Novel BRCA1 Mutations and More Frequent Intron-20 Alteration Found Among 236 Women from Western Poland

Novel BRCA1 mutations and more frequent intron-20 alteration found among 236 women from Western Poland

Krzysztof Sobczak1, Piotr Kozl owski1, Marek Napieral a1, Jakub Czarny1, Marcin WozÂniak1, Mal gorzata KapusÂ cinÂ ska1,2, Mal gorzataL osÂ ko1, Magdalena Koziczak1, Anna JasinÂ ska1, Jolanta Powierska1, Ryszard Braczkowski3, Jan Br borowicz4, Dariusz Godlewski2, . e˛ Andrzej Mackiewicz5 and Wl odzimierz Krzyzosiak1

1Laboratory of Cancer Genetics, Institute of Bioorganic Chemistry, Polish Academy of Sciences, Z. Noskowskiego 12/14, 61 ± 704 PoznanÂ; 2Cancer Epidemiology and Prevention Union, Wielkopolska Cancer Center, Garbary 15, 61 ± 866 PoznanÂ; 35-th Department of Internal Medicine, Silesian University School of Medicine, Bytom; 4Chair of Oncology, University School of Medical Sciences, La˛kowa 1/2 61-878 PoznanÂ; 5Department of Cancer Immunology, University School of Medical Sciences at Wielkopolska Cancer Center, Garbary 15, 61-866 PoznanÂ, Poland

Three dierent novel BRCA1 mutations, ®ve independent most frequently form a favorable genetic background cases of the same 12 bp insertion-duplication in intron-20 for the disease. Also at increased risk are the carriers of and two novel rare BRCA1 sequence variants were one abnormal allele of the ataxia telangiectasia gene identi®ed among 122 Polish women with positive, in most (Swift et al., 1991), the carriers of rare minisatelite cases moderate family history of breast and/or ovarian allele at the HRAS1 locus (Krontiris et al., 1993) and cancer, 80 controls and 34 unselected breast cancer the members of Li Fraumeni families with germline tissue specimens. All mutations and variants were mutation in the p53 gene (Srivastava et al., 1990). germline. The 4153 delA frameshift mutation, the The BRCA1 gene located on chromosome 17q21 Tyr105Cys missense mutation and two cases of the (Hall et al., 1990) is also linked to hereditary ovarian alteration in intron-20 were found in the group of healthy cancer (Narod et al., 1991). More than 400 BRCA1 women with positive family history. Two other cases of sequence variants have been identi®ed thus far and the intronic insertion were found in unselected controls. deposited in Breast Cancer Information Core Database Their carriers had no family history of breast or ovarian (Friend et al., 1995) which now contains about 150 cancer but other cancers occurred in their families. The dierent cancer predisposing mutations. Most of this 1782 Trp/STOP nonsense mutation and one case of the knowledge comes from studies of breast and breast- insertion in intron-20 were ®rst found in tissue specimens ovarian cancer families. Therefore the known muta- of breast cancer patient and breast/ovarian cancer tions are probably biased towards those with more patient, respectively. Their carriers also had no family severe eects on phenotype. It is expected that less history of breast or ovarian cancer. The distribution of expressive mutations will be identi®ed when studies the insertion in intron-20 in analysed groups and results expand and include women with less profound family of RT ± PCR experiments suggest a less prominent role history and women from the general population. It is for this variant considered earlier a splicing mutation. also very likely that the incidence of BRCA1 mutations This study shows also, that more population-oriented in dierent ethnic groups will vary and that dierent research is needed, involving women with less profound mutations will occur in populations from dierent or even no family history of breast and ovarian cancer, geographical regions. First indications of that come to better understand the role and signi®cance of dierent from the studies of Jewish women of Eastern European BRCA1 variants and mutations. descent (Friedman et al., 1995) and from the analysis of Swedish (Johannsson et al., 1996) and Austrian Keywords: BRCA1; mutations and variants; SSCP- (Wagner et al., 1996) breast cancer families. heteroduplex analysis; Polish women; positive family As most women have no close relative with breast or history ovarian cancer, a positive family history of these diseases is one of the primary risk factors. Thus, for women with one aected ®rst-degree relative, the risk of developing the disease is increased by 1.5-3 times Introduction (Thompson, 1994). In this paper we describe the results of BRCA1 gene analysis in a group of Polish women Approximately one in ten women in western countries selected for increased risk of breast and ovarian cancer will develop breast cancer during their lifetime based on their family history. We have also examined a (Wooster & Stratton, 1995). In most cases the disease substantial number of unselected controls and a set of is sporadic; however, some 5 ± 10% of women with breast cancer tissue samples. breast cancer inherit increased susceptibility to the disease (Claus et al., 1991). Within this fraction of breast cancer the germline mutations in BRCA1 (Miki Results et al., 1994) and BRCA2 (Wooster et al., 1995) genes BRCA1 mutations Correspondence: W Krzyz⋅osiak Received 21 August 1996; revised 17 June 1997; accepted 19 June All 24 BRCA1 exons, together with ¯anking intron 1997 fragments containing splice donor and splice acceptor BRCA1 mutations in Polish women K Sobczak et al 1774 sites, were ampli®ed from individual genomic DNA and controls was ampli®ed with primers located in samples and screened for sequence variations by BRCA1 exons 19 and 22, to see whether any shorter or combined single strand conformation polymorphism longer fragment would appear from carrier's cDNA. and heteroduplex analysis ± SSCP ± HDX. The ob- That would re¯ect either exon skipping, cryptic site served variants were sequenced, each along with the activation or intron retention during splicing of RNA corresponding normal fragment. The sequencing results transcript from the mutant allele. However, there was for most interesting variants are shown in Figure 1. no indication of any additional product besides the Three novel BRCA1 mutations identi®ed in this study normal 244 bp fragment. and the characteristics of women in whom these The nonsense Trp4Stop mutation in exon 22 found mutations were found are shown in Table 1. Five in patient 118 truncates the protein of amino acid 1781. independent cases of the same intron-20 alteration are The protein does not contain terminal 82 amino acids also included in the table. Every type of mutation is including 12 residues of the conserved C-terminal represented: frameshift, nonsense, missense and the domain. The carrier was diagnosed with breast cancer potential splicing mutation. The latter is very interesting at the age of 40 and underwent surgery. The mutation as the 12 bp duplication-insertion in intron-20 is was ®rst detected in the tumor, then con®rmed in the observed in our study at unusually high frequency. proband's blood DNA as well as in four out of ®ve In an attempt to clarify the molecular outcome of family members tested. The SSCP analysis of this intronic insertion we have performed the following proband's tumor DNA shows also LOH at the RT ± PCR experiment. The blood cDNA of the carriers BRCA1 locus (Figure 2). Interestingly, in the same

