes and efficient DNA or RNA syn156

8: Analysing the substructure of mammalian nuclei, in vitro thesis. Mg2+ at higher concentrations, as well as inappropriate concentrations of Ca2+ and Cu2+ ions, spermine, and spermidine, 'fix' chromatin (20) so that many more interactions are seen (18). For most cells, Haelll treatment plus electroelution should leave 6-10% of the total DNA associated with residual nuclei, in beads. Because chromatin (not naked DNA) is cut, the limit digest for HeLa nuclei treated with HaeIII gives 6.3% DNA remaining; to obtain limit digests use <2 X 106 cells/ml beads and 1000 U/ml HaeIII in a 10 ml incubation. Enzymes that cut less frequently give larger fragment and more attached DNA—e.g. 15-25% DNA remains following EcoRl digestion.

4.4 The frequency and nature of attachment sites in different nuclear derivatives An analysis of the chromatin domains of HeLa cells extracted using physiological conditions (Protocol 6) is shown in Figure 2. A typical chromatin domain in nucleoskeleton preparations contains —85 kb DNA (18) and the association of active genes with the nucleoskeleton accounts for the majority of these interactions (19). However, these interactions are dynamic and individual sequences are rarely bound in all cells of a population. This is believed to reflect the transient association of transcribed parts of the genome with the nucleoskeleton in vivo. Active genes are also associated with the nuclear cage of nucleoids (4) and the nuclear matrix (5), but some differences in the classes of attachments are seen. Hypertonic treatment reduces the number of transcription-related attachments, while the conditions commonly used to prepare nuclei (a critical step of matrix preparation) tend to increase the number of bound DNA fragments (18). As a result, the typical average loop sizes in nucleoids and matrices are ~125 and ~50 kb, respectively. MARs tend to be AT-rich and commonly contain binding sites for DNA topoisomerase II (17, 20). This class of DNA fragments is also the dominant species (SARs) in nuclear scaffold preparations (12, 22). The average loop size of scaffold preparations is ~15 kb, and it is not clear how these preparations reflect structures found in vivo (23). The reduction in transcription-based interactions, in these preparations, is not surprising as the extraction procedure is known to destroy transcription complexes. FISH has been used to analyse the arrangement of sequences in DNA loops (halos) liberated from the confines of the nucleus or chromosome by hypertonic extraction. These studies have confirmed that genes tend to be more closely associated with the nuclear matrix than intergenic sequences (24, 25). The population of DNA fragments associated with the different nuclear derivatives have also themselves been labelled and hybridized to chromosomes. This has shown that nucleoskeleton preparations contain residual DNA fragments which are predominantly associated with active 157

Figure 2, The size and organization of chromatin loops analysed under physiological conditions. (A) Encapsulated HeLa cells (labelled with [3H]thymidine) were permeabilized and chromatin loops cut with FcoRt and Haelll. Detached chromatin fragments were removed by electroelution, before remaining fragments were further digested with a battery of eight enzymes and the released and attached fragments separated by electroelution in dialysis tubing. Total [lane 1) and attached (lane 2; 4.3%) DNA from the first digestion and released (larte 3; 2.9%) and attached (lane 4; 1.4%} DNA from the second digestion were purified, and 0.5 mg of each separated in 1,7% agarose and visualized after staining with EtBr. The average chromatin loop length can be calculated from the average fragment length and the amount of attached DNA (17). (B) Residual fragments (10% of total) were prepared using EcoRI and Haelll as above; DNA from whole cells were cut to completion with Haelll and 9, 3, and 1 mg total DNA (lanes 1-3) and 1 mg attached DNA (lane 4) separated in 1.2% agarose. After staining, DNA was transferred to nitrocellulose, hybridized to 32;P-labelled probes, and autoradiographed. Probes were prepared from a library of loop attachment sequences (LAS clones; prepared from sample 4) or random HeLa DNA fragments (total clones; prepared from sample 1). Panels—showing the labelled areas of individual filters—show that most LAS clones are enriched in the nucleoskeleton-associated DNA, whereas total clones are not. For example, after hybridizing with LAS clone 10, a 10-d exposure shows that this sequence is -5-fold enriched in lane 4 relative to lanes 1-3, indicating that in -50% cases this sequence is attached to the nucleoskeieton. DNA cut withHindlllland X174 DNA cut with Haelll were used as markers (M). Reproduced from ref. 19 with permission.

genes—found in chromosomal R-bartds—whereas matrix and scaffold preparations contain fragments that hybridize predominantly to the G-bands, that have far fewer genes, and that are more AT-rich (26). This analysis has served to highlight the different nature of interactions probed by different methods. 158

8: Analysing the substructure of mammalian nuclei, in vitro

5. The morphology of different nuclear derivatives Nuclear derivatives prepared using either high- or low-salt extractions, and analysed by EM, have a characteristic tripartite structure, consisting of a peripheral nuclear lamina, residual nucleoli, and an internal fibrogranular network (3-5, 8, 27). Nuclear scaffold preparations retain much more internal material, dependent on the prior stabilization step (8). Nucleoli and chromatin are the dominant morphological features in cells encapsulated in agarose beads and permeabilized under isotonic conditions (13). However, if Protocol 6 is used to remove this chromatin the immense complexity of the underlying structure becomes apparent (28). Resinless electron micrographs give the best available impression of the organization of non-chromatin structures inside mammalian nuclei (Figure 3). Remarkably, even when almost all chromatin is removed a complex residual structure of coated filaments and associated complexes, such as replication and transcription factories and interchromatin granule clusters, can be seen.

Figure 3. Electron micrograph of the nucleoskeleton. Encapsulated HeLa cells were pertneabilized, chromatin cut with nucleases, - 90% DNA removed, and a 500 nm resinless sectioti prepared for examination by EM. Agarose (A) surrounds the cytoplasmic (C) and nuclear remnants that are separated by the nuclear lamina (L). The nucleoskeleton, a diffuse network of coated filaments (arrowheads), connects regions of the nuclear interior such as the nucleolus (NU) replication factories (F) and many dense sites (D), some of which arise from transcription sites. Bar = 1 mm. Reproduced from ref. 34 with permission,

159

Dean A. Jackson

6. Assaying nuclear function and nuclear proteins in permeabilized cells Early experiments suggested close links between nuclear structure and function (4, 5). However, high-salt matrix preparations are not of general utility for studying nuclear function as almost all interesting activities are extracted. While active RNA polymerases remain in matrices extracted with 0.2 M (NH4)2SO4, cells encapsulated in agarose beads and permeabilized under 'physiological' conditions provide by far the most versatile system for analysing nuclear structure and function. Various reagents can be used to permeabilize the cell membrane while leaving the nuclear membrane essentially intact: e.g. Triton X-100, lysolecithin, and saponin used at —0.01-0.05% (w/v) for a few minutes (Protocol 7). 'Biological detergents' such as streptolysin O, a-toxin, and complement can also be used to selectively permeabilize the external membrane of mammalian cells. These small proteins assemble multimeric complexes in cell membranes, generating pores of characteristic structure. While different lysis procedures have advantages and disadvantages, saponin is suitable for most purposes, and gives a reproducibly uniform lysis with good nuclear morphology.

6.1 Labelling sites of replication and transcription in vitro Encapsulated cells permeabilized under isotonic conditions retain replication complexes that were active in the cell at the moment of lysis. This allows sites of DNA synthesis to be labelled under controlled conditions. Although it is routine to label sites of DNA synthesis in vivo using 5-bromo 2'-deoxyuridine (BrdU), the use of dNTP analogues in vitro allows precursor pools to be depleted by washing so that elongation rates can be modified. A range of different precursors can then be incorporated, allowing double immunostaining of replication sites together with a protein of interest (Protocol 8). In addition, fluorescent analogues allow sites of DNA synthesis to be visualized directly, so avoiding fixation (29). The combination of labelling in vitro (Protocol 7A), chromatin extraction (Protocol 6), and immunostaining in combination with resinless EM allowed the first morphological description of replication 'factories' in mammalian cells (Figure 3; ref. 28). Similarly, permeabilized cells supplemented with NTPs and incubated under optimal conditions perform transcription at rates approaching those seen in vivo. This is only possible, however, using salt conditions optimal for RNA polymerase elongation, i.e. ~350 mM (NH4)2SO4. At 'physiological' salt concentrations the polymerase complexes transcribe through nucleosomes with reduced efficiency. The use of encapsulated cells has demonstrated that transcription, like replication, takes place at specialized nuclear sites—transcription 'factories'—where many active units operate together (30,31). 160

8: Analysing the substructure of mammalian nuclei, in vitro Protocol 7. Labelling sites of replication and transcription in vitro Reagents • PB (Protocol 6) • 10 x replication initiation mix (10 x RIM): PB containing 2.5 mM dATP, dCTP, and dGTP, 1 mM CTP, GTP, and UTP, 0.1-1 mM TTP analogue (see Section 6.2) and MgCI2 at a molarity equal to the triphosphates • PB supplemented with 0.5 U/ml human placental ribonuclease inhibitor (HPRI) (Amersham)

• Saponin • 10 x transcription initiation mix (10 x TIM): PB/HPRI containing 1 mM CTP and GTP, 0.02-1 mM UTP analogue (Section 6.2), and MgCI2 at a molarity equal to the triphosphates

A. Replication 1. Prepare synchronized cells, if required, using a standard procedure. Encapsulate cells as described in Protocol 4, at —2-10 x 106 cells/ml beads; use a lower concentration for immunofluorescence and a higher one for analysing 32P incorporation. 2. Recover the beads from the medium by centrifuging at 300 g for 2 min, wash once in PBS and once in PB; after pelleting aspirate supernatant, replace with fresh solution (>10 vols), mix by inverting (5x) and pellet beads at 300 for 2 min. 3. Permeabilize the cells by incubating the beads in PB supplemented with 100-250 mg/ml saponina for 3 min at 0°C, pellet the beads and wash 2 x in PB.b as in step 2 4. Warm the beads in PB to 33°C and initiate replication by adding 1/10th volume of prewarmed 10 x RIM. 5. If immunostaining, wash 3 times in >10 vol. ice-cold PB, fix and immunolabel using standard procedures or as described in Protocol 8. B. Labelling sites of transcription in vitro 1. Follow Protocol 7A using 10 x TIM in place of 10 x RIM.C a To define conditions, use a twofold dilution series of detergent in PB and assess the level of permeabilization using Trypan Blue exclusion (add 50 ml 1% Trypan Blue in PB to 50 ml packed micrpbeads, after 10 min inspect by light microscopy and score % permeabilized, dark-blue cells). Choose the detergent concentration that permeabilizes >95% cells. b If inhibitors are to be used, add them to the beads for 10 min at 0°C before transferring to 33°C. c The different RNA polymerase activities can be separated using inhibitors: polymerase I activity can be inhibited by growing encapsulated cells in medium supplemented with 0.1-0.2 (mg/ml actinomycin D for 15 min, before permeabilization. The activities of polymerases II and III are inhibited by adding 2 and 250 mg/ml a-amanitin, respectively, to permeabilized cells for 10 min at 0°C before transferring to 33°C.

6.2 Technical tips on labelling sites of replication and transcription A range of DNA precursors conjugated with biotin, digoxigenin, and various fluorochromes are available from Amersham International, Boehringer, Du 161

Dean A. Jackson

Pont Ltd, Sigma, and other suppliers. Most modified precursors support replication by endogenous eukaryotic DNA polymerases when added in place of the equivalent unmodified precursor; though the rate of elongation is typically 10-25% of the level seen with the unmodified precursor. Incubating for 15 min with 20 mM biotin- or digoxigenin-coupled precursors gives good indirect immunofluorescence signals (Protocol 8). While 5- and 2-min incubations with 100 mM biotin-16-dUTP allow detection by light and electron microscopy, respectively, using standard detection protocols. Incorporated label can be detected after 30-min incubations with 20 mM fluorescent precursors. Levels of incorporation can be monitored using [a-32P]TTP. A good level of incorporation and high efficiency can be achieved using 2-20 (mM TTP and 50-100 m Ci/ml [a-32P]TTP: incubate the samples at 33 °C for 2-60 min and measure the incorporation of 32P into acid-insoluble material at appropriate intervals. Though many modified NTPs are available, only BrUTP (Sigma) and biotin-14-CTP (Gibco-BRL) have been shown to be incorporated by mammalian cell RNA polymerases. BrUTP is inexpensive and can be used at the same concentration as the unmodified precursors; incorporation is ~75% of the level seen with UTP and immunolabelling can be performed after 5-30 min of synthesis. Biotin-CTP should be used as described for the replication analogues. Levels of incorporation can be monitored using [32P]UTP. Good incorporation with reasonable efficiency can be obtained using 5 mM UTP and 50-100 m,Ci/ml [a-32P]UTP.

6.3 Studying protein distribution relative to sites of transcription or replication in permeabilized cells To analyse how a particular nuclear protein is distributed with respect to sites of replication and transcription, immunofluorescence can be performed on permeabilized cells in which replication or transcription sites have been prelabelled with an appropriate analogue. In some cases it will be necessary to perform experiments on encapsulated cells; however, for many purposes it is convenient to use cells that have been grown on glass coverslips (Protocol 8). Protocol 8. Analysing the distribution of nuclear proteins in relation to sites of transcription or replication Equipment and reagents • • . .

PB (Protocol 6) Saponin 10 X RIM or TIM (Protocol 7) PB containing 0.5% BSA, 0.1% Tween20(polyoxyethylene sorbitan monolaurate) • Paraformaldehyde (pFA) . Appropriate primary and secondary antibodies

. 4',6-diamidine-2-phenylindole dihydrochloride (DAPI) (Boehringer) or TOTO-3 (Molecular Probes

• PBS • Mounting medium (Chapters 4 and 7) • Fluorescent microscope equipped with suitable excitation and emission filters

162

8: Analysing the substructure of mammalian nuclei, in vitro Method 1. Grow adherent cells on 15 mm diameter round glass coverslips by seeding ~105 cells in 2 ml fresh medium in a 60 mm Petri dish. Grow for —24 h or until cells are —50% confluent. 2. Rinse the coverslips8 in PBS. Permeabilize cells by immersing them in ice-cold PB supplemented with 100mg/mlsaponin for 2 min, and rinse 2 x with ice-cold PB. 3. Place the coverslips on a piece of Parafilm in a humidified chamber at 33°C; cover with 250 ml of PB supplemented with the appropriate 10 x IM (Protocol 7) and incubate for 15 min. 4. Rinse the coverslips 3 x in ice-cold PB, fix in ice-cold PB with 4% pFA for 15 min, and rinse 3 x in ice-cold PB containing 0.5% BSA and 0.1% Tween-20. 5. Cover the samples with PB/BSA/Tween-20 containing 1/500-1/1000 dilution of the first antibody (e.g. goat anti-biotin, Sigma cat no. B3640) and incubate at 0°C for 1 h. Rinse the coverslips 3 X in ice-cold PB/BSA/Tween-20. 6. Cover the samples with PB/BSA/Tween-20 containing an appropriate dilution (generally 1/100-1/2000, determined empirically) of the chosen second antibody (usually a mouse monoclonal antibody) and incubate at 0°C for 1 h. Rinse the coverslips 3 x in ice-cold PB/BSA/Tween-20. 7. Cover the samples with PB/BSA/Tween-20 containing 1/500 dilutions of appropriate fluorochrome-coupled secondary antibodies, together (e.g. FITC-donkey anti-goat IgG, Jackson Immunoresearch cat. no. 705-095-131; Cy3-donkey anti-mouse IgG, Jackson Immunoresearch cat. no. 715-165-151) and incubate at 0°C for 1 h. Rinse the coverslips 3 x in ice-cold PB/BSA/Tween-20. 8. Cover the samples with ice-cold PBS/BSA/Tween-20 containing 0.02 mg/ml DAPI or 20 mM TOTO-3 for 10 min, at room temperature. Rinse the coverslips 2 x in PBS, drain, mount with 2.5 ml mounting medium (Chapters 4 and 7) and seal with nail varnish. 9. Visualize on a fluorescent microscope equipped with suitable excitation and emission filters. DAPI is excited by UV, TOTO-3 by far red light. a Encapsulated cells can be labelled in the same way. Wash the beads in PB and immunolabel using the reagents and times given above. For a typical labelling reaction, use 50-100 ml beads, add 200-500 ml ice-cold PB/BSA/Tween-20 containing antibody for 1 h and wash 3 x with 1 ml ice-cold PB/BSA/Tween-20. To wash, pellet the beads by centrifuging at 750 g for 1 min, aspirate the supernatant, add 1 ml fresh buffer, mix by inverting the tube, and stand the sample on ice for 5 min.

