BRCA1 interacts via its BRCT domains with ezrin, radixin and moesin, and colocalizes with F-actin at the plasma membrane to control cell spreading and motility.

For well over a decade, the exact mechanism through which breast cancer susceptibility 1 (BRCA1) functions as a tumor suppressor has evaded scientists. However, research by Coene et al., outlined in The Journal of Cell Biology, has uncovered an unexpected function for full-length BRCA1 beyond its known nuclear roles — in suppressing cell invasion.

Full-length BRCA1 contains an amino-terminal RING domain with E3 ubiquitin ligase activity and a pair of BRCA1 carboxy-terminal (BRCT) domains, which are essential for its tumor-suppressive activity. The authors probed HeLa postnuclear membrane pellet fractions (to exclude nuclear BRCT-interacting proteins) using a region of BRCA containing the two BRCT domains tagged with enhanced green fluorescent protein — EGFP-CTD. They identified the proteins ezrin, radixin and moesin (ERM), which are known to link the plasma membrane to the actin cytoskeleton during cell migration, as independent binding partners. EGFP-CTD colocalized not only with the ERM complex but also with filamentous actin (F-actin) in membrane extensions and at points of cell adhesion with the extracellular matrix.

This interaction led the authors to hypothesize that BRCA1 might affect migration, and when they induced migration in HeLa cells, colocalization of F-actin, ERM and EGFP-CTD increased in membrane ruffles. As full-length endogenous BRCA1 also associated with F-actin and ERM, Coene et al. speculated that the recombinant truncated protein might compete with native BRCA1. Indeed, expressing high levels of EGFP-CTD in CHO cells (CHO-High) displaced endogenous BRCA1 from along the plasma membrane where it normally colocalized with F-actin in wild-type CHO cells (CHO-WT).

How, then, might the absence of the N-terminal domain and central region of BRCA1 affect cell behavior? More CHO-WT cells than CHO-High cells were spread out 3 hours after plating, although there were no differences after 24 hours. CHO-High cells moved faster than CHO-WT cells and CHO cells expressing low levels of EGFP-CTD (CHO-Low), indicating that endogenous full-length BRCA1 might regulate cell migration. Consistent with this notion, MCF-7 cells express full-length functional BRCA1, show low motility and are not metastatic; these cells moved slower than HCC1937 cells, which contain a null BRCA1 allele and a truncating mutation in BRCA1 and are highly motile and invasive. Expressing EGFP-CTD in MCF-7 cells increased motility, whereas restoring expression of full-length BRCA1 in HCC1937 cells reduced motility. However, a point mutation in BRCA1 that abolishes its E3 ubiquitin ligase activity abrogated its effect in HCC1937 cells. Finally, individual EGFP-CTD-expressing CHO cells displayed haphazard and fast movement in wound healing assays compared with CHO-WT cells.

These results suggest that full-length, functional BRCA1 inhibits metastasis and invasion. Mutations that disrupt its localization at the plasma membrane (with ERM or, potentially, other membrane proteins) or that abrogate E3 ubiquitin ligase activity might confer an invasive phenotype by failing to control (through ubiquitylation) the levels of proteins that are present at cell-matrix and/or cell-cell contacts and that promote migration.

source: nature cell biology