Germline MLH1 and MSH2 mutations in Italian pancreatic cancer ...

Familial Cancer (2009) 8:547–553 DOI 10.1007/s10689-009-9285-1

Germline MLH1 and MSH2 mutations in Italian pancreatic cancer patients with suspected Lynch syndrome S. Gargiulo Æ M. Torrini Æ S. Ollila Æ S. Nasti Æ L. Pastorino Æ R. Cusano Æ L. Bonelli Æ L. Battistuzzi Æ L. Mastracci Æ W. Bruno Æ V. Savarino Æ S. Sciallero Æ G. Borgonovo Æ M. Nystro¨m Æ G. Bianchi-Scarra` Æ C. Mareni Æ P. Ghiorzo

Published online: 1 September 2009 Springer Science+Business Media B.V. 2009

Abstract Lynch syndrome is an inherited cancer syndrome caused by germline mutations in mismatch repair (MMR) genes MLH1, MSH2, MSH6 and PMS2. LS predisposes to high risk of early-onset colorectal, endometrial and other tumors. Patients with Lynch syndrome have also been shown to have an elevated risk for pancreatic cancer (PC). In this study, we aimed to estimate the frequency of suspected Lynch syndrome among a series of 135 PC patients. Further, we wanted to determine the frequency of MMR gene mutations in the suspected Lynch syndrome cases. We also aimed to verify the pathogenicity of any novel non-truncating variants we might detect with a functional assay. Based on personal and/or familial cancer history, 19 patients were classified as suspected Lynch syndrome cases. DNA material for mutation analysis was available for eleven of them. Four patients were found to carry a total of five MLH1 or MSH2 variants. Of these, MSH2-Q402X, MSH2-G322D, and MLH1-K618A had been previously reported, while the MSH2-E205Q and MSH2-V367I variants were novel. MSH2-Q402X is a known stop mutation and reported here

for the first time here in association with PC. MLH1-K618A was found in the unaffected branch of a kindred, suggesting that it may be a polymorphism or a low penetrance variant. MSH2-G322D likely does not cause a MMR defect, although this variant has also been associated with breast cancer as indeed seen in our patient. The novel variants MSH2-E205Q and MSH2-V367I were found in the same patient. Both novel variants were however functional in the applied MMR assay. Our findings suggest that only a small subset of pancreatic cancer patients carry pathogenic MMR mutations.

S. Gargiulo S. Nasti L. Pastorino R. Cusano L. Battistuzzi W. Bruno G. Bianchi-Scarra` P. Ghiorzo (&) Department of Oncology, Biology and Genetics, University of Genoa, Viale Benedetto XV, 6, 16132 Genoa, Italy e-mail: [email protected]

L. Mastracci Department of Anatomic Pathology, University of Genoa, Genoa, Italy

M. Torrini C. Mareni Department of Internal Medicine, University of Genoa, Genoa, Italy S. Ollila M. Nystro¨m Department of Biological and Environmental Sciences, Genetics, University of Helsinki, Helsinki, Finland

Keywords Lynch syndrome Hereditary non-polyposis colorectal cancer MLH1 Mismatch repair genes MSH2 MSH6 Pancreatic cancer Abbreviations PC Pancreatic cancer (pancreatic adenocarcinoma) s-LS Suspected-Lynch syndrome

V. Savarino Division of Gastroenterology, Department of Internal Medicine, University of Genoa, Genoa, Italy S. Sciallero Medical Oncology Unit, San Martino Hospital, Genoa, Italy G. Borgonovo Department of Surgical and Morphological Disciplines and Integrated Methodologies, University of Genoa, Genoa, Italy