163

Dean A.Jackson As with all immunodetection systems, the quality of the reagents will determine the success of multiple-labelling experiments. It is usually necessary to titrate the reagents to establish the concentration that gives the best signal with little or no background of non-specific binding, and to use controls to confirm that different reagents are not cross-reacting. It is also important that the order in which antibodies are added does not influence the extent of binding—if trials show that the bound first antibody reduces the signal from the second then both should be added together. Cells encapsulated in agarose beads afford excellent opportunities for the analysis of proteins that interact with different nuclear compartments. Even when all protein has been extracted from encapsulated cells the remaining DNA is protected from damage and remains supercoiled (32). Consequently, it is a simple matter to extract encapsulated cells using a wide range of reagents, either separately or sequentially, and to analyse the structure of the active compartments and determine how these relate to the distribution of different proteins under the conditions used. Following extraction, the organization (after immunolabelling) or composition (using PAGE) of the residual structures can be analysed using standard procedures.

6.4 A typical example Figure 4 (C-E) shows three images that indicate the distribution of DNA, SC35 splicing factor, and transcription sites in the same confocal microscope section of a HeLa cell labelled and processed as described in Protocol 8. For comparison, the distribution of DNA and SC35 in a cell that was fixed without permeabilization is also shown (Figures 4A, B); note that the texture of the nuclear staining is not affected by the permeabilization and in-vitro labelling conditions used. It is important to remember, however, that immunostaining will not correlate directly with function; for many nuclear proteins the majority of molecules are non-functional at any time. Most SC35 (perhaps as much as 90%) is concentrated in speckles that appear to be sites of storage and not active splicing; most active splicing takes place at the site of transcription (33).

7. Conclusions It is now generally accepted that the nuclei of mammalian cells are compartmentalized, so that different nuclear functions are performed at specialized sites (6, 7, 16), where the components required to perform a particular task are concentrated within synthetic 'factories'. Chromatin is also organized, with each chromosome occupying a discrete nuclear 'territory', with predominantly active or inactive regions occupying different nuclear domains that correspond to specific subchromosomal 'bands', during mitosis. Given the complexity of mammalian nuclei, it is remarkable that the active compartments maintain their spatial organization when almost all chromatin, representing roughly 50% of the nuclear material, is removed (Figure 3). This 164

8: Analysing the substructure of mammalian nuclei, in vitro

TOTO-3

SC35

Bio-RNA

Figure 4, Nuclear compartments rich in splicing proteins and nascent transcripts. HeLa cells growing on coverslips were either fixed (A-B) or permeabilized with saponin in an isolonic buffer and sites of RNA synthesis labelled for 15 min with biotin-CTP (C-E). Sites containing SC35 (B, D) or biotin-RNA IE) were indirectly imrnunolabelled and DNA stained with TOTO-3 (A, C), as described in Protocols. Optical sections (-700 nm) show a classical distribution of SC35 antigens, with major sites 'speckles' and dispersed minor foci; nucleoli are blank. In the nucieoplasm, most sites of transcription lie adjacent to minor foci (E), Note that in (E), anti-taiolin antibodies label an extensive mitochondrial network. This indicates the preservation of cellular structure under the conditions used. Bar = 2.5 mm. Figure kindly supplied by Ana Pombo, see ref. 31 for details.

suggests that the synthetic factories are structures in their own right and that they are organized through their association with the nucieoskeleton; this organization could not be maintained if the active compartments contained chromalin domains with many unrelated active complexes. Furthermore, as DNA associated with factories is bound indirectly to the nucieoskeleton, active sequences will be recovered from preparations that preserve these structures. While different classes of bound DNA are associated with different nuclear derivatives, it seems likely that the residual network seen in Figure 3 will be related to the nuclear matrices or nuclear scaffolds that result when nuclei are extracted in appropriate ways. The extent to which these different preparations are related and precisely how they reflect the organization existing in vivo are intriguing questions that we are now able to address.

References 1. Mirsky, A,E. and Ris, H, (1951).,J. Cm Pliysiol., 34. 475. 2. Oeorgiev, G.P and Chcnlsoy, J.S. (1%2). Exp Cell Res., 27, 572. 165

Dean A.Jackson 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34.

Berezney, R. and Coffey, D.S. (1974). Biochem. Biophys. Res. Commun., 60,1410. Jackson, D.A., McCready, S.J., and Cook, P.R. (1984). J. Cell Sri., Suppl. 1, 59. Verheijen, R., van Venrooij, W., and Ramaekers, F. (1988). J. Cell Sci., 90, 11. Berezney, R., Mortillaro, M.J., Ma, H., Wei, X., and Samarabandu, J. (1996). Int. Rev. Cytol., 162A, 1. van Driel, R., Wansink, D.G., van Steensel, B., Grande, M.A., Schul, W., and de Jong, L. (1995). Int. Rev. Cytol., 162A, 151. Belgrader, P., Siegel, A.J., and Berezney, R. (1991). J. Cell Sci., 98, 281. Cook, P.R. and Brazell, LA. (1975). /. Cell Sci., 19, 261. Mirkovitch, J., Mirrault, M.-E., and Laemmli, U.K. (1984). Cell, 39, 223. Capco, D.G., Wan, K., and Penman, S. (1982). Cell, 29, 847. Laemmli, U.K., Kas, E., Poljak, L., and Adachi, Y. (1992). Curr. Opin. Genet. Dev., 2, 275. Nilsson, K., Scheirer, W., Merten, O.W., Ostberg, L., Liehl, E., Katinger, H.W.D., and Mosbach, K. (1983). Nature, 302, 629. Jackson, D.A. and Cook, P.R. (1985). EMBO J., 4, 913. Jackson, D.A. (1997). Mol. Biol. Rep., 24, 209. Koob, M. and Szybalski, W. (1992). In Methods in enzymology, Vol. 216, (ed.), p. 13. Bode, J., Schlake, T., Rios-Ramirez, M., Mielke, C., Stengert, M., Kay, V., and Klehr-Wirth, D. (1996). Int. Rev. Cytol., 162A, 389. Jackson, D.A., Dickinson, P., and Cook, P.R. (1990). EMBO J., 9, 567. Jackson, D.A., Bartlett, J., and Cook, P.R. (1996). Nucl. Acids Res., 24, 1212. Guo, X.-W. and Cole, R.D. (1989). J. Biol. Chem., 264,16873. Boulikas, T. (1995). Int. Rev. Cytol., 162A, 279. Gasser, S.M. and Laemmli, U.K. (1987). Trends Genet., 3,16. Jack, R.S. and Eggert, H. (1992). Eur. J. Biochem., 209,503. Gerdes, M.G., Carter, K.C., Moens, P.T., and Lawrence, J.B. (1994). /. Cell Biol., 126, 289. Bickmore, W.A. and Oghene, K. (1996). Cell, 84, 95-104. Craig, J.M., Boyle, S., Perry, P., and Bickmore, W.A. (1997). J. Cell Sci., 110, 2673. Nickerson, J.A., Krockmalnic, G., Wan, K.M., and Penman, S. (1997). Proc. Natl Acad. Sci. USA, 94, 4446. Hozak, P., Hassan, A.B., Jackson, D.A, and Cook, P.R. (1993). Cell, 73, 361. Hassan, A.B. and Cook, P.R. (1993). /. Cell Sci., 105, 541. Jackson, D.A., Hassan, A.B., Errington, R.J, and Cook, P.R. (1993). EMBO J., 12,1059. Iborra, F.J., Pombo, A., Jackson, D.A, and Cook, P.R. (1996). /. Cell Sci., 109, 1427. Cook, P.R. (1984). EMBO J., 3, 1837. Pombo, A. and Cook, P.R. (1996). Exp. Cell Res., 229,201. Hozak, P, Jackson, D.A, and Cook, P.R. (1994). J. Cell Sci., 107, 2191.

166

9

Chromosome assembly in vitro using Xenopus egg extracts JASON R. SWEDLOW

1. Introduction This chapter presents techniques for the generation of mitotic chromosomes in vitro using Xenopus egg extracts. Discussions of technical details required for the successful use of the system have been included. Methods for immunodepletion of specific factors from extracts and immunolocalization of specific proteins on chromosomes are also presented. An excellent compilation of detailed discussions and methods for the study of nuclear and cell cycle dynamics using Xenopus oocyte and egg extracts has been compiled by Kay and Peng (1).

2. Chromosome structure and biochemistry The replication, transcription, and segregation of DNA in eukaryotes occurs within the context of chromatin and chromosomes. Chromatin shifts from condensed to decondensed states during transcriptional activation and during the transition from mitosis to interphase. Two extremes, the 10000-fold linear compaction of the DNA in heterochromatin or the mitotic chromosome, and the completely unfolded DNA present at replication forks and highly transcribed loci, define the limits of a range of structures that can occur simultaneously in a single cell. To mediate these changes, a host of proteins are available that promote efficient packaging by neutralizing the high negative charge of the DNA-phosphate backbone and forming specific nucleoprotein structures (reviewed in ref. 2). The DNA is wrapped twice around a disc-shaped octamer formed by the histones H2A, H2B, H3, and H4. Numerous non-histone proteins also serve to set up the structure of chromatin and chromosomes and mediate the various processes that utilize an organism's DNA.

Jason R. Swedlow

3. Preparation of Xenopus egg extracts for chromatin and chromosome assembly in vitro 3.1 Xenopus egg maturation In amphibians, immature oocytes are induced to mature, i.e. proceed through meiosis and arrest at metaphase of meiosis II, by the action of progesterone (3). The eggs are arrested at the metaphase state by the action of cytostatic factor (CSF), at least one component of which is the proto-oncogene c-mos (4). Once the eggs are laid, fertilization causes a transient increase in intracellular Ca2+. This triggers the degradation of CSF and subsequent exit from metaphase arrest, completion of meiosis, and entry into the embryonic cell cycle. In the laboratory, egg maturation is induced by injecting Xenopus females with pregnant mare serum gonadotropin (PMSG) to induce progesterone release. Subsequent injection with human chorionic gonadotropin (hCG) induces ovulation.

3.2 Xenopus egg extracts The large number and size of eggs (1-2 mm) produced by Xenopus makes it possible to prepare highly concentrated cytoplasmic extracts that support many in-vitro reactions. Eggs are packed in the presence of an inert oil by low-speed centrifugation to eliminate as much buffer as possible and minimize the volume surrounding the eggs. The eggs are then crushed by centrifuging them at a higher speed. During this step, the egg cytoplasm separates from pigment granules, lipid, and nuclei and can subsequently be isolated as a reasonably clean, highly concentrated (~60 mg/ml) cytoplasmic preparation. If the eggs are crushed in the presence of a Ca2+ chelator, CSF degradation is prevented and the resulting cytoplasmic extract continues to be arrested in the metaphase state (5). This 'CSF extract' has high H1 kinase activity, and, when supplemented with exogenous nuclei, will support the formation of condensed chromosomes and mitotic spindles (5, 6). Alternatively, eggs can be 'activated', i.e. induced to exit CSF arrest, by treatment in an electric field or with a calcium ionophore. Extracts made from these eggs will be in an interphase state: when supplemented with exogenous nuclei, these extracts will support DNA replication and the selective import of nuclear proteins (7-9). Sperm nuclei, or even naked DNA, will be assembled into nuclei surrounded by a nuclear envelope (10,11). These extracts will also synthesize cyclin protein from mRNA stored in eggs and will eventually enter mitosis (12). A stable interphase extract can be generated by inhibiting protein synthesis, and thus inhibiting the synthesis of cyclins. These extracts contain all the components of cytoplasm including membranes, mitochondria, and ribosomes and are therefore not an ideal substrate for biochemical manipulation. It is often convenient to pellet these structures 168

9: Chromosome assembly in vitro using Xenopus egg extracts and use a cytosol containing only soluble factors. Many of the activities associated with the extracts do not survive freezing, but clarified cytosol made from CSF or activated extracts supports chromatin assembly and chromosome condensation and can be stored frozen for >3 months (5, 13, 14). The clarified CSF cytosol has been used successfully to identify and characterize a number of constituents of mitotic chromosomes (14,15). It is important to note that these extracts are made from eggs arrested at metaphase of meiosis II. Sister chromatids are disjoined at exit of this metaphase, so functionally this metaphase is similar to a true mitotic metaphase. However, this metaphase state is distinguished from a true mitosis in that it is stabilized by CSF acting through the ERK2 signalling pathway (16). Therefore, nuclei added to a CSF extract enter cytoplasm that approximates, but is not identical to, a mitotic state (see Section 3.3). Moreover, they enter mitosis without previously passing through S phase. Most of the proteins assembled into chromosomes in a CSF-clarified cytosol are also found on mitotic chromosomes (13-15, 17), so a CSF extract provides a good model system for chromosome assembly. However, a recent analysis of the localization of kinetochore proteins showed that some are not targeted to discrete sites on chromosomes made in vitro from sperm nuclei in CSF extracts (18). In short, the Xenopus in-vitro chromosome assembly system has provided a powerful approach to study the formation and maintenance of chromosome structure, but the protocols described below may not duplicate every aspect of chromosome dynamics in vivo and further modifications may be required to exactly duplicate an in-vivo chromosome.

3.3 Chromosome assembly extracts—technical tips Protocols 1, 2, and 3 describe the preparation of extracts for in vitro chromatin and chromosome assembly. The protocols were developed based on those described by Andrew Murray (19), none the less there are significant modifications that improve the suitability of the extracts for chromosome assembly. Protocol 1. Induction of egg maturation and ovulation Equipment and reagents • Sterile 1 ml syringe fitted with a 27 gauge 1/2 inch needle • 100 U/ml pregnant mare serum gonadotropin (PMSG, Calbiochem) in H20; store at -20°C • MMR: 5 mM Hepes pH 7.8, 100 mM NaCI, 2 mM KCI, 1 mM MgCI2, 2 mM CaCI2, 0.1 mM EdTA (can be prepared as a 25 x stock solution)

• Xenopus females, 7-9 cm in length • 16°C incubator, or (for significantly less cost) a standard lab. refrigerator refitted with a thermostat that is accurate at 16°C and a small fan to help circulate cool air inside. • 1000 U/ml human chorionic gonadotropin (hCG, Sigma) H2o; store at 4°C. Discard dissolved hormone after 5 days.

169

Jason R. Swedlow Protocol 1. Continued Method 1. On day 1, inject each female frog with 50 U PMSG. Typically, inject 4-8 animals for each cytosol preparation. Inject the solution subcutaneously into the dorsal sac, slightly anterior to the intersection of the stitched patterns at the base of the animal's back. Injected animals can be kept together in tanks. 2. (Optional). On day 3, inject each animal again with 25 U PMSG. The need for this second injection generally depends on the health of the animals. If the animals are thriving, the increase in yield of eggs is not worth the additional stress to the animal from a second injection. 3. On days 5-19, inject each frog with 500 U hCG and place each animal in an individual plastic container with 2 I of MMR. Store the frogs at 16°C for 20-22 h, preferably in the dark.

