Clinical Chemistry 45:7 976 –981 (1999)
Molecular Diagnostics and Genetics
BRCA1 Gene Mutations in Sporadic Ovarian Carcinomas: Detection by PCR and Reverse Allele-specific Oligonucleotide Hybridization Dan Tong,1 Margit Stimpfl,1 Alexander Reinthaller,1 Norbert Vavra,1 Silvia Mu¨llauer-Ertl,2 Sepp Leodolter,1,3 and Robert Zeillinger1*
Background: Although germline mutations in BRCA1 play a central role in familial breast and ovarian cancers, to date, no somatic mutations in BRCA1 have been reported in sporadic breast cancer, and only five somatic mutations have been identified in the sporadic ovarian carcinomas. Because loss of heterozygosity appears frequently at the BRCA1 locus in nonfamilial breast and ovarian carcinomas, we searched for mutations in the BRCA1 gene in sporadic ovarian tumors. Methods: We developed a detection system based on PCR and reverse allele-specific oligonucleotide hybridization on membrane strips for the simultaneous detection of 17 frequently occurring mutations in the BRCA1 gene. Results: As little as 2% mutant DNA in a sample could be detected. Two of 122 DNA samples isolated from sporadic ovarian tumor biopsies contained the Cys61Gly mutation. Both mutations were germline mutations. One of these was an ovarian metastasis of a primary fallopian tube carcinoma. The tubal carcinoma was also confirmed to contain the Cys61Gly mutation. Conclusions: This is the first report that a germline BRCA1 mutation is associated with primary tubal carcinoma. The 17 specific mutations in the BRCA1 gene do not play a major role in the tumorigenesis and progression of sporadic ovarian cancer.
The BRCA1 gene contains 22 coding exons and encodes a nuclear phosphoprotein of 1863 amino acid residues (1, 2 ). In human breast cancer cell lines and spermatocytes, BRCA1 protein binds to Rad51, which functions in DNA recombination and repair, suggesting a role for BRCA1 specifically during mitosis and meiosis (3 ). Recent studies have suggested that BRCA1 contributes to cellcycle arrest and growth suppression through the induction of p21 (4 ). BRCA1 and p53 cooperatively induce apoptosis of cancer cells, and BRCA1 may coordinately regulate gene expression together with p53 as tumor suppressors (5 ). The BRCA1 gene frequently is mutated in familial breast and/or ovarian cancers. To date, .500 distinct mutations, polymorphisms, and unclassified variants have been identified [Breast Cancer Information Core database (BIC), 1998]. Most small deletions, insertions, and point mutations lead to premature termination of translation and, thus, truncated proteins. Germline mutations in BRCA1 are thought to be responsible for ;45% of familial breast cancers and for .80% of the inherited breast and ovarian cancer syndrome (6 ). Although BRCA1 mutations are involved in the etiology of hereditary breast and/or ovarian cancers, they seem to play no important role in the tumorigenesis of sporadic breast and ovarian cancer. To date, no somatic mutations in BRCA1 have been reported in sporadic breast cancer, and only five somatic mutations have been identified in sporadic ovarian carcinomas (7, 8 ). However, this is not consistent with the observations that loss of heterozygosity frequently appears at the BRCA1 locus in nonfamilial breast and ovarian carcinomas (9, 10 ). Therefore, we were interested in searching for mutations in the BRCA1 gene in sporadic ovarian tumors. At present, mutation screening in BRCA1 has been performed mostly by the use of the single-strand conformation polymorphism method, the protein truncation test, allele-specific oligonucleotide (ASO) hybridization,
© 1999 American Association for Clinical Chemistry
