UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Characterization of the role of HACE1 in Wilms' tumours Anglesio, Michael Stephen 2006

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_2007-266700.pdf [ 28.88MB ]
Metadata
JSON: 831-1.0100318.json
JSON-LD: 831-1.0100318-ld.json
RDF/XML (Pretty): 831-1.0100318-rdf.xml
RDF/JSON: 831-1.0100318-rdf.json
Turtle: 831-1.0100318-turtle.txt
N-Triples: 831-1.0100318-rdf-ntriples.txt
Original Record: 831-1.0100318-source.json
Full Text
831-1.0100318-fulltext.txt
Citation
831-1.0100318.ris

Full Text

CHARACTERIZATION OF THE ROLE OF HACE1 IN WILMS' TUMOURS by M I C H A E L S T E P H E N A N G L E S I O B.Sc. M c G i l l University, 1998 A THESIS S U B M I T T E D I N P A R T I A L F U L F I L L M E N T OF T H E R E Q U I R E M E N T S F O R T H E D E G R E E O F D O C T O R OF P H I L O S O P H Y in T H E F A C U L T Y OF G R A D U A T E S T U D I E S (Pathology and Laboratory Medicine) T H E U N I V E R S I T Y O F B R I T I S H C O L U M B I A N O V E M B E R 2006 © MICHAEL STEPHEN ANGLESIO, 2006 ABSTRACT Our research group recently identified the novel ubiquitin-protein ligase gene, H A C E 1 ( H E C T domain and Ankyr in repeat Containing E3 ubiquitin-protein ligase 1), on Chr. 6 near a unique t(6;15)(q21;q21) translocation in a Wi lms ' tumour. HACE1 levels are reduced in - 7 5 % of Wilms ' tumours and a number of other malignancies compared to patient-matched normal tissue. Furthermore, aberrant methylation patterns within C p G islands upstream of the HACE1 gene, in Wi lms ' tumours, correlate increased levels of methylation with lower expression (60%). This has led us to hypothesize that HACE1 functions as a tumour suppressor gene. To explore this hypothesis, I have established two retroviral model systems. First, I have observed that Wilms' tumour derived cell lines grown in soft agar show reduced colony number and size when cells were engineered to stably over-express H A C E 1 compared with empty vector or a non-functional H A C E 1 mutant. In vivo tumour formation in nude mice injected with a Wi lms ' tumour cell line overexpressing H A C E 1 is also highly attenuated compared with empty vector control. Second, I have identified a number of small interfering R N A (s iRNA) sequences that functionally knock-down H A C E 1 expression. These s i R N A s have been used in a stable lentiviral system to reduce H A C E 1 expression by as much as 80% in a number of cell lines, mimicking the low level of H A C E 1 observed in Wi lms ' tumours. Knocking down H A C E 1 levels in H E K 2 9 3 cells resulted in an increase in both the number and size of soft agar colonies. H A C E 1 appears to affect several different growth and survival related pathways. Although direct targets of H A C E 1 are not yet validated, changes in A K T and G S K 3 P activation and the levels of cyclin D l are consistently affected by H A C E 1 expression. Taken together, these data support a role for H A C E 1 as a novel tumour suppressor whose reduced expression can contribute to malignant progression. Keywords: Wi lms ' tumour, nephroblastoma, 6q21, translocation, tumour suppressor gene, cancer, ubiquitin, E3 , ubiquitin-protein ligase, HACE1 i i TABLE OF CONTENTS Page A B S T R A C T i i T A B L E OF C O N T E N T S ii i LIST OF T A B L E S v i LIST OF FIGURES vi i LIST OF A B B R E V I A T I O N S ix A C K N O W L E D G E M E N T S xiv C H A P T E R I: I N T R O D U C T I O N 1 1.1 SYNOPSIS AND RATIONALE FOR THE THESIS 1 1.2 WILMS' TUMOURS 3 1.2.1 Clinical Features 3 1.2.2 Familial & Syndromic Forms 5 1.2.3 Molecular Genetics of Sporadic Wilms ' Tumour 7 1.2.4 Epigenetics & Imprinting in Wilms ' Tumour 12 1.3 MECHANISMS OF ONCOGENESIS 13 1.3.1 The Cel l Cycle.... 14 1.3.3 Tumour Suppressor Genes 23 1.3.3.1 Epigenetics & Silencing. 26 1.4 MECHANISMS AND ROLES OF UBIQUITINATION 28 1.5 AIMS AND OBJECTIVES 31 C H A P T E R II: M A T E R I A L S A N D M E T H O D S 33 2.1 PRIMARY TISSUES 3 3 2.2 DNA AND RNA ISOLATION 33 2.3 SOUTHERN AND NORTHERN BLOT ANALYSIS 34 2.4 FLUORESCENCE IN SITU HYBRIDIZATION (FISH) 35 2.5 BAC AND COSMID PROBES 36 2.6 RT-PCR, QUANTITATIVE RT-PCR AND ANALYSIS 37 2.7 BIOINFORMATICS 39 2.7.1 G E N S C A N gene prediction software 39 i i i 2.7.2 Human Genome Browser Tools 40 2.7.3 Protein Domain Prediction Tools 40 2.8 PREPARATION OF PROTEIN LYSATES AND IMMUNOBLOTTING... 42 2.9 CLONING & GENERATION OF TAGGED HACE1 CONSTRUCTS 45 2.9.1 pBluesc r ip t l l -HACEl 46 2.9.2 pET-15b-(HIS)-HACEl 47 2.9.3 p c D N A 3 H A - H A C E l 48 2.9.4 pMSCVhygro- H A H A C E 1 49 2.9.5 Site-Directed Mutagenesis (HACE1-C876S) 50 2.20 CELL CULTURE 50 2.11 RETROVIRUS CONSTRUCTION 52 2.11.1 M S C V based retroviruses 52 2.11.2 Lentivirus based retroviruses & Gateway cloning 53 2.22 RNA INTERFERENCE OF HACE1 54 2.13 EPIGENETIC ANALYSIS 55 2.13.1 5 -AZ Treatment, Methylation Sensitive Restriction Digests & P C R 55 2.13.2 Chromatin Immunoprecipitation (ChIP) 56 2.24 SUBCELLULAR LOCALIZATION BY IMMUNOFLUORESCENCE 58 2.15 BIOCHEMICAL ASSAYS 58 2.15.1 Recombinant H A C E 1 purification 58 2.15.2 Thio-ester bond formation assay 59 2.26 GENOMIC DNA SEQUENCING 59 2.16.1 Primer Design and Sequencing 60 2.16.2 D N A sequence assembly & data analysis 61 C H A P T E R III: D I S C O V E R Y OF A N O V E L E3 UBIQUITIN-PROTEIN L I G A S E G E N E A T A t(6; 15)(q21; q21) T R A N S L O C A T I O N IN S P O R A D I C W I L M S ' T U M O U R 62 3.2 INTRODUCTION 62 3.2 RESULTS , 64 3.2.1 Cytogenetic Analysis 64 3.2.2 Identification of the Breakpoint Region by FISH, Southern Blotting, and In Silico Mapping 67 3.2.3 Gene Prediction and Validation 73 3.2.4 Functional Predictions & Validation Of Function 84 3.2.5 Subcellular Localization of H A C E 1 87 3.3 SUMMARY : 89 C H A P T E R IV: HACE1 EXPRESSION IS A L T E R E D IN S P O R A D I C W I L M S ' T U M O U R 92 iv INTRODUCTION 92 4.1 RESULTS 94 4.1.1 Quantitative RT-PCR & Protein Expression Analysis of the Index Case and Other Wilms ' Tumours 94 4.1.2 Genomic Sequencing of H A C E 1 in Sporadic Wilms ' Tumours 101 4.1.3 Methylation Analysis & Epigenetic Control of the H A C E 1 locus 104 4.1.4 Expression of H A C E 1 in other tumours 110 4.2 SUMMARY 115 C H A P T E R V: E F F E C T S OF A L T E R E D HACE1 EXPRESSION IN VITRO A N D IN VIVO 118 5.1 INTRODUCTION , 118 5.2 RESULTS 120 5.2.1 Expression of H A C E 1 and HACE1(C876S) Mutant i n SKNEP1 Wilms ' Tumour Cells 120 5.2.2 R N A Interference & Stable Knock-Down Of H A C E 1 125 5.2.3 A hacel Mutant Mouse Model 130 5.2.4 Altered Signal Transduction and Interactions wi th H A C E 1 130 5.3 SUMMARY 138 C H A P T E R VI: S U M M A R Y A N D C O N C L U S I O N S 142 6.1 HACE1 EXPRESSION AND SILENCING 142 6.2 HACE1, CELL CYCLE CONTROL, AND THE UBIQUITIN PROTEASOME PATHWAY 146 6.3 GENERAL COMMENTS AND FUTURE DIRECTIONS 252 R E F E R E N C E S 157 A P P E N D I X A : Stock Solutions 180 A P P E N D I X B: A hacel mutant mouse model 183 A P P E N D I X C: Animal Care Certificate 197 A P P E N D I X D: Publication 200 v LIST OF TABLES Page T A B L E 1 - 1 : S E L E C T E D O N C O G E N E S I N C A N C E R 2 2 T A B L E 2 - 1 : A N T I B O D I E S 4 4 T A B L E 2 - 2 : M E D I A F O R M U L A T I O N S 51 T A B L E 4 - 1 : S E Q U E N C I N G S U M M A R Y FOR HACE1 I N W I L M S ' T U M O U R S 1 0 3 T A B L E 4 - 2 : N C I - 6 0 HACE1 Q R T - P C R R E S U L T S 1 1 4 T A B L E 5 - 1 H A C E 1 I N T E R A C T I N G P R O T E I N S I D E N T I F I E D B Y M A S S S P E C T R O M E T R Y 1 3 7 T A B L E B - 1 : S P O N T A N E O U S T U M O U R D E V E L O P M E N T I N H A C E I D E F I C I E N T M I C E 1 9 2 T A B L E B - 2 : S P O N T A N E O U S T U M O U R I N C I D E N C E I N M I C E C A R R Y I N G M U T A N T P 5 3 A N D H A C E I A L L E L E S 1 9 4 vi LIST OF FIGURES Page F I G U R E 1 - 1 : W T 1 S T R U C T U R E 9 F I G U R E 1 - 2 : T H E C E L L C Y C L E 1 6 F I G U R E 1 - 3 : M E C H A N I S M S O F U B I Q U I T I N A T I O N 3 0 F I G U R E 2 - 1 : P B S I I - H A C E I 4 6 F I G U R E 2 - 2 : P E T - 1 5 B - ( H I S ) H A C E 1 4 7 F I G U R E 2 - 3 : P C D N A 3 H A - H A C E 1 4 8 F I G U R E 2 - 4 : P M S C V H Y G R O - H A - H A C E 1 4 9 F I G U R E 3 - 1 : C Y T O G E N E T I C I D E N T I F I C A T I O N O F A T ( 6 ; 1 5 ) ( Q 2 1 ; Q 2 1 ) T R A N S L O C A T I O N 6 5 F I G U R E 3 - 2 : C H R O M O S O M E P A I N T I N G O F A R E C I P R O C A L T ( 6 ; 1 5 ) ( Q 2 1 ; Q 2 1 ) T R A N S L O C A T I O N . . . . 6 6 F I G U R E 3 - 3 : F I N G E R P R I N T E D C O N T I G S ( F P C ) T I L I N G P A T H 6 8 F I G U R E 3 - 4 : F L U O R E S C E N C E I N - S I T U H Y B R I D I Z A T I O N ( F I S H ) 6 9 F I G U R E 3 - 5 : T H E 6 Q 2 1 B R E A K P O I N T R E - A R R A N G E M E N T I D E N T I F I E D B Y S O U T H E R N B L O T T I N G . . . 71 F I G U R E 3 - 6 : T H E 6 Q 2 1 B R E A K P O I N T L O C U S 7 2 F I G U R E 3 - 7 : G E N S C A N W E B G E N E P R E D I C T I O N 7 5 F I G U R E 3 - 8 : R T - P C R V A L I D A T I O N O F G E N S C A N G E N E P R E D I C T I O N S 7 7 F I G U R E 3 - 9 : T H E P R E D I C T E D D O M A I N A R C H I T E C T U R E F O R H A C E I 8 0 F I G U R E 3 - 1 0 : E X P R E S S I O N P R O F I L E O F THE HACEI G E N E P R O D U C T 8 3 F I G U R E 3 - 1 1 : IN VITRO U B I Q U I T I N L I G A S E A C T I V I T Y O F H A C E I 8 6 F I G U R E 3 - 1 2 : H A C E I IN VIVO U B I Q U I T I N L I G A S E A C T I V I T Y 8 7 F I G U R E 3 - 1 3 : L O C A L I Z A T I O N O F H A C E I 8 8 F I G U R E 4 - 1 : P R I N C I P L E S O F T A Q M A N 5 ' E X O N U C L E A S E B A S E D Q U A N T I T A T I V E P C R A S S A Y . . . . 9 5 F I G U R E 4 - 2 : Q U A N T I T A T I V E R T - P C R F R O M T H E I N D E X C A S E 9 7 F I G U R E 4 - 3 : HACEI E X P R E S S I O N I N W I L M S ' T U M O U R A N D P A T I E N T M A T C H E D K I D N E Y 1 0 0 vii F I G U R E 4- 4: M E T H Y L A T I O N A N A L Y S I S O F HACEI I N W I L M S ' T U M O U R 106 F I G U R E 4- 5: I N H I B I T I O N O F M E T H Y L A T I O N I N S K N E P 1 C E L L S 109 F I G U R E 4- 6: HACEI E X P R E S S I O N I N M U L T I P L E T U M O U R T Y P E S 113 F I G U R E 5-1: H A - H A C E 1 O V E R E X P R E S S I O N I N S K N E P 1 C E L L S 121 F I G U R E 5- 2: G R O W T H C H A R A C T E R I S T I C S O F HACEI O V E R E X P R E S S I N G S K N E P 1 C E L L S 123 F I G U R E 5- 3: H A C E I G R O W T H S U P P R E S S I V E E F F E C T S I N T R A N S F O R M E D C E L L S 125 F I G U R E 5- 4: HACEI R N A i I N H E K 2 9 3 C E L L S 127 F I G U R E 5- 5: L O N G T E R M T R A N S I E N T S T E A L T H R N A I 129 F I G U R E 5- 6: C E L L C Y C L E A L T E R A T I O N S I N H A C E I O V E R E X P R E S S I N G H E K 2 9 3 T C E L L S 132 F I G U R E 5- 7: R E S C U E O F A L T E R E D S I G N A L T R A N S D U C T I O N I N H E K 2 9 3 T C E L L S 134 F I G U R E 5- 8: H A C E I I N T E R A C T I N G P R O T E I N S 136 F I G U R E 6-1: P O T E N T I A L S I G N A L L I N G P A T H W A Y S A F F E C T E D B Y H A C E I ; 150 F I G U R E B - 1 : G E N E R A T I O N O F H A C E I M U T A N T M I C E 184 F I G U R E B - 2: S P O N T A N E O U S T U M O U R F O R M A T I O N I N H A C E I M U T A N T M I C E 187 F I G U R E B - 3: H A C E I M U T A N T M I C E A R E S U S C E P T I B L E T O U R E T H A N E I N D U C E D L U N G C A N C E R 1 8 9 F I G U R E B - 4: C O - O P E R A T I V I T Y B E T W E E N P53 A N D H A C E I I N T U M O U R S U P P R E S S I O N 191 viii LIST OF ABBREVIATIONS Every effort has been made to use on ly official gene s y m b o l abbreviations as out l ined by the H u m a n Genome C o n s o r t i u m Nomencla tu re Commi t t ee ( H G N C ; h t t p : / / w w w . g e n e . u c l . a c . u k / n o m e n c l a t u r e / ) and pro te in symbols used by the Swiss-prot database (http:/ / w w w . e b i . a c . u k / s w i s s p r o t / V In add i t i on to s tandard H G N C and I U P A C (International U n i o n of Physics and A p p l i e d Chemis t ry ; h t t p : / / w w w . i u p a c . o r g / ) abbreviations, the f o l l o w i n g abbreviations have been used: 5-AZ (5AZ) 5-aza-2-deoxycytidine a- antibody directed against -aCGH array-based comparative genomic hybridization AKT A K R mouse thymoma ATP adenosine triphosphate BAC bacterial artificial chromosome BCR breakpoint cluster region BLAST basic local alignment search tool (http://www.ncbi.nlm.nih.qov/BLAST) BLAT B L A S T like alignment tool (http://qenome.ucsc.edu/) bp base pair BSA bovine serum albumin BWS Beckwith Wiedemann syndrome CDK cyclin-dependent (serine/threonine) kinase cDNA complimentary deoxyribonucleic acid CDS coding sequence CFS congenital fibrosarcoma CG cytosine-guanine (as in C G dinucleotide) CGH comparative genomic hybridization ChIP Chromatin immunoprecipitation ix CIP calf intestinal phosphatase COG Children's Oncology Group CpG cytosine-phosphate-guanine CS calf serum DAPI diamidino-2-phenylindole dihydrochloride hydrate ddH 2 0 Distilled-deionized water DDS Denys-Drash Syndrome DMEM Dulbecco's modified eagle medium DMSO Dimethylsulfoxide DNA deoxyribonucleic acid DNMT DNA methyl-transferase DTT Dithiothreitol EDTA ethylene-diamine-tetraacetic acid elF4E eukaryotic initiation factor 4E ERAD endoplasmic reticulum associated degradation ErbB2 v-erb-b2 erythroblastic leukemia viral oncogene ERK extracellular signal regulated kinase ES embryonic stem EST expressed sequence tag EtOH Ethanol FBS fetal bovine serum FISH Fluorescent In-Situ Hybridization FPC Finger Printed Contigs software FWT familial Wilms' tumour GH growth hormone Grb2 growth factor receptor-bound protein 2 GSK3 glycogen synthase kinase-3 GTP guanosine triphosphate H&E Hematoxilin and Eosin (histological stain) HA Hemagglutinin (epitope tag) x HACE1 H E C T and ankyrin domain containing E 3 ubiquitin-protein ligase HDAC Histone deacetylase HECT homologous to E6-AP carboxy terminus HEPES N-(2-hyroxyethel)piperazine-N'-(2-ethanesulfonic acid) HGNC Human genome nomenclature committee HIF Hypoxia-inducible factors HMM hidden Markov models HUGO human genome organization IGF insulin-like growth factor IGFBP insulin-like growth factor binding protein IGF-IR insulin-like growth factor-l receptor IMAGE Integrated Molecular Analysis of G e n o m e s and their Expression (Consortium) IUPAC international union of pure and applied chemistry kb kilo-base kDa kilo-daltons LIMS Laboratory Information Management Systems LOH loss of heterozygosity LOI Loss of imprinting MAPK mitogen-activated protein kinase MDB Methyl-DNA binding domain M D M 2 murine double minute-2 MEFs mouse embryo fibroblasts MEK map kinase/erk-activating kinase MMTV mouse mammary tumour virus mRNA messenger ribonucleic acid MSCV murine stem cell virus mTOR mammal ian target of rapapmycin NCI National Cance r Institute NEB New England Biolabs nr non-redundant xi nt Nucleotide NWTSG National Wilms' Tumour Study Group OCT optimal cutting temperature (compound) ORF open reading frame PBS phosphate buffered saline PCR polymerase chain reaction PDK phosphoinositide dependent protein kinase PFAM Protein family database/search tool PI3K phosphoinositol-3' kinase PM plasma membrane PMSF phenylmethylsulfonyl fluoride PSF antibiotic-antimicotic (Penicillin, Streptomycin, and Fungizone®) PTEN phosphatase and tensin homolog deleted on chromosome ten PTK protein tyrosine kinase qPCR quantitative PCR qRT-PCR quantitative reverse transcriptase PCR R-point restriction point RB Retinoblastoma RE Restriction enzyme RING really interesting new gene RNA ribonucleic acid RNAi RNA interference RSK ribosomal S 6 kinase RT Room Temperature RTK receptor tyrosine kinase RT-PCR reverse transcriptase PCR S6K S6 kinase SAPK stress activated protein kinase SDS sodium dodecyl sulphate SDS-PAGE SDS polyacrylamide gel electrophoresis xii shRNA short hairpin ribonucleic acid SIOP International Society of Pediatric Oncology siRNA small interfering ribonucleic acid SKNEP1 Sloan-Kettering Cancer Centre nephroblastoma cell line 1 SMART Simple Modular Architecture Research Tool SNP Single Nucleotide Polymorphism TBS Tris-buffered saline TGF transforming growth factor TK tyrosine kinase TSA Trichostatin A TSG Tumour suppressor genes Ub Ubiquitin UCSC University of California Santa Cruz UPR Unfolded protein response VHL von Hippel-Lindau (gene when italic or protein when not italicized) WAGR Wilms' tumour, Aniridia, genitourinary abnormalities, mental retardation WT Wild-type (NOT Wilms' Tumour) WT1 Wilms' tumour suppressor gene 1 WT2 Wilms tumour 2 (locus at 11 p15.5) WTCR PAX6-WT1 Critical Region x i i i ACKNOWLEDGEMENTS I would like to thank all of the members of Dr. Poul Sorensen's laboratory, past and present, who have helped me either directly or indirectly throughout my training, In particular, Dr. Valentina Evdokimova, Dr. Fan Zhang, Dr. Maureen O'Sullivan, and Nataliya Melnyk all of whom have contributed significant data to this thesis. I would also like to thank the faculty, staff and students at the Child and Family Research Institute (formerly BC Research Institute for Children's and Women's Health), the Centre for Molecular Medicine and Therapeutics, The BC Children's Hospital and The BC Cancer Research Centre for technical support, guidance, and insightful scientific discussions. I am grateful to all of the collaborators who have contributed to this work, including Dr. Conrad Fernandez, Dr. Josef Penninger, Dr. Angela Brooks-Wilson, and the Members of the National Wilms' tumour Study Group. Finally, I would like to thank my supervisor, Dr. Poul Sorensen, for allowing me the opportunity to join a highly skilled research team, and my graduate committee, Drs. Marco Marra, Wan Lam, and Cheryl Wellington, for their guidance. The work contained herein has been financially supported by the Canadian Institutes for Health Research, the National Cancer Institute of Canada, the Children's Oncology Group and Johal Program in Basic and Translational Research (via grants to PHBS). xiv CHAPTER I I N T R O D U C T I O N 1.1 SYNOPSIS A N D R A T I O N A L E F O R T H E THESIS Wilms 1 tumour, also known as nephroblastoma, is the most common malignant tumour of the kidney in children and a prime example of the successes in cancer treatments. Wilms' tumour affects approximately 1 in every 10,000 births (1-3), the peak age of occurrence is at 3 years, and it occurs only rarely after age 8 (2-6). The outlook for children with Wilms' tumour has greatly improved in recent decades with cure rates for unilateral Wilms' tumour exceeding 85% (2, 5). This is primarily attributable to the success of a multidisciplinary approach and the efforts of cooperative study groups including the National Wilms' Tumour Study Group (NWTSG) in North America and the International Society of Pediatric Oncology (SIOP) in Europe. Wilms' tumour is known to be associated with several congenital abnormalities: W A G R syndrome (Wilms' tumour, Aniridia, genitourinary abnormalities, mental retardation) is linked to germline mutations at chromosome H p l 3 while Beckwith Wiedemann syndrome is associated with loss of heterozygosity at H p l 5 (2, 3). Aniridia is an autosomal dominant eye anomaly caused by haploinsufficiency of PAX6; this can be caused by base alterations, position effects, and deletions (1, 2, 7, 8). When these deletions involve the neighbouring gene(s), WTl or others in the PAX6-WT1 Critical Region (WTCR), patients are predisposed to Wilms' 1 tumour. Epigenetic Loss of imprinting (LOI) of the IGF2 (insulin-like growth factor 2) gene was observed first in Wilms' tumour (9). LOI was associated with a 2.2 fold increase in IGF2 expression and patients whose Wilms' tumour displayed LOI of IGF2 were statistically significantly older at diagnosis than patients whose tumours displayed normal imprinting (9, 10). Increases in IGF signalling is a survival factor now known to play a role in a number of malignancies (reviewed in (11-13)). Finally, mutations of the tumour-suppressor gene p53 are correlated with anaplasia and advanced-stage disease (2, 3,14-16). However, even given this seemingly large body of knowledge on molecular aspects of Wilms' tumours as many as 97% of cases arise without a known heritable or congenital risk factor or cause (1-3, 17, 18). As with many pediatric malignancies, an underlying genetic mechanism is believed to play a large role. In fact, progress in the field of molecular genetics has allowed us to define cancer as a genetic disease, and molecular genetic tools are currently a large part of initial diagnosis, definition of prognostically distinct patient subgroups, selection of patients for specific therapies, prediction of risk for toxicities to therapy, and monitoring of patients receiving both conventional and novel targeted therapies. Further, despite significant improvements in outcomes, there are sizeable challenges to overcome and many believe that future advances in the care of pediatric cancer patients will not come from conventional approaches. As mentioned previously, Wilms' tumour survival rates are close to 85%. However, it is necessary to define molecular events, both genetic and epigenetic, that further define Wilms' tumour so that this knowledge can be applied to diagnosis, 2 prognosis, and treatment, especially if we are to make further significant improvements. The project described in this dissertation is a new addition to the scope of projects done in Dr. Poul Sorensen's laboratory. This chapter will give a brief overview on Wilms' tumour, molecular mechanisms that are common to oncogenesis, and the ubiquitin proteasome system. In the following chapters I will describe the initial discovery of a unique Wilms' tumour case along with the molecular genetic follow up leading to the discovery and characterization of the novel E3 ubiquitin ligase gene HACEI. I will discuss the general role of H A C E I as it applies to the ubiquitin-. proteasome system. I will further present experimental evidence suggesting that methylation may be a primary mechanism for maintaining low HACEI expression levels in Wilms' tumour. Finally, I will examine the role of HACEI as a tumour suppressor gene and show that it has growth suppressive qualities in multiple cell lines. 1.2 W I L M S ' T U M O U R S 1.2.1 Clinical Features Wilms' Tumour is typically a solitary mass that is sharply demarcated from adjacent renal parenchyma by a fibrous pseudocapsule (1-3, 5,19). Hematuria may be observed in up to 25% of patients, and hypertension may be present due to an increase in renin secretion by the tumour or from displaced kidney or renal artery by tumour growth (1, 2, 5). Occasionally, fever, malaise, and anemia occur as a result of tumour 3 necrosis with intraparenchymal bleeding. Approximately 7% of Wilms' tumours are multicentric, consisting of more than one tumour focus site, and ~ 5% are bilateral, involving both kidneys (1, 5). Wilms' tumour can arise anywhere in the cortex or medulla of the kidney. The tumour may protrude into the calices and sometimes the ureter. In about 6% of cases Wilms' tumour invades the renal vein and sometimes the vena cava (1,5). At a microscopic level, Wilms' tumour exhibits a triphasic lineage of blastemal, stromal and epithelial cells (1-3, 5, 19). The blastemal cells are typically small closely packed basophilic cells with a high nuclear to cytoplasmic ratio. The epithelial cell component may form primitive rosette-like structures, tubules or papillary structures. Stromal cell differentiation most often resembles that of skeletal muscle, cartilage, osteoid tissue, or fat. Approximately 5% of Wilms' tumours show nuclear anaplasia; this is defined as cells with multipolar polyploid mitotic figures and large hyperchromatic nuclei, at least 3 times the size of nuclei elsewhere in the specimen (1, 5). This histological finding defines unfavourable histology and is associated with an adverse outcome. Anaplasia may occur in any of the 3 components: blastemal, stromal or epithelial. Treatment protocols for Wilms' tumour follow two main streams set by the International Society of Pediatric Oncology (SIOP) in Europe and the National Wilms' tumour Study Group (NWTSG; now succeeded by the Children's Oncology Group, COG) in North America (2). Generally, SIOP protocols call for 4 weeks of 4 chemotherapy followed by surgical resection of the primary tumour. In contrast, the N W T S G prefers immediate surgical resection followed by chemotherapy. Modern chemotherapy appears to be extremely effective for Wilms' tumour and first-line chemotherapeutic agents include dactinomycin, vincristine, and doxorubicin (1, 2). Overall Wilms' tumours metastasize infrequently, though typical metastatic sites include lymph nodes, liver, and lungs. Survival for patients with current therapeutic regimens often exceeds 85%. As the specific treatment regimens and clinical progress of Wilms' tumour is beyond the scope of this dissertation the reader is encouraged to refer to the reports of both the C O G (http: / /www.childrensoncologygroup.org/) and SIOP (http: / / www.siop.nl/), as well as to Metzger and Dome's recent review Current Therapy for Wilms' Tumour (2) and other specific references listed throughout this chapter for further information on tumour staging and treatment options. 1.2.2 Familial & Syndromic Forms The frequency of bilaterality and the association with congenital anomalies have led many to believe that Wilms' tumour has a substantial heritable component. However a number of studies have shown that hereditary Wilms' tumour is in fact uncommon. Bonaiti-Pellie et al. (20) analyzed family history for 501 Wilms 1 tumour patients and found that just 2.4% had a positive family history of Wilms' tumours, this is further supported by other studies (21, 22). Pedigree and segregation analyses suggest that Wilms' tumour predisposition is the result of an autosomal dominant allele that is incompletely penetrant. The penetrance of such alleles has been estimated 5 to be between 25 and 60% and both males and females are equally represented amongst affected individuals (23-25). However, despite the small numbers of affected families two familial predisposition genes have been mapped to the genome: FWT1 at 17ql2-q21 (26) and FWT2 at 19ql3.4 (27). Neither putative gene has yet been cloned and, in fact, the proposed model of an autosomal dominant allele, similar to that observed in retinoblastoma, appears to be oversimplified in the case of Wilms' tumour (17, 28). This is exemplified first in the observation of L O H at the FWT2 region at 19ql3.4 in tumours from families whose inherited predisposition was not linked to FWT2 (27). Further, FWT1 in familial predisposition and tumorigenesis also may not conform to original expectations; as a tumour from an FWTl-linked family was observed to have lost the 17q alleles genetically linked to predisposition in the family (29). There are also Wilms' tumour families for which linkage to any familial Wilms' tumour (FWT) gene locus can be ruled out, implying that additional familial predisposition genes have yet to be localized (17, 27,28,30). In addition to -2% of Wilms' tumours being associated with some sort of familial predisposition a number of congenital anomalies are also known to be associated with Wilms' tumour. Aniridia, hemihypertrophy, genitourinary tract anomalies, Beckwith-Wiedemann syndrome (BWS), and Denys-Drash syndrome (DDS) have all been reported to confer an increased risk of the development of Wilms' tumour (1, 31). Interestingly, familial Wilms' tumours do not appear to display features of these congenital syndromes (17). As noted above, Aniridia is an autosomal dominant eye anomaly caused by deletions around PAX6 at l l p l 3 . When these 6 deletions involve the neighbouring gene, WT1 or others in the PAX6-WT1 Critical Region (WTCR), patients are predisposed to Wilms' tumour (1, 2, 7, 8). The related W A G R syndrome is another rare genetic disorder characterized by a de novo deletion of H p l 3 (32) and is clinically associated with Wilms' tumour, Aniridia, genitourinary anomalies, and mental retardation (WAGR). Approximately 4% of children with BWS develop Wilms' tumour (33-35). At a genetic level mutation or altered expression of one or more genes in a cluster of imprinted genes at H p l 5 has been observed in BWS, and many of these genes play a role, either positively and negatively, in growth control (33, 36, 37). In addition, the observation of nephrogenic rests, foci of undifferentiated embryonic cells within the kidney, in patients both with and without BWS imply that interruption of normal growth and differentiation of the embryonic kidney plays a role in Wilms' tumour development (38). Lastly, DDS is characterized by constitutional missense mutations in the WT1 gene, usually resulting in a rapidly progressive nephropathy, male pseudohermaphroditism, and an increased risk for Wilms'tumour (39,40). 1.2.3 Molecular Genetics of Sporadic Wilms' Tumour De novo mutations not associated with a congenital syndrome appear to represent close to 90% of Wilms' tumour cases, though the genetic factors involved have only begun to be described. The most common molecular genetic events appear to involve the WT1 locus, and to date this is the only Wilms' tumour locus where a specific gene has been cloned and characterized (41). As noted above different 7 mutations affecting the WTl locus at H p l 3 can influence the occurrence of a number of congenital abnormalities that accompany a higher risk for Wilms' tumour. However, amongst sporadic forms of Wilms' tumour, only 0.8-4.6% of patients have germline mutations in WTl and varying reports suggest up to 18% carry WTl mutations within the tumours (3, 18, 42). In addition, there are more subtle molecular changes to WTl such as two alternative m R N A splicing events within the gene, one that involves 17 amino acids encoded by exon 5, and another that involves three amino acids (KTS) of exon 9 (43). Some Wilms' tumours express abnormal ratios of these isoforms and data indicate that there are significant functional differences between the WTl isoforms (44-48). The WTl gene is composed of 10 exons and encodes a 52 to 54 kDa protein containing four tandem C-terminal Cys2-His2 zinc finger (ZF) motifs (41). The W T l gene also contains two alternately spliced regions, a variably present exon 5, encoding 17 amino acids, and a second splice donor site at the end of exon 9, resulting in an additional three amino acids (lysine, threonine, serine; "KTS") within the ZF region (18, 44, 47). Altogether WTl encodes four distinct protein isoforms (Figure 1-1). 8 Self recognition " K T S " Zn Fingers (DNA binding) / RNA recognition motif 1 Activation / Repression Exon 5 + / K T S + L Exon 5+/KTS-Exon 5-/KTS+ Exon 5-/KTS-L Nuclear localization exon 5 (17aa) V J J _ J JED Figure 1-1: WT1 structure The WT1 gene encodes four distinct protein isoforms as the result of two alternatively spliced regions, exon 5 (51 bp/17 amino acids) and KTS (alternative splice donor site at the end of exon 9). Each isoform is specifically illustrated in the lower panel. Regions of the protein involved in transcriptional activation and repression, VVT1 protein self-association, nuclear localisation, RNA recognition and nucleic acid binding are indicated in the upper panel. Primarily WT1 acts as a ZF transcription factor that is required during the development of the urogenital system. Interestingly, Frasier syndrome, another rare congenital disorder not usually associated with Wilms' tumour but leading to 9 malformations of the urogenital system, is associated with a suppression of the expression of the WT1 +KTS isoform (18). Both +/-KTS isoforms can bind nucleic acid (44, 47, 48), though the - K T S isoform is generally described as a transcriptional activator (49, 50) while the +KTS isoform has recently been linked to R N A metabolism (44, 47, 51). A n imbalance favouring a deficiency of exon 5 has been described in as many as 7/10 Wilms' tumour by at least one group (52), but has not been validated by others (43, 46, 53), leading to the general consensus that larger numbers of tumours must be examined before any conclusion as to the etiological effects of such mutations on Wilms' tumour can be drawn. A n interesting correlation between P-catenin mutations and WT1 mutations has also been observed. Mutation in P-catenin have been noted in as many as 15% of Wilms' tumours (54). However, these mutations are restricted almost exclusively to tumours that also have mutations in WT1 (54, 55). P-Catenin is a central effector of the Wnt signalling pathway that can activate growth promoting genes such as cyclin D l (56), when an upstream signal is absent, p-catenin is phosphorylated and degraded via the ubiquitin proteasome pathway. Al l of the observed mutations from the above studies involved phosphorylation sites, suggesting that P-catenin can remain available to activate target genes. The WT2 locus at H p l 5 was originally identified through the observation of loss of heterozygosity (LOH) on H p that did not include the H p l 3 WT1 site (57, 58). Although the region is known to contain a number of genes it is still commonly 10 referred to as WTl. Several of these genes, including insulin-like growth factor-2 (IGF2), H19, and p57 K i P 2 (CDKN1C) have demonstrated abnormal expression or mutations in certain Wilms' tumours and have therefore been implicated in Wilms' tumorigenesis (reviewed in (17, 18, 31)). IGF2 is frequently overexpressed in Wilms' tumour, and both gene duplications and epigenetic events (discussed further in the section to follow) have been described. In contrast, both H19 and CDKN1C appear to be expressed at low levels in a number of Wilms' tumours. Cytogenetic, L O H and comparative genomic hybridization (CGH) studies have revealed a number of other loci that appear to be important in Wilms' tumourigenesis (reviewed in (18, 31)) . Although Wilms' tumour exhibits a low frequency of L O H "genome-wide", L O H is frequently observed at 16q, in up to 20% of cases (59), and lp , in up to 10% of cases (60). More recently, higher resolution array C G H (aCGH) has further highlighted gains on l q to be important in Wilms' tumour relapse. These gains correlate strongly with losses at both l p and 16q (61). In addition, expression studies have revealed alterations in a number of genes including telomerase (62), TRKB (63), VEGF (64, 65), survivin and fas (66), and the INK4 family of cyclin-dependent kinase inhibitors (67). Finally, nuclear anaplasia in unfavourable histology Wilms' tumour is highly correlated with mutation in the tumour suppressor p53 (16, 18). Arguably the most commonly mutated gene in human neoplasms, the p53 protein acts as a negative regulator of cell proliferation and a positive regulator of apoptosis in response to D N A 11 damage (reviewed in (68, 69) and others). Almost 75% of Wilms' tumours with anaplastic histology also bear mutations in p53 (16), some with cytogenetically identifiable abnormalities of 17p resulting in loss of the gene (70). As a prognostic factor, immunopositivity of (mutated) p53 correlated significantly to anaplasia and to poor survival (71), suggesting that immunopositivity may be useful in predicting outcome. Loss of p53 function also appears to be linked to Wilms' tumour progression. This is supported by two studies: one finding that p53 mutations are often present in regions of anaplastic histology within tumour containing mixed favourable and unfavourable histology (72) and another showing that favourable histology, clinically aggressive tumours can progressively lose p53 function after being xenografted into nude mice while less aggressive favourable histology tumours retain p53 function (73). 1.2.4 Epigenetics & Imprinting in Wilms' Tumour Epigenetics is described as heritable changes in gene function that occur without a change in the sequence of genomic D N A . In general these changes usually involve methylation of D N A or modification of histone structure to condense or expand a particular region of the chromosome. Epigenetic modification of expression has been well described at the WT2 locus in sporadic Wilms' tumour. Here a number of differentially imprinted genes, in particular IGF2 and H19, appear to undergo loss of imprinting (LOI). Imprinting can restrict expression of one allele versus the other (74, 75). IGF2 is normally expressed exclusively from the paternal allele and is frequently overexpressed in Wilms' tumour after either gene duplication or LOI of the maternal 12 allele (gain of methylation on this allele), allowing biallelic expression (9, 76). Conversely, H19 is normally expressed only from the maternal allele, oppositely imprinted from IGF2. This opposed imprinting and expression pattern is the result of a shared regulatory domain that resides between the two genes. H19 is therefore silenced by the increased methylation on the maternal allele (77, 78), the same LOI phenomenon that triggers expression for IGF2 (76, 79). As mentioned above IGF2 and IGF signalling in general has been well described as pro-cell survival. HI9 however is a relatively unique untranslated gene that appears to have a growth suppressive effect when expressed in tumour cell lines (80). LOI, at least surrounding IGF2/H19, appears to be restricted to sporadic Wilms' tumour cases (81). Further, the frequency of LOI appears to be higher in lower stage tumours while L O H frequency increases with tumour stage (81). This may suggesting that LOI precedes L O H as these tumours progress (82) or that tumours that undergo L O H are biologically distinct from those that undergo LOI. Interestingly, contrary to global hypomethylation that has been reported in other tumours, Wilms' tumours appear to be globally hypermethylated (81, 83). Loss of expression of INK4 family members in as many as 23% of Wilms' tumour has been attributed to increased methylation at least in the case of p!6 (67). 1.3 M E C H A N I S M S O F O N C O G E N E S I S The term "cancer" is used to describe a variety of malignant diseases that result from uncontrolled cell proliferation. The dividing cells form large masses called 13 neoplasms, or tumours, which can invade neighbouring tissues or may metastasize to more distant sites. In contrast, benign proliferations consist of accumulated cells that neither invade other tissues nor metastasize. Weinberg and Hanahan (84) describe that in order for a normal cell to become neoplastic it must undergo a series of genetic and/or epigenetic changes that enable it to [1] evade apoptosis, [2] become self sufficient in growth signals, [3] avoid anti-proliferative signals, and [4] sustain a limitless replication potential . In addition, most malignant tumours will also require [5] angiogenesis to supply the tumour with nutrients and [6] the ability to invade tissue and metastasize to other sites. These qualities can be acquired in a multitude of different ways; in the sections to follow we will discuss two broad classes of genes that contribute to the process of transformation. In general, normal cell growth is regulated by two classes of genes. Tumour suppressor genes (TSGs) prevent tumour formation by inhibiting uncontrolled cell growth. O n the other hand, proto-oncogenes play a role in stimulating cell division and differentiation (85, 86). Consequently, malignant neoplasms can arise when mutations result in the loss of a TSG, the activation of a proto-oncogene, or a combination of both. 1.3.1 The Cell Cycle Pathways contributing to cell proliferation, growth arrest, and cell death ultimately converge upon the cell cycle, which controls D N A synthesis and cell division. Before 14 exploring the role of oncogenes and TSGs an understanding of the cell cycle and its regulation is critical to understanding key elements of oncogenesis. There are four distinct phases within the cell cycle (Figure 1-2; reviewed in (87-89) and elswhere): S-phase, wherein D N A synthesis occurs and the parental cell chromosomal complement is duplicated, M-phase (mitosis) where replicated D N A is divided and two genetically identical daughter cells are formed, and G l and G2 phases, which are gaps between mitosis and S-phase and S-phase and mitosis, respectively. In G l , cells respond to mitogenic signals and prepare for D N A synthesis. In G2, D N A synthesis is terminated and cell growth continues with accumulation of proteins and organelles to be divided between the two daughter cells during mitosis. Non-proliferating cells do not perpetually move through successive cell cycles. Instead, these cells exist in a quiescent state between M and S phases in a phase termed Go- Most cells in normal adult tissues exist in Go. Both cells in Go and G l are considered to be receptive to proliferative signals and cells in Go may enter G l in response to an appropriate signal. 15 •() CD 0 ; \ /X^ . Mitosis M Q^CDK^  _ c f f ' T ^ Cell division hromosome duplication Chromosome separation Figure 1- 2: The Cell Cycle Quiescent adult cells exist in Go. Mitogenic stimulation can cause them to enter the cell cycle where they passage through a tightly regulated sequence of events culminating in mitosis, or cell division. Once the R (restriction) point has been passed, the cell is committed to enter S-phase and cell cycle transition ceases to be mitogen-dependent. Within the cell cycle, passage from one phase to another is strictly regulated by a complex interplay between enablers and inhibitors. The most well-studied checkpoint is that regulating passage between G l and S phases, termed the restriction point (R-point) (87). In cultured cells, once a cell passes the R-point, it is committed to enter S phase. In vivo, however, cells may arrest at different points within G l in response to different inhibitory signals; thus, in reality, there may be several R-points in different cell types that restrict cell cycle progression (87, 89). When cells are stimulated by growth factors to enter G l from Go, they generally require continuous 1 6 mitogenic stimulation to reach the R-point, after which mitogens can be withdrawn and cells will continue through the cell cycle in their absence. Interactions of cyclins and cyclin-dependent kinases (CDKs) play a large role in regulating passage from one phase of the cell cycle to the next. CDKs are inactive until they bind with their respective cyclin to form a holoenzyme complex that is catalytically active (89). Cyclin levels oscillate during the cell cycle due to variations in gene transcription, protein translation and protein degradation, ensuring that levels peak at the time of maximal kinase activation. Mammalian cyclin family members include cyclins A to H , which all share a conserved sequence of about 100 amino acids. In mammalian cells, the C D K family includes 7 members that are conserved in size between 32-40 kDa, and share approximately 40% sequence homology (89). CDKs are expressed at constant levels throughout the cell cycle and, once bound to cyclins, are active serine/threonine kinases. Once activated, the cyclin-CDK holoenzyme complexes can phosphorylate their effectors, facilitating cell cycle progression. Passage through G l into S phase through the R-point is mediated by D and E type cyclins with their respective CDKs (87, 90). Cyclin D family members (Dl , D2 and D3) bind to CDK4 and CDK6 to yield at least 6 possible holoenzymes that are expressed in tissue specific patterns (87). Due to their important position at the R-point of the cell cycle, and the high level of regulation attached to their protein levels, cyclins have been the subject of much study in the context of oncogenesis. D-type cyclins and their partnered CDKs have been observed to be overexpressed in a 17 number of malignancies (91), including Wilms' tumour (92). Recently cyclin D 1 / C D K 4 complexes have undergone a great deal of study in the development of breast cancer, where gene amplification of cyclin D l is seen in 15-20% of human breast carcinomas (93, 94). Mice with expression of cyclin D l under the control of the mouse mammary tumour virus (MMTV) promoter develop mammary adenocarcinomas and, conversely, cyclin D l null mice appear to be resistant to mammary tumours which develop with MMTV-Ras or MMTV-ErbB-2 oncogenes (95). This requirement for cyclin D l in tumorigenesis appears to be due to its activation of CDKs (96). Once a cell has passed through the R-point, cyclin A / C D K 2 activation occurs and is essential for the initiation of and progression through S phase (89, 97). During S phase, levels of the B-cyclins rise and entry into and out of mitosis are controlled by cyclin B / C D K 1 complexes (89, 97). At the completion of mitosis cyclin B is degraded. Degradation of the cyclin proteins, especially cyclin D, is essential for controlled cell growth. Although cyclin D is primarily targeted for ubiquitination and degradation when it has been phosphorylated at threonine residue 286 (Thr-286) (98), it can also be ubiquitinated independently of phosphorylation when it is not bound in a complex with C D K 4 / 6 (99). Phosphorylation of cyclin D l catalyzed by GSK3P also redirects the protein from the nucleus to the cytoplasm (100). Whether phosphorylation dependent ubiquitination in a complex with CDKs or phosphorylation independent "free" cyclin D l ubiquitination occurs it is quickly 18 followed by proteasomal degradation. The ubiquitin proteasome system will be described in more detail in a later section. Briefly, proteasomes are cytoplasmic protease complexes that rapidly degrade proteins into short peptides after they are ubiquitinated by a cascade of E l , E2 and E3 enzymes (101). In addition to cyclins and CDKs positively regulating the cell cycle, negative regulation is accomplished via two distinct families of cyclin dependent kinase inhibitors: the CIP/KIP family of inhibitors, which interact with cyclin-CDK complexes in all phases of the cell cycle, and the INK4 family that specifically inhibits cyclin D-dependent kinases (89, 102, 103). The CIP/KIP family of polypeptide inhibitors includes p21 C I P 1 , p27 K I P 1 and pbT™2. A l l members of this family bind efficiently to cyclin-CDK holoenzyme complexes containing any of the CDKs from 1-6, and inhibition is accomplished by induction of conformational changes within the C D K molecules themselves (102,103). Additionally, all CIP/KIP molecules have been found to directly inhibit CDK2 in a dose dependent fashion such that high inhibitor levels are required to effect growth suppression (104). The INK4 proteins specifically inhibit the cyclin D-dependent kinases, CDK4 and CDK6. There are four known members of the family to date: p l 6 I N K 4 a , p l 5 I N K 4 b , p l 8 I N K 4 c , and p l 9 I N K 4 d . INK4 proteins inhibit the cell cycle by sequestering CDK4 and CDK6 into binary CDK-INK4 complexes (89, 103). Sequestration of CDK4 and CDK6 has two inhibitory effects. First, it prevents interaction between CDKs and cyclin D, thus preventing kinase activation. Second, disruption of cyclin D - C D K complexes effectively releases bound CIP/KIP proteins, making them available to inhibit other cyc l in /CDK complexes. 19 1.3.2 Oncogenes Oncogenes are now well established as major contributors to the development of cancer in humans. Oncogenes may be viral in origin or may be derived from normal cellular genes, referred to as proto-oncogenes. Proto-oncogenes are highly conserved in evolution and their products are important in the regulation of cell growth and differentiation in a diverse range of organisms (105). The expression of cellular proto-oncogenes is tightly regulated in normal cells, but if converted to oncogenes they can induce tumour formation. Examples of cellular proto-oncogenes include many growth factors, growth factor receptors, cytoplasmic protein kinases, G -proteins, cyclins and transcription factors (see Table 1-1). Conversion of proto-oncogenes into oncogenes can occur by several mechanisms including proviral insertion, gene amplification, point mutation, and chromosomal rearrangement (105). Activation of oncogenes by proviral insertion is complex, involving recombination between viral and cellular genomes following infection and integration of a virus into the cell. In this manner, viral sequence is integrated adjacent to a cellular proto-oncogene resulting in alterations that convert the normal gene to its oncogenic counterpart (105-107). This was first described by the activation of c -MYC transcription by avian leukosis virus in bursal lymphomas (107,108). Amplified copies of proto-oncogenes can often be seen microscopically as homogeneously staining chromosomal regions and double-minute chromosomes (105, 109,110). The proto-oncogene may be amplified at its native chromosomal location or 20 m a y be translocated to another locus as a result of ch romosomal rearrangement. W h i l e the precise mechanisms of gene ampl i f ica t ion are not yet completely unders tood, current li terature suggests that it results f r o m several rounds of unscheduled D N A . synthesis d u r i n g a single cell cycle (110). Examples inc lude ampl i f ica t ion of MYCN i n neuroblastoma (109) and c-erbB2 i n breast cancer and other malignancies (111). 