THE GENETIC BASIS OF THE MULTIPLE
COLORECTAL ADENOMA PHENOTYPE
Oliver Sieber
Thesis submitted for the degree of
Doctor of Philosophy
in the
University of London
Cancer Research UK
January 2004 ProQuest Number: U642895
All rights reserved
INFORMATION TO ALL USERS The quality of this reproduction is dependent upon the quality of the copy submitted.
In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if material had to be removed, a note will indicate the deletion. uest.
ProQuest U642895
Published by ProQuest LLC(2016). Copyright of the Dissertation is held by the Author.
All rights reserved. This work is protected against unauthorized copying under Title 17, United States Code. Microform Edition © ProQuest LLC.
ProQuest LLC 789 East Eisenhower Parkway P.O. Box 1346 Ann Arbor, Ml 48106-1346 Abstract
Multiple adenoma patients are characterised by 3-100 colorectal adenomas, resulting in an increased risk of colorectal cancer. The disease can be inherited as a Mendelian trait, either autosomal dominant or recessive, or occur sporadically. Approximately 10% of multiple adenoma patients carry a truncating mutation in the APC gene, usually in exons 1-4, exon 9 or the 3’-half of exon 15, producing a diagnosis of AAPC. Two APC missense variants, I1307K and E1317Q, have also been associated with the disease, and some whole-gene deletions have been reported. However, these latter associations have been inconsistent. In contrast, germline mutations in the central region of APC cause classical FAP, a dominant predisposition to hundreds or thousands of colorectal adenomas, resulting in a near 100% risk of cancer. The molecular mechanisms underlying these genotype-phenotype correlations are poorly understood. The genetic basis of multiple adenomas in patients without APC mutation remains to be identified.
In this thesis, I have excluded a major role of whole-gene APC deletions in causing AAPC/multiple adenomas. Instead, individuals with such mutations were found to develop classical FAP. I have refined the relationship between germline and somatic mutations at the APC locus in AAPC and classical FAP to show that somatic mutations are constrained by selection for an optimal level of Wnt signalling and that this at least partly determines disease severity. In support of this view, I have provided evidence against a role of truncating APC mutations in causing chromosomal instability. I have shown that the genetic basis underlying multiple adenomas is heterogeneous, by establishing the importance of recessive germline mutations in the base excision repair gene MYH. Some individuals with such mutations were also found to develop classical polyposis. I provide evidence that multiple adenomas are not generally attributable to germline mutations in the Wnt and TGFp/BMP signalling pathways. Acknowledgements
I owe thanks to my supervisor Ian Tomlinson for his great support and guidance of my project. He has been of invaluable help in numerous discussions and a continuous source of encouragement. My thanks also go to my collaborators Hanan Lamlum,
Karl Heinimann, Lara Lipton, Kelly Woodford-Richens and Michael Crabtree for their intellectual and practical contributions, as well as to my colleagues from the
Cancer Research UK Family Cancer Clinic, Colorectal Unit and Polyposis Registry at St Mark's Hospital.
I am indebted to the patients for their participation in this study as well as to their respective doctors and pathologists for contributing samples and clinical information.
I would also like to thank the staff from the Cancer Research UK Equipment Park,
Histopathology and FACS Laboratories for their advice and technical support. I am particularly grateful to the Boehringer Ingelheim Fonds for their financial support.
A big thank you goes to all my colleagues in the Molecular and Population Genetics
Laboratory who have kept my spirits up during my studies. In particular, I am
grateful to Andrew Rowan and Pat Gorman who were readily available for advice
and practical help and who also contributed to my work.
Lastly, I would especially like to thank my family without whose support and
encouragement completion of this thesis would have not been possible. Abbreviations
aa amino acid AAPC attenuated adenomatous polyposis coli Alb albumin APC adenomatous polyposis coli BMP bone morphogenic protein BMPRIA bone morphogenic protein receptor type 1A cDNA complementary DNA CGH comparative genomic hybridisation CIN chromosomal instability Co-SMAD common-mediator SMAD C0X2 cyclooxygenase-2 CPKl-a casein kinase 1-a CRC colorectal cancer c^uK CCLWClr ReStOLTclv ttK dHzO distilled water DI DNA index DNA deoxyribonucleic acid Dsh dishevelled ES cell embryonic stem cell FAP familial adenomatous polyposis FCM flow cytometry GSK3-P glycogen synthase kinase 3-p hDLG human homologue of the Drosophila discs large tumour suppressor HMPS hereditary mixed polyposis syndrome HNPCC hereditary nonpolyposis colorectal cancer IPS juvenile polyposis syndrome kb kilo base pairs LOH loss of heterozygosity LRP low-density lipoprotein receptor-related protein mb mega base pairs MCR mutation cluster region MLHl human homologue of E. coli MutL MMR mismatch repair MSH2 human homologue of E. coli MutS MSI microsatellite instability MTHl human homologue of E. coli MutT MYH human homologue of E. coli MutY nm nanometer NSAID nonsteroidal anti-inflammatory drug nt nucleotides OGGI human homologue of E. coli MutM 8-oxo-G 8-oxo-7,8-dihydroxy-2'-deoxyguanosine PCR polymerase chain reaction PI propidium iodide PJS Peutz-Jeghers syndrome PMS2 human homologue of S. cerevisiae postmeiotic segregation increased PPAR peroxisome proliferator-activated receptor PTEN phosphatase and tensin homologue deleted on chromosome 10 PTT protein truncation test RNA ribonucleic acid ROS reactive oxygen species rpm rotations per minute RQM-PCR real-time quantitative multiplex PCR R-SMAD receptor-regulated SMAD SMAD human homologue of Drosophila small mothers against decapentaplegic SSCP single-strand conformation polymorphism TCP T cell factor TGF-p transforming growth factor-^ p-TRCP P-transducin repeat-containing protein
All genes have been italicised throughout the text, whereas proteins are shown in plain text. Table of Contents
Abstract...... 2 Acknowledgements...... 3 Abbreviations ...... 4 Table of Contents...... 6 Table of Figures...... 10 Table of Tables...... 12
Chapter 1 Introduction to Colorectal Cancer ...... 14 1.1 The Epidemiology of Colorectal Cancer ...... 14 1.2 The Natural History of Colorectal Cancer ...... 17 1.2.1 The Anatomy of the Colorectum ...... 17 1.2.2 Adenomatous Polyps ...... 19 1.2.3 Hamartomatous Polyps...... 21 1.2.4 Stages of Colorectal Cancer...... 22 1.3 The Adenoma-Carcinoma Sequence ...... 23 1.4 Hereditary Colorectal Cancer Syndromes ...... 26 1.4.1 Polyposis Syndromes ...... 27 1.4.1.1 Familial Adenomatous Polyposis ...... 27 1.4.1.1.1 The APC Gene and its Functions ...... 29 1.4.1.1.2 APC and the Regulation of Wnt Signalling ...... 32 1.4.1.1.3 Germline Mutations in the APC Gene ...... 35 1.4.1.1.4 Genotype-phenotype Associations in FAP ...... 36 1.4.1.1.5 Somatic Mutations in the APC Gene ...... 37 1.4.1.1.6 Associations between Germline and Somatic APC Mutations.... 38 1.4.1.1.7 Genetic Testing and Clinical Management ...... 39 1.4.1.2 Attenuated Adenomatous Polyposis Coli ...... 40 1.4.1.2.1 APC Missense Variants and AAPC ...... 41 1.4.1.2.2 Clinical Management ...... 42 1.4.1.3 Juvenile Polyposis ...... 42 1.4.1.3.1 The TGF-(3/BMP Signalling Pathway...... 44
6 1.4.1.4 Peutz-Jeghers Syndrome ...... 46 1.4.1.5 Cowden’s Syndrome and Bannayan-Riley-Ruvalcaba Syndrome.... 48 1.4.1.6 Hereditary Mixed Polyposis Syndrome ...... 49 1.4.2 Hereditary Non-polyposis Colorectal Cancer ...... 49 1.5 The Multiple Colorectal Adenoma Phenotype ...... 54 1.5.1 The Base Excision Repair Pathway ...... 55 1.6 Aims of this Thesis...... 58
Chapter 2 Materials and Methods...... 59 2.1 DNA Extraction ...... 59 2.1.1 DNA Extraction from Blood ...... 59 2.1.2 DNA Extraction from Cell Lines ...... 60 2.1.3 DNA Extraction from Fresh-frozen Tissue ...... 61 2.1.4 DNA Extraction from Formalin-fixed, Paraffin-embedded Tissue 61 2.2 Spectrophotometry...... 62 2.3 Polymerase Chain Reaction ...... 63 2.4 Agarose Gel Electrophoresis...... 64 2.5 Mutation Detection ...... 65 2.5.1 Single-strand Conformation Polymorphism Analysis ...... 65 2.5.2 Protein Truncation Test ...... 67 2.5.3 Fluorescence Cycle-sequencing...... 69 2.5.4 Real-time Quantitative Multiplex PCR ...... 71 2.5.5 Loss of Heterozygosity Analysis...... 73 2.6 Genotyping ...... 74 2.6.1 Amplification Refractory Mutation System PCR ...... 74 2.6.2 Restriction Fragment Length Polymorphism Analysis ...... 74 2.7 Comparative Genomic Hybridisation ...... 75 2.8 Flow Cytometry and Ki-67 Immunocytochemistry ...... 78 2.9 Cloning ...... 81 2.10 Cell Culture...... 83 2.11 Solutions and Media ...... 85 Chapter 3 The Contribution of Whole-gene APC Deletions to AAPC/Multiple Colorectal Adenomas and Classical FAP...... 88 3.1 Introduction ...... 88 3.2 Study Population ...... 90 3.3 Development of a RQM-PCR Assay to Detect APC Exon 14 Deletions 91 3.4 Germline APC Deletions in Patients with Multiple Colorectal Adenomas and Classical Polyposis...... 94 3.5 The Extent of the Detected Germline APC Deletions ...... 96 3.6 Discussion ...... 99
Chapter 4 The Relationship between Germline and Somatic Mutations at the C Locus in Patients with Classical FAP...... 103 4.1 Introduction ...... 103 4.2 Study Population ...... 106 4.3 Loss of Heterozygosity ?iXAPC...... 107 4.4 The Mechanism of Allelic Loss at APC ...... 109 4.5 Truncating Somatic Mutations at APC ...... I ll 4.6 Associations of Germline and Somatic APC Mutations in FAP adenomas 114 4.7 Discussion ...... 116
Chapter 5 The Relationship between Germline and Somatic Mutations at the APC locus in Patients with AAPC...... 124 5.1 Introduction ...... 124 5.2 Study Population ...... 127 5.3 Somatic Mutations and Loss of Heterozygosity at APC ...... 128 5.4 Discussion ...... 131
Chapter 6 The Extent of Chromosomal Instability in Human Colorectal
Adenomas with Two Mutational Hits 2AAPC...... 137 6.1 Introduction ...... 137 6.2 Study Population ...... 139 6.3 The Assessment of Chromosomal Instability ...... 142 6.3.1 Flow Cytometry and Ki-67 Immunocytochemistry ...... 142 6.3.2 Loss of Heterozygosity Analysis ...... 145 6.3.3 Comparative Genomic Hybridisation ...... 147 6.4 Discussion ...... 149
Chapter 7 The Contribution of Germline MITT Mutations to Multiple Colorectal Adenomas and Classical FAP ...... 153 7.1 Introduction ...... 153 7.2 Study Population ...... 154 7.2.1 Multiple Colorectal Adenoma Patients and Normal Controls ...... 154 7.2.2 Classical Adenomatous Polyposis Patients ...... 155 7.3 Mutations in Multiple Adenoma Patients and Normal Controls 156 7.4 Somatic APC Mutations and Allelic Loss at MYH in Adenomas from MYH Mutation Carriers ...... 160 7.5 Clinical Features of Multiple Adenoma Patients with MYH Mutations 161 7.6 MTHl and OGGI Mutations in Multiple Adenoma Patients ...... 163 7.7 MT77 Mutations in Patients with Classical Polyposis ...... 163 7.8 Discussion ...... 166
Chapter 8 The Contribution of Wnt and TGF-p/BMP Pathway Genes to Multiple Colorectal Adenomas ...... 170 8.1 Introduction ...... 170 8.2 Study Population ...... 172 8.3 Germline Mutations in the Wnt and TGF-p/BMP Signalling Pathways ...... 173 8.4 Discussion ...... 178
Chapter 9 Summary and General Discussion ...... 180
Publications ...... 188 A ppendix ...... 191 References ...... 199 Table of Figures
Figure 1.1: The nine anatomical regions of the colorectum ...... 18 Figure 1.2: The kinetic organisation of colorectal crypts ...... 18 Figure 1.3: Section of a tubular adenomatous polyp ...... 20 Figure 1.4: Section of a flat adenomatous polyp ...... 20 Figure 1.5: Section of a Peutz-Jeghers polyp...... 21 Figure 1.6: Section of a juvenile polyp ...... 22 Figure 1.7: The adenoma-carcinoma sequence for sporadic colorectal cancer ...... 25 Figure 1.8: A FAP colon carpeted with hundreds of adenomas ...... 28 Figure 1.9: APC protein domains and their putative functions ...... 31 Figure 1.10: The Wnt signalling pathway...... 34 Figure 1.11: Distribution of germline mutations in the APC gene...... 35 Figure 1.12: Distribution of somatic mutations in the C gene ...... 38 Figure 1.13: The TGF-j3/BMP signalling pathway...... 45 Figure 1.14: Microsatellite instability at BAT26 in a colon cancer from a patient with HNPCC...... 53 Figure 1.15: The base excision repair pathway...... 57 Figure 3.1: Relative standard curves for^PC and Albumin amplification by RQM- PCR...... 92 Figure 3.2: RQM-PCR results for 5 healthy controls, 5 deletion controls, 143 AAPC/multiple adenoma patients, and 60 classical polyposis patients ...... 95 Figure 4.1: Mean RQM-PCR results (2EE(-ddCT) values) of five titrations of DNA mixed from a normal individual and an FAP patient with a germline APC deletion ...... 110 Figure 5.1: A model of somatic mutations dXAPC in colorectal adenomas from AAPC patients with a R332X nonsense mutation in exon 9...... 136 Figure 6.1: Representative results of flow cytometry and Ki-67 immuno cytochemistry on one normal biopsy (N-283; A-C) and one near-diploid colorectal adenoma with two mutational hits at APC (335; D-F)...... 144 Figure 6.2: A representative CGH result of one tumour, 131c, which showed allelic loss as the ‘second hit’ at APC hy microsatellite analysis ...... 148
10 Figure 7.1: Selected germline MYH variants ...... 158 Figure 8.1: Loss of heterozygosity at BMPRIA in an adenocarcinoma from a patient with a germline BMPRIA mutation (codon 360 del his) ...... 177
11 Table of Tables
Table 1.1: Criteria for the diagnosis of juvenile polyposis syndrome ...... 43 Table 1.2: Criteria for the diagnosis of Peutz-Jeghers syndrome ...... 47 Table 1.3: Amsterdam I and Amsterdam II criteria for the diagnosis of hereditary non-polyposis colorectal cancer ...... 50 Table 1.3: continued ...... 51 Table 3.1 : Clinical diagnosis and extent of germline APC deletion in polyposis patients reported in the literature ...... 89 Table 3.2: Primers and probes for the RQM-PCR assay ...... 91 Table 3.3: Primer sequences, annealing temperatures, detection methods applied and allele frequencies for the APC polymorphisms...... 97 Table 3.4: Genotype results for seven classical polyposis patients harbouring an^PC exon 14 deletion as determined by RQM-PCR ...... 98 Table 4.1: ‘Just-right’ signalling model for APC mutations in FAP (after Albuquerque et al.)...... 105 Table 4.2: Loss of heterozygosity at^P C...... 108 Table 4.3: Truncating somatic mutations at APC...... 112 Table 4.4: Association between germline and somatic APC mutations in FAP adenomas ...... 115 Table 4.5: Somatic APC mutations in sporadic colorectal cancers ...... 119 Table 4.6: ‘Loose fit’ signalling model for ^PC mutations in FAP and sporadic colorectal tumours...... 123 Table 5.1: Somatic mutations and allelic loss at ^P C in 72 colorectal adenomas from AAPC patients with a R332X germline mutation in APC...... 130 Table 5.1: continued ...... 131 Table 6.1: Patient ID, APC mutation status and size of the colorectal adenomas analysed as well as analytical methods applied ...... 140 Table 6.1: continued ...... 141 Table 6.2: Results of flow cytometry and Ki-67 immunocytochemistry on colorectal adenomas with two mutational hits at APC and two normal controls ...... 143
12 Table 6.3: Results of LOH analysis on colorectal adenomas with allelic loss as the ‘second hit’ atusing microsatellite markers spanning chromosome 5... 146 Table 7.1: Multiple adenoma patients from the United Kingdom with germline MYH mutations ...... 157 Table 7.2: Clinical features of multiple adenoma patients from the United Kingdom in relation to germline MTTf mutation status ...... 162 Table 7.3: Classical adenomatous polyposis patients with germline MYH mutations...... 164 Table 7.4: Clinical features of classical adenomatous polyposis patients in relation to germline A/TT/mutation status ...... 165 Table 8.1: Characteristics of 47 patients with multiple colorectal adenomas ...... 173 Table 8.2: Germline mutations of the BMPRIA, SMAD4, APC2, axin2 and GSK3P genes detected in 47 multiple adenoma patients ...... 176 Table A.l: Primers and PCR conditions for amplification of APC exons 4-14 from fresh-frozen tissue ...... 191 Table A.2: Primers and PCR conditions for amplification of APC exon 15 regions A- I from fresh-frozen tissue ...... 192 Table A.3: Primers and PCR conditions for amplification of codons 1258-1603 of the APC gene from formalin-fixed, paraffin-embedded tissue ...... 193 Table A.4: Primers and PCR conditions for the MYH gene...... 194 Table A.5: Primers and PCR conditions for the MTHl gene...... 195 Table A.6: Primers and PCR conditions for the OGGI gene...... 195 Table A.7: Primers and PCR conditions for the GSK-3P gene...... 196 Table A.8: Primers and PCR conditions for the oxin2 gene...... 197 Table A.9: Primers and PCR conditions for the APC2 gene...... 198
13 Chapter 1 Introduction to Colorectal Cancer
1.1 The Epidemiology of Colorectal Cancer
Cancer of the colon and rectum (colorectal cancer; CRC) is predominantly a disease of industrialised countries. The disease is common in Australia, Japan, the United
States, Western and Northern Europe, and relatively uncommon in the developing countries of Asia, Africa, and South America. In England and Wales, approximately
30,000 cases are diagnosed each year accounting for approximately 17,000 deaths per annum. These figures make CRC the third most common malignancy after lung and prostate cancer in men and breast and lung cancer in women (Quinn et al., 2001).
The age-standardised incidence of CRC is about 30% greater in males than in females. However, both sexes show similar distributions of incidence by age, with a
sharp rise at around the age of 50 and a peak at around the age of 75 (Hayne et al.,
2001). In patients who develop the disease, prognosis primarily depends on the stage
of the tumour - its anatomical extent and spread - at presentation. There has been a
continuous improvement in survival from CRC over the past 30 years (Hayne et al.,
2001). This is likely to reflect improvements in early detection, surgical techniques,
and adjuvant chemo- and radiotherapy. However, surgery has remained the only
effective cure.
Whilst most CRCs arise sporadically, in individuals with no family history of
disease, at least 5% are due to inherited cancer predisposition syndromes. These
‘classical’ syndromes are generally characterised by autosomal dominant inheritance,
14 early ages of cancer onset, multiple primary tumours, and specific extracolonic features. In addition, milder predispositions to CRC exist which do not manifest as dominant inheritance, and even in cases of sporadic CRC there is a 2- to 3-fold increase in risk in their first-degree relatives (Fuchs et al., 1994; Lovett, 1976; St
John et al., 1993). Furthermore, individuals with chronic inflammatory bowel diseases such as ulcerative colitis and Crohn’s disease are at an increased risk of developing CRC (Gillen et al, 1994; Levin, 1992; Sugita et al, 1991).
The wide variation in CRC rates around the world (Quinn et al, 2001) and the observation that migrating populations generally assume the relative risk of the region into which they move (Haenszel and Kurihara, 1968; King and Haenszel,
1973; McMichael and Giles, 1988) are often taken as evidence that the environment affects the incidence of CRC. However, these data must be interpreted with caution
as they may also reflect differences in reporting, the quality of health care and/or
demographic factors such as life expectancy. Nevertheless, diet has been proposed to
account for at least part of the worldwide variation in CRC risk, as the consumption
of fat, animal proteins, vegetables and fibre varies between regions (Armstrong and
Doll, 1975). Attempts to identify specific dietary components have, however, been
largely unsuccessful. The most consistent correlations have been reported for meat
and alcohol consumption, which show a positive association with CRC risk, and for
vegetable consumption, which shows an inverse association (reviewed in REFS
(Potter, 1999; Reddy, 1999; Ritenbaugh, 2000)). Besides diet, smoking has been
identified as a potential risk factor for CRC (reviewed in REF (Potter, 1999)).
15 A protective effect against CRC has been associated with the use of nonsteroidal anti-inflammatory drugs (NSAIDs) such as aspirin, sulindac and celecoxib
(Giovannucci et ah, 1995; Giovannucci et ah, 1994; Thun et ah, 1991; Waddell and
Loughry, 1983). Perhaps the best evidence for this effect has been obtained from randomised clinical trials on patients with familial adenomatous polyposis (FAP).
FAP is an autosomal dominant predisposition to hundreds or thousands of benign colorectal polyps, resulting in a near 100% lifetime risk of CRC. The disease is caused by germline mutations in the adenomatous polyposis coli {APC) gene. FAP patients treated with sulindac or celecoxib showed a substantial reduction in the number and size of polyps as compared to placebo-control patients, a trend which reversed when the therapy was terminated (Giardiello et ah, 1993; Nugent et ah,
1993; Phillips et ah, 2002; Steinbach et ah, 2000). To date, the mechanism of NSAID action is not fully understood, but it likely to involve inhibition of the enzyme cyclooxygenase-2 (C0X2) (reviewed in REFS (Gupta and Dubois, 2001; Mamett and Kalgutkar, 1999)). C0X2 is an inducible regulator of prostaglandin synthesis,
and is upregulated in approximately 83% of CRCs (reviewed in REF (Smith et ah,
2000)). Prostaglandins control processes as diverse as immunity, maintenance of
vascular integrity, and bone metabolism. They exert their effects by binding to G-
protein-coupled receptors on the cell surface or the nuclear envelope, leading to
changes in the levels of cyclic-AMP and Ca^^ (reviewed in REFS (Gupta and
Dubois, 2001; Sugimoto et ah, 2000)). In addition, prostaglandin metabolites can
directly transactivate members of the peroxisome proliferator-activated receptor
(PPAR) family of nuclear hormone receptors (Forman et ah, 1995; Gupta et ah,
2000; Kliewer et ah, 1995). Genetic evidence for the anti-tumour effect of COX2
inhibitors has been provided by studies of Apc^^^^ and Apc^*" mice (Chulada et ah,
16 2000; Oshima et al., 1996). Both mouse strains are predisposed to intestinal tumour development but develop fewer and smaller lesions on a Com2''' than on a Co^2 wild-type background. In addition, NSAIDs may have other pro-apoptotic or growth inhibitory effects. The drugs have, for example, been associated with an upregulation of the pro-apoptotic proteins bax and bak (Zhou et al., 2001a), and inhibition of the
NFkB/IkB (nuclear factor K-B/inhibitor of k-B) survival signalling pathway (Kopp and Ghosh, 1994; Yamamoto et al., 1999; Yin et al., 1998).
1.2 The Natural History of Colorectal Cancer
The histological progression from tumour initiation to metastasis is generally
referred to as the natural history of a cancer. This transformation is particularly well
characterised for CRC as early lesions are readily available for analysis and is
described below.
1.2.1 The Anatomy of the Colorectum
The colorectum can be divided into nine anatomical regions, the caecum, appendix,
ascending colon, hepatic flexure, transverse colon, splenic flexure, descending colon,
sigmoid and rectum (Figure 1.1). The colonic mucosa is composed of invaginating
crypts which are lined by a single layer of columnar absorptive cells interspersed
with mucin-secreting goblet cells, neuroepithelial cells and paneth cells. The mucosa
rests on a basement membrane (muscularis mucosae), lamina propria, and muscularis
propria. Cell division principally occurs in the lower third of the crypt, the so-called
17 proliferative compartment. As cells migrate up the crypt they differentiate, undergo programmed cell death (apoptosis), and are sloughed into the lumen (Figure 1.2).
Cell turnover in colorectal crypts is rapid, taking approximately 3-6 days (Cotran et al., 1999).
Figure 1.1: The nine anatomical regions of the colorectum.
Caecum (1), appendix (2), ascending colon (3), hepatic flexure (4), transverse colon
(5), splenic flexure (6), descending colon (7), sigmoid colon (8), and rectum (9).
Figure 1.2: The kinetic organisation of colorectal crypts.
Cell proliferation occurs in the lower third of the crypt. As cells migrate up the crypt they differentiate, undergo apoptosis, and are sloughed into the lumen. Cell turnover takes approximately 3-6 days.
Apoptosis
Crypt Differentiation
Proliferation
Muscularis mucosae
18 1.2.2 Adenomatous Polyps
Adenomatous polyps are discrete regions of dysplastic (or neoplastic) epithelium bounded by the underlying basement membrane (reviewed in (Cummings, 2000)).
They develop when the homeostatic balance between epithelial cell proliferation and apoptosis is lost. Histologically, dysplastic epithelium is characterised by cellular crowding, loss of polarity, stratification, and abnormal nuclear features such as enlargement, elongation and hyperchromatism. The increase in nuclear size results in
a loss of cytoplasmic volume. The proliferative compartment of dysplastic crypts is
enlarged and may extend to the surface. Colorectal adenomas probably start as
unicryptal adenomas which initially grow by crypt fission (Chang and Whitener,
1989; Wasan et al., 1998; Wong et al., 1999), presumably to form microadenomas
(clusters of up to 20 dysplastic crypts). As lesion size increases, a defined
adenomatous polyp develops, and there is some evidence that dysplastic epithelium
may begin to grow down into adjacent normal crypts (Preston et al., 2003).
Adenomatous polyps can be divided into three morphological types, tubular,
tubulovillous, or villous. Tubular adenomas are polyps with a less than 25% villous
component (Figure 1.3). A villous component is an area of adjacent, non-branching,
dysplastic fronds that are twice the height of a normal crypt. Villous adenomas
contain a greater than 75% villous component whilst tubulovillous adenomas have an
intermediate morphology. In general, as polyp size increases, the degree of dysplasia
worsens and the villous component increases (O'Brien et al., 1990). A large adenoma
which has severe atypia but has not penetrated the basement membrane is called a
carcinoma in situ. Besides exophytic adenomas, flat adenomas exist (Wolber and
19 Owen, 1991) (Figure 1.4). These lesions are more difficult to detect at endoscopy as they only show as a slight puckering or a depression in the mucosa (Hart et al.,
1998a).
Figure 1.3: Section of a tubular adenomatous polyp.
Taken from REF (Whitelaw et al., 1997).
Figure 1.4: Section of a flat adenomatous polyp.
Taken from REF (Whitelaw et al., 1997).
20 1.2.3 Hamartomatous Polyps
In contrast to adenomatous polyps, hamartomatous polyps do not show the classical features of dysplasia. However, adenomatous dysplasia and invasive carcinoma can develop within hamartomatous polyps (Hizawa et al., 1993b; Jass et al., 1988; Narita et al., 1987). Two types of hamartomatous polyps are of particular importance as they characterise two hereditary CRC predispositions, Peutz-Jeghers syndrome and juvenile polyposis syndrome. Peutz-Jeghers polyps are composed of histologically normal mucosa with an arborizing smooth-muscle core (Figure 1.5). Juvenile polyps lack a smooth muscle core and are characterised by prominent, mucin-filled cysts in the stroma which are lined with flattened epithelium (Figure 1.6).
Figure 1.5: Section of a Peutz-Jeghers polyp.
Normal appearing mucosa is supported by tree-like branched muscularis mucosae.
Taken from REF (Oberhuber and Stolte, 2000).
21 Figure 1.6: Section of a juvenile polyp.
The stroma shows typical prominent cystic dilations. Taken from REF (Aaltonen,
2000 ).
1.2.4 Stages of Colorectal Cancer
CRCs typically present due to a change of bowel habit, rectal bleeding and anaemia.
They can be classified according to a four-stage system which is based on an original proposal made by Dukes in the 1930s. Dukes stage A includes all CRCs which have reached but not penetrated the lamina propria, stage B all those which have broken into the lamina propria but have not invaded the lymph nodes, and stage C all those which have spread to the lymphatic system. Dukes stage D, which was added in
1967, encompasses all CRCs which have metastasised to other organs.
22 1.3 The Adenoma-Carcinoma Sequence
The progression from adenoma to invasive carcinoma is driven by successive genetic and epigenetic changes in tumour suppressor genes and oncogenes. This multi-step evolutionary process is often referred to as the adenoma-carcinoma sequence.
In their normal state, tumour suppressor genes inhibit cell proliferation. Growth
inhibition is lost when both alleles are inactivated by mutation and/or epigenetic
changes, for example promoter méthylation resulting in gene silencing. Tumour
suppressor genes therefore broadly conform to Knudson’s classical two-hit
hypothesis (Knudson, 1971). In contrast, proto-oncogenes, the non-mutant version of
oncogenes, act by promoting cell proliferation. Oncogenic mutations lead to
excessive or inappropriate expression of the protein, resulting in uncontrolled cell
proliferation when only one allele is mutated.
The adenoma-carcinoma sequence for sporadic colorectal cancer is probably most
commonly initiated by bi-allelic mutation of the APC tumour suppressor gene
(Figure 1.7). In this context, APC can therefore be classified as a gatekeeper gene
(Ichii et al., 1993; Levy et al., 1994). Somatic APC mutations have been identified in
microadenomas (Otori et al., 1998; Smith et al., 1994a) and approximately 60-80%
of early sporadic adenomas and carcinomas (Cottrell et al., 1992; Miyoshi et al.,
1992b; Nakamura et al., 1992; Powell et al., 1992). Whilst bi-allelic inactivation of
APC appears to trigger tumour formation - probably by deregulating Wnt signalling
(see Section 1.4.1.1.2) - changes in further genes are required for adenoma growth
and progression. Activating mutations in the oncogenes K-ras or BRAF, both
23 members of the MAPK (mitogen activated protein kinase) pathway, are found at the transition from an early to an intermediate-type adenoma in approximately 50% and
10% of cases, respectively (Bos et ah, 1987; Forrester et ah, 1987; Rajagopalan et ah, 2002; Yuen et ah, 2002). Further progression to a late-type adenoma is associated with loss of chromosome 18q, which is detected in approximately 50% of large adenomas and 75% of carcinomas (Law et ah, 1988; Vogelstein et ah, 1988;
Vogelstein et ah, 1989). The two main tumour suppressor genes which are targeted by this loss are probably SMAD2/MADR2 (small mothers against
decapentaplegic/mothers against decapentaplegic related) and SMAD4/DPC4 (small
mothers against decapentaplegic/deleted in pancreatic cancer), both members of the
TGF-p/BMP (transforming growth factor-p/bone morphogenic protein) signalling
pathway (see Section 1.4.1.3.1). Accordingly, point mutations and deletions have
been identified in SMAD2 and SMAD4 in CRCs (Bppert et ah, 1996; Hahn et ah,
1996; Thiagalingam et ah, 1996). The adenoma-carcinoma transition appears to be
associated with loss of chromosome 17q (Akiyama et ah, 1998; Fearon et ah, 1987;
Rodrigues et ah, 1990; Vogelstein et ah, 1988), although some recent data have been
inconsistent (Hao et ah, 2002). The p53 tumour suppressor gene is the most likely
target of this loss, as 17q LOH correlates with missense and truncating mutations in
p53 (Baker et ah, 1989; Baker et ah, 1990). Tumour invasion and metastasis are,
amongst other changes, associated with loss of E-cadherin function, a component of
adherens-junctions (Christofori and Semb, 1999; Hao et ah, 1997).
24 Figure 1.7: The adenoma-carcinoma sequence for sporadic colorectal cancer.
Inactivation of the APC tumour suppressor gene - or less frequently oncogenic mutation of p-catenin - results in defective Wnt signalling and initiates tumour formation. Subsequent progression towards malignancy is accompanied by mutation in the oncogenes K-ras or BRAF, loss of chromosome 18q harbouring the tumour
suppressor genes SMAD2 and SMAD4, and loss of chromosome 17p harbouring the tumour suppressor gene p53. Tumour invasion and metastasis is associated with loss
of E-cadherin function.
Normal Early Intermediate Late mucosa adenoma adenoma adenoma Carcinoma H Metastasis Î APC K-ras SMAD2, SMAD4 p53 E-cadherin -catenin BRAF Chromosome Chromosome others 18q LOH 17p LOH
A large proportion of sporadic CRCs have aneuploid karyotypes, displaying
numerous numerical and structural chromosomal rearrangements. In addition, about
10-20% of sporadic CRCs show microsatellite instability (MSI), increased somatic
changes in the length of simple tandem nucleotide repeats (Cunningham et al., 1998;
Herman et al., 1998). MSI is the hallmark of defective DNA mismatch repair -
caused, for example, by hypermethylation of the mismatch repair gene MLHl
(Cunningham et al., 1998) - and some aneuploidy is probably also due to underlying
genetic instability. The genes responsible for these two types of genomic instability
are generally referred to as mutator genes. In their normal state, mutator genes act as
25 caretakers of the genome by maintaining genomic integrity and the fidelity of information transfer. Hypermutation appears to result most commonly when both alleles are inactivated by mutation or epigenetic changes. Hypermutation may accelerate carcinogenesis by targeting tumour suppressor genes and oncogenes, in particular when cell-cycle checkpoints have already been compromised (I have reviewed this topic in REF (Sieber et al., 2003b)).
The progression from adenoma to invasive carcinoma probably takes 10 to 40 years
(Ilyas et al., 1999), but there is evidence that not all lesions undergo malignant transformation. The frequency of adenomas in the general population is higher than the frequency of carcinomas, taking into account the fact that some patients will die before the adenomas have had sufficient time to progress to carcinoma (Ransohoff and Lang, 1991). Furthermore, some adenomas which occur in the residual rectum of
FAP patients after ileorectal anastomosis undergo spontaneous regression (Nicholls et al., 1988), and similar findings have been reported for small sporadic tumours which were followed endoscopically over a three year period (Hofstad et al., 1996).
1.4 Hereditary Colorectal Cancer Syndromes
Hereditary colorectal cancer predisposition syndromes can be divided into two main categories depending on the presence or absence of multiple benign precursor lesions. The most common polyposis syndromes, FAP, Peutz-Jeghers syndrome and juvenile polyposis syndrome, account for approximately 1% of the total colorectal cancer burden (Aaltonen, 2000; Lynch and Lynch, 1998). Hereditary non-polyposis colorectal cancer (HNPCC), on the other hand, may account for approximately 3%
26 (Lynch and de la Chapelle, 2003) although conservative estimates range from 1-6%
(Aaltonen et al., 1998; Lynch and de la Chapelle, 1999; Salovaara et al., 2000).
1.4.1 Polyposis Syndromes
In general, polyposis syndromes show autosomal dominant inheritance of either
multiple adenomatous or hamartomatous polyps and CRC. The risk of CRC in
patients with adenomatous polyposis is typically higher than that in patients with
hamartomatous polyposis, although this may reflect a higher overall tumour burden
rather than a polyp-for-polyp risk.
1.4.1.1 Familial Adenomatous Polyposis
Familial adenomatous polyposis (FAP; MIM: 175100) is an autosomal dominant
syndrome with high penetrance, which affects approximately 1 in 13,500 individuals
(Bisgaard et al., 1994). The most prominent clinical manifestation of the disorder is
the development of hundreds to thousands of colorectal adenomas, typically in the
second or third decades of life (Figure 1.8). For clinical diagnosis, it is often
considered that a patient has to have at least 100 polyps although this number can be
smaller in young patients (Vasen et al., 1993). Whilst each adenoma has a low
probability of progressing to malignancy, the large number causes a near 100%
lifetime risk of CRC. If left untreated, CRC usually develops by the early forties.
Despite the strong selective disadvantage associated with FAP, the incidence of the
27 disease is maintained by the occurrence of de novo mutations, which contribute about 20-25% of all cases (Bisgaard et al., 1994; Bulow, 1986).
Figure 1.8: A FAP colon carpeted with hundreds of adenomas.
In addition to colorectal adenomas, FAP patients often develop extracolonic features.
These include upper gastrointestinal tumours, congenital hypertophy of the retinal pigment epithelium (CHRPE), papillary thyroid carcinoma, hepatoblastoma and dental abnormalities (Blair and Trempe, 1980; Heiskanen and Jarvinen, 1996; Herve et al., 1995). Patients with epidermoid cysts and mandibular osteomas are historically diagnosed with Gardner’s syndrome. A further complication of FAP is desmoid tumours, which often develop after surgery or in females after pregnancy. They usually arise in the abdominal wall or bowel mesentery, and can grow to considerable sizes. Some polyposis patients also develop tumours of the central nervous system (medulloblastomas) producing a diagnosis of Turcot’s syndrome type 1 (Hamilton et al., 1995; Paraf et al., 1997; Van Meir, 1998).
28 1.4.1.1.1 The APC Gene and its Functions
FAP is caused by germline mutations in the adenomatous polyposis coli (APC) tumour suppressor gene. The first hint as to the position of the ^PC gene was a patient with colorectal polyposis, mental retardation and other abnormalities who carried a deletion of chromosome band 5q21 (Herrera et al., 1986). Following this clue, the APC gene was mapped to the same region in FAP families (Bodmer et al.,
1987), and subsequently cloned and characterised (Groden et al., 1991; Kinzler et al.,
1991).
The APC gene consists of 16 translated exons and its commonest isoform encodes a
2843 amino acid protein (Groden et al., 1991; Horii et al., 1993; Sulekova and
Ballhausen, 1995). Exon 10A, located downstream of exon 10, is subject to alternative splicing and adds an additional 18 amino acids to the protein when transcribed (Sulekova and Ballhausen, 1995). APC appears to be expressed in a variety of epithelial and mesenchymal cells of several foetal and adult human tissues
(Midgley et al., 1997) and comprises a number of functional domains (Figure 1.9).
The most amino-terminal domain (amino acids 6 to 57) is a series of heptad repeats
that probably mediate APC homo-dimer formation (Groden et al., 1991; Joslyn et al.,
1993). Amino acids 453 to 767 show some homology to the central repeat region of
the Drosophila segment polarity protein armadillo. This domain binds to Asef,
thereby enhancing A sef s interaction with Rac (a member of the Rho family of small
GTPases) which controls cell adhesion and motility via modulation of the actin
cytoskeleton (Kawasaki et al., 2000). The next two motifs, an imperfect repeat of 15
29 amino acids (occurring three times between residues 1020-1169) and a repeat of 20 amino acids (occurring seven times between residues 1262-2033), are involved in |3- catenin binding and downregulation (Rubinfeld et al., 1993; Su et al., 1993). (3- catenin is a central component of the Wnt signalling pathway (see Section 1.4.1.1.2).
Each of the 20 amino acid repeats contains an SXXXS consensus site that acts as a substrate for glycogen synthase kinase 3(3 (GSK3|3) phosphorylation (Munemitsu et al., 1995; Rubinfeld et al, 1996). Interspersed between the 20 amino acid repeats are three SAMP amino acid repeats that mediate binding to the scaffold protein axin
(Kishida et al, 1998). Both GSK3(3 and axin are again members of the Wnt signalling pathway. In addition, several signal motifs are present in APC which are thought to mediate nuclear import and export of the protein (Rosin-Arbesfeld et al,
2000). APC has been shown to bind microtubules when transiently over-expressed in epithelial cells and to trigger tubulin polymerisation in vitro (Munemitsu et al, 1994;
Smith et al, 1994b). Association with microtubules takes place via a region between amino acids 2200 and 2400 (Deka et al, 1998). In addition, the carboxy-terminus of
APC (residues 2560 to 2843) binds the microtubule-associated protein EBl (Su et al, 1995), and the two proteins have recently been shown to co-localise at the ends of microtubules embedded in kinetochores at mitosis (Fodde et al, 2001a; Kaplan et al, 2001). Furthermore, recent studies indicate that mouse embryonic stem (ES) cells harbouring bi-allelic Ape mutations are chromosomally unstable, suggestive of a role of APC in chromosomal segregation (Fodde et al, 2001a; Kaplan et al, 2001).
Consistent with this putative function, three DNA binding domains [S(T)PXX repeats] are present in APC, one within the 20 amino acid repeats, and two in the carboxy-terminal third of the protein (Deka et al, 1999). The carboxy-terminus of
APC has also been shown to bind hDLG, the human homologue of the Drosophila
30 discs large tumour suppressor, and the protein tyrosine phosphatase PTP-BL. This binding occurs via a VTSV amino acid motif in APC, which is a consensus binding motif for the PDZ domain found in both hDLG and PTP-BL (Erdmann et al., 2000;
Matsumine et al., 1996).
Figure 1.9: APC protein domains and their putative functions.
Protein domains of APC include the heptad repeats which mediate APC homo-dimer formation, the armadillo domain which binds to Asef, the 15 and 20 amino acids repeats that are involved in binding and regulating p-catenin, the SAMP amino acid motifs which mediate axin binding, signal-motifs which mediate nuclear localisation and export, several domains for microtubule, EBl and DNA binding that are potentially involved in chromosomal segregation, and a VTSV amino acid motif for the binding of hDLG and PTP-BL.
S(T)PXX repeats Nuclear (DNA binding) export motifs Basic domain Heptad repeats 3x 15-amino acid repeats SAMP-repeats (Microtubule binding, EBl (dim érisation ) (P-catenin binding) (Axin binding) tubulin polymerisation) binding
I I 1020 1169 1262 2200 2400
Armadillo repeats 7x 20-amino acid repeats hDLG and (A sef binding ) (P-catenin binding, GSK3P PTP-BL binding phosphorylation)
31 Due to this diversity of functional domains, APC has been implicated in a variety of cellular processes including cytoskeletal organisation, cell migration, intracellular
adhesion, cell-cycle control and apoptosis (I have reviewed putative APC functions
in detail in REF (Sieber et al., 2000)). However, several lines of evidence suggest
that it is the regulation of Wnt signalling by controlling the cellular level of P-catenin
which is the most important function of APC with respect to tumorigenesis; mutant
APC cannot effectively downregulate p-catenin, thus resulting in clonal expansion.
For example, colorectal tumours without APC mutations sometimes harbour
oncogenic mutations of P-catenin which prevent protein destruction and thus have
effects similar to mutation of APC (Ilyas et al., 1997; Morin et al., 1997; Sparks et
al., 1998), and associations of germline and somatic APC mutations exist which
appear to reflect selection for an optimal level of Wnt signalling (see Section
1.4.1.1.6). However, the newly discovered putative role of APC in chromosomal
segregation could also be of central importance for tumour development, as
chromosomal instability resulting from APC mutation is predicted to accelerate the
multi-step process of colorectal tumorigenesis by targeting downstream tumour
suppressor genes.
1.4.1.1.2 APC and the Regulation of Wnt signalling
The Wnt pathway is primarily effected via cellular p-catenin levels (Figure 1.10). In
the absence of Wnt signalling, APC forms a complex with axin, glycogen synthase
kinase 3-P (GSK3-P) and P-catenin, which facilitates phosphorylation of p-catenin
by GSK3-P (Behrens et al., 1998; Hart et al., 1998b; Ikeda et al., 1998; Kishida et al..
32 1998; Sakanaka et al., 1998; Yamamoto et al., 1998). p-catenin phosphorylation is primed by casein kinase 1-a (CPKl-a) and occurs at four amino-terminal serine/threonine residues (Hinoi et al., 2000; Liu et al., 2002). The phosphorylated protein is a target for ubiquitination by P-transducin repeat-containing protein (P-
TRCP) and subsequent proteasomal degradation (Aberle et al., 1997).
Wnt signalling is initiated by binding of extracellular Wnt ligand to both its transmembrane receptor frizzled and an LRP (low-density lipoprotein receptor- related protein) co-receptor. Binding triggers translocation of the scaffold protein axin to the membrane, where it interacts with the intracellular tail of the LPR co receptor, leading to destabilisation of the p-catenin degradation complex (Bhanot et
al., 1996; He et al., 1997; Mao et al., 2001; Pinson et al., 2000; Tamai et al., 2000;
Wehrli et al., 2000). In parallel, the dishevelled (Dsh) protein is hyperphosphorylated
and inhibits GSK3-p, further repressing P-catenin degradation (Yanagawa et al.,
1995). The resulting accumulation of cytoplasmic and nuclear p-catenin causes an
up-regulation of gene expression. In the nucleus, p-catenin recruits the proteins
Legless/BCL9 and pygopus and interacts with TCF/LEF (T cell factor/lymphoid
enhancer factor) transcription factors, activating transcription of a large number of
target genes including the proto-oncogenes cyclin D1 and c-myc (Belenkaya et al.,
2002; He et al., 1998; Korinek et al., 1997; Kramps et al., 2002; Parker et al., 2002;
Tetsu and McCormick, 1999; Thompson et al., 2002; Van de Wetering et al., 1996).
As a result, Wnt signalling drives cell proliferation in the intestinal epithelium (Pinto
et al., 2003).
33 Figure 1.10: The Wnt signalling pathway.
Adapted from REF (Giles et al., 2003).
A. Absence of Wnt signalling.
Frizzled receptor Plasma membrane
LRP co-receptor
APC Dsh GSK3P 1 p-catenin
p-TrCP
[ p-catenin |(p Nuclear membrane
^ Transcription
Target gene
B. Presence of Wnt signalling.
Frizzled receptor Wnt ligand
Plasma membrane LRP co-receptor
ATP ADP
GSK3P
p-catenin
Nuclear membrane
p-catenin ] ► Transcription M TCF 1 Target gene
34 1.4.1.1.3 Germline Mutations in the APC Gene
Germline APC mutations are detectable in approximately 70% of FAP patients and
are scattered in the 5’-half of the gene (Cottrell et al., 1992; Miyoshi et al., 1992a)
(Figure 1.11). Almost all (-95%) of these changes are nonsense or frameshift mutations that result in protein truncation (Beroud and Soussi, 1996). Two codons,
1061 and 1309, appear to be mutational hot spots accounting for about 11% and 17% of all germline mutations. In addition, some polyposis patients carry an interstitial deletion of chromosome 5q encompassing the APC locus, although the frequency of this change remains to be determined (Barber et al., 1994; Cross et al., 1992; Hockey et al., 1989; Hodgson et al., 1993).
Figure 1.11: Distribution of germline mutations in the gene.
