Search

BRCA1 is a breast and ovarian cancer susceptibility gene. An inferred germline regulatory mutation was previously reported in the BRCA1-linked kindred K2035, based on the absence of transcripts from the BRCA1 allele associated with the cancer susceptibility haplotype. In this study, the promoter region of BRCA1 was examined in individuals from K2035 for evidence of a mutation which could halt transcription. Evaluation of a polymorphism located within intron 2 of BRCA1 gave results consistent with the presence of a large deletion in K2035 mutation carriers. Southern blot analysis identified unique restriction fragments which occurred as a result of a 14 kb deletion that removed both of BRCA1's transcription start sites (exons 1a and 1b) as well as exon 2. Sequencing indicated that unequal crossover between Alu repeats was the likely cause of the deletion. Similar deletions may be responsible for other reported inferred regulatory mutations, as well as unidentified mutations in families linked to BRCA1.

BRCA1 and BRCA2, genes predisposing to early-onset breast cancer, have been isolated and are characterized for mutation spectrum, risks of cancer, and function. The different methodologies to screen for mutations in BRCA1 and BRCA2 are briefly discussed including DNA-based methodologies and potential new assays. The numbers and types of mutations identified to date are described, including the problems of ascribing risk to missense mutations. Recurring, possibly founding mutations have been identified in several populations including Ashkenazi Jews, Icelanders, Swedes, and African Americans. From population-based studies, estimates are that 6%-10% of breast cancers are due to mutations in BRCA1 and BRCA2. Knowledge of mutation status raises additional questions including the interpretation of negative tests and the risks of breast and other cancers associated with positive test results.

BRCA1 germline mutations confer susceptibility to familial breast and ovarian cancer. Mutational hot spots have never been detected in BRCA1 cDNA. Some mutations have been reported several times whereas some others appear to be population-related. In this study a group of 36 Italian families were analysed for BRCA1 germline mutations. All of them were screened by allele-specific oligonucleotide hybridization (ASO) for three recurrent mutations (185delAG, 5382insC, nt332-T>G). Twenty families, selected because of their high risk of carrying BRCA1 mutations, were subjected to analysis of the entire coding sequence of the gene. A total of eight mutations were found. ASO screening demonstrated only one known mutation in one patient, whereas cycle sequencing revealed five new mutations. Three of these new mutations were frameshifts: one occurred in exon 11 (1499insA), one in exon 16 (4873delCA) and one in the splice site of exon 3 (252delAAgt). Two were missense mutations (Cys64Arg; Asn158Tyr). The same frameshift mutation, 1499insA, was detected in three unrelated families. Haplotype analysis supported the hypothesis that two of these families may have had common ancestors, whereas in the third family the analysis was uninformative. BRCA1 germline mutations were found in one out of two families with ovarian cancer, in five out of eight families with breast-ovarian cancer, and in two out of 11 families with breast cancer. All three families with 1499insA mutations included at least one case of ovarian cancer. The majority of the ovarian cancers (4/5) associated with detectable BRCA1 germline mutations were of serous histotype.

Certain germline mutations in either BRCA1 or BRCA2 confer a lifetime risk of developing breast cancer that may approach 90%. The BRCA1 185delAG mutation was found in 20% and the BRCA2 6174delT mutation in 8% of Ashkenazi Jewish women with early-onset breast cancer. The 185delAG mutation was observed in 0.9% of 858 Ashkenazi Jews unselected for a personal or family history of cancer. Assuming comparable age-specific penetrances, a carrier frequency of 0.3% was estimated for the 6174delT BRCA2 mutation. To test this hypothesis, we performed a population survey of 1,255 Jewish individuals. In two independent groups, a prevalence of approximately 1% (C.I. 0.6-1.5) was observed for the 6174delT mutation. The relative risk of developing breast cancer by age 42 was estimated to be 9.3 (C.I. 2.5-22.5) for 6174delT, compared to 31 (C.I. 11-77) for 185delAG. Analysis of 107 Ashkenazi Jewish women with breast cancer and a family history of breast or ovarian cancer confirmed a four-fold greater prevalence for the BRCA1 185delAG mutation compared to the BRCA2 6174delT mutation. Our findings suggest a difference in cumulative life-time penetrance for the two mutations. Genetic counseling for the one in 50 Ashkenazi Jewish individuals harbouring specific germline mutations in BRCA1 or BRCA2 must be tailored to reflect the different risks associated with the two mutations.

