6 pages

Germline mutations in RAD51D confer susceptibility to ovarian cancer

of 6
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Recently, RAD51C mutations were identified in families with breast and ovarian cancer1. This observation prompted us to investigate the role of RAD51D in cancer susceptibility. We identified eight inactivating RAD51D mutations in unrelated
     ©   2   0   1   1   N  a   t  u  r  e   A  m  e  r   i  c  a ,   I  n  c .   A   l   l  r   i  g   h   t  s  r  e  s  e  r  v  e   d . NATURE GENETICS   ADVANCE ONLINE PUBLICATION 1 LETTERS Recently, RAD51C  mutations were identified in families withbreast and ovarian cancer 1 . This observation prompted us toinvestigate the role of  RAD51D in cancer susceptibility. Weidentified eight inactivating RAD51D mutations in unrelatedindividuals from 911 breast-ovarian cancer families compared withone inactivating mutation identified in 1,060 controls ( P = 0.01).The association found here was principally with ovarian cancer,with three mutations identified in the 59 pedigrees with three ormore individuals with ovarian cancer ( P = 0.0005). The relative riskof ovarian cancer for RAD51D mutation carriers was estimatedto be 6.30 (95% CI 2.86–13.85, P = 4.8 × 10 –6 ). By contrast, weestimated the relative risk of breast cancer to be 1.32 (95% CI0.59–2.96, P = 0.50). These data indicate that RAD51D mutationtesting may have clinical utility in individuals with ovariancancer and their families. Moreover, we show that cells deficientin RAD51D are sensitive to treatment with a PARP inhibitor,suggesting a possible therapeutic approach for cancers arising in RAD51D mutation carriers. Homologous recombination is a mechanism for repairing stalledreplication forks, DNA interstrand crosslinks and double-strandbreaks 2 . Constitutional inactivating mutations in several genes thatencode proteins crucial for DNA repair by homologous recombinationhave been shown to predispose to cancer 3 . In particular, these muta-tions have a strong association with female cancers, and mutations ingenes such as BRCA1 , BRCA2 ,  ATM  , BRIP1 , CHEK2 , PALB2 , RAD50  and RAD51C have been shown to confer susceptibility to breast and/orovarian cancer 1,4 . Indeed, the analysis of families with breast andovarian cancer was crucial to mapping BRCA1 (ref. 5). For many years, it was widely believed that the genetic contribution to familieswith breast and ovarian cancer was largely attributable to mutationsin BRCA1 and BRCA2 (refs. 6–8). Last year, however, researchers 1  identified mutations in RAD51C in breast-ovarian cancer families.This suggested that analysis of such families may still have utility inthe discovery of cancer predisposition genes.In eukaryotic cells, DNA repair by homologous recombinationinvolves several proteins, of which a central player is the DNA Germline mutations in RAD51D confer susceptibility toovarian cancer Chey Loveday  1,30 , Clare Turnbull 1,30 , Emma Ramsay  1 , Deborah Hughes 1 , Elise Ruark  1 , Jessica R Frankum 2 ,Georgina Bowden 1 , Bolot Kalmyrzaev  1 , Margaret Warren-Perry  1 , Katie Snape 1 , Julian W Adlard 3 ,Julian Barwell 4 , Jonathan Berg 5 , Angela F Brady  6 , Carole Brewer 7 , Glen Brice 8 , Cyril Chapman 9 , Jackie Cook  10 ,Rosemarie Davidson 11 , Alan Donaldson 12 , Fiona Douglas 13 , Lynn Greenhalgh 14 , Alex Henderson 15 , Louise Izatt 16 ,Ajith Kumar 17 , Fiona Lalloo 18 , Zosia Miedzybrodzka 19 , Patrick J Morrison 20 , Joan Paterson 21 , Mary Porteous 22 ,Mark T Rogers 23 , Susan Shanley  24 , Lisa Walker 25 , Breast Cancer Susceptibility Collaboration (UK) 26 ,Diana Eccles 27 , D Gareth Evans 28 , Anthony Renwick  1 , Sheila Seal 1 , Christopher J Lord 2 , Alan Ashworth 2 ,Jorge S Reis-Filho 2 , Antonis C Antoniou 29 & Nazneen Rahman 1 1 Section of Cancer Genetics, The Institute of Cancer Research, Sutton, UK. 2 The Breakthrough Breast Cancer Research Centre, The Institute of Cancer Research,London, UK. 3 Yorkshire Regional Centre for Cancer Treatment, Cookridge Hospital, Leeds, UK. 4 Leicestershire Genetics Centre, University Hospitals of LeicesterNational Health Service (NHS) Trust, Leicester, UK. 5 Human Genetics, Division of Medical Sciences, University of Dundee, Dundee, UK. 6 North West ThamesRegional Genetics Service, Kennedy Galton Centre, London, UK. 7 Peninsula Regional Genetics Service, Royal Devon & Exeter Hospital, Exeter, UK. 8 South