Data were retrieved from the APC mutation database at http://perso.curie.fr/Thierry.
Soussi/APC. html.
8 0 -
160 -
4 0 -
120 -
(K) -
8 -
APC codon no.
35 L4.1.1.4 Genotype-phenotype Associations in FAP
Interestingly, the severity of FAP is at least in part determined by the position of the germline mutation along the APC gene. Germline mutations between codons 168 and
1580 (except for the alternatively spliced part of exon 9) are associated with the
‘classical’ FAP phenotype, with mutations between codons 1250 and 1464 causing particularly severe polyposis (often more than 5000 colorectal polyps) (Miyoshi et al., 1992a; Nagase and Nakamura, 1993). Germline mutations between codons 457 to 1444 are most strongly associated with the development of CHRPE (Caspari et al.,
1995), whilst mutations between codons 1395 and 1560 correlate with the highest
frequency of mandibular osteomas and desmoid tumours (Wallis et al., 1999) (all of
these boundaries are approximate). Perhaps the strongest effect of APC mutation
position on disease severity is observed for germline mutations in the 5’ (codons 78
to 167) and 3’ (after codon 1581) regions and in the alternatively spliced part of exon
9 (codons 312 to 412). These patients develop a milder form of the disease known as
attenuated adenomatous polyposis coli (AAPC or AFAP). AAPC is characterised by
fewer than 100 adenomas and a later onset of colorectal polyposis and cancer. In
addition, there is evidence that phenotypic severity of FAP is influenced by a variety
of genetic and environmental factors, since individuals with identical germline APC
mutations frequently develop dissimilar phenotypes (Paul et al., 1993).
36 1.4.1.1.5 Somatic Mutations in the APC Gene
In most FAP tumours, the second allele of APC is either disrupted by another truncating mutation, or, less frequently, lost. Tumorigenesis in FAP patients is therefore considered to be a good model for sporadic tumorigenesis, as most sporadic tumours are probably also initiated by bi-allelic mutation of APC (see Section 1.3).
In both familial and sporadic tumours, the somatic APC mutations are mainly
confined to the 5’ half of the gene. However, unlike germline mutations,
approximately 80% of somatic mutations are concentrated between codons 1284 and
1580 (Figure 1.12), an area termed the mutation cluster region (MCR), the limits
having been extended slightly from the original ones set by Miyoshi et al (Miyoshi et
al., 1992b). Three somatic mutational hot spots seem to occur at codon positions
1309, 1450 and 1554 accounting for approximately 7%, 8% and 5% of all somatic
mutations respectively.
37 Figure 1.12: Distribution of somatic mutations in the C gene.
Data were retrieved from the APC mutation database at http://perso.curie.fr/Thierry.
Soussi/APC. html.
ISO 160 I U 120 o ,100
NI
o 8
APC c(xJ()n no.
1.4.1.1.6 Associations between Germline and Somatic APC Mutations
Analysis of somatic mutations including allelic loss in early colorectal adenomas of
FAP patients has revealed an interesting correlation with the position of the germline
APC mutation. Germline mutations close to codon 1300 appear to be associated with
loss of the wild-type APC allele. In contrast, patients with germline mutations 3’ and
5’ to these codons predominantly show truncating ‘second hits’ in the mutation
cluster region (Lamlum et al., 1999). A very similar association also exists between
the ‘two hits’ at APC in sporadic colorectal cancers (Rowan et al., 2000). As APC
mutations appear to target domains required for P-catenin binding and
downregulation, these findings probably reflect selection for an optimal level of Wnt
signalling, with mutations in the MCR and specifically around codon 1300 providing
38 the greatest selective advantage. Patients with germline mutations close to codon
1300 already posses the most strongly selected mutation and their adenomas lose the wild-type APC allele as a ‘second hit’. In contrast, patients with germline mutations outside the strongly selected region acquire ‘second hits’ by truncating mutations in the MCR. It has recently been reported that the spectrum of ‘second hits’ is even
more restricted in FAP patients with germline mutations away from codon 1300,
leading to the hypothesis that the level of Wnt signalling has to be ‘just right’ for
tumour development (Albuquerque et al., 2002a). Patients with germline mutations
earlier in the APC were found to generally acquire later truncating somatic
mutations, and vice versa.
1.4.1.1.7 Genetic Testing and Clinical Management
Genetic testing for germline APC mutations in FAP patients is usually restricted to
the 5’ half of the gene. One frequently employed method, the protein truncation test
(PTT), utilises the fact that most germline mutations result in formation of a
translation termination signal. The coding region oiAPC is amplified from cDNA in
overlapping fragments and protein is synthesised in vitro. Truncating germline
mutations are detected as they result in premature termination of the synthetic
peptide. If direct genetic testing is not available, predictive testing using polymorphic
markers linked to the APC gene is useful in families with two or more affected
members (Aaltonen, 2000).
Patients who carry a germline APC mutation or who have one or more first-degree
relatives with FAP and/or an identified APC mutation are at high risk and should be
39 screened with flexible sigmoidoscopy by the age of 10 to 12 years. Patients with colonic polyps and/or a verified APC mutation will require annual endoscopic examination. If too many colonic polyps develop - which is often the case by the early 20s - prophylactic subtotal colectomy is recommended. In addition, patients need annual follow-up by endoscopy as polyps or desmoids may develop in the remaining rectum or at extracolonic sites. Regular endoscopic surveillance of the upper gastrointestinal tract is recommended, but screening for the other less common extracolonic cancers is difficult (with the exception of papillary thyroid carcinoma) and is therefore not generally performed (Lynch and de la Chapelle, 2003).
Chemoprevention trials for the management of polyposis are currently being conducted and have yielded positive preliminary results (see Section 1.1).
1.4.1.2 Attenuated Adenomatous Polyposis Coli
Patients with attenuated adenomatous polyposis coli (AAPC or AFAP; MIN:
175100) are characterised by a milder course of disease than patients with classical
FAP. Affected individuals typically present with fewer than 100 colorectal adenomas and a delay in onset of CRC. The spectrum of extracolonic manifestations is probably also more restricted, although gastric and duodenal polyps are often reported. The penetrance of AAPC, whilst lower than that of FAP, may still be high
(Knudsen et al., 2003; Lynch and Lynch, 1998; Soravia et al., 1998). AAPC is caused by specific truncating germline mutations in the APC gene (see Section
1.4.1.1.4). In addition, some families have been reported to carry a partial or whole- gene APC deletion (Hodgson et al., 1994; Pilarski et al., 1999; Su et al., 2002), although this latter association has been inconsistent.
40 1.4.1.2.1 APC Missense Variants and AAPC
Besides specific truncating germline APC mutations, two missense variants, I1307K and E1317Q, have been associated with attenuated polyposis. The I1307K variant occurs in about 6-9% of Ashkenazi Jews, but is very rare outside this ethnic group. It leads to the creation of an Ag tract that is hypermutable and tends to undergo somatic
slippage to produce frameshift mutations (Laken et al., 1997). This apparently open-
and-shut case may, however, be more complex (Sieber et al., 2003a). Whilst some
studies on Ashkenazi patients with colorectal cancer and/or adenomatous polyps
have associated a relative risk of 1.5-2.0 with the I1307K variant (Frayling et al.,
1998; Gryfe et al., 1999), studies on unselected Israelis and Ashkenazim have found
no significant increase in risk (Figer et al., 2001; Woodage et al., 1998). Moreover,
only about 50% of tumours from 11307K carriers show slippage of the Ag tract
(Gryfe et al., 1998; Laken et al., 1997; Zauber et al., 2003). Other oligonucleotide
tracts within the APC gene are apparently not hypermutable, and 11307K lies at a site
at which germline mutations are associated with very severe FAP and might
therefore have a direct functional effect. Perhaps, therefore, hypermutability is only
one factor in the action of 11307K, and the possibility that the variant has some direct
functional effect remains. The data may also indicate that some 11307K carriers carry
an additional unidentified genetic predisposition to colorectal cancer, presumably
within APC or in tight linkage disequilibrium.
The E1317Q variant is found in more ethnic groups than 11307K, yet the former has
no clear effects on hypermutability (Lamlum et al., 2000a). There is limited evidence
to suggest that E1317Q has direct effects on APC function, perhaps by relatively
41 subtle effects on p-catenin sequestration or degradation (Frayling et al., 1998; Popat et al., 2000; White et al., 1996). However, the association of El 317Q with colorectal adenomas and cancer has been inconsistent and it remains possible the E1317Q is a non-functional rare polymorphism (Evertsson et al., 2001; Popat et al., 2000).
1.4.1.2.2 Clinical Management
Clinical management of AAPC often requires full colonoscopy. It should begin at the
age of 20-25 years and be continued throughout life. Depending on the extent of
polyposis, prophylactic colectomy with ileorectal anastomosis is recommended in
most patients. In addition, regular endoscopy is usually performed for the
management of upper-gastrointestinal disease (Knudsen et al., 2003).
1.4.1.3 Juvenile Polyposis
Juvenile polyposis (JPS; MIM: 174900) is a rare autosomal dominant syndrome. The
disease is characterised by multiple hamartomatous polyps in the colon and
infrequently the stomach and small intestine (Hofting et al., 1993; Smilow et al.,
1966; Veale et al., 1966). Juvenile polyps have a characteristic histology (see Section
1.2.3), and diagnosis is typically made at less than 10 years of age, often as a result
of rectal bleeding, severe diarrhoea and abdominal pain (Heiss et al., 1993).
Although JPS patients develop fewer benign lesions than FAP patients, they remain
at an increased risk of CRC at a young age (Giardiello et al., 1991; Howe et al.,
1998a). Other features which have been associated with JPS include mental
42 retardation, macrocephaly with hypertelorism, heart defects and polydactyly. To aid the diagnosis of JPS, a set of four criteria has been proposed which is summarised in
Table 1.1. Clinical management of JPS sometimes requires prophylactic surgery
(Heiss et al., 1993; Scott-Conner et al., 1995), but regular colonoscopies may be sufficient in cases where the number of polyps is small (Howe et al., 1998a).
Table 1.1: Criteria for the diagnosis of juvenile polyposis syndrome.
Criteria are from REF (Aaltonen, 2000).
1. At least three juvenile polyps of the colorectum. 2. Juvenile polyps throughout the gastrointestinal tract. 3. Any number of juvenile polyps with a family history of JPS. 4. Other syndromes which display hamartomatous gastrointestinal polyps should have been excluded.
JPS is a heterogeneous disorder, and to date germline mutations have been reported in two tumour suppressor genes, SMAD4/DPC4 ?Lnd BMPR1A/ALK3 (bone morphogenic protein receptor 1 A/activin receptor-like kinase 3) (Howe et al., 1998b;
Lynch et al., 1997; Olschwang et al., 1998; Sayed et al., 2002). Both SMAD4 and
BMPRl A are members of the TGF-(3/BMP signalling pathway, which is also a main target in sporadic colorectal tumorigenesis (see Section 1.3).
43 1.4.1.3.1 The TGF-|3/BMP Signalling Pathway
The TGF-|3/BMP signalling pathway is controlled via the activation of receptor- regulated SMAD (R-SMAD) proteins (Figure 1.13). Signalling is initiated when a member of the TGF-(3 superfamily of secreted polypeptide factors binds to a serine threonine kinase receptor complex consisting of two transmembrane proteins, known as the type I and type II receptors. Ligand binding to the type II receptor stimulates recruitment and phosphorylation of the type I receptor, thereby activating its kinase domain (Wrana et al., 1992; Wrana et al., 1994). The activated type I receptor in turn phosphorylates downstream R-SMADs (SMADsl, -2, -3, -5 and -8) at two conserved
C-terminal serine residues (Nakao et al., 1997; Souchelnytskyi et al., 1997).
Phosphorylation of R-SMADs serves a number of functions: it unmasks their nuclear import signals, and leads to their association with the common-mediator SMAD (co-
SMAD), SMAD4. In the case of the R-SMADs SMAD2 and SMAD3, it also triggers their release from the membrane-localised protein SARA, which helps present these substrates to the type I receptor (Souchelnytskyi et al., 1997; Tsukazaki et al., 1998;
Xu et al., 2000). The heteromeric R-SMAD/co-SMAD complexes then translocate into the nucleus - probably via interaction with FG-repeat containing nucleoporins
(Xu et al., 2003; Xu et al., 2002) - and associate with various transcriptional coactivators or corepressors, thereby positively or negatively regulating gene expression (Inman and Hill, 2002; Liu et al., 1997a; Zhou et al., 1998). A number of
TGF-j3/BMP targets have been identified, including p21,PAIl and JUNE
(Beauchamp et al., 1992; Gerwin et al., 1990; Li et al., 1995). Depending on the cell type, TGF-|3/BMP signalling can stimulate or inhibit cell proliferation, differentiation, motility, adhesion or apoptosis.
44 Vertebrates have seven different type I receptors which can associate with five different type II receptors. Three type I receptors transduce signals from TGF-(3-Iike ligands by activating the R-SMADs SMAD2 and SMAD3, whilst four type I receptors transduce signals from BMP-like ligands by phosphorylating the R-
SMADs SMAD I, SMAD5 and SMAD8. TGF-|3/BMP signalling is negatively regulated by the inhibitory SMADs, SMAD6 and SMAD7, and by the ubiquitin- proteasome pathway (reviewed in REF (Attisano and Wrana, 2002)).
Figure 1.13: The TGF-p/BMP signalling pathway.
Adapted from REF (Fodde et al., 2001b).
TGF-p/BMP ligand Plasma membrane TGF-p/BMP receptor complex
ATP ADP
R-SMAD R-SMAD ( P )
Co-SMAD
SMAD
Nuclear Nucleoporin membrane
:o-factor
Ubiquitination Co-SMAD > Transcription and degradation R-SMAD
Target gene
45 1.4.1.4 Peutz-Jeghers Syndrome
Peutz-Jeghers syndrome (PJS; MIN: 175200) is an autosomal dominant disorder characterised by multiple hamartomatous polyps with a distinct histology (see
Section 1.2.3), which predominantly occur in the small intestine but may spread into the colon. In addition, PJS patients often present with freckle-like spots of mucocutaneous melanin pigment which typically involve the face, lips, mouth and anal region. These spots usually develop before the age of five, but may fade as the individual enters puberty. PJS patients are often diagnosed in adolescence with
abdominal pain and anaemia due to intestinal obstruction and bleeding (Utsunomiya
et al., 1975). Although the greatest risk of cancer is in the gastrointestinal tract, PJS
patients may develop tumours at other sites including the breast, ovary, pancreas and
endometrium (Boardman et al., 1998; Giardiello et al., 2000). In addition, rare
tumours may occur such as minimal deviation adenocarcinoma (adenoma malignum)
of the uterine cervix and sex cord tumour with annular tubules (Giardiello et ah,
1987; Hizawa et al., 1993a). A recently proposed set of diagnostic criteria for PJS is
summarised in Table 1.2.
46 Table 1.2: Criteria for the diagnosis of Peutz-Jeghers syndrome.
Criteria are from REF (Aaltonen, 2000).
1. At least three histologically confirmed Peutz-Jeghers polyps. 2. Any number of Peutz-Jeghers polyps with a family history of PJS. 3. Characteristic, prominent mucocutaneous pigmentation with a family history ofPJS. 4. Any number of Peutz-Jeghers polyps and characteristic, prominent, mucocutaneous pigmentation. 5. Other syndromes which display hamartomatous gastrointestinal polyps should have been excluded.
A substantial proportion of PJS families (-50%) and sporadic cases (-30-60%) have
detectable germline mutations which inactivate the serine/threonine kinase LKBl
(STKl 1), often causing truncation or complete loss of the protein (Hemminki et al.,
1998; Jenne et al., 1998). In addition, many familial cases who have no detectable
germline mutation of LKBl are positive for linkage to chromosome 19pl3.3, the
region comprising the LKBl locus. This may indicate the presence of cryptic LKBl
mutations, epigenetic changes, or alterations in another causative gene in close
proximity to LKBl.
The function of the LKBl gene is not yet fully understood. The protein seems to be
required for foetal development and is present in apoptotic intestinal cells (Karuman
et al., 2001; Ylikorkala et al., 2001). In addition, LKBl has been implicated in the
regulation of p53-dependent apoptosis and the control of cell proliferation (reviewed
in REF (Yoo et al., 2002)).
47 1.4.1.5 Cowden’s Syndrome and Bannayan-Riley-Ruvalcaba Syndrome
The most prominent feature of Cowden’s syndrome (MIN: 158350), an autosomal dominant disorder, is the development of multiple mucocutaneous lesions. These include facial trichilemmomas, acral keratoses and oral mucosal papillomatosis
(Brownstein et al., 1979; Starink et al., 1985). About 40% of patients develop hamartomatous polyps that may affect the entire gastrointestinal tract (Hizawa et al.,
1994; Waite and Eng, 2002). Polyps are often inflamed or fibrotic or contain
excessive smooth muscle in the centre (Oberhuber and Stolte, 2000). In addition,
Cowden’s syndrome patients are at an increased risk of developing tumours of the
thyroid, breast and genito-urinary tract (Salem and Steck, 1983; Starink et al., 1986).
Patients with Bannayan-Riley-Ruvalcaba syndrome (MIN: 153480), on the other
hand, are typified by macrocephaly, lipomatosis, hemangiomatosis and a speckled
penis and may also develop hamartomatous polyps (Gorlin et al., 1992). Due to the
presence of hamartomatous polyps, an increased risk of CRC has been suggested in
both syndromes but not yet been demonstrated.
Both disorders are caused by germline mutations in the PTEN (phosphatase and
tensin homologue deleted on chromosome 10) tumour suppressor gene (Liaw et al.,
1997; Marsh et al., 1997; Marsh et al., 1999). PTEN is a dual-specificity phosphatase
which appears to have a variety of pro-apoptotic functions including negative
regulation of the phoinositol-3-kinase (PI3K)/AKT and the protein-
phosphatase/MAPK survival signalling pathways (reviewed in REF (Waite and Eng,
2002)).
48 1.4.1.6 Hereditary Mixed Polyposis Syndrome
Hereditary mixed polyposis syndrome (HMPS; MIM: 601228) is an autosomal dominant disorder characterised by the development of a variety of different colorectal tumours which appear to be confined to the large bowel. These include atypical juvenile polyps, hyperplastic polyps with areas of dysplasia (serrated adenomas), classical adenomas, and carcinoma (Whitelaw et al., 1997). The HMPS locus has recently been mapped to chromosome 15ql3-ql4 in a large family of
Ashkenazi decent, and has been found to be allelic to a previously identified
Ashkenazi susceptibility locus CRACl (Jaeger et al., 2003). Notably, some
Ashkenazi families sharing the minimal HMPS/CRACl region haplotype have been reported to only develop multiple, classical colorectal adenomas and carcinoma
(Jaeger et al., 2003). Whether HMPS/CRACl is also important outside of the
Ashkenazi population remains to be determined.
1.4.2 Hereditary Non-polyposis Colorectal Cancer
Hereditary non-polyposis colorectal cancer (HNPCC; MIN: 114500) was originally described as two distinct autosomal dominant disorders (Lynch Syndrome I and II), but is now considered a single entity. HNPCC is the most common hereditary predisposition to CRC known to date (see Section 1.4), and has a penetrance of 70-
90%. In contrast to AAPC and FAP, HNPCC patients do not develop multiple adenomatous polyps. Besides colorectal cancer, patients with HNPCC frequently develop cancer of the endometrium, ovary, stomach (particularly in Asian countries such as Japan and Korea), ureter or renal-pelvis, brain, small-bowel, hepatobiliary
49 tract, and skin (sebaceous tumours) (Lynch et al., 1993; Watson et al., 1994). Due to this phenotypic variability two sets of diagnosis criteria have been proposed by the
International Collaborative Group on HNPCC (Table 1.3). The first set of criteria is solely based on the inheritance of CRC, whilst the second set also considers extracolonic tumours. In addition, some patients develop multiple sebaceous skin tumours and keratoacanthomas producing a diagnosis of Muir-Torre syndrome
(Lynch and Fusaro, 1999), or glioblastomas producing a diagnosis of Turcot’s
syndrome type 2 (Hamilton et al., 1995; Paraf et al., 1997; Van Meir, 1998). One
important feature of HNPCC is the low age of CRC onset. Patients often present
between 40 and 45 years of age (or younger), which is about 20-30 years earlier than
the average sporadic case (Lynch et al., 1991; Mecklin and Jarvinen, 1991; Vasen et
al., 1990).
Table 1.3: Amsterdam I and Amsterdam II criteria for the diagnosis of
hereditary non-polyposis colorectal cancer.
Criteria are from REFS (Vasen et al., 1991; Vasen et al., 1999).
Amsterdam I criteria At least three relatives must have histologically verified colorectal cancer (CRC): One must be a first-degree relative of the other two. At least two successive generations must be affected. In at least one of the affected relatives, CRC must have been diagnosed before the age of 50 years. Polyposis syndromes must have been excluded.
50 Table 1.3: continued
Amsterdam II criteria At least three relatives must have a cancer associated with HNPCC (colorectal, endometrial, stomach, ovary, ureter or renal-pelvis, brain, small-bowel, hepatobiliary tract, or skin [sebaceous tumours]): One must be a first-degree relative of the other two. At least two successive generations must be affected. In at least one affected relative, the HNPCC-associated cancer should have been diagnosed before the age of 50 years. Other polyposis syndromes should have been excluded in any relative with CRC.
In contrast to sporadic CRCs that predominantly arise in the distal colorectum
(-65%), HNPCC cancers tend to arise in the proximal colorectum (-60-70%) (Lynch
et al., 1988; Lynch et al., 1993; Vasen et al., 1990). Histologically, CRCs of HNPCC
patients are often poorly differentiated and about 40% are of a mucinous type as
compared to about 20% of sporadic CRCs (Lynch and Lynch, 1998; Mecklin et al.,
1986). Furthermore, HNPCC cancers appear to have a relatively good prognosis
(Sankila et al., 1996).
Defects in four different DNA mismatch repair genes, MLHl (human homologue of
E. coli M utL\ MSH2 (human homologue of E.coli MutS), MSH6 and PMS2 (human
homologue ofS. cerevisiae postmeiotic segregation increased) have been identified
in HNPCC families (Akiyama et al., 1997; Bronner et al., 1994; Leach et al., 1993;
Miyaki et al., 1997; Nicolaides et al., 1994; Papadopoulos et al., 1994). Germline
mutations of M LH l and MSH2 account for approximately 90% of all mutation-
51 positive cases reported to date in the HNPCC mutation database
(http://www.nfdht.nlV Mutations in MSH6 account for the majority of the remaining
cases, but may also be involved in atypical or more benign forms of the syndrome
(Miyaki et al., 1997; Wagner et al., 2001). In addition, germline changes in MLH3 have been reported in a single study (Wu et al., 2001). Immunohistochemical
analysis of DNA mismatch repair proteins in the tumour can provide clues as to
which DNA mismatch repair gene is involved in tumour pathogenesis, if staining for
one of the proteins is very weak or absent (Lindor et al., 2002).
Bi-allelic inactivation of any of these DNA mismatch repair genes appears to lead to
the rapid accumulation of unrepaired mutations throughout the genome. This defect
manifests as microsatellite instability (MSI; Figure 1.14), somatic changes in the
lengths of simple tandem nucleotide repeats, and probably accelerates tumorigenesis
by targeting tumour suppressor genes such as the TGF-P type II receptor (Aaltonen
et al., 1993; Ionov et al., 1993; Thibodeau et al., 1993). Typically, at least half of all
tumours from HNPCC patients show MSI at more than 50% of loci when using the
Bethesda panel of markers, and more than 90% of tumours when using the marker
BAT26 (Loukola et al., 2001). Studies by flow cytometry have revealed that HNPCC
carcinomas are often diploid, suggesting that the two forms of instability are
mutually exclusive (Kouri et al., 1990).
52 Figure 1.14: Microsatellite instability at BAT26 in a colon cancer from a patient with HNPCC.
Fluorescently-labelled PGR product was run on an automated sequencer. Germline
DNA from blood shows a single peak, indicating that the patient is homozygous for the mononucleotide repeat marker BAT26. Tumour DNA shows, in addition to the normal allele (single arrow), a new allele (double arrow) that has lost approximately five nucleotides. The figure is adapted from REF (Lynch and de la Chapelle, 2003).
Germ-Line DNA from Blood
Homozygous for 1 allele (120 bp) ------^ .
BAT26 marker
DNA from Tumor Containing >5096 Tumor Cells
ft t BAT26 marker
Due to the accelerated carcinogenesis in HNPCC, 1-3 yearly colonoscopy is recommended in patients with strong clinical evidence or documented germline mutations in MLH2, MSH2, or MSH6 (or a combination), starting between the ages of 20 and 25 years. Upper-gastrointestinal endoscopy is performed if there is evidence of gastric or small bowel cancer in the family. For the management of endometrial cancer, annual endometrial aspiration and transvaginal ultrasonography
53 is recommended, starting at the age of 30 years. Transvaginal ovarian ultrasonography and CA-125 screening is usually also performed from the age of 30 years for the detection of ovarian tumours. Screening for the other less common extracolonic malignancies is more difficult and generally considered on a case-by-
case basis (Lynch and de la Chapelle, 2003).
1.5 The Multiple Colorectal Adenoma Phenotype
The multiple colorectal adenoma phenotype - as defined herein - is characterised by
3 to 100 synchronous or metachronous adenomatous polyps in the colorectum, late
onset of colorectal cancer and generally the absence of extracolonic features. The
disease can be inherited as a Mendelian trait, either autosomal dominant or recessive,
but can also occur in the form of isolated cases. About 10% of patients with multiple
adenomas carry a truncating germline mutation in specific regions of the APC gene,
and therefore have AAPC. In addition, two APC missense variant I1307K (in
patients of Ashkenazi origin) and E1317Q appear to be involved. Some patients have
also been reported to harbour partial or whole-gene APC deletions, although this
association has been inconsistent and no data on the frequency of large APC
deletions aavailable (see Section 1.4.1.2). The genetic basis underlying the
remaining cases is largely unknown, although there is evidence that some multiple
adenoma patients of Ashkenazi origin carry HMPS/CRACl mutations (see Section
1.4.1,6). Further good candidates include members of the WNT and TGFp/BMP
signalling pathways, both of which are central to colorectal tumorigenesis (see
Sections 1.4.1.1.2 and 1.4.1.3.1). In addition, a recent study has indicated that
recessive germline mutations in the base excision repair gene MYH (human
54 homologue ofE.coli MutY) may play a role (Al-Tassan et al., 2002). Such changes were identified in a single Welsh family and associated with somatic G:C-»T:A
transversion in the APC gene, a mutation-signature characteristic of defective base
excision repair.
1.5.1 The Base Excision Repair Pathway
The base excision repair pathway is utilised by cells to prevent and repair base-pair
mutations induced by reactive oxygen species (ROS). ROS such as superoxide,
hydrogen peroxide and hydroxyl radicals are the product of normal aerobic
metabolism and are usually detoxified by chemical antioxidants (e.g. glutathione,
ascorbate, and uric acid) and antioxidant enzymes (e.g. superoxide dismutase,
catalase, and glutathione peroxidase). However, cellular redox homeostasis may be
disrupted by exposure to exogenous toxins (e.g. smoking), radiation or chronic
inflammation. At increased levels, ROS induce DNA damage such as single- and
double-strand breaks, abasic sites and a variety of nucleotide modifications
(reviewed in REFS (Halliwell, 1996; Wang et al., 1998)). One particularly mutagenic
product of oxidative DNA damage is the stable guanine adduct 8-oxo-7,8-dihydroxy-
2’-deoxy guano sine (8-oxo-dG). 8-oxo-dG readily mispairs with adenine residues
(Shibutani et al., 1991) and has been shown to cause spontaneous G:C-»T:A
transversion mutations in both BER-deficient bacteria and yeast cells (Michaels and
Miller, 1992; Nghiem et al., 1988; Thomas et al., 1997). Levels of 8-oxo-dG have
also been found to be significantly increased in carcinomas of the breast (Malins and
Haimanot, 1991), lung (Jaruga et al., 1994; Olinski et al., 1992) and kidney
(Okamoto et al., 1994).
55 The products of three human base excision repair genes, MTHl (human homologue of E.coli MutT),OGGI (human homologue of E.coli MutM) and MYH, act synergistically to prevent 8-oxo-dG induced mutagenesis (Figure 1.15). In the nucleotide pool, MTHl hydrolyses 8-oxo-dGTP to 8-oxo-dGMP (Kang et al., 1995;
McLennan, 1999; Mo et al., 1992; Oda et al., 1997; Sakumi et al., 1993); OGGI
detects and removes 8-oxoG adducts that are incorporated into DNA during DNA
replication (Lu et al., 1997; Nishioka et al., 1999; Radicella et al., 1997; Roldan-
Arjona et al., 1997; Shinmura et al., 1997); and MYH, an adenine-specific DNA
glycosylase, scans the daughter strand after DNA replication and removes adenines
mispaired with 8-oxo-dG, possibly in a complex with MutSa (a heterodimer of
MSH2 and MSH6) (Ohtsubo et al., 2000; Slupska et al., 1996; Slupska et al., 1999;
Takao et al., 1999). All three base excision repair proteins act in both the nucleus and
mitochondria, probably in different isoforms resulting from alternative splicing
(Kang et al., 1995; Nishioka et al., 1999; Oda et al., 1997; Ohtsubo et al., 2000;
Takao et al., 1999).
56 Figure 1.15: The base excision repair pathway.
Taken from REF (Gu et al., 2002).
nnMcMi dGTP IÏÏTÏÏÏÏ ^ ÏÏÏÏTÏÏT
Oxidative damage Oxidative damage Replication
RepUcaflonjzi d cfxP c| RepUcaüon \ ^ > HIIIIH a 11111111
hOGGl Repiication 1 hMTH (hMYI^
nnzEciniiin nmniTnnnn dG°MP ÏÏÏÏÏÏÏÏ - ÏÏÏÏÏÏÏÏ
Replication 7 (hO G Cl) (hOGGljN ^ V mnnifinnîiî nnnnlniM TÏÏÏÏTIÎ o-oxo-gaanine
57 1.6 Aims of this Thesis
This project was aimed at characterising the genetic basis of the multiple colorectal adenoma phenotype. Studies were designed to clarify the role of germline APC mutations, and to assess the role of putative candidate genes:
1. To clarify the inconsistent association of whole-gene APC deletions and
AAPC/multiple adenomas, a real-time quantitative multiplex (RQM) PCR
assay was developed and used to screen APC mutation-negative patients with
both attenuated and classical polyposis.
2. To identify the molecular mechanism(s) causing attenuated rather than
classical disease in patients with specific germline APC mutations,
associations of germline and somatic APC changes were investigated in both
patients with AAPC/multiple adenomas and classical TAP.
3. To assess putative relationships between APC mutations and chromosomal
instability (GIN), adenomas with characterised bi-allelic APC mutations were
investigated for evidence of such changes.
4. To determine the contribution of germline mutations in MYH and related base
excision repair genes to multiple adenomas and possibly classical polyposis,
multiple adenoma patients and APC mutation-negative, classical polyposis
patients were analysed.
5. To investigate the role of germline mutations in the WNT (besides APC) and
TGF-(3/BMP signalling pathways in causing multiple adenomas, selected
members of these pathways were screened in multiple adenoma patients.
58 Chapter 2 Materials and Methods
Recipes for solutions and media shown in bold are given at the end of this chapter.
2.1 DNA Extraction
2.1.1 DNA Extraction from Blood
DNA extraction from fresh-frozen blood was performed using an ammonium acetate method. The procedure involved cell lysis by hypotonic shock, nuclei lysis (and further cell lysis) by detergent action and proteinase K digestion, and precipitation of
DNA with ammonium acetate and ethanol.
A total of 9ml of fresh-frozen, EDTA-containing blood was thawed and transferred into a 50ml polypropylene tube (Coming). The cells were lysed by adding ice-cold distilled water (dH20) to a final volume of 50ml. The sample was mixed by inversion, and cells were harvested by centrifugation at 2300rpm for 20 minutes at
4°C in a swing-out rotor centrifuge (CR412 Jouan). The supernatant was discarded, and the tube was placed upside down on a paper towel to remove the last traces of supernatant. The cell pellet was washed in 25ml 0.1% NP-40 at 4°C. Nuclei were lysed by adding 3ml nuclei lysis buffer, 200pl 10% SDS, and 600pl proteinase K solution. The sample was vortexed until the pellet was completely resuspended and incubated for 2 hours at 60°C. 1ml of saturated ammonium acetate was added and the tube was vortexed for 15 seconds. The sample was incubated at room
59 temperature for 10-20 minutes and centrifuged at 2300rpm for 20 minutes at room temperature. The supernatant was transferred to a clean 50ml polypropylene tube and two volumes of ice-cold 100% ethanol (BDH) were added. The DNA was precipitated by gently inverting the tube, and spooled out using a plastic inoculating loop. The spooled DNA was dipped into an Eppendorf tube containing 70% ethanol to remove excess salt, and transferred to a labelled screw-capped Eppendorf tube.
After air-drying, the DNA was re-suspended in 1ml dHiO.
2.1.2 DNA Extraction from Cell Lines
DNA was extracted from lymphoblastoid or colorectal cancer cell lines using a high salt method, thus avoiding the use of toxic phenol to separate DNA from protein debris. The protocol involved cell lysis by detergent action and proteinase K digestion, degradation of ribonucleic acids (RNAs) as these can interfere with some downstream applications, separation of DNA from protein debris using sodium chloride and chloroform, and precipitation of DNA with isopropanol.
Approximately 5x10^ cells were transferred to a 50ml polypropylene tube (Coming).
The sample was centrifuged at 1500rpm for 10 minutes at room temperature (CR412
Jouan centrifuge), and the supernatant was discarded. The remaining cell pellet was washed twice in PBS and resuspended in 15ml SE buffer. To degrade RNAs, 50|nl of lOmg/ml RNaseA was added and the mixture was incubated for 1 hour at 37°C.
Proteinase K (BDH) was added to a final concentration of 200p,g/ml, and cells were lysed by overnight incubation at 55°C. To separate DNA from protein debris, 4.5ml of 5M sodium chloride (final concentration 1.5M) and 20ml of chloroform (BDH)
60 were added. The tube was rotated for 30 minutes and centrifuged at 2000rpm for 10 minutes at room temperature. The DNA-containing aqueous layer was transferred to a new 50ml tube, and an equal volume of isopropanol (BDH) was mixed in for 5 minutes to precipitate the DNA. After spooling out the DNA using a plastic inoculating loop, it was washed in 70% ethanol for 1 hour to remove excess salt.
After air-drying, the DNA was resuspended in 1ml dHiO.
2.1.3 DNA Extraction from Fresh-frozen Tissue
DNA was extracted from fresh-frozen tissue using the QIAamp DNA Mini Kit
(Qiagen) according to manufacturer’s instructions. This method permits isolation of
10-30pg of DNA from 25mg of tissue. Briefly, the tissue was cut into small pieces, lysed with proteinase K, treated with RNaseA to degrade RNAs, precipitated with ethanol and added to a spin column to bind the DNA. After a series of washes to remove protein debris, the DNA was eluted from the column with dHzO.
2.1.4 DNA Extraction from Formalin-fixed. Paraffin-embedded Tissue
DNA extraction from formalin-fixed, paraffin-embedded tissue involved manual dissection of dewaxed tissue from unstained slides and cell lysis by proteinase K digestion. As formalin-fixation causes cross-linking and breakage of DNA, this method generally yields highly fragmented product of poor quality.
61 Depending on the size of the tissue sample, 5-8 sections of 10p,m were cut from each paraffin-block onto non-coated glass slides using a microtome (Reichert-Jung). The tissue was dewaxed in xylene (BDH) for 10 minutes, and washed twice in 100% ethanol (BDH) for 10 minutes. One section was stained with haemotoxylin and eosin
(Merck) according to manufacturer’s instructions, and inspected using an Eclipse
400 light microscope (Nikon). Haemotoxilin, a natural dye produced from the heartwood of the tree Haematoxylon campechianum, stains nuclei in purple, whilst eosin, a halogenated derivative of fluorescein, stains cytoplasm in red. Areas of interest were dissected from unstained slides using a sterile needle and a small volume of proteinase K buffer [500p,l lOx Mg^^-free PCR buffer (Promega), 50p,l
20mg/ml proteinase K (BDH), 4.45ml dH20]. The tissue was transferred into a screw-capped Eppendorf tube, and lysed in 50-100p,l of proteinase K buffer by incubation for 2 days at 55°C. Proteinase K was inactivated by heating to 95°C for
10 minutes. The sample was centrifuged at 13000rpm for 15 minutes in a 5415C bench-top centrifuge (Eppendorf). The DNA-containing supernatant was transferred to a fresh Eppendorf tube.
2.2 Spectrophotometry
To assess DNA quantity and quality, aliquots were diluted in dH20 and analysed by spectrophotometry at 260nm and 280nm using a SPECTRAmax PLUS spectrophotometer. As an optical density (OD) of 1 at 260nm corresponds to about
50pg/ml double-stranded DNA, sample concentrations in pig/ml were calculated as
‘dilution factor x 50 x OD 260 • OD260/OD280 ratios were calculated to assess the DNA quality. Pure DNA samples have an OD 260/OD280 ratio of 1.8. RNA contamination 62 will increase the ratio towards 2.0 while protein contamination will decrease the ratio towards 1.4.
2.3 Polymerase Chain Reaction
The polymerase chain reaction (PCR) is an in vitro technique for DNA synthesis, which allows selective amplification of a specific DNA sequence. The reaction requires two oligonucleotide primers, which hybridise to opposite strands and flank the target region in the template DNA. In addition, the reaction requires the four deoxyribonucleoside triphosphates dATP, dTTP, dGTP and dCTP, and a thermostable DNA polymerase. A PCR reaction typically consists of 35 cycles, and each cycle has three steps: (i) Dénaturation of the DNA template, (ii) annealing of the primers to the separated template strands, and (iii) extension of the primers by
DNA polymerase. The first PCR cycle creates two new strands of variable length per
DNA template, which can act as targets in the second cycle. The third cycle produces double-stranded DNA molecules which comprise precisely the target region defined by the two primers. The following cycles result in an exponential doubling of this target fragment which soon becomes the predominant reaction product. PCR amplification is not infinite, and the desired fragment gradually stops to accumulate.
Several factors determine the point at which the reaction plateau is reached including the utilisation of substrates (dNTPs or primers), the stability of DNA polymerases, and the competition for reactants by non-specific products or primer-dimers.
PCR primers were designed using the PrimerB programme of the Whitehead Institute of Biomedical Research fhttp://www-senome.wi.mit.edu/cgi-bin/r)rimer/primer3
63 WWW.c g i). A typical 25pl PCR reaction contained 30-50ng of template DNA,
0.4pmoles of each primer, Ix Mg^^-free PCR buffer (Promega), 2mM MgCb
(Promega), 0.2mM of each dNTP (Amersham) and 1 unit of PIC Taq DNA polymerase (Cancer Research UK). DNA samples were aliquoted into a 96-well plate (ABgene), and a PCR master mix was made up with the remaining reaction components. The master mix was vortexed and added to the samples. The plate was sealed with Thermowell Sealers (Coming) and immediately mn on a Tetrad PCR
Machine (MJ Research). A standard PCR reaction consisted of an initial dénaturation at 94°C for 5min, 35 cycles of 94°C for Imin, 55°C for Imin, and 72°C for Imin, and a final extension at 72°C for lOmin. Depending on the primer pair and target size, variations of these conditions (Mg^^ concentration, additives, annealing temperature, extension time) were necessary. These are listed together with the primer sequences in the Appendix of this thesis. PCR products were analysed by using agarose gel electrophoresis and ethidium bromide staining.
2.4 Agarose Gel Electrophoresis
Agarose gel electrophoresis permits separation of DNA molecules according to their size. In solution, DNA molecules carry a negative charge and hence migrate to the anode of an electrical field. In a 3-D matrix such as an agarose gel, the speed of migration depends on the size of the DNA molecule and the effective ‘pore’ diameter. Large molecules migrate slower than small molecules. DNA molecules can be visualised by ethidium bromide staining. Ethidium bromide is a dye which intercalates between the base pairs of DNA and fluoresces under UV light of 302nm.
64 Agarose gel electrophoresis was performed using 1-2% gels depending on the size of
DNA fragments to be analysed. An appropriate amount of agarose powder
(Invitrogen) was weighed and dissolved by boiling in Ix TBE buffer. The molten agarose was allowed to cool to approximately 60°C, and ethidium bromide was added to a final concentration of 0.25|ig/ml. The solution was poured into a gel- casting tray (Bio-rad) with a comb in position and left to set. The comb was removed, the gel was placed in an electrophoresis chamber (Bio-rad) and covered with Ix TBE buffer. For a standard PCR product, lOpl of DNA were combined with
3jxl of 5x Orange G Loading Dye (Trevigen) and loaded into a well of the gel. 10p,l of Ikb ladder (Gibco BRL) prepared according to manufacturer’s instructions were loaded into an adjacent well to allow sizing of the PCR product. The gel was connected to a DC current Device (Bio-rad PowerPac300), and typically run at 130V for 20 minutes. The DNA was visualised by placing the gel under UV light of 302nm in a White/2UV Transilluminator (UV Products), and photographs were taken using a UV products camera.
2.5 Mutation Detection
2.5.1 Single-strand Conformation Polymorphism Analysis
Single-strand conformation polymorphism (SSCP) analysis allows detection of nucleotide substitutions, deletions, insertions and inversions in DNA fragments ranging from 100 to 500bp in size. The technique involves PCR amplification of the
DNA target, dénaturation of the double-stranded PCR product by heating in the presence of a denaturing chemical, and separation of the generated single-strands in a
65 non-denaturing gel matrix. The non-denaturing gel conditions ensure that the single stranded DNA fragments form unique secondary structures which are determined by their primary sequence and are stabilised by hydrogen-bonding. Since complementary strands have different primary sequences, they adopt different conformations. Mutation detection relies on electrophoresis under conditions where mutant strands have different conformations and hence different migration speeds than wild-type strands. Since the formation of secondary structures is highly dependent on variable such as temperature and ionic strength, some mutations may only be detected under certain running conditions. Notably, single-stranded fragments may adopt more than one conformation.
For SSCP analysis, PCR amplification was performed using fluorescently labelled forward and reverse primers (FAM, TET or HEX dye-labels). PCR products were diluted with dHiO and multiplexed depending on size and dye-label as appropriate.
4^1 aliquots were then transferred into a 96-well optical reaction plate (PE-Applied
Biosystems), and 0.5^1 0.3M sodium hydroxide, 0.5^1 TAMRA-350 size standard
(PE-Applied Biosystems), and lOpl formamide (BDH) were added. The samples were denatured at 95°C for five minutes, plunged onto ice, and then run on the
ABI3100 sequencer (PE-Applied Biosystems). The ABI3100 sequencer is a capillary-based system which provides for higher throughput and greater detection sensitivity than gel-based systems. All samples were run at two different temperatures, 18°C and 24°C, using 36cm capillaries containing 5% native polymer
(PE-Applied Biosystems), 10% glycerol (BDH) and Ix TBE. The Ix TBE running buffer contained 10% glycerol. Samples were analysed for band-shift using
GENESCAN and GENOTYPER software (PE-Applied Biosystems). Mutant
66 samples were reamplified and sequenced in both forward and reverse orientation (see
Section 2.5.3).
2.5.2 Protein Truncation Test
The protein truncation test (PTT) involves PCR amplification of coding DNA (often cDNA) using a forward primer tagged with a T7 RNA polymerase promoter, a ribosome binding site, and an in-frame start codon. Jin -coupled transcription and translation (IVTT) is performed on the tagged PCR product and the resulting peptide is run on a polyacrylamide gel. The presence of a nonsense or truncating frameshift mutation can be detected as these cause premature termination of protein translation.
The protein product is visualised by incorporation of into the IVTT reaction and exposure to X-ray film.
PCR primers amplifying coding DNA were designed using the PrimerS programme
(see Section 2.3). The following oligonucleotide tag was added to the forward primer; the T7 RNA polymerase promoter, the ribosomal binding site and the in frame start codon are indicated:
5'GGATCC7W^7WCG/4CrC4C7W7WGGAACAGACCACCATGGAACAAAAATT
AATATCGGAAGAGGATTTGAAT3 ’
IVTT reactions were performed using the TNT Rabbit Reticulocyte Lysate System
(Promega) according to manufacturer’s instructions. In brief, a master mix for 40 reactions was prepared from 200pl lysate, 16p,l TNT buffer, 8pi amino acids minus methionine, 8 pi T7 RNA polymerase, 8 pi RNase inhibitor (Boehringer Manheim),
67 and 16^1 methionine. 6pil PTT master mix were added to 4pil PCR product in a
96-well plate (ABgene). The plate was sealed, covered with 3MM Watman paper to absorb any evaporation, and incubated for 1-2 hours at 30°C on a Tetrad PCR machine (MJ Research).
The IVTT products were separated by polyacrylamide gel electrophoresis (PAGE).
Each polyacrylamide gel consisted of a lower resolving gel and an upper stacking gel. A master mix sufficient for two 12% resolving gels was prepared using 8ml acrylamide (40% w/v, acryl/bis 37.5:1; Amresco), 5ml Protogel Resolving Buffer
(National Diagnostics) and 7ml dHiO. For each gel, two clean plates were assembled with a gasket acting as a spacer, and two bulldog clips holding the assembly together. lOOp.1 of 20% w/v ammonium persulphate and 20p,l TEMED (Sigma) were added to the resolving gel mix, the solution was swirled and poured into the glass plate assembly. SOOpl of dH20 were added to the top of each resolving gel to create a straight edge. While the two resolving gels polymerised, 1% stacking gels were prepared by mixing 2.5ml acrylamide (40% w/v, acryl/bis 37.5:1; Amresco), 2.5ml
Proto Gel Buffer (National Diagnostics) and 8ml dHzO. The dHiO was discarded from the top of the resolving gels, and each gel was washed with 500pl of the unpolymerised stacking gel mix. A 12-well shark-tooth comb was added at an angle to each gel assembly. 20pl TEMED (Sigma) and lOOpl 20% w/v ammonium persulphate were added to the remaining stacking gel mix before pouring the solution onto the lower gels. The combs were straightened and excess solution was wiped away. Once the two gels had set, clips and gaskets and combs were carefully removed, and the wells were rinsed and straightened using a syringe filled with PTT running buffer. The gels were clamped into an electrophoresis chamber (GRI) and
68 the top and bottom reservoir were filled with PTT running buffer. Before loading the IVVT products, 10|li1 of PTT loading buffer were added, the samples were re sealed, covered with 3MM Watman paper and denatured at 95°C for 5 minutes. 15pl of denatured sample were loaded per well, with 7.5pl of multicoloured protein marker (NEN) serving as a size standard. The gels were run at 60mA for 1-1.5 hours until the loading buffer had reached the bottom.