An integrated approach involving physical mapping, identification of transcribed sequences, and computational analysis of genomic sequence was used to generate a detailed transcription map of the 1. 0-Mb region containing the breast cancer susceptibility locus BRCA2 on chromosome 13q12-q13. This region is included in the genetic interval bounded by D13S1444 and D13S310. Retrieved sequences from exon amplification or hybrid selection procedures were grouped into physical intervals and subsequently grouped into transcription units by clone overlap. Overlap was established by direct hybridization, cDNA library screening, PCR cDNA linking (island hopping), and/or sequence alignment. Extensive genomic sequencing was performed in an effort to understand transcription unit organization. In total, approximately 500 kb of genomic sequence was completed. The transcription units were further characterized by hybridization to RNA from a series of human tissues. Evidence for seven genes, two putative pseudogenes, and nine additional putative transcription units was obtained. One of the transcription units was recently identified as BRCA2 but all others are novel genes of unknown function as only limited alignment to sequences in public databases was observed. One large gene with a transcript size of 10.7 kb showed significant similarity to a gene predicted by the Caenorhabditis elegans genome and the Saccharomyces cerevisiae genome sequencing efforts, while another contained a motif sequence similar to the human 2',3' cyclic nucleotide 3' phosphodiesterase gene. Several retrieved transcribed sequences were not aligned into transcription units because no corresponding cDNAs were obtained when screening libraries or because of a lack of definitive evidence for splicing signals or putative coding sequence based on computational analysis. However, the presence of additional genes in the BRCA2 interval is suggested as groups of putative exons and hybrid selected clones that were transcribed in consistent orientations could be localized to common physical intervals.

The lifetime risk of breast cancer may approach 80-90% in women who have germline mutations of either of two genes, BRCA1 or BRCA2. A single BRCA1 mutation, 185delAG, has been noted in approximately 20% of Ashkenazi Jewish women with early onset breast cancer and in 0.9% of the Ashkenazi population. We recently detected a 6174delT frameshift mutation in BRCA2 in an hereditary breast cancer kindred of Ashkenazi Jewish ancestry. Here, we investigated the frequency of this mutation in 200 women with early-onset breast cancer. Six of 80 Ashkenazi Jewish women (8%) diagnosed with breast cancer before the age of 42, wer heterozygous for the 6174delT mutation, compared to none of 93 non-Jewish women diagnosed with breast cancer at the same age (P = .005). These cases were ascertained without regard to family history. Two of 27 (7%) additional Jewish families in which the proband was diagnosed with breast cancer at age 42 to 50 and had a family history of breast or ovarian cancer had germline 6174delT mutations. The results of this report suggest that a recurrent mutation of BRCA1 and a recurrent mutation BRCA2 together may account for over a quarter of all early-onset breast cancer in the setting of a personal or family history of ovarian cancer in Ashkenazi Jewish women.