1 Department of Obstetrics and Gynecology, Division of Gynecology, and Department of Clinical Pathology, General Hospital of Vienna, University of Vienna, A-1090 Vienna, Austria. 3 Ludwig-Boltzmann Institute for Oncology and Fertility Treatment, A-1090 Vienna, Austria. *Address correspondence to this author at: General Hospital of Vienna, Department of Obstetrics and Gynecology, Molecular Oncology Group, Wa¨hringer Gu¨rtel 18-20, EBO 05, A-1090 Vienna, Austria. Fax 43-1-40400-7832; e-mail
[email protected]. Revision received March 15, 1999; accepted April 27, 1999. 2
976
977
Clinical Chemistry 45, No. 7, 1999
Table 1. Histopathologic data of the 122 sporadic ovarian tumors. Histopathologic type
Number
Serous carcinoma Mucinous carcinoma Endometrioid carcinoma Clear cell carcinoma Undifferentiated carcinoma Other carcinoma Total
56 15 12 7 7 25 122
denaturing gradient gel electrophoresis, and direct sequencing (11–14 ). These techniques have different detection limits. Because tumor biopsies usually contain varying proportions of healthy tissue, it is questionable whether the very low rate of BRCA1 mutations found in ovarian carcinomas is attributable to the detection limits of these techniques. Thus, we chose reverse ASO hybridization, a method that has been reported to detect as little as 2.5% mutant DNA (15 ). Reverse ASO was originally developed to facilitate the simultaneous screening of different allelic variants at an amplified locus (16 ). Because BRCA1 is a large gene, screening for all possible mutations through the whole coding sequence is timeconsuming and costly. The results of mutation screening in the BRCA1 gene suggest that different mutations have different carrier frequencies. Certain mutations have been found repeatedly, whereas others have been reported only once. We chose 17 frequently occurring mutations in BRCA1, which covered .50% of all reported mutations when this study was initiated (BIC, 1996). We developed a system for the simultaneous detection of these mutations by the use of PCR and reverse ASO hybridization techniques and evaluated the system, using DNA samples containing known mutations. The detection limit of the system was tested with control plasmids containing cloned BRCA1 fragments. We then screened 122 DNAs from sporadic ovarian carcinomas for the existence of these mutations.
mutations were obtained from the Department of Pathology, University of Cambridge, Cambridge, UK.
pcr A total of 17 mutations in the BRCA1 gene were chosen because of their prevalence (Table 2, based on data from BIC, 1996). The sense and antisense primers for the amplification of the fragments containing these mutations are shown in Table 2. All antisense primers were labeled with biotin. DNA (50 ng) from peripheral blood or tumor biopsy was used as the template for PCR in a total volume of 50 mL. The reaction mixture included 25 pmol each of the sense and antisense primers, 250 mmol/L dNTPs (ViennaLab), 5 mL of 103 amplification buffer (10 mmol/L Tris-HCl, pH 9.0, 50 mmol/L KCl, 0.1 g/L gelatin, 1.5 mmol/L MgCl2, and 1.0 mL/L Triton X-100; ViennaLab) and 1.6 U of Super Taq Polymerase (HT Biotechnology). PCR was performed in a Perkin-Elmer GeneAmp PCR system 9600 with 40 cycles at 94 °C for 30 s, 50 °C for 30 s, and 72 °C for 30 s. All reactions were preceded by a primary denaturation step at 94 °C for 1 min. The PCR product (10 mL) was then resolved on 4% agarose gels containing 0.1 mL/L SYBR Green I (Molecular Probes). The gels were visualized with 254 nm transillumination and photographed.
Table 2. Primers used for PCR. Mutation
185delAG 188del11 C61G I5 splice 1135insA 1294del40 1411insT 1675delA
Materials and Methods tumor samples Fresh biopsies from 122 sporadic ovarian carcinomas (without known family histories) and the corresponding EDTA blood samples were collected in the Department of Obstetrics and Gynecology, University of Vienna, from 1994 to 1997. The histologic diagnoses (17 ) of the 122 patients are shown in Table 1.