Following PAGE, the PTT running buffer was carefully discarded in the hood, and the plates were dissembled on paper towels to prevent isotope contamination of the bench surface. The gels were placed into PTT fixing solution until the bromophenol blue in the sample buffer had turned green. A gel dryer was heated to 80°C, the gels were placed between two 3MM Whatman papers and dried under vacuum for 1 hour.
After drying, the gels were taped into an X-ray cassette, and overlaid with X-ray film
(Hyperfilm MP, Pharmacia) in the darkroom. After overnight exposure, the film was developed in an automatic developer. Mutant samples were reamplified and sequenced in both forward and reverse orientation.
2.5.3 Fluorescence Cycle-sequencing
Fluorescence cycle-sequencing is an automated method of DNA sequencing which allows to use double-stranded PCR products as template. The technique is based on the dideoxy procedure which was developed by Sanger et al. in 1977 (Sanger et al.,
1977). The procedure relies on the ability of DNA polymerase to synthesise complementary copies of a single-stranded target DNA. DNA synthesis is performed in the presence of one specific primer, the four dNTPs and a low concentration of
69 their dideoxy analogues, each of which is labelled with a different fluorophore. The standard reaction involves 34 cycles of heat dénaturation, primer annealing and extension. Since dideoxy nucleotides lack the 3’-hydroxyl group required for extension, chain termination occurs whenever one of them is incorporated into the growing chain. Due to the low concentration of dideoxy nucleotides, termination events occur at each nucleotide along the target sequence. The generated strands thus have a common 5’-end defined by the primers but vary in length to their base- specific 3’-ends. The different strands can be separated by PAGE. Since the chain- terminating events have specifically attached one of the four fluorophores, the DNA sequence can be read as the strands pass a fluorescence detector.
PCR products for DNA sequencing were purified from unincorporated primers and nucleotides using the QIAquick PCR Purification Kit (Qiagen) according to the manufacturer’s instructions. This kit exploits the differential binding properties of
DNA to a silica-gel membrane under different salt conditions; proteins and oligonucleotides smaller than lOObp are washed through the membrane. The purified
PCR products were subjected to standard DNA sequencing reactions using the same forward or reverse primer as for the original PCR. Each 20pl reaction contained 8pi
BigDye Terminator Ready Reaction Mix (PE-Applied Biosystems), 0.5pl primer
(20pM; forward or reverse), and 5-lOpl purified PCR product. Cycle conditions on a
Tetrad PCR machine (MJ Research) consisted of an initial dénaturation at 94°C for
4min, 25 cycles of 94°C for 30sec, 50°C for lOsec and 60°C for 4min, and a final extension at 60°C for 7min. The sequencing products were purified from excess dye- labelled dideoxy nucleotides using the QIAquick PCR Purification Kit, but with a final elution in 8pi microSTOP loading buffer (PE-Applied Bioscience). Samples
70 were denatured at 94°C for 4min and run on a 5% polyacrylamide gel on a semi automated ABI Prism 377XL DNA sequencer (PE-Applied Biosystems). Sequences analysis was performed using the Semi-adaptive Base Calling option of the
Sequencing Analysis Programme Version 2.1 (PE-Applied Biosystems). The NCBI
BLAST database was searched to ensure correct origin of the sequenced fragment rhttp://www.ncbi.nlm.nih.gov/blast/), and all sequences were analysed by eye for alterations.
2.5.4 Real-time Quantitative Multiplex PCR
Real-time quantitative multiplex (RQM) PCR determines dosage of a DNA target by monitoring PCR amplification in real-time and making use of the 5' exonuclease activity of Taq DNA polymerase. In addition to measuring relative transcript abundance, the method can be used to detect both hetero- and homozygous gene deletions. Besides the standard PCR components, this assay requires an excess of an oligonucleotide probe which is specific to a sequence between the two primers. The probe is labelled with two different fluorophores, a reporter dye (FAM or VIC) at the
5'-end and a quencher dye (TAMRA) at the 3'-end. While the probe is intact, the emission of the reporter dye is absorbed by the quencher dye. During the extension phase of each PCR cycle, Taq DNA polymerase cleaves the oligonucleotide probes annealed between the two primers. Probe cleavage separates the reporter from the quencher dye. The resulting increase in reporter emission can be detected using the
ABI PRISM 7700 Sequence Detection System (PE-Applied Biosystems) and is directly proportional to the amount of product being generated each PCR cycle. By
71 using probes with different reporter dyes, different amplicons can be amplified
(multiplexed) in the same reaction.
A RQM-PCR assay was developed to detect exon 14 deletions (see Chapter 3).
Exon 12 of human serum albumin {Alb), another single-copy gene, served as internal control. The APC exon 14 and exon 12 probes were labelled at the 5’-ends with the reporter dyes FAM and VIC, respectively, and the 3’-ends with the quencher dye
TAMRA. For RQM-PCR, DNA aliquots of 15ng/pl were prepared for each sample.
The assay was carried out in 96-well optical reaction plates sealed with optical adhesive covers (PE-Applied Biosystems). Each experiment comprised in triplicate
3 normal controls, 1 deletion control, 27 patient samples and 1 no-template control.
Each 25pi reaction contained Ix TaqMan Universal PCR Master Mix (PE-Applied
Biosystems), 50nM APC exon 14 and Alb exon 12 forward primers, 300nM APC exon 14 and exon 12 reverse primers, 200nM^PC exon 14 probe, 175nM ^4/6 exon 12 probe, and 2pl of DNA. To reduce the risk of cross-contamination, samples and controls were added after aliquoting the master mix. The plates were centrifuged at ISOOrpm for 1 minute, and the reactions performed on an ABI PRISM 7700
Sequence Detection System (PE-Applied Biosystems). The thermal cycling conditions were 50°C for 2min and 95°C for lOmin, followed by 40 cycles of dénaturation at 95°C for 15sec and simultaneous annealing and extension at 60°C for
Imin. The data obtained were analysed using a modified comparative Ct method (see
Chapter 3).
72 2.5.5 Loss of Heterozygosity Analysis
As described in Chapter 1, tumour suppressor genes act negatively on cell growth and division, and one intact allele is generally sufficient to accomplish this task.
Tumour growth occurs, when both alleles are disrupted by mutation. The first inactivating mutation is usually a small-scale DNA change confined to the tumour suppressor gene itself. The second event can either be a small-scale mutation or epigenetic change, or take the form of allelic loss involving part of or the entire chromosome. Mechanisms underlying loss of heterozygosity (LOH) probably include nondisjunction of homologs during mitosis, mitotic recombination, or de novo interstitial deletions.
For LOH analysis, matched DNA samples from peripheral blood and tumour tissue were genotyped at a set of microsatellite markers close to the tumour suppressor gene under investigation. Microsatellite markers are short tandem repeats of mono-, di-, tri- or tetra-nucleotides which are frequently polymorphic in the population. The forward primer of each PCR reaction was labelled with a fluorochrome (FAM, HEX or TET). The PCR products were run on a semi-automated ABI Prism 377XL DNA sequencer (PE-Applied Biosystems) together with GENESCAN-350 TAMRA size marker (PE-Applied Biosystems). The results were analysed for allelic loss using
GENESCAN and GENOTYPER software (PE-Applied Biosystems). For heterozygous samples, the area under each allele peak was determined, and the ratio of the areas of the two alleles was calculated. The LOH value was calculated by dividing the allelic ratio of the normal tissue by the allelic ratio of the matched tumour tissue. LOH values smaller or equal than 0.5 or larger or equal to 2.0 were
73 scored as allelic loss, assuming that the tumour contained more than 50% neoplastic material as confirmed by histopathological examination.
2.6 Genotyping
2.6.1 Amplification Refractory Mutation System PCR
Amplification Refractory Mutation System (ARMS) PCR is an efficient method to genotype single nucleotide polymorphisms (SNFs). The assay makes use of the fact that Taq DNA polymerase only extends from a primer with a correctly matched base pair at its 3’-end. Thus by choosing a PCR primer whose 3’-end overlies a SNP, allele-specific amplification can be achieved. A typical ARMS-PCR assay consists of two reactions, one specific for each allele. Details of ARMS-PCR primers and conditions are given in the relevant chapter (see Chapter 3). Reaction products were run on a 2% agarose gel containing ethidium bromide and genotypes were called from the gel image.
2.6.2 Restriction Fragment Length Polymorphism Analysis
Restriction fragment length polymorphism (RFLP) analysis exploits the ability of type II restriction endonucleases to cleave DNA molecules at specific sites.
Restriction endonucleases are produced by bacteria to monitor the origin of incoming
DNA, degrading it if it is recognised as foreign. The vast majority of type II restriction endonucleases recognise and cleave DNA within palindromic repeats of tetra-, penta-, hexa- or hepta-nucleotides. Some DNA polymorphisms create or 74 delete recognition sites of restriction endonucleases. Successful digestion of a PCR product encompassing one of these polymorphisms thus depends on the allele status, and the genotype can be called following agarose gel electrophoresis and ethidium bromide staining.
Restriction enzyme digestion of PCR products was performed in 20p.l volumes according to manufacturer’s instructions (New England Biolabs). A typical reaction contained lOpd of PCR product, Ix restriction enzyme buffer, and 10 units of restriction enzyme. Bovine serum albumin (NEB) was added to a Ix concentration as appropriate. To ensure complete digestion, the sample was incubated overnight at the recommended temperature. The resulting fragments were resolved in a 2% agarose gel.
2.7 Comparative Genomic Hybridisation
Comparative genomic hybridisation (CGH) allows'lo screen an entire genome for chromosomal gains and losses in a single reaction, and the technique has been widely used to study tumour karyotypes (Kallioniemi et al., 1992). DNA from tumour and normal tissue are labelled with green and red fluorochromes, respectively, and hybridised onto normal metaphase spreads. Images of 5-10 metaphases are captured and green:red fluorescence ratios are quantified for each chromosome using digital image analysis software. Regions of chromosomal loss or gain can then be identified by a comparison of the red:green fluorescence ratios. Loss of chromosomal material in the tumour will result in a bias towards red fluorescence, whereas gain of material will result in a bias towards green fluorescence. The detection sensitivity of CGH is
75 limited, requiring loss of at least 10Mb or five-fold amplification of at least 2Mb in an otherwise diploid genome (Kallioniemi et al., 1996).
A total of 50-100ng of tumour (containing at least 60% neoplastic material) and reference DNA were amplified by degenerate oligonucleotide primed PCR (DOP-
PCR). DOP-PCR permits amplification of genomic DNA by using a degenerate oligonucleotide primer (6MW; 5'-ccgactcgagnnnnnnatgtgg-3'). The reaction has two stages: (i) low stringency cycles during which the specific bases at the 3'-end of the oligonucleotide theoretically prime every 4kb along the template DNA, and (ii) high stringency cycles during which the oligonucleotide-primed DNA from the initial cycles is amplified. Each 25p,l DOP-PCR reaction contained Ix Mg^^-free PCR buffer (Promega), 4mM MgClz (Promega), 0.2mM of each dNTP (Pharmacia),
20p-M 6MW primer (Cancer Research UK), 1 unit Taq DNA polymerase (Cancer
Research UK), and 50-100ng DNA template. DOP-PCR was performed using a
Tetrad PCR Machine and consisted of an initial dénaturation at 94°C for 5min, 8 cycles of 94°C for Imin, 30°C for 1.5min and 72°C for 3min, 25 cycles of 94°C for
Imin, 62°C for Imin and 72°C for 1.5min, and a final extension at 72°C for Smin.
DOP-PCR products were fluorescently labelled by using nick-translation. Nick- translation utilises the simultaneous activity of two enzymes, DNase I and E.coli
DNA polymerase I. In the presence of Mg^^ ions, DNase I creates random nicks in a double-stranded DNA molecule. Starting at these nicks, E. coli DNA polymerase I can then ‘translate’ the single strands in a 5’->3’ direction. Whilst its 5'->3' exonuclease activity removes nucleotides from the 5’-end of a nick, its 5'->3' polymerase activity adds fluorescently labelled nucleotides to the 3'-end. When nicks
76 on opposite strands meet, the DNA molecule breaks. Tumour and reference DNA were labelled with FITC-12-dUTP (Vysis) and Texas Red-5-dUTP (Vysis), respectively. Each nick-translation reaction contained 25p,l DOP-amplified DNA,
Ipl fluorochrome, 5p.l dNTPs, lOpi DNA polymerase I/DNase I mix (GibcoBRL), and 9|li1 dHzO. The reaction was incubated at 15°C for two hours, and placed on ice while 5p-l were run on a 1% agarose gel. If translation products formed a smear ranging in size from 500-2000bp, the reaction was stopped by adding 5p,l of 0.5M
EDTA.
The nick-translation products of tumour and reference DNA were combined, and added to 50pg Cotl DNA (GibcoBRL), 0.1 volumes 3M sodium acetate, and 2 volumes cold 100% ethanol (BDH). Cotl DNA blocks repetitive DNA sequences.
The DNA was precipitated on dry ice for 30 minutes, and the supernatant was removed after centrifugation at 15000 rpm for 20 minutes at 4°C. The DNA pellet was resuspended in lOpI hybridisation buffer, denatured for 5 minutes at 75°C, and left to pre-anneal for 30 minutes. The mixture was then applied to a slide of denatured metaphase spreads of normal male peripheral blood lymphocytes (Vysis).
Denatured metaphase spreads were prepared by incubating the slide in denaturing solution under a 22x50mm coverslip at 73 °C. The coverslip was flicked off, the slide placed in ice cold 70% ethanol for 3 minutes, and dehydrated through an ethanol series (95%, 100%) of 3 minutes each. lOpl of denatured probe were added per half-side, covered with a 22x22 mm coverslip, sealed with rubber cement, and left to hybridise in a moist chamber for 48-72 hours at 37°C. Following hybridisation, the coverslip was removed from the slide. The slide was washed for 5 minutes in formamide wash solution at 42°C and for 5 minutes in 2x SSC at room
77 temperature. The slide was then dehydrated through an ethanol series (70%, 95%,
100%) and left to air dry. After air-drying the slide was counterstained with approximately 20p,l 4,6-diamino-2-phenylindole (DAPI) under a 22x50mm coverslip, and either stored in a cardboard folder at 4°C or captured immediately.
Images were captured with a CCD camera attached to a Zeiss axioskop microscope and analysed using Quips (Vysis) software. Between 5 to 10 metaphases were analysed for each tumour. The metaphases were karyotyped using the digitally inverted DAPI image which gave a G-banded pattern. After karyotyping the relative intensities of the red and green signals were analysed and an average obtained for multiple metaphases. CGH experiments were considered successful if enough fluorochrome had been incorporated to give smooth intense colour that was granular in appearance. Negative control hybridisations were included in each batch of experiments. A chromosomal region was considered to be lost or gained if the mean hybridisation ratio between tumour and normal DNA was <0.85:1 or >1.15:1, respectively (Roylance et al., 1999).
2.8 Flow Cytometry and Ki-67 Immunocytochemistry
Flow cytometry (FCM) can be used to monitor the DNA content of individual cells, and can be applied to any tissue that can be reduced to a single-cell suspension. After disaggregation, the cells are treated with RNase (to degrade RNAs) and stained with a fluorescent dye that binds stoichometrically to DNA. As individual cells are passed though a laser beam, dye-emission can be measured which is proportional to the
DNA content. Typically, 10,000 or more cells can be analysed in minutes, and the
78 results can be expressed as a frequency curve of DNA content. From thcs^ data, estimates of the number of cells with 2C (C=haploid genome), intermediate and 4C
DNA content can be obtained, corresponding to the GO/G], S and G 2 +M phases of the cell cycle. When comparing normal and tumour samples, aneuploidy or polyploidy can be identified as extra peaks, provided that more than 5% of chromosomal material have been gained or lost (approximately 1 large or two medium/small sized chromosomes). The ratio of the medians of GO/Gi peaks of tumour and normal sample is referred to as DNA index (DI). FCM for DNA content can also be combined with stains for cellular proteins such as Ki-67, a cell proliferation marker. Ki-67 is expressed in all phases of the cell-cycle except Go and early Gi, with maximum amounts during G 2 and M phase.
FCM was performed on both fresh-frozen and formalin-fixed, paraffin-embedded tumours as well as respective normal biopsies. All fresh frozen tissue was simultaneously assayed for expression of the Ki-67 antigen using FITC-labelled monoclonal mouse antibody (DAKO). The appropriate FITC-labelled mouse IgGl antibody (DAKO) was used as isotype control.
Fresh-frozen tissue was prepared by disaggregating a small piece (about 4mm^) into a cellular suspension using the DAKO Medimachine System according to manufacturer’s instructions. The cells were harvested by centrifugation at ISOOrpm for 5 minutes at room temperature, and fixed by drop-wise addition of 1ml of ice- cold 70% ethanol whilst vortexing. The sample was incubated for 40 minutes at
4°C, washed twice in 2ml PBS/0.5% Tween 20 and re-suspended in 80p,l of
PBS/0.5% Tween/0.5% BSA. Incubation with 20 |li1 of antibody was performed for
79 30 minutes at 4°C. Cells were washed twice in 2ml PBS/0.5% Tween 20/0.5%
BSA, resuspended in 50p-l, and treated with 50pil lOO^ig/ml RNaseA and 200p,l
50ng/ral propidium iodide (PI). After a 15 minute incubation, the suspension was filtered through a 70pim nylon filter and immediately analysed on a FACSCalibur
(Becton Dickinson). Cells were excited by the argon laser emitting at 488nm.
Fluorescence from FITC-labelled antibodies was detected using a 530/30nm band pass filter and PI fluorescence was detected using a 670nm long pass filter. Forward and right angle light scatter were used to set a gate including all cells, but excluding debris. A second gate set on area and width of PI fluorescence was used to further define the single cell population. Both FITC and PI fluorescence were collected in linear mode, and acquisition was stopped after 8000 gated events had been acquired.
Data were analysed for proliferative activity (% of Ki-67 positive cells normalised against the isotype control) and aneuploidy (DNA index or DI) using dedicated
Modfit software (Verity Software House). The DNA diploid peak was set using the normal samples.
Formalin-fixed, paraffin-embedded tissue was prepared by cutting a 50pm section from each block, placing it into a histopathology cassette between two sheets of
3MM filter paper, and dewaxing it in 100% xylene (BDH) overnight. The section was rehydrated through an ethanol series (100%, 95%, 90%, 70%, 50%)and rinsed twice in water. The tissue was removed from the slide with a scalpel and digested with pepsin solution for 30 minutes at 37°C. After centrifugation at 1500 rpm for 5 minutes at room temperature, cells were washed twice in PBS before FCM analysis.
80 2.9 Cloning
Cloning permits amplification of a single DNA fi-agment such as a PCR product to a very high amount. The technique involves ligation of the DNA fragment into an appropriate vector which, after transformation, can be propagated in a suitable bacterial host. Recombinant clones are selected based on vector-encoded genes, grown up to a high density, and recombinant vector is purified.
PCR products were cloned using the pGEM-T Easy Vector System (Promega). This system utilises a high copy number vector which carries a gene conferring resistance to the antibiotic ampicillin. The vector is provided ready for ligation, having been prepared by cutting with the restriction endonuclease EcoRW and by adding a single
3’-thymidine to both of the generated blunt ends. The EcoRV restriction site is located within an open reading frame encoding the a-peptide of (3-galactosidase
(lacZ'). When intact, this enzyme can convert the colourless substrate X-gal into a blue product upon induction with IPTG.
Ligation into the pGEM-T Easy Vector requires the insert to have 3’-overhangs of adenines. PCR product was tailed with 3’-adenines using DNA polymerase and an excess of dATP. 25pi of PCR product were denatured at 95°C for 10 minutes and immediately placed on ice. 2.5pl of 2mM dATP (Amersham) and 2.5 units of PIC
Taq DNA polymerase (Cancer Research UK) were added, and the mixture was incubated at 70°C for 15 minutes. Ligation into pGEM-T Easy Vector was performed in a lOpl volume by overnight incubation at 4°C. Each 20pl reaction
81 contained Ix rapid ligation buffer, 10 units of T4 DNA ligase, SOOp-g pGEM-T Easy
Vector, and 3p,l tailed PCR product.
Following ligation, recombinant vector was transformed into Epicurian Coli Sure
Competent Cells (Stratagene) which lack endogenous p-galactosidase. On ice, 50p.l of Epicurian Coli Sure Competent Cells were added to each ligation mixture in a
Falcon 2059 polypropylene tube (Becton Dickinson). The mixture was incubated for
30 minutes. To trigger DNA uptake, the bacteria were heat shocked for 45 seconds at
42°C and immediately placed back on ice for 2 minutes. To aid recovery from the heat shock and permit expression of vector-encoded genes, the cells were then incubated shaking at 37°C in 1ml of non-selective LB medium.
Recombinant clones were selected by growth on plates containing ampicillin, the inducer IPTG, and the colourless substrate X-gal. 400ml LB agar were dissolved by boiling and cooled to 55°C. Ampicillin, IPTG and X-gal (Promega) were added to a final concentration of lOOpg/ml, 0.5mM, and 16pg/ml, respectively. After mixing,
15-20ml agar were poured into sterile Petri Dishes (Becton Dickinson) in a
BioMAT-2 Safety Cabinet. The plates were left to set and covered with their lids.
The bacteria were harvested by centrifugation at 3000rpm for 5 minutes at room temperature (CR412 Jouan centrifuge), and 850p,l of supernatant were discarded.
The bacteria were resuspended in the remaining medium, and lOOpl and lOp.1 aliquots were distributed onto the AMP/IPTG/X-Gal LB plates using autoclaved glass balls. The glass balls were discarded, and the plates were incubated upside- down for overnight at 37°C. Bacteria which had taken up recombinant vector carried a lacZ' gene disrupted by the insert, could not convert X-gal, and produced white
82 colonies. Bacteria which had taken up religated vector carried a functional lacZ' gene, could convert X-gal, and produced blue colonies.
White clones were picked using a plastic inoculating loop, and grown up in 5ml liquid LB medium containing lOOp-g/ml ampicillin by shaking incubation for overnight at 37°C. The bacteria were harvested by centrifugation at 3000rpm for 5 minutes at room temperature, and the supernatant was discarded. Recombinant vector was extracted using the R.E.A.L Prep 96 BioRobot Kit and a BioRobot 9600
(Qiagen) according to manufacturer’s instructions. Briefly, bacteria were lysed using a modified alkaline lysis procedure. Denatured and precipitated genomic DNA, proteins, and carbohydrates were removed by filtration. Plasmid DNA in the filtrate was concentrated by isopropanol precipitation. The sequence of the cloned insert was verified by fluorescence cycle-sequencing
2.10 Cell Culture
Epstein-Barr virus (EBV) transformed lymphoblastoid cell lines were grown in 50ml
RPMI/10% FCS medium by incubation at 37°C in 10% CO 2 . The acronym RPMI stands for Roswell Park Memorial Institute, the site at which this medium was first developed. The lymphoblastoid cells were regularly examined under a light microscope to determine cell viability and density. When necessary, about half of the cell suspension was replaced with fresh medium in a sterile BioMAT-2 Safety
Cabinet.
83 Colorectal cancer cell lines were grown in 50ml E4/10% FCS medium by incubation at 37°C in 10% CO 2 . CRC cells grow as a monolayer and were split once confluency reached about 90%. 10ml old medium was transferred to a 50ml polypropylene tube (Coming) in a sterile BioMAT-2 Safety Cabinet. The rest of the medium was discarded. To detach cells from the tissue culture flask, 10ml trypsin solution (CRUK) was added for 2 minutes. The solution was pipetted against the sides of the tissue culture flask to remove any remaining cells, and transferred to the polypropylene tube containing the old medium to stop the trypsin digestion. The sample was centrifuged at 1500 rpm for 5 minutes, and the supernatant was discarded. The cell pellet was redissolved in 5ml E4/10% FCS medium, and 1ml of the solution was transferred to a new tissue culture flask containing 50ml of fresh
E4/10% FCS medium.
To ensure that cell lines remained a permanent resource, cyrostocks were prepared from all cultures. Pellets of 50ml cultures were prepared as appropriate for each cell type. Each cell pellet was resuspended in 2ml freeze mix and transferred into a sterile cyrotube (Coming) labelled with the cell line name, date, volume processed and initials. The tubes were placed overnight at -80°C and subsequently stored in liquid nitrogen.
84 2.11 Solutions and Media
Ammonium acetate, saturated 148g NH 4AC (BDH), 50ml dHgO Ammonium persulphate, 20% 2g ammonium persulphate (Pharmacia), 10ml dHzO
Ampicillin, 100mg/ml 1g ampicillin (Sigma), 10ml dH 2 0 , stored at -20°C Denaturing soln. (2x SSC, 70% lOOpI 20x SSC, 700pil formamide (BDH), 200pl formamide) dHzO
DTT, 1M 1.5g dithothreitol (Sigma), 10ml dH 2 0 , stored at -20°C E4/10% FCS medium 180ml E4 medium (CRUK), 20ml foetal calf serum (GibcoBRL), stored at 4°C EDTA, 0.5M, pH8.0 93g ethylenediamine tetraacetate (BDH), 400ml
dH2 0 , adjusted to pH8.0 with NaOH pellets
(BDH), made up to 500ml with dH 2 0 , autoclaved
Ethanol, 50% 50ml 100% ethanol (BDH), 50ml dH 2 0
Ethanol, 70% 70ml 100% ethanol (BDH), 30ml dH 2 0
Ethanol, 90% 90ml 100% ethanol (BDH), 10ml dH 2 0
Ethanol, 95% 95ml 100% ethanol (BDH), 5ml dH 2 0
Ethidium bromide, 10mg/ml 0.1 g ethidium bromide (Pierce), 10ml dH 2 0 , stored in dark Formamide wash soln. (2x 50ml 20x SSC, 250 ml formamide (BDH), made
SSC, 50% formamide) up to 500ml with dH 2 0 Freeze mix (90% FCS, 10% 9ml foetal calf serum (GibcoBRL), 1ml dimethyl DMSG) sulphoxide (Sigma) Hybridisation buffer (2x SSC, 1ml 20x SSC, 5ml formamide (BDH), 1g dextran
50% formamide, 10% dextran sulphate (BDH), made up to 10ml with dH 2 0 , sulphate) stored at -20°C
IPTG (0.1 M) 1.2g IPTG (Promega), 50ml dH 2 0 , filter-sterilised and stored at -20°C
NP-40, 0.1% 200pl nonidet P40 (BDH), 200ml dH 2 0 Nuclei lysis buffer (lOmM Tris 500pl 1M Tris pH8.0, 4ml 5M NaCI, 200pl 0.5M pH8.0, 400mM NaCI, 2mM EDTA pH8.0, 45.3ml dH 2 0 EDTA pH8.0)
85 PBS (phosphate buffered 8g NaCI (BDH), 0.2g KCI (BDH), 1.44g Na 2 HP0 4 saline) (BDH), 0.24g KH 2 PO4 (BDH), 800ml dH2 0 , adjusted to pH7.2 with HCI (BDH), made up to 1L
with dH2 0 PBS/5% Tween 20 9.5ml PBS, 500pil Tween 20 (Sigma) PBS/0.5% Tween/0.5% BSA 9ml PBS, 500|uil Tween 20 (Sigma), 500pil BSA (NEB) Pepsin soln. (0.5% pepsin, 0.25mg pepsin (Sigma), 50ml 0.9% NaCI pH1.5, 0.9% NaCI, pH1.5) stored at -20°C
Propidium iodide, 50|ig/ml Img propidium iodide (Sigma), 20ml dH 2 0 , stored at 4°C Proteinase K buffer (2mM 200^1 0.5M EDTA pH8.0, 5ml 10% SDS, 44.8ml
EDTApHS.O, 1%SDS) dH2 0 Proteinase K solution 2mg proteinase K (BDH), 1ml proteinase K buffer, stored at -20°C PTT fixing solution (10% 100ml 100% methanol (BDH), 100ml acetic acid methanol, 1 0 % acetic acid) (BDH), 800ml dH2 0 PTT loading buffer 900pil PTT sample buffer, 100|il 1M DTT, stored at -20°C PTT running buffer (0.025M 3g Tris base (Sigma), 14.4g glycine (BDH), 10ml
Tris, 0.192M glycine, 0.1% 10% SDS, made up to 1L with dH 2 0 SDS)
PTT sample buffer (50mM Tris- 10ml 0.5M Tris-HCI pH 6 .8 , 4g SDS (BDH), 20g
HCI pH 6 .8 , 4% SDS, 20% sucrose (BDH), 2g bromophenol blue (Sigma), sucrose, 2 % bromophenol blue) made up to 1 0 0 ml with dH2 0 RNaseA, lOmg/ml lOOmg RNaseA (Advanced Biotechnologies),
10ml dH2 0 , stored at -20°C
RNaseA, lOOpig/ml lOOpil 10 mg/ml RNaseA, 9.9ml dH 2 0 , stored at - 20°C RPMI/10% FCS medium 180ml 3.7% RPMI (CRUK), 20ml foetal calf serum (GibcoBRL), stored at 4°C SDS, 10% lOOg sodium dodecyl sulphate (BDH), 900ml
dH2 0 , adjusted to pH7.2 with HCI, made up to 1L
with dH2 0
86 SE buffer (75mM NaCI, 25mM 7.5ml 5M NaCI, 2.5ml 0.5M EDTA, 50ml 10% EDTA, 1% SDS) SDS, made up to 500ml with dH20, sterilised through 0.22pim pore filter (Millipore) Sodium acetate, 3M 61.52g sodium acetate (BDH), 200ml dh20, pH adjusted to 6.0, made up to 250ml with dH20, autoclaved Sodium chloride, 5M 73.1g NaCI (BDH), made up to 250ml with dH20, autoclaved Sodium chloride, 0.9%, pH1.5 4.5g NaCI, 400ml dH20, adjusted to pH1.5 with HCI (BDH), made up to 500ml with dH20 Sodium hydroxide, 0.3M 1.2g NaOH pellets (BDH), 100ml dH20 SSC, 2x 50ml 20x SSC, 450ml dH20 SSC, 20x (0.3M Trisodium 88.23g Trisodium citrate (BDH), 175.32g NaCI citrate, 3M NaCI) (Sigma), made up to 1L with dH20 (final pH7-8) TBE, 5x (0.45M Tris, 0.45M 54g Tris base (Sigma), 27.5g boric acid powder Borate, 0.01 M EDTA) (BDH), 20ml 0.5M EDTA pH8.0, made up to 1L
with dH2 0 , autoclaved Tris, 1M, pHS.O 60.55g Tris base (Sigma), 400ml dH20, adjusted to pH8.0 with NaOH pellets (BDH), made up to 500ml with dH20, autoclaved. Tris-HCI, 0.5M, pH6.8 78.8g TRIZMA hydrochloride (Sigma), 400ml dH20, adjusted to pH8.0 with HCI (BDH), made up to 500 ml with dH20
87 Chapter 3 The Contribution of Whole-gene APC Deietions to
AAPC/Muitiple Coiorectal Adenomas and Ciassicai FAR
3.1 Introduction
Routine mutation detection techniques (for example, SSCP, PTT, denaturing gradient gel electrophoresis (DGGE), and DNA sequencing) identify germline APC mutations in approximately 10% of individuals with AAPC/multiple adenomas and 70% of individuals with classical FAP (Cottrell et al., 1992; Lamlum et al., 2000a; Miyoshi et al., 1992a). The underlying molecular genetic basis therefore remains to be determined in a considerable proportion of apparently APC mutation-negative polyposis patients. The failure to detect germline APC mutations might also, however, be due to methodological difficulties, as has recently been shown for another colorectal cancer predisposition, HNPCC, where over 5% of cases in some populations result from exon-spanning genomic deletions in MSH2 (Wijnen et al.,
1998). By analogy, such submicroscopic deletions may be missed in a substantial fraction o f A P C mutation-negative patients with AAPC/multiple adenomas or classical polyposis.
In support of this hypothesis, interstitial deletions of chromosome 5q which encompass all or part of the APC locus have been identified in a limited number of individuals diagnosed with attenuated or classical polyposis (Table 3.1). However, no comprehensive studies have been performed to establish the frequency of germline APC deletions in these patients. Notably, some - but not all - individuals with large-scale APC deletions have been reported to develop extracolonic features and mental retardation.
Table 3.1: Clinical diagnosis and extent of germline APC deletion in polyposis patients reported in the literature.
DPI, deleted in polyposis \;MCC, mutated in colorectal cancer; N.a., not available;
\Cross et al., 1992), ^(Hockey et al., 1989), ^(Hodgson et al., 1993) "^(Barber et al.,
1994), ^(Pilarski et al., 1999) ^(Herrera et al., 1986) ^(De Rosa et al., 1999) ^(Joslyn et al., 1991) ^Mandl et al., 1996) et al., 2002) ’\S u et al, 2000b)
Clinical Extent of Germline Extracolonic Features Other Ref. Diagnosis APC Deletion FAP, familial 46XY del(5)(q15-q23.2) Epidermoid cysts, Low intelligence/ 1 duodenal polyps, mentai retardation CHRPE FAP, familial 46XXdel(5)(q15-q22) Epidermoid cysts Low intelligence/ 2 mental retardation FAP, sporadic 46XY del(5)(q15-q23.1) CHRPE Mental retardation, 3 Caroli's syndrome FAP, sporadic 46XXdel(5)(q15-q22.3) ‘Minor dysmorphic Mental retardation 4 features’ FAP, sporadic 46XY del(5)(q21.3-q23.1) Epidermoid cysts, Low intelligence 3 desmoids, duodenal & gastric polyps AAPC, sporadic 46XY del(5)(q22-q23.3) CHRPE Mental retardation, 5 Hyperlipidaemia, Hypothyrodism G ard ner’s, 46XY del(5)(q21) Epidermoid cysts, Mental retardation 6 sporadic osteom as FAP, familial del {APC, DP1 and MCC) CHRPE Normal intelligence 7
FAP, familial del {APC and DP1) CHRPE Normal intelligence 7
FAP, familial del {APC and DP1) CHRPE Normal intelligence 7
FAP, sporadic del {APC and DP1) N .a. N.a. 8
FAP, sporadic del {APC and DP1) N.a. N.a. 8 AAPC, familial del (APC X11-3') CHRPE Normal intelligence 9
FAP, familial del (APCx15) Epidermoid cysts, N.a. 10 duodenal & gastric polyps AAPC, familial del (APCx15) None N.a. 10 AAPC or FAP, del (APC x11-12) N.a. N.a. 11 familial AAPC or FAP, del (APCx14) N.a. N.a. 11 familial AAPC or FAP, del (APC x14) N.a. N.a. 11 familial
89 This survey aimed (i) to develop a real-time quantitative multiplex (RQM) PCR assay to detect germline APC deletions, (ii) to determine their frequency in apparently APC mutation-negative patients with AAPC/multiple adenomas as well as classical FAP, and (iii) to fine-map all detected APC deletions using a set of polymorphic markers spanning the entire APC gene. The RQM-PCR assay was developed in collaboration with Dr. Karl Heinimann.
3.2 Study Population
This study examined 203 unrelated individuals clinically diagnosed with colorectal polyposis and in whom no germline APC mutation had been identified using one or more standard mutation detection techniques (SSCP, PTT, DGGE, DNA sequencing). The patients came from diagnostic laboratories in the United Kingdom
(Edinburgh, London, Oxford; n=167), Denmark (Copenhagen; n=ll), Switzerland
(Basel; n=ll), Portugal (Lisbon; n=9) and Italy (Naples; n=4). Overall, 143 (70%) patients displayed between 3 and 100 colorectal adenomas (multiple adenoma/AAPC phenotype), and 60 (30%) patients more than 100 colorectal adenomas (classical polyposis). Additional phenotypic details (gender, age at diagnosis, family history, extracolonic disease, occurrence of colorectal cancer) were available for 165 (81%) individuals.
90 3.3 Development of a RQM-PCR Assay to DetectAPC Exon 14
Deletions
A real-time quantitative multiplex (RQM-) PCR assay was developed to detect APC exon 14 deletions. The principle of this assay is described in detail in Chapter 2.5.4.
Exon 14 of APC was chosen as target for the assay, as it is encompassed by the majority of submicroscopic germline APC deletions reported to date (Table 3.1).
Exon 12 of human serum albumin (Alb), another single-copy gene, was chosen as internal control. Alb is located on chromosome 4qll-ql3, a region which is not expected to be deleted or amplified in the germline of colorectal adenoma patients.
Primers and probes were designed using Primer Express software (PE-Applied
Biosystems), and checked for specificity using the NCBI BLAST programme
(www.ncbi.nlm.nih.gov/BLAST). To permit amplification and detection in the same reaction, APC exon 14 and Alb exon 12 probes were labelled with different reporter dyes, F AM and VIC, respectively (Table 3.2). All reactions were performed on an
ABI PRISM 7700 Sequence Detection System (PE-Applied Biosystems).
Table 3.2: Primers and probes for the RQM-PCR assay.
Amplicon Primer/probe name Sequence (S' to 3') size, bp APC exon 14 forward GCCAGACAAACACTTTAGCCATTA 91 APC exon 14 reverse TACCTGTGGTCCTCATTTGTAGCTAT APC exon 14 probe CTGGACACATTCCGTAATATCCCACCTCC (5'-FAM, 3'-TAMRA) Alb exon 12 forward TTGCATGAGAAAACGCCAGTA 70 Alb exon 12 reverse GTCGCCTGTTCACCAAGGA Alb exon 12 probe TCTGTGCAGCATTTGGTGACTCTGTCAC (5’-VIC, 3'-TAMRA)
91 The RQM-PCR assay was optimised following the instructions in User Bulletin #5
(PE-Applied Biosystems). Using 20ng of DNA in a 25pl reaction, the optimal concentrations for APC exon 14 and Alb exon 12 were 50nM for both forward and
300nM for both reverse primers. The optimal probe concentrations proved to be
200nM for^PC exon 14 and 175nM fox Alb exon 12. DNA concentrations between lOng and 50ng were within the linear dynamic range of the system (Figure 3.1).
Figure 3.1: Relative standard curves for APC and Albumin amplification by
RQM-PCR.
DNA concentrations between lOng and 50ng are within the linear dynamic range of the system. Reactions were performed in triplicate for each DNA concentration.
y - -3.8492X + 34.494 - 0.9885
u
y = -3.5546X + 30.414 - 0.9865
1.5 1 -0.5 0 0.5 1 1.5 2
log [DNA conc. (ng)]
Ct (APC) ■ Ct (Albumin)
92 Each RQM-PCR experiment comprised in triplicate 3 normal controls, 1 deletion control, 27 patient samples and 1 no-template control. The data obtained were analysed using the comparative Ct method (as described in User Bulletin #2 (PE-
Applied Biosystems) and REF (Aarskog and Vedeler, 2000)). After normalisation against the internal Alb control, this method allowed determination of APC copy number of a given patient by comparison to a normal calibrator. While the original
Ct method relied only on 1 normal sample as calibrator, I used the mean of 3 normal samples to obtain a more consistent calibrator value between experiments. The calibrator value (CV) was calculated as follows, with CS denoting the normal calibrator sample: CV = {[ACt^/6 (CS 1) - ACt.4PC (CS 1)] + [ACt Alb (CS 2) -
ACt APC (CS 2)] + [ACt Alb (CS 3) - ACt APC (CS 3)]} / 3. APC gene copy number was expressed as a value, where AACt = CV - [d^CjAlb (patient sample) - ACt APC (patient sample)]. ACt represented the mean Ct value of each sample-triplicate, where Ct was defined as the cycle number during the exponential phase at which the amplification plot passed a fixed normalised emission intensity threshold between 0.03 and 0.13 (mean 0.08). Samples without deletions were expected to give values close to 1 and samples with APC deletions 2'^^^^^ values close to 0.5. The reliability of the assay was confirmed using DNA samples of
5 unrelated, healthy controls and 5 unrelated FAP patients with previously characterised yfPC deletions encompassing exon 14 (Mandl et al., 1996). The healthy controls displayed 2'^^^^^^ values ranging from 0.83 to 0.92, while APC deletion controls showed 2'^^^^^^ values ranging from 0.54 to 0.61 (Figure 3.2).
93 3.4 Germline APC Deletions in Patients with Multiple Colorectal
Adenomas and Classical Polyposis
Analysis of 143 AAPC/multiple adenoma and 60 classical polyposis patients with no apparent germline APC mutation using the RQM-PCR assay identified 7 (3.4%) individuals harbouring germline deletions encompassing exon 14, with values ranging from 0.43 to 0.62 (Figure 3.2). Notably, all APC deletion patients came from the classical polyposis group, accounting for 11.7% (7/60) of these cases, compared to none (0/143) in the AAPC/multiple adenoma group (p<0.001, Fisher’s exact test). The phenotypic features of the APC deletion patients compared to the remaining APC mutation-negative classical polyposis patients were very similar, with comparable median age at diagnosis (26.5 V5 30 years) and similar proportions of positive family history (80.0% wild-type. 85.1%) and extracolonic disease (28.6% wild-type. 22.6%).
94 Figure 3.2: RQM-PCR results for 5 healthy controls, 5 deletion controls, 143
AAPC/multiple adenoma patients, and 60 classical polyposis patients.
Patients with 2 copies of AFC exon 14 display values between 0.78 and 1.00, whereas patients with 1 copy only display values between 0.43 and 0.62 (overall, 2'
(a a c t ) consistent with a normal distribution and all AFC deletion patients displayed values >2.9 standard deviations from the mean).
1.10
1.00
0.90 u na 0.80 i 0.70 fN
0.60
0.50
0.40 Patient ID
• Healthy controls ♦ Deletion controls A Multiple adenoma patients ■ Classical polyposis patients
95 3.5 The Extent of the Detected GermlineAPC Deletions
In order to determine the extent of the germline APC deletions identified, 6 intragenic and 5 polymorphic markers flanking the APC gene were analysed (Table
3.3). The following markers were intragenic to APC: (i) an A/G polymorphism
(NCBI SNP cluster ID: rs2019720) located within the promoter region, (ii) an A/T polymorphism (rsl914) located within intron 7, (iii) a T/C polymorphism located within exon 11 (Almeida et al., 1996; Groden et al., 1991), (iv) an A/G polymorphism located within exon 151 (Davies and Snover, 1994), (v) an A/G polymorphism located within exon 15J (Almeida et al., 1996; Kraus and Ballhausen,
1992), and (vi) a T/C polymorphism located within the 3' untranslated region
(Almeida et al., 1996; Heighway et al., 1991). The markers flanking the C gene were an A/G (rs748628) and a C/T polymorphism (rs 1922665) located about 110 kb and 37 kb 5' of APC, as well as 3 microsatellite markers, D5S346, D5S656 and
D5S421 located about 32 kb, 396 kb and 628 kb 3' of APC, respectively (UCSC genome browser April 1, 2001 freeze at http://genome.ucsc.edu). Allele frequencies for the rs2019720, rsl914, rs748628, and polymorphisms rs 1922665 were estimated using DNA from 28 unrelated, healthy European controls.
96 Table 3.3: Primer sequences, annealing temperatures, detection methods
applied and allele frequencies for iXitAPC polymorphisms.
Ta = annealing temperature, ^Forward primers for A and G allele, ^^untranslated
region, ^determined in this study, ^(Almeida et al., 1996; Groden et al., 1991),
^(Davies and Snover, 1994), "^(Almeida et al., 1996; Kraus and Ballhausen, 1992),
^(Almeida et al., 1996; Heighway et al., 1991)
APC region Primer sequence(S’ to 3’) Ta (°C) Detection Allele 1 Allele 2 (freq.) (freq.) rs748628 CTTTTCTTTTTCTTTTTCt* 53 ARMS-PCR A (0.45) G (0.55)^
CTTTTCTTTTTCTTTTTCc*
CTTACTACATTCAAGGGGAT
rs19 2 2 6 6 5 CTTCCCTGTTCTGCCAATCT 53 Sequencing C (0.46) T (0.54)^
TCTGTTGGTGGTCTCC
Prom oter TGGGGATGAGAGAAAGAGGAGGA 60 Rsal digest A (0.45) G (0 .5 5 )’
CGCAAAAAGCCACTACCACTG
Intron 7 CAGGTTTGAGCCATCATGC 60 Sequencing A (0.46) T (0.54)’
ATCCAATCCCTAAGCTTGACTG
Exon 11 GATGATTGTCTTTTTCCTCTTGC 55 Rsai digest I (0.48) C (0.52)^
CTGAGCTATCTTAAGAAATACATG
Exon 151 AGTAAATGCTGCAGTTCAGAGG 56 BsaJ1 digest A (0.62) G (0.38)^
CCGTGGCATATCATCCCCC
Exon 15J CCCAGACTGCTTCAAAATTACC 55 Sequencing A (0.57) G (0.43)'’
GAGCCTCATCTGTACTTCTGC
3' U T R ** GCATTAAGAGTAAAATTCCTCTTAC 58 Sspi digest T (0.54) C (0.48)®
ATGACCACCAGGTAGGTGTATT
Fluorescent D 5 S 3 4 6 ACTCACTCTAGTGATAAATCGGG 55 N.a. N.a. genotyping AGCAGATAAGACAGTATTACTAGTT
Fluorescent D 5 S 6 5 6 GCTAAGAAAATACGACAACTAAATG 55 N.a. N.a. genotyping CATAATAAACTGATGTTGACACAC
Fluorescent D 5S421 TGGAAATAGAATCCAGGCTT 55 N.a. N.a. genotyping TCTATCGTTAACTTTATTGATTCAG
97 Six (85.7%) out of 7 deletion patients were apparently homozygous (suggesting hemizygosity) at markers spanning the coding region of the APC gene, consistent with whole-gene deletions. In 3 of these cases (602, 236.vi.10 and BAN1773), homozygosity at D5S656 suggested that the deletion might also encompass the MCC
(mutated in colorectal cancers) locus. In at least one case (1749-1), the germline deletion did not involve the whole gene (Table 3.4). Assessment of the polymorphic markers 5' and 3' of APC allowed me to determine the maximum extent in 6 (85.7%) of the 7 deletion cases and suggested that at least 4 (57.1%) were restricted to the
APC gene.
Table 3.4: Genotype results for seven classical polyposis patients harbouring an
APC exon 14 deletion as determined by RQM-PCR.
Bold type indicates areas excluded from the respective APC deletions, het, heterozygosity; hom, homozygosity (suggesting hemizygosity); del, deletion; MCC, mutated in colorectal cancers.
ID 367.VI.4 1749-1 236.vi.11 2350 BAN1773 602 JS rs748628 het hom het het het hom het rs1922665 hom hom het hom hom het het
APC promoter het het hom hom hom hom hom
APC intron 7 hom het hom hom hom hom hom
APC exon 11 hom het hom hom hom hom hom
APC exon 14 del del del del del del del
APC exon 151 hom hom hom hom hom hom hom
APC exon 15J hom hom hom hom hom hom hom
APC 3' UTR hom hom hom hom hom hom hom
D5S346 hom het het hom hom hom hom
MCC locus del? not del not del del? del? del? del? D5S656 het het hom hom hom hom het D5S421 het het het hom het het hom
98 It was then assessed whether or not this analysis was likely to have missed any large intragenic deletions 5’ or 3’ of exon 14 in 50 of the patients with classical FAP.