21 Table 1-1: Selected Oncogenes in Cancer (compiled from (105,112,113)) Oncogene Chromosome location Method of Identification Human Neoplasm(s) Mechanism of Activation v-sis 22q12.3-13.1 Sequence homology Glioma/fibrosarcoma Constitutive expression int2 11q13 Proviral insertion (in mice) Mammary carcinoma Constitutive expression KS3 11q13.3 DNA transfection Kaposi sarcoma Constitutive expression HST 11q13.3 DNA transfection Stomach carcinoma Constitutive expression EGFR 7p1.1-1.3 DNA amplification Squamous cell carcinoma Gene amplification TRK 1q32-41 DNA transfection Colon/thyroid carcinomas DNA rearrangement NEU 17q11.2-12 Point mutation/DNA amplification Neuroblastoma/breast carcinoma Gene amplification Mas 6q24-27 DNA transfection Epidermoid carcinoma Rearrangement of non-coding regions SRC 20p12-13 Viral homolog Colon carcinoma Constitutive activation RAS 12p11.1-12.1 Viral homolog/ DNA transfection AML, thyroid carcinoma, melanoma Point mutation Dbl Xq27 DNA transfection Diffuse B-cell lymphoma DNA rearrangement Vav 19p13.2 DNA transfection Hematopoietic cells DNA rearrangement v-raf 3p25 (RAF-1) Viral homolog Sarcoma Constitutive activation pim-1 6p21 Insertional mutagenesis T-cell lymphoma Constitutive activation v-myc 8q24.1 Viral homolog Carcinoma, myelocytomatosis Deregulated activity v-myb 6q22-24 Viral homolog Myeloblastosis Deregulated activity v-fos 14q21-22 Viral homolog Osteosarcoma Deregulated activity v-ets-1 11p23-q24 Viral homolog Erythroblastosis Deregulated activity v-erbA 1 17p11-21 Viral homolog Erythroblastosis Deregulated activity BCL2 18q21.3 Chromosomal translocation B-cell lymphomas Constitutive activity MDM2 12q14 DNA amplification Sarcomas Gene amplification ETV6-NTRK t(12;15) Chromosomal translocation Congenital fibrosarcoma Ligand independent activation/fusion protein BCR-ABL t(9;22) Chromosomal translocation Chronic myelogenous leukemia active fusion protein/transcription factor PAX7-FKHR t(1;13) Chromosomal translocation Rhabdomyosarcoma active fusion protein/transcription factor EWS-FLI1 t(11;22)(q24;q1 2) Chromosomal translocation Ewing Sarcoma active fusion protein/transcription factor Poin t mutat ions i n proto-oncogenes can also result i n oncogene activation, such as occurs w i t h the RAS mutat ions that are characteristic of epithel ial malignancies. A c t i v a t i o n of the NRAS gene by point muta t ion occurs i n about 15% of a l l h u m a n melanomas (114). Such single-base mutat ions alter the amino ac id 22 sequences of the RAS proteins, causing diminished intrinsic GTPase activity and constitutive activation of RAS and its downstream proliferative pathways (105,113). Finally, proto-oncogene activation can arise when a chromosomal rearrangement, such as a translocation, places a proto-oncogene downstream of a promoter (107). This results in the constitutive expression of an otherwise tightly regulated gene. A n example of this is seen with the MYC gene and its translocation to an IgG locus in Burkitt's lymphoma due to a t(8;14) (115). Alternatively a translocation can produce a fusion gene where part of gene A is fused to part of gene B creating a fusion gene with mixed functions attributable to each of the donor genes (116,117). 1.3.3 Tumour Suppressor Genes Generally, a tumour suppressor gene is a gene that functions to reduce the probability that a cell will become neoplastic. Following this, a mutation or deletion of such a gene will increase the probability of the formation of a tumour. Most tumour suppressor genes follow a "two-hit" hypothesis stating that both alleles of a tumour suppressor must be functionally inactivated before an effect is manifested, as originally outlined by Knudson and Strong (24). Possibly the best known example of a tumour suppressor is the TP53 (Tumour Protein 53 or p53) gene. Patients with germline p53 mutations are predisposed to breast cancers, sarcomas, brain tumours, lymphomas and Li-Fraumeni syndrome (118, 119). The p53 protein exists as a tetramer and acts as a tumour suppressor gene because it can block cells at the Gl -S transition, preventing them from entering the S 23 phase of the cel l cycle (120,121). Other evidence has s h o w n that p53 can also block the t ransi t ion f rom G 2 to M v i a b i n d i n g to p 2 1 C I P 1 (122, 123). Muta t ions i n the p53 gene represent the most frequently encountered genetic aberrations observed i n h u m a n malignancies (124,125). M u l t i p l e mechanisms of p53 inact ivat ion have been observed (119, 124-126) i n c l u d i n g L O H and deletions, nonsense mutat ions that truncate the protein, missense mutat ions resul t ing i n dominant-negat ive effect, and functional inact ivat ion by degradat ion, such as the ubiqui t in-proteasome media ted degradat ion effected by h u m a n pap i l l omav i rus E6 or overexpression of MDM2. A number of molecular approaches that have typ ica l ly been used to identify t umour suppressor genes, i n c l u d i n g l inkage analysis to determine w h i c h region-specific markers are l i n k e d to the disease (which helps lead to the identif icat ion of a gene or genomic locus); cytogenetic analysis to determine the extent of ch romosomal loss (deletions) or rearrangements; and L O H analysis, w h i c h detects the loss of an allele or molecular marker. Bo th L O H and chromosome loss effectively delete a region of D N A and any gene(s) w i t h i n it. Rearrangements o n the other hand are more complex. A s noted above i n the case of oncogene activation, recombinat ion can create a n e w mul t i func t iona l fusion oncoprote in or increase the expression of a proto-oncogene by p lac ing it under control of a foreign promoter . In the case of TSGs , genes m a y be rendered non-functional w h e n a translocation or recombinat ion event truncates or deletes an allele of the gene. This is the case for FHIT, a t umour suppressor gene ident if ied i n hereditary renal cel l carc inoma (127,128). The FHIT gene is w i t h i n the F R A 3 B reg ion frequently affected by ch romosomal rearrangement and 24 breakages and is disrupted and inactivated by the familial kidney cancer-associated t(3;8)(pl4.2;q24) translocation (129, 130). Other fragile regions including FRA6F on chromosome 6q are also suspected of harbouring tumour suppressor genes and their expression is also likely to be affected by rearrangements and translocations (131). Recombination and translocation events may also result in distance effects, in which a translocation does not directly affect the gene by intersecting it but results in changes in the expression of the gene (132). Although not commonly reported for tumour suppressor genes, these distance effects cannot be ignored as a potential silencing mechanism. Distance effects from translocations >800kb away are hypothesised to deactivate expression of SOX9 contributing to campomelic dysplasia (133, 134). Reduced expression of Sonic hedgehog (SHH) has been associated with chromosomal rearrangements occurring 15-250 kb from the S H H locus, resulting in holoprosencephaly (135-137). Further, cytogenetically detectable rearrangements at H p l 3 surrounding, but at least 85kb away from, the PAX6 gene have also been attributed to its inactivation and are cited as underlying mechanism of disease, as no mutations were detectable within the PAX6 gene itself (138). A number of exceptions to the TSG "two hit" hypothesis have also been described (139), including haploinsufficiency of TP53 (140,141), BRCA1 (142) and the ubiquitin-protein ligase gene FBXW7 (143). These exceptions open the door for a greater number of yet undescribed TSGs to be identified based on lower rather than absent expression without the necessity of mutation of the other allele. Altogether, inactivation of a TSG can occur via a variety of mechanisms including loss of the gene 25 (deletion), mutation (point, missense, or nonsense), chromosomal rearrangements (affecting expression) or epigenetic silencing. Almost all described cases TSG inactivation affects both alleles. 1.3.3.1 Epigenetics & Silencing Recently epigenetics and specifically methylation has been intently studied to explain silencing of tumour suppressor genes (74, 144-146). Epigenetics itself is the study of heritable changes in expression that are not connected to change of the D N A sequence. Usually, expression changes are due to altered methylation patterns within the D N A and chromatin structure (144, 147). These epigenetic changes occur within a larger context of alterations to chromatin structure in cancer cells in comparison with the normal cells from which they are derived. These can involve both losses and gains of D N A methylation as well as altered patterns of histone modifications (147). A n ever-growing number of histone-modifications have been identified. Serine and threonine residues can be phosphorylated, arginine residues methylated and lysine residues acetylated, methylated or ubiquitinated (148). These modifications affect chromatin structure and regulate chromatin activity. This "histone code" can function as master on/ off switches to determine whether or not particular genes are expressed. In general, acetylation of specific lysine residues at the N-terminus of histones H3 and H4 weaken the interactions with D N A and result in a destabilization of nucleosomal structures, this correlates with increased gene transcription. Deacetylation of histones is associated with condensation of nucleosomal structures 26 and transcriptional silencing. Methylated D N A is known to attract methyl-DNA binding domain containing proteins (MBD proteins) which subsequently recruit histone deacetylases (HDACs) and cause chromatin condensation (149). The D N A methyl-transferase D N M T 1 can also interact directly with H D A C 1 and H D A C 2 (150, 151). This relationship of methylated D N A to chromatin modifying enzymes supports the theory that chromatin structure and methylation are intricately linked (152). D N A methylation is by far the most commonly studied epigenetic modification, however, methylation is complicated by two apparently contrasting phenomena within a cancer cell: [1] a profound loss of global methylation (5-methylcytosine content) with [2] focal regions of dense hypermethylation (144, 146, 147). This is however understandable, as hypomethylation is restricted predominantly to repetitive sequences and has been linked to chromosomal instability (144, 146). In contrast, hypermethylation occurs in C p G islands usually located in the promoter regions of certain TSGs, such as CDKN2A, BRCA1 or hMLH, and has been attributed to gene silencing (146, 147). In fact, although both mutation and methylation have been reported for CDKN2A, loss of gene function due to methylation is 10 times more prevalent than mutation (153, 154). This is also observed in sporadic breast cancer where low expression of BRCA1 correlates highly with methylation in as many as 21% of cases (155,156). Although methylation is usually associated with silencing this is not always the case. In fact, LOI at the IGF2/H19 locus observed in Wilms' tumour and ovarian 27 cancer gives an example where increased methy la t ion i n specific regions increases the expression of IGF2 (9, 76, 157). This increased expression of IGF2 is due to a shared regulatory reg ion between H19 and IGF2, as described above. 1.4 M E C H A N I S M S A N D R O L E S O F U B I Q U I T I N A T I O N Ubiqu i t i na t i on (or Ubiqui t i la t ion) refers to the post-translational modi f ica t ion of a prote in by the covalent attachment (via an isopept ide bond) of one or more ub iqu i t i n monomers (101, 158, 159). U b i q u i t i n , o r ig ina l ly named Ubiqu i tous Immunopoie t ic Polypept ide , was first ident i f ied i n 1975 as an 8.5 k D a prote in of u n k n o w n funct ion expressed un iversa l ly i n eukaryotic cells. The basic functions of ub iqu i t i n and the components of the ub iqu i t ina t ion pa thway were ou t l ined i n the early 1980s by A a r o n Ciechanover , A v r a m H e r s h k o and I r w i n Rose (160, 161), w h o were awarded the N o b e l Pr ize i n Chemis t ry i n 2004 for their w o r k (162). This m u l t i -step process involves a cascade of three different classes of enzymes. The A T P -dependent ubiqui t in-ac t iva t ing E l enzyme (UBE1) first forms a thio-ester b o n d w i t h ub iqu i t in . U B E 1 then transfers the activated u b i q u i t i n moie ty to one of a handful of different E2 ubiqui t in-conjugat ing ( U B C ) enzymes. F ina l ly , an E3 ubiqui t in-pro te in ligase catalyzes the transfer of ub iqu i t i n f rom the associated E2 to a lys ine o n the substrate prote in (out l ined i n F igure 1-3). F r o m genome w i d e scans it is thought that there may be hundreds of E3 enzymes (163), i n general fa l l ing into two m a i n categories: the H E C T d o m a i n E3s and the R I N G (Really Interesting N e w Gene) d o m a i n E3s (101,159). There is also a subgroup of R I N G - l i k e proteins k n o w n as U-box 28 E3s, which appear to be both structurally and functionally similar to the RING E3s without the full complement of Z n 2 + binding domains (163,164). This latter E3 family, including both the RING domain and U-box E3s (165), is not characterized as being able to bind ubiquitin directly. Instead, RING domain E3s facilitate ubiquitination by coupling the E2 enzyme and the substrate in close proximity, thereby allowing the E2 to transfer the ubiquitin (101, 159,163-165). In contrast the H E C T domain E3s directly bind ubiquitin from an E2 via a transient thio-ester bond on a highly conserved C -Terminal cysteine residue (101, 159, 163, 166) and then pass this ubiquitin to the substrate (Figure 3-11). Binding of substrate proteins appears to be mediated by the N -terminal protein-protein interaction domains in H E C T E3 ligases (101, 166, 167). In contrast, the substrate recognition properties of RING domain E3s can be much more complex, owing to a very diverse range of sizes and domain architecture (163). 29 Figure 1- 3: Mechanisms of Ubiquitination A highly conserved cascade of enzyme is required for ubiquitination of target proteins. Ubiquitin is transferred to a thio-ester bond on UBE1, the ubiquitin El , in an ATP dependent reaction. Ubiquitin is then passed to another thio-ester bond on a UBC E2 enzyme. Finally, an E3 ubiquitin protein ligase adds specificity to the cascade targeting various cellular proteins. RING and U-box E3s act to bring together the E2 and target protein for ubiquitination. HECT domain E3s bind the ubiquitin from the E2 in another thio-ester bond, subsequently transferring it to a lysine residue (K) on the target protein. The process is repeated to obtain a poly-ubiquitinated protein destined for degradation at the 26s proteasome. E3 ubiqui t in -pro te in ligases have been l i n k e d to oncogenesis as both oncogenes and TSGs . M o s t notably, the MDM2 proto-oncogene encodes an E3 ligase whose ampl i f ica t ion and subsequent overexpression leads to degradat ion of p53 (168). The v o n H i p p e l - L i n d a u ( V H L ) prote in is another E3 ligase w h i c h no rma l ly ubiquit inates and regulates H I F s (hypoxia induc ib le factors), i n part icular H I F - a and H I F-P subunits. S tabi l iza t ion of these factors a l lows for consti tutive transcript ional s t imula t ion of genes i n v o l v e d i n angiogenesis, red cel l p roduc t ion and glycolys is (168). 30 Loss of function mutations and deletions of VHL tumour suppressor gene predispose individuals to certain cancers such as renal clear cell carcinomas, cerebellar hemangioblastomas, retinal angiomata, pheochromocytomas, all of which have a major vascular component. Finally, the ubiquitin proteasome pathway is already being targeted in cancer treatments. Proteasome inhibitors such as bortezomib (Velcade® or PS-341) are already in clinical use and have been found to have significant anti-tumour effects (169, 170), though the exact mechanism of how this is accomplished clearly needs further examination. Overall, the ubiquitin proteasome pathway appears to be as important as kinase cascades and growth factor signalling in oncogenic progression and will surely be the topic of intense research in the coming years. 1.5 A I M S A N D OBTECTIVES As outlined in the first section of this chapter this dissertation will focus on the characterization and implications of a novel and unique t(6;15)(q21;q21) translocation in Wilms' tumour (171). This particular translocation appears to be remarkable as it involves the same region previously described in other unique Wilms' tumour translocations (172-174) and falls within a region of common chromosome breakages, rearrangements and L O H (131, 175-177) suggesting it is the possible site of an important TSG. The specific objectives I will address are as follows: 1) Mapping the translocation breakpoint on chromosome 6 to specifically identify any genes that may be affected. Affected genes may be deleted, truncated, fused to an unknown partner gene at 31 15q21, or influenced indirectly causing changes in normal expression. This work has lead to the discovery of the HACE1 gene which is described in detail in the following chapters. Identification and evaluation of the effects of the translocation on HACE1 in the index case and investigation of whether a similar molecular genetic pattern is commonly observed in other Wilms' tumours. Identification of the function of the HACE1 gene product and evaluation of whether this protein may have an effect, as a TSG or proto-oncogene, in the development of Wilms' tumour. Begin to examine the implications of the H A C E 1 in other malignancies, focusing first on those where 6q21 is affected. 32 CHAPTER II MATERIALS A N D METHODS 2.1 P R I M A R Y TISSUES Frozen Wilms' tumour tissues with patient matched normal kidney samples were obtained from either British Columbia's Children's Hospital or the National Wilms' Tumour Study Group (NWTSG) tumour bank. Normal fetal tissues were obtained from the Birth Defects Research Laboratory at the University of Washington. 2.2 D N A A N D R N A I S O L A T I O N Primary Wilms' tumour, normal tissue, or cell lines (as listed in text) were used as sources of D N A and R N A . Tissues were embedded and frozen in O C T compound then cut into -15 pm sections. 10-15 sections were used in extraction of D N A or R N A . Cell lines were pelleted, washed with PBS and then used for extraction of D N A or R N A . D N A was extracted from these cells using standard methods (178): immersion of cells or tissue in D N A lysis buffer (50 m M Tris p H 8.0,100 m M E D T A , 100 m M NaCl, 1% SDS), proteinase K (Invitrogen) digestion followed by phenolxhloroform extraction, ethanol precipitation and re-suspension in lOmM Tris, I m M E D T A (TE, p H 7.5). Total R N A was extracted using Trizol (Invitrogen) with an acid guanidium thiocyanate phenol/chloroform method (178, 179), as recommended by the manufacturer. 3 3 2.3 S O U T H E R N A N D N O R T H E R N B L O T A N A L Y S I S Southern and Northern blot analyses were performed as previously described (178, 180). For Southern analysis, 10pg of genomic D N A from each sample was digested with Hindlll, BamHl, EcoRI, or Pstl (as noted in text). For Northern analysis, at least 15pg total R N A was used (human multi-tissue and neural-tissue Northern Blot membranes were obtained commercially; BD Clontech). Briefly, a 0.8% agarose gel was used for Southern analysis, while a 1.2% formaldehyde-agarose gel was used for Northern analysis. D N A or R N A was transferred onto nylon membrane filters (HYBOND-N, Amersham) by upward capillary transfer and fixed to the membrane by baking at 80°C for 2 hours followed by U V crosslinking in a Stratagene Stratalinker with the "Autocrosslink" setting. Probes were radiolabeled with [a- 3 2 P]dATP (50pCi; Amersham) using random primer extension (EZ-strip D N A kit, Ambion) followed by nick-column purification (Amersham). The D N A and R N A membranes were pre-hybridized for at least 1 hour in hybridization solution (~5 ml per 10 cm 2 UltraHyb; Ambion) and then hybridized overnight at 42°C with the radiolabeled probes. Blocking of non-specific nucleic acid interactions was enhanced when probing Southern blot membranes by using sheared (sonicated), denatured Salmon testis D N A (Sigma; 100 pg/ml Hyb solution). Membranes were washed the next day in progressively lower concentrations of SSC (2x - O.lx) and SDS (0.1%), 45-55°C and autoradiographed at - 7 0 ° C with intensifying screens using standard methods (180). To control for equal loading of R N A for Northern analysis, membranes were stripped 34 and re-probed with a $-ACTIN c D N A probe, D N A gels were photographed with ethidium bromide staining prior to transfer of D N A to nylon. When required, blots were stripped of radiolabeled probe using the EZ-strip D N A kit (Ambion) chemical cleavage and stripping protocol recommended by the manufacturer. 2.4 F L U O R E S C E N C E IN SITU H Y B R I D I Z A T I O N (FISH) Cell suspensions of normal cultured lymphocytes, and Index case primary Wilms' tumour cells were processed according to standard cytogenetic procedures (171). Metaphase chromosome spreads and interphase nuclei were prepared on glass microscope slides by Nataliya Melnyk in the Sorensen lab and by Cytogenetic Technicians at IWK Children's and Women's Health Centre in Halifax, NS (171) using standard techniques. Prior to use, the slides were passed through a series of room temperature ethanol washes (2 minutes each in 70%, 90% and 100% to dehydrate the slides). The chromosomes were denatured by submerging them in 70% Formamide (Sigma)/2X SSC (pH 7.0) for 2 minutes. The slides were immediately run through a -20°C ethanol series as above. The slides were allowed to air dry before the probe was applied. The D N A probes for FISH (BAC or cosmid) were labelled with either biotin-14-dATP (Roche) or digoxigenin-ll-dUTP (Roche) using a Nick Translation Labelling Kit (Roche) and purified by ethanol precipitation according to the manufacturer's instructions. Centromeric and telomeric probes (Vysis), prelabelled with SpectrumGreen™ or SpectrumRed™ (Vysis) were used for reference. The purified 3 5 probe was then dissolved in Hybridization solution (Vysis), denatured at 75°C for 10 minutes and allowed to pre-anneal at 37°C for 1 hour. The pre-annealed probe was then applied directly to the denatured slides, sealed with a coverslip and rubber cement and allowed to hybridize overnight for at least 16 hours at 37°C in a humidified chamber. After hybridization, the slides were washed in 50% formamide/2X SSC (pH 7.0) at 45°C for 5 minutes, and another wash in 2X SSC at 45°C for 5 minutes. Detection of the signal then proceeded according to the manufacturer's instructions (Roche). The chromosomes were counter-stained in diamidino-2-phenylindole dihydrochloride hydrate (DAPI; Vysis) and visualized using a 100X oil immersion objective through a Zeiss Axioplan II epifluorescence microscope. Images were captured with a C O H U High Performance camera with Quips software (League City, TX). Images were converted to PICT or TIFF files and then processed with Adobe Photoshop prior to printing. 2.5 B A C A N D C O S M I D PROBES Bacterial artificial chromosomes (BACs), from the RPCI-11 human B A C (male, white blood cell) library mapping within specific areas of interest were kindly provided by Dr. Marco Marra from the BC ancer agency, Genome Sciences Centre, Vancouver, BC. Clones were selected from the RPCI-11 tiling path generated with FPC software (181, 182). These clones included: 809N15, 111D22, 139N14, 68M3, 483H13, 36 28504, and 49F2. Other B A C s were obtained f rom the C T D h u m a n spe rm l ibrary (Cal i fornia Institute of Technology; d is t r ibuted by Invitrogen: clone CTD-3221G2) and C h r o m o s o m e 6 c o s m i d l ibrary L A 0 6 N C 0 1 (Los A l a m o s Na t iona l Laboratory; dis t r ibuted by GeneService / Invi t rogen) . A l l probes were conf i rmed to be non-chimeric by F I S H analysis of no rma l metaphases i n our laboratory. B A C s and cosmids were g r o w n and mainta ined i n our laboratory u s ing s tandard methods. Probes were label led w i t h either b io t in or d igoxygen in u s ing a commerc ia l ly available k i t according to the manufacturer 's instructions (see above). B A C and c o s m i d probes were extracted f rom bacteria u s ing a M i d i - or M a x i - p r e p C o l u m n K i t (Qiagen) as per manufacturer 's recommendat ions. 2.6 R T - P C R , Q U A N T I T A T I V E R T - P C R A N D A N A L Y S I S Total R N A (1-2 pg) isolated w i t h T r i z o l reagent (Invitrogen) f rom p r i m a r y t umour samples, tissues or cel l l ines was used as a template for r a n d o m p r i m e d c D N A synthesis u s ing the Superscr ip t II Reverse Transcriptase k i t (Invitrogen) according to the manufacturers ' recommendations. P C R pr imers for the HACE1 c D N A were: (forward) 5 ' - A C C G T G G A G T T G C C C G A G G A T A A T - 3 ' and (reverse) 5 ' - T G C T G C A C A A G G T C A T G G A G T A G T - 3 ' , p r o d u c i n g a produc t of 377bp Quant i ta t ive R T - P C R (qRT-PCR) was done u s ing the T a q M a n 5' exonuclease assay. A l l p r i m e r / p r o b e sets were designed us ing P r imer Select software ( A p p l i e d Biosystems) and spanned exons to el iminate the r isk of template contaminat ion by 37 genomic D N A and the need for DNasel pre-treatment of R N A . PCR primers and a probe specific for the human HACEI gene were used: (forward) T C T T A C A G T T T G T T A C G G G C A G T T , (probe+) [ 6 F A M ] C A A A C C C A C C A T G T G G G A C C C T G [ T A M R A ] , (reverse) C A A T C C A C T T C C A C C C A T G A T . The reaction was optimized by limiting concentrations of primers (primer-limited), as per manufacturer's recommendations for multiplexing, so that it would not excessively deprive a second (control) PCR amplification reaction carried out in parallel. Final concentrations of primers within each reaction were 300 n M (forward & reverse primers) and 150 n M (probe). These were then multiplexed with a VIC-MGB labelled endogenous control primer set (supplied as pre-diluted reagents; Applied Biosystems). A number of endogenous controls were tested including $-ACTIN, 18S ribosomal R N A , TATA-box binding protein (TBP), Glyceraldehyde phosphate dehydrgenase (GAPDH), and beta-2-microglobulin (P2M). For initial screening of primary Wilms' tumours and patient matched normal kidney both TBP and $-ACTIN were used, while fi-ACTIN was used for all other samples. Both these endogenous control reagents gave the most consistent results in our experiments and are widely reported to be favourable for comparison amongst different tissues and cell types (183-185). Quantitative PCR reactions were amplified using TaqMan universal PCR master mix (Applied Biosystems) and standard conditions in an ABI 7000 sequence detection system (Applied Biosystems). Data were further analyzed using Microsoft Excel. Each PCR reaction was performed in quadrupicate, and all cultured cell line samples were analyzed independently at least three times. 38 2.7 B I O I N F O R M A T I C S Bioinformatics tools were used for data analysis , genetic m a p p i n g , gene predict ions a n d funct ional prote in predictions. 2.7.1 GEN SCAN gene prediction software A t the outset of this project the h u m a n genome sequence conta ined a number of large gaps and m a n u a l assembly of port ions s u r r o u n d i n g the 6q21 reg ion was necessary for complete m a p p i n g of the reg ion and for examina t ion of genes possibly affected by the index case t(6;15)(q21;q21) translocation. Sequences cor responding to RPCI-11 B A C 809N15, 111D22 and 285004 o n 6q21 were assembled m a n u a l l y f rom part ial sequences avai lable th rough the Na t iona l Center for Bio technology Information ( N C B I ; h t t p : / / w w w . n c b i . n l m . n i h . g o v / ) , n o w avai lable as Genbank Access ion AL357315. U s i n g B L A S T and 2-way B L A S T tools (http:/ / w w w . n c b i . n l m . n i h . g o v / B L A S T ) it was established that no k n o w n genes were present i n this region. Gene predic t ion was done by inpu t t i ng this sequence into the G E N S C A N w e b tool ( h t t p : / / g e n e s . m i t . e d u / G E N S C A N . h t m l (186,187)). A number of predicted c D N A s were generated (see Chapter III: Table 3-1) a n d pr imers were designed to check for expression of m R N A f rom predic ted c o d i n g sequences. G E N S C A N is an onl ine p rog ram designed to predict complete gene structures i n genomic sequences: exons, introns, promoter, a n d po ly -adeny la t ion signals. It differs f rom the majori ty of exis t ing gene f i n d i n g a lgor i thms i n that i t a l l o w s for part ial genes as w e l l as complete genes and for the occurrence of mu l t i p l e genes i n a g iven 39 sequence, on either or both strands of D N A . G E N S C A N may have certain biases as it is primarily designed on the Burset/Guigo test set of short genes with relatively simple exon/intron structure (188) ). Internal exons are likely to be predicted more accurately than initial or terminal exons, and exons are predicted more accurately than polyadenylation or promoter signals (186). 2.7.2 Human Genome Browser Tools 2 core platforms were used as Human genome browser tools: The Ensembl Human Genome Browser (http: / / www.ensembl.org/) (189-191) and the University of California: Santa Cruz Genome browser (http://genome.ucsc.edu/) (192, 193). The later was used primarily as it appeared to be more user intuitive and customizable to the requirements of this project. The region surrounding HACE1 and the 6ql6.3-6q21 region could easily be graphically illustrated and many surrounding features of the region could be mapped, such as, C p G islands, repetitive sequence elements, transcription factor binding sites, overlay of gene predictions from G E N S C A N and other tools, ESTs, and BACs. Within the UCSC genome browser the B L A T alignment tool (194) was also used to verify sequence clones of HACE1, PCR primer pair alignment on genomic and cDNA, and locations of original G E N S C A N predictions on the most up to date human genome build. 2.7.3 Protein Domain Prediction Tools Three primary tools were used in predictions of functional domains and sequence motifs within predicted coding sequences generated in G E N S C A N . 40 Similarity to known domain architecture would give hints as to the functions of a novel gene product and lack of similarity could help eliminate some false positive coding sequences from our list of G E N S C A N genes. We first used Pfam or Protein Family search tool (http://pfam.wustl.edu/) (195-197) to screen for known functional domains. Pfam is a database of multiple alignments of conserved protein regions or "domains". The- alignments represent evolutionary conserved structures which may suggest a protein's function. The database search tool uses "Profile hidden Markov models (profile HMMs)" from these alignments to help recognize if a new protein belongs to an existing protein family, even if the homology is weak. Unlike standard pairwise alignment methods (e.g. BLAST), Pfam is able to make alignments with multidomain proteins. Use of the online software is relatively straightforward. A translated coding sequence is input and the multiple alignment is generated. Default settings for searching "glocal" and "local" alignments were used with an "E-value" of 1.0 as a cut-off. Detailed descriptions of settings can be obtained at http://pfam.wustl.edu/; however the use of both "glocal" and "local" alignments in our case was particularly useful. Whereas a glocal alignment can search for a "complete" domain, the "local" alignment can recognize partial domains. In our case the "local" alignment of the G E N S C A N prediction for HACEI revealed 3V2 Ankyrin repeats. This partial domain alignment from Pfam suggested that the original G E N S C A N prediction was likely to be incomplete. This was later revealed with the discovery of a longer (full length) expressed sequence tag (EST)/full length c D N A for HACEI. 41 Simi la r d o m a i n archi tec ture /prote in fami ly search tools i n c l u d i n g the S M A R T database (http: / / smart .embl-heidelberg.de/) and I N T E R P r o database ( h t t p : / / w w w . e b i . a c . u k / i n t e r p r o / ) were also used to con f i rm f indings f rom pfam. 2 . 8 P R E P A R A T I O N O F P R O T E I N L Y S A T E S A N D I M M U N O B L O T T I N G A l l cells l ines, i n c l u d i n g those transfected or engineered to stably express HACEI or mutants, were g r o w n un t i l -80-90% confluent (unless otherwise noted) p r io r to col lect ion of cells and lysates. For col lect ion of cells a n d lysates m e d i a was aspirated a n d cells were r insed w i t h PBS, t ryps in ized (0.25% T r y p s i n 50 m M E D T A ; Invitrogen), and collected to the bot tom of a 15 m l conical tube (BD Falcon) b y brief centrifugation (~3 m i n at 1000 rpm). Cel l s were re-suspended i n ice-cold P B S and transferred to a microfuge tube and again collected by centr ifugation (~ 1 m i n at 5000 r p m ; Benchtop microfuge; Hereas instruments). Ce l l s were f inal ly re-suspended i n an appropriate v o l u m e (10 c m d i s h = -500 p l ; 6-wel l = ~ 50 pl) of Lys i s buffer (1.5 m M M g C b , 150 m M N a C l , 50 m M H E P E S , 10 m M N a F , 10 m M Na4P207, 2 m M Na 3 V0 4 , 2 m M E D T A , 10% G l y c e r o l , 0.5% NP-40 , w i t h freshly a d d e d protease inhibi tors: L e u p e p t i n (1:1000 d i l u t i o n of 2 m g / m l stock made i n H2O), A p r o t i n i n (1:1000 d i l u t i o n of l O m g / m L stock made i n H2O), Pheny lmethy l su l fony l F l u o r i d e (PMSF)) . Pro te in concentrat ion was estimated f rom a 10 p l a l iquot of w h o l e cel l lysate (B ioRad Dc prote in assay kit) . Lysates were d i lu ted as required to establish equal concentrat ion across samples be ing ana lyzed and m i x e d (5:1) w i t h 5x l o a d i n g dye (0.5 M T r i s - H C l 42 (pH 6.8 at 25°C) / 10% SDS, 0.05% bromophenol blue, 50% glycerol, 0.5 M DTT) and boiled for 10 min prior to loading. For mini-gel format (7.5 x 10 cm) SDS-PAGE, 30-50 pg of total protein was used per lane. 10% SDS-PAGE mini-gels were run for ~2 hrs (until the bromophenol blue dye had progressed to the bottom of the gel) in a mini-Protean III electrophoresis system (BioRad) at 100-150 Volts according to standard methods (178). Transfer of the proteins from the gel to pure nitrocellulose (BioRad or Pal-Gelmon) or PVDF (Immobilon-P; Millipore) was accomplished with a Mini Trans-Blot cell in Towbin transfer buffer for 1 hour at 0.35 amps. The membranes were stained post-transfer with Ponsceau S stain (Sigma) to verify transfer and check for equal loading. Blocking was done with 5% milk powder in TBST (IX TBS, 0.05% Tween-20) for one hour at room temperature with gentle agitation. Membranes were then incubated with primary antibodies diluted in 0.5% Milk in TBST overnight at 4°C with gentle agitation (see Table 2-1 below). Membranes were washed following primary antibody twice for 5 minute intervals in TBST and then incubated for 1 hour at room temperature with appropriate secondary antibody conjugated to horse radish peroxidase (HRP). The blots were then washed as above prior to visualization by enzymatic chemiluminescence (ECL) (Amersham/Pharmacia) according to the manufacturer's instructions. After E C L detection, the membrane was placed in between two individual acetate sheets, placed into an X-ray cassette and exposed to film for 10 seconds -1 hour. 43 Table 2-1: Antibodies Antibody Distributor dilution secondary 20s Proteasomal Subunit Calbiochem IP Actin Santa Cruz 1:1000 Rabbit AKT Cell Signalling 1:1000 Rabbit Anti-mouse-HRP Cell Signalling 1:3000 Anti-Rabbit-HRP Santa-Cruz Biotech 1:3000 beta-Catenin BD Transduction Labs 1:5000 BiP/Grp78 CDK4 CDK6 Cyclin A Cyclin B Cyclin D1 Cyclin D1/D2 Cyclin E Cyclin G1 Cyclin H ERK 1/2 Grb2 BD biosciences 1:1000 GSK3(3 GST HA Babco 1:2000 HACE1 (mAb) In-house (mouse monoclonal) 1:5000 Mouse HACE1 (mP) In-house (mouse polyclonal serum) 1:1000 Mouse HACE1 (RP) In-house (rabbit polyclonal serum) 1:3000 Rabbit Histone H3 (Acetyl-Lys9-H3) Upstate ChIP rabbit Histone H3 (dimethyl-Lys79-H3) Upstate ChIP rabbit Histone H3 (total) Upstate 1:2000 rabbit Hsc70 IGFRp IRS1 MEK p21 p 2 1 CIP1 p27 K I P 1 PABP In-house (rabbit polyclonal serum) 1:2000 rabbit 44 phospho-AKT (Ser-473) Cell Signalling 1 1000 rabbit Phospho-ERK Cell Signalling 1 1000 rabbit Phospho-GSK3(3 Cell Signalling 1 1000 rabbit Phospho-MEK Cell Signalling 1 1000 rabbit phospho-PKR Cell Signalling 1 1000 Rabbit PKR Cell Signalling 1 1000 Rabbit UbcH7 Boston Biochem Ubiquitin BD Transduction Labs 1:1000 Mouse Ubiquitin Santa Cruz 1:1000 Rabbit VCP 2.9 C L O N I N G & G E N E R A T I O N O F T A G G E D H A C E 1 C O N S T R U C T S The H u m a n HACE1 c D N A was obtained f rom the I . M . A . G . E . consor t ium (h t tp : / / image . l ln l .gov) as part of the M a m m a l i a n Gene Co l l ec t i on ( N I H - M G C Project U R L : h t tp : / /mgc .nc i .n ih .gov) , clone identif icat ion M G C : 27112 I M A G E : 4838835 (Series: I R A K Plate: 34 R o w : o C o l u m n : 20) or Genbank accession number BC034982. H u m a n HACE1 c D N A was received i n a mod i f i ed pBluescr ip t -R vector i n D H 1 0 B E.coli. The clone was sequenced to ensure accuracy u s i n g M 1 3 - f o r w a r d and Ml3 - r eve r se s tandard sequencing pr imers . A l l p lasmids were g r o w n i n s tandard lab strains of E.coli (DH5ot, TOP10 , or D H 1 0 B ; Invitrogen) unless otherwise noted. P lasmids were extracted f rom bacteria after overnight g r o w t h at 37°C i n a shak ing incubator (350 rpm) i n L B bro th u s ing Q iagen m i n i - or m a x i - prep ki ts , depend ing o n scale of p l a s m i d D N A required, according to the manufacturers instructions (Qiagen). 45 2.9.1 pBluescriptll-HACEl Bgli (473) Kprl (4066) Xhd (4047) EccRl (4014) Sail (4041) CM (4033), BffZI (54 X m d (696) \ P s f l (706) ^\ EccRl (708) HACE1 Shyl (1236) Ned (1236) Psfl (717) 5hJ(ii43) X7ioI (3960) &7/II (2688) Figure 2-1: p B S I I - H A C E l Vector map of pBIuescript II(ks+)-HACEl including a selection of common unique and 2 cutting restriction enzymes. The H A C E 1 cod ing sequence wi thou t an H A - t a g was excised f rom p M S C V h y g r o - H A H A C E l by digest ing w i t h EcoRI. The fragment was separated by agarose gel electrophoresis, cut a n d pur i f ied f rom the gel (Qiagen G e l extraction Ki t ) and l igated into pBIuescript II(ks+) (Stratagene) at the EcoRI sit of the po ly l inker . 4 6 2.9.2 pET-15b-(HIS)-HACEl EccRl (8927) AfluK.4489) £ccRV(4043) Figure 2- 2: pET-15b-(His)HACEl Vector map of pET-15b-(His) H A C E I including a selection of common unique, 2 cut, and 3 cutting restriction enzymes. HACEI was excised from pBluescript-R (image clone: 4838835) by digesting with AscI and BsaAI (double digest). Following the restriction digest all fragments were rendered "blunt-ended" with T4 D N A polymerase (NEB) using standard methods. The HACEI fragment was separated by agarose gel electrophoresis, cut from the gel and purified (Qiagen Gel Extraction kit). The recipient plasmid, pET-15b, was cut with Xhol and ends rendered blunt with T4 D N A polymerase. HACEI was then ligated in-frame to the N-Terminal His tag in pET-15b at the blunted Xhol site. 47 2.9.3 pcDNA3HA-HA CE1 Sad (4287) Figure 2- 3: pcDNA3HA-HACEl Vector map of pcDNA3HA-HACEl including a selection of common unique and 2 cutting restriction enzymes. HACE1 was excised from pBluescript-R (image clone: 4838835) by digesting with Ascl and BsaAl (double digest). Following restriction digest all fragments were rendered "blunt-ended" with T4 D N A polymerase (NEB) using standard methods. The HACE1 fragment was separated by agarose gel electrophoresis, cut from the gel and purified (Qiagen Gel Extraction kit). The recipient plasmid, p c D N A 3 r i A , was cut with Xhol and ends rendered blunt with T4 DNA polymerase and gel purified. HACE1 was then ligated in-frame to the N-Terminal His tag in P C D N A 3 H A at the blunted Xhol site. Note: P C D N A 3 H A is a modified version of p c D N A 3 (Invitrogen) with a single N-48 terminal hemagglu t in in (HA) tag inserted at the HmdII I site and the Ndel site deleted w i thou t affecting C M V promoter act ivi ty (Received as a generous gift f rom Dr . Valen t ina E v d o k i m o v a ) . 2.9.4 pMSCVhygro- HAHACE1 Xmd (399) EccRI (4893) Figure 2- 4: pMSCVhygro-HA-HACEl Vector map of pMSCVhygro-HA-HACEl including a selection of common unique and 2 cut restriction enzymes. H A - H A C E 2 was excised f rom p c D N A 3 H A - H A C E l us ing SacII, retr ict ion ends were b lun ted u s ing T4 D N A polymerase and the fragment of interest was gel pur i f ied us ing s tandard methods. The recipient p l a smid , pMSCVhygro (BD Clontech) was l inear ized w i t h Hpal, and treated w i t h C a l f Intestinal Phosphatase (CIP; NEB) to prevent re- l igat ion of empty vector o n blunt ends. HA-HACE1 was then l igated into 49 pMSCVhygro us ing T4 D N A ligase and checked for proper orientat ion w i t h two i n d i v i d u a l restr ict ion digests: E c o R I a n d Bglll (not shown). 2.9.5 Site-Directed Mutagenesis (HACE1-C876S) In order to generate a non-functional , or ligase dead, ve r s ion of H A C E I the conserved C- te rmina l cysteine residue was muta ted to serine, prevent ing the format ion of a thio-ester b o n d between H A C E I and ub iqu i t in . The Q u i c k C h a n g e k i t (Stratagene) (primers: fo rward , T C A A G C A C A T C C A T C A A C A T G ; reverse, C A T G T T G A T G G A T G T G C T T G A ) was used according to manufacturers recommendat ions (198). In order to facilitate the mutagenesis P C R reaction the 3' Xbal fragment of p c D N A 3 H A - H A C E l was subcloned into the Xbal site of pBSII(ks+). This smal l fragment of H A C E I is easier to mutate u s ing the site directed mutagenesis ( S D M ) kit . After muta t ing the requi red base pairs w i t h the S D M k i t the insert was sequence ver i f ied and shutt led back into the p c D N A 3 H A - H A C E l vector (now p c D N A 3 H A -C876S). The muta t ion and orientat ion i n the fragment was again ver i f ied by sequencing. The muta t ion d i d not introduce any k n o w n restr ict ion enzyme recogni t ion sites and this construct c o u l d then be m o v e d to p M S C V i n the same manner as the wi ld - type H A C E I as described above. 2.10 C E L L C U L T U R E A l l cel l l ines were main ta ined at 37°C i n a 5% CO2 h u m i d i f i e d incubator. A n y w o r k done w i t h cel l l ines was conducted w i t h i n a Class B biosafety cabinet w h i l e observ ing aseptic techniques. In general, cel l l ines were passaged i n fresh complete 50 media every 3 days, as required, or as otherwise specified for individual experiments. To passage, cells were gently rinsed once with lx PBS and then trypsinized (a small volume of 0.25% Tryps in /EDTA (1-3 ml); Invitrogen) to detach cells from culture flasks/plates and each other. Trypsin was neutralized with an equal volume of complete media and cells were collected by brief low speed centrifugation in a conical vial. Cells were then split at a ratio between 1:3 and 1:10 depending on the cell line and re-plated into new, sterile culture plates or flasks. See Table 2-2 for complete media formulations (all media and supplement from Invitrogen unless otherwise noted). Table 2- 2: Media Formulations Cell Lines Base Media Serum Other supplements HEK293, 293T (and derivates), 293FT, PhoenixA, Bosc23, MEFs, SK-N-MC, SK-N-SH Dulbelco's Modified Eagle Medium (DMEM) 10 % Fetal Bovine Serum (FBS) 1% Penicillin, Streptomycin, and Fungizone® (PSF; antibiotic-antimicotic mix) SKNEP1 (and derivates) DMEM 15 % FBS 1 % P S F 1% non-essential amino acids NIH3T3 (and derivates) DMEM 10 % Calf Serum (CS) 1 % P S F IMR-32 (and derivates), SAN-2, Birch, CT-10, KCNR RPMI 1640 15% FBS 1 % P S F 51 2.11 R E T R O V I R U S C O N S T R U C T I O N 2.11.1 MSCV based retroviruses Derivatives of murine (NIH3T3) and human (293T, SKNEP1, and IMR-32) cell lines were generated to stably express HA-tagged H A C E 1 using retroviruses based on the Murine Stem Cell Virus (MSCV; BD Clontech) (as described in (198-200)). Briefly, pMSCVhygro vectors with H A C E 1 were generated as described above. Vectors were transfected into Phoenix A packaging cells (Orbigen) at ~60% confluence using FuGENE6 (Roche) according to the manufacturers' protocol, in the presence of serum, Chloroquine (10 pM; Sigma), HEPES (50 m M , p H 7.5; Invitrogen), and 1% non-essential amino acids. PhoenixA cells were selected for packaging as they contain an amphitrophic envelope (ENV) protein that is integrated onto the recombinant retroviruses. This allows them to infect a broad range of host cells, including murine and human cell lines. Media was changed after ~8 hr with supplements: HEPES (50 m M , p H 7.5), and 1% non-essential amino acids. Virus containing supernatant (spent media) was collected 48-60hrs post transfection, filtered (0.45 pm), and mixed 1:1 with complete media (appropriate for cells to be infected, refer to Table 2-2) and 1 pg/ml polybrene (Sigma). Target cells at ~20-50% confluence were infected with virus/media mixture for at least 24 hrs. At 48 hrs post infection target cells were selected using 100-1000 pg/ml Hygromycin B (Invitrogen) for 5-10 days, or 1 day longer than required to kill off all mock infected cells. Dose response curves for target cells were generated for 52 each cel l l ine p r io r to infection to ensure an efficient concentrat ion of selective antibiotic (not shown). 2.11.2 Lentivirus based retroviruses & Gateway cloning Stable reduct ion of H A C E I expression i n H E K 2 9 3 cel l l ines by R N A interference was performed us ing the Block- iT Len t i v i r a l R N A i k i t (Invitrogen) according to manufacturer 's protocols. A l l R N A i sequences were compared w i t h the h u m a n genome sequence us ing the B L A S T and B L A T tools to rule out significant h o m o l o g y to non-target sequences. Brief ly, H A C E I specific double s t randed short h a i r p i n D N A oligos ( N I : G T A T A G C G C T G A T G T C A A C A T T G A G A A A T G T T G A C A T C A G C G C T A T A ) , or non-specific " sc rambled" (mismatches i n lowercase) ol igos (SI: Gatat c g G C T aggtc t A A C A T T G A G A A A T G TTaga c c t A G ccgat A T ) were c loned into pre- l inear ized p E N T R / U 6 ( included w i t h kit) . Short h a i r p i n constructs were then transferred to the len t iv i ra l packag ing construct p L e n t i 6 / B l o c k - i T - D E S T us ing L R clonase as out l ined i n the manufacturers ' protocols. Lent iv i ruses were p roduced by lipofectamine-2000 co-transfection of p L e n t i 6 / B l o c k - i T - D E S T constructs w i t h p L P l , p L P 2 , and p L P / V S V G (Vi rapower support ki t , Invitrogen) into 293FT cells (Invitrogen) for packaging of v i rus . V i r a l supernatants were collected after 72 hrs, fil tered (0.45 pm), a n d placed und i l u t ed o n H E K 2 9 3 target cells for 24 hrs w i t h 6 p g / m l polybrene (Sigma). Ce l l s were selected w i t h Blas t i c id in (6-10 p g / m l ) for 7-10 days, un t i l a l l mock-infected cells had d ied . Express ion of H A C E I was assayed by 53 Western blot and q P C R before and after i n v i t ro and i n v i v o experiments to ensure stable k n o c k - d o w n (see Chapter V ) . 2.12 R N A I N T E R F E R E N C E O F HACEI Transient R N A interference u s ing s i R N A s was achieved us ing the N I (sense: U A U A G C G C U G A U G U C A A C A ( T T ) ; antisense: U G U U G A C A U C A G C G C U A U A(TT)) , N 2 (sense: G G U C U G U U U C U G A A C U A C U(TT) ; antisense: A G U A G U U C A G A A A C A G A C C(TT)) and S I (aua ucg G C U agg u c u A A C A ( T T ) ; antisense: U G U U a g acc u A G C c g aua u(TT), a manua l ly scrambled vers ion of N I w i t h no k n o w n homology to h u m a n genes) sequences i n an synthetic s i R N A fo rm (Invitrogen). Long- t e rm transient R N A i was achieved us ing synthetic mod i f i ed Stealth R N A i molecules ("stealth select R N A i " ; HSS126385 (385), HSS126386 (385), HSS126387 (387), and l o w G C content control (ctrl); sequences available at http: / /wrww. inv i t rogen .com/ rna i ) . A l l o l igo s i R N A s were transfected into cells us ing Lipofectamine 2000 (Invitrogen) as per the manufacturer 's protocols. Express ion of H A C E I was ver i f ied by Wes te rn blot at the start and end of each experimental t ime point to ensure consistent k n o c k - d o w n . For cell l ines w i t h stable R N A i k n o c k d o w n of H A C E I refer to the section above (2.11.2 Lentivirus based retroviruses & Gateway cloning) 54 2.13 E P I G E N E T I C A N A L Y S I S 2.13.1 5-AZ Treatment, Methylation Sensitive Restriction Digests & PCR To assess the effect of C p G i s land methyla t ion o n HACE1 expression efficiency, S K N E P 1 cells were treated w i t h methy la t ion inhibi tor 5 - A Z (Sigma). Treatment was done twice for 24 h over a 7-day pe r iod (on days 2 and 5) as p rev ious ly described (198, 201). D N A and R N A were isolated o n day 6 and HACE1 expression and methy la t ion of the ups t ream C p G islands were assayed by q R T - P C R (as described above) and semi-quanti tat ive P C R (described below), respectively. M e t h y l a t i o n sensitive restriction enzyme digests fo l l owed by semi-quanti tat ive P C R have been performed as described (198, 202). Brief ly, 100 n g of D N A , extracted f rom tumour and no rma l k idneys frozen tissue sections, was digested to comple t ion (overnight) w i t h EcoRI + Haell, EcoRI + BssHll or EcoRI + Hpall (NEB) as indicated i n Figure 6. 10 n g of digested D N A was used for semi-quantitative P C R w i t h ampl icons su r round ing C p G islands near the HACE1 locus (see Chapter III; F igure 3-6). The 5' reg ion of the M J C 2 gene, p rev ious ly s h o w n to be unmethyla ted (203), was used as a control for complete digest ion. A l l reactions were per formed us ing "hot start" P l a t i n u m T a q (Invitrogen) w i t h the i n c l u d e d buffer, supplemented w i t h 10% D M S O , and carr ied out w i t h the f o l l o w i n g condit ions: C p G - 2 9 (794 bp forward: G G A A A C A A A A G C A A A G C G A C C C A A C T A T , reverse: G G C G G C C G A G A C C T G A G A C C , 100 n M each), 50 m M d N T P s , 1.0 m M M g C l 2 , 9 4 °C 30 s, 56 °C 10s, 72 °C 60 s, 30 cycles. C p G - 1 7 7 (804 bp, fo rward : T G T G C T G T T C G G A A T G A T G T , reverse: C T A G C C T G G G 55 T G T G A G A G G G , 100 n M each), 100 m M d N T P s , 1.5 m M M g C h , 94 °C 45 s, 58 °C 20 s, 72 °C 120 s, f ive cycles, 94 °C 30s, 56 °C 10s, 72 °C 90 s, 30 cycles. C p G - 8 8 (939 bp, fo rward : C C C C G A T G C A G C T T A A A G T A , reverse: G A G G G T A G G A G G A G C A G G G , 100 n M each), 50 m M d N T P s , 1.0 m M M g C l 2 , 94 °C 45s, 58 °C 10s, 72 °C 90 s, 30 cycles. MIC2 (412 bp , fo rward : A G T A T C T G T C C T G C C G C C , reverse: T T T G C A A C T C C G A C A A C A A A C G C , 100 n M each), 50 m M d N T P s , 1.0 m M M g C l 2 , 94 °C 45 s, 55 °C 15 s, 72 °C 45 s, 35 cycles. The op t ima l number of cycles for each reaction was determined empi r i ca l ly by testing a range of 25-40 cycles. The condi t ions above are those i n w h i c h the most reproducible differences i n ampl i f ica t ion w i t h i n the exponent ia l phase of ampl i f ica t ion were achieved. 