DEFINING REGIONS WITHIN THE HUMAN GENOME THAT SEEMINGLY HARBOR NOVEL BREAST CANCER SUSEPTIBILITY GENES IN HIGH RISK

Transcription

1 TEL AVIV UNIVERSITY SACKLER FACULTY OF MEDICINE THE DR. MIRIAM AND SHELDON G. ADELSON DEPARTMENT OF HUMAN GENETICS AND MOLECULAR MEDICINE DEFINING REGIONS WITHIN THE HUMAN GENOME THAT SEEMINGLY HARBOR NOVEL BREAST CANCER SUSEPTIBILITY GENES IN HIGH RISK ASHKENAZI BREAST CANCER FAMILIES IN ISRAEL THESIS SUBMITTED FOR THE DEGREE DOCTOR OF PHILOSOPHY BY TAL DISTELMAN-MENACHEM SUBMITTED TO THE SENATE OF TEL AVIV UNIVERSITY JUNE 2009

2 This work was carried out under the supervision Of Prof. Eitan Friedman, head of the Oncogenetics unit, Sheba Medical Center, Tel Hashomer

3 Acknowledgements I am very grateful to my supervisor Professor Eitan Friedman for his guidance, for his willingness to help and advice; and for giving me the tools to manage with this study and others. My appreciation is endless. I thank Professor Eitan Friedman's chief of lab, Yael Laitman, for her steadfast assistance, moral and professional support along the way. I thank Dr. Tali Shapira for her long hours of work and assistance. I thank my fellow students Sigal Starinsky-Elbaz, Tair Paley-Kontorovich, Roni Milgrom, Shimrit Cohen and Gili Galor-Haskel for their friendship and support, and for sharing with me information and ideas. I thank my wonderful father and mother in law for their help and support. I thank my beloved parents and sister for their never ending love, care and support. A special thank you to my cherished husband Dror and my two amazing children Tamar and Alon, for being there for me. Always. This research project was in part funded by a grant from the Silva De Gomez foundation from the Tel-Aviv University Sackler School of Medicine and by a donation from Menachem Einan, President of Azrieli group, for which I am truly grateful. Thank you all from the bottom of my heart.

4 Table of contents Introduction... 1 Breast cancer... 1 Risk factors for developing breast cancer... 1 Ovarian cancer... 8 BRCA1 and BRCA Major gene rearrangements in BRCA1 and BRCA SNPs Microsatellites Haplotype Linkage disequilibrium Association studies Research goals Materials and Methods Patients and Families DNA extraction Polymerase chain reaction (PCR) Long range PCR Selecting and assigning pathogenicity of missense mutations in BRCA Genotyping for seemingly pathogenic BRCA1 missense mutations Haplotype analysis Multiplex Ligation-dependent Probe Amplification (MLPA) Quantitative PCR Association studies Direct sequencing... 49

5 Results Patients' Characteristics Assessing the possible contribution of germeline mutations in BRCA1 and BRCA2 genes to inherited breast/ovarian cancer in the study population General Identification of the 185delAG mutation in exon 2 of BRCA1 gene Identification of the mutation 5382insC in exon 20 of BRCA Identification of the mutation 6174delT in exon 11 of BRCA Haplotype analysis using intragenic markers in BRCA1 and BRCA2 genes Haplotype analysis of 185delAG BRCA1 mutation carriers Haplotype analysis of BRCA1 of the study participants Haplotype analysis of 6174delT BRCA2 mutation carriers Haplotype analysis of BRCA2 of the study participants Genotyping for seemingly pathogenic missense mutations in BRCA Evaluating the existence of major gene rearrangements in BRCA1 and BRCA2 using Multiplex Ligation-dependent Probe Amplification (MLPA) BRCA1 major gene rearrangements BRCA2 major gene rearrangements Quantitative PCR for BRCA1 and BRCA qpcr for BRCA qpcr for BRCA Long-range PCR for BRCA Direct sequencing for the BRCA genome-wide association study (GWAS) GWAS in 249 Familial Breast Cancer Cases... 75

6 9.2 Replication Analysis Using a Custom SNP Array Direct Sequencing of the RNF146 and ECHDC1 genes as candidates for inherited breast and ovarian cancer in Jewish Ashkenazi women Direct sequencing of the RNF146 gene Direct sequencing of the ECHDC1 gene Discussion Determination of BRCA1 and BRCA2 genes' involvement in familial breast cancer in the Jewish Ashkenazi population Genome-wide association study (GWAS) Direct Sequencing of The RNF146 and ECHDC1 genes as candidates for inherited breast and ovarian cancer in Jewish Ashkenazi women References Appendixes (1-7)...

7 Table of Figures Figure 1: Protein structure of BRCA Figure 2: Protein structure of BRCA Figure 3: The roles of BRCA1 and BRCA2 in DNA repair Figure 4: Relationship between mutational events and formation of breast or ovarian cancers Figure 5: BRCA1 and BRCA2 predominant Jewish Ashkenazi mutations Figure 6: The position of the SNP's rs and rs on the FGFR2 gene Figure 7: PCR procedure Figure 8: The ABI PRISM 310 Genetic Analyzer is an automated single-capillary genetic analyzer Figure 9: The principles of MLPA Figure 10: Wild type DNA, deletion of exon 13 and deletion of exon 22 respectively Figure 11: Quantitative PCR procedure Figure 12: The amount of fluorescence obtained in each amplification cycle for each reaction Figure 13: Restriction enzyme analysis of the 185delAG mutation in BRCA Figure 14: Restriction enzyme analysis of the 5382insC mutation in BRCA Figure 15: Restriction enzyme analysis of the 6174delT mutation in BRCA Figure 16: Haplotype of the intragenic marker D17S1323 in BRCA1 of five carriers Figure 17: Haplotype of the intragenic marker D17S1323 of three samples from the study participants in BRCA Figure 18: The pedigree of the samples that were used on the haplotype of the intragenic marker D17S1323 in BRCA

8 Figure 19: The genotype of the intragenic marker D13S171 in the BRCA2 of three carriers Figure 20: Haplotype of the intragenic marker D13S171 of three samples from the study participants in BRCA Figure 21: The pedigree of the samples that were used on the haplotype of the intragenic marker D13S171 in BRCA Figure 22: Enzyme restriction products of the missense mutation M1008I in BRCA Figure 23: Sequencing of one of the samples to ensure that it is mutated for the M1008I mutation Figure 24: Sequencing of the S1040N in the mutated sample of M1008I Figure 25: Sequencing of the E1038G in the mutated sample of M1008I Figure 26: Restriction enzyme analysis of PCR products to detect the missense mutation R866C in BRCA Figure 27: Restriction enzyme analysis of PCR products to detect the missense mutation R841W in BRCA Figure 28: Restriction enzyme analysis of PCR products to detect the missense mutation Y179C in BRCA Figure 29: Restriction enzyme analysis of PCR products to detect the missense mutation C61G in BRCA Figure 30: Restriction enzyme analysis of PCR products to detect the missense mutation R331S in BRCA Figure 31: Dosage changes found in exons 17 and 19 in BRCA Figure 32: Amplification found in exon 21 in BRCA Figure 33: Dosage changes found in the internal control segments Figure 34: Examples of large genomic rearrangements in BRCA2 detected by MLPA... 69

9 Figure 35: The 40's probe of the SALSA P045 kit binds only in the presence of the mutation 1100delC in CHEK2 gene Figure 36: Sequencing of the 1100delT CHEK2 mutation in one of the mutated samples.. 70 Figure 37: qpcr for BRCA1 exon 19 sample Figure 38: qpcr for BRCA1 exon 19 sample Figure 39: qpcr for BRCA2 exon 1b sample Figure 40: Long-range pcr for exon 11 in BRCA Figure 41: LD image of linkage disequilibrium analysis of SNP's rs and rs in the RNF146 gene Figure 42: LD image of linkage disequilibrium analysis of SNP's rs and rs in the ECHDC1 gene Figure 43: Sequencing of one of the samples containing the T to C change in intron 3 of the gene, 47 nucleotides into the intron -IVS3 + 47T/C... 80

10 Table of Tables Table 1: Incidence rate of ovarian cancer... 8 Table 2: Major gene rearrangements in BRCA Table 3: Major gene rearrangements in BRCA Table 4: The PCR program for the predominant mutations in the Jewish Ashkenazi population in BRCA1 BRCA2 genes Table 5: The annealing step for the predominant mutations in the Jewish Ashkenazi population in BRCA1 BRCA2 genes Table 6: Conditions of the restriction reaction for the predominant mutations in the Jewish Ashkenazi population in BRCA1 BRCA2 genes Table 7: The restriction enzymes for the common mutation in BRCA1/2 in Ashkenazi Jews Table 8: Mutation designation, primer sequences for the BRCA1 missense mutation Table 9: Annealing step temperature and duration in the PCR program for the BRCA1 missense mutation Table 10: Restriction enzymes and incubation time for the BRCA1 missense mutation Table 11: Volume of restriction enzymes for the BRCA1 missense mutation Table 12: Markers fluorescent label Table 13: The amounts of each reagent needed for optimal PCR reaction Table 14: Primers for the intragenic markers in BRCA1 and BRCA Table 15: The annealing step of the PCR reaction for each of the markers Table 16: PCR products dilutions Table 17: Hybridization reaction of MLPA Table 18: Ligation reaction of MLPA Table 19: PCR reaction of MLPA

11 Table 20: Primers and probes for BRCA1 in the qpcr reaction Table 21: Primers and probes for BRCA2 in the qpcr reaction Table 22: RNF146 gene exon and primer sequences Table 23: ECHDC1 gene exon and primer sequences Table 24: RNF146 and ECHDC1 annealing step temperature and duration Table 25: A summary of the large genomic rearrangements in BRCA1 detected by MLPA Table 26: A summary of the large genomic rearrangements in BRCA2 detected by MLPA Table 27: Number of significant associations after initial AJ familial GWAS Table 28: Regions of the genome that showed the strongest associations with AJ breast cancer Table 29: Haplotype statistics Table 30: SNP's location

12 Table of abbreviations AJ - ASR - ATM - BCAC - BMI - BRCA1 - BRCA2 - BRCT - Ashkenazi Jewish age standtardized rate ataxia telangiectasia mutated Breast Cancer Association Consortium body mass index breast cancer 1, early onset breast cancer 2, early onset BRCA carboxy terminal BRIP1 - BRCA1 interacting protein C-terminal helicase 1 CHEK2 - CI - C t - CHK2 checkpoint homolog confidence interval threshold cycle CtIP (RBBP8) - retinoblastoma binding protein 8 DGGE - Denaturing Gradient Gel electrophoresis EAv3 - Affymetrix GeneChip Early Access Version 3 ECHDC1 - Enoyl-coenzyme A hydratase domain containing 1 ER - FA - F-CSGE - estrogen receptor Fanconi anemia fluorescent-conformation sensitive gel electrophoresis FGFR2 - fibroblast growth factor receptor 2 GWAS - genome-wide association study HCN1 - hyperpolarization activated cyclic nucleotide-gated potassium channel 1 HR - HRT - homologous recombination hormone replacement therapy

13 HWE - LD - LOD - LOH - Hardy Weinberg equilibrium Linkage disequilibrium logarithm of odds loss of heterozygosity LSP1 - lymphocyte-specific protein 1 MAP3K1 - mitogen-activated protein kinase kinase kinase 1 MLPA - MRPS30 - NBS1 (NBN) - NYBCS - OR - PALB2 - PCNA - PCR - PR - PTEN - PTT - qpcr - Multiplex ligation-dependent probe amplification mitochondrial ribosomal protein S30 nibrin New York Breast Cancer Study odds ratio partner and localizer of BRCA2 proliferating cell nuclear antigen Polymerase chain reaction Progesteron receptor phosphatase and tensin homolog Protein Truncation Test Quantitative PCR RNF146 - RING finger protein 146 SNPs - TDT - single nucleotide polymorphisms transmission disequilibrium test TNRC9 (TOX3) - TOX high mobility group box family member 3

14 Abstract Breast cancer is the most prevalent malignancy in the western world, with a life time risk of 13% for Western world women. Most of breast cancer cases are sporadic, but ~10% of the cases are thought to have a hereditary-familial basis. In some of the inherited cases germline mutation(s) can be detected in either the BRCA1 or BRCA2 genes. In the presence of a family history of early onset breast and ovarian cancer, the probability of harboring a mutation in either gene is 75-85%, whereas only 40-50% of site specific breast cancer families harbor BRCA1/2 mutations. Thus, the majority of multiple case breast cancer families are unaccounted for by mutations in these 2 genes. In Ashkenazi Jews, there is a high population frequency (2.5%) of three founder mutations in BRCA1 (185delAG, 5382insC) and BRCA2 (6174delT). Except for the predominant mutations in BRCA1 and BRCA2 in Jewish individuals, there are only a handful of pathogenic mutations within these genes. It may imply novel genes may underlie inherited susceptibility to breast/ovarian cancer in Jewish individuals. There are a number of options to account for the paucity of BRCA1 BRCA2 mutations in Jewish high risk families: existence of mutations in unscreened regions (i.e, promoter area), major gene rearrangement not detectable by PCR and the existence of new gene(s) in different chromosomal locations. The objective of the present study was to identify the rate of Major gene rearrangements in BRCA1 and BRCA2 in Jewish Ashkenazi high risk families and to define the contribution of novel candidate genes within new regions in the human genome that are seemingly associated with breast / ovarian cancer risk, by searching for germline mutations in Jewish Ashkenazi high risk breast / ovarian cancer families. The current study focused on Ashkenazi Jews for several reasons: breast cancer is more common among Ashkenazi women than other ethnic origin in Israel, the spectrum of mutations in BRCA1 and BRCA2 is limited and well defined and Ashkenazi Jews are a classic example of genetic isolate. The study was carried out in several steps as described:

15 1. Defining the criteria for the high risk families: All participants had breast or ovarian cancer, and in addition, one or more of the eligibility criteria: 1) under 40 years of age at time of diagnosis of breast cancer or 2) over 40 years of age at time of diagnosis of breast cancer and at least one of the following: a) bilateral breast cancer or ovarian cancer; b) at least one first degree relative with breast cancer; c) one or more first or second degree relatives with bilateral breast cancer or ovarian cancer; d) at least two relatives of the paternal side with breast cancer and e) at least one relative diagnosed for breast cancer under the age of 40 years. 2. Haplotyping of the selected Jewish Ashkenazi high risk breast / ovarian cancer families with intragenic BRCA1 and BRCA2 markers. 3. Genotyping for seemingly pathogenic missense mutations in BRCA1. 4. Analysis of large genomic rearrangements in BRCA1/2 5. Association analysis of the entire human genome- GWAS- genome wide association study. 6. Genotyping for germline mutations within candidate genes in the regions within the human genome, which showed association with breast / ovarian cancers. Overall, 105 Ashkenazi families, and 20 non Ashkenazi families met the eligibility criteria of the study, and all participants were found to be negative for the three predominant "Ashkenazi" mutations in BRCA1 and BRCA2. Haplotype analysis with 3 intragenic BRCA1markers and 3 intragenic BRCA2 markers showed that breast cancer patients in this study and their relatives do not share the same alleles in either gene. These results strongly support the notion that BRCA1 and BRCA2 genes\mutations do not account for the observed familial aggregation of breast cancer in the families that formed the basis of the study. The second method used to assess the possible contribution of mutations in BRCA1 to the observed familial clustering of breast cancer was to genotype for putatively pathogenic and/or common missense mutations in BRCA1 (R866C, R331S, R841W, Y179C, C61G, M1008I). Except for two women who carried the M1008I mutation, a seemingly rare polymorphism, no other genotyped individual carried any of the other tested genetic variants. The next step was to evaluate

16 if and to what extent large genomic rearrangements in BRCA1 BRCA2 exist in the Jewish Ashkenazi high risk population using a combined approach of Multiplex Ligation-dependent Probe Amplification (MLPA), Quantitative PCR and Long Range PCR. In BRCA1, an amplification of exons 17 and 19 had significant odds ratio that was statistically significant (p<0.1) in only two samples, and among the samples that reach partial statistical significance (p<0.1), amplifications were detected in exons 5, 17, 19 and 21 in 3 patients. In BRCA2, dosage changes with statistically significant (p<0.1) were found in exons 1b, 3 (amplification) and 11a, 25 (deletion), in 6 patients. These dosage changes had significant odds ratio. Yet, these results could not be confirmed or validated independently by another technique. After eliminating the possibility that family history and clustering of breast cancer in the study population is attributed to mutations in BRCA1 and BRCA2 genes, an association study to find new chromosomal loci in the genome associated with breast cancer was carried out. DNA samples were sent to Dr. Kenneth Offit from the Memorial Sloan Kettering Cancer Center in New York, who combined our families with additional American Jewish Ashkenazi, high-risk families. In phase 1, 249 (including 45 families from the Sheba Medical center) probands from high risk AJ kindreds and 299 cancer-free AJ controls (including 29 controls from the Sheba Medical Center) were genotyped. In phase 2, a cohort that consisted of a fully independent set of 950 consecutive AJ breast cancer cases and 979 age-matched cancer-free AJ controls were genotyped, and in the third phase 243 sporadic Jewish breast cancer patients and 187 asymptomatic controls were also genotyped and a pooled analysis was eventually done. The results of the association study show that the association with the RNF14 and ECHDC1 region at 6q22 was the strongest and most consistent in this study. Direct sequencing of the RNF146 and ECHDC1 genes, showed only a rare intronic polymorphism in 4% of the samples in the ECHDC1 gene but no bone fide mutations in any of the 105 samples genotyped. Based on these results, it seems that germline mutations in both the ECHDC1 and RNF146 genes contribute only marginally, if at all, to the burden of inherited predisposition to breast cancer among AJ.

17 Introduction Breast cancer Breast cancer is the most common cancer among women, other than skin cancer. It is the second leading cause of cancer death in women, after lung cancer (1). The lifetime risk for developing invasive breast cancer in the Western world is about 12%, and 1 in 33 women will die of breast cancer (2). Every year, about a million new cases are diagnosed in the world, and 4000 new cases in Israel (3). Most breast cancer cases are sporadic, with no discernable family history of cancer. In ~10%- 15% of the cases there is a familial clustering of cancer (4). In a subset of these familial cases, inherited breast cancer is diagnosed, hallmarked by multiple breast cancer cases in more than one generation, an autosomal dominant pattern of inheritance, and at times also manifesting phenotypic features such as breast cancer diagnosed at a young age (under 40 years of age) (5,6,7), bilateral breast cancer or co-occurrence of breast and ovarian cancer in the same patient or in her family (5,6). Like other cancer types, breast cancer is a multifactorial disease that results from a combination of many factors - genetic and environmental- converging to result in the tumorous phenotype. There are known (and putative) risk factors whose existence is associated with an altered breast cancer risk, including: Gender, Age, Menstrual periods, nulliparity, hormone replacement therapy HRT,Body Mass Index, Family history and additional genetic risk factors. Risk factors for developing breast cancer Gender Breast cancer is more common among women than among men. Its frequency is less than one male to 100 women. In Israel 50 cases or less of male breast cancer are diagnosed every year (3). Age Breast cancer is more common among women at or over the age of 50 years. According to the Israeli ministry of health statistics, about 74% of the patients are diagnosed at or over this age. 1

18 About 23% are diagnosed between years and only about 3% are diagnosed when they are younger than 40 years old (8). Family history of cancer Women with family history which includes more than one first-degree relative with breast cancer are times more likely to develop breast cancer (risk is correlated with the number of relatives diagnosed with breast cancer, their degree of relatedness, age at diagnosis, and the diagnosis of bilateral breast cancer) (9). Personal history of cancer A woman who had unilateral breast cancer is 2-3 times more likely than the general population to develop cancer in the contralateral breast or in another location of the previously affected breast (10). Fertility associated breast cancer risk factors One of the major risk factors for breast cancer are factors associated with gynecological and hormonal events in a woman's life (11,12,13,14,15,16,17). The common theme to these risk factors is extended exposure to endogenous estrogens during a woman's life, such as noted in women with a prolonged fertility period. For example, women whose menstrual cycles start at an early age (11-14 years old) are at higher risk (10%-30%) to develop breast cancer during their lifetime compared with women who menstruated later in life (e.g., age of 16). Similarly, the same effect on breast cancer risk is observed for the timing of cessation of menstrual periods: women who experience menstruation termination at the age of 55 years or older have a 50% increased risk to develop breast cancer compared with women whose menstrual cycles stopped between years. Furthermore, women whose menstruation ended at an earlier age (at or under 45 years) have a 30% lower risk factor to develop breast cancer than the average age group (14). 2

19 In line with these associations of estrogen exposure and breast cancer risk, oophorectomy performed before menstruation termination (mainly before the age of 40 years) reduces breast cancer risk by 60% presumably due to low exposure to estrogen (18). Moreover, childbirth at an early age (at or before the age of 20 years) significantly reduces breast cancer risk, compared with not giving birth at all or giving first birth after the age of 30 years, factors that elevate breast cancer risk by ~2 fold (17,18,19). Lastly, breast cancer risk is reduced by 7% for each baby a woman gives birth to (20). Breast-feeding is also associated with reduced breast cancer risk: for every 12 months a woman breast feeds (not necessarily consecutively) she decreases the risk by a 4.3% (21).The presumed mechanism is reduced estrogen exposure as a result of the increased levels of prolactin (20). Exogenous hormones Oral contraceptives Pike at al. and McPherson et al. reported an increased risk of breast cancer that is correlated with the duration of oral contraceptive use before 1st full-term pregnancy (22, 23). A study published in 2002 disputes this "well known" association between oral contraceptives and breast cancer risk: a total of 4575 women with breast cancer and 4682 controls, years of age, were interviewed. The relative risk did not increase consistently with longer periods of use or with higher doses of estrogen. The results were similar among white and black women. Use of oral contraceptives by women with a family history of breast cancer was also not associated with an increased risk of breast cancer, nor was the initiation of oral contraceptive use at a young age (24). HRT Hormone replacement therapy Hormone replacement therapy, primarily combined Estrogen and Progesterone preparations, if taken for 5 years or longer, was reportedly associated with an increase risk (Relative risk RR 1.35; 95% confidence interval [CI], ) to develop breast cancer (25). Following that publication the rate of women who use HRT in the USA has dramatically dropped. Several studies investigated 3

20 whether the reduction of HRT use in the USA affected breast cancer incidence and Ravdin et al. showed a 1-year 6.7% decrease in the rates of breast cancer and attributed that decrease to the lower rates of use of HRT (26,27,28). Obesity Obesity (as measured by BMI) reportedly increases breast cancer risk mainly in postmenopausal women. The risk increases by 1.2 times between the lowest BMI quarintile and highest (29,30,31) Alcohol consumption An established association was reported between increased alcohol consumption and breast cancer risk: the relative risk for breast cancer was 1.32 for women who drank gr (about 4 drinks/day) compared with those who consumed no alcohol, with the risk increasing by 7% for each additional 10gr of daily consumption of alcohol (32). Physical activity Physical activity may be protective against breast cancer, with a reported 20% risk reduction for developing breast cancer among average risk women who were physically active 5 times a week (33,34,35) Pre malignant process Benign or pre malignant lesions in the breast tissue (especially, Atypical ductal hyperplasia) increase the risk of developing breast cancer: risk increases by ~1.5 fold depending on the histopathological features of the tumor (36). Ethnic origin The highest morbidity rates related to breast cancer are in Western countries, North America and Northern Europe (37). Based on cancer data register at the Ministry of Health the morbidity rate in Israel among Jewish women as compared to Arab women is almost double (86/100,000 among Jews Vs 44/100,000 among Arabs) (the data are published in 2006, reflect the reality in 2004, and presented as age standtardized rate- ASR). During the past few years there is a decrease in 4

21 morbidity of breast cancer among Jewish women, and an increase among Arab women. In addition, among Jews, the women that has the highest morbidity of breast cancer are natives of Israel (107/100,000) followed by Jews of Ashkenazi origin (76/100,000) and then Non Ashkenazim of Asian and North African origin (62/100,000) (38). Genetic factors Most breast cancer cases are sporadic. In about 10% of breast cancer cases there is a family history of breast or other cancer types (39). In up to 40% of familial breast cancer cases, germline mutations are found in either the BRCA1 (MIM#113705) or BRCA2 (MIM#600185) genes. The estimated cumulative lifetime risk of breast cancer for women who carry mutations in either of these genes ranges from as high as 80% (40,41) to less than 45% (42,43). In a meta-analysis of 22 studies, among first-degree relatives of 500 index patients with BRCA mutations, the average lifetime cumulative risks for breast cancer were 65% (95% CI: 44-78%) and 45% (31-56%) in BRCA1 and BRCA2 mutation carriers, respectively. There was evidence for variation in risk by mutation position for both genes (44). Much higher estimates were reported in the published New York Breast Cancer Study (NYBCS), where the lifetime risk of breast cancer was estimated in relatives of Ashkenazi Jewish probands with breast cancer who were confirmed mutation carriers. This study demonstrated a lifetime risk of developing breast cancer of 82% among BRCA mutation carriers from low and high incidence families (45). The specific cancer risks vary from publication to publication and many factors affect that variation (recruitment strategy, specific populations, mutation type and location within the genes, methods of ascertainment, to name a few). The numbers and the percentage ranges that are quoted reflect that variety. Germline mutations in other genes also increase breast cancer risks but generally (with the exception of P53 and PTEN) (46) to a more moderate level over that of the general, average risk population. Large case-control association studies have identified variants in the DNA repair genes CHEK2, ATM, NBS1, RAD50, BRIP1 and PALB2 that confer an approximately twofold risk of breast cancer, but these variants are 5

22 rare in the population (47,48,49,50,51,52,53,46). In a study that evaluated whether and to what extent CHEK2 germline mutations contribute to a breast cancer predisposition in Ashkenazi Jewish high risk families, 172 high risk Ashkenazi Jewish women which are not carriers of the predominant Jewish mutations in BRCA1/BRCA2 were genotyped. No truncating mutations were noted and four previously described missense mutations were detected (R3W 1.2%, 1157T 1.2%, R180C 0.6% and S428F 5%), one silent polymorphism (E84E 20.5%) and one novel missense mutation (Y424H 1.2%). Segregation analysis of the 1157T and S428F mutations (shown to deleteriously affect protein function) with the cancer phenotype showed concordance for the CHK2*1157T mutation, as did two of three families with the CHK2*S428F mutation. This study concluded that CHEK2 missense mutations may contribute to breast cancer susceptibility in Ashkenazi Jews (54). Another study assessed the rate of the CHEK2*1100delC variant in Israeli male breast cancer patients. Fifty-four male breast cancer cases diagnosed in Israel were available for study, one of whom had bilateral disease. Twelve patients had at least one first- or seconddegree relative affected with breast cancer (22%), and 11 additional patients reported family history of other cancers. Forty-one (75.9%) patients were of Ashkenazi or Eastern European descent, and none carried any of the 3 predominant Jewish mutations in BRCA1 or BRCA2. No male breast cancer patient carried the CHEK2*1100delC variant. This study concluded that this variant accounts for a relatively small proportion (no more than 3%) of all cases in Israel (55). To evaluate the p53 gene putative involvement in inherited predisposition to breast/ovarian cancer in Jewish high-risk women, mutational analysis of the p53 gene (exons 4-9) was carried out. No inactivating p53 germline mutations were detected in 132 Jewish breast cancer patients non-brca1/2 mutation carriers. Allele distribution of the R72P missense mutation between ethnically diverse Jewish breast cancer cases and average risk controls showed significant differences: among non-ashkenazi breast cancer cases, 62.5%, 33.3% and 4.2% were homozygous, heterozygous and homozygous for the Arg72, Arg72Pro and the Pro72 polymorphism, respectively, whereas for controls, the distribution 6

23 was 22.4%, 65.4% and 12.2%, respectively (P= ), and among Ashkenazi breast cancer cases, allele distribution was 68.5%, 29.6% and 1.9%, whereas for controls, the distribution was 50%, 40% and 10%, respectively (P=0.0125). This study concluded that arginine homozygosity at codon 72 of the p53 gene is associated with a significant increased breast cancer risk in Jewish high-risk population (56). Germline mutations in PTEN underlie two inherited syndromes: Cowden disease (CD) and Bannayan-Riley-Ruvalcaba syndrome (BRRS). The known association of CD with breast cancer risk made it plausible that germline mutations within PTEN may play a role in inherited predisposition to breast cancer. The nine coding exons of the PTEN gene were screened for harboring germline mutations using denaturing gradient gel electrophoresis (DGGE) complemented by sequencing, in 89 women with an apparent inherited predisposition to breast cancer, some with salient features of CD. Among 89 high-risk women, two missense mutations were detected in exon 4: A to C change at cdna position 1279 resulting in a change of aspargine to threonine at codon 82 (N82T), and a G to an A alteration in 1269 which alters threonine to alanine at codon 78 (T78A), a non-conservative missense mutation. This study suggests that PTEN does not play a major role in predisposing to hereditary breast cancer in Israeli women (57). In a study yet unpublished, carried out in the lab, 149 high risk Ashkenazi Jewish women non carriers of the predominant Jewish mutations in BRCA1/BRCA2 were genotyped for mutations in the BRIP1 gene. No inactivating mutations were found in all coding exons. These studies that focused on the Ashkenazi high risk population that were all conducted in the Oncogetics unit lab are consistent with other studies published on ethnically diverse populations that conclude that these additional breast cancer susceptibility genes contribute minimally and marginally to the overall burden of inherited predisposition to breast cancer. 7

24 Ovarian cancer Ovarian cancer is the fifth most commonly diagnosed cancer among women in the world, and causes more deaths per year than any other cancer of the female reproductive system. Ovarian cancer is the most frequent cause of death attributed to gynecological malignancies in Europe and the USA. Between 70 and 75% of ovarian carcinomas are not diagnosed until a late stage III or later (58). The median age at diagnosis for ovarian cancer is 63 years and 1.39% of women born today will be diagnosed with cancer of the ovary at some point in time during their lifetime (59). Caucasian women have higher incidence and mortality rates than other racial and ethnic groups (60). Race/Ethnicity Female All Races 13.1 per 100,000 women White 13.8 per 100,000 women Black 10.1 per 100,000 women Asian/Pacific Islander 9.9 per 100,000 women American Indian/Alaska Native 10.8 per 100,000 women Hispanic 11.3 per 100,000 women Table 1: Incidence rate of ovarian cancer (59). There are 320 new cases of ovarian cancer diagnosed in Israel annually (61). The precise etiology of ovarian cancer is poorly understood, but there are known, established risk factors whose presence alter and affect ovarian cancer risk including: 1. age (ovarian cancer risk increases with more recent year of birth, and for each later decade of birth, risk increased by ~1.2-fold) (62) 2. Having a family history of the disease in a first-degree relative increases the risk of developing ovarian cancer (OR, 2.4; 95% CI, ) (63). 3. Infertility, postmenopausal hormone-replacement therapy and lifestyle factors such as cigarette smoking and alcohol consumption also mildly affect ovarian cancer risk (64). 4. Use of oral contraceptives and increasing parity are protective and 8

25 decrease ovarian cancer cancer risk (65). Other moderately protective factors include salpingooophorectomy, lactation, incomplete pregnancies, hysterectomy and tubal ligation (64). As mentioned earlier, most breast and ovarian cancers cases occur sporadically, namely there are no additional family members who were diagnosed with breast/ovarian cancer. In approximately 10% of breast cancer cases familial clustering of breast and/or ovarian cancer is noted (39).Women with a family history of breast and ovarian cancers, or ovarian cancer in more than one first degree relative, or ovarian cancer in a relative that was diagnosed under the age of 45 years, are at high risk of developing ovarian cancer. The most likely explanation for these familial-inherited cases is inherited mutations in autosomal-dominant transmitted cancer susceptibility genes. Indeed, between 5% to 10% of breast cancers are linked to germline mutations in one of several genes. The most common genetic alterations are those found in either the BRCA1 or the BRCA2 genes which are highly penetrant autosomal dominant susceptibility genes (36,66,40). BRCA1 and BRCA2 BRCA1 (MIM#113705) and BRCA2 (MIM#600185) are human genes that belong to a class of genes known as tumor suppressors, which regulate the cell cycle and prevent uncontrolled proliferation. The proteins made by both genes are essential for repairing damaged DNA (67). BRCA1 By performing a genetic linkage analysis in 23 early-onset breast cancer families, the BRCA1 gene was first localized to chromosome 17q21, the gene was subsequently cloned and sequenced in 1994 (68). The gene spans ~81,000 base pairs (from base pair 38,449,843 to base pair 38,530,933, gene bank accesion number NG_ ) (67) encodes a 7.8-kb transcript composed of 24 coding exons (68,69). The BRCA1 protein, is a 220-kd nuclear protein of 1863 amino acids, that contains phosphorylated residues (70) and several recognizable protein motifs. These highly conserved functional motifs of BRCA1 include the N-terminal ring finger domain involved in interaction with BARD1, two nuclear localization signals within exon 11, an "SQ" cluster, and two C-terminal 9

26 BRCT (=BRCA carboxy terminal) domains, which are found in many DNA damage repair proteins and are involved in numerous protein-protein interactions with binding partners such as BRIP1, Abraxas (ABRA1) and CtIP (71,72). BRC A RING BRCT domain repeats BARD1 binding NLS Rad51 binding Transactivation Figure 1 : Protein structure of BRCA1: highly conserved functional motifs of BRCA1 include the N- terminal ring finger domain (includes zinc finger) involved in interaction with BARD1 (BRCA1 associated RING domain 1), two C-terminal BRCT (=BRCA carboxy terminal) domains, which are found in many DNA damage repair proteins and are involved in numerous protein-protein interactions with binding partners such as BRIP1, Abraxas (ABRA1) and CtIP, in the center of the protein there is a domain that create a complex with Rad51 protein and two NLS (nuclear localization signals) which direct the protein to the nucleus. Thus, BRCA1 is a large protein with multiple functional domains and interacts with numerous proteins that are involved in many important biological processes/pathways. Mounting evidence indicates that BRCA1 is involved in all phases of the cell cycle and regulates orderly events during cell cycle progression. BRCA1 deficiency consequently causes abnormalities in the S-phase checkpoint, the G 2 /M checkpoint, and the spindle checkpoint and centrosome duplication. The genetic instability caused by BRCA1 deficiency, however, also triggers cellular responses to DNA damage that blocks cell proliferation and induces apoptosis. Thus BRCA1 mutant cells cannot develop further into full-grown tumors unless this cellular defense is impaired (69). BRCA1 functions as a checkpoint mediator that collectively facilitates interactions between ATM/ATR and their substrates by mediating the assembly of damage-specific multiprotein complexes at and around repair sites (73). The transcription of BRCA1 is induced late in the G 1 phase of the cell cycle and remains elevated during the S phase (74). 10

27 BRCA1 functions in the maintenance of genomic integrity and in homologous recombination mediated DNA repair. Other functions of this protein include chromatin remodelling, DNA decatenation and ubiquitylation (75,76,77,78), and transcription regulation of some genes, such as the estrogen receptor (75,78). BRCA2 The BRCA2 gene was first localized to chromosome 13q12 q13 by analyzing 15 high-risk breast cancer families that did not show linkage to the BRCA1 locus on chromosome 17q21 (79,80), and was subsequently cloned in 1996 (81). The BRCA2 gene spans ~84,000 basepairs (from base pair 31,787,616 to base pair 31,871,804, gene bank accesion number AC_ ), encodes a 10.4-kb transcript, composed of 27 exons, which codes for a 380-kd nuclear protein of 3,418 amino acids with several phosphyrylated residues (81).There are residue motifs clusterd to the region encoded by exon 11 that mediate binding of BRCA2 protein to RAD51. BRCA2 controls the RAD51 recombinase in reactions that lead to homologous DNA recombination (HDR). BRCA2 binds RAD51 via eight conserved BRC repeat motifs of approximately 35 amino acids, each with a varying capacity to bind RAD51. BRC repeats both promote and inhibit RAD51 assembly on different DNA substrates to regulate HDR (82). BRC repeats BRCA Transactivation Rad51 binding Rad51 binding Figure 2: Protein structure of BRCA2: Eight BRC repeats which mediate binding of BRCA2 protein to RAD51 in exon 11. BRCA2 protein is normally located in the nucleus and contains phosphorylated residues (83). BRCA2 transcription is induced late in the G 1 phase of the cell cycle and remains elevated during 11

28 the S phase (74). BRCA2 functions primarily in homologous recombination (HR) and HR based DNA double-strand break repair (DSBR) (84) and has additional functions in cell growth and DNA damage checkpoint control (85). BRCA2 also has transcriptional and co-transcriptional activities (86,87). In addition, BRCA2 is a Fanconi anemia (FA) protein FA-D1 (88). Figure 3: The roles of BRCA1 and BRCA2 in DNA repair-(a) This model suggests that a macromolecular complex consisting of BRCA1, BRCA2, BARD1 and Rad51 functions to repair damaged DNA. Complex formation is preceded by phosphorylation of BRCA1 by the kinase ATM. In response to DNA damage, the complex relocates to chromosomal regions undergoing DNA replication marked by proliferating cell nuclear antigen (PCNA). (b) Loss of BRCA1 and/or BRCA2 12