a b 433 A/G – Tyr105Cys 4153 del A – frameshift A C G T A C G T A C G T A C G T

normal mutant normal mutant

c d 5465 G/A – Trp1782Stop Insertion 12 bp in intron 20 A C G T A C G T A C G T A C G T

C C T C A C C T T A T G normal mutant normal mutant Figure 1 Sequencing results for BRCA1 mutations: (a) missense mutation in case 91; (b) frameshift mutation in case 98; (c) nonsense mutation in patient 118 and (d) for 12 bp insertion in intron-20, representative for cases 58, 69, 75, 82 and patient 271 BRCA1 mutations in Polish women K Sobczak et al 1775 Table 1 Characteristics of BRCA1 mutations or variants exon/ codon/ Case intron nucleotide Mutation Age Medical history Family history M: BrCa (47/48) frameshift MA1: BrCa (52/55), 98 exon 11 1345/4153 delA 38 None MA2: BrCa (45/66) STOP 1465 MA3:OvCa (62/66) GM: OvCa (59/60) nonsense 118 exon 22 1782/5465 TGG/TGA 43 BrCa (40/7) F: CoCa (60/62) Tyr/STOP F:LuCa (68/70) missense M: UtCa (42/43) 91 exon 7 105/433 TAT/TGT 36 None MA: BrCa (54/56) Tyr/Cys B: LuCa (47/48) MU: BCa (59/60) F: Mel (57/59) 58 36 None M: UtCa (61/ ± ) MA: LuCa (46/47)

