Characterizing Class I WW domains defines key specificity determinants and generates mutant domains with novel specificities

Abstract

Introduction: WW domains are small protein interaction modules found in a wide range of eukaryotic signaling and structural proteins. Five classes of WW domains have been annotated to date, where each class is largely defined by the type of peptide ligand selected, rather than by similarities within WW domains. Class I WW domains bind Pro–Pro–Xxx–Tyr containing ligands, and it would be of interest to determine residues within the domains that determine this specificity.Results: Fourteen WW domains selected Leu/Pro–Pro–Xxx–Tyr containing peptides ligands via phage display and were thus designated as Class 1 WW domains. These domains include those present in human YAP (hYAP) and WWP3, as well as those found in ubiquitin protein ligases of the Nedd4 family, including mouse Nedd4 (mNedd4), WWP1, WWP2 and Rsp5. Comparing the primary structures of these WW domains highlighted a set of highly conserved residues, in addition to those originally noted to occur within WW domains. Substitutions at two of these conserved positions completely inhibited ligand binding, whereas substitution at a non-conserved position did not. Moreover, mutant WW domains containing substitutions at conserved positions bound novel peptide ligands.Conclusions: Class I WW domains contain a highly conserved set of residues that are important in selecting Pro–Xxx–Tyr containing peptide ligands. The presence of these residues within an uncharacterized WW domain can be used to predict its ability to bind Pro–Xxx–Tyr containing peptide ligands.