The time specified in Protocol 1 for full ovulation is 20-22 h at 16°C. This time is longer than that used for cycling or spindle-assembly extracts (19). Chromosome assembly is unaffected by the longer laying time as long as the eggs are kept at 16 °C, but the yields are significantly increased, even if the animals are squeezed after 16 h. Animals can be squeezed gently by stroking their abdomen, but by 22 h, most animals have laid all their eggs. Protocol 2. Preparation of CSF Xenopus egg cytosol Equipment and reagents • Clinical centrifuge with a 4- or 6-place rotor . .5 M K+EGTA pH 7.7 2 mM MgCI2 stock (IEC) solution . TLS-55, SW55, SW50.1 (Beckman), HB-4, or . XBE2: XB, 5 mM K+EGTA pH 7.7. Prepare HB-6 (Sorvall) rotors (or equivalent) from stock solutions. Adjust to pH 7.7. . Ultraclear 13 x 51 mm and 11 x 34 mm . Dejellying solution: XB + 2% cysteine free tubes (Beckman or equivalent) base (Sigma). Adjust the pH to 7.8 with 5 M « 3 ml syringes, 18G and 20G needles KOH. Make this solution immediately • 10 mg/ml cytochalasin D (Sigma) in anhybefore starting the prep. drous DMSO; store in aliquots at -20°C. . Versilube F-50 Silicone Fluid (Andpak-EMA, • LPC: leupeptin, pepstatin A, chymostatin cat. no. MIL-S-81-087C) (Chemicon) dissolve together at 10 mg/ml . Frog fix: 67 ml 80% glycerol, 10 ml 10 x each in anhydrous DMSO; store in aliquots MMR (Protocol 1 gives 1 x MMR), 23 Ml at-20°C. 16% CH20 (Ted Pella, MeOH free), 1ml0.1 • 20 x CSF energy mix: 20 mM ATP, 20 mM mg/ml Hoechst 33258 or DAPI (Sigma) MgCI2, 150 mM phosphocreatine; store in . open-bore transfer pipettes (2-4 mm openaliquots at-20°C. ing polypropylene transfer pipette or fire. 20 x XB salts: 2 M KCI, 20 mM MgCI2, 2 mM polished glass Pasteur pipette with its tip CaCI2 cut off) . 1 M K+Hepes pH 7.7 stock solution . 15mlpolypropylene culture tubes (Falcon) . XB: 10 mM K+Hepes pH 7.7, 50 mM . Fluorescence and phase-contrast microsucrose, 1 x XB salts. Prepare from stock scopes solutions. Adjust to pH 7.7. . 22 x 22 mm coverslips • 2 M sucrose stock solution

170

9: Chromosome assembly in vitro using Xenopus egg extracts Method 1. Before beginning to process the eggs from 4-5 healthy frogs (Protocol 1), prepare all the reagents above and the following: 2 I MMR, 500 ml dejellying solution, 500 ml XB, 350 ml XBE2. 2. Examine the eggs laid by the frogs. Eggs should appear spherical, with clear boundaries between the hemispheres of the animal pole (brown in hue, usually faces up when settled) and vegetal (white, usually faces down) poles. Discard all the eggs from a frog if more than 10% are defective, e.g. irregular shape or boundaries or mottled colouring, white balls, egg strings, etc. Remove any defective eggs, skin, faecal matter, or food using an open-bore transfer pipette.3 3. Pool the eggs from all the frogs into a 2 I beaker and wash 4-5 times with MMR to remove any foreign material. After each wash, allow the eggs to settle to the bottom of the beaker and pour off the supernatant.6 Wash until all pieces of detritus are removed. 4. Wash the eggs with a small amount of dejellying solution, then add —400 ml of this solution. Swirl the eggs gently every 1-2 min and allow them to settle at room temperature. The eggs will gradually settle and pack into a layer on the bottom of the beaker with their animal poles facing up.c 5. While the eggs are dejellying, add 100 ml LPC to 100 ml XBE2 and add 1 ml of this solution to each 13 x 51 mm centrifuge tube. Add 10mlof 10 mg/ml cytochalasin D to each tube, mix the solution vigorously with the pipette tip to prevent precipitation. Eggs from one frog will fill one to two 13 x 51 mm tubes. 6. Pour off the dejellying solution and wash the eggs 4-5 times with XB, removing any defective eggs that appear.01 Wash the eggs 4 times with the remaining XBE2, then with the XBE2 + LPC (from step 5). Save 1-5 mi of this solution in case an extra tube is needed for excess eggs. 7. Transfer the eggs from the beaker to the centrifuge tubes using an open-bore transfer pipette. Minimize the amount of buffer introduced into the pipette by putting its tip deep into the eggs before aspirating—this minimizes the dilution of cytochalasin in the tube. Fill each tube; the eggs will settle over the next 2 min leaving ~0.5 ml of buffer. 8. Aspirate as much of the buffer as possible from the top of the tube and replace it with Versilube, filling the tube to its top. Load the centrifuge tube into a 15 ml polypropylene culture tube (Falcon) and insert into a clinical centrifuge. Centrifuge at 190 g for 1 min, then increase to 750 g for 30 sec. Remove the excess buffer and oil. This procedure packs the eggs, expels the buffer between them, and so minimizes any dilution of the extract by buffer.8

171

Jason R. Swedlow Protocol 2.

Continued

9. Load the tubes into the SW55 (or equivalent) rotor and crush them by centrifuging at 12500 r.p.m. for 15 min at 16°C. Best results have been obtained with minimum brake; the exact setting varies with different centrifuges. In general, an ultracentrifuge is preferred because the acceleration and deceleration are faster and more accurate. When using an ultracentrifuge, wait for the vacuum to reach the level where the centrifuge will not pause during acceleration before starting the run. This maximizes the amount of time spent at the highest speed. 10. Remove the tubes from the centrifuge and place them on ice, keep the extract on ice for the remainder of the protocol. The contents of the eggs will have separated into three layers: a thick black pellet containing pigment granules, yolk, and nuclei; a brown, turbid cytoplasmic layer in the middle; and a yellow lipid layer on top. Using a 3 ml syringe and an 18G needle, puncture the side of the tube and slowly remove the brown, turbid crude cytoplasm. To prevent blockage, use a new needle to remove the cytoplasm from each tube. Combine the cytoplasmic fractions from all the tubes into a single prechilled 15 ml polypropylene culture tube. 11. Add LPC and cytochalasin D to final concentrations of 10 mg/ml and 1/20th volume of 20 X CSF energy mix. Mix by gently inverting the tube. This is the crude extract (also called the 'low-speed supernatant' or 'low-speed extract'). 12. All extracts should be tested by adding nuclei to the crude extract and following the formation of interphase nuclei or mitotic chromosomes as an indication of the cell-cycle state of the extract. Add 1 ml Xenopus sperm nuclei (106/ml) to 20 ml of the crude extract (19) and incubate at 18°C. After 30 min, spot 2 ml of the reaction on a glass slide, overlay with 4 ml of frog fix and squash with a 22 x 22 mm coverslip. Using a fluorescence microscope, look for the appropriate chromatin morphology; mitotic chromosomes will be condensed masses of tangled threads <0.5 mn wide, interphase nuclei will appear as decondensed balls. After a 1-h incubation nuclear envelopes should be visible by phase-contrast microscopy. 13. To make the cytosol for chromosome assembly reactions, load the low-speed supernatant into Ultraclear 11 x 34 mm tubes (Beckman or equivalent) and spin in a TLS-55 rotor (Beckman or equivalent) at 214000 g for 2 h at 4°C. For larger volumes (>8 ml), an SW50.1 or SW55 rotor works well. The crude extract will have separated into: an orange gelatinous pellet containing glycogen and ribosomes; a greyish brown layer of mitochondria, membranes, and annulate lamellae; a clear supernatant; and a floating layer of white lipid. Using a syringe and a 20G needle, puncture the side of the tube and remove the clear

172

9: Chromosome assembly in vitro using Xenopus egg extracts supernatant between the pelleted membranes and floating lipid. Pool the supernatants from the tubes and centrifuge again in a TLS-55 rotor at 214000 g for 30 min at 4°C to remove any contaminating heavy membranes. Remove the clear supernatant above the membrane pellet with a syringe and a 20G needle. This is the clarified cytosol. Freeze 10-100 ml aliquots immediately in liquid N2, and store at-80°C until use. a Do not hesitate to sacrifice good eggs while removing bad eggs. The removal of defective eggs is much more important than the loss of some good ones. b When adding fresh MMR, pour it down the side of the beaker to minimize disturbance to the eggs—too much jostling will cause them to activate. c This should take 5-8 min to complete. Longer times are usually indicative of poor eggs or inactive cysteine and will likely result in poorly performing extracts. d The supernatant of each wash should be clear—if it is cloudy, egg lysis is occuring. Either too many degenerate eggs are present or the eggs are being handled too vigorously. Again, this will result in a poorly performing extract. e From the start of step 5 to the end of step 8 should take no more than about 23 min.

Most published protocols for extract preparation specify centrifugation at 12152 g for 10 min to crush the eggs (19). For chromosome assembly extracts, the speed is increased to 18 988 g and the run extended to 15 min to improve the yield of extract. The resulting extract does not advance through the cell cycle, but works well for chromosome assembly. Most protocols specify the use of cytochalasin B in extracts, this prevents gelling upon egg lysis due to the polymerization of actin. Cytochalasin B can be used if the extract will only be used for the assembly of chromosomes. If the extract is to be used for the biochemical analysis of isolated chromosomes or immunodepletion, cytochalasin D at 10 mg/ml must be used instead (as in Protocol 2). This will prevent the polymerization of actin and the contamination of isolated chromosomes with actin and actin-binding proteins. All chromosome assembly extracts are made in a Hepes/KCl buffer system (XB of Protocol 2 (19)) that approximates the physiological concentration of at least some ions. In some protocols, extracts have been made using a p-glycerophosphate/EGTA-based buffer ('EB' (20)). (b-glycerophosphate inhibits protein phosphatase 1, thus the activity of mitotic kinases is conveniently stabilized in EB-based extracts. Indeed, H1 kinase levels are not as high in XB-based extracts as in those made with EB (J.R. Swedlow, unpublished results). Occasionally, CSF-XB extracts will inadvertently activate during the preparation of the cytosol by centrifugation. We have found that the addition of MgATP and phosphocreatine to crude extracts before further centrifugation completely prevents the activation of CSF-XB extracts during centrifugation. To ensure that adequate amounts of ATP are being added, the concentration of ATP stock solutions is always determined spectrophotometrically at 259 nm using = 15400 M/cm. The CSF extract produced in Protocol 2 approximates a mitotic state and 173

Jason R. Swedlow can be used to generate condensed chromosomes. Mitotic extracts may also be generated by incubating an interphase extract with a non-degradable form of cyclin (e.g. A90-cyclin B) (12, 21). The amount of this reagent to add depends on the protein being used, and can be determined empirically by measuring the minimum amount of cyclin necessary to achieve maximal H1 kinase levels. When converting an interphase clarified cytosol to a mitotic state, the extract must be allowed to incubate (usually for 20 min) after H1 kinase levels have reached a maximum before nuclei are added (J.R. Swedlow, unpublished results). This allows downstream targets of the cdc2 kinase to be fully activated and then produces well-condensed mitotic chromosomes bearing mitotic chromosomal proteins. Protocol 3. Preparation of interphase Xenopus egg cytosol from activated eggs Equipment and reagents • MMR (Protocol 1) • Dejellying solution (Protocol 2); prepared immediately before starting the preparation « MMR/Cx: MMR + 0.1 mg/ml cycloheximide (Sigma) • Activation buffer: MMR/Cx + 2 mg/ml calcium ionophore A23187 (Sigma) • LPC (Protocol 2)

• XB/Cx: XB (Protocol 2) + 0.1 mg/ml cycloheximide • 13 x 51 mm centrifuge tubes (Protocol 2) • 10 mg/ml cytochalasin D (Protocol 2} . Open-bore transfer pipettes (Protocol 2) . Centrifuge, rotors, and centrifuge tubes (Protocol 2)

Method 1. Before beginning to process the eggs from 4-8 frogs (Protocol 1), prepare all the reagents above and the following solutions: 2 litre MMR, 500 ml dejellying solution, 500 ml MMR/Cx, 100 ml activation buffer, 600 ml XB/Cx. 2. Pool, wash, and dejelly the eggs (Protocol 2, steps 3 and 4). While the eggs are dejellying, add 100 ml LPC to 100 ml XB/Cx. Add 1 ml of this solution to each 13 x 51 mm tube. Add 10 ml of 10 mg/ml cytochalasin D to each tube, mixing the solution vigorously with the pipette tip to prevent precipitation. 3. Wash the eggs 4 times with MMR/Cx. 4. Wash the eggs twice with a small amount of the activation solution, then add the remaining solution. Incubate the eggs at room temperature for 15 min, swirling every 2 min. After about 3 min, the vegetal pole will contract around the animal pole—this is the result of the degradation of CSF and exit from meiotic arrest. 5. Wash the eggs 4 times with XB/Cx. Wash the eggs with XB/Cx + LPC (step 2). Proceed to make the interphase cytosol as described in Protocol 2, steps 7-13.

174

9: Chromosome assembly in vitro using Xenopus egg extracts

4. Chromatin and chromosome assembly in vitro Chromatin and chromosomes are complex, irregular, and delicate structures. Isolation from tissue or cells often introduces structural perturbations and cytoplasmic contaminants. The development of an in vitro system based on cytoplasmic extracts from Xenopus laevis' eggs has allowed the generation and isolation of mitotic chromosomes and the identification and characterization of a number of chromosomal proteins (14,15, 22). Assembled chromatin and chromosomes can easily be separated from cytoplasmic contaminants, allowing biochemical analysis of the differences between interphase chromatin and mitotic chromosomes. Because the assembly reaction works in vitro, the molecular basis of chromatin assembly and chromosome condensation is experimentally accessible in a functional system.

4.1 Assembly and isolation of chromatin and chromosomes—technical tips The successful assembly of chromatin and chromosomes and purification from cytoplasm requires careful attention to a number of different parameters. The morphological criteria for the formation of chromosomes is illustrated in Figure 1. In all cases, the progression through the time-course shown in Figure 1 should be fairly synchronous in the majority (>80%) of nuclei on a slide. The formation of fully condensed chromosomes takes 2-3 h, depending

Figure 1. The time course of interphase chromatin and mitotic chromosome assembly from Xenopus sperm nuclei. Interphase chromatin or mitotic chromosomes are made by adding sperm nuclei to interphase or mitotic CCSF'l cytoplasmic extracts. After incubating at 18°C for the indicated time, reactions were fixed with 2% CH20 and stained with DAPI.