WestThames Regional Genetics Service, St. George’s Hospital, London, UK. 9 West Midlands Regional Genetics Service, Birmingham Women’s Hospital, Birmingham,UK. 10 Sheffield Regional Genetics Service, Sheffield Children’s NHS Foundation Trust, Sheffield, UK. 11 West of Scotland Regional Genetics Service, FergusonSmithCentre for Clinical Genetics, Glasgow, UK. 12 South Western Regional Genetics Service, University Hospitals of Bristol NHS Foundation Trust, Bristol, UK. 13 NorthernGenetics Service, Newcastle upon Tyne Hospitals NHS Foundation Trust, Newcastle upon Tyne, UK. 14 Cheshire and Merseyside Clinical Genetics Service, AlderHey Children’s NHS Foundation Trust, Liverpool, UK. 15 Northern Genetics Service (Cumbria), Newcastle upon Tyne Hospitals NHS Trust, Newcastle upon Tyne, UK. 16 South East Thames Regional Genetics Service, Guy  s and St. Thomas NHS Foundation Trust, London, UK. 17 North East Thames Regional Genetics Service, GreatOrmond St. Hospital, London, UK. 18 University Department of Medical Genetics & Regional Genetics Service, St. Mary’s Hospital, Manchester, UK. 19 University ofAberdeen and North of Scotland Clinical Genetics Service, Aberdeen Royal Infirmary, Aberdeen, UK. 20 Northern Ireland Regional Genetics Service, Belfast Healthand Social Care (HSC) Trust & Department of Medical Genetics, Queen’s University Belfast, Belfast, UK. 21 East Anglian Regional Genetics Service, CambridgeUniversity Hospitals NHS Foundation Trust, Cambridge, UK. 22 South East of Scotland Clinical Genetics Service, Western General Hospital, Edinburgh, UK. 23 AllWales Medical Genetics Service, University Hospital of Wales, Cardiff, UK. 24 Royal Marsden NHS Foundation Trust, London, UK. 25 Oxford Regional Genetics Service,Oxford Radcliffe Hospitals NHS Trust, Oxford, UK. 26 A full list of members appears in the Supplementary Note . 27 Faculty of Medicine, University of Southampton,Southampton University Hospitals NHS Trust, Southampton, UK. 28 University Department of Medical Genetics & Regional Genetics Service, St. Mary’s Hospital,Manchester, UK. 29 Center for Cancer Genetic Epidemiology, Department of Public Health and Primary Care, University of Cambridge, Cambridge, UK. 30 These authorscontributed equally to this work. Correspondence should be addressed to N.R. ( 3 March; accepted 1 July; published online 7 August 2011;doi:10.1038/ng.893     ©   2   0   1   1   N  a   t  u  r  e   A  m  e  r   i  c  a ,   I  n  c .   A   l   l  r   i  g   h   t  s  r  e  s  e  r  v  e   d . 2 ADVANCE ONLINE PUBLICATION NATURE GENETICS LETTERS recombinase RAD51, the ortholog of bacterial recA 9 . RAD51 forms heli-cal filaments on DNA and catalyzes DNA strand invasion and exchange.Many other proteins are involved in these processes, including five RAD51paralogs (RAD51B, RAD51C, RAD51D,XRCC2 and XRCC3) 10 . Here, through a case-control mutation study, we show that mutationsin RAD51D (also known as RAD51L3 ) predis-pose to cancer in humans.We sequenced the full coding sequenceand intron-exon boundaries of  RAD51D inDNA from unrelated probands from 911breast-ovarian cancer families and 1,060 pop-ulation controls ( Supplementary Table 1 ).The breast-ovarian cancer families includedat least one case of breast cancer and at leastone case of ovarian cancer and all familieswere negative for mutations in BRCA1 and BRCA2 ( Supplementary Table 2 ).We identified inactivating mutations in RAD51D in 8 of the 911 cases and in 1 of the1,060 controls ( P  = 0.01) ( Table 1 ). The muta-tions were not equally distributed within theseries, with a higher prevalence of mutationspresent in families with more than one case of ovarian cancer: we detected four mutations in 235 families with two ormore cases of ovarian cancer ( P  = 0.005) and three mutations in the 59families with three or more cases of ovarian cancer ( P  = 0.0005) ( Fig. 1 ). Table 1 Cancer history and pathology in RAD51D  mutation carriers Family IDMutation; protein alteration a Person IDCancer history (age) b PathologyTumor analysisFam1c.363delA1Breast cancer, left (34)Invasive ductal carcinoma of no special type, grade 3NABreast cancer, right (52)Invasive ductal carcinoma of no special type, grade 3Loss of wild-type alleleFam2c.803G>A; p.Trp268X1Ovarian cancer (58)Bilateral serous adenocarcinomaLoss of wild-type alleleFam3c.556C>T; p.Arg186X1Ovarian cancer (38)NANA2Breast cancer (39)High grade ductal comedo