L. Bonelli Secondary Prevention and Screening, National Cancer Institute, Genoa, Italy

123

548

Introduction Exocrine pancreatic cancer (PC) is the deadliest among common cancers and one of the main causes of cancer death in Italy. Radical surgical resection remains the only method for potential cure, and applies to no more than 10– 20% of patients. In addition, 5-year survival does not exceed 20%; novel strategies for early detection and therapy are thus urgently needed. Identifying individuals at significantly increased risk of developing PC may facilitate early diagnosis of those who could benefit from treatment [1, 2]. A history of PC and other cancers among first-degree relatives is an important risk factor for the disease, and PC has been associated with several hereditary cancer syndromes, including Lynch syndrome (LS), also termed hereditary non-polyposis colorectal cancer (HNPCC). LS is an autosomal dominant condition characterized by early age of cancer onset, proximal predominance of colorectal cancer, excess of synchronous and metachronous tumors, and an extracolonic tumor spectrum that includes endometrial, ovarian, gastric, small bowel, hepatobiliary, brain, urothelial as well as pancreatic neoplasms [3, 4]. Susceptibility to LS is dominantly inherited with germline defects in the mismatch repair (MMR) genes MLH1, MSH2, MSH6 and PMS2. Truncating mutations in one of these genes, combined with loss of the wild-type allele, compromise MMR and thus promote genetic instability and tumor formation. Conversely, non-truncating mutations can either be neutral variations or lead to a highly increased cancer risk and LS. Determining whether a given variant is pathogenic is crucial, as identifying the causative mutation in a family enables genetic counseling and surveillance of mutation carriers, which has been shown to significantly lower mortality. The first aim of this study was to establish the frequency of suspected LS among our set of 135 PC patients. The second aim was to understand how many PC patients with suspected LS carry MLH1, MSH2 or MSH6 mutations. The third aim was to determine the pathogenicity of any novel non-truncating variants identified among these patients.

Materials and methods

S. Gargiulo et al.

The patients were administered a questionnaire on their personal and family cancer history; family pedigrees were constructed to include all first-, second- and third-degree relatives. PC diagnoses were confirmed by histology, cytology or image studies and verified with the local cancer registry. Other cancers were confirmed in all of the PC patients and where possible also among relatives, either through the local cancer registry or through medical records. All of the individuals enrolled provided their written informed consent to the study under ethics committeeapproved protocols. PC patients were selected for molecular testing if they could be classified as suspected LS (s-LS) for displaying a personal or family history (up to the second degree) of two colorectal cancers or other LS-associated tumors [5] with onset at any age, or one (colorectal cancer or other LS-associated tumor) if either the PC or the colorectal or LS-associated tumor was diagnosed at a young age (\50 years). Mutational analysis of MLH1, MSH2 and MSH6 Mutational analysis was performed in the PC patients identified as s-LS. Total genomic DNA was extracted using standard methods from total blood samples or EBVtransformed lymphocytes. Lymphoblastoid cell lines were available for 87 of the 135 patients enrolled since they were not established for the first 48 patients enrolled. Polymerase chain reaction (PCR) and direct sequencing analysis were used to screen the coding regions of the MLH1, MSH2 and MSH6 genes. Primers for MLH1 and MSH2 have been described elsewhere [6]. For MSH6 screening, primers were designed and are available upon request. PCR products were bi-directionally sequenced using an ABI PRISM Big Dye Terminator Cycle Sequencing Kit (Applied Biosystem, Foster City, CA) and the products were analyzed on an ABI 3100 DNA sequencer (Applied Biosystem, Foster City, CA). Segregation analysis of variations was performed if DNA from family members was available. One hundred healthy controls, selected within the framework of the ongoing hospital-based case–control study, were screened for presence of the novel variants first found in this study.

Patients We studied a series of 135 confirmed PC patients who were consecutively enrolled at the National Cancer Institute and San Martino Hospital in Genoa, within the framework of an ongoing case–control study aimed at identifying the contribution of hereditary cancer syndromes to the development of PC.

123

Immunohistochemical analysis of MMR protein expression and microsatellite-instability analysis Since tumor tissue samples were not available, immunohistochemical analysis of MMR protein expression and microsatellite-instability (MSI) testing were not conducted. The sole exception was a mutation-positive sample with

MLH1 and MSH2 mutations in Italian pancreatic cancer patients

metastatic liver paraffin-embedded tissue block available. IHC analysis was performed according to standard procedures along with MSI analysis, which was conducted on DNA extracted from normal and tumor tissue [7]. RNA analysis RNA analysis was conducted to assess the correct splicing of mRNA. Lymphoblastoid cell lines derived from MMR gene variation carriers, were analyzed when available. Total RNA was extracted according to standard procedures (Trizol, Life Technologies) from the lymphoblastoid cell lines. cDNA was then reverse transcribed from RNA using the Advantage RT-for-PCR kit (Clontech) and amplified using primers designed to specifically amplify MLH1 from exon 14 to 18, MSH2 from exon 1 to 4 and exon 5 to 8, to include the sequences surrounding the variations. Control PCR reactions were carried out using primers to amplify the GAPDH housekeeping gene. Sequencing was performed with the same primers used to amplify cDNA and exonic internal primers to confirm the presence of variations and to verify the exact splicing sites. Primer sequences are available upon request.

