Characterization of BRCA1-induced genomic instability and identification of a novel PARP1 interactor

Germline mutations in BRCA1 gene confer the highest lifetime risks to develop breast and ovarian cancers and are the most frequently mutated gene in women with these cancers. Among all breast cancer 5-10% are hereditary, with a germline mutation in one susceptibility gene and BRCA1 mutation are inherited in an autosomal dominant pattern. BRCA1 is a tumor suppressor gene which encodes for a nuclear protein involved in a wide array of cellular functions such as: DNA repair, cell cycle regulation, genomic stability maintenance, transcription regulation, replication and chromatin remodeling. Among these, it is involved mainly in DNA double-strand breaks repair through homologous recombination pathway. To correctly perform this mechanism, BRCA1 must interact at various level, with several homologous recombination repair proteins such as MRE11, RAD50, RAD51 or RAD51, and also mismatch repair proteins such as MSH2 or MSH6. On the basis of results that expression of BRCA1 cancer related missense variants increases spontaneous homologous recombination in diploid strain of the yeast Saccharomyces cerevisiae, in this study we used this model organism to study contribution of BRCA1 partners in genomic instability caused by BRCA1 pathogenic variants expression. Since, even if it lacks the homologous of BRCA1 gene, preserves the same repair pathways as in mammals and it contains homologous genes for almost all DNA repair partners of BRCA1. While expression of BRCA1 pathogenic variants causes increase of recombination in wild-type yeast, in yeast strains deleted for genes involved in DNA repair cause a drastic reduction in recombination events, meaning that missing of these partners cause reduction of BRCA1 recombination induction. We decided to switch to gene reversion analysis which is a direct indication of genome instability in the same strains and, in this case, deletion of specific partners of BRCA1 such as RAD50 or MSH2 cause a drastic increase in gene reversion frequency. Next step it was to analyze effect of BRCA1 expression in phase-arrested cells on gene reversion, in order to eventually relate gene reversion events increases caused by expression of some variants to a particular phase of the cell cycle, therefore giving indication about its pathogenicity to mechanisms active in specific phase. Interestingly pathogenic missense mutations located in the domain mainly involved in DNA repair determine this increase in S phase arrested cells, therefore when DNA replication is in progress, on the other hand, a pathogenic mutation on the opposite domain involved in E3 ubiquitinates activity cause increase in G1 phase. Last thing analyzed about BRCA1, it was its effect on recombination in presence of DNA damage agent methyl-methane sulfonate (MMS). Interestingly, MMS induced-recombination is strongly reduced when two pathogenic variants are expressed. All these data together confirm the power of this simple model to give indication about BRCA1 missense variants pathogenicity, giving a clearer idea of BRCA1 network of interaction and relate specific cell phase activity to this pathogenicity. Therefore, integrating these data with available genetic data could help to understand cancer predisposition of BRCA1 variants and could also help to develop personalized therapy considering specific partner missing interaction or specific activity deficiency. A second task of this work was to analyze new candidates as PARP1 interactors. PARP1 is a nuclear protein that attaches a poly(ADP-ribose) polymer (PAR) to itself and other target proteins dealing with DNA repair and related pathways. Inhibition of this protein is used in breast cancer therapy, where BRCA1 functionality is deficient, but resistance to these drugs could arise. Therefore, identification of new functional interactors could help to overcome development of resistance and could identify new targets to treat in a combinational approach. Starting from a wide-genome screening performed in yeast by our group, I integrated data coming from this screening with two other screenings performed by other groups. I identified SART1 as best candidate interactor. Then, I confirmed interaction between SART1 and PARP1 in two human cell lines by co-immuno-precipitation and, analyzed the effect of SART1 inhibition on PARP1 activity identifying a deficiency in binding of poly(ADP-ribose) polymer to itself