Thirteen (26.0%) of these patients were homozgous for all intragenic markers. Two
(4.0%) additional patients were homozygous for the 3 intragenic markers 5’ of exon
14. There was no evidence for an over-representation of homozygotes at any of the polymorphisms studied (data not shown). Whilst the possibility cannot be excluded that this analysis failed to identify a small number of patients with APC deletions, the data suggest that it did not miss large numbers of deletions and, therefore, such changes cannot account for all the APC mutation-negative patients with classical
FAP.
3.6 Discussion
Applying the RQM-PCR assay for the detection of APC exon 14 germline deletions,
7 deletion patients were identified in a set of 143 AAPC/multiple adenoma patients and 60 classical polyposis with no apparent germline APC mutation. Although previous studies of germline APC deletions have not consistently reported classical disease in these patients, all of my deletion patients displayed classical polyposis, exhibiting more than 100 adenomas. In contrast to the multiple adenoma subgroup (3 to 100 polyps), where no deletion could be detected, the frequency of APC deletions in patients with classical polyposis amounted to 11.7% of mutation-negative cases.
Fine-mapping of the detected APC deletions (including available affected and unaffected relatives) using 6 intragenic, polymorphic markers suggested a whole- gene deletion in 6 (85.7%) of 7 patients (Table 3.4).
99 These results make an interesting comparison with previous data from AAPC patients. It has been suggested that in some cases AAPC can result from instability of the mutant APC protein (van der Luijt et al., 1996). Van der Luijt et al., for example, reported that truncating germline APC mutations in the 3 ’ half of exon 15 (codons
1862 and 1987) resulted in unstable protein, as determined by western blotting, and an AAPC phenotype (van der Luijt et al., 1996). Since an unstable protein and a whole-gene deletion might be expected to be functionally equivalent, this difference in disease severity suggests the presence of residual mutant APC protein not detectable by western blotting in these AAPC patients.
Of the seven ^PC-deletion patients, three had had formal polyp counts at colectomy
(Crabtree et al., 2001). These counts (800, 1425 and 1899) were extremely similar to those typically reported for TAP patients with truncating APC mutations between codons 168 and 1250 or codons 1400 and 1580 approximately (Crabtree et al, unpublished). Perhaps germline mutations between codons 168 and 1250 or codons 1400 and 1580 are functionally equivalent to null changes. Germline mutations between codons 1250 and 1400, the so-called mutation cluster region
(MCR), are associated with more severe colonic polyposis ((Nugent et al., 1994), also Crabtree et al, unpublished), possibly indicating that the mutant protein is more effective at causing tumorigenesis. There are alternative explanations for the phenotypic similarity between patients with APC deletions and those with truncating mutations between codons 168 and 1250 or codons 1400 and 1580. For example, the
‘second hit’ at APC may be rate-limiting for tumorigenesis and hence the main determinant of disease severity, given the evidence to show that TAP polyps with germline mutations between codons 168 and 1250 (before the first 20aa (3-catenin
100 binding and degradation repeat) tend to acquire truncating mutations after codon
1400 (the second 20aa repeat) and vice versa (Albuquerque et al., 2002a; Lamlum et al., 1999). As germline deletion patients also do not express any 20aa repeats from the inherited mutant allele, they may acquire similar somatic changes than patients with germline mutations between codons 168 and 1250, and thus would be expected to have similar disease severity.
The RQM-PCR assay which has been developed for this study proved to be a fast and reproducible method to detect gene dosage at APC exon 14. Given its potential for high-throughput analysis, its reliability as well as the small amounts of DNA needed (30ng per reaction), this assay may be of potential use in a routine diagnostic setting. In view of the appreciable frequency (11.7%) of APC deletions in APC mutation-negative classical polyposis patients, this technique may be particularly valuable for this group of patients. Disadvantages of the technique, however, may include the high costs for the detection device and the consumables as well as the laborious assay setup. Although my assay will detect whole-gene APC deletions it cannot exclude partial deletions 5’ and 3’ of exon 14.
In conclusion, a RQM-PCR assay to detect APC exon 14 deletions was developed.
Applying this technique to a set of apparently APC mutation-negative polyposis patients, germline deletions were exclusively found in individuals with classical polyposis. Fine-mapping of the region suggested that the majority of these deletions encompass the entire APC locus. Given the frequency of 11.7%, screening of APC mutation-negative individuals with classical polyposis for germline deletions seems
101 warranted. In contrast, whole-gene APC deletions appear to be a rare cause of
AAPC/multiple colorectal adenomas.
102 Chapter 4 The Relationship between Germiine and Somatic
Mutations at the APC Locus in Patients with Ciassicai FAR
4.1 Introduction
It has previously been established that the association between germline and somatic
mutations at the APC locus is non-random (Lamlum et ah, 1999). In patients with
germline mutations close to codon 1300, LOH is the usual ‘second hit’, although the
boundaries of the LOH-associated region have not been precisely defined. In other
FAP patients, however, the ‘second hit’ tends to be a truncating mutation in the
MCR. These findings have implications for the mechanism of tumorigenesis in FAP,
and indeed for colorectal cancer in general, because a similar ‘first hit-second hit’
association at APC is also seen in sporadic tumours (Rowan et al., 2000). The data
suggest thatAPC mutations are selected for their ability to produce an optimal level
of Wnt signalling in the tumour cell, as they evidently target the functional domains
required for p-catenin binding and degradation (see Chapter 1.4.1.1.6). This can be
achieved by LOH in FAP patients with germline mutations near codon 1300, but
other patients must acquire somatic truncating mutations to achieve the same effect.
Similar ‘first hit-second hit’ associations have been described for desmoids and
upper-gastrointestinal tumours, although a different region of APC (distal to codon
1400) was found to be involved (Groves et al., 2002).
Importantly, these observations appear to explain some genotype-phenotype
associations in FAP. Patients with germline mutations near codon 1300 tend to
103 develop particularly severe colorectal polyposis, whilst patients with germline mutations after codon 1400 tend to develop particularly severe duodenal polyposis
(Groves et ah, 2002; Lamlum et ah, 1999). A more detailed characterisation of the
‘first hit-second hit’ associations in FAP patients may thus be a pre-requisite for understanding the molecular genetic mechanism(s) causing attenuated polyposis
(AAPC).
In a recent study of FAP tumours, Albuquerque et al. have refined the spectrum of
‘second hits’ associated with germline APC mutations away from codon 1300
(Albuquerque et ah, 2002a). They reported that patients with germline mutations earlier in the APC gene tended to acquire later truncating somatic mutations, and vice versa. On the basis of these data, Albuquerque et ah proposed a ‘just-right’ signalling model for APC (Table 4.1). In this model, the total number of intact 20aa P-catenin binding and degradation repeats remaining in the proteins encoded by each mutant allele was the factor which determined the APC genotype, because the correct level of P-catenin activity ensued. Most FAP tumours tended to have a pair of mutant
alleles which between them encoded proteins with a total of just a single 20aa repeat.
104 Table 4.1: ‘Just-right’ signalling model for APC mutations in FAP (after
Albuquerque et al).
‘Total’ repeat number is summed over both APC alleles.
Germline APC 20aa repeat SomaticAPC 20aa repeat Total 20aa
mutation number mutation number repeat number
< codon 1285 0 > codon 1285 1 (or 2) 1 (or2) codon 1285-1398 1 LOH 0 1
> codon 1399 2 < codon 1284 0 2
The ‘just right’ hypothesis is intriguing and consistent with the original suggestion that APC mutations are selected to produce an optimal level of p-catenin activity, but certain issues remain unresolved. First, Albuquerque et al. studied relatively large numbers of polyps, but these came from only six FAP patients: in particular, their data on ‘second hits’ associated with germline mutations after codon 1400 came from a single individual and another of their patients had a highly atypical mutation involving a translocation breakpoint within the APC gene. Second, Albuquerque et
al. found undoubted variation in the location of somatic mutations in different polyps
from the same individual, suggesting that the term ‘just-right’ may be a misnomer.
Third, the number of remaining p-catenin repeats was calculated on the implicit
assumption that the mechanism of LOH was deletion, when many forms of LOH do
not involve a change in gene dosage (see Chapter 2.5.5). Fourth, Albuquerque et al.
did not fully characterise all their mutations and may therefore have mis-assigned
some mutation locations.
105 In addition to the reported ‘first hit-second hit’ associations at APC, it can be hypothesised that the type of ‘second hit’ might vary between individuals with the same germline APC mutation. For example, functional germline polymorphisms in
DNA mismatch repair genes might lead to a tendency to acquire ‘second hits’ at^PC by ffameshifts rather than nonsense changes. If such variation among FAP patients exists, and if the frequency of ‘second hits’ is rate-limiting for adenoma development, different severities of disease might be associated with different
‘second hit’ spectra. Patients with very severe disease, for example, might have some underlying tendency to a particular type of mutation (or have experienced some specific environment), manifest in a spectrum of ‘second hits’ which was biased towards a particular mutation type and which differed from that generally observed in FAP.
In order further to assess associations between germline and somatic mutations in
FAP, I undertook a screen of 213 colorectal adenomas from 36 FAP patients in collaboration with Dr. Michael Crabtree. The variation in somatic APC mutation
spectrum among individuals was analysed to assess associations between somatic
mutations and disease severity.
4.2 Study Population
All patients analysed in this study had classical FAP (>100 adenomatous polyps) and
either a known germline APC mutation (in 29 cases) or a dominant family history of
disease on routine screening. Polyp counts were systematically obtained at
106 prophylactic colectomy as previously described (Crabtree et al., 2002). Fresh-frozen or archival samples of colorectal adenoma were taken and H&E-stained sections were cut from each tumour to confirm the presence of at least 60% neoplastic tissue.
DNA was extracted from tumours and paired blood samples using standard methods.
Patients were chosen for analysis primarily on the basis of available samples, although a spectrum of germline mutations along the APC gene was ensured. Eight patients (#603.1, #603.2, #603.3, #531, #564, #304, #468.1, #468.2) were selected specifically for LOH analysis owing to the location of their germline mutations close to codon 1300; archival material only was available from these patients’ tumours.
4.3 Loss of Heterozygosity atAPC
LOH analysis was performed on all tumours in the study, using microsatellites
D5S346, D5S656 and D5S421 which map close to APC. Allelic loss was identified
in 52/213 (24%) of the polyps screened (Table 4.2); in all cases where assessment
was possible, loss targeted the germline ‘wild-type’ allele. Thtxdata confirmed the
association between LOH and germline mutations close to codon 1300, specifically
in the region between the first and second 20aa repeats (codons 1285-1378). LOH
was found in 46/78 (59%) polyps from 7 patients with germline mutations in this
region (at codons 1288, 1303, 1309 and 1323), but in only 6/135 (4%) other patients’
tumours (p<0.0001, Fisher’s exact test). The patients with predominant LOH had
more severe disease than the others in the study (median 3250 wild-type. 1370
polyps, p=0.007, Kruskal-Wallis test).
107 Table 4.2: Loss of heterozygosity 2itAPC,
Patients shown are those from this study (*=only screened for LOH), plus two patients (**) previously reported who provide a 3’ boundary to the LOH-associated region. Interestingly, these patients’ S1392X mutation lies within the second P- catenin degradation repeat, but the low LOH frequency suggests that the repeat is at least partly functional. Indeed, the protein encoded by this allele retains 7 of the 9 amino acids which are strongly conserved among the seven 20aa repeats.
Patient ID Polyp Germline 20aa repeats In germline No. polyps with count mutation mutant allele LOH/total Informative 5 5 4 1077 170FS 0 0/5 136 842 R 2 1 3 X 0 0 /4 6 13 200 2 2 2F S 0 0/1 421 1626 Q 2 3 3 X 0 0 /2 6 23 1425 Intron 6 -3 C/G 0 0 /5 3 52 1370 4 9 5 F S 0 0 /4 561 1954 R 4 99 X 0 1/25 5 5 6 1998 58 2 F S 0 0 /6 6 5 3 1450 6 07 F S 0 0/7 5 7 6 344 6 14 F S 0 0 /3 2 96 N/K Y 9 3 5 X 0 0 /4 152 230 1061F S 0 0/1 2 30 800 1061F S 0 0 /4 634.1 1070 1062F S 0 0 /3 6 3 4 .2 4 0 20 1062F S 0 0/1 100 800 1067F S 0 0/1 6 0 3 .1 * N /K S 1 2 0 1 X 0 2 /4 6 0 3 .2 * N/K S1201X 0 1/10 6 0 3 .3 * N /K S 1 2 0 1 X 0 0/1 5 3 1 * 1500 1 2 61 FS 0 0/2 2 7 2 608 2 1288F S 1 1/1 3 62 3 500 Q 1 3 0 3 X 1 2 /2 2 9 5 3 00 0 1309F S 1 1/1 19 3 50 0 1309F S 1 212 86 1000 1309F S 1 2 /2 560 6 4 85 1309F S 1 1/10 5 6 4* 2 7 90 1323FS 1 37/60 3 0 4 * 3000 1326F S 1 0/0 4 6 8 .1 ** 266 0 S 1 3 9 2 X 1/2 0/6 4 6 8 .2 ** N/K S 13 9 2X 1/2 1/10 4 9 3 2 000 1410F S 2 0/5 501.1 1700 1464F S 2 1/18 5 0 1 .2 500 1464F S 2 0/3 562 N/K Unknown ? 0 /1 2 5 9 0 900 Unknown ? 0/2 6 0 0 123 Unknown ? 0 /7
108 4.4 The Mechanism of Allelic Loss APCat
Fifteen adenomas with unambiguous allelic loss at APC were analysed for a reduction in APC copy number using RQM-PCR. To establish the sensitivity of the assay in detecting a reduction of APC copy number in tumour biopsies containing contaminating normal tissue, titrations of DNA extracted from an FAP patient carrying a germline APC deletion and a normal individual were performed (Figure
4.1). Whilst control samples from the APC deletion patient showed 2EE(-ddCt) values (expected value for normal ~ 1, expected value for single copy deletion ~ 0.5) between 0.54 and 0.69 (mean=0.63, s.d.=0.05), samples from the normal individual showed values between 0.79 and 0.94 (mean=0.86, s.d.=0.05). Reduction in APC copy number was readily detectable in samples containing up to 30% of
'contaminating' normal DNA (2EE(-ddCtO) range=0.62 to 0.72, mean=0.66, s.d.=0.08), with no overlap between ±1.96s.d. of the '30% contaminated' and '100% normal' samples. Accordingly, all tumours chosen for RQM-PCR analysis had LOH values below 0.3 (mean=0.24, range=0.16-0.30) averaged over three microsatellite markers, D5S346, D5S656 and D5S421, indicating that the contamination with normal tissue was less than 30%.
109 Figure 4.1: Mean RQM-PCR results (2EE(-ddCT) values) of five titrations of
DNA mixed from a normal individual and an FAP patient with a germline APC deletion.
Bars denote 1.96x standard deviation. Samples containing up to 30% 'contaminating' normal DNA are readily detectable as carrying an APC deletion.
1.00
0.95
0.90
0.85
0.80
0.75 ♦ Mean of 2EE-ddCt values of five titrations 0.70
0.65
0.60
0.55
0.50 0% 20% 30% 40% 50% 100% Percentage of 'contaminating' normal DNA
All 15 tumours with allelic loss which were analysed by RQM-PCR showed 2EB(- ddCt) values between 0.79 and 0.97 (mean=0.85), consistent with diploid APC copy number. In agreement with this finding, two of these tumours analysed by comparative genomic hybridisation (CGH) displayed normal profiles on 5q (see
Chapter 6). Based on data showing that only a part of chromosome 5q showed LOH
(see Chapter 6), gene conversion resulting from homologous mitotic recombination in mitosis was the likely cause.
110 4.5 Truncating Somatic Mutations atAPC
In all fresh-frozen polyps, exons 4-14 and exon 15 A-I of APC were screened using fluorescence-SSCP analysis. Exon 151 extends past the first APC SAMP repeat
(codon 1580) which may be taken as a 3’ boundary for truncated proteins both in the germline for classical FAP and the soma in all types of tumour. Samples with bandshifts were sequenced in duplicate from a new PCR product and mutations were identified. A total of 52 (44%) truncating somatic APC mutations were detected in
141 polyps (Table 4.3). No tumour had ‘three hits’. Mutations were classified according to the number of intact 20aa repeats which resulted. Of the truncating changes, 7 (13%) were prior to the 3’-end of the first 20aa repeat (before codon
1284), 16 (31%) were between the first and second repeats (codons 1285-1378), four
(8%) lay within the second repeat, 24 (46%) were between the second and third repeats (codons 1399-1494), one (2%) was within the third repeat and none was after the third repeat (codons 1515-1580). Thirty-six (69%) mutations were frameshifts and 16 (31%) were nonsense changes. Of the latter, 11 (69%) were C->T transitions and 5 (31%) were G->T transversions.
I l l Table 4.3: Truncating somatic mutations at/lPC.
Mutations are shown by site (codon) and type. FS = frameshift, LOH = allelic loss.
Mutations were assumed to leave no 20aa repeat if they occurred prior to codon
1265, one repeat if they occurred between codons 1285 and 1378, two repeats if they were between codons 1399 and 1494 and three repeats if they were between codons
1515 and 1580 (3’ limit of mutation screening). Specific assumptions were made about the effects of mutations within 20aa repeats on the basis of the precise position: (i) 1387FS leaves one repeat intact; (ii) 1397FS, E1397X, 1380FS, 1398FS and 1495FS leave 2 repeats intact; and (iii) 1509FS leaves three repeats intact.
Tumours with LOH and tumours screened only for LOH are not shown herein (see
Table 4.2). In addition to the results shown here, 24 polyps from 12 FAP patients were screened for somatic mutations (including LOH), but no changes were found at all in these patients; they had the following germline mutations: codon 222FS;
Q233X; codon 1062FS (2 patients); codon 1067FS; large duplication involving much of exon 15H; 1410FS; 1464FS; and unknown (4 patients).
112 Patient Polyp No. Germline 20aa Polyp Somatic Type of somatic Total 20aa ID count polyps mutation repeats in ID mutation change repeats in screened germline soma mutant allele 554 1077 5 170FS 0 1 1398FS 4192delAG 2 2 1439FS 4316delC 2 3 Q1371X C->T 1 136 842 4 R213X 0 1 1307FS 3919delA 1 2 1387FS 4160insAT 1 623 1425 5 Intron 6 - 0 1 1465FS 4393delAG 2 3 C/G 2 1309FS 3927delAAAGA 1 3 1309FS 3927delAAAGA 1 352 1370 4 495FS 0 1 1435FS complex ins/del 2 561 1954 25 R499X 0 1 1489FS 4466insAC 2 2 1465FS 4393delAG 2 3 Q1338X C->T 1 4 1439FS 4317delT 2 5 1309FS 3927delAAAGA 1 6 E1306X G->T 1 7 1450FS 4348d0lCGAG 2 8 1309FS 3927delAAAGA 1 9 1309FS 3927delAAAGA 1 10 Q1406X C->T 2 11 1309FS 3927del/\AAGA 1 12 1426FS 4277delG 2 13 1489FS 4466delT 2 14 1439FS 4316delC 2 556 1998 6 582FS 0 1 1509FS 4527lnsCT 3 2 1308FS 3924delAG 1 3 E1322X G->T 1 4 Q1429X C->T 2 653 1450 7 607FS 0 1 1414FS 4240delG 2 576 344 3 614FS 0 1 1495FS 4484lnsA 2 2 1436FS 4306del13bp 2 3 1482FS 4446del10bp 2 296 N/K 4 Y935X 0 1 Q1338X C->T 1 152 230 1 1061FS 0 1 1465FS 4393delAG 2 230 800 4 1061FS 0 1 1438FS 4312delA 2 2 1302FS 3905delT 1 560 6485 10 1309FS 1 1 S874X 2621delCAAAGC 1 2 R283X C->T 1 3 R302X C->T 1 4 R302X C->T 1 5 R283X C->T 1 6 1163FS 3488insTC 1 501.1 500 18 1464FS 2 1 1303FS 3908insC 3 562 N/K 12 Unknown ? 1 1488FS 4463delT 2+ 2 R1450X C->T 2+ 3 1438FS 4312delA 2+ 4 1465FS 4393delAG 2+ 5 E1397X G->T 2+ 6 1397FS 4188delT 2+ 7 R1450X C->T 2+ 590 900 2 Unknown ? 1 E1305X G->T 1 + 2 1380FS 4138delAC 2+ 600 123 7 Unknown ? 1 E1151X G->T 1 +
113 4.6 Associations of Germline and SomaticAPC Mutations in FAP
adenomas
I searched for associations between germline and somatic mutations based on the total number of 20aa repeats remaining in both mutant alleles. There was a strong tendency (Table 4.4) for patients with no 20aa repeats in their germline allele to acquire ‘second bits’ which left one or, more commonly, two repeats. By contrast, patients with one repeat in the germline largely acquired a ‘second bit’ by LOH
(Table 4.2); on the assumption that LOH bad occurred by mitotic recombination, the mutant alleles would have encoded a total of two repeats between them. ‘Second bits’ were detected infrequently in the patients whose germline mutations encoded a protein with two repeats (Tables 4.2 and 4.3), suggesting the presence of cryptic somatic mutations outside the MCR in these individuals’ tumours.
In general, no clear evidence (Table 4.3) was found to show that individuals tended to acquire specific types of ‘second bit’ (whether nonsense vs frameshift mutations, or C->T vs G->T changes). There was, however, a tendency for exceptional patients to show clustering of particular mutation types (Tables 4.2 and 4.3). One patient
(#560) with a codon 1309 germline mutation did not adhere to the ‘rule’ that bis polyps should almost all show LOH. Only one of bis ten adenomas showed LOH.
The 6 other polyps with identifiable mutations bad a frameshift change at codon
1163, one unusual 6bp deletion producing a nonsense change at codon 874, and four
nonsense C->T mutations in exon 8. One other patient (#576) showed evidence of a
distinctive mutation spectrum. All of bis polyps with detected mutations bad
acquired frameshift changes, and two of these tumours had deletions (lObp and
114 13bp) which were larger than those found in any of the other polyps in the study
(Table 4.3). Omitting the exceptional patient #560, disease severity was not associated with the predominant type of mutation (frameshift vs nonsense or C->T vs
G->T) in individuals’ polyps (p=0.43 and p=0.46 respectively, Kruskal-Wallis test).
The two individuals with unusual somatic mutation spectra did not have notably severe disease for their germline mutations.
Table 4.4: Association between germline and somatic APC mutations in FAP adenomas.
Only tumours screened for both LOH and truncating mutations are included. This table shows the number of tumours with zero, one, two or three 20aa repeats encoded by the germline and somatic alleles. LOH is assumed to result in an allele identical to the germline mutant. Compared with expectation, there is a deficiency of polyps with a total of zero or three remaining 20aa repeats encoded by their APC alleles, a corresponding excess of those with a total of two repeats, and a smaller excess of those with one repeat (x^6=18.6 (Yates’s correction), p<0.005)).
Somatic
0 1 2 3
0 2 13 20 1 36
Germline 1 6 9 0 0 15
2 0 1 2 0 3
8 23 22 1 54
115 4.7 Discussion
This study has addressed four issues with respect to germline and somatic mutations in FAP: (i) confirmation and refinement of the association between LOH, germline mutations near codon 1300 and more severe colonic disease; (ii) an assessment of the
‘just right’ hypothesis regarding truncating germline and somatic APC mutations;
(iii) an assessment of the tendency of individual FAP patients to acquire specific types of somatic mutations, independent of the germline mutation; and (iv) a search for evidence that the severity of FAP is associated with the somatic mutation spectrum in an individual’s polyps and hence with modifier genes acting through specific hypermutation pathways.
It has been confirmed that FAP patients with germline mutations close to codon 1300 have a strong tendency to undergo LOH in their colorectal polyps (Lamlum et al.,
1999). This region has been extended to include that between codons 1288 and 1323, and it has been shown that the LOH-associated region does not include codon 1261 or codon 1392. Thus, LOH generally occurs in patients with germline mutations between the first and second 20aa repeats of APC which are involved in p-catenin binding and degradation. Furthermore, it has been shown that LOH in these tumours tends to occur by a mechanism which does not result in a net change in APC gene dosage.
It has previously been hypothesised that the severe polyposis of patients with
germline mutations near codon 1300 might occur because LOH both produced a
strongly selected genotype and occurred spontaneously at a higher frequency than
116 truncating mutations close to codon 1300 (Lamlum et al., 1999). Data from this study
(from patients with mutations between codons 1288 and 1326) support this mechanism. However, the hypothesis also predicts that patients with germline mutations between codons 1399 and 1494 should develop more severe disease than patients with germline mutations before codon 1264: in the former case, polyps can
acquire a total of two 20aa repeats owing to somatic mutations almost anywhere
before codon 1264; but the latter group of patients must acquire somatic mutations in
a small region (codons 1285 and 1378) to give a total of one or two repeats, and
presumably the chances of such an event are lower. The findings of this study on 149
FAP patients support this suggestion. Of the 137 patients with mutations before
codon 1264 (excluding exons 1-4 and 9), median polyp number was 800,
significantly fewer than the median of 1850 for the 12 patients with mutations
between codons 1399 and 1494 (p<0.043, Kruskal-Wallis test).
The data support the original suggestions (Groves et al., 2002; Lamlum et al., 1999;
Rowan et al., 2000) that the ‘two hits’ at APC are non-independent and that the
resulting level of p-catenin activity must be neither too high nor too low. In this
respect, Albuquerque et al. (Albuquerque et al., 2002a) and this study agree.
However, the data differ in a small, but important detail. First, given that LOH
appears most often to occur by mitotic recombination, patients with LOH and
mutations near codon 1300 generally have a total of two 20aa repeats (summing over
both APC alleles) in their polyps. This observation may explain why patients with
mutations after codon 1400 rarely show LOH, because such an event would
generally result in a total of four repeats remaining. Second, no strong tendency was
found for polyps from patients with germline mutations before codon 1284 to acquire
117 somatic mutations which retained one p-catenin binding repeat rather than two; indeed, mutations which left two repeats remaining were actually more common in this samples. Thus, even if one assumes that a simple summation of the remaining
20aa repeats in APC can be equated to P-catenin activity, some variation in this activity can be tolerated in colorectal tumorigenesis (Table 4.4). The level must be right, but perhaps not ‘just right’.
In some patients, ‘second hits’ were hard to find (Tables 4.2 and 4.3). These patients tended to have germline mutations with two remaining 20aa repeats and were therefore predicted to acquire a second hit which causes absent protein or protein
truncated before the first P-catenin degradation repeat. Given that a sensitive
mutation detection method was used and that all tumours are expected to have ‘two
hits’ at APC, it is likely that the latter group of patients had a high frequency of
relatively large-scale changes, or mutations in introns, or even promoter méthylation.
It is not necessarily to be expected that the ‘first hit-second hit’ association for APC
is identical in FAP and sporadic cancers. Nevertheless, the data on FAP tumours are
supported by those from sporadic colorectal cancers. An analysis of all sporadic
colorectal tumours with two reported somatic mutations (http://perso.curie.ff/APC/l
shows an association of LOH with mutations around codon 1300, as expected
(details not shown). There is also, however, a clear over-representation of a genotype
comprising one mutation before codon 1285 and the other after codon 1399 (Table
4.5), resulting in a total of two 20aa repeats (ignoring the potential confounding
effect of polyploidy). The total number of 20aa repeats in sporadic colorectal cancers
is approximately normally distributed (p>0.63, Shapiro-Wilk test) with a mean and
118 median of two repeats in total, but with considerable numbers of tumours with one
(or zero or three) repeats.
Table 4.5: Somatic APC mutations in sporadic colorectal cancers.
Data are derived from the APC mutation database and REF (Rowan et al., 2000).
Positions of the APC mutations are shown by their codon. Tumours with LOH have been omitted from the analysis and mutations between codons 1399 and 1580 are pooled owing to relatively small class sizes. Expected frequencies are calculated assuming the two hits occur randomly. The ‘two hits’ are strongly associated
(X^=16.2, d.f.=4, p<0.005), largely owing to the over-representation of cancers with one mutation before codon 1285 and one after codon 1399. Note that mutations between codons 1286 and 1398 are under-represented in this Table, since most are associated with LOH. For simplicity, mutations within the 20aa repeats are assumed in this analysis to disrupt the repeat, although omitting these cancers from the analysis does not alter the conclusions (details not shown).
APC genotype Total 20aa repeats No. cancers Expected no. cancers <1285, <1285 0 4 9.5 <1285, 1286-1398 1 9 8.6 <1285, >1399 2 27 16.4 1286-1398, 1286- 2 4 1.96 1398 1286-1398, >1399 3 3 7.5 >1399, >1399 4 4 7.1
119 The MCR in colorectal tumours in FAP can therefore be explained as follows.
Classical FAP is associated with germline mutations between codons 160 and 1580
approximately. Mutations at codons 1061 and 1309 are particularly common, but
germline changes are otherwise evenly spread throughout the gene, presumably by
chance. Patients with one 20aa repeat encoded by the germline mutant tend to
acquire ‘second hits’ by LOH and therefore contribute relatively little to the MCR
data. Patients with germline mutations after the second 20aa repeat (codons 1399-
1580) are uncommon, because this is a relatively small region. Thus the bulk of the
MCR data comes from patients with mutations proximal to codon 1264, and these
individuals tend to acquire somatic mutations which leave one or two 20aa repeats
intact; these mutations occur between codons 1285 and 1494, essentially equivalent
to the MCR. A similar reasoning applies to sporadic colorectal tumours.
Limited evidence was found to show that the type of somatic APC mutation tended
to vary among patients independent of the site of the germline mutation. Within the
constraints of the sample size, most patients had no notable excess of any particular
type of mutation. Disease severity was not associated with the proportions of (i)
frameshift vs nonsense or (ii) C->T vs G->T mutations in individuals’ polyps, thus
providing no clear evidence for the action of genes which modify disease severity in
this way. Two of the patients did, however, show clustering of somatic mutations.
Patient #560 with a codon 1309 germline mutation had predominantly nonsense
somatic mutations rather than the expected LOH. Thus the ‘first hit-second hit’ rule
for APC is hot invariable. Any explanation for the exceptional patient must be
largely speculative. No germline sequence variants were found in or around exon 8
to explain the apparent clustering of ‘second hits’ in this region in the patient’s
120 tumours. Perhaps this individual has some unexplained tendency to hypermethylation which preferentially affects cytosine residues in exon 8 of APC\ these residues may then undergo deamination to produce thymidine. Other possibilities are that variation elsewhere in the genome means that this patient’s polyps are tightly constrained to contain a total of only one p-catenin repeat, or that the patient carried a deleterious, recessive mutation in another gene on the same chromosome as the APC mutation (in which case LOH would create a homozygous mutant and would therefore be selected against). Patient #576 also showed evidence of a distinctive mutation spectrum, with two relatively large deletions. Inspection of the data of Albuquerque et al
(Albuquerque et al., 2002a), moreover, reveals a patient who had R->X changes in
7/10 polyps, compared with 8/55 of these mutations in their other patients (p<0.01,
Fisher’s exact test). These data emphasise the dangers of basing conclusions regarding APC mutation patterns on small numbers of patients.
There may also be a tendency for somatic APC mutations to cluster in a different way, in that the same mutation may tend to occur in different polyps from the same
individual. Patient #560 had two identical mutations at codon 283 and two at codon
302. Again, inspection of the data of Albuquerque et al. (Albuquerque et al., 2002a)
provides evidence of this mutation clustering. One of their patients (#2) had four
nonsense mutations at codon 1306, another (#4) had two nonsense mutations at
codon 1291, and yet another (#6) had nonsense changes at codons 805 (2 polyps),
876 (3 polyps) and 1114 (2 polyps). Using mutation frequencies from the APC
Mutation Database, all these clusterings occurred significantly more often than
expected (p<0.02 for all examples, exact binomial test, details not shown). The
underlying reasons remain unclear, although multiple sampling from single polyps
121 may not be the only explanation. It is possible, for example, that the polyps with the same somatic mutation had a single origin, but presented as distinct lesions having undergone some process such as crypt fission and migration (Wasan et al., 1998).
In conclusion, the number of remaining 20aa P-catenin degradation repeats appears to be a crucial determinant of the selective advantage conferred by the mutant APC protein. Germline mutations which truncate the protein leaving one such repeat are associated with LOH in colorectal tumours, and this association may be sufficient to explain why these patients have severe polyposis. LOH tends to occur by mitotic recombination and generally leaves a total of two 20aa repeats intact in the two mutant proteins. Other APC ‘first hits’ are associated not with LOH, but with second truncating mutations in specific sites. In general, these ‘second hits’ leave a total of one or two 20aa repeats encoded by the two APC alleles. These ‘first hit-second hit’ associations at APC are also found in sporadic colorectal tumorigenesis. The constraints on the type and site of APC ‘second hits’ strongly suggest corresponding constraints on the resulting level of p-catenin activity in the cell, even allowing for the view that summation of the number of 20aa repeats encoded by the two mutant alleles is perhaps a simplistic way of assessing protein expression and fimction. The data clearly show, however, that some variation in APC mutation is tolerable. APC mutations thus seem to follow a ‘loose fit’ model (Table 4.6) rather than a ‘just right’
model.
122 Table 4.6: ‘Loose fît’ signalling model for APC mutations in FAP and sporadic colorectal tumours.
The model shows the most common mutation combinations; there are many exceptions to these ‘rules’. The association for mutations within the 20aa repeats is not currently clear; it seems likely that some mutations leave repeat function intact, whereas others disrupt it.
Germline or first 20aa repeat Somatic or 20aa repeat Total 20aa
mutation number second mutation number repeat number
< codon 1265 0 > codon 1285 1 or 2 1 or 2
Codon 1285-1378 1 LOH 1 2
Codon 1399-1494 2 < codon 1265 0 2
Codon 1515-1580 3 < codon 1265 0 3
123 Chapter 5 The Relationship between Germiine and Somatic
Mutations at the APC iocus in Patients with AAPC
5.1 Introduction
The spectrum of somatic APC mutations in FAP patients has been shown to be dependent on the position of the germline APC mutation. These ‘first hit-second hit’ associations appear to reflect selection for an optimal level of Wnt signalling and correlate with disease severity in FAP (see Chapter 4). These observations raise the possibility that similar molecular genetic mechanisms underlie the association of specific germline APC mutations with AAPC.
Two studies have already reported specific somatic APC mutation spectra in AAPC patients. Spirio et al. studied somatic APC mutations in colorectal tumours from a single AAPC family with a frameshift mutation in APC exon 4 (nt426delAT) (Spirio et al., 1998). This exon has been shown to be subject to alternative splicing, suggesting that the frameshift mutation is by-passed in a proportion of transcripts, resulting in near full-length protein retaining at least partial tumour suppressor function (Samowitz et al., 1995). Eight of 66 tumours were found to show allelic loss at APC, but - unlike the situation in classical FAP - LOH consistently targeted the inherited mutant rather than the wild-type allele. Six of these tumours were found to also harbour truncating somatic changes, which were assumed, but not proven, to target the germline mutant allele. Mutation analysis on 35 further tumours without
allelic loss revealed single truncating mutations in 21 lesions, which in four cases
124 were formally shown to target the wild-type allele. One tumour (#18) had two truncating somatic APC mutations, one in each allele. Notably, eight of the somatic mutations detected were nt4664insA (codons 1554-6), a change which occurs rarely in sporadic colorectal tumours and very rarely in tumours from FAP patients.
Based on th?^data, Spirio et al. proposed that germline APC mutations in the 5’- region encode proteins with residual activity which modulates the spectrum and frequency of mutations that lead to adenoma formation; either directly by reducing the frequency of loss of the wild-type allele and affecting the frequency and type of small somatic mutations, or indirectly as a result of selection. As model for the latter scenario, Spiro et al. proposed that loss of the wild-type allele did not occur, because the residual function of the inherited mutant allele was enough to retain tumour suppressor activity. Frameshift and stop mutations targeting the wild-type allele would have to be of a type (such as dominant-negative) which reduced the residual function of the inherited mutant allele. Together, these constraints would reduce the frequency and limit the type of mutations that could inactivate the wild-type allele and hence lead to adenoma formation.
The evidence for this hypothesis is, however, not entirely conclusive. The model is based on the assumptions that LOH at APC is a near-universal feature of FAP tumours, and that the eight unusual somatic mutations detected, nt4664insA, indeed target the wild-type allele. LOH at APC is, however, largely restricted to FAP patients with germline mutations between codons 1285-1378, and the single nt4664insA change (tumour #18) formally assessed was shown to target the inherited mutant allele. The five truncating changes which were shown to target the wild-type
125 allele had similar locations to those commonly found in tumours from FAP patients with germline APC mutations prior to codon 1265. Furthermore, only a limited region of APC was screened for somatic mutations, raising the possibility that the proportion of tumours with ‘third hits’ was underestimated.
Su et al. studied the somatic APC mutation spectrum in colorectal tumours from an
AAPC family with a nonsense mutation (R332X) in APC exon 9, a change which again is subject to alternative splicing (van der Luijt et al, 1995). Su et al. showed that the wild-type isoprotein derived from the transcript lacking exon 9 retained its ability to regulate (3-catenin mediated transcription (Su et al, 2000a). Analysis of nine adenomas revealed at least one truncating APC change in each tumour. Four tumours carried two truncating somatic mutations, and one tumour carried one truncating somatic mutation and loss of the inherited mutant allele. An excess of adenine insertions (nt4666insA) was observed at the same Ag-tract as in the study by
Sprio et al. and this change was shown to target the inherited mutant allele. The nt4666insA mutation did not abolish - but possibly reduced - the ability of the full- length protein or the isoprotein lacking exon 9 to regulate (3-catenin mediated transcription.
Based on these data. Su et al. proposed that carriers of germline mutations in the alternatively spliced part of APC exon 9 develop fewer adenomas than do classical
FAP patients, because somatic inactivation of both APC alleles is required for tumorigenesis in the former. However, they contended, these patients develop colorectal tumours more frequently than the general population, because the inherited mutant allele - owing to the low level of isoprotein lacking exon 9 - is
126 more easily inactivated by mutations that do not significantly affect the tumour- suppressor function of wild-type APC, such as the observed nt4666insA change at an apparently hypermutable Ae-tract. Su et al. suggested that this model might also explain the findings by Spirio et al., providing for a common explanation for AAPC associated with 5’-end and exon 9 APC mutations.
The model of Su et al. was based on elegant functional studies, but on relatively few tumours. The model did not, moreover, take into account the requirement for the mutant APC alleles to produce an optimal level of Wnt signalling rather than simple gene inactivation. In order to assess further the association between germline and somatic mutations in AAPC, I analysed 72 colorectal adenomas from eight AAPC patients with a R332X nonsense mutation for somatic changes and LOH at APC. The mutation data were compared to those of Spiro et al. and Su et al., as well as to my data from classical FAP patients.
5.2 Study Population
Eight AAPC patients with a R332X nonsense mutation in the alternatively spliced region of APC exon 9 were recruited for this study. Six patients were members of the same family, and four of these were known to also carry a germline missense variant
(T905R) of unknown functional significance in the mismatch repair gene MSH2.
Table 5.1 shows the APC and status of each patient. Detailed polyp counts were generally not available. A total of 72 formalin-fixed, paraffin-embedded adenomas (median size=3mm; range= 1.5-17mm) were collected from these patients.
127 and the presence of at least 60% neoplastic tissue was confirmed by analysis of haemotoxylin and eosin-stained sections.
5.3 Somatic Mutations and Loss of Heterozygosity APCat
LOH analysis was performed on 72 colorectal adenomas from eight AAPC patients with a R332X germline mutation using three microsatellite markers, D5S346,
D5S656 and D5S421, that map close to APC. As linkage information was available for the microsatellites studied, the allele targeted by the allelic loss could be determined. APC mutation detection at codons 1258 to 1603, encompassing the somatic MCR, was performed on all tumours using fluorescence-SSCP analysis.
Samples with bandshifts were sequenced in duplicate from a new PGR product.
Somatic APC changes were detected in 49 of 72 (68%) colorectal adenomas (Table
5.1). LOH was found in 12 of 72 (17%) lesions, in nine cases involving the germline wild-type allele and in three cases, the mutant allele. I found truncating somatic mutations in 46 of 72 (64%) tumours; nt4661insA was the most frequent somatic mutation (33 of 49, 67%).
Two somatic changes were detected in twelve (17%) polyps. In classical FAP and sporadic colorectal tumours, by comparison, ‘third hits’ are almost never detected
(see Chapter 4). The combinations of somatic mutant APC alleles in the R332X tumours were of three main types: (i) one 20aa repeat plus loss of germline mutant (3 tumours); (ii) three 20aa repeats plus loss of germline wild-type (6 tumours); and (iii)
128 three 20aa repeats plus one 20aa repeat (2 tumours). In addition, in one tumour the mutations left zero 20aa repeats plus two 20aa repeats.
Of the tumours with one detected somatic mutation, 25 were 4661insA, three other mutations left three 20aa repeats and one occurred within the third repeat. Four mutations left two 20aa repeats and four left one 20aa repeat. Three tumours had loss of the wild-type allele, but no detected truncating mutation. Overall, the proportion of truncating APC mutations between codons 1258-1603 which left three repeats in the protein was significantly higher than in classical FAP (36/49 1/45, p<0.001,
Fisher’s exact test) (see Chapter 4) and sporadic colorectal cancers (36/49 6/278, p<0.001, Fisher’s exact test) {APC mutation database at http://perso.curie.fr/Thierry.
Soussi/APC. html).
To exclude any functional effect of the MSH2 variant, 14 adenomas were analysed for microsatellite instability (MSI) using a further highly sensitive microsatellite marker, BAT26 (Loukola et al., 2001). From patients carrying both APC R332X and
MSH2 T905R mutations, none of 14 adenomas analysed at BAT26 and none of 55 adenomas analysed at D5S346, D5S656 and D5S421 showed MSI (data not shown).
Furthermore, the frequency and type of truncating somatic mutations in polyps from
R332X/T905R patients was similar to that of R332X patients; both groups showed a similar excess of adenine-insertions at the Ae-tract at nucleotides 4661-6 (27/39
6/10, p>0.5, Fisher’s exact test). Taken together, these data provide good evidence against a functional effect of the MSH2 missense variant.
129 Table 5.1: Somatic mutations and allelic loss 2iiA P C in 72 colorectal adenomas from AAPC patients with a R332X germline mutation in APC.
FS, frameshift; NL, no loss; LOH mut/wt, loss of heterozygosity of the mutant/wild- type allele; N/d, none detected.
Patient Germline Total Polyp Somatic Type of 20aa repeats Allelic loss ID APCIMSH2 polyps ID APC somatic retained by at APC mutation analysed mutation change somatic mutation 578.fpl R332X 6 2 1554-6FS 4661-6 insA 3 NL 3 1462FS 4386 delGA 2 NL 4 1394FS 4182delTA 1 LOH mut 5 1554-6FS 4661-6 insA 3 NL 7 1554-6FS 4661-6 InsA 3 NL 578.IV.1 R332X/ 14 9 1554-6FS 4661-6 InsA 3 NL T 9 0 5 R 11 1505FS 4514 dup7bp 2 NL 14 1554-6FS 4661-6 InsA 3 NL 15 K 1469X 4 4 0 5 C > T 2 NL 16 1554-6 F S 4661-6 insA 3 NL 18 1554-6FS 4661-6 InsA 3 NL 19 1554-6FS 4661-6 InsA 3 NL 20 15 5 4-6 F S 4661-6 InsA 3 NL 21 1554-6 F S 4661-6 InsA 3 LOH wt 22 N/d N/d N/d LO H wt 23 N/d N/d N/d LOH wt 24 N/d N/d N/d LOH wt 578.33 R332X/ 17 26 155 4-6F S 4661-6 InsA 3 NL T 9 0 5 R 27 E 12 8 6X 3856 G>T 1 LOH mut 32 1554-6FS 4661-6 insA 3 NL 33 1554-6FS 4661-6 insA 3 NL 34 1372FS 4114 deIG 1 NL 15 5 4-6F S 4661-6 InsA 3 36 1554-6F S 4661-6 InsA 3 NL 38 1554-6FS 4661-6 InsA 3 NL 39 S1315X 3944 C>A 1 LOH mut 42 15 5 4-6F S 4661-6 InsA 3 LOH wt 43 1554-6FS 4661-6 InsA 3 LOH wt 5 78.se R 3 3 2 X / 15 46 E 1 2 6 5 X 3793 G>T 0 NL T 9 0 5 R 1462FS 4386 delGAGA 2 49 1554-6FS 4661-6 insA 3 LOH wt 50 1554-6F S 4661-6 insA 3 NL 51 155 4-6F S 4661-6 InsA 3 NL 52 1554-6FS 4661-6 InsA 3 NL 53 1554-6FS 4661-6 InsA 3 LOH wt 55 1554-6FS 4661-6 insA 3 NL 58 155 4-6F S 4661-6 InsA 3 LOH wt 59 1462F S 4386 delGA 2 NL
130 Table 5.1: continued.
Patient Germline Total Polyp Somatic Type of 20aa repeats Allelic loss ID APCIMSH2 polyps ID APC somatic retained by at APC mutation analysed mutation change somatic mutation 578.bl R 3 3 2 X / 9 63 1516FS 4546 delAT 3 NL T 9 0 5 R 64 1554-6FS 4661-6 InsA 3 NL 66 1531F S 4591 insC 3 NL 67 1372FS 4117delC 1 NL 1554-6FS 4661-6 InsA 3 68 155 4-6F S 4661-6 InsA 3 NL 69 1554-6FS 4661-6 InsA 3 NL 70 1335FS 4004 del31bp 1 NL 72 1554-6FS 4661-6 InsA 3 NL 5 7 8 .aa R 3 32 X 4 74 155 4-6F S 4661-6 InsA 3 NL 1571.Ü.2 R332X 5 82 1554-6FS 4661-6 InsA 3 NL 91 1462F S 4386 delGA 2 NL 93 1537FS 4611 delAG 3 NL DFAP81 R332X 2 95 1554-6FS 4661-6 InsA 3 NL
5.4 Discussion
My findings on somatic APC changes in colorectal adenomas from AAPC patients with a R332X (exon 9) germline mutation provide clues as to the mechanism of disease causation in these patients. It is evident that the association between germline and somatic APC mutations, which exists in FAP and sporadic cancers, also applies to these patients. Compared with classical FAP and sporadic colorectal cancers, our patients show a greatly increased frequency of ‘third hits’ (second somatic APC mutations) and of changes which leave three of the 20aa |3-catenin binding and degradation repeats in the protein. A frameshift mutations at codons 1554-6 (nt4661-
6insA) is especially common.