The BRCA2 gene on chromosome 13 has been shown to be associated with familial male and female breast cancer. Here we describe a study on BRCA2 in 21 Icelandic families, including 9 with male breast cancer. We have previously reported linkage to the BRCA2 region in an Icelandic male breast cancer family and subsequently found a strong indication of linkage to BRCA2 and the same BRCA2 haplotype in breast cancer cases from 15 additional families, indicating a common origin. We describe a five base-pair deletion in exon 9 of BRCA2 in an affected male from the male breast cancer family. The same mutation occurs in all the families with the shared BRCA2 haplotype indicating a founder effect. Among mutation carriers there are 12 males with breast cancer, which accounts for 40% of all males diagnosed with breast cancer in Iceland over the past 40 years. Three of them have no family history of breast cancer indicating that this mutation may have variable penetrance. The same BRCA2 mutation appears to be associated with different cancer phenotypes in this population including male and female breast cancer, prostate cancer, pancreas cancer and ovarian cancer.

The breast cancer susceptibility gene, BRCA2 on chromosome 13q12-13, was recently isolated. Mutations in BRCA2 are thought to account for as much as 35% of all inherited breast cancer as wall as a proportion of inherited ovarian cancer. Many BRCA2-linked families also contain cases of male breast cancer. We have analysed germline DNA from 50 males with breast cancer (unselected for family history) and 26 individuals from site-specific female breast and breast-ovarian cancer families for mutations in BRCA2. All 17 breast-ovarian cancer families have been screened for BRCA1 coding region mutations and none were detected. Conformation-sensitive gel electrophoresis (CSGE) analysis of PCR-amplified DNA followed by direct sequencing was used to detect sequence variants. Three of eleven individuals carry the same mutation, all are of Ashkenazi Jewish descent, supporting the observation by Neuhausen et al. in this issue that there is a common mutation in this population. Eleven truncating mutations and nine polymorphisms were identified -- all were coding region variants. No loss-of-transcript mutations were identified in the sixteen samples for which this analysis was possible. Seven of the nine disease-associated mutations were detected in the 50 men with breast cancers; for thus in our series, BRCA2 mutations account for 14% of male breast cancer, all but one of which had a family history of male and/or female breast cancer.

Colorectal cancer (CRC) has a strong familial component. Candidate genes for colorectal cancer have been identified through mutations in four mismatch repair genes (hMSH2, hMLH1, hPMS1, and hPMS2) and genes that are deleted or mutated in tumors (DCC, APC, and p53). Linkage analysis of candidate loci/regions was performed in 10 kindreds ascertained for common colorectal cancer from the Utah Population Database. Evidence for linkage to candidate genes was assessed using two- or three-point logarithm of the odds ratio scores with markers spanning the region of localization. One kindred is linked to hMSH2 and also fits the criteria for hereditary nonpolyposis colorectal cancer, having early age of onset and high penetrance for CRC. The remaining nine kindreds are unlinked to the candidate genes tested. These kindreds have a later age of onset and a lower penetrance than hereditary nonpolyposis colorectal cancer kindreds. these results indicate that further unmapped susceptibility loci may be responsible for much of the familial aggregation of CRC.

Several BRCA1 mutations have now been found to occur in geographically diverse breast and ovarian cancer families. To investigate mutation origin and mutation-specific phenotypes due to BRCA1, we constructed a haplotype of nine polymorphic markers within or immediately flanking the BRCA1 locus in a set of 61 breast/ovarian cancer families selected for having one of six recurrent BRCA1 mutations. Tests of both mutations and family-specific differences in age at diagnosis were not significant. A comparison of the six mutations in the relative proportions of cases of breast and ovarian cancer was suggestive of an effect (P = .069), with 57% of women presumed affected because of the 1294 del 40 BRCA1 mutation having ovarian cancer, compared with 14% of affected women with the splice-site mutation in intron 5 of BRCA1. For the BRCA1 mutations studied here, the individual mutations are estimated to have arisen 9-170 generations ago. In general, a high degree of haplotype conservation across the region was observed, with haplotype differences most often due to mutations in the short-tandem-repeat markers, although some likely instances of recombination also were observed. For several of the instances, there was evidence for multiple, independent, BRCA1 mutational events.