2.13.2 Chromatin Immunoprecipitation (ChIP) S K - N E P - 1 or H E K 2 9 3 cells were treated w i t h or w i t h o u t 5 m M 5 - A Z as described above. Immunoprec ip i ta t ion of D N A representing active versus inactive chromat in was performed us ing a C h I P assay k i t (Upstate) according to the manufacturer 's protocol : D N A and prote in were cross- l inked i n culture w i t h 1% formaldehyde. For each reaction, 10 6 cells were collected i n 200 m l S D S lysis buffer (1% SDS, 10 m M E D T A , 50 m M T r i s - H C l , p H 8.1, 1 m M P M S F , 1 m g / m l aprot in in , 1 m g / m l pepstat in A ) and sonicated to shear genomic D N A to a size range of -300-1000 bp. Lysates were d i lu t ed 1:10 w i t h C h I P buffer (0.01% S D S , 1.1% T r i t o n X-100,1 .2 m M E D T A , 16.7 m M T r i s - H C l , p H 8.1, 167 m M N a C l ) and immunoprec ip i t a ted w i t h the appropriate ant ibody overnight at 48°C w i t h agitation. Ant i -Ace ty l -h i s tone H 3 56 (Lys9/Lysl4) antibodies (Upstate) were used to immunoprecipitate active chromatin fragments (204), whereas anti-dimethyl-histone H3 (Lys79) antibodies (Upstate) were used to immunoprecipitate inactive chromatin (205) with protein A beads. Protein A agarose/Salmon sperm D N A slurry (Upstate) added to lysates without antibody was used as a negative control (data not shown). D N A was recovered by proteinase K digestion and phenol/chloroform extraction after washing and reversal of cross-links. Equal volumes were utilized to maintain the quantitative nature of the assay. The recovered D N A pellet was air dried and re-suspended in 25 ml of T E (lOmM Tris -HCl , 1 m M E D T A , p H 8.0). Immunoprecipitated HACE1 levels were determined by using semi-quantitative PCR amplifying the 5' end of the gene, overlapping the CpG-88 island. PCR was carried out with Platinum Taq (Invitrogen) with the included buffer, supplemented with 10% DMSO, and the following conditions: 2.5 ml recovered ChIP D N A , primers (297 bp product, forward: C G G C T C A C C C T C G G G C A A C T C C , reverse: C G G C G G C G G G T G T A C T G T A G G T G G T C, 200 n M each), 100 m M dNTPs, 2 m M MgC12, 94 °C 30 s, 55 °C 30 s, 72 °C 45 s, 32 cycles. Input D N A amounts were verified by PCR using the same conditions as mentioned using D N A extracted from a 100 ml aliquot of 'ChIP buffer' diluted cell lysate taken prior to addition of antibodies, as recommended by the ChIP assay kit (Upstate). Input was also verified with 2.5 ml of recovered immunoprecipitated D N A using the a PCR assay for the 5' end of the unmethylated MIC2 gene (202, 203), as described earlier. The optimal number of cycles for each reaction was determined empirically by testing a range of 25-40 cycles. The 57 conditions mentioned above are those in which the most reproducible differences in amplification within the exponential phase of amplification were achieved. 2.14 S U B C E L L U L A R L O C A L I Z A T I O N BY I M M U N O F L U O R E S C E N C E Exponentially growing cells were plated on coverslips in the well of a 6-well dish. After the cell had adhered (at least overnight), they were rinsed with PBS and fixed with -20 °C cold methanol for 10 min. To detect Hacel and BiP, coverslips were incubated overnight with a mixture of rabbit a-Hacel (1:2000 dilution) and mouse a-BiP antibodies (1:1000 dilution) followed by the secondary antibodies Rhodamine Red-X-conjugated goat a-rabbit and Oregon Green 514-conjugated goat a-mouse (Molecular Probes). Slides were counterstained with DAPI and analyzed using a Zeiss Axioplan epifluorescence microscope equipped with a C O H U - C C D camera. Control staining with pre-immune antibodies or no primary antibodies showed no signal (data not shown). 2.15 B I O C H E M I C A L A S S A Y S 2.15.1 Recombinant HACEI purification H A C E I was expressed as His-tagged fusion proteins in Escherichia coli BL21(DE3) (Invitrogen), using the pET-15b vector (see map above, Figure 2-2) and recombinant protein purification system as per the manufacturer's protocols. After 4 h of IPTG induction, bacteria were lysed by sonication in buffer containing 2M NaCl, 10 m M Tr i s -HCl (pH 7.6) and 0.5 m M PMSF. Cell debris was removed by centrifugation 58 at lOOOOg for 20 m i n at 48 ° C The supernatant was d i lu t ed 4-fold w i t h 10 m M T r i s -H C l ( p H 7.6), 0.5 m M P M S F and passed th rough a N i + - S e p h a r o s e c o l u m n (Qiagen). After w a s h i n g the c o l u m n w i t h l o a d i n g buffer [500 m M N a C l , 10 m M T r i s - H C l ( p H 7.6)], b o u n d proteins were eluted w i t h the same buffer conta in ing 300 m M imidazo le and d i a l y z e d against 200 m M K C 1 , 1 0 m M T r i s - H C l ( p H 7.6). 2.15.2 Thio-ester bond formation assay L-[ 3 5 S]-Meth ion ine (Amersham) label led H A C E 1 prote in was synthesized i n v i t ro i n a coup led t ranscr ip t ion/ t rans la t ion system (Promega) u s i n g p c D N A 3 - H A -H A C E 1 p lasmids c o d i n g for the w i ld - t ype and C876S mutant H A C E 1 proteins. React ion mixtures contained 2 p l of the translat ion reaction, G S T - U b (1 mg) , 50 m M K C 1 , 5 m M MgC12, 2 m M A T P , 10 m M creatine phosphate, 0.5 U of creatine phosphokinase, 1 m M D T T , and 20 m M T r i s - H C l ( p H 7.8). Reactions were also supplemented w i t h E l (0.5 mg) and the respective E2s (as noted i n F igure 3-11; 0.4 mg) (Boston-Biochem). After incubat ion for 30 m i n at 30°C, reactions were s topped w i t h 2 x S D S - P A G E sample buffer i n the absence or presence of 300 m M (3-mercaptoethanol. Reactions conta in ing p-mercaptoethanol were then bo i l ed for 3 m i n , and those l ack ing P-mercaptoethanol were incubated at r o o m temperature for 20 m i n before load ing . React ion products were resolved by S D S - 7 % - P A G E and v i sua l i zed by autoradiography overnight at -70°C w i t h intensifying screens. 2.16 G E N O M I C D N A S E Q U E N C I N G 59 2.16.1 Primer Design and Sequencing Genomic D N A from Wilms' tumours, patient matched normal kidney, and controls was extracted as described above. Sequencing and analysis was done in collaboration with the laboratory of Dr. Angela Brooks-Wilson at the BC Cancer Agency, Genome Sciences Centre. Primers for the studies were designed using the "Primer3" Web tool (http: / / www-genome.wi.mit.edu/ cgi-bin/primer/primer3 www.cgi). Forward primers had the -21M13F sequence ( T G T A A A A C G A C G G C C A G T ) added to them and the reverse primers had the M13R sequence ( C A G G A A A C A G C T A T G A C ) added to facilitate later sequencing of the PCR products. Al l primers were ordered at 10 n M scale from Invitrogen Life Technologies. PCR reactions were carried out in a volume of 20pl containing lOng genomic D N A , I m M MgSC>4, each primer at 0.5uM, 2mM (each) dNTPs, lx supplied Pfx Amplification Buffer and 0.25 units Platinum Pfx D N A polymerase (Invitrogen). A programmable thermal cycler (MJ Research PTC-225) was used for the PCR reactions which have a total of 30 cycles (94°C - 30s, 50-65°C - 30s and 68°C - 60s). A 5ul Aliquot of each PCR reaction was run on a 2% agarose gel to confirm amplification and quality of the product (178, 180). The remaining 15pl of PCR product was purified with Ampure magnetic beads (Agencourt Bioscience Corporation) and eluted in a volume of 30ul of T E (10 m M Tris p H 7.5; I m M E D T A p H 8.0). These 30 ul purified D N A samples were then cycle sequenced using Big Dye Terminator Mix V.3 at 0.25x chemistry in a total volume of 4ul (Applied Biosystems). Both forward (-21M13F 60 primer) and reverse (M13R pr imer) directions were sequenced. C y c l e sequencing reactions consisted of 30 cycles (96°C - 30s, 52°C - 5s (-21M13F) or 43°C - 5s (M13R), and 60°C - 3min) . U p o n comple t ion , reactions were precipi tated w i t h I sopropyl A l c o h o l , washed once w i t h 70% Ethanol , and a l l o w e d to air dry . Each reaction was then re-suspended i n 8pl of sterile d d F b O . C o m p l e t e d cycle sequencing reactions were loaded on ABI-3700 capi l lary sequencers to collect sequence data. 2.26.2 DNA sequence assembly & data analysis A n appl ica t ion was created to help facilitate the extraction of P C R sequence reads f rom the laboratory Information Management Sys tem (LIMS) and to l i n k sample and exon informat ion to the i n d i v i d u a l sequences. A number of Per l scripts and modules facilitate a p ipel ine where the reads are g rouped together and assembled. First, trace data was m o v e d to specific project directories and read names were mod i f i ed to help identify i n d i v i d u a l patient and specific exon regions. Sequence reads were then base-called w i t h Ph red and sequence reads were assembled a long w i t h reference sequences u s ing Phrap (206, 207). The reference sequences consisted of both exon sequence and a short stretch of sequence f l ank ing the exons. Cont igs were created that corresponded to each exon reg ion of the gene of interest. F ina l ly , P o l y P h r e d (208) was used to examine sequence traces and detect heterozygotes and the contigs were v i sua l i zed i n C o n s e d (209) to verify S N P s . Custom software and support was provided by Yawn Butterfield in the laboratories of Dr. Angela Brooks Wilson and Dr. Marco Marra at the BC Cancer Agency Genome Sciences Centre. 61 CHAPTER III D I S C O V E R Y O F A N O V E L E3 UBIQUITIN-PROTEIN LIGASE G E N E A T A t(6; 15)(q21; q21) T R A N S L O C A T I O N IN SPORADIC WILMS' T U M O U R The project described in this chapter was conceived and initiated by the laboratory of Dr. Poul H.B. Sorensen wi th information on a case study provided by Dr. Conrad Fernandez at the IWK Children's and Women's Health Centre, Halifax, N.S. Portions of this chapter have been published in two peer reviewed publications. In the later publication I generated the majority of the data, and contributed to the writing: Fernandez, C.V. , Lestou, V.S., Wildish, J., Lee, C L . and Sorensen, P .H . (2001) Detection of a novel t(6;15)(q21;q21) in a pediatric Wilms' tumour. Cancer Genet Cytogenet, 129,165-7 Anglesio, M.S., Evdokimova, V . , Melnyk, N . , Zhang, L. , Fernandez, C.V. , Grundy, P.E., Leach, S., Marra, M . A . , Brooks-Wilson, A.R. , Penninger, J. and Sorensen, P .H . (2004) Differential expression of a novel ankyrin containing E3 ubiquitin-protein ligase, Hacel, in sporadic Wilms' tumour versus normal kidney. H u m M o l Genet, 13, 2061-74. Acknowledgements for the data contained in this chapter that was generated by other members of the Dr. Poul H.B. Sorensen's Lab and collaborators is noted in the figure legends. 3.1 I N T R O D U C T I O N A number of clinical observations suggest that the genetics of Wilms ' tumour are complex. These include: the occurrence of bilateral Wilms ' tumour in only 5-10% of affected individuals; the high incidence of Wilms ' tumour in specific congenital syndromes, such as Denys-Drash syndrome; and the recognized phenomenon of familial Wilms ' tumour kindreds (210). In addition, few non-random cytogenetic alterations have been reported in Wilms ' tumour, mainly involving the W T l gene on chromosome H p l 3 and the WT2 locus on chromosome H p l 5 (4, 174, 211-213). 62 Although WT1, WT2 and other identified loci appear to be very important in familial and syndromic forms of Wilms' tumour, there is currently little evidence that they are commonly mutated in sporadic Wilms' tumour (reviewed in (18)). Germline WT1 mutations occur at a low frequency 0.8-4.6% of sporadic cases (42, 214), while non-constitutional mutations, occurring at the tumour site, are observed in 10-18% of sporadic cases (215-217). In addition, a specific gene has not been clearly identified at the WT2 locus. Recurrent abnormalities on 16q, lp , and 7p detected by C G H , and more recently high resolution array C G H (aCGH), suggest that other chromosomal regions may harbour Wilms' tumour genes (210, 212, 213, 218-221). We identified a non-constitutional t(6;15)(q21;q21) translocation in a sporadic Wilms' tumour occurring in a 5 month old boy (171, 198). In this case a Caucasian male without congenital anomalies or evidence of an overgrowth syndrome was found to have a radiographic stage I Wilms' tumour. He had a complete nephrectomy and the resected mass weighed 253 g. The pathology was that of a favourable histology; stage I Wilms' tumour without anaplasia. There were no nephrogenic rests apparent in the adjacent normal kidney tissue. He received vincristine and actinomycin as per the National Wilms' Tumour Study V protocol and after 12 months off therapy remains in remission (171). This t(6;15) translocation is a novel finding in this disease, however the 6q21 region has previously been involved in other Wilms' tumour translocations, including a t(5;6)(q21;q21) and a t(2;6)(q35;q21) (172-174). Further, abnormalities on 6q have 63 been widely reported in a number of malignancies (175, 176, 222, 223) and 6q21 specifically has been suggested to contain one or more tumour suppressor genes (131). This suggested that aberrations at 6q21 might represent a recurrent alteration in Wilms' tumour. The t(6;15) translocation in the index case might have led to an oncogenic gene fusion, as has been previously described in pediatric malignancies (116, 224-230). Alternately, such a rearrangement could result in the inactivation of a tumour suppressor gene (TSG) or the activation of a proto-oncogene. We wished to identify genes surrounding this common 6q breakpoint locus and establish a link to Wilms 1 tumour. 3.2 R E S U L T S 3.2.1 Cytogenetic Analysis G-band karyotyping was performed at IWK Health Center, Halifax Nova Scotia, where the nephrectomy was also performed, using standard methods (171). A 46,XY,t(6;15)(q21;q21) karyotype was demonstrated in 49 of 64 metaphases analysed from cultured tumour cells ((171) and Fig. 3-1). The remaining 15 cells had a near tetraploid complement with 2 to 3 copies of the t(6;15)(q21;q21). Normal kidney tissue did not demonstrate the translocation but curiously 25% of the 100 normal metaphases analyzed had a near tetraploid complement. A balanced reciprocal translocation involving chromosomes 6 and 15 was confirmed in the majority of diploid tumour cells by fluorescence in situ hybridization analysis using whole chromosome painting probes (Fig. 3-2). 64 Figure 3-1: Cytogenetic Identification of a t(6;15)(q21;q21) translocation G-banding karyotype from a specimen of the Wilms' tumour index case showing the reciprocal and balanced t(6;15)(q21;q21). G-band karyotyping was performed at the IWK Grace Health Centre for Women's and Children's Health, Halifax, N.S. 65 Figure 3- 2: Chromosome Painting of a reciprocal t(6;15)(q21;q21) translocation Metaphase cell preparations of the Wilms' tumour case, counterstained with DAPI (blue, cell DNA). Whole chromosomal probe 6 is labelled with SpectrumGreen and whole chromosomal probe 15 is labelled with SpectrumRed. Diploid clone of the Wilms' tumour case demonstrating the balanced and reciprocal t(6;15) by fluorescence in situ hybridization. Chromosome painting was carried out by Dr. Valia Lestou in Dr. Poul Sorensen's lab (171). 6 6 3.2.2 Identification of the Breakpoint Region by FISH, Southern Blotting, and In Silico Mapping I focused on the 6q21 arm of the translocation in the search for candidate genes affected by the t(6;15) translocation. The 6q21 locus has been previously described in at least 2 other Wilms' tumour associated translocations. Further, 6q21 has also been implicated in rearrangements, deletions (or loss of heterozygosity; LOH) and chromosome breakages in a number of other tumours and is thought to contain one or more tumour suppressor genes. Fluorescent In-Situ Hybridization (FISH) using metaphase nuclei from the index case and fluorescently labelled bacterial artificial chromosome (BAC) probes from the 6q21 region were used to map the t(6;15) translocation breakpoint in our Wilms' tumour index case (198). To begin, a tiling path of BACs surrounding the 6q21 region was designed using the RPCI-11 human B A C library (231, 232) and Finger Printed Contigs (FPC) software (181, 182, 233) (see Figure 3-3). Individual BACs were then tested in pairs using two colour FISH: closing in from the centromeric and telomeric sides simultaneously until a breakpoint spanning B A C clone was found. Once a clone spanning the breakpoint had been identified (Figure 3-4), a number of other B A C clones with overlapping sequence from the RPCI-11 and other B A C libraries (see Chapter II: Materials and Methods) were then used to tile a minimal overlapping region. 67 Whole Zoom: in Out 2.0 Hidden: Buried Configure Display Clone: EditContig B | Clear All Merge Analysis snCtgCheck snCtgStep 4-D4S3098 RH91B58 RH119552 6-SHGC-9237 15-015S895 6-AL024505 RH29556 6-stSG51691 15-W1-13387 KAM33 15-W1-20S19 RH28685 RH76145 RH64987 7-SWSS2140 N0257E13 N0714KJ3 N0590G06' N0438N24, VPO244F01 N0629K18 N0771N03" N0705N02 •023O12 ,7103 K07* 1H04-M0784C06 N0706F18 N0421G18-M2310F18 N0442A21 N0666E05 N0080G04 ePO408F19 VP0112L06 N0798E24 N0300A01" N0157E23 N0374A07 N0548J13 eP0405B04 N0517M17 M0115P17 N0194O10' N0703E02 M2015G03 N0380P19 l_ N0318F03-14 N0639C2V N04S0H08 N0294N20* N0444D08-N0409E17 N0781E04-N0652M17-N0002O21 6SHGC-7137 6-stSG83371 RH81133 SHGC-15815 I N0271J19 N0159G14 N0635P23-N0360I1T i N02S4A16-I N0408M22-. N0427EW 6-AFMAI25XC1 6-SISG24757 6-D6S1945 RH29293 6-WI-15170 6-RM111544 RH123473 RH108187 6-RH121S87 6-Z95329 stdJ128H2T/ 6-stdJ128l12SP6 8stSG48354 6-S1SG6369 6-D6S1563 N0355M14 M2285I14 N0654J0V N0811K05-N0398H08-N0089H08 M0152N17 N o e s s G n -M2009P09 N0065O16 M2010F09 eFW39H4 L-N013W14 1 N0483H14 I NPinMfc i W08O8N1S-! N0285OO4 ] N0023E12 N009SL11 N0764N21-N0742O22 NO78OK07-N0S89K10-N0062OO6 VPO522O02 AC009907:N0438N24 UWF-8q16.3-21 COX_6 AC019308 N0429P14 AC027012:N0398H08 ACO 19002:N0570O10 B-stSG53308 17-stSG28023 6-S1SG24682 6-W1-14937 AC020592:N0159G14 AC024292:N0023E14 JMF-6q21 -22 COX_8 COX_6 AC025348:N0139N14 6-AA219556 6SGC38363 6stSG6256 6-stSG6373 6-SGC34572 6-stSG21922 6-S1SG24663 15-WH3387 15-WI-20619 6-stSG51691 6-stSG63371 6-stSG6369 Figure 3- 3: Finger Printed Contigs (FPC) tiling path FPC alignment of adjacent and overlapping B A C clones based on individual B A C s restriction digests pattern is used to construct the FPC database. Above is the FPC output of the B A C clones covering the 6q21 region. Colour highlighted and boxed BACs were chosen for FISH analysis on tl index Wilms' tumour case (and controls) to map the breakpoint region 68 Figure 3- 4: Fluorescence In-Situ Hybridization (FISH) Metaphase chromosome spreads from index case Wilms' tumour with t(6;15)(q21;q21) translocation. The breakpoint region is highlighted in each panel with green labelled B A C probe 111D22 (top) and 809N15 (bottom). Bottom panel also shows red labelled telomeric Chr. 6 probe. Chromosomes are counterstained with DAPI (blue). The above figure clearly illustrates the t(6;15) translocation. FISH experiments illustrated above were done by Nataliya Melnyk in the laboratory of Dr. Poul Sorensen. A map of the breakpoint region o n 6q21 was constructed us ing over l app ing and non-over lapp ing B A C clones. To further micro-map the breakpoint reg ion a s imi lar strategy was app l i ed for southern blot t ing. In this case sequence was retr ieved f rom Genbank ( h t t p : / / w w w . n c b i . n l m . n i h . g o v / ) a n d sma l l P C R ampl icons were 69 designed to be used as Southern blot probes. Areas of repetit ive sequence were filtered u s ing "Repea tMasker" software (Smit, A F A , H u b l e y , R & Green, P . RepeatMasker Open-3.0. 1996-2004; h t t p : / / w w w . r e p e a t m a s k e r . o r g / ) and exc luded so as to a v o i d potential false posi t ive re-arrangement signals f rom poor ly designed probes. A single probe (SB1) cor responding to a C p G i s land was found to be rearranged i n Psf l digested genomic D N A f rom the index case tumour but not n o r m a l k idney tissue or control D N A samples (Figure 3-5) (198). Southern b lo t t ing w i t h other genomic probes, either more centromeric or telomeric f rom the C p G - 2 9 C p G is land, a n d other restr ict ion enzyme digests d i d not s h o w re-arrangements (data not shown) . A more exact pos i t ion for each B A C was determined later u s ing the complete sequence made available as part of the h u m a n genome project and B A C end sequences i n the genome database (h t tp : / /genome.ucsc .edu/ ) (198). U s i n g a l l of this, a precise m a p of the 6q21 breakpoint reg ion i n our W i l m s ' t umour index case was constructed (Figure 3-6), conta in ing k n o w n genes (identified p r io r to this research project), B A C posit ions, sites p robed for Southern blot re-arrangements, and genomic features (such as C p G islands). 70 Mol. Wt. (kb) 23 10 9.0 8.0 ^ / / Figure 3- 5: The 6q21 breakpoint re-arrangement identified by Southern Blotting Southern blotting using SB1 probe (position as in indicated Figure 3-6) demonstrates re-arrangement of genomic D N A in Pstl digested genomic D N A from the index Wilms' tumour case but not in adjacent normal kidney tissue or in peripheral blood from two independent control samples (control 1 and 2). Normal digest pattern is apparent with the SB1 probe hybridization (indicated with solid black arrows) while re-arrangement at the 6q21 locus is apparent only in index case Wilms' tumour D N A (indicated with dashed arrow). 71 Index case 116; 15) reciprocal translocation breakpoint January 2000 Genscan CDSA2RF5 IMAGE clone 4838835 November 2006 LIN28B I I I 1 HACE1 Figure 3- 6: The 6q21 breakpoint locus Detailed map of the 6q21 breakpoint region in the index Wilms' tumour case: BACs used for mapping of the breakpoint region were identified using Finger printed contigs (FPC) software and sequence information from the NCBI and UCSC genome databases. BACs spanning the t(6;15)(q21;q21) breakpoint are indicated by double green lines; BACs that do not span the breakpoint are indicated by single black lines. Green boxes denote CpG islands while the "SB1" southern blot probe generated by PCR is indicated by double purple line at C p G island "CpG-29". The G E N S C A N predicted ORF/CDS 1 and 5 (red; see also Figure 3-7) and I M A G E clone (blue) corresponding to the novel HACE1 gene (IMAGE Clone 4838835) are shown. Approximate position of exons and genomic features are calculated from the human genome browser tool. The lowest panel shows the details available from NCBI data available in January 2000, at the beginning of this project, and the known genes in the region in November 2006 as indicated by the UCSC genome database (http://genome.ucsc.edu/). At the time this work was ongoing, a contiguous sequence for the human genome was not yet available and a number of gaps were present at the 6q region. Exhaustive BLAST searches were required using repeatmasked segments of the partly sequenced, breakpoint spanning BAC, 809N15 and others, against the Genbank non-redundant (nr) and human EST databases. Using this strategy it was hoped that a candidate gene affected by the t(6;15) translocation would be identified, while reducing the number of false positive "hits" in repeat rich genomic sequence. 72 Unfortunately no known genes could be confirmed to be present in this region (data not shown). 3.2.3 Gene Prediction and Validation Although no previously characterized transcripts were identified from database searches, we continued to look for novel coding sequence with an in silico strategy. As the contiguous genomic sequence of the region was made available through the Human genome project over the course of our studies, we were able to manually assemble the region and test gene prediction software over our expanding contig. G E N S C A N (35, 36) gene prediction software was used primarily to search for novel coding transcripts. A number of coding sequences (CDS) were identified with initial screens (See G E N S C A N output: Figure 3-7). Several of these CDSs were homologous to repetitive sequence elements found throughout the genome and were discounted from further study. Two of the CDSs showed some level of similarity to known genes when the sequence was translated. At the time no ESTs in the Genbank database matched the predicted sequences, therefore, in order to validate whether or not these CDSs were indeed bona fide expressed sequences a number of RT-PCR primer pairs were designed based on the original G E N S C A N predictions. Each RT-PCR primer set was tested against cell line c D N A and c D N A from the index case using a wide range of PCR conditions (Figure 3-8 and data not shown; see also Chapter II). 73 SBNSCANW output Cor sequence 809N15 GENSCAN 1.0 Date run: 28-Dec-00 Time: 13:42:09 Sequence 809N15 : 202189 bp : 38.74% C+G : Isochore 1 ( 0 - 4 3 C+G%) Parameter matrix: Humanlso.smat Predicted genes/exons: O B . E x T y p e S . B e g i n ...End . L e n F r Ph I / A c Do/T c o d R g P . . . . T s c r . . 1 .05 Intr - 146B4 1 .04 Intr - 24809 1 .03 Intr - 32507 1 .02 Intr - 33882 1 .01 Init - 41153 1 .00 Prom - 47005 14553 132 2 0 16 24734 76 1 1 113 32418 90 1 0 73 33828 55 1 1 85 41078 76 2 1 80 46966 40 127 122 0.655 8.82 92 39 0.805 5.40 69 56 0.540 0.39 109 41 0.946 3.02 84 100 0.508 8.35 -4.25 2.00 Prom + 49148 2.01 Init + 52277 2.02 Term + 54776 2.03 PlyA + 55541 49187 40 52348 72 1 0 51 55261 486 1 0 31 55546 6 -3.95 78 66 0.782 3.02 32 179 0.179 0.31 1.05 3 .00 Prom + 56301 56340 40 3 .01 Init + 66825 67239 415 3 .02 Intr + 67292 67420 129 3 .03 Intr + 84182 84237 56 3 .04 intr + 90610 90763 154 3 .05 Intr + 111988 112059 72 3 .06 Term + 112835 112965 131 3 .07 PlyA + 112970 112975 6 -3.65 2 1 39 80 339 0.695 24.88 0 0 0 80 115 0.520 1.65 0 2 105 72 39 0.102 1.78 0 1 1 87 112 0.070 1.12 2 0 97 95 51 0.541 5.16 0 2 68 43 67 0.418 -2.44 1.05 4.03 PlyA - 113753 4.02 Term - 122945 4.01 Init - 134906 4.00 Prom - 136423 113748 6 122830 116 0 2 55 134351 556 2 1 35 136384 40 1.05 46 137 0.116 3.95 83 341 0.633 23.48 -4.75 5.00 Prom + 137064 137103 40 5.01 Init + 138762 138771 10 5.02 Intr + 139610 139797 188 5.03 Term + 178005 178049 45 5.04 PlyA + 178949 178954 6 -8.15 2 1 116 33 9 0 786 -0.91 0 2 122 82 200 0 846 21.29 2 0 82 44 64 0 044 -2.37 1.05 6.03 PlyA - 179391 179386 6 1.05 6.02 Term - 185847 185744 104 1 2 83 47 50 0.226 -2.14 6.01 Intr - 191203 191117 87 2 0 43 67 133 0.278 5.72 E x p l a n a t i o n : Gn.Ex : gene number, exon number (for reference) Type : Init = I n i t i a l exon (ATG to 5' splice site) Intr = Internal exon (3* splice site to 5' splice site) Term = Terminal exon (3' splice site to stop codon) Sngl = Single-exon gene (ATG to stop) Prom = Promoter (TATA box / initation site) PlyA - poly-A signal (consensus: AATAAA) S : DNA strand (+ = input strand; - = opposite strand) Begin : beginning of exon or signal (numbered on input strand) End : end point of exon or signal (numbered on input strand) Len : length of exon or signal (bp) Fr : reading frame (a forward strand codon ending at x has frame x mod 3) Ph : net phase of exon (exon length modulo 3) I/Ac : initiation signal or 3' splice site score (tenth bit units) Do/T : 5' splice site or termination signal score (tenth bit units) CodRg : coding region score (tenth bit units) P : probability of exon (sum over a l l parses containing exon) Tscr : exon score (depends on length., I/Ac, Do/T and CodRg scores) 74 B G E N S C A N predicted genes in sequence 809N15 0 48.0 I I I [ I J i • . i • . i i _ L _ i I • • » • • • • i •• ! I kb 0.0 6.0 12.0 18.0 24.0 30.0 36.0 42.0 48.0 54.0 60.0 I I I • ! , ! ! ! ! ! ! ] ' I . I • I . i • ! . kb 60.0 66.0 72.0 78.0 84.0 90.0 %.0 102.0 108.0 114.0 120.0 ] I • ! ! I ; ! I ! _ 1 ' 1 • ' , I . I . I , I fct) 120.0 126.0 132.0 138.0 144.0 150.0 156.0 162.0 168.0 174.0 180.0 41 I 180.0 186.0 192.0 198.0 i I H Optimal exon KCVI ' n ' u a ' Internal Terminal f^lk Singlc-cxon « ™ « « " £<™ • SubopiiiralcNon Figure 3- 7: GENSCAN Web gene prediction (A) (previous page) GENSCAN (http:^genes.mit.edu/GENSCAN.html) output table showing six predicted coding sequences (CDS) within the complete sequence of the index case breakpoint spanning BAC clone 809N15.The SCORE of a predicted feature (e.g., exon or splice site) is a log-odds measure of the quality of the feature based on local sequence properties. For example, a predicted 5' splice site with score > 100 is strong; 50-100 is moderate; 0-50 is weak; and below 0 is poor (more than likely not a real donor site). The PROBABILITY of a predicted exon is the estimated probability under GENSCAN's model of genomic sequence structure that the exon is correct. This probability depends in general on global as well as local sequence properties, e.g., it depends on how well the exon fits with neighboring exons. It has been shown that predicted exons with higher probabilities are more likely to be correct than those with lower probabilities. (B) Lower/second panel show a graphical layout of the predicted CDSs (see also 6q21 map Figure 3-6) 75 Our preliminary RT-PCR screen highlighted one predicted CDS as being a genuinely expressed sequence (Figure 3-8). Importantly, the translation of this CDS showed significant similarity to Ankyrin repeat containing cyclin-dependent kinase inhibitors pl6 and pl9. We were only able to find strong EST evidence and validate a single CDS representing a novel gene whose start lies within -80 kb of the predicted 6q21 breakpoint. The predicted CDS was soon matched to the 5' end of a non-annotated, I M A G E Consortium (234) expressed sequence tag (EST; I M A G E clone 4838835, Genbank Accession # BC034982) as further information was made available in public sequence databases. We obtained this c D N A clone and sequenced it in its entirety, revealing not only 100% sequence identity with our short predicted ORF, at the 5' end, but also that the full length R N A transcript was significantly longer than our original G E N S C A N prediction. A n exact position for the full length EST was obtained by aligning it on the 6q21 genomic sequence using the U C S C B L A T tool (192, 194) (http://genome.ucsc.edu/; data not shown). 76 Genscan 0RF/CDS1 Figure 3- 8: RT-PCR validation of G E N S C A N gene predictions Non-quantitative RT-PCR using primer sets designed from G E N S C A N gene predictions. G E N S C A N CDS1 (upper panels) was found to be almost ubiquitously expressed while CDS5 (lower panel) could not be amplified. A number of different PCR conditions and templates were tested (not shown). 77 We next looked in detail at the potential protein architecture of the predicted gene. Protein signature databases have become vital tools for identifying relationships in novel sequences and can be used for the classification of protein sequences and for inferring their function (235-237). Each of these databases consists of multiple alignments of protein domains or conserved protein regions. The alignments represent evolutionarily conserved structures which could have implications for a protein's function. Models, often termed hidden Markov models (HMMs), built from individual alignments can be used to recognize when a new protein belongs to an existing family even when the homology is weak. In contrast to standard pairwise alignment methods, like BLAST alignments, H M M s from protein signature databases can align multidomain proteins (237). Using domain prediction tools such as P F A M (195, 196), S M A R T (238-240) and InterPRO (241, 242) we predicted that this novel gene encodes a 103 KDa protein possessing six N-terminal Ankyrin repeats and a C-terminal catalytic H E C T (Homologous to E6-AP Carboxy Terminus) domain characteristic of E3 ubiquitin ligases (101, 159) (Figure 3-9). We therefore designated this new gene HACE1, for H E C T domain and Ankyrin repeat Containing E3 ubiquitin ligase 1 (198). Many H E C T domain containing E3 ligases have been described with a modular architecture where the N-terminus contains a domain for substrate recognition while the C-terminus contains the ubiquitin transferring H E C T domain (101, 166, 167). The predicted H A C E 1 protein appears to be consistent with this model. Ankyrin repeats have been widely implicated in mediating protein-protein interaction (243), and therefore may be responsible for substrate specificity and complex interactions of 78 H A C E 1 . B y us ing the B L A S T a l ignment tool ( h t t p : / / v v w w . n c b i . n l m . n i h . g o v / B L A S T ) w e can see that the A n k y r i n repeats encoded by HACE1 bear a h i g h degree of s imi la r i ty to those of C D K N 2 D ( p l 9 I N K 4 D ) and C D K N 2 A ( p l 6 I N K 4 A ) . To date, A n k y r i n repeats have not been found i n any other E3 ligase, ind ica t ing potent ial ly un ique features of this prote in (198). W i t h i n the H E C T domain , H A C E 1 shares > 50% sequence s imi la r i ty w i t h the archetypal E 6 - A P ub iqu i t i n ligase as w e l l as almost 60% (or more) s imi lar i ty w i t h N e d d 4 , S M U R F 1 and S M U R F 2 (Figure 3-9; al ignments done w i t h pair -wise B L A S T ) . E 6 - A P was first ident i f ied as a specific E3 ligase p r o m o t i n g ub iqu i t ina t ion and degradat ion of p53 i n pap i l l omav i rus infected cells (244, 245). Important ly, H A C E 1 contains a conserved cysteine residue w i t h i n the last 34 amino acids, and a hydrophob ic s t r ing of six amino acids o n the extreme C-terminus , found i n a l l H E C T d o m a i n proteins, i n c l u d i n g E 6 - A P , bo th of w h i c h are requi red for substrate pro te in ub iqu i t ina t ion (101,159,166). 79 Oaa 300 600 900 Cys876 * Ankyrin Repeats p 1 9 I N K 4 D ( 5 1 % ) * NEDD4 (64%) p 1 6 I N K 4 A (4 7o / o ) * SMURF1/SMURF2 (59%) E6-AP (53%) Figure 3- 9: The predicted domain architecture for HACEI The HACEI protein is composed of a unique combination of six N-terminal Ankyrin repeats for protein-protein interactions, coupled with a C-terminal HECT domain for transfer of ubiquitin to target molecules. Each of the domains of HACEI has significant similarity to previously described proteins: Ankyrin repeats are highly similar to those of pl6 and pl9, while the HECT domain aligns with a number of other E3 family members. Regions of similarity are denoted by hashed bars with % similar amino acid composition noted below. The position of the HECT domain conserved C-terminal Cysteine residue is marked by an asterisk (*) in HACEI and other E3 family members. H a v i n g ident if ied an interesting candidate gene at the 6q21 locus, w e set out to val idate the gene by l o o k i n g at its expression profile. U s i n g the HACEI I M A G E clone as a probe w e ana lyzed m R N A expression across two mult i- t issue N o r t h e r n blots (Figure 3 -10A,B) . HACEI m R N A was detected almost ub iqu i tous ly across tissues tested (198). Important ly, moderate to h i g h expression was seen i n no rma l k idney , the proposed tissue of o r ig in for W i l m s ' t umour (4 ) . H i g h expression was also observed i n Hear t and across most areas of the C N S . W e next chose to assay the m R N A expression of HACEI i n a panel of pediatr ic t umour cell l ines (Figure 3-10C). A l t h o u g h a w i d e spect rum of expression was seen it was noted that bo th Neurob las toma cell lines and 80 the anaplastic Wilms' tumour cell line SKNEP1* (246-248) showed low to undetectable levels of HACE1 expression (see Figure 3-10). To demonstrate that the HACE1 c D N A did indeed encode the full-length protein, the 2727-bp ORF was cloned into a number of expression vectors. HACE1 was His-tagged and cloned into pET-15b for expression in bacteria and simple purification of the recombinant protein on Nickel-Sepharose columns. The open reading frame (ORF) was also HA-tagged in P C D N A 3 H A (and later subcloned into pMSCVhygro) for expression in a coupled rabbit reticulocyte in vitro transcription/translation system as well as for transfection into mammalian cells. A protein of 103KDa, consistent with the predicted size of the H A C E 1 protein, could be detected by autoradiography or Western blot with an anti-HA antibody (Figure 3-10D,E). To detect H A C E 1 protein directly a number of antibodies were produced: a rabbit polyclonal antibody against the H A C E 1 peptide " C L V L L L K K G A N P N Y Q DISG", corresponding to the second Ankyrin repeat, as well as mouse polyclonal and mouse monoclonal antibodies produced against the full length recombinant protein (198). In addition to detecting purified recombinant protein, Western blotting performed on a panel of pediatric cancer cell lines previously assayed for HACE1 m R N A expression revealed a similar pattern of protein expression and a protein of the expected molecular weight (198)(Figure 3-10F). Taken together this not only confirmed the specificity of the anti-* Our laboratory has recently become aware that the SKNEP1 cell line has been found to express the Ewings family tumour t(ll;22)(q24;ql2) translocation associated EWS-FLI1 fusion transcript. As such, it should be re-classified as a peripheral primitive neuroectodermal tumour (pPNET) of renal origin. Although SKNEP1 is referred to as a Wilms' tumour cell line throughout this dissertation, the reader should refer to conclusions and further discussions in Chapter 6 for implications of the EWS-FLI1 fusion in SKNEP1 cells. 81 H A C E 1 antibodies but also p r o v i d e d evidence of H A C E 1 p ro te in expression across mul t ip l e cel l types. 82 B -HACE1 •p-Actin ' HACE1 •fi-Actin in vitro transcribed/ Translated lactacystin - + KOa 119 91 51.4 34.7 35S-HACE1 -103 KDa NIH3T3cells HA-vector HACE1 S P S P HA-HACE1 -103KDa KOa 119 91 HACE1 Grb2 Figure 3-10: Expression profile of the HACE1 gene product (A) & (B) Multi-tissue Northern blots, probed with a full-length HACE1 c D N A probe, show widespread expression in normal tissues. (C) Northern blot expression profile of HACE1 in multiple pediatric cancer cell lines and primary Wilms' tumours. (Neuroblastoma: K C N R , IMR-32, SAN-2, S K N S H ; Ewing Sarcoma: S K N M C ; Rhabdomyosarcoma: Birch, CT-10; Adenovirus transformed human embryonic kidney: HEK293T; Wilms' tumour: SKNEP1) (D) Rabbit reticulocyte lysate in vitro transcribed/translated HACE1 protein labelled with 3 5 S-methionine with or without the 20s proteasomal inhibitor lactacystin. Both lanes show a band of the expected molecular weight for full length HACE1. (E) NIH3T3 mouse fibroblasts transfected with empty vector or HA-HACE1. When probed with an ct-H A antibody, a band of the expected molecular weight is seen in H A - H A C E transf ected cells, but not vector alone. Lysates were separated into soluble (s) and pellet (p) fractions. (F) HACE1 protein expression in a panel of pediatric tumour cell lines (as noted in (C)) and recombinant (rec.) protein detected with a - H A C E l antibodies. Experiments in (D), (E), and (F) were performed by Dr. Valentina Evdokimova 83 3.2.4 Functional Predictions & Validation Of Function As described above the H A C E I protein is predicted to contain two recognizable functional domain groups: Ankyrin repeats at the N-terminus and a H E C T domain at the C-terminus. Although this is a unique combination of domains the structure is consistent with a H E C T domain containing E3 ubiquitin-protein ligase. In order to confirm that H A C E I is functionally able to ubiquitinate target proteins a number of criteria must be met. As outlined in the first chapter, the process of protein ubiquitination involves a highly conserved pathway in which one or more ubiquitin polypeptides become conjugated to specific substrate proteins (101,158,159). The H A C E I protein is predicted to belong to the H E C T domain family of E3 ligases. We first chose to see if it could interact with various E2 enzymes. A commercially available kit of purified E l and E2 enzymes was purchased for in vitro association and biochemical experiments. First, we assessed whether or not H A C E I could interact directly with any of the E2 enzymes. As a negative control a C -terminally truncated form of H A C E I was used: p75, lacking the last 204 amino acids of the H E C T domain (see Figure 3-9). As the H E C T domains appear to be required for E2-E3 interactions (101,159,163,166, 249) this truncation, missing the majority of the H A C E I H E C T domain, should not be able to bind an E2 enzyme. By mixing the E2s and H A C E I in vitro we were able to see that H A C E I could interact with several E2s, including UbcH5a, UbcH5b, UbcH5c, and UbcH7 without the apparent need for additional co-factors (Figure 3-llA&B)(198). In keeping with other H E C T domain 84 containing E3s, H A C E 1 is able to form a transient thio-ester bond in vitro with ubiquitin only when additional components are added, specifically: ATP, the ubiquitin E l (UBE1) and an E2 enzyme. Al l H E C T domain containing proteins posses a conserved Cys residue 30-33 amino acids from their C-terminus, this residue is specifically responsible for the thio-ester bond, temporarily binding ubiquitin to the E3. In the case of H A C E 1 , mutation of the conserved C-terminal ubiquitin, at position 876, results in the inability of the protein to act as a ubiquitin-protein ligase (Figure 3-11). In addition, H A C E 1 specifically requires UbcH7 as its cognate E2 when forming a thio ester bond in vitro (Figure 3-llC&D)(198). As expected, the thio-ester bond was destroyed in the presence of a reducing agent, P-mercaptoethanol (Figure 3-11D). 85 * & £ & jf # f/4 fff/fff/ Ubc: _H2_ H5a H5b H5c H7 p75 -+ - + . + . + . + Hacel + • • - + - + - + 119 35.8 29.0 -• -• S B 1 2 3 4 5 6 7 8 9 10 E2 E1 + + //////// + + + + + + + Ub-Hace1 119. Hacel Hacel 119 Wild-Type C876S mutant 3 4 6 7 8 9 5S-Hace1: Wild-type C876S mutant Ub-Hace1 Hacel 1 2 3 4 5 6 7 8 Figure 3-11: In vitro ubiquitin ligase activity of H A C E I Experiments (A)-(D) were performed by Dr. Vaelntina Evdokimova (A) Commercially obtained recombinant E2 enzymes stained with coomassie brilliant blue (1 ug of each). (B) In vitro pull-down assay showing the interaction of H A C E I with different E2s. Recombinant His-tagged H A C E I or truncated p75 (2 ug of each) were immobilized on Ni-Sepharose beads and incubated with the corresponding E2s (0.5 ug of each) in 200 m M KC1, 2 m M DTT, 0.2% NP-40, 20 m M Tris-HCl (pH 7.8) and 5% glycerol for 60 min at room temperature. Bound proteins were resolved by SDS/12%-PAGE and visualized by immunoblotting using a-Hacel (upper panel) or mixture of ct-GST and a-UbcH7 (lower panel) antibodies. (C) E l and various E2 enzymes were mixed with HACEI or ligase dead mutant C876S, ATP and GST-ubiquitin (Ub) before separation by PAGE. A band shift indicative of a bond between GST-Ub and HACEI , but not mutant, is apparent when in the presence of UbcH7 with wild-type but not ligase-dead (C876S) mutant H A C E I (D) Same assay as in (C) using only UbcH7 as an E2. Here the dependence of the H A C E I - GST-Ub bond formation on ATP and an E l are demonstrated as well as the ability of a reducing agent (p-mercaptoethanol) to destroy the bond. Again, the ligase dead mutant (C876S) is unable to forma bond with ubiquitin under any of the same conditions Due to the omnipresence of ubiquitin, the ubiquitin cascade enzymes ( U B E 1 and UBCs), and the activity of the proteasome in cells we were unable to look directly for thio-ester bond formation in vivo. Instead, we were able to pul l down smears of 86 ubiquitinated proteins when H A - H A C E 1 was immunoprecipitated indicating that H A C E 1 is associated wi th ubiquitinated proteins (Figure 3-12B). In addition we were able to show an association of UbcH7 with H A C E 1 in vivo (Figure 3-12A)(198). Input IP:aHA Input IP:aHA 1'—^  j 9 1.**-] M H H B B V ^ i y O LJ^K__EL__BHEJM 1 2 3 1 2 3 4 1 2 3 4 Figure 3-12: HACE1 in vivo ubiquitin ligase activity Experiments in (A) & (B) were performed by Dr. Valentina Evdokimova (A) The E2 enzyme UbcH7 can be co-immunoprecipitated with HACE1 from HA-HACE1 transf ected 293T cells, BiP is shown as a negative control (see also section 3.2.5) (B) Immunoprecipitation of HA-HACE1 pulls down a smear of ubiquitinated proteins, typical for E3 ligases and other ubiquitin interacting proteins (right panel). HA-HACE1 is also detectable when this blot is re-probed with ct-HACEl antibodies (left panel) 3.2.5 Subcellular Localization ofHACEl Having already established a profile of expression and validated the basic function of H A C E 1 we set out to examine where in the cell the H A C E 1 protein is expressed. The subcellular localization of H A C E 1 may give information about the general roles of this E3 ligase in normal and cancerous cells. Either wild-type or H A -H A C E 1 transfected HEK293T cells were examined for the localization of H A C E 1 using both a n t i - H A C E l antibodies and anti-HA antibodies. We found that the H A C E 1 protein was abundant throughout the cytoplasm with particularly intense signal 87 surrounding the nucleus (198). This peri-nuclear staining is consistent with the location of the endoplasmic reticulum (ER) and when an ER resident chaperone protein BiP (250, 251) was detected as a control, H A C E 1 was confirmed to co-localize with BiP (Figure 3-13A). The Subcellular localization was further confirmed by fractionation of cell lysates in a sucrose gradient (198). Here the H A C E 1 protein was predominant in the membranous ER fraction (ER) and the soluble (S) cytoplasmic fraction, but not in the nuclear (N) or the mitochondrial (M) fractions (Figure 3-13B). HACE1 BiP merged Cell fraction 119 119 91 21.1 21.1 29.0 «9 I— 1 **« • •< m • -»< UbcH7 Figure 3-13: Localization of HACE1 (A) Immuno-staining of HACE1 (red) and ER resident chaperone BiP (green) both showing predominant perinuclear staining. Overlayed images are shown with DAPI (blue) staining in the nucleus (lower panels). Immunofluorescence microscopy was performed by Nataliya Melnyk (B) Fractionation of cell lysate confirms the localization of HACE1 to the soluble (cytoplasmic) and ER fractions. BiP, UbcH7, Histone H3, and Grb2 are shown as controls for ER/Soluble, ER/Soluble, Nuclear, and cytoplasmic fractions respectively. Fractionation experiments were conducted by Dr. Valentina Evdokimova 88 3.3 S U M M A R Y We have identified a novel t(6;15)(q21;q21) balanced reciprocal translocation in Wilms' tumour (171) (Figure 3-1). Although a near tetraploid complement is seen in a smaller fraction of tumour (and normal) cells, the translocation appears to be the only chromosomal abnormality. We must therefore consider the possibility that this polyploidy is an artefact of short term culturing. As the 6q21 region is commonly involved in chromosomal rearrangement, breakages and deletions in a number of cancers, including Wilms' tumour, we concentrated on mapping this area precisely and looking for candidate genes affected by this translocation. Having mapped the specific region of the translocation to a gene-poor region of 6q21 (Figures 3-3, 3-4, 3-5, & 3-6) (198), we then used a variety of bioinformatic tools to predict novel genes. Although the genome sequence available at the time for G E N S C A N analysis was limited, two predicted CDSs (Figure 3-7) possessed significant similarity to known protein families and we were able to show using RT-PCR that one of these CDSs was expressed almost ubiquitously (Figure 3-8). We later found a full length c D N A clone extending our original coding sequence into a genomic region that we had not previously been able to use for G E N S C A N analysis. We were then able to classify the novel gene based on a prediction of its component domains, N-terminal Ankyrin repeats and a C-terminal H E C T domain (Figure 3-9), designating this novel gene HACEI (HECT and Ankyrin repeat Containing E3 ubiquitin-protein ligase 1) (198). HACEI m R N A appears to be expressed in a wide variety of normal 89 tissues but is not detectable in either Neuroblastoma or Wilms' tumour cell lines (Figure 3-10A-C). In vitro expression confirms a protein of the expected molecular weight to be transcribed and translated from the HACEI ORF (Figure 3-10D&E) and, in agreement with the Northern blot studies, the H A C E I protein is also widely expressed (Figure 3-10F). Validation of the ubiquitin ligase function was accomplished by showing that H A C E I is able to bind with ubiquitin in vitro with a transient bond typical for H E C T domain-containing proteins (Figure 3-11). In vivo H A C E I associates with particular E2 enzymes and ubiquitinated proteins (Figure 3-12). Although the ubiquitin-proteasome pathway is typically thought to target ubiquitinated protein for degradation by the proteasome, more recently the role of protein ubiquitination has been expanded to include other regulatory functions such as transcriptional activation, signal transduction and protein localization, likening this process to phosphorylation in post-translational control of protein activity (101, 252-260). It remains to be seen how H A C E I may affect its targets. Interestingly, we found that in vitro H A C E I was able to bind ubiquitin in the presence of UbcH7, but was also able to interact with UbcH5a, H5b and H5c. UbcH7, H5a and H5b have all been implicated in the regulation of p53 protein levels (261-263), most notably in the complex of E6-AP and UbcH7 (262, 263). UbcH7 has also been linked to ubiquitination of histones (264, 265), although H A C E I appears to be absent from the nucleus under normal conditions. The localization of H A C E I to the ER may also suggest an underlying function in the ER quality control system, ER associated degradation (ERAD) (266), and/or stress responses such as the 90 Unfolded Protein Response (UPR) (267). More detailed examination of the role of H A C E I will be examined and discussed in subsequent chapters. To conclude, cytogenetic and molecular mapping of the region involved in a unique and novel translocation in Wilms' tumour, combined with new information from the human genome project, yielded a novel candidate gene that may play a significant role in oncogenesis, not only in Wilms 1 tumour, but across a wide spectrum of human malignancy. Further work, including expression profiling of Wilms' tumours and other neoplasms and functional characterisation of the novel HACEI gene product, are the topics of the following chapters. 