131 I identified three basic tumorigenic pathways in my patients with a R332X germline mutation (Figure 5.1): (i) 1-repeat mutation on the wild-type allele followed by (a) loss of mutant or (b) 3-repeat mutation on the R332X allele; (ii) 2-repeat mutation on the wild-type allele, followed by a 0-repeat mutation on the R332X allele; and (iii) loss of the wild-type allele, followed by a 3-repeat mutation on the R332X allele. I have deduced the order of events in pathways (ib) and (ii).
However, I also found a large number of tumours with 3-repeat somatic mutations and no detected ‘third hit’. If these mutations targeted the R332X allele, as Su et al. and I suggest, at least some of those tumours probably harboured genetic or epigenetic APC changes outside the region I screened for mutations. In support of this contention, whilst my material only allowed me to screen codons 1258-1603 of
APC, Su et al. found somatic APC mutations prior to codon 1258 in 2/5 polyps which also had mutations leaving three 20aa repeats. Thus, a fourth genetic pathway must be added to those above (Figure 5.1): (vi) 0-repeat mutation on the wild-type allele, followed by a 3-repeat mutation on R332X.
The model of Su et al. stated that AAPC tumours “require ‘three hits’, but [tumours
are more frequent] than in sporadic cases because R332X is more easily inactivated
than is wild-type APC, by mutations that do not significantly (or weakly) affect the
tumour-suppressor function”. In other words, R332X can be inactivated by almost
any protein-truncating APC mutation (perhaps within the first 1580 codons). This
model may in its essence be correct, but if it is, the Ag-tract around nt4661 must be
extraordinarily unstable relative to other mutation sites, for these mutations account
for at least 25% of all truncating ‘second/third hits’ in R332X tumours.
132 An alternative to the Su model may be formulated in terms of the general finding that true inactivation of APC is rare, and of the specific suggestions of the ‘just right’ and
‘loose fit’ hypotheses. Notably, mutations at and around nt4661 on the R332X allele do not actually “inactivate” APC, but leave three 20aa repeats intact in the minor isoprotein which lacks exon 9. (Mutations elsewhere 3’ of codon 332 may leave different numbers of repeats intact.) It is unknown as to how much p-catenin degradation activity results firom the combination of transcripts encoded by the allele carrying R332X and this somatic mutation, but it can be envisaged that it might be equivalent to an allele carrying a single mutation which leaves one or two 20aa repeats. In that case, a near-optimal genotype according to the ‘just right’ and ‘loose fit’ models would result if such a change followed loss of the germline wild-type allele.
An alternative to the Su model is a stepwise process of tumorigenesis in which tumours broadly follow the ‘just right/loose fit’ models. I propose that the increase in polyp number as compared to sporadic patients could be explained if initial mutation of the wild-type allele were sufficient to allow for modest clonal expansion (in terms of rate and/or cell number), perhaps only under specific conditions such as tissue growth or regeneration. Modest clonal expansion could sufficiently increase the effective mutation rate (by increasing cell number) to cause occasional occurrence of
a ‘third hit’ targeting the inherited mutant allele and producing an optimal APC
genotype, as described above. The stepwise model could also explain the marked variability in disease severity between AAPC patients with the same exon 9
mutation. Any environment or genetic predisposition affecting the likelihood of
clonal expansion after initial mutation of the wild-type allele - for example, tissue
133 injury or a functional polymorphism in a mutator gene - would affect the probability of a ‘third hit’ in APC and hence the polyp number. Notably, this model does not assume an absolute requirement of two somatic APC mutations; mutation of the wild-type allele might be sufficient for adenoma development in some cases, especially if supplemented by a selected mutation elsewhere (for example, K-ras or
BRAF). The requirement for a ‘third hit’ could also vary between R332X patients if these expressed different levels of transcripts lacking exon 9.
This model might also explain the findings of Sprio et al. on exon 4-mutant patients.
However, selective constraints do not appear to be identical in polyps from exon 4- and exon 9-mutant patients (although these data must be interpreted with some caution as they are only derived from a small number of individuals). The frequency of allelic loss observed by Sprio et al. is comparable to that observed in our sample of R332X polyps (8/66 vs 12/72, p=0.63, Fisher’s exact test), and a similar proportion of exon 4- and exon 9-mutant polyps without LOH carried somatic mutations leaving three repeats in the protein (9/34 vs 30/60, p=0.16, Fisher’s exact test). Allelic loss in exon 4 patients, however, exclusively targeted the germline mutant allele. Furthermore, the positions of additional truncating changes in exon 4 adenomas with loss of the germline mutant allele were less restricted, leaving either zero, one, or two 20aa repeats in the protein. These differences might be partly due to chance, but they could also reflect differences in the level of alternative transcripts lacking exon 4 or 9.
In conclusion, this study has provided evidence that a significant proportion of macroscopic polyps from AAPC patients with a truncating germline mutation in the
134 alternatively spliced region o f AP C exon 9 have acquired two somatic APC mutations, one in the wild-type and one in the inherited mutant allele. The inherited mutant allele appears to be specifically targeted by mutations leaving three p-catenin degradation repeats in the isoprotein, with the most common change being an adenine insertion at a hypermutable Ag-tract at codons 1554-6 (nt4661-6). Whilst a
(at least partial) requirement of two somatic mutations could explain the reduced polyp number as opposed to classical FAP patients, hypermutation at codons 1554-6 would have to be extreme to fully account for the significantly increased polyp number as compared to sporadic cases. I propose an alternative stepwise model under which mutation of the wild-type allele leads to limited clonal expansion, and a
‘third hit’ at APC can then produce a genotype which is more strongly selected.
135 Figure 5.1: A model of somatic mutations 2itAPC in colorectal adenomas from
AAPC patients with a R332X nonsense mutation in exon 9.
‘N-repeat mutation’ refers to a truncating somatic mutation leaving N 20aa (3-catenin binding and degradation repeats in the protein. (A) 0-repeat mutation on the wild- type allele, followed by a 3-repeat mutation on the R332X allele. (B) 1-repeat mutation on the wild-type allele, followed by loss of the inherited mutant allele or a
3-repeat mutation on the R332X allele. (C) 2-repeat mutation on the wild-type allele, followed by a 0-repeat mutation on the R332X allele. (D) Loss of the wild-type allele, followed by a 3-repeat mutation on the R332X allele.
Heptad repeats 15-aa repeals SAMP repeats Basic domain (dimérisation) ({Vcatenin binding) (axin binding) (Microtubule binding)
I I 6 57 453 767 1020 1169 1264 1378 1580 2033 2200 2400 2560 2843
Armadillo 20-aa repeats EB1 and HDLG KddzA repeats (|Vcatenin bindin^degradation) tending sites
A W T allele MT allele
B WT allele MT allele or
C W T allele MT allele
D WT allele MT allele
136 Chapter 6 The Extent of Chromosomal Instability in Human
Colorectai Adenomas with Two Mutationai Hits atAPC
6.1 Introduction
The molecular genetic studies described above strongly support the view that APC
mutations are selected, at least in part, for their effects on cellular p-catenin levels
(see Chapters 4 and 5). This contention is further supported by the finding that
colorectal tumours without APC mutations sometimes harbour mutations of p-
catenin which prevent protein destruction and thus have effects similar to mutation of
APC (Ilyas et al., 1997; Morin et al., 1997; Sparks et al., 1998). It is unlikely,
however, that inactivating APC mutations and activating P-catenin mutations are
functionally identical, given, for example, that the former seem to be associated with
a higher probability of progression from colorectal adenoma to carcinoma (Samowitz
et al., 1999). Several groups have suggested, therefore, that loss of C-terminal APC
functions due to truncating mutations provides a selective advantage additional to
that which arises from constitutive Wnt signalling (Fodde et al., 2001a; Ishidate et
al., 2000; Kaplan et al., 2001; Kawasaki et al., 2000; Nathke et al., 1996).
Fodde et al. and Kaplan et al. independently reported that APC may have a role in
chromosomal segregation (Fodde et al., 2001a; Kaplan et al., 2001). Both groups
studied mouse embryonic stem (ES) cells homozygous for a truncating Ape mutation
(Min ES cells) and detected a marked increase in numerical and structural
chromosome aberrations as well as disorganized spindle micro tubules. During
137 mitosis, wild-type APC was found to be localised along kinetochore microtubules and at their ends adjacent to kinetochores. Together, these observations led the authors to suggest a new role for APC in kinetochore microtubule-chromosome attachment and therefore chromosome segregation, with mutations in Ape disrupting this function and resulting in chromosomal (or karyotypic) instability (CESi). In this way, APC mutations may be selected not only directly - through their effects on the
Wnt pathway - but also indirectly, through the hypermutation which they engender in the form of CIN.
An alternative hypothesis proposed by Shih et al. suggests that CIN precedes and contributes to genetic changes at APC (Shih et al., 2001). Shih et al. analysed 32 sporadic colorectal adenomas for loss of heterozygosity (LOH) using digital SNP-
PCR and found relatively high frequencies of LOH on chromosome 5q (55%), Ip
(10%), 8p (19%), 15q (28%) and I8q (28%). Whilst digital SNP-PCR may provide some increase in sensitivity over microsatellite-based methods owing to its use of confidence interval thresholds for scoring LOH, in reality most tumours with LOH in
Shih et al.’s study showed one of the SNP alleles to be at a frequency of 66% or more. This is equivalent to an allelic ratio of 2:1, the usual threshold for scoring
LOH using microsatellites. Most cases of LOH detectable using digital SNP-PCR should, therefore, also be detectable using microsatellite-based LOH.
Evidently, the findings of Fodde et al. and Kaplan et al. using in vitro methods may not apply in vivo and Shih et al. made no direct assessment of chromosomal-scale changes in their tumours. I have therefore studied a set of 55 colorectal adenomas
(average size=5mm; range=l-13mm) from 18 FAP patients with a variety of
138 germline mutations and characterised second hits at APC. Using several analytical methods - flow cytometry (FCM), loss of heterozygosity (LOH) analysis and comparative genomic hybridisation (CGH) - I searched for evidence of aneuploidy and polyploidy in these tumours. In addition, Ki-67 immunocytochemistry was used to assess the proliferative activity in a subset of these lesions. In order to elucidate the molecular mechanism underlying allelic loss at APC, the extent of LOH on chromosome 5 was determined in 20 adenomas.
6.2 Study Population
This study examined 55 colorectal adenomas and four normal biopsies from 18 patients diagnosed with FAP and with a known germline APC mutation. All tumours were tubular adenomas with mild dysplasia (median size=5mm; range=l-13mm) which had either been fresh-frozen at colectomy (n=47) or fixed in formalin and embedded in paraffin (n=8). A minimum of 60% neoplastic material was present in each biopsy as assessed by the analysis of haemotoxylin and eosin-stained sections.
The second hit at APC had been determined in all lesions using standard mutation detection techniques (fluorescence-SSCP analysis, DNA sequencing) and LOH analysis at microsatellite markers close to the APC locus (D5S346, D5S656 and
D5S421) (see Chapter 4). Details of the colorectal adenomas are summarised in
Table 6.1.
139 Table 6.1: Patient\D,APC mutation status and size of the colorectal adenomas analysed as well as analytical methods applied.
LI5/18, loss of heterozygosity analysis at chromosomes 15q/18q; FCM, flow cytometry; Ki-67, Ki-67 immunocytochemistry; CGH, comparative genomic hybridisation; FS denotes frameshift mutations.
Patient Adenoma Germline APC mutation SomaticAPC mutation Adenoma Analytical ID ID (nucleotide; codon) (nucleotide; codon) size (mm) method
N -1 1 4 4 249a 502 A>T; R168X 4192 del 2bp; 1398 FS 4 L I 5/18
249b 502 A>T; R168X 4316 del Ibp; 1439 FS 3.5 L I 5/18
315 502 A>T; R168X 4132 OT; Q1371X 5 .5 LI 5/18, FCM. Ki-67 N -1 1 5 4 2 0 3 1495 OT; R499X 4466 ins 2bp; 1489 FS 4.5 LI 5/18. FCM. Ki-67 298 1495 C>T; R499X 4393 del 2bp; 1465 FS 8 LI 5/18. FCM. Ki-67 300 1495 OT; R499X 4012 OT; Q1338X 6 L I 5/18
301 1495 OT; R499X 4317 del Ibp; 1439 FS 5 LI 5/18. FCM. Ki-67 312 1495 OT; R499X 3927 del 5bp; 1309 FS 7.5 LI 5/18. FCM. Ki-67. C G H 340 1495 C>T; R499X 3916 G>T, E1306X 5 L I5/18. FCM. Ki-67 352 1495 C>T; R499X 4348 del 4bp; 1450 FS 5 LI 5/18. FCM. Ki-67. C G H 240 1495 OT; R499X 3927 del 5bp; 1309 FS 6.5 LI5/18. FCM. Ki-67 243 1495 C>T; R499X 3927 del 5bp; 1309 FS 5.5 L I 5/18
259 1495 C>T; R499X 4216 C>T; Q1406X 6 LI 5/18. FCM. Ki-67 264 1495 C>T; R499X 3927 del 5bp; 1309 FS 5 LI 5/18
350 1495 OT; R499X 4277 del Ibp; 1426 FS 6 LI 5/18. FCM. Ki-67 374 1495 O T; R499X 4466 del Ibp; 1489 FS 5 LI5/18. FCM. Ki-67 194b 1495 C>T; R499X 4316 del Ibp; 1439 FS 6 L I 5/18
N -1 1 7 155a 1842 ins 1 bp; 614 FS 4484 ins Ibp; 1495 FS 2.5 LI 5/18
155b 1842 ins 1 bp; 614 FS 4306 del 13bp; 1436 FS 2.5 LI 5/18
155c 1842 ins 1 bp; 614 FS 4446 del lObp; 1482 FS 1 LI 5/18
N -1 2 6 3 17 3183 del 5 bp; 1061 FS 4312 del Ibp; 1438 FS 2 LI 5/18
135b 3183 del 5 bp; 1061 FS 3905 del Ibp; 1302 FS 1.5 LI 5/18
N -1 0 1 6 308 3863 del Ibp; 1287 FS LOH 5.5 FCM. Ki-67
N -6 09 1 3887 ins 13 bp; 1296 FS LOH 3 LI 5/18
2 3887 ins 13 bp; 1296 FS LOH 3 L I 5/18
5 3887 ins 13 bp; 1296 FS LOH 3 LI 5/18
6 3887 ins 13 bp; 1296 FS LOH 3 L I 5/18
140 Table 6.1: continued.
Patient Adenoma Germline APC mutation SomaticAPC mutation Adenoma Analytical ID ID (nucleotide; codon) (nucleotide; codon) size (mm) method N -609 8 3887 ins 13 bp; 1296 FS LOH 3 L I 5/18
10 3887 ins 13 bp; 1296 FS LOH 3 L I 5/18
N -1 02 6 258 3907 OT; Q1303X LOH 4 .5 L15/18, FCM, Ki-67 335 3907 OT; Q1303X LOH 6 LI 5/18. FCM, Ki-67 N-283 347 3927 de! 5bp; 1309 FS LOH 4.5 LI5, FCM, Ki- 67 N -1 06 6 187 3927 de! 5bp; 1309 FS LOH 7 L I 5/18
N -1 63 3 292 3927 de! 5bp; 1309 FS LOH 6 LI 5/18
N -220 399 3927 de! 5bp; 1309 FS LOH 7 FCM, Ki-67
N -127 128a 3927 del 5bp; 1309 FS LOH 5 L I 5/18
128b 3927 del 5bp; 1309 FS LOH 5 L I 5/18
128c 3927 del 5bp; 1309 FS LOH 5 L15/18, CGH
129 3927 del 5bp; 1309 FS LOH 1 L I5/18, CGH
130 3927 del 5bp; 1309 FS LOH 5 L I 5/18
131b 3927 del 5bp; 1309 FS LOH 3 L I 5/18
131c 3927 del 5bp; 1309 FS LOH 3 LI 5/18, CGH
131d 3927 del 5bp; 1309 FS LOH 3 LI 5/18
N -907 3 3927 del 5bp; 1309 FS LOH 3 L I 5/18
5 3927 del 5bp; 1309 FS LOH 3 L I 5/18
8 3927 del 5bp; 1309 FS LOH 3 L I 5/18
N-205 206 3927 del 5bp; 1309 FS LOH 1 L I 5/18
207 3927 del 5bp; 1309 FS LOH 1 L I 5/18
208 3927 del 5bp; 1309 FS LOH 1 L I 5/18
2 09 3927 del 5bp; 1309 FS LOH 1 L I 5/18
1974/92/B 3927 del 5bp; 1309 FS LO H 11 FCM
N -458 1 3927 del 5bp; 1309 FS LO H 12 FCM
52701 2 92 9-M 3927 del 5bp; 1309 FS LOH 6 FCM
N -6 10 5 3927 del 5bp; 1309 FS LOH 12.5 FCM
N-351 2 4392 del 2bp; 1464 FS LOH 3 L I 5/18
141 6.3 The Assessment of Chromosomal Instability
A total of 55 colorectal adenomas with two characterised mutational hits dXAPC was analysed for CIN. A variety of techniques was used depending on the amount and type of material available. Where possible, the proliferative activity of these lesions was also assessed (Table 6.1).
6.3.1 Flow cytometry and Ki-67 Immunocytochemistry
Four paraffin-embedded and 16 fresh-frozen tumours were analysed for aneuploidy/polyploidy by multiparameter flow cytometry (Table 6.2; Figure 6.1).
Results were of similar quality for paraffin-embedded and fresh-frozen material, with the mean coefficient of variation (CV) of the Gl/GO peak being 4.9±1.2 and 4.1+0.5 respectively. Three of 20 (15%) adenomas contained sub-populations of cells displaying changes in ploidy, with two being near-diploid (DI =1.2 and DI = 0.8) and one being hypotetraploid (DI =1.8).
In addition, Ki-67 immunocytochemistry was performed on all fresh-frozen lesions as well as two normal biopsies to assess their proliferative activity (% Ki-67 positive cells) (Table 6.2; Figure 6.1). No apparent difference in the proportion of Ki-67
expressing cells was observed between diploid tumours and normal colonic tissue.
However, one of the two aneuploid polyps studied displayed an increase in the
proportion of Ki-67 expressing cells (53.1%) as compared to the euploid biopsies
(average = 16.8%; range = 5.0-35.7%).
142 Table 6.2: Results of flow cytometry and Ki-67 immunocytochemistry on colorectal adenomas with two mutational hits at APC and two normal controls.
CV, coefficient of variation
Adenoma ID DNA Index (DI) CV of the Gl/GO peak Percentage of KI-67 positive cells Normal (N-283) 1.0 3.7 21.3 Normal (N-220) 1.0 4.39 11.6 315 1.0 4.07 15.2 203 1.0 3.81 21.6 298 0.8 3.54 29.2 301 1.0 5.32 26.1 312 1.0 3.68 20.8 340 1.0 4.26 11.0 352 1.0 4.29 14.5 240 1.0 3.8 7.0 259 1.0 4.14 14.2 350 1.0 3.45 5.0 374 1.0 4.56 6.3 308 1.0 4.29 17.2 258 1.0 4.25 18.3 335 1.2 3.19 53.1 347 1.0 3.57 35.7 399 1.0 4.42 22.2 1974/92/B 1.0 4.9 / 1 1.0 5.35 / 2929-M 1.8 2.96 / 5 1.0 6.52 /
143 Figure 6.1: Representative results of flow cytometry and Ki-67 immuno cytochemistry on one normal biopsy (N-283; A-C) and one near-diploid colorectal adenoma with two mutational hits at APC (335; D-F).
Panels A and D show propidium iodide (PI) fluorescence against Ki-67-FITC fluorescence of the isotype control sample. The box represents Ki-67 positive events.
Panels B and E show PI fluorescence against Ki-67-FITC fluorescence for the monoclonal antibody-stained samples. Panels C and F show the PI histograms. The near-diploid sub-population is indicated by the arrow.
A c0 s 1 Î ë I li.
R2
4D0 (OD 8E0 1(X8 200 4)0 600 eOO 1000 Propidium iodide emission Propidium iodide emission B
4&0' ■ ' '6&o' ’ ' ' ’ 'tw o 200 4(o' ' ' 'sio' ■ ’ 'sio lOOO Propidium iodide emission Propidium iodide emission
§ Ü s-
' \ 6 o ' 200 4(0 600 800 1000 Propidium iodide emission Propidium iodide emission
144 6.3.2 Loss of Heterozygosity Analysis
Loss of heterozygosity (LOH; allelic loss) analysis was performed at microsatellite markers on chromosome 15q (D15S995 and D15S1007; close to the CRACl locus) and chromosome 18q (D18S46 and D18S470; close to the SMAD4/MADH4 locus) on 49 adenomas. Importantly, all of these lesions contained less than 40% contaminating normal tissue and 27 had shown unequivocal LOH as the second hit at
APC. LOH was detected on other chromosomes in these adenomas, but at a low frequency: whilst 2 of 49 (4%) informative tumours showed LOH at markers on chromosome 15q, none of 48 (0%) showed LOH at markers on chromosome 18q.
Previously published data had shown evidence of LOH on chromosome Ip in 1 of 21
(5%) of these polyps (Lamlum et al., 2000b). No evidence of microsatellite instability (MSI) was found at any marker in any polyp.
The extent of allelic loss on chromosome 5 was determined in 20 adenomas with
LOH as the ‘second hit’ at APC using six microsatellite markers (D5S2845, 5pl4.3;
D5S1470, 5pl3.3; D5S82, 5q21.3; D5S489, 5q22.3; D5S2117, 5q31.1; D5S1456,
5q35.1) (Table 6.3). Nineteen (95%) tumours showed LOH at all informative markers spanning chromosome 5q, with the minimal detectable region of allelic loss encompassing -20Mb. The remaining tumour displayed a normal dosage at the telomeric marker D5S2117. In contrast, none of the 20 adenomas showed allelic loss at markers on chromosome 5p.
145 Table 6.3: Results of LOH analysis on colorectal adenomas with allelic loss as the ‘second hit’ at using microsatellite markers spanning chromosome 5.
LOH, loss of heterozygosity; NX, non informative; NL, no loss; /, no data. LOH at
APC has been scored based on three markers in close proximity to the locus,
D5S346, D5S656 andD5S421.
Patient Adenoma D5S2845 D5S1470 D5S82 APC D5S489 D5S2117 D5S1456 ID ID 5p14.3 5p13.3 5q21.3 5q22.2 5q22.3 5q31.1 5q35.1 D5S346 D5S656 D5S421
N-127 128a NL NL NI NI LOH LOH LOH LOH NI 128b NL NL NI NI LOH LOH LOH LOH NI 128c NL NL NI NI LOH LOH LOH LOH NI 129 NL NL NI NI LOH LOH LOH LOH NI 130 NL NL NI NI LOH LOH LOH NL NI 131b NL NL NI NI LOH LOH LOH LOH NI 131c NL NL NI NI LOH LOH LOH LOH NI
131d NL NL NI NI LOH LOH LOH LOH NI N-907 3 NL NL LOH LOH LOH LOH LOH LOH LOH
5 NL NLLOH LOH LOH LOH LOH LOH LOH 8 NL NL LOH LOH LOH LOH LOH LOH LOH N-609 1 NL / LOH LOH LOH LOH LOH LOH LOH 2 NL NL LOH LOH LOH LOH LOH LOH LOH
5 NL NL / LOH LOH LOH LOH LOH LOH 6 NL NL LOH / LOH LOH LOH LOH LOH
8 NL NL LOH LOH LOH LOH LOH LOH LOH 10 NL NLLOHLOH LOH LOHLOH LOH LOH N-205 207 NL NLLOH LOH LOH LOH LOH LOH LOH 208 NL NL / LOH LOH LOH LOH LOH LOH
209 NL NL/LOH LOH LOH LOH LOH LOH
146 6.3.3 Comparative Genomic Hybridisation
CGH analysis on five fresh-frozen tumours containing at least 60% neoplastic material, two of which had been shown to be euploid by FCM, revealed normal CGH profiles (Figure 6.2). Interestingly, three of the polyps had shown LOH at APC, with allelic loss involving at least 20Mb of chromosome 5q (Table 6.3). One of these polyps had also shown LOH on chromosome 15q. Whilst the region of loss at 15q may have been below the resolution of CGH, the failure to detect a deletion of chromosome 5q is in accordance with my real-time quantitative multiplex PCR
(RQM-PCR) results, showing that LOH at APC does not result from physical loss of material (see Chapter 4).
147 Figure 6.2: A representative CGH result of one tumour, 131c, which showed allelic loss as the ‘second hit’ 2it APC by microsatellite analysis.
Male tumour DNA was co-hybridised with female reference DNA onto normal male metaphase spreads. The composite CGH profile shows 95% confidence intervals of the mean values from six metaphase spreads, with threshold values for chromosomal gain and loss of 1.15 and 0.85, respectively. Gains are indicated by bars to the right, losses by bars to the left of the chromosomal ideograms. Tumour 131c shows no chromosomal imbalances with the exception of the sex-mismatch control, a relative gain of the Y and loss of the X chromosome.
T 1 1 1
2
i p Ij,
7 10 11 12 if i 1 k
13 14 15 16 17 18
Î-
Î ! § ■
19 20 21 22
148 6.4 Discussion
My results using a variety of experimental methods show that the majority of adenomas with two mutational hits at APC are diploid or near-diploid. Just 3 of 20
(15%) tumours displayed evidence of chromosomal (or karyotypic) instability as determined by FCM and CGH analysis, two being near-diploid and one being hypotetraploid. Three further samples showed normal DNA profiles by CGH analysis indicating an absence of gross unbalanced karyotypic rearrangements.
Furthermore, CIN was not associated with the type of ‘second hit’ (truncating mutation or LOH) at APC. One of the two near-diploid lesions showed an increase in the proportion of Ki-67 expressing cells as compared to all euploid biopsies. Given that models of colorectal tumorigenesis predict increasingly aggressive features as tumours progress, it is likely that this tumour had become aneuploid as part of its progression rather than as a direct result of inactivation of APC.
Complementing these findings, two mutational hits at APC have previously been identified in near-diploid, MSI-positive colorectal cancer cell lines with only a few
(< 5) chromosomal rearrangements (Abdel-Rahman et al., 2001; Rowan et al., 2000).
Two of these cell lines (LoVo, VAC05) harboured bi-allelic truncating APC mutations, and two showed a truncating APC mutation and LOH (DLDl,
GP2d/GP5d).
Apart from at the APC locus, LOH was uncommon in the polyps of this study, being found at a frequency of about 5% at chromosomes Ip (Lamlum et al., 2000b) and
15q, but being absent at chromosome 18q. In 20 analysed adenomas, allelic loss at
149 APC was associated with loss at markers on chromosome 5q but not 5p, with the minimal region of allelic loss encompassing ~20Mb. CGH analysis of three of these polyps, however, revealed no unbalanced chromosomal rearrangements. Together with my RQM-PCR findings on colorectal adenomas showing that LOH at APC does not result from deletion of material (see Chapter 4), these data suggest that the molecular mechanism of allelic loss at APC is nearly always mitotic recombination.
My data are in agreement with a recent report by Haigis et al. (Haigis et al., 2002), who analysed 18 colorectal polyps from Apc^^'* mice and six human adenomas from patients without FAP which had uncertain APC mutation status. Haigis et al. used interphase FISH analysis on selected mouse/human chromosomes and found no evidence of chromosomal gains or losses. In tumours from Apc^^'' mice, this result was confirmed by karyotypic analysis. Furthermore, allelic loss at Ape was shown to commonly occur by somatic recombination in Min adenomas. Other studies have found early colorectal adenomas to be near-diploid in most cases, although larger and/or more dysplastic lesions tend to become aneuploid/polyploid (Giaretti et al.,
1988; Goh, 1987; Hamada et al., 1987; Quirke et al., 1986; Suzuki et al., 1995; van
den Ingh et al., 1985; Weiss et al., 1985).
As in the study of Shih et al. (Shih et al., 2001), LOH was detected at sites on
chromosomes Ip and 15q in a minority of colorectal adenomas, although LOH
occurred at a much lower frequency in my sample (10% vs 5% at Ip, p=0.64; 28% vs
4% at 15q, p=0.004, Fisher’s exact tests) and was not found on chromosome 18q
(28% vs 0%, p<0.001, Fisher’s exact test). It is unlikely that this difference is due to
an increased amount of contaminating normal tissue, since 27 of our 49 tumours had
150 previously been shown to have LOH at the APC locus. The probable explanation is partly chance, but also that the two studies used different methods (with different specificity and sensitivity) and analysed tumours of different origin (from sporadic cases and FAP patients, respectively). My data are, however, inconsistent with the view of Shih et al. that the LOH results indicate that karyotypic instability is common in early colorectal adenomas. First, only three cases of aneuploidy/polyploidy were found in 20 tumours analysed by FCM and CGH analysis. Second, three polyps with detectable LOH were normal by CGH analysis.
Third, LOH diX APC was shown not to result from physical loss of material but probably from somatic recombination, inconsistent with Shih et aL’s view that CIN precedes APC inactivation.
My results do not support the hypothesis that APC mutations are selected for effects
on chromosomal mis-segregation manifesting as karyotypic instability in early stages
of colorectal tumorigenesis, although a minor tendency to CIN cannot be excluded. It
is evident, moreover, that at least some near-diploid colorectal carcinomas have two
mutational hits at APC. Thus, whilst APC may well have a role in interacting with, or
perhaps controlling the mitotic spindle, loss of this C-terminal function does not
inevitably lead to spindle dysfunction and genomic instability, even in late lesions in
which cell cycle checkpoints are likely to be deranged.
In summary, my data and earlier results (Lamlum et al., 1999; Rowan et al., 2000) -
together with the findings of Haigis et al. (Haigis et al., 2002) - show thatAPC
mutations are common in colorectal tumours because they provide cells with a direct
selective advantage. The nature of that advantage probably primarily involves
151 changes in p-catenin levels. The APC protein may physically associate with components of the mitotic spindle, but its role, if any, in chromosomal segregation is not yet characterised. The model cell systems previously used to study the association of APC with chromosomal mis-segregation are themselves prone to spontaneous changes in chromosome number and structure, even in the presence of wild-type APC (Liu et al., 1997b; Longo et al., 1997). It cannot yet be excluded that
APC mutations increase the tendency for chromosomal mis-segregation to occur in human tumours in vivo, but if mutant APC does have this effect, its consequences do not generally manifest until later-stage tumorigenesis.
152 Chapter 7 The Contribution of Germline MYH Mutations to
Muitiple Coiorectai Adenomas and Ciassicai FAP
7.1 Introduction
The importance of germline APC mutations - including large-scale deletions - in causing AAPC/multiple adenomas and classical polyposis is well-established (see
Chapter 3), but there is good evidence that other genes may be involved. Recently,
A1 Tassan et al. (Al-Tassan et al., 2002) reported a single Welsh family with three affected individuals and recessive inheritance of multiple colorectal adenomas and carcinoma. Patients’ tumours had an excess of somatic G:C-»T:A mutations in the adenomatous polyposis coli {APC) gene, typical of changes caused by oxidative
DNA damage (Halliwell, 1996; Michaels and Miller, 1992; Nghiem et al., 1988;
Shibutani et al., 1991; Thomas et al., 1997; Wang et al., 1998). This damage produces the stable guanine adduct 8-oxo-7,8-dihydroxy-2’-deoxyguanosine (8-oxo- dG), which tends to mispair with adenine, leading to the observed mutation (see
Chapter 1.5.1). A1 Tassan et al. (Al-Tassan et al., 2002) therefore tested oxidative
repair genes for germline changes in their family. They found that the affected
individuals carried two missense variants, Y165C and G382D, in the base excision
repair gene MYH.
Besides MYH, two further base excision repair genes, M T H l and OGGI, are
involved in preventing 8-oxo-dG induced mutagenesis (see Chapter 1.5.1). All three
base excision repair genes are therefore good candidates for causing multiple
153 colorectal adenomas and possibly also ylPC-mutation negative cases of classical polyposis.
In order to assess the contribution of BER defects, both a group of patients with multiple colorectal adenomas and a set of APC mutation-negative polyposis patients were analysed for germline mutations in MYH. Selected patients’ tumours were screened for somatic APC mutations and a sub-set of multiple adenoma patients was analysed for mutations in MTHl and OGGI. All mutation screens were performed in collaboration with Dr. Lara Lipton.
7.2 Study Population
7.2.1 Multiple Colorectal Adenoma Patients and Normal Controls
For this study, 152 patients with multiple (3 to 100) synchronous or metachronous colorectal adenomas were identified from Genetics Departments in the United
Kingdom (St Mark’s Hospital, Harrow; Oxford; and Guy’s Hospital, London).
Patients had been referred either owing to a family history of colorectal tumours or after they themselves had presented symptomatically with multiple polyps, suggesting a genetic disease. Clinicopathological data were ascertained from patients’ records to confirm diagnoses. In all cases, either a precise adenoma count had been reported, or, more rarely, a rounded or approximate count was given. To
the date of follow-up (3U* December 2001), the mean number of adenomas
developed was 16 (median=7, range=3-100). Family histories (of tumours or other
154 major disease) were recorded as reported by the patients, and were confirmed, where possible, from hospital records, although precise adenoma counts in patients’ relatives were rarely available. No patient reported consanguineous origins. In addition, 26 patients with multiple adenomas from Finland and Denmark were analysed, all of whom had been reported as having between 3 and 100 adenomas. A set of 107 samples derived from tumour-free UK individuals from a national study unrelated to cancer served as normal control.
7.2.2 Classical Adenomatous Polyposis Patients
Polyposis Registries in the United Kingdom, Switzerland, Finland, Portugal and
Denmark were contacted with a request to study all APC mutation-negative patients with more than 100 adenomas (synchronous or metachronous). A total of 107 probands were ascertained and it was confirmed that local laboratories had rigorously excluded germline APC changes. Full clinicopathological details and family history were obtained. For some patients, exact polyp counts at colectomy had been recorded; for others, counts were given in a range (for example, “100-1000” or
“several thousand” to distinguish mild and severe classical polyposis respectively); and for yet others, counts were provided essentially for diagnostic purposes (that is,
>100).
155 7.3 MYH Mutations in Multiple Adenoma Patients and Normal Controls
Coding regions and exon-intxon boundaries of MYH were screened in 152 multiple adenoma patients by fluorescence-SSCP analysis. Six multiple adenoma patients carried bi-allelic MYH mutations (Table 7.1, Figure 7.1). Of these, three were compound heterozygotes, as shown by sequencing of cloned PCR products; the remaining three were presumed homozygotes. The previously reported missense changes, Y165C (nt494 A>G) and G382D (ntll45 G>A), were the most common alterations. Both Y165C and G382D target highly conserved residues, the former mapping to the helix-hairpin-helix protein domain which confers specificity of mismatch recognition and the latter affecting the catalytic core of the glycosylase
(residues 366-497). In addition, novel frameshift changes, ntll03delC (codon 368) and ntl419delC (codon 473) which are predicted to abolish glycosylase function were identified. Six patients were heterozygous for an MYH mutation and the wild- type allele (Table 7.1). In these individuals, the entire MYH gene was sequenced, but no further changes were found.
156 Table 7.1: Multiple adenoma patients from the United Kingdom with germline
MYH mutations.
Note that two of the single mutation carriers had novel MYH variants, R83X (nt247
C>T) and R295C (nt883 C>T). It is not known whether or not these variants would be pathogenic as compound heterozygotes or homozygotes, although R83X is likely to be so. CRC = colorectal cancer; N = none reported; n/d = none detected.
Patient MYH MYH Gender Age at Polyp CRC Family history ID mutation 1 mutation 2 diagnosis number of adenomas and/or CRC 1 Y165C Y165C M 52 40 N Y 2 Y165C G382D F 45 100 Y N 3 Y165C nt1419delC F 57 18 Y Y 4 ntllOSdeIC G382D M 55 70 NY 5 G382D G382D M 56 40 Y Y 6 G382D G382D F 59 100 N Y 7 R83X n/d F 69 6 Y Y 8 Y165G n/d M 72 3 N Y 9 Y165C n/d F 58 5 N Y 10 R295C n/d F 25 3 N N 11 G382D n/d M 52 3 Y Y 12 G382D n/d F 70 12 NN
157 Figure 7.1: Selected germline MYH variants.
(A.) ntll03delC (reverse sequence); (B.) ntl419delC; (C.) 247 O T , R83X; and (D.) nt883 O T , R295C.
CCCCAAGGNNNNNNNNNNN GCTGTTTCNNNNNNNNNN
158 None of 107 United Kingdom controls carried two MYH mutations. Y165C was found in just two controls and none of the other disease-associated mutations was present in the control group. The previously described MYH polymorphisms in exon
2 (nt64 G>A; V22M), exon 12 (nt972 G>C; Q324H) and exon 16 (ntl502 C>T;
S501F) were detected with respective allele frequencies of 10%, 21% and 2% in the patients, similar to the control population previously reported (Al-Tassan et al.,
2002 ).
In order to provide further evidence of the pathogenicity of the MYH mutations, available relatives were screened for the changes which the proband (Table 7.1) carried. In all cases, the results were consistent with recessive inheritance. For patient 1, an affected sister was a mutant homozygote. For patient 2, an unaffected daughter was a heterozygote. For patient 4, her sister carried both mutations and had developed multiple adenomas and two colorectal cancers. For patient 6, only unaffected siblings survived and they were both heterozygotes.
All of the United Kingdom patients with bi-allelic MYH mutations were of northern
European ancestry. In order to determine the frequency of Y165C and G382D in
different populations, 26 Finnish and Danish individuals with multiple adenomas
were investigated. Two (8%) of the patients were compound heterozygotes for
Y165C and G382C. In the entire patient set, Y165C and G382C were not
consistently associated with specific alleles at the microsatellite marker D1S2667
(2.5 kb from MYH\ details not shown), thus providing no evidence for these being
ancestral rather than recurrent changes.
159 7.4 Somatic APC Mutations and Allelic Loss atMYH in Adenomas
from MYH Mutation Carriers
Twenty-five formalin-fixed, paraffin-embedded adenomas from three MYH compound heterozygotes (Y165C/ntl419delC; ntll03delC/G382D; Y165C/G382D) were screened for somatic APC mutations by fluorescence-SSCP analysis (exon
15G/H). Three mutations were detected (one per patient), all G:C-»T:A transversions (nt4230 C>A, C1410X; nt4381 G>T, E1461X; nt4480 G>T, E1494X).
In addition, all adenomas were screened for allelic loss at MYH using the microsatellite marker D1S2677. Loss of heterozygosity 3iMYH was found in 3 of 25 adenomas from the MYH mutant compound heterozygotes. Given that G382D reportedly retains some enzyme activity (Al-Tassan et al., 2002), the low frequency of allelic loss may indicate that absent MYH function is not necessary for tumorigenesis.
Seven adenomas from an individual heterozygous for the MYH missense variant
G382D were also screened for somatic APC mutations. Two changes were identified,
one C>G transversion (S1346X) and one Ibp deletion (nt4244delG). Neither change
was suggestive of defective MYH activity.
160 7.5 Clinical Features of Multiple Adenoma Patients with MYH
Mutations
The six patients with bi-allelic MYH mutations presented at a similar age
(median=56, range=45-59) to the other multiple adenoma patients in our study
(median=56, range=19-77) (Tables 7.1 and 7.2). Five of these individuals had
symptoms at presentation and one was found to have polyps at colonoscopy
performed owing to a family history of colorectal tumours. Polyps were
predominantly small, mildly dysplastic tubular adenomas, with a minority of
tubulovillous adenomas and very few hyperplastic polyps. Three of the six patients
had colorectal cancer at presentation. Five of the six had a family history of
colorectal cancer, in all cases involving more than one generation. Carriers of bi-
allelic MYH mutations were no more likely to have a family history of colorectal
cancer than other patients in the study (p=0.10, Fisher's exact test). Two MYH
patients had confirmed family histories of multiple adenomas, but these were found
only in siblings, consistent with recessive inheritance of MT//-associated disease. No
notable extra-colonic tumours or other clinical features were reported.
The one phenotypic difference which clearly distinguished patients with bi-allelic
MYH mutations from single mutation carriers and MYH mutation-negative
individuals in our multiple adenoma patient set was the number of tumours (medians
of 55 vs A vs 1 respectively; p=0.023, Kruskal-Wallis test) (Table 7.2). Of 21 patients
with 15 to 100 adenomas, six (29%) harboured bi-allelic MT/7 mutations.
161 The colorectal cancer frequency (3 out of 6) in bi-allelic MYH mutation carriers was greater than that in the other patients in our study (p=0.05; Fisher’s exact test) as well as the general population (3.86% for 0 to 80 years of age; http://seer.cancer.gov/).
These data suggest an increased risk of bowel cancer in bi-allelic MYH mutants, although this result should be interpreted with a little caution, because, whilst our patients were recruited on the basis of their adenoma phenotype alone, they were more likely to have come to clinical attention if they had a synchronous carcinoma.
Table 7.2: Clinical features of multiple adenoma patients from the United
Kingdom in relation to germline MYH mutation status.
Patients Germline MHY Negative (%) Single Bi-aiieiic mutation status mutation (%) mutation (%) Total Number 140 6 6 Age at Known 139 (99) 6(100) 6(100) Presentation Median 56 64 56 Range 18-77 25-72 45-59 Polyp number Precise count given 114(81) 6(100) 6(100) Median 7 4 55 Range 3-100 3-12 18-100 Coiorectai cancer Yes 19(14) 2(33) 3(50) None reported 121 (86) 4(67) 3(50) Family history of Positive 66 (47) 4(67) 5(83) coiorectai cancerNone reported 74 (53) 2(33) 1(17)
162 7.6 MTHl and OGGI Mutations in Muitiple Adenoma Patients
Coding regions and exon-intron boundaries of M THl and OGGI were screened by fluorescence-SSCP analysis. No carriers of obviously pathogenic or bi-allelic MTHl ox OGGI mutations were detected in 127 and 42 multiple adenoma patients respectively. For M T H l, the allele frequencies of previously described polymorphisms (Al-Tassan et al., 2002) were not significantly different from controls (details not shown). In addition, a novel MTHl missense variant (nt92 G>A;
R31Q) was identified in one patient, but not in the controls, but this mutation did not co-segregate with the multiple adenoma phenotype. For OGGI, besides the well- described polymorphism in exon 7 (nt977 C>G; S326C) (Hanaoka et al., 2001;
Kohno et al., 1998; Wikman et al., 2000), no further DNA alterations were detected.
7.7 MYH Mutations in Patients with Classical Polyposis
Eight of 107 (7.5%) probands with classical polyposis were found to carry bi-allelic, pathogenic M7/7 mutations using fluorescence-SSCP analysis (Table 7.3). Y165C and G382D were again the most common changes. Three other mutations were found: a frameshift (nt252delG, codon 84); and an unusual in-frame duplication
(nt411 dupATGGAT, codon 137 insIW); and a non-conservative missense change
(nt694 G>T, V232F). Four patients carried single MYH mutations (two Y165C,
I209V and G382D).
163 Table 7.3: Classical adenomatous polyposis patients with germline MYH mutations.
CRC, colorectal cancer; CHRPE, congenital hypertrophy of the retinal pigment epithelium; N, none reported; n/d, none detected.
Patient MYH MYH Gender Age at Polyp CRC Family Extracolonic ID mutation mutation diagnosis number history features 1 2 1 Y165C Y165C M 41 100-1000 N Y CHRPE II Y165C G382D F 50 100-1000 N Y N III G382D G382D M 30 100-1000 N N Duodenal adenomas IV nt252delG 137inslW M 38 100-1000 Y N Duodenal adenomas V Y165C Y165C F 45 100-1000 Y Y N VI Y165C V232F M 70 100-1000 N NN VII Y165C G382D M 51 210 Y NN VIII Y165C G382D F 69 115 Y YN IX G382D n/d M 32 100-1000 N Y N X 1209V n/d F 54 100-1000 N N Osteoma XI Y165C n/d M 30 100-1000 NN N XII Y165C n/d M 30 750 N YN
164 All polyposis probands with bi-allelic MYH mutations had family histories compatible with recessive inheritance, in that only the proband or siblings in a single generation were affected by polyposis. Although it is necessary to exercise some caution owing to variation in clinical practice and precision of polyp counting among centres, it is probable that patients with MYH mutations had mild classical adenomatous polyposis (Tables 7.3 and 7.4): none had more than 1000 polyps; two had exact counts of 115 and 210 adenomas; and none had early-onset cancer. All patients in the double MYH mutation group had been treated by total colectomy with ileorectal anastomosis or ileal pouch, at a mean age of 47.6 years (median=47, range=30-70), compared with 28 years (median=23, range=13-65) for St Mark’s classical polyposis patients with APC mutations (details not shown).
Table 7.4: Clinical features of classical adenomatous polyposis patients in relation to germline MYH mutation status.
Patients Germline M /// Negative (%) Single Bi-aileiic mutation status mutation (%) mutation (%) Total Number 95 4 8 Age at Known 55 (58) 4(100) 8(100) presentation Median 30 31 47.5 Range 7-72 30-54 30-70 Polyp number 100-1000 68 (72) 4(100) 8(100) >1000 27 (28) 0(0) 0(0) Family history of Positive 29 (31) 0(0) 4(50) coiorectai cancerNone reported 66 (69) 4(100) 4(50)
165 In several respects, the clinicopathological features of patients with bi-allelic MYH mutations were the same as those of patients with polyposis resulting from APC mutations: macroadenoma morphology was the same (largely small tubular lesions with mild dysplasia); oligocryptal adenomas were present, despite previously being thought to be pathognomonic of familial adenomatous polyposis; and some patients had extra-colonic disease. Patient III developed severe (Spigelman stage IV) duodenal polyposis and patient IV had duodenal polyps at diagnosis. Congenital hypertrophy of the retinal pigment epithelium was diagnosed in patient I (although it was not specifically noted to be of a polyposis-associated type). No desmoid tumours were reported.
7.8 Discussion
This study has shown that germline MYH mutations predispose to recessive inheritance of multiple colorectal adenomas and classical adenomatous polyposis in a variety of European populations. All patients with bi-allelic MYff mutations probably have a raised risk of colorectal cancer. Of patients with 3-100 adenomas, about 5% had disease ascribable to MYH, and of those with over 15 adenomas, nearly one third had bi-allelic MYH mutations. In patients with a phenotype of classical polyposis and
no APC mutation, about 8% of cases had two germline MYH mutations. Extra
colonic disease was present in some patients with MTff-associated disease, showing
that these features are not restricted to those with germline APC mutations. The
presence of extra-colonic disease is consistent with the model of mutant MYH action
in the colon, namely hypermutability oiAPC (and probably of other genes also).
