Color bar coding the BRCA1 gene on combed DNA: a useful strategy for detecting large gene rearrangements.

Genetic linkage data have shown that alterations of the BRCA1 gene are responsible for the majority of hereditary breast and ovarian cancers. BRCA1 germline mutations, however, are found less frequently than expected. Mutation detection strategies, which are generally based on the polymerase chain reaction, therefore focus on point and small gene alterations. These approaches do not allow for the detection of large gene rearrangements, which also can be involved in BRCA1 alterations. Indeed, a few of them, spread over the entire BRCA1 gene, have been detected recently by Southern blotting or transcript analysis. We have developed an alternative strategy allowing a panoramic view of the BRCA1 gene, based on dynamic molecular combing and the design of a full four-color bar code of the BRCA1 region. The strategy was tested with the study of four large BRCA1 rearrangements previously reported. In addition, when screening a series of 10 breast and ovarian cancer families negatively tested for point mutation in BRCA1/2, we found an unreported 17-kb BRCA1 duplication encompassing exons 3 to 8. The detection of rearrangements as small as 2 to 6 kb with respect to the normal size of the studied fragment is achieved when the BRCA1 region is divided into 10 fragments. In addition, as the BRCA1 bar code is a morphologic approach, the direct observation of complex and likely underreported rearrangements, such as inversions and insertions, becomes possible.

Screening of male breast cancer and of breast-ovarian cancer families for BRCA2 mutations using large bifluorescent amplicons.

41 breast cancer or breast-ovarian cancer families, including 12 families with at least one affected first-degree male relative, were screened for mutations in the BRCA2 gene. Mutations had not been found in the BRCA1 gene of these families. Chemical cleavage of mismatch was used to identify nucleotide changes within large PCR products (average size 1.2 kb) that carried strand-specific fluorescent end-labels. 15 amplicons were sufficient to scan 18 exons, including the large exon 11. The remaining 9 small exons were examined by Denaturing Gradient Gel Electrophoresis. The high sensitivity of this approach was documented by the detection, in these 41 patients, of all 9 exonic single nucleotide polymorphisms reported with heterozygosity >0.1. Truncating BRCA2 mutations were found in 7 of the 41 families. 3 of them were in the group of 12 families comprising cases of male breast cancer. Since the methods used here have no bias for particular types of mutations, these data confirm the high proportion of frameshifts among mutations in BRCA2. However, relevant single nucleotide substitutions were also found: one resulting in a stop codon and another one, present in a male patient, was the previously reported change Asp2723His, that affects a highly conserved region of the BRCA2 protein. This study indicates a BRCA2 contribution of 10% (95% CI 2.5-17.5) to our original cohort of 59 breast-ovarian cancer families, whereas the contribution of BRCA1 had been estimated at 46% (95% CI 33-59).

Genetic testing for breast cancer predisposition.

The recent identification of the BRCA1 and BRCA2 genes has improved our understanding of the association between breast and ovarian cancers in certain families. Carriers of predisposing germline mutations must decide on different options for management, including close follow-up or prophylactic surgery. Further studies are needed to elucidate the optimal management of these patients and to identify the factors that modify their risk for developing breast cancer. Finally, we must work to prevent any form of discrimination against those who, following genetic testing, are found to be at increased risk for breast cancer.

No evidence for germline PTEN mutations in families with breast and brain tumours.

Germline mutations of the PTEN gene are involved in Cowden disease, a genetic condition associated with an increased risk of breast cancer. Further somatic PTEN mutations have been found in glioblastomas and to a lesser extent in meningiomas. Therefore, PTEN germline mutations were searched for in a series of 20 unrelated women with breast cancer who also had a personal or familial breast-brain tumour history. Inclusion criteria were 1. family history of breast cancer; 2. absence of germline BRCA1 and p53 mutation; and 3. at least one case of brain tumour (glioblastoma, meningioma, or medulloblastoma) in either the index case or one of their first or second degree relatives. Any stigmata of Cowden disease was an exclusion criteria. Screening of the PTEN gene for point mutations or small rearrangements were performed using the denaturing gradient gel electrophoresis method on the 9 coding exons. No disease-associated mutation of the PTEN gene has been detected in our series. It is, thus, unlikely that PTEN is a significant BRCA predisposing locus. However, one might ask whether breast cancer cases resulting from germline PTEN mutation could occur without any mammary histological feature of Cowden disease.

Denaturing high-performance liquid chromatography detects reliably BRCA1 and BRCA2 mutations.

Denaturing high-performance liquid chromatography (DHPLC) is a recently developed method of comparative sequencing based upon heteroduplex detection. To assess the reliability of this method, 180 different mutations (54 deletions, 12 insertions, and 117 single base substitutions) in BRCA1 and BRCA2 were tested. Second, 25 index individuals with complete DHPLC analysis of BRCA1 were reanalyzed by dye-terminator sequencing. Third, 41 index individuals were analyzed concomitantly by both DGGE and DHPLC. Of the 180 different BRCA1 and BRCA2 mutations, 179 showed heterozygous DHPLC elution profiles. Dye-terminator sequencing of the entire BRCA1 gene, including 5592 bp of coding sequence and 5206 bp of flanking noncoding sequence, in 25 index individuals did not reveal additional variants missed by DHPLC. The concomitant analysis of 41 index cases showed that 4 probably disease-associated mutations were identified by DHPLC while only 3 of those 4 sites were detected by denaturing gradient gel electrophoresis. We conclude that DHPLC is a sensitive and cost-effective method for the screening of BRCA1 and BRCA2.