Helsinki University Biomedical Dissertations No. 43

COLORECTAL CANCER AND DEFECTS IN THE MISMATCH REPAIR AND

BASE EXCISION REPAIR SYSTEMS

Susa Enholm

Department of Medical Genetics Molecular and Cancer Biology Programme

Haartman Institute and Biomedicum Helsinki University of Helsinki

Finland

Academic dissertation

To be publicly discussed, with the permission of the Medical Faculty of the

University of Helsinki, in the lecture hall 3, Biomedicum Helsinki, Haartmaninkatu 8, on 16 January 2004, at 12 noon.

Helsinki 2004

Supervised by Lauri A. Aaltonen, MD, PhD Research Professor of the Finnish Academy of Sciences Department of Medical Genetics Haartman Institute and Biomedicum Helsinki University of Helsinki Finland Auli Karhu, PhD Department of Medical Genetics Haartman Institute and Biomedicum Helsinki University of Helsinki Finland Reviewed by Anne Kallioniemi, MD, PhD Docent Laboratory of Cancer Genetics Institute of Medical Technology University of Tampere Finland Minna Nyström-Lahti, PhD Docent Division of Genetics Department of Biosciences University of Helsinki Finland Official opponent Tapio Visakorpi, MD, PhD Professor of Cancer Genetics Institute of Medical Technology University of Tampere Finland ISBN 952-10-1493-8 (paperback) ISBN 952-10-1494-6 (PDF) ISSN 1457-8433 http://ethesis.helsinki.fi Yliopistopaino Helsinki 2004

CONTENTS ABBREVIATIONS.......................................................................................................5 LIST OF ORIGINAL PUBLICATIONS .....................................................................7 ABSTRACT ..................................................................................................................8 REVIEW OF THE LITERATURE .............................................................................9

1. SUSCEPTIBILITY TO CANCER......................................................................................9 1.1 Oncogenes and tumour suppressor genes .......................................................9 1.2 Epigenetic mechanisms ...............................................................................10 1.3 High- and low-penetrance genes..................................................................10

2. COLORECTAL CANCER ............................................................................................11 2.1 Non-polyposis and polyposis syndromes .....................................................11

2.1.1 Hereditary non-polyposis colorectal cancer ........................................11 2.1.2 Familial adenomatous polyposis.........................................................13 2.1.3 MYH-associated polyposis..................................................................14

2.2 Low-penetrance genes for colorectal cancer.................................................14 3. MODELS FOR COLORECTAL CANCER FORMATION .....................................................15

3.1 From adenoma to carcinoma........................................................................15 3.2 Tumour clock model ...................................................................................17

4. DNA REPAIR..........................................................................................................18 4.1 Mechanisms of DNA repair .........................................................................18 4.2 DNA repair disorders ..................................................................................19

5. COLORECTAL CANCER-CAUSING GENES IN MISMATCH REPAIR AND BASE EXCISION

REPAIR SYSTEMS ........................................................................................................19 5.1 Mismatch repair mechanism........................................................................19

5.1.1 Defects in mismatch repair genes .......................................................22 5.1.1.1 MLH1 and MSH2 ...................................................................22 5.1.1.2 MSH6 and MSH3 ...................................................................22 5.1.1.3 PMS1 and PMS2 ....................................................................23 5.1.1.4 MLH3 ....................................................................................23 5.1.1.5 EXO1 .....................................................................................23

5.1.2 Inactivation of mismatch repair genes.................................................24 5.1.3 Microsatellite instability.....................................................................24

5.1.3.1 Microsatellite instability target genes......................................25 5.2 Base excision repair mechanism ..................................................................27

5.2.1 Defects in base excision repair genes..................................................28 5.2.1.1 MYH ......................................................................................29

AIMS OF THE STUDY..............................................................................................30 MATERIALS AND METHODS ................................................................................31

1. PATIENTS AND DNA SAMPLES ................................................................................31 1.1 MLH1-/- patient (I) .......................................................................................31 1.2 Colorectal carcinoma patients (II-V)............................................................31

1.2.1 Searching for microsatellite instability target genes (II) ......................31

1.2.2 MLH3 analysis (III)............................................................................31 1.2.3 EXO1 analysis (IV) ............................................................................31 1.2.4 MYH analysis (V)...............................................................................32

1.3 Control individuals (I, III-V) .......................................................................32 1.4 DNA extraction (I-V) ..................................................................................32

2. STUDY OF THE MLH1-/- PATIENT (I).........................................................................33

2.1 MLH1 mutation 1 analysis...........................................................................33 2.2 Microsatellite analysis .................................................................................33

3. SEARCH FOR MICROSATELLITE INSTABILITY TARGET GENES (II) ...............................34 3.1 Mononucleotide repeats in introns and known microsatellite instability target genes.......................................................................................................34 3.2 Search for putative microsatellite instability target genes .............................34 3.3 Statistical analysis .......................................................................................35

4. MLH3 MUTATION ANALYSIS (III) ............................................................................35 4.1 Single-strand conformational polymorphism analysis ..................................35 4.2 Radioactive polyacrylamide gel electrophoresis...........................................35

5. EXO1 MUTATION ANALYSIS (IV) ............................................................................35 5.1 Direct sequencing, single-strand conformational polymorphism analysis, and restriction analysis ......................................................................................35 5.2 Denaturing gradient gel electrophoresis, conformation-sensitive gel electrophoresis, and denaturing high-performance liquid chromatography .........35

6. MYH FOUNDER MUTATION ANALYSIS (V) ................................................................36 6.1 Solid-phase minisequencing in detecting G382D and Y165C variants..........36 6.2 Direct sequencing, single-strand conformational polymorphism analysis, and denaturing high-performance liquid chromatography ..................................36

RESULTS ...................................................................................................................37 1. MICROSATELLITE INSTABILITY ANALYSIS (I) ...........................................................37 2. MUTATIONS OF MONONUCLEOTIDE REPEATS (II)......................................................37 3. MLH3 MUTATION ANALYSIS (III) ............................................................................38 4. EXO1 MUTATION ANALYSIS (IV) ............................................................................39 5. MYH GERMLINE MUTATION CARRIERS (V)...............................................................39

DISCUSSION..............................................................................................................41 1. MICROSATELLITE MUTATIONS PRECEDE VISIBLE TUMORIGENESIS (I) ........................41 2. MICROSATELLITE INSTABILITY TARGET GENES (II)...................................................43

2.1 Background mutations in intronic mononucleotide repeats and mutability of known microsatellite instability target genes .................................................43 2.2 Putative microsatellite instability target genes..............................................43

3. POSSIBLE LOW-RISK ALLELES FOR COLORECTAL CANCER (III, IV) ............................45 3.1 MLH3 (III) ..................................................................................................45 3.2 EXO1 (IV)...................................................................................................47

4. BIALLELIC MUTATIONS IN MYH PREDISPOSE TO MULTIPLE ADENOMAS (V) ...............49 CONCLUSIONS AND FUTURE PROSPECTS........................................................52 ACKNOWLEDGEMENTS........................................................................................53 REFERENCES ...........................................................................................................55

5

ABBREVIATIONS ACF aberrant crypt foci APC adenomatous polyposis coli ATM ataxia-telangiectasia mutated BER base excision repair BLM Bloom syndrome gene bp base pair BRCA1 and -2 breast and ovarian cancer genes 1 and 2 CDH1 cadherin 1 CHK2 checkpoint kinase 2 CIMP CpG island methylated phenotype CIN chromosomal instability COX-2 cyclo-oxygenase-2 CRC colorectal cancer CSGE conformation-sensitive gel electrophoresis CtIP CtBP-interacting protein, retinoblastoma-binding protein 8 DAPK death-associated protein kinase DCC deleted in colorectal cancer DGGE denaturing gradient gel electrophoresis dHPLC denaturing high-performance liquid chromatography DNA deoxyribonucleic acid DSB double-strand break in DNA EB1 microtubule end binding EXO1 exonuclease 1 FANC Fanconi anaemia genes FAP familial adenomatous polyposis GI gastrointestinal GSK3β glycogen synthase kinase 3β GSTM1 and -T1 glutathione-S transferases Mu 1 and Theta 1 IDL insertion/deletion loops HNPCC hereditary non-polyposis colorectal cancer IGFII insulin-like growth factor II kb kilobase K-RAS gene coding for Harvey sarcoma virus homologue, Kirsten type LKB1 serine/threonine protein kinase 11 LOH loss of heterozygosity LOI loss of imprinting MDE mutation detection enhancement MGMT O6-methylguanine methyltransferase Min multiple intestinal neoplasia MLH1 and -3 human mutL (E. coli) homologues 1 and 3 MMR mismatch repair

6

MRE11 meiotic recombination protein 11 (S. cerevisiae) homologue MSH2, -3, -6 human mutS (E. coli) homologues 2, 3, and 6 MSI microsatellite instability, microsatellite-instable MSI-H high microsatellite instability MSI-L low microsatellite instability MSS microsatellite stable MTH1 human MutT (E. coli) homologue 1 MYH human MutY (E. coli) homologue NAT1 and -2 N-acetyl transferases 1 and 2 NBS1 mutated in Nijmegen breakage syndrome NER nucleotide excision repair NF neurofibromatosis OBR leptin receptor OGG1 8-oxoguanine DNA glycosylase O6-Meg O6-methylguanine 8-oxoG oxidized guanine lesion p short arm of the chromosome p16 cyclin-dependent kinase inhibitor-2A PAGE polyacrylamide gel electrophoresis PCNA proliferating cell nuclear antigen PCR polymerase chain reaction PLA2G2A phospholipase A2, group 2A PMS1 and -2 postmeiotic segregation increased 1 and 2 PPAR peroxisome proliferative activated receptor PRL-3 protein-tyrosine phosphatase 3 q long arm of the chromosome RB retinoblastoma gene RECQL2 and -4 RECQ protein-like 2 and 4, DNA helicases RNA ribonucleic acid ROS reactive oxygen species SMAD2 and -4 human homologues of Drosophila melanogaster Mad genes 2

and 4 SRC human homologue of Rous sarcoma virus gene SSCP single-strand conformational polymorphism analysis STK15 serine/threonine kinase 15 TGF-β transforming growth factor-β TGF-βRII transforming growth factor-β receptor type II TP53 tumour protein 53 TTDA mutated in trichothiodystrophy UV ultraviolet XPA-G xeroderma pigmentosum syndrome genes

7

LIST OF ORIGINAL PUBLICATIONS This thesis is based on the following original articles referred to in the text by Roman numerals I-V: I. Vilkki S, Tsao JL, Loukola A, Pöyhönen M, Vierimaa O, Herva R, Aaltonen LA,

Shibata D (2001). Extensive somatic microsatellite mutations in normal human tissue. Cancer Research 61, 4541-4544.

II. Vilkki S, Launonen V, Karhu A, Sistonen P, Västrik I, Aaltonen LA (2002).

Screening for microsatellite instability target genes in colorectal cancers. Journal of Medical Genetics 39, 785-789.

III. Loukola A, Vilkki S, Singh J, Launonen V, Aaltonen LA (2000). Germline and

somatic mutation analysis of MLH3 in MSI-positive colorectal cancer. American Journal of Pathology 157, 347-352.

IV. Jagmohan-Changur S*, Poikonen T*, Vilkki S*, Launonen V, Wikman F, Orntoft

TF, Møller P, Vasen H, Tops C, Kolodner RD, Mecklin J-P, Järvinen H, Bevan S, Houlston RS, Aaltonen LA, Fodde R, Wijnen J, Karhu A (2003). EXO1 variants occur commonly in normal population: evidence against a role in hereditary nonpolyposis colorectal cancer. Cancer Research 63, 154-158.

V. Enholm S, Hienonen T, Suomalainen A, Lipton L, Tomlinson I, Kärjä V,

Eskelinen M, Mecklin J-P, Karhu A, Järvinen HJ, Aaltonen LA (2003). Proportion and phenotype of MYH-associated colorectal neoplasia in a population-based series of Finnish colorectal cancer patients. American Journal of Pathology 163, 827-832.

* Equal contribution

8

ABSTRACT Exogenous and endogenous mutagenesis occurs continuously in DNA. To prevent accumulation of mutations and subsequent development of cancer, different cell mechanisms repair the DNA damage. Two DNA repair systems, mismatch repair (MMR) and base excision repair (BER), were evaluated in this study. Defects in these mutation repair systems are known to lead to colorectal cancer (CRC). DNA polymerases make errors in DNA replication by adding or deleting nucleotides in long repeat units. The normal MMR system removes these “slippages”, but if defective, mutations occur in the DNA. Accumulation of mutations in repeat tracts can be detected as microsatellite instability (MSI). Repeat sequences of numerous genes may be targets for frameshift mutations as a result of MMR deficiency, leading to truncated proteins. Mutations in MMR genes (mostly MLH1 and MSH2) have been identified as causing hereditary non-polyposis colorectal cancer (HNPCC), which accounts for approximately 1-5% of all CRCs. Aerobic cell metabolism creates reactive oxygen species that damage DNA. BER is the primary mechanism responsible for repairing these oxidative damages. Recently, inherited defects in the BER system have been reported, and MYH mutations have been associated with colorectal neoplasia. We studied somatic microsatellite mutations in MMR-deficient cells. The phenotypically normal tissues of a MLH1-/- patient were evaluated, and large numbers of mutated non-coding microsatellites were observed preceding any signs of tumour formation. To identify new MSI target genes, intronic and coding repeat sequences were examined in colorectal tumours. Comparison of mutations in intronic and exonic repeats revealed a novel MSI target gene, CtIP. Also studied was the role of MMR genes MLH3 and EXO1 in CRC. No MLH3 germline mutations were found in MSI or microsatellite-stable CRC patients, and somatic mutations in MSI tumours were not selected. A prominent role of MLH3 in HNPCC was thus excluded. Nor did EXO1 appear to predispose to HNPCC. However, the role of six EXO1 amino acid changes in CRC could not be excluded. Defective BER and founder mutations in MYH were evaluated in an unselected, population-based set of CRC samples, and an allele frequency of 0.006 was detected. Biallelic mutations were present in 0.4% of patients, and these biallelic mutation carriers seemed to have more polyps than patients with only one mutated allele. MYH-associated CRC was found to be as common as familial adenomatous polyposis-associated CRC in the Finnish population.

9

REVIEW OF THE LITERATURE 1. SUSCEPTIBILITY TO CANCER Tumorigenesis and tumour metastasis require several genetic mutations. A single cell acquires mutations and loses control of the cell cycle, differentiation, and apoptosis, giving the mutated cell a growth advantage over normal cells. Deregulated cell proliferation results in the accumulation of mutated cells ultimately dependent on the formation of blood vessels. Tumour growth is subsequently followed by tissue invasion and metastasis to distant organs (Hanahan and Weinberg 2000). More than one hundred forms of cancers, which can be further divided into various subtypes, are known. The total number of genes and pathways involved in the formation of different cancers can not be specified because in each case of tumorigenesis several different mutated genes and pathways are involved (Hanahan and Weinberg 2000). However, the genes with major effects on tumour initiation can be classified into two groups: oncogenes and tumour suppressor genes. 1.1 Oncogenes and tumour suppressor genes Proto-oncogenes are responsible for controlled cellular proliferation and differentiation. Because of their dominant nature, one mutated proto-oncogene allele is sufficient to lead to uncontrolled proliferation. Oncogenes can act by increasing the amount of gene product or the function by mutations, insertional mutagenesis, or gene amplification. Chromosomal translocations may even form proteins with new growth-enhancing properties (Weinberg 1994). More than one hundred oncogenes have been identified to date (http://www.ncbi.nlm.nih.gov/dbEST/CancerGene.html, Futreal et al. 2001). Proto-oncogenes may function as growth factors, growth factor receptors, intracellular signalling transduction factors, DNA-binding nuclear proteins, and cell cycle factors. Three classes of tumour suppressor genes have been introduced: gatekeepers, caretakers and landscapers (Kinzler and Vogelstein 1997, 1998). Mutations of gatekeepers, such as proliferation-inhibiting or apoptosis-promoting genes, are essential for tumour formation. Deficiency of caretakers, such as DNA repair genes, leads to an accumulation of mutated genes, further assisting the development or progression of cancer. Landscapers modify the microenvironment, and thus, mutations affect the characters of a tumour. Tumour progression requires loss of both alleles of the tumour suppressor gene in the target tissue, known as Knudson’s two-hit model (Knudson 1971, 1985). A recessive effect of mutations may be achieved by inherited or somatic hits. After the first mutation, loss of heterozygosity (LOH) or epigenetic mechanisms are needed to inactivate the gene (Devilee et al. 2001, Tomlinson et al. 2001).

10

Recently, haploinsufficiency of tumour suppressor genes has been recognized. A study of the Peutz-Jeghers syndrome gene LKB1 showed that only one mutated Lkb1 allele resulted in gastrointestinal polyposis in mice (Rossi et al. 2002). Even though heterozygous mutations in tumour suppressor genes are sufficient for tumour formation, biallelic inactivation leads to faster tumour progression, producing more and larger tumours (Bai et al. 2003). 1.2 Epigenetic mechanisms Not only genetic but also epigenetic changes occur during carcinogenesis. Hypermethylation of CpG islands, producing the CpG island methylated phenotype (CIMP), has been detected in colorectal cancers (CRC) (Toyota et al. 1999). DNA hypermethylation may act alone or in combination with genetic alterations to delete the function of tumour suppressor genes. Methylation of the mismatch repair gene MLH1 promoter has been found in the majority of sporadic microsatellite-instable (MSI) colon cancers, leading to transcriptional repression (Herman et al. 1998, Kuismanen et al. 1999, Costello and Plass 2001). In addition to tumour suppressor gene silencing, expression of imprinted genes can be modified by aberrant DNA methylation. Loss of imprinting has been found for instance in the IGF2 gene. Expression of both IGF2 alleles leads to tissue overgrowth and tumorigenesis (Ogawa et al. 1993), and loss of IGF2 imprinting has been suggested as a marker of CRC risk (Cui et al. 2003). Furthermore, genes such as p16, DAPK, and MGMT are methylated in multiple types of tumours (Costello and Plass 2001). Other epigenetic mechanisms include acetylation, deacetylation, and methylation of histones. Removal of acetyl groups from histone H3 results in an increased positive charge and a more condensed chromatin structure, leading to impaired transcription factor binding (Kuo and Allis 1998, Guschin et al. 2000). Mihaylova et al. (2003) have recently shown that the expression of MLH1 in mammalian cells is reduced by histone deacetylation under hypoxia. 1.3 High- and low-penetrance genes High-penetrance or low-penetrance genes cause susceptibility to cancers. Generally, inherited disorders, such as hereditary non-polyposis colorectal cancer (HNPCC) or familial adenomatous polyposis (FAP), are caused by mutations in high-penetrance genes and are often inherited in an autosomal dominant fashion. Hereditary cancer syndromes are mostly a result of mutations in tumour suppressor genes. Mutations in high-penetrance genes directly increase the cancer risk of the carrier. In low-penetrance cases, the polymorphic allele or combinations of different polymorphic alleles possibly act together with environmental factors, leading to increased cancer risk (de Boer 2002). However, high-penetrance genes may also contain low-penetrance mutations (Fraylin et al. 1998).

11

2. COLORECTAL CANCER As is the case for many other cancers, CRC is a multifactorial disease. Diet, life-style, ageing (Boyle and Leon 2002), and genetic predisposition are known to affect its development. Approximately 5% of the Western population suffers from CRC, and the disease is common in both women and men (Kinzler and Vogelstein 1996). CRC incidence and mortality rank among the top three cancers world-wide (http://www-dep.iarc.fr/globocan/globocan.htm). Sporadic forms account for 65-95% of CRCs (Cannon-Albright et al. 1988, Houlston et al. 1992, Lichtenstein et al. 2000), and mainly affect persons older than 50 years (Boland et al. 1998). 2.1 Non-polyposis and polyposis syndromes The inherited syndromes, consisting of non-polyposis and polyposis syndromes (Table 1), are mainly passed down in an autosomal dominant fashion. The non-polyposis syndromes are HNPCC and its variants. Polyposis syndromes are further divided into adenomatous and hamartomatous polyposis. The former includes FAP, attenuated FAP, and MYH-associated polyposis. Turcot syndrome can occur with or without polyps as a result of mutations in the APC or mismatch repair (MMR) genes. HNPCC and MYH-associated polyposis will be introduced in further detail in the following sections (2.1.1, 2.1.3, 5.1, and 5.2) as the sequelae of defects in DNA repair. Rare polyposis syndromes with histological characteristics of hamartomatous polyps are juvenile polyposis, Peutz-Jeghers syndrome, and Cowden syndrome. 2.1.1 Hereditary non-polyposis colorectal cancer Dr. Warthin (1913) showed that CRC occurred frequently in one family. He also reported cancers of the endometrium and stomach in the same family. Later, inheritance of the syndrome was demonstrated to be autosomal dominant (Lynch et al. 1966). Tumours of HNPCC patients were found to contain unstable microsatellites (Aaltonen et al. 1993). Mutations in genes MLH1, MSH2, PMS2, and MSH6 were subsequently identified (Fishel et al. 1993, Leach et al. 1993, Bronner et al. 1994, Nicolaides et al. 1994, Papadopoulos et al. 1994, Miyaki et al. 1997). HNPCC accounts for approximately 1-5% of all CRCs (Aaltonen et al. 1998, Salovaara et al. 2000, Samowitz et al. 2001). HNPCC patients are about 20 years younger than sporadic CRC patients at diagnosis, the mean age for HNPCC onset being 44 years. In HNPCC patients, the progression of endoscopically flat adenomas to cancer can be rapid, taking only a couple of years (Boland et al. 1998). Approximately 90% of tumours in HNPCC patients display microsatellite instability (MSI) (Aaltonen et al. 1994, Lu et al. 1995, Liu et al. 1996). MSI CRCs occur slightly more frequently in women than in men, and when compared with microsatellite-stable (MSS) colorectal tumours, they are often poorly differentiated and muscinous, proximally located, and have peritumoral lymphocyte infiltration (Ionov et al. 1993, Thibodeau et al. 1993, Boland et al. 1998, Elsaleh et al. 2000, Gryfe et al. 2000, Smyrk et al. 2001, Ward et al. 2001). Metastasis occurs less frequently in patients with MSI tumours than in patients with MSS tumours

12

Table 1. Non-polyposis (A), adenomatous polyposis (B), and hamartomatous polyposis (C) syndromes (Hamilton et al. 2000, Hampel and Peltomäki 2000). Syndrome Incidence Clinical features Genes

responsible A.

Hereditary non-polyposis colorectal cancer (HNPCC)

1/1000 (1-6% of CRCs)

Adenomas and colon cancer, tumours of the endometrium, stomach, ovary, small bowel, biliary tract, uro-epithelium, kidney, central nervous system

MLH1, MSH2, MSH6, PMS2

Muir-Torre syndrome (variant of HNPCC)

Sebaceous adenomas of the face and scalp, keratoa-canthomas, HNPCC cancers, basal cell carcinomas of the skin, breast cancer

MLH1, MSH2

Turcot syndrome (variant of HNPCC)

Adenomas in the colon and rectum, colorectal tumours and glioblastomas

MLH1, PMS2

B. Familial adenomatous polyposis (FAP)

1/8-10 000 (<1% of CRCs)

Hundreds to thousands of adenomatous polyps in the colon and rectum, colorectal and duodenal cancers, papillary thyroid carcinoma, hepatoblastoma, congenital hypertrophy of the retinal pigment epithelium, epidermoid cysts, desmoids, osteomas, dental anomalies

APC

Attenuated FAP (variant of FAP)

Dozens of polyps in the colon and rectum, colorectal tumours

APC

Turcot syndrome (variant of FAP)

Adenomas in the colon and rectum, colorectal tumours, medulloblastomas

APC

MYH-associated polyposis unknown

(0.4-1.3% of CRCs) From a few to over a hundred polyps in the colon and rectum, colorectal cancer

MYH

C. Juvenile polyposis 1/100 000 Hamartomatous polyps in the stomach

and large bowel, congenital heart defects, cleft lip or palate, malrotations, tumours of the colon, stomach, duodenum, pancreas

SMAD4, BMPR1A

Peutz-Jeghers syndrome 1/200 000 Hamartomatous polyps in the GI tract and

peri-oral pigmentation, tumours of the colon, small intestine, stomach, pancreas, ovary

LKB1

Cowden syndrome 1/300 000 Hamartomatous polyps in the

gastrointestinal tract, fibrocyctic and fibroadenomatous breasts, multinodular goiter of the thyroid and endometrial fibroids, skin findings, lipomas, macrocephaly, dolicocephaly, keratoses, tumours of the breast, thyroid, endometrium, skin, kidney, brain

PTEN

Recessive inheritance of the syndrome is italicized.