carcinoma in situ  NA3Breast cancer (58)Invasive carcinoma with medullary featuresNA4Breast cancer (53)Invasive ductal carcinoma of no special typeNAFam4c.480+1G>A1Breast cancer (51)Invasive ductal carcinoma of no special type, grade 3NAFam5c.345G>C; p.Gln115His c 1Ovarian cancer (45)Bilateral serous adenocarcinomaNA2Ovarian cancer (74)NANAFam6c.556C>T; p.Arg186X1Breast cancer (35)Invasive ductal carcinoma of no special type, grade 3NAFam7c.757C>T; p.Arg253X1Ovarian cancer (51)Differentiated endometrioid adenocarcinomaNA2Breast cancer (47)NANAFam8c.270_271dupTA1Ovarian cancer (58)Differentiated adenocarcinomaLoss of mutant alleleBreast cancer (65)Invasive ductal carcinoma of no special type, grade 3Reduction of wild-type alleleControlc.748delCNANANA NA, not available. a Mutation nomenclature corresponds to Ensembl Transcript ID ENST00000345365. b Age at which cancer occurred in years. c This mutation is at the final base of exon 4, disrupts the splice siteand results in skipping of exons 3 and 4. Person IDs correspond to those shown in Figure 1 . OCCRC OCBC bilatBC 79LC 54BC 70OC1OC 58p.Trp268XPrC 68CRC 40PrC 60BC bilat.34 52c.363delA 1OC 56Family 1Family 2CRCCRC 51OCOC 65OC 62BC 58CRC 52BC 39BC 53OC 49p.Arg186X234Family 3OC 381p.Arg186Xp.Arg186Xp.Arg186XOC 54NHL 49BC 43BC 511Family 4c.480+1G>A OC 74OC 45OC 37CRC 74PaC 75CRC1BC 36OCOC 46BC 35CRC1Family 5Family 6p.Gln115Hisp.Gln115Hisp.Arg186XOC 58BC 651PaC 78BC 47OC 511Family 7Family 82p.Arg253Xp.Arg253Xc.270_271dupTA Figure 1 Abridged pedigrees of eight familieswith RAD51D  mutations. Individuals withovarian cancer are shown as red circles;individuals with breast cancer are shown asblack circles. Other cancers are shown asunfilled circles or squares. Where known,the age of cancer diagnosis is listed underthe individual, with two ages given formetachronous bilateral breast cancers. Therelevant RAD51D  mutation is listed underthe affected individuals analyzed, but notthe unaffected individuals, to preserveconfidentiality. BC, breast cancer; BC bilat.,bilateral breast cancer; OC, ovarian cancer;CRC, colorectal cancer; LC, lung cancer; NHL,non-Hodgkin lymphoma; PaC, pancreaticcancer; PrC, prostate cancer.     ©   2   0   1   1   N  a   t  u  r  e   A  m  e  r   i  c  a ,   I  n  c .   A   l   l  r   i  g   h   t  s  r  e  s  e  r  v  e   d . NATURE GENETICS   ADVANCE ONLINE PUBLICATION 3 LETTERS All the mutations are predicted to result in protein truncationthrough frameshifting insertions or deletions ( n = 3), the genera-tion of nonsense codons ( n = 4) or splice defects ( n = 2) ( Table 1 ).We also identified 5 intronic, 3 synonymous and 15 nonsynonymous variants. Three coding variants, rs9901455 (p.Ser78Ser), rs4796033(p.Arg165Gln) and rs28363284 (p.Glu233Gly), have minor allele fre-quencies >1%; none of these variants was associated with risk of breastor ovarian cancer in our dataset ( Supplementary Table 3 ). Of theremaining rare variants, three were present in both cases and controls,nine were detected in a single case and eight were detected in a singlecontrol ( Supplementary Table 4 ). Thus, there was no overall differencein the frequency of nontruncating RAD51D variants between casesand controls. Moreover, there was no difference in the position or pre-dicted functional effects of these variants, and it is noteworthy that anequal number ( n = 5) of nonsynonymous variants detected in casesand controls are predicted to affect function ( Supplementary Fig. 1  and Supplementary Table 4 ). These data indicate that mutations thatresult in inactivation of RAD51D function predispose to cancer butthat other variants are likely to be predominantly nonpathogenic.We tested for the family mutations in samples from 13 relatives. Thisshowed that five of five individuals affected with ovarian or breast cancercarried the family mutation, whereas six of eight unaffected relativesdid not carry the family mutation. Severalother cancers were present in relatives, suchas pancreatic, prostate and colorectal cancer( Fig. 1 ). However, the mutation status of theseindividuals is not known, and additionalstudies will be required to evaluate whether RAD51D mutations predispose to other can-cers. Pathology information was available forfour ovarian cancers from RAD51D mutationcarriers; three of the cancers were serous ade-nocarcinoma and one was an endometrioidcancer. Pathology information was availablefor