29 function (indicated by dotted outlines) leads to inability to repair damaged DNA. When damage occurs to critical checkpoint genes, such as p53 (dotted outline), checkpoints such as p21 cannot be activated and cells proliferate (from 89). Both the BRCA1 and BRCA2 genes are classical tumor suppressor genes. Consistent with this notion allelic loss was demonstrated in BRCA1 BRCA2 associated tumors. Whenever tested, the wild type allele was somatically deleted in cancer from individuals with a disease-causing mutation, resulting in a high susceptibility to malignant transformation (69). Loss of function mutations and hence loss of protein expression of either BRCA1 or BRCA2 results in defective DNA repair, defects in transcription, abnormal centrosome duplication, defective G2/M cell cycle checkpoint regulation, impaired spindle checkpoint, and sustained chromosomal damage (90,91,92). Figure 4: Relationship between mutational events and formation of breast or ovarian cancers. In response to estrogen present in all women from puberty through pregnancy until menopause, breast epithelial cells proliferate. During such periods of rapid cellular proliferation, it is likely that the repair capacity of cells is compromised. Among women with a germline mutation in BRCA1 or BRCA2, somatic inactivation of the remaining allele will result in repair-deficient cells. These cells will be unable to repair damaged DNA in the following cell cycle and will ultimately die due to activation of critical cell cycle checkpoints. The rare repair-deficient cell that escapes death by checkpoint may acquire mutations at other sites including critical checkpoint genes, ultimately resulting in tumor formation. In inherited BRCA tumorigenesis, somatic inactivation of the second 13

30 BRCA allele likely occurs early in a young woman s life, leading to early onset of breast cancer. In women without a BRCA germline mutation (=sporadic tumorigenesis), somatic alteration of one allele of BRCA1 or BRCA2 in breast epithelial cells is likely to occur during estrogen-mediated proliferation, due to the density of repetitive elements in these genes. A second somatic inactivation of the second allele may occur by one of four mechanisms: somatic genomic alteration or transcriptional silencing, or loss of BRCA function attributable to haplo-insufficiency (resulting in DNA damage) or failure of post-translational modification. Inactivation of a second BRCA allele in a cell would generally result in cell death. The rare BRCA-deficient cell that escapes death may acquire inactivating mutations in critical checkpoint genes, allowing it to proliferate. In BRCAmediated sporadic tumorigenesis, these events must occur in the same cell and, therefore, tumors would occur much later in a woman s life (from 93). Mutations in both the BRCA1 and BRCA2 genes substantially increase the carrier's risk of developing specific cancer types. Compared with a lifetime risk of about 13.2 % for women in the general population to develop breast cancer, the cumulative lifetime risk of breast cancer in women who carry BRCA1 or BRCA2 gene mutations is approximately 60% to 80%, and these cancers often occur at a younger age. The lifetime risk for ovarian cancer in women with BRCA1 mutations is estimated to be 40% to 50% and is slightly lower (10% to 20%), in women who carry BRCA2 gene mutations (44,45,94,95). These rates were reported from different world populations and in women ascertained using different methodologies, and hence the wide range of reported mutation-associated risks. In a study that focused on Ashkenazi Jewish women with inherited mutations in BRCA1 and BRCA2, risks appear to be increasing with time: breast cancer risk by age 50 among mutation carriers born before 1940 was 24%, but among those born after 1940 it was 67%. Lifetime risks of ovarian cancer were 54% for BRCA1 and 23% for BRCA2 mutation carriers (45). Risks for developing breast or ovarian cancer to women with BRCA1 or BRCA2 mutations in an Israeli cohort were significantly associated with the cancer site in the index case. By age 75, combined cancer risk for BRCA1 mutation carriers was 87% among relatives of breast cancer index cases and 87% among relatives of ovarian cancer index cases, and for BRCA2 carriers combined risks were 78% and 88%, respectively (62).Women with BRCA1/2 gene mutations are also be at higher risk for developing additional malignancies: a relative risk of two- to four fold for colon carcinoma in BRCA1 carriers 14

31 was reported in a number of studies (94,96,97), but it is still debatable (98,99,100). Malignant melanoma was originally reported as at excess relative risk (RR = 2.58, 95% confidence interval [CI] = ; P=.01) in the BRCA2 Linkage Consortium study (97), but not confirmed in a smaller Dutch study (101), thus, the risk may be higher in specific BRCA2 families. A pancreatic cancer risk for female carriers was also reported to be elevated. In the BRCA2 Linkage Consortium study, estimated cumulative risk for pancreatic cancer by 70 years of age was 1.5% (95% CI = ) (97). A smaller BRCA2 carrier study in the Dutch population noted a four-fold lifetime risk for pancreatic cancer (101). For female BRCA1 carriers, the cumulative risk for pancreatic cancer by age 70 was significant (1.26%, 95% CI = ) (96). The presence of a mutation in the BRCA2 gene in men leads to a risk of BC of 5% 10% and an increased risk of developing prostate cancer, while there is a much lower risk in male carriers of BRCA1 mutations (100,102). Furthermore, the relative risk to male BRCA2 mutation carriers is high before age 65 years, largely attributable to breast, prostate, and pancreatic cancers. Lastly, BRCA2 mutation carriers were also reported to be at risk of stomach cancer and melanoma (of the skin and eye) (100). Since their cloning, approximately 3800 mutations have been described in BRCA1 and BRCA2. The majority of the clearly pathogenic mutations are inactivating mutations that lead to frameshift and a premature stop codon or are nonsense mutations leading to a truncated, inactivated, non-functional protein (103). The incidence of mutations in high-risk families varies widely among different populations; some present a wide spectrum of different mutations, while in particular ethnic groups, specific mutations show a high frequency due to a founder effect. Founders are small groups of people who have remained isolated with consequent interbreeding and the result that a normally rare mutation continues to be present and becomes more common within the population (104). Populations where founder mutations have been identified in the BRCA1 and BRCA2 genes are: Iraqi/Iranian Jews (105), Icelanders (106,107), Norwegians (108,109,110), Finns (111), Swedes (112,113,114) French (115.) Dutch (116,117) Italians (Calabria) (118,119) Sardinians (120,121) 15

32 French-Canadians (Quebec) (122,123,124,125) Hispanics (Southern California) (126), Hispanics (Columbia)(127,128), South Africans (129), Chinese (130) Japanese (131,132), Malaysians (133), Filipinos (134), Pakistanis (135) and Ashkenazi Jews. In Ashkenazi Jews, three predominant mutations have been described in high risk breast/ovarian cancer families: 185delAG and 5382insC in BRCA1 and 6174delT in BRCA2. The combined population frequency of these mutations in Ashkenazi Jews approaches 2.5% (106,136,137,138,104). 185delAG Tyr978X 5382insC 8765delAG 6174delT Figure 5: BRCA1 and BRCA2 predominant Jewish Ashkenazi mutations (green stars) and non Ashkenazi mutations (red stars). A study which attempted to determine the spectrum of BRCA1 and BRCA2 mutations in high-risk Jewish individuals, non-carriers of any of the predominant Jewish mutations, screened 47 high-risk Jewish individuals, 26 Ashkenazim, and 21 non-ashkenazim. Overall, 13 sequence alterations in BRCA1 and eight in BRCA2 were detected: nine neutral polymorphisms and 12 missense mutations, including five novel ones. The novel missense mutations did not co-segregate with disease in BRCA1 and were detected at rates ranging from 6.25% to 52.5% in the general population for BRCA2. The study findings suggest that except for the predominant mutations in BRCA1 and 16

33 BRCA2 in Jewish individuals, there are only a handful of pathogenic mutations within these genes in Jewish high risk families. It may imply novel genes may underlie inherited susceptibility to breast/ovarian cancer in Jewish individuals (139). Another study which screened 208 Ashkenazi Jewish women with ovarian cancer, found 87 mutations in the BRCA1 and BRCA2 genes. Of the 87 identified mutations, 86 were founder mutations, and the single nonfounder mutation was inherited from a non-jewish ancestor. This suggests that most of the familial clustering of ovarian cancer in the Jewish population is attributable to the three founding mutations (140). In a study that screened 160 Ashkenazi Jewish breast-ovarian cancer families for the entire BRCA1 gene, one novel frameshift mutation in BRCA1 in exon 14 was detected. In addition, three missense variants of unknown significance were found in single families. This study also supports the notion that nonfounder BRCA1 mutations are rare in Ashkenazi Jewish breast/ovarian cancer families (141). Another study aimed to characterize the clinical predictors, spectrum, and frequency of BRCA1 and BRCA2 mutations in an ethnically diverse, high-risk clinic derived population, sequenced 29 Ashkenazi Jewish families. The three founder mutations accounted for 85% of the 20 mutations detected. One individual of Ashkenazi Jewish ancestry had a C61G mutation, a BRCA1 mutation that has been reported as a founder mutation in Poland (142). A study aimed at defining the prevalence and spectrum of point mutations and genomic rearrangements in BRCA genes in a large U.S. high-risk clinic population of both non-ashkenazi and Ashkenazi Jewish descent, found that out of 136 non-ashkenazi Jewish probands, 36 (26%) had BRCA point mutations and 8 (6%) genomic rearrangements (7 in BRCA1 and 1 in BRCA2). Forty-seven of the 115 (40%) Ashkenazi Jewish probands had point mutations (46 founder mutations and only one deleterious non founder point mutation in the BRCA2 gene) and no genomic rearrangements (143). Despite their undisputed role in breast/ovarian cancer predisposition, only 20-40% of the familial aggregation of breast cancers can be accounted for by mutations in BRCA1, BRCA2 or other 17

34 identified breast cancer susceptibility genes (40) like: p53 (144), CHK2 (145,146,147), PTEN (148) and ATM (149). Hence it seems clear that additional yet unidentified genes are associated with breast cancer predisposition. Several research groups focused on different populations, in the hope to find new region within the genome that is associated with breast cancer predisposition. A study focusing on 28 Japanese high risk families, suggested linkage to chromosome 3p22-p25. They performed genome-wide linkage analysis in 58 patients and nine unaffected members among these 28 high risk families who carried no mutation in BRCA1 or BRCA2, employing a set of 410 microsatellite markers. To explore whether the candidate gene in this 3p22 p25 region contributed to carcinogenesis of familial ovarian cancer in a similar fashion to the tumor suppressor gene, they performed loss of heterozygosity (LOH) analysis. It was observed that the frequency of LOH at four markers in this region was >50% only in tumor tissues from patients with no mutation in BRCA1 or BRCA2, not in those with a BRCA1 mutation (150). These observations suggested a novel tumor suppressor gene involved in breast/ovarian cancer predisposition that localizes to 3p22-p25. Interestingly, in 2009 a large consortium reported that a single SNP that localizes to chromosome 3p24 (rs ) was associated with an increased breast cancer risk (151). Targeted linkage analysis was carried out in a set of 77 breast cancer families (57 Finnish, 12 Icelandic, and 8 Swedish) with no detected BRCA1 and BRCA2 mutations. A maximum parametric two-point and under heterogeneity logarithm of odds (LOD) scores were 2.76 and 3.46, respectively for a marker at 13q21 (D13S1308, θ = 0.10). The results were further evaluated by simulation to assess the probability of obtaining significant evidence in favor of linkage by chance as well as to take into account the possible influence of the BRCA2 locus, located at a recombination fraction of 0.25 from the new locus (152). However despite the considerable time that elapsed since the publication of this study, no further studies have been published that replicated the linkage to that 18

35 specific chromosomal region and no mutations have been reported in genes that localize to that region. Another study performed linkage analysis in two German breast cancer families, showing negative LOD scores with 17q and 13q markers, using seven adjacent microsatellite markers from 8p12-p22. Incorporating LOH data from tumors of the affected family members, a maximum cumulative three-point LOD score of 3.30 was obtained with the D8S137 and D8S131 markers (153). Yet, similar to the fate of the previously published linkage studies, no validation was ever reported for this initial study. In 2007 a consortium of researchers conducted a three-stage genome-wide association study. In the first stage, they used a panel of 227,876 SNPs, to genotype 390 cases and 364 controls. The cases were selected to have a strong family history of breast cancer. For the second stage 12,711 SNPs, approximately 5% of those typed in stage 1, were selected on the basis of the significance of the difference in genotype frequency between cases and controls. These SNPs were then genotyped in a further 3,990 invasive breast cancer cases and 3,916 controls. Those two stages were followed by a third stage in which 30 of the top hit SNPs were tested for confirmation in 21,860 cases and 22,578 controls from 22 studies. SNPs in five novel independent loci exhibited strong and consistent evidence of association with breast cancer (P < 10(-7)). Four of these contain plausible causative genes (FGFR2, TNRC9, MAP3K1 and LSP1). Furthermore, 1,792 SNPs that were significant at the P < 0.05 level was also highlighted, indicating that many additional common susceptibility alleles may be identifiable by this approach (154). Another genome-wide association study (GWAS) of breast cancer conducted in 2007 by genotyping 528,173 SNPs in 1,145 postmenopausal women of European ancestry with invasive breast cancer and 1,142 controls. In that study four SNPs in intron 2 of FGFR2 that were highly associated with breast cancer were identified. This association was further confirmed by genotyping 1,776 affected individuals and 2,072 controls from three additional studies. Across the four studies, the association 19

36 with all four SNPs was highly statistically significant (P trend for the most strongly associated SNP (rs ) = ; population attributable risk = 16%) (155). Interestingly while both studies identified SNPs from within intron 2 of the FGFR2 gene as associated with breast cancer susceptibility, the specific SNPs that showed the strongest association were distinct: rs (in the Easton et al study) localizes to a 25Kb region that contains most of exon 2 of the FGFR2 gene, whereas the SNP in the Hunter et al study, was rs Figure 6: The position of the SNP's rs and rs on the FGFR2 gene. Combined these studies suggest that while undoubtedly there are multiple genomic regions that are associated with breast cancer risk and/or susceptibility, there is no single region that accounts for a large proportion of familial cases unaccounted for by BRCA1/BRCA2 mutations. Major gene rearrangements in BRCA1 and BRCA2 The majority of pathogenic mutations in BRCA1 and BRCA2 are point mutations or small deletions and insertions (BIC database). In most populations tested, the observed frequencies of BRCA1 variations in high-risk breast and/or ovarian cancer families are lower than predicted by linkage analysis. This finding suggests that methods generally used for mutation scanning fail to detect certain types of germline defects, such as large genomic rearrangements. Most of mutation screening methods are PCR-based, and hence there is an inherent inability to detect large genomic rearrangements such as partial or complete exon loss or amplification (156). Multiplex ligationdependent probe amplification (MLPA) is widely used as a highly sensitive method for detecting the relative copy number of all BRCA1 BRCA2 exons in a high-throughput format (157). 20

37 Major gene rearrangements in BRCA1 and BRCA2 are shown in tables 2 and 3. Reference Gad et al. Oncogene (158) Major gene rearrangements 1. dup ex dup ex18-20 Montagna et al. Hum Mol Genet (159) Belogianni et al. BMC Cancer (160) Woodward et al. J Med Genet (157) Agata et al. J Med Genet (161) Preisler-Adams et al. Cancer Genet Cytogenet (162) 1. del ex del ex del ex del ex20 1. del ex20 1. del ex3 2. del ex5 3. del ex21-23 de la Hoya et al. Clin Chem (163) 1. del ex del ex14 3. dup ex dup ex20 5. dup ex21-22 Peixoto et al. Fam Cancer (164) Armaou et al. Eur J Cancer (165) Buffone et al. Breast Cancer Res Treat (166) Lim et al. Clin Genet (167) no major gene rearrangements in BRCA1 1. del ex5 2. del ex del ex17 1. del ex more than 60 different genomic rearrangements 1. del ex dup ex13 2. del ex13-15 Zikan et al. Mutat Res (168) 1. complex genomic rearrangement exons 21, dup ex13 3. dup ex15 Palanca Suela et al. Breast Cancer Res Treat (169) 1. del ex del ex del ex del ex20 Ratajska et al. Oncol Rep (170) 1. del ex1a-1b-2 2. del ex17-19 Petrij-Bosch et al. Nat Genet (171) 1. del ex22 (510bp) 2. del ex13 (3835bp) 3. del 14kb Puget et al. Cancer Res (172) 1. del ex17 (1kb) Puget et al. Cancer Res (173) 1. del ex15 2. del ex del ex1-2 Rohlfs et al. Hum Genet (174) 1. Complex rearrangement ex

38 Reference Payne et al. Genes Chromosomes Cancer (175) Unger et al. Am J Hum Genet (176) Gad et al. Oncogene (158) Hogervorst et al. Cancer Res (177) Hofmann et al. Hum Mutat (178) Montagna et al. Hum Mol Genet (179) Bunyan et al. BJC (180) Hartmann et al. Hum Mutat (116) Barrios et al. Clin Genet (181) Belongianni et al. BMC cancer (182) Tancredi et al. Eur J Hum Genet (183) Table 2: Major gene rearrangements in BRCA1. Major gene rearrangements (del8.2kb, ins420bp, del3.7kb) 2. del ex21-22(3.4kb) 1. del ex 3 (complex rearrangement including inversion, duplicaton and deletion) 1. dup ex13 2. del ex ex17-19 (complex rearrangement) 4. ex1a+b(complex rearrangement) 1. del ex dup ex dup ex del ex8 2. del ex dup ex13 4. dup ex triplication ex dup ex13 1. del ex1a+b ex2 2. del ex del ex l ex20 1. del ex20 2. del ex del ex del ex del ex dup ex del ex 1A+B+2 2. del ex dup ex dup ex del ex del ex promoter region 5. del ex del ex del 5' end of BRCA1 1. del ex20 1. del ex14-19 Reference Nordling et al. Cancer Res (184) Wang et al. Genes Chromosomes Cancer (185) Tournier et al. Cancer Res (186) 1. Peelen et al. BJC (187) 2. Lahti-Domenici et al. Cancer Genet Major gene rearrangements 1. del ex3 1. del ex del ex12-13 No major gene rearrangements in BRCA2 22

39 Reference Cytogenet (188) 3. Gad et al. Oncogene (158) Agata et al. J Med Genet (161) Ramus et al. Hum Mutat (189) Lim et al. Clin Genet (167) Gutiérrez-Enríquez et al. Breast Cancer Res Treat (190) Agata et al. J Med Genet (161) Major gene rearrangements 1. del ex del ex8-11a 3. del ex20 1. del ex dup ex del ex2 2. del ex del ex dup ex20 1. del ex8-11a 2. del ex del ex20 Woodward et al. J Med Genet (157) 1. del ex del ex Tournier et al. Cancer Res (186) Table 3: Major gene rearrangements in BRCA2. 1. dup ex del ex complete deletion of BRCA2 Major gene rearrangements have been identified in BRCA1 and BRCA2 in several populations. Specifically, large genomic deletions in BRCA1 were noted in the French, Italian, and Dutch populations (158,159,171). More than 60 different BRCA1 germline rearrangements with mapped breakpoints have been reported to date (165). These are scattered throughout the whole gene, and most of them are deletions, but duplication, triplication or combined deletion/insertion events (191,192) have also been described. Less is known about the occurrence rate of large genomic rearrangements in BRCA2 in hereditary breast/ovarian cancer families. To date, few studies have been published revealing less than 20 (157,161,167,189,190) different genomic rearrangements in BRCA2 in hereditary breast cancer patients from diverse European populations and the value of BRCA2 screening for such alterations is still undetermined. 23

40 Several molecular biological tools, techniques and terms are used in this thesis and for the sake of being comprehensive are being briefly presented below: Single Nucleotoide Polymorphism - SNPs The most stable and abundant variation in the genome occurs in the form of single nucleotide polymorphisms (SNPs) and are a resource for mapping complex genetic traits (193). SNPs represent about 90% of the common variation in the genome (insertion/deletion polymorphisms and variable repeat elements providing the rest). Common SNPs, by definition, have a minor allele frequency greater than 1%. This variation arises through a single mutation event in the history of the population. The likelihood of recurrent mutation at the same site is low. Consequently, SNPs are stable. They are di-allelic, the 2 alleles represent the "ancestral" or "wildtype" and the variant form (194). In the "Common disease- common variant" model, SNPs that are rare or exclusive to a subpopulation will have no value because they cannot account for the distribution or prevalence of disease (194). Because of their abundance and distribution in the human genome SNP are widely used in genome wide association studies, as was done in the present study. Microsatellites Microsatellites are simple sequence repeats, the most common classes being dinucleotide, trinucleotide, and tetranucleotide, and they occur at a rate of 1 in every 10kb in a wide range of eukaryotic genomes. Human microsatellites are used for linkage studies and they average at least ten alleles with very high heterozygosity per locus of over 80%. Although they have a much higher mutation rate than the standard sequence, they are not densely distributed (195). In this study dinucleotide markers were employed in the haplotype analysis. Haplotype Human genetic diversity appears to be limited not only at the level of individual polymorphisms, but also in the specific combinations of alleles (haplotypes) observed at closely linked sites. These 24

41 common variants are responsible for most heterozygosity in the population. If limited haplotype diversity is general, it should be practical to define common haplotypes using a dense set of polymorphic markers, and to evaluate each haplotype for association with disease (195). The selection of markers to represent not only genes, but also the common haplotypes of genes provides a more powerful approach to detect disease association (196,197). Linkage disequilibrium Linkage disequilibrium (LD) is the increased sharing of alleles above that expected by random sharing (194). The magnitude and positional pattern of distribution of LD in the human population influences the total number of test markers required to identify disease loci in LD association studies. If LD is well preserved, then a single marker may be a reliable reporter for extensive adjacent polymorphism. However, if it is low, little or no reduction in marker density may be possible. If it is a positionally heterogeneous mixture of regions of high and low LD, then successful exploitation of LD will require a modified approach encompassing greater knowledge of the regional pattern and extent of LD distribution across the genome (194). Estimation of the extent of LD in the human population has been obtained by sampling LD in various limited genomic regions (198). The average distance in which useful LD is preserved (e.g., the distance over which 50% of pair-wise markers demonstrate association) is 50 to 60 kilobases (kb). With a uniformly distributed marker set, LD of this magnitude means that about SNPs would be required for a whole-genome association study (199). However, in some chromosomal loci, LD extends as far as 500 kb, whereas in other loci no useful disequilibrium exists. Because linkage disequilibrium is highly variable across the genome the use of average distances to direct marker selection means that disease susceptibility variation residing in loci containing low LD would usually not be identified (194). 25

42 Disruption of allelic association by crossing over during recombination is not the only mechanism that can erode LD. Gene conversion is an additional meiotic process of sequence alteration in which one allele directs the conversion of a partner allele to its own form. The Holliday 4-way DNA recombination junction is the physical element by which both crossing over and gene conversion occur. Gene conversion may be explained by repair of heteroduplex DNA formed in short tracts (up to 1 kb) adjacent to the Holliday junction. Gene conversion represents a further problem for association mapping which seeks to exploit LD. Gene conversion disrupts LD over short distances (200). Thus, chromosomal regions which, on the basis of allelic association over long range, appear to retain high levels of LD, may in fact harbor short regions where LD has been disrupted by gene conversion. A study by Ardlie and colleagues reports that LD between tightly linked markers in humans is much less than expected (201). A plausible explanation for this observation is that LD has been reduced by gene conversion. This phenomenon represents an important impediment to LD mapping studies because LD markers selected to survey extensive regions of LD that also harbor regions affected by gene conversion may fail to detect allelic association (194). Association studies Linkage analysis uses collections of related individuals with members who manifest a trait (e.g., hypertension, breast cancer) to examine the co-inheritance with the trait of widely distributed markers in order to infer the genomic position of alleles contributing to the trait. In contrast, association studies tests for an association between genetic markers with a trait or a disease, using a different methodology rationale. The scientific rationale of that approach relies on the linkage disequilibrium between genetic markers and closely related disease genes/loci. Thus, by comparing the rate of occurrence of a polymorphic genetic marker in the affected group of individuals (=cases) to the rate of occurrence of that marker in the group of matched controls enables deduction of the existence of a disease gene in linkage disequilibrium with the marker at the specified chromosomal location. In addition, association studies can also generate data by genotyping unrelated individuals. 26

43 In order to directly identify the polymorphisms contributing to disease susceptibility, very large collections of markers will be required to ensure that markers interrogating the disease-causing variants are represented. Genotyping many individuals for SNPs is a daunting and expensive task. However, such genome-wide association studies may be made more feasible if the number of markers to be typed can be reduced by exploiting the preservation of nonrandom association between a polymorphism contributing to a trait and adjacent markers (linkage disequilibrium) (194). Association between chromosomal region and disease phenotype recognize regions within the genome which probably contain genes related to the specific phenotype (=susceptibility genes), so there is no need to have hypothesis about these genes. As the relevant chromosomal regions have been found, one should do positional cloning or screening for intragenic mutations within the genes in these regions. For the purpose of gene discovery, SNPs are considered to be the most predominant segregating form of variation at the molecular level because of their frequent occurrence throughout the genome, and they can be useful in association studies. Although there are many limitations, association studies are perhaps best suited for mapping of polygenic complex disease loci. However, this type of study requires a large number of patients and an adequate control group to achieve over 80% power to detect a locus. Once a significant site is identified, one can use either a pedigree-based transmission disequilibrium test (TDT), which measures the transmission of alleles from a heterozygous parent to the affected offspring (an unequal transmission of SNP alleles to affected and unaffected siblings), or a case-control population sampling, which measures the association between SNPs and the disease in a large population (202,203). TDT detects both linkage and association, and SNPs are usually used. These methods are based on the assumption that SNP variants account for population susceptibility to certain disorders. These methods have several limitations, such as difficulties associated with 27

44 population structure, different levels of linkage disequilibrium in loci, allelic and nonallelic heterogeneity of phenotypes, and epistatic interaction of alleles, all of which have been previously discussed by others (204,205,206) (195). The disadvantage of association studies is the possibility of false positive results which are the result of two main reasons: 1. The existence of allelic heterogeneity in the population, a phenomenon which can be characterized and corrected by stratification, focusing on ethnically homogeneous populations (like Ashkenazi Jews). 2. The analysis of a large number of markers enables statistical mistakes so there is a need for correction of multiple factors to reduce false discovery rates (FDR) such as the Boneffroni correction test. The current study focused on Ashkenazi Jews. The main reasons for focusing on the Ashkenazi population are: breast cancer is more common among Ashkenazi women than other ethnic origins in Israeli women, the spectrum of mutations in BRCA1 and BRCA2 is well-defined and limited, and Ashkenazi Jews are a classic example of a relative genetic isolate with multiple examples of founder mutations (104,207). It is a well accepted and documented fact that Ashkenazim do represent a relative genetic isolate: the limited spectrum of germline mutations in autosomal recessive disorders (208), the shared haplotype along long stretches of DNA (209), combined with historical documentation of social and cultural isolation from their surrounding, non-jewish neighbors are all indicative of the relative genetic isolation of Ashkenazi Jews. 28

45 Research goal The major goal of this thesis was to define new regions within the human genome that are likely to harbor novel breast cancer susceptibility genes. This study is based on a number of basic assumptions: 1. Ashkenazi Jews represent a relative genetic isolate. 2. The spectrum of mutations in BRCA1 and BRCA2 is well defined in the Jewish Ashkenazi population (there is only a limited number of founder mutations). 3. Additional, yet unidentified genes are associated with breast cancer predisposition. Specific aims 1. Haplotype comparison of the BRCA1 and BRCA2 genes in Jewish Ashkenazi high risk breast / ovarian cancer families. 2. Analysis of large genomic rearrangements in BRCA1 and BRCA2 in Jewish Ashkenazi high risk breast / ovarian cancer families. 3. Association analysis of the entire human genome for the Jewish Ashkenazi high risk breast / ovarian cancer families. 4. Genotyping candidate genes in the regions within the human genome which show association with breast / ovarian cancers for germline mutations in the Jewish Ashkenazi high risk breast / ovarian cancer families. 29

46 Materials and Methods Patients and Families Patients were ascertained from one of two sources: the high risk clinic at the Oncogenetics unit, "Sheba" medical center, and an ongoing project that genetically tests and counsels consecutive Jewish Ashkenazi women diagnosed with breast cancer conducted jointly by the Sheba and "Shaare Tzedek" medical centers. Over the course of activity of the Oncogenetics unit at the Sheba Medical center (since the late 1990's) ~8000 individuals (belonging to ~5000 families) were assessed to evaluate their cancer risk. Of these individuals and families, more than 1000 were found to harbor one of the founder mutations in Jewish women in BRCA1 BRCA2, and a few high risk women were genotyped for all possible mutations in both genes and few were found to harbor "private" BRCA1/2 mutations. All individuals with founder (or non founder) mutations were excluded from the current study. All study participants were naturally aware that they do not harbor BRCA1 BRCA2 mutations. Of all the individuals counseled over the years at the oncogenetics unit only 105 were selected for participation in this study. The participants in this study underwent two genotyping sessions for the Jewish founder mutations: originally when they were screened clinically and then, as part of the inclusion criteria for this study. All participants had breast or ovarian cancer, and in addition, one or more of the eligibility criteria: 1) under 40 years of age at time of diagnosis of breast cancer or 2) over 40 years of age at time of diagnosis of breast cancer and at least one of the following: a) bilateral breast cancer or ovarian cancer; b) at least one first degree relative with breast cancer; c) one or more first or second degree relatives with bilateral breast cancer or ovarian cancer; d) at least two relatives of the paternal side with breast cancer (this criterion was specifically added to add high risk individuals for whom the first degree relative is usually breast cancer free) and e) at least one relative diagnosed for breast 30

47 cancer under the age of 40 years. Inclusion criteria were not met by 8 patients who did not fit these above mentioned criteria but had a personal and family history highly suspicious for an inherited predisposition and were also analyzed herein (210,211,212). DNA extraction Genomic DNA was isolated from peripheral blood leucocytes by the PUREGene kit (Gentra Inc Minenapolis MN) using the manufacturer's recommended protocol. 10 ml of blood were collected from each patient, in tubes containing EDTA. Cells are lysed with anionic detergents (cell lysis solution and RBC lysis solution) in the presence of a DNA stabilizer. The DNA stabilizer limits the activity of intracellular DNases and also DNases found elsewhere in the environment. RNA is then removed by treatment with an RNA digesting enzyme (RNase A solution). Other contaminants, such as proteins, are removed by salt precipitation. Finally, the genomic DNA is recovered by precipitation with alcohol (70% ethanol) and dissolved in a buffered solution containing a DNA stabilizer. Polymerase chain reaction (PCR) The polymerase chain reaction is a technique for amplifying a specific region in the human genome or a particular piece of DNA in the test tube bypassing the need to clone that piece of DNA. Application of this procedure, enables making virtually unlimited copies of a single DNA molecule, even though it is initially present in a mixture containing many different DNA molecules. 31

48 Figure 7: PCR procedure. PCR was preformed in a 25-50µl reaction, containing ng genomic DNA. For known mutations in the BRCA1 BRCA2 genes (=the predominant mutations in the Jewish Ashkenazi population) PCR primers were designed to flank the mutated area and a restriction enzyme was used to differentiate the wild type from the mutant allele. The PCR reaction contains PCR buffer (Fisher Biotec, Australia), 2.5 mm MgCl2, 200nM dntps, 10pmol of each primer and 0.2 U FB1 DNA Polymerase (Fisher Biotec). Amplification was carried out as follows: an initial denaturation step of 5 minutes at 94 C, followed by 35 cycles of 94 C for 30 seconds, annealing step which was different for each mutation, 72 C for 30 seconds, and a final extension step at 72 C for 10 minutes. The modifications between the mutations were primarily in the annealing 32

49 temperature, annealing times, the magnesium concentrations and the additions of non-ionic detergents such as DMSO Triton and Formamide. PCR products were analyzed in 3% Agarose gel, with pbr322 DNA/AluI marker running in a separate lane to facilitate allele size determination. PCR: 94ºC 5 min 94ºC 20 sec 42-68ºC 1-3 min go to step 2, 35 times 72ºC 20 sec 4ºC 3 hours Table 4: The PCR program for the predominant mutations in the Jewish Ashkenazi population in BRCA1 BRCA2 genes. Mutation 185delAG BRCA1 5382insC BRCA1 6174delT BRCA2 Annealing step 56 ºC-1 min 55 ºC -1 min 54 ºC-1 min Table 5: The annealing step for the predominant mutations in the Jewish Ashkenazi population in BRCA1 BRCA2 genes. The mutant allele creates a novel restriction enzyme site in the PCR product that is not generated when only the wild type, non-mutated allele is present. Thus, specific restriction enzyme digests distinguish the wild type from the mutant allele, by the absence or presence of the restriction site. Gene Exon Mutation Volume of restriction enzyme l) Temperature Incubation time (hours) Wild type allele Mutant allele BRCA1 Exon insC ºC bp 251 bp BRCA1 Exon 2 185delAG ºC bp 136 bp 33

50 BRCA2 Exon delT 3 37ºC bp 143 bp Table 6: Conditions of the restriction reaction for the predominant mutations in the Jewish Ashkenazi population in BRCA1 BRCA2 genes. Gene Exon Mutation Restriction enzyme Digested allele BRCA1 Exon insC MvaI mutant BRCA1 Exon 2 185delAG HinfI Wild type BRCA2 Exon delT Eco72I Wild type Table 7: The restriction enzymes for the common mutation in BRCA1/2 in Ashkenazi Jews. Long range PCR The Long Range PCR is aimed for efficient amplification of long targets of DNA of up to 40 kb. The following program was performed in order to amplify exon 11 of BRCA2 - a ~8500 bp fragment. The PCR reaction contains LA PCR TM buffer ll (TAKARA BIO Inc, Japan), dntps Mixture (2.5 mm each), 10pmol of each primer and TaKaRa LA Taq TM 5 units/ µl (TAKARA BIO Inc, Japan). Amplification was carried out as follows: an initial denaturation step of 5 minutes at 95 C, followed by 30 cycles of 98 C for 30 seconds, annealing step 68 C for 15 minutes, and a final extension step at 72 C for 10 minutes. PCR products were analyzed in 1.5% Agarose gel, with MassRuler DNA ladder high range ready to use SM0393 marker running in a separate lane to facilitate allele size determination. Selecting and assigning pathogenicity of missense mutations in BRCA1 We searched the BIC database ( for missense mutations in BRCA1 gene reported in the Ashkenazi population or individuals of East European origin. Missense mutations within conserved domains (R866C; R841W), those with a proven pathogenic effect (C61G), and those that scored moderately high on the align- GVGD algorithm (213) (Y179C) or are extremely rare (R331S) were selected. 34

51 Genotyping for seemingly pathogenic BRCA1 missense mutations PCR primers (Table 8) were designed to flank the genomic regions containing the missense mutations, and subsequently a restriction enzyme was used to differentiate the wild type from the mutant allele. PCR was preformed in a 25µl reaction, containing ng genomic DNA, 2.5 µl 10X PCR buffer (Fisher Biotec, Australia), 2.5 mm MgCl2, 200nM dntps, 10pmol of each primer and 0.2 U FB1 DNA Polymerase (Fisher Biotec). Amplification was carried out as follows: an initial denaturation step of 5 minutes at 94 C, followed by 35 cycles of 94 C for 30 seconds, annealing step, with a specific distinct temperature for each fragment- (Table 9), 72 C for 30 seconds, and a final extension step at 72 C for 10 minutes. Mutation Forward primer Reverse primer R866C AGAATACATTCAAGGTTTCACA TAGACAAAACCTAGAGCCTCCT R331S CCAGAAAAGTATCAGGGTAGTTC CTTTTTTTCTGTGCTGGGAGACC R841W ATTGGGACATGAAGTTAACCAGAG CTGTCTGTACAGGCTTGATATTAG Y179C TACAACCTCAAAAGACGTCTGTC TGTGAGCCACCATGCCTGGCC C61G TGGCTCTTAAGGGCAGTTGTGAGA CTGTGGTTGCTTCCAACCTAGCAT M1008I TCAGAGGCAACGAAACTGGACTCA TGGAGCCCACTTCATTAGTACTGG Table 8: Mutation designation, primer sequences for the BRCA1 missense mutation. Mutation R866C R331S R841W Y179C C61G M1008I Annealing step 53ºC-1 min 57ºC-1 min 56ºC-1 min 57ºC-1 min 56ºC-1 min 59ºC-1 min Table 9: Annealing step temperature and duration in the PCR program for the BRCA1 missense mutation. 35

52 Mutation R866C R331S R841W Y179C C61G M1008I Restriction enzyme TaaI- 16 hours Cfr10I-16 hours Sch1-16 hours Xmil-16 hours AvaII-16 hours Tsp45I-16 hours Table 10: Restriction enzymes and incubation time for the BRCA1 missense mutation. About 80% of the PCR products were subsequently digested with a specific restriction enzyme (Table 10). The incubation conditions were 2.5 ml enzyme buffer and 1-5U restriction enzyme (Fermentas, Lithuania) (according to the mutation) per sample. Restriction enzyme Volume of restriction enzyme l TaaI 0.1 Cfr10I 0.1 Sch1 0.1 Xmil 0.5 AvaII 0.1 Tsp45I 0.1 Table 11: Volume of restriction enzymes for the BRCA1 missense mutation. Differentiation between the wild type and the mutant alleles was done by running the digested PCR products (50% of the digested PCR product) on a 3% agarose gels at 120 volts for 30 minutes and visualizing with UV light after ethidium bromide staining. 36