166 Patients with bi-allelic MYH mutations thus tend to have mild disease relative to most patients with classical familial adenomatous polyposis, but more severe disease than most multiple adenoma patients. It is difficult to distinguish between patients with APC and bi-allelic MYH mutations on the basis of clinicopathological features, although family history can be useful. MYH mutations appear to be a more frequent cause of the multiple adenoma (or ‘attenuated polyposis’) phenotype than are APC mutations (Lamlum et al., 2000a), but are evidently a less frequent cause of classical adenomatous polyposis. Compared with hereditary non-polyposis colon cancer, carriers of bi-allelic MYH mutations develop more tumours, hut progression of adenoma to carcinoma appears to be slower.
Ten patients were identified who only had one mutated MYH allele. Given that allelic loss on chromosome arm Ip - the location of the MYH locus - is apparently an early event in colorectal tumorigenesis (Lothe et al., 1995; Praml et al., 1995; Tanaka et al., 1993), this may indicate that MYH mutant heterozygotes have a weak susceptibility to bowel tumours. However, in this patient set, there was no over representation in general of heterozygotes as compared to our and published controls
(Al-Tassan et al., 2002); and the two somatic APC mutations in polyps from an MYH heterozygote individuals were not G:C->T:A transversions. Nevertheless, several of
our patients had family histories suggesting dominant inheritance of colorectal
cancer (although not of multiple adenomas). Formal exclusion of MYH as a weak
susceptibility allele will require a large set of colorectal cancer cases and controls.
The data derived from this study do not yet answer the question of why MYH,
rather than MTHl or OGGI, is important for tumour predisposition. Specifically, the
167 possibility cannot be excluded that carriers of bi-allelic M THl or OGGI mutations are predisposed to tumours, although no such individuals were found.
The results suggest that genetic testing forMYH changes should be performed in patients with tens or hundreds of colorectal adenomas, subject to the proviso that almost all patients with MTTf-associated polyposis will have a family history consistent only with recessive inheritance of multiple adenomas (although the family may appear to have dominant inheritance of colorectal cancer). Screening of APC and MYH may be performed in parallel in some patients, such as isolated cases.
Evidently, if bi-allelic MYH mutations are identified in a proband, testing of siblings is worthwhile, even if they are asymptomatic in their 6* or 7* decade. It should, however, be borne in mind that in 2-3% of cases, a carrier of two MYH mutations will produce children with a spouse/partner who carries a single mutation; in this case the disease will appear to be dominantly inherited. The question remains open as to whether it is worthwhile to undertake genetic testing in the spouses/partners of
patients.
All but three of the patients with bi-allelic MYH mutations proceeded to colectomy, since their disease could not be controlled by colonoscopic polypectomy. For patients with relatively mild disease, regular screening by colonoscopy may be used initially - although optimal screening intervals must be determined empirically - and may prevent colectomy, as it has in the three patients with fewer than 50 adenomas.
Unfortunately, the recessive nature of the disease means that it will prove difficult to
identify MYH mutation carriers at an early enough age to prevent colectomy in all
cases. In the absence of any evidence for an increased risk of colorectal tumours in
168 MYH heterozygous mutation carriers, there is currently little justification for aggressive colonoscopic screening of any heterozygote family member. The occurrence of duodenal polyposis in some bi-allelic MYH mutation carriers indicates, that all such individuals should have regular endoscopy of the upper-gastrointestinal tract.
In conclusion, molecular methods should be used to classify disease in patients with multiple adenomas or adenomatous polyposis. Patients with identified germline mutation(s) should be classified as having APC- or MITf-associated disease. Risks for relatives and the likely severity of disease can then be accurately assessed.
Patients with polyposis but no identified germline mutation may then be further classified as presumed familial adenomatous polyposis if they have a dominant family history of classical disease and/or severe polyposis (>1000 colorectal adenomas). Other patients with no detected germline mutation in APC or MYH, and with either mild polyposis (100-1000 adenomas) or multiple (<100) adenomas should be classed as having ‘polyposis/multiple adenomas of unknown origin’. The genetic basis underlying disease in these remaining patients remains to be determined.
169 Chapter 8 The Contribution of Wnt and TGF-p/BMP Pathway
Genes to Multipie Coiorectai Adenomas
8.1 introduction
Bi-allelic germline mutations of the base excision repair gene MYH and specific mono-allelic germline mutations of the Wnt pathway gene APC have been shown to predispose to multiple colorectal adenomas, but together are unlikely to account for more than half of all cases (see Chapters 3 and 7). Whilst other key members of the base excision repair pathway, MTHl and OGGI, have been excluded as playing a major role in causing multiple colorectal adenomas (see Chapter 7), further members of the Wnt signalling pathway remain good candidates.
Two good candidates are the kinase GSK-Sp and the scaffold-protein axin which form a complex with APC essential for regulation of the Wnt pathway (see Chapter
1.4.1.1.2 for details). In the absence of Wnt signalling, this complex binds and phosphorylates p-catenin, thereby targeting p-catenin for proteasomal degradation.
Wnt signalling activates a cascade, which inhibits GSKSp, allowing p-catenin to
escape degradation and to activate transcription of genes involved in cellular
proliferation, invasion and metastasis. Notably, vertebrates carry two homologous
axin genes, axinl and axin 2laxil/conductin, both of which are negative regulators of
Wnt signalling (Behrens et al., 1998; Hart et al., 1998b; Ikeda et al., 1998; Kishida et
al., 1998; Sakanaka et al., 1998; Yamamoto et al., 1998). However, only axin2 is
170 induced by active Wnt signalling and probably acts in a negative feedback loop (Jho et al., 2002; Lustig et al., 2002; Yan et al., 2001).
Another good candidate is the recently identified homologue of APC, APC2/APCL
(adenomatous polyposis coli 2/adenomatous polyposis coli like) (Nakagawa et al.,
1998; van Es et al., 1999). The human APC2 gene localises to chromosome band
19pl3.3 and encodes a 2302 amino acid protein. The protein seems to be ubiquitously expressed in normal tissues, with a particularly high level in the central nervous system (Nakagawa et al., 1998; van Es et al., 1999). The N-terminus and central region of APC2 displays high homology to APC including the dimérisation domain, armadillo repeats, SAMP repeats and 20aa repeats. Accordingly, APC2 has been shown to regulate cellular p-catenin levels (Nakagawa et al., 1998; van Es et al., 1999). Differential functions of APC2 and APC may, however, be mediated by their distinct C-termini. This suggestion is supported by the observation that the C- terminus of APC2, but not APC, interacts with p53-binding protein 2 (Nakagawa et al., 2000). Given the high homology between APC2 and APC, it seems possible that
APC2 participates in Wnt signalling and is involved in tumorigenesis. A recent report of allelic loss in a region near APC2 in ovarian carcinomas is consistent with this hypothesis (Jarrett et al., 2001).
In addition, members of the TGF-p/BMP signalling pathway (see Chapter 1.4.1.3.1 for details) may be involved in causing multiple colorectal adenomas. Germline mutations in the TGF-p/BMP pathway components SMAD4 and BMPRIA cause juvenile polyposis syndrome (JPS) (see Chapter 1.4.1.3). JPS is typified by characteristic colorectal polyps with inflammatory infiltrate, mucin-filled cysts and
171 flattened epithelium. These polyps are usually multiple and can occur throughout the intestine, conferring a greatly increased risk of gastrointestinal cancer. However, it has been reported that JPS is not solely associated with juvenile polyps, but also with colorectal adenomas (Al-Jaberi and El-Shanti, 2002). To date, all carriers of SMAD4 and BMPRIA mutations with colorectal adenomas have also had a personal or family history of juvenile polyps, but the possibility remains that some individuals presenting with multiple adenomas harbour germline SMAD4 or BMPRIA mutations.
This survey was aimed at assessing the contribution of Wnt and TGF-p/BMP pathway genes to the multiple adenoma phenotype by screening GSK3-P, cain2,
APC2, SMAD4 and BMPRIA for germline mutations in a cohort of 47 multiple adenoma patients. Alterations of the two known predisposition genes APC and MYH had previously been excluded in these patients. All analyses were performed in
collaboration with Dr. Lara Lipton and Dr. Kelly Woodford-Richens.
8.2 Study Population
A total of 47 patients was recruited through the Cancer Research UK Family Cancer
Clinic at St Mark’s Hospital (London UK) and the Guy’s Hospital Clinical Genetics
Unit (London UK). Inclusion criteria were between 3 and 100 synchronous or
metachronous colorectal polyps, with or without colorectal cancer, and histological
confirmation of adenomatous features in the polyps. Additional phenotypic details
(gender, age at presentation, polyp number and histology) were available for all
patients and are summarised in Table 8.1. Germline mutations had previously been
172 excluded in ^PC (AAPC-associated regions) and MYH using single strand conformational polymorphism (SSCP) analysis.
Table 8.1: Characteristics of 47 patients with multiple colorectal adenomas.
Note: Characteristics of 48 additional patients screened for BMPRIA mutations were not significantly different.
Characteristics Muitipie Adenoma Patients
(N=47)
Age at presentation - Median / Range 50 /18-72
Adenoma count - Median / Range 12/3-100
Hyperplasic polyps - No. of patients (%) 12 (25%)
Colorectal cancer - No. of patients (%) 9(19%)
Family history - No. of patients (%) 39 (83%)
8.3 Germline Mutations in the Wnt and TGF-p/BMP Signalling
Pathways
A total of 47 multiple adenoma patients without germline APC and MYH mutations
was screened for alterations in selected members of the Wnt (APC2, axin2, GSK3-P)
and TGF-p/BMP {BMPRIA, SMAD4) signalling pathways, using a combination of
fluorescence-SSCP analysis and the protein truncation test (PTT). For fluorescence-
SSCP analysis, overlapping primer pairs amplifying coding regions and exon-intron
boundaries were designed for ^PC2 (exons 1-13), axin2 and GSK3~p. As a complete
173 genomic axin2 sequence was not available on a public database, it was derived by performing a BLAST search of cDNA against the HTGS sequence database. The last exon of APC2 (exon 14) was screened up to codon 1388 using the protein truncation test (PTT). This region included the 20aa repeats required for p-catenin binding and degradation. SSCP analysis of the BMPRIA and SMAD4 genes was performed using previously published primers and conditions (Houlston et al., 1998; Zhou et al.,
2001b). Samples with aberrant patterns were re-amplified and sequenced.
No clearly pathogenic changes were detected in any of the three Wnt signalling components. In the 5’ and central region of APC2, which comprises the majority of the known P-catenin binding/regulation domains, two silent (nt687 G-»A and ntl317
C->T) and two novel missense mutations (nt419 T->C, F140S and nt967 G->C,
G322R) were identified (Table 8.2). The F140S variant affected an amino acid that is not evolutionary conserved and is therefore unlikely to be of functional significance.
The G322R variant could formally be excluded as disease-causing, being absent in other affected family members. In axin2, only two polymorphisms were detected.
One occurred in intron 6 (IVS6+19 G->T) and was present in 8 of 47 (17%) of patients. The other occurred in exon 6 (nt635 C-*T; silent) and was present in 3 of
47 (3%) of patients. No DNA sequence changes were detected in GSKS-p.
Screening of the TGF-p/BMP pathway members, BMPRIA and SMAD4 identified a
single putative pathogenic change in BMPRIA, a three base-pair deletion at codon
360 resulting in the loss of a histidine residue. This amino acid is highly conserved
and lies within the kinase domain of the protein, which is essential for the activation
of the downstream targets SMADl and SMAD5 (Massague, 2000). In accordance
174 with a pathogenic effect, LOH at three microsatellite markers close to the BMPRIA locus (D10S573, ALK3CA, ALK3GGAA) was detected in an adenocarcinoma from this patient (Figure 8.1). Besides this alteration, two intronic polymorphisms (IVS6
-26 T-*A and IVSl 1 -11 T ^C ), two silent point mutations (nt435 G-^A and nt777
G->A), and one previously reported missense polymorphism (nt4 C->A; T2P)
(Howe et al., 2001) were detected in BMPRIA. No pathogenic sequence changes or
polymorphisms were detected in the SMAD4 gene. To further clarify the involvement
of BMPRIA mutations in causing multiple adenomas, a second cohort of 48 APC and
MYH mutation-negative patients was screened for BMPRIA changes, but no
additional pathogenic variants were identified.
175 Table 8.2: Germline mutations of the BMPRIA, SMAD4, APC2, oxin2 and
GSK3P genes detected in 47 multiple adenoma patients.
Mutation positions are based on Genbank accession numbers gi|4757853, gi|2865656, gi|4514536, gi|20558133, and gi|21361339 for BMPRIA, SMAD4,
APC2, axin2 and GSK3P, respectively. The asterisk (*) indicates patients screened in the second cohort of 48 multiple adenoma patients.
Patient Gene Germline mutation D1 APC2 nt419T->C; F130S E6 APC2 nt 687 G-»A; silent A1 APC2 nt 967, G->C; G323R D3, H6 APC2 nt 1317 C->T; silent A1.A4, A5, B4. C2. F4, H1, Axin2 nt 635 C-^T; silent H2 D2, D6, E3 Axin2 IVS6+19 G->T A1,F2,A7*. E6 BMPR1A nt 4 C-^A; T2P H1,H3, F4 BMPR1A IVS6 -26 T->A H2 BMPR1A nt 435 G^A; silent A8* BMPR1A nt 777 G->A; silent D1 BMPR1A nt1079 del3bp; loss of histidine residue at codon 360 H2 BMPR1A IVS11 -11 T-*-C; homozygote E3 BMPR1A IVS11 -11 T-*C; heterozygote
176 Figure 8.1: Loss of heterozygosity at BMPRIA in an adenocarcinoma from a patient with a germline BMPRIA mutation (codon 360 del his).
Shown are the results for microsatellites (a) ALK3CA and (b) ALK3GGAA located
43Mb and 73Mb proximal to the BMPRIA gene, respectively. The allele which is lost in the carcinoma is marked by the arrow.
Adenocofclnoma
B Normal
Adenocarcinoma
177 8.4 Discussion
This analysis of selected members of the Wnt and TGF-p/BMP pathways in 47 patients provided little evidence for the involvement of germline mutations in these two signalling cascades, other than in APC, in the pathogenesis of multiple colorectal adenomas. This finding is consistent with a number of previous studies which (i) excluded germline p-catenin mutations in multiple adenoma patients (Albuquerque et al., 2002b), (ii) failed to detect somatic APC2 mutations in brain tumours
(Nakagawa et al., 1999), and (iii) failed to detect somatic GSK-p mutations in colorectal adenomas and carcinomas (Sparks et al., 1998). In contrast, somatic mutations in axin2 have been found at mononucleotide repeats in colorectal cancers, but all cases also showed microsatellite-instability (Liu et al., 2000).
The histology of polyps in the investigated APC and MYH mutation-negative multiple adenoma patients did not differ in any obvious way from that of sporadic or
FAP adenomas and thus gives no clue as to the genetic aetiology (Table 8.1). The one individual who was found to have an apparently pathogenic mutation in
BMPRIA had no juvenile polyps reported, indicating that this gene may contribute,
albeit rarely, to a subset of patients who present only with multiple colorectal
adenomas. The absence of mutations in genes involved in the Wnt and TGF-p/BMP
signalling suggests that other as yet unidentified predisposition genes and pathways
exist. Possible candidates include the HMPS/CRACl locus on chromosome 15ql3-
ql4, which has been associated with the inheritance of colorectal adenomas and
carcinoma in Ashkenazi families (Jaeger et al, 2003; Tomlinson et al, 1999).
178 In conclusion, the multiple adenoma phenotype cannot generally be attributed to germline mutations in genes involved in the Wnt and TGF-p/BMP signalling pathways. Patients and families who present in this manner could be offered mutation analysis of the APC (AAPC-associated mutations), MYH, and perhaps
BMPRIA genes, although the number of pathogenic mutations identified in the last of these will be relatively small. Further work is needed to unravel the genetic basis of multiple colorectal adenomas, but given the homogeneity of the phenotype, this is likely to rely on a combination of genetic linkage analysis and examination of candidate genes in alternative pathways leading to adenoma formation.
179 Chapter 9 Summary and General Discussion
Multiple adenoma patients - as defined herein - typically develop 3-100 adenomatous polyps in the large bowel, resulting in an increased risk of CRC. The disease can be inherited as a Mendelian trait, either autosomal dominant or recessive, but can also occur in the form of isolated cases. Extracolonic features such as desmoid tumours, and gastric or duodenal adenomas may occur in some cases. Due to this clinical heterogeneity, the true incidence of the disease is unknown.
Approximately 10% of multiple adenoma patients carry a truncating germline mutation in specific regions of the APC gene, usually in exons 1-4, exon 9 or the 3’- half of exon 15, producing a diagnosis of AAPC (Spirio et al., 1993; van der Luijt et al., 1996). Two APC missense variants, I1307K and E1317Q, have also been associated with multiple colorectal adenomas (Laken et al., 1997; Lamlum et al.,
2000a), and whole-gene APC deletions have been identified in some patients
(Hodgson et al., 1994; Pilarski et al., 1999; Su et al., 2002). However, these latter associations have been inconsistent. In contrast, germline mutations in the central region of APC cause classical FAP, a dominant predisposition to hundreds or thousands of colorectal adenomas resulting in a near 100% lifetime risk of CRC. The molecular mechanisms underlying these genotype-phenotype correlations are to date poorly understood.
Whilst germline APC mutations clearly do play a role in causing AAPC/multiple adenomas, it is evident that a large proportion of multiple adenoma patients do not carry pathogenic APC mutations (Albuquerque et al., 2002b; Lamlum et al., 2000a).
180 This thesis focused on characterising the genetic basis of the multiple colorectal adenoma phenotype by clarifying the contribution of germline APC mutations (in particular, the role of whole-gene APC deletions), by investigating the molecular mechanisms underlying the association of specific germline APC changes and
AAPC, and by evaluating candidate predisposition genes.
The identification of germline APC mutations in patients with colorectal adenomas generally relies on routine mutation detection techniques such as SSCP, DGGE, PTT or DNA sequencing. These techniques are suitable to detect most small-scale DNA changes, but may miss large-scale alterations such as partial or whole-gene APC deletions. Consistent with a potential role of such changes, large-scale APC deletions have been identified in a limited number of patients with either AAPC/multiple adenomas or classical polyposis (see Table 3.1 in Chapter 3). My work has clarified this inconsistent association by showing that large-scale APC deletions rarely contribute to the multiple colorectal adenoma phenotype (see Chapter 3). Instead, such changes were found in approximately 12% of APC mutation-negative patients with classical polyposis. Given this appreciable frequency, screening of mutation-
negative classical polyposis patients for germline APC deletions seems warranted.
Although some FAP patients with cytogenetically visible interstitial deletions of
chromosome 5q develop mental retardation, all of my APC deletion patients were of
normal intelligence. Fine-mapping of the chromosomal breakpoints in these former
patients could therefore aid in the identification of the gene(s) responsible for this
disability.
181 My data suggest that the contribution of exonic germline APC mutations to
AAPC/multiple adenomas does not greatly exceed the previous estimate of 10%
(Lamlum et al., 2000a). However, it remains possible that germline APC mutations exist within regulatory regions and/or introns, although I identified no changes within the promoter region of the gene (see REF (Lamlum et al., 2000a)).
Although the spectrum of germline APC mutations causing AAPC/multiple adenomas is well-characterised, little is known about the molecular mechanisms which underlie the association of specific APC changes and attenuated disease.
However, a previous study on adenomas from FAP patients has revealed intriguing associations between germline and somatic APC mutations which appear to explain at least some of the phenotypic variability in FAP (Lamlum et al., 1999). The data suggest that somaticAPC mutations are selected for their ability to produce an optimal level of Wnt signalling in the tumour cell. This can be achieved by allelic loss in FAP patients with germline mutations near codon 1300, but other patients must acquire truncating somatic mutations to achieve the same effect. Notably, patients with germline mutations near codon 1300 - who probably have the highest effective mutation rate dX APC - tend to develop particularly severe colorectal polyposis. In addition, a recent study has indicated that even in patients with
mutations away from codon 1300 the ‘second hit’ is non-random, leading to the
suggestion that the level of Wnt signalling has to be ‘just right’ for tumorigenesis
(Albuquerque et al., 2002a). Assuming that allelic loss atAPC occurred by deletion,
most FAP tumours were found to have a pair of mutant alleles which between them
encoded proteins with a total of just a single 20aa p-catenin binding and degradation
repeat. I have further refined the association of germline and somatic APC mutations
182 in FAP to provide clues for understanding the molecular genetic mechanisms causing attenuated polyposis (see Chapter 4). I found that in FAP polyps allelic loss at APC occurred by mitotic recombination leaving two identical alleles, and that the ‘first hit-second hit’ associations at APC follow a ‘loose’ rather than a ‘just right fit’: truncated APC proteins in the tumour were found to generally retain a total of one or two 20aa P-catenin binding and degradation repeats, but several exceptions to this rule were found.
The suggestion from FAP adenomas that the spectrum of somatic APC mutations is constrained by selection for an optimal level of Wnt signalling and therefore influences disease severity is consistent with my (and previous) data on the relationship between germline and somatic APC mutations in AAPC patients with a
R332X mutation in the alternatively spliced region of exon 9 (see Chapter 5). In contrast to FAP adenomas, a significant proportion of macroscopic polyps from these patients harboured two somatic APC mutations, one in the wild-type and one in the inherited mutant allele. Furthermore, the inherited mutant allele appeared to be specifically targeted by mutations leaving three 20aa p-catenin binding/degradation repeats in the isoprotein, with the most common change being an adenine insertion at a hypermutable Ag-tract at codons 1554-6 (nt4661-6). It seems likely that the (at least partial) requirement of two somatic APC mutations explains the lower polyp number in exon 9-mutant patients as opposed to classical FAP patients. However, hypermutation at codons 1554-6 would have to be extreme to fully account for the
significantly increased polyp number as compared to sporadic cases. Instead, tumorigenesis in exon 9-mutant patients may follow a stepwise model, in which
mutation of the wild-type APC allele initiates limited tumour growth, which results
183 in an expanded clone of cells with ‘two hits’ at APC, but a sub-optimal genotype.
The increase in size of the tumour clone, however, increases the likelihood of a ‘third hit’ to produce the optimal genotype. Thus AAPC patients are expected to develop
fewer tumours than FAP patients, but more than the general population.
Whilst this model may also apply to AAPC patients with germline mutations in the
alternatively spliced exons 1-4, it cannot explain the attenuated phenotype in AAPC
patients with 3’-end mutations. To date, no comprehensive data on the somatic APC
mutation spectrum are available for these patients, and resolution of this issue will
require future studies.
Besides its well-established role in Wnt-signalling, APC has been implicated in a
variety of cellular processes including cell migration, cell-cell adhesion and
organisation of the microtubule cytoskeleton. Furthermore, it has recently been
proposed that APC acts in chromosomal segregation and that APC inactivation leads
to chromosomal instability (CIN). This potential function could be important for
tumour development, as CIN is predicted to accelerate the multi-step process of
colorectal tumorigenesis by targeting downstream tumour suppressor genes. In
contrast, an alternative hypothesis based on allelic loss studies in small colorectal
adenomas proposes that CIN precedes and causes APC mutations. My data from
flow cytometry, comparative genomic hybridisation and LOH analysis on colorectal
adenomas with characterised bi-allelic APC mutations permit testing of these models
(Chapter 6). The apparently limited extent of CIN in these lesions together with the
finding that allelic loss at APC generally occurs by mitotic recombination does not
support the hypothesis that CIN precedes APC mutations in tumorigenesis.
184 Regarding the model in which APC mutations lead directly to CIN, if APC mutations
do have this effect in vivo, it must be subtle or be suppressed by intact cellular DNA
damage checkpoints. Alternatively, CIN associated with APC mutations might be
essentially an in vitro phenomenon.
Whilst germline APC mutations clearly do play a role in causing AAPC/multiple
adenomas, it is evident that a large proportion of multiple adenoma patients do not
carry pathogenic APC mutations (Albuquerque et al., 2002b; Lamlum et al., 2000a).
Recently, recessive missense mutations in the base excision repair gene MYH have
been identified in a kindred with multiple colorectal adenomas and carcinoma (Al-
Tassan et al., 2002). Together with two related base excision repair genes, MTHl and
O GGI, MYH is involved in the repair of 8-oxo-7,8-dihydro2’deoxyguanosine (8-
oxo-G) lesions induced by oxidative DNA damage. Consistent with a defect in base
excision repair, tumours from patients with bi-allelic MYH mutations displayed an
excess of somatic G:C->T:A mutations in the APC gene. I have shown that bi-allelic
MYH mutations account for up to 30% of UK patients with 15-100 adenomas, and
approximately 8% of APC mutation-negative patients with classical polyposis
(Chapter 7). MYH mutations thus appear to be a more frequent cause of the multiple
adenoma phenotype than are APC mutations (Lamlum et al., 2000a), but are
evidently a less frequent cause of classical polyposis. Extracolonic disease was
present in some patients with MTH-associated polyposis (MAP), showing that these
manifestations are not restricted to patients with germline APC mutations. Two MYH
missense variants, Y165C and G382D, were particularly common in my sample.
Recent data on lymphoblastoid cell lines confirm that these variants are indeed
defective in DNA binding and repair (J. Eshelman, personal correspondence), but it
185 remains to be determined whether or not these mutants have any additional effects.
There is currently little evidence for a role of single MYH mutations, but the
possibility of a mild predisposition to colorectal adenomas should be addressed in a
future large case-control study. Although I did not detect pathogenic MTHl and
OGGI mutations in our multiple adenoma patients, a role of these genes can
currently not be excluded.
Bi-allelic MYH and specific mono-allelic APC germline mutations have thus been
shown to predispose to multiple colorectal adenomas, but together are unlikely to
account for more than half of all cases. Obvious candidates for the remaining cases
are members of the Wnt and TGF-p/BMP pathways, as both of these signalling
cascades are central to colorectal tumorigenesis. However, I detected no pathogenic
germline mutations in the Wnt pathway genes APC2, axin2 and GSK3P or the TGF-
p/BMP pathway gene SMAD4 (Chapter 8). Only one of 95 (1%) patients carried a
putatively pathogenic mutation in the TGF-p/BMP pathway gene BMPRIA. In
addition, a previous study had failed to identify pathogenic germline changes in p-
catenin (Albuquerque et al., 2002b). It therefore seems unlikely that the disease in
the remaining multiple adenoma patients is generally attributable to germline
changes in the Wnt (besides APC) and TGF-p/BMP signalling pathways. The
identification of further predisposition gene(s) will probably rely on a combination of
genetic linkage analysis and examination of candidate genes in alternative DNA
repair or signalling pathways.
In conclusion, I have established that the genetic basis of the multiple colorectal
adenoma phenotype is heterogeneous, with two major predisposition genes being
186 APC and MYH. Germline mutations in the Wnt (besides APC) and TGF-p/BMP signalling pathways, on the other hand, are unlikely to play a major role. Germline
APC mutations appear to primarily drive colorectal tumorigenesis by deregulating
Wnt signalling rather than disrupting chromosomal segregation. In contrast, bi-allelic
MYH germline mutations appear to drive tumorigenesis by generating a mutator phenotype, although it remains puzzling as to why a FAP-like disease results. The spectrum of somatic APC mutations appears to be constrained by selection for an optimal level of Wnt signalling in both AAPC and classical FAP and this probably explains the difference in disease severity. Additional multiple adenoma loci are likely to exist, and their identification will probably rely on linkage analysis and further candidate screens.
187 Publications
1. Lipton L., Halford S.E., Johnson V., Novelli M.R., Jones A., Cummings G.,
Barclay, E, Sieber O.M., Sadat, A., Bisgaard M-L., Hodgson S.V., Aaltonen L.A.,
Thomas H.J.W., Tomlinson LP.M. Carcinogenesis in MYH-associated polyposis follows a distinct genetic pathway. Cancer Res. 2003, 63(22):7595-9.
2. Sieber O.M., Heinimann K., Tomlinson LP.M. Genomic instability - the engine of tumorigenesis? Nature Reviews Cancer. 2003; 3:7-14.
3. Crabtree M., Sieber O.M., Lipton L., Hodgson S.V., Lamlum H., Thomas H.J.W.,
Neale K., Phillips R.K.S., Heinimann K., Tomlinson LP.M. Refining the relationship between 'first hits' and 'second hits' at the APC locus: the 'loose fit' model and evidence for differences in somatic mutation spectra among patients. Oncogene.
2003; 22:4257-65.
4. Lipton, L, Fleischmann, C., Sieber, O.M., Thomas, H.J.W., Hodgson, S.V.,
Tomlinson, LP.M., Houlston, R.S. Contribution of the CHEK2 llOOdelC variant to risk of multiple colorectal adenoma and carcinoma. Cancer Lett. 2003; 200:149-52.
5. Halford S.E., Rowan A.J., Lipton L., Sieber O.M., Pack K., Thomas H.J.W.,
Hodgson S.V., Bodmer W.F., Tomlinson LP.M. Germline mutations but not somatic changes at the MYH locus contribute to the pathogenesis of unselected colorectal
cancers. Am J Pathol. 2003; 162(5): 1545-8.
188 6. Sieber O.M., Lipton L., Crabtree M., Heinimann K., Fidalgo P., Phillips R.K.S.,
Bisgaard M-L., Omtoft T.F., Aaltonen L.A., Hodgson S.V., Thomas H.J.W.,
Tomlinson LP.M. Multiple colorectal adenomas, classic adenomatous polyposis, and germline mutations 'mMYH. New Eng J Med. 2003; 348:791-9.
7. Lipton L., Sieber O.M., Thomas H.J.W., Hodgson S.V., Tomlinson LP.M.,
Woodford-Richens K. Germline mutations in the TGF-beta and Wnt signalling pathways are a rare cause of the "multiple" adenoma phenotype. J Med Genet. 2003;
40(4):e35.
8. Sieber O.M., Lipton L., Heinimann K., Tomlinson LP.M. Colorectal tumorigenesis in carriers of the APC I1307K variant: lone gunman or conspiracy? J
Pathol. 2003; 199(2):137-9.
9. Sieber O.M., Heinimann K., Gorman P., Lamlum H., Crabtree M., Simpson C.A.,
Davies D., Neale K., Hodgson S.V., Roylance R.R., Phillips R.K., Bodmer W.F.,
Tomlinson LP.M. Analysis of chromosomal instability in human colorectal adenomas with two mutational hits at APC. Proc Natl Acad Set USA. 2002;
99(26): 16910-5.
10. Sieber O.M., Lamlum H., Crabtree M.D., Rowan A.J., Barclay B., Lipton L.,
Hodgson S., Thomas H.J., Neale K., Phillips R.K., Farrington S.M., Dunlop M.G.,
Mueller H.J., Bisgaard M-L., Bulow S., Fidalgo P., Albuquerque C., Scarano M.I.,
Bodmer W., Tomlinson LP.M., Heinimann K. Whole-gene APC deletions cause
189 classical familial adenomatous polyposis, but not attenuated polyposis or "multiple"
colorectal adenomas. Proc Natl Acad Sci USA. 2002; 99(5):2954-8.
11. Sieber O.M., Tomlinson I.P.M., Lamlum H. The adenomatous polyposis coli
(APC) tumour suppressor - genetics, function and disease. Mol Med Today. 2000;
6(12):462-9.
12. Lamlum H., A1 Tassan N., Jaeger E., Frayling I., Sieber O.M., Reza F.B., Eckert
M., Rowan A., Barclay E., Atkin W., Williams C., Gilbert J., Cheadle J., Bell J.,
Houlston R., Bodmer W., Sampson J., Tomlinson LP.M. Germline AAC variants in
patients with multiple colorectal adenomas, with evidence for the particular
importance of E1317Q. Hum Mol Genet. 2000; 9(15):2215-21.
190 Appendix
Table A.1: Primers and PCR conditions for amplification of APC exons 4-14 from fresh-frozen tissue.
APC Primer Amplicon Finai MgCb Annealing Primer sequence (S' to 3') region name size, bp cone., mM temp., *C
Exon 4 4F ATAGGTCATTGCTTCTTGCTGAT -185 1 57
4R TGAATTTTAATGGATTACCTAGGT
Exon 5 5F CTTTTTTTGCTTTTACTGATTAACG -248 2 57
5R TGTAATTCATTTTATTCCTAATAGCTC
Exon 6 6F GGTAGCCATAGTATGATTATTTCT -200 1.5 55
6R CTACCTATTTTTATACCCACAAAC
Exon 7 7F AAGAAAGCCTACACCATTTTTGC -250 2 57
7R GATCATTCTTAGAACCATCTTGC
Exon 8 8F ACCTATAGTCTAAATTATACCATC -190 2 57
8R GTCATGGCATTAGTGACCAG
Exon 9/9a 9F AGTCGTAATTTTGTTTCTAAACTC -495 1.5 55 9R CTTTGAAACATGCACTACGA
Exon 10 IGF AAACATCATTGCTCTTCAAATAAC -220 2 57
ICR TACCATGATTTAAAAATCCACCAG
Exon 10A 10AF AATCATAACAGGAAGGGGAT -290 1.5 55
10AR CAGTCACACCCCCAGAGG
Exon 11 11F GATGATTGTCTTTTTCCTCTTGC -220 1.5 55
11R CTGAGCTATCTTAAGAAATACATG
Exon 12 12F TTTTAAATGATCCTCTATTCTGTAT -180 2 60
12R ACAGAGTCAGACCCTGCCTCAAAG
Exon 13 13F TTTCTATTCTTACTGCTAGCATT -300 1.5 55
13R ATACACAGGTAAGAAATTAGGA
Exon 14 14F TAGATGACCCATATTCTGTTTC -325 1.5 55
14R CAATTAGGTCTTTTTGAGAGTA
191 Table A.2: Primers and PCR conditions for amplification of APC exon 15 regions A-I from fresh-frozen tissue.
* Denotes addition of Q-solution (Qiagen).
APC Primer Amplicon Finai MgCb Annealing Primer sequence (S' to 3') region name size, bp conc., mM temp., *C
Exon 15A 15AF GTTACTGCATACACATTGTGAC 417 1.5 55
15AR GCTTTTTGTTTCCTAACATGAAG
Exon 15B 15BF AGTACAAGGATGCCAATATTATG 346 1.5 55
15BR ACTTCTATCTTTTTCAGAACGAG
Exon 15C 15CF ATTTGAATACTACAGTGTTACCC 399 2 59
15CR CTTGTATTCTAATTTGGCATAAGG
Exon 15D 15DF CTGCCCATACACATTCAAACAC 382 1.5 55
15DR TGTTTGGGTCTTGCCCATCTT
Exon 15E 15EF AGTCTTAAATATTCAGATGAGCAG 420 1.5 55
15ER GTTTCTCTTCATTATATTTTATGCTA
Exon 15F 15FF AAGCCTACCAATTATAGTGAACG 428 1.5 55
15FR AGCTGATGACAAAGATGATAATG
Exon 15G 15GF AAGAAACAATACAGACTTATTGTG 382 1.5 55
15GR ATGAGTGGGGTCTCCTGAAC
Exon 15H 15HF ATCTCCCTCCAAAAGTGGTGC 420 1.5 55
15HR TCCATCTGGAGTACTTTCTGTG Exon 151 15IF AGTAAATGCTGCAGTTCAGAGG 508 r 60 15IR CCGTGGCATATCATCCCCC
192 Table A 3: Primers and PCR conditions for amplification of codons 1258-1603 of the APC gene from formalin-fixed, paraffin-embedded tissue.
Codons 1258-1603 encompass the somatic mutation cluster region of APC.
* Denotes addition of Q-solution (Qiagen).
APC Primer Amplicon Final MgCh Annealing Primer sequence (S' to 3') region name size, bp conc., mM temp., ®C
Exon 15 F MCR IF CAAGCAGTGAGAATAGGTCCA 134 1 55 MCR 1R TTTCTTGGTTAATAGAAGAAACTTTGC Exon 15 F-G MCR2F GCCACTTGCAAAGTTTCTTCT 153 2* 55 MCR2R TGCTTCCTGTGTCGTCTGA Exon 15 G MCR3F TTCATTATCATCTTTGTCATCAGC 177 1 55* MCR3R GGATTTGGTTCTAGGGTGCT Exon 15 G MCR4F GATCCTGTGAGCGAAGTTCC 149 1 58 MCR4R CTGAGCACCACTTTTGGAGG Exon 15 G-H MCR5F AGAATCAGCCAGGCACAAAG 173 1 55 MCR5R GCAATCGAACGACTCTCAAA Exon 15 H MCR6F CACTATGTTCAGGAGACCCCA 149 1 58 MCR6R TGGAAGATCACTGGGGCTTA Exon 15 H MCR7F GTGAACCATGCAGTGGAATG 135 1 60 MCR7R ACTTCTCGCTTGGTTTGAGC Exon 15 H MCR8F AAACACCTCCACCACCTCCT 145 1 58 MCR8R AGCATCTGGAAGAACCTGGA Exon 15 l-H MCR9F CCTAAAAATAAAGCACCTACTGCTG 165 3 60 MCR9R CACTCAGGCTGGATGAACAA Exon 15 1 MCR 10F ACATTTTGCCACGGAAAGTA 158 2 55 MCR 10R GGCTGCTCTGATTCTGTTTC Exon 15 1 MCR 11F CAGGAAAATGACAATGGGAAT 168 2 55 MCR 11R TGGCATGGCAGAAATAATACA Exon 15 l-J MCR 12F GGACCTATTAGATGATTCAGATGATG 148 2 55 MCR 12R ACTTGGTTTCCTTGCCACAG
193 Table A.4: Primers and PCR conditions for the M F/f gene.
* Denotes addition of Q-solution (Qiagen).
MYH Primer Amplicon Final MgCb Annealing Primer sequence(S' to 3') region Name size, bp conc., mM temp., ®C Exon 1 IF TGAAGGCTACCTCTGGGAAG 141 1 55 1R AGGAGACGGACCGCAAGT Exon 2 2F GGCTGGGTCTTTTTGTTTCA 164 1 60
2R GGGCCACAACCTAGTTCCTT Exon 3 3AF CTGTGTCCCAAGACCCTGAT 187 3* 55 3AR TTGGTCGTACCAGCTTAGCA
Exon 3 3BF AGCTGAAGTCACAGCCTTCC 110 1 60
3BR CACCCACTGTCCCTGCTC
Exon 4 4F CCTCCACCCTAACTCCTCATC 110 1 55
4R AAAGTGGCCCTGCTCTCAG Exon 5 5F CAGGTCAGCAGTGTCCTCAT 152 2 60 5R GTCTGACCCATGACCCTTCC Exon 6 6F GTCTCTTTCTGCCTGCCTGT 125 2 60 6R TCACCCGTCAGTCCCTCTAT Exon 7 7F CGGGTGATCTCTTTGACCTC 134 2 60 7R GTTCCTACCCTCCTGCCATC Exon 8 8F TCTTGAGTCTTGCACTCCAATC 164 1 55 8R AAAGTGGGGGTGGGCTGT Exon 9 9F GCTAACTCTTTGGCCCCTCT 150 3* 60 9R CACCCTTGTTACCCCAACAT
Exon 10 IGF CTGCTTCACAGCAGTGTTCC 208 2* 60
10R GACTTCTCACTGCCCCTTCC Exon 11 11F ACACTCAACCCTGTGCCTCT 147 1 60
11R GGAATGGGGCTTCTGACTG Exon 12 12AF CTTGGCTTGAGTAGGGTTCG 179 1* 60
12AR GGCTGTTCCAGAACACAGGT Exon 12 12BF GAGTGGTCAACTTCCCCAGA 146 1 60 12BR CACGCCCAGTATCCAGGTA Exon 13 13F AGGGAATCGGCAGCTGAG 209 1 60 13R GCTATTCCGCTGCTCACTTA Exon 14 14F AGGCCTATTTGAACCCCTTG 210 1 60 14R CAACAAAGACAACAAAGGTAGTGC Exon 15 15F CCCTCACCTCCCTGTCTTCT 159 3* 55 15R TGTTCACCCAGACATTCGTT Exon 16 16AF CTACAAGGCCTCCCTCCTTC 158 r 55 16AR GCTGCACTGTTGAGGCTGT Exon 16 16BF GCCAGCAAGTCCTGGATAAT 181 2 60 16BR ACATAGCGAGACCCCCATCT
194 Table A.5: Primers and PCR conditions for the M TH l gene.
* Denotes addition of Q-solution (Qiagen).
Primer Amplicon Final MgCIa Annealing MTH1 region Primer sequence (5' to 3') Name size, bp conc., mM temp., “C Exon 2 2F GCCTTATCGCAAGGACAGAG 241 3* 60
2R TGATGGAACCATGAGTTTGG
Exon 3 3F TGACTCTGCCCTCTCACCTT 2 24 1 60
3R CGGTTCTATGGCCAGACCT
Exon 4 4 F TCCCTGGGCTGTGTGTAGAT 270 1 60
4 R GAGCAGGCCCTGTGAGACT
Exon 5 5F CTCTTCCCCCATTGGTACAG 240 1 60
5R CTGTTCAGCAGCCACGTCT
Table A.6: Primers and PCR conditions for the OGGI gene.
* Denotes addition of Q-solution (Qiagen).
Primer Amplicon Final MgCl2 Annealing OGGI region Primer sequence (S' to 3') Name size, bp conc., mM temp., *C Exon 1 I F GGGTAGGCGGGGCTACTAC 266 1 55
1R CCCCAAATTCCTTTGTACCC
Exon 2 2F AATTGAGTGCCAGGGTTGTC 313 1 55
2R CTAACCCCAGCCCAGGTC
Exon 3 3F CAGCAGGTACCTCTCCTACCC 269 2* 55
3R TCTTGAAAGCTGATGGAAGG
Exon 4 4 F TTGAAGATGCCTGATGCTTG 258 1 55
4 R TAGAGAGGGCAGCTCCTACC
Exon 5 5F TCTTCCACAAGGGCTCATTC 232 1 55
5 R TCTACCATCCCAGCCCACT
Exon 6 6F TCACAGAAGGGGTCAGATAACTT 197 3 60
6 R GGCTGGAGAGTCCTTTAGGG
Exon 7 7F GACCCCAGTGTACCCTCCTC 311 1 55
7R ATATCCCCCACCCCATCTT
Exon 8 BF ATTCTCCATGCTGCCTTCCT 346 1 55
8 R GTAAGCTGGCTTGCATCACA
195 Table A.7: Primers and PCR conditions for the GSK-3p gene.
* Denotes addition of Q-solution (Qiagen).
GSKSbeta Primer Amplicon Final MgCb Annealing Primer sequence (S' to 3') region Name size (bp) conc. (mM) temp. (®C)
Exon 1 1F AAAGGTGATTCGCGAAGAGA 164 2 55
1R TATTTTAAGGGCGAGGTGGA
Exon 2 2F TTCCTTACAGTTCTCGTGAAAAA 284 3* 50
2 R TGTATTACGGCAGCACAAAAA
Exon 3 3F TGTTATTTCAGAAGCATCTTGTTCA 245 2* 60
3R TATGAGCGGTGGGGAGATTA
Exon 4 4F AAGAGGCTCTCCTTGGTTCA 228 1 55
4 R GACCAGTTTCTAATCTGTTTTGTTTTT
Exon 5 5F CATTTGATACTGTGAAAGGATAGCA 259 3 55
5R AAATTACCAAAATCAAGAAGCA
Exon 6 6F CAAACACTTTTCAGCTCAAGCTAT 206 1 60
6R GCACAAAAATTCCTTCCAGTTC
Exon 7 7F CAGTGAATCCAATGCCTGAA 2 93 1 55
7R CATGCAGAACTGCTTGTATTCAT
Exon 8 BF ATGTTTCTGTACATGGTGGGAAT 343 2 55
8 R ACCCAGGGTGTAGCTTTCCT
Exon 9 9F GCCCGGCATAAACTGGTAGT 209 1 60
9 R GCTGCTGTGGCATTTGTG
196 Table A.8: Primers and PCR conditions for the axinl gene.
* Denotes addition of Q-solution (Qiagen).
A xin2 P rim e r A m p lic o n Final M g C l2 A n n e a lin g Primer sequence (S' to 3') reg io n N a m e s ize (b p ) c o n c. (m M ) te m p . (®C)
Exon 1 1AF CCAGACTCAGTGGGAAGAGC 2 5 4 1 55 1AR CAGAGGGGAATCCGGAGAT
Exon 1 1BF CGGAACGAAGATGGGTTG 323 2* 55 1BR TGCTTCTTGATGCCATCTCTT
Exon 1 1CF CCTGCCACCAAGACCTACAT 2 6 6 2 60 1C R AAGTCGGCACAAGTCCACTC
Exon 1 1D F CTCCCCACCTTGAATGAAGA 191 2* 55 1D R TCTGCAAAGTCCACAGCATC
Exon 2 2F GGTTTCTGGTGAGGGTCAGT 25B 1 60 2R TCTGGCTAAGTGCTCAGGTG
Exon 3 3F CCCATGTGATGTGCTGAAAA 2 1 3 1 60 3R TCAACATGGCAGAAAACAGC
Exon 4 4F CTCCTCCCTTTCCTTTCCAC 2 0 4 2* 60 4 R GACGGAAGCAGGAAGAAGG
Exon 5 5AF CCTTCTGACGTCTTCCCTTTC 2 6 0 1 55 5 A R GGAGTGGTACTGCGAATGGT
Exon 5 5B F GCCAGTCTCCAGGCGTAG 2 1 5 1 55 5 B R GGCATGGTGGTGGATGTAGT
Exon 5 5 C F CTTTGTGACCAAGCAGACGA 2 1 0 1 60 5 C R GGCACTTACCCAAACTGCTC
Exon 5 5D F AATGCAAAAGCCACTCCAAG 2 2 5 1 55 5 D R GCTGGTGACACGAAAGACC
Exon 5 5E F TTCTAAGAGGGCGGCAGAG 300 1 60 5 E R TGCGACCTGTCTCCTTCC
Exon 5 5FF CAAACGCAATGGGAAAGG 2 45 1* 55 5 F R CAGCCATTCCCACAATACCT
Exon 6 6F TCTGTTTTCTCTCTGCTCATTCC 2 97 2 60 6 R GCCTCAACCTAGGACCCTTC
Exon 7 7F AATTGCTCTGGGGACAACAG 195 1 60 7R AGCACTCGGCAGATCTCAGT
Exon 8 BF TGCTAAACTTGTTCCATTCCA 2 60 2 60 BR TCTGGCTCTTGGTTCTGAGC
Exon 9 9F CCTCAGTCCTCCATGTTGGT 241 2* 60 9R CAGTTCACCAAAGCCAGACC
197 Table A.9: Primers and PCR conditions for the APC2 gene.
* Denotes addition of Q-solution (Qiagen). ** Denotes a 2 minute extension time.
