Description

Proteins encoded by the breast cancer susceptibility genes BRCA1 and BRCA2 and their partners form several networks.1 One such protein complex is responsible for the maintenance of the genome by repairing damaged DNA and preventing harmful gene rearrangements.2, 3 Deleterious germ-line mutations in BRCA1 and BRCA2 are mainly associated with increased risks for breast and ovarian cancers, but pancreas and prostate cancers are also observed at increased frequency in BRCA2 carriers. Nonetheless, most cases of these cancers are not attributable to mutations in these genes. To date, known protein partners of BRCA1 and BRCA2 have been investigated to determine possible associations between their genetic variants and an increase in the risk of developing cancers.4 One example is PALB2, which interacts with both BRCA1 and BRCA2 proteins, and multiple different constitutional mutations have been found to be associated with an elevated risk of breast and pancreatic cancers.5, 6, 7 These and other studies show the importance of investigating additional genes from the BRCA network as potential breast cancer susceptibility genes.

Recently, MRG15 (also known as MORF4L1), a new BRCA complex-interacting protein, has been characterized. It participates in the response to double-strand breaks by recruiting the BRCA complex to sites of damaged DNA.8 MRG15 is encoded by a gene located on 15q24. It is a 41.5 kDa chromodomain protein that directly binds to PALB2.9 This interaction is essential for homologous recombination regulation during DNA repair. This functional relationship with PALB2 means that MRG15 is a candidate breast cancer susceptibility gene. Consequently, we have formally investigated this hypothesis by screening germ-line DNA from a set of BRCA1/2 mutation-negative breast cancer cases for potential cancer-related genetic alterations.

Approval to perform the study was obtained from the local ethical review boards. DNA from blood of affected index cases of 232 breast or breast/ovarian cancer families was screened for MRG15 mutations. All families were negative for the relevant BRCA1/2 founder mutations in the Finnish (n=148), French-Canadian (n=45) and Ashkenazi Jewish groups (n=9) and/or were negative for BRCA1/2 mutations by full sequencing. The average age at diagnosis of the index case was 48.12 years (range 19–89 years) and 225 of 232 (97%) probands presented with breast cancer (including three men), with the remainder presenting with ovarian cancer. A total of 179 families (77%) had three or more affected individuals, 44 families (19%) had two affected individuals and nine index cases (4%) presented with either breast cancer <30 years (n=3) or bilateral breast cancer (n=6), with no family history.

Mutation screening of the coding and exon–intron boundaries of the gene was carried out by a combination of high-resolution melting mutation scanning analysis10 and DNA-sequencing analysis using an ABI 3130 XL DNA analyzer (Applied Biosystems, Foster City, CA, USA). Primer information is summarized in Table 1. Electropherograms were aligned using Lasergene-SeqMan software (DNASTAR, Madison, WI, USA) and every exon was screened for variants by comparing with MRG15 reference sequence (NC_000015.9, range 79165171–79190074).

Table 1 Primer information

In total, we identified 21 variants: 17 intronic changes, 3 coding SNPs and 1 variant localized in the 3′UTR (Table 2). Six out of the 17 intronic variants plus the one localized in the 3′UTR were referenced neither in dbSNP nor Ensembl SNP databases. These seven variants were each detected in not more than 1% of the patients indicating they could be rare and potentially harmful variants (Table 2). To gain further insight into possible negative effects of these non-referenced variants, we performed in silico analyses using accurate splicing prediction software: BDGP,11 NetGene 2 server12, 13 and GENSCAN.14 None of these intronic changes were predicted to impact on the splicing of the pre-mRNA. Analysis of the 3′UTR, using TargetScanHuman 5.1 software,15 revealed that the site where the variant in the 3′UTR lies is not a known target of miRNAs. Consequently, it is very unlikely that this variant has any effect on the regulation of MRG15 mRNA expression.

Table 2 Variants detected in MRG15

In conclusion, we found no evidence that mutations in MRG15 strongly predispose to breast cancer. Given our relatively small sample size, it remains possible that rare, or very rare, variants in this gene will be found to increase the risk for breast cancer. Moreover, sequencing of other novel genes whose products interact with BRCA proteins and their partners should be undertaken as members of this network are some of the most important contributors to the inherited susceptibility to breast cancer.