53 Haplotype analysis In order to further exclude BRCA1 BRCA2 genes' contribution to the analyzed high risk families, 3 intragenic polymorphic dinucleotide markers in each of the two genes were selected. BRCA1: D17S855, D17S1322 and D17S1323, and for BRCA2: D13S171, D13S1698 and D13S1699. In every set of primers, one is fluorescently labeled by one of the colors: FAM-blue, TET- green or HEX-yellow. Fluorescently labeled dyes are attached to ACGT extension products in DNA sequencing reactions. Dye labels are incorporated using either 5 -dye labeled primers or 3 -dye labeled dideoxynucleotide terminators. Polymerases such as AmpliTaq FS are used for primer extension. Marker BRCA1 D17S855 BRCA1 D17S1322 BRCA1 D17S1323 BRCA2 D13S171 BRCA2 D13S1698 BRCA2 D13S1699 Fluorescent label FAM TET TET TET TET TET Table 12: Markers fluorescent label. The sequencing reaction sample tubes are placed in a tray in the instrument s autosampler. The autosampler successively brings each sample into contact with the cathode electrode and one end of a glass capillary filled with polymer (POP-4). An anode electrode at the other end of the capillary is immersed in buffer. A portion of the sample enters the capillary as current flows from the cathode to the anode. This is called electrokinetic injection. The end of the capillary near the cathode is then 37

54 placed in buffer. Current is applied again to continue electrophoresis. When the nucleotides reach a detector window in the capillary coating, a laser excites the fluorescent dye labels. Emitted fluorescence from the dyes is collected by a CCD camera. GeneScan Analysis Software analyzes raw data to quantify the DNA fragments and determine the size of the fragments by comparing them to fragments contained in a size standard (TAMRA 500). Figure 8: The ABI PRISM 310 Genetic Analyzer is an automated single-capillary genetic analyzer. Before the samples were analyzed, they were prepared by the conditions listed in table 13. Marker Water μl Enzyme μl Primers μl Mgcl2 μl Buffer(x10) μl DNTP's μl D17S D17S D17S D13S D13S D13S Table 13: The amounts of each reagent needed for optimal PCR reaction. Marker Primer Sequence D17S855 D17S855-F 5` GGA TGG CCT TTT AGA AAG TGG 3` D17S855-R 5` ACA CAG ACT TGT CCT ATG CC 3` 38

55 D17S1322 D17S1322-F 5` CTA GCC TGG GCA ACA AAC GA 3` D17S1322-R 5` GCA GGA AGC AGG AAT GGA AC 3` D17S1323 D17S1323-F 5` TAG GAG ATG GAT TAT TGG TG 3` D17S1323-R 5` AAG CAA CTT TGC AAT GAG TG 3` D13S171 D13S171-F 5` CCTACCATTGACACTCTCAG 3` D13S171-R 5` TAGGGCCATCCATTCT 3` D13S1698 D13S1698-F 5` AACCTCAGGCTAATAGTCTCA 3` D13S1698-R 5` GTCCATACCACTAAGTCTGAC 3` D13S1699 D13S1699-F 5` AGACAGAGAATCTCAACTGG 3` D13S1699-R 5` TTTGATTTTCACAGCAGATG 3` Table 14: Primers for the intragenic markers in BRCA1 and BRCA2. Amplification was carried out as follows: an initial denaturation step of 5 minutes at 94 C, followed by 35 cycles of 94 C for 30 seconds, annealing step which was different for each fragment, 72 C for 30 seconds, and a final extension step at 72 C for 10 minutes. Marker D17S855 D17S1322 D17S1323 D13S171 D13S1698 D13S1699 Annealing step 57ºC-1 min 58ºC-1 min 54ºC-1 min 58ºC-1 min 51ºC-3 min 60ºC- SD Table 15: The annealing step of the PCR reaction for each of the markers. Then each of the PCR products was diluted according to the marker. 39

56 Marker Dilution D17S855 1:5 D17S1322 1:20 D17S1323 1:30 D13S171 1:5 D13S1698 1:30 D13S1699 1:30 Table 16: PCR products dilutions. Then 1 l of the diluted PCR products + 1 l standart size TAMRA GENESCAN 500 TM + Formamide 15 l. The samples were denaturated at 95 C for 5 min and than were stored at -20 C for 5 min. The PCR products were analyzed by using GeneScan software in the ABI 310 Prism (PE Biosystems, Foster City, CA). Multiplex Ligation-dependent Probe Amplification (MLPA) Deletions and amplifications of a gene (or a portion of it), are usually not detected by PCR based technologies of amplified gene fragments, as a non-mutated copy is still present (156). Several alternatives have been suggested and employed for analysis of major gene rearrangements: Southern blotting (171,214,215), Comparative genomic hybridization, (216) long range PCR (217), and pulsed-field gel electrophoresis (PFGE) (218). Analysis by Multiplex Ligation-dependent Probe Amplification (MLPA) is a suitable alternative that is capable of detecting major gene rearrangements involving BRCA1 BRCA2 in the form of deletions and amplifications (157). Basically, MLPA is a method which makes a nucleic acid sample suitable for a multiplex PCR reaction, in which up to 45 specific nucleic acid sequences are amplified simultaneously using a single PCR primer-pair. The resulting amplification products are separated by sequence-type 40

57 electrophoresis. MLPA tests are designed so that the length of each amplification product is unique. The length increases in a stepwise-fashion by 6 or 9 nucleotides, with the total size range lying between nucleotides. This size range provides an optimal fragment separation and a low background on sequence type gels. MLPA probes are also capable discriminating between sequences that differ in only one nucleotide (219,220,221,222,223). In MLPA it is not the DNA sample which is amplified, but only the probes that anneal to the sample DNA that are amplified. MLPA probes consists of two separate oligonucleotides, each of which containing one of the PCR primer sequences. It is only when these two probe parts both hybridize to their target sequence, that they can be ligated to each other. Only ligated probes will be amplified exponentially in a PCR reaction. The number of probe ligation products depends on the number of target sequences in the sample. Although the MLPA amplification reaction is very reproducible, there will always be some sequences whose amplification rate per cycle will be 1-2% lower than others, thus resulting in a lower final peak area. Hence, a single MLPA amplification profile is not sufficient to determine whether there are any copy number changes; a patient s peak profile should always be compared to that of a reference sample. Compared with this reference sample, the relative peak area of each amplification product then reflects the relative copy number of the probe s target sequence in the analyzed patient sample. A deletion of one or more exons in a patient thus becomes apparent as a decrease in relative peak area of the amplification products of the probes corresponding with these exons. 41

58 Figure 9: The principles of MLPA: each probe has a different stuffer sequence. The two parts of each probe hybridise, than they are ligated by thermostable ligase. All probe ligation products are amplified by PCR using only one primer pair. The amplification products differ by their length as a result of the differences in the length of the stuffer (from 219). MLPA protocol: 1. Denaturation of ng DNA by heating to 98oC in a thermocycler. 2. Addition of the MLPA probes and overnight incubation at 60 o C for hybridization. 42

59 3. Adding the ligase and ligase buffer to ligate at 54 o C for 15 minutes. 4. Inactivation of the ligase by heating at 98 o C. Adding PCR primers, dntps and polymerase and starting the PCR. 5. Analysis of the PCR products by (capillary) electrophoresis. MLPA analysis was performed by the Salsa P002 BRCA1 and Salsa P045 BRCA2 MLPA probe mix assay as described by the manufacturer (MRC-Holland Amsterdam, the Netherlands). The Salsa P002 for BRCA1 gene contains 24 sets of probes for the 22 coding exons of the gene. Exon 1 and 11 have two sets of probes and exon 4 which does not encodes, does not have a probe. In addition, there are probes for 10 additional fragments on other chromosomes: 2q14, 2p14, 3p21, 4q26, 5q31, 6p21, 7q, 11p13, 12p12 and 12p13. These fragments are used as an internal control for the genotyped sample. The Salsa P045 for BRCA2 contains 40 sets of probes for 23 exons. Exons 1 and 11 have two sets of probes, exons 5, 6, 23 and 26 do not have probes. The kit contains 3 sets of probes for fragments in CHEK2 gene: exon 2, 12 and another probe for the mutation 1100delC. The internal controls are directed to chromosomes 2, 4 and 8 which have two sets of probes and chromosomes 5, 9, 11 and 17 which have only one set of probes. In addition, there are two pairs of control probes for chromosome 13 which are directed to the 3' and 5' of the gene. Peak heights from each patient were then exported to an excel spreadsheet, which was designed by Dr. Andrew Wallace from the National Genetics Reference Laboratories, Manchester, to assess the ratios of each test peak relative to all other peaks for that individual. Each test and control sample's data is normalized (by summing the total control peak height and dividing each ligation product's peak height by this Figure), the dosage quotients are calculated and the three hypotheses (normal, duplicated, deleted) are tested comparing to 5 normal controls. For normal sequence a dosage quotient of 1.0 ( ) is expected; if a deletion or duplication is present, the dosage quotient should be and , respectively. A dosage quotient of and is equivocal. In addition, the sample quality is assessed by measuring the standard deviation of the 43

60 internal controls against the investigated exon. A sample in which the Standard deviation was less than 0.1 shows no overlap between normal, duplicated and deleted ranges. The following program covers the complete MLPA reaction: Dilution of 250ng DNA with Tris-EDTA to a final volume 5μl. 98 C 5 minutes 25 C hold 95 C 1 minute 60 C hold Table 17: Hybridization reaction of MLPA. 54 C Hold 54 C 15 minutes 98 C 5 minutes 4 C Hold Table 18: Ligation reaction of MLPA. 60 C hold 95 C 30 seconds seconds seconds Go to step 2, 38 times 72 C 20 minutes 4 C hold Table 19: PCR reaction of MLPA. Electrophoresis- After diluting the DNA with 1μl Genescan-Tamra 500 size standard and 15μl formamide, electrophoresis was done by the ABI310 capillary sequencer using Genescan software. In order to 44

61 improve the reading of the results, we extended the injection time to 7 seconds and the run time to 30 seconds. We used the Genotyper software to read the results. In a case of duplication there will be a high reading, while in a case of deletion there will be a low reading. Figure 10: Wild type DNA, deletion of exon 13 and deletion of exon 22 respectively. Note these are only examples of what an abnormality would look like, and these Figures are taken (from 219). Quantitative PCR Dosage changes detected and heralded by MLPA analysis were re-evaluated (for ascertainment and validation) by quantitative PCR (qpcr). Quantitative, real-time PCR is based on detection of a fluorescent signal produced proportionally during the amplification of a PCR product. The chemistry is the key to the detection system. A probe (i.e., TaqMan) is designed to anneal to the target sequence between the traditional forward and reverse primers. The probe is labeled at the 5' end with a reporter fluorochrome and a quencher fluorochrome added at any T position or at the 3' 45

62 end. The probe is designed to have a higher T m than the primers, and during the extension phase, the probe must be 100% hybridized for success of the assay. As long as both fluorochromes are on the probe, the quencher molecule stops all fluorescence by the reporter. However, as Taq polymerase extends the primer, the intrinsic 5' to 3' nuclease activity of Taq degrades the probe, releasing the reporter fluorochrome. The amount of fluorescence released during the amplification cycle is proportional to the amount of product generated in each cycle. Figure 11: quantitative PCR procedure. (1) In intact probes, reporter fluorescence is quenched. (2) Probes and the plementary DNA strand are hybridized and reporter fluorescence is still quenched. (3) During PCR, the probe is degraded by the Taq polymerase and the fluorescent reporter released (from 224). The ABI7700 detection system was used for qpcr analysis. The sensitivity of detection allows acquisition of data when PCR amplification is still in the exponential phase. This is determined by identifying the cycle number at which the reporter dye emission intensities rises above background noise; this cycle number is called the threshold cycle (C t ) (Figure 12). The C t is determined at the most exponential phase of the reaction. The C t is inversely proportional to the copy number of the target template; the higher the template concentration, the lower the threshold cycle measured. 46

63 Figure 12: The amount of fluorescence obtained in each amplification cycle for each reaction. The threshold cycle (C t ) is shown by the darker horizontal line (from 225). A set of FAM-labeled primers and probes were designed for each exon or gene fragment that showed an abnormality on MLPA analysis. The primers were designed to flank the MLPA probe, and the design of the Taqman probe was similar to the MLPA probe. Exon Primer Probe Exon 5 F-TGCTGAAACTTCTCAACCAGAAGA CTTCACAGTGTCCTTTAT R-CCAACCTAGCATCATTACCAAATTATA Exon 17 F-ACACGTGTAGAACGTGCAGGATTG CAGATGCTCGTGTACAAGTTTGCCAGA R-CCAAAGTGCTGCGATTACAGGCAT Exon 19 F- GCTCTTTTGTGAATCGCTGA TCTCTTTAGGGGTGACCCAGTCTATT R- GGTGCATTGATGGAAGGAAG Exon 21 F- TTCCCCTGTCCCTCTCTCTT TCTTCAGGGGGCTAGAAATCTGTTGC R- CCATCGTGGGATCTTGCTTA Table 20: Primers and probes for BRCA1 in the qpcr reaction. 47

64 Exon Primer Probe Exon 11a F- TTTAGTGAATGTGATTGATGGTACTTT AGAAGCTGTTCACAGAATGATTCTGAA R- TTCCTCAGAATTGTCCCAAAA Exon 11b F- ACTACTTTGAAACAGAAGCAGTAG ACTGCCAAGTCATGCCACACATTCT R- AGTGATTGGCAACACGAAAGGT Exon 2 F- GGAATGCATCCCTGTGTAAGTGCAT AAGACACGCTGCAACAAAGCAGGT R- CTAAGCAACACTGTGACGTACTGG Exon 3 F- AGCTTCTGAAACTAGGCGGCAGA CAGCTGGCTTCAACTCCAATAATATTC R- GGCTCGAATTTGCTTGGCGAAA AA Exon 25 F- GTCAGACGAATGTTACAATTTACTGGCA TCCAGTGGCGACCAGAATCCAAATCA R- GTTGAATGTCTCTTGAAAGTGGCCC Table 21: Primers and probes for BRCA2 in the qpcr reaction. A mix containing VIC-labeled primers and probe for RNaseP (Roche diagnostics, Manheim, Germany), a single copy gene, was used as an internal standard. Amplification/deletion reaction were performed in a volume of 20μl containing 10μl of TaqMan universal PCR mastermix (Roche diagnostics), 1μl of each BRCA-specific primers, 1μl of each BRCA-specific labeled hybridization probe, 0.5μl of the RNaseP primer-probe mix and 1μl of genomic DNA. Thermocycling was as followed: 95 0 C for 15 min, followed by 40 cycles of 95 0 C for 15 sec, 60 0 C for 10 sec. Data analysis was carried out using ABI Prism 3100 Sequence Detection Software. Since the standard curve for sequential concentrations of the two different probes was similar, a difference between the Ct of the examined BRCA exon and that of the RNaseP indicates a deletion / duplication. Association studies As part of the first phase of a GWAS, 249 Ashkenazi Jewish (AJ) breast cancer patients participated of whom 45 were patients from the cases who formed the basis of the present study from the Sheba Medical Center. As controls to phase 1, 299 healthy AJ women were enrolled who were cancer free 48

65 and did not have a family history of breast cancer. The controls also included 29 healthy AJ women enrolled at Sheba Medical Center, Tel-Hashomer, Israel. For the controls, any woman who indicated a prior diagnosis of breast cancer, atypical hyperplasia, or lobular carcinoma in situ was not eligible for this study. Genotyping was carried out by using Affymetrix GeneChip Early Access Version 3 (EAv3) Human Mapping Arrays. Use of Affymetrix EAv3 chips for genotyping was performed as described in the Gtype 4.0 manual (226), except that 150 ng of all genomic DNA samples were evaluated for quality by gel electrophoresis. After qualification of the DNA samples, each sample was then divided into two aliquots. Sequence complexity was reduced by restriction enzyme digestion with either NspI or StyI, and a biotin-labeling primer amplification assay was performed on each DNA aliquot. Hybridization of the amplified probes was then performed on specific NspI or StyI arrays, as appropriate. Direct sequencing PCR primers were designed to flank all coding exons of ECHDC1 and RNF146 genes. PCR was preformed in a 25µl reaction, containing ng genomic DNA, PCR buffer (Fisher Biotec, Australia), 2.5 mm MgCl2, 200nM dntps, 10pmol of each primer and 0.2 U FB1 DNA Polymerase (Fisher Biotec). Amplification was carried out as follows: an initial denaturation step of 5 minutes at 94 C, followed by 35 cycles of 94 C for 30 seconds, annealing step (unique temperature for each exon) (table 24), 72 C for 30 seconds, and a final extension step at 72 C for 10 minutes. Primers sequences are listed in tables 22 and 23. Exon Forward primer Reverse primer Exon 1a CCCATTGTCTAGTATGTGGCTTGC AGTCCAGCTACTCCCTTCTTTGGT Exon 1b ATGAACATGGACGTCGCAGGAAGA ACCTTGAGCATGGAGCTGAAGACA Table 22: RNF146 gene exon and primer sequences. 49

66 Exon Forward primer Reverse primer Exon 1 GTCACTGGAGCTTTACTTGGAGT GATTAGAGCATACAGAGAGAGTTAC Exon 2 GGTTCAAGCTTCTGCTATCCTTTG GAGCAAGTTAATTATATTTGTGCAAG Exon 3 TGATTATCCTCTACGGCACAA GGAGCCCTAGGATATCAACAGTCA Exon 4 TGTTCCTCCTATAATCCTACCA GTATACTCCATAGTGCCAGTGCAG Exon 5 ACATTCTGTCACCACAGCAAGGA TTCCCAACTCAACAGAAGGGTCCA Table 23: ECHDC1 gene exon and primer sequences. Gene RNF146 exon 1a RNF146 exon 1b ECHDC1 exon 1 ECHDC1 exon 2 ECHDC1 exon 3 ECHDC1 exon 4 ECHDC1 exon 5 Annealing step 57ºC-1 min 58ºC-1 min 61ºC-1 min 58ºC-1 min 56ºC-1 min 57ºC- 1 min 60ºC- 1 min Table 24: RNF146 and ECHDC1 annealing step temperature and duration. PCR fragments were subject to sequence analysis using the big Dye terminator chemistry and kit (PE Biosystems, Foster city, CA), and using the ABI Prism 3100 semiautomatic DNA sequencer (PE Biosystems). PCR fragments were sequenced using the cycle sequencing technique using either the forward or reverse primers, as recommended in the ABI Prism 3100 protocol. The resulting sequences were manually read and compared with the published reference sequences (Echdc1- NCBI Reference Sequence: NC_ , Rnf146- NCBI Reference Sequence: NC_ ) 50

67 Results 1. Patients' Characteristics Overall, 105 Jewish Ashkenazi women participated in the study: 103 were diagnosed with breast cancer [mean age at diagnosis (+ SD) was years (range years)], one with ovarian cancer (61 years) and one patient had both ovarian and breast cancers (ovarian cancer at 43 years and breast cancer at 48 years). Twenty one (19.9%) of the participants were diagnosed with breast cancer under the age of 40 years, and 8 (7.6%) had bilateral breast cancer. All participants had between one to six family members who were diagnosed with breast cancer, and 16 (15.2%) had family members who were diagnosed with ovarian cancer. 2. Assessing the possible contribution of germline mutations in BRCA1 and BRCA2 genes to inherited breast/ovarian cancer in the study population 2.1 General All study participants were Ashkenazi Jews at high risk for developing breast cancer, an ethnic population where three predominant mutations in BRCA1 (185delAG 5382InsC) and BRCA2 (6174delT) can be detected in a substantial proportion of breast/ovarian cancer families (136). Family-specific pathogenic mutations in both genes can be detected in only a minority (up to 5%) of high risk Ashkenazim (227). Thus, initially all participants were screened for the presence of these predominant BRCA1 and BRCA2 mutations. 51

68 2.1.1 Identification of the 185delAG mutation in exon 2 of BRCA1 gene All participants were non-carriers of the 185delAG mutation in the BRCA1 gene M WT AluI marker Figure 13: Restriction enzyme analysis of the 185delAG mutation in BRCA1. Lanes 2, 8 and 10 represent positive controls, namely known carriers of the 185delAG mutation, where two bands can be seen, representing the mutated and the wild type alleles. PCR products of several participants from the study population are seen in the other lanes. These samples exhibit only one band, which represents two wild type alleles, namely these samples are non-carriers of that particular mutation Identification of the mutation 5382insC in exon 20 of BRCA1 All study participants were non-carriers of the 5382insC mutation in the BRCA1 gene Figure 14: Restriction enzyme analysis of the 5382insC mutation in BRCA1. Lanes 2,6,7 and 8 represent positive controls, namely known carriers of the 5382insC mutation where two bands can be seen representing the mutated and the wild type alleles. PCR products of several participants from the study population are seen in the other lanes. These samples exhibit only one band, which represents two wild type alleles, namely these samples are wild type, non carriers of that specific mutation. 52

69 2.1.3 Identification of the mutation 6174delT in exon 11 of BRCA2 All study participants were non-carriers of the 6174delT mutation in BRCA2 gene M WT Figure 15: restriction enzyme analysis of the 6174delT mutation in BRCA2. Lanes 2, 5 and 8 represent positive controls namely, known carriers of the 6174delT mutation, where two bands can be seen representing the mutated and the wild type alleles. PCR products of several participants of the study population are seen in the other lanes. These samples have only one band, which represents two wild type alleles, namely these samples are do not carry this particular mutation. Using these techniques, none of the study participants was a carrier of any of the three predominant Jewish Ashkenazi mutations: BRCA1 (185delAG, 5382InsC) or BRCA2 (6174DelT). To emphasize this was one of the entry criteria to the study population. 3. Haplotype analysis using intragenic markers in BRCA1 and BRCA2 genes Another, non-mutually exclusive way to eliminate the possibility that mutations in BRCA1 and BRCA2 genes underlie the observed familial clustering of breast/ovarian cancer among study participants, was to genotype participants and their affected relatives with intragenic BRCA1 and BRCA2 markers, and look for haplotype sharing. To that end, three BRCA1 intragenic markers (D17S1323, D17S1322 and D17S855) and three BRCA2 intragenic markers (D13S171, D13S1698 and D13S1699) were used to genotype 161 participants (105 Ashkenazi patients, 36 of their relatives and 20 non Ashkenazim) all non carriers of the predominant mutations in BRCA1 and BRCA2 genes participants. The subjects included in the haplotype analysis section encompass 36 53

70 breast cancer patients who were family members of one of the 105 Ashkenazi participants in the core study. These included 13 pedigrees overall: 5 pedigrees with 2 affected individuals genotyped, 6 pedigrees included 3 relatives each who were all genotyped and 2 pedigrees included 4 affected relatives who were genotyped with intragenic markers. As another type of controls, 20 non Ashkenazi breast cancer patients were also genotyped with the same markers. The haplotype analysis was used as an additional tool to exclude the contribution of BRCA1 BRCA2 to the disease burden in these specific families. While admittedly the haplotype analysis did not include all families, it should be pointed out that these data only supplement previously published data from our own lab (228) as well as other centers (217) showing that in high risk Ashkenazi families the contribution of non founder Jewish mutations in BRCA1 BRCA2 is marginal at best. In addition, we genotyped 23 Iraqi carriers and 11 Ashkenazi carriers of the 185delAG mutation in BRCA1 and 10 Ashkenazi carriers of the 6174delT mutation in BRCA2. Some of the pedrees are shown in appendix 7. 54

71 3.1 Haplotype analysis of 185delAG BRCA1 mutation carriers. We compared the haplotype of the 185delAG BRCA1 mutation carriers to see if they share a common allele for the intragenic markers. Figure 16: Haplotype of the intragenic marker D17S1323 in BRCA1 of five carriers of the mutation 185delAG in BRCA1 gene mutation. All carriers (Iraqi- rows 1,3 and 4, Ashkenazi- rows 2 and 5) display a common allele for this marker whose size is 157bp (arrow). Similar results were obtained using the two other intragenic markers of BRCA1 gene in all 185delAG mutation carriers: The common allele for the D17S855 intragenic marker is 139bp and that for the D17S1322 intragenic marker - 121bp (data not shown). 55

72 3.2 Haplotype analysis of BRCA1 of the study participants. An example of the genotyping results in of the three intragenic markers within BRCA1 of the non carriers study participants is shown below: Figure 17: Haplotype of the intragenic marker D17S1323 in BRCA1 of two relatives from the study population. 56

73 BC at 45 BC at 43 Bil. BC (LT) at 42 (RT) at 55 Figure 18: The pedigree of the samples that were used on the haplotype of the intragenic marker D17S1323 in BRCA1 in figure 17. These samples represent two breast cancer patients who are related to each other (cousins) and are part of the study group, and hence do not carry the mutation 185delAG or 5382InsC mutations in the BRCA1 gene. As is clearly visible, they do not share the same allele using an intragenic BRCA1 marker. The fact that both were diagnosed with breast cancer, and they are part of the same family and yet do not share an allele in the BRCA1 gene, effectively excludes this gene as underlying or even contributing to the familial aggregation of breast cancer cases in that family. Similarly all of the study participants where affected family members were available for analysis did not share an allele with any of the intragenic markers in BRCA1 (13 families with overall 36 affected family members). 57

74 3.3 Haplotype analysis of 6174delT BRCA2 mutation carriers. Ten individuals who are 6174delT BRCA2 mutation carriers were genotyped using 3 intragenic BRCA2 markers, to assess allele sharing. Figure 19: The genotype of the intragenic marker D13S171 in the BRCA2 gene of three carriers of the mutation 6174delT in BRCA2. All three mutation carriers share a common allele for this marker - which is 235bp (arrow). Similar results were obtained when genotyping of all 6174delT BRCA2 mutation carriers using the two other intragenic markers of BRCA2 gene: The common allele for the D13S1699 intragenic marker is 155bp and for the D13S1698 intragenic marker- 177bp (data not shown). 58

75 3.4 Haplotype analysis of BRCA2 of the study participants. An example of genotyping results using one of the three intragenic BRCA2 markers of the non carrier's participants: Figure 20: Haplotype of the intragenic marker D13S171 of three samples from the study participants. BC at 55 BC at 43 BC at 40 BC at 28 BC at 30 Figure 21: The pedigree of the samples that were used on the haplotype of the intragenic marker D13S171 in BRCA2 in figure

76 The first two samples (top rows) represent two breast cancer patients who are relatives (cousins once removed), from the same family and the third sample belongs to an unrelated breast cancer patient participant in this study. None of the three individuals genotyped harbored the same allele. The fact that the two individuals from the same family who are diagnosed with breast cancer, do not exhibit the same allele for this marker, suggests that the BRCA2 gene is not the reason for the familial aggregation of breast cancer cases in this family. None of the study participants genotyped shared an allele for the intragenic markers in BRCA2 with her tested affected family member. (13 families with overall 36 affected family members). These data are consistent with the notion that sequence alterations and ("private"") mutations in both BRCA1 and BRCA2 genes are not the reason for the observed familial clustering of breast cancer in our study population. 60

77 4. Genotyping for seemingly pathogenic missense mutations in BRCA1. All study participants were genotyped for 6 seemingly pathogenic missense mutations in BRCA1: R866C, R331S, R841W, Y179C, C61G, M1008I. Two women were found to be carriers of the M1008I mutation. The first patient displaying the sequence alteration is a breast cancer patient diagnosed at age 62 years, her sister was diagnosed with breast cancer at age 56 years, their mother was diagnosed with breast cancer at age 44 years, and their maternal aunt was diagnosed with breast cancer around the age of 50 years. The other patient was diagnosed with breast cancer at age 45 years and her mother was diagnosed with breast cancer and colorectal cancer at age 62 years. There are no other relatives on the maternal side. In addition to the M1008I mutation, sequencing of these two samples revealed 2 neutral polymorphisms: E1038G which is not clinically important, and S1040N which is of unknown significance. None of the other missense mutations was detected in any of the study participants. Figure 22: Enzyme restriction products of the missense mutation M1008I in BRCA1. Two of the study participants are clearly carriers of the M1008I mutation. 61

78 Figure 23: Sequencing of one of the samples to ensure that it is mutated for the M1008I mutation. ATG (methionine) is altered to ATA (isoleucine). Figure 24: Sequencing of the S1040N in the mutated sample of M1008I. AAC (serine) turns into AGC (asparagine). 62

79 Figure 25: Sequencing of the E1038G in the mutated sample of M1008I. GAA (glutamic acid) turns into GGA (glycine). Figure 26: Restriction enzyme analysis of PCR products to detect the missense mutation R866C in BRCA1. Only one band indicating homozygosity for the wild type allele is seen in all samples. M Figure 27: Restriction enzyme analysis of PCR products to detect the missense mutation R841W in BRCA1. Only one band indicating homozygosity for the wild type allele is seen in all samples. M= mutant (control) sample. 63

80 M Figure 28: Restriction enzyme analysis of PCR products to detect the missense mutation Y179C in BRCA1. Only one band indicating homozygosity for the wild type allele is seen in all samples. M= mutant (control) sample.. M Figure 29: Restriction enzyme analysis of PCR products to detect the missense mutation C61G in BRCA1. Only one band indicating homozygosity for the wild type allele is seen in all samples. M= mutant (control) sample. M Figure 30: Restriction enzyme analysis of PCR products to detect the missense mutation R331S in BRCA1. Only one band indicating homozygosity for the wild type allele is seen in all samples. M= mutant (control) sample. 64

81 5. Evaluating the existence of major gene rearrangements in BRCA1 and BRCA2 using Multiplex Ligation-dependent Probe Amplification (MLPA). The next step was to evaluate if and to what extent large genomic rearrangements in BRCA1 BRCA2 exist in the Jewish Ashkenazi high risk population. These types of genetic abnormalities are missed by most methods commonly used for mutation screening, as most of these methods are PCR-based, and hence there is an inherent inability to detect large genomic rearrangements such as partial or complete exon loss or amplification. Hence, in order to assess the existence of large genomic rearrangements - MLPA was employed. 5.1 BRCA1 gene rearrangements MLPA analysis - MLPA reaction for BRCA1 had statistically significant results (p<0.1) in only 48 samples of the total 105 genotyped. In 16 of these samples, no dosage changes were found. Among the remaining 32 samples with statistically significant data, several dosage changes were found, but none in sequential exons. In only two samples, an amplification of exons 17 and 19 had significant odds ratio (1:2232, 1:421 for the first sample and 1:85, 1:74 for the second sample).the first patient displaying the abnormality is a breast cancer patient diagnosed at age 52 years, her sister was diagnosed with breast cancer at age 57 years, and their paternal cousin was diagnosed with breast cancer at age 50 years. There are no other women in the family. The other patient was diagnosed with breast cancer at age 28 years and her sister was diagnosed with breast cancer at age 43 years. There are no cancer cases at their maternal family and on the paternal side there are no women relatives. 65

82 Figure 31: Dosage changes found in exons 17 and 19 in BRCA1. On the upper sample dosage changes were found in exons 17 and 19 in BRCA1. As is evident, the columns of these exons reach 1.45 and 1.6 compared with the control sample (lower sample) where the columns of these exons reach 1.1 and 1.2. Among the remaining 57 samples, 21 samples did reach partial statistical significance (p<0.1). In this set of genotyped samples, amplifications were detected in exons 5, 17, 19 and 21 in 3 patients. Peak: Scan 4143 Size Height 294 Area !4181CON.FSA 14 Blue !4647SZ.FSA 2 Blue Figure 32: Amplification found in exon 21 in BRCA1. The sample shown on the upper panel was used as a control for MLPA reaction for BRCA1 gene. On the lower panel sample amplification was found in exon 21 (marked with an arrow). 66

83 The other 36 samples did not reach statistical significance. In these samples amplifications and deletions were found in the sections used as internal control. Because of the dosage changes in the control segments, it was impossible to determine if there are changes in the BRCA1 exons !3305.FSA 3 Blue !4181CON.FSA 14 Blue Figure 33: Dosage changes found in the internal control segments. The sample shown on the lower panel was used as a control for MLPA reaction for BRCA1 gene. On the upper panel sample dosage changes were found in the internal control segments (marked with an arrow): on the right, a segment on the short arm of chromosome 6 and on the left, a segment on the short arm of chromosome 12. Ex 5 Ex 17 Ex 19 Ex Amp (2232:1) Amp (421:1) 3965 Amp (9:1) 3575 Amp (288:1) Amp (49:1) 5079 Amp (85:1) Amp (74:1) Amp (18:1) 5571sz Amp (3217:1) 4849sz Amp (13:1) Table 25: A summary of the large genomic rearrangements in BRCA1 detected by MLPA in the present study. In the left column sample number. The exons in which dosage changes were found, are at the head of the table. Amp= amplification, Del= deletion. The number in brackets is the odds ratio for the dosage change. 67

84 5.2 BRCA2 major gene rearrangements MLPA analysis- The MLPA reaction for BRCA2 had statistically significance results (p<0.1) in only 31 of the 105 samples. Out of the 31 samples, 11 samples had no dosage changes. Among the remaining 20 samples with significant data, dosage changes were found in exons 1b, 3 (amplification) and 11a, 25 (deletion), in 6 patients. These dosage changes had significant odds ratio. 5' Ex 1B Ex 2 Ex 3 Ex 11A Ex sz Del (222:1) 5821sz Del (1338:1) 5265sz Amp (2:1) Del (74:1) 3965 Amp (4777:1) Amp (102084:1) Amp (40:1) 3572 Amp Amp (23:1) (58361:1) 4332 Amp Amp (2:1) Amp (8:1) (12926:1) Table 26: A summary of the large genomic rearrangements in BRCA2 detected by MLPA in the present study. In the left column sample number. The exons in which dosage alterations were found, are at the head of the table. Amp= amplification, Del= deletion. The number in brackets is the odds ratio for the dosage change. 68

85 Peak: Scan 4169 Size Height 54 Area !3572(BR.FSA 17 Blue !5265SZB.FSA 11 Blue !5313CON.FSA 13 Blue !5329SZB.FSA 16 Blue !5821SZB.FSA 1 Blue Figure 34: Examples of large genomic rearrangements in BRCA2 detected by MLPA. The third sample (5313) is a control sample for the MLPA reaction. In the first sample (3572) there is amplification of exon 1B on the left and amplification of exon 3 on the right. In the second sample (5265sz) there is a deletion of exon 25. In the forth sample (5329sz) there is amplification of exon 2 with odds ratio of 1:14 which is not high enough. In the fifth sample (5821sz) there is a deletion of exon 11A. All abnormalities are indicated by an arrow. Among the remaining 74 samples, 24 samples did reach partial statistical significance (p<0.1). In this set of genotyped samples, amplifications were detected in same exons as in the previous set of samples with the statistical significance. The amplification of exon 2 which had low odds ratio in the first group of samples, had higher chance in this group. The MLPA kit for BRCA2 includes a probe for the CHEK2*1100delC mutation as an internal control. Three samples were positive for this mutation, one in a statistically significant score. 69

86 Peak: Scan 4502 Size Height 856 Area !4181CON.FSA 12 Blue !R4675SZ.FSA 11 Blue Figure 35: The 40's probe of the SALSA P045 kit binds only in the presence of the mutation 1100delC in CHEK2 gene. The sample at the top panel is a control. In the second sample there is an additional pick (arrow) which suggests the existence of the 1100delC mutation in CHEK2 gene. Sequencing confirmation of the 1100delT CHEK2 mutation- The 3 samples for which the MLPA reaction indicated the presence of the CHEK2*1100delC mutation, were sequenced to ensure the existence of that mutation. Only one of the 3 samples was positive for the mutation and the other two samples, including the one showing the most statistically significant finding on MLPA, did not harbor that mutation. Figure 36: Sequencing of the 1100delT CHEK2 mutation in one of the mutated samples. 70

87 6. Quantitative PCR for BRCA1 and BRCA2 In order to validate the results of the MLPA, another method was employed as a validation tool - Quantitative PCR (qpcr). 6.1 qpcr for BRCA1 Due to the low rate of statistically significant results, qpcr for confirmation of the 4 amplified regions detected by MLPA, was performed for the whole cohort. None of the samples showed any of the amplifications in exons 5, 17, 19 and 21, by qpcr. Specifically, the 5 samples that showed the MLPA abnormality did not reveal any abnormality using the qpcr probe. Figure 37: qpcr for BRCA1 exon 19 in sample The RNAseP serves as a control sample, represented only once in the genome. The Ct for the RNAseP is lower than the Ct for exon 19 in sample 3364 (which has the highest chance of being amplified by the MLPA), which implies, that exon 19 in sample 3364 is not amplified. 71

88 Figure 38: qpcr for BRCA1 exon 19 in sample The RNAseP serves as a control sample, represented only once in the genome. We expected to see the same Ct for the RNAseP and for the sample 4849, and a lower Ct for the sample which was amplified by the MLPA (5079). As is clearly seen the Ct for the research sample is the highest, which means that it is not amplified. 72