eight breast cancers, of which seven wereductal in srcin and one was a carcinomawith medullary features. Receptor status wasavailable from five breast cancers, of whichthree were estrogen-receptor positive and twowere estrogen-receptor negative. Tumor mate-rial was available from two ovarian cancersand two breast cancers. We detected loss of the wild-type allele in oneovarian and one breast cancer and reduction of the proportion of thewild-type allele in a further breast cancer. In the last ovarian cancer,the mutant allele was lost and the wild-type allele was retained ( Table 1  and Supplementary Fig. 2 ).These characteristics are typical of the intermediate-penetrancecancer predisposition genes that we and others have described inbreast cancer 1,4,11–14 . To estimate directly the risks associated with RAD51D mutations, we performed modified segregation analysis by modeling the risks of ovarian and breast cancer simultaneously andincorporating the information from the controls and full pedigreesof both mutation-positive and mutation-negative breast-ovariancancer families. The ovarian cancer relative risk for RAD51D  mutation carriers was estimated to be 6.30 (95% CI 2.86–13.85, P  = 4.8 × 10 –6 ) ( Fig. 2 ). By contrast, the association with breastcancer risk was not statistically significant (relative risk = 1.32, 95%CI 0.59–2.96, P  = 0.50).To further explore the role of  RAD51D mutations in breast cancerrisk, we sequenced the gene in an additional series of 737 unrelatedindividuals from pedigrees in which there was familial breast cancerbut no ovarian cancer. We did not identify any inactivating mutations(0 out of 737 cases compared to 1 out of 1,060 controls had inacti- vating mutations; P  = 1.0). Although at first glance these data may seem surprising, they are consistent with the results of the segregationanalysis. In particular, if  RAD51D mutations confer a sizeable rela-tive risk of ovarian cancer but only a small or no increase in breastcancer risk, the frequency of  RAD51D mutations in a series of breastcancer families selected on the basis of not containing ovarian cancerwould be anticipated to be very low. The data are also consistent withthe detection of  RAD51D mutations in seven individuals with breastcancer in the breast-ovarian cancer families, as we specifically ascer-tained the ovarian cancer cases because of their close family history of breast cancer. This selection will inevitably result in an enrichmentof breast cancer in relatives of  RAD51D -mutation–positive ovariancancer cases irrespective of whether or not such mutations confera risk of breast cancer. To formally refine the risk of breast cancerassociated with RAD51D mutations will likely be very challengingbecause the population frequency of  RAD51D mutations is so low.Assuming a population mutation frequency of 0.1% and a relative risk of breast cancer of 1.3, full mutational analysis of  RAD51D in 275,000cases and 275,000 controls would be required to have 90% power toshow the association.    C  u  m  u   l  a   t   i  v  e  r   i  s   k 030 40 50Age60 70 BRCA1BRCA2RAD51D Population Figure 2 Average age-related cumulative risk of ovarian cancer in RAD51D  mutation carriers, BRCA1 and BRCA2  mutation carriers 22 andthe general population 23 . 1.0 a b c    S  u  r  v   i  v   i  n  g   f  r  a  c   t   i  o  n 0.200 –9 –8 –7Olaparib (M)–6siCONTROLsiBRCA2siRAD51D OLIGO1RAD51D WTRAD51D MUTsiRAD51D OLIGO2siRAD51D POOL–    S  u  r  v   i  v   i  n  g   f  r  a  c   t   i  o  n 0.200 –9 –8 –7Olaparib (M)–6 –    S  u  r  v   i  v   i  n  g   f  r  a  c   t   i  o  n 0.200 –9 –8 –7Olaparib (M)–6 –5 Figure 3 Effect of RAD51D  silencing on olaparib sensitivity. ( a , b ) We transfected CAL51 ( a ) orMCF7 ( b ) cells with siCONTROL, siRNA directed against RAD51D  or siRNA directed against BRCA2  . For siRNA targeting RAD51D  , cells were transfected with one of two individual siRNAs ora pool of both siRNAs combined. We then treated transfected cells with olaparib for 7 days beforeassaying for cell viability. ( c ) We treated wild-type (WT) CHO cells or CHO cells mutated (MUT) in RAD51D  with olaparib for 7 days before assaying for cell viability. M, molar concentration.     ©   2   0   1   1   N  a   t  u  r  e   A  m  e  r   i  c  a ,   I  n  c .   A   l   l  r   i  g   h   t  s  r  e  s  e  r  v  e   d . 