13

(Lothe et al. 1993, Thibodeau et al. 1993, Gryfe et al. 2000), and patients with MSI tumours seem to benefit from adjuvant chemotherapy (Elsaleh et al. 2000). HNPCC mutation carriers have a very high lifetime risk (>80%) of developing CRC (Vasen et al. 1996). Furthermore, the risk for a second primary colon cancer as well as endometrial, ovary, and stomach cancers is increased. Extracolonic tumours may also occur in the kidney, urinary tract, biliary tract, small intestine, brain, and skin (Watson and Lynch 1993, Aarnio et al. 1995). CRC risk has been found to be higher in male mutation carriers than in their female counterparts (Dunlop et al. 1997, Vasen et al. 2001). Risk of developing CRC is also increased among MSH2 mutation carriers as compared with MLH1 mutation carriers (Vasen et al. 1996, 2001, Dunlop et al. 1997). Endometrial cancer has been shown to develop more frequently in MSH6 mutation carriers (Wijnen et al. 1999), and CRC caused by MSH6 mutations to have a late onset (>60 years) and an MSS phenotype (Kolodner et al. 1999, Wu et al. 1999b). HNPCC patients are offered regular colonoscopy. In HNPCC families, CRC incidence and mortality rate decreased by more than 60% when colonoscopy was carried out at three-year intervals. Extensive subtotal colectomy is recommended to prevent further colonic tumours (Järvinen et al. 2000). The clinical selection criteria to identify HNPCC families were defined in 1990 by the International Collaborative Group on HNPCC. Amsterdam Criteria I was revised in 1999 to include all HNPCC-associated cancers (Vasen et al. 1991, 1999). Amsterdam Criteria II consists of three items: 1) at least three relatives with an HNPCC-associated cancer (cancers of the colorectum, endometrium, small bowel, ureter, or renal pelvis), one being a first-degree relative of the other two, 2) cases covering at least two generations, and 3) at least one cancer case diagnosed before age 50 years. In addition, FAP should be excluded and tumours verified by pathological examination. Advantages of the modified Amsterdam criteria include the possibility to use it in small families, to discover families with cancers other than CRC, and to test people with later onset of cancer. More than half of HNPCC families who fulfil the clinical Amsterdam criteria have defects in MLH1 or MSH2 (Liu et al. 1996, Wijnen et al. 1997). However, all HNPCC mutation carriers do not fulfil these criteria (Salovaara et al. 2000, Samowitz et al. 2001). Thus, testing for replication errors is recommended for patients with an early onset of CRC (<50 years), a suspect family history of CRC or endometrial cancer, or a history of multiple colorectal or endometrial cancers (Aaltonen et al. 1998). 2.1.2 Familial adenomatous polyposis FAP is an autosomal dominant trait caused by mutations in the APC gene. The rectum, colon, and the upper gastrointestinal tract of FAP patients display hundreds or thousands of adenomatous polyps. At least a hundred polyps must be present to meet the clinical diagnosis of FAP. Without surgery, most of these patients develop colon cancer originating in polyps by the age of 40 years. Surveillance by sigmoidoscopy begins at the age of ten, and later prophylactic and total colectomies are offered (Talbot et al. 2000).

14

Depending on the location of the mutation in the gene, colonic and extracolonic features vary. Severe disease counted by the number of polyps is caused by mutations in the central region of the gene, between codons 1250 and 1464, and the most severe disease by mutations around codon 1300 (Nugent et al. 1994). Attenuated FAP, clinically diagnosed by fewer polyps at a later age, is mainly caused by mutations in exons 1-4 (codons 1-158) (Spirio et al. 1993). The second somatic hit in APC depends on the type of the first mutation. The germline mutation detected between codons 1194 and 1392 often leads to allelic loss of APC which is the second hit. If the germline mutation occurs outside codons 1194-1392, the second mutation will probably hit a mutation cluster region between codons 1286 and 1513, resulting in a truncated protein (Lamlum et al. 1999). Desmoids in FAP patients are associated with mutations after codon 1444 (Caspari et al. 1995), and congenital hypertrophy of the retinal pigment epithelium (CHRPE) between codons 463 and 1387 (Olschwang et al. 1993). One-fifth of all clinically diagnosed FAP patients carry no APC mutations (Hamilton et al. 2000). 2.1.3 MYH-associated polyposis MYH-associated polyposis, as recently reported by Al-Tassan et al. (2002), is an autosomal recessive trait. Almost one-third of patients with 15-100 polyps have biallelic MYH mutations (Sieber et al. 2003). MYH double mutations were also found in 7% of patients with a clinical diagnosis of classic adenomatous polyposis but without an APC germline mutation (Sieber et al. 2003). Sampson et al. (2003) studied polyposis registries in UK and screened MYH mutations in patients selected by the following criteria: family history showing no vertical transmission of polyposis, at least ten colorectal adenomas, and no pathogenic mutation in the APC gene. Biallelic mutations were found almost in one-fourth of the cases. 2.2 Low-penetrance genes for colorectal cancer APC polymorphisms I1307K (Laken et al. 1997) and E1317Q (Fraylin et al. 1998) have been reported to increase CRC risk. These APC alleles do not, however, cause the polyposis typical of many other APC mutations. Missense mutations in CDH1 (cadherin 1) have also been associated with colorectal and gastric cancers (Kim et al. 2000). Various genes can modify the predisposing effect to the phenotype of a given form of cancer. Activation of the peroxisome proliferative activated receptor (PPAR) –γ has been shown to promote the development of colon tumours in heterozygous APC (APCmin/+) mice (Lefebvre et al. 1998). Mutated phospholipase a2 (PLA2G2A) also increased the number of intestinal polyps in polypotic mice (Cormier et al. 1997). The role of cyclo-oxygenase 2 enzyme (COX-2) in CRC was recognized following the discovery of production of COX-2 being increased in colon cancer cells (Kargman et al. 1995). Furthermore, deletion of COX-2 in APC knock-out mice reduced the number and size of tumours in the intestine (Oshima et al. 1996). COX-2 inhibitor is also known to diminish polyps in FAP patients (Giardiello et al. 1993). The components of the colonic luminal contents, which are affected by diet, were recently demonstrated to induce COX-2 transcription in human colon carcinoma cells (Glinghammar and Rafter 2001).

15

An association between polymorphic carcinogen metabolism genes and increased risk of CRC has been suggested. Some alleles of N-acetyl transferases (NAT1 and NAT2) and glutathione-S transferases (GSTM1 and GSTT1) that detoxify arylamines or mutagenic electrophiles have been proposed to have such characters (Houlston and Tomlinson 2001). In addition, polymorphism of 5,10-methylenetetrahydrofolate reductase MTHFR has been shown to affect DNA methylation (Houlston and Tomlinson 2001). Up to 35% of all CRC cases may be due to genetic predisposition (Cannon-Albright et al. 1988, Houlston et al. 1992, Lichtenstein et al. 2000), but mutations of genes predisposing to CRC have not been found in the majority of familial CRC cases (Burt 2000). 3. MODELS FOR COLORECTAL CANCER FORMATION 3.1 From adenoma to carcinoma Approximately 40% of the population in Western countries will develop adenomas of the colon (DiSario et al. 1991, Lieberman and Smith 1991, Rex et al. 1991), but only a minor proportion develops cancer. Alterations in the Wnt signalling pathway initiate the formation of colon cancer in most cases. Adenomatous polyps harbour mutations or loss of the APC gene. Mutations in the K-ras, TP53, and TGF-β signalling pathway further promote tumour progression (Fearon and Vogelstein 1990) (Figure 1). Tumour progression of adenomas results in increasing size and dysplastic features of the polyp. A clear carcinoma is characterized by invasion of the basement membrane and metastasis. This process develops over years or decades (Muto et al. 1975). Chromosomal 5q 12p 18q 17p location SMAD2 SMAD4 Other

APC K-RAS DCC TP53 changes normal dysplastic early inter- late epithelium aberrant adenoma mediate adenoma carcinoma metastasis

crypt foci adenoma

Figure 1. Accumulation of mutations leading to the formation of CRC in a multistep model. Modified from Fearon and Vogelstein (1990) and Kinzler and Vogelstein (1996). The tumour suppressor gene product APC is known to interact with a number of proteins, including β-catenin (CTNNB1), glycogen synthase kinase (GSK)-3β, end binding protein (EB)1, axis inhibitors (axin), tubulin (TUB), the human homologue of Drosophila discs’ large tumour suppressor protein (hDLG) (Kinzler and Vogelstein 1996), and budding uninhibited by benzimidazoles (Bub) kinases (Kaplan et al. 2001). The main tumour promoting character of mutated APC is the insufficient degradation of β-catenin. Consequently, more β-catenin enters the nucleus and overactivates transcription of such

16

tumour promoting genes as c-myc, cyclin D1 (CCND1), matrilysin (MMP7), and gastrin (GAS) (Wong and Pignatelli 2002). APC mutations are observed in 40-80% of CRCs and at a similar rate in adenomas (Vogelstein et al. 1988, Leslie et al. 2002). In addition to APC, mutations of β-catenin in the Wnt signalling pathway are early events in CRC (Samowitz et al. 1999). The proto-oncogene K-RAS is involved in pathways of normal proliferation and differentiation, mediating signalling from growth factor receptors to signal transduction pathways. K-RAS mutations are displayed by 35-42% of CRCs and large adenomas. Moreover, lack of K-RAS mutations in small adenomas illustrates the role of this oncogene in providing a growth advantage for mutated cells (Vogelstein et al. 1988, Leslie et al. 2002). TP53, a transcription factor with tumour suppressor activity, is involved in the transition of adenoma to carcinoma. TP53 alterations have been shown in 4-26% of adenomas and 50-75% of adenocarcinomas. TP53 can block the cell cycle for DNA repair, promote apoptosis, or further inhibit angiogenesis (Levine 1997, Leslie et al. 2002). Different mutations in the TGF-β pathway affect transcription of genes involved in cell growth, differentiation, matrix production, immune surveillance, and apoptosis (Massague 1996). Smad proteins are also involved in this cascade. Loss of chromosome 18q occurs in approximately 70% of CRCs and frequently also in adenomas (Vogelstein et al. 1988, Fearon and Vogelstein 1990). Tumour suppressor genes mutated in colon cancers in this area include SMAD4, SMAD2, and DCC (Fearon et al. 1990, Cho et al. 1994, Eppert et al. 1996, Takagi et al. 1996, Fazeli et al. 1997). Moreover, mutations in the TGF-β type II receptor occur in several CRCs (Markowitz et al. 1995, Lu et al. 1998), and these mutations correlate with progression of adenoma to carcinoma (Grady et al. 1998). The oncogene HER-2/neu is overexpressed in most CRCs, with expression correlating with stage of the disease (Kapitanovic et al. 1997). Mutations of the oncogenic tyrosine kinase SRC were reported in 12% of advanced CRCs from North America (Irby et al. 1999), but not in CRCs from other populations (Laghi et al. 2001). In addition, increased expression or amplification of the gene tyrosine phosphatase PRL-3 occur in metastatic CRCs (Saha et al. 2001), and the protein has been shown to promote cell motility, invasion activity, and metastasis (Zeng et al. 2003). Some genetic alterations lead to chromosomal (CIN) or microsatellite instability (MSI) as accumulation of mutations increases. Most of the CRCs display CIN such as gains or losses of chromosomal material or alterations in chromosome number (Lengauer et al. 1997). The molecular events underlying CIN are known to be multifactorial, and they include loss of telomere functions and genetic alterations in genes that control chromosome segregation. Recently, defects in APC, BUB kinases, TP53, STK15, RB1, and BRCA1 have been shown to be associated with chromosomal aberrations (Albertson et al. 2003). The MSI phenomenon occurs in only a small group of CRCs (12-16%) (Aaltonen et al. 1993, Ionov et al. 1993, Thibodeau et al. 1993). MSI is currently suggested to be an

17

alternative and rapid pathway in colon cancer progression and a cause of DNA repair defects. Deficiency in the mismatch repair (MMR) system causes accumulation of mutations because polymerase errors in DNA replication are not repaired. Thus, the progression to carcinoma is mainly caused by mutations in short repeat tracts of MSI target genes (for more detail, see section 5.1.4) (Ionov et al. 1993). In these cases, mutations are secondary due to MMR deficiency. In the case of HNPCC, progression of a normal cell to carcinoma could develop over only a few years instead of taking 10-15 years (Muto et al. 1975). Sporadic cancers with defects in the DNA repair system may also progress in a similar rapid way (Boland et al. 1998). 3.2 Tumour clock model As described earlier, tumorigenesis is a multistep process. In the classical model of CRC formation, a carcinoma develops after a gatekeeper mutation and through a visible adenoma (Figure 2A). In HNPCC, loss of MMR leading to somatic microsatellite mutations and a gatekeeper mutation occurs before visible tumour formation (Shibata 2001) (Figure 2B and C). A. Normal cell B. Loss of MMR Gatekeeper mutation C. Adenoma Cancer Microsatellite mutations before a gatekeeper mutation

Figure 2. According to the adenoma-carcinoma sequence, gatekeeper mutation and visible adenoma formation lead to CRC (A). Loss of MMR activity results in the accumulation of mutations that leads to further progression of HNPCC tumours (B and C) (Tsao et al. 1999, 2000). Genetic events may be deduced by microsatellite sequence comparisons. Modified from Shibata (2001). A tumour arises from a single cell after clonal expansion. Thus, mutations from that founder cell will be present in all tumour cells. Furthermore, each cell in the tumour divides and acquires new mutations. Microsatellite alterations have been used to estimate the number of cell divisions in phases of MSI tumorigenesis. “Slippage” occurs during DNA replication, leading to deletions or insertions of repeat units. A single repeat unit is typically changed at one time (Bhattacharyya et al. 1994, Shibata et al. 1994). Thus, greater numbers of microsatellite mutations around the germline allele sizes or around the allele sizes of the final founder cell indicate a longer time since the loss of MMR or a

18

clonal expansion, respectively. This tumour clock model was initially described by Shibata et al. (1996), who also proposed that MSI carcinoma may develop without preceding formation of a visible adenoma. Similar mutation frequencies of microsatellites in both adenomas and carcinomas show that an approximately equal progression period has occurred (Figure 2C) (Tsao et al. 1999, 2000, Kim et al. 2002). 4. DNA REPAIR 4.1 Mechanisms of DNA repair Exogenous and endogenous mutagenesis occurs continuously in DNA. DNA is damaged by ultraviolet irradiation, ionizing irradiation, alkylating agents, oxygen, and free radicals. DNA replication and recombination also produce mistakes. In the process of evolution, mutations are necessary, but they can also lead to disorders. Cells prevent the development of cancers by repairing these damaged or mutated DNA regions. Five DNA repair mechanisms with affinity for certain types of damage have been described. 1) O6-methylguanine (O6-MeG) DNA methyl transferase (MGMT) transfers the alkyl group of guanine to cysteine, thus preventing pairing of O6-MeG with thymine in DNA replication. 2) Double-strand break (DSB) repair is needed because breaks of both DNA strands are lethal if not repaired. Homologous recombination or non-homologous end joining may be used in repairing DSB. The normal strand is used as a template to repair the broken area of DNA, or DNA ends are joined together without sequence homology from the normal DNA strand. 3) The nucleotide excision repair (NER) system repairs abnormal structures of DNA, including thymidine dimers after UV irradiation as well as chemical modifications of bases and large insertion/deletion loops (>15 base pairs) caused by defective MMR. 4) Base excision repair (BER) takes care of hydrolytic, oxidative, and alkylation damages. 5) The mismatch repair (MMR) system corrects mismatches and insertions/deletions and prevents heterologous DNA exchanges. NER, BER, and MMR mechanisms all use similar ways of repairing the damaged bases of DNA: splicing out the damaged region, inserting new bases, and then ligating the pieces (Friedberg 2003, Mohrenweiser et al. 2003). DNA damage checkpoint pathways control activation of DNA repair pathways but are also involved in, for example, the composition of telomeric chromatin and telomere length, movement of DNA repair proteins to DNA damage, and transcription of various genes. Cells are able to arrest the cell cycle and repair the DNA damage, but if the damage in DNA is too great, cells undergo apoptosis (Zhou and Elledge 2000). Recently, more than 70 genes involved in DNA damage repair or surveillance have been reported (Wood et al. 2001). Despite different efficient DNA repair systems, special DNA polymerases in the damage-tolerance mechanism may allow the DNA replication to continue without repairing the damage (Friedberg 2003, Mohrenweiser et al. 2003).

19

4.2 DNA repair disorders Mutations in DNA repair genes have been identified in several different human disorders, including cancers. Reduced expression of MGMT has been detected in tumour cells (Costello and Plass 2001), but the syndrome-causing germline mutations have not been found. Defects in the other four DNA repair mechanisms have been reported in various disorders (Table 2). DSB repair is affected in many disorders. In chromosome instability disorders, ATM, NBS1, and MRE11 genes are responsible for maintaining resistance to ionizing radiation (Michelson and Weinert 2000). Mutations in the RecQ family of DNA helicases (BLM, RECQL2, RECQL4) affect both DNA replication and homologous recombination (Hickson 2003). Response to the DSB (BRCA1, BRCA2, FANCs, TP53, CHK2) is defective, for example, at cell cycle checkpoints (Welcsh and King 2001, D'Andrea and Grompe 2002, Thompson and Schild 2002). Genes in the NER pathway (XPA-XPG, CSA, CSB, TTDA) also harbour disorder-causing mutations, and defective transcription coupled repair or global genomic repair has been detected in three different syndromes (de Boer and Hoeijmakers 2000) (Table 2). Recently, inherited defects in the BER system have been reported, and mutations of MYH have been associated with colorectal neoplasia (Al-Tassan et al. 2002). Moreover, mutations in MMR genes (mostly MLH1 and MSH2) have been found to cause hereditary non-polyposis colorectal cancer (HNPCC) (Fishel et al. 1993, Leach et al. 1993, Bronner et al. 1994, Nicolaides et al. 1994, Papadopoulos et al. 1994, Miyaki et al. 1997) (Table 2). BER and MMR are described in detail in the following sections. 5. COLORECTAL CANCER-CAUSING GENES IN MISMATCH REPAIR AND BASE EXCISION REPAIR SYSTEMS 5.1 Mismatch repair mechanism The MMR system is well conserved in different organisms. Six E. coli MutS homologues (Msh1-6) have been found in yeast, and only MSH1 is lacking in mammals. Both in yeast and in human cells, the MSH2-MSH6 complex (MutSα) recognizes mispaired bases and single base insertion/deletion loops (IDL), while the MSH2-MSH3 complex (MutSβ) takes care of larger IDLs (2-4 bases) (Marsischky et al. 1996, Pochart et al. 1997) (Figure 3). The eukaryotic MSH4-MSH5 complex is not required for MMR, but it has a role in meiotic recombination (Edelmann et al. 1999, Kneitz et al. 2000). Four E. coli MutL homologues have been described in both mammals and yeast: MLH1, PMS1, PMS2, and MLH3 (Jacob and Praz 2002). MLH1 can form complexes with other MutL homologues such as PMS1, PMS2, and MLH3. The MLH1-PMS2 complex (MutLα) interacts with mismatch recognition complexes MSH2-MSH6 and MSH2-MSH3, repairing the mismatch or IDL in concert with other proteins (Wang et al. 1999b).

20

Table 2. Hereditary syndromes with defective DNA repair mechanisms (de Boer and Hoeijmakers 2000, Michelson and Weinert 2000, Welcsh and King 2001, D'Andrea and Grompe 2002, Thompson and Schild 2002, Hickson 2003). Hereditary syndrome Main clinical features Cancer

predisposition Genes responsible

Defective DSB repair

Ataxia-telangiectasia Neuronal degeneration, immune deficiency, radiation sensitivity, sterility

Lymphoma, leukemia, various solid tumours

ATM

AT-like disorder Immunodeficiency, radiosensitivity Lymphoma MRE11 Nijmegen breakage syndrome

Immunodeficiency, radiosensitivity Lymphoma, various tumours NBS1

Bloom`s syndrome Restricted growth, sun-induced

erythaema, type II diabetes, narrow face and prominent ears, male fertility

Lymphoma, leukemia, tumours of the breast, gut, and skin, most cancer types

BLM

Werner syndrome Premature ageing, short stature, leg

ulceration, soft-tissue calcification Soft-tissue and osteogenic sarcomas, melanoma, thyroid cancer

RECQL2

Rothmund-Thomson syndrome

Restricted growth, skeletal abnormalities, skin pigmentation changes, skin atrophy

Osteogenic sarcomas, various cancers

RECQL4

Fanconi anaemia Bone-marrow failure, congenital

anomalies, dysmorphic features, growth retardation

Leukemia, squamous cell carcinoma

FANCA, C, D2, E, F, G, BRCA2

Hereditary breast and ovarian cancers

Breast, ovarian, prostate, colon, and pancreatic cancers

BRCA1, BRCA2

Li-Fraumeni syndrome Sarcomas, breast, lung, and

colon cancers, brain tumours, leukemia

TP53, CHK2

Li-Fraumeni-like syndrome

Sarcomas, brain tumours, adrenocortical carcinoma

TP53

Defective NER Xeroderma pigmentosum Sensitivity to ultraviolet light, skin

atrophy, irregular pigmentation, telangiectasia, keratosis

Melanoma, basal cell carcinoma

XPA-XPG

Cockayne syndrome Sensitivity to many DNA-damaging

agents, physical and mental retardation CSA,

CSB Trichothiodystrophy Sulphur-deficient brittle hair, ichthyosis,

mental and physical retardation, sun sensitivity

XPD, XPB, TTDA

Defective MMR Hereditary nonpolyposis colorectal cancer (HNPCC)

Adenomas and colon cancer, tumours of the endometrium, stomach, ovary, small bowel, biliary tract, uro-epithelium, kidney, central nervous system

MLH1, MSH2, MSH6, PMS2

Defective BER MYH-associated polyposis

From a few to over a hundred polyps in the colon and rectum

Colorectal cancer MYH

21

In some cases, the MLH1-MLH3 complex may also participate in repair of IDLs (Lipkin et al. 2000). The role of PMS1 in mammalian MMR remains unclear (Prolla et al. 1998, Räschle et al. 1999). E. coli endonuclease MutH nicks the hemimethylated DNA (Welsh et al. 1987, Hall et al. 1999), and excision of the DNA fragment requires exonucleases such as recJ, ExoVII, ExoI, and ExoX (Tran et al. 1999, Burdett et al. 2001). After removal of nucleotides around the mismatch or IDL, the gap is filled again with the correct sequence. In humans, the MutH homologue has not been defined, and only one exonuclease (EXO1) has been found to act in MMR. In addition to MutS and MutL homologues and exonucleases, at least proliferating cell nuclear antigen (PCNA), DNA polymerases, replication factors, and helicases are needed in the process of MMR (Jacob and Praz 2002). Many as yet undefined genes may also have important functions in repairing mismatches or IDLs.

1 bp mismatch short insertion/deletion loop A A A A A A A A GTGTGTGTGTG T T T T G T T T CACACACACAC CA

A A A A A A A A GTGTGTGTGTG T T T T G T T T CACACACACAC

Figure 3. Mammalian DNA mismatch repair complexes: MSH2/MSH6/MLH1/PMS2 participates in repairing mismatches and single base insertion/deletion loops and MSH2/MSH3/MLH1/PMS2 in repairing short insertion-deletion loops. Modified from Chung and Rustgi (2003). In addition to repair of replication mistakes or recombination, MMR proteins are involved in double-strand break repair (MSH2, MSH3, EXO1) and together with NER proteins in the transcription-coupled repair pathway (MSH2, PMS2) (Mellon and Champe 1996, Mellon et al. 1996, Sugawara et al. 1997, Tsubouchi and Ogawa 2000, Rizki and Lundblad 2001). Furthermore, the MMR system has been proven to act in repairing mutations caused by mutagens, such as alkylating agents or heterocyclic amines (Andrew et al. 1998, de Wind et al. 1998, Kawate et al. 1998, Qin et al. 1999). MLH1 and MSH2 participate in cell cycle regulation by inducing G2/M-arrest or apoptosis in

MSH2

MSH2 MSH2

MSH2 MSH3

PMS2

MSH6

MSH6

MLH1 PMS2 MLH1

MSH3

22

response to DNA damage (Duckett et al. 1999, Hickman and Samson 1999, Toft et al. 1999, Wu et al. 1999, Zhang et al. 1999). 5.1.1 Defects in mismatch repair genes 5.1.1.1 MLH1 and MSH2 Most of the germline mutations of mismatch repair genes causing HNPCC are in MLH1 (55%) (Bronner et al. 1994, Papadopoulos et al. 1994) or MSH2 (40%) (Fishel et al. 1993, Leach et al. 1993, http://www.nfdht.nl/database/mdbchoice.htm, http://archive. uwcm.ac.uk/uwcm/mg/search/203983.html) (Table 3). The MLH1 mutation cluster locates in exons 15 and 16, but MSH2 mutations are distributed along the entire gene. In Finland, two MLH1 founder mutations account for the majority of HNPCC mutations (Nyström-Lahti et al. 1996, Aaltonen et al. 1998, Salovaara et al. 2000). Some cases of biallelic mutations in MMR genes in the human germline have also been recognized. MLH1-/- patients in two families (Ricciardone et al. 1999, Wang et al. 1999a) and one MSH2-/- patient (Whiteside et al. 2002) have been diagnosed with signs of neurofibromatosis (NF) and leukemia. In Mlh1-/- cell extracts, single base mispairs or IDLs were not corrected. Moreover, T-cell lymphomas and GI tumours were detected in Mlh1-/- mice (Baker et al. 1996, Edelmann et al. 1996). Because Mlh1-/- mice are sterile (Baker et al. 1996) and human MLH1 interacts with the meiotic recombination protein MSH4 (Santucci-Darmanin et al. 2000), MLH1 protein is suggested to also be involved in meiotic recombination. Tissues in Msh2-/- mice displayed high mutation frequencies, and tumours showed MSI (Andrew et al. 2000). In these mice, T-cell lymphomas occurred first and GI tumours later (>6 months and >10 months, respectively) (Lowsky et al. 1997). Sebaceous gland tumours were similar to those of human patients with Muir-Torre syndrome (Kruse et al. 1998). 5.1.1.2 MSH6 and MSH3 MSH6 germline mutations have also been observed in HNPCC patients (Akiyama et al. 1997, Miyaki et al. 1997, http://www.nfdht.nl/database/mdbchoice.htm) (Table 3). In Msh6-/- cells, single base mismatches were not corrected, and further tumours in Msh6-/- mice did not display typical MSI (Edelmann et al. 1997, 2000). In these mice, T-cell and B-cell lymphomas occurred first, followed by GI tumours (Edelmann et al. 1997, de Wind et al. 1999). Tumours of the skin and endometrium were also detected (de Wind et al. 1999). Germline mutations of MSH3 have not been detected (Huang et al. 2001). However, in Msh3-/- cell extracts, dinucleotide IDLs were not repaired, and some Msh3-/- mice developed GI tumours at a late age. In mice, defects in Msh3 resulted in a mild mutator but not in a significant cancer phenotype (de Wind et al. 1999, Edelmann et al. 2000).