69 45 None M: BrCa (62/65) 12bp 48 bp from M: UtCa (60/7) 75 intron 20 end of duplication/insertion 46 None F: StCa (65/65) exon 20 GTATTCCACTCC M: BrCa (67/75) 82 51 None S: BrCa (44/7), B: OsSa (40/7) FA: Leu (35/35)

271 BrCa D37 none OvCa (39/39)

BrCa ± breast cancer, OvCa ± ovarian cancer, UtCa ± uterine cancer, LuCa ± lung cancer, StCa ± stomach cancer, OsSa ± osteosarcoma, Leu ± leukemia, Mel ± melanoma, BCa ± brain cancer (age of diagnosis/age of death) M ± mother, F ± father, S ± sister, B ± brother, GM ± grandmother, MA ± maternal aunt, MU ± maternal uncle

tumor sample the p53 gene mutation Pro278A1a was . earlier found (Krzyzosiak et al., unpublished) and con®rmed to be somatic. The frameshift mutation at codon 1345 which triggers the protein synthesis termination at codon 1465 was found in the DNA of a healthy 38-year-old woman, case 98. Her mother and the mother's two sisters had breast cancer diagnosed respectively at the age of 47, 52 and 45. Another mother's sister and a grandmother had ovarian cancer, diagnosed respectively at the ages of 62 and 59. The DNA samples from some aected and unaected members of this family could be obtained, and the results of their analysis for codon 1345 mutation are shown in the family pedigree in Figure 3a. The missense mutation Tyr105Cys in case 91 was the ®rst alteration found in BRCA1 exon 7 (Friend et al., 1995). We detected the A to G transition at codon 105 in only one out of 236 DNA samples tested. The carrier, a 36-year-old healthy woman, had a father who died of lung cancer at the age of 70 and a brother who died of the same type of cancer at the age of 48. Her mother died of uterine cancer at the age of 43 and the mother's sister died of breast cancer at the age of 56 (Figure 3b). The segregation of this amino acid change with the disease in the carrier's family could not be Figure 2 Segregation of the Trp1782/Stop mutation in a family of patient 118 and the corresponding results of SSCP ± HDX examined as the appropriate DNA samples were not analysis of exon 22: control DNA (lane C); DNA from tumor available for analysis. The tyrosine at position 105 in tissue of patient 118 (lane T); blood DNA of patient 118 (lane 3); BRCA1 protein lies outside the two regions conserved blood DNAs of relatives of patient 118 (lanes 1, 2, 4, 5). The in human and mouse sequences (Abel et al., 1995). In comparison of lanes T and 3 shows LOH in the BRCA1 locus in murine Brca1 the position 105 is occupied by Phe, the tumor sample. The SSCP bands corresponding to mutant single strands are indicated in lane T. Other symbols used are similar to Tyr. However, in canine sequence, like in explained in legend to Figure 3 human, tyrosine 105 is present (Szabo et al., 1996). The BRCA1 mutations in Polish women K Sobczak et al 1776 biological signi®cance of the Tyr105Cys change is the human migration due to historical events has been therefore uncertain. rather low. The analysis of the BRCA1 gene in 236 women revealed three novel mutations, one novel common polymorphism, two new rare sequence Novel common polymorphism and rare sequence variants variants and relatively high frequency of the alteration In this study we have also identi®ed one common in intron-20. Our data may be considered one of the polymorphism, which has not been described earlier. ®rst insights into the future results of BRCA1 mutation This is the C4T transition in intron-7, located 34 bp analysis when the research extends to many other upstream of the start of exon 8. Its frequency in the ethnic groups and to women with less profound family studied group is 0.22. One of the rare BRCA1 sequence history of breast and ovarian cancer. variants was found in exon 11. It is a silent base change The incidence of BRCA1 mutations and the intron- C4T at the third position of serine codon 378. Another 20 alteration in the analysed groups is shown in Figure rare variant, the T to C transition was found in intron-2, 4. The observed high frequency of the 12 bp insertion- 37 bp from the end of exon 2. Both were detected only duplication in intron-20 and its distribution in the once among the 236 DNA samples tested. In a family of studied groups is intriguing. In earlier reports this the woman with the alteration in intron-2 there was no insertion was in each case associated with the serious history of breast or ovarian cancer. medical history of malignant conditions and was not detected in numerous controls (Takahashi et al., 1995; Langston et al., 1996a; Langston et al., 1996b). In one Discussion of these studies the carrier was a woman with both breast and ovarian cancer who had ®ve maternal In Poland the country-average mortality rate for breast relatives with breast cancer (Takahashi et al., 1995). In cancer is approximately half of that reported for the the other study this insertion was found in a young 31- United States. The participants of this study, except for year-old women diagnosed with breast and cervical patient 271 who was from Upper Silesia, were from the cancer (Langston et al., 1996a). Most recently two PoznanÂ region in Western Poland where the incidence other cases of this insertion were found in men with of breast cancer is the highest in the country (ZatonÂ ski, prostate cancer (Langston et al., 1996b). Their ages 1993). The studied population is representative of the were 46 and 51 at diagnosis. Only one out of ®ve ethnic group inhabiting the Wielkopolska region where Polish women with this insertion, patient 271, resembles in this respect the earlier described cases. This patient had breast cancer diagnosed at the age of 37 and ovarian cancer at the age of 39. The other four carriers are healthy, however, they have a signi®cant but diverse family history of various cancers. The pedigrees of Polish families with this insertion are shown in Figure 5. As shown in this study the 12 bp insertion- duplication does not appear to cause aberrant splicing of RNA transcript in blood tissue. Nevertheless it is possible that it impairs splicing in other tissues e.g.

