Assessing the RNA effect of 26 DNA variants in the BRCA1 and BRCA2 genes

• DOI: 10.1007/s10549-011-1661-5
• Publication Date: 2011-07-06T12:58:28Z
• Authors: Mireia Menéndez; Joan Castellsagué; Marc Mirete; Eva Pros; Lídia Feliubadaló; Ana Osorio; Mónica Calaf; Eva Tornero; Jesús del Valle; Juana Fernández-R...; Francisco Quiles; Mónica Salinas; Angela Velasco; Alex Teulé; Joan Brunet; Ignacio Blanco; Gabriel Capellá; Conxi Lázaro
• Keywords: BRCA2 Protein; 0301 basic medicine; Base Sequence; Models, Genetic; BRCA1 Protein; Sequence Analysis, RNA; Molecular Sequence Data; Alternative Splicing; 03 medical and health sciences; Mutation; Hereditary Breast and Ovarian Cancer Syndrome; Humans; Protein Isoforms; Computer Simulation; Female; Genetic Predisposition to Disease; RNA, Messenger

Comprehensive genetic testing of the breast cancer susceptibility genes BRCA1 and BRCA2 identified approximately 16% of variants of unknown significance (VUS), a significant proportion of which could affect the correct splicing of the genes. Our aim is to establish a workflow for classifying VUS in these complex genes, the first stage of which is splicing analysis. We used a combined approach consisting of five in silico splicing prediction programs and RT-PCR analysis for a set of 26 variants not previously studied at the mRNA level and six variants that had already been studied, four of which were used as positive controls as they were found to affect the splicing of these genes and the other two were used as negative controls. We identified a splicing defect in 8 of the 26 newly studied variants and ruled out splicing alteration in the remaining 18 variants. The results for the four positive and the two negative control variants were consistent with results presented in the literature. Our results strongly suggest that the combination of RNA analysis and in silico programs is an important step towards the classification of VUS. The results revealed a very high correlation between experimental data and in silico programs when using tools for predicting acceptor/donor sites but a lower correlation in the case of tools for identifying ESE elements.