23

5.1.1.3 PMS1 and PMS2 Only one HNPCC patient harbouring the PMS1 mutation has been found (Nicolaides et al. 1994, http://www.nfdht.nl/database/mdbchoice.htm) (Table 3). However, an MSH2 germline mutation was later detected from the same patient and another affected person in the family. This notion confirmed that CRC was due to mutated MSH2 (Liu et al. 2001). Thus, the role of PMS1 in HNPCC is questionable. Further, while Pms1-/--deficient mice showed a little instability at mononucleotide repeats, they developed no tumours (Prolla et al. 1998). HNPCC patients with a deletion in the PMS2 gene have been reported (Nicolaides et al. 1994, http://www.nfdht.nl/database/mdbchoice.htm) (Table 3). Mice models support the role of deficient Pms2 in causing a mutator phenotype. Mutation frequencies at mononucleotide repeat tracts in Pms2-/- tissues were lower than in Mlh1-/- tissues (Narayanan et al. 1997, Yao et al. 1999), and Pms2-/- mice developed lymphomas and sarcomas at six months of age (Baker et al. 1995, Prolla et al. 1998). A role for Pms2 in meiosis has also been suggested since Pms2-/- mice are sterile (Baker et al. 1995). 5.1.1.4 MLH3 The role of MLH3 mutations in CRC is unclear (Table 3). Germline mutations have been found in CRC patients, including two frame shift mutations causing a premature stop codon (Wu et al. 2001b, Liu et al. 2003). However, the nature of missense mutations remains obscure (Lipkin et al. 2001, Wu et al. 2001b, Hienonen et al. 2003, Liu et al. 2003). The importance of Mlh3 in the MMR system has recently been studied. Overexpression of truncated Mlh3 induced MSI in mammalian cells (Lipkin et al. 2000), but MSI was not detectable in Mlh3-/- fibroblasts (Lipkin et al. 2002). In addition to MMR, the role of Mlh3 in meiosis has been demonstrated in Mlh3-/- mice (Lipkin et al. 2002). 5.1.1.5 EXO1 In humans, a double-stranded DNA-specific 5’ to 3’ exonuclease EXOI was found to bind MSH2, MSH3, and MLH1 and was thus suggested to act in the MMR system (Schmutte et al. 1998, 2001, Tishkoff et al. 1998). EXO1 germline mutations have been sought in CRC patients: 14 mutations were found, including one splice site mutation (Wu et al. 2001a). The data from a yeast study indicated that Exo1 defects lead to accumulation of base substitutions and frameshift mutations in mononucleotide runs (Tishkoff et al. 1997). In mice, Exo1 has been shown to function both in the MMR system and in meiosis. Exo1-deficient mice harboured unstable mononucleotide repeats and lymphoma (Wei et al. 2003).

24

Table 3. MMR genes affecting the phenotype of yeast and mammals. Only germline mutations in humans are taken into account. Modified from Lipkin et al. (2000) and Peltomäki (2001).

DNA mismatch repair gene

Chromosomal location

Yeast Mammals

MSI Mouse tumours Human cancers MLH1 3p21-23 ++ Lymphoma (T-cell)

GI tumours CRC

MSH2 2p21 ++ Lymphoma (T-cell) GI and skin cancer

CRC, sebaceous gland tumours and internal

malignancies PMS2 (yeast Pms1) 7p22 ++ Lymphomas and sarcomas

(myxoid and fibroid sarcoma, leiomyosarcoma)

CRC

MSH6 2p21 + Lymphoma (B- and T-cell) GI, uterine, and skin

tumours

CRC, tumours of the endometrium

MLH3 14q24.3 + - CRC? EXO1 1q42-43 + Lymphomas CRC? MSH3 5q11-12 + Late GI tumours -

PMS1 (yeast Mlh2) 2q31-33 - - - ++ Severe MSI phenotype in yeast + Less intense MSI phenotype - Not detected ? More data needed GI Gastrointestinal Mammalian PMS2 is most closely related to yeast Pms1. Mammalian PMS1 is most closely related to yeast Mlh2, which is not associated with MSI. Mice heterozygous for these mutations do not display increased tumour formation.

5.1.2 Inactivation of mismatch repair genes In HNPCC, the inactivation of MMR genes involves the first inherited mutation, and the second allele is later somatically mutated in the epithelial cells of the target tissue. This second change, a somatic deletion, a point mutation, or methylation of the promoter, can inactivate the wild-type allele and further act as a driving force for tumour development. In MLH1 and MSH2 mutation carriers, LOH is detected in approximately 30% and 10% of tumours, respectively (Hemminki et al. 1994, Potocnik et al. 2001). In MLH1 mutation carriers, 46% of gene inactivation is caused by promoter hypermethylation (Kuismanen et al. 2000). Point mutations in MLH1 mutation carriers are rare when compared with MSH2 mutation carriers (Yuen et al. 2002). In contrast to HNPCC tumours, 78% of sporadic MSI CRCs were found to have reduced MLH1 expression mainly due to promoter methylation (Kane et al. 1997, Kuismanen et al. 2000). Approximately 6-15% of tumours showed reduced MSH2 expression, and LOH or mutation in MSH2 was detected (Kuismanen et al. 2000, Cunningham et al. 2001). 5.1.3 Microsatellite instability Microsatellites consist of six or fewer nucleotides, and form repeat tracts of up to 100 base pairs (Levinson and Gutman 1987). These repetitive elements are targets for replication mistakes made by DNA polymerases. “Slippage” of the polymerase leads to insertions or deletions of repeat units. If MMR genes are defective, these DNA changes

25

are not repaired, and microsatellite mutations accumulate. The lengths of microsatellite repeats were found to be unstable in unselected and familial CRCs (Aaltonen et al. 1993, Ionov et al. 1993, Thibodeau et al. 1993, Peltomäki et al. 1993). The Bethesda panel of five microsatellite markers has been proposed for the detection of microsatellite instability (MSI) (Boland et al. 1998). Three dinucleotide repeat markers (D2S123, D5S346, D17S250) and two mononucleotide repeat markers (BAT25, BAT26) are used to distinquish MSI-L (low) and MSI-H (high) instability from microsatellite-stable (MSS) samples. At least two markers with instability confirm MSI-H tumours, and one marker showing instability defines MSI-L tumours (Boland et al. 1998). BAT26 alone is sufficient to distinguish MSI-H tumours, when compared to corresponding normal tissues (Hoang et al. 1997, Aaltonen et al. 1998). MSI-H tumours have a lower frequency of K-ras, TP53, and β-catenin mutations and less LOH on chromosomes 5q, 17p, and 18q than MSS tumours (Jass et al. 1999). Jass et al. (1999) showed that MSI-L tumours could be distinguished from MSS tumours by a higher frequency of K-ras mutations and a lower frequency of 5q LOH. Recently, Laiho et al. (2002) suggested that both MSS and MSI-L tumours could be classified by MSS status. K-RAS mutations, MGMT and MLH1 promoter methylation, and LOH at APC, TP53 or SMAD4 loci did not differ between MSS and MSI-L tumours (Laiho et al. 2002). 5.1.3.1 Microsatellite instability target genes Repeat sequences of numerous genes may be targets for frameshift mutations as a result of MMR deficiency. These sequences are mostly mononucleotide repeats. The frameshift mutations lead to truncated proteins with impaired function, possibly contributing to tumour development (Table 4). The first MSI target genes proposed in HNPCC were TGFβ-RII (Markowitz et al. 1995), IGFIIR (Souza et al. 1996), and BAX (Rampino et al. 1997). Since then, many other genes with mutations in repeat tracts have been reported to be results of MMR defects. In 2002, a total of 110 publications contained 245 coding and non-coding mononucleotide microsatellite analyses of 177 genes in MSI colorectal, gastric, or endometrial cancers (Woerner et al. 2003). MSI target genes encode proteins destined for such different roles as signal transduction, apoptosis, DNA repair, transcriptional regulation, protein translocation and modification, and immune surveillance (Woerner et al. 2003). The somatic mutation frequencies of proposed MSI target genes vary from <10% to >60%, with the exception of TGFβ-RII, which is mutated in approximately 90% (Markowitz et al. 1995, Parsons et al. 1995) of MSI CRCs (Table 4). The character of microsatellites affects the mutation rate. Differences in length of the repeat unit, base composition of the repeat, number of repeat units, possible interruptions in the tract, and type of flanking sequences cause variable mutation frequencies of repeats (Boyer et al. 2002). Thus, different models have been provided to distinguish real target genes from bystander or background mutations. The nature of real target genes can be confirmed by high mutation frequency and by in vitro or in vivo functional studies (Boland et al. 1998, Duval et al. 2001). Other criteria have been discussed and

26

Table 4. Mutations of coding mononucleotide repeats in MSI CRCs (at least 9% mutation frequency reported in the first publication). Gene Coding

repeat Mut %

Function Reference

Β2-microglobulin A5 14 Immune surveillance Bicknell et al. 1996 PTEN A6 x2 19 Cell cycle Guanti et al. 2000 AXIN-2 A6 x2, G7, C6 24 Wnt signalling Liu et al. 2000 FAS T7 10 Proapoptotic factor Yamamoto et al. 2000 CYSLT1 A8 9 Inflammatory response Mori et al. 2001 TPRDI A8 9 Putative extracellular matrix component Mori et al. 2001 Actin depolymerizing factor

T8 10 Acts in actin depolymerization Mori et al. 2001

CDC25C A8 11 Triggers entry into mitosis Park et al. 2002 Protein-tyrosine-phosphatase D1

A8 12 Intracellular signalling Mori et al. 2001

APAF-1 A8 13 Proapoptotic factor Yamamoto et al. 2000 BCL-10 A8 13 Proapoptotic factor Yamamoto et al. 2000 Coagulation factor VIII:C

A8 15 Coagulation factor Mori et al. 2001

BDH A8 16 Maintains cytoskeletal structure Shin et al. 2003 RACK7/PRKCBP1 A8 19 Intracellular signalling Park et al. 2002 GARE T8 20 Endocrine signalling Laghi et al. 2002 hG4-1 A8 21 Cell cycle Mori et al. 2001 MSH3 A8 39 DNA repair Malkhosyan et al. 1996 ACTRII A8 58 Growth factor signalling Mori et al. 2001 PRRG1 C8 9 Homology to coagulation factors Mori et al. 2001 IGFIIR G8 9 Growth factor signalling Souza et al. 1996 SEX G8 14 Growth factor signalling Mori et al. 2001 TFE3 C8 23 Transcription factor Potocnik et al. 2003 RGS12 C8 29 Regulator of G protein signalling Potocnik et al. 2003 MSH6 C8 30 DNA repair Malkhosyan et al. 1996 TEF4 C8 32 Transcription enhancer factor Potocnik et al. 2003 TCF1 C8 32 Transcription factor Potocnik et al. 2003 BAX G8 51 Promotes apoptosis Rampino et al. 1997 ERCC5 A9 10 DNA repair Park et al. 2002 CHK1 A9 10 Cell cycle Bertoni et al. 1999 MCT T9 11 Intracellular transport Mori et al. 2001 RECQL1 A9 12 DNA repair Park et al. 2002 Exportin t T9 14 Translation Mori et al. 2001 CBF2 A9 16 Transcription factor Park et al. 2002 BLM A9 18 Response to DNA damage Calin et al. 1998 RHAMM A9 19 Cell motility Duval et al. 2001 GRK4/GPRK2L A9 21 Intracellular signalling Park et al. 2002 NADH-UOB T9 28 Cell metabolism Mori et al. 2001 WISP-3 A9 31 Growth factor Thorstensen et al. 2001 RAD50 A9 33 Response to DNA damage Kim et al. 2001 GRB-14 A9 34 Growth factor signalling Duval et al. 2001 RIZ A9, A8 37 Cell cycle Chadwick et al. 2000 TCF-4 A9 39 Transcription factor Duval et al. 1999 SEC63 A9, A10 49 ER membrane protein Mori et al. 2001 SLC4A3 C9 33 Anion exchanger Woerner et al. 2001 SLC23AI C9 45 Nucleobase transporter Woerner et al. 2001 GART A10 24 Cell metabolism Woerner et al. 2001 MBD4 A10 26 DNA repair Riccio et al. 1999 DNAPK/PRKDC A10 33 DNA repair and recombination Park et al. 2002 MAC30X A10 33 Cytosceletal functions Woerner et al. 2001 PRKDC A10 35 Intracellular signalling Woerner et al. 2001 OGT T10 40 Cell metabolism Woerner et al. 2001 AIM2 A10 48 Inflammatory response Mori et al. 2001 Caspase-5 A10 62 Proapoptotic factor Schwartz et al. 1999 TGF-βRII A10 82 Growth factor signalling Markowitz et al. 1995

27

questioned. First, a role for candidate target genes in cellular pathways was speculated to be important. However, this role cannot necessarily be explained since its cellular mechanism may as yet be unknown. Second, inactivation of both alleles has been emphasized. Inactivation of the target gene may occur either by biallelic alteration of the repeat or by imprinting (Boland et al. 1998, Perucho et al. 1999, Duval et al. 2001, Duval and Hamelin 2002). Haploinsufficiency or the dominant negative effect of MSI target genes (Liu et al. 2000) explains the importance of only one mutated allele. To cover all types of genes with mutations in repeat tracts, Duval and Hamelin (2002) divided MSI target genes into four groups. First, normal expression of “survivor” genes is needed in the cell, and thus mutations will be avoided. Second, the “hibernator” genes do not have important functions in affected cells, and their expression is down-regulated. Mutations in these genes are not selected for, and the mutation rate would show up as background mutations. Third, mutations in “co-operator” genes could have importance in combination with other mutations in the same pathway. Variable mutation frequencies of these genes can be obtained. Fourth, mutations of “transformator” genes provide a positive growth advantage for MSI cells and have the highest mutation frequencies. Most of the MSI target gene mutations accumulate in cells, eventually leading to uncontrolled growth. In MSI tumour progression, TGFβ-RII inactivation was often found to be the second event after MMR gene mutation, and in some cases, MSH3 was the third target of mutation. The MMR phenotype is modified by the mutated MSH3, which further increases mutations in nucleotide repeats (Grady et al. 1998, Duval et al. 2001). 5.2 Base excision repair mechanism Reactive oxygen species (ROS) mainly produced by aerobic cell metabolism but also by radiation, macrophages, and chemical agents damages DNA (Aruoma et al. 1989, Ames et al. 1993, Henle and Linn 1997). BER is used in the repair of methylated or deaminated DNA bases and is also proposed to be the primary mechanism responsible for the repair of oxidative damage in DNA. In addition to mutagenesis and further carcinogenesis, oxidized DNA bases are known to play a role in ageing and in neurological disorders (Ames et al. 1993). Like many other DNA repair systems, the BER system is well conserved in different organisms. DNA glycosylases identify specific types of base damage. There are over ten known DNA glycosylases (Mohrenweiser et al. 2003) which hydrolyse the N-glycosidic bond between the base and sugar-phosphate backbone, thus releasing the damaged base. An apurinic/apyrimidinic (AP) endonuclease nicks the sugar-phosphate backbone at the position of the missing base. Exonucleases remove some nucleotides from the DNA strand and the DNA polymerase fills the gap. In all, over thirty different genes involved in BER have been described (Mohrenweiser et al. 2003). Repair of oxidized guanine lesions is an example of BER system function. Oxidized guanine is highly mutagenic and one of the main causes for spontaneous mutations. In DNA replication, an adenine residue is more frequently added opposite to an oxidized guanine lesion 8-oxoG than is a cysteine residue. This leads to GC→TA transversions

28

(Cooke et al. 2003). E. coli glycosylases mutY and mutM have been reported to repair oxidized guanine lesions 8-oxoG in DNA (Cabrera et al. 1988, Michaels et al. 1992, Nghiem et al. 1998). Human homologues of mutM (OGG1) (Roldan-Arjona et al. 1997) and mutY (MYH) (Slupska et al. 1996) have also been identified. The role of OGG1 is to remove the 8-oxoG opposite to a cysteine residue from the DNA (Figure 4). All damage is not immediately repaired, but after replication, MYH removes an adenine residue across from 8-oxoG (Boldogh et al. 2001). Polymerases insert a cysteine residue in place of the removed adenine, and OGG1 can then repair the lesion. The role of MutT (MTH1) is to hydrolyse the oxidized dGTPs to unusable monophosphates, thus preventing the formation of a mutated DNA strand (Maki and Sekiguchi 1992, Sakumi et al. 1993). Proteins other than MYH, OGG1, and MTH1 have also been detected with repair activity against 8-oxoG in vitro. For example, the mammalian alkylpurine DNA glycosylase (Bessho et al. 1993) and the human NER system (Reardon et al. 1997, Le Page 2000, Tuo 2001) have been reported to have such functions. The human Nei-like glycosylase 1 (NEIL1) has also been shown to preferentially remove the 8-oxoG opposite to an adenine (Hazra et al. 2002). C G Oxidative damage C GO Replication GO A OGG1 GO C A MYH Repair GO C G Repair C GO

Figure 4. Oxidative damage results in the formation of a 8-hydroxyguanine (GO), which can be removed by the OGG1 protein. During replication GO mispairs with an adenine residue, which can be removed by MYH. Modified from Slupska et al. (1996). 5.2.1 Defects in base excision repair genes Mutations and polymorphisms of OGG1 on chromosome 3p26 have been detected in the lung, kidney (Chevillard et al. 1998), ovarian and endometrial tumours, and one leukemia cell line (Pieretti et al. 2001, Hyun et al. 2002). Recently, it was also suggested that polymorphism of OGG1 may alter the impact of environmental factors on colon cancer development (Kim et al. 2003). Studies of animal models showed that mutation

29

frequency increased slightly in some tissues of Ogg1-/- mice, but no tumours were detected. This indicates the role of various enzymes in repairing 8-oxoGs (Klungland et al. 1999, Minowa et al. 2000). Increased expression of MTH1 on chromosome 7p22 and accumulation of 8-oxoG have been detected in human brain tumours (Okamoto et al. 1996, Iida et al. 2001); however, the role of MTH1 in CRC has not been shown. Mice models lacking the Mth1 gene harbour spontaneous carcinogenesis in the lung, liver, and stomach (Tsuzuki et al. 2001). 5.2.1.1 MYH Mutations in MYH on chromosome 1p32-p34 were recently detected in CRC cases. Al-Tassan et al. (2002) showed that multiple somatic transversions in the APC gene were the result of MYH deficiency in patients with colorectal adenomas and carcinomas. Two amino acid variants of MYH were identified from three affected siblings, and compound heterozygosity of mutations revealed a recessive trait. Recent studies have reported more biallelic germline mutations in MYH and described multiple adenomas in these patients (Jones et al. 2002, Halford et al. 2003, Sieber et al. 2003). Functional evidence of mutated MYH was represented by a greater than 80% reduction of enzyme function. MYH mutations almost totally reduced adenine removal from G:A substrates in E. coli (Al-Tassan et al. 2002). The mutations were suggested to occur in conserved areas of glycosylase, probably affecting catalytic activities and the recognition of mismatches.

30

AIMS OF THE STUDY Evaluation of somatic microsatellite mutations caused by mismatch repair deficiency: 1. Evaluation of normal tissues in an MLH1-/- patient. 2. Detection of the mutation frequencies of mononucleotide repeats in colorectal

cancers. Evaluation of germline mutations in DNA repair genes and their association with colorectal cancers: 3. Identification of the mutation status of mismatch repair genes MLH3 and EXO1. 4. Evaluation of the frequency of MYH founder mutations and further

characterization of the phenotype of mutation carriers.

31

MATERIALS AND METHODS 1. PATIENTS AND DNA SAMPLES 1.1 MLH1-/- patient (I) A four-year-old girl with café au lait spots, glioma, and strong family history of HNPCC (both parents mutation carriers) was suspected of MLH1 deficiency. The patient died of intracranial haemorrhage. Tissue samples of the brain, kidney, heart, liver, bladder, lymph node, and glioma were obtained at autopsy. The samples were received from Oulu University Hospital. The study was approved by the Ethics committee of the Department of Medical Genetics, University of Helsinki. 1.2 Colorectal carcinoma patients (II-V) A set of 1042 unselected Finnish CRC samples was the basis for CRC Studies II-V (Table 5). This set has been described in detail elsewhere (Aaltonen et al. 1998, Salovaara et al. 2000). The diagnoses of cancers and the possible connection to identified HNPCC families were obtained from the Finnish Cancer Registry, the Finnish HNPCC Registry and the Finnish Polyposis Registry. The study was approved by the Ethics committee of the Department of Medical Genetics, University of Helsinki. 1.2.1 Searching for microsatellite instability target genes (II) A panel of 93 MSI colorectal tumours was used to study intronic mononucleotide repeats and known MSI target genes IGFIIR and BLM. Thirty MSI colorectal tumours were used in screening for mutations of eight possible MSI target genes. An additional 79 MSI tumour samples were used to sequence the repeat tract of CtIP. A summary of all CRC samples used in Studies II-V is presented in Table 5. 1.2.2 MLH3 analysis (III) Normal DNA of 46 MSI and six MSS CRC patients was used for MLH3 germline analysis. Of these patients, 43 were selected because of at least one first- or second-degree relative with cancer and three for their young age. Six patients whose MSS colorectal tumours showed 14q deletions were also used for MLH3 analysis. These patients had at least one first-degree relative with CRC. Ninety-three MSI tumours were screened for repeats of MLH3 (Table 5). 1.2.3 EXO1 analysis (IV) The following three groups of Finnish CRC patients were selected for EXO1 germline mutation analysis (Table 5): 1) Normal DNA from 19 CRC patients (18 families), 17 of

32

which had at least one first-degree relative, were sequenced for EXO1 mutations. Tumours of eight patients were MSI with no germline MLH1 or MSH2 mutation found. 2) Samples from 111 MSS or MSI-L CRC patients with a first-degree relative with CRC, 67 MSS CRC patients with early age of onset (≤50 years) and unknown family history, and 122 CRC patients aged over 50 years and without family history of CRC were used in EXO1 single-strand conformational polymorphism (SSCP) analysis. 3) Samples from 80 patients with familial CRC and 3 patients with early onset of cancer (≤50 years) were studied for the IVS12-1 variant. CRC patients from the Netherlands, Norway, the UK, and Denmark were included in Study IV (Table 5). Of the 116 Dutch and 39 Norwegian families (1 sample/family), 55 fulfilled the revised Amsterdam criteria, 68 were suspected HNPCC families, and 32 were identified with late onset of CRC. MSH2, MLH1, or MSH6 mutations were excluded. Four hundred and eighteen CRC cases from the UK were diagnosed before patients reached the age of 56 years. Danish CRC samples were from 97 suspected HNPCC patients. 1.2.4 MYH analysis (V) Analysis of MYH mutations was based on a set of 1042 unselected DNA samples from CRC patients (Aaltonen et al. 1998, Salovaara et al. 2000). MSI was discovered in 130 tumours, and MLH1 or MSH2 germline mutations were detected in 29 samples (Aaltonen et al. 1998, Salovaara et al. 2000, unpublished data). Normal DNA from 1003 patients was available for the MYH study (Table 5). 1.3 Control individuals (I, III-V) Colon harbouring a normal MMR mechanism was used as control tissue in Study I. The Finnish Red Cross of Helsinki provided control DNA samples from anonymous blood donors in Studies I, III, IV, and V. In addition to Finnish controls, anonymous controls or spouses of CRC patients from the Netherlands, Norway, the UK, and Denmark were used in Study IV. 1.4 DNA extraction (I-V) DNA from fresh frozen tumours and blood samples was extracted using standard procedures (Lahiri and Nurnberger 1991). Histological analysis of tumour samples was used to confirm that a maximum amount of tumour tissue was used for DNA extraction. DNA from the brain, kidney, heart, lymph node, colon, and bladder epithelium (Study I) obtained at autopsy was extracted according to Shibata et al. (1996).