4 ADVANCE ONLINE PUBLICATION NATURE GENETICS LETTERS Our data clearly show that RAD51D is an ovarian cancer predispo-sition gene, but further studies in familial and sporadic ovarian cancerseries would be of value to further clarify the risks of ovarian cancer. RAD51D mutation analysis in individuals with Fanconi anemia andFanconi-like disorders would also be of interest, given that biallelicmutations in BRCA2 , PALB2 , BRIP1 and RAD51C have been shownto cause these phenotypes 15–18 .Our discovery has potential clinical utility both for individuals withcancer and their relatives. For example, cancer patients with RAD51D  mutations may benefit from specific therapies such as poly (ADP-ribose) polymerase (PARP) inhibitors, which have shown efficacy inpatients with impaired homologous recombination caused by muta-tions in BRCA1 or BRCA2 (ref. 19). To investigate this, we used RNAinterference (RNAi) and assessed the relationship between RAD51D  loss of function and the sensitivity of tumor cells to a clinical PARPinhibitor, olaparib (AstraZeneca). Short interfering RNAi (siRNAi)reagents targeting RAD51D caused olaparib sensitivity of a magnitudesimilar to that achieved following silencing of  BRCA2 ( Fig. 3a , b ), anobservation in keeping with the homologous recombination defectobserved in Rad51d  -null rodent cell lines 20 . To extend this analysis,we also observed the RAD51D -selective effect of olaparib in RAD51D -deficient Chinese hamster ovary (CHO) cells in which both alleles of  RAD51D had been rendered dysfunctional by gene targeting ( Fig. 3c ) 20 .These data suggest that PARP inhibitors may have clinical utility inindividuals with RAD51D mutations. We estimate that only ~0.6% of unselected individuals with ovarian cancer will harbor RAD51D muta-tions, but as we enter an era in which genetic testing will become rou-tine, such individuals will be readily identifiable. Their identificationwill also be of potential value to female relatives, as those relatives withmutations will be, on average, at an approximately sixfold increased risk of ovarian cancer, which equates to an ~10% cumulative risk by age 80.An appreciable proportion of women at this level of risk may considerstrategies such as laparoscopic oophorectomy, which is well toleratedand is undertaken in many women with BRCA mutations 21 . URLs. Centre for Longitudinal Studies, National Child DevelopmentStudy,;Mutation Surveyor software, METHODS Methods and any associated references are available in the online version of the paper at Accession codes. RAD51D mutation nomenclature corresponds toEnsembl Transcript ID ENST00000345365. Note: Supplementary information is available on theNature Geneticswebsite. ACKNOWLEDGMENTS We thank all the subjects and families that participated in the research. We thank A. Hall, D. Dudakia, J. Bull, R. Linger and A. Zachariou for their assistance inrecruitment, B. Ebbs for assistance in DNA extraction and running the ABIsequencers, L. Thompson for the provision of cell lines and A. Strydom forassistance in preparing the manuscript. We are very grateful to all the cliniciansand counselors in the Breast Cancer Susceptibility Collaboration UK (BCSC)that have contributed to the recruitment and collection of the Familial BreastCancer Study (FBCS) samples. The full list of BCSC contributors is providedin the Supplementary Note . This work was funded by Cancer Research UK(C8620/A8372 and C8620/A8857), US Military Acquisition (ACQ) Activity,Era of Hope Award (W81XWH-05-1-0204), Breakthrough Breast Cancer and theInstitute of Cancer Research (UK). We acknowledge NHS funding to the RoyalMarsden/Institute of Cancer Research National Institute for Health Research(NIHR) Specialist Cancer Biomedical Research Centre. C.T. is a Medical ResearchCouncil (MRC)-funded Clinical Research Fellow. A.C.A. is a Cancer ResearchUK Senior Cancer Research Fellow (C12292/A11174). We acknowledge the useof DNA from the British 1958 Birth Cohort collection funded by the MRC grantG0000934 and the Wellcome Trust grant 068545/Z/02. AUTHOR CONTRIBUTIONS N.R., C.L. and C.T. designed the experiment. M.W.-P., C.T. and N.R. coordinatedrecruitment to the FBCS. J.W.A., J. Barwell, J. Berg, A.F.B., C.B., G. Brice, C.C.,J.C., R.D., A.D., F.D., D.G.E., D.E., L.G., A.H., L.I., A.K., F.L., Z.M., P.J.M., J.P., M.P.,M.T.R., S. Shanley and L.W. coordinated the FBCS sample recruitment from theirrespective Genetics centers. C.L., E. Ramsay, D.H., G. Bowden, B.K., K.S., A.R.and S. Seal performed sequencing of  RAD51D . J.R.F., C.J.L. and A.A. designed andconducted drug sensitivity experiments. J.S.R.