549

Mismatch repair assay The mismatch repair (MMR) assay was used to assess the in vitro mismatch repair activity of the MSH2 E205Q and V367I proteins. The assay was conducted as previously described [11] with minor modifications. The repair substrate used was a pGEM (Promega)-derived circular plasmid, which contained a G•T mismatch in a BglII restriction enzyme cleavage site and a single-stranded nick 390 bp 50 from the mismatch, directing the repair to the correct DNA strand. 75 lg of MSH2-deficient LoVo colon carcinoma cell line nuclear extract (NE) was complemented with 1 ll (WT) or 2 ll (E205Q and V367I) Sf 9 TEs, including *500 ng of recombinant MutSa, and incubated with 100 ng of substrate DNA. The reaction was incubated for 30 min at 37C, after which the DNA was extracted. Upon successful MMR, the G•T mismatch in the BglII restriction site was converted into A•T base pair, allowing BglII cleavage. Repair efficiency was assessed by BsaI—BglII double digestion and agarose gel electrophoresis. LoVo NE without added MutSa was used as a negative control, and LoVo NE supplemented with WT MutSa as a positive control.

Functional analysis of the novel MSH2 variants

Results

Protein expression

Mutational analysis of suspected LS patients

The recombinant wild-type (WT) MSH2, MSH2 E205Q and MSH2 V367I proteins were expressed together with their cognate partner WT MSH6 with previously published methods [8, 9]. Briefly, the mutations corresponding to the patient-derived MSH2 mutations E205Q (c. 613 G[C) and V367I (c. 1099 G[A) were constructed on MHS2 WT cDNA using site-directed mutagenesis. The PCR conditions and primer sequences are available upon request. The WT MSH2 and WT MSH6 cDNAs cloned into pFastBac1 plasmid vector (Invitrogen) were obtained from Professor Josef Jiricny. The MSH2-MSH6 heterodimer (MutSa) was produced using the Bac-to-Bac baculovirus expression system (Invitrogen) in Sporoptera frugiperda 9 (Sf 9) cells following the manufacturer’s instructions. The total soluble protein content (total extract, TE) of the cells was collected and the expression of MSH2 and MSH6 was verified with Western blotting using anti-MSH2 (MSH2 Ab-2, NA27, Calbiochem, Darmstadt, Germany, dilution 1:250) and anti-MSH6 (MSH6/GTBP, Clone 44, BD Transduction Laboratories, Erembodegem, Belgium, dilution 1:1,000) antibodies. 1 ll of each identically prepared Sf 9 TE was loaded on 6% SDS–PAGE gels and 300 ng of purified MutSa [10] was used as a reference for MutSa concentration.

Nineteen PC patients of the 135 enrolled were identified as being s-LS, based on their personal and family cancer history (see ‘‘Materials and methods’’). Among these 19 patients, two had previously been found to be positive for the CDKN2A-G101W germline mutation. In one family there were two colorectal cancers in second degree relatives, in the second family where the propositus was diagnosed with PC before age 50, a melanoma was present in a first degree relative and a colorectal cancer in a second degree relative [12]. Since the identified CDKN2A mutation was responsible for inherited cancer susceptibility in these two families, these patients were no longer considered for MLH1, MSH2 or MSH6 testing. Mutational analysis could be performed in 11 of the 17 other s-LS patients as lymphoblastoid cell lines were not available as a source of genomic DNA for six. Four of the eleven patients tested were found to carry MLH1 (K618A) or MSH2 (Q402X, G322D, E205Q and V367I) variants, only one of them was clearly deleterious (MSH2 Q402X). The MLH1-K618A variant was found in a female patient diagnosed with PC at the age of 37. Segregation analysis was performed in the available family members (Fig. 1a), and the mutation was seen to have been inherited from the unaffected branch of the family: the healthy father tested