175

Jason R. Swedlow on the particular extract preparation. After condensation is complete, individual chromosomes (Figure 1, 180 min) should dissociate from condensing nuclei (Figure 1, 30 and 60 min). The exact dimensions of fully condensed chromosomes depends on the source of the nuclei, but they are generally at least 0.4 mm thick. The assembly of well-condensed chromosomes depends on diluting the extract with 1 volume of an XB-based buffer (Protocol 4). Further dilution does not produce well-separated condensed chromosomes. For isolation, the dilution is critical for minimizing contamination from cytoplasmic proteins, especially actin and actin-binding proteins. For biochemical analysis, extracts are also diluted with 1 volume of buffer. The structure of chromatin is also highly sensitive to the concentration of divalent cations. The formation of well-separated chromosomes is very sensitive to the concentration of Mg2+. All the stock solutions for Protocol 4 contain Mg2+, but each investigator should establish the ideal concentration for their assembly system. When using Xenopus sperm nuclei as a substrate, we dilute the cytosol with XB supplemented with 5 mM EGTA and a final MgCl2 concentration in the buffer, after the addition of an ATP regeneration system, of 5 mM ('XBE5', see Protocol 4, step 1). The exact concentration of Mg2+ does not affect the presence of chromosome proteins detected after isolation. An ATP regeneration system is added to the reaction to support the assembly reaction in Protocol 4 during the lengthy incubation. This is added to the dilution buffer immediately before extract dilution. An active ATP regenerating system is critical for the efficient formation of well-separated, condensed chromosomes and for preventing contamination of isolated chromosomes with cytoplasmic proteins. To ensure that adequate amounts of ATP are being added, the concentration of ATP stock solutions is always determined spectrophotometrically at 259 nm using = 15400 M/cm. The clarified cytosolic extracts of Protocols 2 and 3 will assemble chromatin and chromosomes from a variety of DNA substrates. Sperm nuclei are the natural substrate for this reaction, and are therefore used in many cases. An excellent protocol for purifying and demembranating Xenopus sperm nuclei has been described (19). The assembly of naked DNA substrates into chromatin by these and oocyte extracts is well known (13, 23, 24), but it is important to note that this assembly occurs in the absence of replication. The efficiency of chromatin assembly on linear and circular DNA substrates is significantly reduced from that on sperm, and the structure of the assembled chromatin is significantly different. Besides chromatin assembly, naked DNA is also a good substrate for nucleases and ligases. Efficient chromatin assembly can be achieved in vitro using replication-dependent chromatin assembly (25, 26). Care is therefore required if naked DNA is to be used in this system. 176

9: Chromosome assembly in vitro using Xenopus egg extracts Protocol 4. Chromosome assembly and isolation Equipment and reagents • 10 x XBE3: 10 x XBE2 (see Protocol 2 for 1 • Prechilled 0.65 ml polypropylene tubes, x XBE2), but with 30 mM MgCI2 instead of prachilled 1.6 ml polypropylene conical 20 mM MgCI2. Prepare from stock solutubes, and 11 x 34 mm polypropylene tions. Adjust to pH 7.7. tubes • 10 x XB: 10 x XB (see Protocol 2 for 1 x . Frog fix (Protocol 2) XB) • Fluorescence microscope with 40 x/0.75 • XBE2 (Protocol 2} NA dry Fluor lens or equivalent • 20 x ATP regeneration system: 20 mM ATP • Wide-bore pipette tips (Sigma), 20 mM MgCI2, 200 mM phospho. 30% sucrose/XBE2 (w/v) creatine (Boehringer Mannheim), 1 mg/ml . HB-4 rotor (Sorvall or equivalent) fitted with creatme kinase (Boehringer Mannheim); an adaptor to take 1.6 ml tubes store in 10 and 50 ml ahquots at-80°C. . . • TLS-55 rotor (Beckman or equivalent) • Refrigerated microcentrifuge . 18 °C water-bath

Method 1. Prepare the dilution buffers for the extracts. For CSF extracts (Protocol 2), use the 10 X XBE3 and 20 x ATP regeneration stocks to make a volume of buffer sufficient to dilute enough extract for an experiment, e.g.: 20 ml 10 x XBE3, 20 ml 20 x ATP regeneration system, 160mlH2O. This solution is XBE5 and contains 2 mM ATP, 20 mM phosphocreatine, 0.1 mg/ml creatine kinase, thus the concentration of the ATP regeneration system is 2 x. After diluting the extracts 1:1 (step 2), the final concentration is 1 x. For interphase extracts (Protocol 3), use 10 x XB for the buffer stock solution instead of 10 x XBE3. 2. Quickly thaw aliquot(s) of the extract in a water-bath at room temperature. If necessary (e.g. for larger scale biochemical preps) combine aliquots. Dilute 1:1 with XBE5 (step 1). Mix by pipetting up and down, but do not introduce air bubbles. 3. Centrifuge in refrigerated microcentrifuge at 12000 g for 10 min at 4°C. Carefully remove the supernatant and place in new, prechilled 0.65 ml polypropylene tube. Unstable pellets are indicative of poor cytosol.8 4. Add nuclei or DNA. The ratio of nuclei to extract volume that produces the best chromosomes varies with the source of nuclei. For morphological assays, dilute a stock of demembranated sperm nuclei (typically 3 x 107 nuclei/ml) 1:30 in XBE2 (1 X 106 nuclei/ml). Mix gently by flicking the tube with a finger (excessive pipetting will shear nuclei). Add 1 ml of this suspension to 20 ml diluted extract. For protein analysis, use 25 ml 3 x 107 nuclei/ml for 200 ml of diluted extract. One can also use naked DNA substrates for chromatin assembly, use no more than 10 (mg DNA/ml extract. If using lambda DNA, decatenate before adding to the extract by heating to 65°C for 10 min, then immediately chill on ice.

177

Jason R. Swedlow Protocol 4.

Continued

5. Incubate in a 18°C water-bath. The incubation temperature is not absolutely critical, but if the temperature in the lab. varies by more than 3°C, use the water-bath. 6. To examine the morphology of chromosomes during or after incubation, open the tube, stir with a pipette tip (but no up-and-down pipetting), and remove a 2 ml sample. Place this on a clean slide, overlay with 4 ml frog fix and squash with a 22 x 22 mm coverslip. (For quick viewing, coverslips need not be sealed. For extended storage, seal with nail polish and store at -20°C). Observe under UV on a fluorescence microscope; we generally use a 40 x /0.75 NA dry Fluor lens for viewing these samples. 7. To isolate the chromosomes generated from nuclei for biochemical analysis, chill the assembly reaction for 15 min on ice. Using a widebore pipette tip, layer the chilled assembly reaction on 800 ul of 30% sucrose/XBE2 in a prechilled polypropylene 1.6 ml conical tube. Centrifuge at 10000 g for 15 minutes at 4°C in an HB-4 rotor fitted with an adaptor for 1.6 ml tubes. For analysing chromatin assembled from naked DNA, layer the chilled reaction onto 1 ml of 30% sucrose/XBE2 in an 11 X 34 mm polypropylene tube and centrifuge at 137088 g for 30 min at 4°C in a TLS-55 rotor. 8. Remove the tubes from the centrifuge and gently aspirate the cytosol from the surface of the cushion. Wash the top of the cushion 5 times with XBE2, then aspirate the remaining cushion. The chromatin pellet is invisible, but adheres well to the surface of the tube. Process the chromatin as appropriate for analysis. If the chromatin will be denatured in small volumes (e.g. for electrophoresis), add the appropriate sample buffer and quickly freeze the pellet in liquid nitrogen to shear the DNA before subsequent handling. a This step is absolutely necessary for protein analysis, but unnecessary for morphological analysis.

5. Immunofluorescence of in vitro assembled chromosomes Immunofluorescence is commonly used to determine the distribution of a protein within the cell. Localization of proteins to in vitro assembled chromatin and chromosomes is also possible and specific sublocalizations have been detected (13,14). However, it is important to note that, in this case, the localization of proteins is determined after centrifuging chromosomes onto a coverslip, slightly flattening the chromosomes. Furthermore, localization is 178

9: Chromosome assembly in vitro using Xenopus egg extracts determined in a sample containing only chromosomes—there are no other cellular structures available to provide an internal control for the specificity of localization. Therefore, it is very important to perform specificity controls (preimmune sera, blockade of primary antibody with antigen) to provide convincing evidence of antibody specificity. It is also important to vary the fixation conditions (concentration of fixative, length of fixation) to determine the optimal fixation protocol for a given antigen with isolated chromosomes. The fixative concentration and incubation given below should be a good starting point. Protocol 5. Immunofluorescence of in vitro assembled chromatin and chromosomes Equipment and reagents • Whatman 3MM paper • 1 mg/ml poly-L-lysine (M, >300000, Sigma) • Coverslip spin-down tubes: modified 15 ml Corex tube (Aladdin or equivalent) that has a glass plug with a flat surface glued into the bottom of the tube. A cylindrical glass chock with a flat top surface and a groove running along one side and continuing across the bottom surface is dropped onto the glass plug and serves as a support for the coverslip The coverslip is removed by slipping a wire hook into the groove in the chock and pulling it out of the bottom of the tube.

• 18 mm diameter circular coverslips, acidwashed and coated with polylysine • XBE2 (Protocol 2) • 30% glycerol in XBE2 (v/v) . 2% CH2O in XBE2: use high-quality formaldehyde for chromosome fixation. Good results have been obtained using methanol-free 16% CH2O (Ted Pella). . wide.bore pipette tips . ' HB'4 or HS-4rotor (or equivalent) • Humid chamber • Wash buffer: 10 mM Tris-HCI pH 7.5, 0.15 M NaCI, 0.1% Triton X-100

Method 1. Prepare coverslips. Put a large number of circular 18 mm diameter, #1 1/2 thickness coverslips into 1 M HCI. Heat to 60-65°C for 16 h. Wash the coverslips extensively in distilled H2O, then in deionized H20. Dry the coverslips separately on Whatman 3MM paper. Coat the coverslips by gently swirling in —10 ml 1 mg/ml poly-L-lysine for 30 min at room temperature. Remove the coverslips from the poly-L-lysine solution (it can be reused up to 4 times if stored -20°C) and wash them 5 times with deionized H2O. Dry the coverslips separately on Whatman paper. Store in a clean, covered plastic dish. Do at least 100 coverslips at a time to save time and effort. 2. Assemble chromatin or chromosomes as described in Protocol 4. 3. At the end of the assembly incubation, gently pipette 400 ml of 2% CH2O/XBE2 into the reaction tube. Immediately, and very gently, mix the reaction by pipetting up and down once through a wide-bore pipette tip. This prevents the chromosomes from clumping together, but it must be done gently to prevent breakage of the chromosomes due to shear. Incubate at room temperature for 10 min.

179

Jason R. Swedlow Protocol 5.

Continued

4. Layer the fixed chromosomes on an ice-cold, 5 ml 30% glycerol/XBE2 cushion in a coverslip spin-down tube containing a circular 18 mm poly-L-lysine coated coverslip. Centrifuge the chromosomes onto the coverslip at 6000 g for 10 min in an HB-4 or HS-4 rotor (or equivalent). 5. Aspirate to remove the cushion and recover the coverslip by pulling the chock out of the tube with a wire hook. Remove the coverslip from the top surface of the chock with fine forceps and place in a humid chamber. Wash with 10 mM Tris pH 7.5, 0.15 M NaCI, 0.1% Triton X100. 6. Process for immunofluorescence using appropriate antibodies (see Chapters 5 and 6).

6. Functional analysis of the role of specific proteins in chromatin and chromosome structure by immunodepletion The Xenopus in vitro assembly system provides the ability to analyse the function of specific factors on chromatin and chromosome assembly and structure. The roles of different proteins in chromosome condensation can be assessed by depleting, individually or in combination, each protein from the clarified cytosol using specific antibodies, thus determining the degree of condensation that occurs in the absence of that component (Protocol 6). The specificity of the observed phenotype is confirmed by adding purified or expressed proteins back to the depleted cytoplasm. A number of studies have used this technique to study the role of various individual proteins and protein complexes in chromatin and chromosome assembly (14, 17, 27, 28). Addition of antibody has also been used to block the function of a cytosolic protein (15). It is important to remember that the ultimate fate of the antibody-antigen complex is often not clear in such an experiment. In addition, antibodies are multivalent and can effectively cross-link antigens, so antibody addition experiments should be interpreted with caution. Protocol 6. Immunodepletion of specific antigens from clarified cytosol Equipment and reagents Clarified cytosol Tube rotator Affi-Prep Protein A beads (Bio-Rad) 0.6 ml polypropylene tubes TBS: 10 mM Tris-HCI pH 7.5, 0.15 M NaCI

• Rabbit IgG (Jackson ImmunoResearch, species as appropriate) . TBSTw: 10 mM Tris-HCI pH 7.5, 0.15 M NaCI, 0.05% Tween-20 • XBE2 (Protocol 2}

180

9: Chromosome assembly in vitro usingXenopus egg extracts Method 1. Perform all steps on ice or in cold room. Take 25 ml of the Affi-Prep bead slurry and place in a 0.6 ml polypropylene tube. Wash the beads 3 times with 0.5 ml TBSTw. For each wash, add buffer, mix, let the beads settle, and aspirate the supernatant. 2. Bring up beads in 500 ml TBSTw and add an appropriate amount of antibody or control IgG. Incubate in the cold room, rotating the tube end over end, for 1 h. The beads must not settle, but should stay suspended in the buffer. 3. Wash the beads 3 times with TBSTw, 3 times with XBE2. Briefly centrifuge the beads to pack them tightly. Aspirate all excess buffer. 4. Add 100 ml of clarified cytosol. Rotate the rube end over end in the cold room for 1 h.a 5. Pellet the beads in a microcentrifuge at 4°C. Remove the supernatant and use for assembly reactions. Take a sample of the depleted extract for immunoblotting to test the degree of depletion. 6. Wash the beads 3 times with TBS (the beads can be washed with TBSTw, but this is generally unnecessary). Process the beads for SDSPAGE to determine the specificity of the imuunodepletion. In ideal cases, only those proteins recognized by the antibody or antibodies will be bound to the beads. "To minimize the effect on the extract, I have kept the tube containing cytosol and beads on ice and simply used a fine pipette tip (a gel loading tip works well) to gently resuspend the beads in the extract every 5 min. Quantitative immunodepletions have been obtained using this method and control IgG-depleted extracts produce good chromosomes.

Antibody specificity is important for unequivocal results in immunodepletions. In general, antibodies should be affinity purified. Control immunodepletions using random IgG should be run in parallel with the antibody being tested. Since the physical manipulations required for immunodepletion can affect the activity of the extract, disturbance to the extract should be minimized. The amount of antibody necessary for complete immunodepletion depends on the quality of antibody and amount of antigen. Trial depletions should be run, varying the amounts of antibodies and examining the amount of antigen remaining after immunodepletion by immunoblotting.

Acknowledgements Thanks are due to Inke S. Nathke for critically reading the manuscript. The protocols for making chromosome assembly extracts and chromosomes in vitro were originally developed by Tatsuya Hirano, and I am indebted for his 181

Jason R. Swedlow help in learning these techniques. Arshad Desai developed the fixation protocol for in-vitro assembled chromosomes. I am grateful for the support of Dr Timothy J. Mitchison, in whose laboratory this work was performed. I am supported by a Fellowship from the Damon Runyon-Walter Winchell Foundation Cancer Research Fund.

References 1. Kay, B.K. and Peng, H.B. (ed.) (1991). Xenopus laevis: practical uses in cell and molecular biology, Academic Press, San Diego, CA. 2. Koshland, D. and Strunnikov, A.V. (1996). Annu. Rev. Cell Dev. Biol., 12, 305. 3. Masui, Y. (1967). J. Exp. Zool., 166,365. 4. Sagata, N., Watanabe, N., Vande Woude, G.F., and Ikawa, Y. (1989). Nature, 342, 512. 5. Lohka, M.J. and Mailer, J.L. (1985). J. Cell Biol., 101, 518. 6. Sawin, K.E. and Mitchison, T.J. (1991). /. Cell Biol., 112, 925. 7. Blow, J.J. and Laskey, R.A. (1986). Cell, 47,577. 8. Hutchison, C.J., Cox, R., Drepaul, R.S., Gomperts, M., and Ford, C.C. (1987). EMBOJ., 6, 2003. 9. Newmeyer, D.D., Finlay, D.R., and Forbes, D.J. (1986). J. Cell Biol., 103, 2091. 10. Lohka, M.J. and Masui, Y. (1983). Science, 220, 719. 11. Newport, J. (1987). Cell, 48, 205. 12. Murray, A.W. and Kirschner, M.W. (1989). Nature, 339, 275. 13. Hirano, T. and Mitchison, T.J. (1991). J. Cell Biol, 115,1479. 14. Hirano, T. and Mitchison, T.J. (1993). /. Cell Biol., 120, 601. 15. Hirano, T. and Mitchison, T.J. (1994). Cell, 79, 449. 16. Sagata, N. (1997). Bioessays, 19, 13. 17. Adachi, Y., Luke, M., and Laemmli, U.K. (1991). Cell, 64,137. 18. Desai, A., Deacon, H.W., Walczak, C.E., and Mitchison, T.J. (1997). Proc. Natl Acad. Sci. USA, 94,12378. 19. Murray, A.W. (1991). In Methods in cell biology, Vol. 36 (eds B. K. Kay and H. B. Peng), p. 581. Academic Press, San Diego, CA. 20. Wu, M. and Gerhart, J.C. (1980). Dev. Biol., 79, 465. 21. Murray, A.W., Solomon, M.J., and Kirschner, M.W. (1989). Nature, 339, 280. 22. Strick, R. and Laemmli, U.K. (1995). Cell, 83,1137. 23. Rodriguez-Campos, A., Shimamura, A., and Worcel, A. (1989). J. Mol. Biol., 209, 135. 24. Shimamura, A., Tremethick, D., and Worcel, A. (1988). Mol. Cell. Biol., 8, 4257. 25. Almouzni, G. and Mechali, M. (1988). EMBO J., 7, 665. 26. Wolffe, A.P. and Schild, C. (1991). In Methods in cell biology, Vol. 36 (eds B. K. Kay and H. B. Peng), p. 541. Academic Press, San Diego, CA . 27. Philpott, A., Leno, G.H., and Laskey, R.A. (1991). Cell, 65,569. 28. Hirano, T., Kobayashi, R., and Hirano, M. (1997). Cell, 89, 511.