89 6.2 qpcr for BRCA2 Due to the low rate of statistically significant results, qpcr for the search of the 4 dosage changes found by MLPA in the BRCA2 gene was performed for the whole cohort. None of the amplifications were detected and validated by qpcr, including the 6 samples that showed the MLPA associated abnormality. Figure 39: qpcr for BRCA2 exon 1b in sample The RNAseP serves as a control sample, represented only once in the genome. The Ct for the RNAseP is the same as the Ct for the sample 3965 (which has an high chance of being amplified by the MLPA), which implies, that the sample 3965 is not amplified. 73

90 7. Long-range PCR for BRCA2 A deletion was detected in two samples in exon 11A in BRCA2 gene by using MLPA. To confirm the MLPA results, we designed a reaction for exon 11 in BRCA2 by using long-range PCR technique for the all cohort. The results of the long-range PCR showed that in 3 samples there was an extra band (only one of the samples was identical to one of the two that had a suggestive MLPA pattern) bp 1500 bp 8500 bp Figure 40: Long-range pcr for exon 11 in BRCA2. In all five samples a band of 8500bp is ptresent and in three samples, an extra band in the size of 1600bp~ is noted. We designed an extra set of primers flannking the 11 exon in a different position, but this time there was not an extra band in any of the samples. We repeated the experiment several times and the only band was the one indicates on the whole exon. Another deletion was detected in exon 25 in BRCA2 gene by using MLPA. To confirm the MLPA results, we designed a reaction for exon 25 in BRCA2 by using long-range PCR technique for the all cohort. We designed several assays, but none of them yielded any meaningful interpretable results (data not shown). 8. Direct sequencing for the BRCA2 A deletion was detected in exon 25 in BRCA2 gene by using MLPA. To confirm the MLPA results, we designed a reaction for exons in BRCA2 by using direct sequencing technique for the all cohort. The results of the direct sequencing showed no mutations or polymorphisms in any of the samples (data not shown). 74

91 The results of the genotyping for missense mutations in BRCA1 and the MLPA analyses, were published in the peer reviewed journal "Familial Cancer" (appendix 1). 9. Genome-wide association study (GWAS) After eliminated the possibility that family history and clustering of breast cancer in the study population is attributed to mutations in BRCA1 and BRCA2 genes (except for the rare possibility of private mutations) (227), new chromosomal loci in the genome associated with breast cancer were sought. In order to do so, DNA samples were sent to Dr. Kenneth Offit from the Memorial Sloan Kettering Cancer Center in New York, who combined our families with additional American Jewish Ashkenazi, high-risk families. 9.1 GWAS in 249 Familial Breast Cancer Cases In phase 1, 435,632 genotypes of SNPs were analyzed in 249 (including a subset of families from the Sheba Medical center that formed the basis of the current study) probands from high risk AJ kindreds and in 299 cancer-free AJ controls (including 29 controls from the Sheba Medical Center). As an initial data quality control, SNPs that were out of Hardy Weinberg equilibrium (HWE) in the controls were filtered out; quantile quantile plot analysis showed that 44,165 SNPs with Fisher exact test P values <0.02 were not in HWE, leaving 391,467 "viable" reliable SNPs. Next, allele frequencies in cases versus controls were compared. 2 and Cochran Armitage tests produced closely comparable results. To ensure genotype accuracy, any SNP with more than two no calls was excluded from further analysis. Setting the miscall threshold at two or less restricted the analysis to those SNPs with >99.7% call rate. This procedure reduced the effective size of the survey to 150,080 SNPs. The minimal minor allele frequency (MAF) of all these SNPs was 10%. All results were corrected for false discovery rates using the Benjamni Hochberg algorithm (229). 75

92 Levels of significance Observed * Expected Ratio ,232 5, ,438 1, < All P < ,825 7, Table 27: Number of significant associations after initial AJ familial GWAS *Observed includes only values with H-W 2 in 299 controls with P > 0.02; the number of finite P values of this category were 150,080 of the 167,676 SNPs with call rates >99.7% surveyed. 9.2 Replication Analysis Using a Custom SNP Array. In phase 2, the top-ranking 123 chromosomal regions (each region spanning 200 kb) were selected for further analysis. To achieve satisfactory density of SNPs in candidate regions for haplotype analysis, 2 to 4 additional SNPs per region and an additional 18 SNPs that also showed strong association (P < 0.001) and mapped within the distance of 200 kb from the top 123 loci were added to the genotyped SNPs. In total, there were 343 SNPs selected for genotyping in a larger replication cohort that consisted of a fully independent set of 950 consecutive AJ breast cancer cases and 979 age-matched cancer-free AJ controls. This analysis was performed on the Illumina "GoldenGate" platform. For these 343 SNPs, allele frequencies were compared between breast cancer cases and controls using both the 2 test and the Cochran Armitage trend test, which produced closely comparable results. The results of the combined analysis (of familial cases and sporadic cases and controls) are summarized in table 28. Following this phase an additional third phase that encompassed 243 sporadic Jewish breast cancer patients and 187 asymptomatic controls were also genotyped and a pooled analysis was eventually done. 76

93 SNP * Chr Gene/region 2 P value of allele frequency (top rank no.) 2 P value of allele frequency (top rank no.) 2 P value of allele frequency (top rank no.) 2 P value aggregate OR allele for risk (95% CI) Phase 1 Phase 2 Phase3 rs q22.33 ECHDC1; RNF x 10 3 (1,358) 9.8 x 10 5 (1) 2.2 x x ( ) rs q22.33 ECHDC1; RNF x 10 3 (742) 9.9 x 10 5 (2) 2 x x ( ) rs q22.33 ECHDC1; RNF x 10 4 (205) 1.1 x 10 4 (3) 1.8 x x ( ) rs q22.33 ECHDC1; RNF x 10 3 (616) 1.2 x 10 4 (4) 2 x x ( ) rs q26.13 FGFR2 4.5 x 10 2 (10,062) 8.6 x 10 4 (5) 4 x x ( ) rs q13.1 PHKA1; 2.1 x x x 10 2 NS NS HDAC8 (4,751) (6) rs p13.3 A2BP1 2.5 x 10 2 (5,573) 3.3 x 10 3 (7) x ( ) Table 28: Regions of the genome that showed the strongest associations with AJ breast cancer after phase 2 of the study. NS, not significant; dom., dominant; rec., recessive; MAFcs, minor allele frequency in cases; MAFcn, minor allele frequency in controls. *Included are all SNPs that had P < 0.01 based on the analysis of the phase 2 case-control data (see ). Chromosome position by cytogenetic band. Genes identified in the genome browser that are within 100 kb on either side of the SNP indicated. Phase 1 consisted of 249 AJ probands from multiplex families in whom a mutation in BRCA1 and BRCA2 was excluded versus 299 cancer-free AJ controls; phase 2 consisted of 950 consecutive AJ breast cancer cases versus 979 cancer-free AJ controls; phase 3 consisted of an additional 243 AJ breast cancer cases from MSKCC and an independent 187 AJ cancer-free controls (rank no. represents the top order of the SNP of the total SNPs analyzed in phases 1 and 2). Aggregate data consisted of the combined phases 1, 2, and 3. OR calculated by using the aggregate data and based on the test of the alleles inverted to express the risk allele when necessary; 95% CI. **OR calculated by using the aggregate data and based on the 2 test of the dominant model. OR calculated by using the aggregate data and based on the 2 test of the recessive model. As shown in table 28, the association with the RNF146; ECHDC1 region at 6q22 was the strongest and most consistent in this study. The major haplotype (H1) composed of the four SNPs was found to be protective at 5.53 x 10 5 level of significance (OR 0.564, 95% CI ) (table 29). 77

94 Haplotype Genotype Phase 1 Phase 2 Aggregate study P n OR (CI) P n OR (CI) P n OR (CI) H1 ACGT 5.53 x 1.30 x , ( ) 10 4 ( ) ( ) ( ) H2 GTAA 2.79 x 1.11 x 1.33 x 10 2 ( ) 10 4 ( ) 10 7 ( ) Table 29: Haplotype statistics using rs , rs , rs , and rs in the RNF146; ECHDC1 locus. In the phase 1 and aggregate studies, 4% of the subjects were imputed to possess five rare haplotypes not shown here. NS, not significant This haplotype was confirmed in phase 2 at a significance threshold of 4 x 10 5 with nearly identical ORs and 95% CIs as the familial study. Because of the high linkage disequilibrium (LD) in the region, it could not be determined whether the signal was arising from the RNF146 or the ECHDC1 gene. SNP Gene Position MAF rs RNF rs RNF rs ECHDC rs ECHDC Table 30: SNP's location. Figure 41: LD image of linkage disequilibrium analysis of SNP's rs and rs in the RNF146 gene. 78

95 Figure 42: LD image of linkage disequilibrium analysis of SNP's rs and rs in the ECHDC1 gene. 10. Direct Sequencing of the RNF146 and ECHDC1 genes as candidates for inherited breast and ovarian cancer in Jewish Ashkenazi women. The results of the genome-wide association study seemingly identified a novel locus on chromosome 6 as putatively containing breast cancer susceptibility genes. At that locus two novel candidate genes: RNF146 and ECHDC1 are located. Hence, to further expand the results of the GWAS and assess the possible contribution of these two genes to inherited predisposition to breast cancer, study participants were genotyped to look for germline mutations in these two genes, by direct sequencing Direct sequencing of the RNF146 gene No mutations or polymorphisms were detected in the RNF146 gene in any of the 105 samples genotyped in this study Direct sequencing of the ECHDC1 gene No mutations or polymorphisms were detected in the ECHDC1 gene in any of the 105 samples genotyped in this study. 79

96 A sequence variant (a T to C change in intron 3 of the gene, 47 nucleotides into the intron -IVS3 + 47T/C) was detected in 4/105 women (*4%) genotyped in the ECHDC1 gene. Figure 43: Sequencing of one of the samples containing the T to C change in intron 3 of the gene, 47 nucleotides into the intron -IVS3 + 47T/C. The results of this latter, sequencing of the two genes study have been published in the peer reviewed journal "Familial Cancer" (Appendix 2). 80

97 Discussion The major aims of this study were to explore and define the extent of involvement of BRCA1 and BRCA2 genes in familial clustering of breast cancer in Jewish Ashkenazi, high risk families, and in parallel, define new regions within the human genome that are likely to harbor novel breast cancer susceptibility genes. Determination of BRCA1 and BRCA2 genes' involvement in familial breast cancer in the Jewish Ashkenazi population In this part of the study we evaluated the putative contribution of BRCA1 and BRCA2 mutations to familial breast cancer clustering among Ashkenazi high risk families using several approaches: initially, an indirect approach using haplotype sharing estimation in the tested families, using markers that are tightly associated with BRCA1 and BRCA2. Using that technique, it became obvious that affected individuals from the same family do not share a common haplotype using intragenic markers in BRCA1 and BRCA2 genes. These results strongly support the notion that BRCA1 and BRCA2 genes\mutations do not account for the observed familial aggregation of breast cancer in the families that formed the basis of the study. The second method used to assess the possible contribution of mutations in BRCA1 to the observed familial clustering of breast cancer was to genotype for putatively pathogenic and/or common missense mutations in BRCA1 (R866C, R331S, R841W, Y179C, C61G, M1008I). One of the genotyped missense mutations in BRCA1 (M1008I) was detected in two of the 105 study participants (~2%). The M1008I missense mutation is a conservative mutation due to a G -> A transition at nucleotide 3143 in exon 11 changing codon 1008 (ATG) encoding the non-polar Met into ATA encoding Ile, another non-polar residue. This Met is changed for a Ser in the mouse BRCA1 protein (230). The M1008I mutation was reported in ethnically diverse populations: Ashkenazim, Central/Eastern Europe, Western European population, Latin American/Caribbean, and Puerto Rico populations (BIC database). Given these considerations and the fact that this 81

98 variant was previously classified as neutral by two independent groups of investigators (231,232) and that it could be found in trans with a clearly deleterious BRCA1 mutation (232) it seems likely that this variant does not have a deleterious effect on BRCA1 protein function and is merely a rare polymorphism. The selection of the other 5 specific mutations was based on several criteria that made it plausible that these may indeed be pathogenic. We searched the BIC database for missense mutations in BRCA1 gene reported in the Ashkenazi population or individuals of East European origin. Missense mutations within conserved domains (R866C; R841W), those with a proven pathogenic effect (C61G), those that scored moderately high on the align- GVGD algorithm (213) (Y179C) or are extremely rare (R331S) were thus selected for further analysis in the present study. Yet no individual tested was found to harbor these mutations. The results of this analysis were published in Familial Cancer (appendix 1). Several explanations may account for this lack of detected missense mutations: inadequate assignment of the pathogenicity of these mutations, rarity of these mutations among high risk Jewish Ashkenazi families, even if they are truly pathogenic, small sample size, inclusion of noninherited cases among genotyped families. Indeed, at least one missense mutation tested (C61G) is a known pathogenic mutation (170). The lack of this mutation in the present study deserves special emphasis, as a substantial proportion of Ashkenazim originated in Poland, and intuitively, one would expect to find at least some representation of this mutation (prevalent in non-jewish individuals) among Jewish families. This expectation is based on the existence of the 5382InsC BRCA1 mutation in Jewish and non- Jewish individuals from the same geographical region (170). The lack of this mutation may be accounted for by a later date of origin of this mutation, or its limited geographical location to regions that were not densely populated with Jews in Poland. Other studies screened ethnically diverse populations searching for germline and missense mutations, to investigate the contribution of BRCA1 gene mutations to familial breast cancer. A 82

99 study done in Tunisia performed BRCA1 mutation analysis by DNA sequencing of all BRCA1 exons. Thirty two unrelated patients who had at least one first degree relative affected with breast and/or ovarian cancer were analyzed. Five of thirty two familial cases (15.6%) had deleterious BRCA1 mutations. Fifteen additional cases carried unclassified variants (UV) or single nucleotide polymorphisms (SNPs) (233). In order to evaluate the role of BRCA1 and BRCA2 germline mutations in Korean patients with sporadic breast cancer, 97 patients with sporadic breast cancer were analyzed for mutations in the BRCA1 and BRCA2 coding regions, by using a combination of fluorescent-conformation sensitive gel electrophoresis (F-CSGE) and direct sequencing. Fifty-five distinct sequence variants were detected, which included three pathogenic truncating mutations, 15 missense mutations, 16 polymorphisms, and 21 intronic variants. Twenty-six of these variants have never been previously reported and may be of Korean-specific origin. Two pathogenic BRCA1 mutations (c.922_924delinst, c.5445g>a) and one pathogenic BRCA2 mutation (c.2259delt) were observed, and two of these (BRCA1 c.5445g>a and BRCA2 c.2259delt) are novel. That study estimated the total prevalence of germline pathogenic mutations in BRCA1 and/or BRCA2 in Korean sporadic breast cancer to be about 3.1% (234). Another study done on cohort of sixty Korean women who developed breast cancer by age 40 years, revealed nine patients (9/60-15%) with 11 deleterious mutations (six in BRCA1 and five in BRCA2) and seven missense mutations of unknown significance. That study concluded that the prevalence of BRCA1 and BRCA2 mutations in Korean women with breast cancer at a young age was high (235). A study reported results of a full mutational screening analysis for germ line BRCA1 and BRCA2 mutations in 161 Southern Swedish patients with invasive epithelial ovarian carcinomas. Deleterious mutations were detected in 13/161 (8%) of the patients. Four additional missense variants (one in BRCA1 and three in BRCA2) with a possible association with an increased risk to ovarian cancer were revealed, resulting in a total frequency of BRCA gene alterations of 17/161 (11%) (236). A mutational analysis of the BRCA1 and BRCA2 genes in 124 Spanish women diagnosed with breast cancer before the age 41 and who 83

100 were not selected for a family history of this disease was carried out. The genetic study was performed by PCR-SSCP analysis and DNA sequencing. Six pathogenic BRCA mutations were identified in 7 unrelated probands 1 in BRCA1 and 5 in BRCA2. Additionally, two new missense mutations were identified: one in BRCA1 and the other in BRCA2. This data support the notion that Spaniards represent a heterogeneous population with its own spectrum of BRCA mutations, some of which appear as founder mutations (237). In Japan, the cumulative frequency of BRCA1 and BRCA2 mutations is 32%, similar to that of Caucasian populations and two founder mutations have been identified in BRCA1, one in BRCA2 (132). The third method that was used to assess the relative contribution of BRCA1 and BRCA2 mutations to the burden of cancer in families with a seemingly inherited predisposition to breast cancer was by analyzing for major gene rearrangements in both BRCA1 and BRCA2 using a combined approach of Multiplex Ligation-dependent Probe Amplification (MLPA), Quantitative PCR and Long Range PCR. MLPA analysis of Jewish women with inherited predisposition to breast cancer was suggestive of the existence of a few genomic rearrangements involving BRCA1 and BRCA2. Yet, these results could not be confirmed or validated independently by another technique- quantitative PCR. Several studies reported the MLPA as a high sensitivity technique by using a positive control that was originally detected by Southern blotting, the gold standard method for dosage changes (171,214,215). Other studies focusing on the presence of major gene rearrangements in BRCA1 and BRCA2 either used a MLPA kit with a different set of probes and primers (163) and qpcr for the validation of their findings. Based on these published studies that focused on non Jewish populations and the present study, it seems that an approach that uses MLPA as a screening technique and as a guide for further analysis and validation of the findings by an independent technique should be recommended, at least for the set of probes applied for BRCA1 and BRCA2. The results of this analysis were published in Familial Cancer (appendix 1). 84

101 The lack of finding of consistent major gene rearrangements involving BRCA1 and/or BRCA2 in high risk Jewish Ashkenazi families in the present study is in line with the previous study that focused on Jewish families. In that study, only one major gene rearrangement in BRCA2 in a non- Ashkenazi high risk family (238), and notably no gene rearrangements involving Ashkenazim (n=30) were reported. Similarly, two studies that have applied MLPA analysis have failed to show any BRCA1 and BRCA2 gene rearrangement in 135 French-Canadian (214) and no BRCA2 MLPAdetectable abnormalities in 36 Finnish (239) high risk families. The limitations of this part of the study should be borne in mind. This is a limited study from a single medical center and the high risk families analyzed may not be representative of the spectrum of high risk families, even of Ashkenazi origin. Only a subset of analyzed samples were adequately scored by the MLPA analysis, so that the actual number of analyzed individuals is even smaller regarding an assessment of the contribution of major gene rearrangement to inherited predisposition to breast/ovarian cancer among Ashkenazim. Lastly the lack of confirmation by Southern blotting (predominantly related to the lack of sufficient amounts of high quality DNA), detracts from the firmness of the conclusions. These results that show little or no contribution of pathogenic BRCA1 and BRCA2 mutations to inherited predisposition to breast cancer in Ashkenazim are in line with previously published data on that ethnic subset. Kauff et al (2002) have genotyped a large set of Jewish Ashkenazi women from high risk families that do not carry any of the three predominant mutations, and detected a pathogenic mutation (using full sequencing of both genes) in 3 of 70 probands from families analyzed, with only one case of a mutation detected when the family history did not include ovarian cancer (227). Another study which attempted to determine the spectrum of BRCA1 and BRCA2 mutations in high-risk Jewish individuals, non-carriers of any of the predominant Jewish mutations, screened 47 high-risk Jewish individuals, 26 Ashkenazim, and 21 non-ashkenazim. Overall, 13 sequence alterations in BRCA1 and eight in BRCA2 were detected, but none was a bone fide 85

102 pathogenic mutation. That study's findings suggest that except for the predominant mutations in BRCA1 and BRCA2 in Jewish individuals, there are only a handful of pathogenic mutations within these genes (139). Palma et al (2008) failed to detect genomic rearrangements involving either the BRCA1 or BRCA2 genes in 115 Ashkenazi Jewish patients representing high risk families, only one deleterious non founder point mutation in the BRCA2 gene and none in the BRCA1 gene (1/ %) (143). Phelan et al screened 160 Ashkenazi Jewish breast and breast-ovarian cancer families for the entire BRCA1 coding region by a combination of Denaturing Gradient Gel electrophoresis (DGGE) and Protein Truncation Test (PTT) and determined that non-founder BRCA1 mutations are rare in Ashkenazi Jewish breast/ovarian cancer families (141), and Frank et al screened 2,539 individuals of Ashkenazi ancestry. Of the 74 mutations identified through full sequence analysis of both BRCA1 and BRCA2, 16 (21.6%) were nonfounder mutations, including seven in BRCA1 and nine in BRCA2 (240). Thus it seems reasonable to state, based on the present study as well as from previously published data, that among Jewish Ashkenazi high risk families that do not harbor any of the predominant mutations the contribution of both BRCA1 BRCA2 genes to the familial clustering of breast/ovarian cancer is minimal. Genome-wide association study (GWAS) This part of the study constitutes one of the first genome-wide analyses that targeted breast cancer prone families in the hope to define novel chromosomal loci that harbor breast cancer susceptibility genes. Compared with the study of Easton and coworkers (154), where 390 familial breast cancer cases were initially screened with 227,876 SNPs, this study encompassed 249 familial breast cancer cases at approximately the same SNP density. Ovarian cancer is a rare tumor overall, in the general population and even among high risk families. Furthermore, the association of breast and ovarian cancer in the same individual or the same family is for the most part accounted for by BRCA1/ BRCA2 mutations. Hence the majority of high risk breast and ovarian cancer families are carriers of 86

103 BRCA1 or BRCA2 mutations. This is the reason why in the GWAS there were only breast cancer patients. Notably ~50 of these families were families contributed by the Oncogenetics unit at the Sheba Medical Center, who formed the basis of this thesis. In contrast with the Easton et al study, the current GWAS focused on a relatively genetically isolated population, Ashkenazi Jews, in which a significant increase in power to detect founder mutations in the BRCA1 BRCA2 as well as other genes was demonstrated (241). Hunter and coworkers (155) performed another GWAS that genotyped 528,173 SNPs in 1,145 postmenopausal women of European ancestry with sporadic invasive breast cancer. Both GWAS's performed by Easton et al and Hunter et al noted a significant contribution of FGFR2 alleles to breast cancer morbidity in familial and sporadic cases. Our Ashkenazi Jewish Genome wide association study AJ-GWAS confirmed breast cancer association with the FGFR2 locus noted in these two prior studies. Raskin et al estimated the relative risk and contribution of FGFR2 alleles to breast cancer risk in diverse ethnic groups within Jewish and other non-jewish Middle Eastern populations. They genotyped four FGFR2 SNPs and tested for association of these SNPs and haplotypes with breast cancer risk in a population-based case-control study of 1,529 women with breast cancer and 1,528 controls. They found significant associations between breast cancer risk and all four studied SNPs in FGFR2 (P trend for all SNPs < ). In ethnicity-specific analysis, all four SNPs were significantly associated with breast cancer risk in Ashkenazi and Sephardic (=non- Ashkenazi) Jews, with a similar but not significant trend in Arabs. They concluded that genetic variations in FGFR2, identified by rs , may account for a substantial fraction of breast cancer in Arab (12%), Ashkenazi (15%), and non-ashkenazi Jewish (22%) populations (242). In addition to the association between the FGFR2 gene locus, Easton and coworkers implicated 10 additional SNPs located to seven other genomic regions as associated with breast cancer risk. These additional loci for the most part could not be confirmed by the data generated by the AJ GWAS and by an additional breast cancer GWAS reported by Hunter and co-workers. Three of the additional 87

104 SNPs that were seemingly associated with breast cancer risk by Easton et al, were in or near the TNRC9 locus on chromosome 16q. In the AJ GWAS study, there were two reliable SNPs (rs and rs ) that were located close to that gene at the 5 end of the gene with allele P values in the range of P 0.01, and in the Hunter et al. study, there was only one SNP within 200 kb of TNRC9 (rs ) that showed a significant association with breast cancer risk, with P Similarly, at the MAP3K1 locus on chromosome 5q, where Easton reported association with breast cancer risk with P values in the range of 10 6 to 10 20, no significant SNPs by allele test was seen in the AJ study, and Hunter et al found only one SNP (rs726501) in that chromosomal region that was associated with breast cancer risk with a P value in the range of P 0.01 by allele test. Near the LSP1 region, AJ GWAS data showed two SNPs (rs , rs498337) associated with breast cancer risk with P values in the range of P 0.01 by allele test, whereas Easton et al reported P values in the range 10 5 to In accordance with the data from the AJ GWAS the Hunter et al. data provided evidence for one SNP (rs ) in the LSP1 region with a P value In the nearby H19 region on chromosome 11p, where Easton et al. reported P values in the range , no signal was detected in the AJ GWAS, whereas Hunter et al. found two SNPs (rs , rs ), with association P values in the range of 0.01, with one additional SNP, rs217228, with a P value in the range of Thus, with the exception of the common findings regarding the FGFR2 locus, there is relatively little overlap in findings between each of the two published studies and the AJ GWAS study. Four additional chromosomal regions (2p, 5p, 5q, and 8q) implicated by Easton and co-workers as putatively containing breast cancer susceptibility genes with P values in the range of , were screened in the AJ GWAS enriched with SNPs spanning 200Kb around the original SNPs. With one exception, the results of an apparent association between breast cancer and these four chromosomal regions could not be confirmed. A modest signal was noted in the AJ GWAS near SNP rs981782, (P 0.02 at rs ) which is near HCN1 (hyperpolarization activated cyclic 88

105 nucleotide-gated potassium channel 1), but Hunter's data revealed a P value of at rs in HCN1. It is possible that the differences observed between these three studies as described above are the result of several factors: population stratification, sample-size differences, genetic heterogeneity in the setting of differing genotyping platforms, and different algorithms for filtering the data. The differences observed in the three studies to date may also be a result of differential choice of SNPs used in the Perlegen, Affymetrix, and Illumina platforms used in these studies. In the study using the Perlegen platform (155), and in our study using the Affymetrix platform, filtering removed from 25% to 75% of the SNPs present on the original arrays. Although this resulted in significantly reduced likelihood of false positive associations, it could also have allowed false negatives, i.e., missed associations. The results of these GWAS were subsequently used to highlight additional aspects of breast cancer phenotype. A report from the Breast Cancer Association Consortium (BCAC), a consortium spearheaded by Prof Doug Easton, aimed at evaluating whether common variants in FGFR2, TNRC9, MAP3K1, 8q24, and LSP1 affect the clinical presentation of breast cancer and survival after diagnosis. To that end, these susceptibility loci were genotyped and correlated with clinically relevant somatic tumor characteristics in 23,039 invasive breast cancer cases and 26,273 controls of European or Asian origin from 20 centers. In addition, the association of these SNPs with overall survival was evaluated in 13,527 breast cancer cases from 13 centers. The most notable findings were that the genetic variants in the fibroblast growth factor receptor 2 (FGFR2) gene was more strongly related to ER-positive (per-allele OR (95%CI) = 1.31 ( )) than ER-negative (1.08 ( )) disease (P for heterogeneity = 10(-13)), and to PR-positive and low grade tumors (P = 10(-5), 10(-8), respectively). The SNP within the 8q24 region was also stronger for ER-positive, PR-positive, and low grade tumors (P = 0.001, and 10(-4), respectively). The SNPs in these loci did not significantly affect survival, after stratifying for known prognostic factors. Analyses 89

106 indicated that common genetic variants influence the pathological subtype of breast cancer and provide further support for the hypothesis that ER-positive and ER-negative diseases are biologically distinct tumors (243). Another study carried out by Fletcher et al (244) genotyped SNPs that all localize to the 8q24 region: rs , rs , rs variants associated with prostate cancer, rs associated with both prostate and colorectal cancer, and rs associated with breast cancer, in a series of 1,499 breast cancer cases and 1,390 controls ascertained through the English and Scottish cancer registries. That study provides further evidence for the relationship between rs and breast cancer risk, with heterozygote odds ratio (OR) % confidence interval (CI) and homozygote OR 1.52 (95% CI, ; P trend = ). The Fletcher et al study observed a protective effect of rs for breast cancer (allelic OR, 0.88; 95% CI, ; P = 0.02), which is supported by the Cancer Genetic Markers of Susceptibility data (pooled allelic OR, 0.88; 95% CI, ; P = 0.003) (245). The fact that a protective effect for breast cancer of one variant (rs ) that has previously been reported to be associated with a 1.25-fold increased risk of prostate cancer (246), may indicate that the effects of the risk alleles clustered at 8q24, are cancer-site-specific (244). Huijts et al also analyzed the correlation between disease characteristics and the patient genotypes of seven SNPs in five genomic loci: The FGFR2 gene (SNP rs ), the TNRC9 gene (rs , rs , rs ), the MAP3K1 gene (rs889312), the LSP1 gene (rs ), and a geneless linkage disequilibrium block of ~110 kb on 8q24 (rs ).these researchers genotyped these SNPs in an unselected prospective cohort of 1,267 consecutive patients with primary breast cancer. In their study heterozygote carriers and minor allele homozygote carriers for SNP rs in the MAP3K1 gene were significantly less likely to be lymph node positive at breast cancer diagnosis compared with major allele homozygote carriers (P = 0.044). Heterozygote carriers and minor allele homozygote carriers for SNP rs near the TNCR9 gene were more likely to be diagnosed 90

107 before the age of 60 years (P = 0.025) relative to major allele homozygote carriers. A correlation between the number of minor alleles of rs in FGFR2 and the average number of firstdegree and second-degree relatives with breast cancer and/or ovarian cancer was also noted in that study (P = 0.05). All other disease characteristics, including tumor size, grade, and estrogen or progesterone receptor status, were not significantly associated with any of these 7 sequence variants (247). Liang and co-workers assessed whether polymorphisms of FGFR2 are associated with breast cancer risk in Chinese women and whether these associations are stronger in women with a reproductive history suggestive of greater exposure to endogenous estrogens. They genotyped three FGFR2 polymorphisms (rs , rs , rs ) in 1049 breast cancer patients and 1073 cancerfree controls. Each of the three SNPs was significantly associated with increased breast cancer risk in a dose-dependent manner. Compared with women with 0-2 risk loci, those with 3 risk loci had a 1.36-fold increased odds of breast cancer (95% confidence interval = , P = 0.001). In stratified analyses, associations between the presence of 3 risk loci and breast cancer were stronger among women with ER- and/or PR-positive cancers, pre-menopausal women and women with an older age at first live birth. Furthermore, there was a significant additive interaction between risk genotypes and menopausal status (P for multiplication interaction/additive interaction: 0.083/0.037). These findings indicate that genetic variants in FGFR2 may contribute to breast cancer occurrence in Chinese women, possibly through pathways related to estrogen and/or progesterone (248). None of the three other SNPs reported by Easton and co-workers that are located to 2p, 5p, 5q, was found to be associated with breast cancer risk in the AJ GWAS. The present GWAS noted a risk locus at 6q22.33, a locus not detected by either of the two previous GWAS performed on non- Jewish individuals. 91

108 The association with the RNF146; ECHDC1 gene region at 6q22.33 was the strongest and most consistent in this study. However, it could not be determined whether the association noted in this study was arising from the ECHDC1 gene, RNF146, or another locus in linkage disequilibrium. The limitations of the Ashkenazi Jewish GWAS should be pointed out. This study was performed on a relatively isolated genetically homogeneous population. Residual effects of population heterogeneity were addressed through use of a principal components analysis. It remains to be seen whether further study in non-ashkenazi cohorts, using these and other SNPs, will confirm the association with the RNF146/ECHDC1 region. It is important also to emphasize the clinical challenges posed by the relatively modest magnitude of relative risk associated with SNPs in that particular chromosomal region. The 1.4 relative risk documented here is substantially small compared with the 20- to 40-fold increase in risk for early-onset breast cancer associated with BRCA mutations. The high frequency of these risk alleles in the population (23% of the population studied carried the H2 haplotype), combined with the observed relative risk produces a calculated proportion of breast cancer attributable to this risk factor in this population of 7%. However, such calculations do not take into account possible interactions between multiple loci. It will be important to look for multiplicative effects involving these and other low-penetrance alleles and cancer risk. Such interactions, which will impact both the attributable fraction and the relative risk of breast cancer, have thus far not been observed among known candidate loci (154,249). In the absence of such interactive effects, the finding of individual low-penetrance genetic risk factors for breast cancer, such as the chromosome 6q22 locus, will be of limited clinical utility compared with known high-penetrance mutations of genes such as BRCA1 or BRCA2 (207,250). Other studies using the same or similar genotyping strategy that focused on relatively isolated populations to look for breast cancer susceptibility loci were reported in the past few years. Stacey et al focusing on the Iceland population, genotyped approximately 300,000 SNPs in 1,600 Icelandic individuals with breast cancer and 11,563 controls using the Illumina Hap300 platform. In that 92

109 study, two SNPs consistently showed an association with breast cancer: rs on chromosome 2q35 and rs on 16q12. Risk from both alleles was confined to estrogen receptor-positive tumors. No gene(s) in the linkage disequilibrium block containing rs are noted. rs is located close to the 5' end of TNRC9, a high mobility group chromatinassociated protein whose expression is implicated in breast cancer metastasis to bone (251). A year later the same group of investigators reported another genome-wide association study of breast cancer predisposition with replication and refinement studies involving 6,145 cases and 33,016 controls. In that follow up study, two SNPs (rs and rs ) on 5p12 were reportedly associated with breast cancer risk, preferentially for estrogen receptor (ER)-positive tumors (OR = 1.27, P = 2.5 x 10(-12) for rs ). The nearest gene, MRPS30, was previously implicated in apoptosis, ER-positive tumors and favorable prognosis (252). Kibriya et al reported their own somewhat limited efforts using the GWAS approach targeting early-onset breast cancer. They included 30 randomly sampled incident invasive breast cancer cases aged <45 years without deleterious mutations in the BRCA1 or BRCA2 genes, and 30 population controls. Several possible significant regions harboring biologically meaningful known candidate genes, such as genes encoding fibroblast growth factor, transforming growth factor, epidermal growth factor, and estrogen synthesis enzymes seem to be associated with early-onset breast cancer. However, the sample size and the populational heterogeneity detract from the scientific value of these findings (253). To date GWAS s have identified about 250 common variants associated with risk alleles that contribute to a wide range of diseases (254,255). So far, most of these impart small effects on disease risk, and, in addition, even when extremely large studies have been performed, the vast majority of the genetic contributions to disease risk remain unexplained (256). These findings suggest that individually rare variants with relatively large effect may account for a large fraction of this missing trait variance. Indeed, studies addressing this question have documented the presence 93

110 of individually rare variants with relatively large effect (257,258). Consistent with the Mendelian model, coding variants have proven to be prevalent sources of such rare variants. Moreover, our current ability to interpret the functional consequences of sequence variation outside coding regions is highly limited. Thus, these concerns inspire execution of methods to sequence complete coding regions of genomes, namely the exome. Direct Sequencing of The RNF146 and ECHDC1 genes as candidates for inherited breast and ovarian cancer in Jewish Ashkenazi women In the current study, no truly pathogenic germline mutations were detected in two seemingly biologically plausible candidate genes (RNF146 and ECHDC1) in a large set of Jewish Ashkenazi women at high risk for developing breast/ovarian cancer. The plausibility of these genes being involved in predisposing to breast/ovarian cancer stems from two independent sources: the known functions of the gene products and their chromosomal localization. The ECHDC1 gene - Enoyl-coenzyme A hydratase domain containing 1 involved in lipid metabolism. The gene product has a related domain to the mitochondrial enoyl-coa hydratase/3- hydroxyacyl-coa dehydrogenase/3-ketoacyl-coa thiolase, a key trifunctional protein in mitochondrial fatty acid oxidation (259). The tri-functional enzyme comprises three consecutive steps in the mitochondrial beta-oxidation of long-chain acyl-coa esters: 2-enoyl-CoA hydratase, 3- hydroxyacyl-coa dehydrogenase and 3-ketoacyl-CoA thiolase. Deficiencies in either 3- hydroxyacyl-coa dehydrogenase activity, or all three activities, are important causes of human disease. The dehydrogenase and thiolase have a requirement for NAD+ and CoA respectively, whose levels are conserved within the mitochondrion and thus provide possible means for control and regulation of beta-oxidation. Although ECHDC1 gene mutations have not been reported to occur somatically in breast cancer, its possible involvement in breast cancer pathogenesis stems from the observation that fatty acid 94