33

Table 5. CRC samples used in Studies II-V. Number of samples

Family history of cancer Age of onset

MSI status of tumour

Mutation analysis

Study Gene

93 - - MSI PAGE II MSI targets, intronic repeats

30+79 - - MSI Direct sequencing II MSI targets 43 At least one first- or

second-degree relative with cancer

- MSI SSCP III MLH3

3 - Mean 40 y

MSI SSCP III MLH3

6 At least one first-degree relative with CRC

- MSS SSCP III MLH3

93 - - MSI PAGE III MLH3, OBR 19 17 had at least one first-

degree relative with CRC 45-84 y MSI or MSS

Direct sequencing IV EXO1

111 At least one first-degree relative with CRC

- MSS or MSI-L

SSCP IV EXO1

67 - ≤50 y MSS SSCP IV EXO1 122 No family history >50 y Sporadic SSCP IV EXO1 80 Familial CRC - MSI or MSS Restriction

analysis IV EXO1

3 - ≤50 y - Restriction analysis

IV EXO1

55 Revised Amsterdam criteria

- - DGGE IV EXO1

68 Suspected HNPCC families

- - DGGE IV EXO1

32 - Late - DGGE IV EXO1 418 Family history of CRC <56 y CSGE IV EXO1 97 Suspected HNPCC

families - - dHPLC IV EXO1

1003 - - MSI or MSS Solid phase minisequencing

V MYH

- Unknown data

2. STUDY OF THE MLH1-/- PATIENT (I) 2.1 MLH1 mutation 1 analysis Mutation 1 of MLH1 is an in-frame 165 base pair (bp) deletion of exon 16. This 3.5 kb genomic deletion in MLH1 was detected by polymerase chain reaction (PCR) and agarose gel electrophoresis. Three primers (Nyström-Lahti et al. 1995) were used to identify the 634 bp mutated allele (primers 1+2) and the 475 bp normal allele (primers 2+3). With primers 1 and 2, the normal allele is about 4 kb and thus not produced. 2.2 Microsatellite analysis DNA from the heart, kidney, lymph node, and bladder epithelium was diluted for radioactive microsatellite analysis with markers DXS556, DXS1060, and DXS453. PCR reactions were labelled by incorporating (α-33P)dCTP (NEN) in the synthesized strands, and products were analysed by polyacrylamide gel electrophoresis (PAGE) (Tsao et al.

34

2000). Between 42 and 56 molecules were genotyped for each tissue sample at each locus. Allele sizes of markers DXS556, DXS1060, and DXS453 in different tissues were compared with allele sizes in normal tissue. Variances of allele size distributions under the stepwise mutation model were considered to represent the numbers of mutations at the loci (Tsao et al. 2000). The number of cell divisions equals the number of mutations per mutation rate. Mutations from already mutated alleles back to normal allele size cannot be excluded or estimated in this quantitative analysis. TGFβRII primers were used to detect frameshifts in the A10 repeat of the tumour suppressor gene (Loukola et al. 1999) in normal lymphocytes of the MLH1-/- patient and controls. DNA samples from the MLH1-/- patient were diluted to the point at which PCR amplification detected alleles from a single cell. 3. SEARCH FOR MICROSATELLITE INSTABILITY TARGET GENES (II) 3.1 Mononucleotide repeats in introns and known microsatellite instability target genes Background mutations of 14 intronic (A/T)6-9 and (G/C)6-9 repeats were studied by radioactive PCRs and PAGE (Table 5). The set of 93 MSI colorectal tumours was used to detect deletions or insertions of mononucleotide repeats. Radioactive (α-32P)dCTP (Amersham Pharmacia Biotech) PCR products were run on 6% polyacrylamide gels. After exposure, all aberrant bands were sequenced and somatic mutation was confirmed by analysing the corresponding normal DNA. The known MSI target genes IGFIIR and BLM were screened for mutations in the G8 and A9 mononucleotide repeats, respectively, using the same method. 3.2 Search for putative microsatellite instability target genes Bioinformatic tools were used in the search for novel candidate MSI target genes. Mononucleotide repeats of eight base pairs or more were sought from coding sequences added to the GenBank since 1997. Eighty-three keywords related to cancer were combined with found repeat sequences and their Medline records, resulting in 251 sequences. The other search was based on sequence homology of repeat-containing genes to 34 known tumour suppressor or CRC-related genes. This effort resulted in 841 hits. A small group of candidate MSI target genes was selected according to abstracts (Study II, Table 2). Thirty MSI CRCs were used in sequencing the repeat area of eight possible MSI target genes. The primer sequences and PCR conditions are described in Study II (Table 2). An additional 79 MSI samples were used to sequence the repeat tract of CtIP. The nature of somatic mutations was confirmed in the presence of normal DNA.

35

3.3 Statistical analysis Logistic regression analysis was used in estimating the correlation of the independence and significance of mutation frequency with repeat length in the intron. χ2 test was used for other analyses. p<0.05 was considered statistically significant. 4. MLH3 MUTATION ANALYSIS (III) 4.1 Single-strand conformational polymorphism analysis Single-strand conformational polymorphism (SSCP) analysis was performed with normal DNA of 52 CRC patients (Table 5). MLH3 was amplified in 24 fragments (Study III, Table 1). The PCR products were denaturated and run on a 0.6x mutation detection enhancement (MDE) gel (FMC BioProducts). MDE gels were silver-stained according to standard procedure, and all aberrant bands were sequenced. 4.2 Radioactive polyacrylamide gel electrophoresis Somatic mutation analysis was carried out by radioactive PAGE analysis as described earlier (Section 3.1). Ninety-three MSI tumours (Table 5) were screened for coding A6-A9 mononucleotide repeats of MLH3 (6xA6, 1xA8, 1xA9) (Study III, Table 2) and non-coding A8 and A9 repeats of OBR. 5. EXO1 MUTATION ANALYSIS (IV) 5.1 Direct sequencing, single-strand conformational polymorphism analysis, and restriction analysis Nineteen CRC patients from 18 families (Table 5) and 20 controls were sequenced for 14 exons of EXO1 using an ABI 3100 DNA sequencer (AB). Eight candidate mutations detected by Wu et al. (2001a) were screened from 300 Finnish CRC patients and 373 controls by SSCP (method described earlier in Section 4.1). Detected variants were confirmed by direct sequencing. The NlaIV restriction enzyme (New England Biolabs) was used according to manufacturer’s instructions to screen the ninth mutation, truncating variant IVS12-1 (G/C), from 83 Finnish CRC patients and 504 controls. 5.2 Denaturing gradient gel electrophoresis, conformation-sensitive gel electrophoresis, and denaturing high-performance liquid chromatography Denaturing gradient gel electrophoresis (DGGE) (Fodde et al. 1992) was used to screen the EXO1 gene in 155 Dutch/Norwegian families. Twenty-four primer pairs were used according to the study by Wu et al. (2001a). PCR products were run on a 9% polyacrylamide gel with a denaturing gradient of 30-60% urea/formamide. Conformation-sensitive gel electrophoresis (CSGE) (Ganguly et al. 1993) was used to detect three candidate mutations of exon 11 (Ser610Gly, Pro640Ser, Pro640Ala) in the

36

UK cohort. Primers were described by Wu et al. (2001a). PCR products were run on a 6% polyacrylamide gel containing 15% formamide and 10% ethylene glycol. The DGGE and CSGE gels were stained with ethidium bromide, and bands were detected in UV light. Danish samples and controls were screened by denaturing high-performance liquid chromatography (dHPLC) for the three (Ser610Gly, Pro640Ser, Pro640Ala) exon 11 mutations. dHPLC heteroduplex analysis was carried out on the Transgenomic Wave genetic analyser (Transgenomics). All variants detected by the various screening methods were sequenced to identify the mutation. Furthermore, EXO1 sequences of Homo sapiens, Mus musculus, Drosophila melanogaster, S. cerevisiae, and S. pompe were aligned using the Clustal W (1.82) multiple sequence alignment program (Thompson et al. 1994). 6. MYH FOUNDER MUTATION ANALYSIS (V) 6.1 Solid-phase minisequencing in detecting G382D and Y165C variants Solid-phase minisequencing in streptavidin-coated wells (Labsystems) was carried out to detect Caucasian G382D and Y165C (Al-Tassan et al. 2002) MYH variants in CRC patients. Genomic DNA from 16-22 patients was pooled for PCR, for a total of 52 pools covering DNA from 1003 patients (Table 5). Detection primers were hybridized next to the analysed nucleotide in PCR products, and bound specific H3-labelled nucleotides (Amersham) described the amount of normal or mutant alleles (Syvänen et al. 1993, Suomalainen and Syvänen 2000). PCR and minisequencing were repeated for positive pools. Double-positive pools were divided into smaller pools containing ten DNA samples each. After two positive results from small pools, reactions with individual DNA were carried out. χ2 test was used for examining MYH mutation distribution between CRC patients and controls. p<0.05 was considered statistically significant. 6.2 Direct sequencing, single-strand conformational polymorphism analysis, and denaturing high-performance liquid chromatography To search for additional MYH defects, patients with only one mutated allele in the germline were sequenced for all exons in both normal and tumour DNA with primer sequences as described earlier by Al-Tassan et al. (2002). Mutation cluster (nt 1286-1513) of the APC gene was examined in tumours of MYH mutation carriers. Exons 15A to 15L were screened by fluorescent SSCP and dHPLC analyses. For more detail, see Study V.

37

RESULTS 1. MICROSATELLITE INSTABILITY ANALYSIS (I) A patient with glioma, a neurofibromatosis phenotype, and a HNPCC family history was diagnosed with a homozygous 165 bp deletion in MLH1 exon 16. Both parents harboured the heterozygous mutation. DNA from the heart, kidney, lymph node, and bladder epithelium was diluted to amplify alleles from a single cell. DNA from the brain was too degraded for analysis. CA-repeats detected by markers DXS556, DXS1060, and DXS453 showed allele size variation in different autopsy tissues. From seven to 15 mutant alleles were detected, when between 42 and 56 molecules were genotyped for each tissue sample at each locus (Study I, Table 1). Somatic mutations were present in 15-32% of alleles in different autopsy tissues. Control tissue, colon harbouring a normal MMR mechanism, showed only one mutant allele (marker DXS556). Thus, less than 5% of mutant alleles were detected by the three microsatellite markers. The allele sizes, presented as a number of CA-repeats, varied from 24 to 31 as detected by the unimodal marker DXS556, from 18 to 26 by bimodal marker DXS1060, and from 16 to 28 by bimodal marker DXS453 (Study I, Figure 3A). Microsatellite allele frequency distributions were identical between tissues. Variances of allele frequency distributions ranged between 0.26 and 1.4: in the heart 0.36-1.0, in the kidney 0.26-0.59, in lymph nodes 0.32-0.95, and in bladder epithelium 0.38-1.4 (Study I, Table I). In other words, approximately equal numbers of mutations seemed to occur in different tissues of the MLH1-/- patient. A total of 0.26-1.4 mutations per locus per cell had occurred, assuming that variances of microsatellite mutation frequency distributions are proportional to numbers of microsatellite mutations (Tsao et al. 2000). Numbers of cell divisions in tissues (= number of mutations/mutation rate) were calculated by using the microsatellite mutation rate 5x10-3 per cell division (Bhattacharyya et al. 1994, Shibata et al. 1994). Thus, MLH1-/- cells had divided approximately 52-280 times. An A10 mononucleotide repeat of TGFβRII was studied in DNA samples aquired from peripheral lymphocytes of the MLH1-/- patient. Diluted DNA showed amplification in 134 PCRs out of a total of 221. One base pair deletion of the repeat was found in one TGFβRII PCR, but also in one out of 96 control PCRs from blood donor DNA. 2. MUTATIONS OF MONONUCLEOTIDE REPEATS (II) The mutation frequencies of 14 intronic (A/T)6-9 and (G/C)6-9 repeats are summarized in Study II (Table 1). Mutation frequencies varied between 0% and 5.7%. By logistic regression analysis, the length of G/C and A/T intronic mononucleotide repeats correlated positively with the number of mutations (p=0.0020 and p=0.0012, respectively).

38

The known MSI target genes IGFIIR (coding G8 repeat) and BLM (coding A9 repeat) showed mutations in nine out of 92 (9.8%) and 18 out of 93 cases (19.4%), respectively. Mutation frequencies in BLM (19.4%) and intronic A/T9 repeats (4.3-5.7%) differed significantly (p≤0.001), and when results of earlier studies (Souza et al. 1996, Ouyang et al. 1997, Calin et al. 2000) were combined with findings made here, the IGFIIR mutation rate (28/216=13%) was also found to differ significantly from the intronic G/C8 mutation rate (2.5%) (p≤0.01). A search for putative MSI target genes using keywords related to cancer and tumour suppressors or CRC-related sequence homology, resulted in 251 and 841 hits, respectively. The eight genes BCL10, PTP-BAS, CtIP, BAI1, p73, GIOT-2, PAIP, and RIP3 were screened for mutations in exonic mononucleotide repeats (Study II, Table 2). CtIP had mutations in 22.9% of MSI tumours (25/109). No mutations were detected in the other seven genes. Comparison of CtIP with intronic A/T9 mutations (4.3-5.7%) revealed a significant difference (p≤0.001). Combined data of CtIP mutations from this study and Ikenoue et al. (2001) (25/122=20.5%) were also compared with intronic A9 mutations, and the significant difference remained (p≤0.001). No mutations in other exonic 5-6 bp mononucleotide repeats of CtIP were found. Tumours of CtIP mutation carriers did not differ in their histology from other MSI CRCs. In addition, age distributions at diagnosis and tumour site distributions were similar. CtIP mutations were observed in similar numbers in men (14/49) and women (11/60). 3. MLH3 MUTATION ANALYSIS (III) No MLH3 germline mutations were found in 52 CRC patients by SSCP. Two exonic variants, G1258A (Val420Ile) and G4377A (Gln1421Gln), were detected both in CRC patients (allele frequencies 2.9% and 46%, respectively) and healthy controls (allele frequencies 2.7% and 49%, respectively). A third exonic change, A1234G (Lys412Glu, allele frequency 3.8%), was detected as a result of sequencing somatic mutations in tumours; it was also found in controls (allele frequency 1.1%). One intronic variant, IVS9+55 G to C (allele frequency 2.9%), was revealed by SSCP, and another intronic change, IVS8+66 A to G (allele frequency 50%), was uncovered by sequencing SSCP variants. Somatic deletions of MLH3 mononucleotide repeats were found in eight out of 93 cases (allele frequency 8.6%) when studying eight mononucleotide repeats (6xA6, 1xA8, 1xA9) in coding regions. One mutation was found in an A8 repeat (nt 2128), and seven were detected in an A9 repeat (nt 1861), resulting in a premature stop codon. The OBR gene was used as a control in the screening of somatic mutations in intronic mononucleotide repeats: three deletions in an A8 repeat (allele frequency 3.2%) and three deletions in an A9 repeat (allele frequency 3.2%) were present in the same set of 93 MSI tumours.

39

4. EXO1 MUTATION ANALYSIS (IV) A summary of all EXO1 variants found in Finnish, Dutch, Norwegian, British, and Danish data sets and in the study by Wu et al. (2001a) is presented in Study IV (Table 1). Comparing EXO1 variants in 19 Finnish CRC patients and 20 cancer-free controls revealed that a Pro640Ser change was the only remaining putative mutation. To further evaluate nine mutations reported by Wu et al. (2001a), 970 familial and sporadic CRC cases (Table 5) and 1007 controls were studied by SSCP, DGGE, CSGE, dHPLC, and restriction analyses. Five of the mutations (Ser610Gly, Pro640Ser, Pro640Ala, truncating variant IVS12-1 G/C, Gly759Glu) presented in CRC cases in Wu et al. (2001a) were found both in CRC patients (allele frequencies 0.3%, 0.6%, 0.4%, 0.2%, and 0.8%, respectively) and controls (allele frequencies 0.07%, 0.5%, 0.3%, 0.2%, and 1.9%, respectively) in the present study. Four mutations found by Wu et al. (2001a), Val27Ala, Glu109Lys, Leu410Arg, and Pro770Leu, were not detected in controls. Neither were three of these mutations (Val27Ala, Glu109Lys, Pro770Leu) found in CRC probands of this study. Only one Leu410Arg mutation was found in CRC patient (allele frequency 0.2%). LOH as a second hit in the tumour was not detected in exon 11 variants (Ser610Gly, Pro640Ser, Gly670Gly) from seven Finnish CRC patients, when comparing alleles in tumours and normal tissues by sequence intensities. In addition, five new unpublished variants were detected: one in controls, three in CRC patients only, and one in both controls and CRC patients. A G/A change in place of truncating variant IVS12-1 G/C was found in a Finnish control. An Ala827Val change was found in a Dutch CRC patient and control. New CRC patient variants not present in controls were Ala137Ser, Phe438Cys, and a silent Asp661Asp variant. These were detected in Dutch/Norwegian patients. Combined data from this study and from Wu et al. (2001a) identify six amino acid changes detected in only CRC probands (Val27Ala, Glu109Lys, Ala137Ser, Leu410Arg, Phe438Cys, Pro770Leu). According to the EXO1 sequence alignment of Homo sapiens, Mus musculus, Drosophila melanogaster, S. cerevisiae, and S. pompe, all six changes affected non-conserved amino acids of the EXO1 protein. 5. MYH GERMLINE MUTATION CARRIERS (V) Three Y165C/G382C compound heterozygotes and one G382D homozygous mutation carrier were found when searching for MYH variants in 1003 samples (0.4%) from a population-based series of 1042 Finnish CRC patients. Furthermore, five patients harboured either mutation Y165C or G382C (Study V, Table 2). These patients with only one mutant allele showed no other mutations of the gene in the germline or in the tumour as evidenced by sequencing. Allele frequency of mutated MYH was 0.006 (13/2006) when the Y165C mutation was present in four alleles (0.002) and the G382D mutation in nine alleles (0.004). MYH founder mutations were not found among the 424 cancer-free controls, and the difference between CRC patients and controls reached significance (p<0.025).

40

The phenotype of mutation carriers displayed some variation, but four patients with two mutated alleles had five to one hundred polyps, and patients with only one mutated allele had either none, one, or more than six polyps (Study V, Table 2). No particular type of malignancy was observed in first-degree relatives of the nine CRC patients with MYH mutations. Eight out of nine mutation carriers had at least one first-degree relative with some form of cancer, and of these, two probands had three and five first-degree relatives suffering from cancer (Study V, Table 2). Tumours other than CRC were not available for this study. Of the set of 1042 Finnish CRC cases, 35 (3.4%) were shown to have a genetic predisposition to cancer (Study V, Table 3). Mutations in five genes (MLH1, MSH2, APC, ALK3, MYH) have been detected. According to the Finnish Polyposis Registry, four FAP cases were diagnosed during the CRC collection from May 1994 to July 1998. During this same period, four patients with MYH double mutations were detected. Thus, the frequency of APC and MYH mutations seems to be equal in the Finnish population.

41

DISCUSSION 1. MICROSATELLITE MUTATIONS PRECEDE VISIBLE TUMORIGENESIS (I) The MLH1-/- (exon 16 del) female patient died at the age of four years of a haemorrhage caused by glioma. Café au lait spots and axillary freckles were also detected. The phenotype of the MLH1-/- patient was similar to other cases described earlier (Gly67Trp, Arg226Stop) (Ricciardone et al. 1999, Wang et al. 1999a). In two families, five patients were found with café au lait spots, dermal neurofibromas, pseudoarthrosis, and axillary freckles common in neurofibromatosis type 1 (Ricciardone et al. 1999, Wang et al. 1999a). In four patients, lymphomas or leukemias were diagnosed before the age of three. In one patient, leukemia was diagnosed at the age of six. However, haematological malignancies were not diagnosed in the patient of Study I. In addition to homozygous MLH1 mutation carriers, a woman with compound heterozygous missense mutations was reported by Hackman et al. (1997). This patient was diagnosed with breast cancer at the age of 35, and the mild phenotype was suggested to be the result of remaining activity of the MLH1 allele. Recently, a patient with a homozygous MSH2 mutation was also reported. This patient had MLH1-/--like symptoms: café au lait spots and leukemia (Whiteside et al. 2002). Gliomas, usually astrocytomas or glioblastomas, may associate with HNPCC (Hamilton et al. 2000). However, glioma of the MLH1-/- patient is not necessarily a HNPCC tumour. DNA from the brain or glioma was not available for Study I. MMR-deficient patients and mice have similar phenotypes of haematological malignancy, and thus, MMR-deficient mice models are important for studies of MMR-deficient patients. Mlh1-/- mice have been found to develop early invasive lymphomas, and the majority of mice surviving more than half a year developed adenomas and adenocarcinomas of the intestine (Edelmann et al. 1996, Prolla et al. 1998). The phenotype of Msh2-/- mice was similar to that of Mlh1-/- mice (de Wind et al. 1995, Reitmair et al. 1995, 1996). A rate of 2 x 10-7 mutations per locus per cell division has been estimated in normal somatic cells (Loeb 2001). Similar mutation rates have also been observed in MSS sporadic cancers (Wang et al. 2002). Deficiencies in DNA repair do, however, lead to an increased rate of somatic mutations, and in MMR-deficient cells, base substitutions and short mononucleotide repeat deletions are increased approximately 100-fold (Narayanan et al. 1997, Andrew et al. 2000). CA-repeats at microsatellite loci have been shown to mutate even more (Bhattacharyya et al. 1994, Shibata et al. 1994). In this study, microsatellites in different tissues (heart, kidney, lymph node, bladder epithelium) of the MLH1-/- patient mutated between 0.26 and 1.4 times per microsatellite locus per cell. This reveals a great mutation frequency compared with MMR-deficient tissue culture studies (approximately 5 x 10-3 mutations/locus/cell) (Bhattacharyya et al. 1994, Shibata et al.

42

1994). Consistent with the findings of this MLH1-/- patient study, phenotypically normal cells from Mlh1-/- mice (Edelmann et al. 1996) and humans (Wang et al. 1999a) have shown genetic instability. Despite the large numbers of non-coding microsatellite mutations in the MLH1-/- patient, differences in mutations between tissues were not detected. Differences could possibly have occurred over time due to the character of the several tissues and cell types used. However, because the patient died at the age of four years, there may have been insufficient time for mutations to accumulate. A molecular tumour clock analysis assuming the stepwise mutation model has shown that the majority of somatic microsatellite mutations accumulate before tumour formation (Tsao et al. 2000). Thus, the time for accumulation of mutations before a gatekeeper mutation could be long. Patterns of microsatellite mutations in CA-repeats have shown that loss of MMR occurs on average six years before tumour removal (Tsao et al. 2000), explaining the absence of tumours in the four-year-old MLH1-/- patient. When tissues of the MLH1-/- patient and MMR-deficient mice (Pms2) (Tsao et al. 1997) were compared, only at one year of age did the MMR-deficient mice show greater variances of microsatellite allele sizes in the intestine than the four-year-old MLH1-/- patient in any tissue. Furthermore, MLH1-/- human cancer (Tsao et al. 2000) showed higher variance of microsatellite allele frequency distribution than phenotypically normal tissues of the MLH1-/- patient. Taken together, these data indicate the importance of time in the accumulation of mutations and the potential usefulness of the molecular tumour clock model in the determination of occult changes. Different studies have shown that lack of MMR leads to haematological malignancy. In both MLH1-/- and MSH2-/- patients, downstream mutations could occur in genes such as NF1, causing haematological malignancies (Ricciardone et al. 1999). Various other target genes could also mutate in MMR-deficient patients, and tumours would likely progress over time. However, the role of possible mutations in the formation of tumours before loss of MMR and their absence in MMR-deficient patients remains to be elucidated (Huang et al. 1996, Miyaki et al. 1999, Kim et al. 2002). In this study, the putative mutability of repeat tracts in lymphocytes was also examined. TGFβ-RII is a gene mutated in approximately 90% of MSI CRCs (Markowitz et al. 1995, Parsons et al. 1995), but the mutation frequency of a coding A10 repeat in blood lymphocytes of the MLH1-/- patient was not increased when compared with the cancer-free control. This result is consistent with a recent study in which several microsatellite markers displayed no instabilities in lymphocyte DNA of an MMR-deficient patient (Whiteside et al. 2002). Mutations due to MMR deficiency occur during DNA replication. Thus, if lymphocytes divide infrequently, mutations in repeat tracts must be rare. Furthermore, mononucleotide repeats have been shown to mutate less frequently than microsatellite loci with CA repeats (Bhattacharyya et al. 1994, Shibata et al. 1994). In conclusion, in this study, large numbers of mutated non-coding microsatellites were shown to occur in phenotypically normal tissues of an MLH1-/- patient before any signs of tumour formation.