Loss of heterozygosity (LOH) was evaluated in 174 breast and ovarian tumors derived from 94 families with at least 3 first-degree relatives affected with either of these cancers. By linkage analysis 26 families were identified as having a high posterior probability of being due to BRCAl (the breast/ovarian cancer susceptibility locus on 17q12-21) with lod scores varying from 0.51 to 9.49. Tumor genotypes were determined at at least 2 constitutionally heterozygous markers flanking BRCAl in a total of 58 tumors from these families. These tumors were derived from 52 patients, the BRCAl mutation carrier status of which was evidenced by DNA sequencing in 20, and inferred by reconstructing haplotypes in the remainder. LOH was detected in 50 (86%) tumors, and invariably involved the wild-type allele. Where informative, this allele was of paternal origin in 33 cases and of maternal origin in 10 cases. These results strongly suggest that BRCAl is a tumor suppressor gene and that LOH is greatly favored to fully inactivate it.

The q21 region of chromosome 17 contains the gene BRCA1, which is involved in familial early-onset breast and ovarian cancers. A physical map of a region that extends from a distal boundary of the BRCA1 region, D17S78, to GP2B has been constructed. The map consists of 30 STSs, including 2 new short tandem repeat polymorphic markers. The contig is composed of a mixture of 7 YACs, 5 P1 plasmids, and 14 cosmids and was ordered by STS-content mapping.

Interest in the genetics of breast cancer has intensified with the discovery of a breast cancer susceptibility locus, BRCA1, on chromosome 17q. In this paper, we describe updated information on a large breast cancer kindred (K107) that has been extensively studied since 1948. Specifically, we have identified many new cases of cancer in the family and have shown that this family is unlinked to BRCA1 as well as a number of other genes considered as candidates for breast cancer. In a collaborative study between the University of Utah and the Institute of Cancer Research in the United Kingdom, we have collected a set of families with a predisposition to breast and ovarian cancers that have been reliably excluded from linkage to BRCA1 and evaluated their usage in a genomic search for other breast cancer loci. This effort led to the discovery of a second breast cancer locus located on chromosome 13q, BRCA2, which is responsible for the increased incidence of breast cancer in Kindred 107.

BRCA1, a breast-ovarian cancer susceptibility gene which has been localized to 17q21, appears to be a tumor suppressor gene based on evidence from loss of heterozygosity (LOH) studies. We analyzed 14 ovarian and breast tumors from BRCA1 carriers and 1 sporadic breast tumor from 3 kindreds for 17q21 LOH. Thirteen of the 14 tumors from gene carriers exhibited LOH of the wild-type allele. Tumors from one gene carrier and the sporadic breast case did not exhibit any LOH in the region. There was loss of the wild-type allele from both maternally and paternally derived chromosomes, therefore excluding the possibility of genomic imprinting and providing further evidence that BRCA1 is a tumor suppressor. Three tumors showed interstitial LOH in the region, and thus established the utility of familial tumors in refining a region surrounding a tumor suppressor gene in a manner analogous to using genetic recombinants.

BRCA1, a breast and ovarian cancer susceptibility locus, has been isolated and maps to 17q21. A physical map of the BRCA1 region which extended from the proximal boundary at D17S776 to the distal boundary at D17S78 was constructed and consists of 51 sequence tagged sites (STSs) from P1 and YAC ends, nine new short-tandem repeat (STR) polymorphic markers, and eight identified genes. The contig, which spans the estimated 2.3 Mb region, contains 29 P1s, 11 YACs, two BACs, and one cosmid. Based on key recombinants in two linked families, BRCA1 was further localized to a region bounded by D17S1321 on the proximal side and D17S1325 on the distal side. Within this estimated 600 kb region, the contig was composed completely of P1s and BACs ordered by STS-content mapping and confirmed by DNA restriction fragment fingerprinting.