123

550

S. Gargiulo et al.

Fig. 1 Pedigrees of s-LS mutation positive families. Family probands are indicated by arrows. Black symbols indicate malignant disease. Age at diagnosis and type of cancer are indicated under each symbol: Pan pancreatic cancer, Lung lung cancer, Col colorectal cancer, Br breast cancer, Kid kidney cancer, End endometrial cancer, Stom stomach cancer, Liv liver cancer, Thyr thyroid cancer, Unk unknown cancer; Brain brain cancer, BCC basal cell carcinoma; Blad bladder cancer; ?, - mutation positive and mutation negative individuals

positive while the mother, who had developed breast and kidney cancer at the age of 60 and had a family history suggestive of LS, tested negative. The proband was negative for mutations in MSH2 and MSH6. IHC and MSI analysis from liver metastatic tissue showed that MMR protein expression was conserved and MSI was absent. The MSH2 Q402X mutation was found in a female patient diagnosed with PC at the age of 51 (Fig. 1b) whose son had been diagnosed with brain cancer at age 20, whose mother had developed colorectal cancer at the age of 42 and whose sister developed endometrial cancer at the age of 43. Mutational analysis was conducted in the proband, a healthy brother and the sister; only the latter was mutationpositive. The MSH2 G322D mutation was identified in a female patient diagnosed with PC at the age of 69, patient who had previously developed breast and colorectal cancer at the age of 62 and 69, respectively. The proband’s mother had died of breast cancer at 80 years of age (Fig. 1c). The novel MSH2-E205Q and MSH2-V367I variants were detected in a female patient who was diagnosed with PC at the age of 59. Mutation analysis was not possible in family members since they had died of endometrial cancer diagnosed at the age of 45 and colorectal cancer at the age of 61 (Fig. 1d). Neither variant was found in 100 healthy

123

controls. None of the patients carried MSH6 mutations or variants. RNA analysis RT-PCR analysis showed no difference in the PCR product in MLH1 K618A, MSH2 Q402X and MSH2 E205Q/V367I carriers versus controls in terms of size, nor the presence of additional PCR products by agarose gel analysis (data not shown). The patient who harbored the G322D variant could not be tested because RNA was unavailable. Sequencing results showed biallelic RNA expression, with no evidence of second mutations or of alternative splicing products, which may have been undetectable by agarose gel analysis. These results indicate that no gross RNA rearrangements, including exon skipping were conferred by these variants. No relevant RNA decay resulted from the MSH2-Q402X mutation. Protein expression MSH2 WT, E205Q and V367I were co-expressed with WT MSH6, to give rise to the functional MSH2-MSH6 heterodimer MutSa. All MutSa dimers were produced


551

In vitro repair efficiency of MSH2 variants

Discussion

In vitro MMR assay was used to assess the repair activity of MSH2 E205Q and V367I as compared to WT MutSa. WT and both mutated proteins were able to partially convert the G•T mismatch to the G•C base pair, allowing the BglII digestion of the substrate molecule (Fig. 2b), visualized as double digestion of the repair substrate. Thus, both MSH2 E205Q and V367I were functional in the in

Approximately 10% of PC patients are estimated to have an inherited predisposition to the disease [2]. A fraction of these patients belong to kindreds with familial pancreatic cancer, but most are affected by inherited cancer syndromes predisposing them to a wider spectrum of tumor types. While the cumulative risk for the development of PC until the age of 70 years in these syndromes varies between 3 and 40% [2, 13], it is generally considered to be \5% in LS kindreds [2, 14]. According to a recent study [15], PC risk is increased sevenfold in both MLH1 and MSH2 carriers belonging to LS families, especially at young ages, as previously noted by Lynch et al. [16] as early as in 1991. The contribution of LS susceptibility genes in consecutive series of PC has been rarely investigated. In a study by Lal et al. 17], samples for immunohistochemistry were available for only 3/9 patients who met the criteria for LS, and these were tested for MSH2/MLH1 expression. As all three tumors demonstrated intact expression of these mismatch repair proteins, the cases were not tested for presence of germ-line mutations in MSH2 and MLH1. One of the primary aims of our study was therefore to investigate how many PC patients in our unselected consecutively enrolled series had a personal or family cancer history that could be classified as s-LS. Our definition of sLS is a modified version of the revised Bethesda criteria [5] that takes into account PC as the main tumor type. Because tumor specimens were not available for testing, presence of MSI was not one of our criteria and the PC patients who met them could only be suspected of being affected by LS. We found that 19/135 (14%) of our PC patients were sLS, based on their personal and family cancer history. This high frequency is likely explained by the specific criteria we adopted. Lymphoblastoid cell lines as a source of genomic DNA were available for 11 of them. The eight sLS cases who were not tested were similar to those tested for familial characteristics: the frequency of s-LS was similar in tested and untested cases (11/87 = 12.6% e 8/ 48 = 16.7%, P = 0.70). Of the 11 s-LS PC patients for whom lymphoblastoid cell-lines were available and who could thus be tested for presence of germline mutations, four were found to carry MLH1 or MSH2 variants, all of which of debated or unknown functional significance, except for the Q402X mutation. Q402X is a stop mutation that was previously identified in an LS kindred from Poland [18] and we have identified it here for the first time here in association with PC.