43

2. MICROSATELLITE INSTABILITY TARGET GENES (II) 2.1 Background mutations in intronic mononucleotide repeats and mutability of known microsatellite instability target genes Intronic mononucleotide repeats with different nucleotides and lengths have not been adequately studied for mutation rates. Only 25 non-coding microsatellites in 22 different human genes were presented in 33 publications in 2002 (Woerner et al. 2003). Here, the length of eight A/T6-9 and six G/C6-9 repeats correlated positively with the mutation rate of these repeats. The results also demonstrated that mutations occur sporadically in sequence repeats due to MMR deficiency, but unless the mutations are selected for, the mutation frequency appears to be low (<6%). Indeed, according to Suzuki et al. (2002), non-coding nucleotide repeat tracts (A8-9, G7-9) displayed few mutations (mutation frequencies 0-6.7%) in MMR-deficient tumours. However, intronic G8 repeats mutated more frequently than A8 repeats when a total of 29 repeats were examined (Zhang et al. 2001). Because monomorphic G8 and G9 repeats were difficult to find, only one non-coding repeat of each length was analysed. Comparisons of mutability between A and G repeats would have required more data on analysed repeats. However, independent tracts did mutate at different frequencies, similar to earlier reports (Zhang et al. 2001, Suzuki et al. 2002). Sequence or chromatin structure around the tract could possibly explain these differences (Zhang et al. 2001). In the previously described MSI target genes BLM (A9) and IGFIIR (G8), low mutation rates of 16-18% and 9.3-22% (Souza et al. 1996, Calin et al. 1998, 2000), respectively, were confirmed in this study by similar mutation rates (19.4% and 9.8%). Mutation frequencies of repeats in BLM and IGFIIR were significantly different from corresponding intronic repeats, demonstrating selection for these mutations in coding regions. The mutation rate, the biallelic inactivation, and functional studies in vitro have confirmed the MSI target gene character of IGFIIR. Mutated IGFIIR is not able to bind its ligand IGFII, disabling further activation of proteins involved in the control of cellular proliferation and apoptosis (Motyka et al. 2000). Germline mutations in the BLM gene have been detected in patients suffering from a hereditary cancer syndrome called Bloom´s syndrome, and heterozygous mutations of BLM have been reported to slightly increase CRC risk (Gruber et al. 2002). Mutations in the BLM gene that affect DNA unwinding lead, for example, to abnormal DNA replication and an elevated level of homologous recombination (Hickson 2003). 2.2 Putative microsatellite instability target genes Mononucleotide repeats of eight candidate MSI target genes BCL10 (T7), PTP-BAS (A8), CtIP (A9), BAI1 (C7), P73 (C6), GIOT-2 (T7), PAIP (T6), RIP3 (A6) were studied in MSI CRCs. The mutation frequency of 22.9% in the CtIP gene suggests that CtIP could be a putative novel MSI target gene. No mutations were detected in the other seven genes. Comparing the results of CtIP with A9/T9 intronic repeats used in this study indicated the potential of CtIP in MSI tumorigenesis. Mutation frequencies of CtIP and corresponding intronic repeats differed significantly. This suggests that CtIP mutations are advantageous for the cell and thus are selected for. The mutation rate of the A9 repeat

44

in CtIP was even higher than in the MSI target gene BLM (19.4%). Comparison of CtIP with previously proposed MSI target genes containing mutable A9 repeats (mutation frequencies 10-49%) suggests that CtIP is a reasonable target for MSI. However, CtIP-mutated patients could not be distinguished from other MSI CRC patients by histopathology of the tumour, sex of the patient, age at tumour occurrence, or site of the tumour. Many different cellular pathways may be affected by MSI simultaneously, and thus, the phenotype of a tumour is the result of accumulated mutations. While we propose that CtIP is a target for mutations in mononucleotide repeats caused by MMR deficiency, Li et al. (2000) showed that the DNA damage response may occur through CtIP. A large complex acting in DNA damage consists of DNA repair proteins (MSH2, MSH6, MLH1, ATM, BLM), RAD50-MRE11-NBS1 protein complex, DNA replication factor C, and BRCA1 (Wang et al. 2000). From this complex, at least MSH6 (Malkhosyan et al. 1996), ATM (Ejima et al. 2000), BLM (Calin et al. 1998), and RAD50 (Kim et al. 2001) genes are targets for MSI. Defects in response to DNA damage caused by mutations in these genes lead to increased genomic instability and tumour progression. ATM is known to phosphorylate proteins involved in cell cycle progression and DNA damage repair. For example, phosphorylation of p53 activates the protein and leads to transcription of p21 and GADD45 (Banin et al. 1998, Canman et al. 1998). Interestingly, after ionizing radiation, ATM also phosphorylates CtIP at serine residues (Ser664, Ser745). This leads to dissociation of the CtIP-CtBP complex from BRCA1 and further to BRCA1-dependent transcription of the GADD45 gene (Li et al. 2000). As in the case of mutated serine residues (Li et al. 2000), truncated CtIP might also prevent GADD45 transcription and the response to DNA damage. The DNA damage activated pathway mediated by ATM is a factor in the progression of MSI CRC. CtIP binds Rb, a key regulator of G1/S transition of the cell cycle, and Rb-mediated repression of transcription can be relayed through CtIP. E2F binds to the target gene promoter and CtIP interacts with Rb/p130/E2F complex and CtBP, allowing CtBP to mediate repression (Meloni et al. 1999). Thus, lack of CtIP could lead to transcription of genes encoding proteins for cell cycle progression or DNA replication. CtIP may belong to the group of “co-operator” genes described by Duval and Hamelin (2002). Mutations of different “co-operator” genes show variable mutation frequencies, but the same pathway may be affected in all cases. Not only genes with high mutation frequencies but also rarely mutated repeat tracts in different CRCs might have an important task in MSI tumorigenesis. Inactivated proteins lead to deficiency of the pathway, and thus, low-frequency mutations could also be considered as relevant (Perucho 2003). If the question of mutation frequency in repeats is difficult, also the evidence of biallelic mutations is controversial. Biallelic mutations in CtIP were not detected. However, a biallelic mutation alone does not necessarily prove the impact of the phenotype. Recently, haploinsufficiency or a dominant negative effect of MSI target gene products has also been discussed (Perucho 2003). According to Perucho (2003), only functional evidence is adequate to prove the authenticity of novel MSI target genes. Indeed, functional studies confirming the nature of CtIP as a true MSI target gene are warranted.

45

3. POSSIBLE LOW-RISK ALLELES FOR COLORECTAL CANCER (III, IV) The efficiency and accuracy of DNA repair can be affected by polymorphisms or defects in many genes. Proteins that do not directly act in DNA repair but interact with repair proteins or are mediators in a signal pathway may have an influence on repair processes. Not only loss of function but also mild reductions in DNA repair genes may cause cancer predisposition. Association studies are used in detecting low-penetrance genes. The frequencies of gene variants are compared among patients and controls, and the ethnicity of cases must be taken into consideration. Combinations of variants in different genes may also affect the phenotype. Furthermore, environmental factors can have a great impact on the phenotype of low-penetrance mutation carriers. The age-structure of cases and controls is relevant in, for example, determining probable exposure to carcinogens (Houlston and Tomlinson 2000, Mohrenweiser et al. 2003). Mutations in at least MLH1, MSH2, MSH6, and PMS2 seem to cause CRC. Several novel human genes involved in the MMR system have recently been cloned and studied in HNPCC. According to Studies III and IV, germline mutations in MLH3 or EXO1 do not predispose affected individuals to HNPCC. In general, studies of sequence variants in these genes have failed to demonstrate their importance in CRC, but the role of some of the variants or low-penetrance mutations could neither be confirmed nor excluded (Lipkin et al. 2001, Wu et al. 2001a, 2001b, Hienonen et al. 2003, Liu et al. 2003). 3.1 MLH3 (III) In yeast, Mlh3 interacts with Mlh1 and is involved in the repair of insertions or deletions longer than 2 bp (Flores-Rozas and Kolodner 1998). Overexpression of mutated Mlh3 has been shown to lead to MSI in mouse cell lines (Lipkin et al. 2000), but the role of the MLH1-MLH3 complex in humans needs to be studied further. Indeed, we did not find germline mutations in 46 samples from CRC patients with MSI tumours or in six CRC patients harbouring 14q deletions in MSS tumours. The three variants G1258A (Val420Ile), A1234G (Lys412Glu), and G4377A (Gln1421Gln) in exons 1 and 11, detected in CRC samples, were also found in controls at a similar frequency. These results suggest that the germline mutations in MLH3 are likely not involved in HNPCC. Furthermore, the importance of MLH3 in MMR can be questioned. Yeast and mouse studies have demonstrated that Mlh3 does not have a significant role in MMR (Harfe and Jinks-Robertson 2000, Lipkin et al. 2002). Mlh3 has been shown to be associated with Msh4 in mammalian meiotic cells, supporting its role in meiotic recombination (Santucci-Darmanin et al. 2002). Somatic mutations of MLH3 in 93 MSI tumours occurred in eight A8-A9 repeats. However, while the mutation frequency (8.6%) remains at a level similar to that of intronic A8 and A9 control repeats (6.5%), somatic MLH3 mutations seem not to be selected for. Thus, they do not play a key role in the formation of tumours. Akiyama et al. (2001) and Lipkin et al. (2001) also detected somatic frameshift mutations in MLH3 mononucleotide repeats (14.8% and 25%, respectively). In contrast to the present study, Lipkin et al. (2001) found biallelic inactivation containing both A8 and A9 repeat

46

mutations in four of the six tumours. Lipkin et al. (2001) also reported two stop codon causing mutations in 14q LOH MSS tumours. In summary, the present study together with Lipkin et al. (2001) found no MLH3 germline mutations in CRC patients. Hienonen et al. (2003) detected a missense variant, not present in cancer-free controls. Wu et al. (2001b) and Liu et al. (2003), by contrast, showed both missense and frameshift mutations. Wu et al. (2001b) reported 10 MLH3 germline mutations from 39 HNPCC patients as well as 288 HNPCC suspects, one being a frameshift 2578delA. Liu et al. (2003) found the frameshift mutation 885delG twice in the same family and 11 missense mutations in 70 suspected familial CRC cases (Table 6). Most tumours with MLH3 mutations were MSI-negative (Wu et al. 2001b, Liu et al. 2003), similar to animal studies (Harfe and Jinks-Robertson 2000, Lipkin et al. 2002). Polymorphisms seemed to vary between studies and possibly between populations (Liu et al. 2003). Furthermore, different mutation screening methods (SSCP, DGGE, dHPLC) were used in these five studies. In conclusion, some of the mutated MLH3 genes may have a minor effect on CRC, possibly with some other low-risk alleles (Liu et al. 2003). Table 6. At least 33 different germline variants have been reported in the coding sequence of MLH3. Nucleotide Amino acid substitution Reference C70G Gln24Glu Wu et al. 2001 T344C* Phe74Leu* Lipkin et al. 2001 T561C* Tyr149Tyr* Lipkin et al. 2001 A691C Lys231Gln Liu et al. 2003 885delG Termination 37 bp downstream Liu et al. 2003 C1111T* Asn333Asn* Lipkin et al. 2001 G1153C* Asp385His* Hienonen et al. 2003 T1168A* Phe390Ile* Hienonen et al. 2003 A1234G* Lys412Glu* Study III, Liu et al. 2003 G1258A* Val420Ile* Study III, Hienonen et al. 2003 A1496G Asn499Ser Wu et al. 2001 G1870C* Glu624Gln* Wu et al. 2001, Hienonen et al. 2003, Liu et al. 2003 C1939T Arg647Cys Wu et al. 2001 G2221T Val741Phe Liu et al. 2003 A2425G Met809Val Hienonen et al. 2003 A2449G Ser817Gly Wu et al. 2001 C2531T* Pro844Leu* Liu et al. 2003 2578delA Termination 49 bp downstream Wu et al. 2001 G2590A* Glu841Asn* Lipkin et al. 2001 G2797T Gly933Cys Liu et al. 2003 C2825T* Thr942Ile* Hienonen et al. 2003, Liu et al. 2003 C2838A* Ser947Ser* Liu et al. 2003 T2896C* Ser966Pro* Hienonen et al. 2003, Liu et al. 2003 G2911A Val971Ile Liu et al. 2003 G2941A Gly981Ser Wu et al. 2001 A3020G Asn1007Ser Wu et al. 2001 C3315A* Asp1105Glu* Hienonen et al. 2003 T3826C Trp1276Arg Liu et al. 2003 G4180A Ala1394Thr Wu et al. 2001 G4351A Glu1451Lys Wu et al. 2001, Liu et al. 2003 G4356C* Gln1414His* Lipkin et al. 2001 G4362A* Lys1416Lys* Lipkin et al. 2001 G4377A* Gln1421Gln* Study III * Polymorphisms

47

3.2 EXO1 (IV) Wu et al. (2001a) presented strong evidence for the truncating variant of EXO1 (IVS12-1 G/C) as a predisposing mutation for CRC. In Study IV, this variant was found in a Dutch CRC sample. The CRC patient fulfilled the Amsterdam criteria and had an MSI-L colorectal tumour. The patient’s brother, who also suffered from CRC, harboured the same truncating mutation. Despite this, the change was considered polymorphic because it was also found in three Dutch controls (Table 7). More polymorphism in the place of a truncating variant was detected when a Finnish control revealed a IVS12-1 G/A change. RNA from this patient was unavailable, and the splicing effect could not be confirmed. Thus, the splicing variant reported by Wu et al. (2001a) in the family with HNPCC appears to be a polymorphism. In addition to the truncating variant, four other candidate mutations (Ser610Gly, Pro640Ser, Pro640Ala, Gly759Glu) of the nine reported by Wu et al. (2001a) were detected in both CRC samples and controls of this study, and deemed polymorphisms. The role of the four variants Val27Ala, Glu109Lys, Leu410Arg, and Pro770Leu in CRC reported by Wu et al. (2001a) could not be confirmed in the present study. Three of the mutations were not detected in either CRC samples or controls. The fourth variant Leu410Arg was found in a Dutch/Norwegian putative HNPCC patient but could not be detected in controls (Table 7). Moreover, the same CRC patient carried a MSH2 missense mutation of unknown pathogenicity, suggesting that mutations in different genes together could cause a more severe phenotype than either mutation alone. Indeed, although mutations in S. cerevisiae Exo1 cause a mutator phenotype (Tishkoff et al. 1997), Exo1 mutant yeast models have shown that Exo1 mutations together with defects in Mlh1, Pms1, Msh2, Msh3, and Pol30 cause stronger mutator phenotypes than single mutants (Amin et al. 2001). Four new EXO1 variants (Ala137Ser, Phe438Cys, Asp661Asp, Ala827Val) were uncovered in addition to the

Table 7. Germline variants have been reported in the coding sequence of EXO1 (Wu et al. 2001a, Study IV). Nucleotide Amino acid substitution T80C Val27Ala G325A Glu109Lys G409T Ala137Ser G745A* Asp249Asn* G820A* Gly274Arg* A836G* Asn279Ser* G1061A* Arg354His* T1229G Leu410Arg T1313G Phe438Cys C1316T* Thr439Met* G1372A* Val458Met* G1378C* Val460Leu* A1765G* Lys589Glu* A1828G* Ser610Gly* C1918T* Pro640Ser* C1918G* Pro640Ala* C1983T Asp661Asp G2009A* Gly670Glu* T2167C* Cys723Arg* T2175A* Ser725Ser* IVS12-1* G to C, 194 bp deletion,

termination* IVS12-1* G to A* T2270C* Leu757Pro* G2276A* Gly759Glu* C2309T Pro770Leu C2480T* Ala827Val* * Polymorphisms

48

possible truncating variant (IVS12-1 G/A). Three of the variants (Ala137Ser, Phe438Cys, silent Asp661Asp) were detected in CRC patients but not in controls, and thus, the role of two amino acid changing variants in CRC could not be excluded (Table 7). The Ala137Ser alteration was found in a Norwegian patient from a late onset CRC family and the Phe438Cys variant in a Norwegian patient from a family fulfilling the revised Amsterdam criteria. The fourth new variant Ala827Val was found in both a CRC sample and a control and was considered a polymorphism. Both MSI and MSS tumour-harbouring patients with or without family histories as well as patients with late onset tumour development were used to cover different CRC groups. CRC susceptibility mutations could not be determined in the wide set of 989 European samples when compared to 1027 controls. The sample set of Wu et al. (2001a) included patients from 33 families fulfilling the Amsterdam criteria, 225 suspected HNPCC patients, and 235 controls. In these two studies, altogether seven EXO1 variants (Pro770Leu, Asp661Asp, Phe438Cys, Leu410Arg, Ala137Ser, Glu109Lys, Val27Ala) were not found in controls, and thus, their effect on CRC cannot be excluded. Six amino acid changes were all located in the non-conserved areas of the EXO1 protein. Val27Ala and Phe438Cys variants change a hydrophobic residue into a less hydrophobic one. Ala137Ser and Leu410Arg variants change a hydrophobic residue into a small polar residue and a charged residue, respectively, whereas Glu109Lys changes the charge of the residue. A causative role for the two variants Glu109Lys and Pro770Leu not found in controls was confirmed in vitro by Sun et al. (2002). Nuclease activity assays were carried out for eight of the EXO1 missense mutations described by Wu et al. (2001a), excluding the truncating variant. The sequence variants Glu109Lys and Leu410Arg resulted in loss of exonuclease activity. Pro640Ser, Gly759Glu, and Pro770Leu showed about 60% reduced binding between EXO1 and MSH2. Logically, the mutated amino acids are located in the MSH2 interaction domain. Mutants Val27Ala, Ser610Gly, and Pro640Ala did not show any phenotype. In addition to exonuclease activities and interactions with Mlh1 and Msh2, Exo1 is involved in repairing UV damage together with the Mre11-Rad50-Xrs2 complex (Tsubouchi and Ogawa 2000) and has a role in meiosis (Wei et al. 2003). With further studies, the role of specific mutations in different pathways is expected to be elucidated (Sun et al. 2002). In conclusion, it is likely that the EXO1 polymorphism is common in humans and affects the functions of the EXO1 protein in vitro. In humans, the function of the protein may, however, be modified by, for example, redundant replacement of mutated EXO1 by other exonucleases (Tran et al. 1999, Burdett et al. 2001). The effect of various mutations occurring together in different CRC-predisposing genes could modify the resultant phenotype.

49

4. BIALLELIC MUTATIONS IN MYH PREDISPOSE TO MULTIPLE ADENOMAS (V) MYH-associated polyposis has recently been added to the list of recessive inheritance syndromes. Compound heterozygote or homozygote mutations of MYH have been shown to cause polyposis and possible CRC (Al-Tassan et al. 2002, Sieber et al. 2003). In this study, the founder mutations (Y165C, G382D) were screened for in 1003 DNA samples available from a group of 1042 unselected Finnish CRC patients. Four (0.4%) patients were found with biallelic and five patients with heterozygous MYH mutations. All together, mutated MYH alleles in the population were not frequent (13/2006). Because mutations were not found in 424 cancer-free controls in Finland and each mutation was found once in 100 British controls (Al-Tassan et al. 2002), the mutated alleles in the normal population are rare. Comparison of the results from this study with data from the National Finnish Polyposis Registry revealed that MYH-associated CRC appears to be as common as FAP-associated CRC. FAP has been shown to account for less than 1% of all CRC cases (Talbot et al. 2000). In another study with unselected samples, the corresponding figure of biallelic MYH mutation carriers in a set of 75 sporadic CRCs was 1.3% (Halford et al. 2003). The frequency of MYH-associated CRC was higher in that study than in ours, but their number of samples was also smaller. Two MYH mutations were screened for in our study, but the possibility of patients harbouring other MYH mutations must also be taken into account. In addition to the Y165C and G382D mutations, other variants of MYH have been reported in polyposis cases (Halford et al. 2003, Sampson et al. 2003, Sieber et al. 2003) (Table 8). Furthermore, some MYH variants have been detected at similar frequencies in CRC patients and controls (Al-Tassan et al. 2002, Halford et al. 2003, Sieber et al. 2003). Interestingly, Pakistanian and Indian patients were found with biallelic MYH mutations in Y90X and E466X, respectively; thus, specific mutations of MYH are likely to occur in different ethnic populations (Jones et al. 2002, Sampson et al. 2003). Abundant somatic G:C→T:A transversions in APC have been detected in MYH patients (Al-Tassan et al. 2002, Jones et al. 2002, Sieber et al. 2003). Here, transversion in APC was found in only one patient with heterozygous MYH mutation, revealing the effect of deficient BER. According to the study of Halford et al. (2003), biallelic MYH mutations also occurred without somatic APC transversions, indicating that other genes could be mutated. In this study, two of the four biallelic MYH mutation carriers had five or several polyps and two had 50-100 adenomas. Importantly, the phenotype of all biallelic MYH mutation carriers in the Finnish sample set was not as polypotic as in previous studies. Sieber et al. (2003) showed that none out of six biallelic mutation carriers had less than 18 polyps. They also reported that almost one-third of patients harbouring 15-100 polyps and 7.5% of classical polyposis patients (APC mutations excluded) were biallelic MYH mutation carriers. However, the numbers of mutated patients in these studies were small, and thus, not necessarily comparable. Sampson et al. (2003) described a compound heterozygous patient with only three adenomas, showing that biallelic MYH mutations can rarely cause only a few adenomas. More data on MYH patients are needed to distinguish the mutation

50

Table 8. Variants of MYH reported in patients with colorectal adenomas or cancer (Al-Tassan et al. 2002, Jones et al. 2002, Halford et al. 2003, Sampson et al. 2003, Sieber et al. 2003, Study V). Nucleotide Amino acid

substitution Biallelic mutation with

Number of patients

Number of adenomas in patient

CRC Ethnicity

G64A* Val22Met* - Allele freq. 9%

Not known Yes/No Caucasian

T182A Val61Glu - 1 None Yes Caucasian C247T Arg83Stop - 1 6 Yes Caucasian 252delG Termination 137insIleTrp 1 100-1000 Yes Caucasian C270A Tyr90Stop Tyr90Stop 1 >400 Yes Pakistani IVS5-1 G to A Tyr165Cys 1 14 No Caucasian T349A Trp117Arg Tyr165Cys 1 >100 No Caucasian 411dupATGGAT 137insIleTrp nt252delG ● ● ● ● A494G Tyr165Cys Tyr165Cys

IVS5-1 Trp117Arg Val232Phe IVS10+3 Gly382Asp nt1419delC -

10 ● ● 1 1 16 1 several

Numerous-1000 ● ● 100-1000 Numerous Numerous- >1000 18 0-1000

Yes/No ● ● No No Yes/No Yes Yes/No

Caucasian ● ● Caucasian Caucasian Caucasian Caucasian Caucasian

A625G Ile209Val - 1 100-1000 No Caucasian G694T Val232Phe Tyr165Cys ● ● ● ● G777A Arg260Gln Ser501Phe* 1 3 Yes Caucasian C883T Arg295Cys - 2 0-3 Yes/No Caucasian IVS10+3 A to C Tyr165Cys,

Gly382Asp

● 1

● Multiple

● No

● Caucasian

C970T Gln324Stop Gly382Asp 1 >30 No Caucasian G972C* Gln324His* - Allele freq.

24-29% Not known Yes/No Caucasian

1103delC Termination Gly382Asp 1 70 No Caucasian G1145A Gly382Asp Gly382Asp

IVS10+3 Gln324Stop nt1103delC 465del -

5 ● ● ● 1 several

5-1000 ● ● ● Multiple 0-1000

Yes/No ● ● ● Yes Yes/No

Caucasian ● ● ● Caucasian Caucasian

1395-7delGGA 465del Gly382Asp ● ● ● ● G1396T Glu466Stop Glu466Stop 4 >100- >200 Yes/No Indian 1419delC Termination Tyr165Cys ● ● ● ● C1502T* Ser501Phe* Arg260Gln

-

● Allele freq. 1-3%

● Not known

● Yes/No

● Caucasian

* Polymorphisms ● Mentioned earlier in the table

carriers from other polyposis patients. Polyp number and family history help the prediction of MYH-associated polyposis in most cases. Treatment to prevent CRC, corresponding to the number of polyps, is expected to develop over time. Interestingly, classical polyposis patients with biallelic MYH mutations were treated by total colectomy later than FAP patients (mean age 47.6 and 28 years, respectively) (Sieber et al. 2003). Only one out of five heterozygous mutation carriers had more than five polyps, whereas in the study of Sampson et al. (2003), three heterozygous patients had 15, 20, and

51

numerous adenomas. The second hits in MYH were absent when the tumours of these heterozygous patients were sequenced. Thus, it seems that the formation of adenomas can vary to some extent (Sampson et al. 2003). Heterozygous MYH mutation carriers have not been shown to have increased risk of CRC, and the risk of MYH patients or mutation carriers to develop other cancers or disorders has not been defined. Interestingly, the present study revealed various solid tumours in the families of MYH patients, and thus, the predisposition of MYH mutations to other types of cancers warrants investigation.