Figure 3 Pedigrees of two families, 98 (a) and 91 (b), with BRCA1 frameshift mutation 4153de1A and missense mutation Tyr105Cys, respectively. Circles represent females and squares represent males. Left- and right-side-blackened symbols represent women aected with breast and ovarian cancer, respectively. The upper-side-blackened symbols represent any other cancer. The type of cancer is given below the circle and square symbols: BrCa ± breast cancer, OvCa ± ovarian cancer, BCa ± brain cancer, UtCA ± uterine cancer, LuCa ± lung cancer. The slashed symbols indicate individuals who are deceased. Age at diagnosis and age at death, or current age of the analysed patients are given in bracket under the patient symbol. Black vertical line through a symbol Figure 4 The incidence of BRCA1 mutations in the analysed indicates carriers of mutation, while Wt indicates analysed cases sample: women with positive family history of breast and/or who are noncarriers. Arrowhead indicates proband in each family ovarian cancer (FH); tumor samples (T); controls (C) BRCA1 mutations in Polish women K Sobczak et al 1777 at the stage IIA. As mentioned earlier in this paper, the tumor tissue DNA of this patient harbors also the somatic mutation in the p53 gene. This C to G transversion replaces Pro for Ala at codon 278. The Pro 278 nucleates the H2 helix in p53 core domain crystal structure (Cho et al., 1994), is present in a conserved region V and is identical in p53 proteins from diverse species (Friend, 1994). This missense mutation seems therefore to be functionally signi®cant. Our ®nding gives new insight into the molecular mechanism of tumor development in epithelial cells of the breast and shows that during the process of multistep carcinogenesis, the p53 gene mutation can also occur in the background of mutated BRCA1. From the perspective of genetic testing, the spectrum of BRCA1 mutations identi®ed thus far looks complicated. The distribution of mutations, although uneven in various BRCA1 sequence regions, does not allow focusing the analysis on some exons while omitting the other. Of the methods used for mutation detection, besides direct sequencing, the protein truncation test gives good results, as about 95% of known mutations lead to truncated proteins (Friend et al., 1995). Also the SSCP technique (Orita et al., 1989) when combined with heteroduplex analysis is capable of detecting the majority of missense, nonsense and splicing mutations and Figure 5 Pedigrees of families 58, 69, 75, 82 and 271, carrying the 12 bp insertion-duplication in intron 20. The symbols used are perhaps all frameshift mutations. Simplicity is one of as described in legend to Figure 3, except for: StCa ± stomach the advantages of the latter technique which is cancer, Le ± leukemia, OsCa ± osteosarcoma, Mel ± melanoma amenable to multiplexing and to analysing PCR products obtained from pooled genomic DNA samples (Kozl owski et al., 1996). These factors will ovary or breast. Another possibility is that the become important if the search for dispersed insertion aects only the kinetics of splicing and mutations in BRCA1 and BRCA2 genes is scaled results in a lower level of normal mRNA from the up, before the DNA chip technology (Hacia et al., mutant allele. In this context it should be recalled that 1996) becomes mature and widely available. markedly decreased levels of BRCA1 mRNA were Further research on dierent ethnic groups should observed in sporadic invasive breast cancers (Thomp- involve a higher number of women with positive son et al., 1995). Alternatively, the 12 bp insertion is family history of breast and ovarian cancer, divided only a neutral marker that may or may not be linked into several