91 CHAPTER IV HACEI EXPRESSION IS ALTERED IN SPORADIC W I L M S ' T U M O U R The project described in this chapter was conceived and initiated by the laboratory of Dr. Poul H.B. Sorensen. Portions of this chapter have been published in two peer reviewed publications: Anglesio, M.S., Evdokimova, V., Melnyk, N . , Zhang, L. , Fernandez, C.V., Grundy, P.E., Leach, S., Marra, M.A. , Brooks-Wilson, A.R., Penninger, J. and Sorensen, P.H. (2004) Differential expression of a novel ankyrin containing E3 ubiquitin-protein ligase, Hacel, in sporadic Wilms' tumour versus normal kidney. H u m Mol Genet, 13, 2061-74. In this publication I generated the majority of the data, contributed to the writing. Zhang L*, Anglesio MS*, O'Sullivan M , Zhang F, Yang G, Sarao R, Nghiem MP, Cronin S, Hara H , Melnyk N , L i L , Arya S, Wada T, Liu PP, Sorensen P H , and Penninger J. (2006) The E3 ligase H A C E I is a critical chromosome 6q21 tumour suppressor involved in multiple cancers. Manuscript submitted In this publication I generated the majority of in vitro data, and in cooperation with Dr. Josef Penninger and Dr. Poul Sorensen, contributed to the writing. * These authors contributed equally to this work Acknowledgements for the data contained in this chapter that was generated by other members of the Dr. Poul H.B. Sorensen's Lab and collaborators is noted in the figure legends. I N T R O D U C T I O N In the previous chapter I described the molecular characterization of the 6q21 region targeted by the Wilms' tumour t(6;15) translocation. Mapping. experiments revealed that the 6q21 breakpoint occurred directly upstream of a novel gene. This new protein contains six Ankyrin protein-protein interaction motifs linked to a H E C T (Homologous to E6-associating protein Carboxyl Terminus) domain (166, 268), a 92 feature of a subclass of E3 ubiquitin-protein ligases. We therefore designated this gene H E C T domain and Ankyrin domain Containing E3 ubiquitin-protein ligase 1, or HACEI. As this novel gene has been found in proximity to a chromosomal abnormality, our lab was interested in finding out if its expression was altered in the index case or more universally in sporadic Wilms' tumour. Previous literature has described situations where rearrangement can result in the inactivation of a tumour suppressor gene (TSG) (269-271), the activation of a proto-oncogene (272, 273) or the creation of a chimeric protein with mixed functions (116, 224, 225, 229, 272, 273). The translocation in the Wilms' tumour index case appeared to be distant from any described genes, including our novel HACEI gene. This translocation is unique among Wilms' tumours and in fact other Wilms' tumour translocations that involve this same region on 6q21 are exceptionally rare making a gene fusion event very unlikely. Therefore if this translocation is to affect our novel H A C E I E3 ligase it must be acting to either enhance or impair its expression. We will show in this chapter that HACEI is expressed at low or undetectable levels in the majority of sporadic Wilms' tumours at both the m R N A and protein levels and that this apparent downregulation appears to correlate with elevated methylation. Further, HACEI expression can be restored in Wilms' tumour cell culture models. Sequencing of HACEI performed on a cohort of primary Wilms' tumour specimens has failed to uncover mutations, deletions, or L O H in the region, in agreement with 93 other reports i n the literature (172). Despite this f ind ing , the 6q21 reg ion is c o m m o n l y deleted i n a number of other t umour types and i n fact a w i d e r screen for expression has revealed that HACEI expression is exceptional ly l o w across a majority of cancer cel l lines. 4.1 R E S U L T S 4.1.1 Quantitative RT-PCR & Protein Expression Analysis of the Index Case and Other Wilms' Tumours The nove l HACEI gene appeared to be the most p r o x i m a l expressed transcript to our unique t(6;15)(q21;q21) translocation, occur r ing approximate ly 80kb upst ream and 5', to the start of the HACEI gene (see also Chapter I). To evaluate the m R N A expression quanti tat ively we chose to use a T a q M a n 5' exonuclease based quanti tat ive R T - P C R assay (274, 275). U s i n g this methodology , a f luorogenic probe complementary to a target sequence is added to a P C R reaction mix . The probe consists of an ol igonucleot ide w i t h a reporter and quencher dye attached. D u r i n g P C R , if HACEI is present, the probe w i l l anneal specifically between the f o r w a r d and reverse p r imer sites. The nucleolyt ic act ivi ty of T a q polymerase cleaves the probe, separating the reporter a n d quencher dyes and results i n an increased fluorescence intensity of the reporter dye. This process occurs w i t h every P C R cycle and can be moni to red w i t h a C C D camera w i t h i n the sequence detection system's thermal cycler (Figure 4-1). 94 polymerization strand displacement cleavage —3' —5' Two fluorescent dyes, a reporter (R) and a quencher (Q). are attached to the probes used with the TaqMan PCR Reagent Kit. The 3' end of the probe is blocked, so it is not extended during the PCR reaction. When both dyes are attached to the probe, reporter dye emission is quenched due to fluorescence energy transfer from the reporter dye to the quencher dye. During each extension cycle, the probe is displaced at the 5' end by the DNA polymerase. 5-3-Taq DNA polymerase then cleaves the reporter dye from the probe via its 5'-3' 5 exonuclease. -3' polymerization complete -3 Once separated from the quencher, the reporter dye emits its characteristic fluorescence which can then be ^ measured. The amount of fluorescence measured is proportional to the amount of PCR product made. Figure 4-1: Principles of TaqMan 5' Exonuclease Based Quantitative PCR Assay (Adapted from applied biosystems http^/www.appliedbiosystems.com) U s i n g the index case as a test sample, w e first assayed the expression of H A C E 1 us ing our quanti tat ive P C R technique. T o ensure a re la t ive ly u n i f o r m sample, fresh frozen tissue f r o m the index case ( tumour and adjacent normal) was embedded i n O C T c o m p o u n d (276) and ten 12-15 p m thick sections were cut and placed immedia te ly i n T r i z o l for R N A extraction. A section of ~ 7 p m was taken f rom about halfway th rough the ten sections, app l i ed to a glass sl ide, f ixed, and stained w i t h H e m a t o x i l i n a n d E o s i n ( H & E ) . This section was used to con f i rm that the sample was compr i sed of at least 80% tumour cells (or was d e v o i d of t umour cells i n the case of the 95 normal sample; Figure 4-2A). The sample was processed for quantitative RT-PCR and evaluated with beta-Actin as an endogenous control (Figure 4-2B and Figure 4-3). Interestingly, we found that expression was significantly reduced in the Wilms' tumour when compared to adjacent non-tumour kidney cells. 96 A Figure 4- 2: Quantitative RT-PCR from the index case (A) Histological examination of frozen sections confirmed the presence of > 80% tumour cells in the tumour sample. Normal kidney histology was confirmed in the normal sample with no apparent trace of tumour cells. Photographs and histological evaluation was done by Dr. Maureen O'Sullivan, Dr. Poul Sorensen and Dr. David Grynspan in the laboratory of Dr. Poul Sorensen. (B) Screen capture from Sequence Detection System software tracking the amplification of HACEI and $-Actin (endogenous control) in each sample. The endogenous control amplification is relative similar for both samples. The HACEI PCR using a tumour sample does not enter a exponential phase until 3-4 cycles after the normal sample, indicating that HACEI expression is as low as 2H) fold lower in tumour tissue compared with normal (6.25% of normal). 97 W e next evaluated a larger panel of W i l m s ' tumours to test whether or not l owered HACEI expression migh t be a recurrent molecular al teration found i n sporadic W i l m s ' tumour . W e obtained 17 sporadic W i l m s ' tumours f rom the Na t iona l W i l m s ' T u m o u r S tudy G r o u p ( N W T S G ) and 8 cases f rom B C C h i l d r e n ' s Hosp i t a l . A total of 26 cases were processed as described above (see also Chapter II). W e found that a l though there was considerable var iab i l i ty i n the absolute level of expression amongst the samples, 20 of 26 showed a reduc t ion of HACEI expression i n the W i l m s ' t umour tissue w h e n compared w i t h patient matched n o r m a l tissue (Figure 4-3A, see also Table 4-1). M o s t of the samples were also evaluated u s ing a second endogenous control gene, TBP ( T A T A - b o x b i n d i n g protein), s h o w i n g expression changes v i r tua l ly ident ical to those obtained w h e n us ing beta-Actin as the endogenous control (data not shown). To verify H A C E I prote in expression, prote in lysates were prepared f rom tissue processed i n the same manner as described above (see also Chapter II). In this case, frozen sections of tissue were p laced i n Lys i s buffer rather than T r i z o l . Western blot analysis of prote in samples m i r r o r e d the results f rom the quanti tat ive R T - P C R w i t h 20 of 26 tumours , i n c l u d i n g our index case, exhib i t ing l ower or undetectable levels of H A C E I prote in (Figure 4-3B, see also Table 4-1). H a v i n g described l o w or absent expression of HACEI i n - 7 5 % of tested sporadic W i l m s ' tumours , and compar ing this to k n o w n molecular genetic alterations i n W i l m s t umour (see Chapter I), w e conc luded that loss of HACEI expression may be 98 much more common than WTl alterations in sporadic Wilms' tumour. However it should be noted that fetal kidney is presumed to be the tissue of origin for Wilms' tumour (3, 4, 277), and it is also possible that a subpopulation of developing kidney cells may not express HACE1 and may be the same population that also give rise to Wilms' tumours. Although our current study does not address this issue directly, we also tested fetal and pediatric kidney tissue for HACE1 expression. We found that although levels were quite variable between samples, especially at the m R N A level, both fetal and pediatric kidney sample expressed detectable levels of H A C E 1 protein comparable to that seen in normal kidney tissue that was adjacent to the Wilms' tumours tested in the cohort discussed above (Figure 4-3C&D and (198)). 99 628 321_ 395 . 624_ NN55 277 654 785 015 555 728 586 520 432 I N D E X NN18 LL38 HH41 EE7 CC66 N T N T N T N T N T N T N T N T N T J N T N T N T N T N T N T N T N T N T N T N T HACE11 Grb2 ! ... I i 3 0 • [ ] a fk1 fk2 fk3 ped HEK — — — kidney 293 fetal kidney samples kDa 119 "f 51.4 fetal kidney samples j j . jc fK1 fK2 fK3 H Hacel actin Figure 4- 3: HACE1 Expression in Wilms' Tumour and Patient Matched Kidney (A) Quantitative RT-PCR results showing relative mRNA expression as a fold difference from patient matched normal kidney. The index case with the unique t(6;15)(q21;q21) translocation is highlighted in red. The last 6 lanes show normal total R N A (NTR) from kidney, brain and cell lines 293T (teal) and SKNEP1 (yellow) (expression relative to NTR-Kidney). (B) Representative Western blot analysis of Wilms' tumours (T) and patient matched normal kidney (N) probed with a mouse polyclonal ant i-HACEl antibody. Grb2 is show as a loading control. (C) HACE1 qRT-PCR result of fetal kidney samples (fkl-3) showing mRNA expression relative to pediatric kidney and HEK293 cells (D) Western blot analysis showing HACE1 protein expression from samples shown in (C) 100 4.1.2 Genomic Sequencing ofHACEl in Sporadic Wilms' Tumours Current literature suggests that Wilms' tumour most likely derives from incomplete and aberrant differentiation of the fetal mesenchyme (3, 4, 277) and we have observed strong H A C E 1 expression in both mature pediatric and fetal kidney but not in the Wilms' tumours samples evaluated above. It is therefore possible HACE1 down-regulation or inactivation is a common and potentially etiologic event in this disease. We therefore wished to assess the Wilms' tumour cohort for potential genetic mechanisms of HACE1 down-regulation. Unfortunately, of the 26 Wilms' tumours analyzed for HACE1 expression, aside from the index case, no karyotypes were available for analysis. However, it should be noted that there are only occasional reports of chromosome 6q21 alterations in Wilms' tumour (18, 172-174), making it unlikely that chromosomal rearrangements affecting this region are a > major mechanism of HACE1 down-regulation in Wilms' rumour. Given this, we decided to search for alternative mechanisms that might underlie altered HACE1 expression in Wilms' tumour. To do this, we screened 24 cases and matched normal kidney as well as the SKNEP1 cell line for HACE1 mutations by direct sequencing. Al l exons and intron-exon boundary regions were sequenced from genomic D N A of each sample. No sequence discrepancies between tumour and normal D N A were found for any of the cases. Furthermore, no mutations were found in the SKNEP1 cell line. Sequence data was compared to the U C S C human genome database and a total of twelve HACE1 polymorphisms were identified within our cohort (see Table 4-1). Of these, only three of the polymorphisms were within coding sequences, but only one represented an 101 amino acid change. However, the latter (Asp to Gly at amino acid 399) does not affect any identifiable functional domains within the H A C E I protein (see Figure 3-9). This polymorphism occurred only in one sample pair and when a non-diseased control cohort of 95 Individuals was tested for this polymorphism it was present in only 1 sample (data not shown). Therefore, it most likely represents a low frequency polymorphism in the general population. Within our sample group virtually all contained one or more heterozygous polymorphisms in tumour and normal D N A , suggesting that L O H of the HACEI locus is unlikely in this cohort. This appears to agree with current literature reports for this region (172). However, our sequencing strategy cannot rule out micro-deletions of whole exons where heterozygosity was not observed or other genomic sequences located between these heterozygous polymorphisms. In addition, the extensive 3' untranslated region (UTR) of HACEI was not sequenced neither was the putative promoter region 5' to the first exon. Both of these sections could harbour regulatory sequences important for both expression and stability of the HACEI mRNA. Taken together, the data indicates that a genetic mechanism involving point mutations to the coding region or large scale deletions is unlikely to explain the low HACEI expression observed in our sporadic Wilms' tumour cohort. 102 Table 4-1: Sequencing Summary for HACE1 in Wilms' tumours Samples labelled -1 are Wilms' tumours, while samples labelled - N are patient-matched normal kidney. Expression is indicated relative to matching normal. Methylation status at HACE1 -associated CpG islands is also indicated (see also Figure 4-4 and 3-6). Position of each SNP detected by sequencing is given within, (in green) or relative to, (in blue) the nearest exon. Positions are indicated as exon (X#) or intervening sequence (IVS#) (eg. X8 = exon 8, IVS1 = intervening sequence between exon 1 and 2). The number represents the base position within the coding sequence of HACE2, for exons, or the position from the from 5' end (+) or 3' end (-) of the IVS. The two letters in each column represent the base of each allele, except in the case of IVS21 (-5) where a 3bp deletion was detected: WT indicates wild-type and "3bp DEL (het)" represents that the gene is heterozygous for the deletion at this position. In this latter case the deletion does not appear to affect the splice acceptor site, or frame, of the HACE1 gene. DNA sequencing and subsequent analysis were done in collaboration with the laboratory of Dr. Angela Brooks Wilson at the BC Cancer Agency Genome Sciences Centre. 103 4.1.3 Methylation Analysis & Epigenetic Control of the HACEI locus Methylation involves the addition of a methyl group to the cytosine base in a cytosine-guanine (CG) dinucleotide. Methylation, often within C p G islands, either in coding or non-coding sequences of specific genetic loci generally results in gene silencing (147, 278). Recently it has been suggested that promoter hypermethylation may be as frequent as inactivating mutations for reducing tumour suppressor gene expression in human malignancies (147). Since our previous effort to find a genetic basis for the downregulation of HACEI was unsuccessful, we decided to investigate the methylation status of C p G islands surrounding the HACEI locus in our Wilms' tumour cohort and the SKNEP1 Wilms' tumour cell line. There are three C p G islands located within or near the HACEI gene, one at the putative HACEI promoter extending through exon 1 which contains 88 C G dinucleotides (CpG-88), and two located ~80 kb upstream of the HACEI transcriptional start site containing 177 (CpG-177) and 29 (CpG-29) C G dinucleotides repeats, respectively (see Chapter III, Fig. 3-6). To assess whether altered methylation might influence H A C E I expression, we analyzed the methylation status at each of these C p G islands by digestion with the methylation sensitive restriction enzymes Haell or BssHII followed by semi-quantitative PCR (see also Chapter II). Surprisingly, we were unable to find any evidence of methylation at CpG-88, the most proximal C p G island to H A C E I , in any of the tumour or normal kidney samples. However, methylation of either CpG-177 or CpG-29, or both C p G Islands, was observed in 12 out 104 of 20 W i l m s ' tumours w i t h l o w H A C E 1 m R N A expression as w e l l as i n S K N E P 1 cells (Figure 4-4 and Table 4-1). O f the 6 r ema in ing cases that d i d not s h o w reduced H A C E 1 expression, four d i d not exhibit a discernable change i n methy la t ion and two were not able to be evaluated (see Table 4-1). M e t h y l a t i o n was not observed i n any of the n o r m a l k idney samples. 105 A x f f CpG-88 1 CpG-29 Hpall 44f 44 CpG-177 HpaU ' 1 1 1 1 B - + - + - + - + <-A//C2 5 ' region p - | - - * CpG-88 M M M H M — CpG-29 * * • « M I H M I CpG-177 tumour normal tumour normal Figure 4- 4: Methylation Analysis of HACEI in Wilms' Tumour (A) Restriction enzyme (RE) mini-map of each PCR amplicons assayed by methylation sensitive restriction digestion followed by PCR (not to scale). If any of the indicated methylation sensitive RE sites are not protected from digestion by methylation of a C G dinucleotide the region of amplification wi l l be cut and a PCR product cannot be formed. (B) Representative PCR results from each HACEI -associated CpG island (see also Figure 3-6) after digestion of genomic D N A from Wilms' tumour and patient matched normal kidney with EcoRI and the indicated methylation sensitive RE. As a control for complete digestion the 5' region of the mic2 gene, known to be un-methylated, was also assayed for each sample. See Table 4-1 for a summary of results. 106 To explore the role of methylation at the HACEI locus, we continued experiments using the Wilms' tumour cell line SKNEP1, which exhibited methylation at both CpG-29 and CpG-177 as well as low HACEI expression. We tested whether pharmacological demethylation of these upstream C p G islands could influence expression levels of HACEI. SKNEP1 cells were treated with the methylation inhibitor 5-aza-2-deoxycytidine (5AZ) (279, 280) and then assessed for HACEI expression and methylation status at the HACEI C p G islands (Figure 4-5). Treatment with 5AZ reduced methylation at both CpG-29 and CpG-177, particularly at 5 p M 5AZ, no effect was observed at CpG-88. To increase the sensitivity of our test an additional methylation sensitive restriction enzyme, Hpall, was used in this experiment. In addition to demethylating the HACEI C p G islands (Figure 4-5A) 5AZ treatment was associated with at least a 4-fold increase in HACEI m R N A expression (Figure 4-5B). These findings suggest that methylation status of CpG-29 and CpG-177 can influence HACEI expression. Next, to examine the functional consequences of demethylation at the HACEI locus, we performed chromatin immunoprecipitation (ChIP). We wished to analyze whether the HACEI locus exists in an active or inactive chromatin conformation in SKNEP1 cells. This was assessed by comparing the relative levels of acetylated-Histone H3 versus dimethyl-(Lys79)-Histone H3 bound to the HACEI promoter. Acetylated-Histone H3 is known to associate with D N A of active chromatin, while 107 dimethyl-(Lys79)-Histone H3 is present in the inactive conformation of heterochromatin or silenced genes (205, 281, 282). The promoter region of the HACEI gene exists predominantly in an inactive chromatin conformation in normally growing SKNEP1 cells (i.e. associated with dimethyl-(Lys79)-Histone H3). However, treatment with 5 u M 5AZ reverses this pattern, switching the HACEI promoter to an active chromatin conformation associated predominantly with acetylated-Histone H3 (Figure 4-5C, right panel). As a control, we performed identical studies with HEK293 cells which abundantly express HACEI. Consistent with its high expression, the HACEI locus in these cells exists in an active chromatin state and no apparent change in chromatin structure is evident after treatment with 5AZ (Figure 4-5C, left panel). Although opening of the chromatin conformation may be expected with increased HACEI transcription it does directly indicate a functional consequence at the HACEI locus upon treatment with a "global" methylation inhibitor. Taken together, our data are consistent with a role for CpG-29 and CpG-177 hypermethylation in influencing HACEI expression. 108 A J> J^J^jP * * i f J> * * * * <r <r </ w & «F <r <r # EcoRI EcoRI EcoRI EcoRI + Hae II + BssH II + Hpa II B relative expression of HACE1 CD CO c sz o o • • DMSO IMM 5AZ 5MM 5AZ HEK293 SK -NEP -1 ChIP: acetyl-histone H3 ChIP: methyl-histone H3 SK-NEP-1 cells input fraction Figure 4- 5: Inhibition of Methylation in SKNEP1 Cells (A) SKNEP1 Wilms' tumour cells were treated with 5-aza-2-deoxycytidine (5AZ) to inhibit methylation. Semi-quantitative PCR was then used to assess methylation status of CpG islands associated with HACE1 (see Chapter II and Figure 4-4 above). After treatment with 5AZ, methylation of CpG-29 and CpG-177 islands is decreased. No apparent affect was noted at CpG-88 (B) Quantitative RT-PCR showing an increase in HACE1 mRNA transcripts upon treatment of SKNEP1 cells with increasing doses of 5AZ. A maximal response, 4-fold change in mRNA expression, was observed at 5 u M 5AZ. (C) Chromatin immunoprecipitation (ChIP) assays to detect active or inactive chromatin were performed using antibodies against either acetyl-histone H3 (acetyl H3) or dimethyl-(Lys79)-histone H3 (methyl H3), respectively. SKNEP1 and HEK293 cells treated with or without 5 u M 5AZ prior to ChIP studies. PCR products for each cell line correspond to the transcriptional start of HACE1 (see Chapter II), indicate that SKNEP1 cells, but not HEK293 cells, alter their association with acetylated of methylated histone H3. 109 4.1.4 Expression of HACEI in other tumours HACEI m R N A transcripts are w i d e l y expressed i n n o r m a l h u m a n tissues (see Figure 3-10), i n fact observations f rom R T - P C R and q R T - P C R results indicate that expression of H A C E I m a y be ubiqui tous (not shown; see also F igure 3-3). W e have already s h o w n that the expression i n W i l m s ' t umour is considerably lower than n o r m a l k idney . To further explore whether HACEI has a w i d e r role i n h u m a n oncogenesis, w e assessed relative m R N A expression i n a broader range of cancer types. First, w e assayed expression of HACEI u s ing a c D N A dot blot composed of pa i red h u m a n patient t umour and organ-matched n o r m a l tissue samples f rom mul t ip l e t umour types u s ing a B D Clon tech Cancer P ro f i l i ng A r r a y (see also Chapter II). I nd iv idua l c D N A spots o n the array are n o r m a l i z e d for concentrat ion of 4 housekeeping genes by the manufacturer, these inc lude U b i q u i t i n , 23-kDa H i g h l y basic protein, beta-Act in , and Glu tamate dehydrogenase (see Cancer P ro f i l ing A r r a y user manua l : h t t p : / / w w r w . c l o n t e c h . c o m / c l o n t e c h / t e c h i n f o / m a n u a l s / P D F / P T 3 5 7 8 -l .pdf) . A r r a y s were p robed us ing a ful l - length HACEI c D N A a n d s ignal for "no rma l tissue" was compared to match ing t umour tissue (Figure 4-6A). This demonstrated variable but decreased expression of this gene i n several cancer types: 4 / 1 0 breast carcinomas, 5 /10 renal carcinomas, 6 /10 t hy ro id carcinomas, 4 / 5 v u l v a r carcinomas, and 2 / 3 hepatocellular carcinomas, a l though not a l l cancer types examined showed a lower level of HACEI expression w h e n compared w i t h matched n o r m a l tissue. 110 N e x t w e used q R T - P C R to assay the expression of HACEI i n the NCI -60 panel of cancer cel l l ines. The N C I - 6 0 is a set of 59 h u m a n cancer cel l l ines de r ived f rom diverse tissues; bra in , b l o o d and bone mar row, breast, colon, k idney , l u n g , ovary, prostate and sk in . The col lect ion was assembled i n 1992 and since then these cell l ines have been subjected to extensive pharmacologica l characterization, chromosome k a r y o t y p i n g and gene expression analysis. W e found that HACEI transcripts are expressed at very l o w levels i n almost a l l l ines f rom the NCI -60 panel relative to H E K 2 9 3 h u m a n embryonic k idney cells w h i c h abundant ly express HACEI (Figure 4-6B, Table 4-2; Data i n F igure 4-6 was generated by Dr . M a u r e e n O ' S u l l i v a n ( A & B ) and D r . P a u l C l a r k s o n (B)). A l t h o u g h H E K 2 9 3 cells were used as a control , they are not representative of a no rma l tissue of o r i g i n for most of these cel l lines. Unfor tunately n o r m a l tissues a n d / o r non-transformed cel l l ines for most of the cancer lines i l lustrated here were not available to our research group for compar ing HACEI expression. H o w e v e r , u s ing q R T - P C R the levels of HACEI expression i n four w e l l defined breast cancer cells l ines were compared to an immor t a l i z ed breast epi thel ial cel l l ine M C F 1 0 A (Figure 4-6C). Here I was able to compare transformed and non-transformed cells of the same or ig in ; again w e see substantially lower expression of HACEI. I l l A Breast Ovary Colon Stomach Lung Kidney Bladder Vulva Prostate * • m • * • • • • • • • » * « f • • • • * * • • • • • • • • - «, • * • • • « • • • • • • • • • • • • • •* • « • • • • • • • • • • • • * • • * • • • • • < Trachea Liver NT • • * • # * m g * • * • • • • • * • • • * • • • « • • • NT N T NT NT N T N T NT NT Uterus Cervix Rectum Thyroid Testes Skin S.Bowel Pancreas • HeU • Daudi • t <* * • • • • ft « • KS62 • * • « • • *  meo • G361 t m • • • • • • • t • * • A549 • • • • • • MOLT4 • • NT NT  SW480 • Raji NT N T NT NT NT N T E u o ' £ (A IB V 11 m X V HACE1 Expression in Breast Cancer Cell Lines 1.4, 1 0.6 0.2 ^%o^5>^^ 1 **** 112 Figure 4- 6: HACE1 Expression in Multiple Tumour Types (A) cDNA dot blot array of tumour and patient matched normal tissues. HACE1 transcript is expressed at lower levels in many tumour types, most notably in Breast, Kidney, Vulva, Liver, and Thyroid derived tumour (boxed in red). Lower expression can also be seen in many other individual pairs from across this spectrum. cDNA dot blot experiments conducted by Dr. Maureen O'Sullivan (B) Quantitative RT-PCR of the NCI-60 panel of tumour cell lines show very low level of expression in virtually all of the cell lines analyzed. Total R N A was normalized with beta-Actin and HACE1 expression was calibrated relative to 293T cell which abundantly express HACE1 (see also Table 4-2). qRT-PCR experiments conducted and analyzed in collaboration with Dr. Maureen O'Sullivan and Dr. Paul Clarkson (C) qRT-PCR for HACE1 expression in four breast cancer cell lines compared with the MCF10A immortalized breast epithelial cell line show that HACE1 is expressed at low levels in transformed breast cancer cell lines and is consistent with the data in (B) 113 Table 4- 2: NCI-60 HACE1 qRT-PCR results HACE1 in NCI-60 cell lines (illustrated in Figure 4-6B) are categorized based on tissue of origin; order of samples within each category is the same as in Figure 4-6B with relative expression +/- error. Tissues of Origin Cell Line Rel. Expression + -Callibrator 293T 1.0000 BREAST MCF7 0.0679 0.0131 0.0110 HS 578T 0.4005 0.0561 0.0492 NCI/ADR-RES 0.1473 0.0386 0.0306 MDA-MB-435 0.0533 0.0082 0.0071 MDA-MB-231/ATCC 0.1832 0.0252 0.0222 BT-549 0.6462 0.0320 0.0305 T-47D 0.0429 0.0068 0.0059 CNS SNB-75 0.7846 0.2089 0.1650 SF-268 0.1381 0.0440 0.0334 SF-295 0.2169 0.0526 0.0423 U251 0.4398 0.0416 0.0380 SF-539 0.2822 0.0317 0.0285 SNB-19 0.0611 0.0143 0.0116 COLON HCC-2998 0.1355 0.0268 0.0224 HCT-15 0.3601 0.0715 0.0596 SW620 0.2902 0.0682 0.0552 KM12 0.2612 0.0623 0.0503 COLO 205 0.1047 0.0228 0.0187 HT29 0.1502 0.0174 0.0156 HCT-116 0.0701 0.0091 0.0081 OVARIAN OVCAR-4 0.0924 0.0057 0.0054 IGROV1 2.3950 0.3376 0.2959 OVCAR-5 0.1191 0.0126 0.0114 SK-OV-3 1.0070 0.1339 0.1182 OVCAR-8 25.5455 3.7603 3.2778 OVCAR-3 0.0998 0.0127 0.0113 LEUKEMIA MOLT-4 0.5932 0.2716 0.1863 K562 0.8919 0.1194 0.1053 RPMI 8226 0.0661 0.0269 0.0191 CCRF-CEM 0.7846 0.1264 0.1089 SR 0.0665 0.0171 0.0136 HL-60 (TB) 9.8948 1.6055 1.3813 Tissues of Origin Cell Line Rel. Expression + -LUNG NCI-H322M 0.0305 0.0076 0.0061 NCI-H460 0.1877 0.0320 0.0274 HOP-92 0.0965 0.0200 0.0166 NCI-H226 0.2748 0.0625 0.0509 NCI-H23 0.1604 0.0345 0.0284 HOP-62 0.2066 0.0411 0.0343 EKVX 0.0612 0.0203 0.0152 NCI-H522 0.0506 0.0056 0.0051 A549/ATCC 0.1670 0.0228 0.0200 MELANOMA SK-MEL-2 0.1285 0.0467 0.0342 MALME-3M 0.4988 0.1139 0.0927 UACC-257 0.0282 0.0173 0.0107 SK-MEL-5 0.1303 0.0238 0.0201 SK-MEL-28 0.2207 0.0874 0.0626 M14 0.0683 0.0153 0.0125 LOX IMVI 0.1527 0.0350 0.0285 UACC-62 0.3515 0.0567 0.0489 PROSTATE DU-145 1.0943 0.3069 0.2397 PC-3 0.3186 0.0343 0.0309 RENAL A498 0.0273 0.0019 0.0018 SN12C 0.2274 0.0477 0.0394 786-0 0.4633 0.0943 0.0783 CAKI-1 0.0842 0.0173 0.0143 TK-10 0.1058 0.0380 0.0280 UO-31 0.0589 0.0119 0.0099 RXF393 0.1263 0.0018 0.0017 ACHN 0.5905 0.1209 0.1004 114 4.2 S U M M A R Y In the previous chapter w e ident if ied the H E C T fami ly E3 ubiqui t in -pro te in ligase H A C E 1 by c lon ing the chromosome 6q21 breakpoint of a t(6;15)(q21;q21) rearrangement that had occurred i n a ch i l dhood W i l m s ' t umour patient (198). Interestingly, deletions or L O H of h u m a n chromosome 6q21 have been described for a w i d e spect rum of t umour types i n c l u d i n g prostate, breast, or ovar ian cancers as w e l l as leukemias and l y m p h o m a s suggest ing that this regions encompasses one or more major t umour suppressor genes (175-177, 222, 223). For example, one recent s tudy found that ~50% of prostate cancers have deletions of the 6q21 reg ion (223). A number of candidate metastasis suppressor genes have also been m a p p e d to chromosome 6q21 (283). Here w e have demonstrated that greater than 75% of sporadic W i l m s ' t umour tested showed l o w or undetectable levels of HACE1, bo th at the R N A and prote in level . A large number of other p r i m a r y tumours also exhibi ted l o w or undetectable HACE1 expression w h e n compared w i t h patient matched n o r m a l tissue. Moreover , a majority of t umour cel l lines f rom the N C I - 6 0 col lect ion appear to have exceptionally l o w HACE1 expression. A l t h o u g h it shou ld be noted that our calibrator sample, 293T cells, is not representative of a l l no rma l tissues it does express H A C E 1 levels w i t h i n a reasonable range of no rma l k idney (see Figure 4-3C) and the results are also consistent amongst a separately calibrated group of breast cancer cel l l ines (see F igure 4-6C). Dele t ion or L O H m a y indeed p lay a large role i n some of these tumours; however a mechan i sm for the downregu la t ion of HACE1 i n W i l m s ' t umour appears to be 115 somewhat complex. Literature reports have demonstrated that 6q21 is rarely deleted in Wilms' tumour (172) and our own sequencing results support this (198). In fact, through the sequencing of D N A from 24 primary tumours with patient matched normal tissue, we were unable to find any evidence of mutations that may explain the downregulation of HACE1. As an alternate strategy, we also investigated whether methylation at the H A C E 1 locus might influence the level of expression. We found that 12/20 Wilms' tumour cases with lower HACE1 expression also had methylation at one or both of two C p G islands located ~80kb upstream (see Figure 3-6). Methylation was not observed at the most proximal C p G island or in any cases where expression was higher or unchanged in tumour tissue. It should be noted however that the assay used here is not tremendously sensitive. Future work in the lab is now looking in more detail at all three HACE1 related C p G islands using a bisulphite-sequencing strategy (284, 285) to examine each individual C G dinucleotide in the Wilms' tumour cohort discussed above. These methylation studies are being done in collaboration with Dr. Robert Arceci, Johns Hopkins University, Baltimore, M D . Methylation has been cited to be as common as mutations in silencing tumour suppressor genes (147). Gene regulation from a great distance has been described (286, 287) in the literature, and in fact there are cases that show methylation can control a gene from a great distance upstream (288, 289). The predominant theory for silencing leads from the fact that chromatin condensation is self-propagating and can spreads into neighbouring gene loci (287, 290). In vitro experiments with SKNEP1 Wilms' tumour cells support a link 116 between the HACE1 ups t ream C p G islands a n d the regula t ion of the HACE1 gene. In these cells the use of a methy la t ion inhibi tor not on ly reduces the level of methyla t ion at ups t ream C p G islands but also restores HACE1 expression and opens the chromat in conformat ion at the start of the HACE1 gene. In this case, u n d o i n g or prevent ing the spreading of condensed chromat in c o u l d release the HACE1 gene f rom repression. This chromat in condensat ion m o d e l may expla in one major mechan i sm for s i lencing HACE1 i n W i l m s ' t umour both f rom upst ream C p G islands and th rough i m p o r t i n g of heterochromatin f rom translocation events near the HACE1 locus. 117 CHAPTER V EFFECTS OF ALTERED HACE1 EXPRESSION IN VITRO A N D IN VIVO The project described in this chapter was conceived and initiated by the laboratory of Dr. Poul H.B. Sorensen. Portions of this chapter have been published in two peer reviewed publications: Anglesio, M.S., Evdokimova, V., Melnyk, N . , Zhang, L. , Fernandez, C.V., Grundy, P.E., Leach, S., Marra, M.A. , Brooks-Wilson, A.R., Penninger, J. and Sorensen, P.H. (2004) Differential expression of a novel ankyrin containing E3 ubiquitin-protein ligase, Hacel , in sporadic Wilms' tumour versus normal kidney. H u m Mol Genet, 13, 2061-74. In this publication I generated the majority of the data, contributed to the writing. Zhang L*, Anglesio MS*, O'Sullivan M , Zhang F, Yang G, Sarao R, Nghiem MP, Cronin S, Hara H , Melnyk N , L i L , Arya S, Wada T, Liu PP, Sorensen P H , and Penninger J. (2006) The E3 ligase H A C E 1 is a critical chromosome 6q21 tumour suppressor involved in multiple cancers. Manuscript submitted In this publication I generated the majority of in vitro data, and in cooperation with Dr. Josef Penninger and Dr. Poul Sorensen, contributed to the writing. * These authors contributed equally to this work Acknowledgements for the data contained in this chapter that ivas generated by other members of the Dr. Poul H.B. Sorensen's Lab and collaborators is noted in the figure legends. 5.1 I N T R O D U C T I O N The previous chapters described the molecular characterization of the 6q21 region involved in the Wilms' tumour t(6;15) translocation, the cloning of a novel E3 ubiquitin-protein ligase, HACE1, and the altered expression of HACE1 in Wilms' tumours and other neoplasms. Evidence has been presented suggesting. that low HACE1 expression in Wilms' tumour is linked to methylation of two upstream C p G 118 islands, and that the common loss of expression across a variety of tumour types may be an etiologically significant event in cancer progression. In this Chapter I will examine the potential role of HACEI as a tumour suppressor gene in Wilms' tumour. Transformation and cancer growth are regulated by the coordinate actions of oncogenes and tumour suppressor genes (84, 168, 291, 292). Characterization of tumour suppressor proteins has revealed diverse functions in the regulation of cell growth, including roles in cell cycle checkpoint responses, detection and repair of D N A damage, protein ubiquitination and degradation, apoptosis, differentiation, migration, or tumour angiogenesis (168). We have specifically examined a number of growth characteristics related to HACEI expression using two models: first, through overexpression in a Wilms' tumour derived cell line, and second, using R N A i mediated HACEI downregulation in a kidney derived cell line. In addition, I will discuss results generated in the laboratory of Dr. Josef Penninger whose collaboration on this project has involved the generation of a hacel mutant mouse (see Appendix B). I will describe how, using these models, we will examine changes in the growth characteristics of cells with altered HACEI expression and changes in a number of cell survival and proliferation pathways that were observed. Evidence will be provided that H A C E I affects the stability of cyclin D l and the level of activation of A K T and that these changes in signalling pathways are likely major contributors to H A C E I mediated growth attenuation in vitro and possibly tumourigenesis in vivo. 119 5.2 R E S U L T S 5.2.1 Expression of HACEI and HACEKC876S) Mutant in SKNEP1 Wilms' Tumour Cells To assess the potential role of H A C E I i n suppress ing cel l g r o w t h of W i l m s ' t umour cells, w e used a M S C V based retrovirus to stably integrate an H A - t a g g e d HACEI gene, or ligase dead C876S mutant, into S K N E P 1 cells (Figure 5-1 A). The morpho logy of the cells d i d not appear to be altered by the expression of H A - H A C E 1 . W e therefore decided to assess the g r o w t h rates of the transfected cells. W h e n cells were plated at l o w confluence and assayed by direct cel l count ing d u r i n g the exponent ia l g r o w t h phase, there was no significant difference i n g r o w t h rate (Figure 5-1B). H o w e v e r , w h e n cells were plated at higher densities and the g r o w t h rate was moni to red as the cells passed a confluent state a clear difference between HACEI transfected and empty vector cells was more apparent (Figure 5-1C). 120 a-HACE1 •* a-Actin B 4 5 6 7 Days in culture • II CD o e O MSCV • HA-HACE1 U n P < 0.02 Days in culture Figure 5-1: HA-HACE1 Overexpression in SKNEP1 Cells (A) Western blot showing overexpression of HA-HACE1 or HA-C876S mutant in SKNEP1 cells. Note low level expression of HACE1 in vector alone (mscv) cells, band is visible only because of extended exposure of film. (B) SKNEP1 cells (plated at 10 000 cells/well) expressing HA-HACE1 were plated in 6-well dishes and counted over 8 days. Arrow denotes the day when cells became fully confluent in the dish. Growth curves separate as the cells become confluent. (C) SKNEP1 cells (plated at 5000 cells/well) were plated in 96-well dishes and tritiated thymidine incorporation was measured every day, over 4 days. As above, arrow denotes the day when cells became fully confluent in the dish. Note thymidine incorporation increases at a faster rate in vector alone (MSCV) cells. Thymidine incorporation assays were performed by Dr. Valentina Evdokimova. 121 We next analyzed effects of HACEI or ligase dead mutant HACEI (C876S) overexpression on anchorage-independent growth of transformed cells. This C876S mutant is unable to bind ubiquitin as the conserved cysteine residue required for this function has been replaced with a serine residue unable to form a thio-ester bond (See also Chapter III, and Figure 3-11). We found in these experiments that wild-type, but not mutant, H A - H A C E 1 significantly reduced the ability of cells to form colonies in soft agar (Figure 5-2A & B). We found that the overexpression of H A C E I not only reduced the number of colonies formed but also the size of colonies that were formed. To assess the growth suppressive effects of Hacel in vivo, SKNEP1 cells stably overexpressing wild-type H A - H A C E 1 or C876S mutant were assayed for their ability to form tumours after injection into nude mice. Both SKNEP1 vector alone and C876S mutant expressing cells formed visible tumours 10-15 days after injection, which continued to increase in size over time up to the termination of the experiment (Fig. 5-2C & D). This is consistent with previous reports for this cell line (246, 247). In contrast, tumorigenicity was profoundly inhibited for SKNEP1 cells overexpressing wild-type H A C E I . Only rare very small nodules were observed in mice injected with these cells over the entire course of the experiment. 122 MSCV HA-HACE1 C D SKNEP1 Tumour Growth in Nude Mice Days post injection Figure 5- 2: Growth Characteristics of HACEI Overexpressing SKNEP1 Cells (A) Soft agar assay showing reduced colony formation with the expression of wild-type HA-HACE1, but not mutant HA-C876S or empty vector (MSCV) (B) Colonies formed with H A C E I overexpression were less abundant and smaller in size than cells containing empty vector (MSCV) (C) SKNEP1 cells (lxlO 7 cells/site) were injected subcutaneously into nude mice. Cells expressing wild-type HA-HACE1 were impaired in their ability to form tumours over a 30 day time period. (D) Representative tumours formed in (C) can be seen on the backs of nude mice only when expressing empty vector (MSCV) or mutant (HA-C876S; not shown) but not with HA-HACE1. Animals were monitored by Mary Bowden in the laboratory of Dr. Martin Gleave. 123 Interestingly, these results do not appear to be restricted to W i l m s ' rumour de r ived cells. A s imi la r set of experiments were also conducted i n K-ras t ransformed N I H 3 T 3 cells. C e l l cycle analysis by PI s ta ining and f l o w cytometry showed that w h e n the cells were over-confluent a greater percentage of K - r a s / H A C E I expressing cells accumulated w i t h a 2 N D N A content, consistent w i t h G o / G i (Figure 5-3A). This suggested that a greater por t ion of these cells are either exi t ing the cel l cycle, entering G o , as w o u l d be expected i n confluent non-transformed cells, or are be ing act ively s topped at a G i checkpoint . Moreover , the p ropor t ion of mitot ic (G2; 4 N D N A content) K - r a s / H A C E I expressing cells was marked ly decreased (~5%) i n compar i son to their K-ras alone counterparts (12%), as was the fraction of S-phase cells. Further, neuroblastoma cel l l ine I M R - 3 2 also exhibi ted H A C E I dependent g r o w t h suppress ion w h e n cells were plated i n soft agar (Figure 5-3B). 124 Cell cycle profile K-ras -HACE1 0 200 400 0 200 400 Fluorescence Intensity c o *s <s o u. >. c o o u 3? 40 30 20 10 0 IMR-32 Soft Agar Assay • Figure 5- 3: HACEI Growth Suppressive Effects in Transformed Cells (A) FACS cell cycle profile of K-ras transformed NIH3T3 cells with or without HACEI overexpression. Note higher proportion of mitotic cells (G2:12%) in K-ras alone cells vs. HACEI overexpressing cells (G2: 4.75%). There is also a very slight decrease in S-phase and accumulation of cells in the G i / G 0 peak with HACEI overexpression. FACS experiments were conducted by Natalyia Melnyk and Dr. Valentina Evdokimova. (B) Soft agar assay showing wild-type HA-HACE but not mutant (HA-C876S) or vector alone (MSCV) suppresses anchorage independent growth in IMR-32 neuroblastoma cells. Experiment conducted by Dr. Maureen O'Sullivan. 5.2.2 RNA Interference & Stable Knock-Doum Of HACEI Since HACEI expression was or ig ina l ly observed to be downregula ted i n W i l m s ' tumour , w e dec ided to see if w e c o u l d recreate this downregu la t ion i n a cel l l ine m o d e l u s ing R N A i . Trad i t iona l transfected sma l l interfering ( s i ) R N A molecules typ ica l ly have an effectiveness over a course of on ly a few days (293). I therefore chose to use a len t iv i ra l based system to stably express short ha i rp in R N A s against HACEI. U s i n g this me thod I c o u l d then use our k n o c k d o w n cells for longer culture experiments wi thou t the need for repeated transfections w i t h s i R N A s (see also Chapter II)(293). 125 Before u s ing the len t iv i ra l system a number of potential HACE1 specific R N A i molecules were screened by transiently transfecting them into H E K 2 9 3 cells (Figure 5-4 A ) . T w o of these sequences ( N I and N2) were chosen to adapt into a Len t iv i r a l -s h R N A vector, and scrambled versions of each sequence were generated to serve as negative controls i n R N A i experiments (see also Chapter II). Lent ivi ruses were constructed u s ing the Invitrogen's B lock- iT Len t iv i rus construct ion k i t and app l i ed to target cells. H E K 2 9 3 cells were selected for this m o d e l as they expressed abundant levels of HACE1, were de r ived f rom h u m a n embryonic k i d n e y (s imilar ly to W i l m s ' tumour) , and g rew s l o w l y i n bo th soft agar and i n nude mouse models (294, 295). Once the cells had undergone selection they were assayed for H A C E 1 expression. O n l y the N I s h R N A construct appeared to rel iably k n o c k d o w n H A C E 1 expression w h e n conver ted to an s h R N A (Figure 5-4B, see also Chapter II). U s i n g these k n o c k d o w n cells, and their respective scrambled contro l (SI; see Chapter II for "scrambled" sequence) cells w e assayed for g r o w t h i n soft agar a n d t umour format ion i n nude mice (Figure 5-4C-E; Data i n figure 5-4C generated i n part by D r . F a n Zhang) . H A C E 1 k n o c k d o w n correlated w i t h an increase i n soft agar co lony format ion of H E K 2 9 3 cells expressing N I s h R N A but not the S I scrambled construct. Further, w h i l e cells expressing the scrambled s h R N A construct were non- tumour igenic i n nude mice, as expected, those expressing HACE1 s h R N A s h o w e d a dramatic increase i n tumorigenic i ty in vivo (Figure 5-4E). 126 B i HEK293 S1 N1 HACE1 Actin HACE1 Actin HEK293-S1 HEK293-N1 E HEK293 Growth in Nude mice 6001 Time (Days post injection) Figure 5- 4: HACE1 RNAi in HEK293 Cells (A) Transient transfection with oligo siRNAs were used to screen a number of HACE1 specific target sequences. Western blot shows HACE1 downregulation following treatment with one or more siRNAs. Western blot performed by Dr. Fan Zhang. (B) NI sequence adapted into a shRNA and stably introduced into HEK293 cells using a Lentivirus vector reduces HACE1 expression > 60% compared with scrambled control SI. (C) Soft agar colony formation is dramatically increased in HEK293 cells carrying the NI shRNA compared with control SI shRNA containing cells. Soft agar assays were performed in collaboration with Liheng Li and Dr. Fan Zhang (D) Representative photographs of soft agar colonies formed in (C) at lOOx magnification (E) Tumour formation in Nude mice is enhanced when HACE1 expression is suppressed using NI shRNA in HEK293 cells. Control SI shRNA containing cells fail to form measurable tumours. Animals were monitored by Mary Bowden in the laboratory of Dr. Martin Gleave. 127 More recently commercially available, chemically modified, long-term, transient s iRNA molecules have been developed (296, 297). In order to complement the aforementioned data and in an < effort to rule out off-target effects of the R N A i treatment we obtained three chemically modified "stealth" s iRNA molecules specific for H A C E 1 along with a control. Using these modified s iRNA molecules we consistently achieved knockdown of H A C E 1 in HEK293 cells for more than 7 days (Figure 5-5A). Unfortunately this time frame was not long enough for traditional soft agar assays which typically take 18-21 days (199, 200). We therefore employed a more quantitative soft agar assay in which equal numbers of cells are plated in 96-well format, suspended in a media:soft agar mix, grown over only 1 week, and then lysed. The lysates are assayed for nucleic acid content using a fluorogenic dye and the fluorescence is used as a direct proxy for cell number (see Chapter II). Using this method we found that the effect of H A C E 1 knockdown on cell division was almost dose dependent, with constructs able to knockdown H A C E 1 to the lowest levels exhibiting the largest gains in soft agar growth. One of our three stealth siRNA (387) was only able to knockdown H A C E 1 protein -25-30% and achieves a - 1.5 fold increase in soft agar growth. Two other stealth siRNAs (385 and 386) achieved > 90% knockdown and up to 2.5 fold increase in soft agar growth (Figure 5-5). Taken together, these data show the ability of H A C E 1 to inhibit soft agar colony formation and tumorigenicity, suggesting that overexpression may oppose the oncogenic qualities of transformed cells. At the same time, knock-down of H A C E 1 increases 128 growth in soft agar and tumour growth in nude mice, supporting that HACE1 possesses tumour suppressor like qualities, without which cells may have a growth advantages under specific conditions. # # 4 & 4 4 c / ° / Day 1 HACE1 Actin Day 7 B Long Term Transient RNAi Soft agar cell growth after 7 days in culture Figure 5- 5: Long Term Transient Stealth R N A i (A) Stealth R N A i oligos were used to knockdown HACE1 expression in HEK293 cells, Western blot analysis demonstrates that HACE1 knockdown starts as early as day 1 post-transfection and continues through day 7. (B) HEK293 cells treated with stealth R N A i (as in (A)) were used in a quantitative soft agar assay and analysed for cell growth (as a function of fluorescence; see text and Chapter II) after 7 days soft agar culture. 129 5.2.3 A hacel Mutant Mouse Model To study the in vivo function of H A C E I and to aide in providing genetic proof that this gene is indeed involved in cancer pathogenesis, hacel mutant mice were generated. Please refer to Appendix B for all data concerning the hacel mutant mouse model. The data presented were generated exclusively by a post-doctoral fellow, Liyong Zhang, in the laboratory of Dr. Josef Penninger at the University of Toronto, Toronto, Ontario, Canada and at the Institute for Molecular Biotechnology of the Austrian Academy of Sciences (1MBA), Vienna, Austria (unless otherwise stated). 