52

CONCLUSIONS AND FUTURE PROSPECTS The roles of two different DNA repair systems, mismatch repair (MMR) and base excision repair (BER), in colorectal cancers (CRC) were investigated. Germline mutations of genes acting on the MMR and BER systems have previously been shown to cause hereditary non-polyposis colorectal cancer (HNPCC) and MYH-associated polyposis, respectively. To further evaluate the defects in MMR, phenotypically normal tissues of MLH1-/- patient were studied, candidate microsatellite instability target genes were sought in CRCs, and the involvement of MMR genes MLH3 and EXO1 in CRC was elucidated. Moreover, the defective BER and the phenotype of MYH mutation carriers were evaluated by searching for founder mutations in an unselected, population-based set of CRC samples. 1. MMR-deficient normal phenotype tissues showed that a large number of

mutations occur without a visible tumour. The molecular tumour clock model helps to clarify the genetic changes in cells before tumour formation and the time required for the accumulation of occult mutations. This model can be used in studies on the characters of different MMR-deficient, phenotypically normal tissues as well as tumours.

2. A new microsatellite instability target gene candidate CtIP was proposed, and the

length of intronic mononucleotide repeats was found to affect the frequency of repeat background mutations. Knowledge of different mutated cellular pathways leading to tumour formation is essential for understanding the behaviour of cancer and assigning specific treatments. Thus, increasing data on microsatellite instability target genes could shed light on the mutation spectrum of MMR-deficient tumours.

3. Neither MLH3 nor EXO1 seemed to occur frequently in colorectal tumour

formation. Mutations in different mismatch repair genes can explain some of the HNPCC or sporadic colon tumours. Some as yet unidentified genes in the MMR system may also have a role in colon tumour formation. Discovery of these MMR-related genes, followed by detection of harboured mutations could provide new insights into the biology of tumours.

4. Biallelic MYH mutations were found in Finnish CRC patients, and these biallelic

mutation carriers had more polyps than patients with only one mutated allele. The usefulness of MYH mutation screening of CRC patients and the optimal treatment for MYH-associated polyposis patients need to be evaluated. Characterization of the phenotype of mutation carriers could help to identify the syndrome. The cancer risk of heterozygous mutation carriers should also be determined.

53

ACKNOWLEDGEMENTS This study was carried out at the Department of Medical Genetics, Haartman Institute and Biomedicum Helsinki, University of Helsinki, as part of the Academy of Finland, Finnish Centre of Excellence Programme (2000-2005). The former and present heads of the Department of Medical Genetics, Pertti Aula, Anna-Elina Lehesjoki, Leena Palotie, and Kristiina Aittomäki, are thanked for providing excellent research facilities. My supervisor Lauri Aaltonen is warmly thanked for his continuous support and relevant comments. His enthusiasm revealed scientific possibilities almost everywhere, and showed the importance of motivation in all of the studies. I am also grateful to my other supervisor Auli Karhu for her calming influence and guidance. Virpi Launonen is thanked for help with thousands of problems during this project. I am deeply grateful to both Virpi and Auli for always being there at times of crises. Anne Kallioniemi and Minna Nyström-Lahti are thanked for their encouraging and instructive comments on this manuscript. Juha Kere and Päivi Peltomäki of the thesis committee are also gratefully acknowledged. Special thanks go to my Biomedicum room-mates: Päivi Laiho and Sini Marttinen for everything from help with Cancer Registries to extracurricular activities, and Rainer Lehtonen for insightful comments and PC/Mac support. Always positive and sunny Anniina Leskinen is thanked for excellent technical assistance and friendship, Tuija Hienonen, Taija Poikonen, and Jaskiran Singh for fruitful collaboration, and Anu Loukola for assistance during the first steps in Lauri’s lab. Mikko Aho, Pia Alhopuro, Diego Arango, Carita Eklund, Maija Kiuru, Antti Kokko, Annika Korvenpää, Kirsi Laukkanen, Heli Lehtonen, Nina Nupponen, Reijo Salovaara, Heli Sammalkorpi, Inga-Lill Svedberg, Sakari Vanharanta, Sanna Virta, and Iina Vuoristo are thanked for the inspiring atmosphere in the lab. Co-authors and collaborators Stephen Bevan, Matti Eskelinen, Riccardo Fodde, Riitta Herva, Richard Houlston, Shantie Jagmohan-Changur, Heikki Järvinen, Richard Kolodner, Vesa Kärjä, Lara Lipton, Jukka-Pekka Mecklin, Pål Møller, Torben Orntoft, Minna Pöyhönen, Darryl Shibata, Pertti Sistonen, Anu Suomalainen, Ian Tomlinson, Carli Tops, Jen-Lan Tsao, Hans Vasen, Outi Vierimaa, Imre Västrik, Juul Wijnen, and Friedrik Wikman are warmly acknowledged. All of my friends are thanked for providing a good balance between work and personal time. My heartfelt gratitude goes to my dear family: my mother Riitta and father Markku for support and encouragement, and my sister Sini for endless discussions and understanding.

54

Finally, I am deeply grateful to my husband Berndt for his patience and love. His constructive criticism and explanations have often made me see scientific matters in a new light. And my sweet, stubborn son Oskar has made me a part of an inspiring science project of his own! This study was supported by the Helsinki Biomedical Graduate School and by grants from the Emil Aaltonen Foundation, the Finnish Cancer Society, the Finnish Cultural Foundation, the Ida Montin Foundation, the Paulo Foundation, and the Research and Science Foundation of Farmos.

55

REFERENCES Aaltonen LA, Peltomäki P, Leach F, Sistonen P, Pylkkänen L, Mecklin J-P, Järvinen H, Powell S, Jen J, Hamilton SR, et al. (1993). Clues to pathogenesis of familial colorectal cancer. Science 260, 812-6.

Aaltonen LA, Peltomäki P, Mecklin J-P, Järvinen H, Jass JR, Green JS, Lynch HT, Watson P, Tallqvist G, Juhola M, et al. (1994). Replication errors in benign and malignant tumors from hereditary nonpolyposis colorectal cancer patients. Cancer Res 54, 1645-8.

Aaltonen LA, Salovaara R, Kristo P, Canzian F, Hemminki A, Peltomäki P, Chadwick RB, Percesepe A, Kääriäinen H, Ahtola H, et al. (1998). Incidence of hereditary nonpolyposis colorectal cancer, and molecular screening for the disease. N Engl J Med 338, 1481-7.

Aarnio M, Mecklin J-P, Aaltonen LA, Nyström-Lahti M, Järvinen HJ (1995). Life-time risk of different cancers in hereditary nonpolyposis colorectal cancer (HNPCC) syndrome. Int J Cancer 64, 430-3.

Akiyama Y, Nagasaki H, Nakajima T, Sakai H, Nomizu T, Yuasa Y (2001). Infrequent frameshift mutations in the simple repeat sequences of hMLH3 in hereditary nonpolyposis colorectal cancers. Jpn J Clin Oncol 31, 61-4.

Akiyama Y, Sato H, Yamada T, Nagasaki H, Tsuchiya A, Abe R, Yuasa Y (1997). Germline mutation of the hMSH6/GTBP gene in an atypical hereditary nonpolyposis colorectal cancer kindred. Cancer Res 57, 3920-3.

Albertson DG, Collins C, McCormick F, Gray JW (2003). Chromosome aberrations in solid tumors. Nat Genet 34, 369-76.

Al-Tassan N, Chmiel NH, Maynard J, Fleming N, Livingston AL, Williams GT, Hodges AK, Davies DR, David SS, Sampson JR, Cheadle JP (2002). Inherited variants of MYH associated with somatic G:C T:A mutations in colorectal tumors. Nat Genet 30, 227-32.

Ames BN, Shigenaga MK, Hagen TM (1993). Oxidants, anti-oxidants and the degenerative diseases of aging. Proc Natl Acad Sci USA 90, 7915-22.

Amin NS, Nguyen M-N, Oh S, Kolodner RD (2001). exo1 dependent mutator mutations: model system for studying functional interactions in mismatch repair. Mol Cell Biol 21, 5142-55.

Andrew SE, McKinnon M, Cheng BS, Francis A, Penney J, Reitmair AH, Mak TW, Jirik FR (1998). Tissues of MSH2 deficient mice demonstrate hypermutability on exposure to a DNA methylating agent. Proc Natl Acad Sci USA 95, 1126-30.

Andrew SE, Xu XS, Baross-Francis A, Narayanan L, Milhausen K, Liskay RM, Jirik FR, Glazer PM (2000). Mutagenesis in PMS2- and MSH2-deficient mice indicates differential protection from transversions and frameshifts. Carcinogenesis 21, 1291-5.

Aruoma OI, Halliwell B, Dizdaroglu M (1989). Iron ion-dependent modification of bases in DNA by the superoxide radical-generating system hypoxanthine/xanthine oxidase. J Biol Chem 264, 13024-8.

Bai F, Pei X-H, Godfrey VL, Xiong Y (2003). Happloinsufficiency of p18INK4c sensitizes mice to carcinogen-induced tumorigenesis. Mol Cell Biol 23, 1269-77.

56

Baker SM, Bronner CE, Zhang L, Plug AW, Robatzek M, Warren G, Elliott EA, Yu J, Ashley T, Arnheim N (1995). Male mice defective in the DNA mismatch repair gene PMS2 exhibit abnormal chromosome synapsis in meiosis. Cell 82, 309-20.

Baker SM, Plug AW, Prolla TA, Bronner CE, Harris AC, Yao X, Christie DM, Monell C, Arnheim N, Bradley A, et al. (1996). Involvement of mouse Mlh1 in DNA mismatch repair and meiotic crossing over. Nat Genet 13, 336-42.

Banin S, Moyal L, Shieh S-Y, Taya Y, Anderson CW, Chessa L, Smorodinsky NI, Prives C, Reiss Y, Shiloh Y, Ziv Y (1998). Enhanced phosphorylation of p53 by ATM in response to DNA damage. Science 281, 1674-7.

Bertoni F, Codegoni AM, Furlan D, Tibiletti MG, Capella C, Broggini M (1999). CHK1 frameshift mutations in genetically unstable colorectal and endometrial cancers. Gene Chrom Cancer 26, 176-80.

Bessho T, Roy R, Yamamoto K, Kasai H, Nishimura S, Tano K, Mitra S (1993). Repair of 8-hydroxyguanine in DNA by mammalian N-methylpurine-DNA glycosylase. Proc Natl Acad Sci USA 90, 8901-4.

Bhattacharyya NP, Skandalis A, Ganesh A, Groden J, Meuth M (1994). Mutator phenotypes human colorectal carcinoma cell lines? Proc Natl Acad Sci USA 91, 6319-23.

Bicknell DC, Kaklamanis L, Hampson R, Bodmer WF, Karran P (1996). Selection for B2-microglobulin mutation in mismatch repair defective colorectal carcinomas. Curr Biol 6, 1695-7.

Boland CR, Thibodeau SN, Hamilton SR, Sudransky D, Eshleman JR, Burt RW, Meltzer SJ, Rodrigues-Bigas MA, Fodde R, Ranzani GN, Srivastava S (1998). A National Cancer Institute workshop on microsatellite instability for cancer detection and familial predisposition: development of international criteria for the determination of microsatellite instability in colorectal cancer. Cancer Res 58, 5248-57.

Boldogh I, Milligan D, Lee MS, Bassett H, Lloyd RS, McCullough AK (2001). hMYH cell cycle-dependent expression, subcellular localization and association with replication foci: evidence suggesting replication-coupled repair of adenine:8-oxoguanine mispairs. Nucleic Acids Res 29, 2802-9.

Boyer JC, Yamada NA, Roques CN, Hatch SB, Riess K, Farber RA (2002). Sequence dependent instability of mononucleotide microsatellites in cultured mismatch repair proficient and deficient mammalian cells. Hum Mol Genet 11, 707-13.

Boyle P, Leon ME (2002). Epidemiology of colorectal cancer. Br Med Bull 64, 1-25.

Bronner CE, Baker SM, Morrison PT, Warren G, Smith LG, Lescoe MK, Kane M, Earabino C, Lipford J, Lindblom A, et al. (1994). Mutation in the DNA mismatch repair gene homologue hMLH1 is associated with hereditary non-polyposis colon cancer. Nature 368, 258-61.

Burdett V, Baitinger C, Viswanathan M, Lovett ST, Modrich P (2001). In vivo requirement for RecJ, ExoVII, ExoI, and ExoX in methyl-directed mismatch repair. Proc Natl Acad Sci USA 98, 6765-70.

Burt RW (2000). Colon cancer screening. Gastroenterology 119, 837-53.

Cabrera M, Nghiem Y, Miller JH (1988). MutM a second mutator locus in Escherichia coli that generates G:C T:A transversions. J Bacteriol 170, 5405-7.

Calin G, Cafa R, Tibiletti MG, Herlea V, Becheanu G, Cavazzini L, Barbanti-Brodano G, Nenci I, Negrini M, Lanza G (2000). Genetic progression in microsatellite instability high (MSI-H) colon cancers correlates

57

with clinico-pathological parameters: a study of the TGFBRII, BAX, HMSH3, HMSH6, IGFR and BLM genes. Int J Cancer 89, 230-5.

Calin G, Herlea V, Barbanti-Brodano G, Negrini M (1998). The coding region of the Bloom syndrome BLM gene and of the CBL proto-oncogene is mutated in genetically unstable sporadic gastrointestinal tumors. Cancer Res 58, 3777-81.

Canman CE, Lim DS, Cimprich KA, Taya Y, Tamai K, Sakaguchi K, Appella E, Kastan MB, Siliciano JD (1998). Activation of the ATM kinase by ionizing radiation and phosphorylation of p53. Science 281, 1677-9.

Cannon-Albright LA, Skolnick MH, Bishop DT, Lee RG, Burt RW (1988). Common inheritance of susceptibility to colonic adenomatous polyps and associated colorectal cancers. N Engl J Med 319, 533-7.

Caspari R, Olschwang S, Friedl W, Mandl M, Boisson C, Boker T, Augustin A, Kadmon M, Moslein G, Thomas G (1995). Familial adenomatous polyposis: desmoids tumours and lack of ophthalmic lesions (CHRPE) associated with APC mutations beyond codon 1444. Hum Mol Genet 4, 337-40.

Chadwick RB, Jiang GL, Bennington GA, Yuan B, Johnson CK, Stevens MW, Niemann TH, Peltomäki P, Huang S, de la Chapelle A (2000). Candidate tumor suppressor RIZ is frequently involved in colorectal carcinogenesis. Proc Natl Acad Sci USA 97, 2662-7.

Chevillard S, Radicella JP, Levalois C, Lebeau J, Poupon MF, Oudard S, Dutrillaux B, Boiteux S (1998). Mutations in OGG1, a gene involved in the repair of oxidative DNA damage, are found in human lung and kidney tumors. Oncogene 16, 3083-6.

Cho KR, Oliner JD, Simons JW, Hedrick L, Fearon ER, Preisinger AC, Hedge P, Silverman GA, Vogelstein B (1994). The DCC gene: structural analysis and mutations in colorectal carcinomas. Genomics 19, 525-31.

Chung DC, Rustgi AK (2003). The hereditary nonpolyposis colorectal cancer syndrome: Genetics and clinical implications. Ann Intern Med 138, 560-70.

Cooke MS, Evans MD, Dizdaroglu M, Lunec J (2003). Oxidative DNA damage: mechanisms, mutation, and disease. FASEB J 17, 1195-214.

Cormier RT, Hong KH, Halberg RB, Hawkins TL, Richardson P, Mulherkar R, Dove WF, Lander ES (1997). Secretory phospholipase Pla2g2a confers resistance to intestinal tumorigenesis. Nat Genet 17, 88-91.

Costello JF, Plass C (2001). Methylation matters. J Med Genet 38, 285-303.

Cui H, Cruz-Correa M, Giardiello FM, Hutcheon DF, Kafonek DR, Brandenburg S, Wu Y, He X, Powe NR, Feinberg AP (2003). Loss of IGF2 imprinting: A potential marker of colorectal cancer risk. Science 299, 1753-5.

Cunningham JM, Kim CY, Christensen ER, Tester DJ, Parc Y, Burgort LJ, Halling KC, McDonnell SK, Schaid DJ, Walsh Vockley C, et al. (2001). The frequency of hereditary defective mismatch repair in a prospective series of unselected colorectal carcinomas. Am J Hum Genet 69, 780-90.

D'Andrea AD, Grompe M (2002). The fanconi anemia/BRCA pathway. Nature 3, 23-34.

de Boer J, Hoeijmakers JHJ (2000). Nucleotide excision repair and human syndromes. Carcinogenesis 21, 453-60.

58

de Boer JG (2002). Polymorphisms in DNA repair and environmental interactions. Mut Res 509, 201-10.

de Wind N, Dekker M, Berns A, Radman M, te Riele H (1995). Inactivation of the mouse Msh2 gene results in mismatch repair deficiency, methylation tolerance, hyperrecombination, and predisposition to cancer. Cell 82, 321-30.

de Wind N, Dekker M, Claij N, Jensen L, van Klink Y, Radman M, Riggins G, van der Valk M, van 't Wout K, te Riele H (1999). HNPCC like cancer predisposition in mice through simultaneous loss of Msh3 and Msh6 mismatch repair protein functions. Nat Genet 23, 359-62.

de Wind N, Dekker M, van Rossum A, van der Valk M, te Riele H (1998). Mouse models for hereditary nonpolyposis colorectal cancer. Cancer Res 58, 248-55.

Devilee P, Cleton-Jansen AM, Cornelisse CJ (2001). Ever since Knudson. Trends Genet 17, 569-73.

DiSario JA, Foutch G, Mai HD, Pardy K, Manne RK (1991). Prevalence and malignant potential of colorectal polyps in asymptomatic, average-risk men. Am J Gastroenterol 86, 941-5.

Duckett DR, Bronstein SM, Taya Y, Modrich P (1999). hMutSalpha- and hMutLalpha- dependent phosphorylation of p53 in response to DNA methylator damage. Proc Natl Acad Sci USA 96, 12384-8.

Dunlop MG, Farrington SM, Carothers AD, Wyllie AH, Sharp L, Burn J, Liu B, Kinzler KW, Vogelstein B (1997). Cancer risk associated with germline DNA mismatch repair gene mutations. Hum Mol Genet 6, 105-110.

Duval A, Gayet J, Zhou XP, Iacopetta B, Thomas G, Hamelin R (1999). Frequent frameshift mutations of the TCF-4 gene in colorectal cancers with microsatellite instability. Cancer Res 59, 4213-5.

Duval A, Hamelin R (2002). Mutations at coding repeat sequencies in mismatch repair-deficient human cancers: Towards a new concept of target genes for instability. Cancer Res 62, 2447-54.

Duval A, Rolland S, Compoint A, Tubacher E, Iacopetta B, Thomas G, Hamelin R (2001). Evolution of instability at coding and non-coding repeat sequences in human MSI-H colorectal cancers. Hum Mol Genet 10, 513-8.

Edelmann W, Cohen PE, Kane M, Lau K, Morrow B, Bennett S, Umar A, Kunkel T, Cattoretti G, Chaganti R, et al. (1996). Meiotic pachytene arrest in MLH1 deficient mice. Cell 85, 1125-34.

Edelmann W, Cohen PE, Kneitz B, Winand N, Lia M, Heyer J, Kolodner R, Pollard JW, Kucherlapati R (1999). Mammalian MutS homologue 5 is required for chromosome pairing in meiosis. Nat Genet 21, 123-7.

Edelmann W, Umar A, Yang K, Heyer J, Kucherlapati M, Lia M, Kneitz B, Avdievich E, Fan K, Wong E, et al. (2000). The DNA mismatch repair genes Msh3 and Msh6 cooperate in intestinal tumor suppression. Cancer Res 60, 803-7.

Edelmann W, Yang K, Umar A, Heyer J, Lau K, Fan K, Liedtke W, Cohen PE, Kane MF, Lipford JR, et al. (1997). Mutation in the mismatch repair gene Msh6 causes cancer susceptibility. Cell 91, 467-77.

Ejima Y, Yang L, Sasaki MS (2000). Aberrant splicing of the ATM gene associated with shortening of the intronic mononucleotide tract in human colon tumor cell lines: a novel mutation target of microsatellite instability. Int J Cancer 86, 262-8.

Elsaleh H, Joseph D, Grieu F, Zeps N, Spry N, Iacopetta B (2000). Association of tumor site and sex with survival benefit from adjuvant chemotherapy in colorectal cancer. Lancet 355, 1745-50.

59

Eppert K, Scherer S, Ozcelik H, Pirone R, Hoodless P, Kim H, Tsui LC, Bapat B, Gallinger S, Andrulis IL, et al. (1996). MADR2 maps to 18q21 and encodes a TGFB-regulated MAD-related protein that is functionally mutated in colorectal carcinoma. Cell 86, 543-52.

Fazeli A, Dickinson SL, Hermiston ML, Tighe RV, Steen RG, Small CG, Stoeckli ET, Keino-Masu K, Masu M, Rayburn H, et al. (1997). Phenotype of mice lacking functional deleted in colorectal cancer (Dcc) gene. Nature 386, 796-804.

Fearon E, Cho K, Nigro J, Kern S, Simons J, Ruppert J, Hamilton S, Preisinger A, Thomas G, Kinzler K, Vogelstein B (1990). Identification of a chromosome 18q gene that is altered in colorectal cancers. Science 247, 49-56.

Fearon ER, Vogelstein B (1990). A genetic model for colorectal tumorigenesis. Cell 61, 759-67.

Fishel R, Lescoe MK, Rao MRS, Copeland NG, Jenkins NA, Garber J, Kane M, Kolodner R (1993). The human mutator gene homolog MSH2 and its association with hereditary nonpolyposis colon cancer. Cell 75, 1027-38.

Flores-Rozas H, Kolodner RD (1998). The Saccharomyces cerevisiae MLH3 gene functions in MSH3-dependent suppression of frameshift mutations. Proc Natl Acad Sci USA 95, 12404-9.

Fodde R, van der Luijt R, Wijnen J, Tops C, van der Klift H, van Leeuwen-Cornelisse I, Griffioen G, Vasen H, Khan PM (1992). Eight novel inactivating germ line mutations at the APC gene identified by denaturing gradient gel electrophoresis. Genomics 13, 1162-8.

Fraylin IM, Beck NE, Ilyas M, Dove-Edwin I, Goodman P, Pack K, Bell JA, Williams CB, Hodgson SV, Thomas HJW, et al. (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, 10722-7.

Friedberg EC (2003). DNA damage and repair. Nature 421, 436-40.

Futreal PA, Kasprzyk A, Birney E, Mullikin JC, Wooster R, Stratton MR (2001). Cancer and genomics. Nature 409, 850-2.

Ganguly A, Rock MJ, Prockop DJ (1993). Conformation-sensitive gel electrophoresis for rapid detection of single-base differences in double-stranded PCR products and DNA fragments: Evidence for solvent-induced bends in DNA heteroduplexes. Proc Natl Acad Sci USA 90, 10325-9.

Giardiello FM, Hamilton SR, Krush AJ, Piantadosi S, Hylind LM, Celano P, Booker SV, Robinson CR, Offerhaus GJ (1993). Treatment of colonic and rectal adenomas with sulindac in familial adenomatous polyposis. N Engl J Med 328, 1313-6.

Glinghammar B, Rafter J (2001). Colonic luminal contents induce cyclooxygenase 2 transcription in human colon carcinoma cells. Gastroenterology 120, 401-10.

Grady WM, Rajput A, Myeroff L, Liu DF, Kwon K, Willis J, Markowitz S (1998). Mutation of the type II transforming growth factor beta receptor is coincident with the transformation of human colon adenomas to malignant carcinomas. Cancer Res 58, 3101-4.

Gruber SB, Ellis NA, Rennert G, Offit K (2002). BLM heterozygosity and the risk of colorectal cancer. Science 297, 2013.

Gryfe R, Kim H, Hsieh ET, Aronson MD, Holowaty EJ, Bull SB, Redston M, Gallinger S (2000). Tumor microsatellite instability and clinical outcome in young patients with colorectal cancer. N Engl J Med 342, 69-77.

60

Guanti G, Resta N, Simone C, Cariola F, Demma I, Fiorente P, Gentile M (2000). Involvement of PTEN mutations in the genetic pathways of colorectal cancerogenesis. Hum Mol Genet 9, 283-7.

Guschin D, Wade PA, Kikyo N, Wolffe AP (2000). ATP-dependent histone octamer mobilization and histone deacetylation mediated by the Mi-2 chromatin remodeling complex. Biochemistry 39, 5238-45.

Hackman P, Tannergard P, Osei-Mensa S, Chen J, Kane MF, Kolodner R, Lambert B, Hellgren D, Lindblom A (1997). A human compound heterozygote for two MLH1 missense mutations. Nat Genet 17, 135-6.

Halford SE, Rowan AJ, Lipton L, Sieber OM, Pack K, Thomas HJ, Hodgson SV, Bodmer WF, Tomlinson IP (2003). Germline mutations but not somatic changes at the MYH locus contribute to the pathogenesis of unselected colorectal cancers. Am J Pathol 162, 1545-8.

Hall MC, Matson SW, Yamaguchi M, Dao V, Modrich P (1999). The Escherichia coli MutL protein physically interacts with MutH and stimulates the MutH associated endonuclease activity. J Biol Chem 274, 1306-12.