-F. undertook examination anddissection of pathological specimens. C.T., E. Ruark and A.C.A. performedstatistical analyses. C.L., C.T. and N.R. drafted the manuscript with substantialinput from D.G.E., D.E., A.C.A., A.A. and J.S.R.-F. C.T. and N.R. oversaw andmanaged all aspects of the study. COMPETING FINANCIAL INTERESTS The authors declare no competing financial interests. Published online at and permissions information is available online at reprints/index.html. 1. Meindl, A. et al. Germline mutations in breast and ovarian cancer pedigrees establish RAD51C  as a human cancer susceptibility gene. Nat. Genet.   42 , 410–414 (2010).2. Heyer, W.-D., Ehmsen, K.T. & Liu, J. Regulation of homologous recombination ineukaryotes. Annu. Rev. Genet.   44 , 113–139 (2010).3. Futreal, P.A. et al. A census of human cancer genes. Nat. Rev. Cancer    4 , 177–183(2004).4. Turnbull, C. & Rahman, N. Genetic predisposition to breast cancer: past, present,and future. Annu. Rev. Genomics Hum. Genet.   9 , 321–345 (2008).5. Easton, D.F., Bishop, D.T., Ford, D. & Crockford, G.P. Genetic linkage analysis infamilial breast and ovarian cancer: results from 214 families. The Breast CancerLinkage Consortium. Am. J. Hum. Genet.   52 , 678–701 (1993).6. Gayther, S.A. et al. The contribution of germline BRCA1 and BRCA2  mutations tofamilial ovarian cancer: no evidence for other ovarian cancer-susceptibility genes. Am. J. Hum. Genet.   65 , 1021–1029 (1999).7. Ramus, S.J. et al. Contribution of BRCA1 and BRCA2  mutations to inherited ovariancancer. Hum. Mutat.   28 , 1207–1215 (2007).8. Antoniou, A.C., Gayther, S.A., Stratton, J.F., Ponder, B.A. & Easton, D.F. Risk modelsfor familial ovarian and breast cancer. Genet. Epidemiol.   18 , 173–190 (2000).9. Shinohara, A. et al. Cloning of human, mouse and fission yeast recombination geneshomologous to RAD51 and recA. Nat. Genet.   4 , 239–243 (1993).10. Masson, J.Y. et al. Identification and purification of two distinct complexescontaining the five RAD51 paralogs. Genes Dev.   15 , 3296–3307 (2001).11. Meijers-Heijboer, H. et al. Low-penetrance susceptibility to breast cancer due to CHEK2  (*)1100delC in noncarriers of BRCA1 or BRCA2  mutations. Nat. Genet.   31 ,55–59 (2002).12. Renwick, A. et al.   ATM  mutations that cause ataxia-telangiectasia are breast cancersusceptibility alleles. Nat. Genet.   38 , 873–875 (2006).13. Rahman, N. et al.   PALB2  , which encodes a BRCA2-interacting protein, is a breastcancer susceptibility gene. Nat. Genet.   39 , 165–167 (2007).14. Seal, S. et al. Truncating mutations in the Fanconi anemia J gene BRIP1 are low-penetrance breast cancer susceptibility alleles. Nat. Genet.   38 , 1239–1241 (2006).15. Howlett, N.G. et al. Biallelic inactivation of BRCA2  in Fanconi anemia. Science    297 ,606–609 (2002).16. Levitus, M. et al. The DNA helicase BRIP1 is defective in Fanconi anemiacomplementation group J. Nat. Genet.   37 , 934–935 (2005).17. Reid, S. et al. Biallelic mutations in PALB2  cause Fanconi anemia subtype FA-Nand predispose to childhood cancer. Nat. Genet.   39 , 162–164 (2007).18. Vaz, F. et al. Mutation of the RAD51C  gene in a Fanconi anemia-like disorder. Nat.Genet.   42 , 406–409 (2010).19. Fong, P.C. et al. Inhibition of poly(ADP-ribose) polymerase in tumors from BRCAmutation carriers. N. Engl. J. Med.   361 , 123–134 (2009).20. Hinz, J.M. et al. Repression of mutagenesis by Rad51D-mediated homologousrecombination. Nucleic Acids Res.   34 , 1358–1368 (2006).21. Rebbeck, T.R., Kauff, N.D. & Domchek, S.M. Meta-analysis of risk reductionestimates associated with risk-reducing salpingo-oophorectomy in BRCA1 or BRCA2   mutation carriers. J. Natl. Cancer Inst.   101 , 80–87 (2009).22. Antoniou, A. et al. Average risks of breast and ovarian cancer associated with BRCA1  or BRCA2  mutations detected in case series unselected for family history: a combinedanalysis of 22 studies. Am. J. Hum. Genet.   72 , 1117–1130 (2003).23. International Agency for Research on Cancer. Cancer incidence in five continents.Volume VIII. IARC Sci. Publ. 1–781 (2002).     ©   2   0   1   1   N  a   t  u  r  e   A  m  e  r   i  c  a ,   I  n  c .   