more homogenous risk groups, and more to some speci®c high-risk haplotype or to some other controls representing the general population. A high cancer predisposing mutation. This problem clearly number of unselected, consecutive breast and ovarian needs further investigation. cancer tissue specimens should be subjected to a The example of the relatives of patient 118 with the similar analysis. This will allow an experimental de®nite mutation in exon 22 shows that in the family veri®cation of the crude estimates regarding the with no reported history of breast or ovarian cancer contribution of hereditary factors to the overall the BRCA1 mutation carriers were identi®ed through incidence of these cancers. A systematic search for the analysis of tumor tissue of the aected family somatic mutations in other tumor suppressor genes member (Figure 2). Thus, the screening of cancer and oncogenes in these tissue specimens, will help samples ± even those unselected for family history ± may distinguishing between the cancers with and without be considered as an eective way of identifying those at the hereditary background. high risk who are at increased need for a preventive intervention. The contribution of such carriers to the overall incidence of BRCA1 mutations is as yet unknown. In two recent studies the incidence of mutations was found to be 13% in women with Materials and methods breast cancer diagnosed before the age of 30 Patients and cancer tissue characteristics (FitzGerald et al., 1996) and 10% in women diagnosed with breast cancer under the age of 35 The volunteers of this study were among the women who (Langston et al., 1996a). Two out of six mutation underwent the mammography screening at the Wielkopol- carriers from the latter study did not report any family ska Cancer Center in winter 1995. The initial selection was history of breast or ovarian cancer. The average age of based on direct interviews, and the number of reported cases of breast or ovarian cancers among ®rst- and second- breast cancer diagnosis in our sample of 34 women was degree relatives was con®rmed wherever possible by 56, and this group included only four women with medical records. The predominant part of the analysed cancer diagnosed at the age of 40 or earlier. Patient 118 sample were 122 healthy women of an average age of 38, was one, diagnosed with in®ltrating lobular carcinoma range 18 ± 67, with a varying degree of family history of BRCA1 mutations in Polish women K Sobczak et al 1778 breast and ovarian cancer. This group included women in Table 2 Primers used for ampli®cation of exon 11 fragments whose families one case of breast cancer occurred among Fragment Length Sequence the ®rst and second degree relatives (70), and women from families in which one case of ovarian cancer occurred (4). 11.01 265 bp 5'-CCACCTCCAAGGTGTATGAA-3' Therewerealso48womenwithamoresigni®cantfamily 5'-GAGCTGGCATGAGTATTTGTG-3' 11.02 240 bp 5'-GGCATCCAGAAAAGTATCAGGGTAG-3' history. Among their relatives occurred two cases of breast '- - ' cancer (20), two cases of ovarian cancer (2), one case of 5 TGGGAGTCCGCCTATCATTACAT 3 11.03 190 bp 5'-AAGGCTGAATTCTGTAATAAAAGCA-3' breast and one of ovarian cancer (13), three breast cancers 5'-GATTCTCTGAGCATGGCAGTT-3' or two breast and one ovarian (10), four and ®ve cases of 11.04 202 bp 5'-GATCTGAATGCTGATCCCCTGTGTG-3' breast and ovarian cancer (3). The control group was 5'-GGCATTTGATTCAGACTCCCCATC-3' composed of 80 women unselected for family history of 11.05 244 bp 5'-TGAGTGGTTTTCCAGAAGTGATG-3' breast or ovarian cancer. The blood samples were collected 5'-GGTTTTCCCAAATATTTTGTCTTCA-3' and the con®dentiality of the patients was preserved in 11.06 207 bp 