182

10 Chromosome fragmentation in vertebrate cell lines CHRISTINE J. FARR

1. Introduction Experimental approaches to understanding chromosome structure and function fall into two main categories: the first involves assembling candidate DNA elements in vitro to construct an artificial chromosome. This approach was used to create yeast artificial chromosomes in both Saccharomyces cerevisiae and Schizosaccharomyces pombe. More recently a related approach has lead to the generation of mammalian minichromosomes in the human somatic cell line HT1080, although the precise DNA organization and requirements of these mammalian minichromosomes has yet to be fully elucidated (1). The alternative approach involves manipulation of an existing chromosome to generate smaller minichromosome derivatives suitable for further study/fragmentation. This chapter will focus on the use of short stretches of the vertebrate terminal repeat DNA sequence (TTAGGG)n to 'seed' the formation of de novo telomeres when introduced into vertebrate cell lines, simultaneously truncating the chromosome at the integration site (2, 3). Combined with elaborate selection schemes and homologous recombination this approach has led to the generation of a series of minichromosomes derived from native human chromosomes (4,5).

2. Telomere-associated chromosome fragmentation When plasmid DNA is introduced into established mammalian cell lines the DNA may become stably integrated into the genome as the result of nonhomologous recombination. If the introduced DNA is a linear molecule capped at one end by a correctly orientated vertebrate telomere repeat array (i.e. with the G-rich strand running 5'-3' towards the end of the linear DNA molecule) then, as in yeast, the introduced DNA is capable of generating new chromosome ends at previously interstitial sites in the recipient genome. The mechanism by which (TTAGGG)n repeat arrays seed new telomeres appears to involve their recognition and extension by telomerase. A detailed analysis

Christine J. Farr Table 1. Frequency of telomere seeding in vertebrate cell lines Cell line

Organism and tissue of origin

Frequency of seeding

HeLa

Human cervical carcinoma (epithelial)

36-55 45-69

EC27C4

Human teratocarcinoma

63

HT1080

Human fibrosarcoma

55 23-69

Reference

Palin and Farr, unpublished

10-FT

Human primary fibroblasts

Not detected (0/54)

3

ES EFC-I

Mouse embryonic stem-cell line

36

3

C3H10T1/2

Mouse embryo fibroblast (pluripotent, non-transformed)

2 (1/52)

3

A9

Mouse L cell line (fibroblast)

6(1/17)

Palin and Farr, unpublished

RAG

Mouse renal adenocarcinoma (epithelial)

Not detected (0/11)

Palin and Farr, unpublished

Wg3H

Chinese hamster lung cell line (fibroblast)

20-70

2; Palin and Farr unpublished

DT40

Chicken ALV-transformed pre-Bcell

>90

Farr, unpublished

The assay used involves the transfection of cells with linearized (TTAGGG)n-constructs based on various selectable markers. Genomic DNAs have been analysed by Southern blotting and BAL 31 nuclease digestion. Percentage estimates are derived from the number of telomeric integration events identified.

of the cis-acting requirements for telomere formation in the HeLa cell line by de Lange and colleagues demonstrated that the process is a highly sequencespecific interaction, with heterologous telomeric DNAs (such as (TTAGGC)n and (TTTAGGG)n) seeding new telomeres only very inefficiently, if at all (6). Another crucial aspect affecting the frequency with which new chromosome ends are created is the recipient cell line, and although a variety of established vertebrate cell lines will seed new telomeres efficiently in this assay, some do not (see Table 1). The factors affecting telomere seeding have not been fully established, but may include telomerase activity, the presence of other telomere-associated proteins, levels of nuclease activity, and the cell's DNA damage checkpoints and repair pathways. Although following random integration many of the telomere-seeding events observed by Barnett et al. in the HT1080 line and mouse embryonic stem-cell line EFC-1 appeared to occur in the vicinity of existing telomeres (3), the application of powerful selection schemes and targeting of telomere-seeding events to specific chromosomal loci by homologous recombination has resulted in the removal of 184

10: Chromosome fragmentation in vertebrate cell lines whole arms from non-essential chromosomes (usually human chromosomes maintained on a rodent background) (7, 8). These observations underlie the use of the (TTAGGG)n repeat array as a tool for dissecting mammalian chromosomes and creating reagents for studies into chromosome structure and function relationships.

3. Experimentally induced de novo telomere formation 3.1 Design of the telomere-seeding construct A typical construct for de novo telomere formation is shown in Figure la. The critical components are: • a dominant selectable marker for use in mammalian cell lines (such as the bacterial hygromycin B kinase gene placed under the SV40 early promoter); • a few hundred base pairs (bp) of (TTAGGG)n DNA (the lower limit has not been established and may vary depending on the recipient cell line; in general, arrays of around 500-1000 bp have been used); and • a unique restriction enzyme site immediately distal to the repeat array. This allows the plasmid to be linearized prior to transfection. It appears that for optimal telomere seeding this restriction enzyme site should lie no more than 100 bp distal to the start of the terminal repeat array, although this may be partly dependent on the recipient cell line and mode of transfection. Long stretches (more than a few hundred bp) of non-terminal repeat DNA distal to the (TTAGGG)n array appear to diminish seeding efficiencies. In early versions of telomere-seeding constructs additional 'tag' DNA sequences were included. These were 5-10 kb stretches of human genomic DNA designed simply to increase the total size of the introduced DNA molecule in order to allow direct visualization of the integration sites by fluorescence in situ hybridization (FISH) (2). Today, however, with improved tyramide signal-amplification systems, the inclusion of additional DNA sequences is no longer essential (9). If there is a requirement to characterize the endogenous DNA at the integration site this can be achieved by plasmid-rescue of the flanking DNA. This approach requires the plasmid backbone (i.e. the bacterial antibiotic resistance marker and origin of replication) to be positioned at the opposite end of the ingoing linear DNA molecule from the (TTAGGG)n repeat array (as shown in Figure la). At least one (and ideally a choice of) restriction enzyme sites should be available for releasing the plasmid DNA and the sequences immediately adjacent (see the EcoR1 site in pHyTMl). This DNA flank can then be recovered by recircularizing the released genomic DNA fragment and transforming it into a bacterial host, such as DH5a. 185

BamHITRF >8.9 kb Figure 1. Telomere seeding constructs, (a) A (TTAGGG)n-construct, pHyTMl, designed to integrate and seed new telomeres randomly in the recipient genome. It is based on a pSV2 plasmid backbone into which has been inserted 1.6 kb of the vertebrate terminal repeat sequence (TTAGGG)n, a dominant selectable marker and 4.5 kb of 'tag' DNA sequences. A unique Not1 site immediately distal to the (TTAGGG)n repeat array has been used to linearize the plasmid. The DNA probe suitable for detecting the SamH1 terminal restriction fragment (TRF) is indicated, (b) A generic targeting-breakage construct designed to seed a new telomere at a specific chromosomal locus. The positions of the endogenous centromere and telomere are indicated. The predicted product of homologous recombination between the construct and the target locus is shown.

3.2 Transfection of the telomere-seeding construct The linearized (TTAGGG)n-construct can be introduced into the recipient cell line using a variety of transfection techniques. The most commonly used are lipofection and electroporation. Electroporation can be used with virtually any cell type and is particularly good at generating cell lines which have integrated only one, or a few, copies of the ingoing plasmid DNA; this technique is discussed in more detail below. Electroporation uses an electric field to open up pores in the cell membrane through which the DNA can diffuse into the cell. Optimal electroporation conditions have to be established empirically for each cell line, varying in turn the voltage and capacitance used. Detailed electroporation protocols have been collated by Bio-Rad. In general for human cells, conditions which result in a long time constant (e.g. capacitance 960 mF and voltage 550 V giving a time constant of about 10 msec) are selected, while for rodent cells conditions resulting in a very short time pulse seem optimal (e.g. capacitance 25 mF and voltage 1.2 kV giving a time constant of 0.3 msec) (Protocol 1). For the avian DT40 cell line electroporation conditions of 25 mF and 550 V are effective. Protocol 1. Electroporation of an established vertebrate cell line with a telomere-seeding construct Equipment and reagents • (TTAGGG)n-construct • 175 cm3 tissue culture flask

Plasmid DNA • Haemocytometer

186

10: Chromosome fragmentation in vertebrate cell lines • Appropriate restriction enzyme and buffer as recommended by manufacturer • 0.4 mm electroporation cuvettes (e.g. BioRad) • TE buffer pH 7.4: 10 mM Tns-HCI pH 7.4, 1 mM EDTA pH 8.0 • Multiwell dishes

• Capacitance discharge device/ electroporater (e.g. Bio-Rad Gene Pulsar) • Phosphate-buffered saline without Mg2+ or Ca2+ (PBSA) pH 7.2 • Non-selective tissue culture medium . Appropriate selection reagents

Method 1. Linearize the (TTAGGG)n-construct with the appropriate restriction enzyme (allow 20-30 jj-g of plasmid DNA per ml of recipient cells). Heat (or otherwise) inactivate the restriction enzyme. Precipitate the linear DNA and resuspend at 1 M-g/n-l in sterile TE buffer. 2. Harvest exponentially growing recipient cells for transfection (replate adherent cells, or refeed suspension cells, 24-48 h prior to transfection. Estimate the cell number (e.g. using a haemocytometer). 3. Wash the cells once with PBSA then resuspend in PBSA at an appropriate cell concentration. In general, for electroporation, the cell concentration should be between 1 x 106/ml and 2 x 107/ml depending on the transfectability and plating efficiency of the particular cell line.9 4. Aliquot 1 ml of cells into a sterile 0.4 mm electroporation cuvette and add 20-30 |xg of plasmid DNA. Mix gently and leave at room temperature for 5 min. 5. Subject the DNA/cell suspension to an electric pulse using a capacitance discharge device. 6. After a further 5 min at room temperature transfer the cells to 25 ml of non-selective medium in a 175 cm3 tissue culture flask and incubate at 37°Cfor18-24h. 7. Add selection to the cells the next day, e.g. for isolating pSV2 Hygro (Figure 1). CHO transfectants use a hygromycin B concentration of 500 jig l^r1 (equivalent to 500 u,/ml,Calbiochem). For adherent cells the selective medium is added directly to the flask. For cell lines which grow in suspension the transfected cells should be pelleted (1000 grfor 5 min), resuspended in selective medium, and plated into multiwell dishes, (e.g. for DT40 cells 2 x 107 transfected cells should be divided among 4 x 96-well microtitre dishes). For adherent cells dead cells and cellular debris can be removed after 3-4 days and fresh selective medium applied. Drug-resistant colonies should be visible after 2-3 weeks. 8. Pick off colonies and expand into cell lines. Suspension colonies should be transferred initially to 24-well plates. Adherent colonies can be picked from the 175 cm3 flask using bent Pasteur pipettes and transferred into 25 cm3 flasks. "For example, for Chinese hamster ovary-derived cell lines (CHO) the number of stable transfectants generated is in the range of 102to 103per 10" cells electroporated; while the numbers generated for the human cell line HT1080 are significantly less (10-100 per 106).

187

Christine J. Fair

3.3 Screening stable transfectants for de novo telomere formation In brief, the criteria for de novo telomere formation by the introduced construct are the following: • A heterogeneous terminal restriction fragment (TRF) must be detected on a Southern blot. • For mapping purposes the telomere can be regarded as a universal restriction enzyme site. Therefore in any cell line in which the construct DNA is providing a functional telomere a restriction analysis can be used to detect a series of heterogeneous DNA smears of increasing size. • Terminal DNA sequences must be sensitive to BAL 31 nuclease (Protocol 2). In contrast, a chromosome-internal sequence will be resistant to such degradation. • A terminal signal must be detected when the construct DNA is used as a probe for in situ hybridization on metaphase chromosomes. 3.3.1 Molecular analysis The terminal restriction fragments of a chromosome are heterogeneous in length. This is caused by variability in the precise amount of (TTAGGG)n present at any particular chromosome end across a population of cells. Therefore, as an initial screen for de novo telomere formation, 20-30 stable transfectants are picked and expanded into cell lines. Genomic DNA is extracted from these cell lines using a standard procedure and is analysed on a Southern blot. A 20 ug aliquot of genomic DNA should be digested for 6-16 h with a restriction enzyme that will define a TRF if the construct has seeded a new chromosome end (e.g. in the plasmid illustrated in Figure la the BamHl site would be suitable). The digested genomic DNA is size-fractionated on a 0.5% agarose gel, transferred to a nylon membrane, and hybridized with a diagnostic probe (Figure la). Transfectants in which the construct has integrated interstitially without seeding a new chromosome end will display one, or a few (if more than one copy of the plasmid has become incorporated into the genome) discrete hybridization bands of variable size (due to the random nature of the integration events). The ease with which a TRF can be detected varies depending on the quality of the blot and the particular probe DNA used. With probes giving a higher level of background hybridization, the diffuse signal associated with a TRF can be difficult to detect. An example of the hybridization signals generated by a series of CHO-derived cell lines transfected with the pHyTMl telomere-seeding construct is shown in Figure 2a. Examples of both diffuse hybridization smears, characteristic of TRFs, and of discrete bands characteristic of interstitial integration events are illustrated. Note that both events are sometimes present within the same clonal cell line. 188

10: Chromosome fragmentation in vertebrate cell lines Although a heterogeneous hybridization signal is characteristic of TRFs derived from a functional telomere, the definitive test for the terminality of any DNA is its sensitivity to digestion by the nuclease BAL 31 as described in Protocol 2. BAL 31 is predominantly a 3' exonuclease that removes mononucleotides from the 3' termini of linear DNA molecules. It also has an endonuclease activity which removes the single-stranded DNA generated by the 3' exonuclease activity. Thus BAL 31 degrades both the 5' and 3' termini resulting in a controlled shortening of the DNA, and so a fragment of DNA present at a naturally occurring DNA end (a telomere) will be reduced in size before a more internal one. Degradation is absolutely dependent on the presence of Ca2+, and so the reaction can be stopped at different stages by the addition of the chelating agent EGTA. Protocol 2.