5.2.4 Altered Signal Transduction and Interactions with HACEI We have previously shown that overexpression of H A C E I can suppress growth, particularly in an anchorage independent environment. We have also shown that downregulation or loss of H A C E I can contribute to increased growth in vitro and tumorigenesis in vivo. In order to uncover the mechanisms by which H A C E I may be affecting cell growth, we have examined a number of growth related signal transduction pathways. Interestingly, in H A - H A C E 1 overexpressing HEK293T cells we see a very consistent downregulation of cyclin D l and very low levels of phospho-AKT (Serine 473) (Figure 5-6A). As a screen for cell cycle regulators, cells were synchronized by thymidine/aphidicolin in fresh medium to block cells at the Gl -S phase, released by washing and medium change, and then assayed for expression of cell cycle proteins by 130 Western blot t ing. There was no difference i n the levels of cycl ins A , cyc l i n B , or cyc l i n G l a m o n g the different cel l l ines over the t ime pe r iod ana lyzed after release f rom the G l - S phase b lock (Figure 5-6B). A s noted above, w e observed a m a r k e d reduc t ion i n c y c l i n D l i n t h y m i d i n e / a p h i d i c o l i n treated cells overexpressing H A G E 1 compared w i t h those expressing H A C E 1 - C 8 7 6 S or vector alone. C y c l i n s D 2 a n d D 3 were either not affected or were expressed at undetectable levels i n these cells (data not shown). Since D- type cycl ins are m a x i m a l l y expressed i n the early G l cel l cycle phase and are induced by mitogens (298), w e next examined their expression i n cells s tarved of se rum overnight and then s t imulated for var ious t ime points w i t h 10% se rum (Figure 5-6C). W i t h the exception of c y c l i n D l , there was no difference between cel l l ines i n the levels of other cycl ins . In t r iguingly , activated A K T ( p h o s p h o - A K T ; Serine 473), but not total A K T , was also at a m u c h lower level i n cells overexpressing wi ld - type H A -H A C E 1 , as was the downs t ream effector p G S K 3 p and the c y c l i n dependent kinase C D K 4 . In addi t ion , no changes i n p21 or p27 were observed and, cons ider ing the s lower g r o w t h of H A C E I overexpressing cells, it seems counter intui t ive to note a sl ight increase i n p h o s p h o - M E K , an effector whose act ivat ion is n o r m a l l y associated w i t h cell g r o w t h and s u r v i v a l (299, 300). Important ly, not on ly were cyc l i n D l and p h o s p h o - A K T marked ly reduced i n cells overexpressing H A C E I compared to vector alone cells under exponential g r o w t h condi t ions (Figure 5 - 6 A & C ) , but s tarved H A C E I overexpressing cells fai led to up-regulate c y c l i n D l or activate A K T after mi togen re-s t imula t ion th rough the add i t i on of fresh se rum (Figure 5-6C). 131 8 8 I j mrmti* cyclinD2 *» ~ •* cvclinDI | - > PAM jjj^Akt pGSK3p L Z H p E R K B Time after release, hr: -4 Cyclin A > O m fg -» Cyclin B — — — — «*!•«! Cyclin D1 Cyclin G1 «* — « cdk6 o u I •4 Cyclin A », — —1-< Cyclin B |-< Cyclin D1 •4 Cyclin G1 cdk6 W tp r~ co O — _ < Cyclin A , m» Cyclin B - m< Cyclin D1 >la — •* Cyclin G1 • «• «H"< cdk6 MSCV HA-Hace1 e e e e e e e e 5 £ £ £ £ £ £ £ £ £ Serum, hr: o - « • « > » O - N <o to •11 • - — I , •< HA-Hace1 •< Cyclin A < Cyclin B -< Cyclin D1 _ _ .. -< Cyclin E -< Cyclin G1 < cdk4 * ••§>•«•#» . • • > « • • •* P27/KIP • •»«••»•> • • » • » • • •< P21/CIP •< IGFRR •* pAKT . . . •< total Akt < pGSK3|! < total GSK3|i „ - - - » - * — pMEK PABP 1 2 3 4 5 6 7 8 9 10 11 12 Figure 5- 6: Cell Cycle Alterations in H A C E I Overexpressing HEK293T Cells (A-C) Experiments were conducted by Dr. Valentina Evdokimova and Dr. Fan Zhang with Liheng Ei. (A) Western blot illustrating expression of selected cell cycle and growth related signalling molecules in exponentially growing HEK293T cells over expressing H A - H A C E (or empty vector, MSCV). (B) Expression of cell cycle molecules in HEK293T cells stably overexpressing HACEI , ligase dead H A C E I (C876S), or vector alone (MSCV). A l l cells were synchronized with thymidine/aphidicolin and released from the cell cycle arrest by washing and application of fresh media with 10% FBS. Expression of cell cycle molecules was determined by Western blotting at the indicated time points in hours after release (hrs). (C) Western blot illustrating expression of cell cycle molecules in HEK293T cells stably overexpressing H A C E I or transfected with vector alone (MSCV). Cells were stimulated for 0 (dashed line), 2,4, 6, or 8 hrs with 10% serum after overnight serum starvation. Ctrl refers to control cells under exponential growth conditions in 10% serum. 132 To verify these observations, we next attempted to rescue cyclin D l expression in H A C E I overexpressing HEK293T cells by reducing H A C E I levels with R N A interference. Down-regulation of H A C E I expression using specific siRNAs but not a scrambled control indeed resulted in increased cyclin D l , phospho-AKT and phospho-GSK3(5 expression in these cells (Figure 5-7A). Further, treatment of cells with the translation inhibitor cyclohexamide demonstrated that cyclin D l stability is comparable to that of cyclin B and cyclin E over a 10 hr time course, but that the stability of cyclin D l is markedly reduced when H A C E I is overexpressed (Figure 5-7B). Cyclin D l levels became stabilized in H A C E I overexpressing cells treated with the proteasome inhibitor MG132 (Figure 5-7C). In contrast, levels of phospho-AKT do not appear to be restored by MG132 (data not shown). These data suggest that loss of cyclin D l in these cells occurs through degradation by the 26S proteasome and not through alterations in cyclin D l translation. 133 RNAi HA-HACE1 pAKT Total AKT pGSK3|l Cyclin D1 293T HA-HACE1 B MSCV HA-Hace1 CHX, hr: 100 N <t iom ? . IN 10 to J •* HA-Hace1 i Cyclin A 54.4 J** * * * * * * * < | ^ ! M 4 ) i — • <— - ~ - » ^ | — <^ Cyclin B ""U Cyclin D1 38.7 -} 54.4 { Cyclin E 1 2 3 4 5 6 7 8 91011 12 MSCV HA-Hace1 MG132, hr: • 100 -C CM * (0 CO ? . N 4 » • r r —< HA-Hace1 54.4 -E 54.4 38.7 -E 54.4 -F" • Cyclin A >«»•• •<n»|^ Cyclin B • •»—•]•< Cyclin D1 Cyclin E 1 2 3 4 5 6 7 8 9 10 11 12 Figure 5- 7: Rescue of Altered Signal Transduction in HEK293T Cells (A) Western blot illustrating that reduced phospho-AKT, phospho-GSK3(3, and cyclin D l can all be recovered by downregulation of HACE1 using a the s iRNA NI , but not the scrambled control SI. Experiments conducted by Dr. Fan Zhang. (B) Protein stability of indicated cyclins in M S C V and HA-HACE1 transf ected HEK293T cells following treatment with 10 u,g/ml of the translation inhibitor cyclohexamide (CHX). Stability of cyclin D l is markedly reduced when HACE1 is overexpressed. Experiments conducted by Dr. Valentina Evdokimova. (C) Protein expression of the indicated cyclins in M S C V and HA-HACE1 transfected HEK293T cells following treatment with 10 uM of the proteasome inhibitor MG132. Virtually all of the cyclin proteins, including cyclin D l , are at steady state levels over the time course observed. Experiments conducted by Dr. Valentina Evdokimova. Last ly , w e have identif ied a number of b i n d i n g partners for H A C E 1 th rough immunoprec ip i t a t ion and mass spectrometry analysis. A number of different proteins were identif ied and can be classified as R N A b i n d i n g proteins, proteins b o u n d to the ER, prote in chaperones, and other proteins (Table 5-1). W e have also been able to val idate interactions w i t h V C P / p 9 7 , H S C 7 0 , the proteasome a n d u b i q u i t i n by co-immunoprec ip i t a t ion (Figure 5-8). Nei ther V C P nor H S C 7 0 appear to be a target for 134 H A C E 1 ubiquitination (data not shown), however, they may be parts of larger complexes interacting with or bringing target proteins in proximity to H A C E 1 for subsequent ubiquitination and degradation via the proteasome. Interestingly, one of the chaperone proteins interacting with H A C E 1 , Valosin containing protein ( V C P / p97), is already known to interact with ubiquitinated proteins and is thought to deliver them to the 26S proteasome. 135 B IP HA-Hace1 1 2 3 4 normal Poly-Ub. conjugates IgG -+-VCP IgG Hsc70 IgG - HA-Hace1 IgG Figure 5- 8: HACE1 interacting proteins (A) Immunoprecipitation of components of the ubiquitin-proteasome pathway also co-precipitates both HA-HACE1 and V C P from HA-HACE1 overexpressing HEK293T cell lysates. Lane 1 (preim; pre-immune antibodies) serve as a negative control for immunoprecipitation, lane 2 (a ubiq) using anti-ubiquitin antibodies, lane 3 (a 20S) using antibodies against the 20S proteasomal subunit, and lane 4 (a Hacel) using HACE1 antibodies serves as a positive control. Upper panel is probed with anti-HA antibodies. Lower panel is probed with anti-VCP antibodies. (B) Immunoprecipitation of HA-HACE1 co-precipitates interacting proteins V C P and Hsc70 from HA-HACE1 overexpressing HEK293T cell lysates. Antibodies used for Immunoprecipitations are indicated at the top (preim; pre-immune antibodies serve as the negative control). Each panel was immuno-blotted with antibodies against the indicated protein (a-HA used for HA-HACE1 and cc-ubiquitin used for poly-Ub conjugates). (A-B) Experiments were conducted by Dr. Valentina Evdokimova 136 Table 5- 1 HACE1 Interacting Proteins Identified by mass spectrometry in the laboratory of Dr. Karl Mechtler RNA-binding proteins Proteins bound to endoplasmic reticulum Protein Chaperones Others 1. Poly(A) Binding Protein (PABP) 1. Sorting Nexin 9 (SH3 and PX Domain-Containing Protein SH3PX1) 1. Valosin Containing Protein (VCP/p97) 1. Interleukin Enchancer Binding Factor 2 2. Cytotoxic Granule-Associated RNA-Binding Protein (TIA-1) 2. Yeast Sec Op-related Protein 2. Heat Shock 70 kDa protein 8 (Hsp73) 2. Thyroid Autoantigen 70 kDa 3. Heterogeneous Nuclear Ribonuclear Protein (hnRNPAl) 3. Yeast Sec31 p-related Protein 3. dTDP-4-keto-6-deoxy-D-glucose 4-reductase 4. Heterogeneous Nuclear Ribonuclear Protein (hnRNP A2) 5. Heterogeneous Nuclear Ribonuclear Protein H3 (hnRNP H3) 6. Similar to Nucleolin 7. IGF-II mRNA Binding Protein 1 8. IGF-II mRNA Binding Protein 3 9. Eukaryotic Translation Initiation Factor 3, 30909-kDa subunit 10. Eukaryotic Translation Initiation Factor 3, 36479-kDa subunit 11. Phenylalanyl-tRNA Synthetase, 3-subunit 12. Acidic Ribosomal Phosphoprotein PO 13. Ribosomal Protein L5 137 5.3 S U M M A R Y In this chapter w e have explored the role of HACEI as a t umour suppressor gene. Overexpress ion of H A C E I seems to have a general g r o w t h suppressive effect i n mul t ip l e cel l types both in vitro and in vivo. N I H 3 T 3 cells t ransformed by activated K -Ras h a d marked ly reduced g r o w t h rates w h e n engineered to co-express H A C E I . Moreover , anchorage independent g r o w t h of S K N E P 1 h u m a n W i l m s ' t umour and I M R 3 2 neuroblastoma cells, bo th of w h i c h show very l o w endogenous H A C E I expression, was dramat ica l ly suppressed w h e n engineered to express H A C E I . Important ly, in vivo tumor igenic i ty i n nude mice was p ro found ly inh ib i t ed i n S K N E P 1 cells overexpressing H A C E I . Converse ly , w h e n endogenous H A C E I levels were reduced by mul t ip l e R N A interference constructs i n h u m a n embryonic k idney H E K 2 9 3 cells, this correlated w i t h a p ro found increase i n soft agar co lony format ion and a dramatic increase i n tumorigenic i ty in vivo compared to w i l d type cells or scrambled controls. O u r data also demonstrate that the g r o w t h suppressive functions of H A C E I depend on its E3 ubiqui t in -pro te in ligase act ivi ty, as our H A C E I mutant (C876S) does not have any effects o n cell g r o w t h in vitro or in vivo. Thus , our data identify H A C E I as an E3 ligase dependent t umour suppressor that regulates g r o w t h and tumorigenic i ty of h u m a n cancer cells. Further, a mouse knockou t m o d e l of hacel gave rise to increased spontaneous tumour format ion w i t h a diverse range of t umour types compared to w i l d type littermates. Hacel deficient mice were more susceptible to D N A damag ing agents 138 including ionizing radiation and the alkylating agent urethane. Interestingly, crossing hacel mutant mice with p53 mutant mice resulted in more rapid tumour formation and again a broader range in tumour types than are normally associated with p53 null mice. However, crossing hacel mutations with a p21 tumour suppressor gene null line did not appear to increase the rate of onset or spectrum of tumours observed in hacel mutant mice. These data not only support a tumour suppressor gene role for hacel but implies that it may cooperate with p53 in transformation. O n the other hand, the lack of change in tumour incidence or tumour spectrum in hacel mutant versus hacel/p21 double mutant mice may suggest that the same or parallel pathways are affected by loss of function in either of these to genes. For example, both cyclin D l and, to a lesser extent, CDK4 appear to be downregulated in H A C E I overexpressing cells (Figure 5-6) while p21 has been shown to inhibit the activity of cyclin D / C D K 4 complexes (97, 301, 302). It is therefore conceivable that in the case of hacel, p21, or hacel/p21 mutants a primary, and common, deficiency may be in their ability to regulate cyclin D / C D K 4 activity, although more work remains to done to explore this possibility. At the molecular level, H A C E I appears to regulate cell cycle progression by modifying the stability of cyclin D l , and possibly the activity of A K T and other molecules. Roles of D-type cyclins in oncogenesis are well established (303), and overexpression of cyclin D l has been described in a wide range of human malignancies (refer to Appendix B). Similarly, overexpression and activation of A K T has been cited as a marker of poor prognosis in a number of malignancies (304-306). Further, active A K T is sufficient to effect transformation in vitro and in vivo (307-310). 139 Lastly, decreased expression of CDK4 has been linked to accumulation of cells in G l / S independent of changes to other cell cycle regulatory proteins such as cyclin D l and Rb (311). We found that synchronized HEK293T cells overexpressing H A C E 1 exhibited a profound defect in their ability to re-express cyclin D l and CDK4, and phosphorylate A K T when released from synchronization or when serum-stimulated, compared to cells expressing vector alone or a ligase dead H A C E 1 mutant. Other cell cycle regulatory proteins remained relatively unaffected. Cyclin D l appeared to be less stable in H A C E 1 overexpressing cells; however, treatment with the proteasome inhibitor MG132 or administration of HACE1 specific R N A i restored cyclin D l levels and enhanced the activation of A K T . These findings suggest a possible mechanism whereby H A C E 1 can regulate exit from the cell cycle in response to various forms of cell stress by reducing cyclin D l and CDK4 protein levels in an E3 ligase and proteasome dependent manner. In accordance with this notion, at least one recent study has demonstrated that MEFs lacking D-type cyclins show reduced susceptibility to oncogenic transformation by various oncogenes, and that these cells have an attenuated ability to respond to extracellular mitogenic stimulation (312). Taken together the data presented in this chapter support the notion that HACE1 is a novel tumour suppressor gene that is able to attenuate cell proliferation in a wide variety of cell types. The tumour suppressor activity of H A C E 1 is dependent on its E3 ligase activity, as cell growth is not significantly affected by a ligase dead mutant (C876S). Finally, H A C E 1 appears to specifically regulate cell cycle progression through its effects on A K T , cyclin D l , and possibly CDK4, all of which are associated with 140 progression through the G l / S cell cycle checkpoint (313-318). Further discussions on cooperation with other tumour suppressor such as p53 and p21 as well as potential models of H A C E 1 activity will be explored in the final chapter (see also Appendix B). 141 C H A P T E R V I S U M M A R Y A N D CONCLUSIONS 6.1 HACEI EXPRESSION A N D S I L E N C I N G In the previous chapters I have described the molecular characterization of the 6q21 reg ion ident if ied near the W i l m s ' t umour t(6;15) translocation. A s ment ioned: deletions, L O H , breakages and rearrangements of h u m a n chromosome 6q21 have been described for a w i d e spect rum of t umour types i n c l u d i n g prostate, breast, or ovar ian cancers as w e l l as leukemias and l y m p h o m a s , suggest ing that this reg ion encompasses one or more major rumour suppressor genes (131, 175-177, 222, 223). A l t h o u g h L O H and chromosomal rearrangements at this reg ion are rare i n W i l m s ' t umour at least two other un ique translocations have been described that i nvo lve the same 6q21 locus (172-174). M a p p i n g experiments revealed that the 6q21 breakpoint i n the W i l m s ' t umour (index case) occurred direct ly ups t ream of a nove l gene predic ted to encode a prote in w i t h a p rev ious ly unrepor ted d o m a i n architecture. M y results s h o w that this nove l gene, HACEI, is a member of the H E C T fami ly of E3 ubiqui t in -pro te in ligases. I have s h o w n that HACEI expression is except ional ly l o w or down-regula ted i n mul t ip l e h u m a n tumour types, not on ly W i l m s ' tumour , at bo th the R N A and pro te in levels. I next l ooked for a possible genetic mechan i sm for downregula t ion . Sequencing of the genomic D N A i n 25 W i l m s ' tumours and patient matched controls suppor ted that L O H was not l i ke ly a cause for loss of expression, a l though, this 142 particular series of experiments cannot exclude smaller scale deletions of whole exons. Further, no apparent "loss of function mutations" were found. In fact, all of the SNPs found during the sequencing project were constitutional changes, affecting both normal and tumour D N A . Only 2 changes within the coding region of HACE1 were discovered: one silent, affecting 2 cases, and the other falling outside of any identifiable functional domain or motif. Overall this indicates that L O H , deletions or other mutations are unlikely to be the cause of downregulation in Wilms' rumour. However, given the frequency of L O H and rearrangements at this region in other tumours it seems reasonable that genetic mechanism like these could have some effect on the HACE1 locus. Clearly further investigation is warranted, especially in the above mentioned tumours where abnormalities of 6q are commonplace. Unfortunately, our sequencing strategy did not examine the 5' promoter region for mutations that may result in the inability of transcription factors and enhancers to bind leading to lowered transcription, as has been reported for RUNX2 (135, 319) and PTEN (320). The 3' UTR was also omitted from initial analysis, and should be examined more closely in future work. In this case, deletion or mutations of this region may alter the stability of the HACE1 m R N A (321, 322), and have not been ruled out in Wilms' tumour. After having examined the HACE1 locus for potential genetic mechanism of downregulation the potential for epigenetic changes was explored. There are 3 C p G islands located in proximity to the HACE1 gene. I measured the level of methylation at each of these using a methylation sensitive restriction digest and semi-quantitative PCR strategy. We found increased methylation at two upstream islands, however, the 143 assay was unable to detect methylation at the most proximal CpG-88 island (refer to Chapter IV and Figure 3-9). Methylation correlates well with decreased expression: 20/26 cases show decreased expression, 12/20 cases with low expression also have increased methylation, and 0/6 cases with increased or no difference in expression detectably increases in methylation. In vitro use of the methylation inhibitor 5-AZ on SKNEP1 cells was also able restore expression of HACEI. While detailed analysis of the chromatin conformation by ChIP confirmed that the structure of chromatin at the HACEI locus opened in response to 5-AZ treatment. Taken together this strongly suggests that methylation surrounding the HACEI locus exerts some control on the expression of this gene. However, it must also be noted that our experimental design is not sensitive enough to detect all methylation changes and is likely to miss some regions of methylation resulting in false negatives. This is especially true for C p G dinucleotides residing outside of the methylation sensitive RE recognition site, of which there are many. More detailed examination of these C p G islands, could provide further insight on epigenetic transcriptional control at the HACEI locus. As mentioned in the summary for Chapter IV, Dr. Sorensen's lab is currently collaborating with Dr. Robert Arceci, of Johns Hopkins University, in order to evaluate individual methyl-cytosines within each HACEI associated C p G island. In fact, preliminary data appears to support our original finding that the CpG-88 island has no detectable methylation, the CpG-177 island is hypermethylated in Wilms' tumour samples, and the CpG-29 island has varying levels of methylation, when comparing Wilms' tumour and patient-matched normal kidney (Dr. Robert Arceci, personal communication). Although 144 methylation is widely accepted to directly effect chromatin structure and transcriptional activity of many genes, including TSGs, one particular problem that exists in virtually all epigenetic studies is the difficulty in proving a cause and effect relationship between methylation and expression changes in complex organism/mammalian systems. In keeping with this, all of our studies, even with more in depth analysis, will likely remain correlative. Finally, it should be noted that differences in HACE1 m R N A expression were somewhat variable, ranging from 20% to over 90% less HACE1 m R N A in tumour samples compared with patient matched normal kidney. However, H A C E 1 protein levels in primary Wilms' tumours and patient matched normal kidney were much more consistent in their differences, with H A C E 1 protein being almost absent or undetectable in tumours with reduced expression. This can be seen when comparing results of quantitative PCR in Figure 4-3A to Western blot results in Figure 4-3B. This same variability in HACE1 m R N A expression is also apparent in normal fetal kidney control tissues when comparing Figures 4-3C and 4-3D. Taken together, this may suggest that post-transcriptional control of HACE1 is important to its expression and may influence levels of H A C E 1 protein observed Wilms' tumours. This post-transcriptional control may be related to a number of factors such as m R N A binding proteins. Relative levels of putative binding proteins may be altered in Wilms' tumours, these binding proteins may be mutated, or otherwise sequestered by other factors and mRNAs. Alternately, binding of these proteins may be affected by mutations to the UTRs in HACE1, which have yet to be evaluated for mutations in 145 W i l m s ' tumour , as ment ioned above. This area of s tudy m a y be of part icular interest i n exp lo r ing HACEI regula t ion especially i n W i l m s ' t umour cases w i t h l owered H A C E I expression, no muta t ions /de le t ions , and no apparent epigenetic changes. Post-translational events m a y also p lay a role i n stabili ty of the H A C E I protein. Post-translational events such as the avai labi l i ty of chaperone proteins for H A C E I fo ld ing or even the avai labi l i ty of H A C E I interactors c o u l d sequester a n d / o r shorten the half life of the protein. Other H E C T fami ly E3 u b i q u i t i n ligases have been s h o w n to control their o w n ac t iv i ty th rough auto-ubiqui t inat ion (163). H A C E I cou ld be us ing the same mechanism, aberrantly auto-ubiqui t inat ing itself, resul t ing i n degradat ion and lower prote in levels. C lea r ly further invest igat ion into a mechan i sm of downregu la t ion i n W i l m s ' t umour and other neoplasms is warranted . 6.2 H A C E I , C E L L C Y C L E C O N T R O L , A N D T H E UBIQUITIN P R O T E A S O M E P A T H W A Y H A C E I is not the first t umour suppressor gene w i t h a direct l i nk to the ubiqui t in-proteasome pa thway (168, 169). V H L , B R C A 1 , F B W 7 and E D D are a l l E3 ligases w i t h p roposed t umour suppressor functions. Ubiqui t in-proteasome pa thway related oncogenes have also been identif ied, most notably M D M 2 , w h i c h control the degradat ion of the p53 tumour suppressor. Further, the ubiqui t in-proteasome system is integral to cell cycle control . Al tera t ions i n key regulatory complexes, i n c l u d i n g the A P C / C and the S C F complex, have been observed i n a large number of malignancies (323). 146 The H A C E 1 protein possesses many of the qualities of a tumour suppressor gene. By increasing the level of H A C E 1 expression we were able to attenuate the tumour forming ability of SKNEP1 cells, both in vitro and in vivo. Further, this effect is not restricted solely to the SKNEP1 cell line as it is particularly evident in anchorage independent conditions in both neuroblastoma and osteosarcoma cell lines (Dr. Paul Clarkson; unpublished observations/personal communication). It should also be noted that although the SKNEP1 cell line is referred to throughout this dissertation and in current literature as a Wilms' tumour cell line, it has recently become apparent that it expresses the Ewing family tumour, oncogenic EWS-FLI1 fusion transcript (Dr. Peter Houghton/personal communication). Therefore SKNEP1 should more accurately be described as a pPNET of renal origin. Although this particular tumour is rare, a number of pPNETs of renal origin have previously been described in the literature. It has also been suggested that pPNETs of renal origin may be under-represented in favour of the more commonly recognized and sometimes histologically similar diagnosis of monophasic Wilms' tumour or Clear Cell Sarcoma of the kidney (324-326). In as far as our model of H A C E 1 tumour suppression, the SKNEP1 cell line still represents one of the few renally derived tumour cell line models. In fact, there are currently no commercially available Wilms' tumour cell lines. Interestingly, a rhabdoid tumour of the kidney cell line, G401 - previously thought to be Wilms' tumour, also expresses exceptionally low levels of H A C E 1 (data not shown). Further, 147 the growth suppressive activity of H A C E I does not appear to be restricted to the SKNEP1 cell line, but also has a significant effect in Neuroblastomas and Osteosarcomas (see Chapter V). At this time it is still unclear whether H A C E I effects may be restricted to particular cell types. However, it is certainly worthy of note that hacel mutant mice developed a high proportion of sarcomas (refer to Appendix B) and that both Osteosarcoma cell lines and SKNEP1 (considered as a member of the Ewing sarcoma family) showed significantly reduced growth in anchorage independent and nude mouse models when overexpressing H A C E I . Hacel mutant mice appear very sensitive to D N A alkylating agents, while clinically it has been noted that some Ewing sarcoma family tumours show benefit from intensification of treatment with alkylating agents (327). It will certainly be worthwhile to further investigate the relationship between sarcomas and H A C E I expression and even whether or not H A C E I may have direct effects on EWS-FLI1 or other EWS-ETS fusion oncoproteins. In HEK293 cells that normally express higher levels of H A C E I , reduction of H A C E I levels with various R N A i constructs consistently increased these cells' ability to grow in anchorage independent conditions. Finally, loss of H A C E I by gene inactivation in mice results in spontaneous, late onset cancers and renders mice sensitive to both environmental and genetic cancer triggers. Interestingly, hacel cooperates with p53, but not p21, in cancer formation and mutation of both hacel and p53 genes dramatically increased tumour incidence and significantly expanded the profile of cancer types from that normally observed in p53 mutant mice. 148 The activity of H A C E 1 is clearly linked to its ubiquitin ligase activity and it has been shown to interact with a number of components of the ubiquitin proteasome system, including the 20S proteasomal subunit, supporting that H A C E 1 has direct role in protein degradation. However, the roles of protein ubiquitination cannot exclude regulatory functions outside of the proteasomal degradation pathway. Molecular signals such as transcriptional activation, signal transduction and protein localization can be accomplished by differential ubiquitination and modification by ubiquitin-like molecules (328), overall likening this process to phosphorylation in post-translational control of protein activity (101, 252-260). Therefore we must recognize that H A C E 1 could have multiple target and these targets are not necessarily bound for proteasomal degradation. The mechanism by which H A C E 1 suppresses growth is not altogether clear. H A C E 1 targets are obviously linked to the process; however, this project has not yielded potential targets for H A C E 1 ubiquitination. We have identified a number of important proliferation-related pathways that are affected by H A C E 1 . In particular, phospho-AKT, phospho-GSK3(3, cyclin D l and to a lesser extent CDK4 levels are all reduced when H A C E 1 levels are high. Reduction of H A C E 1 expression also rescues this affect, restoring cyclin D l and phospho-AKT levels. The effects on cyclin D l and CDK4 both suggest H A C E 1 is linked to cell cycle progression. The changes in activation of A K T and GSK3P also suggest that H A C E 1 targets something upstream of A K T . Figure 6-1 outlines some potential signalling pathways affected by H A C E 1 . In addition, the recent identification of EWS-FLI1 expression in SKNEP1 cells, as well as 149 strong g r o w t h suppressive effect seen i n these cells w i t h H A C E 1 overexpression, m a y suggest overlap between pa thways that are activated and inhib i ted by E W S - F L I 1 . A s ment ioned above, potential direct interaction a n d / o r ub iqu i t ina t ion of E W S - F L I 1 shou ld also be investigated i n the future. PM Levels affected by HACE1 Increased translation of mRNA Figure 6-1: Potential Signalling Pathways Affected by HACE1 A number of signalling molecules may be affected either directly or indirectly by HACE1 activity. Some of the pathways involving A K T / p A K T , GSK3pypGSK3B, and Cyclin D l are illustrated above with potential HACE1 interactions highlighted. In add i t i on to the observed changes i n molecular effectors such as cyc l i n D , C D K 4 , A K T and G S K 3 P ment ioned above, cooperat ion w i t h the tumour suppressor p53 and lack of cooperat ion w i t h p21 m a y also give clues as to the act ivi ty of H A C E 1 (refer to A p p e n d i x B). First, as discussed i n Chapter I, p53 is able to d is rupt the cel l 150 cycle at either the G l / S or G 2 / M transitions (90,123,168). W h i l e the p 2 1 C I P 1 acts as an inhibi tor of cel l cycle progress ion i n association w i t h C D K 2 complexes by inh ib i t ing kinase act ivi ty and block progress ion th rough G l / S (89, 90, 329). W e k n o w that H A C E I affects levels of cyc l i n D l and C D K 4 (refer to Chapter V ) and that cyc l i n D / C D K 4 complexes are cri t ical i n progress ion past the R-point at the G l / S transi t ion (89, 90). Further, breeding of hacel mutant mice w i t h p21 mutant mice d i d not appear to affect the spect rum or overa l l t umour incidence, w h i l e breeding hacel mutant mice w i t h p53 mutant mice signif icantly accelerated tumour format ion (refer to A p p e n d i x B). This m a y suggest that hacel and p21 n u l l mice inact ivate/act ivate s imi lar pa thways l ead ing to oncogenic transformation. H o w e v e r , this hypothesis clearly needs more research i n order to val idate or revoke it. 6.3 G E N E R A L C O M M E N T S A N D F U T U R E D I R E C T I O N S A s the first w o r k on character izing H A C E I , the informat ion here w i l l serve as a b u i l d i n g b lock for future research w i t h i n the laboratory of D r . P o u l Sorensen and others. The data contained i n this dissertation identify HACEI as a candidate tumour suppressor gene that maps to a reg ion of chromosome 6q21 impl ica ted i n the development of mul t ip l e h u m a n tumour types. HACEI expression is d o w n regulated i n mul t ip l e h u m a n tumours . Loss of H A C E I in vitro and in vivo correlates w i t h increased tumourigenesis , w h i l e increased levels of H A C E I suppress g rowth , par t icular ly i n anchorage independent condi t ions. In con t inu ing to characterize the role of H A C E I i n n o r m a l cel lular homeostasis and cancer progress ion I believe two 151 main pathways should be undertaken: first, defining H A C E I targets, and second defining the scope and mechanisms of H A C E I downregulation in cancer. The lack of knowledge around H A C E I targets is arguably the most important factor holding back the characterization of H A C E I function. By finding targets of H A C E I we will be able to define its role in cell growth and proliferation. Interactions with Hsc70, V C P , and components of the ubiquitin proteasomal pathway suggest that part of the role of H A C E I is linked to protein quality control and/or stress responses such as E R A D and the UPR. This is also supported by predominant ER localization of the H A C E I protein. E3 ubiquitin ligases are thought to control the specificity of ubiquitination in the cascade of E1-E2 and E3 enzymes. Genome scans estimating > 500 different functional E3s suggest that most of these enzymes must target multiple proteins (163). Some E3 ligases may have more general functions, affecting entire protein populations while other may be much more specific. Following this, it seems likely that H A C E I has multiple targets, and these targets may control different aspects of cellular homeostasis. Again, our data showing a growth suppressive function for H A C E I suggest that some of its targets are integral to cell cycle progression, especially surrounding G l / S phase transition and stability of cyclin D l and CDK4. Co-immunoprecipitation experiments with H A C E I are going to be essential to the identification of more interacting proteins and targets. In particular, the use of the C876S H A C E I mutant protein, which is unable to bind and transfer ubiquitin, may allow for potential targets to be identified before they are rapidly degraded by the ubiquitin proteasome system. Current research in Dr. Sorensen's lab being done by Dr. 152 / Fan Zhang has already recognized this and is focussing on a number of systems to identify complexes and potential target that associate with H A C E I . Further work will also be required in defining both the scope of tumour in which HACEI functions are deregulated and the mechanism by which HACEI functions are deregulated. Although a few shortcomings have been discussed in the above sections it seems that epigenetic control of this locus is a primary mechanism of downregulation in Wilms' tumour. More recent data from the Human Genome Project and preliminary work by Dr. Maureen O'Sullivan (unpublished observations/personal communication) in Dr. Poul Sorensen's lab also indicates the presence of at least one other gene, LIN28B, in proximity to HACEI and in fact overlapping the two upstream C p G islands that are involved with the index case t(6;15) translocation (refer to http://genome.ucsc.edu/ and Chapter III for the position of the C p G islands). The possibility therefore exists that these C p G islands represent a locus control region (330-332) and that epigenetic changes may affect all of the genes in this region. Further, the role of maternal and paternal imprinting on this locus has not yet been examined and LOI could also play a role in expression from this region. Ongoing collaborations with Dr. Robert Arceci, mentioned above, involving more detailed examination of the all three C p G islands may begin to give more information on potential epigenetic regulation of the HACEI locus, as will further expression studies on other genes in the region. Unfortunately, as discussed in the previous section, these epigenetic studies may have some inherent shortcoming. This seems especially problematic considering the distance (~75-80kb) between the HACEI 153 gene and ups t ream methyla ted C p G is land. A l t h o u g h some distance effect have been described i n the literature (as noted above), it seems m u c h more w i d e l y accepted that C p G islands influence t ranscr ipt ion of m u c h more p r o x i m a l l y located genes. M o r e recent evalua t ion of the ESTs annotated i n the U C S C genome database suggest a sma l l number of transcript cor responding w i t h the d i rec t ion of t ranscr ipt ion of the H A C E 1 gene, but direct ly ove r l app ing the upst ream C p G - 1 77/ -29 islands (see h t tp : / /genome.ucsc .edu) . W o r k is n o w progressing to assess whether these may be distant alternate t ranscript ional start sites for the HACE1 gene, if this is the case it c o u l d strengthen the correlat ion between methyla t ion and lowered HACE1 expression. D u e to the relat ively l o w incidence of W i l m s ' tumours , larger studies o n epigenetics, i m p r i n t i n g a n d / o r other genetics factors i n W i l m s ' t umour that may w o r k i n t andem w i t h H A C E 1 loss of function, w i l l require partnerships w i t h groups such as the C O G and S I O P i n order to achieve statistically significant results. In add i t i on to epigenetic controls ment ioned above, the incidence of L O H , delet ion and rearrangements at 6q21 i n rumours , other than W i l m s ' tumour , suggests that a genetic mechan i sm of inact ivat ion also p lays a role i n HACE1 expression, however this has not yet been specifically investigated. M o r e impor tant ly , a l though l o w levels of HACE1 transcripts have been observed i n a large number of cancer cel l l ines baseline levels i n n o r m a l tissues and even c l in ica l samples, aside f rom W i l m s ' tumours , have not been w i d e l y examined. W h i l e the effects of H A C E 1 m a y be g r o w t h suppressive i n mul t ip l e tissues and our mouse m o d e l clearly shows an increased susceptibi l i ty to a b road spect rum of tumours , the relevance to h u m a n disease m a y be 154 more restricted. Es tabl ishing this relevance w i l l require a number of larger studies to look at both diseased and n o r m a l control tissues. F ina l ly , it is interesting to note that orthologues of H A C E I appear to be present on ly i n vertebrates, w h e n com par ing current ly sequenced genomes ((198) and http: / /genome.ucsc .edu or h t tp : / /wwrw.ensembl .org) . In add i t ion , the H A C E I prote in appears to be ve ry h igh ly conserved, w i t h the degree of amino ac id s imi la r i ty r ang ing f rom >80% between h u m a n and the putat ive F u g u H A C E I to >95% between h u m a n a n d mouse H A C E I proteins as measured u s ing 2-way B L A S T comparisons. This m a y suggests that it p lays a crucia l role. Since it appears that H A C E I plays a role i n the development of W i l m s ' tumour , it is reasonable to suggest that H A C E I m a y also p lay a role i n the n o r m a l development of vertebrate p ronephro i , mesonephroi a n d / o r metanephroi . The no rma l pattern of expression of H A C E I , w i t h par t icular ly s trong expression i n the b ra in and k idney , m a y also suggest that H A C E I plays a larger role i n the g r o w t h of m a n y structures that have complex development i n vertebrate species. Cur ren t ly , w o r k is be ing done by D r . Dieter F ink , i n Dr . Sorensen's laboratory, to generate tissue specific and induc ib le R N A i media ted k n o c k d o w n mouse models w i l l help define h o w and w h e n hacel functions. M o n i t o r i n g and evaluat ion of hacel expression a n d / or epigenetic changes of the hacel locus i n mul t ip l e tissues d u r i n g development m a y give clues as to where H A C E I invo lvement is cri t ical . In spite of the l imita t ions of some of the experiments l is ted above, the HACEI gene clearly has a role i n n o r m a l cel l function. The unique nature of this gene, its 155 domain architecture, its presence in only vertebrate genomes, and its high degree of conservation across various organisms all suggest that it has a vital role. In conclusion, the data contained herein and future studies on the HACE1 gene w i l l undoubtedly contribute to our overall understanding of cell homeostasis and cancer development. 156 R E F E R E N C E S 1. K i m , S. and C h u n g , D . H . (2006) Pediatr ic so l i d malignancies: neuroblastoma and W i l m s 1 tumor. Surg Clin North Am, 86,469-87, x i . 2. Metzger , M L . and D o m e , J.S. (2005) Cur ren t therapy for W i l m s ' tumor. Oncologist, 10, 815-26. 3. Coppes , M . J . and Pritchard-Jones, K . (2000) Pr inc ip les of W i l m s 1 t umor bio logy. Urol Clin North Am, 27,423-33, v i i i . 4. Pritchard-Jones, K . and Hast ie , N . D . (1990) W i l m s 1 t umour as a p a r a d i g m for the relat ionship of cancer to development . Cancer Surv, 9, 555-78. 5. H a r t m a n , D.J. and M a c L e n n a n , G.T . (2005) W i l m s tumor. / Urol, 173,2147. 6. Beckwi th , J.B. (1998) N a t i o n a l W i l m s T u m o r Study: an update for pathologists. Pediatr Dev Pathol, 1, 79-84. 7. ChaO, L . Y . , Huff , V . , Strong, L . C . and Saunders, G . F . (2000) M u t a t i o n i n the P A X 6 gene i n twenty patients w i t h ani r id ia . Hum Mutat, 15,332-9. 8. Chao , L . Y . , M i s h r a , R., Strong, L . C . and Saunders, G . F . (2003) Missense mutat ions i n the D N A - b i n d i n g reg ion and terminat ion c o d o n i n P A X 6 . Hum Mutat, 21,138-45. 9. O g a w a , O. , Eccles, M . R . , Szeto, J., M c N o e , L . A . , Y u n , K . , M a w , M . A . , Smi th , P.J. and Reeve, A . E . (1993) Relaxat ion of insu l in - l ike g r o w t h factor II gene i m p r i n t i n g impl ica ted i n W i l m s ' tumour . Nature, 362, 749-51. 10. Ravenel , J.D., Broman , K . W . , Per lman , E.J., N i e m i t z , E . L . , Jayawardena, T . M . , Be l l , D . W . , Haber , D . A . , Uej ima, H . and Feinberg, A . P . (2001) Loss of i m p r i n t i n g of insu l in - l ike g r o w t h factor-II (IGF2) gene i n d i s t ingu ish ing specific b iologic subtypes of W i l m s tumor. / Natl Cancer Inst, 93,1698-703. 11. Pol lak , M . N . (2004) Insul in- l ike g r o w t h factors and neoplasia. Novartis Found Sym-p, 262,84-98; d iscuss ion 98-107, 265-8. 12. Engs t rom, W . , Shokra i , A . , Otte, K . , Granerus , M . , Gessbo, A . , Bierke, P., Madej , A . , S jo lund, M . and W a r d , A . (1998) Transcr ip t ional regula t ion a n d bio logica l significance of the i n s u l i n l ike g r o w t h factor II gene. Cell Prolif, 31,173-89. 13. L e R o i t h , D . , M c G u i n n e s s , M . , Shemer, J., Stannard, B. , Lanau , F. , Faria , T . N . , Ka to , H , Werner , H , A d a m o , M . and Roberts, C .T . , Jr. (1992) Insul in- l ike g r o w t h factors. Biol Signals, 1,173-81. 14. H i l l , D . A . , Shear, T . D . , L i u , T., B i l lups , C . A . , S ingh , P . K . and D o m e , J.S. (2003) C l i n i c a l and biologic significance of nuclear unrest i n W i l m s tumor . Cancer, 97, 2318-26. 15. Laho t i , C , Thorner , P., M a l k i n , D . and Yeger , H . (1996) Immunohis tochemica l detection of p53 i n W i l m s ' tumors correlates w i t h unfavorable outcome. Am J Pathol, 148,1577-89. 157 16. M a l k i n , D . , Sexsmith, E . , Yeger , H . , W i l l i a m s , B.R. and Coppes , M . J . (1994) Muta t ions of the p53 tumor suppressor gene occur infrequently i n W i l m s 1 tumor. Cancer Res, 54,2077-9. 17. Ruteshouser, E . C . and Huff , V . (2004) F a m i l i a l W i l m s tumor. Am ] Med Genet C Semin Med Genet, 129,29-34. 18. D o m e , J.S. and Coppes , M . J . (2002) Recent advances i n W i l m s tumor genetics. Curr Opin Pediatr, 14,5-11. 19. Pritchard-Jones, K . and H a w k i n s , M . M . (1997) B io logy of W i l m s 1 tumour . Lancet, 349, 663-4. 20. Bonai t i -Pel l ie , C , Chompre t , A . , Tournade , M . F . , Hochez , J., M o u t o u , C , Zucker , J . M . , Steschenko, D . , Bruna t -Ment igny , M . , Roche, H . , T ron , P. et al. (1992) Genetics a n d ep idemio logy of W i l m s ' rumor: the French W i l m s ' tumor study. Med Pediatr Oncol, 20,284-91. 21. Pastore, G . , C a r l i , M . , Lemerle , J., Tournade, M . F . , Vou te , P. A . , Rey, A . , Burgers, J . M . , Zucker , J . M . , Burger, D . , de Kraker , J. et al. (1988) E p i d e m i o l o g i c a l features of W i l m s ' tumor: results of studies by the International Society of Paediatric O n c o l o g y (SIOP). Med Pediatr Oncol, 16, 7-11. 22. Bres low, N . E . and Beckwi th , J.B. (1982) Ep idemio log i ca l features of W i l m s ' tumor: results of the Na t iona l W i l m s ' T u m o r Study. / Natl Cancer Inst, 68,429-36. 23. B r o w n , W . T . , Puran ik , S.R., A i r m a n , D . H . and H a r d i n , H . C . , Jr. (1972) W i l m s ' tumor i n three successive generations. Surgery, 72, 756-61. 24. K n u d s o n , A . J . and Strong, L . C (1972) M u t a t i o n a n d cancer: a m o d e l for W i l m s ' tumor of the k idney . / Natl Cancer Inst, 48,313-24. 25. Matsunaga , E . (1981) Genetics of W i l m s ' tumor . Hum Genet, 57, 231-46. 26. Rahman , N . , A r b o u r , L . , T o n i n , P., Renshaw, J., Pelletier, J., Baruchel , S., Pritchard-Jones, K . , Stratton, M . R . and N a r o d , S .A . (1996) Evidence for a fami l ia l W i l m s ' t umour gene (FWT1) o n chromosome 17ql2-q21. Nat Genet, 13,461-3. 27. M c D o n a l d , J . M . , Douglass , E . C , Fisher, R., Geiser, C .F . , K r i l l , C . E . , Strong, L . C , V i r s h u p , D . and Huff , V . (1998) L inkage of fami l ia l W i l m s ' tumor predispos i t ion to chromosome 19 and a two-locus m o d e l for the et iology of fami l ia l tumors. Cancer Res, 58,1387-90. 28. Rapley , E . A . , Barfoot, R., Bonai t i -Pel l ie , C , Chompre t , A . , Foulkes , W . , Perusinghe, N . , Reeve, A . , Royer -Pokora , B. , Schumacher , V . , She l l ing , A . et al. (2000) Evidence for susceptibi l i ty genes to fami l ia l W i l m s tumour i n add i t i on to W T l , F W T 1 and F W T 2 . Br J Cancer, 83,177-83. 29. Rahman , N . , A r b o u r , L . , T o n i n , P., Baruchel , S., Pritchard-Jones, K . , N a r o d , S .A. and Stratton, M . R . (1997) The fami l ia l W i l m s ' t umour susceptibi l i ty gene, F W T 1 , may not be a t umour suppressor gene. Oncogene, 14,3099-102. 30. Rahman , N . , A b i d i , F. , F o r d , D . , A r b o u r , L . , Rapley, E . , T o n i n , P., Barton, D . , Batcup, G . , Berry, J., Cotter, F. et al. (1998) Conf i rma t ion of F W T 1 as a W i l m s ' t umour susceptibi l i ty gene and phenotypic characteristics of W i l m s ' t umour attributable to F W T 1 . Hum Genet, 103, 547-56. 158 31. Coppes, M.J. and Egeler, R .M. (1999) Genetics of Wilms' tumor. Semin Urol Oncol, 17,2-10. 32. Miller, R.W., Fraumeni, J.F., Jr. and Manning, M.D. (1964) Association of Wilms's Tumor with Aniridia, Hemihypertrophy and Other Congenital Malformations. N Engl J Med, 270, 922-7. 33. Enklaar, T., Zabel, B.U. and Prawitt, D. (2006) Beckwith-Wiedemann syndrome: multiple molecular mechanisms. Expert Rev Mol Med, 8,1-19. 34. Wiedemann, H.R., Tolksdorf, M . , Hansen, H . G . and Klose, K. (1964) [Chromosomal Studies in "Partial Gigantism".]. Monatsschr Kinderheilkd, 112, 281-2. 35. Sotelo-Avila, C , Gonzalez-Crussi, F. and Starling, K. A. (1980) Wilms' rumor in a patient with an incomplete form of Beckwith-Wiedemann syndrome. Pediatrics, 66,121-3. 36. Li , M . , Squire, J.A. and Weksberg, R. (1998) Overgrowth syndromes and genomic imprinting: from mouse to man. Clin Genet, 53,165-70. 37. Li , M . , Squire, J.A. and Weksberg, R. (1997) Molecular genetics of Beckwith-Wiedemann syndrome. Curr Opin Pediatr, 9, 623-9. 38. Beckwith, J.B. (1998) Nephrogenic rests and the pathogenesis of Wilms tumor: developmental and clinical considerations. Am J Med Genet, 79, 268-73. 39. Kucinskas, L. , Rudaitis, S., Pundziene, B. and Just, W. (2005) Denys-Drash syndrome. Medicina (Kaunas), 41,132-4. 40. Coppes, M.J., Huff, V. and Pelletier, J. (1993) Denys-Drash syndrome: relating a clinical disorder to genetic alterations in the tumor suppressor gene W T l . / Pediatr, 123, 673-8. 41. Haber, D.A., Buckler, A.J., Glaser, T., Call, K . M . , Pelletier, J., Sohn, R.L., Douglass, E.C. and Housman, D.E. (1990) A n internal deletion within an H p l 3 zinc finger gene contributes to the development of Wilms' rumor. Cell, 61,1257-69. 42. Diller, L. , Ghahremani, M . , Morgan, J., Grundy, P., Reeves, C , Breslow, N. , Green, D., Neuberg, D., Pelletier, J. and Li , F.P. (1998) Constitutional WT1 mutations in Wilms' tumor patients. / Clin Oncol, 16, 3634-40. 43. Haber, D.A., Sohn, R.L., Buckler, A.J., Pelletier, J., Call, K . M . and Housman, D.E. (1991) Alternative splicing and genomic structure of the Wilms tumor gene W T l . Proc Natl Acad Sci USA, 88,9618-22. 44. Little, M . , Holmes, G. and Walsh, P. (1999) W T l : what has the last decade told us? Bioessays, 21,191-202. 45. Hewitt, S.M. and Saunders, G.F. (1996) Differentially spliced exon 5 of the Wilms 1 tumor gene W T l modifies gene function. Anticancer Res, 16, 621-6. 46. Pritchard-Jones, K. (1995) Mutations in the Wilms' tumour gene, W T l . What do they mean? Eur ] Cancer, 31A, 2138-40. 47. Wagner, K.D. , Wagner, N . and Schedl, A . (2003) The complex life of W T l . / Cell Sci, 116,1653-8. 159 48. Davies , R., M o o r e , A . , Schedl , A . , Bratt, E . , M i y a h a w a , K . , Ladomery , M . , M i l e s , C , M e n k e , A . , v a n H e y n i n g e n , V . and Hast ie , N . (1999) M u l t i p l e roles for the W i l m s 1 t umor suppressor, W T l . Cancer Res, 59,1747s-1750s; d iscuss ion 1751s. 49. E l l i sen , L . W . (2002) Regula t ion of gene expression by W T l i n development and tumorigenesis . Int ] Hematol, 76,110-6. 50. D r u m m o n d , I. A . , Rupprecht , H . D . , Rohwer-Nut te r , P., L o p e z - G u i s a , J . M . , M a d d e n , S.L., Rauscher, F.J., 3 rd and Sukhatme, V . P . (1994) D N A recogni t ion by sp l i c ing variants of the W i l m s 1 tumor suppressor, W T l . Mol Cell Biol, 14, 3800-9. 51. Bor , Y . C . , Swar tz , J., M o r r i s o n , A . , Rekosh , D . , L a d o m e r y , M . and Hammarsk jo ld , M . L . (2006) The W i l m s 1 t umor 1 ( W T l ) gene (+KTS isoform) functions w i t h a C T E to enhance translat ion f rom an unsp l iced R N A w i t h a retained intron. Genes Dev, 20,1597-608. 52. S imms , L . A . , A l g a r , E . M . and Smi th , P.J. (1995) Sp l i c ing of exon 5 i n the W T l gene is d i s rupted i n W i l m s 1 rumour . Eur ] Cancer, 31A, 2270-6. 53. Brenner, B. , W i l d h a r d t , G . , Schneider, S. and Royer -Pokora , B. (1992) R N A polymerase chain reaction detects different levels of four al ternatively sp l iced W T l transcripts i n W i l m s ' tumors . Oncogene, 7,1431-3. 54. M a i t i , S., A l a m , R., A m o s , C.I . and Huff , V . (2000) Frequent association of beta-catenin and W T l mutat ions i n W i l m s tumors. Cancer Res, 60, 6288-92. 