A   l   l  r   i  g   h   t  s  r  e  s  e  r  v  e   d . NATURE GENETICS doi:10.1038/ng.893 ONLINE METHODS Cases. We used lymphocyte DNA from 1,648 families with breast-ovariancancer or breast cancer only. These families were ascertained from 24 geneticscenters in the UK through the Genetics of Familial Breast Cancer Study (FBCS), which recruits women  18 years of age who have had breast and/orovarian cancer and have a family history of breast and/or ovarian cancer.At least 97% of the families studied are of European ancestry. Index cases fromeach family were screened and were negative for germline mutations, includinglarge rearrangements, in BRCA1 and BRCA2 . Informed consent was obtainedfrom all participants and the research was approved by the London MulticentreResearch Ethics Committee (MREC/01/2/18). Breast-ovarian cancer pedigrees. We included 911 unrelated index casesfrom breast-ovarian cancer pedigrees. The index cases were diagnosed withbreast and/or ovarian cancer. Each family contained an individual with bothbreast and ovarian cancer or contained at least one case of breast cancer andat least one case of ovarian cancer with  1 intervening unaffected femalerelative. Cases of ovarian cancer below the age of 20 were excluded from theanalysis, as an appreciable proportion of these cases are likely to represent non- epithelial ovarian tumors, for example germ cell cancers. Of the 911 probands,271 had ovarian cancer (with or without breast cancer) and 617 had breast can-cer only. The number of family members (including the probands) diagnosedwith breast cancer and/or ovarian cancer in the 911 breast-ovarian cancerpedigrees included in the analysis is shown in Supplementary Table 2.Breast-cancer–only pedigrees. We included 737 unrelated index cases frombreast cancer–only pedigrees. The index case from each family was diagnosedwith breast cancer and had bilateral disease and/or a family history of breastcancer. There was no known case of ovarian cancer in any of these pedigrees.The number of family members (including the probands) diagnosed withbreast cancer in the 737 breast cancer–only pedigrees included in the analysisis shown in Supplementary Table 2 . The six cases of isolated breast cancerall had bilateral disease. Samples and pathology information from mutation-positive families.  For families in which a mutation in RAD51D was detected, we sought DNAsamples from relatives, and we genotyped all obtainable samples for the fam-ily mutation. We also requested tumor material, pathology information andreceptor status in probands and affected relatives from the hospitals wherethey had been treated. Controls. We used lymphocyte DNA from 1,060 population-based controlsobtained from the 1958 Birth Cohort Collection, an ongoing follow-up of persons born in Great Britain during one week in 1958. Biomedical assessmentwas undertaken during 2002–2004, at which blood samples and informedconsent were obtained for the creation of a genetic resource; phenotype datafor these individuals were not available (see URLs). At least 97% of the controlswere of European ancestry. Mutation analysis of  RAD51D. We analyzed genomic DNA extracted fromlymphocytes for mutations by direct sequencing of the full coding sequenceand the intron-exon boundaries of  RAD51D . Primer sequences and PCRconditions are given in Supplementary Table 1 . The PCR reactions wereperformed in multiplex using the QIAGEN Multiplex PCR Kit (QIAGEN)according to the manufacturer’s instructions. Amplicons were unidirectionally sequenced using the BigDye Terminator Cycle sequencing kit and an ABI3730automated sequencer (ABI PerkinElmer). Sequencing traces were analyzedusing Mutation Surveyor software (see URLs) and by visual inspection.All mutations were confirmed by bidirectional sequencing from a fresh aliquotof the stock DNA. Samples from members of  RAD51D -mutation–positivefamilies were tested for the family mutation by direct sequencing of theappropriate exon. In silico analyses of identified variants. We computed the predicted effectsof  RAD51D missense variants on protein function using PolyPhen 24 andSIFT 25 . All variants (intronic and coding) were analyzed for their poten-tial effect on splicing. Variants were analyzed using two splice predictionalgorithms, NNsplice 26 and MaxEntScan 27 , using the Alamut software inter-face (Interactive Biosoftware). If both the NNsplice and MaxEntScan scoreswere altered by >20% (that is, a wild-type splice-site score decreased