5'-AGACTTACTGGCCAGTGATCCTCAT-3' accordance with the guidelines for studies of human 5'-AGGGGACGCTCTTGTATTATCTGTG-3' 11.07 231 bp 5'-CCTCCCCAACTTAAGCCATGTAACT-3' subjects (Hoskins et al., 1995). Women who participated '- - ' in this study, were informed about its goals, and oered 5 TTCATCACTTGACCATTCTGCTCC 3 11.08 183 bp 5'-AGAAAGCAGATTTGGCAGTTCA-3' counseling and education about the known risk factors of 5'-AAGCAGATTCCTTTTCGAGTGATTC-3' cancer and potential risks as well as the bene®ts of genetic 11.09 249 bp 5'-GGTCATGAGAATAAAACAAAAGGTG-3' testing (Collins, 1996). 5'-AGGTGGGCTTAGATTTCTACTGACT-3' This study included also 34 breast cancer tissue 11.10 236 bp 5'-TCCACAATTCAAAAGCACCTAAAAA-3' specimens that were earlier analysed for mutations in the 5'-GGCTCCAGTTGCAGGTTCTTTAC-3' p53 and p16 genes. The tissue specimens were from 11.11 228 bp 5'-AAAGTACAACCAAATGCCAGTCAGG-3' patients, unselected for the age of diagnosis, who under- 5'-TCTCTTGGAAGGCTAGGATTGACAA-3' 11.12 210 bp 5'-CATGACAGCGATACTTTCCCAGAGC-3' went surgery at the Wielkopolska Cancer Center in PoznanÂ . '- - ' The average age at diagnosis was 56 years (range 29 ± 82). 5 TTCAGTTTGCAAAACCCTTTCTCCA 3 11.13 211 bp 5'-AAACAGTTAAAGTGTCTAATAATGC-3' They were diagnosed at stage I (5), IIA (13), IIB (9), IIIA 5'-GCACACTGACTCACACATTT-3' (5) and IIIB (2). 11.14 334 bp 5'-AAACAGTTAAAGTGTCTAATAATGC-3' 5'-ATTTCTATGCTTGTTTCCCGACTGT-3' 11.15 235 bp '- - ' PCR ± SSCP ± HDX analysis and sequencing 5 GTTGTTCCAAAGATAATAGAAATGA 3 5'-TGTTTCTTTAAGGACCCAGA-3' Forty-nine PCR primer pairs were used, those for exons 11.16 219 bp 5'-CGCCAGTCATTTGCTCCGTTTT-3' 1 ± 10 and 12 ± 24 were adapted from the literature 5'-TGGCTTATCTTTCTGACCAACCACA-3' 11.17 248 bp 5'-TGAACAAAAGGAAGAAAATCAAGGA-3' (Friedman et al., 1994). Exon 11 was covered with 26 '- - ' newly designed primer pairs giving partially overlapping 5 TGGGAAAAAGTGGTGGTATACGAT 3 11.18 226 bp 5'-TCAGTTCAGAGGCAACGAAA-3' PCR products of the average size 230 bp (Table 2). The 5'-AATGTTATTACGGCTAATTGTGCTC-3' protocol for rapid and economical PCR was used in all 11.19 237 bp 5'-AGAGGAAAACTTTGAGGAACATTCA-3' ampli®cations (Sobczak et al., 1995). All variant bands 5'-GGCCCTCTGTTTCTACCTAGTTCTG-3' except for those originating from known common 11.20 269 bp 5'-GTACTAATGAAGTGGGCTCCAGTAT-3' polymorphisms were con®rmed in three independent 5'-GTATGCTTACAATTACTTCCAGG-3' PCR-SSCP-HDX experiments. 11.21 229 bp 5'-ATTGAATGCTATGCTTAGATTAGGG-3' PCR ampli®cations were carried out in a volume of 5 ml 5'-ATCTAACAGGTCATCAGGTGTCTC-3' 11.22 242 bp 5'-AGATTTCTCTCCATATCTGATTTCA-3' containing 25 ng of genomic DNA, 1.5 mM MgCl ,50mM 2 5'-GGTAACCCTGAGCCAAATGT-3' KCl, 10 mM Tris/HCl (pH 8.3), 200 mM of each dNTP and 11.23 234 bp 5'-AAAAGCGTCCAGAAAGGAGAG-3' 0.125 U Taq polymerase (Perkin Elmer/Cetus). PCR primers 5'-TGTGTTCTTAGACAGACACTCGGTA-3' used at concentration 1 mM were 5'-end labeled with 11.24 190 bp 5'-GAGCTTCCCTGCTTCCAACACTTGT-3' [g-32P]ATP and T4 polynucleotide kinase (New England 5'-GGTGATGTTCCTGAGATGCCTTTGC-3' BioLabs) prior to PCR (Hayashi et al., 1989). Amplifications 11.25 230 bp 5'-ATTGAAGAATAGCTTAAATGACTGC-3' done on a Perkin Elmer-Cetus 480 Thermal Cycler involved 5'-CTTTGTCACTCAGACCAACTC-3' 35 cycles: 1s/948C, 1s/558C and 10s/728C. The annealing 11.26 252 bp 5'-TGACTGCAAATACAAACACCC-3' temperature was dierent: 508C for exon 8, 578C for exons 1, 5'-GCTCCCCAAAAGCATAAACA-3' 3, 4 and 648C for fragment four of exon 11. The SSCP-HDX analysis was carried out as described earlier (Kozl owski et al., 1996). The con®rmed variant bands were cut o from the gel, reampli®ed and sequenced from both ends using amplimers GAAAGCTC-3', ampli®cation conditions: 1s/948C, 1s/ as sequencing primers. 578C, 10s/728C, 35 cycles.