MutSα V3 367I

MutSα S WT

A

urifieed Mu utSα pu

vitro MMR assay. LoVo NE without recombinant MutSa failed to repair the substrate molecule.

MutSα S E205 E Q

successfully and efficiently with over 300 ng/ll TE concentrations, as concluded by comparing the band intensities to the purified MutSa (Fig. 2a).

MSH6

MSH2

B 1. 2 2. 3. 4.

LoVo NE (MSH2 -/-) LoVo + MutSα WT LoVo + MutSα E205Q LoVo + MutSα V367I

Unrepaire i d R Repaired i d

-

+

+

+

Fig. 2 a The Western blot of Sf 9 TEs containing either WT MutSa, MutSa E205Q or MutSa V367I. 1 ll of TE derived from baculovirusinfected Sf 9 cells were run on an SDS–PAGE gel, blotted to nitrocellulose membrane and detected with MSH2 and MSH6 antibodies. 300 ng of purified MutSa preparation was used as a reference for the concentration of MutSa in TE. All MutSa preparations were expressed efficiently, with [300 ng recombinant MutSa/ll TE concentrations. b Mismatch repair efficiency of MutSa E205Q or MutSa V367I. LoVo (MSH2-/-) NE was incubated with MutSa and nicked heteroduplex plasmid substrates. The efficient complementation of LoVo NE by recombinant MutSa results in the correction of the mismatch in the substrate plasmid, restoring a BglII restriction site. Successful repair is visualized as BsaI-BglII doubledigested 1,830 and 1,360 bp DNA fragments, whereas unrepaired plasmids are only cut by BsaI and migrate at 3,190 bp. Both MutSa E205Q and MutSa V367I proteins repaired the heteroduplex plasmid, whereas LoVo NE alone (negative control) did not exhibit detectable repair

123

552

Whether K618A is pathogenic remains a debated question. According to Perera and colleagues it decreases the stability of the protein [19], but the functional assays conducted by Raevaara [20] indicate that it should be classified as non pathogenic. We found this variant in the unaffected branch of a kindred, suggesting that it may be a polymorphism or possibly a low penetrance variant. Indeed, our findings seem to confirm the results of Blasi et al. [21] who found that K618A did not segregate with the disease in one Italian LS family and derived from the unaffected father in another, leading them to conclude that this variant may behave as a low penetrance mutation. We tested MSI in the liver metastasis of the PC patient for this particular variant, and found that expression was intact and there was no MSI (data not shown), apparently confirming that this variant does not impact directly on protein function, while other studies obtained contrasting results. However, the same analysis should be performed on the primary tumor to confirm this finding. The G322D variant is also much discussed: in yeast it has been considered to be non pathogenic [22] but in an assay of mismatch release it showed slightly, if not significantly, reduced mismatch binding and/or release efficiencies compared to wt [10]. This variant has also been associated with breast cancer [23] as indeed seen in our patient who had a personal history of breast cancer and whose mother developed breast cancer at the age of 80. The third aim of this study was to determine the pathogenicity of any novel non-truncating variants identified among these patients. The two new variants that we detected in the same patient (E205Q e V367I) do not to seem pathogenic, although, as they were analyzed separately, a subtle combined effect of the two cannot be ruled out. RNA analysis revealed that no gross RNA rearrangements, including exon skipping were conferred by these and other variants tested. Overall, one variant (Q402X) of the four detected seems to be pathogenic. This frequency [1/11 patients with s-LS (9%), and 1/87 consecutively enrolled PC patients (1.1%)] is consistent with other estimates on the contribution of mutations in MMR genes to susceptibility to PC to be less than 5% [2]. It must be noted, however, that those results derived from familial studies of high-risk patients, while our PC patients were consecutively enrolled within the framework of a case–control study. One limitation of this study is that IHC and MSI testing were conducted for a single sample (K618A variant), which however, showed that MMR protein expression was conserved and MSI was absent. IHC and MSI analyses were not a goal of our study which proceeded through comprehensive mutational analysis of the candidate genes. One other limitation is that MLPA analysis was not