APC2 Primer Amplicon Final MgCb Annealing Primer sequence(S' to 3') region Name size (bp) conc. (mM) temp. (®C)
Exon 1 IF ACCATCAGCTGAACCCTCTG 220 1 55 1R TAGCCCACCTCCTCCATGT
Exon 2 2F CAGTGCAGTGGCTGATGG 243 2* 55 2R TCCATCTGTAGGGCTGGATG
Exon 3 3F GGAGCAGCTGAAGGGTGA 365 1* 55
3R GCGAATTAGAGGGCAGAGC
Exon 4 4 F GCCATGTTGCCTGTTAACCT 361 1* 55 4 R ATCGAGAACTGCTGGCAAAG
Exon 5 5F CACTCAGGGTGCGGGAAG 285 1.5* 60 5R CCGCCCATAATATCTGCACT Exon 6 6F GGGTCAGCTCCAGCACTTG 193 1* 60
6R CGGGCAGAACTCGAGGAG Exon 7 7F TGGGTGCTCTGGGACTGTA 2 05 2* 55
7R ACCTGATCCCCCTTCCAG Exon 8 8AF GACCGGGTTTCCAGGTGT 323 2* 55
8A R CTGCGAGAAGACGATGTTGT Exon 8 8B F TCTGCTGCTGCAAATCCTC 317 1* 55 8B R AGCCCTAGCTGCACGTTAAT
Exon 9 9F GTCACCTGGGACATTTCCTG 172 2* 60
9 R AGAGGACCTGGGACACTCAC
Exon 10 IG F ACCTTGTTGGGTCCTCACAG 2 13 1* 55
ICR TCCCTGGAAGACTGGATGAG
Exon 11/12 11F GCTGCATAACCCCCAACAG 4 36 1 60 12R CCAGTGACTGAAGAAGCATCC
Exon 13 13F CCAACTTCACTGAATGTGAGC 306 2 55 13R GGGCCGACCTGTAGTCC
Exon 14 PTTAF GGATCCTAATACGACTCACTATAGGAACAGACCACCATGCTCCGGGATCACAACTGTCT
PTTAR CTACGCTGCAAGGTTCACTGG 1699 1.5 55**
Exon 14 PTTBF GGATCCTAATACGACTCACTATAGGAACAGACCACCATGAGCTTCGAGAGCCCGTCCAT
PTTBR CTACTTTGAGCCGTCCTTGTCTG 940 1* 55**
198 References
Aaltonen, L.A. (2000) Hereditary intestinal cancer. Semin Cancer Biol, 10, 289-298. Aaltonen, L.A., Peltomaki, P., Leach, F.S., Sistonen, P., Pylkkanen, L., Mecklin, J.P., Jarvinen, H., Powell, S.M., Jen, J., Hamilton, S.R. and et al. (1993) Clues to the pathogenesis of familial colorectal cancer. Science, 260, 812- 816. Aaltonen, L.A., Salovaara, R., Kristo, P., Canzian, P., Hemminki, A., Peltomaki, P., Chadwick, R.B., Kaariainen, H., Eskelinen, M., Jarvinen, H., Mecklin, J.P. and de la Chapelle, A. (1998) Incidence of hereditary nonpolyposis colorectal cancer and the feasibility of molecular screening for the disease. N Engl J Med,33S, 1481-1487. Aarskog, N.K. and Vedeler, C.A. (2000) Real-time quantitative polymerase chain reaction. A new method that detects both the peripheral myelin protein 22 duplication in Charcot-Marie-Tooth type 1A disease and the peripheral myelin protein 22 deletion in hereditary neuropathy with liability to pressure palsies. Hum Genet, 107,494-498. Abdel-Rahman, W.M., Katsura, K., Rens, W., Gorman, P.A., Sheer, D., Bicknell, D., Bodmer, W.F., Arends, M.J., Wyllie, A.H. and Edwards, P.A. (2001) Spectral karyotyping suggests additional subsets of colorectal cancers characterized by pattern of chromosome rearrangement. Proc Natl Acad Sci USA, 98,2538- 2543. Aberle, H., Bauer, A., Stappert, J., Kispert, A. and Kemler, R. (1997) beta-catenin is a target for the ubiquitin-proteasome pathway. Embo J, 16, 3797-3804. Akiyama, Y., Sato, H., Yamada, T., Nagasaki, H., Tsuchiya, A., Abe, R. and Yuasa, Y. (1997) Germ-line mutation of the hMSH6/GTBP gene in an atypical hereditary nonpolyposis colorectal cancer kindred. Cancer Res, 57, 3920- 3923. Akiyama, Y., Yagi, O.K., Ishikawa, T., Nagasaki, H., Saitoh, K. and Yuasa, Y. (1998) Genetic alterations are frequent in APC but rare in the TGF-beta type II receptor gene in cancer in adenomas of the colon. Cancer Lett, 125, 89-96.
199 Al-Jaberi, T.M. and El-Shanti, H. (2002) Diversity in polyp pathology and distribution of Familial Juvenile Polyposis Syndrome. Saudi Med J, 23, 328- 331. Al-Tassan, N., Chmiel, N.H., Maynard, J., Fleming, N., Livingston, A.L., Williams, G.T., Hodges, A.K., Davies, D.R., David, S.S., Sampson, J.R. and Cheadle, J.P. (2002) Inherited variants of MYH associated with somatic G:C-->A mutations in colorectal tumors. Nat Genet, 30, 227-232. Albuquerque, C., Breukel, C., van der Luijt, R., Fidalgo, P., Lage, P., Slors, F.J., Leitao, C.N., Fodde, R. and Smits, R. (2002a) The 'just-right' signaling model: APC somatic mutations are selected based on a specific level of activation of the beta-catenin signaling cascade. Hum Mol Genet, 11,1549- 1560. Albuquerque, C., Cravo, M., Cruz, C., Lage, P., Chaves, P., Fidalgo, P., Suspiro, A. and Nobre Leitao, C. (2002b) Genetic characterisation of patients with multiple colonic polyps. J Med Genet, 39, 297-302. Almeida, R., Morton, N., Fidalgo, P., Leitao, N., Mira, C., Rueff, J. and Monteiro, C. (1996) APC intragenic haplotypes in familial adenomatous polyposis. Clin Genet, 50,483-485. Armstrong, B. and Doll, R. (1975) Environmental factors and cancer incidence and mortality in different countries, with special reference to dietary practices. Int J Cancer, 15, 617-631. Attisano, L. and Wrana, J.L. (2002) Signal transduction by the TGF-beta superfamily. Science, 296, 1646-1647. Baker, S.J., Fearon, E.R., Nigro, J.M., Hamilton, S.R., Preisinger, A.C., Jessup, J.M., vanTuinen, P., Ledbetter, D.H., Barker, D.F., Nakamura, Y. and et al. (1989) Chromosome 17 deletions and p53 gene mutations in colorectal carcinomas. Science, 244, 217-221. Baker, S.J., Preisinger, A C., Jessup, J.M., Paraskeva, C., Markowitz, S., Willson, J.K., Hamilton, S. and Vogelstein, B. (1990) p53 gene mutations occur in combination with 17p allelic deletions as late events in colorectal tumorigenesis. Cancer Res, 50, 7717-7722. Barber, J.C., Ellis, K.H., Bowles, L.V., Delhanty, J.D., Ede, R.F., Male, B.M. and Eccles, D M. (1994) Adenomatous polyposis coli and a cytogenetic deletion
200 of chromosome 5 resulting from a maternal intrachromosomal insertion. J Med Genet, 31, 312-316. Beauchamp, R.D., Sheng, H.M., Ishizuka, J., Townsend, C.M., Jr. and Thompson, J.C. (1992) Transforming growth factor (TGF)-beta stimulates hepatic jun-B and fos-B proto-oncogenes and decreases albumin mRNA. Ann Surg, 216, 300-307; discussion 307-308. Behrens, J., Jerchow, B.A., Wurtele, M., Grimm, J., Asbrand, C., Wirtz, R., Kuhl, M., Wedlich, D. and Birchmeier, W. (1998) Functional interaction of an axin homolog, conductin, with beta-catenin, APC, and GSK3beta. Science, 280, 596-599. Belenkaya, T.Y., Han, C., Standley, H.J., Lin, X., Houston, D.W. and Heasman, J. (2002) pygopus Encodes a nuclear protein essential for wingless/Wnt signaling. Development, 129,4089-4101. Beroud, C. and Soussi, T. (1996) APC gene: database of germline and somatic mutations in human tumors and cell lines. Nucleic Acids Res, 24, 121-124. Bhanot, P., Brink, M., Samos, C.H., Hsieh, J.C., Wang, Y., Macke, J.P., Andrew, D., Nathans, J. and Nusse, R. (1996) A new member of the frizzled family from Drosophila functions as a Wingless receptor. Nature, 382, 225-230. Bisgaard, M.L., Fenger, K., Bulow, S., Niebuhr, B. and Mohr, J. (1994) Familial adenomatous polyposis (FAP): frequency, penetrance, and mutation rate. Hum Mutat, 3, 121-125. Blair, N.P. and Trempe, C.L. (1980) Hypertrophy of the retinal pigment epithelium associated with Gardner's syndrome. Am J Ophthalmol, 90, 661-667. Boardman, L.A., Thibodeau, S.N., Schaid, D.J., Lindor, N.M., McDonnell, S.K., Burgart, L.J., Ahlquist, D.A., Podratz, K.C., Pittelkow, M. and Hartmann, B.C. (1998) Increased risk for cancer in patients with the Peutz-Jeghers syndrome. Ann Intern Med, 128, 896-899. Bodmer, W.F., Bailey, C.J., Bodmer, J., Bussey, H.J., Ellis, A., Gorman, P., Lucibello, F.C., Murday, V.A., Rider, S.H., Scambler, P. and et al. (1987) Localization of the gene for familial adenomatous polyposis on chromosome 5. Nature, 328, 614-616. Bos, J.L., Fearon, E.R., Hamilton, S R., Verlaan-de Vries, M., van Boom, J.H., van der Eb, A.J. and Vogelstein, B. (1987) Prevalence of ras gene mutations in human colorectal cancers. Nature, 327, 293-297.
201 Bronner, C E., Baker, S.M., Morrison, P.T., Warren, G., Smith, L.G., Lescoe, M.K., Kane, M., Earabino, C., Lipford, J., Lindblom, A. and et al. (1994) Mutation in the DNA mismatch repair gene homologue hMLHl is associated with hereditary non-polyposis colon cancer. Nature, 368, 258-261. Brownstein, M.H., Mehregan, A.H., Bikowski, J.B., Lupulescu, A. and Patterson, J.C. (1979) The dermatopathology of Cowden's syndrome. Br J Dermatol, 100, 667-673. Bulow, S. (1986) Clinical features in familial polyposis coli. Results of the Danish Polyposis Register. Dis Colon Rectum, 29,102-107. Caspari, R., Olschwang, S., Friedl, W., Mandl, M., Boisson, C., Boker, T., Augustin, A., Kadmon, M., Moslem, G., Thomas, G. and et al. (1995) Familial adenomatous polyposis: desmoid tumours and lack of ophthalmic lesions (CHRPE) associated with APC mutations beyond codon 1444. Hum Mol Genet, 4, 337-340. Chang, W.W. and Whitener, C.J. (1989) Histogenesis of tubular adenomas in hereditary colonic adenomatous polyposis. Arch Pathol Lab Med, 113, 1042- 1049. Christofori, G. and Semb, H. (1999) The role of the cell-adhesion molecule E- cadherin as a tumour-suppressor gene. Trends Biochem Set, 24, 73-76. Chulada, P.C., Thompson, M.B., Mahler, J.F., Doyle, C.M., Gaul, B.W., Lee, C., Tiano, H.F., Morham, S.G., Smithies, O. and Langenbach, R. (2000) Genetic disruption of Ptgs-1, as well as Ptgs-2, reduces intestinal tumorigenesis in Min mice. Cancer Res, 60, 4705-4708. Cotran, R., Collins, K.V. and Robbins, T. (1999) Pathologic Basis o f Disease. W. B. Saunders Company, Philadelphia. Cottrell, S., Bicknell, D., Kaklamanis, L. and Bodmer, W.F. (1992) Molecular analysis of APC mutations in familial adenomatous polyposis and sporadic colon carcinomas. Lancet, 340, 626-630. Crabtree, M.D., Tomlinson, I.P., Hodgson, S.V., Neale, K., Phillips, R.K. and Houlston, R.S. (2002) Explaining variation in familial adenomatous polyposis: relationship between genotype and phenotype and evidence for modifier genes. Gut, 51,420-423. Crabtree, M.D., Tomlinson, I P., Talbot, I.C. and Phillips, R.K. (2001) Variability in the severity of colonic disease in familial adenomatous polyposis results from
202 differences in tumour initiation rather than progression and depends relatively little on patient age. Gut, 49, 540-543. Cross, I., Delhanty, J., Chapman, P., Bowles, L.V., Griffin, D., Wolstenholme, J., Bradbum, M., Brown, J., Wood, C., Gunn, A. and et al. (1992) An intrachromosomal insertion causing 5q22 deletion and familial adenomatous polyposis coli in two generations. J Med Genet, 29,175-179. Cummings, O.W. (2000) Pathology of the adenoma-carcinoma sequence: from aberrant crypt focus to invasive carcinoma. Semin Gastrointest Dis, 11, 329- 237. Cunningham, J.M., Christensen, E.R., Tester, D.J., Kim, C.Y., Roche, P.C., Burgart, L.J. and Thibodeau, S.N. (1998) Hypermethylation of the hMLHl promoter in colon cancer with microsatellite instability. Cancer Res, 58, 3455-3460. Davies, S.M. and Snover, D C. (1994) Frequent polymorphism in exon 15 of the adenomatous polyposis coli gene. Hum Genet, 93, 329-330. De Rosa, M., Scarano, M.I., Panariello, L., Carlomagno, N., Rossi, G.B., Tempesta, A., Borgheresi, P., Renda, A. and Izzo, P. (1999) Three submicroscopic deletions at the APC locus and their rapid detection by quantitative-PCR analysis. EurJHum Genet, 7, 695-703. Deka, J., Herter, P., Sprenger-Haussels, M., Koosch, S., Franz, D., Muller, K.M., Kuhnen, C., Hoffmann, I. and Muller, O. (1999) The APC protein binds to A/T rich DNA sequences. Oncogene, 18, 5654-5661. Deka, J., Kuhlmann, J. and Muller, O. (1998) A domain within the tumor suppressor protein APC shows very similar biochemical properties as the microtubule- associated protein tau. Bur J Biochem, 253, 591-597. Eppert, K., Scherer, S.W., Ozcelik, H., Pirone, R., Hoodless, P., Kim, H., Tsui, L.C., Bapat, B., Gallinger, S., Andrulis, I.E., Thomsen, G.H., Wrana, J.L. and Attisano, L. (1996) MADR2 maps to 18q21 and encodes a TGFbeta- regulated MAD-related protein that is functionally mutated in colorectal carcinoma. Cell, 86, 543-552. Erdmann, K.S., Kuhlmann, J., Lessmann, V., Herrmann, L., Eulenburg, V., Muller, O. and Heumann, R. (2000) The Adenomatous Polyposis Coli-protein (APC) interacts with the protein tyrosine phosphatase PTP-BL via an alternatively spliced PDZ domain. Oncogene, 19, 3894-3901.
203 Evertsson, S., Lindblom, A. and Sun, X.F. (2001) APC I1307K and B1317Q variants are rare or do not occur in Swedish colorectal cancer patients. Eur J Cancer, 37, 499-502. Fearon, E.R., Hamilton, S.R. and Vogelstein, B. (1987) Clonal analysis of human colorectal tumors. Science, 238, 193-197. Figer, A., Shtoyerman-Chen, R., Tamir, A., Geva, R., Irmin, L., Flex, D., Theodor, L., Sulkes, A., Sadetzki, S., Bar-Meir, S. and Friedman, E. (2001) Phenotypic characteristics of colo-rectal cancer in I1307K APC germline mutation carriers compared with sporadic cases. Br J Cancer, 85, 1368-1371. Fodde, R., Kuipers, J., Rosenberg, C., Smits, R., Kielman, M., Caspar, C., van Es, J.H., Breukel, C., Wiegant, J., Giles, R.H. and Clevers, H. (2001a) Mutations in the APC tumour suppressor gene cause chromosomal instability. Nat Cell Biol, 3, 433-438. Fodde, R., Smits, R. and Clevers, H. (2001b) APC, signal transduction and genetic instability in colorectal cancer. Nature Rev Cancer, 1, 55-67. Forman, B.M., Tontonoz, P., Chen, J., Brun, R.P., Spiegelman, B.M. and Evans, R.M. (1995) 15-Deoxy-delta 12,14-prostaglandin J2 is a ligand for the adipocyte determination factor PPAR gamma. Cell, 83, 803-812. Forrester, K., Almoguera, C., Han, K., Grizzle, W.E. and Perucho, M. (1987) Detection of high incidence of K-ras oncogenes during human colon tumorigenesis. Nature, 327, 298-303. Frayling, I.M., Beck, N.E., Ilyas, M., Dove-Edwin, I., Goodman, P., Pack, K., Bell, J.A., Williams, C.B., Hodgson, S.V., Thomas, H.J., Talbot, I.C., Bodmer, W.F. and Tomlinson, I.P. (1998) The APC variants I1307K and E1317Q are associated with colorectal tumors, but not always with a family history. Proc Natl Acad Sci USA, 95, 10722-10727. Fuchs, C.S., Giovannucci, E.L., Colditz, G.A., Hunter, D.J., Speizer, F.E. and Willett, W.C. (1994) A prospective study of family history and the risk of colorectal cancer. J Med, 331, 1669-1674. Gerwin, B.I., Keski-Oja, J., Seddon, M., Lechner, J.F. and Harris, C.C. (1990) TGF- beta 1 modulation of urokinase and PAI-1 expression in human bronchial epithelial cells. Am J Physiol, 259, L262-269. Giardiello, F.M., Brensinger, J.D., Tersmette, A.C., Goodman, S.N., Petersen, G.M., Booker, S.V., Cruz-Correa, M. and Offerhaus, J.A. (2000) Very high risk of
204 cancer in familial Peutz-Jeghers syndrome. Gastroenterology, 119, 1447- 1453. Giardiello, F.M., Hamilton, S.R., Kern, S.E., Offerhaus, G.J., Green, P.A., Celano, P., Krush, A.J. and Booker, S.V. (1991) Colorectal neoplasia in juvenile polyposis or juvenile polyps. Arch Dis Child, 66, 971-975. Giardiello, P.M., Hamilton, S.R., Krush, A.J., Piantadosi, S., Hylind, L.M., Celano, P., Booker, S.V., Robinson, C.R. and Offerhaus, G.J. (1993) Treatment of colonic and rectal adenomas with sulindac in familial adenomatous polyposis. NEnglJMed,32S, 1313-1316. Giardiello, P.M., Welsh, S.B., Hamilton, S.R., Offerhaus, G.J., Gittelsohn, A.M., Booker, S.V., Krush, A.J., Yardley, J.H. and Luk, G.D. (1987) Increased risk of cancer in the Peutz-Jeghers syndrome. NEngl JMed, 316,1511-1514. Giaretti, W., Sciallero, S., Bruno, S., Geido, E., Aste, H. and Di Vinci, A. (1988) DNA flow cytometry of endoscopically examined colorectal adenomas and adenocarcinomas. Cytometry, 9, 238-244. Giles, R.H., van Es, J.H. and Clevers, H. (2003) Caught up in a Wnt storm: Wnt signaling in cancer. Biochim Biophys Acta, 1653,1-24. Gillen, C.D., Andrews, H.A., Prior, P. and Allan, R.N. (1994) Crohn's disease and colorectal cancer. Gut, 35, 651-655. Giovannucci, E., Egan, K.M., Hunter, D.J., Stampfer, M.J., Colditz, G.A., Willett, W.C. and Speizer, P.E. (1995) Aspirin and the risk of colorectal cancer in women. NEngl JMed, 333, 609-614. Giovannucci, E., Rimm, E.B., Stampfer, M.J., Colditz, G.A., Ascherio, A. and Willett, W.C. (1994) Aspirin use and the risk for colorectal cancer and adenoma in male health professionals. Ann Intern Med, 121, 241-246. Goh, H.S. (1987) Plow cytometry and colorectal neoplasia. Ann Acad Med Singapore, 16, 535-538. Gorlin, R.J., Cohen, M.M., Jr., Condon, L.M. and Burke, B.A. (1992) Bannayan- Riley-Ruvalcaba syndrome. Am JM ed Genet, 44, 307-314. Groden, J., Thliveris, A., Samowitz, W., Carlson, M., Gelbert, L., Albertsen, H., Joslyn, G., Stevens, J., Spirio, L., Robertson, M. and et al. (1991) Identification and characterization of the familial adenomatous polyposis coli gene. Cell, 66, 589-600.
205 Groves, C., Lamlum, H., Crabtree, M., Williamson, J., Taylor, C., Bass, S., Cuthbert- Heavens, D., Hodgson, S., Phillips, R. and Tomlinson, I. (2002) Mutation cluster region, association between germline and somatic mutations and genotype-phenotype correlation in upper gastrointestinal familial adenomatous polyposis. Am J Pathol, 160,2055-2061. Gryfe, R., Di Nicola, N., Gallinger, S. and Redston, M. (1998) Somatic instability of the APC I1307K allele in colorectal neoplasia. Cancer Res, 58,4040-4043. Gryfe, R., Di Nicola, N., Lai, G., Gallinger, S. and Redston, M. (1999) Inherited colorectal polyposis and cancer risk of the APC I1307K polymorphism. Am J Hum Genet, 64, 378-384. Gu, Y., Parker, A., Wilson, T.M., Bai, H., Chang, D.Y. and Lu, A.L. (2002) Human MutY homolog, a DNA glycosylase involved in base excision repair, physically and functionally interacts with mismatch repair proteins human MutS homolog 2/human MutS homolog 6.JBiol Chem, 277, 11135-11142. Gupta, R.A. and Dubois, R.N. (2001) Colorectal cancer prevention and treatment by inhibition of cyclooxygenase-2. Nat Rev Cancer, 1, 11-21. Gupta, R.A., Tan, J., Krause, W.F., Geraci, M.W., Willson, T.M., Dey, S.K. and DuBois, R.N. (2000) Prostacyclin-mediated activation of peroxisome proliferator-activated receptor delta in colorectal cancer. Proc Natl Acad Sci USA, 97,13275-13280. Haenszel, W. and Kurihara, M. (1968) Studies of Japanese migrants. I. Mortality from cancer and other diseases among Japanese in the United States. J Natl Cancer Inst, 40,43-68. Hahn, S.A., Schutte, M., Hoque, A.T., Moskaluk, C.A., da Costa, L.T., Rozenblum, E., Weinstein, C.L., Fischer, A., Yeo, C.J., Hruban, R.H. and Kem, S.E. (1996) DPC4, a candidate tumor suppressor gene at human chromosome 18q21.1. Science, 271, 350-353. Haigis, K.M., Caya, J.G., Reichelderfer, M. and Dove, W.F. (2002) Intestinal adenomas can develop with a stable karyotype and stable microsatellites. Proc Natl Acad Sci USA, 99, 8927-8931. Halliwell, B. (1996) Mechanisms involved in the generation of free radicals. Pathol Biol (Paris), 44, 6-13.
206 Hamada, S., Namura, K., Itoh, R. andFujita, S, (1987) Characteristics of colorectal epithelia and adenomas as revealed by DNA cytofluorometry. Jpn J Cancer Res, 78, 826-832. Hamilton, S R., Liu, B., Parsons, R.E., Papadopoulos, N., Jen, J., Powell, S.M., Krush, A.J., Berk, T., Cohen, Z., Tetu, B. and et al. (1995) The molecular basis of Turcot's syndrome. NEngl JMed, 332, 839-847. Hanaoka, T., Sugimura, H., Nagura, K., Diara, M., Li, X.J., Hamada, G.S., Nishimoto, L, Kowalski, L.P., Yokota, J. and Tsugane, S. (2001) hOGGl exon7 polymorphism and gastric cancer in case-control studies of Japanese Brazilians and non-Japanese Brazilians. Cancer Lett, 170, 53-61. Hao, X., Palazzo, J.P., Ilyas, M., Tomlinson, I. and Talbot, I.C. (1997) Reduced expression of molecules of the cadherin/catenin complex in the transition from colorectal adenoma to carcinoma. Anticancer Res, 17, 2241-2247. Hao, X.P., Frayling, I.M., Sgouros, J.G., Du, M.Q., Willcocks, T.C., Talbot, I.C. and Tomlinson, I.P. (2002) The spectrum of p53 mutations in colorectal adenomas differs from that in colorectal carcinomas. Gut, 50, 834-839. Hart, A.R., Kudo, S., Mackay, E.H., Mayberry, J.F. and Atkin, W.S. (1998a) Flat adenomas exist in asymptomatic people: important implications for colorectal cancer screening programmes. Gut, 43, 229-231. Hart, M.J., de los Santos, R., Albert, I N., Rubinfeld, B. and Polakis, P. (1998b) Downregulation of beta-catenin by human Axin and its association with the APC tumor suppressor, beta-catenin and GSK3 beta. Curr Biol, 8, 573-581. Hayne, D., Brown, R.S., McCormack, M., Quinn, M.J., Payne, H.A. and Babb, P. (2001) Current trends in colorectal cancer: site, incidence, mortality and survival in England and Wales. Clin Oncol (R Coll Radiol), 13, 448-452. He, T.C., Sparks, A.B., Rago, C., Hermeking, H., Zawel, L., da Costa, L.T., Morin, P.J., Vogelstein, B. and Kinzler, K.W. (1998) Identification of c-MYC as a target of the APC pathway [see comments]. Science, 281,1509-1512. He, X., Saint-Jeannet, J.P., Wang, Y., Nathans, J., Dawid, I. and Varmus, H. (1997) A member of the Frizzled protein family mediating axis induction by Wnt- 5A. Science, 275, 1652-1654. Heighway, J., Hoban, P R. and Wyllie, A.H. (1991) SspI polymorphism in sequence encoding 3' untranslated region of the APC gene. Nucleic Acids Res, 19, 6966.
207 Heiskanen, I. and Jarvinen, H.J. (1996) Occurrence of desmoid tumours in familial adenomatous polyposis and results of treatment. Int J Colorectal Dis, 11, 157-162. Heiss, K.F., Schaffher, D., Ricketts, R.R. and Winn, K. (1993) Malignant risk in juvenile polyposis coli: increasing documentation in the pediatric age group. J Pediatr Surg, 28, 1188-1193. Hemminki, A., Markie, D., Tomlinson, I., Avizienyte, E., Roth, S., Loukola, A., Bignell, G., Warren, W., Aminoff, M., Hoglund, P., Jarvinen, H., Kristo, P., Pelin, K., Ridanpaa, M., Salovaara, R., Toro, T., Bodmer, W., Olschwang, S., Olsen, A.S., Stratton, M.R., de la Chapelle, A. and Aaltonen, L.A. (1998) A serine/threonine kinase gene defective in Peutz-Jeghers syndrome. Nature, 391,184-187. Herman, J.G., Umar, A., Polyak, K., Graff, J R., Ahuja, N., Issa, J.P., Markowitz, S., Willson, J.K., Hamilton, S.R., Kinzler, K.W., Kane, M.F., Kolodner, R.D., Vogelstein, B., Kunkel, T.A. and Baylin, S B. (1998) Incidence and functional consequences of hMLHl promoter hypermethylation in colorectal carcinoma. Proc Natl Acad Sci USA, 95, 6870-6875. Herrera, L., Kakati, S., Gibas, L., Pietrzak, E. and Sandberg, A.A. (1986) Gardner syndrome in a man with an interstitial deletion of 5q. Am JM ed Genet, 25, 473-476. Herve, R., Farret, O., Mayaudon, H., Helie, C., Denee, J.M., Bauduceau, B. and Molinie, C. (1995) [Association of Gardner syndrome and thyroid carcinoma (letter) (see comments)]. Presse Med, 24,415. Hinoi, T., Yamamoto, H., Kishida, M., Takada, S., Kishida, S. and Kikuchi, A. (2000) Complex Formation of Adenomatous Polyposis Coli and Axin Facilitates Glycogen Synthase Kinase-3 {beta}-dependent Phosphorylation of {beta}-Catenin and Downregulates {beta}-Catenin. JB iol Chem. Hizawa, K., lida, M., Matsumoto, T., Kohrogi, N., Kinoshita, H., Yao, T. and Fujishima, M. (1993a) Cancer in Peutz-Jeghers syndrome. Cancer, 72,2777- 2781. Hizawa, K., lida, M., Matsumoto, T., Kohrogi, N., Suekane, H., Yao, T. and Fujishima, M. (1994) Gastrointestinal manifestations of Cowden's disease. Report of four cases. / C/m Gastroenterol, 18,13-18.
208 Hizawa, K., lida, M., Matsumoto, T., Kohrogi, N., Yao, T. and Fujishima, M. (1993b) Neoplastic transformation arising in Peutz-Jeghers polyposis. Dis Colon Rectum, 36, 953-957. Hockey, K.A., Mulcahy, M.T., Montgomery, P. and Levitt, S. (1989) Deletion of chromosome 5q and familial adenomatous polyposis. JM ed Genet, 26, 61- 62. Hodgson, S.V., Coonar, A.S., Hanson, P.J., Cottrell, S., Scriven, P.N., Jones, T., Hawley, P.R. and Wilkinson, M.L. (1993) Two cases of 5q deletions in patients with familial adenomatous polyposis: possible link with Caroli's disease. JM ed Genet, 30, 369-375. Hodgson, S.V., Fagg, N.L., Talbot, I.C. and Wilkinson, M. (1994) Deletions of the entire APC gene are associated with sessile colonic adenomas [letter]. JM ed Genet, 31,426. Hofstad, B., Vatn, M.H., Andersen, S.N., Huitfeldt, H.S., Rognum, T., Larsen, S. and Osnes, M. (1996) Growth of colorectal polyps: redetection and evaluation of unresected polyps for a period of three years. Gut, 39, 449-456. Hofting, I., Pott, G., Schrameyer, B. and Stolte, M. (1993) [Familial juvenile polyposis with predominant stomach involvement]. Z Gastroenterol, 31, 480- 483. Horii, A., Nakatsuru, S., Ichii, S., Nagase, H. and Nakamura, Y. (1993) Multiple forms of the APC gene transcripts and their tissue-specific expression. Hum Mol Genet, 2, 283-287. Houlston, R., Bevan, S., Williams, A., Young, J., Dunlop, M., Rozen, P., Eng, C., Markie, D., Woodford-Richens, K., Rodriguez-Bigas, M.A., Leggett, B., Neale, K., Phillips, R., Sheridan, E., Hodgson, S., Iwama, T., Eccles, D., Bodmer, W. and Tomlinson, I. (1998) Mutations in DPC4 (SMAD4) cause juvenile polyposis syndrome, but only account for a minority of cases. Hum Mol Genet,!, 1907-1912. Howe, J R., Bair, J.L., Sayed, M.G., Anderson, M.E., Mitros, F.A., Petersen, G.M., Velculescu, V.E., Traverso, G. and Vogelstein, B. (2001) Germline mutations of the gene encoding bone morphogenetic protein receptor 1A in juvenile polyposis. Nat Genet, 28, 184-187. Howe, J R., Mitros, F.A. and Summers, R.W. (1998a) The risk of gastrointestinal carcinoma in familial juvenile polyposis. Ann Surg Oncol, 5, 751-756.
209 Howe, J R., Roth, S., Ringold, J.C., Summers, R.W., Jarvinen, H.J., Sistonen, P., Tomlinson, I.P., Houlston, R.S., Bevan, S., Mitros, F.A., Stone, E.M. and Aaltonen, L.A. (1998b) Mutations in the SMAD4/DPC4 gene in juvenile polyposis. Science^ 280,1086-1088. Ichii, S., Takeda, S., Horii, A., Nakatsuru, S., Miyoshi, Y., Emi, M., Fujiwara, Y., Koyama, K., Furuyama, J., Utsunomiya, J. and et al. (1993) Detailed analysis of genetic alterations in colorectal tumors from patients with and without familial adenomatous polyposis (FAP). Oncogene, 8, 2399-2405. Ikeda, S., Kishida, S., Yamamoto, H., Murai, H., Koyama, S. and Kikuchi, A. (1998) Axin, a negative regulator of the Wnt signaling pathway, forms a complex with GSK-3beta and beta-catenin and promotes GSK-3beta-dependent phosphorylation of beta-catenin. 17,1371-1384. Ilyas, M., Straub, J., Tomlinson, I.P. and Bodmer, W.F. (1999) Genetic pathways in colorectal and other cancers. Eur J Cancer, 35, 1986-2002. Ilyas, M., Tomlinson, I P., Rowan, A., Pignatelli, M. and Bodmer, W.F. (1997) Beta- catenin mutations in cell lines established from human colorectal cancers. Proc Natl Acad Sci USA, 94, 10330-10334. Inman, G.J. and Hill, C.S. (2002) Stoichiometry of active smad-transcription factor complexes on DNA. JB iol Chem, 277, 51008-51016. Ionov, Y., Peinado, M.A., Malkhosyan, S., Shibata, D. and Perucho, M. (1993) Ubiquitous somatic mutations in simple repeated sequences reveal a new mechanism for colonic carcinogenesis. Nature, 363, 558-561. Ishidate, T., Matsumine, A., Toyoshima, K. and Akiyama, T. (2000) The APC- hDLG complex negatively regulates cell cycle progression from the GO/Gl to S phase. Oncogene, 19, 365-372. Jaeger, E.E., Woodford-Richens, K.L., Lockett, M., Rowan, A.J., Sawyer, E.J., Heinimann, K , Rozen, P., Murday, V.A., Whitelaw, S.C., Ginsberg, A., Atkin, W.S., Lynch, H.T., Southey, M.C., Debinski, H., Eng, C., Bodmer, W.F., Talbot, I.e., Hodgson, S.V., Thomas, H.J. and Tomlinson, I.P. (2003) An ancestral Ashkenazi haplotype at the HMPS/CRACl locus on 15ql3-ql4 is associated with hereditary mixed polyposis syndrome. Am J Hum Genet, 72, 1261-1267.
210 Jarrett, C.R., Blancato, J., Cao, T., Bressette, D.S., Cepeda, M., Young, P.E., King, C.R. and Byers, S.W. (2001) Human APC2 localization and allelic imbalance. Cancer Res, 61, 7978-7984, Jaruga, P., Zastawny, T.H., Skokowski, J., Dizdaroglu, M. and Olinski, R. (1994) Oxidative DNA base damage and antioxidant enzyme activities in human lung cancer. FEBSLett, 341, 59-64. Jass, J.R., Williams, C.B., Bussey, H.J. and Morson, B.C. (1988) Juvenile polyposis- -a precancerous condition. Histopathology, 13, 619-630. Jenne, D.E., Reimann, H., Nezu, J., Friedel, W., Loff, S., Jeschke, R., Muller, O., Back, W. and Zimmer, M. (1998) Peutz-Jeghers syndrome is caused by mutations in a novel serine threonine kinase. Nat Genet, 18, 38-43. Jho, E.H., Zhang, T., Domon, C., Joo, C.K., Freund, J.N. and Costantini, F. (2002) Wnt/beta-catenin/Tcf signaling induces the transcription of Axin2, a negative regulator of the signaling pathway.Mol Cell Biol, 22, 1172-1183. Joslyn, G., Carlson, M., Thliveris, A., Albertsen, H., Gelbert, L., Samowitz, W., Groden, J., Stevens, J., Spirio, L., Robertson, M. and et al. (1991) Identification of deletion mutations and three new genes at the familial polyposis locus. Cell, 66, 601-613. Joslyn, G., Richardson, D.S., White, R. and Alber, T. (1993) Dimer formation by an N-terminal coiled coil in the APC protein. Proc Natl Acad Sci USA, 90, 11109-11113. Kallioniemi, A., Kallioniemi, O.P., Sudar, D., Rutovitz, D., Gray, J.W., Waldman, F. and Pinkel, D. (1992) Comparative genomic hybridization for molecular cytogenetic analysis of solid tumors. Science, 258, 818-821. Kallioniemi, A., Visakorpi, T., Karhu, R., Pinkel, D. and Kallioniemi, O.P. (1996) Gene Copy Number Analysis by Fluorescence in Situ Hybridization and Comparative Genomic Hybridization. Methods, 9,113-121. Kang, D., Nishida, J., lyama. A., Nakabeppu, Y., Furuichi, M., Fujiwara, T., Sekiguchi, M. and Takeshige, K. (1995) Intracellular localization of 8-oxo- dGTPase in human cells, with special reference to the role of the enzyme in mitochondria. JBiol Chem, 270, 14659-14665. Kaplan, K.B., Burds, A.A., Swedlow, J.R., Bekir, S.S., Sorger, P.K. and Nathke, I S. (2001) A role for the Adenomatous Polyposis Coli protein in chromosome segregation.Nat Cell Biol, 3, 429-432.
211 Karuman, P., Gozani, 0., Odze, R.D., Zhou, X.C., Zhu, H., Shaw, R., Brien, T.P., Bozzuto, C D., Ooi, D., Cantley, L.C. and Yuan, J. (2001) The Peutz-Jegher gene product LKBl is a mediator of p53-dependent cell death. Mol Cell, 7, 1307-1319. Kawasaki, Y., Senda, T., Ishidate, T., Koyama, R., Morishita, T., Iwayama, Y., Higuchi, O. and Akiyama, T. (2000) Asef, a link between the tumor suppressor APC and G-protein signaling. Science, 289,1194-1197. King, H. and Haenszel, W. (1973) Cancer mortality among foreign- and native-born Chinese in the United States. J Chronic Dis, 26, 623-646. Kinzler, K.W., Nilbert, M.C., Su, L.K., Vogelstein, B., Bryan, T.M., Levy, D.B., Smith, K.J., Preisinger, A.C., Hedge, P., McKechnie, D. and et al. (1991) Identification of FAP locus genes from chromosome 5q21. Science, 253, 661- 665. Kishida, S., Yamamoto, H., Ikeda, S., Kishida, M., Sakamoto, I., Koyama, S. and Kikuchi, A. (1998) Axin, a negative regulator of the wnt signaling pathway, directly interacts with adenomatous polyposis coli and regulates the stabilization of beta-catenin. JBiol Chem, 273,10823-10826. Kliewer, S.A., Lenhard, J.M., Willson, T.M., Patel, I., Morris, D C. and Lehmann, J.M. (1995) A prostaglandin J2 metabolite binds peroxisome proliferator- activated receptor gamma and promotes adipocyte differentiation. Cell, 83, 813-819. Knudsen, A.L., Bisgaard, M.L. and Bulow, S. (2003) Attenuated familial adenomatous polyposis (AFAP): a review of the literature. Fam Cancer, 2, 43-55. Knudson, A.G., Jr. (1971) Mutation and cancer: statistical study of retinoblastoma. Proc Natl Acad Sci USA, 68, 820-823. Kohno, T., Shinmura, K., Tosaka, M., Tani, M., Kim, S.R., Sugimura, H., Nohmi, T., Kasai, H. and Yokota, J. (1998) Genetic polymorphisms and alternative splicing of the hOGGl gene, that is involved in the repair of 8- hydroxyguanine in damaged DNA. Oncogene, 16, 3219-3225. Kopp, E. and Ghosh, S. (1994) Inhibition of NF-kappa B by sodium salicylate and aspirin. Science, 265, 956-959. Korinek, V., Barker, N., Morin, P.J., van Wichen, D., de Weger, R., Kinzler, K.W., Vogelstein, B. and Clevers, H. (1997) Constitutive transcriptional activation
212 by a beta-catenin-Tcf complex in APC-/- colon carcinoma [see comments]. Science, 275, 1784-1787. Kouri, M., Laasonen, A., Mecklin, J.P., Jarvinen, H., Franssila, K. and Pyrhonen, S. (1990) Diploid predominance in hereditary nonpolyposis colorectal carcinoma evaluated by flow cytometry. Cancer, 65,1825-1829. Kramps, T., Peter, O., Brunner, E., Nellen, D., Froesch, B., Chattel]ee. S., Murone, M., Zullig, S. and Basler, K. (2002) Wnt/wingless signaling requires BCL9/legless-mediated recruitment of pygopus to the nuclear beta-catenin- TCF complex. Cell, 109, 47-60. Kraus, C. and Ballhausen, W.G. (1992) Two intragenic polymorphisms of the APC- gene detected by PCR and enzymatic digestion. Hum Genet, 88, 705-706. Laken, S.J., Petersen, G.M., Gruber, S.B., Oddoux, C., Ostrer, H., Giardiello, F.M., Hamilton, S.R., Hampel, H., Markowitz, A., Klimstra, D., Jhanwar, S., Winawer, S., Offit, K., Luce, M.C., Kinzler, K.W. and Vogelstein, B. (1997) Familial colorectal cancer in Ashkenazim due to a hypermutable tract in APC. Nat Genet, 17, 79-83. Lamlum, H., A1 Tassan, N., Jaeger, E., Frayling, I., Sieber, O., Reza, F.B., Eckert, M., Rowan, A., Barclay, E., Atkin, W., Williams, C., Gilbert, J., Cheadle, J., Bell, J., Houlston, R., Bodmer, W., Sampson, J. and Tomlinson, I. (2000a) Germline APC variants in patients with multiple colorectal adenomas, with evidence for the particular importance of El 317Q. Hum Mol Genet, 9,2215-
2221. Lamlum, H., Ilyas, M., Rowan, A., Clark, S., Johnson, V., Bell, J., Frayling, I., Efstathiou, J., Pack, K., Payne, S., Roylance, R., Gorman, P., Sheer, D., Neale, K., Phillips, R., Talbot, I., Bodmer, W. and Tomlinson, I. (1999) The type of somatic mutation at APC in familial adenomatous polyposis is determined by the site of the germline mutation: a new facet to Knudson's ’two-hit' hypothesis. Nat Med, 5, 1071-1075. Lamlum, H., Papadopoulou, A., Ilyas, M., Rowan, A., Gillet, C., Hanby, A., Talbot, I., Bodmer, W. and Tomlinson, I. (2000b) APC mutations are sufficient for the growth of early colorectal adenomas. Proc Natl Acad Sci USA, 97, 2225- 2228.
213 Law, D.J., Olschwang, S., Monpezat, J.P., Lefrancois, D., Jagelman, D., Petrelli, N.J., Thomas, G. and Feinberg, A.P. (1988) Concerted nonsyntenic allelic loss in human colorectal carcinoma. Science, 241, 961-965. Leach, F.S., Nicolaides, N.C., Papadopoulos, N., Liu, B., Jen, J., Parsons, R., Peltomaki, P., Sistonen, P., Aaltonen, L.A., Nystrom-Lahti, M. and et al. (1993) Mutations of a mutS homolog in hereditary nonpolyposis colorectal cancer. Cell, 75,1215-1225. Levin, B. (1992) Inflammatory bowel disease and colon cancer. Cancer, 70,1313- 1316. Levy, D.B., Smith, K.J., Beazer-Barclay, Y., Hamilton, S.R., Vogelstein, B. and Kinzler, K.W. (1994) Inactivation of both APC alleles in human and mouse tumors. Cancer Res, 54, 5953-5958. Li, C.Y., Suardet, L. and Little, J.B. (1995) Potential role of WAFl/Cipl/p21 as a mediator of TGF-beta cytoinhibitory effect. JBiol Chem, 270,4971-4974. Liaw, D., Marsh, D.J., Li, J., Dahia, P.L., Wang, S.I., Zheng, Z., Bose, S., Call, K.M., Tsou, H.C., Peacocke, M., Eng, C. and Parsons, R. (1997) Germline mutations of the PTEN gene in Cowden disease, an inherited breast and thyroid cancer syndrome. Nat Genet, 16, 64-67. Lindor, N.M., Burgart, L.J., Leontovich, O., Goldberg, R.M., Cunningham, J.M., Sargent, D.J., Walsh-Vockley, C., Petersen, G.M., Walsh, M.D., Leggett, B.A., Young, J.P., Barker, M.A., Jass, J.R., Hopper, J., Gallinger, S., Bapat, B., Redston, M. and Thibodeau, S.N. (2002) Immunohistochemistry versus microsatellite instability testing in phenotyping colorectal tumors. J Clin Oncol, 20, 1043-1048. Liu, C., Li, Y., Semenov, M., Han, C., Baeg, G.H., Tan, Y., Zhang, Z., Lin, X. and He, X. (2002) Control of beta-catenin phosphorylation/degradation by a dual kinase mechanism. Cell, 108, 837-847. Liu, F., Pouponnot, C. and Massague, J. (1997a) Dual role of the Smad4/DPC4 tumor suppressor in TGFbeta-inducible transcriptional complexes. Genes Dev, 11,3157-3167. Liu, W., Dong, X., Mai, M., Seelan, R.S., Taniguchi, K., Krishnadath, K.K., Hailing, K.C., Cunningham, J.M., Boardman, L.A., Qian, C., Christensen, E., Schmidt, S.S., Roche, P.C., Smith, D.I. and Thibodeau, S.N. (2000)
214 Mutations in AXIN2 cause colorectal cancer with defective mismatch repair by activating beta-catenin/TCF signalling. Nat Genet, 26,146-147. Liu, X., Wu, H., Loring, J., Hormuzdi, S., Disteche, C.M., Bomstein, P. and Jaenisch, R. (1997b) Trisomy eight in ES cells is a common potential problem in gene targeting and interferes with germ line transmission. Dev Dyn, 209, 85-91. Longo, L., Bygrave, A., Grosveld, E.G. and Pandolfi, P.P. (1997) The chromosome make-up of mouse embryonic stem cells is predictive of somatic and germ cell chimaerism. Transgenic Res, 6, 321-328. Lothe, R.A., Andersen, S.N., Hofstad, B., Meling, G.I., Peltomaki, P., Heim, S., Brogger, A., Vatn, M., Rognum, T.O. and Borresen, A.L. (1995) Deletion of Ip loci and microsatellite instability in colorectal polyps. Genes Chromosomes Cancer, 14, 182-188. Loukola, A., Bklin, K., Laiho, P., Salovaara, R., Kristo, P., Jarvinen, H., Mecklin, J.P., Launonen, V. and Aaltonen, L.A. (2001) Microsatellite marker analysis in screening for hereditary nonpolyposis colorectal cancer (HNPCC). Cancer Res, 61,4545-4549. Lovett, B. (1976) Family studies in cancer of the colon and rectum. B rJ Surg, 63, 13-18. Lu, R., Nash, H.M. and Verdine, G.L. (1997) A mammalian DNA repair enzyme that excises oxidatively damaged guanines maps to a locus frequently lost in lung cancer. Curr Biol, 7, 397-407. Lustig, B., Jerchow, B., Sachs, M., Weiler, S., Pietsch, T., Karsten, U., van de Wetering, M., Clevers, H., Schlag, P.M., Birchmeier, W. and Behrens, J. (2002) Negative feedback loop of Wnt signaling through upregulation of conductin/axin2 in colorectal and liver tumors. Mol Cell Biol, 22,1184-1193. Lynch, B.D., Ostermeyer, B.A., Lee, M.K., Arena, J.F., Ji, H., Dann, J., Swisshelm, K., Suchard, D., MacLeod, P.M., Kvinnsland, S., Gjertsen, B.T., Heimdal, K., Lubs, H., Moller, P. and King, M.C. (1997) Inherited mutations in PTBN that are associated with breast cancer, cowden disease, and juvenile polyposis, y //mw Genet, 61,1254-1260. Lynch, H.T. and de la Chapelle, A. (1999) Genetic susceptibility to non-polyposis colorectal cancer. JM ed Genet, 36, 801-818.
