The tumor suppressor breast cancer 1 (BRCA1) contains two functional domains—an N-terminal RING domain that has ubiquitin E3 ligase activity when bound to its partner BARD1 (BRCA1-associated RING domain protein 1) and a pair of C-terminal BRCT motifs that interact with phosphorylated serine resides in DNA repair complexes. Mutations associated with hereditary forms of breast and ovarian cancer map to both domains, and it has therefore been assumed that both domains contribute to BRCA1’s tumor suppressor activity. Using a BRCA1 mutant with an Ile26→Ala (I26A) substitution in the RING domain that disrupts ligase activity without affecting binding to BARD1, Shakya et al. determined that the RING domain was dispensable for the tumor-suppressive activity of BRCA1. BRCA1I26A was as effective as wild-type (WT) BRCA1 in suppressing tumor formation in pancreatic and mammary cancer mouse models. Mice in which the WT Brca1 allele had been replaced with Brca1I26A were healthy and showed no greater incidence or severity of tumor formation than control animals but exhibited male sterility characterized by smaller testes and absence of spermatids and spermatozoa. In contrast, an allele of Brca1 carrying a mutation in the phosphorecognition BRCT domain failed to act as a tumor suppressor. The authors generated mice in which the WT Brca1 gene was replaced with an allele (Brca1S1598F) that corresponds to a human familial breast cancer susceptibility missense mutation (Ser1598→Phe) that disrupts the interaction between BRCA1 and phosphorylated DNA repair proteins. These animals were more prone to developing tumors, and embryonic fibroblasts from these mutants exhibited chromosomal instability and reduced recruitment of repair complexes to sites of DNA damage. These results revealed that removing the phosphorecognition ability of the BRCT motifs was as detrimental as completely deleting Brca1 in the pancreatic and mammary tumor models. Thus, contrary to widely held assumptions, the ubiquitin E3 ligase activity of BRCA1 is not required for tumor suppression but is required for some aspect of spermatogenesis, perhaps for meiotic sex chromosome inactivation or silencing satellite repeats, as suggested in the commentary by Greenberg.
R. Shakya, L. J. Reid, C. R. Reczek, F. Cole, D. Egli, C.-S. Lin, D. G. deRooij, S. Hirsch, K. Ravi, J. B. Hicks, M. Szabolcs, M. Jasin, R. Baer, T. Ludwig, BRCA1 tumor suppression depends on BRCT phosphoprotein binding, but not its E3 ligase activity. Science 334, 525–528 (2011). [Abstract] [Full Text]