Get into the groove

Many mutations in the BRCA1 tumour suppressor gene that lead to an increased susceptibility to breast and ovarian cancers are located in the tandem BRCA1 carboxy-terminal (BRCT) domains. These domains are known to function together as a phosphoserine/phosphothreonine-binding module, but how do they recognize phosphopeptides? And, might cancer-causing mutations in these domains specifically disrupt protein–protein interactions?

In Nature Structural and Molecular Biology, papers from Smerdon, Yaffe and colleagues and the Glover group now describe crystal structures of the BRCA1 BRCT domains bound to phosphopeptides that contain the pSer-X-X-Phe recognition motif (where pSer represents phosphoserine and X represents any amino acid). Both papers show that each BRCT domain forms a compact unit and that the phosphopeptides bind to a groove between these domains. The pSer residue of the motif binds to a basic pocket in the amino-terminal BRCT domain, whereas the Phe binds to a hydrophobic pocket between the domains. Smerdon, Yaffe and colleagues showed that a set of cancer-related BRCT mutations disrupt BRCA1–phosphopeptide interactions in vitro and BRCA1–phosphoprotein binding in vivo. The Glover group also showed that a large set of cancer-causing BRCT mutations disrupt phosphopeptide binding by using peptide-binding assays and by determining the crystal structures of two BRCT variants. These papers have therefore shown that a specific reduction in the phosphopeptide-binding ability of the BRCT domains, rather than a general disruption of the BRCT fold, might explain the increased cancer risks that are associated with BRCT mutations. REFERENCES Clapperton, J. A. et al. Structure and mechanism of BRCA1 BRCT domain recognition of phosphorylated BACH1 with implications for cancer. Nature Struct. Mol. Biol. 9 May 2004 (doi:10.1038/nsmb775) Williams, R. S. et al. Structural basis of phosphopeptide recognition by the BRCT domain of BRCA1. Nature Struct. Mol. Biol. 9 May 2004 (doi:10.1038/nsmb776)

No go for H2O

Aquaporin-0 (AQP0) — a lens-specific water pore — is the only aquaporin that is known to form membrane junctions. In Nature, Walz and colleagues now describe the electron-crystallographic structure of an AQP0 membrane junction, which has given us our first insights into water-pore closure and gating.

The structure shows that the extracellular surface of AQP0 is relatively flat, and that the junction is formed by three specific interactions between AQP0 molecules in adjacent membranes. These interactions are mainly mediated by proline residues that are conserved in AQP0 molecules, but are not present in most other aquaporins. Compared to AQP1, the constriction site in AQP0 is narrower and longer, and prevents water permeation. Furthermore, AQP0 has a further, new constriction site in the cytoplasmic half of the pore, which is again too narrow for water to traverse. A tyrosine residue in this new constriction site seems mobile, which might constitute a gating mechanism for the AQP0 pore. Indeed, the authors suggest that junction formation might induce a series of conformational changes that result in pore narrowing at both constriction sites — a hypothesis for pore closure that can now be tested. REFERENCE Gonen, T. et al. Aquaporin-0 membrane junctions reveal the structure of a closed water pore. Nature 429, 193–197 (2004)