Hamilton SR, Vogelstein B, Kudo S, Riboli E, Nakamura S, Hainaut P, Rubio CA, Sobin LH, Fogt F, Winawer SJ, et al. (2000). Carcinoma of the colon and rectum. In: Pathology and genetics of tumors of the digestive system, World Health Organization Classification of tumors. Eds. Hamilton SR, Aaltonen LA. Lyon, IARC Press, 103-43.

Hampel H, Peltomäki P (2000). Hereditary colorectal cancer: risk assessment and management. Clin Genet 58, 89-97.

Hanahan D, Weinberg RA (2000). The hallmarks of cancer. Cell 100, 57-70.

Harfe BD, Jinks-Robertson S (2000). DNA mismatch repair and genetic instability. Annu Rev Genet 34, 359-99.

Hazra TK, Izumi T, Boldogh I, Imhoff B, Kow YW, Jagura P, Dizdaroglu M, Mitra S (2002). Identification and characterization of a human DNA glycosylase for repair of modified bases in oxidatively damaged DNA. Proc Natl Acad Sci USA 99, 3523-8.

Hemminki A, Peltomäki P, Mecklin J-P, Järvinen H, Salovaara R, Nyström-Lahti M, de la Chapelle A, Aaltonen LA (1994). Loss of the wild type MLH1 gene is a feature of hereditary nonpolyposis colorectal cancer. Nat Genet 8, 405-10.

Henle ES, Linn S (1997). Formation, prevention and repair of DNA damage by iron/hydrogen peroxide. J Biol Chem 272, 19095-8.

Herman JG, Umar A, Polyak K, Graff JR, Ahuja N, Issa J-P, Markowitz S, Willson JKV, Hamilton SR, Kinzler KW, et al. (1998). Incidence and functional consequences of hMLH1 promoter hypermethylation in colorectal carcinoma. Proc Natl Acad Sci USA 95, 6870-5.

Hickman MJ, Samson LD (1999). Role of DNA mismatch repair and p53 in signaling induction of apoptosis by alkylating agents. Proc Natl Acad Sci USA 96, 10764-9.

Hickson ID (2003). RecQ helicases: caretakers of the genome. Nature 3, 169-78.

Hienonen T, Laiho P, Salovaara R, Mecklin J-P, Järvinen H, Sistonen P, Peltomäki P, Lehtonen R, Nupponen N, Launonen V, et al. (2003). Little evidence for involvement of MLH3 in colorectal cancer predisposition. Int J Cancer 106, 292-6.

61

Hoang J-M, Cottu PH, Thuille B, Salmon RJ, Thomas G, Hamelin R (1997). BAT-26, an indicator of the replication error phenotype in colorectal cancers and cell lines. Cancer Res 57, 300-3.

Houlston RS, Collins A, Slack J, Norton NE (1992). Dominant genes for colorectal cancer are not rare. Ann Hum Genet 56, 99-103.

Houlston RS, Tomlinson IPM (2001). Polymorphisms and colorectal tumor risk. Gastroenterology 121, 282-301.

Huang J, Kuismanen SA, Liu T, Chadwick RB, Johnson CK, Stevens MW, Richards SK, Meek JE, Gao X, Wright FA, et al. (2001). MSH6 and MSH3 are rarely involved in genetic predisposition to nonpolypotic colon cancer. Cancer Res 61, 1619-23.

Huang J, Papadopoulos N, McKinley AJ, Farrington SM, Curtis LJ, Wyllie AH, Zheng S, Willson JKV, Markowitz SD, Morin P, et al. (1996). APC mutations in colorectal tumors with mismatch repair deficiency. Proc Natl Acad Sci USA 93, 9049-54.

Hyun JW, Cheon GJ, Kim HS, Lee YS, Choi EY, Yoon BH, Kim JS, Chung MH (2002). Radiation sensitivity depends on OGG1 activity status in human leukemia cell lines. Free Radic Biol Med 32, 212-20.

Iida T, Furuta A, Kawashima M, Nishida J, Nakabeppu Y, Iwaki T (2001). Accumulation of 8-oxo-2'-deoxyguanosine and increased expression of hMTH1 protein in brain tumors. Neuro-oncology 3, 73-81.

Ikenoue T, Togo G, Nagai K, Ijichi H, KatoJ, Yamaji Y, Okamoto M, Kato N, Kawabe T, Tanaka A, et al. (2001). Frameshift mutations at mononucleotide repeats in RAD50 recombinational DNA repair gene in colorectal cancers with microsatellite instability. Jpn J Cancer Res 92, 587-91.

Ionov Y, Peinado MA, Malkhosyan S, Shibata D, Perucho M (1993). Ubiquitous somatic mutations in simple repeated sequences reveal a new mechanis for colonic carcinogenesis. Nature 363, 558-61.

Irby RB, Mao W, Coppola D, Kang J, Loubeau JM, Trudeau W, Karl R, Fujita DJ, Jove R, Yeatman TJ (1999). Activating SRC mutation in a subset of advanced human colon cancers. Nat Genet 21, 187-90.

Jacob S, Praz F (2002). DNA mismatch repair defects: role in colorectal carcinogenesis. Biochimie 84, 27-47.

Jass JR, Biden KG, Cummings MC, Simms LA, Walsh M, Schoch E, Meltzer SJ, Wright C, Searle J, Young J, Leggett BA (1999). Characterisation of a subtype of colorectal cancer combining features of the suppressor and mild mutator pathways. J Clin Pathol 52, 455-60.

Jones S, Emmerson P, Maynard J, Best JM, Jordan S, Williams G, Sampson JR, Cheadle JP (2002). Biallelic germline mutations in MYH predispose to multiple colorectal adenoma and somatic G:C T:A mutations. Hum Mol Genet 11, 2961-7.

Järvinen H, Aarnio M, Mustonen H, Aktan-Collan K, Aaltonen LA, Peltomäki P, de la Chapelle A, Mecklin JP (2000). Controlled 15-year trial on screening for colorectal cancer in families with hereditary nonpolyposis colorectal cancer. Gastroenterology 118, 829-34.

Kane MF, Loda M, Gaida GM, Lipman J, Mishra R, Goldman H, Jessup JM, Kolodner R (1997). Methylation of the hMLH1 promoter correlates with lack of expression of hMLH1 in sporadic colon tumors and mismatch repair defective human tumor cells. Cancer Res 57, 808-11.

Kapitanovic S, Radosevic S, Kapitanovic M, Andelinovic S, Ferencic Z, Tavassoli M, Primorac D, Sonicki Z, Spaventi S, Pavelic K, Spaventi R (1997). The expression of p185HER/neu correlates with the stage of disease and survival in colorectal cancer. Gastroenterology 112, 1103-13.

62

Kaplan KB, Burds AA, Swedlow JR, Bekir SS, Sorger PK, Näthke IS (2001). A role for the adenomatous polyposis coli protein in chromosomal segregation. Nat Cell Biol 3, 429-32.

Kargman SL, O'Neill GP, Vickers PJ, Evans JF, Mancini JA, Jothy S (1995). Expression of prostaglandin G7H synthase-1 and -2 protein in human colon cancer. Cancer Res 55, 2556-9.

Kawate H, Sakumi K, Tsuzuki T, Nakatsuru Y, Ishikawa T, Takashi S, Takano H, Noda T, Sekiguchi M (1998). Separation and killing tumorigenic effects of an alkylating agent in mice defective in two of the DNA repair genes. Proc Natl Acad Sci USA 95, 5116-20.

Kim HC, Wheeler JMD, Kim JC, Ilyas M, Beck NE, KIM BS, Park KC, Bodmer WF (2000). The E-cadherin gene (CDH1) variants T340A and L599V in gastric and colorectal cancer patients in Korea. Gut 47, 262-7.

Kim JI, Park YJ, Kim KH, Kim JI, Song BJ, Lee MS, Kim CN, Chang SH (2003). hOGG1 Ser326Cys polymorphism modifies the significance of the environmental risk factor for colon cancer. World J Gastroenterol 9, 956-60.

Kim K-M, Salovaara R, Mecklin J-P, Järvinen HJ, Aaltonen LA, Shibata D (2002). PolyA deletions in hereditary nonpolyposis colorectal cancer. Am J Pathol 160, 1503-6.

Kim N-G, Choi YR, Baek MJ, Kim YH, Kang H, Kim NK, Min JS, Kim H (2001). Frameshift mutations at coding mononucleotide repeats of the hRAD50 gene in gastrointestinal carcinomas with microsatellite instability. Cancer Res 61, 36-8.

Kinzler KW, Vogelstein B (1996). Lessons from hereditary colorectal cancer. Cell 87, 159-70.

Kinzler KW, Vogelstein B (1997). Cancer-susceptibility genes. Gatekeepers and caretakers. Nature 386, 761-3.

Kinzler KW, Vogelstein B (1998). Landscaping the cancer terrain. Science 280, 1036-7.

Klungland A, Rosewell I, Hollenbach S, Larsen E, Daly G, Epe B, Seeberg E, Lindahl T, Barnes DE (1999). Accumulation of premutagenic DNA lesions in mice defective in removal of oxidative base damage. Proc Natl Acad Sci USA 96, 13300-5.

Kneitz B, Cohen PE, Avdievich E, Zhu L, Kane MF, Hou H Jr, Kolodner RD, Kucherlapati R, Pollard JW, Edelmann W (2000). MutS homolog 4 localization to meiotic chromosomes is required for chromosome pairing during meiosis in male and female mice. Genes Dev 14, 1085-97.

Knudson AG Jr (1971). Mutation and cancer: statistical study of retinoblastoma. Proc Natl Acad Sci USA 68, 820-3.

Knudson AG Jr (1985). Hereditary cancer, oncogenes and antioncogenes. Cancer Res 45, 1437-43.

Kolodner RD, Tytell JD, Schmeits JL, Kane MF, Gupta RD, Weger J, Wahlberg S, Fox EA, Peel D, Ziogas A, et al. (1999). Germline Msh6 mutations in colorectal cancer families. Cancer Res 59, 5068-74.

Kruse R, Rutten A, Lamberti C, Hosseiny-Malayeri HR, Wang Y, Ruelfs, Jungck M, Mathiak M, Ruzicka T, Hartschuh W, et al. (1998). Muir-Torre phenotype has a frequency of DNA mismatch repair gene mutations similar to that in hereditary nonpolyposis colorectal cancer families defined by the Amsterdam criteria. Am J Hum Genet 63, 63-70.

63

Kuismanen SA, Holmberg MT, Salovaara R, de la Chapelle A, Peltomäki P (2000). Genetic and epigenetic modification of MLH1 accounts for a major share of microsatellite-unstable colorectal cancers. Am J Pathol 156, 1773-9.

Kuismanen SA, Holmberg MT, Salovaara R, Schweizer P, Aaltonen LA, de la Chapelle A, Nyström-Lahti M, Peltomäki P (1999). Epigenetic phenotypes distinguish microsatellite- stable and -unstable colorectal cancers. Proc Natl Acad Sci USA 96, 12661-6.

Kuo MH, Allis CD (1998). Roles of histone acetyltransferases and deacetylases in gene regulation. Bioessays 20, 615-26.

Laghi L, Bianchi P, Orbetegli O, Gennari L, Roncalli M, Malesci A (2001). Lack of mutation at codon 531 of SRC in advanced colorectal cancers from Italian patients. Br J Cancer 84, 196-8.

Laghi L, Ranzani GN, Bianchi P, Mori A, Heinimann K, Orbetegli O, Spaudo MR, Luinetti O, Francisconi S, Roncalli M, et al. (2002). Frameshift mutations of human gastrin receptor gene (hGARE) in gastrointestinal cancers with microsatellite instability. Lab invest 82, 265-71.

Lahiri DK, Nurnberger JI Jr (1991). A rapid non-enzymatic method for the preparation of HMW DNA from blood for RFLP studies. Nucleic Acids Res 19, 5444.

Laiho P, Launonen V, Lahermo P, Esteller M, Guo M, Herman JG, Mecklin J-P, Järvinen H, Sistonen P, Kim K-M, et al. (2002). Low-level microsatellite instability in most colorectal carcinomas. Cancer Res 62, 1166-70.

Laken SJ, Petersen GM, Gruber SB, Oddoux C, Ostrer H, Giardiello FM, Hamilton SR, Hampel H, Markowitz A, Klimstra D, et al. (1997). Familial colorectal cancer in Askenazim due to a hypermutable tract in APC. Nat Genet 17, 79-83.

Lamlum H, Ilyas M, Rowan A, Clark S, Johnson V, Bell J, Frayling I, Efstathiou J, Pack K, Payne S, et al. (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, 1071-5.

Le Page F, Kwoh EE, Avrutskaya A, Gentil A, Leadon SA, Sarasin A, Cooper PK (2000). Transcription-coupled repair of 8-oxoguanine: requirement for XPG, TFIIH, and CSB and implications for Cockayne syndrome. Cell 101, 159-71.

Leach FS, Nicolaides NC, Papadopoulos N, Liu B, Jen J, Parsons R, Peltomäki P, Sistonen P, Aaltonen LA, Nyström-Lahti M, et al. (1993). Mutations of a mutS homolog in hereditary nonpolyposis colorectal cancer. Cell 7, 1215-25.

Lefebvre A-M, Chen I, Desreumaux P, Najib J, Fruchart J-C, Geboes K, Briggs M, Heyman R, Auwerx J (1998). Activation of the peroxisome proliferator-activated receptor gamma promotes the development of colon tumors in C57BL/6J-APCMin/+mice. Nat Med 4, 1053-7.

Lengauer C, Kinzler KW, Vogelstein B (1997). Genetic instability in coloretal cancers. Nature 386, 623-6.

Leslie A, Carey FA, Pratt NR, Steele RJC (2002). The colorectal adenoma-carcinoma sequence. Br J Surg 89, 845-60.

Levine AJ (1997). p53, the cellular gatekeeper for growth and division. Cell 88, 323-31.

Levinson G, Gutman GA (1987). Slipped-strand mispairing: a major mechanism for DNA sequence evolution. Mol Biol Evol 4, 203-21.

64

Li S, Ting NSY, Zheng L, Chen P-L, Ziv Y, Shiloh Y, Lee E, Lee W-H (2000). Functional link of BRCA1 and ataxia telangiectasia gene product in DNA damage response. Nature 406, 210-5.

Lichtenstein P, Holm NV, Verkasalo PK, Iliadou A, Kaprio J, Koskenvuo M, Pukkala E, Skytthe A, Hemminki K (2000). Environmental and heritable factors in the causation of cancer- analyses of cohorts of twins from Sweden, Denmark, and Finland. N Engl J Med 343, 78-85.

Lieberman DA, Smith FW (1991). Screening for colon malignancy with colonoscopy. Am J Gastroenterol 86, 946-51.

Lipkin SM, Moens PB, Wang V, Lenzi M, Shanmugarajah D, Abigail G, Thomas J, Cheng J, Touchman JW, Green ED, et al. (2002). Meiotic arrest and aneuploidy in MLH3-deficient mice. Nat Genet 31, 385-90.

Lipkin SM, Wang V, Jacoby R, Banerjee-Basu S, Baxevanis AD, Lynch HT, Elliot RM, Collins FS (2000). MLH3: a DNA mismatch repair gene associated with mammalian microsatellite instability. Nat Genet 24, 27-35.

Lipkin SM, Wang V, Stoler DL, Anderson GR, Kirsch I, Hadley D, Lynch HT, Collins FS (2001). Germline and somatic mutation analysis in the DNA mismatch repair gene MLH3: Evidence for somatic mutation in colorectal cancers. Hum Mutat 17, 389-96.

Liu B, Parsons R, Papadopoulos N, Nicolaides NC, Lynch HT, Watson P, Jass JR, Dunlop M, Wyllie A, Peltomäki P, et al. (1996). Analysis of mismatch repair genes in hereditary nonpolyposis colorectal cancer patients. Nat Med 2, 169-74.

Liu H-X, Zhou X-L, Liu T, Werelius B, Lindmark G, Dahl N, Lindblom A (2003). The role of hMLH3 in familial colorectal cancer. Cancer Res 63, 1894-9.

Liu T, Yan H, Kuismanen S, Percesepe A, Bisgaard ML, Pedroni M, Benatti P, Kinzler KW, Vogelstein B, Ponz De Leon M, et al. (2001). The role of hPMS1 and hPMS2 in predisposing colorectal cancer. Cancer Res 61, 7798-802.

Liu W, Dong X, Mai M, Seelan RS, Taniguchi K, Krishnadath KK, Halling KC, Cunningham JM, Boardman LA, Qian C, et al. (2000). Mutations in AXIN2 cause colorectal cancer with defective mismatch repair by activating beta-catenin/TCF signaling. Nat Genet 26, 146-7.

Loeb LA (2001). A mutator phenotype in cancer. Cancer Res 61, 3230-9.

Lothe RA, Peltomäki P, Meling GI, Aaltonen LA, Nyström-Lahti M, Pylkkänen L, Heimdal K, Andersen TI, Moller P, Rognum TO, et al. (1993). Genomic instability in colorectal cancer: relationship to clinicopathological variables and family history. Cancer Res 53, 5849-52.

Loukola A, Salovaara R, Kristo P, Moisio A-L, Kääriäinen H, Ahtola H, Eskelinen M, Härkönen N, Julkunen R, Kangas E, et al. (1999). Microsatellite instability in adenomas as a marker for hereditary nonpolyposis colorectal cancer. Am J Pathol 155, 1849-53.

Lowsky R, DeCoteau JF, Reitmair AH, Ichinohasama R, Dong WF, Xu Y, Mak TW, Kadin ME, Minden MD (1997). Defects of the mismatch repair gene MSH2 are implicated in the development of murine and human lymphoblastic lymphomas and are associated with the aberrant expression of rhombotin-2 (Lmo-2) and Tal-1 (SCL). Blood 89, 2276-82.

Lu SL, Akiyama Y, Nagasaki H, Saitoh K, Yuasa Y (1995). Mutations of the transforming growth factor-B type II receptor gene and genomic instability in hereditary nonpolyposis colorectal cancer. Biochem Biophys Res Comm 216, 452-7.

65

Lu S-L, Kawabata M, Imamura T, Akiyama Y, Nomizu T, Miyazono K, Yuasa Y (1998). HNPCC associated with germline mutation in the TGF-beta type II receptor gene. Nat Genet 19, 17-8.

Lynch HT, Shaw MW, Magnuson CW, Larsen AL, Krush AJ (1966). Hereditary factors in cancer. Study of two large midwestern kindreds. Arch Intern Med 117, 206-12.

Maki H, Sekiguchi M (1992). MutT protein specifically hydrolyzes a potent mutagenic substrate for DNA synthesis. Nature 355, 273-5.

Malkhosyan S, Rampino N, Yamamoto H, Perucho M (1996). Frameshift mutator mutations. Nature 382, 499-500.

Markowitz S, Wang J, Myerhoff L, Parsons R, Sun L, Lutterbaugh J, Fan RS, Zborowska E, Vogelstein B, Brattain M, Willson JKV (1995). Inactivation of the type II TGF-beta receptor in colon cancer cells with microsatellite instabillity. Science 268, 1336-8.

Marsischky GT, Filosi N, Kane MF, Kolodner R (1996). Redundancy of Saccharomyces cerevisiae MSH3 and MSH6 in MSH2-dependent mismatch repair. Genes Dev 10, 407-20.

Massague J (1996). TGF-β signaling:receptors, transducers, and mad proteins. Cell 85, 947-50.

Mellon I, Champe GN (1996). Products of DNA mismatch repair genes mutS and mutL are required for transcription-coupled nucleotide-excision repair of the lactose operon in Escherichia coli. Proc Natl Acad Sci USA 93, 1292-7.

Mellon I, Rajpal DK, Koi M, Boland CR, Champe GN (1996). Transcription-coupled repair deficiency and mutations in human mismatch repair genes. Science 272, 557-60.

Meloni AR, Smith EJ, Nevins JR (1999). A mechanism for Rb/p130-mediated transcription repression involving recruitment of the CtBP corepressor. Proc Natl Acad Sci USA 96, 9574-9.

Michaels ML, Cruz C, Grollman AP, Miller JH (1992). Evidence that mutM and mutY combine to prevent mutations by an oxidatively damaged form of guanine in DNA. Proc Natl Acad Sci USA 89, 7022-25.

Michelson RJ, Weinert T (2000). Closing the gaps among a web of DNA repair disorders. Bioessays 22, 966-9.

Mihaylova VT, Bindra RS, Yuan J, Campisi D, Narayanan L, Jensen R, Giordano F, Johnson RS, Rockwell S, Glazer PM (2003). Decreased expression of the DNA mismatch repair gene Mlh1 under hypoxic stress in mammalian cells. Mol Cell Biol 23, 3265-73.

Minowa O, Arai T, Hirano M, Monden Y, Nakai S, Fukuda M, Itoh M, Takano H, Hippou Y, Aburatani H, et al. (2000). Mmh/Ogg1 gene inactivation results in accumulation of 8-hydroxyguanine in mice. Proc Natl Acad Sci USA 97, 4156-61.

Miyaki M, Iijima T, Kimura J, Yasuno M, Mori T, Hayashi Y, Koike M, Shitara N, Iwama T, Kuroki T (1999). Frequent mutations of β-catenin and APC genes in primary colorectal tumors from patients with hereditary nonpolyposis colorectal cancer. Cancer Res 59, 4506-9.

Miyaki M, Konishi M, Tanaka K, Kikuchi-Yanoshita R, Muraoka M, Yasuno M, Igari T, Koike M, Chiba M, Mori T (1997). Germline mutations of MSH6 as the cause of hereditary nonpolyposis colorectal cancer. Nat Genet 17, 271-2.

66

Mohrenweiser HW, Wilson DM, Jones IM (2003). Challenges and complexities in estimating both the functional impact and the disease risk associated with the extensive genetic variation in human DNA repair genes. Mutat Res 526, 93-125.

Mori Y, Yin J, Rashid A, Leggett BA, Young J, Simms L, Kuehl PM, Langenberg P, Meltzer SJ, Stine OC (2001). Instabilotyping: Comprehensive identification of frameshift mutations caused by coding region microsatellite instability. Cancer Res 61, 6046-9.

Motyka B, Korbutt G, Pinkoski MJ, Heibein JA, Caputo A, Hobman M, Barry M, Shostak I, Sawchuk T, Holmes CFB, et al. (2000). Mannose 6-phosphate/insulin-like growth factor II receptor is a death receptor for granzyme B during cytotoxic T cell-induced apoptosis. Cell 103, 491-500.

Muto T, Bussey HJR, Morson BC (1975). The evolution of the cancer of the rectum. Cancer 36, 2251-76.

Narayanan L, Fritzell JA, Baker SM, Liskay RM, Glazer PM (1997). Elevated levels of mutation in multiple tissues of mice deficient in the DNA mismatch repair gene Pms2. Proc Natl Acad Sci USA 94, 3122-7.

Nghiem Y, Cabrera M, Cupples CG, Miller JH (1998). The mutY gene: a mutator locus in Escherichia coli that generates G:C to A:T transversions. Proc Natl Acad Sci USA 85, 9163-6.

Nicolaides NC, Papadopoulos N, Liu B, Wei YF, Carter KC, Ruben SM, Rosen CA, Haseltine WA, Fleischmann RD, Fraser CM, et al. (1994). Mutations of two PMS homologues in hereditary nonpolyposis colon cancer. Nature 371, 75-80.

Nugent KP, Phillips RK, Hodgson SV, Cottrell S, Smith-Ravin J, Pack K, Bodmer WF (1994). Phenotypic expression in familial adenomatous polyposis: partial prediction by mutation analysis. Gut 35, 1622-3.

Nyström-Lahti M, Kristo P, Nicolaides NC, Sheng-Yung C, Aaltonen LA, Moisio A-L, Järvinen HJ, Mecklin J-P, Kinzler KW, Vogelstein B, et al. (1995). Founding mutations and Alu-mediated recombination in hereditary colon cancer. Nat Med 1, 1203-6.

Nyström-Lahti M, Wu Y, Moisio A-L, Hofstra RMW, Osinga J, Mecklin J-P, Järvinen H, Leisti J, Buys CHCM, de la Chapelle A, Peltomäki P (1996). DNA mismatch repair gene mutations in 55 kindreds with verified or putative hereditary nonpolyposis colorectal cancer. Hum Mol Genet 5, 763-9.

Ogawa O, Becroft DM, Morison IM, Eccles MR, Skeen JE, Mauger DC, Reeve AE (1993). Constitutional relaxation of insulin-like growth factor II gene imprinting associated with Wilms' tumor and gigantism. Nat Genet 5, 408-12.

Okamoto K, Toyokuni S, Kim WJ, Ogawa O, Kakehi Y, Arao S, Hiai H, Yoshida O (1996). Overexpression of human mutT homologue gene messenger RNA in renal-cell carcinoma: evidence of persistent oxidative stress in cancer. Int J Cancer 65, 437-41.

Olschwang S, Tiret A, Laurent -Puig P, Muleris M, Parc R, Thomas G (1993). Restriction of ocular fundus lesions to a specific subgroup of APC mutations in adenomatous polyposis coli patients. Cell 75, 959-68.

Oshima M, Dinchuk JE, Kargman SL, Oshima H, Hancock B, Kwong E, Trzaskos JM, Evans JF, Taketo MM (1996). Suppression of intestinal polyposis in APC delta716 knockout mice by inhibition of cyclooxygenase 2 (COX-2). Cell 87, 803-9.