55. Kusafuka , T., M i a o , J., K u r o d a , S., Uda t su , Y . and Yoneda , A . (2002) C o d o n 45 of the beta-catenin gene, a specific muta t ional target site of W i l m s ' tumor. Int J Mol Med, 10, 395-9. 56. Smal ley, M . J . and Dale , T . C (1999) W n t s igna l l ing i n m a m m a l i a n development and cancer. Cancer Metastasis Rev, 18,215-30. 57. Besnard-Guer in , C , N e w s h a m , I., W i n q v i s t , R. and Cavenee, W . K . (1996) A c o m m o n reg ion of loss of heterozygosi ty i n W i l m s ' t umor a n d embryona l rhabdomyosarcoma dis ta l to the D11S988 locus o n chromosome l l p l 5 . 5 . Hum Genet, 97,163-70. 58. D o w d y , S.F., Fasching, C . L . , Araujo , D . , L a i , K . M . , L ivanos , E . , Weissman , B .E . and Stanbridge, E.J. (1991) Suppress ion of tumor igenic i ty i n W i l m s tumor by the p l 5 . 5 - p l 4 reg ion of chromosome 11. Science, 254,293-5. 59. M a w , M . A . , G r u n d y , P .E . , M i l l o w , L .J . , Eccles, M . R . , D u n n , R.S., Smi th , P.J., Feinberg, A . P . , L a w , D.J. , Paterson, M . C . , Te lze row, P . E . et al. (1992) A th i rd W i l m s ' tumor locus o n chromosome 16q. Cancer Res, 52,3094-8. 60. G r u n d y , P .E . , Te lzerow, P .E . , Bres low, N . , Moksness , J., Huff , V . and Paterson, M . C . (1994) Loss of heterozygosi ty for chromosomes 16q and l p i n W i l m s ' tumors predicts an adverse outcome. Cancer Res, 54,2331-3. 61. Natrajan, R., W i l l i a m s , R . D . , H i n g , S . N . , M a c k a y , A . , Reis -Fi lho , J.S., Fenwick , K . , I ravani , M . , Valge i rsson , H . , Gr igor i ad i s , A . , Langfo rd , C F . et al. (2006) A r r a y C G H prof i l ing of favourable his tology W i l m s tumours reveals nove l gains and losses associated w i t h relapse. / Pathol. 160 62. D o m e , J.S., C h u n g , S., Bergemann, T., Umbr ich t , C . B . , Saji, M . , Carey , L . A . , G r u n d y , P .E . , Pe r lman , E.J., Bres low, N . E . and Sukumar , S. (1999) H i g h telomerase reverse transcriptase (hTERT) messenger R N A level correlates w i t h tumor recurrence i n patients w i t h favorable h is to logy W i l m s ' tumor . Cancer Res, 59,4301-7. 63. Eggert, A . , Grotzer , M . A . , Ikegaki , N . , Zhao , H , Cnaan , A . , Brodeur , G . M . and Evans , A . E . (2001) Express ion of the neuro t rophin receptor T r k B is associated w i t h unfavorable outcome i n W i l m s ' tumor . / Clin Oncol, 19,689-96. 64. K a y t o n , M . L . , R o w e , D . H . , O 'Toole , K . M . , Thompson , R .B. , Schwarz , M . A . , Stolar, C.J . and K a n d e l , J.J. (1999) Metastasis correlates w i t h p roduc t ion of vascular endothel ia l g r o w t h factor i n a mur ine m o d e l of h u m a n W i l m s ' tumor . / Pediatr Surg, 34, 743-7; d iscuss ion 747-8. 65. R o w e , D . H . , H u a n g , J., K a y t o n , M . L . , T h o m p s o n , R., Troxe l , A . , O 'Toole , K . M . , Yamash i ro , D . , Stolar, C.J . and K a n d e l , J.J. (2000) A n t i - V E G F ant ibody suppresses p r i m a r y tumor g r o w t h and metastasis i n an exper imental m o d e l of W i l m s ' tumor. / Pediatr Surg, 35,30-2; d iscuss ion 32-3. 66. T a k a m i z a w a , S., Scott, D . , W e n , J., G r u n d y , P., Bishop , W . , K i m u r a , K . and Sandler, A . (2001) The survivinrfas ratio i n pediatr ic renal tumors. / Pediatr Surg, 36,37-42. 67. Arce l lana-Panl i l io , M . Y . , Egeler, R . M . , Ujack, E . , P in to , A . , Demetr ick, D.J. , Robbins , S . M . and Coppes , M . J . (2000) Decreased expression of the I N K 4 fami ly of cycl in-dependent kinase inhibi tors i n W i l m s tumor . Genes Chromosomes Cancer, 29, 63-9. 68. A r t a n d i , S.E. and A t t a r d i , L . D . (2005) Pa thways connect ing telomeres and p53 i n senescence, apoptosis, and cancer. Biochem Biophys Res Commun, 331, 881-90. 69. Soussi , T. and Lozano , G . (2005) p53 muta t ion heterogeneity i n cancer. Biochem Biophys Res Commun, 331,834-42. 70. Peres, E . M . , Savasan, S., C u s h i n g , B. , A b e l l a , S. and M o h a m e d , A . N . (2004) C h r o m o s o m e analyses of 16 cases of W i l m s tumor: different pattern i n unfavorable his tology. Cancer Genet Cytogenet, 148,66-70. 71. Beniers, A . J . , Efferth, T., Fuzes i , L . , Granzen , B. , Mertens , R. a n d Jakse, G . (2001) p53 expression i n W i l m s 1 tumor: a possible role as prognost ic factor. Int} Oncol, 18,133-9. 72. Bardeesy, N . , Beckwi th , J.B. and Pelletier, J. (1995) C l o n a l expansion and attenuated apoptosis i n W i l m s ' tumors are associated w i t h p53 gene mutations. Cancer Res, 55, 215-9. 73. H u a n g , J., Softer, S.Z., K i m , E.S., Y o k o i , A . , M o o r e , J.T., M c C r u d d e n , K . W . , M a n l e y , C , M i d d l e s w o r t h , W . , O 'Toole , K . , Stolar, C . et al. (2002) p53 accumula t ion i n favorable-histology W i l m s tumor is associated w i t h angiogenesis and c l in ica l ly aggressive disease. / Pediatr Surg, 37,523-7. 74. Feinberg, A . P . (2004) The epigenetics of cancer et iology. Semin Cancer Biol, 14, 427-32. 161 75. Ohlsson , R., Tycko , B. and Sapienza, C . (1998) Monoa l l e l i c expression: 'there can on ly be one'. Trends Genet, 14,435-8. 76. Reik , W . , Constancia , M . , Dean, W . , Davies , K . , B o w d e n , L . , M u r r e l l , A . , Fe i l , R., Wal ter , J. and Kelsey , G . (2000) Igf2 i m p r i n t i n g i n development and disease. Int / Dei? Bio/, 44,145-50. 77. Z h a n g , Y . , Shields, T., Crenshaw, T., H a o , Y-, M o u l t o n , T. and Tycko , B. (1993) Impr in t ing of h u m a n H19: allele-specific C p G methyla t ion , loss of the active allele i n W i l m s tumor , and potential for somatic allele swi tch ing . Am J Hum Genet, 53,113-24. 78. Dao, D . , W a l s h , C P . , Y u a n , L . , Gore lov , D . , Feng, L . , Hensle , T., N i s e n , P., Yamash i ro , D.J. , Bestor, T . H . and Tycko , B. (1999) M u l t i p o i n t analysis of h u m a n chromosome H p l 5 / m o u s e distal chromosome 7: inc lus ion of H 1 9 / I G F 2 i n the m i n i m a l W T 2 region, gene specificity of H 1 9 s i lencing i n W i l m s ' tumorigenesis and methyla t ion hyper-dependence of H 1 9 i m p r i n t i n g . Hum Mol Genet, 8,1337-52. 79. Reik , W . and M u r r e l l , A . (2000) Genomic impr in t i ng . Silence across the border. Nature, 405,408-9. 80. H a o , Y . , Crenshaw, T., M o u l t o n , T., N e w c o m b , E . and Tycko , B. (1993) T u m o u r -suppressor act ivi ty of H 1 9 R N A . Nature, 365, 764-7. 81. Y u a n , E . , L i , C M . , Yamash i ro , D.J. , K a n d e l , J., Thaker , H , M u r t y , V . V . and Tycko , B. (2005) Genomic p ro f i l i ng maps loss of heterozygosi ty and defines the t i m i n g and stage dependence of epigenetic and genetic events i n W i l m s ' tumors. Mol Cancer Res, 3,493-502. 82. Okamoto , K . , M o r i s o n , I . M . , Tan iguchi , T. and Reeve, A . E . (1997) Epigenetic changes at the insu l in- l ike g r o w t h factor I I / H I 9 locus i n deve lop ing k i d n e y is an early event i n W i l m s tumorigenesis. Proc Natl Acad Sci USA, 94,5367-71. 83. Y u a n , E . , H a g h i g h i , F. , Whi te , S., Costa , R., M c M i n n , J., C h u n , K . , M i n d e n , M . and Tycko , B. (2006) A single nucleot ide p o l y m o r p h i s m chip-based me thod for combined genetic and epigenetic prof i l ing : va l ida t ion i n decitabine therapy and t u m o r / n o r m a l comparisons. Cancer Res, 66,3443-51. 84. Hanahan , D . and Weinberg , R . A . (2000) The ha l lmarks of cancer. Cell, 100,57-70. . 85. M i l l e r , D . M . , B lume , S., Borst, M . , Gee, J., Polansky, D . , Ray , R., R o d u , B. , Shrestha, K . , Snyder , R., Thomas, S. et al. (1990) Oncogenes, mal ignant transformation, and m o d e r n medicine. Am J Med Sci, 300,59-69. 86. Bishop , J . M . (1991) M o l e c u l a r themes i n oncogenesis. Cell, 64,235-48. 87. Zetterberg, A . , Larsson, O . and W i m a n , K . G . (1995) W h a t is the restriction point? Curr Opin Cell Biol, 7,835-42. 88. Skehan, P. (1988) C o n t r o l models of cel l cycle transit, exit, and arrest. Biochem Cell Biol, 66,467-77. 89. Sherr, C.J . (2000) The Pezcol ler lecture: cancer cel l cycles revisi ted. Cancer Res, 60,3689-95. 90. Sherr, C.J . (2000) C e l l cycle control and cancer. Harvey Lect, 96, 73-92. 162 91. D i e h l , J .A . (2002) C y c l i n g to cancer w i t h c y c l i n D l . Cancer Biol Titer, 1,226-31. 92. Fauss i l lon , M . , M o n n i e r , L . , Junien, C . and Jeanpierre, C . (2005) Frequent overexpression of c y c l i n D2 /cyc l in -dependen t kinase 4 i n W i l m s ' tumor . Cancer Lett, 221,67'-75. 93. Buckley , M . F . , Sweeney, K . J . , H a m i l t o n , J .A. , S in i , R . L . , M a n n i n g , D . L . , N i c h o l s o n , R.I. , deFazio , A . , Watts , C . K . , M u s g r o v e , E . A . and Suther land, R . L . (1993) Express ion and ampl i f ica t ion of c y c l i n genes i n h u m a n breast cancer. Oncogene, 8,2127-33. 94. D i c k s o n , C , Fant l , V . , Gi l le t t , C , Brookes, S., Bartek, J., Smi th , R., Fisher, C , Barnes, D . and Peters, G . (1995) A m p l i f i c a t i o n of chromosome band H q l 3 and a role for cyc l i n D l i n h u m a n breast cancer. Cancer Lett, 90,43-50. 95. W a n g , T .C . , Cardiff , R . D . , Zukerberg , L . , Lees, E . , A r n o l d , A . and Schmidt , E . V . (1994) M a m m a r y hyperp las ia and carc inoma i n M M T V - c y c l i n D l transgenic mice. Nature, 369, 669-71. 96. Land i s , M . W . , P a w l y k , B.S., L i , T., S ic insk i , P. and H i n d s , P . W . (2006) C y c l i n D l -dependent kinase act ivi ty i n mur ine development and m a m m a r y tumorigenesis . Cancer Cell, 9,13-22. 97. O lashaw, N . , Bagu i , T . K . and Pledger, W.J . (2004) C e l l cycle control : a complex issue. Cell Cycle, 3,263-4. 98. D i e h l , J. A . , Z i n d y , F. and Sherr, C.J . (1997) Inhib i t ion of c y c l i n D l phosphory la t ion o n threonine-286 prevents its r a p i d degradat ion v i a the ubiqui t in-proteasome pathway. Genes Dev, 11, 957-72. 99. G e r m a i n , D . , Russel l , A . , T h o m p s o n , A . and Hend l ey , J. (2000) Ub iqu i t i na t i on of free c y c l i n D l is independent of phosphory la t ion o n threonine 286. / Biol Chem, 275,12074-9. 100. D i e h l , J .A. , Cheng , M . , Roussel , M . F . and Sherr, C.J . (1998) G l y c o g e n synthase kinase-3beta regulates c y c l i n D l proteolysis and subcel lular local iza t ion. Genes Dev, 12,3499-511. 101. Pickart , C M . (2001) Mechan i sms u n d e r l y i n g ubiqui t ina t ion . Annu Rev Biochem, 70,503-33. 102. Hengst , L . and Reed, S.I. (1998) Inhibitors of the C i p / K i p family . Curr Top Microbiol Immunol, 227,25-41. 103. Harper , J .W. (1997) C y c l i n dependent kinase inhibi tors . Cancer Sum, 29,91-107. 104. Hash imoto , Y . , K o h r i , K . , Kaneko , Y . , M o r i s a k i , H . , Ka to , T., Ikeda, K . and Nakan i sh i , M . (1998) Cr i t i c a l role for the 310 hel ix reg ion of p57(Kip2) i n cyc l in -dependent kinase 2 inh ib i t i on and g r o w t h suppression. / Biol Chem, 273,16544-50. 105. T o d d , R. and W o n g , D.T . (1999) Oncogenes. Anticancer Res, 19,4729-46. 106. K l e i n , G . (2002) Perspectives i n studies of h u m a n tumor viruses. Front Biosci, 7, d268-74. 107. M a k o w s k i , D.R. , Rothberg, P . G . and A s t r i n , S . M . (1984) The role of promoter inser t ion i n the i nduc t ion of neoplasia. Surv Synth Pathol Res, 3,342-9. 163 108. H a y w a r d , W.S. , N e e l , B . G . and A s t r i n , S . M . (1981) A c t i v a t i o n of a cel lular one gene by promoter inser t ion i n A L V - i n d u c e d l y m p h o i d leukosis . Nature, 290, 475-80. 109. Schwab, M . (2004) M Y C N i n neuronal tumours . Cancer Lett, 204,179-87. 110. Schwab, M . (1999) Oncogene ampl i f ica t ion i n so l id tumors . Semin Cancer Biol, 9, 319-25. 111. Lofts, F.J. and G u l l i c k , W.J . (1992) c-erbB2 ampl i f ica t ion and overexpression i n h u m a n tumors. Cancer Treat Res, 61,161-79. 112. K o p n i n , B.P. (2000) Targets of oncogenes and tumor suppressors: key for unders tanding basic mechanisms of carcinogenesis. Biochemistry (Mosc), 65, 2-27. 113. Go ldbe rg , Y . P . , Parker, M . I . and Gevers , W . (1991) The genetic basis of cancer. S Afr Med 7,80,99-104. 114. Borner, C . , Schlagbauer W a d l , H . , Fel lay, I., Selzer, E . , Polterauer, P . and Jansen, B. (1999) M u t a t e d N-ras upregulates Bcl-2 i n h u m a n me lanoma i n v i t ro and i n S C I D mice. Melanoma Res, 9,347-50. 115. Per l , A . , W a n g , N . , W i l l i a m s , J . M . , H u n t , M . J . , Rosenfeld, S.I., C o n d e m i , J.J., Packman , C . H . and A b r a h a m , G . N . (1987) Aber ran t i m m u n o g l o b u l i n and c-myc gene rearrangements i n patients w i t h nonmal ignant monoc lona l c ryoglobul inemia . / Immunol, 139,3512-20. 116. K n e z e v i c h , S.R., M c F a d d e n , D . E . , Tao, W . , L i m , J.F. a n d Sorensen, P . H . (1998) A nove l E T V 6 - N T R K 3 gene fus ion i n congenital f ibrosarcoma. Nat Genet, 18,184-7. 117. Sorensen, P . H . and Triche, T.J. (1996) Gene fusions encoding chimaeric t ranscr ipt ion factors i n so l i d tumours . Semin Cancer Biol, 7,3-14. 118. K n u d s o n , A . G . (1993) Ant ioncogenes and h u m a n cancer. Proc Natl Acad Sci U S A, 90,10914-21. 119. M a l k i n , D . , L i , F.P. , Strong, L . C , F raumeni , J.F., Jr., Ne l son , C . E . , K i m , D . H . , Kassel , J., G r y k a , M . A . , Bischoff, F .Z . , Ta insky , M . A . et al. (1990) G e r m l ine p53 mutat ions i n a fami l ia l syndrome of breast cancer, sarcomas, and other neoplasms. Science, 250,1233-8. 120. Di l l e r , L . , Kassel , J., Ne l son , C . E . , G r y k a , M . A . , L i t w a k , G . , Gebhardt , M . , Bressac, B. , O z t u r k , M . , Baker, S.J., Vogels te in , B. et al. (1990) p53 functions as a cel l cycle control prote in i n osteosarcomas. Mol Cell Biol, 10,5772-81. 121. M i c h a l o v i t z , D . , H a l e v y , O . and Oren , M . (1990) C o n d i t i o n a l i nh ib i t i on of t ransformation and of cel l prol i ferat ion by a temperature-sensitive mutant of p53. Cell, 62,671-80. 122. Paules, R.S., Levedakou , E . N . , W i l s o n , S.J., Innes, C . L . , Rhodes, N . , Tlsty, T .D . , G a l l o w a y , D . A . , Donehower , L . A . , Ta insky , M . A . and Kaufmann , W . K . (1995) Defective G 2 checkpoint funct ion i n cells f rom i n d i v i d u a l s w i t h fami l ia l cancer syndromes. Cancer Res, 55,1763-73. 123. Stewart, N . , H i c k s , G . G . , Paraskevas, F. and M o w a t , M . (1995) Evidence for a second cel l cycle b lock at G 2 / M by p53. Oncogene, 10,109-15. 164 124. Ho l l s t e in , M . , Hergenhahn , M . , Y a n g , Q. , Bartsch, H . , W a n g , Z . Q . and Hainaut , P. (1999) N e w approaches to unders tanding p53 gene tumor muta t ion spectra. Mutat Res, 431,199-209. 125. Ho l l s t e in , M . , S idransky, D . , Vogels te in , B. and Har r i s , C C . (1991) p53 mutat ions i n h u m a n cancers. Science, 253,49-53. 126. Vogels te in , B. and K i n z l e r , K . W . (1992) p53 funct ion and dysfunct ion. Cell, 70, 523-6. 127. Pekarsky, Y . , Palamarchuk, A . , Huebner , K . and Croce , C M . (2002) F H I T as tumor suppressor: mechanisms and therapeutic opportunit ies . Cancer Biol Ther, 1,232-6. 128. G l o v e r , T . W . and Stein, C . K . (1988) C h r o m o s o m e breakage and recombinat ion at fragile sites. Am J Hum Genet, 43,265-73. 129. Hadaczek , P., S ip rashv i l i , Z . , M a r k i e w s k i , M . , Domaga la , W . , Druck , T., M c C u e , P . A . , Pekarsky, Y . , Ohta , M . , Huebner , K . and L u b i n s k i , J. (1998) Absence or reduc t ion of Fhi t expression i n most clear cel l renal carcinomas. Cancer Res, 58, 2946-51. 130. C l i f fo rd , S C . , Prowse, A . H . , Affara , N . A . , Buys , C . H . and Maher , E .R. (1998) Inact ivat ion of the v o n H i p p e l - L i n d a u ( V H L ) t umour suppressor gene and allelic losses at chromosome a r m 3p i n p r i m a r y renal cell carcinoma: evidence for a V H L - i n d e p e n d e n t pa thway i n clear cel l renal tumourigenesis . Genes Chromosomes Cancer, 22,200-9. 131. M o r e l l i , C , Ka ray i ann i , E . , M a g n a n i n i , C , M u n g a l l , A J . , Thor l and , E . , N e g r i n i , M . , Smi th , D.I . and Barbant i-Brodano, G . (2002) C l o n i n g and characterization of the c o m m o n fragile site F R A 6 F harbor ing a replicat ive senescence gene and frequently deleted i n h u m a n tumors. Oncogene, 21, 7266-76. 132. Bedel l , M . A . , Jenkins, N . A . and C o p e l a n d , N . G . (1996) G o o d genes i n bad neighbourhoods . Nat Genet, 12,229-32. 133. Velagale t i , G . V . , B ien-Wi l lne r , G . A . , N o r t h u p , J .K., Lockhar t , L . H . , H a w k i n s , J . C , Jalal, S . M . , Wi thers , M . , L u p s k i , J.R. and Stankiewicz , P. (2005) Pos i t ion effects due to chromosome breakpoints that map approximate ly 900 K b ups t ream and approximate ly 1.3 M b downs t ream of S O X 9 i n two patients w i t h campomel ic dysplas ia . Am J Hum Genet, 76,652-62. 134. Pfeifer, D . , Kis t , R., Dewar , K . , D e v o n , K . , Lander , E.S., Bi r ren , B. , K o r n i s z e w s k i , L . , Back, E . and Scherer, G . (1999) C a m p o m e l i c dysplas ia translocation breakpoints are scattered over 1 M b p r o x i m a l to S O X 9 : evidence for an extended control region. Am JHum Genet, 65,111-24. 135. Fernandez, B . A . , Siegel-Bartelt, J., Herbr ick , J .A. , Teshima, I. and Scherer, S.W. (2005) Holoprosencepha ly and c le idocranial dysplas ia i n a patient due to two position-effect mutations: case report and r ev i ew of the literature. Clin Genet, 68, 349-59. 136. Be l lon i , E . , M u e n k e , M . , Roessler, E . , Traverso, G . , Siegel-Bartelt, J., F r u m k i n , A . , M i t c h e l l , H . F . , Donis -Ke l le r , H . , H e l m s , C , H i n g , A . V . et al. (1996) Identification 165 of Sonic hedgehog as a candidate gene responsible for holoprosencephaly. Nat Genet, 14,353-6. 137. Roessler, E . , W a r d , D . E . , Gaudenz , K . , Be l lon i , E . , Scherer, S.W., D o n n a i , D . , Siegel-Bartelt, J., Tsu i , L . C . and M u e n k e , M . (1997) Cytogenet ic rearrangements i n v o l v i n g the loss of the Sonic H e d g e h o g gene at 7q36 cause holoprosencephaly. Hum Genet, 100,172-81. 138. Fantes, J., Redeker, B. , Breen, M . , Boyle , S., B r o w n , J., Fletcher, J., Jones, S., Bickmore , W . , F u k u s h i m a , Y . , Mannens , M . et al. (1995) Anir id ia-associa ted cytogenetic rearrangements suggest that a pos i t ion effect m a y cause the mutant phenotype. Hum Mol Genet, 4,415-22. 139. Q u o n , K . C . and Berns, A . (2001) Haplo- insuff ic iency? Let me count the ways . Genes Dev, 15,2917-21. 140. L y n c h , C.J. and M i l n e r , J. (2006) Loss of one p53 allele results i n four-fold reduct ion of p53 m R N A and protein: a basis for p53 haplo-insufficiency. Oncogene, 25,3463-70. 141. Vaslet , C . A . , Messier , N . J . and Kane , A . B . (2002) Accelerated progression of asbestos-induced mesothel iomas i n heterozygous p53+/ - mice. Toxicol Sci, 68, 331-8. 142. Staff, S., Isola, J. a n d Tanner, M . (2003) Haplo- insuff ic iency of B R C A 1 i n sporadic breast cancer. Cancer Res, 63,4978-83. 143. M a o , J .H . , Perez-Losada, J., W u , D . , Delrosar io , R., Tsunematsu, R., N a k a y a m a , K . I . , B r o w n , K . , Bryson , S. and Ba lma in , A . (2004) F b x w 7 / C d c 4 is a p53-dependent, haploinsufficient t umour suppressor gene. Nature, 432, 775-9. 144. Feinberg, A . P . and Tycko , B. (2004) The his tory of cancer epigenetics. Nat Rev Cancer, 4,143-53. 145. B a y l i n , S.B. and O h m , J.E. (2006) Epigenetic gene s i lencing i n cancer - a mechan i sm for early oncogenic pa thway addict ion? Nat Rev Cancer, 6,107-16. 146. Esteller, M . (2005) Aber ran t D N A methyla t ion as a cancer- inducing mechanism. Annu Rev Pharmacol Toxicol, 45,629-56. 147. Jones, P . A . and B a y l i n , S.B. (2002) The fundamental role of epigenetic events i n cancer. Nat Rev Genet, 3,415-28. 148. Cosgrove , M . S . and Wolberger , C . (2005) H o w does the histone code w o r k ? Biochem Cell Biol, 83,468-76. 149. Jones, P . L . , Veenstra , G.J . , W a d e , P . A . , Ve rmaak , D . , Kass , S .U. , Landsberger , N . , S t rouboul is , J. and Wolffe , A . P . (1998) Me thy la t ed D N A and M e C P 2 recruit histone deacetylase to repress t ranscript ion. Nat Genet, 19,187-91. 150. Fuks , F. , Burgers, W . A . , B rehm, A . , Hughes-Davies , L . and Kouzar ides , T. (2000) D N A methyltransferase D n m t l associates w i t h histone deacetylase activity. Nat Genet, 24, 88-91. 151. Rountree, M . R . , Bachman, K . E . and B a y l i n , S.B. (2000) D N M T 1 b inds H D A C 2 and a n e w co-repressor, D M A P 1 , to f o r m a complex at repl ica t ion foci. Nat Genet, 25, 269-77. 166 152. El -Osta , A . (2003) D N M T cooperat ivi ty~the deve lop ing l inks between methyla t ion , chromat in structure and cancer. Bioessays, 25,1071-84. 153. Sharpless, N . E . , A l s o n , S., C h a n , S., Si lver , D .P . , Cas t r i l lon , D . H . and DeP inho , R . A . (2002) p l 6 ( I N K 4 a ) and p53 deficiency cooperate i n tumorigenesis. Cancer Res, 62,2761-5. 154. A u e r k a r i , E.I. (2006) M e t h y l a t i o n of tumor suppressor genes p l 6 ( I N K 4 a ) , p 2 7 ( K i p l ) and E-cadher in i n carcinogenesis. Oral Oncol, 42,5-13. 155. Birgisdot t i r , V . , Stefansson, O . A . , Bodvarsdot t i r , S .K. , H i lmarsdo t t i r , H , Jonasson, J .G. and Eyfjord, J.E. (2006) Epigenetic s i lencing and dele t ion of the B R C A 1 gene i n sporadic breast cancer. Breast Cancer Res, 8, R38. 156. Mat ros , E . , W a n g , Z . C . , Lode i ro , G . , M i r o n , A . , Iglehart, J .D. and Richardson, A . L . (2005) B R C A 1 promoter methyla t ion i n sporadic breast tumors: relat ionship to gene expression profiles. Breast Cancer Res Treat, 91,179-86. 157. M u r p h y , S.K. , H u a n g , Z . , W e n , Y . , Sp i l lman , M . A . , Whi taker , R.S., S ime l , L . R . , N icho l s , T .D . , M a r k s , J.R. and Berchuck, A . (2006) Frequent I G F 2 / H 1 9 d o m a i n epigenetic alterations and elevated IGF2 expression i n epi thel ia l ovar ian cancer. Mol Cancer Res, 4, 283-92. 158. Weissman , A . M . (2001) Themes and variat ions o n ubiqui ty la t ion . Nat Rev Mol Cell Biol, 2,169-78. 159. Passmore, L . A . and Barford, D . (2004) Ge t t ing into posi t ion: the catalytic mechanisms of prote in ub iqui ty la t ion . Biochem J, 379,513-25. 160. He r shko , A . , Ciechanover , A . , He l le r , H , Haas , A . L . and Rose, L A . (1980) Proposed role of A T P i n pro te in b reakdown: conjugation of prote in w i t h mul t ip l e chains of the po lypep t ide of ATP-dependen t proteolysis . Proc Natl Acad Sci USA, 77,1783-6. 161. W i l k i n s o n , K . D . , U rban , M . K . and Haas , A . L . (1980) U b i q u i t i n is the A T P -dependent proteolysis factor I of rabbit reticulocytes. / Biol Chem, 255,7529-32. 162. V o g e l , G . (2004) N o b e l Prizes. G o l d meda l f rom cel lular trash. Science, 306,400-1. 163. A r d l e y , H . C . and Robinson , P . A . (2005) E3 u b i q u i t i n ligases. Essays Biochem, 41, 15-30. 164. Ha takeyama, S., Y a d a , M . , Mat sumoto , M . , Ishida, N . and N a k a y a m a , K . I . (2001) U box proteins as a n e w fami ly of ub iqui t in -pro te in ligases. / Biol Chem, 276,33111-20. 165. Ha takeyama, S. and N a k a y a m a , K . L (2003) U-box proteins as a n e w fami ly of ub iqu i t i n ligases. Biochem Biophys Res Commun, 302,635-45. 166. Huibregtse , J . M . , Scheffner, M . , Beaudenon, S. and H o w l e y , P . M . (1995) A fami ly of proteins structural ly and funct ional ly related to the E 6 - A P ub iqu i t in -pro te in ligase. Proc Natl Acad Sci USA, 92,2563-7. 167. Ingham, R.J., G i s h , G . and Pawson , T. (2004) The N e d d 4 fami ly of E3 ub iqu i t i n ligases: functional d ivers i ty w i t h i n a c o m m o n m o d u l a r architecture. Oncogene, 23,1972-84. 168. Sherr, C.J . (2004) Pr inc ip les of tumor suppression. Cell, 116,235-46. 167 169. Sun , Y . (2003) Target ing E3 u b i q u i t i n ligases for cancer therapy. Cancer Biol Ther, 2,623-9. 170. O r l o w s k i , R . Z . and Dees, E . C . (2003) The role of the ubiqui t inat ion-proteasome pa thway i n breast cancer: a p p l y i n g drugs that affect the ubiqui t in-proteasome pa thway to the therapy of breast cancer. Breast Cancer Res, 5,1-7. 171. Fernandez, C . V . , Lestou, V . S . , W i l d i s h , J., Lee, C L . and Sorensen, P . H . (2001) Detect ion of a nove l t(6;15)(q21;q21) i n a pediatr ic W i l m s tumor. Cancer Genet Cytogenet, 129,165-7. 172. Bruce, C . K . , H o w a r d , P., N o w a k , N . J . and H o b a n , P.R. (2003) Molecu l a r analysis of r eg ion t(5;6)(q21;q21) i n W i l m s tumor. Cancer Genet Cytogenet, 141,106-13. 173. H o b a n , P.R., C o w e n , R . L . , M i t c h e l l , E . L . , Evans , D . G . , K e l l y , M . , H o w a r d , P.J. and H e i g h w a y , J. (1997) Phys ica l local isat ion of the breakpoints of a const i tut ional t ranslocation t(5;6)(q21;q21) i n a c h i l d w i t h bilateral W i l m s ' tumour . J Med Genet, 34,343-5. 174. Solis , V . , Pr i tchard , J. and C o w e l l , J .K. (1988) Cytogenet ic changes i n W i l m s ' tumors. Cancer Genet Cytogenet, 34,223-34. 175. Utada , Y . , H a g a , S., Kaj iwara , T., K a s u m i , F. , Sakamoto, G . , N a k a m u r a , Y . and E m i , M . (2000) M a p p i n g of target regions of allelic loss i n p r i m a r y breast cancers to 1 -cM intervals o n genomic contigs at 6q21 and 6q25.3. Jpn J Cancer Res, 91,293-300. 176. Z h a n g , Y . , Matthiesen, P., Harder , S., Siebert, R., Cas to ld i , G . , Calasanz, M . J . , W o n g , K . F . , Rosenwald , A . , Ott , G . , A t k i n , N . B . et al. (2000) A 3 - c M c o m m o n l y deleted region i n 6q21 i n leukemias and l y m p h o m a s del ineated by fluorescence i n s i tu hybr id iza t ion . Genes Chromosomes Cancer, 27, 52-8. 177. Inoue, M . , M a r x , A . , Zet t l , A . , Strobel, P., M u l l e r - H e r m e l i n k , H . K . and Starostik, P. (2002) C h r o m o s o m e 6 suffers frequent and mul t ip l e aberrations i n thymoma. Am J Pathol, 161,1507-13. 178. A u s u b e l , F . M . (ed.) (2002) Current Protocols in Molecular Biology, upda ted 2001 ed. John W i l e y and Sons, Inc., N e w Y o r k . 179. C h o m c z y n s k i , P. and Sacchi, N . (1987) Single-step me thod of R N A isola t ion by ac id g u a n i d i n i u m thiocyanate-phenol-chloroform extraction. Anal Biochem, 162, 156-9. 180. Sambrook, J. and Russel , D . W . (eds.) (2001) Molecular Cloning: a laboratory manual. T h i r d ed. C o l d S p r i n g Ha rbo r Laboratory Press, C o l d S p r i n g Harbor , N Y . 181. Engler , F . W . , Ha t f ie ld , J., Ne l son , W . and Soder lund , C . A . (2003) Loca t ing sequence o n F P C maps and selecting a m i n i m a l t i l i ng path. Genome Res, 13, 2152-63. 182. Soder lund , C , L o n g d e n , I. and Mot t , R. (1997) F P C : a system for b u i l d i n g contigs f rom restriction f ingerpr inted clones. Comput Appl Biosci, 13, 523-35. 183. Lupberger , J., Kreuzer , K . A . , Baskaynak, G . , Peters, U . R . , le Coutre , P. and Schmidt , C . A . (2002) Quant i ta t ive analysis of beta-actin, beta-2-microglobul in 168 and po rphob i l inogen deaminase m R N A and their compar i son as control transcripts for R T - P C R . Mol Cell Probes, 16,25-30. 184. Hama la inen , H . K . , T u b m a n , J.C., V i k m a n , S., K y r o l a , T., Y l i k o s k i , E . , War r ing ton , J .A. and Lahesmaa, R. (2001) Identification and va l ida t ion of endogenous reference genes for expression p ro f i l i ng of T helper cell differentiation by quanti tat ive real-time R T - P C R . Anal Biochem, 299, 63-70. 185. Gorze ln iak , K . , Janke, J., Enge l i , S. and Sharma, A . M . (2001) V a l i d a t i o n of endogenous controls for gene expression studies i n h u m a n adipocytes and preadipocytes. Horm Metab Res, 33, 625-7. 186. Burge, C . and K a r l i n , S. (1997) Pred ic t ion of complete gene structures i n h u m a n genomic D N A . / Mol Biol, 268,78-94. 187. Burge, C . B . and K a r l i n , S. (1998) F i n d i n g the genes i n genomic D N A . Curr Opin Struct Biol, 8, 346-54. 188. Burset, M . and G u i g o , R. (1996) Eva lua t ion of gene structure pred ic t ion programs. Genomics, 34,353-67. 189. Birney, E . , A n d r e w s , D . , Bevan, P., Caccamo, M . , C a m e r o n , G . , C h e n , Y . , C la rke , L . , Coates, G . , C o x , T., Cuff,. J. et al. (2004) Ensembl 2004. Nucleic Acids Res, 32, D468-70. 190. Birney, E . , A n d r e w s , D . , Caccamo, M . , C h e n , Y . , C la rke , L . , Coates, G . , C o x , T., C u n n i n g h a m , F., C u r w e n , V . , Cut ts , T. et al. (2006) Ensembl 2006. Nucleic Acids Res, 34, D556-61. 191. H u b b a r d , T., A n d r e w s , D . , Caccamo, M . , Cameron , G . , C h e n , Y . , C l a m p , M . , C la rke , L . , Coates, G . , C o x , T., C u n n i n g h a m , F. et al. (2005) Ensembl 2005. Nucleic Acids Res, 33, D447-53. 192. H in r i chs , A S . , Karo lch ik , D . , Baertsch, R., Barber, G .P . , Bejerano, G . , C l a w s o n , H . , Diekhans , M . , Furey, T.S., Harte , R. A . , H s u , F. et al. (2006) The U C S C Genome Browser Database: update 2006. Nucleic Acids Res, 34, D590-8. 193. Kent , W.J . , Sugnet, C . W . , Furey, T.S., Rosk in , K . M . , Pr ingle , T . H . , Zahler , A . M . and Haussler , D . (2002) The h u m a n genome browser at U C S C . Genome Res, 12, 996-1006. 194. Kent , W.J . (2002) B L A T - t h e B L A S T - l i k e a l ignment tool . Genome Res, 12,656-64. 195. Bateman, A . , C o i n , L . , D u r b i n , R., F i n n , R . D . , H o l l i c h , V . , Griffiths-Jones, S., Khanna , A . , M a r s h a l l , M . , M o x o n , S., Sonnhammer , E . L . et al. (2004) The Pfam pro te in families database. Nucleic Acids Res, 32, D138-41. 196. F i n n , R . D . , M i s t r y , J., Schuster-Bockler, B. , Griffiths-Jones, S., H o l l i c h , V . , Lassmann , T., M o x o n , S., M a r s h a l l , M . , K h a n n a , A . , D u r b i n , R. et al. (2006) Pfam: clans, w e b tools and services. Nucleic Acids Res, 34, D247-51. 197. Sonnhammer , E . L . , E d d y , S.R. and D u r b i n , R. (1997) Pfam: a comprehensive database of pro te in d o m a i n families based on seed al ignments. Proteins, 28,405-20. 198. Ang les io , M . S . , E v d o k i m o v a , V . , M e l n y k , N . , Z h a n g , L . , Fernandez, C . V . , G r u n d y , P .E . , Leach, S., M a r r a , M . A . , B rooks -Wi l son , A . R . , Penninger , J. et al. (2004) Differential expression of a nove l a n k y r i n conta in ing E3 ub iqu i t i n -169 prote in ligase, H a c e l , i n sporadic W i l m s ' tumor versus n o r m a l k idney . Hum Mol Genet, 13,2061-74. 199. Tognon , C , Knezev i ch , S.R., H u n t s m a n , D . , Roskel ley, C D . , M e l n y k , N . , Mathers , J .A. , Becker, L . , Carne i ro , F., MacPherson , N . , H o r s m a n , D.-et al. (2002) Express ion of the E T V 6 - N T R K 3 gene fus ion as a p r i m a r y event i n h u m a n secretory breast carcinoma. Cancer Cell, 2,367-76. 200. W a i , D . H . , K n e z e v i c h , S.R., Lucas , T., Jansen, B. , K a y , R.J. and Sorensen, P . H . (2000) The E T V 6 - N T R K 3 gene fus ion encodes a chimeric pro te in tyrosine kinase that transforms N I H 3 T 3 cells. Oncogene, 19, 906-15. 201. Sarkar, S., Roy , B . C . , Hatano, N . , A o y a g i , T., Gohj i , K . and K i y a m a , R. (2002) A nove l a n k y r i n repeat-containing gene (Kank) located at 9p24 is a g r o w t h suppressor of renal cel l carcinoma. / Biol Chem, 277,36585-91. 202. A n d e r s o n , C L . and B r o w n , C.J . (2002) Var i ab i l i t y of X chromosome inact ivat ion: effect o n levels of T I M P 1 R N A and role of D N A methyla t ion . Hum Genet, 110, 271-8. 203. Goodfe l l ow , P.J., M o n d e l l o , C , D a r l i n g , S . M . , P y m , B. , Li t t le , P. and Goodfe l l ow , P . N . (1988) Absence of methyla t ion of a C p G - r i c h reg ion at the 5' end of the M I C 2 gene on the active X , the inactive X , and the Y chromosome. Proc Natl Acad Sci USA, 85,5605-9. 204. L u o , R . X . , Post igo, A . A . and Dean , D C . (1998) R b interacts w i t h histone deacetylase to repress transcript ion. Cell, 92,463-73. 205. N g , H . H . , Feng, Q. , W a n g , H , Erdjument-Bromage, H , Tempst , P., Z h a n g , Y . and St ruhl , K . (2002) Lys ine methyla t ion w i t h i n the g lobular d o m a i n of histone H 3 by D o t l is impor tant for telomeric s i lencing and Sir pro te in association. Genes Dev, 16,1518-27. 206. E w i n g , B. , H i l l i e r , L . , W e n d l , M C . and Green , P. (1998) Base-cal l ing of automated sequencer traces u s ing phred . I. Accu racy assessment. Genome Res, 8, 175-85. 207. E w i n g , B. and Green , P. (1998) Base-cal l ing of automated sequencer traces us ing phred . II. Er ror probabili t ies. Genome Res, 8,186-94. 208. N icke r son , D . A . , Tobe, V . O . and Taylor , S.L. (1997) P o l y P h r e d : au tomat ing the detection and genotyp ing of single nucleot ide substitutions u s i n g fluorescence-based resequencing. Nucleic Acids Res, 25,2745-51. 209. G o r d o n , D . , Abaj ian, C . and Green , P. (1998) Consed : a graphical tool for sequence f in ishing. Genome Res, 8,195-202. 210. Sheng, W . W . , Soukup , S., Bove , K . , Go twa l s , B. and L a m p k i n , B. (1990) C h r o m o s o m e analysis of 31 W i l m s ' tumors. Cancer Res, 50,2786-93. 211. Betts, D .R. , Koesters, R., Pluss, H . J . and N i g g l i , F . K . (1997) Rout ine k a r y o t y p i n g i n W i l m s tumor. Cancer Genet Cytogenet, 96,151-6. 212. Slater, R . M . and Mannens , M . M . (1992) Cytogenetics a n d molecular genetics of W i l m s 1 t umor of ch i ldhood . Cancer Genet Cytogenet, 61,111-21. 213. Soukup , S., Go twa l s , B . , B lough , R. and L a m p k i n , B. (1997) W i l m s tumor: s u m m a r y of 54 cytogenetic analyses. Cancer Genet Cytogenet, 97,169-71. 170 214. Li t t le , S.E., H a n k s , S.P., K i n g - U n d e r w o o d , L . , Jones, C , Rapley , E . A . , Rahman , N . and Pritchard-Jones, K . (2004) Frequency and heri tabi l i ty of W T l mutat ions i n nonsyndromic W i l m s 1 t umor patients: a U K Chi ld ren ' s Cancer S tudy G r o u p Study. / Clin Oncol, 22,4140-6. 215. C o w e l l , J .K., W a d e y , R.B. , Haber , D . A . , C a l l , K . M . , H o u s m a n , D . E . and Pr i tchard , J. (1991) Structural rearrangements of the W T l gene i n W i l m s ' t umour cells. Oncogene, 6,595-9. 216. Coppes , M . J . , Liefers, G.J. , Pau l , P., Yeger , H . and W i l l i a m s , B.R. (1993) H o m o z y g o u s somatic W t l po in t mutat ions i n sporadic uni la teral W i l m s tumor. Proc Natl Acad Sci USA, 90,1416-9. 217. Lee, S.B. and Haber , D . A . (2001) W i l m s tumor and the W T l gene. Exp Cell Res, 264,74-99. 218. Li t t le , S.E., V u o n o n v i r t a , R., Reis-Fi lho , J.S., Natrajan, R., Iravani , M . , Fenwick , K . , M a c k a y , A . , A s h w o r t h , A . , Pritchard-Jones, K . and Jones, C . (2006) A r r a y C G H us ing w h o l e genome ampl i f ica t ion of fresh-frozen and formal in-f ixed, paraff in-embedded tumor D N A . Genomics, 87,298-306. 219. H i n g , S., L u , Y . J . , Summersg i l l , B. , K i n g - U n d e r w o o d , L . , N i c h o l s o n , J., G r u n d y , P., G r u n d y , R., Gessler, M . , Ship ley , J. and Pritchard-Jones, K . (2001) G a i n of l q is associated w i t h adverse outcome i n favorable his tology W i l m s ' tumors. Am ] Pathol, 158,393-8. 220. L u , Y . J . , H i n g , S., W i l l i a m s , R., P inker ton , R., Ship ley , J. and Pritchard-Jones, K . (2002) C h r o m o s o m e l q expression p rof i l ing and relapse i n W i l m s ' tumour . Lancet, 360,385-6. 221. Rapley , E . A . , Hargrave , D . , Persinguhe, N . , Barfoot, R., M o o r e , I., Radford , M . , Stratton, M . R . , Rahman , N . and Pritchard-Jones, K . (2001) Case of intersti t ial 12q delet ion i n association w i t h W i l m s tumor. Am J Med Genet, 104, 246-9. 222. Orphanos , V . , M c G o w n , G . , H e y , Y . , Thorncroft , M . , Santibanez-Koref, M . , Russel l , S.E., H i c k e y , I., Atkinson, .R.J . and Boyle , J . M . (1995) A l l e l i c imbalance of chromosome 6q i n ovar ian tumours . Br J Cancer, 71,666-9. 223. H y y t i n e n , E.R. , Saadut, R., C h e n , C , P a u l l , L . , Ko iv i s t o , P . A . , Vessel la , R . L . , Fr ierson, H . F . , Jr. and D o n g , J.T. (2002) De f in ing the region(s) of delet ion at 6ql6-q22 i n h u m a n prostate cancer. Genes Chromosomes Cancer, 34,306-12. 224. Sorensen, P . H . , L y n c h , J.C., Q u a l m a n , S.J., Tirabosco, R., L i m , J.F., Maure r , H . M . , Br idge , J .A. , Cr is t , W . M . , Triche, T.J. and Barr, F . G . (2002) P A X 3 - F K H R and P A X 7 - F K H R gene fusions are prognost ic indicators i n alveolar rhabdomyosarcoma: a report f rom the children's oncology group. / Clin Oncol, 20,2672-9. 225. Alessandr i , A . J . , Re id , G.S. , Bader, S .A. , M a s s i n g , B . G . , Sorensen, P . H . and Schul tz , K . R . (2002) E T V 6 ( T E L ) - A M L l pre-B acute lymphoblas t ic l eukaemia cells are associated w i t h a dist inct antigen-presenting phenotype. Br ] Haematol, 116,266-72. 226. Ginsberg , J.P., de A l a v a , E . , L a d a n y i , M . , Wexler , L . H . , K o v a r , H . , Paulussen, M . , Zoubek , A . , D o c k h o r n - D w o r n i c z a k , B. , Juergens, H . , W u n d e r , J.S. et al. 171 (1999) E W S - F L I 1 and E W S - E R G gene fusions are associated w i t h s imi la r c l in ica l phenotypes i n Ewing ' s sarcoma. / Clin Oncol, 17,1809-14. 227. de A l a v a , E . , K a w a i , A . , Hea ley , J . H , F l i gman , I., Meyers , P. A . , H u v o s , A . G . , Ge ra ld , W . L . , Jhanwar, S C . , A r g a n i , P., Antonescu , C .R . et al. (1998) E W S - F L I 1 fus ion transcript structure is an independent determinant of prognosis i n Ewing ' s sarcoma. / Clin Oncol, 16,1248-55. 228. Argatoff, L . H . , O ' C o n n e l l , J .X., Mathers , J. A . , G i l k s , C . B . and Sorensen, P . H . (1996) Detect ion of the E W S / W T 1 gene fus ion by reverse transcriptase-polymerase cha in reaction i n the diagnosis of in t ra-abdominal desmoplast ic sma l l r o u n d cel l tumor. Am } Surg Pathol, 20,406-12 . 229. Sorensen, P . H . , Lessnick, S.L., Lopez-Terrada , D . , L i u , X . F . , Triche, T.J. and Denny , C .T . (1994) A second Ewing ' s sarcoma translocation, t(21;22), fuses the E W S gene to another ETS-fami ly t ranscr ipt ion factor, E R G . Nat Genet, 6,146-51. 230. Sorensen, P . H . , L i u , X . F . , Delattre, O. , R o w l a n d , J . M . , Biggs , C . A . , Thomas, G . and Triche, T.J. (1993) Reverse transcriptase P C R ampl i f ica t ion of E W S / F L I - 1 fus ion transcripts as a diagnostic test for per ipheral p r imi t i ve neuroectodermal tumors of ch i ldhood . Diagn Mol Pathol, 2,147-57. 231. Osoegawa, K . , M a m m o s e r , A . G . , W u , C , Frengen, E . , Z e n g , C , Catanese, J.J. and de Jong, P.J. (2001) A bacterial artificial chromosome l ibrary for sequencing the complete h u m a n genome. Genome Res, 11,483-96. 232. Osoegawa, K . , W o o n , P . Y . , Zhao , B. , Frengen, E . , Tateno, M . , Catanese, J.J. and de Jong, P.J. (1998) A n i m p r o v e d approach for construct ion of bacterial artificial chromosome libraries. Genomics, 52,1-8. 233. Ness, S.R., Terpstra, W . , K r z y w i n s k i , M . , M a r r a , M . A . a n d Jones, S.J. (2002) A s s e m b l y of f ingerprint contigs: para l le l ized F P C . Bioinformatics, 18,484-5. 234. Lennon , G . , Auff ray , C , Po lymeropou los , M . and Soares, M . B . (1996) The I . M . A . G . E . Conso r t ium: an integrated molecular analysis of genomes and their expression. Genomics, 33,151-2. 235. Heger , A . and H o l m , L . (2000) Towards a cover ing set of pro te in fami ly profiles. Prog Biophys Mol Biol, 73,321-37. 236. Bateman, A . , Bi rney, E . , D u r b i n , R., E d d y , S.R., F i n n , R . D . and Sonnhammer , E . L . (1999) P f a m 3.1:1313 mul t ip l e al ignments and profi le H M M s match the majority of proteins. Nucleic Acids Res, 27, 260-2. 237. Sonnhammer , E . L . , E d d y , S.R., Bi rney, E . , Bateman, A . and D u r b i n , R. (1998) Pfam: mu l t ip l e sequence al ignments and H M M - p r o f i l e s of prote in domains . Nucleic Acids Res, 26,320-2. 238. Letunic , I., C o p l e y , R.R. , Schmidt , S., Ciccare l l i , F . D . , Doerks , T., Schul tz , J., Pont ing , C P . and Bork, P. (2004) S M A R T 4.0: towards genomic data integration. Nucleic Acids Res, 32, D142-4. 239. Schul tz , J., C o p l e y , R.R. , Doerks , T., Pont ing , C P . and Bork , P. (2000) S M A R T : a web-based tool for the s tudy of genetically mobi le domains . Nucleic Acids Res, 28,231-4. 172 240. Pont ing , C P . , Schul tz , J., M i l p e t z , F. and Bork, P. (1999) S M A R T : identif icat ion and annotat ion of domains f rom s igna l l ing and extracellular prote in sequences. Nucleic Acids Res, 27,229-32. 241. M u l d e r , N . J . , A p w e i l e r , R., A t t w o o d , T . K . , Bai roch, A . , Bateman, A . , Binns , D . , Bradley , P., Bork, P., Bucher, P., Cerut t i , L . et al. (2005) InterPro, progress and status i n 2005. Nucleic Acids Res, 33, D201-5. 242. M u l d e r , N . J . , A p w e i l e r , R., A t t w o o d , T . K . , Bai roch, A . , Bateman, A . , Binns , D . , Biswas , M . , Bradley, P. , Bork , P., Bucher, P. et al. (2002) InterPro: an integrated documenta t ion resource for prote in families, domains and funct ional sites. Brief Bioinform, 3, 225-35. 243. Bork , P. (1993) H u n d r e d s of ankyr in - l ike repeats i n funct ional ly diverse proteins: mobi le modules that cross p h y l a hor izonta l ly? Proteins, 17,363-74. 244. Scheffner, M . , Huibregtse , J . M . , Viers t ra , R . D . and H o w l e y , P . M . (1993) The H P V - 1 6 E6 and E 6 - A P complex functions as a ubiqui t in -pro te in ligase i n the ub iqu i t ina t ion of p53. Cell, 75,495-505. 245. Y a n g , Y . , L i , C C . and Weissman , A . M . (2004) Regula t ing the p53 system th rough ubiqui t ina t ion . Oncogene, 23, 2096-106. 246. Fogh , J. (1978) Cu l t i va t i on , characterization, and ident if icat ion of h u m a n tumor cells w i t h emphasis o n k idney , testis, and bladder tumors. Natl Cancer Inst Monogr, 5-9. 247. Fogh , J., Fogh , J . M . and Orfeo, T. (1977) One h u n d r e d and twenty-seven cul tured h u m a n tumor cel l l ines p r o d u c i n g tumors i n nude mice. / Natl Cancer Inst, 59, 221-6. 248. O r k i n , S .H . , G o l d m a n , D.S. and Sallan, S.E. (1984) Deve lopment of homozygos i ty for chromosome H p markers i n W i l m s ' tumour . Nature, 309,172-4. 249. W a n g , M . and Pickart , C M . (2005) Different H E C T d o m a i n u b i q u i t i n ligases emp loy distinct mechanisms of p o l y u b i q u i t i n cha in synthesis. Embo J, 24,4324-33. 250. Bole, D . G . , D o w i n , R., Dor i aux , M . and Jamieson, J.D. (1989) Immunocytochemica l loca l iza t ion of B i P to the r o u g h endoplasmic re t iculum: evidence for prote in sor t ing by selective retention. / Histochem Cytochem, 37, 1817-23. 251. N i s h i k a w a , S., H i ra t a , A . and N a k a n o , A . (1994) Inhib i t ion of endoplasmic r e t i cu lum (ER)- to-Golgi transport induces re local iza t ion of b i n d i n g pro te in (BiP) w i t h i n the E R to f o r m the B i P bodies. Mol Biol Cell, 5,1129-43. 252. K a r i n , M . and Ben-Ner iah , Y . (2000) Phosphory la t ion meets ubiqui t ina t ion: the control of NF-[kappa]B act ivi ty. Annu Rev Immunol, 18,621-63. 253. Pickart , C M . and Edd ins , M . J . (2004) U b i q u i t i n : structures, functions, mechanisms. Biochim Biophys Acta, 1695,55-72. 254. Pickart , C M . and Cohen , R .E . (2004) Proteasomes a n d their k i n : proteases i n the machine age. Nat Rev Mol Cell Biol, 5,177-87. 173 255. Pickart , C M . (2002) D N A repair: r ight on target w i t h ub iqu i t in . Nature, 419,120-1. 256. Pickart , C M . (2001) U b i q u i t i n enters the n e w m i l l e n n i u m . Mol Cell, 8,499-504. 257. H a g l u n d , K . , D i Fiore, P.P. and D i k i c , I. (2003) Dis t inct m o n o u b i q u i t i n signals i n receptor endocytosis. Trends Biochem Sci, 28,598-603. 258. D i Fiore, P.P. , Po lo , S. and H o f m a n n , K . (2003) W h e n ub iqu i t i n meets ub iqu i t i n receptors: a s igna l l ing connection. Nat Rev Mol Cell Biol, 4,491-7. 259. Passmore, L . A . (2004) The anaphase-promoting complex ( A P C ) : the s u m of its parts? Biochem Soc Trans, 32, 724-7. 260. Urbe , S. (2005) U b i q u i t i n and endocytic prote in sort ing. Essays Biochem, 41, 81-98. 261. Savi l le , M . K . , Sparks, A . , X i r o d i m a s , D.P . , W a r d r o p , J., Stevenson, L . F . , B o u r d o n , J . C , W o o d s , Y . L . and Lane, D .P . (2004) Regula t ion of p53 by the ubiqui t in-conjugat ing enzymes U b c H 5 B / C i n v i v o . / Biol Chem, 279,42169-81. 262. Schwarz , S.E., Rosa, J.L. and Scheffner, M . (1998) Character iza t ion of h u m a n hect d o m a i n fami ly members and their interaction w i t h U b c H 5 and U b c H 7 . / Biol Chem, 273,12148-54. 263. Nube r , U . , Schwarz , S., Kaiser , P., Schneider, R. and Scheffner, M . (1996) C l o n i n g of h u m a n ubiqui t in-conjugat ing enzymes U b c H 6 and U b c H 7 (E2-F1) and characterization of their interaction w i t h E 6 - A P and RSP5. / Biol Chem, 271, 2795-800. 264. Kaiser , P., Seufert, W . , Hofferer, L . , Kofler , B. , Sachsenmaier, C , H e r z o g , H . , Jentsch, S., Schweiger , M . and Schneider, R. (1994) A h u m a n ub iqu i t i n -conjugating enzyme homologous to yeast U B C 8 . / Biol Chem, 269,8797-802. 265. Blumenfe ld , N . , G o n e n , H . , M a y e r , A . , Smi th , C . E . , Siegel , N . R . , Schwartz , A . L . and Ciechanover , A . (1994) Pur i f ica t ion and characterizat ion of a nove l species of ubiqui t in-carr ier protein, E2, that is i n v o l v e d i n degradat ion of non-"N-end rule" prote in substrates. / Biol Chem, 269, 9574-81. 266. Jarosch, E . , Lenk , U . and Sommer , T. (2003) Endop la smic ret iculum-associated pro te in degradation. Int Rev Cytol, 223,39-81. 267. K o u m e n i s , C . (2006) E R stress, h y p o x i a tolerance a n d tumor progression. Curr Mol Med, 6,55-69. 268. Huibregtse , J . M . , Scheffner, M . and H o w l e y , P . M . (1993) C l o n i n g a n d expression of the c D N A for E 6 - A P , a prote in that mediates the interaction of the h u m a n pap i l l omav i rus E6 oncoprote in w i t h p53. Mol Cell Biol, 13,775-84. 269. Yamamoto , K . , Ito, M . , M i n a g a w a , K . , Urahama , N . , Sada, A . , O k a m u r a , A . and M a t s u i , T. (2005) A der(13)t(7;13)(pl3;ql4) w i t h monoal le l ic loss of R B I and D13S319 i n myelodysplas t ic syndrome. Cancer Genet Cytogenet, 162,160-5. 270. C o w e l l , J .K., L a D u c a , J., Ross i , M . R . , Burkhardt , T., N o w a k , N J . and M a t s u i , S. (2005) M o l e c u l a r characterization of the t(3;9) associated w i t h immor ta l i za t ion i n the M C F 1 0 A cel l l ine. Cancer Genet Cytogenet, 163,23-9. 271. N i s h i o , J., Gent ry , J.D., Neff, J.R., Ne l son , M . , Daniels , W . , Perry, D . , Gatal ica , Z . and Br idge , J. A . (2006) Monoa l l e l i c delet ion of the p53 gene th rough 174 chromosomal translocation i n a sma l l cel l osteosarcoma. Virchows Arch, 448, 852-6. 272. Rabbitts, T . H . (1997) C h r o m o s o m a l breakpoints hi t the spot. Nat Med, 3,496-7. 273. Rabbitts, T . H . (1994) C h r o m o s o m a l translocations i n h u m a n cancer. Nature, 372, 143-9. 274. G i b s o n , U . E . , H e i d , C . A . and W i l l i a m s , P . M . (1996) A nove l me thod for real t ime quanti tat ive R T - P C R . Genome Res, 6,995-1001. 275. H e i d , C . A . , Stevens, J., L i v a k , K . J . and W i l l i a m s , P . M . (1996) Real t ime quanti tat ive P C R . Genome Res, 6, 986-94. 276. Sun , Y . , Hegamyer , G . and C o l b u r n , N . H . (1992) A s imple me thod u s i n g P C R for direct sequencing of genomic D N A f rom frozen tumor tissue embedded i n op t ima l cut t ing temperature c o m p o u n d . Biotechniques, 12, 639-40. 277. Pritchard-Jones, K . (1999) The W i l m s t umour gene, W T 1 , i n n o r m a l and abnormal nephrogenesis. Pediatr Nephrol, 13, 620-5. 278. Jain, P . K . (2003) Epigenetics: the role of methyla t ion i n the mechan i sm of act ion of tumor suppressor genes. Ann N Y Acad Sci, 983,71-83. 279. M o m p a r l e r , R . L . and Bovenz i , V . (2000) D N A methy la t ion and cancer. / Cell Physiol, 183,145-54. 280. Jones, P . A . and Taylor , S . M . (1980) Ce l lu l a r differentiation, cy t id ine analogs and D N A methyla t ion . Cell, 20,85-93. 281. I izuka , M . and Smi th , M . M . (2003) Funct iona l consequences of histone modificat ions. Curr Opin Genet Dev, 13,154-60. 282. Berger, S.L. (2002) His tone modif icat ions i n t ranscript ional regulat ion. Curr Opin Genet Dev, 12,142-8. 283. Y o s h i d a , B . A . , Sokoloff, M . M . , W e l c h , D .R . and Rinker-Schaeffer, C . W . (2000) Metastasis-suppressor genes: a r ev iew and perspective o n an emerging f ie ld. / Natl Cancer Inst, 92,1717-30. 284. C l a r k , S.J., Ha r r i son , J., Pau l , C L . and Frommer , M . (1994) H i g h sensi t ivi ty m a p p i n g of methyla ted cytosines. Nucleic Acids Res, 22,2990-7. 285. L i , E . , Bestor, T . H . and Jaenisch, R. (1992) Targeted muta t ion of the D N A methyltransferase gene results i n embryonic lethali ty. Cell, 69,915-26. 286. West , A . G . and Fraser, P. (2005) Remote control of gene transcript ion. Hum Mol Genet, 14 Spec No 1, R101-11. 287. G r e w a l , S.I. and M o a z e d , D . (2003) Hete rochromat in and epigenetic control of gene expression. Science, 301,798-802. 288. Ginjala , V . , H o l m g r e n , C , Ul leras , E . , K a n d u r i , C , Pant, V . , Lobanenkov , V . , F r a n k l i n , G . and Ohlsson , R. (2002) M u l t i p l e cis elements w i t h i n the Ig f2 /H19 insulator d o m a i n organize a distance-dependent silencer. A caut ionary note. / Biol Chem, 277,5707-10. 289. Srivastava, M . , Hs i eh , S., Gr inbe rg , A . , Wi l l i ams -S imons , L . , H u a n g , S.P. and Pfeifer, K . (2000) H 1 9 a n d Igf2 monoal le l ic expression is regulated i n two distinct w a y s by a shared cis act ing regulatory reg ion ups t ream of H19 . Genes Dev, 14,1186-95. 175 290. . G r e w a l , S.I. and E l g i n , S.C. (2002) Heterochromat in : n e w possibil i t ies for the inheritance of structure. Curr Opin Genet Dev, 12,178-87. 