BAL 31 sensitivity assay

Equipment and reagents • 30°C water-bath • 5 x BAL 31 reaction buffer 100 mM Tris-HCI pH 8.0, 50 mM CaCI2, 50 mM MgClz, 3 M NaCI, 5 mM EDTA pH 8.0 . 0.1 M EGTA

• BAL 31 nuclease (New England BioLabs) • phenol:chloroform:octan-2-ol (25:24:1 v/v) . chloroform:octan-2-ol 24:1 v/v . TE buffer (Protocol 1)

Method 1. Aliquot out 200 ug of intact genomic DNA from cell lines which appear, from the initial Southern analysis, to be potential single-copy integrants in which the construct provides a functional telomere. This provides sufficient DNA for the analysis of five time points.a 2. Make up to 500 ul in 1 x BAL 31 reaction buffer. Preincubate the DNA at 30°C for 2 h in an uncovered water-bath. 3. Prelabel a series of Eppendorf tubes with various time points; a suitable range would be 0, 10, 20, 40, and 80 min exposure to BAL 31 nuclease activity. Aliquot 20 ul 0.1 M EGTA per Eppendorf tube (this will give a final concentration of between 15 and 20 mM with the addition of 100 ul DNA BAL 31 nuclease solution). 4. Add 20 U BAL 31 nuclease to the genomic DNA; mix by inversion and incubate the open tube at 30°C. Immediately remove 100 u1 and stop the reaction by adding EGTA. 5. Continue the incubation and remove further samples at the appropriate time intervals. 6. When the incubation step is complete remove the nuclease by a phenol/ chloroform:octan-2-ol extraction followed by a chloroform:octan-2ol extraction step. Ethanol-precipitate the DNA and resuspend it in sterile TE buffer. 189

Christine J. Farr Protocol 2. Continued 7. Digest the nuclease-treated DNA with a restriction endonuclease suitable for TRF identification and analyse the DNA by Southern blotting and hybridization.b a BAL 31 nuclease is sensitive to RNA contamination, so 'good' quality DNA preparations are required. b Some restriction enzymes are inhibitad by EGTA.

An example of BAL 31 digestion of DNA from a cell line with a functioning (TTAGGG)n construct is illustrated in Figure 2b. BAL 31 degrades AT-rich DNA sequences significantly more rapidly than it degrades GC-rich DNA. This may be responsible for the end products of some BAL 31 digestions appearing as a discrete hybridi/ation band at late time points (see Figure 2b). 3.3.2 Cytogenetic analysis Finally, the location of the exogenous DNA can be examined cytogenetically by fluorescence in situ hybridization using the introduced telomere- see ding

Figure 2. Molecular analysis of transfectants. (a) Southern blot analysis of a series of CHO-derived cell lines transfected with Not1-linearized pHyTM1, Genomic DNAs have been digested with BamH1 and probed with the DNA fragment indicated in Figure 1a. (b) BAL 31 sensitivity of DMA from one transfectant in which the (TTAGGG)n-construct has seeded a new telomere.

190

10: Chromosome fragmentation in vertebrate cell lines construct, minus the (TTAGGG)n repeat array itself, as the probe. Appropriate cytogenetic techniques for the preparation of fixed metaphase chromosomes from Colcemid-blocked cells are described in Chapter 4. The exogenous DNA is detected by FISH using the (TTAGGG)n-containing construct as the probe, minus the terminal repeat (since this would result in the probe DNA hybridizing to all endogenous chromosome ends). It should be noted that even if transfectants are scored cytogenetically as having integrated the construct at apparently terminal locations this is not definitive, and the molecular analyses described in Section 3.3.1 must also be performed. A region of DNA may appear cytogenetically to be terminal, but in fact may lie several megabases away from the actual chromosome end; conversely, DNA shown by the molecular analyses to be at the very end of the chromosome and sensitive to the action of BAL 31 nuclease may appear subtelomeric on metaphase chromosomes. The reason for the latter observation is unclear, but it has been speculated that it may reflect loops of DNA extending out and beyond the telomeres, resulting in the appearance of DNA beyond the site of hybridization. Visualization of the integration site by FISH therefore provides supportive data, but is not in itself proof of terminality.

4. Targeted de novo telomere formation The (TTAGGG)n-construct shown in Figure la is designed to seed telomeres randomly in the recipient genome. Two strategies have been used to obtain truncation events on a particular chromosome: • the use of a randomly integrating (TTAGGG)n-construct combined with a powerful selection scheme to enrich for truncation events of interest; and • the use of homologous recombination to target the (TTAGGG)n-construct to a particular chromosomal location. The first approach is reliant on the presence of two selectable markers on the chromosome of interest: a negative or bidirectional selectable marker (e.g. HSV-Tk, or HPRT) distally located on the chromosome arm to be truncated; and a dominant selectable marker (e.g. hygror, neor, or hisDr) located on the segment of chromosome which will be retained. Using this 'PushmiPullyu' counterselection system a panel of somatic cell hybrids, in which the long arm of the human X chromosome is truncated to various extents, has been created (7). Selection against the human HPRT gene at Xq26 was used to isolate cells in which the introduced hygromycin-based (TTAGGG)nconstruct had replaced the long-arm telomere. Out of 85 cell lines isolated, 42 displayed apparently random terminal truncations of Xq, extending from qter to the a-satellite array at the centromere. Presumably, the stringent biochemical selection permits the isolation of the desired truncation events from a much larger number of truncations generated by the random integration of 191

Christine J. Farr the (TTAGGG)n-construct. Although a powerful approach, its application is strictly limited due to the absolute requirement for strategically placed selectable markers. Since there are relatively few endogenous biochemically selectable markers in the mammalian genome, this necessitates retrofitting the chromosome of interest using homologous recombination. The feasibility of creating defined truncation events was first demonstrated by Itzhaki et al. who seeded a de novo telomere within the interferoninducible 6-16 gene located on the distal tip of chromosome Ip in the HT1080 cell line (10). In addition to targeting unique loci, (TTAGGG)n-constructs have also been targeted to the repetitive a-satellite DNA at the centromere of the human Y chromosome, resulting in the creation of both long-arm and short-arm derivatives (8). In both instances, the basic (TTAGGG)n-construct is modified to include a stretch of genomic DNA homologous to the locus being targeted. When linearized, the targeting (TTAGGG)M-construct has the vertebrate terminal repeat array at one end and a stretch of DNA homologous to the locus of interest at the other (Figure Ib). This approach requires either that the orientation of the locus being targeted relative to the centromere and telomere is known (so that the (TTAGGG)n-array can be orientated correctly relative to it) or, alternatively, the use of two targeting constructs with the region of homology in both possible orientations. Targeted chromosome breakage has been used extensively in the mapping and characterization of both natural and artificial chromosomes in yeast, where the integration of exogenous DNA by homologous recombination is very efficient. The major limitation to the use of this approach for the functional analysis and modification of mammalian chromosomes has been the very low levels of homologous recombination observed in mammalian somatic cell lines.

5. Targeted truncation events in the recombinationproficient avian cell line DT40 The chicken pre-B cell line DT40 is unusual for a vertebrate cell line in that it exhibits high levels of homologous recombination, the reasons for which are as yet unclear (11). Human chromosomes can be transferred into DT40 cells using microcell-mediated chromosome transfer (MMCT) and appear to segregate in a relatively stable manner. Moreover, it has recently been demonstrated that the transferred chromosomes can also be efficiently modified by homologous recombination in the DT40 hybrids (12, 13). Systematic manipulation of human chromosomes in the DT40 background can therefore be undertaken, with subsequent transfer of the modified human chromosome back into a mammalian cell line by MMCT for further analysis. To construct microcell hybrids, donor cells are subjected to a prolonged mitotic block, which induces micronucleation (partitioning of individual or a 192

10: Chromosome fragmentation in vertebrate cell lines few chromosomes into subnuclear packets). The micronuclei are extruded from the cells by centrifugation in the presence of cytochalasin B, forming microcells (micronuclei surrounded by a plasma membrane) and fused to intact recipient cells. Hybrids are selected using a metabolic or transgenic marker located on the desired chromosome. A detailed protocol for the production of DT40-human hybrids by MMCT is provided by Dieken and Fournier (14). Microcell fusion is technically demanding, temperamental, and requires proficiency in tissue culture techniques. Moreover, the transfer of human chromosomes into DT40 cells is probably an order of magnitude more difficult than fusions between mammalian donor and recipient cell lines. Success is dependent upon careful organization and monitoring of each step, and on patience and perseverance (15). Fortunately, laboratories highly proficient at microcell hybrid production have already generated several DT40-human hybrids and, although a full panel representing all human chromosomes is not yet available, several hybrids have been described in the literature (e.g. DT40-human microcell hybrids carrying chromosomes 1, 2, 3, and 11) (12, 13). The frequency of transfer of the modified human chromosome back into a mammalian cell line for further study is much higher than in the reciprocal cross. However, after each transfer step the chromosome under investigation will need to be thoroughly examined, both at the molecular and cytogenetic level, to determine whether or not it has been transferred intact, or, as frequently happens during MMCT, has undergone rearrangement. Since homologous recombination is very efficient in DT40 cells and hybrids, vector construction is straightforward, with no requirement for elaborate schemes to enrich, or screen, for targeting events. A simple targeting construct with a single block of homology at one end and screening of a small number of transfectants by Southern blotting is generally sufficient, although this may depend on the type of locus being targeted (e.g. unique or repetitive, euchromatic or heterochromatic) and on the nature of the targeted product (e.g. targeting a functional chromosomal element may affect stability, which will in turn affect the number of targeting events isolated).

6. The characterization of chromosomes modified by de novo telomere formation and fragmentation Characterization of the modified chromosome product will involve both molecular and cytogenetic analyses. FISH is essential for confirming that the integrated (TTAGGG)n-construct is in the predicted location and for determining the effect of the de novo seeding event on the overall morphology of the targeted chromosome, including whether it still exists as an unrearranged, independently segregating entity. If the target chromosome has been fragmented within the a-satellite DNA sequences, which are localized at the 193

Christine J. Farr primary constriction of all naturally occurring human centromeres, pulsedfield gel electrophoresis (PFGE) will be required to estimate the amount of a-satellite DNA remaining and its integrity compared with the parental chromosome. In addition, some measure of the functional status of the centromere can be obtained by determining the presence of centromere-binding proteins associated with activity, such as CENP-C and CENP-E, by immunofluorescence (Chapter 5).

6.1 Estimation of minichromosome size If the target chromosome has been subjected to several rounds of telomereassociated chromosome fragmentation, in order to create minichromosomes based on human centromeric sequences, the total size of the minichromosome (if it has a linear structure) can be investigated by PFGE of undigested, intact, genomic DNA (circular structures are thought not to resolve under these conditions) as in Protocol 3. To resolve linear DNA molecules in the 2-10 Mbp size range, long runs (5 days or longer) and long pulse times are required on a suitable apparatus (e.g. BioMetra's Rotaphor system). Size markers covering the Mbp range are available commercially (Bio-Rad and BioMetra); chromosomes in S. pombe are approximately 3.5, 4.6, and 5.7 Mbp in size, while the Hansenula wingei genome provides markers in the 1 to 3.1 Mbp range, overlapping with chromosomes in 5. cerevisiae (0.2-2.2 Mbp). Following electrophoresis, the DNA is transferred to a nylon membrane and hybridized with a repeat probe, such as the appropriate chromosome-specific a-satellite DNA. A DNA probe which hybridizes to a high copy-number repeat is required since only a very small proportion of the DNA molecules migrate into the agarose. Estimation of minichromosome size by PFGE is much easier if the human-derived minichromosome is in a non-human background, since the presence of other human chromosomes will give rise to some level of background hybridization from inevitable degradation arising during block preparation and electrophoresis. Protocol 3. Pulsed-field gel electrophoresis of large DNA Caution: PMSF must be treated with extreme caution. It is highly toxic and volatile and should always be handled in a fume hood. It is unstable and must be freshly prepared. Equipment and reagents • Pulsed-field block formers (Bio-Rad, BioMetra) • Water-baths at 48, 50, and 65°C • Pulsed-field gel electrophoresis apparatus (e.g. Bio-Rad CHEF DRII or BioMetra Rotaphor system) • Stratalinker UV crosslinker (Stratagene)

• Pulsed-field grade and chromosomal grade agarose (Bio-Rad) « SE: 75 mM NaCI, 25 mM EDTA, pH 8.0 • PBSA (Protocol 7) . 50 ml Falcon tubes . ESP: 0.5 M EDTA pH 8.0, 1% sodium lauryl sarcosinate, 2 mg/ml proteinase K

194

10: Chromosome fragmentation in vertebrate cell lines • 0.5 M EDTA pH 8.0 • TE buffer (Protocol 1) . Phenylmethylsulfonyl (Sigma)

fluoride

(PMSF)

• 10 mg/ml ethidium bromide • Equipment and reagents for Southern blotting, hybridization, and autoradiography

Method 1. Prepare 1% pulsed-field grade agarose in SE. Place at 65 °C and transfer to a 48°C water-bath 10-15 min before use. Clean and dry pulsed field block formers. 2. Harvest exponentially growing cells and wash once with PBSA. Check the cell number (e.g. with a haemocytometer) and resuspend in PBSA or SE at 2 x 107 cells/ml. 3. Add 1 ml of cooled agarose per ml of cell suspension. Quickly pipette up and down to mix and immediately aliquot into a block-forming mould. Leave to solidify. (If embedding several ml of the cell suspension do so in 1 ml aliquots to prevent the cell/agarose mix from solidifying during the process.) 4. Transfer the blocks to a 50 ml Falcon tube and incubate in 20 ml of ESP for 48 h at 50°C; swirl occasionally during this period. 5. For long-term storage of blocks (for up to 6-12 months) at 4°C replace the ESP with 0.5 M EDTA pH 8.0. 6. Prior to use, wash the blocks thoroughly with multiple (at least 10) changes of TE buffer. (TE supplemented with 1 mM PMSF can be used to inactivate and remove all traces of proteinase K. However, PMSF must be treated with extreme caution. It is highly toxic and volatile and should always be handled in a fume hood. It is unstable and must be freshly prepared.) 7. Load the rinsed and pre-equilibrated (in running buffer, see step 8) sample blocks and suitable size markers, into a 0.6% chromosomal grade agarose gel using a sterile spatula. Precool the gel in the buffer chamber before starting electrophoresis. 8. Run for a 5-10 day period using the conditions recommended by the manufacturer, e.g. on the Bio-Rad CHEF DRII apparatus, which has a fixed 120° angle, reasonable resolution in the 2-6 Mbp range can be obtained using: 75 V, 600-2700 sec pulse, 120 h run at 8°C in 0.5 x TBE. For the BioMetra Rotaphor system 5- and 10-day run conditions are detailed in the user manual. 9. After electrophoresis is complete stain the DNA with ethidium bromide for 30-60 min and photograph the gel. UV-treat the gel to nick the DNA (e.g. deliver 960 mJ of energy using a Stratalinker UV crosslinker). Denature, neutralize, and transfer the gel as usual for Southern blotting. 10. Hybridize the filter with a suitable DNA repeat probe and subject to autoradiography.

195

Christine J. Fan

6.2 Assays for mitotic stability of minichromosomes The assays currently available for estimating mammalian chromosome stability are limited and cumbersome. However, if properly undertaken they do give some information as to the stability of the modified chromosome compared with intact human chromosomes, in the context of a somatic cell hybrid background. To determine whether modified chromosomes and minichromosomes are mitotically stable the cell lines should be grown for prolonged periods both in the presence and absence of selection (1-6 months), with frequent passaging of the cells to ensure that the cultures are kept in an actively proliferating growth state. At monthly time-intervals some measure of stability can be obtained by plating-efficiency assays. Plate 1-2 X 103 cells per 75 cm3 flask (in triplicate) into media both with and without the selective agent. After 5-10 days stain the colonies with Crystal violet and count. This assay is only suitable for adherent cell lines and, although it reveals whether or not the selectable marker being assayed is present/absent, it does not provide any information about copy number or the physical status of the chromosome being monitored. A more informative assay is to analyse the cells by FISH, screening 100-200 metaphases at each time-point off selection and comparing this with stability on selection. The presence or absence of autonomous minichromosomes should be determined using a highly efficient probe, such as a centromeric repeat DNA. If the chromosome displays instability this can be investigated using FISH to anaphase and early telophase cells (8,16). Exponentially growing cells are assayed without prior Colcemid treatment (Protocol 4). The presence of anaphase bridges between separating daughter cells and lagging chromosomes would indicate abnormal mitotic behaviour (see also Chapter 5). Protocol 4. Anaphase analysis for adherent cells Equipment and reagents • • • . •

Sterile coverslips (e.g. see Chapter 6) PBSA (Protocol 1) EDTA MgCI2 Triton X-100

• 1 mg/ml NaBH4 in H2O . 5-10 % goat serum in PBSA • Ethylene glycol-bis(succinimidylsuccinate) (EGS) (Pierce) • 2 x SSC: 0.3 M NaHCI, 0.03 M trisodium

citrate pH 7 . Paraformaldehyde or glutaraldehyde '° . Permeabilization and fixation solution: 10 * 10° ^9/ml RNase A in 2 x SSC mM EDTA, 2 mM MgCI2, 0.1% Triton X-100, • 0.2 M NaOH 3.7% paraformaldehyde (or 0.5% glu- • Equipment and reagents for hybridization taraldehyde) in PBSA and standard FISH analysis

Method 1. Plate cells onto coverslips and grow for 24-48 h until 80% confluent. 2. Simultaneously permeabilize and fix the cells by incubating them for 30 min in the permeabilization and fixation solution.