dna preparation We isolated DNA from blood or tumor tissue, using commercially available kits (DNA Extraction System I and II; ViennaLab). Three DNA samples with different known
Q563X 2804delAA E908X 3121delA 3450del4 4184del4 R1443X R1443G 5382insC a
Exon
2
Sense Antisense 5 Sense Antisense Intron 5 Sense Antisense 11 Sense Antisense 11 Sense Antisense 11 Sense Antisense 11 Sense Antisense 11 Sense Antisense 11 Sense Antisense 11 Sense Antisense 11 Sense Antisense 11 Sense Antisense 13 Sense Antisense 20 Sense Antisense
Name or sequence
Exon 2a Exon 5a Exon 6a Exon 11Ba Exon 11B2a 59-CAAATGCCAAAGTAGCTGATG-39 59-GCTTGCCTTCTTCCGATAG-39 59-AGCATTTGTTACTGAGCCAC-39 59-ATCACTTGACCATTCTGCTCC-39 Exon 11Ea 59-TTAACCACAGTCGGGAAAC-39 59-TGACCAACCACAGGAAAGC-39 Exon 11Ka Exon 11Ma c5-exon 11Pia Exon 13a Exon 20a
Names and sequences were published previously (19).
978
Tong et al.: Mutation Detection in the BRCA1 Gene
construction of control plasmids and evaluation of the detection limit To evaluate the system, plasmids containing cloned mutant fragments for each tested mutation were constructed using the overlapping extension PCR technique (Fig. 1). Briefly, two separate PCRs were carried out to generate two overlapping fragments. One of these fragments contained the mutation at the one end, whereas the other fragment contained the mutation at the other end. The bands of these two PCR products were excised from the agarose gel, placed in H2O, and vortex-mixed. Portions of the solutions were combined, dNTPs and Taq polymerase were added, and 10 cycles of extension were carried out as follows: 30 s at 94 °, 30 s at 50 °, and 30 s at 72 °C. After the extension was completed, the sense and antisense primers were added, and a PCR of 30 cycles of 30 s at 94 °C, 30 s at 50 °C, and 30 s at 72 °C was performed. The mutant PCR product and the PCR product generated from wildtype DNA were cloned into plasmids (TA cloning kit; Invitrogen). To evaluate the detection limit, plasmid DNA containing the wild-type fragment was mixed at different ratios with plasmid DNA containing the corresponding mutant fragment. PCR was carried out as described above.
Fig. 1. Construction of mutation controls. Step 1, two separate PCRs were performed to generate two overlapping fragments that contained the specific mutation in the overlapping region; step 2, the two PCR products were excised from the agarose gel and combined; step 3, dNTPs and Taq polymerase were added, and extension was performed for 10 cycles. After the extension was completed, the sense and antisense primers were added to the reaction, and a PCR of 30 cycles was carried out.
reverse allele-specific hybridization Oligonucleotides were designed to specifically hybridize with either wild-type or mutant PCR products (Table 3). The oligonucleotides were 59 labeled with an NH2 group and immobilized on nylon membranes (15 ). Oligonucleotides were applied as lanes, and the membranes were cut into strips, producing a barcode-like pattern of the oligonucleotides. The biotinylated PCR products containing wild-type sequences and possible specific mutations were hybridized to the membrane strips, using the VARISTRIP Detection Assay kit (ViennaLab). This kit provides alkaline denaturation buffer, a hybridization mixture, wash buffers, a streptavidin-horseradish peroxidase complex, and color developer containing p-nitroblue tetrazolium chloride/5-bromo-4-chloro-3-indolyl-phosphate.