291. Blume-Jensen, P. and Hunte r , T. (2001) Oncogenic kinase s ignal l ing . Nature, 411, 355-65. 292. Lev ine , A . J . , M o m a n d , J. and F in lay , C . A . (1991) The p53 t umour suppressor gene. Nature, 351,453-6. 293. Sandy, P., Ven tura , A . and Jacks, T. (2005) M a m m a l i a n R N A i : a practical guide. Biotechniques, 39, 215-24. 294. H a m i d , T., M a l i k , M . T . and Kaka r , S.S. (2005) Ectopic expression of P T T G 1 / s e c u r i n promotes tumorigenesis i n h u m a n embryonic k i d n e y cells. Mol Cancer, 4,3. 295. K a m e i , D . , M u r a k a m i , M . , Naka tan i , Y . , I shikawa, Y . , Ishii , T. and K u d o , I. (2003) Potential role of mic rosomal pros taglandin E synthase-1 i n tumorigenesis. / Biol Chem, 278,19396-405. 296. Layzer , J . M . , McCaff rey , A . P . , Tanner, A . K . , H u a n g , Z . , K a y , M . A . and Sullenger, B . A . (2004) In v i v o act ivi ty of nuclease-resistant s i R N A s . Rna, 10, 766-71. 297. Sharma, A . , T r i v e d i , N . R . , Z i m m e r m a n , M . A . , Tuveson , D . A . , Smi th , C D . and Robertson, G .P . (2005) M u t a n t V599EB-Raf regulates g r o w t h and vascular development of mal ignant me lanoma tumors. Cancer Res, 65,2412-21. 298. Sherr, C.J . and Roberts, J . M . (1999) C D K inhibi tors: posi t ive and negative regulators of G l - p h a s e progression. Genes Dev, 13,1501-12. 299. C h a n g , F. , Steelman, L .S . , Shelton, J .G. , Lee, J.T., N a v o l a n i c , P . M . , Bla lock, W . L . , F r a n k l i n , R. and M c C u b r e y , J .A. (2003) Regula t ion of cel l cycle progress ion and apoptosis by the R a s / R a f / M E K / E R K pa thway (Review). Int J Oncol, 22,469-80. 300. Roovers , K . and Asso ian , R . K . (2000) Integrating the M A P kinase s ignal into the G l phase cell cycle machinery . Bioessays, 22,818-26. 301. Stein, G . H . and D u l i c , V . (1998) M o l e c u l a r mechanisms for the senescent cell cycle arrest. / Investig Dermatol Symp Proc, 3,14-8. 302. Stein, G . H . and D u l i c , V . (1995) Or ig in s of G l arrest i n senescent h u m a n fibroblasts. Bioessays, 17, 537-43. 303. Ma lumbres , M . and Barbacid , M . (2001) To cycle or not to cycle: a cr i t ical decis ion i n cancer. Nat Rev Cancer, 1,222-31. 304. N icos i a , S.V., Ba i , W . , C h e n g , J.Q., C o p p o l a , D . and K r u k , P . A . (2003) Oncogenic pa thways impl ica ted i n ovar ian epi thel ial cancer. Hematol Oncol Clin North Am, 17,927-43. 305. C la rke , R .B . (2003) p 2 7 K I P l phosphory la t ion by P K B / A k t leads to poor breast cancer prognosis . Breast Cancer Res, 5,162-3. 306. Nie l sen , T .O. , A n d r e w s , H . N . , Cheang , M . , Kucab , J.E., H s u , F . D . , Ragaz, J., G i l k s , C . B . , Makre t sov , N . , Bajdik, C D . , Brookes, C . et al. (2004) Express ion of the insu l in- l ike g r o w t h factor I receptor and urokinase p lasminogen activator i n breast cancer is associated w i t h poor su rv iva l : potential for in tervent ion w i t h 17-al ly lamino ge ldanamycin . Cancer Res, 64,286-91. 176 307. K i m , D . , Cheng , G . Z . , L inds l ey , C . W . , Y a n g , H . and C h e n g , J.Q. (2005) Target ing the phosphat idyl inosi to l -3 k inase / A k t pa thway for the treatment of cancer. Curr Opin Investig Drugs, 6,1250-8. 308. C h e n g , J.Q., L inds l ey , C . W . , Cheng , G . Z . , Y a n g , H . and Nicos i a , S .V. (2005) The A k t / P K B pathway: molecular target for cancer d r u g discovery. Oncogene, 24, 7482-92. 309. K u r i e , J . M . (2004) Role of pro te in kinase B-dependent s igna l ing i n l u n g tumorigenesis. Chest, 125,141S-4S. 310. Fresno V a r a , J .A. , Casado, E . , de Castro, J., Cejas, P., Belda-Iniesta, C . and Gonza lez -Baron , M . (2004) P I 3 K / A k t s igna l l ing pa thway and cancer. Cancer Treat Rev, 30,193-204. 311. Baek, H.J . , K i m , S.S., da S i lva , F . M . , V o l p e , E . A . , Evans , S., M i s h r a , B. , M i s h r a , L . and Bla i r M a r s h a l l , M . (2006) Inact ivat ion of TGF-be ta s igna l ing i n l u n g cancer results i n increased C D K 4 act ivi ty that can be rescued by E L F . Biochem Biophys Res Commun, 346,1150-7. 312. K o z a r , K . , C i emerych , M . A . , Rebel , V . I . , Shigematsu, H . , Z a g o z d z o n , A . , Sic inska, E . , G e n g , Y . , Y u , Q. , Bhattacharya, S., Bronson, R.T. et al. (2004) M o u s e development and cel l prol i ferat ion i n the absence of D-cycl ins . Cell, 118,477-91. 313. Fatrai , S., E lghaz i , L . , Balcazar, N . , Cras -Meneur , C , Kr i t s , I., K i y o k a w a , H . and Berna l -Miz rach i , E . (2006) A k t induces beta-cell prol i ferat ion by regula t ing cyc l i n D l , cyc l i n D 2 , and p21 levels and cycl in-dependent kinase-4 activity. Diabetes, 55,318-25. 314. Gao , N . , F l y n n , D . C . , Z h a n g , Z . , Z h o n g , X . S . , Wa lke r , V . , L i u , K . J . , Sh i , X . and Jiang, B . H . (2004) G l cel l cycle progress ion and the expression of G l cycl ins are regulated by P I 3 K / A K T / m T O R / p 7 0 S 6 K l s igna l ing i n h u m a n ovar ian cancer cells. Am J Physiol Cell Physiol, 287, C281-91. 315. C a p p e l l i n i , A . , Tabe l l in i , G . , Z w e y e r , M . , Bor tu l , R., Tazzar i , P . L . , B i l l i , A . M . , Fala , F. , Cocco , L . and M a r t e l l i , A . M . (2003) The phosphoinos i t ide 3-kinase/ A k t pa thway regulates cel l cycle progress ion of H L 6 0 h u m a n l eukemia cells th rough cytoplasmic re local izat ion of the cycl in-dependent kinase inhibi tor p 2 7 ( K i p l ) and control of c y c l i n D l expression. Leukemia, 17,2157-67. 316. R a d u , A . , Neubauer , V . , A k a g i , T., Hanafusa, H . and Georgescu, M . M . (2003) P T E N induces cel l cycle arrest by decreasing the leve l and nuclear loca l iza t ion of c y c l i n D l . Mol Cell Biol, 23,6139-49. 317. L i , Y . , D o w b e n k o , D . and Lasky , L . A . (2002) A K T / P K B phosphory la t ion of p 2 1 C i p / W A F l enhances prote in stabil i ty of p 2 1 C i p / W A F l and promotes cel l su rv iva l . / Biol Chem, 277,11352-61. 318. Paramio , J . M . , Nava r ro , M . , Segrelles, C , Gomez-Casero , E . and Jorcano, J.L. (1999) P T E N tumour suppressor is l i n k e d to the cel l cycle control th rough the ret inoblastoma protein. Oncogene, 18, 7462-8. 319. Napie ra la , D . , Garcia-Rojas, X . , Sam, K . , W a k u i , K . , C h e n , C , M e n d o z a -L o n d o n o , R., Z h o u , G . , Zheng , Q . and Lee, B. (2005) Muta t ions and promoter 177 S N P s i n R U N X 2 , a t ranscript ional regulator of bone format ion. Mol Genet Metab, 86,257-68. 320. Z h o u , X . P . , Wai te , K . A . , P i l a r sk i , R., H a m p e l , H . , Fernandez, M J . , Bos, C , Dasouk i , M . , Fe ldman , G . L . , Greenberg, L . A . , Ivanovich , J. et al. (2003) Ge rml ine P T E N promoter mutat ions and deletions i n C o w d e n / B a n n a y a n - R i l e y -Ruvalcaba syndrome result i n aberrant P T E N pro te in and dysregula t ion of the phospho inos i to l -3 -k inase /Akt pa thway. Am J Hum Genet, 73,404-11. 321. A u d i c , Y . and Har t ley , R.S. (2004) Post- transcript ional regula t ion i n cancer. Biol Cell, 96, 479-98. 322. Der r igo , M . , Ceste l l i , A . , Savettieri , G . and D i L iegro , I. (2000) R N A - p r o t e i n interactions i n the control of stabili ty and loca l iza t ion of messenger R N A (review). Int J Mol Med, 5,111-23. 323. N a k a y a m a , K . I . a n d ' N a k a y a m a , K . (2006) U b i q u i t i n ligases: cell-cycle control and cancer. Nat Rev Cancer, 6,369-81. 324. Premalata, C.S. , Gaya th r i D e v i , M . , Biswas, S., Mukherjee, G . , Ba lu , S., Sundareshan, T.S. and Prabhakaran, P.S. (2003) P r imi t i ve neuroectodermal tumor of the k idney . A report of two cases d iagnosed by fine needle aspira t ion cytology. Acta Cytol, 47,475-9. 325. M a c c i o n i , F. , De l i a Rocca, C . , Sa lv i , P.F. , Man icone , A . M . , Asca re l l i , A . , L o n g o , F. and Rossi , P. (2000) M a l i g n a n t per ipheral neuroectodermal tumor ( M P N E T ) of the k idney . Abdom Imaging, 25,103-6. 326. Quezado , M . , Benjamin, D .R . and Tsokos, M . (1997) E W S / F L I - 1 fus ion transcripts i n three per ipheral p r imi t ive neuroectodermal tumors of the k idney . Hum Pathol, 28, 767-71. 327. Rodr iguez -Ga l indo , C . , Spunt , S.L. and Pappo , A . S . (2003) Treatment of E w i n g sarcoma fami ly of tumors: current status and out look for the future. Med Pediatr Oncol, 40,276-87. 328. Kerscher, O . , Felberbaum, R. and Hochstrasser, M . (2006) M o d i f i c a t i o n of Proteins by U b i q u i t i n and U b i q u i t i n - L i k e Proteins. Annu Rev Cell Dev Biol. 329. Pestell , R . G . , Albanese, C . , Reutens, A . T . , Segall , J.E., Lee, R J . and A r n o l d , A . (1999) The cycl ins a n d cycl in-dependent kinase inhibi tors i n ho rmona l regula t ion of prol i ferat ion and differentiation. Endocr Rev, 20,501-34. 330. L i , Q . , Barkess, G . and Q i a n , H . (2006) C h r o m a t i n l o o p i n g and the probabi l i ty of t ranscript ion. Trends Genet, 22,197-202. 331. L i , Q . , Peterson, K . R . , Fang, X . and Stamatoyannopoulos , G . (2002) Locus cont ro l regions. Blood, 100,3077-86. 332. Festenstein, R. and Kiouss i s , D . (2000) Locus cont ro l regions and epigenetic ch romat in modif iers . Curr Opin Genet Dev, 10,199-203. 333. Sharpless, N . E . (2005) I N K 4 a / A R F : a mul t i funct ional tumor suppressor locus. Mutat Res, 576,22-38. 334. Sharpless, N . E . , Bardeesy, N . , Lee, K . H . , Carrasco, D . , Cas t r i l lon , D . H . , A g u i r r e , A . J . , W u , E . A . , Horner , J .W. and DeP inho , R . A . (2001) Loss of p l 6 I n k 4 a w i t h retention of p l 9 A r f predisposes mice to tumorigenesis . Nature, 413, 86-91. 178 335. H e r z o g , C .R . , N o h , S., Lan t ry , L . E . , G u a n , K . L . and Y o u , M . (1999) C d k n 2 a encodes funct ional var ia t ion of p l 6 I N K 4 a but not p l 9 A R F , w h i c h confers selection i n mouse l u n g tumorigenesis . Mol Carcinog, 25, 92-8. 336. Jackson, R.J., A d n a n e , J., C o p p o l a , D . , Cantor , A . , Sebti, S . M . and Pledger, W.J . (2002) Loss of the cel l cycle inhibi tors p 2 1 ( C i p l ) and p 2 7 ( K i p l ) enhances . tumorigenesis i n knockou t mouse models . Oncogene, 21,8486-97. 337. Donehower , L . A . , H a r v e y , M . , Slagle, B . L . , M c A r t h u r , M . J . , M o n t g o m e r y , C . A . , Jr., Bute l , J.S. and Bradley , A . (1992) M i c e deficient for p53 are developmenta l ly n o r m a l but susceptible to spontaneous tumours . Nature, 356, 215-21. 338. H a r v e y , M . , M c A r t h u r , M . J . , M o n t g o m e r y , C . A . , Jr., Bradley , A . and Donehower , L . A . (1993) Genetic background alters the spect rum of tumors that develop i n p53-deficient mice. Faseb J, 7, 938-43. 339. H a r v e y , M . , M c A r t h u r , M . J . , M o n t g o m e r y , C . A . , Jr., Butel , J.S., Bradley, A . and Donehower , L . A . (1993) Spontaneous and carcinogen- induced tumorigenesis i n p53-deficient mice. Nat Genet, 5, 225-9. 179 A P P E N D I X A Stock Solutions & Recipes Selected C o m m o n Solut ions are l is ted be low, for add i t iona l Solut ions, Recipes and Protocols see Chapter II: Mater ia ls & Methods or Current Protocols in Molecular Biology (178) and Molecular Cloning: A Laboratory Manual (180). 6x D N A Loading 25 m g B r o m o p h e n o l blue 25 m g Xylene C y a n o l 4g (40 %) Sucrose H 2 0 to 10 m l For RNA Loading Buffer use DEPC treated H2O Coomassie Blue Solution 2 g (0.2%) Coomass ie Blue 75 m l Acet ic ac id 500 m l M e t h a n o l Top u p to 1 L w i t h d d L b O Coomassie De-stain Solution D E P C treated H 2 Q 75 m l Acet ic A c i d 100 m l E thano l Top u p to 1 L w i t h d d E b O 1000 m l H2O (or d d H 2 0 or Ul t raPure H2O) 1 m l (0.1%) Die thylpyrocarbonate ( D E P C ) Let sit at R T for at least 1 hour , Au toc lave E D T A (0.5M p H 8.0) 186 g E D T A H2O u p to 800 m l Adjus t p H s l o w l y u p to ~7 w i t h N a O H pellets (EDTA will not dissolve until neutral pH is reached) Fina l ize to p H 8.0 w i t h 5 M N a O H H 2 0 u p to 1L , Au toc lave L a e m m l i 30 g Tris base Elec t rophores is Buf fe r 144 g G l y c i n e ( lx) 2.9 g E D T A 10 g S D S H2O u p to 10 litres 180 L B Media 5 g Yeast Extract 10 g Tryptone-Peptone 5 g N a C l H 2 0 u p to 1L p H 7.5, Au toc l ave For L B Plates a d d - 1 5 g agar before A u t o c l a v i n g Effective Dose Of Antibiotics For Common Lab Strains Of E.Coli: A m p i c i l l i n 50-100 p g / m l C h l o r a m p h e n i c o l 20 p g / m l K a n a m y c i n (Neomycin) 30 p g / m l Tet racycl in 12-15 p g / m l Nonidet-P40 (NP-40) 20 m M Tris H C 1 p H 8 buffer: 137 m M N a C l 10% glycerol 1% non ide tP-40 2 m M E D T A P B S (lOx) Phosphate Buffered Saline lOx concentrat ion 23.2 g N a 2 H P 0 4 2 g K C l 2 g K3PO4 80 g N a C l (1 L d is t i l l ed water) p H 7 . 4 For l x P B S di lute 10-fold i n d d l r b O and steril ize by fi l t rat ion (0.2pm) or au toc lav ing before use i n cel l culture or sterile applicat ions 181 RIPA buffer (RadioImmunoPrecipit ation Assay) buffer: 5 0 m M Tr is H C 1 p H 8.0 150 m M N a C l 1% NP-40 0.5% s o d i u m Deoxycholate 0.1% S D S SSC (20x) The 10% sodium deoxycholate stock solution (5 g into 50 ml) must be protected from light. 350 g N a C l 176.4 g Na 3 -C i t r a t e H 2 O t o 9 5 0 m l Adjus t p H 7.0 w i t h H C 1 H 2 0 to 1 L T A E (50x) 18.6 g E D T A 242 g Tris base 57 m l Glac i a l Acet ic H 2 Q to 1 L TBS lOx (concentrated TBS) T B S T 24.23 g Tris base 80.06 g N a C l M i x i n 800 m l u l t r a pure water. p H to 7.6 w i t h pure H C 1 . Top u p to 1 L 100 m l of T B S lOx 900 m l ul t ra pure water l m l Tween20 T E (lOx) Western Transfer Buffer (lOx) 100 m l I M Tris p H 8.0 20 m l 0 . 5 M E D T A p H 8.0 H 2 0 to 1 L ( p H 8.0) Au toc lave or filter sterilize (0.2pm) 73 g Tr is base 300 g G l y c i n e optional - 4 ml 20% SDS (PVDF only) H 2 Q to 2 L For l x buffer a l l o w for u p to 20% M e t h a n o l w h e n d i l u t i n g 182 A P P E N D I X B A hacel Mutant Mouse Model Unless otherwise noted, data contained in this appendix was generated in the laboratory of Dr. Josef Penninger at the University of Toronto, Toronto, Ontario, Canada and at the Institute for Molecular Biotechnology of the Austrian Academy of Sciences (1MBA), Vienna, Austria. To s tudy the in vivo funct ion of H A C E 1 and to p rov ide genetic proof that this gene is indeed i n v o l v e d i n cancer pathogenesis, hacel mutant mice were generated. M o u s e (Genbank accession NP_766061) and h u m a n (Genbank accession NP_065822) H A C E 1 proteins share ~96% identi ty, > 60% nucleic ac id identi ty, and v i r tua l ly ident ical genomic organizat ion. S imi la r to humans (198), expression of mouse hacel m R N A was detected i n essentially a l l tissues tested by N o r t h e r n blot analysis (Figure B - 1 A , r ight panel). Hacel m R N A was also expressed throughout embryonic development (Figure B - 1 A , left panel). The hacel gene was d i s rup ted i n mur ine embryonic stem (ES) cells u s i n g a targeting vector i n w h i c h nucleotides encompassing exons 3 and 4 were deleted (Figure B-1B). The targeting construct was electroporated into E14 ES cells (129/Ola) . T w o G418-resistant cel l l ines heterozygous for the muta t ion at the hacel locus were used to generate chimeric mice, w h i c h were backcrossed onto a C 5 7 B L / 6 J background to obtain heterozygous hacel+/- mice. The intercross of hacel+/- mice p roduced h o m o z y g o u s hacel-/- mice, as conf i rmed by Southern blot analysis u s ing a 5' f l ank ing probe (Figure B-1C). Hacel m R N A transcripts were undetectable i n k idneys of hacel-/- mice and haceT/- embryonic fibroblasts by N o r t h e r n blot analyses u s ing a fu l l length probe (Figures B-1D). 183 //////// —<» <- Hace / p-actin 5° flanking probr short aim Ei3 j ' nankins probr froRY EcoRV Spel 5" flanking probe short arm J' nankin; probr long aim ^ Ei2 EcoRV Spel EcoRV ExJ I3 Wild t\ p t allele Targeting construct Ei5 =k Mutant allele Figure B-1: Generation of hacel Mutant Mice (A) Detection of hacel mRNA in different murine tissues and at different days of embryogenesis (E7-E17). Expression was detected with a 5' cDNA probe by Northern blotting. (B) Restriction map of the genomic hacellocus and construction of the pGK-Neo targeting vector. Exons, long and short arms, flanking probes, Neomycin (Neo) resistance cassette and restriction enzymes are indicated. (C) Southern blotting. Genomic D N A from mice heterozygous (+/-) and homozygous (-/-) for the hacel mutation was digested with Spel and analyzed by Southern blotting using the 5' flanking probe shown in B. Wi ld type (WT) and mutant (Neo) bands are indicated. (D) Northern blot of total R N A from kidney of 3 month old wi ld type (+/+)/ hacel +/-, and hacel-/-littermates using full length mouse hacel cDNA as a probe. 184 HaceT/- mice were born at the expected Mendelian frequency, were fertile, and appeared phenotypically healthy. Hacel-/- mice were indistinguishable from wild-type controls and displayed apparently normal morphologies of all organs analyzed. As hacel-/- mice aged, these mutant mice spontaneously developed a remarkable spectrum of tumours such as melanoma, hepatocellular carcinoma, spontaneous lung adenocarcinoma, angiosarcoma, mammary carcinomas, and lymphomas (Figure B-2). The first spontaneous tumours in hacel'/- mice were observed as early as 20 weeks of age and the incidence of spontaneous tumours reached ~12% in 2 year old mice (Table B-l). The incidence of spontaneous tumours in the hacel+/- and hacel+/+ littermate cohorts was ~l-2% at the comparable age and genetic background. These included one angiosarcoma, one hepatocellular carcinoma, and one lymphoma in the hacel+/- mice (Table B-l). 185 A Melanoma B Hepatocellular carcinoma C Lung adenocarcinoma D Angiosarcoma E Mammary carcinoma F Lymphoma 186 Figure B- 2: Spontaneous tumour formation in hacel mutant mice. (A-D) Photomicrographs of hematoxylin and eosin (H&E) stained sections showing tumours which developed spontaneously in aged (> 1 year) hacel-/- mice. Histological evaluation was done by Dr. Maureen O'Sullivan (A) Melanoma with the black arrow indicating intact epidermis and white arrow indicating an ulceration overlying the dermal tumour [X100]. Arrowhead in insert points to melanin pigment production by pleomorphic cells infiltrating the dermis [X400]. (B) Hepatocellular carcinoma with effacement of normal porto-lobular architecture [X40]. Insert [X400] shows carcinoma cells forming (white arrow) pseudoglandular structures. (C) Lung adenocarcinoma. Arrows show interface between adenocarcinoma (right) and lung parenchyma (left) [X40]. Insert shows haphazard arrangement of lung adenocarcinoma cells with mitotic activity (arrows). 35[X400]. (D) Angiosarcoma with formation of blood vessels (black arrow). Insert shows endothelial cell pleomorphism. (E) Mammary carcinoma, with arrow showing border between carcinoma (bottom) and normal breast tissue (top). Insert shows focal keratinization. (F) Lymphoma, with arrow showing starry sky appearance of lymphoma cells (right) expanding into normal pancreatic parenchyma (left). Lymphoma. [X40]. At 400X in the insert, a mitotic figure is circled. 187 Incidence of cancer has been observed to increase dramat ica l ly i n t umour suppressor mutant mice u p o n exposure to a second genetic or envi ronmenta l insult , as is seen w i t h pl6mK4a mutant mice (153, 333, 334). W e speculated that inact ivat ion of hacel migh t also render mice susceptible to a second envi ronmenta l a n d / o r genetic cancer trigger. W e therefore chal lenged hacel mutant and control littermate mice w i t h urethane, a D N A a lky la t ing agent that causes l u n g carcinomas f o l l o w i n g repeated adminis t ra t ion (334, 335). Consistent w i t h prev ious reports (336), a single challenge of urethane d i d not trigger cancer i n w i l d type littermates (Figure B-3). Surpr i s ing ly , the urethane challenge y ie lded on ly - 8 % tumour incidence i n hacel+1- mice. A l t h o u g h this was significantly h igher incidence than i n w i ld - t ype mice it was also considerably lower than i n hacel-/- mice. This sl ight increase i n t umour incidence m a y suggest an elevate sensit ivi ty to oncogenic stress i n hacel haploinsufficient mice. T u m o u r s i n hacel-/- mutant mice were observed as early as 3 months, a lmost 100% hacel mutant mice deve loped l u n g tumours (Figure B-3 A , B). His to log ica l ly , these l u n g tumours were classified as adenocarcinomas (Figure B-3C). D u r i n g the observat ion pe r iod metastatic spread of these p r i m a r y l u n g tumours into other organs was not observed. 188 hacel+/+ Figure B- 3: hacel Mutant Mice Are Susceptible to Urethane Induced Lung Cancer (A) Lung tumours in hacel-/- mice. Three typical cases of lung tumours in hacel-/- mice are shown six months following a single injection of urethane (1000 mg'kg of body weight; i.p.) at 4 weeks of age. All hacel+/+ littermates treated with the same procedure remained tumour free. (B) Lung cancer incidence data are summarized from 30 hacel-/-, 10 hacel+/+/ a n d 20 hacel +/- mice at 6 month after the single urethane injection. (C) Histology of lung tumours in hacel-/- mice. Upper panels: normal lung from a urethane-treated hacel+/+ mouse. Middle and bottom panels show lung adenocarcinomas in hacel-/- mice. H&E: Bars indicate 0.5mm (left panels) and 50um (right panels). 189 These f indings were also extended to a genetic "second h i t " m o d e l to assess whether loss of hacel expression exhibits co-operat ivi ty i n tumorigenesis w i t h other t umour suppressors. To do this the hacel muta t ion was crossed onto a p53 or p21 mutant background . N o significant statistical differences i n tumour development were detected i n hacel/p21 double mutant mice, compared w i t h hacel or p21 s ingle knockou t mice (data not shown). This suggested that t umour incidence i n hacel mutant mice does not appear to be inf luenced by the p21 expression status. C o -operat ivi ty between hacel and the tumour suppressor p53 was also assessed. Heterozygos i ty at the p53 locus d i d not appear to affect t umour development o n a hacel+l- background and hacel-1- p53+/+ showed an incidence of cancer i n older mice s imi lar to that described above (data not shown). A s expected, loss of both p53 alleles resulted i n h i g h incidence of tumour format ion (337, 338). In t r iguingly , loss of a single p53 allele o n a hacel-/- background increased overa l l t umour incidence, w h i l e inact ivat ion of both p53 and hacel alleles (hacel-/- p53-/-) m a r k e d l y increased tumour incidence i n younger mice (Figure B-4 and Table B-2). The spect rum of tumours deve loped appeared to be very diverse compared w i t h wha t has been reported i n p53 mutant mice and inc luded osteosarcoma, angiosarcoma, le iomyosarcoma, m a m m a r y carcinoma, kera t in is ing squamous cel l carcinoma, and others (Table B-2). H o w e v e r , it shou ld be noted that the genetic background of p53 mutant mice does tend to influence the spect rum of tumours that arise (338, 339) and these mice were not backcrossed onto a u n i f o r m genetic background but rather are a complex m i x of backgrounds 190 compr i s ing of C 5 7 B L / 6 J and 1 2 9 / O l a , f rom the hacel mutant mice, and C 5 7 B L / 6 J and 129S2, f rom the p53 mutant mice (B6.129S2- T rP 5 3 t a l Ty)7J). up to 4 month old up to 10 month old Genotypes and ages of mice Figure B- 4: Co-operativity Between p53 and hacel in Tumour Suppression Incidence of spontaneous, macroscopically visible tumours in mice carrying the indicated hacel and p53 genotypes at 4 months and 10 months of age. Numbers per group were: hacel +/-p53+/- n = 85; hacel-/-p53+/+ n = 29; hacel +/-p53-/- n = 11; hacel-/-p53+/- H P 69; hacel-/-p53-/- n = 24. 191 Table B-1: Spontaneous tumour development in hacel deficient mice hace (29 mice with tumours from a total of 252 mice analyzed, tumour incidence 12%) Mouse Sex Age (weeks) Type of Tumour B196 M 36 mammary carcinoma B390 M 70 Lymphoma B395 M 56 melanoma B491 M 96 hepatocellular carcinoma F6 F 105 Lymphoma F26 F 105 Melanoma F156 F 90 hepatocellular carcinoma F191 F 90 Adenocarcinoma of lung angiosarcoma splenic lymphoma F209 M 90 Lymphoma H6 M 78 lung adenocarcinoma H l l F 90 splenic lymphoma H20 M 82 hepatocellular carcinoma H41 M 90 hepatocellular carcinoma H49 F 81 Lymphoma H120 F 80 lymphoma H128 F 90 Angiosarcoma H143 F 95 histiocytic sarcoma H57 M 80 lung carcinoma H60 M 80 Sarcoma thymic lymphoma H76 M 90 hepatocellular carcinoma H190 F 20 fibrohistiocytic sarcoma H200 M 26 Myxosarcoma B527 F 66 soft tissue tumour F69 M 102 soft tissue tumour F126 M 96 soft tissue tumour G60 M 95 soft tissue tumour thymic lymphoma G76 M 100 soft tissue tumour H2 M 90 lung tumour H5 M 80 lung tumour 192 hacel+/- (4 mice with tumours from total 295 mice analyzed, tumour incidence 1.3%) Mouse Sex Age (weeks) Type of Tumour B84 F 90 A n g i o s a r c o m a G36 F 90 hepatocellular carcinoma, l y m p h o m a G71 M 90 L y m p h o m a B58 F 78 soft tissue tumour A l l tumours in black were detected macroscopically and analyzed by microscopy; those listed in green italics were analysed by weight and flow cytometry for lymphomas and by gross examination the remaining cases. Age indicates the age of mice when tumours where detected. 193 Table B- 2: Spontaneous tumour incidence in mice carrying mutant p53 and hacel alleles hacel-/- p53+/- (47 mice with tumours from total 69 mice analyzed, tumour incidence 68%) Mouse Sex Age (weeks) Type of Tumour B102 F 80 osteosarcoma thymic lymphoma B107 M 14 lymphoma B133 F 78 angiosarcoma lymphoma B145 M 44 leiomyosarcoma B148 M 56 osteosarcoma B169 M 28 pleomorphic sarcoma likely R M S histiocytic sarcoma B235 M 25 pleomorphic high grade sarcoma (possible MPNST) B237 F 74 pleomorphic sarcoma w / H P C features lymphoma B240 F 74 angiosarcoma lymphoma B246 M 85 myxosarcoma B265 F 36 osteosarcoma and leiomyosarcoma B268 M 80 angiosarcoma B269 M 60 angiosarcoma splenic lymphoma B270 F 60 osteosarcoma angiosarcoma B279 M 72 pleomorphic sarcoma likely R M S B281 F 24 leiomyosarcoma, lymphoma, M F H , and angiosarcoma thymic lymphoma B286 M 22 leiomyosarcoma B307 M 60 M F H B344 M 48 spindle cell sarcoma B347 F 56 angiosarcoma B348 M 50 M F H B351 M 56 angiosarcoma B359 M 12 sebaceous adenoma B378 M 21 poorly differentiated carcinoma B387 M 60 osteosarcoma B389 M 60 M F H B403 M 42 M F H B405 F 56 basal cell carcinoma B423 M 42 osteosarcoma B426 F 60 high grade lymphoma B427 F 20 poorly differentiated breast 1 9 4 adenocarcinoma B495 M 32 angiosarcoma breast carcinoma B501 F 65 M F H B509 M 30 lymphoma B518 M 28 pleomorphic spindle cell sarcoma likely R M S B523 M 20 liposarcoma B525 M 32 Fibrosarcoma BU6 F 60 Sarcoma B118 M 16 Sarcoma B132 M 12 colon tumour B159 M 75 lung tumour B263 M 80 Sarcoma B340 F 60 Melanoma B379 M 50 Sarcoma B381 M 48 Sarcoma B406 F 60 Sarcoma B470 M 52 Sarcoma hacel*/- p53-/- (11 mice with tumours from total 14 mice analyzed, tumour incidence 79%) Mouse Sex Age (weeks) Type of Tumour B126 M 36 Lymphoma B143 F 78 Lymphoma B257 F 24 Lymphoma B276 M 20 high grade lymphoma B392 F 16 Lymphoma B416 M 20 Lymphoma B98 F 16 soft tissue tumour thymic lymphoma B108 M 16 thymic lymphoma B153 M 58 thymic lymphoma B278 F 12 thymic lymphoma B326 M 37 thymic lymphoma 195 hacel/- p53-/- (24 mice with turnouts from total 24 mice analyzed, tumour incidence 100%) Mouse Sex Age (weeks) Type of Tumour B161 M 24 Angiosarcoma thymic lymphoma B162 M 22 angiosarcoma and leiomyosarcoma lymphoma B222 F 11 pleomorphic RMS high grade lymphoma (Burkitt-like) B271 F 12 high grade lymphoma B275 M 23 Angiosarcoma B284 M 20 Lymphoma B288 M 21 lung adenoma B316 F 21 Lymphoma B332 M 20 Angiosarcoma lymphoma B338 F 20 high grade lymphoma B358 M 12 thymic lymphoma B360 M 12 sebaceous adenoma B361 M 19 infiltrating carcinoma with keratinisation lymphoma B391 M 20 high grade lymphoma B393 F 20 thymic lymphoma B394 M 14 Angiosarcoma B445 F 20 Lymphoma B447 M 20 angiosarcoma B473 F 18 thymic lymphoma B285 F 22 soft tissue tumour thymic lymphoma B362 M 19 soft tissue tumour thymic lymphoma B366 M 19 soft tissue tumour thymic lymphoma B414 M 8 soft tissue tumour B451 M 19 soft tissue tumour A l l tumours in black were analyzed by microscopy; those listed in green italics were analysed by weight and flow cytometry for lymphomas and by gross examination for the remaining cases. Age indicates the age of mice when tumours where detected. Abbreviations: RMS, rhabdomyosarcoma; MPNST, malignant peripheral nerve sheath tumour; HPC, hemangiopericytoma; M F H , malignant fibrous histiocytoma. 196 APPENDIX C Animal Care Certificate 197 UBC THE UNIVERSITY OF BRITISH COLUMBIA ANIMAL CARE CERTIFICATE Application Number: A04-0282 Investigator or Course Director: Poul H . B . Sorensen Department: Pathology & Laboratory Medicine Animals: M i c e 73 Start Date: July 1,2003 Approval Q c t o h e r 5 > 2 Q Q 5 Date: Funding Sources: Grant Agency: Grant Title: Nat iona l Cancer Institute of C a n a d a A l t e r e d expression of a nove l E3 ub iqu i t i n pro te in ligase, W T - A N K , In sporadic W i l m s 1 t umour Grant Agency: Grant Title: Grant Agency: Grant Title: N a t i o n a l Cancer Institute of C a n a d a A l t e r e d Express ion of a nove l E3 U b i q u i t i n pro te in ligase gene, W T -A n k , i n sporadic W i l m s 1 t umour N a t i o n a l Cancer Institute of C a n a d a Studies into the tumour suppressor act ivi ty of the H A C E 1 E3 ubiqui t in -pro te in ligase 198 Unfunded X T / . . . . . N / A title: The Animal Care Committee has examined and approved the use of animals for the above experimental project. This certificate is valid for one year from the above start or approval date (whichever is later) provided there is no change in the experimental procedures. Annual review is required by the C C A C and some granting agencies. A copy of this certificate must be displayed in your animal facility. Office of Research Services and Administration 102, 6190 Agronomy Road, Vancouver, B C V6T 1Z3 Phone: 604-827-5111 Fax: 604-822-5093 199 A P P E N D I X D Publication Angles io , M . S . , E v d o k i m o v a , V . , M e l n y k , N . , Z h a n g , L . , Fernandez, C . V . , G r u n d y , P .E . , Leach, S., M a r r a , M . A . , B rooks -Wi l son , A . R . , Penninger , J. and Sorensen, P . H . (2004) Differential expression of a novel ankyrin containing E3 ubiquitin-protein ligase, Hacel , in sporadic Wilms' tumour versus normal kidney. H u m M o l Genet, 13, 2061-74. 200 Human Molecular Genetics, 2004, Vol. 13, No. 18 2061-2074 doi:10.1093/hmg/ddh215 Advance Access published on July 14, 2004 Differential expression of a novel ankyrin containing E3 ubiquitin-protein ligase, Hacel, in sporadic Wilms' tumor versus normal kidney Michael S. Anglesio 1 , 2 , Valentina Evdokimova 1' 2, Nataliya Melnyk1'2, Liyong Zhang 3 ' 4 , Conrad V. Fernandez5, Paul E. Grundy 6, Stephen Leach 7, Marco A. Marra7, Angela R. Brooks-Wilson 7, Josef Penninger8 and Poul H.B. Sorensen 1 ' 2 * department of Pathology and department of Pediatrics, British Columbia Research Institute for Children's and Women's Health, University of British Columbia, Vancouver, BC, Canada V5Z 4H4, department of Medical Biophysics and department of Immunology, University of Toronto, Toronto, Ontario, Canada, department of Pediatrics, IWK Grace Health Centre, Halifax, Nova Scotia, Canada B3L 3G9, 6 Cross Cancer Institute, Edmonton AB, Canada T6G 1Z2, 7Genome Sciences Centre, British Columbia Cancer Agency, Vancouver, BC, Canada V5Z 4S6 and institute for Molecular Biotechnology of the Austrian Academy of Sciences, 1030 Vienna, Austria Received May 7, 2004; Revised and Accepted July 6, 2004 We have analyzed the chromosome 6q21 breakpoint of a non-constitutional t(6;15)(q21 ;q21) rearrangement in sporadic Wilms' tumor. This identified a novel gene encoding a protein with six N-terminal ankyrin repeats linked to a C-terminal HECT ubiquitin-protein ligase domain. We therefore designated this gene HACE1 (HECT domain and Ankyrin repeat Containing E3 ubiquitin-protein ligase 1). HACE1 is widely expressed in human tissues, including mature and fetal kidney. We show that Hacel protein possesses intrinsic ubiquitin ligase activity, utilizes UbcH7 as a candidate partner E2 enzyme and localizes predominantly to the endoplas-mic reticulum. Although the HACEI locus was not directly interrupted by the translocation in the index Wilms' case, its expression was markedly lower in tumor tissue compared with adjacent normal kidney. Moreover, HACE1 expression was virtually undetectable in the SK-NEP-1 Wilms' tumor cell line and in four of five additional primary Wilms' tumor cases compared with patient-matched normal kidney. We found no evidence of HACE1 mutations or deletions, but hypermethylation of two upstream CpG islands correlates with low HACE1 expression in tumor samples. Our findings implicate Hacel as a novel ubiquitin-protein ligase and demonstrate that its expression is very low in primary Wilms' tumors. INTRODUCTION Analysis of chromosomal translocations in human malig-nancies has led to the characterization of numerous genes involved in oncogenesis. For example, numerous transloca-tions have been found to either activate proto-oncogenes or to generate gene fusions encoding dominantly acting chimeric oncoproteins (1). Less commonly, mapping of translocations has highlighted the positions of novel tumor suppressor genes (TSGs) within deletion hotspots of tumors, including the von Hippel Lindau (VHL) gene of renal carcinoma (2), the NF2 gene in neurofibromatosis type 2 (3), the hSNF5/INIl gene in malignant rhabdoid tumors (4,5) and the BCSC-1 locus in breast and other cancers (6). We recently identified a balanced non-constitutional t(6;15)(q21;q21) translocation in a sporadic Wilms' tumor occurring in a 5-month-old male (7). Although Wilms' tumor is the most common renal neoplasm in children, accounting for ~90% of pediatric kidney tumors and 6% of all childhood cancers, the genetics of sporadic Wilms' tumor remain largely unknown (8). Rearrangements of the 6q21 region have been previously reported in this tumor, including t(5;6)(q21;q21) and t(2;6)(q35;q21) translocations (9-11). L O H of this region appears to be rare in Wilms' tumor (9), *To whom correspondence should be addressed. Tel: Human Molecular Genetics, Vol. 13, No. 18 © Oxford University Press 2004; all rights reserved 201 2062 Human Molecular Genetics, 2004, Vol. 13, No. 18 and there is currently little evidence for a 6q21 Wilms' tumor suppressor locus. However, deletions of 6q21 have been widely reported in human malignancies, including carcinomas of the breast, ovary and prostate, as well as in leukemias and lymphomas (12-15), and this region is hypothesized to harbor one or more TSGs (16). These observations highlight chromo-some 6q21 as a region frequently targeted for genomic altera-tions in human cancer. We therefore postulated that the t(6;15) translocation of the index case might be contributing to Wilms' tumor oncogenesis, either through oncogene acti-vation or by targeting a 6q21 TSGs. This prompted us to search for the 6q21 gene or genes potentially altered by the t(6;15) rearrangement. In this study, we have characterized the 6q21 region inter-rupted by the Wilms' tumor t(6;15) translocation. We report that the 6q21 breakpoint maps to a non-coding region with >200 kb from the nearest known gene, but ~50 kb upstream of a novel open reading frame (ORF). This ORF encodes a protein with a previously unreported domain architecture con-sisting of six ankyrin repeats linked to a HECT (homologous to E6-AP C-terminus) domain. HECT domains have thus far only been described in E3 ubiquitin-protein ligases (17,18). E3 ligases are essential components of a highly conserved pathway involving conjugation of one or more ubiquitin (Ub) polypeptides to specific substrate proteins, leading in most cases to substrate proteasomal degradation (17,19). We therefore designated this gene HACE1 (HECT domain and Ankyrin repeat Containing E3 ubiquitin-protein ligase 1). Hacel protein has in vitro and in vivo ubiquitin ligase activity and represents, to our knowledge, the first documented example of a HECT domain containing ubiquitin ligase pos-sessing N-terminal ankyrin repeats. We found that although HACE1 is ubiquitously expressed in normal tissues, including both adult and fetal kidneys, mRNA and protein levels were almost undetectable in the index case and in four of five insti-tutional Wilms' tumor cases compared with matching adjacent normal kidney, and in the SK-NEP-1 Wilms' tumor cell line. Lower expression was strongly associated with hypermethyla-tion of two CpG islands located upstream of the HACE1 locus. Our findings implicate Hacel as a novel HECT E3 ubiquitin-protein ligase whose lack of expression is associated with sporadic Wilms' tumor. RESULTS Mapping the 6q21 breakpoint in Wilms' tumor To assess candidate genes potentially affected by the t(6;15)(q21;q21) translocation in the index case, we focused on the 6q21 breakpoint as this region has previously been implicated in Wilms' tumors and other human malignancies. Using 6q21 bacterial artificial chromosomes (BACs) for fluor-escence in situ hybridization (FISH), we identified several overlapping BACs that spanned the translocation breakpoint (i.e. that showed three signals as opposed to the normal two; Fig. 1A). On the basis of sequence information from NCBI (http://www.ncbi.nlm.nih.gov) and UCSC (http://genome. ucsc.edu) genome databases, we assembled a precise map of the involved 6q21 region and narrowed the breakpoint to an ~12kb region between the D6S2097 and STSG6084 polymorphic markers (Fig. IC). We next determined the restriction enzyme distribution within this region from pub-licly available genomic sequences. Southern blotting using genomic DNA from the index case and the SB1 probe (Fig. IC) allowed us to confirm the position of the 6q21 breakpoint, as rearrangements were detected in the tumor sample but not in matched normal tissue or control samples (shown for Pstl digests in Fig. IB). Identification of the HACE1 gene Using public database sequences, we next inspected the region surrounding the 6q21 breakpoint for candidate genes poten-tially affected by the rearrangement. No previously character-ized genes were present in this region. We therefore used GenScan gene prediction software to search for novel ORFs. No coding sequences were identified spanning the breakpoint, indicating that the translocation does not directly disrupt a gene in the 6q21 region. However, a novel 2727 bp ORF located ~50 kb downstream was identified by this strategy, a portion of which matched to the 5' end of a non-annotated, IMAGE Consortium expressed sequence tag (EST; GenBank accession no. BC034982). We obtained this cDNA clone and sequenced it in its entirety, revealing 100% sequence iden-tity with the predicted full-length ORF from 6q21. Alignment to publicly available 6q21 genomic sequences using the UCSC genome browser B L A T tool predicts that this ORF is orga-nized into 24 exons (Fig. IC). The nearest known gene, BVES (blood vessel epicardial substance), lies ~200 kb upstream of the breakpoint (Fig. IC). Protein domain analysis predicted that the 6q21 ORF encodes a 909 amino acid protein (~ 103 kDa) possessing six N-terminal ankyrin repeats and a C-terminal HECT domain (Fig. ID). Ankyrin repeats are well-documented to mediate protein-protein interactions (20), whereas HECT is a catalytic domain possessing ubiquitin-protein ligase activity (17,18,21). We therefore designated this gene HACEL Com-parison of the Hacel protein sequence to public databases using NCBI BLAST and DART tools (http://www.ncbi.nlm. nih/BLAST) highlights Hacel as the first documented example of an E3 ligase in which a HECT domain is linked to ankyrin repeats. Interestingly, when the HACE1 coding sequence was aligned with those of other genomes using the B L A T tool at the UCSC genome database (http://genome. ucsc.edu), orthologues of HACE1 appear to be present only in vertebrate species (data not shown). Hacel ankyrin repeats show >47% sequence similarity with those of the cyclin-dependent kinase inhibitors pl6INK4A and pl9INK4D (Fig. ID), both of which are known to be inacti-vated in human malignancies including Wilms' tumor (22). The Hacel HECT domain is also highly conserved, with more than 53% sequence similarity to the HECT domains of well-characterized ubiquitin ligases including E6-AP, which destabilizes p53 in HPV infected cervical carcinoma cells (23), Nedd4, which regulates stability of the epithelial Na+ channel (ENaC) and insulin-like growth factor 1 receptor (21), and Smurf 1 and 2, which are involved in ubiquitination of TGF -3 receptors (24). HECT family E3 ligases are known to contain an invariant C-terminal cysteine residue necessary 202 Human Molecular Genetics, 2004, Vol. 13, No. 18 2063 B ,T N C, C II II I Chromosome 6 1 1 1 1 1 il 1 111D22 809N1S CpG-29 CpG-177 II I I t • IB • II I I CpG-88 HACEI Oaa 300 600 900 Cys876 Ankyrin Repeats p19 I N K*°(51%) —' i ^ — P 1 6 I N K 4 A (47%) NEDD4 (64%) SMURF1/SMURF2 (59%) E6-AP (53%) Figure 1. Identification of a novel gene, HACEI, within the t(6;15)(q21;q21) breakpoint region in the Wilms' tumor index case. (A) FISH using 6q21 BAC probe 809N15 (green) and a telomeric probe for 6q (red). Normal Chr 6 (white arrow), derivative Chr 15 (red arrow) and derivative Chr 6 (yellow arrow) are indicated. (B) Southern blotting using SB 1 probe indicated in (B) demonstrates re-arrangement in Pst I digested genomic DNA from the index tumor (T) case but not in peripheral blood (N) or control samples (C] and C2). (C) Mapping of the 6q21 breakpoint region. Twenty-four exon structure of HACEI as predicted by the UCSC genome browser BLAT tool is shown. BACs used for mapping of the breakpoint region were identified using Finger printed contigs (FPC) software and sequence information from the NCBI and UCSC genome databases and are indicated as follows: BACs spanning the t(6;15)(q21;q21) breakpoint (double green lines); BACs that remain on the derivative Chr 6 (red single lines); BACs detected on the derivative 15 (blue single line), green boxes denote CpG islands. (D) Hacel (GenBank accession number AAH34982) consists of six N-terminal ankyrin repeats and a C-terminal HECT domain as predicted by Pfam and SMART database tools. An invariant Cys residue in the HECT domain is indicated by asterisk. Alignment to a number of proteins containing ankyrin repeats or HECT domains are shown below; hashed region corresponds to the region of significant similarity. Percentage similarity was calculated using BLAST tools. 203 2064 Human Molecular Genetics, 2004, Vol. 13, No. 18 for thioester bond formation with ubiquitin (18); in Hacel this appears to be represented by Cys-876 (Fig. ID). Normal HACE1 expression patterns We next examined HACE1 expression profiles in normal human tissues by northern analysis. Using a full-length cDNA probe, we found that human HACE1 is expressed as a single mRNA species of ~4.6 kb in multiple tissues, includ-ing strong expression in heart, brain and kidney (Fig. 2A). As Wilms' tumor is hypothesized to derive from early metane-phrogenic stem cells (25-28), we also assessed HACE1 expression in three (day 54-122) human fetal kidney tissue samples. We found that, although variable, mRNA transcripts are expressed in these cells at levels that are similar to or greater than in mature pediatric kidney (Fig. 2B). There was no obvious trend in expression patterns with the age of the human fetal kidney samples, although this requires a larger sample size for rigorous evaluation. To examine expression of Hacel protein, we generated polyclonal a-Hacel antibodies directed towards either the second ankyrin repeat or to full-length recombinant protein. Both antibodies detect a protein with the expected size of 103 kDa identical to that of recom-binant Hacel in both fetal and pediatric kidney cells, as well as in the HEK293 human embryonic kidney cell line (Fig. 2C). Hacel possesses ubiquitin ligase activity in vitro and in vivo To determine whether Hacel possesses a ubiquitin ligase activity, we tested the ability of this protein to form a thioester bond with ubiquitin, a well-documented characteristic of HECT E3 ligases (17). In vitro [35S]-labeled Hacel was incubated in the presence of G S T - U b and E l Ub-activating enzyme, along with a panel of E2 Ub-conjugating enzymes (Fig. 3 A). As HECT domains contain a conserved Cys residue which is crucial for Ub transfer to substrate proteins, we mutated Cys-876 to Ser and tested this Hacel-C876S mutant in thioester bond formation. As seen in Figure 3A (top panel), an additional higher molecular mass band corresponding to the expected size of a Hacel - G S T - U b conjugate (~ 140 kDa) was evident in the presence of the UbcH7 E2 enzyme (lane 8). Thioester bond for-mation was completely abolished by the C876S substitution in Hacel (Fig. 3A, bottom panel) or by disulphide bond reducing agent 3-mercaptoethanol (Fig. 3B; compare lanes 1 and 4), indicating specificity of the reaction. These data confirm Hacel ubiquitin ligase activity in vitro and identify UbcH7 as a candidate partner E2. To demonstrate the involvement of Hacel in the ubiqui-tination of proteins in vivo, HA-tagged Hacel was stably expressed in NIH3T3 fibroblasts (Fig. 3C). Cytoplasmic extracts from these cells were then subjected to immunopreci-pitation with a-HA antibodies followed by western blotting with ot-Ub antibodies to assess whether Hacel associates with ubiquitinated proteins. As shown in Figure 3D, overall levels of protein ubiquitination were similar in HA-Hacel and vector alone cells. However, high molecular weight ubiquitinated proteins could only be detected in a-HA immunoprecipitates from the HA-Hace l expressing cells (compare lanes 3 and 4). The level of ubiquitinated proteins in these immunoprecipitates was increased in the presence of | M mm • •+HACE1 •+ACVN B 2 6 £ •o 3 i u • fk 1 fk2 flc3 ped. HEK kidney 293 fetal kidney samples kOa 119 51.4 4 fetal kidney samples « 1 fK2 fK3~ Hacel U actin Figure 2. Hacel expression in normal tissues. (A) Northern blotting demon-strating HACEI mRNA expression in a panel of normal tissues. Membrane was probed with the full-length Hacel cDNA. (B) qRT-PCR showing HACEI expression in three normal fetal kidney samples (fkl-3), pediatric (ped.) kidney and HEK293 (human embryonic kidney) are shown for compari-son. Expression of HACEI is normalized to the ped. kidney sample. (C) Western blotting showing Hacel protein expression fetal kidney samples [from (B)]. Actin serves as a loading control (lower panel). proteasome inhibitors such as lactacystin or MG132 (data not shown), indicating that at least some of the proteins tar-geted for ubiquitination by Hacel are normally degraded by the proteasome. Moreover, we found that Hacel directly inter-acts with the 26S proteasomal complex as Hacel could be immunoprecipitated using antibodies against the 20S core pro-teasomal subunits (Fig. 3E, lane 3). Although the identity of Hacel target proteins in the earlier mentioned immunoprecipi-tates remains to be established, our data strongly support the involvement of Hacel in ubiquitination and degradation of cellular proteins. Subcellular localization of Hacel To further characterize Hacel, we analyzed its localization within the cell. Subcellular fractionation followed by western blotting demonstrated that in exponentially growing NIH3T3 fibroblasts both endogenous and ectopically expressed Hacel are found predominantly in the endoplasmic reticulum (ER) and the cytoplasm, although a small amount of endogenous protein is also present in other fractions (Fig. 4A). ER localization was confirmed by co-immunostaining of 204 Human Molecular Genetics, 2004, Vol. 13, No. 18 2065 //////// Ub-Hace E2 E1 + + + + + + + + + Hacel 1194 Hacel~ Wild-Type C876S mutant 1 2 3 4 5 6 7 8 9 B 35S-Hace1: Wild-type P-me+ E2 + E1 + GST-Ub • Ub-Hace1 • Hacel C876S mutant 1 2 3 4 5 6 7 8 C D E Input IP:aHA Input IP:aHA IP Figure 3. Hacel exhibits E3 ubiquitin ligase activity in vitro and in vivo. (A) Thioester bond formation assay was performed using the [35S]-methionine labeled wild-type or C876S point-mutant Hace l proteins, GST-Ub, El activating enzyme and the respective E2s, as indicated. After 30 min at 30°C, the reactions were stopped with SDS-PAGE sample buffer without B-mercaptoethanol and subjected to SDS-7% PAGE and autoradiography. (B) The reactions were carried out as in (A) except that exclusively UbcH7 has been used as a source of E2. Reactions were stopped with or without B-mercaptoethanol. Note that addition of B-mercaptoethano! (lane l) or mutation of the invariant Cys residue (lane 8) completely abolished thioester formation. (C and D) Cytosolic extracts from NIH3T3 cells ectopically expressing HA-Hacel or vector alone were used for immunoprecipitation (IP) with a-HA antibodies. Of the IP fractions 20% were then subjected to SDS-7% PAGE and western blotting with a-HA (C) or a-Ub (D) antibodies. Input fraction lanes represent 5% of cytosolic cell extract used for IP. (E) Cytosolic extracts obtained from NIH3T3 cells expressing HA-Hacel were used for IP with either rabbit pre-immune, a-Ub, a-Hacel or a-20Sa and B proteasomal subunits antibodies, as indicated. Immu-noprecipitated proteins were then separated by SDS-7% PAGE and analyzed by western blotting using a-HA antibodies. Hacel with the ER resident, BiP (Grp78). As seen in Figure 4B, a predominantly perinuclear reticular staining pattern of the ER and of the nuclear envelop was observed for both Hacel and BiP in NIH3T3 cells using immunofluor-escence microscopy. This suggests that a significant portion of Hacel localizes to the ER. Altered expression of HACEI in sporadic Wilms' tumor As the 6q2l breakpoint in the index case occurred ~50 kb upstream of the HACEI locus, we hypothesized that HACEI expression might be affected by the t(6;15) translocation. More-over, we wondered whether altered HACEI expression might represent a recurrent molecular alteration in sporadic Wilms' tumor. To begin to address this possibility, we assessed HACEI mRNA levels in the index case and in five additional institutional Wilms' tumor cases with patient-matched adjacent normal kidney by quantitative R T - P C R (qRT-PCR). The his-tology of each tumor sample confirmed the presence of at least 80% tumor cells (data not shown). Of the six paired samples, five tumor specimens showed