196

10: Chromosome fragmentation in vertebrate cell lines 3. Rinse the cells on the coverslip 3 x with PBSA. 4. Incubate 2 x with 1 mg/ml NaBH4 in dH20 for 10 min. 5. Wash 2 x with PBSA and incubate for 30 min in 5-10 % goat serum in PBS A at 37 °C. 6. Wash briefly 5 x in PBSA and incubate with the cross-linking agent EGS for 30 min at 37°C (to preserve the microtubule morphology). 7. Wash briefly 5 x in PBSA and treat the coverslip with 100 jig/ml RNase A in 2 x SSC for 30-60 min at 37°C. 8. Wash 3 x in PBSA and denature the DNA in 0.2 M NaOH at room temperature for 4 min. 9. Hybridize in situ with biotinylated probe DNA and visualize with FITC-avidin detection as for standard FISH analysis.

7. Concluding remarks The physical manipulation of mammalian chromosomes and the characterization of their phenotypes demands expertise in a wide range of molecular biology, tissue culture, and cytogenetic methodologies. In this brief chapter it has only been possible to outline current developments in this field; for more detailed protocols the reader is referred to the reference list below.

References 1. Harrington, J.J., van Bokkelen, G., Mays, R.W., Gustashaw, K., and Willard, H.F. (1997). Nature Genet., 15, 345. 2. Farr, C., Fantes, J., Goodfellow, P., and Cooke, H. (1991). Proc. Natl Acad. Sci. USA, 88,7006. 3. Barnett, M.A., Buckle, V.J., Evans, E.P., Porter, A.C.G., Rout, D., Smith, A.G., and Brown, W.R.A. (1993). Nucl. Acid Res., 21,27. 4. Farr, C.J., Bayne, R.A.L., Kipling, D., Mills, W., Critcher, R., and Cooke, H.J. (1995). EMBO J., 14,5444. 5. Heller, R., Brown, K.E., Burgtorf, C., and Brown, W.R.A. (1996). Proc. Natl Acad. Sci. USA, 93,7125. 6. Hanish, J.P., Yanowitz, J.L., and de Lange, T. (1994). Proc. Natl Acad. Sci. USA, 91, 8861. 7. Farr, C.J., Stevanovic, M., Thomson, E.J., Goodfellow, P.N., and Cooke, H.J. (1992). Nature Genet., 2, 275. 8. Brown, K.E., Barnett, M.A., Burgtorf, C., Shaw, P., Buckle, V., and Brown, W.R.A. (1994). Hum. Mol. Genet, 3,1227. 9. Raap, A.K., van de Corput, M.P.C., Vervenne, R.A.W., van Gijlswijk, R.P.M., Tanke, H.J., and Wiegant, J. (1995). Hum. Mol. Genet., 4,529. 10. Itzhaki, J.E., Barnett, M.A., MacCarthy, A.B., Buckle, V.J., Brown, W.R.A., and Porter, A.C.G. (1992). Nature Genet., 2, 283. 197

Christine J. Farr 11. Buerstedde, J.-M. and Takeda, S. (1991). Cell, 67,179. 12. Dieken, E.S., Epner, E.M., Fiering, S., Fournier, R.E.K., and Groudine, M. (1996). Nature Genet., 12,174. 13. Koi, M., Lamb, P.W., Filatov, L., Feinberg, A.P., and Barrett, J.C. (1997). Cytogenet. Cell Genet., 76,72. 14. Dieken, E.S. and Fournier, R.E.K. (1996). In Methods (ed. R.E.K. Fournier), Vol. 9, p.56. Academic Press, New York. 15. Schafer, AJ. and Farr, C.J. (1998). In The ICRF handbook of genome analysis (ed. N. Spurr), Chapter 14. Blackwell Science. 16. Larin, Z., Fricker, M.D., and Tyler-Smith, C. (1994). Hum. Mol. Genet, 3, 689.

198

A1

List of suppliers Advanced Biotechnologies Ltd. Advanced Biotechnologies Ltd., Units B1-B2, Longmead Business Centre, Blenheim Road, Epsom, Surrey KT19 9QQ, UK. Advanced Biotechnologies Ltd., Wendenstrasse 23,20097 Hamburg, Germany. Aladdin Enterprises, 1255 23rd St, San Francisco, CA 94122, USA. Amersham Amersham International pic, Lincoln Place, Green End, Aylesbury, Buckinghamshire HP20 2TP, UK. Amersham Corporation, 2636 South Clearbrook Drive, Arlington Heights, IL 60005, USA. Anachem, Anachem Ltd., Anachem House, Charles Street, Luton, Beds LU2 OEB, UK Andennan and Co. Ltd., 145 London Road, Kingston-Upon-Thames, Surrey KT17 7NH, UK. K. R. Anderson Company Inc., 2800 Bowers Ave, Santa Clara, CA 95051, USA. AndPak-EMA, 1560 Dobbin Drive, San Jose, CA 95133, USA. Applied Imaging International Ltd., Hylton Park, Wessington Way, Sunderland, SR5 3HD, UK. Applied Precision Inc., 1040 12th Ave, Northwest, Issaquam, Washington 98027, USA. Applied Spectral Imaging Ltd., P.O. Box 101, Migdal Haemek, 10551, Israel. AutoQuant Imaging, 877 25th St, Watervliet, NY 12189, USA. Bayer Diagnostics pic, Bayer House, Strawberry Hill, Newbury RG14 1JA, UK. BDH BDH Chemicals UdJMerck Ltd., Merck House, Poole, Dorset BH15 1TD, UK. BDH, Merck Ltd., Hunter Boulevard, Magna Park, Latterworth, Leicester, LE17 4XN, UK. Beckman Instruments Beckman Instruments UK Ltd., Progress Road, Sands Industrial Estate, High Wycombe, Buckinghamshire HP12 4JL, UK.

List of suppliers Beckman Instruments Inc., P.O. Box 3100,2500 Harbor Boulevard, Fullerton, CA 92634, USA. Becton Dickinson Becton Dickinson and Co., Between Towns Road, Cowley, Oxford OX4 SLY, UK. Becton Dickinson and Co., 2 Bridgewater Lane, Lincoln Park, NJ 07035, USA. Bio Bio 101 Inc., do Statech Scientific Ltd., 61-63 Dudley Street, Luton, Bedfordshire LU2 OHP, UK. Bio 101 Inc., P.O. Box 2284, La Jolla, CA 92038-2284, USA. BioMetra Biometra Ltd., Whatman House, St Leonard's Road, 20/20 Maidstone, Kent ME 16 OLS, UK. Biometra Inc., 550 N. Reo St. #101, Tampa, FL 33609, USA. Bio-Rad Laboratories Bio-Rad Laboratories Ltd., Bio-Rad House, Maylands Avenue, Hemel Hempstead HP2 7TD, UK. Bio-Rad Laboratories, Division Headquarters, 3300 Regatta Boulevard, Richmond, CA 94804, USA. Biospec Products Inc., P.O. Box 722, Bartlesville, OK 74005, USA. Biovation, Crombie Lodge, Aberdeen Science and Technology Park, Balgownie Drive, Aberdeen AB22 8GU, UK Boehringer Mannheim Boehringer Mannheim UK (Diagnostics and Biochemicals) Ltd., Bell Lane, Lewes, East Sussex BN17 1LG, UK. Boehringer Mannheim Corporation, Biochemical Products, 9115 Hague Road, P.O. Box 504 Indianopolis, IN 46250-0414, USA. Boehringer Mannheim Biochemica, GmbH, Sandhofer Str. 116, Postfach 310120 D-6800 Ma 31, Germany. Branson Ultrasonic Corp., 41 Eagle Rd., Danbury, CT 06813, USA. British Drug Houses (BDH) Ltd., Poole, Dorset, UK. R. Cadisch and Sons, Arcadia Avenue, Finchley, London, N3 2JW, UK. Calbiochem Calbiochem-Novabiochem, Boulevard Industrial Park, Padge Rd, Beeston, Nottingham NG9 2JR, UK. Calbiochem, Inc., P.O. Box 12087, San Diego, CA 92112-4180, USA. Cambio, 34 Newnham Road, Cambridge, CB3 9EY, UK. Cambridge Bioscience (Molecular Probes), 24-25, Signet Court, Newmarket Road, Cambridge, CB5 8LA, UK. Chance Propper Ltd., West Midlands, B66 1NZ, UK. Chemicon Chemicon International, 2 Bonnersfield Lane, Harrow HA1 2JR, UK. Chemicon International, 28835 Single Oak Dr.,Temecula, CA 92590, USA. 200

List of suppliers Chroma Technology Corporation, 72 Cotton Mill Hill Unit A-9, Brattleboro, VT 05301, US A. Citifluo, Connaught Buildings, City University, Northampton Square, London EC1N OH13, UK. Cytocell Ltd, Somerville Court, Banbury Business Park, Adderbury, Banbury, Oxon OX17 3SN, UK Difco Laboratories Difco Laboratories Ltd., P.O. Box 14B, Central Avenue, West Molesey, Surrey KT8 2SE, UK. Difco Laboratories, P.O. Box 331058, Detroit, MI 48232-7058, USA. Du Pont Dupont (UK) Ltd., Industrial Products Division, Wedgwood Way, Stevenage, Herts, SGI 4Q, UK. Du Pont Co. (Biotechnology Systems Division), P.O. Box 80024, Wilmington, DE 19880-002, USA. Electron Microscopy Sciences, PO Box 251, 321 Morris Road, Fort Washington, PA 19034, USA. European Collection of Animal Cell Culture, Division of Biologies, PHLS Centre for Applied Microbiology and Research, Porton Down, Salisbury, Wiltshire SP4 OJG, UK. Falcon (Falcon is a registered trademark of Becton Dickinson and Co.) Fisher Scientific Co., 711 Forbest Avenue, Pittsburgh, PA 15219-4785, USA. Flow Laboratories, Woodcock Hill, Harefield Road, Rickmansworth, Hertfordshire WD3 1PQ, UK. Fluka Fluka-Chemie AG, CH-9470, Buchs, Switzerland. Fluka Chemicals Ltd., The Old Brickyard, New Road, Gillingham, Dorset SP8 4JL, UK. Gibco BRL Gibco BRL (Life Technologies Ltd.), Trident House, Renfrew Road, Paisley PA3 4EF, UK. Gibco BRL (Life Technologies Inc.), 3175 Staler Road, Grand Island, NY 14072-0068, USA. Growing Point, Toyobo Co Ltd, 2-8 Dojima Hama Z Chome, Kita-Ku, Osaka 530 8230, Japan. Heraeus Equipment Ltd., 9 Wates Way, Brentwood, Essex CM15 9TB, UK. Arnold R. Horwell, 73 Maygrove Road, West Hampstead, London NW6 2BP, UK. Hybaid Hybaid Ltd., 111-113 Waldegrave Road, Teddington, Middlesex TW11 8LL, UK. Hybaid, National Labnet Corporation, P.O. Box 841, Woodbridge, NJ 07095, USA. 201

List of suppliers HyClone Laboratories, 1725 South HyClone Road, Logan, UT 84321, USA. ICN ICN Pharmaceuticals Inc., 3300 Hyland Avenue Costa Mesa, CA 92226, USA. ICN Biochemicals Ltd., Unit 18, Wenman Rd, Thame, Oxfordshire OX9 3XA, UK. International Biotechnologies Inc., 25 Science Park, New Haven, CT 06535, USA. International Equipment Company (IEC), Needham Heights, MA 02194, USA. Improvision, Viscount Centre 11, University of Warwick Science Park, Millburn Hill Road, Coventry, CV4 7HS, UK. Invitrogen Corporation Invitrogen Corporation, 3985 B Sorrenton Valley Building, San Diego, CA 92121, USA. Invitrogen Corporation, do British Biotechnology Products Ltd., 4-10 The Quadrant, Barton Lane, Abingdon, OX14 3YS, UK. Jackson ImmunoResearch Laboratories Inc., P.O. Box 9, 872 West Baltimore Pike, West Grove, PA 19390, USA. Kodak: Eastman Fine Chemicals, 343 State Street, Rochester, NY, USA. Leica Inc. Leica Inc., Ill Deer lake Road, Deerfield, IL 60015, USA. Leica AG, CH-9435 Heerbrugg, Switzerland Leinco Technologies Inc. Leinco Technologies Inc., 14730 Manchester Rd, Ballwin, MO 63011, USA. Universal Biologicals Ltd., 30 Merton Road, London SW18 1QY, UK. Life Sciences International (Shandon UK), Unit 5, The Ringway Centre, Edison Rd, Basingstoke, Hampshire RG21 6YH, UK. Life Sciences Resources, Abberley House, Granham's Road, Great Shelford, Cambridge CB2 5LQ, UK. Life Technologies Inc., 8451 Helgerman Court, Gaithersburg, MD 20877, USA. Merck Merck Industries Inc., 5 Skyline Drive, Nawthorne, NY 10532, USA. Merck, Frankfurter Strasse, 250, Postfach 4119, D-64293, Germany. Millipore Millipore (UK) Ltd., The Boulevard, Blackmoor Lane, Watford, Hertfordshire WD1 8YW, UK. Millipore CorpJBiosearch, P.O. Box 255, 80 Ashby Road, Bedford, MA 01730, USA. Molecular Dynamics Ltd, 5 Beech House, Chiltern Court, Asheridge Road, Chesham, Bucks HP5 2PX, UK. Molecular Probes Inc. Molecular Probes Inc., P.O. Box 22010, Eugene, OR 97402-0469, USA. 202

List of suppliers Cambridge Bioscience (Molecular Probes), 24-25, Signet Court, Newmarket Road, Cambridge, CBS SLA, UK. MSE Scientific Instruments, Manor Royal, Crawley, West Sussex RH10 2QQ, UK. NBL Gene Sciences Ltd., South Nelson Industrial Estate, Cramlington, Northumberland NE23 9HL, UK. New England Biolabs (NBL) New England Biolabs (NBL), 32 Tozer Road, Beverley, MA 01915-5510, USA. New England Biolabs (NBL), c/o CP Labs Ltd., P.O. Box 22, Bishops Stortford, Hertfordshire CM23 3DH, UK. New England Nuclear Du Pont (UK) Ltd., Biotechnology Systems Division, Wedgewood Way, Stevenage, Hertfordshire SGI 4QN, UK. NEN Life Science Products, 549 Albany Street, Boston, MA 02118, USA. Nikon Corporation, Fuji Building, 2-3 Marunouchi 3-chome, Chiyoda-ku, Tokyo, Japan. Olympus America Inc., Two Corporate Center Drive, Melville, NY 11747-3157, USA. Omega Optical Inc., 3 Grove St, P.O. Box 573, Brattleboro, VT 05302-0573, USA. Oncor Appligene Inc. Appligene Oncor, Pinetree Centre, Durham Rd, Birtley, Chester-le-St, Co. Durham DH3 2TD, UK. Oncor Appligene Inc., 209 Perry Parkway, Gaithersburg, MD 20877, USA. Perkin-Elmer Perkin-Elmer Ltd., Maxwell Road, Beaconsfield, Buckinghamshire HP9 1QA, UK. Perkin Elmer Ltd., Post Office Lane, Beaconsfield, Buckinghamshire HP9 1QA, UK. Perkin Elmer-Cetus (The Perkin-Elmer Corporation), 761 Main Avenue, Norwalk, CT 0689, USA. Pharmacia Biosystems Pharmacia Biosystems Ltd., (Biotechnology Division), Davy Avenue, Knowlhill, Milton Keynes MK5 8PH, UK. Pharmacia LKB Biotechnology AB, Bjorngatan 30, S-75182 Uppsala, Sweden. Pharmacia Biotech Europe, Procordia EuroCentre, Rue de la Fuse-e 62, B-1130 Brussels, Belgium. Philip Harris Scientific, Lynn Lane, Shenstone, Lichfield, Staffordshire WS14 9BR, UK. Pierce Pierce and Warriner (UK) Ltd., 44 Upper Northgate St, Chester CHI 4EF, UK. 203

