PH domains in WASP – a bug in the system? Reply

As Insall and Machesky point out, the sequence identity between Wiskott–Aldrich syndrome protein (WASP; and N-WASP) and the consensus pleckstrin-homology (PH) domain sequence is low. Not all PH domains match the consensus well: the PH domain from phospholipase C delta, a well-characterized PH domain, is identical in 21 out of 94 amino acid residues, less than the identity with Sos, although still greater than that with WASP. We first proposed that the N-terminal region of WASP and N-WASP contains a PH domain based on its specific association with phosphatidylinositol (4,5)-bisphosphate but not other phospholipids, a feature characteristic of PH domains, and the observation that conserved acid residues can be aligned with those of typical PH domains1xMiki, H., Miura, K., and Takenawa, T. EMBO J. 1996; 15: 5326–5335PubMedSee all References1. However, it could be premature to call the region a ‘PH domain’ in the absence of structural data.The N-terminal region of WASP interacts with WIP2xRamesh, N. et al. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 14671–14676CrossRef | PubMed | Scopus (257)See all References2. Insall and Machesky emphasize the sequence similarity with Ena–VASP family proteins in the ‘WH1 domain’, but the deletion of residues 1–46 of WASP, which destroys half of the ‘PH domain’ but does not destroy the sequence of the ‘WH1 domain’, also results in strong reduction of the interaction with WIP2xRamesh, N. et al. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 14671–14676CrossRef | PubMed | Scopus (257)See all References2. In addition, the interaction with WIP requires not only intact ‘PH’ and ‘WH1’ domains but also more C-terminal sequences (residues 1–170 in WASP are reported to be sufficient). Thus, it is also questionable whether the ‘WH1 domain’ is really a functional domain.In conclusion, we now believe that a much broader region, including both the ‘PH domain’ and the ‘WH1 domain’, could form a single functional unit that should be appropriately called a ‘domain’. The recent discovery of the Src-homology 2 (SH2) domain structure in the N-terminal region in Cb1, which had not previously been thought to be an SH2 domain because of the lack of enough similarity to the consensus SH2 domain sequence, underlines the importance of the structural analysis3xMeng, W. et al. Nature. 1999; 398: 84–90CrossRef | PubMed | Scopus (220)See all References3. We need to accumulate more functional data and determine the three-dimensional structure to resolve this confusion.