sequencing To confirm the mutations detected by reverse ASO hybridization, we sequenced PCR products, using a 310
Table 3. Hybridization oligonucleotides. Oligonucleotide
Sequence (5*–3*)
185delAG-WTa 185delAG-Mut 188del11-Mut C61G-WT C61G-Mut I5 splice-WT I5 splice-Mut 1135insA-WT 1135insA-Mut 129del40-WT 1294del40-Mut 1411insT-WT 1411insT-Mut 1675delA-WT 1675delA-Mut Q563X-WT Q563X-Mut 2804delAA-WT 2804delAA-Mut E908X-WT E908X-Mut 3121delA-WT 3121delA-Mut 3450del4-WT 3450del4-Mut 4184del4-WT 4184del4-Mut R1443X-WT R1443X-Mut R1443G-Mut 5382insC-WT 5382insC-Mut
AGAAAATCTTAGAGTGTCb GAAAATCTT D AGTGTCCC ATCTTAGADGGTAAGTCAGCA GGCCTTCACAGTGTC GCCTTCACAGGGTC TCTCAAACAATTTAATTT C CTCAAACAAGTTAATTTC CACAGAAAAAAA GGTAG ACAGAAAAAAAAGGTAGA TTAGGTTCTGATGACTC GTGATGAACDAAATGCC GAAAATAGAC TTACTGG GAAAATAGACTTTACTGG TGAGGATTTTATCAAGAAA TGAGGATTTTATCADGAAAG AAGGTGATTCTATTCAGAATGA AAAGGTGATTCTATTTAGAATG CCTTAAAGAAACAAAGTCC CCTTAAAGAAACA D GTCCAA GAACAAAAGGAAGAAAATC GAACAAAAGGAATAAAATCA GCTAGAGGAAAACTTTG TGCTAGAGGDAAACTTTG TAAAAAAGCAAGAATATGA TAAAAAAG D AATATGAAGAAG GGAAGAAAATAATCAAG GGAAGAAAATAA D GAAGAGCA TGAGGACCTGCGAAAT GAGGACCTGTGAAAT GAGGACCTGGGAAAT AGAGAA TCCCAGGACA AGAGAACTCCCAGGAC
a b
WT, wild-type; Mut, mutant. Sequences in bold font are the mutation sites.
Clinical Chemistry 45, No. 7, 1999
979
Genetic Analyzer and BigDyeTM Terminator Cycle Sequencing Kit (Applied Biosystems).
histopathology The specimens were obtained from adnexectomy. Tissues were fixed in buffered formalin and embedded in paraffin. The sections were routinely stained with hematoxylin and eosin.
Results and Discussion Examples of the detection of the PCR products of BRCA1 by reverse ASO hybridization are shown in Figs. 2 and 3. As indicated, PCR products from both mutant and wildtype plasmids hybridized specifically with the corresponding oligonucleotides on the membrane strips. PCR products from DNA containing known mutations also specifically hybridized to the corresponding oligonucleotides. The aim of this study was to detect several possible mutations simultaneously. Therefore, it was important to optimize PCR and ASO hybridization conditions to allow the amplification and detection of fragments containing different mutations in a single assay. PCR optimization can be achieved by selecting primers with similar melting temperatures and amplicons with similar lengths and base composition. The specific hybridization of the PCR products to the immobilized oligonucleotides was dependent on the length of the hybridization oligonucleotides, the positions of mutations in the oligonucleotides, the temperature for hybridization and washing, and the salt concentration in the washing buffer. Because several different hybridization oligonucleotides were immobilized on a strip, the stringency of the hybridization could be adjusted only by the length and base composition of the oligonucleotides. We tested several oligonucleotides
Fig. 2. Examples of hybridization results with PCR products generated from plasmid DNAs containing the wild-type or the mutant fragment for four different mutations. Strip 1, 185delAG and 188del11 wild-type plasmid; strip 2, 185delAG mutant plasmid; strip 3, 188del11 mutant plasmid; strip 4, 1294del40 wild-type plasmid; strip 5, 1294del40 mutant plasmid; strip 6, 5382insC wild-type plasmid; strip 7, 5382insC mutant plasmid; strip 8, H2O control. The oligonucleotides immobilized at the corresponding positions on the strips are indicated on the left (wt, wild-type; mut, mutant).
Fig. 3. Examples of mutation detection in tumor samples. Strips 1 and 2, wild-type DNA probes; strips 3, 4, and 5, DNA samples from known carriers of the 185delAG, 1294del40, and 5382insC mutations, respectively. The oligonucleotides located at the corresponding positions on the strips are indicated on the left (wt, wild-type; mut, mutant).