A small proportion of breast cancer, in particular those cases arising at a young age, is due to the inheritance of dominant susceptibility genes conferring a high risk of the disease. A genomic linkage search was performed with 15 high-risk breast cancer families that were unlinked to the BRCA1 locus on chromosome 17q21. This analysis localized a second breast cancer susceptibility locus, BRCA2, to a 6-centimorgan interval on chromosome 13q12-13. Preliminary evidence suggests that BRCA2 confers a high risk of breast cancer but, unlike BRCA1, does not confer a substantially elevated risk of ovarian cancer.

A familial melanoma susceptibility locus (MLM) was identified on the short arm of chromosome 9 in a set of Utah and Texas kindreds. Subsequent confirmation was reported for a set of 26 Australian kindreds, a set of 7 Dutch kindreds, and a set of 13 NIH melanoma kindreds. The original localization report placed the locus near the IFNA-D9S126 interval on chromosome 9p. We report further localization using D9S171 and a newly developed marker, D9S736, both of which lie in the IFNA-D9S126 interval. Analysis of this set of kindreds places the melanoma susceptibility locus in a 2-cM region that is proximal to D9S736 and distal to D9S171.

The p21 region of human chromosome 9 is thought to contain a gene (MLM) involved in genetic susceptibility to melanoma and a gene or genes that influence progression of certain other tumors. Genomic clones that span a large region in 9p21 surrounding the presumptive tumor suppressor gene(s) have been isolated. A set of sequence-tagged sites in this region has been developed. By using these markers and others previously reported, the 9p21 region has been studied by physical mapping in 84 melanoma cell lines. A putative tumor suppressor gene, perhaps MLM itself, has been localized to a region of less than 40 kb that lies proximal (centromeric) to the alpha-interferon gene cluster.

The objectives of this study were to assess the degree of restriction fragment length polymorphism (RFLP) in Cucumis melo and to determine interrelationships among cultivated varieties. Initial screening of a genomic PstI library revealed that approximately 40% of the clones were repetitive. A total of 162 unique and low-copy sequence clones were hybridized to seven diverse accesions of C. melo and a C. sativus cultivar 'Pacer' to evaluate RFLP variation. Of these, 130 probes (80%) detected a polymorphism between C. melo accessions and C. sativus, and the majority were polymorphic with more than one enzyme digest. In contrast, only 53 probes (33%) were useful in differentiating at least one of the seven accessions. Of those, only 9% were informative with more than one enzyme digest. This indicates that within C. melo, the differences among accessions are due to infrequent base substitutions, whereas between the two species, differences are mainly due to genome rearrangements such as insertions and deletions or numerous base substitutions. Of the informative probes, 34 were used in analyzing 44 C. melo lines to establish a data base of RFLP hybridization patterns. Percent similarity based on RFLP profiles was computed among lines and analyzed by principal component analysis, to visualize relationships among lines. There were clear demarcations among, but not within, muskmelon and honeydew groups.

While compiling genetic linkage maps in several plant species based upon restriction fragment length polymorphisms (RFLPs), it was noted that the incidence of polymorphism differs among species. The basis of this disparity was investigated in this study by examining the nucleotide sequence at homologous loci among distinct cultivars within two species which exhibit considerably different levels of RFLPs. Using the polymerase chain reaction, homologous regions from different cultivars were first amplified and the nucleotide sequence of the products were determined. Four genomic regions of seven maize cultivars and three genomic regions of eight melon cultivars were examined to compare the respective levels of sequence variation between the two species. Levels of variation for both base substitutions and insertions/deletions varied widely among the maize sequences and between maize and melon for base substitutions. Estimates of theta (a measure of polymorphism) ranged from 0 to 0.002 in melon and from 0.006 to 0.040 for base substitutions and from 0.002 to 0.023 for insertions/deletions in maize. Critical value tests and chi-squared tests suggested that in maize the underlying processes generating and maintaining neutral mutations differ among the regions. The results not only suggest that several mechanisms are necessary to explain the variation seen in these two species, but also point to some basic dissimilarities in the organization and maintenance of the genomes of different plant species.