RT ± PCR Acknowledgements Reverse transcription of the total RNA isolated from the This work was supported by grant from State Committee blood of the four healthy women with the 12 bp insertion for Scienti®c Research Nr 6 P207 106 06, by the Social in intron-20 and from two control individuals was done Foundation `People for People' and by donation from with oligo-dT primer and AMV reverse transcriptase Banyu Tsukuba Research Institute in Collaboration with (Promega). The PCR primers used to analyse the eect of Merck Sharp & Dohme Research Laboratories, Banyu this insertion on splicing were F19: 5'-GTGACCCAGTC- Pharmaceutical Co., Ltd., Tsukuba Techno-Park, Ohkubo TATTAAAGAAAGA-3' and R22: 5'-GGGTGAATGAT- 3, Tsukuba 300-33, Japan.

References

Abel KJ, Xu J, Yin G-Y, Lyons RH, Meisler MH and Weber Cho Y, Gorina S, Jerey PD and Pavletich NP. (1994). BL. (1995). Hum. Mol. Genet., 4, 2265 ± 2273. Science, 265, 346 ± 355. BRCA1 mutations in Polish women K Sobczak et al 1779 Claus EB, Risch N and Thompson WD. (1991). Am.J.Hum. Narod SA, Feunteun J, Lynch HT, Watson P, Conway T, Genet., 48, 232 ± 242. Lynch J and Lenoir GM. (1991). Lancet, 338, 82 ± 83. Collins, FS. (1996). N. Engl. J. Med., 334, 186 ± 188. Orita M, Suzuki Y, Sekiya T and Hayashi K. (1989). FitzGerald MG, MacDonald DJ, Krainer M, Hoover L, Genomics, 5, 874 ± 879. . O'Neil E, Unsal H, Silva-Arrieto S, Finkelstein DM, Beer- Sobczak K, Kozl owski P and Krzyzosiak WJ. (1995). Acta Romero P, Englert C, Sgrol DC, Smith BL, Younger JW, Biochim. Pol., 42, 363 ± 366. Garber JE, Duda RB, Mayzel KA, Isselbacher KJ, Friend Srivastava S, Zou ZQ, Pirollo K, Blattner W and Chang EH. SH and Haber DA. (1996). N Engl. J. Med., 334, 143 ± 149. (1990). Nature, 348, 747 ± 749. Friedman LS, Szabo CI, Ostermeyer EA, Dowd P, Butler L, Swift M, Morrel D, Massay R and Chase CL. (1991). N. Park T, Lee MK, Goode EL, Rowell SE and King M-C. Engl. J. Med., 326, 1831 ± 1836. (1995). Am. J. Hum. Genet., 57, 1284 ± 1297. Szabo CI, Wagner LA, Francisco LV, Roach JC, Argonza R, Friedman LS, Ostermeyer EA, Szabo CI, Dowd P, Lynch King M-C and Ostrander A. (1996). Hum. Mol. Genet., 5, ED, Rowell SE and King M-C. (1994). Nature Genet., 8, 1289 ± 1298. 