123

S. Gargiulo et al.

performed; therefore we cannot rule out the presence of large deletions encompassing MMR genes. While the identification of clearly causative mutations such as Q402X, will help us improve genetic counseling, surveillance and follow-up of carriers in LS families, the functional significance of rare, low-penetrance variants like K618A and G322D that, in conjunction with specific MMR polymorphisms [24] may be associated with clinical outcome, needs to be further investigated before finding application in clinical practice. Acknowledgments This work was supported by the 2007 Italian Ministry of Health DGRST.4/4235-P1.9.A.B. We wish to thank the Galliera Genetic Bank—Network of Telethon Genetic Biobanks (project GTB07001) for providing lymphoblastoid cell lines.

References 1. Brand RE, Lynch HT (2006) Genotype/phenotype of familial pancreatic cancer. Endocrinol Metab Clin North Am 35(2): 405–415 2. Habbe N, Langer P, Sina-Frey M, Bartsch DK (2006) Familial pancreatic cancer syndromes. Endocrinol Metab Clin North Am 35(2):417–430 3. Lynch HT, de la Chapelle A (2003) Hereditary colorectal cancer. N Engl J Med 348:919–932 4. Vasen HF, Watson P, Mecklin J, Lynch HT (1999) New clinical criteria for hereditary nonpolyposis colorectal cancer (HNPCC, Lynch syndrome) proposed by the International Collaborative group on HNPCC. Gastroenterology 116:1453–1456 5. Umar A, Boland CR, Terdiman JP, Syngal S, de la Chapelle A, Ru¨schoff J, Fishel R, Lindor NM, Burgart LJ, Hamelin R, Hamilton SR, Hiatt RA, Jass J, Lindblom A, Lynch HT, Peltomaki P, Ramsey SD, Rodriguez-Bigas MA, Vasen HF, Hawk ET, Barrett JC, Freedman AN, Srivastava SJ (2004) Revised Bethesda Guidelines for hereditary nonpolyposis colorectal cancer (Lynch syndrome) and microsatellite instability. J Natl Cancer Inst 96(4):261–268 6. Montera M, Resta N, Simone C, Guanti G, Marchese C, Civitelli S, Mancini A, Pozzi S, De Salvo L, Bruzzone D, Donadini A, Romio L, Mareni C (2000) Mutational germline analysis of hMSH2 and hMLH1 genes in early onset colorectal cancer patients. J Med Genet 37(7):E7 7. Baudhuin LM, Burgart LJ, Leontovich O, Thibodeau SN (2005) Use of microsatellite instability and immunohistochemistry testing for the identification of individuals at risk for Lynch syndrome. Fam Cancer 4(3):255–265 Review 8. Kariola R, Raevaara TE, Lo¨nnqvist KE, Nystro¨m-Lahti M (2002) Functional analysis of MSH6 mutations linked to kindreds with putative hereditary non-polyposis colorectal cancer syndrome. Hum Mol Genet 11(11):1303–1310 9. Ollila S, Sarantaus L, Kariola R, Chan P, Hampel H, HolinskiFeder E, Macrae F, Kohonen-Corish M, Gerdes AM, Peltoma¨ki P, Mangold E, de la Chapelle A, Greenblatt M, Nystro¨m M (2006) Pathogenicity of MSH2 missense mutations is typically associated with impaired repair capability of the mutated protein. Gastroenterology 131(5):1408–1417 [Epub 2006 Aug 22] 10. Ollila S, Dermadi Bebek D, Jiricny J, Nystro¨m M (2008) Mechanisms of pathogenicity in human MSH2 missense mutants. Hum Mutat 29(11):1355–1363 11. Nystro¨-Lahti M, Perrera C, Ra¨schle M, Panyushkina-Seiler E, Marra G, Curci A, Quaresima B, Costanzo F, D’Urso M, Venuta


12.

13.

14. 15.

16.

17.

18.