List of suppliers Pierce, 3747 N. Meridian Rd, P.O. Box 117, Rockford IL 61105, USA. Polysciences, 400 Valley Road, Wallington, PA 18976, USA. Princeton Instruments, 3660 Quakerbridge Road, Trenton, NJ, USA. Promega Promega Ltd., Delta House, Enterprise Road, Chilworth Research Centre, Southampton, UK. Promega Corporation, 2800 Woods Hollow Road, Madison, WI 53711-5399, USA. Qiagen Qiagen Inc., do Hybaid, 111-113 Waldegrave Road, Teddington, Middlesex, TW11 8LL, UK. Qiagen Inc., 9259 Eton Avenue, Chatsworth, CA 91311, USA. Sartorius AG., 37070 Gottingen, Germany. Savant Instruments Inc. 110-103 Bi-county Boulevard, Farmingdale NY11735, USA. Scanalytic Inc., 8550 Lee Highway, Suite 400, Fairfax, Virginia 22031-1515, USA. Schleicher and Schuell Schleicher and Schuell Inc., do Andermann and Company Ltd. Schleicher and Schuell Inc., Keene, NH 03431A, USA. Schleicher and Schuell Inc., D-3354 Dassel, Germany. Scientific Volume Imaging, Alexanderlaan 14, 1213 XS, Hilversum, Netherlands. Shandon Scientific Ltd., Chadwick Road, Astmoor, Runcorn, Cheshire WA7 1PR, UK. Sigma Chemical Company Sigma Chemical Company (UK), Fancy Road, Poole, Dorset BH17 7NH, UK. Sigma Chemical Company, 3050 Spruce Street, P.O. Box 14508, St Louis, MO 63178-9916, USA. Sorenson Bioscience Inc., 6507 South 400 West, Salt Lake City, UT 84107, USA. Sorvall DuPont Company, Biotechnology Division, P.O. Box 80022, Wilmington, DE 19880-0022, USA. Stratagene Stratagene Ltd., Unit 140, Cambridge Innovation Centre, Milton Road, Cambridge CB4 4FG, UK. Stratagene Inc., 11011 North Torrey Pines Road, La Jolla, CA 92037, USA. Stratech Scientific Ltd (Jackson ImmunoResearch), 61-63, Dudley St, Luton, Bedfordshire, LU2 ONP, UK. Techne Techne (Cambridge) Ltd., Duxford, Cambridge CB2 4PZ, UK. Techne Incorporated, 3700 Brunswick Pike, Princeton, New Jersey 08540-6192, USA. Ted Pella, Inc., P.O. Box 492477, Reading, CA 96049-2477, USA. 204

List of suppliers Thomas Scientific, 99 High Hill Road, PO Box 99, Swedesboro, NJ 08085, USA. Tip Top Stahlgruber, D-81675, Muenchen, Germany. United States Biochemical, P.O. Box 22400, Cleveland, OH 44122, USA. VayTek Inc., 305 West Lowe Ave, Suite 109, Fairfield, IA 52556, USA. Vector Vector Laboratories Ltd., 16 Wulfric Square, Bretton, Peterborough PE3 8RF, UK. Vector Laboratories Inc. 30 Ingold Rd, Burlingame, CA 94010, USA. Vysis Vysis (UK) Ltd., Rosedale House, Rosedale Road, Richmond, Surrey TW9 2SZ, UK. Vysis Inc., 3100 Woocreek Drive, Downers Grove, IL 60515-5400, USA. Wellcome Reagents, Langley Court, Beckenham, Kent BR3 3BS, UK. Whatman International Ltd., St Leonard's Road, 20/20 Maidstone, Kent ME16 OLS, UK. Worthington, 7 Tavistock Estate, Ruscombe Business Park, Ruscombe Lane, Twyford, Reading RG10 9NJ, UK. Xillix Technologies Corporation, #300-13775 Commercial Parkway, Richmond, BC V6V 2V4, Canada. Zeiss Optical Systems Inc., One Zeiss Drive, Thornwood, NY 10594, USA.

205

This page intentionally left blank

Index acetone, fixation with 108-9,111 actinomycin D 62-3,77-8,161 agarose beads 151-5,159-60 alkaline agarose gel electrophoresis 42-3,45 alpha satellite 82,89-90,93^t, 100,191-3 AMCA 117,120,130 a-amanatin 161 anaphase 82-3,91-6,98,100,196 antibodies conjugated 119 coupling to protein Sepharose 52,55 antifade agents, see mountants artificial chromosomes 183 ATP 9,155-6,173,176-7 Bal 3l nuclease 184,188-91 bandshift 1 biotin, labelling with 69-70,112,127-8,161 biotin, detection of 712,75,104,116,129-30, 138,141,162-3,165 bithorax complex, Drosophila 31-4 BrdU 62,77,160 Caenorhabditis elegans 126,134 cell cycle stage 104,116 cell wall, digestion of 45-6 centromere 82-3,90,93,108,191-2,194 centromere proteins (CENPs) 82-3,90,194 CHIP (chromatin immunoprecipitation), see immunoprecipitation, chromatin chromosomal proteins 175-6 chromosome assembly in vitro 169,172-3, 175-8 chromosome banding 77-8,158,164 chromosome condensation 60-2,103,134-5, 167,169,174-6,180-1 chromosome paints 59,69,103,105,111-12 chromosome scaffold 67 chromosome stability 93-4,100,196 chromosome territories 103-4,117,164 CISS (chromosomal in situ suppression), see suppression of hybridization from repeats Colcemid 60-1,63-4, 84-5,90,92-^t colony picking 187 competitive PCR 48,55-6 confocal laser scanning microscopy (CLSM) 120-1,128,143,164 Cot-1 DNA 73,112 counterselection 191 coverslips, see slides CpG island 2,17 creatine kinase 177

CREST 82 CsCl gradients 24-8 CSF, see cytostatic factor cyanine dyes 40,69,75,120,129-30,137,163 cyclins 168,174 cytocentrifugation 83-7,94-5,106-7 cytochalasinB and D 912,96-9,170-4,193 cytokinesis 82,91-2,96,99 cytosol, Xenopus egg extract clarified 168-9, 172,176 cytostatic factor (CSF) 168 DAPI, see 4,6-diamidino-2-phenylindole deconvolution 120-1,128,134,143 depurination 66,109-10 detection of hybridized probes 47,75-6,89, 138,141 DHS, see DNase I hypersensitive sites 4,6-diamidino-2-phenylindole 47,77,86,120, 130,138,141,163 dicentric chromosomes 90,94,100 digitonin, permeabilization of cell membrane with 63-5,150 digoxigenin, labelling with 41-2,69-70,112, 127-8,161 digoxigenin, detection of 42-3,47,71-2,75, 129-30,138,141,162 dimethylsulfate, modification of DNA by 4-7 DMS, see dimethylsulfate DNA staining with fluorescent dyes 23,26,47, 77,129-30,163,165 DNase 11,2,4-5,41-2,45,69-70,147,149 DNase I hypersensitive sites (DHS) 2,4 Drosophila embryos, collection and staging of 23 cross-linking of 23-6 preparation of chromatin from 23,27-8 DT40 cells 192-3 EGS, fixation and cross-linking with 196-7 electroblotting 10-11 electroelution 154-8 electron microscopy 159-60,162 electroporation 186-7 end-labelling of DNA 3,9,12,127 ethanol, fixation with 108,135 FISH, see fluorescence in situ hybridization FITC, see fluoroscein isothiocyanate fluorescence in situ hybridization 39,46-7,59, 69,72-7,81,84,87-90,99,113-15,125, 134,137-41,157,185,188,190,193,196-7

Index fluorochromes 69,129,144 flouroscein isothiocyanate 40,75,97,100,120, 130,163 footprinting DNase I 1,4,13 dimethylsulfate 5,12,16-17 formaldehyde, cross-linking and fixation with 21-6,36-7,43,45-6,48-50,53,86-8,96-8, 108-9, 111, 133,135,163,179,196 freeze-fracture 109-10 GAGA factor 22,31-4 genomic sequencing 4 GFP, see green fluorescent protein glutaraldehyde, fixation with 45-6,133,196 green fluorescent protein 49,109 Hl kinase 168,173-4 halos 106-7,157 histones, and modifications of 36-7,49,51, 153,167 Hoechst dyes 26,130 homologous recombination 191-3 HPRT, see hypoxanthine-guanine phosphoribosyl transferase hydrazine, modification of DNA by 4 hygromycin 185,191 hypotonic, swelling of cells with 61-2,64,83-5, 94-5,99,107,148 hypoxanthine-guanine phosphoribosyl transferase 191 immunodepletion 173,180 immunofluorescence 39,49,81,83-7,92,94, 96,116,118,164,178-80 after FISH 117-19,133 before FISH 43,87,90,116,119 immunoprecipitation, of chromatin (CHIP) 21,28-9,31-3,36,39,48-9,51-5 interphase extract, see Xenopus egg extract activated interphase interspersed repetitive sequences (IRS) 59,112 IP, see immunoprecipitation kinasing 9,14-17 kinetochore 81-3,169 LCR, see locus control regions ligation of DNA 3,8,30 ligation-mediated polymerase chain reaction (LMPCR) 1,3-4,8-13,17 lipofection 186 LIS, 3,5-diiodosalicylic acid 68,107,150,153 LMPCR, see ligation-mediated polymerase chain reaction

locus control regions (LCRs) 2 loop size 157-8 low gelling temperature agarose 152-3 matrix-associated regions (MARs) 59,106, 153,157 Maxam and Gilbert chemical cleavage 4 meiosis 126,168-9 metaphase chromosomes extraction of 66-8 preparation from blood 60-3 preparation from cultured vertebrate cells 62-4,84-5 preparation from in vitro assembly reactions 173,175-8 spreading of 65-6,83-5 structure of 67,83,169 treatment with RNase 66-7 methanol, fixation with 94-5,108-9, 111, 133-4 methanol:acetic acid, fixation with 61-2,65-6, 87-8,94,98-9,108,133 methotrexate (MTX) 62-3 methylcytosine 4 microcell-mediated chromosome transfer 192-3 micrococcal nuclease (MNase) 2,5,13-17 microtubules (MTs) 81-3,91,94-7 inhibitors of polymerization, see Colcemid; nocodazole; vinblastine mitotic extract, see Xenopus egg extracts, CSF(mitotic) mitotic index (MI) 60,63-4,84,94 mitotic shake-off 64,94-5 mitotic spindle 60,81-3,94 MNase, see micrococcal nuclease 78 mountants 47,78,86,119-20,137 nick translation 41,69 nocodazole 91,94-5,97-9 nuclear envelope/lamina or periphery 82,108, 117,133,159,172 nuclear matrix 67,106-7,147-9,153,157, 165 nuclear scaffold 148,150,153,157,159,165 nuclear speckles 164-5 nuclease, see DNase I; micrococcal nuclease 2 nuclease hypersensitive sites 2,4,16-17 nuclei, isolation of 13-14,142,149 nuclei, preservation of structure in 103-8,110, 125,133-5,147,151,154 nucleoids 149,153,157 nucleolus 118-19,159,165 nucleoskeleton 147-8,151,157,165 nucleosome ladder 2,14-16, nucleotides, conjugated 41,69,112,127-9, 161-2 optical sectioning microscopy 128,134,165

208

Index paraformaldehyde, see formaldehyde PC, see Polycomb PCR, see polymerase chain reaction pepsin, digestion with 111 peptide nuceic acid (PNA) probes 144 permeabilizing cells nuclei and tissues 24-5, 45-6,107,109,133,151,160-1,163,196 phenylmethylsulfonyl fluoride (PMSF) 29,67, 148,150,194-5 phosphocreatine 173,177 phosphorimaging 56 piperidine, cleavage of DNA with 4-6 plasmid-rescue 185 PMSF, see phenylmethylsulfonyl fluoride polyacrylamide gel electrophoresis (PAGE) 55,164 Polycomb 21,31-4 polylysine, coating of slides or coverslips with 47,106,136,179 polymerase chain reaction 22,28,31—2,48, 70-1 see also ligation mediated polymerase chain reaction; competitive PCR probe denaturation 47,74,89,99,113,138,140 probe labelling by end-labelling 127,131-2 by nick translation 41-2,45,69,112,127 by PCR 70,112,127 by random priming 70,127 probe size for FISH 40-4,69-70,112,127 propidium iodide 77,129-30,138 protein A Sepharose 29,52-3,55 protein G Sepharose 52-3 proteinase K digestion 7,14-15,27-9,53-4, 156,194-5 pulsed-field gel ecetrophoresis 194-5 radiolabelling DNA in vivo 155,162 replication, sites of 148-50,159-62 rhodamine 69,100,120,130 RN A polymerase 160-2 RNase, treatment of slides with 66,137,110, 140

denaturation of 72-4, 87-8,99,113-15,138, 140,142 freezing 66,109-10,136,142 growing and attaching cells on 105-6,135-6, 139,163 hybridizing probes to 73-4,89,99,115,140, 142 incubation of antibodies with 75-7,86,89,140 mounting tissue onto 136,138-9 washing of 75-6,87,89,115,140 somatic cell hybrids 87,111-12,193-4,196 sonication 25-6,28,48,50-1 Southern hybridization 2,11,17,22,31-6,55, 154,156,184,188,190,195 SpI transcription factor 11-13,16-17 spindle pole 82,91 sucrose step gradient 64-5,149-50 suppression of hybridization from repeats 73, 112 Swi6 protein 48-9,52 synchronization of cells 62-3 telomeres, detection of 188-91 telomere formation 184-7,191 telomere structure 183,188 telophase 82-3,92-3,100,196 tetramethylirhodamine isothiocyanate (TRITC) 69 Texas Red 69,97,117,120 topoisomerase II 157 transcription, sites of 148-50,157,159-62, 164-5 transfections, see lipofection; electroporation TRITC, see tetramethylirhodamine isothiocyanate tubulin, see microtubules tyramide 185 vinblastine 91

S phase 62-3,104,116 salt extraction of chromosomes or nuclei 67-9, 106-7,147-50,153,159-60 scaffold-associated regions (SARs) 59,153, 157 scintillation counting 154,156 Sephadex G5O 41,69 Silent Information Regulators 21 SIR, see Silent Information Regulators slides and coverslips cleaning 65,84,105,179 counterstaining 77,86,89,141 dehydration of 66,74,88,99,107,142

whole-mount tissues detection of fluorescence in 130,143 penetration into 128,131,133,135 preparation of for FISH 125,133,135-7, 139-40,143 Xenopus egg extracts activated (interphase) 168,173-5 CSF(mitotic) 168-75 Xenopus eggs, collection of 168-70 Xenopus sperm nuclei 168,172,175-7 yeast cell lysis 50-1

209