that differed in length and in the positions of the mutations within the oligonucleotides to obtain clear positive signals and to avoid cross-hybridization with nonspecific PCR products. The hybridization process took ,2 h, and the results could be visualized directly from the strips. In our system, the oligonucleotides specific for both wildtype and mutant alleles of all 17 selected mutations were divided into three groups and immobilized on three different strips. We could screen 96 samples for all 17 mutations within 1 week. Therefore, this system is very useful to screen for known mutations in many samples. It is also appropriate for the detection of BRCA1 mutations in familial breast and ovarian carcinomas as a prescreening method before analysis of the whole coding sequence. To evaluate the detection limit of this system, we chose three mutations that represented a short deletion (3450del4), an insertion (1135insA), and a single-base exchange (E908X). As shown in Fig. 4, this system can detect as little as 2% mutant DNA in DNA samples. Thus, this combined PCR-reverse ASO hybridization method for the simultaneous detection of several known mutations is also applicable for mutation analysis in tumor samples containing healthy tissue. In addition, this method may be used for mutation detection in genes other than BRCA1. Two of the 122 sporadic ovarian carcinomas evaluated contained the Cys61Gly mutation in exon 5 of BRCA1. However, analysis of the corresponding DNA extracted from peripheral blood confirmed that these two mutations were germline mutations. Further histopathological analysis revealed that one of these ovarian carcinomas was a metastasis of a primary fallopian tube carcinoma. The dilated tube was filled with papillary and solid tumor mass penetrating the muscularis as well as spreading to the ovary. The neoplasm was composed of fine branching
980
Tong et al.: Mutation Detection in the BRCA1 Gene
This project was supported by the Anniversary Fund of the Austrian National Bank for the Promotion of Scientific ¨ NB 5572 and O ¨ NB 6056). We Research and Teaching (O thank Drs. Simon Gayther and Bruce Ponder of the Department of Pathology, Cambridge University, for providing us with three control DNA samples with known BRCA1 mutations. We also thank Eva Schuster for excellent technical assistance.
References
Fig. 4. Evaluation of the detection limit. Strips 1–9 represent mixed DNAs containing 90%, 50%, 20%, 10%, 5%, 2%, 1%, 0.5%, and 0% mutant plasmid DNA, respectively; strip 10, H2O control. The oligonucleotides located at the corresponding positions on the strips are indicated on the left (wt, wild-type; mut, mutant).
papillae covered by one or more layers of epithelium with enlarged pleomorphic and hyperchromatic nuclei with increased and abnormal mitoses. In some areas, the tumor exhibited solid sheets of cells with small or large foci of necrosis. The tumor biopsy from the tubal carcinoma was also confirmed to contain the Cys61Gly mutation by both reverse ASO hybridization and direct sequencing. To our knowledge, this is the first report that a BRCA1 mutation is associated with primary tubal carcinoma. To date, there have been no systematical studies on the BRCA1 mutations and fallopian tube carcinomas. However, a recent study suggested that serous tubal carcinoma may have a molecular pathogenesis similar to those of serous ovarian and uterine carcinomas because these three types of carcinomas have strikingly similarities in the frequency and pattern of chromosomal changes (18 ). At the initiation of this study in 1996, the 17 selected mutations accounted for .50% of all described mutations. However, many novel mutations have been found since then. Thus, the 17 analyzed mutations represent only 30% of the mutations published to date (BIC, February 1999). Although we analyzed 122 sporadic ovarian carcinomas, we could not detect any mutations other than the two germline mutations described above. Thus, the 17 specific mutations in the BRCA1 gene do not play a major role in the tumorigenesis and progression of sporadic ovarian cancer.
