BRCA1 and BRCA2 proteins are crucial for DNA repair through homologous recombination (HR), which predominantly takes place during S and G2 phases. Their expression is tightly regulated to ensure HR occurs exclusively within these phases. While these proteins are well-established tumor suppressors in hereditary breast and ovarian cancers, their inactivation is rare across all sporadic cancers. Counterintuitively, BRCA1 and BRCA2 expression is downregulated rather than upregulated following DNA damage and p53 activation. In this study, we demonstrate that BRCA1 and BRCA2 gene expression is governed by the same transcriptional mechanisms throughout the cell cycle, peaking in the S phase. During G0/G1 and following p53 activation, BRCA1/2 transcription is repressed by DREAM and RB:E2F repressor complexes. Importantly, this transcriptional repression occurs concurrently with the coordinated downregulation of numerous genes involved in cell cycle control and DNA repair pathways. Consistent with previous findings, this suppression notably affects members of the Fanconi anemia group and is mediated through the DREAM complex. Such broad transcriptional suppression facilitates exit from S phase, thereby promoting a fundamental shift in DNA repair mechanisms. Following DNA damage, we demonstrate that BRCA1/2 downregulation occurs indirectly through the p53-p21-DREAM/RB axis, wherein p53-induced p21/CDKN1A expression initiates repression dependent on DREAM and RB. These results, together with observations from previous studies, suggest that DNA repair shifts from HR to the error-prone pathways of non-homologous end joining (NHEJ) and single-strand annealing (SSA), resulting in chromosomal aberrations and cell death, thereby in fact preventing malignant transformation. Our findings elucidate the transcriptional regulation of BRCA1 and BRCA2 expression. These regulatory mechanisms, when considered alongside prior findings and hypotheses, may help explain why BRCA1 and BRCA2 proteins do not exhibit tumor-suppressive functions in most cell types.
Quaas et al. (Fri,) studied this question.