S, Jiricny J (2002) Functional analysis of MLH1 mutations linked to hereditary nonpolyposis colon cancer. Genes Chromosomes Cancer 33(2):160–167 Ghiorzo P, Pastorino L, Bonelli L, Cusano R, Nicora A, Zupo S, Queirolo P, Sertoli M, Pugliese V, Bianchi-Scarra` G (2004) INK4/ARF germline alterations in pancreatic cancer patients. Ann Oncol 15(1):70–78 Bartsch DK, Sina-Frey M, Lang S, Wild A, Gerdes B, Barth P, Kress R, Gru¨tzmann R, Colombo-Benkmann M, Ziegler A, Hahn SA, Rothmund M, Rieder H (2002) CDKN2A germline mutations in familial pancreatic cancer. Ann Surg 236(6):730–737 Lynch HT, Brand RE, Deters CA, Shaw TG, Lynch JF (2001) Hereditary pancreatic cancer. Pancreatology 1(5):466–471 Geary J, Sasieni P, Houlston R, Izatt L, Eeles R, Payne SJ, Fisher S, Hodgson SV (2008) Gene-related cancer spectrum in families with hereditary non-polyposis colorectal cancer (HNPCC). Fam Cancer 7(2):163–172 [Epub 2007 Oct 16] Lynch HT, Smyrk T, Watson P, Lanspa SJ, Boman BM, Lynch PM, Lynch JF, Cavalieri J (1991) Hereditary colorectal cancer. Semin Oncol 18(4):337–366 Lal G, Liu G, Schmocker B, Kaurah P, Ozcelik H, Narod SA, Redston M, Gallinger S (2000) Inherited predisposition to pancreatic adenocarcinoma: role of family history and germ-line p16, BRCA1, and BRCA2 mutations. Cancer Res 60(2):409–416 Kurzawski G, Suchy J, Lener M, Kłujszo-Grabowska E, Kładny J, Safranow K, Jakubowska K, Jakubowska A, Huzarski T, Byrski T, Debniak T, Cybulski C, Gronwald J, Oszurek O, Oszutowska D, Kowalska E, Go´z´dz´ S, Niepsuj S, Słomski R, Pławski A, Łacka-Wojciechowska A, Rozmiarek A, Fiszer-Maliszewska Ł, Bebenek M, Sorokin D, Sasiadek MM, Stembalska A, Grzebieniak Z, Kilar E, Stawicka M, Godlewski D, Richter P, Brozek I,

553

19.

20.

21.

22.

23.

24.

Wysocka B, Limon J, Jawien´ A, Banaszkiewicz Z, Janiszewska H, Kowalczyk J, Czudowska D, Scott RJ, Lubin´ski J (2006) Germline MSH2 and MLH1 mutational spectrum including large rearrangements in HNPCC families from Poland (update study). Clin Genet 69(1):40–47 Perera S, Bapat B (2008) The MLH1 variants p.Arg265Cys and p.Lys618Ala affect protein stability while p.Leu749Gln affects heterodimer formation. Hum Mutat 29(2):332 Raevaara TE, Korhonen MK, Lohi H, Hampel H, Lynch E, Lo¨nnqvist KE, Holinski-Feder E, Sutter C, McKinnon W, Duraisamy S, Gerdes AM, Peltoma¨ki P, Kohonen-Ccorish M, Mangold E, Macrae F, Greenblatt M, de la Chapelle A, Nystro¨m M (2005) Functional significance and clinical phenotype of nontruncating mismatch repair variants of MLH1. Gastroenterology 129(2):537–549 Blasi MF, Ventura I, Aquilina G, Degan P, Bertario L, Bassi C, Radice P, Bignami M (2006) A human cell-based assay to evaluate the effects of alterations in the MLH1 mismatch repair gene. Cancer Res 66(18):9036 Gammie AE, Erdeniz N, Beaver J, Devlin B, Nanji A, Rose MD (2007) Functional characterization of pathogenic human MSH2 missense mutations in Saccharomyces cerevisiae. Genetics 177(2):707–721 Epub 2007 Aug 24 Poplawski T, Zadrozny M, Kolacinska A, Rykala J, Morawiec Z, Blasiak J (2005) Polymorphisms of the DNA mismatch repair gene HMSH2 in breast cancer occurence and progression. Breast Cancer Res Treat 94(3):199–204 Dong X, Jiao L, Li Y, Evans DB, Wang H, Hess KR, Abbruzzese JL, Li D (2009) Significant associations of mismatch repair gene polymorphisms with clinical outcome of pancreatic cancer. J Clin Oncol 27(10):1592–1599 Epub 2009 Feb 23

123