111 synthase-dependent endogenous fatty acid activity is abnormally elevated in a biologically aggressive subset of breast carcinomas, and that inhibition of that activity induces apoptosis in breast cancer cell lines, an effect enhanced 300-fold in TP53 null cell lines (260,261). Recently, ECHDC1 gene expression was also found to be down-regulated in hepatocellular carcinoma, as further supportive evidence to its intimate involvement in the malignant transformation process (262). RNF146 (RING finger protein 146), also called dactylidin, is a 359 amino acid protein that contains one RING-type zinc finger and one WWE domain. It encodes a ubiquitously expressed, cytoplasmic polypeptide containing an amino-terminal C3HC4 RING finger domain that functions as a ubiquitin protein ligase (E3). Abnormalities in the ubiquitin mediated protein degradation pathway affect cellular processes (e.g., cell cycle, apoptosis, etc) that are pivotal in breast cancer pathogenesis (263) Via its RING-type zinc finger, RNF146 may play a role in transcriptional regulation and protein degradation events (264). The RING-type zinc finger motif is present in a number of viral and eukaryotic proteins and is made of a conserved cysteine-rich domain that is able to bind two zinc atoms. Proteins that contain this conserved domain are generally involved in the ubiquitination pathway of protein degradation (264). Other genes that confer breast cancer predisposition (BRCA1, BRCA2, and BARD1) also exhibit ubiquitin like activity or have an indirect effect on ubiquitin mediated protein degradation (265). Another supportive evidence for the putative involvement of RNF146 in the pathogenesis of breast cancer comes from its differential expression in breast epithelial stem cells in parous compared with nulliparous women (266). These lines of evidence, combined with the reported rates of somatic loss of 6q in pre-malignant breast tissue lesions as well as breast cancer in a region that encompass both the RNF146 and the ECHDC1 genes (267), make it plausible for both genes to be involved in predisposing to breast cancer. 95

112 Despite the lack of mutations in both genes, the technique cannot be faulted, as a novel sequence alteration has been detected in ~4% of this subset of women. Furthermore, one or both genes may still be involved in conferring inherited predisposition to breast cancer by other mechanisms: major gene rearrangements that cannot be detected by PCR, sequence variants in gene expression controlling regions, or epigenetic mechanisms. In addition, it is possible that using a larger group of women from the same ethnic origin, rare sequence alterations may be detected. To assess that, a larger study that genotypes more women from additional medical centers that focus on AJ women is warranted. Noteworthy, 45 of the women who were genotyped herein, were also genotyped as part of the original GWAS. Presently, we can conclude based on the results of this preliminary study, that germline mutations in both the ECHDC1 and RNF146 genes contribute only marginally, if at all, to the burden of inherited predisposition to breast cancer among AJ. The results of the GWAS that revealed an association with breast cancer are in all likelihood attributed to other, yet unidentified genes in the 6q region. The next logical step then would be to screen all genes that are located in this region and look for germline mutations that are likely to be associated with breast cancer. If no mutations will be identified in genes that localize to this chromosomal region, then we must assume that the results of the current study indicate that this region is involved in inherited predisposition by other mechanisms: harboring mirna binding sites, containing coding sequence affecting breast cancer risk in ways that are distinct from exonic germline alterations. The way to diagnose and assess these putative alterations would be to sequence the entire chromosomal region (i.e., coding and non coding regions) using next generation methodologies of sequencing and comparing cases and controls. If no sequence alterations that distinguish between cases and controls are detected then one must assume that the results are erroneous and cannot contribute to the understanding of the genetic basis of breast cancer. One way of progressing in that last scenario in the search for identifying breast cancer genes in the Ashkenazi population should be sequencing all of the exons 96

113 in the human genes or even in the human genome, in cases from families with a strong family history of breast cancer. If a mutation is detected within any gene this would facilitate assessing the relative contribution of this gene and the mutations in it to the overall burden of familial breast cancer in Ashkenazim and other ethnically diverse populations. Such a finding would impact both genetic counseling facilitating a more accurate, rational genetic counseling, and an objective assessment of breast cancer risk based on the genotype within these high risk families. Furthermore, non carriers who are members of a family with known mutation can be reassigned an average risk status rather than high risk status, an assignment with obvious clinical and psychological impact. The results of this analysis have been published in the peer reviewed journal "Familial Cancer" (Appendix 2). 97

114 References 1 Parkin DM, Bray F, Pisani P, Ferlay J. Global cancer statistics. CA- A cancer. Journal for Clinicians 1999; 49: Pharoah PDP, Mackay J. Absolute risk of breast cancer in women at increased risk: a more useful clinical measure than relative risk? The Breast 1998; 7: Easton DF. How many more breast cancer predisposition genes are there? Breast Cancer Research. 1999; 1: Shattuck-Eidens D, Oliphant A, McClure M, McBride C, Gupte J, Rubano T, Pruss D, Tavtigian SV, Teng DH, Adey N, Staebell M, Gumpper K, Lundstrom R, Hulick M, Kelly M, Holmen J, Lingenfelter B, Manley S, Fujimura F, Luce M, Ward B, Cannon-Albright L, Steele L, Offit K, Thomas A. BRCA1 sequence analysis in women at high risk for susceptibility mutations. Risk factor analysis and implications for genetic testing. JAMA 1997; 278: Couch FJ, DeShano ML, Blackwood MA, Calzone K, Stopfer J, Campeau L, Ganguly A, Rebbeck T, Weber BL. BRCA1 mutations in women attending clinics that evaluate the risk of breast cancer. N Engl J Med 1997; 336: Hemminki K, Vaittinen P, Kyyronen P. Age-specific familial risks in common cancers of the offspring. International Journal of Cancer 1998; 78: Newman B, Millikan RC, King MC. Genetic epidemiology of breast and ovarian cancers. Epidemiological Reviews 1997; 19: Habel LA, Moe RE, Daling JR, Holte S, Rossing MA, Weiss NS. Risk of contralateral breast cancer among women with carcinoma in situ of the breast. Ann Surg 1997; 225: Harris JR, Lippman ME, Veronesi U, Willett W. Breast cancer. N Engl J Med 1992; 327: Kelsey JL, Berkowitz GS. Breast cancer epidemiology. Cancer Res 1988; 48: Kelsey JL, Gammon MD. Epidemiology of breast cancer. Epidemiol Rev 1990;12: Kelsey JL, Gammon MD, John EM. Reproductive factors and breast cancer. Epidemiol Rev 1993; 15: Spicer DV, Krecker EA, Pike MC. The endocrine prevention of breast cancer. Cancer Invest. 1995; 13:

115 16 Anderson TJ, Ferguson DJ, Raab GM.. Cell turnover in the resting human breast: Influence of parity, contraceptive pill, age, and laterality. Br J Cancer 1982; 46: Daling JR, Malone KE, Voigt LF, White E, Weiss NS. Risk of breast cancer among young women. Relationship to induced abortion. J Natl Cancer Inst 1994; 86: Kelsey JL. A review of the epidemiology of human breast cancer. Epidemiol Rev 1979; 1: White E. Projected changes in breast cancer incidence due to the trend toward delayed childbearing. American Journal of Public Health 1987; 77: Woodman I. Breast feeding reduces risk of breast cancer, says study. BMJ 2002; 325: Newcomb PA, Storer BE, Longnecker MP, Mittendorf R, Greenberg ER, Clapp RW, Burke KP, Willett WC, MacMahon B. Lactation and a reduced risk of premenopausal breast cancer. New England Journal of Medicine 1994; 330: Pike MC, Wynn V, Chilvers CE, McPherson K. Oral contraceptives and breast cancer rates. Lancet 1984; 1(8373): McPherson K, Neil A, Vessey MP, Doll R. Oral contraceptives and breast cancer. Lancet 1983; 1(8364): Marchbanks PA, McDonald JA, Wilson HG, Folger SG, Mandel MG, Daling JR, Bernstein L, Malone KE, Ursin G, Strom BL, Norman SA, Wingo PA, Burkman RT, Berlin JA, Simon MS, Spirtas R, Weiss LK. Oral contraceptives and the risk of breast cancer. N Engl J Med 2002; 346: Nelson HD, Humphrey LL, Nygren P, Teutsch SM, Allan JD. Postmenopausal hormone replacement therapy :scientific review. JAMA 2002; 288: Kurian AW, Clarke CA, Carlson RW. The decline in breast cancer incidence: real or imaginary? Curr Oncol Rep 2009; 11(1): Ravdin PM, Cronin KA, Howlader N, Berg CD, Chlebowski RT, Feuer EJ, Edwards BK, Berry DA. The decrease in breast-cancer incidence in 2003 in the United States. The N Engl J Med 2007; 356(16): Parkin DM. Is the recent fall in incidence of post-menopausal breast cancer in UK related to changes in use of hormone replacement therapy? Eur J Cancer 2009; 45(9): Singletary ES. Rating the Risk Factors for Breast Cancer. Annals of surgery 2003; 237: Romieu I, Lajous M.The role of obesity, physical activity and dietary factors on the risk for breast cancer: Mexican experience. Salud Publica Mex 2009; 51 Suppl 2:s

116 31 Key TJ, Spencer EA, Reeves GK. Symposium 1: Overnutrition: consequences and solutions Obesity and cancer risk. Proc Nutr Soc 2009; 3: Hamajima N, Hirose K, Tajima K, Rohan T, Calle EE, Heath CW Jr, Coates RJ, Liff JM, Talamini R, Chantarakul N, Koetsawang S, Rachawat D, Morabia A, Schuman L, Stewart W, Szklo M, Bain C, Schofield F, Siskind V, Band P, Coldman AJ, Gallagher RP, Hislop TG, Yang P, Kolonel LM, Nomura AM, Hu J, Johnson KC, Mao Y, De Sanjosé S, Lee N, Marchbanks P, Ory HW, Peterson HB, Wilson HG, Wingo PA, Ebeling K, Kunde D, Nishan P, Hopper JL, Colditz G, Gajalanski V, Martin N, Pardthaisong T, Silpisornkosol S, Theetranont C, Boosiri B, Chutivongse S, Jimakorn P, Virutamasen P, Wongsrichanalai C, Ewertz M, Adami HO, Bergkvist L, Magnusson C, Persson I, Chang-Claude J, Paul C, Skegg DC, Spears GF, Boyle P, Evstifeeva T, Daling JR, Hutchinson WB, Malone K, Noonan EA, Stanford JL, Thomas DB, Weiss NS, White E, Andrieu N, Brêmond A, Clavel F, Gairard B, Lansac J, Piana L, Renaud R, Izquierdo A, Viladiu P, Cuevas HR, Ontiveros P, Palet A, Salazar SB, Aristizabel N, Cuadros A, Tryggvadottir L, Tulinius H, Bachelot A, Lê MG, Peto J, Franceschi S, Lubin F, Modan B, Ron E, Wax Y, Friedman GD, Hiatt RA, Levi F, Bishop T, Kosmelj K, Primic-Zakelj M, Ravnihar B, Stare J, Beeson WL, Fraser G, Bullbrook RD, Cuzick J, Duffy SW, Fentiman IS, Hayward JL, Wang DY, McMichael AJ, McPherson K, Hanson RL, Leske MC, Mahoney MC, Nasca PC, Varma AO, Weinstein AL, Moller TR, Olsson H, Ranstam J, Goldbohm RA, van den Brandt PA, Apelo RA, Baens J, de la Cruz JR, Javier B, Lacaya LB, Ngelangel CA, La Vecchia C, Negri E, Marubini E, Ferraroni M, Gerber M, Richardson S, Segala C, Gatei D, Kenya P, Kungu A, Mati JG, Brinton LA, Hoover R, Schairer C, Spirtas R, Lee HP, Rookus MA, van Leeuwen FE, Schoenberg JA, McCredie M, Gammon MD, Clarke EA, Jones L, Neil A, Vessey M, Yeates D, Appleby P, Banks E, Beral V, Bull D, Crossley B, Goodill A, Green J, Hermon C, Key T, Langston N, Lewis C, Reeves G, Collins R, Doll R, Peto R, Mabuchi K, Preston D, Hannaford P, Kay C, Rosero-Bixby L, Gao YT, Jin F, Yuan JM, Wei HY, Yun T, Zhiheng C, Berry G, Cooper Booth J, Jelihovsky T, MacLennan R, Shearman R, Wang QS, Baines CJ, Miller AB, Wall C, Lund E, Stalsberg H, Shu XO, Zheng W, Katsouyanni K, Trichopoulou A, Trichopoulos D, Dabancens A, Martinez L, Molina R, Salas O, Alexander FE, Anderson K, Folsom AR, Hulka BS, Bernstein L, Enger S, Haile RW, Paganini-Hill A, Pike MC, Ross RK, Ursin G, Yu MC, Longnecker MP, Newcomb P, Bergkvist L, Kalache A, Farley TM, Holck S, Meirik O; Collaborative Group on Hormonal Factors in Breast Cancer. Collaborative Group on Hormonal Factors in Breast Cancer. Alcohol, tobacco and breast cancer-collaborative reanalysis of individual data from 53 epidemiological studies, including 58,515 women with breast cancer and 95,067 women without the disease. Int J Cancer 2006; 119: Bernstein L, Patel AV, Ursin G, Sullivan-Halley J, Press MF, Deapen D, Berlin JA, Daling JR, McDonald JA, Norman SA, Malone KE, Strom BL, Liff J, Folger SG, Simon MS, Burkman RT, Marchbanks PA, Weiss LK, Spirtas R. Lifetime recreational exercise activity and breast cancer risk among black women and white women. J Natl Cancer Inst 2005; 97: Fuemmeler BF, Pendzich MK, Tercyak KP. Weight, Dietary Behavior, and Physical Activity in Childhood and Adolescence: Implications for Adult Cancer Risk. Obes Facts 2009; 2(3):

117 35 Bernstein L. Identifying population-based approaches to lower breast cancer risk. Oncogene 2008; 27 Suppl 2:S Szabo CI, King MC. Inherited breast and ovarian cancer. Hum Mol Genet 1995; Parkin DM, Whelan SL, Ferhy J Raymond L. Cancer incidence in five continents. Oxford: International Agency for Research on Cancer, Oxford University press Dapic V, Carvalho MA, Monteiro ANA. Breast cancer susceptibility and the DNAdamage response. Cancer control 2005; 12: D Ford, D F Easton, M Stratton, S Narod, D Goldgar, P Devilee, D T Bishop, B Weber, G Lenoir, J Chang-Claude, H Sobol, M D Teare, J Struewing, A Arason, S Scherneck, J Peto, T R Rebbeck, P Tonin, S Neuhausen, R Barkardottir, J Eyfjord, H Lynch, B A Ponder, S A Gayther, M Zelada-Hedman. Genetic heterogeneity and penetrance analysis of the BRCA1 and BRCA2 genes in breast cancer families: The Breast Cancer Linkage Consortium. Am J Hum Genet 1998; 62: Easton DF, Steele L, Fields P, Ormiston W, Averill D, Daly PA, McManus R, Neuhausen SL, Ford D, Wooster R, Cannon-Albright LA, Stratton MR, Goldgar DE. Cancer risks in two large breast cancer families linked to BRCA2 on chromosome 13q Am J Hum Genet 1997; 61: Struewing JP, Hartge P, Wacholder S, Baker SM, Berlin M, McAdams M, Timmerman MM, Brody LC, Tuckerj MA. The risk of cancer associated with specific mutations of BRCA1 and BRCA2 among Ashkenazi Jews. N Engl J Med 1997; 336: Satagopan JM, Offit K, Foulkes W, Robson ME, Wacholder S, Eng CM, Karp SE, Begg CB. The lifetime risks of breast cancer among Ashkenazi Jewish carriers of BRCA1 and BRCA2 mutations. Cancer Epidemiol Biomarker Prev 2001; 10: Antoniou A, Pharoah PD, Narod S, Risch HA, Eyfjord JE, Hopper JL, Loman N, Olsson H, Johannsson O, Borg A, Pasini B, Radice P, Manoukian S, Eccles DM, Tang N, Olah E, Anton-Culver H, Warner E, Lubinski J, Gronwald J, Gorski B, Tulinius H, Thorlacius S, Eerola H, Nevanlinna H, Syrjäkoski K, Kallioniemi OP, Thompson D, Evans C, Peto J, Lalloo F, Evans DG, Easton DF. Average risks of breast and ovarian cancer associated with BRCA1 or BRCA2 mutations detected in case Series unselected for family history: a combined analysis of 22 studies. Am J Hum Genet 2003; 72: King MC, Marks JH, Mandell JB. New York Breast Cancer Study Group. Breast and ovarian cancer risks due to inherited mutations in BRCA1 and BRCA2. Science 2003; 302: Walsh T, King MC. Ten genes for inherited breast cancer. Cancer Cell 2007; 11(2): Rahman N, Seal S, Thompson D, Kelly P, Renwick A, Elliott A, Reid S, Spanova K, Barfoot R, Chagtai T, Jayatilake H, McGuffog L, Hanks S, Evans DG, Eccles D; Breast Cancer Susceptibility Collaboration (UK), Easton DF, Stratton MR. PALB2, which encodes a 101

118 BRCA2-interacting protein, is a breast cancer susceptibility gene. Nature Genet 2007; 39: Thompson D, Duedal S, Kirner J, McGuffog L, Last J, Reiman A, Byrd P, Taylor M, Easton DF. Cancer risks and mortality in heterozygous ATM mutation carriers. J Natl Cancer Inst 2005; 97: Meijers-Heijboer H, van den Ouweland A, Klijn J, Wasielewski M, de Snoo A, Oldenburg R, Hollestelle A, Houben M, Crepin E, van Veghel-Plandsoen M, Elstrodt F, van Duijn C, Bartels C, Meijers C, Schutte M, McGuffog L, Thompson D, Easton D, Sodha N, Seal S, Barfoot R, Mangion J, Chang-Claude J, Eccles D, Eeles R, Evans DG, Houlston R, Murday V, Narod S, Peretz T, Peto J, Phelan C, Zhang HX, Szabo C, Devilee P, Goldgar D, Futreal PA, Nathanson KL, Weber B, Rahman N, Stratton MR; CHEK2-Breast Cancer Consortium. Low-penetrance susceptibility to breast cancer due to CHEK2*1100delC in noncarriers of BRCA1 or BRCA2 mutations. Nature Genet 2002; 31: Erkko H, Xia B, Nikkilä J, Schleutker J, Syrjäkoski K, Mannermaa A, Kallioniemi A, Pylkäs K, Karppinen SM, Rapakko K, Miron A, Sheng Q, Li G, Mattila H, Bell DW, Haber DA, Grip M, Reiman M, Jukkola-Vuorinen A, Mustonen A, Kere J, Aaltonen LA, Kosma VM, Kataja V, Soini Y, Drapkin RI, Livingston DM, Winqvist R. A recurrent mutation in PALB2 in Finnish cancer families. Nature 2007; 446: Renwick A, Thompson D, Seal S, Kelly P, Chagtai T, Ahmed M, North B, Jayatilake H, Barfoot R, Spanova K, McGuffog L, Evans DG, Eccles D; Breast Cancer Susceptibility Collaboration (UK), Easton DF, Stratton MR, Rahman N. ATM mutations that cause ataxiatelangiectasia are breast cancer susceptibility alleles. Nature Genet 2006; 38: Seal S, Thompson D, Renwick A, Elliott A, Kelly P, Barfoot R, Chagtai T, Jayatilake H, Ahmed M, Spanova K, North B, McGuffog L, Evans DG, Eccles D; Breast Cancer Susceptibility Collaboration (UK), Easton DF, Stratton MR, Rahman N. Truncating mutations in the Fanconi anemia J gene BRIP1 are lowpenetrance breast cancer susceptibility alleles. Nature Genet 2006; 38: The CHEK2 Breast Cancer Case-Control Consortium. CHEK2*1100delC and susceptibility to breast cancer: A collaborative analysis involving 10,860 breast cancer cases and 9,065 controls from ten studies. Am J Hum Genet 2004; 74: Laitman Y, Kaufman B, Lahad EL, Papa MZ, Friedman E. Germline CHEK2 mutations in Jewish Ashkenazi women at high risk for breast cancer. Isr Med Assoc J 2007; 9(11): Ohayon T, Gal I, Baruch RG, Szabo C, Friedman E. CHEK2*1100delC and male breast cancer risk in Israel. Int J Cancer 2004; 108(3): Ohayon T, Gershoni-Baruch R, Papa MZ, Distelman Menachem T, Eisenberg Barzilai S, Friedman E. The R72P P53 mutation is associated with familial breast cancer in Jewish women. Br J Cancer 2005; 92(6): Figer A, Kaplan A, Frydman M, Lev D, Paswell J, Papa MZ, Goldman B, Friedman E. Germline mutations in the PTEN gene in Israeli patients with Bannayan-Riley-Ruvalcaba syndrome and women with familial breast cancer. Clin Genet 2002; 62(4):

119 58 Argento M, Hoffman P, Gauchez AS. Ovarian cancer detection and treatment: current situation and future prospects. Anticancer Res 2008; 28(5B): NCI s SEER Cancer Statistics Review Simchoni S, Friedman E, Kaufman B, Gershoni-Baruch R, Orr-Urtreger A, Kedar-Barnes I, Shiri-Sverdlov R, Dagan E, Tsabari S, Shohat M, Catane R, King MC, Lahad A, Levy-Lahad E. Familial clustering of site-specific cancer risks associated with BRCA1 and BRCA2 mutations in the Ashkenazi Jewish population. Proc Natl Acad Sci U S A 2006; 103(10): Soegaard M, Frederiksen K, Jensen A, Høgdall E, Høgdall C, Blaakaer J, Ramus SJ, Gayther SA, Kjaer SK. Risk of ovarian cancer in women with first-degree relatives with cancer. Acta Obstet Gynecol Scand 2009; 88(4): Permuth-Wey J, Sellers TA. Epidemiology of ovarian cancer. Methods Mol Biol 2009; 472: Modan B, Hartge P, Hirsh-Yechezkel G, Chetrit A, Lubin F, Beller U, Ben-Baruch G, Fishman A, Menczer J, Ebbers SM, Tucker MA, Wacholder S, Struewing JP, Friedman E, Piura B; National Israel Ovarian Cancer Study Group. Parity, oral contraceptives, and the risk of ovarian cancer among carriers and noncarriers of a BRCA1 or BRCA2 mutation. N Engl J Med 2001; 345(4): Claus EB, Schildkraut JM, Thompson WD, Risch NJ. The genetic attributable risk of breast and ovarian cancer. Cancer 1996; 77: Venkitaraman AR. Functions of BRCA1 and BRCA2 in the biological response to DNA damage. Journal of cell science 2001; 114: Miki Y, Swensen J, Shattuck-Eidens D, Futreal PA, Harshman K, Tavtigian S, Liu Q, Cochran C, Bennett LM, Ding W, Bell R, Rosenthal J, Hussey C, Tran T, McClure M, Frye C, Hattier T, Phelps R, Haugen-Strano A, Katcher H, Yakumo K, Gholami Z, Shaffer D, Stone S, Bayer S, Wray C, Bogden R, Dayananth P, Ward J, Tonin P, Narod S, Bristow PK, Norris FH, Helvering L, Morrison P, Rosteck P, Lai M, Barrett JC, Lewis C, Neuhausen S, Cannon-Albright L, Goldgar D, Wiseman R, Kamb A, Skolnick MH. A strong candidate for the breast and ovarian cancer susceptibility gene BRCA1. Science 1994; 266: Deng CX. BRCA1: cell cycle checkpoint, genetic instability, DNA damage response and cancer evolution. Nucleic Acids Research 2006; 34: Chen Y, Farmer AA, Chen CF, Jones DC, Chen PL, Lee WH. BRCA1 is a 220-kDa nuclear phosphoprotein that is expressed and phosphorylated in a cell cycle-dependent manner. Cancer Research 1996; 56:

120 71 Cantor SB, Bell DW, Ganesan S, Kass EM, Drapkin R, Grossman S, Wahrer DC, Sgroi DC, Lane WS, Haber DA, Livingston DM. BACH1, a novel helicase-like protein, interacts directly with BRCA1 and contributes to its DNA repair function. Cell 2001; 105: Wang W. Emergence of a DNA-damage response network consisting of Fanconi anaemia and BRCA proteins. Nat Rev Genet 2007; 8(10): Kastan MB, Bartek J. Cell-cycle checkpoints and cancer. Nature 2004: 432(7015): Rajan JV, Wang M, Marquis ST, Chodosh LA. Brca2 is coordinately regulated with Brca1 during proliferation and differentiation in mammary epithelial cells. Proc Natl Acad Sci USA 1996; 93: Narod SA, Foulkes WD. BRCA1 and BRCA2: 1994 and beyond. Nat Rev Cancer 2004; 4(9): Lou Z, Chen J. BRCA proteins and DNA damage checkpoints. Front Biosci 2003; 1;8:s Gudmundsdottir K, Ashworth A. The roles of BRCA1 and BRCA2 and associated proteins in the maintenance of genomicstability. Oncogene 2006; 25(43): Turner NC, Reis-Filho JS. Basal-like breast cancer and the BRCA1 phenotype. Oncogene 2006; 25(43): Wooster R, Neuhausen SL, Mangion J, Quirk Y, Ford D, Collins N, Nguyen K, Seal S, Tran T, Averill D. Localization of a breast cancer susceptibility gene, BRCA2, to chromosome 13q Science 1994; 265(5181): Wooster R, Bignell G, Lancaster J, Swift S, Seal S, Mangion J, Collins N, Gregory S, Gumbs C, Micklem G. Identification of the breast cancer susceptibility gene BRCA2. Nature 1995; 378(6559): Tavtigian SV, Simard J, Rommens J, Couch F, Shattuck-Eidens D, Neuhausen S, Merajver S, Thorlacius S, Offit K, Stoppa-Lyonnet D, Belanger C, Bell R, Berry S, Bogden R, Chen Q, Davis T, Dumont M, Frye C, Hattier T, Jammulapati S, Janecki T, Jiang P, Kehrer R, Leblanc JF, Mitchell JT, McArthur-Morrison J, Nguyen K, Peng Y, Samson C, Schroeder M, Snyder SC, Steele L, Stringfellow M, Stroup C, Swedlund B, Swense J, Teng D, Thomas A, Tran T, Tranchant M, Weaver-Feldhaus J, Wong AK, Shizuya H, Eyfjord JE, Cannon-Albright L, Tranchant M, Labrie F, Skolnick MH, Weber B, Kamb A, Goldgar DE. The complete BRCA2 gene and mutations in chromosome 13q-linked kindreds. Nat Genet 1996; 12(3): Rajendra E, Venkitaraman AR. Two modules in the BRC repeats of BRCA2 mediate structural and functional interactions with the RAD51 recombinase. Nucleic Acids Res 2010; 38(1):

121 83 Bertwistle D, Swift S, Marston NJ, Jackson LE, Crossland S, Crompton MR, Marshall CJ, Ashworth A. Nuclear location and cell cycle regulation of the BRCA2 protein. Cancer Research 1997; 57: Shivji MK, Venkitaraman AR. DNA recombination, chromosomal stability and carcinogenesis: insights into the role of BRCA2. DNA Repair (Amst) 2004; 3(8-9): Xia B, Sheng Q, Nakanishi K, Ohashi A, Wu J, Christ N, Liu X, Jasin M, Couch FJ, Livingston DM. Control of BRCA2 cellular and clinical functions by a nuclear partner, PALB2. Mol Cell 2006; 22(6): Milner J, Ponder B, Hughes-Davies L, Seltmann M, Kouzarides T. Transcriptional activation functions in BRCA2. Nature 1997; 386(6627): Shin S, Verma IM. BRCA2 cooperates with histone acetyltransferases in androgen receptormediated transcription. Proc Natl Acad Sci U S A 2003; 100(12): Howlett NG, Taniguchi T, Olson S, Cox B, Waisfisz Q, De Die-Smulders C, Persky N, Grompe M, Joenje H, Pals G, Ikeda H, Fox EA, D'Andrea AD. Biallelic inactivation of BRCA2 in Fanconi anemia. Science 2002; 297(5581): Welcsh PL, Owens KN, King MC. Insights into the functions of BRCA1 and BRCA2. Trends in Genetics 2000; 16: Brodie SG and Deng CX. BRCA1-associated tumorigenesis: what have we learned from knockout mice? Trends in genetics 2001; 17:S Deng CX. Tumor formation in Brca1 conditional mutant mice. Environmental and Molecular Mutagenesis 2002; 39: Venkitaraman AR. Cancer susceptibility and the functions of BRCA1 and BRCA2. Cell 2002; 108(2): Welcsh PL and King MC. BRCA1 and BRCA2 and the genetics of breast and ovarian cancer. Human Molecular Genetics 2001; 10: 7: Ford D, Easton DF, Bishop DT, Narod SA, Goldgar DE. Risks of cancer in BRCA1- mutation carriers. Breast Cancer Linkage Consortium. Lancet 1994; 343: Allain DC, Sweet K, Agnese DM. Management options after prophylactic surgeries in women with BRCA mutations: a review. Cancer Control 2007; 14(4): Thompson D, Easton DF; Breast Cancer Linkage Consortium. Cancer incidence in BRCA1 mutation carriers. J Natl Cancer Inst 2002; 94: The Breast Cancer Linkage Consortium. Cancer risks in BRCA2 mutation carriers. J Natl Cancer Inst 1999; 91:

122 98 Risch HA, Narod SA. Re: On the use of familial aggregation in population-based case probands for calculating penetrance. J Natl Cancer Inst 2003; 95: Risch HA, Narod SA. Re: Cancer risks in BRCA1 carriers: time for the next generation of studies. J Natl Cancer Inst 2003; 95: Liede A, Karlan BY, Narod SA. Cancer risks for male carriers of germline mutations in BRCA1 or BRCA2: a review of the literature. J Clin Oncol 2004; 22: van Asperen CJ, Brohet RM, Meijers-Heijboer EJ, Hoogerbrugge N, Verhoef S, Vasen HF, Ausems MG, Menko FH, Gomez Garcia EB, Klijn JG, Hogervorst FB, van Houwelingen JC, van't Veer LJ, Rookus MA, van Leeuwen FE, Netherlands Collaborative Group on Hereditary Breast Cancer (HEBON). Cancer risks in BRCA2 families: estimates for sites other than breast and ovary. J Med Genet 2005; 42: Giordano SH. A review of the diagnosis and management of male breast cancer. Oncologist 2005; 10: BIC database: Ferla R, Calò V, Cascio S, Rinaldi G, Badalamenti G, Carreca I, Surmacz E, Colucci G, Bazan V, Russo A. Founder mutations in BRCA1 and BRCA2 genes. Ann Oncol 2007; 18 Suppl 6:vi Quintana-Murci L, Gal I, Bakhan T, Quach H, Sayar SH, Shiri-Sverdlov R, Baruch RG, McElreavey K, Dagan E, Narod S, Friedman E. The Tyr978X BRCA1 mutation: occurrence in non-jewish Iranians and haplotype in French-Canadian and non-ashkenazi Jews. Fam Cancer 2005; 4: Roa BB, Boyd AA, Volcik K, Richards CS. Ashkenazi Jewish population frequencies for common mutations in BRCA1 and BRCA2. Nat Genet 1996; 14: Thorlacius S, Olafsdottir G, Tryggvadottir L, Neuhausen S, Jonasson JG, Tavtigian SV, Tulinius H, Ogmundsdottir HM, Eyfjörd JE. A single BRCA2 mutation in male and female breast cancer families from Iceland with varied cancer phenotypes. Nat Genet 1996; 13: Møller P, Heimdal K, Apold J, Fredriksen A, Borg A, Hovig E, Hagen A, Hagen B, Pedersen JC, Maehle L, Norwegian Inherited Breast Cancer Group, Norwegian Inherited Ovarian Cancer Group. Genetic epidemiology of BRCA1 mutations in Norway. Eur J Cancer 2001; 37: Heimdal K, Maehle L, Apold J, Pedersen JC, Møller P. The Norwegian founder mutations in BRCA1: high penetrance confirmed in an incident cancer series and differences observed in the risk of ovarian cancer. Eur J Cancer 2003; 39: Rudkin TM, Hamel N, Galvez M, Hogervorst F, Gille JJ, Møller P, Apold J, Foulkes WD. The frequent BRCA1 mutation 1135insA has multiple origins: a haplotype study in different populations. BMC Med Genet 2006; 7:

123 111 Sarantaus L, Huusko P, Eerola H, Launonen V, Vehmanen P, Rapakko K, Gillanders E, Syrjäkoski K, Kainu T, Vahteristo P, Krahe R, Pääkkönen K, Hartikainen J, Blomqvist C, Löppönen T, Holli K, Ryynänen M, Bützow R, Borg A, Wasteson Arver B, Holmberg E, Mannermaa A, Kere J, Kallioniemi OP, Winqvist R, Nevanlinna H. Multiple founder effects and geographical clustering of BRCA1 and BRCA2 families in Finland. Eur J Hum Genet 2000; 8: Bergman A, Einbeigi Z, Olofsson U, Taib Z, Wallgren A, Karlsson P, Wahlström J, Martinsson T, Nordling M. The western Swedish BRCA1 founder mutation 3171ins5; a 3.7 cm conserved haplotype of today is a reminiscence of a 1500-year-old mutation. Eur J Hum Genet 2001; 9: Bergman A, Flodin A, Engwall Y, Arkblad EL, Berg K, Einbeigi Z, Martinsson T, Wahlström J, Karlsson P, Nordling M. A high frequency of germline BRCA1/2 mutations in western Sweden detected with complementary screening techniques. Fam Cancer 2005; 4: Einbeigi Z, Bergman A, Meis-Kindblom JM, Flodin A, Bjursell C, Martinsson T, Kindblom LG, Wahlström J, Wallgren A, Nordling M, Karlsson P. A founder mutation of the BRCA1 gene in western Sweden associated with a high incidence of breast and ovarian cancer. Eur J Cancer 2001; 37: Muller D, Bonaiti-Pellié C, Abecassis J, Stoppa-Lyonnet D, Fricker JP. BRCA1 testing in breast and/or ovarian cancer families from northeastern France identifies two common mutations with a founder effect. Fam Cancer 2004; 3: Hartmann C, John AL, Klaes R, Hofmann W, Bielen R, Koehler R, Janssen B, Bartram CR, Arnold N, Zschocke J. Large BRCA1 gene deletions are found in 3% of German high-risk breast cancer families. Hum Mutat 2004; 24: Zeegers MP, van Poppel F, Vlietinck R, Spruijt L, Ostrer H. Founder mutations among the Dutch. Eur J Hum Genet 2004; 12: Cipollini G, Tommasi S, Paradiso A, Aretini P, Bonatti F, Brunetti I, Bruno M, Lombardi G, Schittulli F, Sensi E, Tancredi M, Bevilacqua G, Caligo MA. Genetic alterations in hereditary breast cancer. Ann Oncol 2004; 15 (Suppl 1): I7 I Baudi F, Quaresima B, Grandinetti C, Cuda G, Faniello C, Tassone P, Barbieri V, Bisegna R, Ricevuto E, Conforti S, Viel A, Marchetti P, Ficorella C, Radice P, Costanzo F, Venuta S. Evidence of a founder mutation of BRCA1 in a highly homogeneous population from southern Italy with breast/ovarian cancer. Hum Mutat 2001; 18: Pisano M, Cossu A, Persico I, Palmieri G, Angius A, Casu G, Palomba G, Sarobba MG, Rocca PC, Dedola MF, Olmeo N, Pasca A, Budroni M, Marras V, Pisano A, Farris A, Massarelli G, Pirastu M, Tanda F. Identification of a founder BRCA2 mutation in Sardinia. Br J Cancer 2000; 82: Tonin PN, Mes-Masson AM, Futreal PA, Morgan K, Mahon M, Foulkes WD, Cole DE, Provencher D, Ghadirian P, Narod SA. Founder BRCA1 and BRCA2 mutations in French Canadian breast and ovarian cancer families. Am J Hum Genet 1998; 63:

124 122 Tonin PN, Mes-Masson AM, Narod SA, Ghadirian P, Provencher D. Founder BRCA 1 and BRCA 2 mutations in French Canadian ovarian cancer cases unselected for family history.clin Genet 1999; 55: Tonin PN, Perret C, Lambert JA, Paradis AJ, Kantemiroff T, Benoît MH, Martin G, Foulkes WD, Ghadirian P. Founder BRCA1 and BRCA2 mutations in early-onset French Canadian breast cancer cases unselected for family history. Int J Cancer 2001; 95: Oros KK, Leblanc G, Arcand SL, Shen Z, Perret C, Mes-Masson AM, Foulkes WD, Ghadirian P, Provencher D, Tonin PN. Haplotype analysis suggest common founders in carriers of the recurrent BRCA2 mutation, 3398delAAAAG, in French Canadian hereditary breast and/ovarian cancer families. BMC Med Genet 2006; 7: Vézina H, Durocher F, Dumont M, Houde L, Szabo C, Tranchant M, Chiquette J, Plante M, Laframboise R, Lépine J, Nevanlinna H, Stoppa-Lyonnet D, Goldgar D, Bridge P, Simard J. Molecular and genealogical characterization of the R1443X BRCA1 mutation in high-risk French-Canadian breast/ovarian cancer families. Hum Genet 2005; 117: Weitzel JN, Lagos V, Blazer KR, Nelson R, Ricker C, Herzog J, McGuire C, Neuhausen S. Prevalence of BRCA mutations and founder effect in high-risk Hispanic families. Cancer Epidemiol Biomarkers Prev 2005; 14: Torres D, Rashid MU, Gil F, Umana A, Ramelli G, Robledo JF, Tawil M, Torregrosa L, Briceno I, Hamann U. High proportion of BRCA1/2 founder mutations in Hispanic breast/ovarian cancer families from Colombia. Breast Cancer Res Treat 2007; 103(2): Pal T, Permuth-Wey J, Holtje T, Sutphen R. BRCA1 and BRCA2 mutations in a study of African American breast cancer patients. Cancer Epidemiol Biomarkers Prev 2004; 13: Reeves MD, Yawitch TM, van der Merwe NC, van den Berg HJ, Dreyer G, van Rensburg EJ. BRCA1 mutations in South African breast and/or ovarian cancer families: evidence of a novel founder mutation in Afrikaner families. Int J Cancer 2004; 110: Khoo US, Chan KY, Cheung AN, Xue WC, Shen DH, Fung KY, Ngan HY, Choy KW, Pang CP, Poon CS, Poon AY, Ozcelik H. Recurrent BRCA1 and BRCA2 germline mutations in ovarian cancer: a founder mutation of BRCA1 identified in the Chinese population. Hum Mutat 2002; 19: Ikeda N, Miyoshi Y, Yoneda K, Shiba E, Sekihara Y, Kinoshita M, Noguchi S. Frequency of BRCA1 and BRCA2 germline mutations in japanese breast cancer families. Int J Cancer 2001; 91: Sekine M, Nagata H, Tsuji S, Hirai Y, Fujimoto S, Hatae M, Kobayashi I, Fujii T, Nagata I, Ushijima K, Obata K, Suzuki M, Yoshinaga M, Umesaki N, Satoh S, Enomoto T, Motoyama S, Tanaka K. Japanese Familial Ovarian Cancer Study Group. Mutational analysis of BRCA1 and BRCA2 and clinicopathologic analysis of ovarian cancer in 82 ovarian cancer families: two common founder mutations of BRCA1 in Japanese population. Clin Cancer Res 2001; 7:

125 133 Lee AS, Ho GH, Oh PC, Balram C, Ooi LL, Lim DT, Wong CY, Hong GS. Founder mutation in the BRCA1 gene in Malay breast cancer patients from Singapore. Hum Mutat 2003; 22(2): De Leon Matsuda ML, Liede A, Kwan E, Mapua CA, Cutiongco EM, Tan A, Borg A, Narod SA. BRCA1 and BRCA2 mutations among breast cancer patients from the Philippines. Int J Cancer 2002; 98: Rashid MU, Zaidi A, Torres D, Sultan F, Benner A, Naqvi B, Shakoori AR, Seidel-Renkert A, Farooq H, Narod S, Amin A, Hamann U. Prevalence of BRCA1 and BRCA2 mutations in Pakistani breast and ovarian cancer patients. Int J Cancer 2006; 119: Levy-Lahad E, Catane R, Eisenberg S, Kaufman B, Hornreich G, Lishinsky E, Shohat M, Weber BL, Beller U, Lahad A, Halle D. Founder BRCA1 and BRCA2 mutations in Ashkenazi Jews in Israel: frequency and differential penetrance in ovarian cancer and in breast-ovarian cancer families. Am J Hum Genet 1997; 60: Struewing JP, Abeliovich D, Peretz T, Avishai N, Kaback MM, Collins FS, Brody LC. The carrier frequency of the BRCA1 185delAG mutation is approximately 1 percent in Ashkenazi Jewish individuals. Nat Genet 1995; 11(2): Oddoux C, Struewing JP, Clayton CM, Neuhausen S, Brody LC, Kaback M, Haas B, Norton L, Borgen P, Jhanwar S, Goldgar D, Ostrer H, Offit K. The carrier frequency of the BRCA2 6174delT mutation among Ashkenazi Jewish individuals is approximately 1%. Nat Genet 1996; 14(2): Shiri Sverdlov R, Oefner P, Green L, Gershoni Baruch R, Wagner T, Kruglikova A, Haitchick S, Hofstra RMW, Papa MZ, Mulder I, Risel S, Bruchim Bar Sade R, Dagan E, Abdeen Z, Goldman B, Fridman E. Mutational Analyses of BRCA1 and BRCA2 in Ashkenazi and Non-Ashkenazi Jewish Women With Familial Beast and Ovarian Cancers. Human Mutation 2000; 16: Moslehi R, Chu W, Karlan B, Fishman D, Risch H, Fields A, Smotkin D, Ben-David Y, Rosenblatt J, Russo D, Schwartz P, Tung N, Warner E, Rosen B, Friedman J, Brunet JS, Narod SA. BRCA1 and BRCA2 mutation analysis of 208 Ashkenazi Jewish women with ovarian cancer. Am J Hum Genet 2000; 66(4): Phelan CM, Kwan E, Jack E, Li S, Morgan C, Aubé J, Hanna D, Narod SA. A low frequency of non-founder BRCA1 mutations in Ashkenazi Jewish breast-ovarian cancer families. Hum Mutat 2002; 20(5): Nanda R, Schumm LP, Cummings S, Fackenthal JD, Sveen L, Ademuyiwa F, Cobleigh M, Esserman L, Lindor NM, Neuhausen SL, Olopade OI. Genetic testing in an ethnically diverse cohort of high-risk women: a comparative analysis of BRCA1 and BRCA2 mutations in American families of European and African ancestry. JAMA 2005; 294(15): Palma MD, Domchek SM, Stopfer J, Erlichman J, Siegfried JD, Tigges-Cardwell J, Mason BA, Rebbeck TR, Nathanson KL. The relative contribution of point mutations and genomic 109

126 rearrangements in BRCA1 and BRCA2 in high-risk breast cancer families. Cancer Res 2008; 68(17): Steele RJC, Thompson PAH, Lane DP. The p53 tumor suppressor gene. Br J Surg 1998; 85: Matsuoka S, Huang M, Elledge S.J. Linkage of ATM to cell cycle regulation by the Chk2 protein kinase. Science 1998; 282: Chaturvedi P, Eng WK, Zhu Y, Mattern MR, Mishra R, Hurle MR, Zhang X, Annan RS, Lu Q, Faucette LF, Scott GF, Li X, Carr SA, Johnson RK, Winkler JD, Zhou BB. Mammalian Chk2 is a downstream effector of the ATM-dependent DNA damage checkpoint pathway. Oncogene 1999; 18: Meijers-Heijboer H, van den Ouweland A, Klijn J, Wasielewski M, de Snoo A, Oldenburg R, Hollestelle A, Houben M, Crepin E, van Veghel-Plandsoen M, Elstrodt F, van Duijn C, Bartels C, Meijers C, Schutte M, McGuffog L, Thompson D, Easton D, Sodha N, Seal S, Barfoot R, Mangion J, Chang-Claude J, Eccles D, Eeles R, Evans DG, Houlston R, Murday V, Narod S, Peretz T, Peto J, Phelan C, Zhang HX, Szabo C, Devilee P, Goldgar D, Futreal PA, Nathanson KL, Weber B, Rahman N, Stratton MR, CHEK2-Breast Cancer Consortium. Low- penetrance susceptibility to breast cancer due to CHEK2*1100delC in noncarriers of BRCA1 or BRCA2 mutations. Nat Gen 2002; 31: Wirtzfeld DA, Petrelli NJ, Rodriguez-Bigas MA. Hamartomatous polyposis syndromes: molecular genetics, neoplastic risk, and surveillance recommendations. Ann Surg Oncol 2001; 8(4): Ripperger T, Gadzicki D, Meindl A, Schlegelberger B. Breast cancer susceptibility: current knowledge and implications for genetic counselling. Eur J Hum Genet 2009; 17(6): Sekine M, Negata H, Tsuji S, Hirai Y, Fujimoto S, Hatae M, Kobayashi I, Fujii T, Negata I, Ushijima K, Obata K, Suzuki M, Yoshinaga M, Umesaki N, Satoh S, Enomoto T, Motoyama S, Tanaka K, The Japanese Familial Ovarian Cancer Study Group. Localization of a novel susceptibility gene for familial ovarian cancer to chromosome 3p22-p25. Human Molecular Genetics 2001; 10(13): Ahmed S, Thomas G, Ghoussaini M, Healey CS, Humphreys MK, Platte R, Morrison J, Maranian M, Pooley KA, Luben R, Eccles D, Evans DG, Fletcher O, Johnson N, dos Santos Silva I, Peto J, Stratton MR, Rahman N, Jacobs K, Prentice R, Anderson GL, Rajkovic A, Curb JD, Ziegler RG, Berg CD, Buys SS, McCarty CA, Feigelson HS, Calle EE, Thun MJ, Diver WR, Bojesen S, Nordestgaard BG, Flyger H, Dörk T, Schürmann P, Hillemanns P, Karstens JH, Bogdanova NV, Antonenkova NN, Zalutsky IV, Bermisheva M, Fedorova S, Khusnutdinova E; SEARCH, Kang D, Yoo KY, Noh DY, Ahn SH, Devilee P, van Asperen CJ, Tollenaar RA, Seynaeve C, Garcia-Closas M, Lissowska J, Brinton L, Peplonska B, Nevanlinna H, Heikkinen T, Aittomäki K, Blomqvist C, Hopper JL, Southey MC, Smith L, Spurdle AB, Schmidt MK, Broeks A, van Hien RR, Cornelissen S, Milne RL, Ribas G, González-Neira A, Benitez J, Schmutzler RK, Burwinkel B, Bartram CR, Meindl A, Brauch H, Justenhoven C, Hamann U; GENICA Consortium, Chang-Claude J, Hein R, Wang-Gohrke S, Lindblom A, Margolin S, Mannermaa A, Kosma VM, Kataja V, Olson JE, Wang X, 110

127 Fredericksen Z, Giles GG, Severi G, Baglietto L, English DR, Hankinson SE, Cox DG, Kraft P, Vatten LJ, Hveem K, Kumle M, Sigurdson A, Doody M, Bhatti P, Alexander BH, Hooning MJ, van den Ouweland AM, Oldenburg RA, Schutte M, Hall P, Czene K, Liu J, Li Y, Cox A, Elliott G, Brock I, Reed MW, Shen CY, Yu JC, Hsu GC, Chen ST, Anton-Culver H, Ziogas A, Andrulis IL, Knight JA; kconfab; Australian Ovarian Cancer Study Group, Beesley J, Goode EL, Couch F, Chenevix-Trench G, Hoover RN, Ponder BA, Hunter DJ, Pharoah PD, Dunning AM, Chanock SJ, Easton DF. Newly discovered breast cancer susceptibility loci on 3p24 and 17q23.2. Nat Genet 2009; 41(5): Kainu T, Juo SH, Desper R, Schaffer AA, Gillanders E, Rozenblum E, Freas-Lutz D, Weaver D, Stephan D, Baily-Wilson J, Kallioniemi OP, Tirkkonen M, Syrjakoski K, Kuukasjarvi T, Koivisto P, Karhu R, Holli K, Arason A, Johannesdottir G, Bergthorsson JT, Johannsdottir H, Egilsson V, Barkardottir RB, Johannsson O, Haraldsson K, Sandberg T, Holmberg E, Gronberg H, Olsson H, Borg A, Vehmanen P, Heikkila P, Pyrhonen S, Nevanlinna H. Somatic deletions in hereditary breast cancers implicate 13q21 as a putative novel breast cancer susceptibility locus. Medical Sciences 2000; 97(17): Seitz S, Rohde K, Bender E, Nothnagel A, Kolble K, Schlag PM, Scherneck S. Strong indication for a breast cancer susceptibility gene on chromosome 8p12-p22: Linkage analysis German breast cancer families. Oncogene 1997; 14: Easton DF, Pooley KA, Dunning AM, Pharoah PD, Thompson D, Ballinger DG, Struewing JP, Morrison J, Field H, Luben R, Wareham N, Ahmed S, Healey CS, Bowman R; SEARCH collaborators, Meyer KB, Haiman CA, Kolonel LK, Henderson BE, Le Marchand L, Brennan P, Sangrajrang S, Gaborieau V, Odefrey F, Shen CY, Wu PE, Wang HC, Eccles D, Evans DG, Peto J, Fletcher O, Johnson N, Seal S, Stratton MR, Rahman N, Chenevix-Trench G, Bojesen SE, Nordestgaard BG, Axelsson CK, Garcia-Closas M, Brinton L, Chanock S, Lissowska J, Peplonska B, Nevanlinna H, Fagerholm R, Eerola H, Kang D, Yoo KY, Noh DY, Ahn SH, Hunter DJ, Hankinson SE, Cox DG, Hall P, Wedren S, Liu J, Low YL, Bogdanova N, Schürmann P, Dörk T, Tollenaar RA, Jacobi CE, Devilee P, Klijn JG, Sigurdson AJ, Doody MM, Alexander BH, Zhang J, Cox A, Brock IW, MacPherson G, Reed MW, Couch FJ, Goode EL, Olson JE, Meijers-Heijboer H, van den Ouweland A, Uitterlinden A, Rivadeneira F, Milne RL, Ribas G, Gonzalez-Neira A, Benitez J, Hopper JL, McCredie M, Southey M, Giles GG, Schroen C, Justenhoven C, Brauch H, Hamann U, Ko YD, Spurdle AB, Beesley J, Chen X; kconfab; AOCS Management Group, Mannermaa A, Kosma VM, Kataja V, Hartikainen J, Day NE, Cox DR, Ponder BA. Genome-wide association study identifies novel breast cancer susceptibility loci. Nature 2007; 447(7148): Hunter DJ, Kraft P, Jacobs KB, Cox DG, Yeager M, Hankinson SE, Wacholder S, Wang Z, Welch R, Hutchinson A, Wang J, Yu K, Chatterjee N, Orr N, Willett WC, Colditz GA, Ziegler RG, Berg CD, Buys SS, McCarty CA, Feigelson HS, Calle EE, Thun MJ, Hayes RB, Tucker M, Gerhard DS, Fraumeni JF Jr, Hoover RN, Thomas G, Chanock SJ. A genome-wide association study identifies alleles in FGFR2 associated with risk of sporadic postmenopausal breast cancer. Nat Genet 2007; 39: Schouten JP, McElgunn CJ, Waaijer R, Zwijnenburg D, Diepvens F, Pals G. Relative quantification of 40 nucleic acid sequences by multiplex ligation-dependent probe amplification. Nucleic Acids Res 2002; 15:30(12):e

128 157 Woodward AM, Davis TA, Silva AG, Kirk JA, Leary JA, kconfab Investigators. Large genomic rearrangements of both BRCA2 and BRCA1 are a feature of the inherited breast/ovarian cancer phenotype in selected families. J Med Genet 2005; 42(5):e Gad S, Caux-Moncoutier V, Pagès-Berhouet S, Gauthier-Villars M, Coupier I, Pujol P, Frénay M, Gilbert B, Maugard C, Bignon YJ, Chevrier A, Rossi A, Fricker JP, Nguyen TD, Demange L, Aurias A, Bensimon A, Stoppa-Lyonnet D. Significant contribution of large BRCA1 gene rearrangements in 120 French breast and ovarian cancer families. Oncogene 2002; 21(44): Montagna M, Dalla Palma M, Menin C, Agata S, De Nicolo A, Chieco-Bianchi L, D'Andrea E. Genomic rearrangements account for more than one-third of the BRCA1 mutations in northern Italian breast/ovarian cancer families. Hum Mol Genet 2003; 12(9): Belogianni I, Apessos A, Mihalatos M, Razi E, Labropoulos S, Petounis A, Gaki V, Keramopoulos A, Pandis N, Kyriacou K, Hadjisavvas A, Kosmidis P, Yannoukakos D, Nasioulas G. Characterization of a novel large deletion and single point mutations in the BRCA1 gene in a Greek cohort of families with suspected hereditary breast cancer. BMC Cancer 2004; 4: Agata S, Dalla Palma M, Callegaro M, Scaini MC, Menin C, Ghiotto C, Nicoletto O, Zavagno G, Chieco-Bianchi L, D Andrea E, Montagna M. Large genomic deletions inactivate the BRCA2 gene in breast cancer families. J Med Genet 2005; 42(10):e Preisler-Adams S, Schönbuchner I, Fiebig B, Welling B, Dworniczak B, Weber BH. Gross rearrangements in BRCA1 but not BRCA2 play a notable role in predisposition to breast and ovarian cancer in high-risk families of German origin. Cancer Genet Cytogenet 2006; 168(1): de la Hoya M, Gutiérrez-Enríquez S, Velasco E, Osorio A, Sanchez de Abajo A, Vega A, Salazar R, Esteban E, Llort G, Gonzalez-Sarmiento R, Carracedo A, Benítez J, Miner C, Díez O, Díaz-Rubio E, Caldes T. Genomic rearrangements at the BRCA1 locus in Spanish families with breast/ovarian cancer. Clin Chem 2006; 52(8): Peixoto A, Salgueiro N, Santos C, Varzim G, Rocha P, Soares MJ, Pereira D, Rodrigues H, Bento MJ, Fráguas A, Moura G, Regateiro F, Castedo S, Teixeira MR. BRCA1 and BRCA2 germline mutational spectrum and evidence for genetic anticipation in Portuguese breast/ovarian cancer families. Fam Cancer 2006; 5(4): Armaou S, Konstantopoulou I, Anagnostopoulos T, Razis E, Boukovinas I, Xenidis N, Fountzilas G, Yannoukakos D. Novel genomic rearrangements in the BRCA1 gene detected in Greek breast/ovarian cancer patients. Eur J Cancer 2007; 43(2): Buffone A, Capalbo C, Ricevuto E, Sidoni T, Ottini L, Falchetti M, Cortesi E, Marchetti P, Scambia G, Tomao S, Rinaldi C, Zani M, Ferraro S, Frati L, Screpanti I, Gulino A, Giannini G. Prevalence of BRCA1 and BRCA2 genomic rearrangements in a cohort of consecutive Italian breast and/or ovarian cancer families. Breast Cancer Res Treat 2007; 106(2):

129 167 Lim YK, Lau PT, Ali AB, Lee SC, Wong JE, Putti TC, Sng JH. Identification of novel BRCA large genomic rearrangements in Singapore Asian breast and ovarian patients with cancer. Clin Genet. 2007; 71(4): Zikan M, Pohlreich P, Stribrna J, Kleibl Z, Cibula D. Novel complex genomic rearrangement of the BRCA1 gene. Mutat Res 2008; 637(1-2): Suela PS, Cardeñosa EE, González BE, Soler OS, de Juan Jiménez I, González CI, Huerta SA, Ponce GC, de Dueñas ME, Gilabert BP, Group for Assessment of Hereditary Cancer of Valencia Community. Identification of a novel BRCA1 large genomic rearrangement in a Spanish breast/ovarian cancer family. Breast Cancer Res Treat 2008; 112(1): Ratajska M, Brozek I, Senkus-Konefka E, Jassem J, Stepnowska M, Palomba G, Pisano M, Casula M, Palmieri G, Borg A, Limon J. BRCA1 and BRCA2 point mutations and large rearrangements in breast and ovarian cancer families in Northern Poland. Oncol Rep 2008; 19(1): Petrij-Bosch A, Peelan T, van Vliet M, van Eijk R, Olmer R, Drusedau M, Hogervorst FB, Hageman S, Arts PJ, Ligtenberg MJ, Meijers Heijboer H, Klijn JG, Vasen HF, Cornelisse CJ, van't Veer LJ, Bakker E, van Ommen GJ and Devilee P. BRCA1 Genomic Deletions Are Major Founder Mutations in Dutch Breast Cancer Patients. Nat Genet 1997; 17(3), pp: Puget N, Torchard D, Serova-Silinkova OM, Lynch HT, Feunteun J, Lenoir GM and Mazoyer S. A 1-kb Alu-Mediated Germ-Line Deletion Removing BRCA1 exon 17. Cancer Res 1997; 57(5), pp: Puget N, Stoppa-Lyonnet D, Sinilnikova OM, Pages S, Lynch HT, Lenoir GM, Mazoyer S. Screening for Germ-Line Rearrangements and Regulatory Mutation in BRCA1 Led to Identification of Four New Deletion. Cancer Res 1999; 59: pp Rohlfs EM, Chung CH, Yang Q, Skrzynia C, Grody WW, Graham ML, Silverman LM. In- Frame Deletions of BRCA1 May Define Critical Functional Domain. Hum Genet 2000; 107, pp: Payne SR, Newman B, King MC. Complex Germline Rearrangement of BRCA1 Associated With Breast and Ovarian Cancer. Genes Chromosomes Cancer 2002; 29(1), pp: Unger MA, Nathanson KL, Calzone K, Antin-Ozerkis D, Shih HA, Martin AM, Lenoir GM, Mazoyer S, Weber BL. Screening for genomic rearrangements in families with breast and ovarian cancer identifies BRCA1 mutations previously missed by conformation-sensitive gel electrophoresis or sequencing. Am J Hum Genet 2000;67(4), pp: Hogervorst FBL, Nederlof PM, Gille JJP, McElgunn CJ, Grippeling M, Pruntel R, Regnerus R, van Welsem T, van Spaendonk R, Menko FH, Kluijt I, Dommering C, Verhoef S, Schouten JP, van't Veer LJ, Pals G. Large Genomic Deletion and Duplication in the BRCA1 Gene Identified by a Novel Quantification Method. Cancer Res 2003; 63, pp: Hofmann W, Gorgens H, John A, Horn D, Huttner C,Arnold N, Scherneck S, Schackert HK. Screening for Large Rearrangements of the BRCA1 Gene in German Breast or Ovarian 113

130 Cancer Families Using Semi-Quantitative Multiplex PCR Method. Hum Mutat 2003; 22(1): Montagna M, Dalla Palma M, Menin C, Agata S, De Nicolo A, Chieco-Bianchi L, D'Andrea E. Genomia Rearrangement Account For More Than One Third of BRCA1 Mutations in Northern Italian Breast/Ovarian Cancer Families. Hum Mol Genet 2003; 12(9), pp: Bunyan DJ, Eccles DM, Sillibourne J, Wilkins E, Simon Thomas N, Shea-Simonds J, Duncan JP, Curtis CE, Robinson DO, Harvey JF, Cross NCP. Dosage Analysis of Cancer Predisposition Genes by Multiplex Ligation-Dependent Probe Amplification. BJC 2004; 91, pp: Barrios M, Bièche I, Mazoyer S, Champème MH, Bressac de Pailleret B, Lidereau R. Realtime PCR-based Gene Dosage Assay for Detection of BRCA1 Rearrangements in Breast- Ovarian Cancer Families. Clin Genet 2004; 65, pp: Belongianni I, Apessos A, Mihalatos M, Razi E, Labropoulos S, Petounis A, Gaki V, Keramopoulos A, Pandis N, Kyriacou K, Hadjisavvas A, Kosmidis P, Yannoukakos D, Nasioulas G. Characterization of Novel Large Deletion and Single Point Mutations in the BRCA1 Gene in Greek Cohort of Families With Suspected Hereditary Breast Cancer. BMC cancer 2004; 4, pp: Tancredi M, Sensi E, Cipollini G, Aretini P, Lombardi G, Di Cristofano C, Presciuttini S, Bevilacqua G, Caligo MA. Haplotype analysis of BRCA1 gene reveals a new gene rearrangement: characterization of a 19.9 KBP deletion. Eur J Hum Genet 2004; 12(9): Nordling M, Karlsson P, Wahlstrom J, Engwall A, Martinsson T. A Large Deletion Disrupts the exon 3 Transcription Activation Domain of the BRCA2 Gene in a Breast/Ovarian Cancer Family. Cancer Res 1998; 58(7), pp: Wang T, Lerer I, Gueta Z, Sagi M, Kadouri L, Peretz T, Abeliovich D. A Deletion/Insertion Mutation in the BRCA2 Gene in a Breast Cancer Family: A Possible Role of Alu-polyA Tail in the Evaluation of the Deletion. Genes Chromosomes Cancer 2001; 31(1), pp: Tournier I, Paillerets BB, Sobol H, Stoppa-Lyonnet D, Lidereau R, Barrois M, Mazoyer S, Coulet F, Hardouin A, Chompret A, Lortholary A, Chappuis P, Bourdon V, Bonadona V, Maugard C, Gilbert B, Nogues C Frebourg T, Tosi M. Significant Contribution of Germline BRCA2 Rearrangements in Male Breast Cancer Families. Cancer Res 2004; 64(22), pp: Peelen T, van Vliet M, Bosch A, Bignell G, Vasen HFA, Klijn JGM, Meijers-Heijboer H, Staratton M, van Ommen G-JB, Cornelisse CJ, Devilee P. Screening for BRCA2 Mutations in 81 Dutch Breast-Ovarian Cancer Families. BJC 2000; 82(1), pp: Lahti-Domenici J, Rapakko K, Pääkkönen K, Allinen M, Nevanlinna H, Kujala M, Huusko P, Winqvist R. Exclusion of Large Deletion and Other Rearrangements in BRCA1 and BRCA2 in Finnish Breast and Ovarian Cancer Families. Cancer Genet Cytogenet 2001; 129, pp:

131 189 Ramus SJ, Harrington PA, Pye C, DiCioccio RA, Cox MJ, Garlinghouse-Jones K, Oakley- Girvan I, Jacobs IJ, Hardy RM, Whittemore AS, Ponder BA, Piver MS, Pharoah PD, Gayther SA. Contribution of BRCA1 and BRCA2 mutations to inherited ovarian cancer. Hum Mutat 2007; 28(12): Gutiérrez-Enríquez S, de la Hoya M, Martínez-Bouzas C, Sanchez de Abajo A, Ramón y Cajal T, Llort G, Blanco I, Beristain E, Díaz-Rubio E, Alonso C, Tejada MI, Caldés T, Diez O. Screening for large rearrangements of the BRCA2 gene in Spanish families with breast/ovarian cancer. Breast Cancer Res Treat 2007; 103(1): Smith TM, Lee MK, Szabo CI, Jerome N, McEuen M, Taylor M, Hood L, King MC. Complete genomic sequence and analysis of 117 kb of human DNA containing the gene BRCA1. Genome Res 1996; 6: Puget N, Gad S, Perrin-Vidoz L, Sinilnikova OM, Stoppa-Lyonnet D, Lenoir GM, Mazoyer S. Distinct BRCA1 rearrangements involving the BRCA1 pseudogene suggest the existence of a recombination hot spot. Am J Hum Genet 2002; 70: Marth GT, Korf I, Yandell MD, Yeh RT, Gu Z, Zakeri H, Stitziel NO, Hillier L, Kwok PY, Gish WR. A general approach to single-nucleotide polymorphism discovery. Nat Genet 1999;23(4): Doris PA. Hypertension genetics, single nucleotide polymorphisms, and the common disease:common variant hypothesis. Hypertension 2002; 39(2 Pt 2): Shastry BS. SNP alleles in human disease and evolution. J Hum Genet 2002; 47(11): Akey J, Jin L, Xiong M. Haplotypes vs single marker linkage disequilibrium tests: what do we gain? Eur J Hum Genet 2001; 9: Stephens JC, Schneider JA, Tanguay DA, Choi J, Acharya T, Stanley SE, Jiang R, Messer CJ, Chew A, Han JH, Duan J, Carr JL, Lee MS, Koshy B, Kumar AM, Zhang G, Newell WR, Windemuth A, Xu C, Kalbfleisch TS, Shaner SL, Arnold K, Schulz V, Drysdale CM, Nandabalan K, Judson RS, Ruano G, Vovis GF. Haplotype variation and linkage disequilibrium in 313 human genes. Science 2001; 293: Abecasis GR, Noguchi E, Heinzmann A, Traherne JA, Bhattacharyya S, Leaves NI, Anderson GG, Zhang Y, Lench NJ, Carey A, Cardon LR, Moffatt MF, Cookson WO. Extent and distribution of linkage disequilibrium in 3 genomic regions. Am J Hum Genet 2001; 68: Miller RD, Kwok PY. The birth and death of human single-nucleotide polymorphisms: new experimental evidence and implications for human history and medicine. Hum Mol Genet 2001; 10(20): Andolfatto P, Nordborg M. The effect of gene conversion on intralocus associations. Genetics 1998; 148:

132 201 Ardlie K, Liu-Cordero SN, Eberle MA, Daly M, Barrett J, Winchester E, Lander ES, Kruglyak L. Lower-than-expected linkage disequilibrium between tightly linked markers in humans suggests a role for gene conversion. Am J Hum Genet 2001; 69: Keavney B, McKenzie C, Parish S, Palmer A, Clark S, Youngman L, Delepine M, Lathrop M, Peto R, Collins R. Large scale test of hypothesized associations between the angiotensinconverting enzyme insertion/deletion polymorphisms and myocardial infarction in about 5000 cases and 6000 controls. Lancet 2000; 355: Spielman RS, McGinnis RE, Ewens WJ. Transmission distortion test for linkage: the insulin gene and susceptibility to insulin dependent diabetes mellitus (IDMM). Am J Hum Genet 1992; 52: Schork NJ, Fallin D, Lanchbury JS. Single nucleotide polymorphism and the future of genetic epidemiology. Clin Genet 2000; 58: Chakravarti A. Population genetics: making sense out of sequence. Nat Genet 1999; 21: Weiss KM, Terwilliger JD. How many diseases does it take to map a gene with SNPs? Nat Genet 2000; 26: Gold B, Kirchhoff T, Stefanov S, Lautenberger J, Viale A, Garber J, Friedman E, Narod S, Olshen AB, Gregersen P, Kosarin K, Olsh A, Bergeron J, Ellis NA, Klein RJ, Clark AG, Norton L, Dean M, Boyd J, Offit K. Genome-wide association study provides evidence for a breast cancer risk locus at 6q Proc Natl Acad Sci U S A 2008; 105(11): Motulsky AG. Jewish diseases and origins. Nat Genet 1995; 9(2): Olshen AB, Gold B, Lohmueller KE, Struewing JP, Satagopan J, Stefanov SA, Eskin E, Kirchhoff T, Lautenberger JA, Klein RJ, Friedman E, Norton L, Ellis NA, Viale A, Lee CS, Borgen PI, Clark AG, Offit K, Boyd J. Analysis of genetic variation in Ashkenazi Jews by high density SNP genotyping. BMC Genet 2008; 9: Claus EB. Risk models used to counsel women for breast and ovarian cancer: a guide for clinicians. Fam Cancer 2001; 1(3-4): Loman N, Johannsson O, kristoffersson U, Olsson H, Borg A. Family history of breast and ovarian cancers and BRCA1 and BRCA2 mutations in population-based series of early-onset breast cancer. J Natl Cancer Inst 2001; 93(16): Hanson H, Hodgson S. Cancer genetics and reproduction. Best Pract Res Clin Obstet Gynaecol 2010; 24(1): Tavtigian SV, Deffenbaugh AM, Yin L, Judkins T, Scholl T, Samollow PB, de Silva D, Zharkikh A, Thomas A. Comprehensive statistical study of 452 BRCA1 missense substitutions with classification of eight recurrent substitutions as neutral. J Med Genet 2006; 43:

133 214 Moisan AM, Fortin J, Dumont M, Samson C, Bessette P, Chiquette J, Laframboise R, Lépine J, Lespérance B, Pichette R, Plante M, Provencher L, Voyer P, Goldgar D, Bridge P, Simard J. No Evidence of BRCA1/2 genomic rearrangements in high-risk French-Canadian breast/ovarian cancer families. Genet Test 2006; 10(2): Puget N, Torchard D, Serova-Sinilnikova OM. A 1-kb Alu-mediated germline deletion removing BRCA1 exon 17. Cancer res 1997: 57(5) pp: Stankiewicz P, Lupski JR. The genomic basis of disease, mechanisms and assays for genomic disorders. Genome Dyn 2006; 1: Vasickova P, Machackova E, Lukesova M, Damborsky J, Horky O, Pavlu H, Kuklova J, Kosinova V, Navratilova M, Foretova L. High occurrence of BRCA1 intragenic rearrangements in hereditary breast and ovarian cancer syndrome in the Czech Republic. BMC Med Genet 2007; 11:8: Potocki L, Chen KS, Park SS, Osterholm DE, Withers MA, Kimonis V, Summers AM, Meschino WS, Anyane-Yeboa K, Kashork CD, Shaffer LG, Lupski JR. Molecular mechanism for duplication 17p11.2- the homologous recombination reciprocal of the Smith-Magenis microdeletion. Nat Genet 2000; 24(1): Kozlowski P, Jasinska AJ, Kwiatkowski DJ. New applications and developments in the use of multiplex ligation-dependent probe amplification. Electrophoresis 2008; 29(23): Hansen TV, Jønson L, Albrechtsen A, Andersen MK, Ejlertsen B, Nielsen FC. Large BRCA1 and BRCA2 genomic rearrangements in Danish high risk breast-ovarian cancer families. Breast Cancer Res Treat 2009; 115(2): Pylkäs K, Erkko H, Nikkilä J, Sólyom S, Winqvist R. Analysis of large deletions in BRCA1, BRCA2 and PALB2 genes in Finnish breast and ovarian cancer families. BMC Cancer 2008; 8: Engert S, Wappenschmidt B, Betz B, Kast K, Kutsche M, Hellebrand H, Goecke TO, Kiechle M, Niederacher D, Schmutzler RK, Meindl A. MLPA screening in the BRCA1 gene from 1,506 German hereditary breast cancer cases: novel deletions, frequent involvement of exon 17, and occurrence in single early-onset cases. Hum Mutat 2008; 29(7): Kauff ND, Perez-Segura P, Robson ME, Scheuer L, Siegel B, Schluger A, Rapaport B, Frank TS, Nafa K, Ellis NA, Parmigiani G, Offit K. Incidence of non-founder BRCA1 and BRCA2 mutations in high risk Ashkenazi breast and ovarian cancer families. Med Genet 2002; 39(8):

134 228 Shiri-Sverdlov R, Oefner P, Green L, Baruch RG, Wagner T, Kruglikova A, Haitchick S, Hofstra RM, Papa MZ, Mulder I, Rizel S, Bar Sade RB, Dagan E, Abdeen Z, Goldman B, Friedman E. Mutational analyses of BRCA1 and BRCA2 in Ashkenazi and non-ashkenazi Jewish women with familial breast and ovarian cancer. Hum Mutat. 2000;16(6): Benjamini Y, Yekutieli D. The control of the false discovery rate in multiple testing under dependency. Ann. Statist 2001; 29: Durocher F, Shattuck-Eidens D, McClure M, Labrie F, Skolnick MH, Goldgar DE, Simard J. Comparison of BRCA1 polymorphisms, rare sequence variants and/or missense mutations in unaffected and breast/ovarian cancer populations. Hum Mol Genet 1996; 5(6): Goldgar DE, Easton DF, Deffenbaugh AM, Monteiro AN, Tavtigian SV, Couch FJ, Breast Cancer Information Core (BIC) Steering Committee. Integrated evaluation of DNA sequence variants of unknown clinical significance: application to BRCA1 and BRCA2. Am J Hum Genet 2004; 75: Judkins T, Hendrickson BC, Deffenbaugh AM, Eliason K, Leclair B, Norton MJ, Ward BE, Pruss D, Scholl T. Application of embryonic lethal or other obvious phenotypes to characterize the clinical significance of genetic variants found in trans with known deleterious mutations. Cancer Res 2005; 65: Troudi W, Uhrhammer N, Romdhane KB, Sibille C, Amor MB, Khodjet El Khil H, Jalabert T, Mahfoudh W, Chouchane L, Ayed FB, Bignon YJ, Elgaaied AB. Complete mutation screening and haplotype characterization of BRCA1 gene in Tunisian patients with familial breast cancer. Cancer Biomark 2008; 4(1): Seo JH, Cho DY, Ahn SH, Yoon KS, Kang CS, Cho HM, Lee HS, Choe JJ, Choi CW, Kim BS, Shin SW, Kim YH, Kim JS, Son GS, Lee JB, Koo BH. BRCA1 and BRCA2 germline mutations in Korean patients with sporadic breast cancer. Hum Mutat 2004; 24(4): Choi DH, Lee MH, Bale AE, Carter D, Haffty BG. Incidence of BRCA1 and BRCA2 mutations in young Korean breast cancer patients. J Clin Oncol 2004; 22(9): Malander S, Ridderheim M, Måsbäck A, Loman N, Kristoffersson U, Olsson H, Nilbert M, Borg A. One in 10 ovarian cancer patients carry germ line BRCA1 or BRCA2 mutations: results of a prospective study in Southern Sweden. Eur J Cancer 2004; 40(3): Martínez-Ferrandis JI, Vega A, Chirivella I, Marín-García P, Insa A, Lluch A, Carracedo A, Chaves FJ, García-Conde J, Cervantes A, Armengod ME. Mutational analysis of BRCA1 and BRCA2 in Mediterranean Spanish women with early-onset breast cancer: identification of three novel pathogenic mutations. Hum Mutat 2003; 22(5): Wang T, Lerer I, Gueta Z, Sagi M, Kadouri L, Peretz T, Abeliovich D. A Deletion/Insertion Mutation in the BRCA2 gene in a Breast Cancer Family: A Possiable Role of the Alu-polyA Tail in the Evolution of the Deletion. Genes Chromosomes cancer 2005; 31: Karhu R, Laurila E, Kallioniemi A, Syrjäkoski K. Large genomic BRCA2 rearrangements and male breast cancer. Cancer Detect Prev 2006; 30(6):