markedly lower HACEI mRNA expression compared with their matching normal kidney counterparts, including the index case (Fig. 5A). Although there was variability in the absolute degree to which HACEI expression was lower, the average decrease in transcript levels was ~ 5-fold. Low expression was confirmed at the protein level by western blotting using a-Hacel antibodies (Fig. 5B), as expression of Hacel protein was very low or undetectable in the index case and in the other four cases with low mRNA levels, compared with matching normal kidney. We next assessed HACEI expression levels and the presence of variant transcripts in the SK-NEP-1 Wilms' tumor cell line by northern analysis. Transcript levels are virtually undetectable in these cells compared with HEK293 human embryonic kidney cells or Ewing tumor and rhabdomyosarcoma cell lines (Fig. 5C). Hacel mRNA expression was also low to non-detectable in neuroblastoma cell lines. In agreement with these findings, Hacel protein levels are extremely low in SK-NEP-1 and neuro-blastoma cell lines compared with HEK293, Ewing tumor and rhabdomyosarcoma cells (Fig. 5D). Absence of HACEI mutations or deletions in sporadic Wilms' tumor Current views hold that Wilms' tumor most likely derives from embryonic kidney (26-28). As we observed strong 205 2066 Human Molecular Genetics, 2004, Vol. 13, No. 18 Cell fraction kDa 119 119 91 21.1 21.1 29.0 CO mm - h • mm* •*< — •< mm — • « - m» 1 2 3 4 UbcH7 B a-Hace1 a-BiP merge Figure 4. Hacel is mainly localized to the ER. (A) Subcellular fractionation of N1H3T3 cell extract. Equal amounts (50 u.g) of nuclear (N), mitochondrial (M), endoplasmic reticulum/polyribosome (ER) and cytosolic (S) fractions were analyzed by western blotting using the respective antibodies, as indicated. Histone H3, BiP and Grb2 were used as markers of nuclear, ER and cytosolic fractions, respectively. (B) Indirect immunofluorescence microscopy of NIH3T3 cells using a-Hacel (red) and BiP (green) antibodies, as indicated. Nuclear staining by propidium iodide is in blue. Hacel expression in both mature pediatric and fetal kidneys but not in most of the Wilms' tumors analyzed, one possibility is that Hacel down regulation or inactivation is a common and potentially etiologic event in this disease. However, it is equally plausible that the specific (as yet unknown) blastemal cell giving rise to Wilms' tumor does not express HACEI and that the observed lack of expression simply reflects the normal expression profile of the putative neoplastic precursor cell. We therefore assessed the Wilms' tumor cohort for potential genetic mechanisms of HACEI down regulation as a strategy for distinguishing between these possibilities, as loss of TSGs in malignancies typically involves inactivation of both the alleles (29). Current literature indicates that chromosomal rearrangements, large scale deletions or L O H of the 6q21 region are rare in Wilms' tumor (8-11). Unfortunately, aside from the index case, karyotypes were not available for the Wilms' tumors analyzed in this study. To search for alternative mechanisms that might underlie altered HACEI expression in Wilms' tumor, we screened the cohort of cases and matched normal kidney as well as the SK-NEP-1 cell line for HACEI mutations. All exons and intron-exon bound-ary regions were sequenced from genomic DNA of each sample. No sequence discrepancies between tumor and normal DNAs were found for any of the six cases. Further-more, no mutations were found in the SK-NEP-1 cell line. Sequence data were compared with the UCSC human genome database and a total of eight HACEI polymorphisms were identified in the earlier mentioned cohort (Supplementary Material, Table SI). Two of the polymorphisms were within the coding sequences, but only one represents an amino acid change. However, the latter (Asp to Gly at amino acid 399) does not affect any identifiable functional domains within Hacel protein, and when a non-disease control cohort of 95 individuals was tested for this polymorphism it was present in 1% of samples. Therefore, it most likely represents a low-frequency polymorphism in the general population. Of the five Wilms' tumor cases with low HACEI expression, at least one heterozygous polymorphism was found in four of the tumor (and normal) DNAs as well as in the SK-NEP-1 cell line, suggesting that L O H of the HACEI locus is unlikely in this cohort. However, our sequencing strategy cannot rule out micro-deletions of exons or other genomic sequences located between these polymorphisms. Therefore, our data indicate that a genetic mechanism involving point mutations or deletions is unlikely to explain the low HACEI expression observed in Wilms' tumor, although larger numbers of cases must be assessed to rigorously test this assumption. Upstream C p G island methylation is associated with low HACEI expression Recent studies indicate that promoter hypermethylation may be as frequent as inactivating mutations for reducing TSG expression in human malignancies (30). Methylation involves the addition of methyl groups to cytosine residues within CpG islands within coding or non-coding sequences of genetic loci, generally resulting in gene silencing (31,32). We therefore investigated the methylation status of CpG islands at the HACEI locus in the primary Wilms' tumor cohort and in the SK-NEP-1 cell line. There are three CpG islands located within or near the HACEI locus, one at the promoter extend-ing into exon 1 which contains 88 C G dinucleotides (CpG-88), and two located ~50 kb upstream of the Hacel transcriptional start site containing 177 (CpG-177) and 29 (CpG-29) CG dinucleotides repeats, respectively (Fig. IC). To assess whether altered methylation might influence HACEI expression, we analyzed methylation status at each of these CpG islands by digestion with the methylation sensitive restriction enzymes Haell, Hpall or Bssllll, followed by semi-quantitative PCR (Fig. 6A). No evidence of methylation at CpG-88 was found in any of the tumor or normal kidney samples (Fig. 6A). However, hypermethylation of either CpG-177 and CpG-29, or both, was observed in four out of five Wilms' tumors with low HACEI mRNA expression (Fig. 6A) as well as in SK-NEP-1 cells (Fig. 6B). It was never observed in any of the normal kidney samples (data not shown). If hypermethylation of CpG-29/CpG-177 is mechanistically related to low HACEI expression, then pharmacological demethylation of these islands would be expected to restore 206 Human Molecular Genetics, 2004, Vol. 13, No. 18 2067 relative expression of HACE1 (mRNA) T1 T2 T3 T4 T5 4s # D 0 0 9 * • primary Wilms' tumour samples B kDa Index 1 2 3 4 5 Z I - Z t - Z H Z I - Z I - Z K £ ^ / / / / / / ACTIN kDa ^ <Sr" # C> # 4 * Hacel Grb2 1 2 3 4 5 6 7 8 9 10 Figure 5. Hacel expression in Wilms' tumors and pediatric cancer cell lines. (A) qRT-PCR showing expression of HACEI mRNA in Wilms' tumor samples relative to patient-matched normal kidney. HACEI expression relative to a normal pediatric kidney sample is shown for SK-NEP-1 and HEK293 (human embryonic kidney) cells. (B) Western blotting for the Hacel protein showing the tumor (T) samples [from (A)] patient-matched normal kidney (N). (C) Northern blotting demonstrating HACEI mRNA expression in a panel of pediatric tumor cell lines including neuroblastomas (SAN-2, IMR-32, SK-N-SH), Ewing's sarcoma (SK-N-MC), rhabdomyosarcomas (Birch, CT-10), Wilms' tumor (SK-NEP-1) and HEK293 cells, Membranes were probed with the full-length Hacel cDNA. (D) Western blotting showing Hacel protein expression in pediatric tumor cell lines [from (C), and an additional neuroblastoma line, KCNR]. Westerns were done using the N-terminal a-Hacel antibody. Recombinant Hacel protein (lane 10) is used to confirm the specificity of the antibody. Grb2 serves as a loading control. expression levels. As SK-NEP-1 Wilms' tumor cells express extremely low HACEI levels and both CpG-29 and CpG-177 are methylated in this line, we treated cells with the methylation inhibitor 5-aza-2-deoxycytidine (5AZ) (33,34) and assessed effects on HACEI expression. As shown in Fig. 6B, treatment with this agent reduced methylation at both CpG-29 and CpG-177, particularly at 5 |XM 5AZ. More-over, this was associated with at least a 4-fold increase in HACEI mRNA expression in the presence of 5 U.M 5AZ (Fig. 6C). These findings strongly suggest that methylation status of CpG-29 and CpG-177 influences HACEI expression. To examine the functional consequences of this observation, we next performed chromatin immunoprecipitation (ChIP) to analyze whether the HACEI locus exists in an active or inactive chromatin conformation in SK-NEP-1 cells. This was assessed by comparing the relative levels of acetylated-histone H3 versus dimethyl-(Lys79)-histone H3 bound to the HACEI promoter. Acetylated-histone H3 is known to asso-ciate with DNA of active chromatin, whereas dimethyl-(Lys79)-histone H3 correlates with chromatin present in the inactive conformation of heterochromatin or silenced genes (35-37). As shown in Figure 6D, the promoter region of the HACEI gene exists predominantly in an inactive chroma-tin conformation in normally growing SK-NEP-1 cells (i.e. associated with dimethyl-(Lys79)-histone H3 by ChIP). However, treatment with 5 u-M 5AZ reverses this pattern, switching the HACEI promoter to an active chromatin confor-mation associated predominantly with the acetylated-histone H3. As a control, we performed identical studies with HEK293 cells (Fig. 6D, left panel) which abundantly express HACEI (Fig. 4). Consistent with its high expression, the HACEI locus exists in an active chromatin organization in these cells, and no change in chromatin structure is evident after treatment with 5AZ. Taken together, our data are consistent with a role for CpG-29 and CpG-177 hyper-methylations in influencing HACEI expression, although additional studies are required to confirm this possibility and whether it represents an acquired alteration in tumor cells versus the normal pattern of methylation in the Wilms' tumor neoplastic precursor cell. D I S C U S S I O N In this study, we demonstrate that a previously uncharacteri-zed E3 ubiquitin-protein ligase gene is located ~50 kb down-stream of the 6q21 breakpoint of a t(6;15) translocation in sporadic Wilms' tumor. This gene, which we designated HACEI, encodes a 103 kDa protein containing six N-terminal ankyrin repeats connected via a linker region to a C-terminal HECT domain. The latter has to date only been described in E3 ligases (17), implicating the Hacel protein as a new member of the HECT family of E3 ubiquitin-protein ligases. The protein ubiquitination process involves a highly conserved pathway in which one or more Ub polypeptides become con-jugated to specific substrate proteins (17,19). This multistep process involves a cascade of three different classes of enzymes. An ATP-dependent Ub-activating E l enzyme first forms a thioester bond with Ub, and then transfers the activated Ub moiety to one of a number of different E2 Ub-conjugating (Ubc) enzymes. Finally, E3 ubiquitin ligases catalyze the transfer of Ub from the cognate E2 to a lysine on the substrate protein. Binding of substrate proteins appears to be mediated by the N-terminal protein-protein interaction domains of E3 ligases (17,18,21). 207 2068 Human Molecular Genetics, 2004, Vol. 13, No. 18 lrrrn—rr I i i 1 1—'i i i i f f CpG-88 Hpall T CpG-29 Hpall li —i 'i 'in i—i i 'i i r CpG-177 Bum HaeW M1C2 5' region CpCi-88 B tumour normal tumour normal J / 4t *t f / 4 T > / relative expression of HAC£ J i 3 € 2 EcoRI + Hae 1 1 • c DMSO luM 5uM 5AZ 5AZ SK-NEP-1 cells HEK293 SK-NEP-1 ChIP: acetyl-histone H3 ChIP: methyl-histone H3 input fraction Figure 6. Methylation analysis of the HACEI gene in Wilms' tumor patient's samples and SK-NEP-1 cell line. (A) Representative sample: Wilms' tumor and patient-matched normal kidney DNA was digested using EcoR I in combination with HaeW, Hpall or BssHl methylation sensitive restriction enzymes (upper panel illustrates restriction map of each PCR amplicon), methylation of three CpG islands surrounding the HACEI locus was then assessed by semi-quatitative PCR. Note that hypermethylation was restricted exclusively to tumor samples in CpG-29 and CpG-177 islands, no methylation was observed at CpG-88. (B) SK-NEP-1 Wilms' tumor cells were treated with 5AZ to inhibit methylation. Semi-quantitative PCR was then used to assess methylation status of CpG islands associated with HACEI as described in (A). After treatment with 5AZ, methylation of CpG-29 and CpG-177 islands is decreased, whereas there is no evidence of methylation at CpG-88. Similar results were observed in three independent experiments. (C) qRT-PCR showing an increase in HACEI mRNA transcripts upon treatment of SK-NEP-1 cells with increasing doses of 5AZ. A maximal response was observed at 5 u.M 5AZ. (D) ChIP assays to detect active or inactive chromatin were performed using antibodies against either acetyl-histone H3 (acetyl H3) or dimethyl-(Lys79)-histone H3 (methyl H3), respectively. SK-NEP-1 and HEK293 cells treated with or without 5 u.M 5AZ prior to ChIP studies. PCR products in the top two panels for each cell line correspond to the transcriptional start of HACEI and is overlapped by CpG-88. Identical results were observed in four independent experiments. HACEI is widely expressed in human tissues, including both fetal and mature kidneys. The Hacel protein forms thio-ester bonds with ubiquitin as expected for a HECT family E3 ligase (17,18), and is associated with high molecular weight ubiquitinated proteins within cells. Among a panel of E2 ubiquitin carrier proteins, thioester bond formation in vitro occurred in the presence the UbcH7 E2 enzyme, and weakly with UbcH6 and UbcH5b, all of which are common partners of other HECT domain containing E3 ubiquitin-protein ligases. We also found that Hacel associates with ubiquiti-nated proteins and components of the 26S proteasomal complex, indicating that at least some of the proteins targeted for ubiquitination by Hacel are likely to be degraded by the proteasome. Using cell fractionation and immunofluorescence microscopy, we found that Hacel localizes predominantly to the ER indicating its possible involvement in ER-associated protein degradation (ERAD). E R A D is a quality control process that selectively directs degradation of misfolded and aberrant proteins through ubiquitination within the ER followed by transport to the proteasome (17,38). More studies are necessary to demonstrate whether Hacel contri-butes to this process. Most HECT E3 ligases characterized to date contain N-terminal WW domains for substrate interaction and target-ing (17,18,21). In fact, the presence of ankyrin repeats within an E3 ligase as in Hacel has not been previously described. Given this modular structure, Hacel may represent a novel sub-family of HECT E3 ubiquitin-protein ligases. Human Molecular Genetics, 2004, Vol. 13, No. 18 2069 Ankyrin motifs are well known to mediate protein—protein interactions (20), and therefore likely serve the same function in Hacel. Alternatively, the ankyrin repeats may be necessary for Hacel to link with non-substrate proteins in higher order signaling complexes, a function that has been described previously for ankyrin motifs (20,39). Interestingly, the Hacel ankyrin repeats show high sequence similarity to those of the cyclin-dependent kinase inhibitors pl6INK4A (CDKN2A) and pl9INK4D (CDKN2D), which inhibit the cell cycle by binding to and inhibiting CDK4 and CDK6. HACEI expression was very low both at the mRNA and protein levels in five of the six sporadic Wilms' tumors com-pared with patient-matched normal kidney, including the index case. Moreover, expression of this gene is virtually undetectable in SK-NEP-1 Wilms' tumor cells. This raises the possibility that loss of HACEI expression may be a recur-rent alteration in Wilms' tumor, and that HACEI inactivation plays a role in the pathogenesis of Wilms' tumor. On the other hand, this may reflect the normal expression profile of HACEI in the Wilms' tumor neoplastic precursor cell; i.e. HACEI may not normally be expressed in the putative blastemal cell of origin of this disease. Typically, inactivation of TSGs in human malignancies involves targeting of both alleles by genetic or epigenetic mechanisms (29). Therefore, we searched for mechanistic evidence of HACEI inactivation in Wilms' tumor. As chromosomal translocations or other cyto-genetic abnormalities involving 6q21 have only infrequently been reported in Wilms' tumors (7,9-11), this is unlikely as a mechanism of HACEI loss. Moreover, in the index case of the present study the 6q21 breakpoint did not directly disrupt the HACEI gene, but occurred outside of the coding region. We also failed to detect HACEI mutations in any of the cases with low HACEI expression, and L O H of this gene was not evident by analysis of several informative markers identified within the HACEI locus. The latter is con-sistent with other studies indicating that L O H of 6q21 is rare in Wilms' tumor (9). Although further studies using larger case cohorts to definitively rule out that mutations or deletions of HACEI occur in sporadic Wilms' tumor, our results to date do not support a role for genetic mechanisms in HACEI loss in Wilms' tumor. Gene inactivation by methylation appears to be at least as frequent as inactivating mutations for disrupting TSG expression in sporadic tumors (30). For example, in sporadic breast and ovarian carcinomas, BRCA1 mutations are highly infrequent and epigenetic silencing by promoter methylation is the only apparent mechanism for loss of function in most cases (40). We therefore examined whether the CpG islands associated with the HACEI locus were differentially methyl-ated in tumor versus normal kidney samples. Although the HACEI promoter CpG island was not methylated, we detected hypermethylation of CpG-29 and CpG-177, located ~50 kb upstream of HACEI exon 1, in four of the five Wilms' tumors with reduced mRNA expression, and also in the SK-NEP-1 cell line. Moreover, treatment with the methyltrans-ferase inhibitor 5AZ not only blocked methylation of these islands but also restored HACEI mRNA expression in SK-NEP-1 cells. One obvious caveat of these experiments is that 5AZ as a pharmacological inhibitor has a global effect on CpG methylation. We therefore performed ChIP analysis of the HACEI promoter region using antibodies to acetylated-histone H3 or dimethyl-(Lys79)-histone H3, which associate with actively transcribed versus inactive, silenced chromatin, respectively (36,37). These experiments indicated that 5AZ treatment of SK-NEP-1 cells shifts the HACEI promoter from an inactive chromatin conformation to that of active chromatin, in keeping with the increased expression of this gene in response to 5AZ. In contrast, the HACEI promoter of HEK293 human embryonic kidney cells, which express high levels of HACEI, exists in an active conformation under normal growth conditions and there is no change in chromatin structure upon treatment with 5AZ. These findings provide compelling evidence that HACEI expression is influ-enced by upstream CpG island methylation. It is important to note, however, that our findings do not establish whether the correlation between CpG island hypermethylation and low HACEI expression is specific to transformed cells. That is, they do not distinguish between acquired hypermethylation leading to HACEI inactivation as an etiologic event in Wilms' tumor oncogenesis, versus hypermethylation leading to low HA CE1 expression as part of the normal transcriptional regulation of this gene in the Wilms' tumor precursor cell. Functional studies are necessary to determine whether the Hacel protein has suppressive effects on cell growth, and whether its loss contributes to Wilms' tumor oncogenesis. Although methylation of CpG dinucleotides outside of pro-moters has been documented to influence gene transcription (41), most studies which describe epigenetic silencing of genes have focused on CpG islands located within promoters or coding regions of genes (30). Recently, distance effects on insulator methylation were reported for the H19 gene, whereby imprinting by CpG methylation at more distant sites regulated mono-allelic expression of both H19 and Igf2 (42,43). In addition, reduced expression of Sonic hedgehog (SHH) has been associated with chromosomal rearrangements occurring 15-250 kb from the SHH locus in holoprosence-phaly (44,45). It is possible that long distance effects of CpG island methylation on transcription may be a more general mechanism for gene silencing in human cancers. This might explain how chromosomal translocation break-points located at long distances from promoters of potential TSGs could influence gene silencing. It is well-documented that heterochromatin has a silencing effect on adjacent euchro-matin, likely through spreading of histone methylation (46-48). It is possible that by juxtaposing heterochromatin in the vicin-ity of potential TSGs, a chromosomal translocation can exert long distance transcriptional suppression on that locus by effecting local methylation changes. For the index case in this study, the 6q21 breakpoint mapped very closely to if not within CpG-29 or CpG-177. Moreover, both of these islands were methylated in tumor tissue but not in adjacent normal kidney tissue. Therefore, the direct effects of the t(6;15) in this case might have been to trigger methylation of these CpG islands, resulting in HACEI silencing. Alterna-tively, effects on methylation due to translocation breakpoints occurring in this region may not be limited to HACEI. It is possible that long distance effects on expression may also affect other genes in the region, and that targeting of these genes for activation or inactivation may actually be contri-buting to Wilms' tumor oncogenesis rather than HACEI. 209 2070 Human Molecular Genetics, 2004, Vol. 13, No. 18 However, the only known genes mapping to within ~500 kb of the breakpoint include the telomeric genes BVES, Popeye domain containing protein 3 (POPDC3) and prolyl endo-peptidase (PREP) (http://genome.ucsc.edu/). None of these genes have been implicated as oncogenes or TSGs, although this remains to be rigorously determined. No known genes are located within at least 1 Mb centromeric to the breakpoint. Interestingly, a recent reported case of Wilms' tumor with a 6q21 breakpoint maps to a 1.3 Mb region we now know includes the HACEI locus (9), suggesting that other rearrange-ments of this region may similarly occur near HACEI and adjacent loci in Wilms' tumor. The 6q21 region is commonly deleted in several other human neoplasms. Examples include malignancies of the prostate, breast and ovaries, as well as leukemias and lympho-mas (12-15). For example, a recent study found that ~50% of prostate cancers had deletions of the 6q21 region (15). A number of candidate metastasis suppressor genes have also been mapped to chromosome 6q21 (49). There is emerging interest in the role of protein degradation in neoplasia, and E3 ligases have been implicated in both tumor formation and suppression (50). In fact, the E6-AP HECT E3 ligase was originally defined by its ability to ubiquitinate and promote p53 degradation after recruitment by the E6 onco-protein in HPV infected cervical carcinoma cells (23,51,52). It will be important to further assess the possibility that the Hacel E3 ligase plays a role in human cancer. MATERIALS AND METHODS FISH and Southern blotting Single color FISH was performed on metaphase and inter-phase nuclei from the index Wilms' tumor case. BAC probes used for mapping were obtained from the RPCI-11 human B A C library, labeled with Spectrum Green or Spec-trum Red kits (Vysis) and hybridized to denatured slides as described previously (53). For Southern blotting, genomic DNA (~15 u,g) extracted from frozen tissue sections was first digested with respective restriction enzymes, and then analyzed by hybridization with [a- 3 2P]dATP labeled genomic probe generated by PCR from a human DNA tem-plate SB1 (primers: forward, G G A A A C A A A A G C A A A G C G A C C C A A C TAT; reverse, GGCGG C C G A G A C C T G A G ACC). Plasmids The HACEI coding region from the IMAGE cDNA clone no. 4838835 (ATCC; GenBank accession no. BC034982) was excised using Ascl and BsaAl, blunted with T4 DNA polymer-ase and ligated in frame into blunted Xhol site of the mamma-lian expression vector pcDNA3-HA (Invitrogen) or pET-15b (Novagen). The HA-tagged (hemagglutinin peptide tag, YPYDVPDYA) construct was excised from pcDNA3-HA with Sacll, blunted with T4 DNA polymerase, and subcloned into the Hpal site of p M S C V h y g r o (Clontech). To construct the plasmid encoding the Cys to Ser mutant (C876S), the 3' Xbal fragment of HACEI was subcloned from pcDNA3-HA-Hacel into the Xbal site of pBluescript-II. Site directed mutagenesis was done using the QuickChange kit (Stratagene) (primers: forward, T C A A G C A C A T C C A T C A A C A T G; reverse, C A T G T T G A T G GATGT GCTTGA). The mutation was sequence verified and the Xbal fragment was than replaced into the original pcDNA3-HA-Hacel construct. The N-term-inal Hacel fragment, p75, was generated by removal of a 3' Xbal fragment of HACEI from the pET-15b-Hacel. Cell lines NIH3T3 cells (ATCC) were grown in Dulbecco's modified Eagle's medium (DMEM) (Invitrogen) containing 10% calf serum (CS) (Invitrogen). SK-NEP-1, SK-N-SH, SK-N-MC, HEK293 and Bosc23 cells were grown in D M E M sup-plemented with 10% fetal bovine serum (FBS) (Invitrogen). KCNR, SAN-2, IMR-32, Birch and CT-10 were maintained in RPMI 1640 (Invitrogen) with 15% FBS. NIH3T3 cell lines stably expressing HA-Hacel were generated using a ret-roviral system as described previously (54). Briefly, ecotropic virus packaging Bosc23 cells were transfected with empty vector (pMSCV h y g r o ) or p M S C V h y g r o - H A - H a c e l plasmid using calcium phosphate. Retrovirus-containing supernatants were collected 48 h after transfection, filtered and incubated with NIH3T3 cells for 24 h. Infected cells were selected for 4-8 days with hygromycin B (600-800 u,g/ml) (Invitrogen). Methylation assays Methylation sensitive restriction enzyme digests followed by semi-quantitative PCR have been performed as described (55). Briefly, 100 ng of DNA, extracted from tumor and normal kidneys frozen tissue sections, was digested to com-pletion (overnight) with £coRI ± HaeW, EcoRI + BssUll or .EcoRI + Hpall (NEB) as indicated in Figure 6. Ten nano-grams of digested DNA was used for semi-quantitative PCR with amplicons surrounding CpG islands near the HACEI locus (Fig. 1). The 5' region of the M1C2 gene, previously shown to be unmethylated (56), was used as a control for com-plete digestion. All reactions were performed using Platinum Taq (Invitrogen) with the included buffer supplemented with 10% DMSO and carried out with the following conditions: CpG-29 (794 bp forward: G G A A A C A A A A G C A A A G C G A C C C A A C TAT, reverse: GGCGG C C G A G A C C T G A G A C C , 100 nM each), 50 U,M dNTPs, 1.0 mM MgCl 2 , 94°C 30 s, 56°C 10 s, 72°C 60s, 30 cycles. CpG-177 (804 bp, forward: TGTGC T G T T C G G A A T GATGT, reverse: C T A G C C T G G G T G T G A GAGGG, 100 nM each), 100 U,M dNTPs, 1.5 mM MgCl 2 , 94°C 45 s, 58°C 20 s, 72°C 120 s, five cycles, 94°C 30s, 56°C 10s, 72°C 90s, 30 cycles. CpG-88 (939 bp, forward: C C C C G A T G C A G C T T A A A G T A , reverse: G A G G G T A G G A G G A G C A G G G , 100 nM each), 50 u-M dNTPs, 1.0 mM MgCl 2 , 94°C 45 s, 58°C 10 s, 72°C 90s, 30 cycles. MIC2 (512 bp, forward: A G T A T CTGTC C T G C C GCC, reverse: T T T G C A A C T C C G A C A A C A A A CGC, 100 nM each), 50 u.M dNTPs, 1.0 mM MgCl 2 , 94°C 45 s, 55°C 15 s, 72°C 45 s, 35 cycles. The optimal number of cycles for each reaction was deter-mined empirically by testing a range of 25-40 cycles. Above conditions are those in which the most reproducible differences in amplification within the exponential phase of 210 Human Molecular Genetics, 2004, Vol. 13, No. 18 2071 amplification were achieved. All reactions were performed in triplicate. To assess the effect of CpG island methylation on HACEI expression efficiency, SK-NEP-1 cells were treated with methylation inhibitor 5-AZ (Sigma). Treatment was done twice for 24 h over a 7-day period (on days 2 and 5) as pre-viously described (57). DNA and RNA were isolated on day 6 and HACEI expression and methylation of the upstream CpG islands were assayed by qRT-PCR and semi-quantitative PCR, respectively. Each experiment was repeated four times. Quantitative RT-PCR Total RNA was extracted using Trizol reagent (Invitrogen) and converted to cDNA using 2 |xg RNA in a random primed synthesis with the Superscript II Reverse transcriptase kit (Invitrogen). qRT-PCR was done using the TaqMan 5' exonuclease assay. All primer/probe sets were designed to span exon boundaries to eliminate the risk of template contamination by genomic DNA and the need for DNasel pre-treatment of RNA. HACEI gene specific primer/probe set (forward: T C T T A C A G T T T G T T A CGGGC AGTT, probe: [6FAMJCAAAC C C A C C A T G T G G G A C C C T G [TAMRA], reverse: C A A T C C A C T T C C A C C CATGAT) was multiplexed with VIC-MGB labeled p-actin endogenous control primer/probe kit (Applied Biosystems) using TaqMan universal PCR master mix (Applied Biosystems) and standard conditions. An ABI 7000 sequence detection system (Applied Biosystems) was used to run the PCR reac-tions and measure fluorescence at each cycle. Generated data were further analyzed using Microsoft Excel. Each PCR reaction was performed in quadruplet and each sample was analyzed independently at least twice. Chromatin immunoprecipitation SK-NEP-1 or HEK293 cells were treated with or without 5 \XM 5AZ as described earlier. Immunoprecipitation of DNA repre-senting active versus inactive chromatin was performed using a ChIP assay kit (Upstate) according to the manufacturer's protocol. Briefly, DNA and protein were cross-linked in culture with 1% formaldehyde. For each reaction 1 x 106 cells were collected in 200 u.1 SDS lysis buffer (1% SDS, 10 mM EDTA, 50 mM Tris-HCl, pH 8.1, 1 mM PMSF, 1 u.g/ml aprotinin, 1 |xg/ml pepstatin A) and sonicated to shear genomic DNA to a size range of ~300-1000bp. Lysates were diluted 1:10 with ChIP buffer (0.01% SDS, 1.1% Triton X-100, 1.2 mM EDTA, 16.7 mM Tris-HCl, pH 8.1, 167 mM NaCl) and immunoprecipitated with the appropri-ate antibody overnight at 4°C with agitation. a-Acetyl-histone H3 (Lys9/Lysl4) antibodies (Upstate) were used to immuno-precipitate active chromatin fragments (58), whereas a-dimethyl-histone H3 (Lys79) antibodies (Upstate) were used to immunoprecipitate inactive chromatin (35) with protein A beads. Protein A agarose/Salmon sperm DNA slurry (Upstate) added to lysates without antibody was used as a negative control (data not shown). DNA was recovered by proteinase K digestion and phenol/chloroform extraction after washing and reversal of cross-links. Equal volumes were utilized to maintain the quantitative nature of the assay. The recovered DNA pellet was air dried and resus-pended in 25 u.1 of T E (10 mM Tris-HCl, 1 mM EDTA, pH 8.0). Immunoprecipitated HACEI levels were determined by using PCR amplifying the 5' end of the gene, overlapping the CpG-88 island. PCR was carried out with Platinum Taq (Invitrogen) with the included buffer, supplemented with 10% DMSO and the following conditions: 2.5 u.1 recovered ChIP DNA, primers (297 bp product, forward: CGGCT C A C C C T C G G G C A A C T CC, reverse: C G G C G GCGGG T G T A C T G T A G GTGGT C, 200 nM each), 100 U.M dNTPs, 2 mM MgCl 2 , 94°C 30 s, 55°C 30 s, 72°C 45 s, 32 cycles. Input DNA amounts were verified by PCR using the same con-ditions as mentioned using DNA extracted from a 100 JJLI aliquot of 'ChIP buffer' diluted cell lysate taken prior to addition of antibodies, as recommended by the ChIP assay kit (Upstate) (shown as 'input fraction' in Fig. 6D). Input was also verified with 2.5 |xl of recovered immunoprecipitated DNA using the a PCR assay for 5' end of the unmethylated MIC2 gene (55,56), as described earlier (data not shown). The optimal number of cycles for each reaction was deter-mined empirically by testing a range of 25-40 cycles. The conditions mentioned earlier are those in which the most reproducible differences in amplification within the exponen-tial phase of amplification were achieved. The entire ChIP experiment was repeated four times for each cell line. Sequence analysis Genomic DNA from Wilms' tumors, patient-matched normal kidney (same cohort used for expression analysis) and the SK-NEP-1 cell line were used as templates for sequencing. Reference sequence for primer design, exon layout and assem-bly was obtained from the UCSC human genome database (http://genome.ucsc.edu/). Forward and reverse primers for each of the 24 HACEI exons contained the 21M13F ( T G T A A A A C G A C G G C C A G T ) or M13R (CAGGAAA-C A G C T A T G A C ) sequences at their 5' ends, respectively. After PCR of each exon from genomic DNA samples auto-mated sequencing was performed by standard methods. Sequence reads were base-called using Phred software and subsequently assembled with reference sequences using Phrap software (59,60). Contigs of sequence traces corres-ponding to each exon were examined using PolyPhred soft-ware (61) for detection of heterozygotes and visualized in Consed software (62) to facilitate verification of sequence variants by examination of individual traces. Thioester bond formation assay L-[35S]-Methionine labeled Hacel proteins were synthesized in vitro in a coupled transcription/translation system (Promega) using pcDNA3-HA-Hacel plasmids coding for the wild-type and mutant Hacel proteins. Reaction mixtures contained 2 JJLI of the translation reaction, G S T - U b (1 |xg), 50 mM KC1, 5 mM MgCl 2 , 2 mM ATP, 10 mM creatine' phos-phate, 0.5 U of creatine phosphokinase, 1 mM DTT, and 20 mM Tris -HCl (pH 7.8). Reactions were also supplemented with E l (~0.5 (xg) and the respective E2s (~0.4 u.g) (Boston-Biochem). After incubation for 30 min at 30°C, reactions were stopped with 2 x S D S - P A G E sample buffer at the 211 2072 Human Molecular Genetics, 2004, Vol. 13, No. 18 absence or presence of 300 mM p-mercaptoethanol. Reactions containing p-mercaptoethanol were then boiled for 3 min, and those lacking P-mercaptoethanol were incubated at room temperature for 20 min before loading. Reaction products were resolved by SDS-7% PAGE and visualized by autoradiography. Purification of the recombinant proteins from bacteria and antibody preparation Hacel or p75, lacking the HECT domain, were expressed as His-tagged fusion proteins in Escherichia coli BL21(DE3). After 4 h of IPTG induction, bacteria were lysed by sonication in buffer containing 2 M NaCl, 10 mM Tris -HCl (pH 7.6) and 0.5 mM PMSF. Cell debris was removed by centrifugation at lOOOOg for 20 min at 4°C. The supernatant was diluted 4-fold with 10 mM Tris -HCl (pH 7.6), 0.5 mM PMSF and passed through a Ni+-Sepharose column (Qiagen). After washing the column with loading buffer [500 mM NaCl, 10 mM Tris -HCl (pH 7.6)], bound proteins were eluted with the same buffer containing 300 mM imidazole and dialyzed against 200 mM KC1, 10 mM Tris -HCl (pH 7.6). a-Hacel antibodies were generated by subcutaneous immunization of rabbits with the full-length Hacel recombinant protein or with the N-terminal Hacel peptide ( C L V L L L K K G A NPNYQ DISG) and used for western blotting at 1:2000 dilution. Immunoprecipitation and western blotting Immunoprecipitation and western blotting were done essen-tially as described previously (53,63). Briefly, immunoprecipi-tations were performed using 500 u.1 of cytoplasmic cell extracts (~500 u.g-1 mg of total protein) and 3 u,l (~3 p-g) of the appropriate antibodies immobilized on 20 |xl of protein (A + G)-Sepharose beads (Qiagen) for 2-3 h at 4°C with rotation. For immunoprecipitation of 26S proteasome, the mixture of antibodies directed against a and P subunits [Calbiochem; 5 |xl of each antibody per 30 u,l of protein(A)-Sepharose] has been used. Monoclonal mouse a-HA (Babco), a-BiP, a-Grb2 and a-UbcH7 (Transduction Labora-tories), a-p97 (VCP; Research Diagnostics Inc.), goat a-actin, rabbit a-Ub (Santa Cruz) and a-acetyl-histone H3 (Upstate) antibodies were used at 1:1000 dilution. Subcellular fractionation To prepare cell extracts, NIH3T3 cells expressing HA-Hacel were trypsinized, washed three times with PBS and lysed with buffer containing 50 mM KC1, 2 mM MgCl 2 , 2 mM DTT, 0.25% NP-40, 20 mM Hepes-KOH (pH 7.8) and the proteasome inhibitors MG132 (20 U,M) or lactacystin (10 U,M). Nuclear, mitochondrial and post-mitochondrial fractions were separated by sequential centrifugation at lOOOg for 5 min followed by centrifugation at lOOOOg for 15 min. Nuclei were then additionally purified from the re-suspended lOOOg pellet fraction by spinning down through a 50% glycerol cushion. The post-ribosomal super-natant was obtained by a further 20 min centrifugation of the post-mitochondrial fraction at lOOOOOrpm in a TLA-100 centrifuge (Beckman). Cell fractions were normalized for protein concentration using Protein Assay kit (BioRad). Immunofluorescence microscopy NIH3T3 cells exponentially growing on coverslips were rinsed with PBS and fixed with — 20°C cold methanol for 10 min. To detect Hacel and BiP, coverslips were incubated overnight with a mixture of rabbit a-Hacel (1:2000 dilution) and mouse a-BiP antibodies (1:1000 dilution) followed by the secondary, antibodies Rhodamine Red-X-conjugated goat a-rabbit and Oregon Green 514-conjugated goat a-mouse (Molecular Probes). Slides were counterstained with DAPI and analyzed using a Zeiss Axioplan epifluorescent micro-scope equipped with a COHU-CCD camera. Control staining with pre-immune antibodies or no primary antibodies showed no signal (data not shown). SUPPLEMENTARY MATERIAL Supplementary Material is available at H M G Online. ACKNOWLEDGEMENTS We would like to thank Seong-Jin Kim for reagents, Joan Mathers and Heather Wildgrove for technical assistance, Francis Ouellette, Yaron Butterfield, Jacqueline Schein, Stefanie Butland and Sohrab Shah for bioinformatics support, as well as Fred Barr, Catherine Anderson, Lola Maksumova, Gregor Reid and Fan Zhang for helpful discus-sions. Normal fetal kidney samples were provided by Alan G. Fantel, Birth Defects Research Laboratory, University of Washington. This study was supported by the National Cancer Institute of Canada and by a Translational Research Grant from the Children's Oncology Group (to P.H.B.S.). This work was also funded by the Johal Program in Pediatric Oncology Basic and Translational Research at the BC Research Institute for Children's and Women's Health. REFERENCES 1. Rabbitts, T.H. (1994) Chromosomal translocations in human cancer. Nature, 372, 143-149. 2. Gnarra, J.R., Tory, K„ Weng, Y„ Schmidt, L., Wei, M.H., Li, H., Latif, F., Liu, S., Chen, F., Duh, F.M. et al. (1994) Mutations of the VHL tumor suppressor gene in renal carcinoma. Nat. Genet., 7, 85-90. 3. Arai, E., Ikeuchi, T. and Nakamura, Y. (1994) Characterization of the translocation breakpoint on chromosome 22ql2.2 in a patient with neurofibromatosis type 2 (NF2). Hum. Mol. Genet., 3, 937-939. 4. Rosty, C , Peter, M., Zucman, J., Validire, P., Delattre, O. and Aurias, A. (1998) Cytogenetic and molecular analysis of a t(l;22)(p36;ql 1.2) in a rhabdoid tumor with a putative homozygous deletion of chromosome 22. Genes Chromosomes Cancer, 21, 82-89. 5. Versteege, I., Sevenet, N., Lange, J., Rousseau-Merck, M.F., Ambros, P., Handgretinger, R., Aurias, A. and Delattre, O. (1998) Truncating mutations of hSNF5/INI 1 in aggressive paediatric cancer. Nature, 394, 203-206. 6. Martin, E.S., Cesari, R., Pentimalli, F., Yoder, K., Fishel, R., Himelstein, A.L., Martin, S.E., Godwin, A.K., Negrini, M. and Croce, C M . (2003) The BCSC-1 locus at chromosome 1 Iq23-q24 is a candidate tumor suppressor gene. Proc. Natl Acad. Sci. USA, 100, 11517-11522. 212 Human Molecular Genetics, 2004, Vol. 13, No. 18 2073 7. Fernandez, C.V., Lestou, V.S., Wildish, J., Lee, C L . and Sorensen, P.H. (2001) Detection of a novel t(6;15)(q21;q21) in a pediatric Wilms tumor. Cancer Genet. Cytogenet., 129, 165-167. 8. Dome, J.S. and Coppes, M.J. (2002) Recent advances in Wilms tumor genetics. Curr. Opin. Pediatr., 14, 5-11. 9. Bruce, C.K., Howard, P., Nowak, N.J. and Hoban, P.R. (2003) Molecular analysis of region t(5;6)(q21;q21) in Wilms tumor. Cancer Genet. Cytogenet, 141, 106-113. 10. Hoban, P.R., Cowen, R.L., Mitchell, E.L., Evans, D.G., Kelly, M., Howard, P.J. and Heighway, J. (1997) Physical localisation of the breakpoints of a constitutional translocation t(5;6)(q21;q21) in a child with bilateral Wilms' tumor. J. Med. Genet., 34, 343-345. 11. Solis, V., Pritchard, J. and Cowell, J.K. (1988) Cytogenetic changes in Wilms' tumors. Cancer Genet. Cytogenet., 34, 223-234. 12. Utada, Y., Haga, S., Kajiwara, T., Kasumi, F., Sakamoto, G., Nakamura, Y. and Emi, M. (2000) Mapping of target regions of allelic loss in primary breast cancers to 1-cM intervals on genomic contigs at 6q21 and 6q25.3. Jpn J. Cancer Res., 91, 293-300. 13. Zhang, Y., Matthiesen, P., Harder, S., Siebert, R., Castoldi, G., Calasanz, M.J., Wong, K.F., Rosenwald, A., Ott, G., Atkin, N.B. et al. (2000) A 3-cM commonly deleted region in 6q21 in leukemias and lymphomas delineated by fluorescence in situ hybridization. Genes Chromosomes Cancer, 27, 52—58. 14. Orphanos, V., McGown, G., Hey, Y., Thorncroft, M., Santibanez-Koref, M„ Russell, S.E., Hickey, I., Atkinson, R.J. and Boyle, J.M. (1995) Allelic imbalance of chromosome 6q in ovarian tumors. Br. J. Cancer, 71, 666-669. 15. Hyytinen, E.R., Saadut, R., Chen, C , Paull, L., Koivisto, P.A., Vessella, R.L., Frierson, H.F., Jr and Dong, J.T. (2002) Defining the region(s) of deletion at 6ql6-q22 in human prostate cancer. Genes Chromosomes Cancer, 34, 306—312. 16. Morelli, C , Karayianni, E., Magnanini, C , Mungall, A.J., Thorland, E., Negrini, M., Smith, D.I. and Barbanti-Brodano, G. (2002) Cloning and characterization of the common fragile site FRA6F harboring a replicative senescence gene and frequently deleted in human tumors. Oncogene, 21, 7266-7276. 17. Pickart, C M . (2001) Mechanisms underlying ubiquitination. Annu. Rev. Biochem., 70, 503-533. 18. Huibregtse, J.M., Scheffner, M., Beaudenon, S. and Howley, P.M. (1995) A family of proteins structurally and functionally related to the E6-AP ubiquitin-protein ligase. Proc. Natl Acad. Sci. USA, 92, 2563-2567. 19. Weissman, A.M. (2001) Themes and variations on ubiquitylation. Nat. Rev. Mol. Cell. Biol., 2, 169-178. 20. Bork, P. (1993) Hundreds of ankyrin-like repeats in functionally diverse proteins: mobile modules that cross phyla horizontally? Proteins, 17, 363-374. 21. Ingham, R.J., Gish, G. and Pawson, T. (2004) The Nedd4 family of E3 ubiquitin ligases: functional diversity within a common modular architecture. Oncogene, 23, 1972-1984. 22. Arcellana-Panlilio, M.Y., Egeler, R.M., Ujack, E., Pinto, A., Demetrick, D.J., Robbins, S.M. and Coppes, M.J. (2000) Decreased expression of the INK4 family of cyclin-dependent kinase inhibitors in Wilms tumor. Genes Chromosomes Cancer, 29, 63-69. 23. Huibregtse, J.M., Scheffner, M. and Howley, P.M. (1993) Cloning and expression of the cDNA for E6-AP, a protein that mediates the interaction of the human papillomavirus E6 oncoprotein with p53. Mol. Cell. Biol., 13, 775-784. 24. Izzi, L. and Attisano, L. (2004) Regulation of the TGFbeta signalling pathway by ubiquitin-mediated degradation. Oncogene, 23, 2071-2078. 25. Horster, M.F., Braun, GS. and Huber, S.M. (1999) Embryonic renal epithelia: induction, nephrogenesis, and cell differentiation. Physiol. Rev., 19, 1157-1191. 26. Pritchard-Jones, K. and Hastie, N.D. (1990) Wilms' tumor as a paradigm for the relationship of cancer to development. Cancer Surv., 9, 555-578. 27. Pritchard-Jones, K. (1999) The Wilms tumor gene, WT1, in normal and abnormal nephrogenesis. Pediatr. Nephrol, 13, 620-625. 28. Coppes, M.J. and Pritchard-Jones, K. (2000) Principles of Wilms' tumor biology. Urol. Clin. North Am., 27, 423-433, viii. 29. Knudson, A.J. and Strong, L.C. (1972) Mutation and cancer: a model for Wilms' tumor of the kidney. J. Natl Cancer Inst., 48, 313-324. 30. Jones, P.A. and Baylin, S.B. (2002) The fundamental role of epigenetic events in cancer. Nat. Rev. Genet., 3, 415-428. 31. Jain, P.K. (2003) Epigenetics: the role of methylation in the mechanism of action of tumor suppressor genes. Ann. NY Acad. Sci., 983, 71-83. 32. Nephew, K.P. and Huang, T.H. (2003) Epigenetic gene silencing in cancer initiation and progression. Cancer Lett., 190, 125-133. 33. Momparler, R.L. and Bovenzi, V. (2000) DNA methylation and cancer. J. Cell. Physiol, 183, 145-154. 34. Jones, P.A. and Taylor, S.M. (1980) Cellular differentiation, cytidine analogs and DNA methylation. Cell, 20, 85-93. 35. Ng, H.H., Feng, Q., Wang, H., Erdjument-Bromage, H., Tempst, P., Zhang, Y. and Struhl, K. (2002) Lysine methylation within the globular domain of histone H3 by Dotl is important for telomeric silencing and Sir protein association. Genes Dev., 16, 1518-1527. 36. Iizuka, M. and Smith, M.M. (2003) Functional consequences of histone modifications. Curr. Opin. Genet. Dev., 13, 154-160. 37. Berger, S.L. (2002) Histone modifications in transcriptional regulation. Curr. Opin. Genet. Dev., 12, 142-148. 38. Friedlander, R., Jarosch, E., Urban, J., Volkwein, C and Sommer, T. (2000) A regulatory link between ER-associated protein degradation and the unfolded-protein response. Nat. Cell Biol, 2, 379-384. 39. Russo, A.A., Tong, L., Lee, J.O., Jeffrey, P.D. and Pavletich, N.P. (1998) Structural basis for inhibition of the cyclin-dependent kinase Cdk6 by the tumor suppressor pl6INK4a. Nature, 395, 237-243. 40. Rice, J.C., Ozcelik, H., Maxeiner, P., Andrulis, I. and Futscher, B.W. (2000) Methylation of the BRCA1 promoter is associated with decreased BRCA1 mRNA levels in clinical breast cancer specimens. Carcinogenesis, 21, 1761-1765. 41. Nguyen, C , Liang, G., Nguyen, T.T., Tsao-Wei, D., Groshen, S., Lubbert, M., Zhou, J.H., Benedict, W.F. and Jones, P.A. (2001) Susceptibility of nonpromoter CpG islands to de novo methylation in normal and neoplastic cells. J. Natl Cancer Inst., 93, 1465-1472. 42. Srivastava, M., Hsieh, S., Grinberg, A., Williams-Simons, L., Huang, S.P. and Pfeifer, K. (2000) H19 and Igf2 monoallelic expression is regulated in two distinct ways by a shared cis acting regulatory region upstream of H19. Genes Dev., 14, 1186-1195. 43. Ginjala, V., Holmgren, C , Ulleras, E., Kanduri, C , Pant, V., Lobanenkov, V., Franklin, G. and Ohlsson, R. (2002) Multiple cis elements within the Igf2/H19 insulator domain organize a distance-dependent silencer. A cautionary note. J. Biol. Chem., 277, 5707-5710. 44. Roessler, E., Ward, D.E., Gaudenz, K , Belloni, E., Scherer, S.W., Donnai, D„ Siegel-Bartelt, J., Tsui, L.C. and Muenke, M. (1997) Cytogenetic rearrangements involving the loss of the Sonic Hedgehog gene at 7q36 cause holoprosencephaly. Hum. Genet., 100, 172-181. 45. Belloni, E., Muenke, M., Roessler, E., Traverso, G., Siegel-Bartelt, J., Frumkin, A., Mitchell, H.F., Donis-Keller, H., Helms, C , Hing, A.V. et al. (1996) Identification of Sonic hedgehog as a candidate gene responsible for holoprosencephaly. Nat. Genet., 14, 353-356. 46. Grewal, S.I. and Moazed, D. (2003) Heterochromatin and epigenetic control of gene expression. Science, 301 798-802. 47. Noma, K , Allis, C D . and Grewal, S.I. (2001) Transitions in distinct histone H3 methylation patterns at the heterochromatin domain boundaries. Science, 293, 1150-1155. 48. Richards, E.J. and Elgin, S.C. (2002) Epigenetic codes for heterochromatin formation and silencing: rounding up the usual suspects. Cell, 108, 489-500. 49. Yoshida, B.A., Sokoloff, M.M., Welch, D.R. and Rinker-Schaeffer, C.W. (2000) Metastasis-suppressor genes: a review and perspective on an emerging field. J. Natl Cancer Inst., 92, 1717-1730. 50. Sakamoto, K.M. (2002) Ubiquitin-dependent proteolysis: its role in human diseases and the design of therapeutic strategies. Mol. Genet. Metab., 11, 44-56. 51. Scheffner, M., Huibregtse, J.M., Vierstra, R.D. and Howley, P.M. (1993) The HPV-16 E6 and E6-AP complex functions as a ubiquitin-protein ligase in the ubiquitination of p53. Cell, 75, 495-505. 52. Yang, Y., Li, C C and Weissman, A.M. (2004) Regulating the p53 system through ubiquitination. Oncogene, 23, 2096-2106. 53. Tognon, C , Knezevich, S.R., Huntsman, D., Roskelley, CD., Melnyk, N., Mathers, J.A., Becker, L., Carneiro, F., MacPherson, N., Horsman, D. et al. (2002) Expression of the ETV6-NTRK3 gene fusion as a primary event in human secretory breast carcinoma. Cancer Cell, 2, 367-376. 54. Tognon, C.E., Kirk, H.E., Passmore, L.A., Whitehead, LP., Der, C.J. and Kay, R.J. (1998) Regulation of RasGRP via a phorbol ester-responsive CI domain. Mol. Cell. Biol, 18, 6995-7008. 213 2074 Human Molecular Genetics, 2004, Vol. 13, No. 18 55. Anderson, C L . and Brown, C.J. (2002) Variability of X chromosome inactivation: effect on levels of TIMP1 RNA and role of DNA methylation. Hum. Genet, 110, 271-278. 56. Goodfellow, P.J., Mondello, C , Darling, S.M., Pym, B., Little, P. and Goodfellow, P.N. (1988) Absence of methylation of a CpG-rich region at the 5' end of the MIC2 gene on the active X, the inactive X, and the Y chromosome. Proc. Natl Acad. Sci. USA, 85, 5605-5609. 57. Sarkar, S., Roy, B.C., Hatano, N., Aoyagi, T., Gohji, K. and Kiyama, R. (2002) A novel ankyrin repeat-containing gene (Kank) located at 9p24 is a growth suppressor of renal cell carcinoma. J. Biol. Chem., 277, 36585-36591. 58. Luo, R.X., Postigo, A.A. and Dean, D.C (1998) Rb interacts with histone deacetylase to repress transcription. Cell, 92, 463-473. 59. Ewing, B. and Green, P. (1998) Base-calling of automated sequencer traces using phred. II. Error probabilities. Genome Res., 8, 186-194. 60. Ewing, B., Hillier, L., Wendl, M C . and Green, P. (1998) Base-calling of automated sequencer traces using phred. I. Accuracy assessment. Genome Res., 8, 175-185. 61. Nickerson, D.A., Tobe, V.O. and Taylor, S.L. (1997) PolyPhred: automating the detection and genotyping of single nucleotide substitutions using fluorescence-based resequencing. Nucl. Acids Res., 25, 2745-2751. 62. Gordon, D., Abajian, C. and Green, P. (1998) Consed: a graphical tool for sequence finishing. Genome Res., 8, 195-202. 63. Evdokimova, V., Ruzanov, P., Imataka, H., Raught, B., Svitkin, Y., Ovchinnikov, L.P. and Sonenberg, N. (2001) The major mRNA-associated protein YB-1 is a potent 5' cap-dependent mRNA stabilizer. EMBOJ., 20, 5491-5502. 214 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0100318/manifest

Comment

Related Items