RESUMEN
BRCA2 is essential for homologous recombination DNA repair. BRCA2 mutations lead to genome instability and increased risk of breast and ovarian cancer. Similarly, mutations in BRCA2-interacting proteins are also known to modulate sensitivity to DNA damage agents and are established cancer risk factors. Here we identify the tumor suppressor CDK5RAP3 as a novel BRCA2 helical domain-interacting protein. CDK5RAP3 depletion induced DNA damage resistance, homologous recombination and single-strand annealing upregulation, and reduced spontaneous and DNA damage-induced genomic instability, suggesting that CDK5RAP3 negatively regulates double-strand break repair in the S-phase. Consistent with this cellular phenotype, analysis of transcriptomic data revealed an association between low CDK5RAP3 tumor expression and poor survival of breast cancer patients. Finally, we identified common genetic variations in the CDK5RAP3 locus as potentially associated with breast and ovarian cancer risk in BRCA1 and BRCA2 mutation carriers. Our results uncover CDK5RAP3 as a critical player in DNA repair and breast cancer outcomes.
RESUMEN
Fanconi anemia (FA) and ataxia telangiectasia (AT) share common traits such chromosomal instability and proneness to hematological cancers. Both AT and FA cell lines, and patients, are characterized by abnormally high levels of oxidative stress markers. The key FA protein FANCD2 is phosphorylated on Ser 222 by ATM after ionizing radiation (IR), thus allowing normal activation of the S-phase checkpoint, and ATM cells are known to be hypersensitive to oxidative damage. In this work we show that FANCD2 deficient cells have a defective S-phase checkpoint after Hydrogen Peroxide (H(2)O(2)) induced oxidative damage. ATM dependent phosphorylation of FANCD2 at the S222 residue is necessary for normal S-phase checkpoint activation after oxidative stress, while FANCD2 monoubiquitination at K561 is dispensable. We also show that FANCD2 is not required for base excision repair of 8-oxoG and other DNA lesions (abasic sites, uracils), while treatments that exclusively induce 8-oxoG, but not DNA double strand breaks, fail to activate FANCD2 monoubiquitination, thus indicating that the known accumulation of 8-oxoG in FA cells reflects an overproduction of ROS rather than defective processing of oxidized bases. We conclude that the handling of DNA damage after H(2)O(2)-induced oxidative stress requires the coordinated action of FANCD2 and ATM.