135 240 Frank TS, Deffenbaugh AM, Reid JE, Hulick M, Ward BE, Lingenfelter B, Gumpper KL, Scholl T, Tavtigian SV, Pruss DR, Critchfield GC. Clinical characteristics of individuals with germline mutations in BRCA1 and BRCA2: analysis of 10,000 individuals. J Clin Oncol 2002; 20(6): Ellis NA, Kirchhoff T, Mitra N, Ye TZ, Chuai S, Huang H, Nafa K, Norton L, Neuhausen S, Gordon D, Struewing JP, Narod S, Offit K. Localization of breast cancer susceptibility loci by genome-wide SNP linkage disequilibrium mapping. Genet Epidemiol 2006; 30: Raskin L, Pinchev M, Arad C, Lejbkowicz F, Tamir A, Rennert HS, Rennert G, Gruber SB. FGFR2 is a breast cancer susceptibility gene in Jewish and Arab Israeli populations. Cancer Epidemiol Biomarkers Prev 2008; 17(5): Garcia-Closas M, Hall P, Nevanlinna H, Pooley K, Morrison J, Richesson DA, Bojesen SE, Nordestgaard BG, Axelsson CK, Arias JI, Milne RL, Ribas G, González-Neira A, Benítez J, Zamora P, Brauch H, Justenhoven C, Hamann U, Ko YD, Bruening T, Haas S, Dörk T, Schürmann P, Hillemanns P, Bogdanova N, Bremer M, Karstens JH, Fagerholm R, Aaltonen K, Aittomäki K, von Smitten K, Blomqvist C, Mannermaa A, Uusitupa M, Eskelinen M, Tengström M, Kosma VM, Kataja V, Chenevix-Trench G, Spurdle AB, Beesley J, Chen X, Australian Ovarian Cancer Management Group, Kathleen Cuningham Foundation Consortium For Research Into Familial Breast Cancer, Devilee P, van Asperen CJ, Jacobi CE, Tollenaar RA, Huijts PE, Klijn JG, Chang-Claude J, Kropp S, Slanger T, Flesch-Janys D, Mutschelknauss E, Salazar R, Wang-Gohrke S, Couch F, Goode EL, Olson JE, Vachon C, Fredericksen ZS, Giles GG, Baglietto L, Severi G, Hopper JL, English DR, Southey MC, Haiman CA, Henderson BE, Kolonel LN, Le Marchand L, Stram DO, Hunter DJ, Hankinson SE, Cox DG, Tamimi R, Kraft P, Sherman ME, Chanock SJ, Lissowska J, Brinton LA, Peplonska B, Klijn JG, Hooning MJ, Meijers-Heijboer H, Collee JM, van den Ouweland A, Uitterlinden AG, Liu J, Lin LY, Yuqing L, Humphreys K, Czene K, Cox A, Balasubramanian SP, Cross SS, Reed MW, Blows F, Driver K, Dunning A, Tyrer J, Ponder BA, Sangrajrang S, Brennan P, McKay J, Odefrey F, Gabrieau V, Sigurdson A, Doody M, Struewing JP, Alexander B, Easton DF, Pharoah PD. Heterogeneity of breast cancer associations with five susceptibility loci by clinical and pathological characteristics. PLoS Genet 2008; 4(4):e Fletcher O, Johnson N, Gibson L, Coupland B, Fraser A, Leonard A, dos Santos Silva I, Ashworth A, Houlston R, Peto J. Association of genetic variants at 8q24 with breast cancer risk. Cancer Epidemiol Biomarkers Prev 2008; 17(3): ref Haiman CA, Patterson N, Freedman ML, Myers SR, Pike MC, Waliszewska A, Neubauer J, Tandon A, Schirmer C, McDonald GJ, Greenway SC, Stram DO, Le Marchand L, Kolonel LN, Frasco M, Wong D, Pooler LC, Ardlie K, Oakley-Girvan I, Whittemore AS, Cooney KA, John EM, Ingles SA, Altshuler D, Henderson BE, Reich D. Multiple regions within 8q24 independently affect risk for prostate cancer. Nat Genet 2007; 39: Huijts PE, Vreeswijk MP, Kroeze-Jansema KH, Jacobi CE, Seynaeve C, Krol-Warmerdam EM, Wijers-Koster PM, Blom JC, Pooley KA, Klijn JG, Tollenaar RA, Devilee P, van 119

136 Asperen CJ. Clinical correlates of low-risk variants in FGFR2, TNRC9, MAP3K1, LSP1 and 8q24 in a Dutch cohort of incident breast cancer cases. Breast Cancer Res 2007; 9(6):R Liang J, Chen P, Hu Z, Zhou X, Chen L, Li M, Wang Y, Tang J, Wang H, Shen H. Genetic variants in fibroblast growth factor receptor 2 (FGFR2) contribute to susceptibility of breast cancer in Chinese women. Carcinogenesis 2008; 29(12): Pharoah PD, Tyrer J, Dunning AM, Easton DF, Ponder BA. Association between common variation in 120 candidate genes and breast cancer risk. PLoS Genet 2007; 3:e Offit K, Garber JE. Time to check CHEK2 in families with breast cancer? J Clin Oncol 2008; 26: Stacey SN, Manolescu A, Sulem P, Rafnar T, Gudmundsson J, Gudjonsson SA, Masson G, Jakobsdottir M, Thorlacius S, Helgason A, Aben KK, Strobbe LJ, Albers-Akkers MT, Swinkels DW, Henderson BE, Kolonel LN, Le Marchand L, Millastre E, Andres R, Godino J, Garcia-Prats MD, Polo E, Tres A, Mouy M, Saemundsdottir J, Backman VM, Gudmundsson L, Kristjansson K, Bergthorsson JT, Kostic J, Frigge ML, Geller F, Gudbjartsson D, Sigurdsson H, Jonsdottir T, Hrafnkelsson J, Johannsson J, Sveinsson T, Myrdal G, Grimsson HN, Jonsson T, von Holst S, Werelius B, Margolin S, Lindblom A, Mayordomo JI, Haiman CA, Kiemeney LA, Johannsson OT, Gulcher JR, Thorsteinsdottir U, Kong A, Stefansson K. Common variants on chromosomes 2q35 and 16q12 confer susceptibility to estrogen receptor-positive breast cancer. Nat Genet 2007; 39(7): Stacey SN, Manolescu A, Sulem P, Thorlacius S, Gudjonsson SA, Jonsson GF, Jakobsdottir M, Bergthorsson JT, Gudmundsson J, Aben KK, Strobbe LJ, Swinkels DW, van Engelenburg KC, Henderson BE, Kolonel LN, Le Marchand L, Millastre E, Andres R, Saez B, Lambea J, Godino J, Polo E, Tres A, Picelli S, Rantala J, Margolin S, Jonsson T, Sigurdsson H, Jonsdottir T, Hrafnkelsson J, Johannsson J, Sveinsson T, Myrdal G, Grimsson HN, Sveinsdottir SG, Alexiusdottir K, Saemundsdottir J, Sigurdsson A, Kostic J, Gudmundsson L, Kristjansson K, Masson G, Fackenthal JD, Adebamowo C, Ogundiran T, Olopade OI, Haiman CA, Lindblom A, Mayordomo JI, Kiemeney LA, Gulcher JR, Rafnar T, Thorsteinsdottir U, Johannsson OT, Kong A, Stefansson K. Common variants on chromosome 5p12 confer susceptibility to estrogen receptor-positive breast cancer. Nat Genet 2008; 40(6): Kibriya MG, Jasmine F, Argos M, Andrulis IL, John EM, Chang-Claude J, Ahsan H. A pilot genome-wide association study of early-onset breast cancer. Breast Cancer Res Treat 2008; 114(3): Altshuler D, Daly MJ, Lander ES. Genetic mapping in human disease. Science 2008; 322: Hirschhorn JN. Genomewide association studies_illuminating biologic pathways. N Engl J Med 2009; 360: Zeggini E, Scott LJ, Saxena R, Voight BF, Marchini JL, Hu T, de Bakker PI, Abecasis GR, Almgren P, Andersen G, Ardlie K, Boström KB, Bergman RN, Bonnycastle LL, Borch- Johnsen K, Burtt NP, Chen H, Chines PS, Daly MJ, Deodhar P, Ding CJ, Doney AS, Duren 120

137 WL, Elliott KS, Erdos MR, Frayling TM, Freathy RM, Gianniny L, Grallert H, Grarup N, Groves CJ, Guiducci C, Hansen T, Herder C, Hitman GA, Hughes TE, Isomaa B, Jackson AU, Jørgensen T, Kong A, Kubalanza K, Kuruvilla FG, Kuusisto J, Langenberg C, Lango H, Lauritzen T, Li Y, Lindgren CM, Lyssenko V, Marvelle AF, Meisinger C, Midthjell K, Mohlke KL, Morken MA, Morris AD, Narisu N, Nilsson P, Owen KR, Palmer CN, Payne F, Perry JR, Pettersen E, Platou C, Prokopenko I, Qi L, Qin L, Rayner NW, Rees M, Roix JJ, Sandbaek A, Shields B, Sjögren M, Steinthorsdottir V, Stringham HM, Swift AJ, Thorleifsson G, Thorsteinsdottir U, Timpson NJ, Tuomi T, Tuomilehto J, Walker M, Watanabe RM, Weedon MN, Willer CJ; Wellcome Trust Case Control Consortium, Illig T, Hveem K, Hu FB, Laakso M, Stefansson K, Pedersen O, Wareham NJ, Barroso I, Hattersley AT, Collins FS, Groop L, McCarthy MI, Boehnke M, Altshuler D. Meta-analysis of genome-wide association data and large-scale replication identifies additional susceptibility loci for type 2 diabetes. Nat Genet 2008; 40: Cohen JC, Kiss RS, Pertsemlidis A, Marcel YL, McPherson R, Hobbs HH. Multiple rare alleles contribute to low plasma levels of HDL cholesterol. Science 2004; 305: Ji W, Foo JN, O'Roak BJ, Zhao H, Larson MG, Simon DB, Newton-Cheh C, State MW, Levy D, Lifton RP. Rare independent mutations in renal salt handling genes contribute to blood pressure variation. Nat Genet 2008; 40: Hashimoto T, Shindo Y, Souri M, Baldwin GS. A new inhibitor of mitochondrial fatty acid oxidation. J Biochem 1996; 119: Menendez JA, Lupu R. RNA interference-mediated silencing of the p53 tumor-suppressor protein drastically increases apoptosis after inhibition of endogenous fatty acid metabolism in breast cancer cells. Int J Mol Med 2005; 15: Zhou W, Simpson PJ, McFadden JM, Townsend CA, Medghalchi SM, Vadlamudi A, Pinn ML, Ronnett GV, Kuhajda FP. Fatty acid synthase inhibition triggers apoptosis during S phase in human cancer cells. Cancer Res 2003; 63: Lee SA, Ho C, Roy R, Kosinski C, Patil MA, Tward AD, Fridlyand J, Chen X. Integration of genomic analysis and in vivo transfection to identify sprouty 2 as a candidate tumor suppressor in liver cancer. Hepatology 2008; 47: Mani A, Gelmann EP. The ubiquitin-proteasome pathway and its role in cancer. J Clin Oncol 2005; 23: Zhang Q, Meng Y, Zhang L, Chen J, Zhu D. RNF13: a novel RING-type ubiquitin ligase over-expressed in pancreatic cancer. Cell Res 2009; 19(3): Chen C, Seth AK, Aplin AE. Genetic and expression aberrations of E3 ubiquitin ligases in human breast cancer. Mol Cancer Res 2006; 4: Russo J, Balogh GA, Chen J, Fernandez SV, Fernbaugh R, Heulings R, Mailo DA, Moral R, Russo PA, Sheriff F, Vanegas JE, Wang R, Russo IH. The concept of stem cell in the mammary gland and its implication in morphogenesis, cancer and prevention. Front Biosc 2006; 11:

138 267 Tibiletti MG, Sessa F, Bernasconi B, Cerutti R, Broggi B, Furlan D, Acquati F, Bianchi M, Russo A, Capella C, Taramelli R. A large 6q deletion is a common cytogenetic alteration in fibroadenomas, pre-malignant lesions, and carcinomas of the breast. Clin Cancer Res 2000; 6:

139 Appendix 1 Distelman-Menachem T, Shapira T, Laitman Y, Kaufman B, Barak F, Tavtigian S, Friedman E, Analysis of BRCA1/BRCA2 genes' contribution to breast cancer susceptibility in high risk Jewish Ashkenazi women, Fam Cancer. 2009;8(2): Epub 2008 Sep 17.

140 Familial Cancer (2009) 8: DOI /s Analysis of BRCA1/BRCA2 genes contribution to breast cancer susceptibility in high risk Jewish Ashkenazi women Tal Distelman-Menachem Æ Tal Shapira Æ Yael Laitman Æ Bella Kaufman Æ Frida Barak Æ Sean Tavtigian Æ Eitan Friedman Published online: 17 September 2008 Ó Springer Science+Business Media B.V Abstract Background Three mutations in BRCA1 (185delAG 5382InsC) and BRCA2 (6174delT) can be detected in a substantial proportion of Jewish Ashkenazi breast/ovarian cancer families. Family-specific pathogenic mutations in both genes can be detected in up to 5% of high risk Ashkenazim. The contribution of major gene rearrangements and seemingly pathogenic missense mutations to inherited breast cancer predisposition has never been systematically evaluated in Ashkenazim. Material and methods High risk, Jewish Ashkenazi women, non-carriers of the predominant Jewish BRCA1/BRCA2 mutations, were genotyped for major gene rearrangements in BRCA1/ BRCA2 using Multiplex ligation-dependent probe Tal Distelman Menachem and Tal Shapira have equally contributed to this manuscript. T. Distelman-Menachem T. Shapira Y. Laitman E. Friedman (&) Susanne Levy Gertner Oncogenetics Unit, The Danek Gertner Institute of Human Genetics, Chaim Sheba Medical Center, Tel-Hashomer 52621, Israel eitan211@netvision.net.il B. Kaufman Susanne Levy Gertner Oncogenetics Unit, The Oncology Institute, Chaim Sheba Medical Center, Tel-Hashomer, Israel B. Kaufman E. Friedman The Sackler School of Medicine, Tel-Aviv University, Tel-Aviv, Israel F. Barak The Oncology Institute, Barzilai Medical Center, Ashkelon, Israel S. Tavtigian Genetic Susceptibility Group, International Agency for Research on Cancer, Cours Albert Thomas, Lyon Cedex 08, France amplification (MLPA), and for the occurrence rate of 6 seemingly pathogenic missense mutations in BRCA1 (R866C, R331S, R841W, Y179C, C61G, M1008I) using a modified restriction enzyme assay. Results Overall, 105 Jewish Ashkenazi high risk women, participated in the study: 104 with breast cancer [age at diagnosis (mean ± SD) ± years], one was affected with ovarian cancer (61 years). Two were found to carry the M1008I mutation in BRCA1 and none harbored any of the other missense mutations. MLPA reveled four changes (amplifications of exons 5, 17, 19 and 21) in BRCA1 in five patients, and six patients exhibited 4 MLPA-detectable abnormalities in BRCA2 (amplifications in exons 1b, 2, and deletions in exons 11a and 25). None of these abnormalities could be confirmed using quantitative PCR (qpcr) analysis. Conclusions Major gene rearrangements involving BRCA1 BRCA2 contribute little to the burden of inherited predisposition of breast cancer in Ashkenazi Jews. Keywords Genetic susceptibility BRCA1 BRCA2 Major gene rearrangements Sequence variants of unknown significance Introduction In approximately 5 10% of breast cancer cases, familial clustering and other features indicative of an inherited predisposition to cancer are noted. A substantial proportion of breast-ovarian cancer families and to a lesser extent, site-specific breast cancer are accounted for by germline mutations in the BRCA1 (MIM#113705) and BRCA2 (MIM#600185) genes. While in the majority of populations, the mutational spectrum of BRCA1 and BRCA2 in high risk families varies from family to family, among 123

141 128 T. Distelman-Menachem et al. Jewish Ashkenazi families there are only a handful of mutations. The three predominant mutations found in this population (185delAG, 5382insC and 6174delT in BRCA1 and BRCA2, respectively) can be detected in up to 12% of incident breast cancer, in 35% of incident ovarian cancer, and in 2.5% of the general Ashkenazi population [1]. These three mutations account for the majority of high risk families of Ashkenazi origin, and the prevalence of nonfounder mutations in Jewish Ashkenazi women undergoing clinical testing for a family history is only *2% [2]. The majority of pathogenic mutations in BRCA1 and BRCA2 are point mutations or small deletions and insertions (BIC database). In most populations tested, the observed frequencies of BRCA1 variations in high-risk breast and/or ovarian cancer families are lower than predicted by linkage analysis. This finding suggests that methods generally used for mutation scanning fail to detect certain types of germline defects, such as large genomic rearrangements. Most of mutation screening methods are PCR-based, and hence there is an inherent inability to detect large genomic rearrangements such as partial or complete exon loss or amplification [3] Several approaches have been reported for detecting BRCA germline gene rearrangements: Southern blot [4 6] long-range PCR [7], color bar coding of the BRCA genes on combed DNA [8, 9], semiquantitative-multiplex PCR [2], and real-time PCR [5, 6]. Multiplex ligation-dependent probe amplification (MLPA) is widely used as a highly sensitive method for detecting the relative copy number of all BRCA1 BRCA2 exons in a high-throughput format [10]. Major gene rearrangements have been identified in BRCA1 and BRCA2 in several populations. Specifically, large genomic deletions in BRCA1 were noted in the French, Italian, and Dutch populations [6, 11, 12]. More than 60 different BRCA1 germline rearrangements with mapped breakpoints have been reported to date [13]. These are scattered throughout the whole gene, and most of them are deletions, but duplication, triplication or combined deletion/insertion events [14, 15] have also been described. The proportion of genomic rearrangements in the BRCA1 mutation spectrum varies from 8 to 15%, depending on population and methodology, with the higher rates associated with founder mutations [1, 6, 7, 16 19]. Less is known about the occurrence rate of large genomic rearrangements in BRCA2 in hereditary breast/ovarian cancer families. To date, few studies have been published revealing less than 20 [10, 20 23] different genomic rearrangements in BRCA2 in hereditary breast cancer patients from diverse European populations. Testing has primarily been carried out by Southern blot analysis. Only the most recent studies have applied MLPA as a comprehensive screening method [3, 4, 7, 12, 21 38] and the value of BRCA2 screening for such alterations is still undetermined. Only one published study focused on the contribution of major gene rearrangements to inherited predisposition in Jewish women. Wang et al. [39] looked for major gene rearrangements by Southern blot hybridization among 47 Jewish breast cancer patients with hereditary features, 30 of whom were of Ashkenazi descent. A deletion/insertion in BRCA2 was detected in a single family of Jewish-Turkish (non-ashkenazi) descent. The deletion included exons 12 and 13, a track of approximately 60 adenine nucleotide residues was inserted at the breakpoint junction [39]. Notably, there were no major gene rearrangements reported among Ashkenazi patients in that study. While the pathogenicity of inactivating, truncating BRCA1 and BRCA2 mutations is well established, the contribution of other BRCA1 and BRCA2 sequence variants to cancer risk remains largely undefined. These variant of unknown significance (VUSs) are mainly missense mutations. The open-access, on-line Breast Cancer Information core (BIC database) contains over 1500 distinct sequence variants that are currently reported as having unknown clinical significance. Several methodologies have been applied to assess the presumed pathogenicity of these sequence variants, but there is no consensus regarding the clinical application of any of these techniques as an aid in assigning a pathogenic role for these VUSs. A notable exception is the C61G BRCA1 mutation, which is one of the most prevalent, clearly pathogenic, germline mutations in Polish high risk families. From the mechanistic point of view, the C61G mutation disrupts homodimer formation in the NH2-terminal BRCA1 RING finger [40], and hence is clearly pathogenic. In the present study, 105 Ashkenazi breast/ovarian cancer patients with inherited predisposition to cancer, who tested negative for the three predominant Ashkenazi mutations in BRCA1 and BRCA2 were genotyped for putative pathogenic missense BRCA1 mutations and for major gene rearrangements in both BRCA1 and BRCA2 using MLPA. Patients, materials and methods Patients and families Patients were ascertained from one of two sources: the high risk clinic at the Oncogenetics unit, Sheba medical center, and an ongoing project that genetically tests and counsels consecutive Jewish Ashkenazi women diagnosed with breast cancer conducted jointly by the Sheba and Shaare Tzedek medical centers. All participants had breast or ovarian cancer, and in addition, one or more of the eligibility criteria: (1) under 40 years of age at time of diagnosis of breast cancer (n = 16) or (2) over 40 years of 123

142 Analysis of BRCA1/BRCA2 genes contribution to breast cancer 129 age at time of diagnosis of breast cancer and at least one of the following: (a) bilateral breast cancer (n = 13) or ovarian cancer (n = 1); (b) at least one-first degree relative with breast cancer (n = 20); (c) one or more first or second degree relatives with bilateral breast cancer (n = 21) or ovarian cancer (n = 7); (d) at least two relatives of the paternal side with breast cancer (n = 11)and e) at least one relative diagnosed for breast cancer under the age of 40 years (n = 8). (3) tested negative for the three predominant Ashkenazi mutations (BRCA1*185delAG, 5382insC, BRCA2*6174delT). Inclusion criteria were not met by 8 patients who did not fit to these above mentioned criteria but had a personal and family history highly suspicious for an inherited predisposition and were also analyzed herein. The study was approved by the local and the Ministry of Health IRB, and each participant signed a written informed consent. DNA extraction Genomic DNA was isolated from peripheral blood leucocytes by the PUREGene kit (Gentra Inc Minenapolis MN) using the manufacturer s recommended protocol. Selecting and assigning pathogenicity of missense mutations in BRCA1 We searched the BIC database for missense mutations in BRCA1 gene reported in the Ashkenazi population or individuals of East European origin. Missense mutations within conserved domains (R866C; R841W), those with a proven pathogenic effect (C61G), those that scored moderately high on the align-gvgd algorithm [41] (Y179C) or are extremely rare (R331S) were selected. Genotyping for the missense mutations PCR primers were designed to flank the mutated area and a restriction enzyme was used to differentiate the wild type from the mutant allele. PCR was preformed in a 25 ll reaction, containing ng genomic DNA, PCR buffer (Fisher Biotec, Australia), 2.5 mm MgCl2, 200nM dntps, 10 pmol of each primer and 0.2 U FB1 DNA Polymerase (Fisher Biotec,). Amplification was carried out as follows: an initial denaturation step of 5 min at 94 C, followed by 35 cycles of 94 C for 30 s, annealing step which was different for each fragment, 72 C for 30 s, and a final extension step at 72 C for 10 min. Mutations and primers sequences are listed in Table 1. Amplification conditions for each fragment are available from the authors upon request. Multiplex ligation-dependent probe amplification (MLPA) MLPA analysis was performed by the Salsa P002 BRCA1 and Salsa P045 BRCA2 MLPA probe mix assay as described by the manufacturer (MRC-Holland Amsterdam, the Netherlands) [3]. Peak heights from each patient were then exported to an excel spreadsheet, which was designed by Dr. Andrew Wallace from the National Genetics Reference Laboratories, Manchester, to assess the ratios of each test peak relative to all other peaks for that individual. Each test and control sample s data is normalized (by summing the total control peak height and dividing each ligation product s peak height by this figure.), the Dosage quotients are calculated and the three hypotheses (normal, duplicated, deleted) are tested comparing to 5 normal controls. For normal sequence a dosage quotient of 1.0 ( ) is expected; if a deletion or duplication is present, the dosage quotient should be and , respectively. A dosage quotient of and is equivocal. In addition, the sample quality is assessed by measuring the standard deviation of all the test ligation products measured against each other. Standard deviation of less than 0.1 shows no overlap between normal, duplicated and deleted ranges [3, 4]. Quantitative PCR Dosage changes found by MLPA analysis were re-evaluated (for ascertainment and validation) by Multiplex quantitative PCR (qpcr). qpcr was performed in the ABI7700 instrument. A set of FAM-labeled primers and Table 1 Mutation designation, primer sequences and restriction Mutation Forward primer Reverse primer Restriction enzyme R866C AGAATACATTCAAGGTTTCACA TAGACAAAACCTAGAGCCTCCT TaaI R331S CCAGAAAAGTATCAGGGTAGTTC CTTTTTTTCTGTGCTGGGAGACC Cfr10I R841 W ATTGGGACATGAAGTTAACCAGAG CTGTCTGTACAGGCTTGATATTAG Sch1 Y179C TACAACCTCAAAAGACGTCTGTC TGTGAGCCACCATGCCTGGCC Xmil C61G TGGCTCTTAAGGGCAGTTGTGAGA CTGTGGTTGCTTCCAACCTAGCAT AvaII M1008I TCAGAGGCAACGAAACTGGACTCA TGGAGCCCACTTCATTAGTACTGG Tsp45I 123

143 130 T. Distelman-Menachem et al. probes were designed for each exon or gene fragment that showed an abnormality on MLPA analysis. The primers were designed to flank the MLPA probe, and the design of the Taqman probe was similar to the MLPA probe. A mix containing VIC-labeled primers and probe for RNaseP (Roche diagnostics, Manheim, Germany), a single copy gene, was used as an internal standard. Amplification/ deletion reaction were performed in a volume of 20 ll containing 10 ll of TaqMan universal PCR mastermix (Roche diagnostics, Manheim, Germany), 1 ll of each BRCA-specific primers, 1 ll of each BRCA-specific labeled hybridization probe, 0.5 ll of the RNaseP primerprobe mix and 1 ll of genomic DNA. Thermocycling was as followed: 95 C for 15 min, followed by 40 cycles of 95 C for 15 s, 60 C for 10 s. Data analysis was carried out using ABI Prism 7700 Sequence Detection Software Since the standard curve for sequential concentrations of the two different probes was similar, a difference between the Ct of the examined BRCA exon and that of the RNaseP indicates a deletion/ duplication. Results Patient characteristics Overall, 105 women participated in the study: 103 were diagnosed with breast cancer [mean age at diagnosis (±SD) was ± years (range years)], one with ovarian cancer (61 years) and one patient had both ovarian and breast cancers (ovarian cancer at 43 years and breast cancer at 48 years). All participants were of Jewish Ashkenazi ancestry, and none was a carrier of any of the three predominant mutations in BRCA1 (185delAG, 5382InsC) or BRCA2 (6174DelT). BRCA1 gene rearrangements MLPA analysis MLPA reaction for BRCA1 had statistically significant results (P \ 0.1) in only 48 samples of the total 105 genotyped. In 16 of these samples, no dosage changes were found. Among the remaining 32 samples with statistically significant data, several dosage changes were found, but none in sequential exons. In only two samples, an amplification of exons 17 and 19 had significant odds ratio (1:2232, 1:421 for the first sample and 1:85, 1:74 for the second sample).the first patient displaying the abnormality is a breast cancer patient diagnosed at age 52, her sister was diagnosed with breast cancer at age 57 years, and their paternal cousin was diagnosed with breast cancer at age 50 years. There are no other women in the family. The other patient was diagnosed with breast cancer at age 28 years and her sister was diagnosed with breast cancer at age 43 years. There are no cancer cases at their maternal family and on the paternal side there are no women relatives. Among the remaining 57 samples, 21 samples did reach partial statistical significance (P \ 0.1). In this set of genotyped samples, amplifications were detected in exons 5, 17, 19 and 21 in three patients. Quantitative PCR Due to the low incidence of statistically significant results, qpcr for confirmation of the four amplified regions detected by MLPA, was performed for the whole cohort. None of the samples showed any of the amplifications in exons 5, 17, 19 and 21, by qpcr. Specifically, the five samples that showed the MLPA abnormality did not reveal any abnormality using the qpcr probe. BRCA2 major gene rearrangements MLPA analysis The MLPA reaction for BRCA2 had statistically significance (P \ 0.1) in only 31 of the 105 samples. In 11 samples no dosage changes were detected. Among the remaining 20 samples, dosage changes were found in exons 1b, 3 (amplification) and 11a, 25 (deletion), in six patients. These dosage changes had significant odds ratio (Table 2). The MLPA kit for BRCA2 includes a probe for the Chek2*1100delC mutation. Three samples were positive for this mutation, one in a statistically significant score. qpcr Due to the low incidence of statistically significant results, qpcr for the search of the four dosage changes found by MLPA in the BRCA2 gene was performed for the whole cohort. None of the amplifications were detected and Table 2 Summary of the MLPA analyses for BRCA2 Family # Ex1b Ex3 Ex11a Ex Amp 1:58361 Amp 1: Amp 1: Amp 1: Amp 1: sz Del 1: sz Del 1: sz Del 1:74 123

144 Analysis of BRCA1/BRCA2 genes contribution to breast cancer 131 validated by qpcr, including the six samples that showed the MLPA associated abnormality. Sequencing confirmation of the 1100delT CHEK2 mutation The three samples for which the MLPA reaction indicated the presence of the Chek2*1100delC mutation, were sequenced to ensure the existence of that mutation. Only one of the three samples was positive for the mutation and the other two samples, including the one showing the most statistically significant finding on MLPA, did not harbor that mutation. Genotyping for missense mutations in BRCA1 All 105 samples were genotyped for 6 missense mutations in BRCA1: R866C, R331S, R841W, Y179C, C61G, M1008I. Two women were carriers of the I1008M mutation. The first patient displaying the abnormality is a breast cancer patient diagnosed at age 62 years, her sister was diagnosed with breast cancer at age 56 years, their mother was diagnosed with breast cancer at age 44 years, and their maternal aunt was diagnosed with breast cancer around the age of 50 years. The other patient was diagnosed with breast cancer at age 45 years and her mother was diagnosed with breast cancer and colorectal cancer at age 62 years. There are no other relatives on the maternal side. In addition to the M1008I mutation, sequencing of these two samples revealed two neutral polymorphisms: E1038G which is not clinically important, and S1040N which is of unknown significance. None of the other missense mutations was detected in any of the samples. Discussion In this study, MLPA analysis of Jewish women with inherited predisposition to breast cancer was suggestive of the existence of a few genomic rearrangements involving BRCA1 and BRCA2. Yet, these results could not be confirmed or validated independently by another techniquequantitative PCR. Several studies reported the MLPA as a high sensitivity technique by using a positive control that was originally detected by Southern blotting, the gold standard method for dosage changes [4 6]. Other studies focusing on the presence of major gene rearrangements in BRCA1 and BRCA2 either used a MLPA kit with a different set of probes and primers [34] or qpcr for the validation of their findings. Based on these studies and the present study, it seems that an approach that uses MLPA as a screening technique and as a guide for further analysis and validation of the findings by an independent technique should be recommended, at least for the set of probes applied for BRCA1 and BRCA2. The lack of finding of consistent major gene rearrangements involving BRCA1 and/or BRCA2 in high risk Jewish Ashkenazi families in the present study is in line with the previous study that focused on Jewish families. In that study, only one major gene rearrangement in BRCA2 in a non-ashkenazi high risk family [39], and notably no gene rearrangements involving Ashkenazim (n = 30) were reported. Similarly, two studies that have applied MLPA analysis have failed to show any BRCA1 BRCA2 gene rearrangement in 135 French-Canadian [4] and no BRCA2 MLPA-detectable abnormalities in 36 Finnish [32] high risk families. In the present study one of the genotyped missense mutations in BRCA1 (M1008I) was detected in two of the study participants. The M1008I missense mutation is a conservative mutation due to a G? A transition at nucleotide 3143 in exon 11 changing codon 1008 (ATG) encoding the non-polar Met into ATA encoding Ile, another non-polar residue. This Met is changed for a Ser in the mouse BRCA1 protein [42]. The M1008I mutation was reported in ethnically diverse populations: Ashkenazim, Central/Eastern Europe, Western European population, Latin American/Caribbean, and Puerto Rico populations, (BIC database). Given these considerations and the fact that this variant was previously classified as neutral by two independent groups of investigators [43, 44] and that it could be found in trans with a clearly deleterious BRCA1 mutation [44], it seems likely that this variant does not have a deleterious effect on BRCA1 protein function and is merely a rare polymorphism. The selection of the other five specific mutations was based on several criteria, that made it plausible that these may indeed be pathogenic. Several explanations may account for this lack of detected missense mutations: inadequate assignment of the pathogenicity of these mutations, rarity of these mutations in high risk Jewish Ashkenazi families, even if they are truly pathogenic, small sample size, inclusion of non-inherited cases among genotyped families. Indeed, at least one missense mutation tested (C61G) is a known pathogenic mutation [27]. The lack of this mutation in the present study deserves special emphasis, as a substantial proportion of Ashkenazim originated in Poland, and intuitively, one would expect to find at least some representation of this mutation (prevalent in non- Jewish individuals) among Jewish families. This expectation is based on the existence of the 5382InsC BRCA1 mutation in Jewish and non-jewish individuals from the same geographical region [27]. The lack of this mutation may be accounted for by a later date of origin of this mutation, or its limited geographical location to regions that were not densely populated with Jews in Poland. 123

145 132 T. Distelman-Menachem et al. The limitations of the study should be borne in mind. This is a limited study from a single medical center and the high risk families analyzed may not be representative of the spectrum of high risk families, even of Ashkenazi origin. The possibility that additional pathogenic BRCA1 BRCA2 mutations, other then the three screened predominant Jewish mutations, do exist in the sample analyzed, should also be considered a limitation. However, based on other studies performed in the Jewish Ashkenazi population, the maximum number of expected private mutations in both genes is less than 5% [17]. Only a subset of analyzed samples were adequately scored by the MLPA analysis, so that the actual number of analyzed individuals is even smaller regarding an assessment of the contribution of major gene rearrangement to inherited predisposition to breast/ovarian cancer among Ashkenazim. Lastly the lack of confirmation by Southern blotting (predominantly related to the lack of sufficient amounts of high quality DNA), detracts from the firmness of the conclusions. In conclusion, no consistent major gene rearrangements involving BRCA1 or BRCA2 and no seemingly pathogenic missense mutations in BRCA1 were detected in a cohort of high risk Jewish Ashkenazi women. Extension and validation of this preliminary conclusion is highly warranted. References 1. King MC, Marks JH, Mandell JB (2003) Breast and ovarian cancer risks due to inherited mutations in BRCA1 and BRCA2. Science 302(5645): doi: /science Antoniou A, Pharoah PD, Narod S et al (2003) Average risks of breast and ovarian cancer associated with BRCA1 or BRCA2 mutations detected in case Series unselected for family history: a combined analysis of 22 studies. Am J Hum Genet 72: doi: / Schouten JP, McElgunn CJ, Waaijer R et al (2002) Relative quantification of 40 nucleic acid sequences by multiplex ligationdependent probe amplification. Nucleic Acids Res 30(12):e57 4. Moisan AM, Fortin J, Dumont M et al (2006) No evidence of BRCA1/2 genomic rearrangements in high-risk French-Canadian breast/ovarian cancer families. Genet Test 10(2): doi: /gte Puget N, Torchard D, Serova-Sinilnikova OM et al (1997) A 1-kb Alu-mediated germline deletion removing BRCA1 exon 17. Cancer Res 57(5): Petrij-Bosch A, Peelan T, van Vliet M et al (1997) BRCA1 genomic deletions are major founder mutations in Dutch breast cancer patients. Nat Genet 17(3): doi: /ng Vasickova P, Machackova E, Lukesova M et al (2007) High occurrence of BRCA1 intragenic rearrangements in hereditary breast and ovarian cancer syndrome in the Czech Republic. BMC Med Genet 11(8):32 8. Gad S, Aurias A, Puget N et al (2001) Color bar coding the BRCA1 gene on combed DNA: a useful strategy for detecting large gene rearrangements. Genes Chromosomes Cancer 31: doi: /gcc Gad S, Klinger M, Caux-Moncoutier V et al (2002) Bar code screening on combed DNA for large rearrangements of the BRCA1 and BRCA2 genes in French breast cancer families. J Med Genet 39: doi: /jmg Woodward AM, Davis TA, Silva AG (2005) Large genomic rearrangements of both BRCA2 and BRCA1 are a feature of the inherited breast/ovarian cancer phenotype in selected families. J Med Genet 42:e31. doi: /jmg Gad S, Caux-Moncoutier V, Pagès-Berhouet S et al (2002) Significant contribution of large BRCA1 gene rearrangements in 120 French breast and ovarian cancer families. Oncogene 3: 21(44): Montagna M, Dalla Palma M, Menin C et al (2003) Genomic rearrangements account for more than one-third of the BRCA1 mutations in northern Italian breast/ovarian cancer families. Hum Mol Genet 12(9): doi: /hmg/ddg Armaou S, Konstantopoulou I, Anagnostopoulos T et al (2007) Novel genomic rearrangements in the BRCA1 gene detected in Greek breast/ovarian cancer patients. Eur J Cancer 43(2): doi: /j.ejca Smith TM, Lee MK, Szabo CI et al (1996) Complete genomic sequence and analysis of 117 kb of human DNA containing the gene BRCA1. Genome Res 6: doi: /gr Puget N, Gad S, Perrin-Vidoz L et al (2002) Distinct BRCA1 rearrangements involving the BRCA1 pseudogene suggest the existence of a recombination hot spot. Am J Hum Genet 70: doi: / Garber JE, Offit K (2005) Hereditary cancer predisposition syndromes. J Clin Oncol 23(2): doi: /jco Kauff ND, Perez-Segura P, Robson ME et al (2002) Incidence of non-founder BRCA1 and BRCA2 mutations in high risk Ashkenazi breast and ovarian cancer families. J Med Genet 39(8): doi: /jmg Swensen J, Hoffman M, Skolnick MH et al (1997) Identification of a 14 kb deletion involving the promoter region of BRCA1 in a breast cancer family. Hum Mol Genet 6: doi: /hmg/ Puget N, Stoppa-Lyonet D, Sinilnikova OM et al (1999) Screening for germ-line rearrangements and regulatory mutations in BRCA1 led to the identification of four new deletions. Cancer Res 59: Ramus SJ, Harrington PA, Pye C et al (2007) Contribution of BRCA1 and BRCA2 mutations to inherited ovarian cancer. Hum Mutat 28(12): doi: /humu Lim YK, Iau PTC, Ali AB et al (2007) Identification of novel BRCA large genomic rearrangements in Singapore Asian breast and ovarian patients with cancer. Clin Genet 71(4): doi: /j x 22. Gutiérrez-Enríquez S, de la Hoya M, Martínez-Bouzas C et al (2007) Screening for large rearrangements of the BRCA2 gene in Spanish families with breast/ovarian cancer. Breast Cancer Res Treat 103(1): doi: /s Agata S, Dalla Palma M, Callegaro M (2005) Large genomic deletions inactivate the BRCA2 gene in breast cancer families. J Med Genet 42(10):e64. doi: /jmg Tournier I, Paillerets BB, Sobol H et al (2004) Significant contribution of germline BRCA2 rearrangements in male breast cancer families. Cancer Res 64: doi: / can Staaf J, Törngren T, Rambech E et al (2008) Detection and precise mapping of germline rearrangements in BRCA1, BRCA2, MSH2, and MLH1 using zoom-in array comparative genomic hybridization (acgh). Hum Mutat 29(4): doi: / humu