and/or acryptic splice-site score increased), three further prediction algorithms wereused; NetGene2 (ref. 28), HumanSplicingFinder 29 and GENSCAN 30 . A consen-sus decrease in a wild-type splice-site score and/or a consensus increase in acryptic splice-site score across all algorithms was considered indicative of disruption of normal splicing. Tumor analysis. Representative tumor sections were stained with NuclearFast Red and microdissected using a sterile needle and a stereomicroscope(Olympus SZ61) to ensure that the proportion of tumor cells was >90%, aspreviously described 31 . DNA was extracted using the DNeasy kit (QIAGEN)according to the manufacturer’s instructions. DNA concentration was mea-sured using the PicoGreen assay (Invitrogen) according to the manufacturer’sinstructions. RAD51D -specific fragments encompassing the relevant muta-tions were PCR amplified using the primers listed in Supplementary Table 1  and bidirectionally sequenced using the BigDye Terminator Cycle sequencingkit and an ABI3730 automated sequencer (ABI PerkinElmer). Sequence tracesfrom tumor DNA were compared to sequence traces from lymphocyte DNAfrom the same individual. Drug sensitivity. We used nonsilencing BRCA2 and RAD51D siGENOMEsiRNAs (Dharmacon). CAL51 and MCF7 cells were grown in Dulbecco’sModified Eagle Medium (DMEM) (Gibco, Invitrogen) supplementedwith 10% (v/v) FCS (Gibco, Invitrogen). Chinese hamster ovary (CHO) RAD51D wild-type (51D1.3 clone) and RAD51D dysfunctional (51D1 clone)cells were grown in  MEM (Gibco, Invitrogen) supplemented with 10%FCS (Gibco, Invitrogen). Cells were siRNA transfected using RNAiMAX(Invitrogen), plated in 96-well microtiter plates and then exposed to a titra-tion of olaparib for 7 days. The media and drug were replenished every 3 days. After 7 days of continuous culture, cell viability was estimated usingCellTiter-Glo reagent (Promega), and surviving fractions were calculatedas previously described 32 . Statistical methods. Statistical analyses were performed using Stata v11software (StataCorp). The frequency of mutations in cases and controls wascompared using a two-sided Fisher’s exact test. We estimated the RAD51D  combined mutation frequency, the breast cancer risk ratio and the ovariancancer risk ratio relative to non- RAD51D mutation carriers simultaneously using modified segregation analysis implemented in the pedigree analysissoftware MENDEL 33 . The analysis was based on breast and ovarian canceroccurrence in the combined dataset of families and controls. All individu-als were censored at age 80 years, the age of their first cancer or their age of death or last observation, whichever occurred first. Females who had bilateralprophylactic mastectomy were censored for breast cancer, and those who hadhad bilateral prophylactic oophorectomy were censored for ovarian cancer.Thus, only information on the first cancer was included in the primary analy-sis. We assumed that the breast cancer incidence depends on the underlyinggenotype through a model of the form:   ( t  )=   0 (t)exp(   x ), where   0 ( t  ) is thebaseline incidence at age t  in nonmutation carriers,   is the log risk ratioassociated with the mutation and x takes value 0 for nonmutation carriersand 1 for mutation carriers. A similar model was assumed for the ovariancancer incidences. Breast and ovarian cancers were assumed to occur inde-pendently, conditional on the genotype 22 . The overall breast and ovariancancer incidences were constrained to agree with the population incidencesfor England and Wales in the period of 1993–1997 (ref. 23), as describedpreviously  34,35 . The models were parameterized in terms of the mutationfrequencies and log-risk ratios for breast and ovarian cancer. Parameterswere estimated using a maximum likelihood estimation. Because RAD51D  mutation screening was carried out in all index cases and controls, we wereable to incorporate information from all controls and the full pedigrees fromall cases (including those without a RAD51D mutation) together with thesegregation information from the families in which a RAD51D mutationwas detected and genotyping was possible in relatives of the index case. Toadjust for ascertainment, we modeled the conditional likelihood of all family phenotypes and mutation status of the index family member and other tested
Related Documents
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!