The RAD51 recombinase is a critical effector of Homologous Recombination (HR), which is an essential DNA repair mechanism for double-strand breaks. The RAD51 protein is recruited onto the DNA break by BRCA2 and forms homopolymeric filaments that invade the homologous chromatid and use it as a template for repair. RAD51 filaments are detectable by immunofluorescence as distinct foci in the cell nucleus, and their presence is a read out of HR proficiency. RAD51 is an essential gene, protecting cells from genetic instability. Its expression is low and tightly regulated in normal cells and, contrastingly, elevated in a large fraction of cancers, where its level of expression and activity have been linked with sensitivity to genotoxic treatment. In particular, BRCA-deficient tumors show reduced or obliterated RAD51 foci formation and increased sensitivity to platinum salt or PARP inhibitors. However, resistance to treatment sets in rapidly and is frequently based on a complete or partial restoration of RAD51 foci formation. Consequently, RAD51 could be a highly valuable therapeutic target. Here, we review the multiple levels of regulation that impact the transcription of the RAD51 gene, as well as the post-translational modifications that determine its expression level, recruitment on DNA damage sites and the efficient formation of homofilaments. Some of these regulation levels may be targeted and their impact on cancer cell survival discussed.

Triple-negative breast cancer (TNBC), representing 15% of breast carcinomas, is an aggressive breast cancer subtype with a high probability of metastasis and limited treatment options. Noticeably, BRCA-deficiency occurs in 25% of the TNBCs and results in deficient homologous recombination (HR) repair. Interestingly, PARP inhibitors (PARPi) have shown synthetic lethality in a BRCA-deficient context; however, their efficacy is frequently hampered by intrinsic or acquired resistance mechanisms involving restoration of the HR. In that regard, the role of some CDKs proven to regulate key HR actors was of interest to us. In this study, we aimed to understand the rewiring pathways determining resistance to PARPi in BRCA-deficient cancers and to assess the role of transcriptional regulating CDKs such as CDK7, CDK9, or CDK12 in the transcriptional regulation of key HR genes. Our ultimate goal was to determine whether and which CDK inhibitors could be effective approaches to repress HR gene expression and induce pharmacological HR-deficiency. As such, these CDK-inhibitors (CDKi) could be molecules of choice allowing sensitization of tumors that would otherwise respond poorly to DNA damaging treatment. With this purpose, we used in vitro and in vivo (PDX) models of TNBC and studied the attenuation of the HR response in tumor cells and PDX models treated with CDK-inhibitors. Our final aim was to determine the most efficient combination of CDKi and PARPi Our HR read outs were RAD51 and BRCA1 foci formation upon PARPi treatment. We also measured the modification of RNA and protein expression levels induced by CDKi treatment on a series of diagnostic HR genes (BRCA2, PALB2, ATR, FANCD2), as a measure of HR repression. We present data comparing the relative efficiency of three 3 CDKi; dinaciclib, NVP-2, and SR-4835, which have different specificities and inhibit different CDKs with variable efficacy.

In triple negative breast cancer (TNBC), chemotherapy is the only systemic treatment and sustained remissions are rare. We propose to widen therapeutic options. About 30% of TNBC tumors are BRCA1-deficient, presenting defective DNA repair and increased sensitivity to genotoxic drugs. We hypothesized that BRCA-deficient TNBC is highly sensitive to replication stress-inducing drugs, thus, opening new therapeutic perspectives. Our preliminary results showed that BRCA1-deficient TNBC cell lines and a CRISPR/Cas9 BRCA1 KO isogenic model display increased sensitivity to gemcitabine. Cell cycle distribution of gemcitabine-treated BRCA1-deficient cells was characterized by an elevated Sub-G1 fraction caused by increased numbers of cells in replication catastrophe. This was illustrated by 80% of BRCA1-deficient cells showing persistent (48–72 h post treatment) gH2AX staining in the absence of RPA32 co-staining, whereas in the isogenic BRCA1 WT model gH2AX and RPA32 positive cell numbers started decreasing at 24 h. Interestingly, we noted that in addition to replication catastrophe, BRCA-deficient cells treated with gemcitabine underwent aberrant mitosis as shown by a clear increase of micro-nuclei. Interestingly, in vivo experiments appear to reproduce in vitro data. Indeed, a BRCA hypermethylated TNBC PDX showed a higher sensitivity to gemcitabine than the BRCA1 WT. In conclusion, our data suggest that BRCA-deficient tumors are more sensitive to the replication poison gemcitabine. Furthermore, this sensitivity seems to be mediated by an accentuated replicative stress response that is not well managed. Upon gemcitabine treatment, the cells undergo important DNA damage that leads to stalled replication forks, and DNA breakage. In the absence of BRCA1, the HR pathway is compromised, which leads to fork collapse and accumulation of single-stranded DNA, therefore exhausting the pool of RPA within the cell and inducing replicative catastrophe. In addition to deficient replication, gemcitabine treated BRCA-deficient, but not BRCA-proficient cells, are subjected to mitotic catastrophe.