146 Analysis of BRCA1/BRCA2 genes contribution to breast cancer Miramar MD, Calvo MT, Rodriguez A et al (2008) Genetic analysis of BRCA1 and BRCA2 in breast/ovarian cancer families from Aragon (Spain): two novel truncating mutations and a large genomic deletion in BRCA1. Breast Cancer Res Treat [Epub ahead of print] 27. Ratajska M, Brozek I, Senkus-Konefka E et al (2008) BRCA1 and BRCA2 point mutations and large rearrangements in breast and ovarian cancer families in Northern Poland. Oncol Rep 19(1): Purnomosari D, Pals G, Wahyono A et al (2007) BRCA1 and BRCA2 germline mutation analysis in the Indonesian population. Breast Cancer Res Treat 106(2): doi: /s Nowee ME, Snijders AM, Rockx DA et al (2007) DNA profiling of primary serous ovarian and fallopian tube carcinomas with array comparative genomic hybridization and multiplex ligationdependent probe amplification. J Pathol 213(1): doi: /path Konecny M, Zavodna K, Vranova V et al (2008) Identification of rare complete BRCA1 gene deletion using a combination of SNP haplotype analysis, MLPA and array-cgh techniques. Breast Cancer Res Treat 109(3): doi: /s Gutiérrez-Enríquez S, Balmaña J, Baiget M et al (2008) Detection of the CHEK2 1100delC mutation by MLPA BRCA1/2 analysis: a worthwhile strategy for its clinical applicability in 1100delC low-frequency populations? Breast Cancer Res Treat 107(3): doi: /s Karhu R, Laurila E, Kallioniemi A et al (2006) Large genomic BRCA2 rearrangements and male breast cancer. Cancer Detect Prev 30(6): doi: /j.cdp Thomassen M, Gerdes AM, Cruger D et al (2006) Low frequency of large genomic rearrangements of BRCA1 and BRCA2 in western Denmark. Cancer Genet Cytogenet 168(2): de la Hoya M, Gutiérrez-Enríquez S, Velasco E et al (2006) Genomic rearrangements at the BRCA1 locus in Spanish families with breast/ovarian cancer. Clin Chem 52(8): doi: /clinchem Hartmann C, John AL, Klaes R (2004) Large BRCA1 gene deletions are found in 3% of German high-risk breast cancer families. Hum Mutat 24(6):534. doi: /humu Bunyan DJ, Eccles DM, Sillibourne J et al (2004) Dosage analysis of cancer predisposition genes by multiplex ligationdependent probe amplification. Br J Cancer 91(6): Belogianni I, Apessos A, Mihalatos M et al (2004) Characterization of a novel large deletion and single point mutations in the BRCA1 gene in a Greek cohort of families with suspected hereditary breast cancer. BMC Cancer 4: Hogervorst FB, Nederlof PM, Gille JJ et al (2003) Large genomic deletions and duplications in the BRCA1 gene identified by a novel quantitative method. Cancer Res 63(7): Wang T, Lerer I, Guetan Z et al (2005) A Deletion/Insertion Mutation in the BRCA2 gene in a Breast Cancer Family: A Possiable Role of the Alu-polyA Tail in the Evolution of the Deletion. Genes Chromosomes Cancer 31: doi: / gcc Brzovic PS, Meza J, King MC et al (1998) The cancer-predisposing mutation C61G disrupts homodimer formation in the NH2-terminal BRCA1 RING finger domain. J Biol Chem 273(14): Tavtigian SV, Deffenbaugh AM, Yin L et al (2006) Comprehensive statistical study of 452 BRCA1 missense substitutions with classification of eight recurrent substitutions as neutral. J Med Genet 43: doi: /jmg Durocher F, Shattuck-Eidens D, McClure M et al (1996) Comparison of BRCA1 polymorphisms, rare sequence variants and/or missense mutations in unaffected and breast/ovarian cancer populations. Hum Mol Genet 5: doi: /hmg/ Goldgar DE, Easton DF, Deffenbaugh AM et al (2004) Integrated evaluation of DNA sequence variants of unknown clinical significance: application to BRCA1 and BRCA2. Am J Hum Genet 75: doi: / Judkins T, Hendrickson BC, Deffenbaugh AM et al (2005) Application of embryonic lethal or other obvious phenotypes to characterize the clinical significance of genetic variants found in trans with known deleterious mutations. Cancer Res 65: doi: / can

147 Appendix 2 Menachem TD, Laitman Y, Kaufman B, Friedman E. The RNF146 and ECHDC1 genes as candidates for inherited breast and ovarian cancer in Jewish Ashkenazi women. Fam Cancer Jun 11.

148 Familial Cancer DOI /s The RNF146 and ECHDC1 genes as candidates for inherited breast and ovarian cancer in Jewish Ashkenazi women Tal Distelman Menachem Æ Yael Laitman Æ Bella Kaufman Æ Eitan Friedman Ó Springer Science+Business Media B.V Abstract Only about 40% of the familial aggregation of breast cancer can be attributed to germline mutations in currently identified genes, primarily BRCA1 and BRCA2.A recent genome-wide association study focusing on Jewish Ashkenazi high risk women identified a novel locus on chromosome 6 as putatively containing breast cancer susceptibility genes, a locus that contains two seemingly novel candidate genes: RNF146 and ECHDC1. To further explore the role of these two genes in inherited predisposition to breast cancer. High risk, cancer affected Jewish Ashkenazi women, were genotyped for harboring germline mutations in the coding exons of both the RNF146 and ECHDC1 genes, using direct sequencing. All participants were Ashkenazim, of high risk families, and affected with cancer: 104 with breast cancer [age at diagnosis (mean ± SD) 51 ± 11.1 years], and one with ovarian cancer (61 years). None was a carrier of the predominant Jewish BRCA1/ BRCA2 mutations. An intronic sequence alteration was detected in 4/105 genotyped patients in intron 3 of the ECHDC1 gene. No other sequence alterations were detected in the genomic regions analyzed of the RNF146 and ECH- DC1 genes in any of the study participants. Mutations in the T. D. Menachem Y. Laitman E. Friedman (&) The Susanne Levy Gertner Oncogenetics Unit, The Danek Gertner Institute of Human Genetics, Chaim Sheba Medical Center, Tel-Hashomer, Israel eitan211@netvision.net.il B. Kaufman The Oncology Institute, Tel-Hashomer and the Sackler School of Medicine, Chaim Sheba Medical Center, Tel-Aviv University, Tel-Aviv, Israel E. Friedman The Sackler School of Medicine, Tel-Aviv University, Tel-Aviv, Israel coding regions of the RNF146 and ECHDC1 genes do not contribute to the burden of inherited predisposition of breast cancer in Ashkenazi high risk women. Keywords Inherited predisposition to breast cancer Genome-wide association study Candidate genes RNF146 ECHDC1 Introduction About 40% of the familial aggregation of breast cancers can be accounted for by mutations in BRCA1, BRCA2, and to a lesser extent, other known breast cancer susceptibility genes [1]. Recently, two groups have carried out independent genome-wide association studies (GWAS) of either selected high risk kindreds and/or consecutive, seemingly sporadic breast cancer cases [2, 3]. Both studies have shown that a locus near the FGFR2 gene (rs ) is associated with an &1.2-fold increased risk for developing breast cancer. Another GWAS [4] that focused specifically on Ashkenazi Jewish high risk families was recently completed. This study encompassed 249 Ashkenazi Jewish kindreds containing multiple cases of breast cancer but lacking any of the predominant mutations in this ethnic group in either the BRCA1 (185delAG 5382InsC) or BRCA2 (6174delT) genes. This latter study successfully validated the previously reported association with the FGFR2 locus, and additionally, identified a novel locus on 6q22.33 as significantly associated with breast cancer susceptibility (OR 1.41, 95% CI for rs ) Within that region, in tight linkage disequilibrium, there are two novel candidate genes: RNF146 and ECHDC1 that could plausibly be implicated as breast cancer susceptibility genes. 123

149 T. D. Menachem et al. To further elucidate the possible role that these genes play in inherited predisposition to breast cancer, direct sequencing of the coding regions (and flanking intronic boundaries) in both genes was carried out in 105 Ashkenazi breast/ovarian cancer patients from high risk families, who all tested negative for the three predominant Ashkenazi mutations in BRCA1 and BRCA2. Patients, materials and methods Patients and families Patients were ascertained from one of two sources: the high risk clinic at the Oncogenetics unit, Sheba medical center, and an ongoing project that genetically tests and counsels consecutive Jewish Ashkenazi women diagnosed with breast cancer conducted jointly by the Sheba and Shaare Tzedek medical centers. All participants had breast or ovarian cancer, and in addition, one or more of the eligibility criteria: (1) under 40 years of age at time of diagnosis of breast cancer (n = 16) or (2) over 40 years of age at time of diagnosis of breast cancer and at least one of the following: (a) bilateral breast cancer (n = 13) or ovarian cancer (n = 1); (b) at least one-first degree relative with breast cancer (n = 20); (c) one or more first or second degree relatives with bilateral breast cancer (n = 21) or ovarian cancer (n = 7); (d) at least two relatives of the paternal side with breast cancer (n = 11) and (e) at least one relative diagnosed for breast cancer under the age of 40 years (n = 8); (3) tested negative for the three predominant Ashkenazi mutations (BRCA1*185delAG, 5382insC, BRCA2*6174delT). Inclusion criteria were not met by eight patients who did not fit to these above mentioned criteria but had a personal and family history highly suspicious for an inherited predisposition and were also analyzed herein. The study was approved by the local and the Ministry of Health IRB, and each participant signed a written informed consent. DNA extraction Genomic DNA was isolated from peripheral blood leucocytes by the PUREGene kit (Gentra Inc., Minenapolis, MN) using the manufacturer s recommended protocol. Direct sequencing PCR primers were designed to flank all coding exons. PCR was preformed in a 25 ll reaction, containing ng genomic DNA, PCR buffer (Fisher Biotec, Australia), 2.5 mm MgCl 2, 200 nm dntps, 10 pmol of each primer and 0.2 U FB1 DNA Polymerase (Fisher Biotec). Amplification was carried out as follows: an initial denaturation step of 5 min at 94 C, followed by 35 cycles of 94 C for 30 s, annealing step which was different for each exon, 72 C for 30 s, and a final extension step at 72 C for 10 min. Primers sequences are listed in Tables 1 and 2. Amplification conditions for each fragment are available from the authors upon request. PCR fragments were subject to sequence analysis using the big Dye terminator chemistry and kit (PE Biosystems, Foster City, CA), and using the ABI Prism 3100 semiautomatic DNA sequencer (PE Biosystems). Results Patient characteristics Overall, 105 women participated in the study: 103 were diagnosed with breast cancer [mean age at diagnosis (±SD) was 51.0 ± 11.1 years (range years)], one with ovarian cancer (61 years) and one patient had both ovarian and breast cancers (ovarian cancer at 43 years and breast cancer at 48 years). All participants were of Jewish Ashkenazi ancestry, and none was a carrier of any of the three predominant mutations in BRCA1 (185delAG, 5382InsC) or BRCA2 (6174DelT). RNF146 and ECHDC1 genes genotyping No mutations or polymorphisms were detected in the genotyped regions of the RNF146 gene (Data not shown). A rare sequence variant (a T to C change in intron 3 of the gene, 47 nucleotides into the intron -IVS3? 47T/C) was detected in 4/105 women (*4%) genotyped in the ECH- DC1 gene (data not shown). Discussion In the current study, no germline mutations were detected in the coding regions and flanking intronic sequences of two seemingly biologically plausible candidate genes (RNF146 and ECHDC1) in a large set of Jewish Ashkenazi women at high risk for developing breast/ovarian cancer. The plausibility of these genes being involved in predisposing to breast/ovarian cancer stems from two independent sources: the known functions of the gene products and their chromosomal localization. The ECHDC1 gene product has a related domain to the mitochondrial enoyl-coa hydratase/3-hydroxyacyl-coa dehydrogenase/3-ketoacyl-coa thiolase, a key protein in mitochondrial fatty acid oxidation [5]. Although ECHDC1 123

150 The RNF146 and ECHDC1 genes Table 1 RNF146 gene exon and primer sequences Exon Forward primer Reverse primer Exon 1a CCCATTGTCTAGTATGTGGCTTGC AGTCCAGCTACTCCCTTCTTTGGT Exon 1b ATGAACATGGACGTCGCAGGAAGA ACCTTGAGCATGGAGCTGAAGACA Table 2 ECHDC1 gene exons and primer sequences (GeneBank accession number NT_025741) Exon Forward primer Reverse primer Exon 1 GTCACTGGAGCTTTACTTGGAGT GATTAGAGCATACAGAGAGAGTTAC Exon 2 GGTTCAAGCTTCTGCTATCCTTTG GAGCAAGTTAATTATATTTGTGCAAG Exon 3 TGATTATCCTCTACGGCACAA GGAGCCCTAGGATATCAACAGTCA Exon 4 TGTTCCTCCTATAATCCTACCA GTATACTCCATAGTGCCAGTGCAG Exon 5 ACATTCTGTCACCACAGCAAGGA TTCCCAACTCAACAGAAGGGTCCA gene mutations have not been reported to occur somatically in breast cancer, its possible involvement in breast cancer pathogenesis stems from the observation that fatty acid synthase-dependent endogenous fatty acid activity is abnormally elevated in a biologically aggressive subset of breast carcinomas, and that inhibition of that activity induces apoptosis in breast cancer cell lines, an effect enhanced 300-fold in TP53 null cell lines [6, 7]. Recently, ECHDC1 gene expression was also found to be downregulated in hepatocellular carcinoma, as further supportive evidence to its intimate involvement in the malignant transformation process [8]. RNF146, also called dactylidin, encodes a ubiquitously expressed, cytoplasmic polypeptide containing an amino-terminal C3HC4 RING finger domain that functions as a ubiquitin protein ligase (E3). Abnormalities in the ubiquitin-mediated protein degradation pathway affect cellular processes (e.g., cell cycle, apoptosis, etc.,) that are pivotal in breast cancer pathogenesis [9]. Other genes that confer breast cancer predisposition (BRCA1, BRCA2, and BARD1) also exhibit ubiquitin like activity or have an indirect effect on ubiquitin-mediated protein degradation [10]. Another supportive evidence for the putative involvement of RNF146 in the pathogenesis of breast cancer comes from its differential expression in breast epithelial stem cells in parous compared with nulliparous women [11]. These lines of evidence, combined with the reported rates of somatic loss of 6q in premalignant breast tissue lesions as well as breast cancer in a region that encompass both the RNF146 and the ECHDC1 genes [12], make it plausible for both genes to be involved in predisposing to breast cancer. Despite the lack of mutations or even any sequence variants in the coding regions of both genes in this subset of women, one or both genes may still be involved in conferring inherited predisposition to breast cancer by other mechanisms: major gene rearrangements that cannot be detected by PCR, sequence variants in gene expression controlling regions, or epigenetic mechanisms. Presently, we can conclude based on the results of this preliminary study, that germline mutations in the coding regions and flanking intronic sequences of both the ECHDC1 and RNF146 genes contribute only marginally, if at all, to the burden of inherited predisposition to breast cancer among AJ. Acknowledgments This research project was in part funded by a grant from the Silva De Gomez foundation from the Tel-Aviv University Sackler School of Medicine to Eitan Friedman (For Tal Distelman-Menachem) and by funds from the Israeli cancer association to Eitan Friedman. References 1. Walsh T, King MC (2007) Ten genes for inherited breast cancer. Cancer Cell 11: doi: /j.ccr Easton DF, Pooley KA, Dunning AM et al (2007) Genome-wide association study identifies novel breast cancer susceptibility loci. Nature 447: doi: /nature Hunter DJ, Kraft P, Jacobs KB et al (2007) A genome-wide association study identifies alleles in FGFR2 associated with risk of sporadic postmenopausal breast cancer. Nat Genet 39: doi: /ng Gold G, Kirchhoff T, Stefanov S et al (2008) Genome-wide association study provides evidence for a breast cancer risk locus at 6q Proc Natl Acad Sci USA 105(11): doi: /pnas Hashimoto T, Shindo Y, Souri M, Baldwin GS (1996) A new inhibitor of mitochondrial fatty acid oxidation. J Biochem 119: Menendez JA, Lupu R (2005) RNA interference-mediated silencing of the p53 tumor-suppressor protein drastically increases apoptosis after inhibition of endogenous fatty acid metabolism in breast cancer cells. Int J Mol Med 15: Zhou W, Simpson PJ, McFadden JM et al (2003) Fatty acid synthase inhibition triggers apoptosis during S phase in human cancer cells. Cancer Res 63: Lee SA, Ho C, Roy R et al (2008) Integration of genomic analysis and in vivo transfection to identify sprouty 2 as a candidate tumor suppressor in liver cancer. Hepatology 47: doi: /hep

151 T. D. Menachem et al. 9. Mani A, Gelmann EP (2005) The ubiquitin-proteasome pathway and its role in cancer. J Clin Oncol 23: doi: / JCO Yarden RI, Papa MZ (2006) BRCA1 at the crossroad of multiple cellular pathways: approaches for therapeutic interventions. Mol Cancer Ther 5: doi: / mct Russo J, Balogh GA, Chen J et al (2006) The concept of stem cell in the mammary gland and its implication in morphogenesis, cancer and prevention. Front Biosci 11: doi: / Tibiletti MG, Sessa F, Bernasconi B et al (2000) A large 6q deletion is a common cytogenetic alteration in fibroadenomas, pre-malignant lesions, and carcinomas of the breast. Clin Cancer Res 6:

152 Appendix 3 Atwal GS, Bond GL, Metsuyanim S, Papa M, Friedman E, Distelman-Menachem T, Ben Asher E, Lancet D, Ross DA, Sninsky J, White TJ, Levine AJ, Yarden R. Haplotype structure and selection of the MDM2 oncogene in humans. Proc Natl Acad Sci U S A Mar 13;104(11):

153 Haplotype structure and selection of the MDM2 oncogene in humans Gurinder Singh Atwal*, Gareth L. Bond*, Sally Metsuyanim, Moshe Papa, Eitan Friedman, Tal Distelman-Menachem, Edna Ben Asher, Doron Lancet, David A. Ross**, John Sninsky**, Tomas J. White**, Arnold J. Levine*, and Ronit Yarden *Institute for Advanced Study, Simons Center for Systems Biology, Princeton, NJ 08540; Cancer Institute of New Jersey, Department of Pediatrics, Robert Wood Johnson Medical School, New Brunswick, NJ 08903; Laboratory of Genomic Applications, Department of Surgical Oncology, Department of Surgical Oncology, and Susanne Levy Gertner Oncogenetics Unit, The Danek Gertner Institute of Human Genetics, Sheba Medical Center, Tel Hashomer 52621, Israel; The Crown Human Genome Center, Department of Molecular Genetics, The Weizmann Institute of Science, Rehovot 76100, Israel; and **Celera Diagnostics, Alameda, CA Contributed by Arnold J. Levine, December 21, 2006 (sent for review August 15, 2006) The MDM2 protein is an ubiquitin ligase that plays a critical role in regulating the levels and activity of the p53 protein, which is a central tumor suppressor. A SNP in the human MDM2 gene (SNP309 T/G) occurs at frequencies dependent on demographic history and has been shown to have important differential effects on the activity of the MDM2 and p53 proteins and to associate with altered risk for the development of several cancers. In this report, the haplotype structure of the MDM2 gene is determined by using 14 different SNPs across the gene from three different population samples: Caucasians, African Americans, and the Ashkenazi Jewish ethnic group. The results presented in this report indicate that there is a substantially reduced variability of the deleterious SNP309 G allele haplotype in all three populations studied, whereas multiple common T allele haplotypes were found in all three populations. This observation, coupled with the relatively high frequency of the G allele haplotype in both and Caucasian and Ashkenazi Jewish population data sets, suggests that this haplotype could have undergone a recent positive selection sweep. An entropy-based selection test is presented that explicitly takes into account the correlations between different SNPs, and the analysis of MDM2 reveals a significant departure from the standard assumptions of selective neutrality. cancer p53 population genetics SNP entropy In response to a wide variety of stresses, such as DNA damage or oncogene activation, the p53 tumor suppressor protein is activated and initiates a transcriptional program leading to cell cycle arrest, cell senescence or apoptosis (1). This eliminates clones of cells that have acquired mutations, which arise at a high frequency when DNA replication or the cell cycle proceeds under stress. When the p53 gene is mutated in either the germ line or in a somatic cell, many types of cancers can arise (2). The p53 protein is regulated by a ubiquitin ligase, the MDM2 protein, which binds to p53, blocking its function as a transcription factor, and polyubiquitinates the p53 protein sending it to the proteiosome for degradation (3). The MDM2 gene in turn is positively regulated by p53-mediated transcription, setting up an autoregulatory loop that keeps both proteins at moderate levels. Stress responses perturb this feedback loop, which leads to the initiation of p53-dependent apoptosis. Functional SNPs in the human genome have been identified in both the p53 and the MDM2 genes (4). In the p53 gene, a SNP (codon 72) results in the change of a proline residue to an arginine at codon 72 of the p53 protein (p53-pro and p53-arg, respectively). Multiple groups have shown that p53-pro is weaker than p53-arg in its ability to both suppress cellular transformation and induce apoptosis in cell culture (5 8), and can associate with an earlier onset of tumor formation and a poorer tumor response to chemotherapy in humans (7, 9, 10). In the MDM2 gene, a SNP (SNP309) results in a nucleotide change from the wild-type thymine (T) to guanine (G) in the intronic promoter/enhancer region (11). The G allele increases the binding of a transcription factor, SP1, which in turn results in higher levels of MDM2 RNA and protein, the attenuation of the p53 pathway and an enhanced early onset of, and increased risk for, tumorigenesis (11 22). More recent studies of MDM2 SNP309 suggest that primarily female specific hormones, like estrogen, either directly or indirectly, allow for the G allele of SNP309 to accelerate tumor formation in women in four different sporadic cancers (diffuse large B cell lymphoma, soft tissue sarcoma, invasive ductal breast carcinoma and colorectal cancer) (14, 23). Together, these data suggest that functional SNPs in the p53 pathway will play a role in regulating the efficiency of the p53 stress response over a lifetime, and as such the efficacy of the p53 pathway in tumor suppression after exposure to stresses. To date all of the associations of the G allele of MDM2 SNP309 with the early onset of cancers in patients have been linked to this locus alone and little is known of the haplotypes that contain the G or the T alleles at SNP309. It remains possible that one or several G haplotypes will be associated with early onset of cancer, and it is important to determine this at an early stage in the genetic epidemiological studies of this allele. For that reason, the haplotype structure of 14 different SNPs in the MDM-2 gene was determined by employing three different racial and ethnic populations: one African American population, one Caucasian not selected for ethnicity, and one Caucasian of the Ashkenazi Jewish ethnic group. These populations were chosen for further haplotype analysis, as it had been previously observed that African Americans have a low G allele frequency, non-jewish Caucasians an intermediate G allele frequency, and Ashkenazi Jewish groups a high G allele frequency (14, 24). The results presented in this report indicate that there are only a few common ( 1%) G allele haplotypes in all three populations studied, one in the African American and Caucasian data sets and two in the Ashkenazi Jewish data set. The SNPs in the G allele haplotype are thus highly correlated. We suggest that the single G allele haplotype in the African American population could have arisen through admixture with other racial groups, like Caucasians, and thus is a relatively recent mutation. In contrast, multiple common T allele haplotypes were found in all three populations, thus exhibiting reduced correlations between the SNPs. This observation coupled with the relatively high frequency of the G allele haplotype in both Caucasian populations suggests that this haplotype could have experienced recent positive selection pressure. To test this hypothesis, an entropy-based selection test is devised that Author contributions: G.S.A., G.L.B., J.S., A.J.L., and R.Y. designed research; G.S.A., S.M., M.P., E.F., T.D.-M., E.B.A., D.L., D.A.R., and T.J.W. performed research; G.S.A. contributed new reagents/analytic tools; G.S.A. analyzed data; and G.S.A., G.L.B., and A.J.L. wrote the paper. The authors declare no conflict of interest. Abbreviation: DLE, differential mean linkage equilibrium. To whom correspondence should be addressed. atwal@ias.edu. This article contains supporting information online at /DC by The National Academy of Sciences of the USA PNAS March 13, 2007 vol. 104 no cgi doi pnas

154 Fig. 1. Schematic diagram of the MDM2 gene and the SNPs genotyped in the present study. The three SNPs common to both the Celera and Sheba data sets are indicated by the dotted lines. compares both the frequency and long-range correlations of the allele with a simulated model where the allele is selectively neutral. The results confirm that the probability that the G allele haplotype is selectively neutral is quite low. Results The MDM2 SNPs genotyped from various populations in this study are depicted in Fig. 1, and their frequencies are detailed in Tables 1 and 2. The observed genotype frequencies of the MDM2 SNPs were found not to deviate significantly from Hardy Weinberg equilibrium within each race and ethnic group, with P values ranging from 0.08 to As expected, the assumption of Hardy Weinberg equilibrium was found to be notably violated when the populations were pooled together. The frequencies of the SNPs common to both studies are, on average, most similar between the Caucasian and Ashkenazi Jewish data sets. However, because the absolute frequencies of the common SNPs across the Caucasian and Ashkenazi Jewish samples do not match exactly, there is no way of combining the haplotypes without incurring severe biases. All haplotypes with an expected frequency of at least 1% are presented in Figs. 2, 3, and 4 for Caucasians, African Americans, and Ashkenazi Jewish data sets, respectively. The marginal haplotype frequencies inferred from this calculation are shown in the Table 1. Major allele frequencies of the 11 SNPs genotyped in the Celera study SNP ID dbsnp ID Major/minor allele African American Frequency Caucasian Cel 1 rs A/G Cel 2 rs C/T Cel 3 rs T/G Cel 4 rs T/A Cel 5 rs A/G Cel 6 rs T/C Cel 7 rs A/G Cel 8 rs G/A Cel 9 rs C/G Cel 10 rs A/G Cel 11 rs T/G The labeling of major and minor allele was determined by pooling both the African American and Caucasian population samples. figures in a SNP309-centric fashion, adding additional SNPs about SNP309 two at a time so as to permit comparisons of subhaplotypes. Previously reported SNP309 G allele frequencies in Northern European Caucasians were 33%, but in African Americans they have been noted to be significantly lower, 11% (14, 24, 25). A salient feature of the African Americans and Caucasians figures is that, in both populations, the G allele of SNP309 is highly correlated with all of the other SNPs across the entire region of MDM2 covered by the Celera Diagnostic SNP set, resulting in only one G haplotype in both populations. The G allele haplotype is identical in both races, just at a much lower frequency in African Americans (43% vs. 10%), possibly suggesting that its presence in African Americans could be due to an admixture (25). If true, this would lead to the prediction that Africans probably do not carry the G allele of SNP309, which then posits the idea that the G allele could have arisen more recently in evolution. Evidence of recombination was determined by the four-gamete test, serving as a guide to the minimum number of recombination events consistent with the data. Of the possible 55 pairs, there were seven positive tests in both the African American and Caucasians populations, and five out of 15 in the Jewish ethnic population. To directly ascertain the correlations among SNPs of the MDM2 gene in the differing population data sets, the linkage disequilibrium of the MDM2 SNPs was estimated by calculating the standard pairwise values of D and r 2 (Fig. 5). The linkage disequilibrium estimation analysis of the MDM2 SNPs revealed that the SNP309 locus is absolutely correlated with two other MDM2 SNPs, Cel 5 and Cel 8. These three SNPs seem to always travel together in a haplotype, and thus collectively they represent only one degree of freedom in genetic variation. Hence, there is no detectable simple single mutation precursor of the G allele haplotype of SNP309 amongst the T allele haplotypes in both populations. Table 2. Major allele frequencies of the six SNPs genotyped in the Sheba study SNP ID dbsnp ID Major/minor allele Ashkenazi Jewish frequency She 1 rs G/A 0.66 She 2 rs T/G 0.73 She 3 rs A/G 0.74 She 4 rs G/T 0.54 She 5 rs C/A 0.63 She 6 rs A/G 0.57 APPLIED MATHEMATICS GENETICS Atwal et al. PNAS March 13, 2007 vol. 104 no

155 Fig. 2. Inferred haplotype frequencies in the Caucasian population. Fig. 4. Inferred haplotype frequencies in the Ashkenazi Jewish population. Independent evolution of subpopulations results in reduced heterozygosity of the total population as detected by Wright s F statistic. For the Caucasian and African American populations, we find that F ST 0.17, mirroring the observation from the allelic Hardy Weinberg tests that mating was essentially random within each race but not between the races. The data across the races was permuted 110 times and the calculated F ST statistic indicated that P 10 5 (26). The average number of pairwise distances between the two races was 4.3 with P Although the G allele occurs at intermediate frequencies, it is a striking observation that the number of different G allele haplotypes across the entire gene is dramatically reduced compared with the number of T allele haplotypes even though the T allele frequency is also intermediate in the non-african-american populations. How significant then is the paucity of the G allele haplotypes and what biological implications, if any, does it suggest? To quantify the total variation of haplotypes within each population for an arbitrary number of SNPs, we appealed to information theory (27) and estimated the entropy, which serves as a unique measure of variability under a few, but very general, mathematical assumptions. To be more specific, we calculated the multientropy H[{X}] from the observed haplotypes arising from a set, {X},ofb 2 SNPs, H[{X}] b i 1 p i (x 1 x 2...x b )log 2 p i (x 1 x 2...x b ), where each random variable x i denotes one of the two alleles of SNP i. Concavity of the entropy function results, on average, in a negative sampling bias, which can be corrected by using a bootstrap resampling procedure to extrapolate to the infinite sample size (28). Under the standard simple assumptions of a large panmictic population the variability of a stretch of the genome increases with time or, more accurately, number of generations, because of increased probability of recombination and mutation. The total Fig. 3. Inferred haplotype frequencies in the African American population. entropy of the African American and Caucasian haplotypes in the Celera data set was calculated to be 3.0 bits and 2.05 bits, respectively, supporting the idea that the Caucasians are a much more recent interbreeding population than the Africans. A large level of linkage disequilibrium across the gene entails a low level of multientropy. It is expected from simple considerations of a neutral model of a population of mutating and recombining chromosomes that low frequency (younger) alleles ought to occur on fewer haplotypes than higher frequency (older) alleles, and thus will be in strong linkage disequilibrium across relatively large stretches of the genome. To capture this intuition, we suggest that, under the assumptions of a neutral model, there ought to be a monotonically increasing relationship between the frequency of the allele and the mean entropy of the associated haplotypes. To show this, we generated samples of polymorphic data from a Monte Carlo simulation within a coalescent framework of neutral mutation and homogeneous recombination (29). In Fig. 6a, we show a plot of the entropy of haplotypes around a particular SNP with distance (base pairs) away from the SNP. For low-frequency alleles, the entropy of associated haplotypes rises slowly with distance away from the allele because low-frequency alleles are usually recently occurring mutations that have not had sufficient time to attain linkage equilibrium with surrounding SNPs. Higher frequency alleles, as expected, exhibit a much greater increase in linkage equilibrium and entropy. To make comparisons between the alleles of a particular SNP, we summarize the linkage equilibrium for each allele by calculating the distance-averaged entropy up to some cutoff distance away from the SNP. The cutoff in our study here has a natural upper bound due to the limited number of SNPs genotyped across the genetic region of interest. More generally, a natural cutoff is given by the accuracy by which we can calculate the entropy, and a simple large-sample calculation of the first-order error in entropy estimation shows that the criteria for accurate entropy estimation is a function of N, the number of samples, and b, the number of SNPs, (i.e., 2 b N 1 1). Decreasing the cutoff arbitrarily results in greater variability of the distribution of linkage equilibrium for a given allele and, conversely, increasing the cutoff compromises the accuracy of entropy, and thus there ought to be an optimal intermediate cutoff where both effects are mitigated. By comparing the differences of average linkage equilibrium for different alleles at a particular SNP, we are able to obviate the systematic error that occurs due to variable recombination rates across the genome. Thus, the summary statistic of the differential mean linkage equilibrium (DLE) for an allele A is given by DLE 1 c 1 2z 0,c H c A H B c z c 1,c H A i H B i z i 1 i 1,i 1, [1] cgi doi pnas Atwal et al.

156 APPLIED MATHEMATICS Fig. 5. Pairwise linkage disequilibrium for Ashkenazi Jewish (a), Caucasian (b), and African American (c) populations. The upper right triangle reports the D measure, and the lower left triangle reports the r 2 measure. SNP309 is highlighted in all populations. where the index B refers to the other allele at the locus, c is the number of SNPs before the cutoff, z i,j is the base pair distance between SNPs situated at loci i and j, and H i A,B is the multientropy for i SNPs extending away from a particular locus, conditional on either allele A or B at locus i 0. In Fig. 6b, we summarize the data of allele frequency and DLE showing the expected relationship under the neutral model, whereby high-frequency alleles have positive DLE and low-frequency alleles have negative DLE. Any deviation from this would be due to a relaxation of the standard assumptions of a neutral model. An allele with a large frequency but low DLE is suggestive of a selection pressure that has rapidly pushed up the frequency of the allele giving it insufficient time to approach equilibration with surrounding SNPs. As proof of principle, we then generated a Monte Carlo sample of polymorphic data where one allele was preassigned to have a weak positive selection pressure (selective advantage per copy per generation 0.1) (30). The summary plot of the DLE values for all SNPs (Fig. 6c) shows that the selected allele does indeed significantly deviate from the mean of the null distribution. Indeed, recent work (31, 32) has provided supportive evidence that such haplotype-based methods are far more sensitive to recent selection detection. The P value was Fig. 6. Monte Carlo simulation of haplotypes. (a) The conditional multientropy of the adjacent SNPs is plotted for a particular locus where the minor allele (lower curve) had a frequency of 0.19 and the major allele (upper curve) had a frequency of The unit of distance is measured in rescaled base pairs. (b) Plot of the DLE values versus frequency of each allele. One thousand Monte Carlo genome samples of size 250,000 base pairs were generated. The mutation rate was set to produce 8,000 SNPs, and the recombination rate was set to1cmmb 1.(c) Selected allele (indicated by arrow) underwent a selective advantage per copy per generation of 0.1 estimated by Monte Carlo simulation of a neutral model (31) with 10 6 trials to provide a fine-scale distribution of neutral DLE for a given allele frequency where the frequencies were quantized into steps of 0.1. Previous tests of selection in the population genetics literature also look for deviations away from the neutrality but do not incorporate information from multiallelic associations even though selection pressure can have a strong effect on levels of linkage disequilibrium, and thus these tests have low power to detect recent selective sweeps (32). Nevertheless, we used a variety of standard selection tests on the population data sets, and the results, as detailed in Table 3, are inconsistent with rejecting the null hypothesis of a neutral model, with each population data set showing significant departure or not from neutrality depending on which selection test is used. Our proposed test of neutrality does not rest on an assumption of independent alleles but is explicitly based on the variability of haplotypes and hence also linkage disequilibrium. GENETICS Atwal et al. PNAS March 13, 2007 vol. 104 no