215 Lynch, H.T. and de la Chapelle, A. (2003) Hereditary colorectal cancer. NEngl J Med, 348, 919-932. Lynch, H.T. and Fusaro, R.M. (1999) The Muir-Torre syndrome in kindreds with hereditary nonpolyposis colorectal cancer (Lynch syndrome): A classic obligation in preventive medicine. J Am Acad Dermatol^ 41, 797-799. Lynch, H.T., Lanspa, S., Smyrk, T., Roman, B., Watson, P. and Lynch, J. (1991) Hereditary nonpolyposis colorectal cancer (Lynch syndromes I & II). Genetics, pathology, natural history, and cancer control. Part I. Cancer Genet Cytogenet, 53, 143-160. Lynch, H.T., Lanspa, S.J., Roman, R.M., Smyrk, T., Watson, P., Lynch, J.F., Lynch, P.M., Cristofaro, G., Rufo, P., Tauro, A.V. and et al. (1988) Hereditary nonpolyposis colorectal cancer—Lynch syndromes I and II. Gastroenterol Clin North Am, 17, 679-712. Lynch, H.T. and Lynch, J.F. (1998) Genetics of colonic cancer. Digestion, 59,481- 492. Lynch, H.T., Smyrk, T.C., Watson, P., Lanspa, S.J., Lynch, J.F., Lynch, P.M., Cavalieri, R.J. and Roland, C.R. (1993) Genetics, natural history, tumor spectrum, and pathology of hereditary nonpolyposis colorectal cancer: an updated review. Gastroenterology, 104,1535-1549. Malins, D.C. and Haimanot, R. (1991) Major alterations in the nucleotide structure of DNA in cancer of the female breast. Cancer Res, 51, 5430-5432. Mandl, M., Caspari, R., Jauch, A., Roker, T., Raschke, H., Sengteller, M., Propping, P. and Friedl, W. (1996) Familial adenomatous polyposis: a submicroscopic deletion at the APC locus in a family with mentally normal patients. Hum Genet, 97, 204-208. Mao, J., Wang, J., Liu, R., Pan, W., Farr, G.H., 3rd, Flynn, C., Yuan, H., Takada, S., Kimelman, D., Li, L. and Wu, D. (2001) Low-density lipoprotein receptor- related protein-5 binds to Axin and regulates the canonical Wnt signaling pathway. Mol Cell, 1, 801-809. Mamett, L.J. and Kalgutkar, A.S. (1999) Cyclooxygenase 2 inhibitors: discovery, selectivity and the future. Trends Pharmacol Sci, 20,465-469. Marsh, D.J., Dahia, P.L., Zheng, Z., Liaw, D., Parsons, R., Gorlin, R.J. and Eng, C. (1997) Germline mutations in PTEN are present in Rannayan-Zonana syndrome. Nat Genet, 16, 333-334.
216 Marsh, D.J., Kum, J.B., Lunetta, K.L., Bennett, M.J., Gorlin, R.J., Ahmed, S.F., Bodurtha, J., Crowe, C., Curtis, M.A., Dasouki, M., Dunn, T., Feit, H., Geraghty, M.T., Graham, J.M., Jr., Hodgson, S.V., Hunter, A., Korf, B.R., Manchester, D., Miesfeldt, S., Murday, V.A., Nathanson, K.L., Parisi, M., Pober, B., Romano, C., Eng, C. and et al. (1999) PTEN mutation spectrum and genotype-phenotype correlations in Bannayan-Riley-Ruvalcaba syndrome suggest a single entity with Cowden syndrome. Hum Mol Genet, 8, 1461-1472. Massague, J. (2000) How cells read TGF-beta signals. Nat Rev Mol Cell Biol, 1,169- 178. Matsumine, A., Ogai, A., Senda, T., Okumura, N., Satoh, K., Baeg, G.H., Kawahara, T., Kobayashi, S., Okada, M., Toyoshima, K. and Akiyama, T. (1996) Binding of APC to the human homolog of the Drosophila discs large tumor suppressor protein [see comments]. Science, 272, 1020-1023. McLennan, A.G. (1999) The MutT motif family of nucleotide phosphohydrolases in man and human pathogens (review). Int J Mol Med, 4, 79-89. McMichael, A.J. and Giles, G.G. (1988) Cancer in migrants to Australia: extending the descriptive epidemiological data. Cancer Res, 48, 751-756. Mecklin, J.P. and Jarvinen, H.J. (1991) Tumor spectrum in cancer family syndrome (hereditary nonpolyposis colorectal cancer). Cancer, 68, 1109-1112. Mecklin, J.P., Jarvinen, H.J. and Peltokallio, P. (1986) Cancer family syndrome. Genetic analysis of 22 Finnish kindreds. Gastroenterology, 90, 328-333. Michaels, M.L. and Miller, J.H. (1992) The GO system protects organisms from the mutagenic effect of the spontaneous lesion 8-hydroxyguanine (7,8-dihydro-8- oxoguanine). yRflcterio/, 174, 6321-6325. Midgley, C.A., White, S., Howitt, R., Save, V., Dunlop, M.G., Hall, P.A., Lane, D.P., Wyllie, A.H. and Bubb, V.J. (1997) APC expression in normal human tissues. J Pathol, 181,426-433. Miyaki, M., Konishi, M., Tanaka, K., Kikuchi-Yanoshita, R., Muraoka, M., Yasuno, M., Igari, T., Koike, M., Chiba, M. and Mori, T. (1997) Germline mutation of MSH6 as the cause of hereditary nonpolyposis colorectal cancer. Nat Genet, 17, 271-272. Miyoshi, Y., Ando, H., Nagase, H., Nishisho, I., Horii, A., Miki, Y., Mori, T., Utsunomiya, J., Baba, S., Petersen, G. and et al. (1992a) Germ-line mutations
217 of the APC gene in 53 familial adenomatous polyposis patients. Proc Natl Acad Sci USA, 89, 4452-4456. Miyoshi, Y., Nagase, H., Ando, H., Horii, A., Ichii, S., Nakatsuru, S., Aoki, T., Miki, Y., Mori, T. and Nakamura, Y. (1992b) Somatic mutations of the APC gene in colorectal tumors: mutation cluster region in the APC gene. Hum Mol Genet, 1,229-233. Mo, J.Y., Maki, H. and Sekiguchi, M. (1992) Hydrolytic elimination of a mutagenic nucleotide, 8-oxodGTP, by human 18-kilodalton protein: sanitization of nucleotide pool. Proc Natl Acad Sci US A, S9,11021-11025. Morin, P.J., Sparks, A.B., Korinek, V., Barker, N., Clevers, H., Vogelstein, B. and Kinzler, K.W. (1997) Activation of beta-catenin-Tcf signaling in colon cancer by mutations in beta-catenin or APC [see comments]. Science, 275, 1787-1790. Munemitsu, S., Albert, I., Souza, B., Rubinfeld, B. and Polakis, P. (1995) Regulation of intracellular beta-catenin levels by the adenomatous polyposis coli (APC) tumor-suppressor protein. Proc Natl Acad Sci USA, 92, 3046-3050. Munemitsu, S., Souza, B., Muller, O., Albert, I., Rubinfeld, B. and Polakis, P. (1994) The APC gene product associates with microtubules in vivo and promotes their assembly in vitro. Cancer Res, 54, 3676-3681. Nagase, H. and Nakamura, Y. (1993) Mutations of the APC (adenomatous polyposis coli) gene. Hum Mutat, 2,425-434. Nakagawa, H., Koyama, K., Monden, M. and Nakamura, Y. (1999) Analysis of APCL, a brain-specific adenomatous polyposis coli homologue, for mutations and expression in brain tumors. Jpn J Cancer Res, 90, 982-986. Nakagawa, H., Koyama, K., Murata, Y., Morito, M., Akiyama, T. and Nakamura, Y. (2000) APCL, a central nervous system-specific homologue of adenomatous polyposis coli tumor suppressor, binds to p53-binding protein 2 and translocates it to the perinucleus. Cancer Res, 60, 101-105. Nakagawa, H., Murata, Y., Koyama, K., Fujiyama, A., Miyoshi, Y., Monden, M., Akiyama, T. and Nakamura, Y. (1998) Identification of a brain-specific APC homologue, APCL, and its interaction with beta-catenin. Cancer Res, 58, 5176-5181. Nakamura, Y., Nishisho, I., Kinzler, K.W., Vogelstein, B., Miyoshi, Y., Miki, Y., Ando, H. and Horii, A. (1992) Mutations of the APC (adenomatous polyposis 218 coli) gene in FAP (familial polyposis coli) patients and in sporadic colorectal tumors. Tohoku J Exp Med, 168,141-147. Nakao, A., Imamura, T., Souchelnytskyi, S., Kawabata, M., Ishisaki, A., Oeda, E., Tamaki, K., Hanai, J., Heldin, C.H., Miyazono, K. and ten Dijke, P. (1997) TGF-beta receptor-mediated signalling through Smad2, Smad3 and Smad4. Embo J, 16, 5353-5362. Narita, T., Eto, T. and Ito, T. (1987) Peutz-Jeghers syndrome with adenomas and adenocarcinomas in colonic polyps. Am J Surg Pathol, 11, 76-81. Nathke, I.S., Adams, C.L., Polakis, P., Sellin, J.H. and Nelson, W.J. (1996) The adenomatous polyposis coli tumor suppressor protein localizes to plasma membrane sites involved in active cell migration. J Cell Biol, 134,165-179. Nghiem, Y., Cabrera, M., Guppies, C.G. and Miller, J.H. (1988) The mutY gene: a mutator locus in Escherichia coli that generates G.G T.A transversions. Proc Natl Acad Sci USA, 85, 2709-2713. Nicholls, R.J., Springall, R.G. and Gallagher, P. (1988) Regression of rectal adenomas after colectomy and ileorectal anastomosis for familial adenomatous polyposis. Br Med J (Clin Res Ed), 296, 1707-1708. Nicolaides, N.G., Papadopoulos, N., Liu, B., Wei, Y.F., Garter, K.G., Ruben, S.M., Rosen, G.A., Haseltine, W.A., Fleischmann, R.D., Fraser, G.M. and et al. (1994) Mutations of two PMS homologues in hereditary nonpolyposis colon cancer. Nature, 371, 75-80. Nishioka, K., Ohtsubo, T., Oda, H., Fujiwara, T., Kang, D., Sugimachi, K. and Nakabeppu, Y, (1999) Expression and differential intracellular localization of two major forms of human 8-oxoguanine DNA glycosylase encoded by alternatively spliced OGGI mRNAs. Mol Biol Cell, 10,1637-1652. Nugent, K.P., Farmer, K.G., Spigelman, A.D., Williams, G.B. and Phillips, R.K. (1993) Randomized controlled trial of the effect of sulindac on duodenal and rectal polyposis and cell proliferation in patients with familial adenomatous polyposis. B rJ Surg, 80, 1618-1619. Nugent, K.P., Phillips, R.K., Hodgson, S.V., Cottrell, S., Smith-Ravin, J., Pack, K. and Bodmer, W.F. (1994) Phenotypic expression in familial adenomatous polyposis: partial prediction by mutation analysis. Gut, 35,1622-1623. O'Brien, M.J., Winawer, S.J., Zauber, A.G., Gottlieb, L.S., Sternberg, S.S., Diaz, B., Dickersin, G.R., Ewing, S., Geller, S., Kasimian, D. and et al. (1990) The
219 National Polyp Study. Patient and polyp characteristics associated with high- grade dysplasia in colorectal adenomas. Gastroenterology, 98, 371-379. Oberhuber, G. and Stolte, M. (2000) Gastric polyps: an update of their pathology and biological significance. Virchows Arch, 437, 581-590. Oda, H., Nakabeppu, Y., Furuichi, M. and Sekiguchi, M. (1997) Regulation of expression of the human MTHl gene encoding 8-oxo-dGTPase. Alternative splicing of transcription products. JBiol Chem, 272,17843-17850. Ohtsubo, T., Nishioka, K., Imaiso, Y., Iwai, S., Shimokawa, H., Oda, H., Fujiwara, T. and Nakabeppu, Y. (2000) Identification of human MutY homolog (hMYH) as a repair enzyme for 2- hydroxyadenine in DNA and detection of multiple forms of hMYH located in nuclei and mitochondria. Nucleic Acids Res, 28, 1355-1364. Okamoto, K., Toyokuni, S., Uchida, K., Ogawa, O., Takenewa, J., Kakehi, Y., Kinoshita, H., Hattori-Nakakuki, Y., Hiai, H. and Yoshida, O. (1994) Formation of 8-hydroxy-2'-deoxyguanosine and 4-hydroxy-2-nonenal- modified proteins in human renal-cell carcinoma. Int J Cancer, 58, 825-829. Olinski, R., Zastawny, T., Budzbon, J., Skokowski, J., Zegarski, W. and Dizdaroglu, M. (1992) DNA base modifications in chromatin of human cancerous tissues. FEES Lett, 309, 193-198. Olschwang, S., Serova-Sinilnikova, O.M., Lenoir, G.M. and Thomas, G. (1998) PTEN germ-line mutations in juvenile polyposis coli. Nat Genet, 18,12-14. Oshima, M., Dinchuk, I.E., Kargman, S.L., Oshima, H., Hancock, B., Kwong, E., Trzaskos, J.M., Evans, J.F. and Taketo, M.M. (1996) Suppression of intestinal polyposis in Ape delta716 knockout mice by inhibition of cyclooxygenase 2 (COX-2). Cell, 87, 803-809. Otori, K., Konishi, M., Sugiyama, K , Hasebe, T., Shimoda, T., Kikuchi-Yanoshita, R., Mukai, K., Fukushima, S., Miyaki, M. and Esumi, H. (1998) Infrequent somatic mutation of the adenomatous polyposis coli gene in aberrant crypt foci of human colon tissue. Cancer, 83, 896-900. Papadopoulos, N., Nicolaides, N.C., Wei, Y.F., Ruben, S.M., Carter, K.C., Rosen, C.A., Haseltine, W.A., Fleischmann, R.D., Fraser, C.M., Adams, M.D. and et al. (1994) Mutation of a mutL homolog in hereditary colon cancer. Science, 263, 1625-1629.
220 Paraf, F., Jothy, S. and Van Meir, E.G. (1997) Brain tumor-polyposis syndrome: two genetic diseases? J Clin Oncol, 15, 2744-2758. Parker, D.S., Jemison, J. and Cadigan, K.M. (2002) Pygopus, a nuclear PHD-fmger protein required for Wingless signaling in Drosophila. Development, 129, 2565-2576. Paul, P., Letteboer, T., Gelbert, L., Groden, J., White, R. and Coppes, M.J. (1993) Identical APC exon 15 mutations result in a variable phenotype in familial adenomatous polyposis. Hum Mol Genet, 2, 925-931. Phillips, R.K., Wallace, M.H., Lynch, P.M., Hawk, E., Gordon, G.B., Saunders, B.P., Wakabayashi, N., Shen, Y., Zimmerman, S., Godio, L., Rodrigues-Bigas, M., Su, L.K., Sherman, J., Kelloff, G., Levin, B. and Steinbach, G. (2002) A randomised, double blind, placebo controlled study of celecoxib, a selective cyclooxygenase 2 inhibitor, on duodenal polyposis in familial adenomatous polyposis. Gut, 50, 857-860. Pilarski, R.T., Brothman, A.R., Benn, P. and Shulman Rosengren, S. (1999) Attenuated familial adenomatous polyposis in a man with an interstitial deletion of chromosome arm 5q. Am JM ed Genet, 86, 321-324. Pinson, K.I., Brennan, J., Monkley, S., Avery, B.J. and Skames, W.C. (2000) An LDL-receptor-related protein mediates Wnt signalling in mice. Nature, 407, 535-538. Pinto, D., Gregorieff, A., Begthel, H. and Clevers, H. (2003) Canonical Wnt signals are essential for homeostasis of the intestinal epithelium. Genes Dev, 17, 1709-1713. Popat, S., Stone, J., Coleman, G., Marshall, G., Peto, J., Frayling, I. and Houlston, R. (2000) Prevalence of the APC E1317Q variant in colorectal cancer patients. Cancer Lett, 149,203-206. Potter, J.D. (1999) Colorectal cancer: molecules and populations. J Natl Cancer Inst, 91, 916-932. Powell, S.M., Zilz, N., Beazer-Barclay, Y., Bryan, T.M., Hamilton, S.R, Thibodeau, S.N., Vogelstein, B. and Kinzler, K.W. (1992) APC mutations occur early during colorectal tumorigenesis. Nature, 359, 235-237. Praml, C., Finke, L.H., Herfarth, C., Schlag, P., Schwab, M. and Amler, L. (1995) Deletion mapping defines different regions in lp34.2-pter that may harbor
221 genetic information related to human colorectal cancer. Oncogene^ 11,1357- 1362. Preston, S.L., Wong, W.M., Chan, A.O., Poulsom, R., Jeffery, R., Goodlad, R.A., Mandir, N., Elia, G., Novelli, M., Bodmer, W.F., Tomlinson, I.P. and Wright, N.A. (2003) Bottom-up histogenesis of colorectal adenomas: origin in the monocryptal adenoma and initial expansion by crypt fission. Cancer Res, 63, 3819-3825. Quinn, M., Babb, P., Brock, A., Kirby, L. and Jones, J. (2001) Cancer Trends in England and Wales. Office for National Statistics, London. Quirke, P., Fozard, J.B., Dixon, M.F., Dyson, J.B., Giles, G.R. and Bird, C.C. (1986) DNA aneuploidy in colorectal adenomas. Br J Cancer, 53,477-481. Radicella, J.P., Dherin, C., Desmaze, C., Fox, M.S. and Boiteux, S. (1997) Cloning and characterization of hOGGl, a human homolog of the OGGI gene of Saccharomyces cerevisiae. Proc Natl Acad Sci USA, 94, 8010-8015. Rajagopalan, H., Bardelli, A., Lengauer, C., Kinzler, K.W., Vogelstein, B. and Velculescu, V.E. (2002) Tumorigenesis: RAF/RAS oncogenes and mismatch- repair status. Nature, 418, 934. Ransohoff, D.F. and Lang, C.A. (1991) Screening for colorectal cancer. NEngl J Med, 325, 37-41. Reddy, B.S. (1999) Role of dietary fiber in colon cancer: an overview. Am JMed, 106,16S-19S; discussion 50S-51S. Ritenbaugh, C. (2000) Diet and prevention of colorectal cancer. Curr Oncol Rep, 2, 225-233. Rodrigues, N.R., Rowan, A., Smith, M.E., Kerr, LB., Bodmer, W.F., Gannon, J.V. and Lane, D P. (1990) p53 mutations in colorectal cancer. Proc Natl Acad Sci USA, 87, 7555-7559. Roldan-Aijona, T., Wei, Y.F., Carter, K.C., Klungland, A., Anselmino, C., Wang, R.P., Augustus, M. and Lindahl, T. (1997) Molecular cloning and functional expression of a human cDNA encoding the antimutator enzyme 8- hydroxyguanine-DNA glycosylase. Proc Natl Acad Sci C75*^, 94, 80lb- 8020. Rosin-Arbesfeld, R., Townsley, F. and Bienz, M. (2000) The APC tumour suppressor has a nuclear export function. Nature, 406, 1009-1012.
222 Rowan, A.J., Lamlum, H., Ilyas, M., Wheeler, J., Straub, J., Papadopoulou, A., Bicknell, D., Bodmer, W.F, and Tomlinson, I.P. (2000) APC mutations in sporadic colorectal tumors: A mutational "hotspot" and interdependence of the "two hits". Proc Natl Acad Sci USA, 97, 3352-3357. Roylance, R., Gorman, P., Harris, W., Liebmann, R., Barnes, D., Hanby, A. and Sheer, D. (1999) Comparative genomic hybridization of breast tumors stratified by histological grade reveals new insights into the biological progression of breast cancer. Cancer Res, 59,1433-1436. Rubinfeld, B., Albert, I., Porfiri, B., Fiol, C., Munemitsu, S. and Polakis, P. (1996) Binding of GSK3beta to the APC-beta-catenin complex and regulation of complex assembly [see comments]. Science, 272,1023-1026. Rubinfeld, B., Souza, B., Albert, I., Muller, O., Chamberlain, S.H., Masiarz, F.R., Munemitsu, S. and Polakis, P. (1993) Association of the APC gene product with beta-catenin. Science, 262, 1731-1734. Sakanaka, C., Weiss, J.B. and Williams, L.T. (1998) Bridging of beta-catenin and glycogen synthase kinase-3beta by axin and inhibition of beta-catenin- mediated transcription. Proc Natl Acad Sci USA, 95, 3020-3023. Sakumi, K., Furuichi, M., Tsuzuki, T., Kakuma, T., Kawabata, S., Maki, H. and Sekiguchi, M. (1993) Cloning and expression of cDNA for a human enzyme that hydrolyzes 8-oxo- dGTP, a mutagenic substrate for DNA synthesis. J Biol Chem, 268, 23524-23530. Salem, O.S. and Steck, W.D. (1983) Cowden's disease (multiple hamartoma and neoplasia syndrome). A case report and review of the English literature. J Am Acad Dermatol, 8, 686-696. Salovaara, R., Loukola, A., Kristo, P., Kaariainen, H., Ahtola, H., Eskelinen, M., Harkonen, N., Julkunen, R., Kangas, E., Ojala, S., Tulikoura, J., Valkamo, E., Jarvinen, H., Mecklin, J.P., Aaltonen, L.A. and de la Chapelle, A. (2000) Population-based molecular detection of hereditary nonpolyposis colorectal cancer. J Clin Oncol, 18, 2193-2200. Samowitz, W.S., Powers, M.D., Spirio, L.N., Nollet, F., van Roy, F. and Slattery, M.L. (1999) Beta-catenin mutations are more frequent in small colorectal adenomas than in larger adenomas and invasive carcinomas. Cancer Res, 59, 1442-1444.
223 Samowitz, W.S., Thliveris, A., Spirio, L.N. and White, R. (1995) Alternatively spliced adenomatous polyposis coli (APC) gene transcripts that delete exons mutated in attenuated APC. Cancer Res, 55, 3732-3734. Sanger, P., Nicklen, S. and Coulson, A.R. (1977) DNA sequencing with chain- terminating inhibitors. Proc Natl Acad Sci USA, 74, 5463-5467. Sankila, R., Aaltonen, L.A., Jarvinen, H.J. and Mecklin, J.P. (1996) Better survival rates in patients with MLR 1-associated hereditary colorectal cancer. Gastroenterology, 110, 682-687. Sayed, M.G., Ahmed, A.F., Ringold, J R., Anderson, M.E., Bair, J.L., Mitros, F.A., Lynch, H.T., Tinley, S.T., Petersen, G.M., Giardiello, P.M., Vogelstein, B. and Howe, J.R. (2002) Germline SMAD4 or BMPRl A mutations and phenotype of juvenile polyposis. Ann Surg Oncol, 9, 901-906. Scott-Conner, C.E., Hausmann, M., Hall, T.J., Skelton, D.S., Anglin, B.L. and Subramony, C. (1995) Familial juvenile polyposis: patterns of recurrence and implications for surgical management. JAm Coll Surg, 181, 407-413. Shibutani, S., Takeshita, M. and Grollman, A.P. (1991) Insertion of specific bases during DNA synthesis past the oxidation- damaged base 8-oxodG. Nature, 349, 431-434. Shih, I.M., Zhou, W., Goodman, S.N., Lengauer, C., Kinzler, K.W. and Vogelstein, B. (2001) Evidence that genetic instability occurs at an early stage of colorectal tumorigenesis. Cancer Res, 61, 818-822. Shinmura, K., Kasai, H., Sasaki, A., Sugimura, H. and Yokota, J. (1997) 8- hydroxyguanine (7,8-dihydro-8-oxoguanine) DNA glycosylase and AP lyase activities of hOGGl protein and their substrate specificity. Mutat Res, 385, 75-82. Sieber, O., Lipton, L., Heinimann, K. and Tomlinson, I. (2003a) Colorectal tumourigenesis in carriers of the APC I1307K variant: lone gunman or conspiracy? y 199, 137-139. Sieber, O.M., Heinimann, K. and Tomlinson, I.P. (2003b) Genomic instability—the engine of tumorigenesis? Nat Rev Cancer, 3, 701-708. Sieber, O.M., Tomlinson, I.P. and Lamlum, H. (2000) The adenomatous polyposis coli (APC) tumour suppressor—genetics, fimction and disease. Mol Med Today, 6, 462-469.
224 Slupska, M.M., Baikalov, C., Luther, W.M., Chiang, J.H., Wei, Y.F. and Miller, J.H. (1996) Cloning and sequencing a human homolog (hMYH) of the Escherichia coli mutY gene whose function is required for the repair of oxidative DNA damage, JBacteriol, 178, 3885-3892. Slupska, M.M., Luther, W.M., Chiang, J.H., Yang, H. and Miller, J.H. (1999) Functional expression of hMYH, a human homolog of the Escherichia coli MutY JBacteriol, 181, 6210-6213. Smilow, P.C., Pryor, C.A., Jr. and Swinton, N.W. (1966) Juvenile polyposis coli. A report of three patients in three generations of one family. Dis Colon Rectum, 9, 248-254. Smith, A.J., Stem, H.S., Penner, M., Hay, K., Mitri, A., Bapat, B.V. and Gallinger, S. (1994a) Somatic APC and K-ras codon 12 mutations in aberrant crypt foci from human colons. Cancer Res, 54, 5527-5530. Smith, K.J., Levy, D.B., Maupin, P., Pollard, T.D., Vogelstein, B. and Kinzler, K.W. (1994b) Wild-type but not mutant APC associates with the microtubule cytoskeleton. Cancer Res, 54, 3672-3675. Smith, W.L., DeWitt, D.L. and Garavito, R.M. (2000) Cyclooxygenases: structural, cellular, and molecular biology. Annu Rev Biochem, 69, 145-182. Soravia, C., Berk, T., Madlensky, L., Mitri, A., Cheng, H., Gallinger, S., Cohen, Z. and Bapat, B. (1998) Genotype-phenotype correlations in attenuated adenomatous polyposis coli. Am JHum Genet, 62, 1290-1301. Souchelnytskyi, S., Tamaki, K., Engstrom, U., Wemstedt, C., ten Dijke, P. and Heldin, C.H. (1997) Phosphorylation of Ser465 and Ser467 in the C terminus of Smad2 mediates interaction with Smad4 and is required for transforming growth factor-beta signaling.JBiol Chem, 111, 28107-28115. Sparks, A.B., Morin, P.J., Vogelstein, B. and Kinzler, K.W. (1998) Mutational analysis of the APC/beta-catenin/Tcf pathway in colorectal cancer. Cancer Res, 58, 1130-1134. Spirio, L., Olschwang, S., Groden, J., Robertson, M., Samowitz, W., Joslyn, G., Gelbert, L., Thliveris, A., Carlson, M., Ottemd, B. and et al. (1993) Alleles of the APC gene: an attenuated form of familial polyposis. Cell, 75, 951-957. Spirio, L.N., Samowitz, W., Robertson, J., Robertson, M., Burt, R.W., Leppert, M. and White, R. (1998) Alleles of APC modulate the frequency and classes of mutations that lead to colon polyps. Nat Genet, 20, 385-388.
225 St John, D.J., McDermott, F.T., Hopper, J.L., Debney, E.A., Johnson, W.R. and Hughes, E.S. (1993) Cancer risk in relatives of patients with common colorectal cancer. Ann Intern Med, 118, 785-790. Starink, T.M., Meijer, C.J. and Brownstein, M.H. (1985) The cutaneous pathology of Cowden's disease: new findings. / Pathol, 12, 83-93. Starink, T.M., van der Veen, J.P., Arwert, F., de Waal, L.P., de Lange, G.G., Gille, J.J. and Eriksson, A.W. (1986) The Cowden syndrome: a clinical and genetic study in 21 patients. Clin Genet, 29,222-233. Steinbach, G., Lynch, P.M., Phillips, R.K., Wallace, M.H., Hawk, E., Gordon, G.E., Wakabayashi, N., Saunders, B., Shen, Y., Fujimura, T., Su, L.K. and Levin, B. (2000) The effect of celecoxib, a cyclooxygenase-2 inhibitor, in familial adenomatous polyposis. N EnglJ Med, 342,1946-1952. Su, L.K., Barnes, C.J., Yao, W., Qi, Y., Lynch, P.M. and Steinbach, G. (2000a) Inactivation of germline mutant APC alleles by attenuated somatic mutations: a molecular genetic mechanism for attenuated familial adenomatous polyposis. Am JHum Genet, 67, 582-590. Su, L.K., Burrell, M., Hill, D.E., Gyuris, J., Brent, R., Wiltshire, R., Trent, J., Vogelstein, B. and Kinzler, K.W. (1995) APC binds to the novel protein EBl. Cancer Res, 55, 2972-2977. Su, L.K., Kohlmann, W., Ward, P.A. and Lynch, P.M. (2002) Different familial adenomatous polyposis phenotypes resulting from deletions of the entire APC exon 15. Hum Genet, 111, 88-95. Su, L.K., Steinbach, G., Sawyer, J.C., Hindi, M., Ward, P.A. and Lynch, P.M. (2000b) Genomic rearrangements of the APC tumor-suppressor gene in familial adenomatous polyposis. Hum Genet, 106,101-107. Su, L.K., Vogelstein, B. and Kinzler, K.W. (1993) Association of the APC tumor suppressor protein with catenins. Science, 262, 1734-1737. Sugimoto, Y., Narumiya, S. and Ichikawa, A. (2000) Distribution and function of prostanoid receptors: studies from knockout mice. Prog Lipid Res, 39, 289- 314. Sugita, A., Sachar, D.B., Bodian, C., Ribeiro, M.B., Aufses, A.H., Jr. and Greenstein, A.J. (1991) Colorectal cancer in ulcerative colitis. Influence of anatomical extent and age at onset on colitis-cancer interval. Gut, 32, 167-169.
226 Sulekova, Z. and Ballhausen, W.G. (1995) A novel coding exon of the human adenomatous polyposis coli gene. Hum Genet, 96,469-471. Suzuki, S., Mizuno, M., Tomoda, J., Ohmori, M. and Tsuji, T. (1995) Flow cytometric analysis of the DNA content in colorectal adenomas with focal cancers. Gastroenterology, 109, 1098-1104. Takao, M., Zhang, Q.M., Yonei, S. and Yasui, A. (1999) Differential subcellular localization of human MutY homolog (hMYH) and the fimctional activity of adenine:8-oxoguanine DNA glycosylase. Nucleic Acids Res, 27, 3638-3644. Tamai, K., Semenov, M., Kato, Y., Spokony, R., Liu, C., Katsuyama, Y., Hess, F., Saint-Jeannet, J.P. and He, X. (2000) LDL-receptor-related proteins in Wnt signal transduction. Nature, 407, 530-535. Tanaka, K., Yanoshita, R., Konishi, M., Oshimura, M., Maeda, Y., Mori, T. and Miyaki, M. (1993) Suppression of tumourigenicity in human colon carcinoma cells by introduction of normal chromosome lp36 region. Oncogene, 8, 2253-2258. Tetsu, O. and McCormick, F. (1999) Beta-catenin regulates expression of cyclin D1 in colon carcinoma cells. Nature, 398, 422-426. Thiagalingam, S., Lengauer, C., Leach, F.S., Schutte, M., Hahn, S.A., Overhauser, J., Willson, J.K., Markowitz, S., Hamilton, S R., Kern, S.E., Kinzler, K.W. and Vogelstein, B. (1996) Evaluation of candidate tumour suppressor genes on chromosome 18 in colorectal cancers. Nat Genet, 13, 343-346. Thibodeau, S.N., Bren, G. and Schaid, D. (1993) Microsatellite instability in cancer of the proximal colon. Science, 260, 816-819. Thomas, D., Scot, A.D., Barbey, R., Padula, M. and Boiteux, S. (1997) Inactivation of OGGI increases the incidence of G . C ->T . A transversions in Saccharomyces cerevisiae: evidence for endogenous oxidative damage to DNA in eukaryotic cells. Mol Gen Genet, 254, 171-178. Thompson, B., Townsley, F., Rosin-Arbesfeld, R., Musisi, H. and Bienz, M. (2002) A new nuclear component of the Wnt signalling pathway. Nat Cell Biol, 4, 367-373. Thun, M.J., Namboodiri, M.M. and Heath, C.W., Jr. (1991) Aspirin use and reduced risk of fatal colon csmcer. N EnglJ Med, 325, 1593-1596. Tomlinson, L, Rahman, N., Frayling, I., Mangion, J., Barfoot, R., Hamoudi, R , Seal, S., Northover, J., Thomas, H.J., Neale, K., Hodgson, S., Talbot, I., Houlston,
227 R. and Stratton, M.R. (1999) Inherited susceptibility to colorectal adenomas and carcinomas: evidence for a new predisposition gene on 15ql4-q22. Gastroenterology, 116, 789-795. Tsukazaki, T., Chiang, T.A., Davison, A.F., Attisano, L. and Wrana, J.L. (1998) SARA, a FYVE domain protein that recruits Smad2 to the TGFbeta receptor. Cell, 95, 779-791. Utsunomiya, J., Gocho, H., Miyanaga, T., Hamaguchi, E. and Kashimure, A. (1975) Peutz-Jeghers syndrome: its natural course and management. Johns Hopkins Mec/y, 136, 71-82. Van de Wetering, M., Castrop, J., Korinek, V. and Clevers, H. (1996) Extensive alternative splicing and dual promoter usage generate Tcf-1 protein isoforms with differential transcription control properties. Mol Cell Biol, 16, 745-752. van den Ingh, H.F., Griffioen, G. and Comelisse, C.J. (1985) Flow cytometric detection of aneuploidy in colorectal adenomas. Cancer Res, 45, 3392-3397. van der Luijt, R.B., Meera Khan, P., Vasen, H.F., Breukel, C., Tops, C.M., Scott, R.J. and Fodde, R. (1996) Germline mutations in the 3' part of APC exon 15 do not result in truncated proteins and are associated with attenuated adenomatous polyposis coli. Hum Genet, 98, 727-734. van der Luijt, R.B., Vasen, H.F., Tops, C.M., Breukel, C., Fodde, R. and Meera Khan, P. (1995) APC mutation in the alternatively spliced region of exon 9 associated with late onset familial adenomatous polyposis. Hum Genet, 96, 705-710. van Es, J.H., Kirkpatrick, C., van de Wetering, M., Molenaar, M., Miles, A., Kuipers, J., Destree, O., Peifer, M. and Clevers, H. (1999) Identification of APC2, a homologue of the adenomatous polyposis coli tumour suppressor. Curr Biol, 9,105-108. Van Meir, E.G. (1998) "Turcot's syndrome": phenotype of brain tumors, survival and mode of inheritance. Int J Cancer, 75, 162-164. Vasen, H.F., Griffioen, G., Tops, C.M., van Haeringen, A., Nagengast, F.M., van Leeuwen-Comelisse, I.S., Khan, P.M. and van Slooten, E.A. (1993) Familial adenomatous polyposis and its clinical surveillance. Neth JMed, 42, 105-108. Vasen, H.F., Mecklin, J.P., Khan, P.M. and Lynch, H.T. (1991) The International Collaborative Group on Hereditary Non-Polyposis Colorectal Cancer (ICG- HNPCC). Dis Colon Rectum, 34,424-425.
228 Vasen, H.F., Offerhaus, G.J., den Hartog Jager, F.C., Menko, F.H., Nagengast, F.M., Griffioen, G,, van Hogezand, R.B. and Heintz, A.P. (1990) The tumeur spectrum in hereditary non-polyposis colorectal cancer: a study of 24 kindreds in the Netherlands. Int J Cancer, 46, 31-34. Vasen, H.F., Watson, P., Mecklin, J.P. and Lynch, H.T. (1999) New clinical criteria for hereditary nonpolyposis colorectal cancer (HNPCC, Lynch syndrome) proposed by the International Collaborative group on HNPCC. Gastroenterology, 116, 1453-1456. Veale, A.M., McColl, I., Bussey, H.J. and Morson, B.C. (1966) Juvenile polyposis coli. JM ed Genet, 3, 5-16. Vogelstein, B., Fearon, E.R., Hamilton, S.R., Kern, S.B., Preisinger, A.C., Leppert, M., Nakamura, Y., White, R., Smits, A.M. and Bos, J.L. (1988) Genetic alterations during colorectal-tumor development. NEngl JMed, 319, 525- 532. Vogelstein, B., Fearon, E.R., Kern, S.E., Hamilton, S.R., Preisinger, A.C., Nakamura, Y. and White, R. (1989) Allelotype of colorectal carcinomas. Science, 244, 207-211. Waddell, W.R. and Loughry, R.W. (1983) Sulindac for polyposis of the colon. J Surg Oncol, 24, 83-87. Wagner, A., Hendriks, Y., Meijers-Heijboer, E.J., de Leeuw, W.J., Morreau, H., Hofstra, R., Tops, C., Bik, E., Brocker-Vriends, A.H., van Der Meer, C., Lindhout, D., Vasen, H.F., Breuning, M.H., Comelisse, C.J., van Krimpen, C., Niermeijer, M.F., Zwinderman, A.H., Wijnen, J. and Fodde, R. (2001) Atypical HNPCC owing to MSH6 germline mutations: analysis of a large Dutch pedigree. JM ed Genet, 38, 318-322. Waite, K.A. and Eng, C. (2002) Protean PTEN: form and function. Am JHum Genet, 70, 829-844. Wallis, Y.L., Morton, D.G., McKeown, C.M. and Macdonald, F. (1999) Molecular analysis of the APC gene in 205 families: extended genotype-phenotype correlations in FAP and evidence for the role of APC amino acid changes in colorectal cancer predisposition. JM ed Genet, 36, 14-20. Wang, D., Kreutzer, D A. and Essigmann, J.M. (1998) Mutagenicity and repair of oxidative DNA damage: insights from studies using defined lesions. Mutat Res, 400, 99-115.
229 Wasan, H.S., Park, H.S., Liu, K.C., Mandir, N.K., Winnett, A., Sasieni, P., Bodmer, W.F., Goodlad, R.A. and Wright, N.A. (1998) APC in the regulation of intestinal crypt fission. J Pathol, 185, 246-255. Watson, P., Vasen, H.F., Mecklin, J.P., Jarvinen, H. and Lynch, H.T. (1994) The risk of endometrial cancer in hereditary nonpolyposis colorectal cancer. Am J Med, 96, 5X6-510. Wehrli, M., Dougan, S.T., Caldwell, K., O'Keefe, L., Schwartz, S., Vaizel-Ohayon, D., Schejter, E., Tomlinson, A. and DiNardo, S. (2000) arrow encodes an LDL-receptor-related protein essential for Wingless signalling. Nature, 407, 527-530. Weiss, H., Wildner, G.P., Jacobasch, K.H., Heinz, U. and Schaelicke, W. (1985) Characterization of human adenomatous polyps of the colorectal bowel by means of DNA distribution patterns. Oncology, 42, 33-41. White, S., Bubb, V.J. and Wyllie, A.H. (1996) Germline APC mutation (Glnl317) in a cancer-prone family that does not result in familial adenomatous polyposis. Genes Chromosomes Cancer, 15, 122-128. Whitelaw, S.C., Murday, V.A., Tomlinson, I P., Thomas, H.J., Cottrell, S., Ginsberg, A., Bukofzer, S., Hodgson, S.V., Skudowitz, R.B., Jass, J R., Talbot, I.C., Northover, J.M., Bodmer, W.F. and Solomon, E. (1997) Clinical and molecular features of the hereditary mixed polyposis syndrome. Gastroenterology, 112, 327-334. Wijnen, J., van der Klift, H., Vasen, H., Khan, P.M., Menko, F., Tops, C., Meijers Heijboer, H., Lindhout, D., Moller, P. and Fodde, R. (1998) MSH2 genomic deletions are a frequent cause of HNPCC [letter]. Nat Genet, 20, 326-328. Wikman, H., Risch, A., Klimek, F., Schmezer, P., Spiegelhalder, B., Dienemann, H., Kayser, K., Schulz, V., Drings, P. and Bartsch, H. (2000) hOGGl polymorphism and loss of heterozygosity (LOH): significance for lung cancer susceptibility in a Caucasian population. Int J Cancer, 88, 932-937. Wolber, R.A. and Owen, D A. (1991) Flat adenomas of the colon. Hum Pathol, 22, 70-74. Wong, W.M., Garcia, S B. and Wright, N.A. (1999) Origins and morphogenesis of colorectal neoplasms. Apmis, 107, 535-544. Woodage, T., King, S.M., Wacholder, S., Hartge, P., Struewing, J.P., McAdams, M., Laken, S.J., Tucker, M.A. and Brody, L.C. (1998) The APCI1307K allele and
230 cancer risk in a community-based study of Ashkenazi Jews. Nat Genet, 20, 62-65. Wrana, J.L., Attisano, L., Carcamo, J., Zentella, A., Doody, J., Laiho, M., Wang, X.F. and Massague, J. (1992) TGF beta signals through a heteromeric protein kinase receptor complex. Cell, 71, 1003-1014. Wrana, J.L., Attisano, L., Wieser, R., Ventura, F. and Massague, J. (1994) Mechanism of activation of the TGF-beta receptor. Nature, 370, 341-347. Wu, Y., Berends, M.J., Sijmons, R.Fl., Mensink, R.G., Verlind, E., Kooi, K.A., van der Sluis, T., Kempinga, C., van dDer Zee, A.G., Hollema, H., Buys, C.H., Kleibeuker, J.H. and Hofstra, R.M. (2001) A role for MLH3 in hereditary nonpolyposis colorectal cancer. Nat Genet, 29, 137-138. Xu, L., Alarcon, C., Col, S. and Massague, J. (2003) Distinct domain utilization by Smad3 and Smad4 for nucleoporin interaction and nuclear import. J Biol Chem. Xu, L., Chen, Y.G. and Massague, J. (2000) The nuclear import function of Smad2 is masked by SARA and unmasked by TGFbeta-dependent phosphorylation. Nat Cell Biol, 2, 559-562. Xu, L., Kang, Y., Col, S. and Massague, J. (2002) Smad2 nucleocytoplasmic shuttling by nucleoporins CAN/Nup214 and Nupl53 feeds TGFbeta signaling complexes in the cytoplasm and nucleus. Mol Cell, 10, 271-282. Yamamoto, H., Kishida, S., Uochi, T., Ikeda, S., Koyama, S., Asashima, M. and Kikuchi, A. (1998) Axil, a member of the Axin family, interacts with both glycogen synthase kinase 3beta and beta-catenin and inhibits axis formation of Xenopus embryos. Mol Cell Biol, 18, 2867-2875. Yamamoto, Y., Yin, M.J., Lin, K.M. and Gaynor, R.B. (1999) Sulindac inhibits activation of the NF-kappaB pathway. J Biol Chem, 274, 27307-27314. Yan, D., Wiesmann, M., Rohan, M., Chan, V., Jefferson, A.B., Guo, L., Sakamoto, D., Caothien, R.H., Fuller, J.H., Reinhard, C., Garcia, P.D., Randazzo, F.M., Escobedo, J., Fantl, W.J. and Williams, L.T. (2001) Elevated expression of axin2 and hnkd mRNA provides evidence that Wnt/beta -catenin signaling is activated in human colon tumors. Proc Natl Acad Sci USA, 98, 14973- 14978.
231 Yanagawa, S., van Leeuwen, F,, Wodarz, A., Klingensmith, J. and Nusse, R. (1995) The dishevelled protein is modified by wingless signaling in Drosophila. Genes Dev^ 9, 1087-1097. Yin, M.J., Yamamoto, Y. and Gaynor, R.B. (1998) The anti-inflammatory agents aspirin and salicylate inhibit the activity of I(kappa)B kinase-beta. Nature, 396, 77-80. Ylikorkala, A., Rossi, D.J., Korsisaari, N., Luukko, K., Alitalo, K., Henkemeyer, M. and Makela, T.P. (2001) Vascular abnormalities and deregulation of VEGF in Lkbl-deficient mice. Science, 293, 1323-1326. Yoo, L.I., Chung, D C. and Yuan, J. (2002) LKBl—a master tumour suppressor of the small intestine and beyond. Nat Rev Cancer, 2, 529-535. Yuen, S.T., Davies, H., Chan, T.L., Ho, J.W., Bignell, G.R., Cox, C., Stephens, P., Edkins, S., Tsui, W.W., Chan, A.S., Futreal, P.A., Stratton, M R., Wooster, R. and Leung, S.Y. (2002) Similarity of the phenotypic patterns associated with BRAF and KRAS mutations in colorectal neoplasia. Cancer Res, 62, 6451-6455. Zauber, N.P., Sabbath-Solitare, M., Marotta, S.P. and Bishop, D.T. (2003) The characterization of somatic APC mutations in colonic adenomas and carcinomas in Ashkenazi Jews with the APC I1307K variant using linkage disequilibrium, y PflZ/zo/, 199, 146-151. Zhou, S., Zawel, L., Lengauer, C., Kinzler, K.W. and Vogelstein, B. (1998) Characterization of human FAST-1, a TGF beta and activin signal transducer. Mol Cell, 2, 121-127. Zhou, X.M., Wong, B.C., Fan, X.M., Zhang, H.B., Lin, M.C., Kung, H.F., Fan, D.M. and Lam, S.K. (2001a) Non-steroidal anti-inflammatory drugs induce apoptosis in gastric cancer cells through up-regulation of bax and bak. Carcinogenesis, 22, 1393-1397. Zhou, X.P., Woodford-Richens, K., Lehtonen, R., Kurose, K., Aldred, M., Hampel, H., Launonen, V., Virta, S., Pilarski, R., Salovaara, R., Bodmer, W.F., Conrad, B.A., Dunlop, M., Hodgson, S.V., Iwama, T., Jarvinen, H., Kellokumpu, I., Kim, J.C., Leggett, B., Markie, D., Mecklin, J.P., Neale, K., Phillips, R., Piris, J., Rozen, P., Houlston, R.S., Aaltonen, L.A., Tomlinson, I P. and Eng, C. (2001b) Germline mutations in BMPRl A/ALK3 cause a
232 subset of cases of juvenile polyposis syndrome and of Cowden and Bannayan-Riley-Ruvalcaba syndromes. Am J Hum Genet, 69, 704-711.
233