1. Miki Y, Swensen J, Shattuck-Eidens D, Futreal PA, Harshman K, Tavtigian S, et al. A strong candidate for the breast and ovarian cancer susceptibility gene BRCA1. Science 1994;266:66 –71. 2. Chen Y, Farmer AA, Chen CF, Jones DC, Chen PL, Lee WH. BRCA1 is a 220-kDa nuclear phosphoprotein that is expressed and phosphorylated in a cell cycle-dependent manner. Cancer Res 1996;56:3168 –72. 3. Scully R, Chen J, Plug A, Xiao Y, Weaver D, Feunteun J, et al. Association of BRCA1 with Rad51 in mitotic and meiotic cells. Cell 1997;88:265–75. 4. Somasundaram K, Zhang HB, Zeng YX, Houvras Y, Peng Y, Zhang HX, et al. Arrest of the cell cycle by the tumour-suppressor BRCA1 requires the CDK-inhibitor p21WAF1/CiPL. Nature 1997;389:187– 90. 5. Zhang HB, Somasundaram K, Peng Y, Tian H, Zhang HX, Bi D, et al. BRCA1 physically associates with p53 and stimulates its transcriptional activity. Oncogene 1998;16:1713–21. 6. Easton DF, Bishop DT, Ford D, Crockford GP. Genetic linkage analysis in familial breast and ovarian cancer: results from 214 families. The Breast Cancer Linkage Consortium. Am J Hum Genet 1993;52:678 –701. 7. Merajver SD, Pham TM, Caduff RF, Chen M, Poy EL, Cooney KA, et al. Somatic mutations in the BRCA1 gene in sporadic ovarian tumours. Nat Genet 1995;9:439 – 43. 8. Hosking L, Trowsdale J, Nicolai H, Solomon E, Foulkes W, Stamp G. et al. A somatic BRCA1 mutation in an ovarian tumour. Nat Genet 1995;9:343– 4. 9. Papp J, Csokay B, Bosze P, Zalay Z, Toth J, Ponder B, Olah E. Allele loss from large regions of chromosome 17 is common only in certain histological subtypes of ovarian carcinomas. Br J Cancer 1996;74:1592–7. 10. Beckmann MW, Picard F, An HX, Van Roeyen CR, Dominik SI, Mosny DS, et al. Clinical impact of detection of loss of heterozygosity of BRCA1 and BRCA2 markers in sporadic breast cancer. Br J Cancer 1996;73:1220 – 6. 11. Roa BB, Boyd AA, Volcik K, Richards CS. Ashkenazi Jewish population frequencies for common mutations in BRCA1 and BRCA2. Nat Genet 1996;14:185–7. 12. Matsushima M, Kobayashi K, Emi M, Saito H, Saito J, Suzumori K, Nakamura Y. Mutation analysis of the BRCA1 gene in 76 Japanese ovarian cancer patients: four germline mutations, but no evidence of somatic mutation. Hum Mol Genet 1995;4:1953– 6. 13. Shattuck-Eidens D, McClure M, Simard J, Labrie F, Narod S, Couch F, et al. A collaborative survey of 80 mutations in the BRCA1 breast and ovarian cancer susceptibility gene. JAMA 1995;273:535– 41. 14. FitzGerald MG, MacDonald DJ, Krainer M, Hoover I, O’Neil E, Unsal H, et al. Germ-line BRCA1 mutations in Jewish and non-Jewish women with early-onset breast cancer. N Engl J Med 1996;334: 143–9. 15. Zhang Y, Coyne MY, Will SG, Levenson CH, Kawasaki ES. Single-
Clinical Chemistry 45, No. 7, 1999
base mutational analysis of cancer and genetic diseases using membrane bound modified oligonucleotides. Nucleic Acids Res 1991;19:3929 –33. 16. Saiki RK, Walsh PS, Levenson CH, Erlich HA. Genetic analysis of amplified DNA with immobilized sequence-specific oligonucleotide probes. Proc Natl Acad Sci U S A 1989;86:6230 – 4. 17. Serov SF, Scully RE, Sobin LH. Histological typing of ovarian tumours. International histological classification of tumours, No. 9. Geneva: World Health Organization, 1973.
981
18. Pere H, Tapper J, Seppa¨la¨ M, Knuutila S, Butzow R. Genomic alterations in fallopian tube carcinoma: comparison to serous uterine and ovarian carcinomas reveals similarity suggesting likeness in molecular pathogenesis. Cancer Res 1998;58: 4274 – 6. 19. Friedman LS, Ostermeyer EA, Szabo CI, Dowd P, Lynch ED, Rowell SE, King MC. Confirmation of BRCA1 by analysis of germline mutations linked to breast and ovarian cancer in ten families. Nat Genet 1994;8:399 – 404.