399 ± 404. Takahashi H, Behbakht K, McGovern PE, Chiu H-Ch, Friend S, Borresen AL, Brody L, Casey G, Devilee P, Couch FJ, Weber BL, Friedman LS, King M-C, Furusato Gayther S, Goldgar D, Murphy P, Weber BL and M, LiVolsi VA, Menzin AW, Liu PC, Benjamin I, Morgan Wiseman R. (1995). Nature Genet., 11, 238 ± 239. MA, King SA, Rebane BA, Cardonick A, Mikuta JJ, Friend S. (1994). Science, 265, 334 ± 335. Rubin SC and Boyd J. (1995). Cancer Res., 55, 2998 ± Hacia JG, Brody LC, Chee MS, Fodor SPA and Collins FS. 3002. (1996). Nature Genet., 14, 441 ± 447. Thompson ME, Jensen RA, Obermiller PS, Page DL and Hall JM, Lee MK, Newman B, Morrow JE, Anderson LA, Holt JT. (1995). Nature Genet., 9, 444 ± 450. Huey B and King M-C. (1990). Science, 250, 1684 ± 1689. Thompson WD. (1994). Cancer, 74, 279 ± 287. Hayashi K, Orita M, Suzuki Y and Sekiya T. (1989). Nucleic Wagner TMU, MoÈ lslinger R, Zielinski C, Scheiner O and Acids Res., 17, 3605. Breiteneder H. (1996). Lancet, 347, 1263. Hoskins KF, Stopfer JE, Calzone KA, Merajver SD, Wooster R and Stratton MR. (1995). TIG, 11, 3±5. Rebbeck TR, Garber JE and Weber BL. (1995). JAMA, Wooster R, Bignell G, Lancaster J, Swift S, Seal S, Mangion 273, 577 ± 585. J, Collins N, Gregory S, Gumbs C, Mickiem G, Barfoot Johannsson O, Ostermeyer EA, Hakasson S, Friedman LS, M, Hamoudi R, Patel S, Rice C, Biggs P, Hashim Y, Smith Johansson U, Sellberg G, Brondum-Nielsen K, Sele V, A, Connor F, Arason A, Gudmundsson J, Ficenec D, Olsson H, King M-C and Borg A. (1996). Am.J.Hum. KeisellD,FordD,ToninP,BishopDT,SpurrNK, Genet., 58, 441 ± 450. Ponder BAJ, Eeles R, Pete J, Devilee P, Cornelisse C, Koz owski P, Sobczak K, Napieral a M, WozÂniak M, Czarny l . Lynch H, Narod S, Lenoir G, Eglisson V, Bjork J and Krzyzosiak WJ. (1996). Nucl. Acids Res., 24, 1177 ± Barkadottir R, Easton DF, Bentley DR, Futreal PA, 1178. Asheorth A and Stratton MR. (1995). Nature, 378, 789 ± Krontiris TG, Devlin B, Karp DD, Robert NJ and Risch N. 792. (1993). N.Engl.J.Med.,329, 42 ± 51. ZatonÂ ski WA. (1993). Cancer in Poland, editor: Oncology Langston AA, Malone KE, Thompson JD, Paling JR and Center: Warszawa. Ostarander EA. (1996a). N.Engl.J.Med.,331, 137 ± 142. Langston AA, Stanford JL, Wicklund KG, Thompson JD, Blazej RG and Ostrander EA. (1996b). Am.J.Hum. Genet., 58, 881 ± 885. Miki Y, Swensen J, Shattusk-Eidens D, Futreal PA, Harshman K, Tavigian S, Liu Q, Cochran C, Tran T, McClure M, Frye C, Hattier T, Phelps R, Haugen-Strano A, Katcher H, Ykumo K, Gholami Z, Shaer D, Stone S, Bayer S, Wray C, Bogden R, Dayanath P, Ward J, Tonin P, Narod S, Bristow PK, Norris FH, Helvring L, Morrison P, Rosteck P, Lai M, Barret JC, Lewis C, Neuhausen S, Cannon-Albright L, Goldgar D, Wiseman R, Kamb A and Skolnick MH. (1994). Science, 266, 66 ± 71.