Ouyang H, Shiwaku HO, Hagiwara H, Miura K, Abe T, Kato Y, Ohtani H, Shiiba K, Souza RF, Meltzer SJ, Horii A (1997). The insulin-like growth factor II receptor gene is mutated in genetically unstable cancers of the endometrium, stomach, and colorectum. Cancer Res 57, 1851-4.

67

Papadopoulos N, Nicolaides NC, Wei YF, Ruben SM, Carter KC, Rosen CA, Haseltine WA, Fleischmann RD, Fraser CM, Adams MD, et al. (1994). Mutation of a MutL homolog in hereditary colon cancer. Science 263, 1625-9.

Park J, Betel D, Gryfe R, Michalickova K, Di Nicola N, Gallinger S, Hogue CWV, Redston M (2002). Mutation profiling of mismatch repair-deficient colorectal cancers using an in silico genome scan to identify coding microsatellites. Cancer Res 62, 1284-8.

Parsons R, Li GM, Longley M, Modrich P, Liu B, Berk T, Hamilton SR, Kinzler KW, Vogelstein B (1995). Mismatch repair deficiency in phenotypically normal human cells. Science 268, 738-40.

Peltomäki P (2001). DNA mismatch repair and cancer. Mutat Res 488, 77-85.

Peltomäki P (2003). Role of DNA mismatch repair defects in the pathogenesis of human cancer. J Clin Oncol 21, 1174-9.

Peltomäki P, Lothe RA, Aaltonen LA, Pylkkänen L, Nyström-Lahti M, Seruca R, David L, Holm R, Ryberg D, Haugen A, et al. (1993). Microsatellite instability is associated with tumors that characterize the hereditary non-polyposis colorectal carcinoma syndrome. Cancer Res 53, 5853-5.

Perucho M (2003). Tumors with microsatellite instability: many mutations, targets and paradoxes. Oncogene 22, 2223-5.

Perucho M, Boland CR, Thibodeau SN, Hamilton SR, Sidransky D, Eshleman JR, Burt RW, Meltzer SJ, Rodrigues-Bigas MA, Fodde R, et al. (1999). Cancer Res 59, 249-56.

Pieretti M, Khattar NH, Smith SA (2001). Common polymorphisms and somatic mutations in human base excision repair genes in ovarian and endometrial cancers. Mutat Res 432, 53-9.

Pochart P, Woltering D, Hollingsworth NM (1997). Conserved properties between functionally distinct MutS homologs in yeast. J Biol Chem 272, 30345-9.

Potocnik U, Glavac D, Golouh R, Ravnik-Glavac M (2001). Causes of microsatellite instability in colorectal tumors: implications for hereditary nonpolyposis colorectal cancer screening. Cancer Genet Cytogenet 126, 85-96.

Potocnik U, Glavac D, Ravnik-Glavac M (2003). Identification of novel genes with somatic frameshift mutations within coding mononucleotide repeats in colorectal tumors with high microsatellite instability. Gene Chrom Cancer 36, 48-56.

Prolla TA, Baker SM, Harris AC, Tsao JL, Yao X, Bronner CE, Zheng B, Gordon M, Reneker J, Arnheim N, et al. (1998). Tumor susceptibility and spontaneous mutation in mice deficient in Mlh1, Pms1 and Pms2 DNA mismatch repair. Nat Genet 18, 276-9.

Qin X, Liu L, Gerson SL (1999). Mice defective in the DNA mismatch gene PMS2 are hypersensitive to MNU induced thymic lymphoma and are partially protected by transgenic expression of human MGMT. Oncogene 18, 4394-400.

Rampino N, Yamamoto H, Ionov Y, Li Y, Sawai H, Reed JC, Perucho M (1997). Somatic frameshift mutations in the BAX gene in colon cancers of the microsatellite mutator phenotype. Science 275, 967-9.

Reardon JT, Bessho T, Kung HC, Bolton PH, Sancar A (1997). In vitro repair of oxidative DNA damage by human nucleotide excision repair system: possible explanations for neurodegeneration in Xeroderma pigmentosum patients. Proc Natl Acad Sci USA 94, 9463-8.

68

Reitmair AH, Redston M, Cai JC, Chuang TC, Bjerknes M, Cheng H, Hay K, Gallinger S, Bapat B, Mak TW (1996). Spontaneous intestinal carainomas and skin neoplasms in MSH2 deficient mice. Cancer Res 56, 3842-9.

Reitmair AH, Schmits R, Ewel A, Bapat B, Redston M, Mitri A, Waterhouse P, Mittrucker HW, Wakeham A, Liu B, et al. (1995). MSH2 deficient mice are viable and susceptible to lymphoid tumors. Nat Genet 11, 64-70.

Rex DK, Lehman GA, Hawes RH, Ulbright TM, Smith JJ (1991). Screening colonoscopy in asymptomatic average-risk persons with negative fecal occult blood tests. Gastroenterology 100, 64-7.

Ricciardone M, Özcelik T, Cevher B, Özdag H, Tuncer M, Gurgey A, Uzunalimoglu Ö, Cetinkaya H, Tanyeli A, Erken E, Özturk M (1999). Human MLH1 deficiency predisposes to hematological malignancy and neurofibromatosis type 1. Cancer Res 59, 290-3.

Riccio A, Aaltonen L, Godwin AK, Loukola A, Percesepe A, Salovaara R, Masciullo V, Genuardi M, Paravatou-Petsotas M, Bassi DE, et al. (1999). The DNA repair gene MBD4 (MED1) is mutated in human carcinomas with microsatellite instability. Nat Genet 23, 266-8.

Rizki A, Lundblad V (2001). Defects in mismatch repair promote telomerase-independent proliferation. Nature 411, 713-6.

Roldan-Arjona T, Wei YF, Carter KC, Klungland A, Anselmino C, Wang RP, Augustus M, Lindahl T (1997). Molecular cloning and functional expression of a human cDNA encoding the antimutator enzyme 8-hydroxyguanine-DNA glycosylase. Proc Natl Acad Sci USA 94, 8016-20.

Rossi DJ, Ylikorkala A, Korsisaari N, Salovaara R, Luukko K, Launonen V, Henkemeyer M, Ristimäki A, Aaltonen LA, Mäkelä TP (2002). Induction of cyclooxygenase-2 in a mouse model of Peutz-Jeghers polyposis. Proc Natl Acad Sci USA 99, 12327-32.

Räschle M, Marra G, Nyström-Lahti M, Schar P, Jiricny J (1999). Identification of hMutLbeta, a heterodimer of hMLH1 and hPMS1. J Biol Chem 5, 32368-75.

Saha S, Bardelli A, Buckhaults P, Velculescu VE, Rago C, Croix B, Romans KE, Choti MA, Lengauer C, Kinzler KW, Vogelstein B (2001). A phosphatase associated with metastasis of colorectal cancer. Science 294, 1343-6.

Sakumi K, Foruichi M, Tsuzuki T, Kakuma T, Kawabata S, Maki H, Sekiguchi M (1993). Cloning and expression of cDNA for a human enzyme that hydrolyzes 8-oxodGTP, a mutagenic substrate for DNA synthesis. J Biol Chem 268, 23524-30.

Salovaara R, Loukola A, Kristo P, Kääriäinen H, Ahtola H, Eskelinen M, Härkönen N, Julkunen R, Kangas E, Ojala S, et al. (2000). A population-based molecular detection of hereditary nonpolyposis colorectal cancer. J Clin Oncol 18, 2193-200.

Samowitz WS, Curtin K, Lin HH, Robertson MA, Schaffer D, Nichols M, Gruenthal K, Leppert MF, Slattery ML (2001). The colon cancer burden of genetically defined hereditary nonpolyposis colon cancer. Gastroenterology 121, 830-8.

Samowitz WS, Powers MD, Spirio LN, Nollet F, van Roy F, Slattery ML (1999). Beta-catenin mutations are more frequent in small colorectal adenomas than in larger adenomas and invasive carcinomas. Cancer Res 59, 1442-4.

69

Sampson JR, Dolwani S, Jones S, Eccles D, Ellis A, Evans DG, Fraylin I, Jordan S, Maher ER, Mak T, et al. (2003). Autosomal recessive colorectal adenomatous polyposis due to inherited mutations of MYH. Lancet 362, 39-41.

Santucci-Darmanin S, Neyton S, Lespinasse F, Saunieres A, Gaudray P, Paquis-Flucklinger (2002). The DNA mismatch-repair MLH3 protein interacts with MSH4 in meiotic cells, supporting a role for this MutL homolog in mammalian meiotic recombination. Hum Mol Genet 11, 1697-706.

Santucci-Darmanin S, Walpita D, Lespinasse F, Desnuelle C, Ashley T, Paquis-Flucklinger (2000). MSH4 acts in conjunction with MLH1 during mammalian meiosis. FASEB J 14, 1539-47.

Schmutte C, Marinescu RC, Sadoff MM, Guerrette S, Overhauser J, Fishel R (1998). Human exonuclease I interacts with the mismatch repair protein hMSH2. Cancer Res 58, 4537-42.

Schmutte C, Sadoff MM, Shimm KS, Acharya S, Fishel R (2001). The interaction of DNA mismatch repair proteins with humman exonuclease I. J Biol Chem 276, 33011-8.

Schwartz S, Yamamoto H, Navarro M, Maestro M, Reventos J, Perucho M (1999). Frameshift mutations at mononucleotide repeats in caspase-5 and other target genes in endometrial and gastrointestinal cancer of the microsatellite mutator phenotype. Cancer Res 39, 2995-3002.

Shibata D (2001). Talking tumors: histories from sequencies. Ann Med 33, 148-52.

Shibata D, Navidi W, Salovaara R, Zhi-Hua L, Aaltonen LA (1996). Somatic microsatellite mutations as molecular tumor clocks. Nat Med 2, 676-81.

Shibata D, Peinado MA, Ionov Y, Malkhosyan S, Perucho M (1994). Genomic instability in simple repeat sequences is a very early somatic event in colorectal tumorigenesis that persists after transformation. Nat Genet 6, 273-81.

Shin J-H, Shin Y-K, Ku J-L, Jeong S-Y, Hong S-H, Park S-Y, Kim W-H, Park J-G (2003). Mutations of the Birt-Hogg-Dube (BHD) gene in sporadic colorectal carcinomas and colorectal carcinoma cell lines with microsatellite instability. J Med Genet 40, 364-7.

Sieber OM, Lipton L, Crabtree M, Heinimann K, Fidalgo P, Phillips RKS, Bisgaard M-L, Orntoft T, Aaltonen LA, Hodgson SV, et al. (2003). Multiple colorectal adenomas, classic adenomatous polyposis and germline mutations in MYH. N Engl J Med 348, 845-7.

Slupska MM, Baikalov C, Luther WM, Chiang J-H, Wei Y-F, Miller JH (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. J Bacteriol 178, 3885-92.

Smyrk TC, Watson P, Kaul K, Lynch HT (2001). Tumor-infiltrating lymphocytes are a marker for microsatellite instability in colorectal carcinoma. Cancer 91, 2417-22.

Souza RF, Appel R, Yin J, Wang S, Smolinski KN, Abraham JM, Zou T-T, Shi Y-Q, Lei J, Cottrell J (1996). Microsatellite instability in the insulin-like growth factor II receptor gene in gastrointestinal tumours. Nat Genet 14, 255-7.

Spirio L, Olschwang S, Groden J, Robertson M, Samowitz W, Joslyn G, Gelbert L, Thliveris A, Carlson M, Otterud B (1993). Alleles of the APC gene: an attenuated form of familial polyposis. Cell 75, 951-7.

Sugawara N, Paques F, Colaiacovo M, Haber JE (1997). Role of Saccharomyces cerevisiae Msh2 and Msh3 repair proteins in double strand break-induced recombination. Proc Natl Acad Sci USA 94, 9214-9.

70

Sun X, Zheng L, Shen B (2002). Functional alterations of human exonuclease 1 mutants identified in atypical hereditary nonpolyposis colorectal cancer syndrome. Cancer Res 62, 6026-30.

Suomalainen A, Syvänen A-C (2000). Quantitative analysis of human DNA sequences by PCR and solid-phase minisequencing. Mol Biotechnol 15, 123-31.

Suzuki K, Dai T, Suzuki I, Dai Y, Yamashita K, Perucho M (2002). Low mutation incidence in polymorphic noncoding short mononucleotide repeats in gastrointestinal cancer of the microsatellite mutator phenotype pathway. Cancer Res 62, 1961-5.

Syvänen A-C, Sajantila A, Lukka M (1993). Identification of individuals by analysis of biallelic DNA markers, using PCR and solid-phase minisequencing. Am J Hum Genet 42, 46-59.

Takagi Y, Kohmura H, Futamura M, Kida H, Tanemura H, Shimokawa K, Saji S (1996). Somatic alterations of the DPC4 gene in human colorectal cancers in vivo. Gastroenterology 111, 1369-72.

Talbot IC, Burt R, Järvinen H, Thomas G (2000). Familial adenomatous polyposis. In: Pathology and genetics of tumors of the digestive system, World Health Organization Classification of tumors. Eds. Hamilton SR, Aaltonen LA. Lyon: IARC Press, 120-5.

Thibodeau SN, Bren G, Shaid D (1993). Microsatellite instability in cancer of the proximal colon. Science 260, 186-9.

Thompson JD, Higgins DG, Gibson TJ (1994). Clustal W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res 22, 4673-80.

Thompson LH, Schild D (2002). Recombinational DNA repair and human disease. Mutat Res 509, 49-78.

Thorstensen L, Diep CB, Meling GI, Aagesen TH, Ahrens CH, Rognum TO, Lothe RA (2001). WNT1 inducible signaling pathway protein 3, WISP-3, a novel target gene in colorectal carcinomas with microsatellite instability. Gastroenterology 121, 1275-80.

Tishkoff DX, Amin NS, Viars CS, Arden KC, Kolodner RD (1998). Identification of a human gene encoding a homologue of Saccharomyces cerevisiae EXOI, an exonuclease implicated in mismatch repair and recombination. Cancer Res 58, 5027-31.

Tishkoff DX, Boerger AL, Bertrand P, Filosi N, Gaida GM, Kane MF, Kolodner RD (1997). Identification and characterization of Saccharomyces cerevisiae EXO1, a gene encoding an exonuclease that interacts with MSH2. Proc Natl Acad Sci USA 94, 7487-92.

Toft NJ, Winton DJ, Kelly J, Howard LA, Dekker M, teRiele H, Arends MJ, Wyllie AH, Margiosn GP, Clarke AR (1999). Msh2 status modulates both apoptosis and mutation frequency in the murine small intestine. Proc Natl Acad Sci USA 96, 3911-5.

Tomlinson IPM, Roylance R, Houlston RS (2001). Two hits revisited again. J Med Genet 38, 81-5.

Toyota M, Ahuja N, Ohe-Toyota M, Herman JG, Baylin SB, Issa JP (1999). CpG island methylation phenotype in colorectal cancer. Proc Natl Acad Sci USA 96, 8681-6.

Tran HT, Gordenin DA, Resnick MA (1999). The 3'-5' exonucleases of DNA polymerases delta and epsilon and the 5'-3' exonuclease EXOI have major roles in postreplication mutation avoidance in Saccharomyces cerevisiae. Mol Cell Biol 19, 2000-7.

71

Tsao JL, Davis SD, Baker SM, Liskay RM, Shibata D (1997). Intestinal stem cell divisions and genetic diversity: a computer and experimental analysis. Am J Pathol 151, 573-9.

Tsao JL, Tavare S, Salovaara R, Jass JR, Aaltonen LA, Shibata D (1999). Colorectal adenoma and cancer divergence: evidence of multi-lineage progression. Am J Pathol 154, 815-24.

Tsao JL, Yatabe Y, Salovaara R, Järvinen HJ, Mecklin JP, Aaltonen LA, Tavare S, Shibata D (2000). Genetic reconstruction of individual colorectal tumor histories. Proc Natl Acad Sci USA 97, 1236-41.

Tsubouchi H, Ogawa H (2000). Exo1 roles for repair of DNA double-strand breaks and meiotic crossing over in Saccharomyces cerevisiae. Mol Biol Cell 11, 2221-33.

Tsuzuki T, Egashira A, Igarashi H, Iwakuma T, Nakatsuru Y, Tominaga Y, Kawate H, Nakao K, Nakamura K, Ide F, et al. (2001). Spontaneous tumorigenesis in mice defective in the MTH1 gene encoding 8-oxo-dGTPase. Proc Natl Acad Sci USA 98, 11456-61.

Tuo J, Muftuoglu M, Chen C, Jaruga P, Selzer RR, Brosh RM Jr, Rodrigues H, Dizdaroglu M, Bohr VA (2001). The Cockayne syndrome group B gene product is involved in general genome base excision repair of 8-hydroxyguanine in DNA. J Biol Chem 276, 45772-9.

Vasen HF, Mecklin J-P, Khan PM, Lynch HT (1991). The International Collaborative Group on Hereditary Nonpolyposis colorectal Cancer (ICG-HNPCC). Dis Colon Rectum 34, 424-5.

Vasen HF, Stormorken A, Menko FH, Nagengast FM, Kleibeuker JH, Griffioen G, Taal BG, Moller P, Wijnen JT (2001). Msh2 mutation carriers are at higher risk of cancer than Mlh1 mutation carriers: a study of hereditary nonpolyposis colorectal cancer families. J Clin Oncol 19, 4074-80.

Vasen HF, Watson P, Mecklin J-P, Lynch HT (1999). New clinical criteria for hereditary nonpolyposis colorectal cancer (HNPCC, Lynch syndrome) proposed by the international Collaborative Group on HNPCC. Gastroenterology 116, 1453-6.

Vasen HF, Wijnen JT, Menko FH, Kleibeuker JH, Taal BG, Griffioen G, Nagengast FM, Meijers-Heijboer EH, Bertario L, Varesco L, et al. (1996). Cancer risk in families with hereditary nonpolyposis colorectal cancer diagnosed by mutation analysis. Gastroenterology 110, 1020-7.

Vogelstein B, Fearon ER, Hamilton SR, Kern SE, Preisinger AC, Leppert M, Nagamura Y, White R, Smits AMM, Bos JL (1988). Genetic alterations during colorectal tumor development. N Engl J Med 319, 525-32.

Wang Q, Lasset C, Desseigne F, Frappaz D, Bergeron C, Navarro C, Ruano E, Puisieux A (1999a). Neurofibromatosis and early onset of cancers in hMLH1-deficient children. Cancer Res 59, 294-7.

Wang TF, Kleckner N, Hunter N (1999b). Functional specificity of MutL homologs in yeast: evidence for three Mlh1-based heterocomplexes with distinct roles during meiosis in recombination and mismatch correction. Proc Natl Acad Sci USA 96, 13914-9.

Wang T-L, Rago C, Silliman N, Ptak J, Markowitz S, Willson JKV, Parmigiani G, Kinzler KW, Vogelstein B, Velculescu VE (2002). Prevalence of somatic alterations in the colorectal cancer cell genome. Proc Natl Acad Sci USA 99, 3076-80.

Wang Y, Cortez D, Yazdi P, Neff N, Elledge SJ, Qin J (2000). BASC, a super complex of BRCA1-associated proteins involved in the recognition and repair of aberrant DNA structures. Genes Dev 14, 927-39.

72

Ward R, Meagher A, Tomlinson I, O'Connor T, Norrie M, Wu R, Hawkins N (2001). Microsatellite instability and the clinicopathological features of sporadic colorectal cancer. Gut 48, 821-9.

Warthin AS (1913). Heredity with reference to carcinoma. Arch Intern Med 12, 546-55.

Watson P, Lynch HT (1993). Extracolonic cancer in hereditary nonpolyposis colorectal cancer. Cancer 71, 677-85.

Wei K, Clark AB, Wong E, Kane MF, Mazur DJ, Parris T, Kolas NK, Russell R, Hou H Jr, Kneitz B, et al. (2003). Inactivation of exonuclease 1 in mice results in DNA mismatch repair defects, increased cancer suscptibility, and male and female sterility. Genes Dev 17, 603-14.

Weinberg RA (1994). Oncogenes and tumor suppressor genes. CA Cancer J Clin 44, 160-70.

Welcsh PL, King M-C (2001). BRCA1 and BRCA2 and the genetics of breast and ovarian cancer. Hum Mol Genet 10, 705-13.

Welsh KM, Lu AL, Clark S, Modrich P (1987). Isolation and characterization of the Escherichia coli mutH gene product. J Biol Chem 262, 15624-9.

Whiteside D, McLeod R, Graham G, Steckley JL, Booth K, Somerville MJ, Andrew SE (2002). A homozygous germ-line mutation in the human MSH2 gene predisposes to hematological malignancy and multiple cafe-au-lait spots. Cancer Res 62, 359-62.

Wijnen J, de Leeuw W, Vasen H, van der Klift H, Moller P, Stormorken A, Meijers-Heijboer H, Lindhout D, Menko F, Vossen S, et al. (1999). Familial endometrial cancer in female carriers of MSH6 germline mutations. Nat Genet 23, 142-4.

Wijnen J, Meera Khan P, Vasen H, van der Klift H, Mulder A, van Leeuwen-Cornelisse I, Bakker B, Losekoot M, Moller P, Fodde R (1997). Hereditary nonpolyposis colorectal cancer families not complying with the Amsterdam criteria show extremely low frequency of mismatch-repair gene mutations. Am J Hum Genet 61, 329-35.

Woerner SM, Benner A, Sutter C, Schiller M, Yuan YP, Keller G, Bork P, Doeberitz MK, Gebert JF (2003). Pathogenesis of DNA repair-deficient cancers: a statistical meta-analysis of putative real common target genes. Oncogene 22, 2226-35.

Woerner SM, Gebert J, Yuan YP, Sutter C, Ridder R, Bork P, von Knebel Doeberitz M (2001). Systematic identification of genes with coding microsatellites mutated in DNA mismatch repair deficient cancer cells. Int J Cancer 93, 12-9.

Wong NA, Pignatelli M (2002). B-catenin-A linchpin in colorectal carcinogenesis. Am J Pathol 160, 389-401.

Wood RD, Mitchell M, Sgouros J, Lindahl T (2001). Human DNA repair genes. Science 291, 1284-9.

Wu J, Gu L, Wang H, Geacintov NE, Li GM (1999a). Mismatch repair processing of carcinogen-DNA adducts triggers apoptosis. Mol Cell Biol 19, 8292-301.

Wu Y, Berends MJ, Mensink RG, Kemoinga C, van der Zee AG, Hollema H, Kleibeuker JH, Buys CH, Hofstra RM (1999b). Association of hereditary nonpolyposis colorectal cancer-related tumors displaying low microsatellite instability with MSH6 germline mutations. Am J Hum Genet 65, 1291-8.

73

Wu Y, Berends MJ, Post JG, Mensink RG, Verlind E, van Der Sluis T, Kemoinga C, Sijmons RH, van der Zee AG, Hollema H, et al. (2001a). Germline mutations of EXO1 gene in patients with hereditary nonpolyposis colorectal cancer (HNPCC) and atypical HNPCC forms. Gastroenterology 120, 1580-7.

Wu Y, Berends MJ, Sijmons RH, Mensink RG, Verlind E, Kooi KA, van Der Sluis T, Kempinga C, van der Zee AG, Hollema H, et al. (2001b). A role for MLH3 in hereditary nonpolyposis colorectal cancer. Nat Genet 29, 137-8.

Yamamoto H, Gil J, Schwartz Jr S, Perucho M (2000). Frameshift mutations in Fas, Apaf-1, and Bcl-10 in gastro-intestinal cancer of the microsatellite mutator phenotype. Cell Death Diff 7, 238-9.

Yao X, Buermeyer AB, Narayanan L, Tran D, Baker SM, Prolla TA, Glazer PM, Liskay RM, Arnheim N (1999). Different mutator phenotypes in Mlh1 versus Pms2 deficient mice. Proc Natl Acad Sci USA 96, 6850-5.

Yuen ST, Chan TL, Ho JWC, Chan ASY, Chung LP, Lam PWY, Tse CW, Wyllie AH, Leung SY (2002). Germline, somatic and epigenetic events underlying mismatch repair deficiency in colorectal and HNPCC-related cancers. Oncogene 21, 7585-92.

Zeng Q, Dong J-M, Guo K, Li J, Tan H-X, Koh V, Pallen CJ, Manser E, Hong W (2003). PRL-3 and PRL-1 promote cell migration, invasion, and metastasis. Cancer Res 63, 2716-22.

Zhang H, Richards B, Wilson T, Lloyd M, Cranston A, Thorburn A, Fishel R, Meuth M (1999). Apoptosis induced by overexpression of hMSH2 or hMLH1. Cancer Res 59, 3021-7.

Zhang L, Yu J, Willson JKV, Markowitz SD, Kinzler KW, Vogelstein B (2001). Short mononucleotide repeat sequence variability in mismatch repair-deficient cancers. Cancer Res 61, 3801-5.

Zhou BBS, Elledge SJ (2000). The DNA damage response: putting checkpoints in perspective. Nature 408, 433-9.