The conformation of the nSrc specificity-determining loops in the Src and Abl SH3 domains are modulated by a WX conserved sequence motif

Abstract

Cellular signaling networks are modulated by multiple protein-protein interaction domains that coordinate extracellular inputs and processes to regulate cellular processes. Several of these domains recognize short linear motifs, or SLiMs, which are often highly conserved and are closely regulated. One such domain, the Src homology 3 (SH3) domain, typically recognizes proline-rich SLiMs and is one of the most abundant SLiM-binding domains in the human proteome. These domains are often described as quite versatile, and indeed, SH3 domains can bind ligands in opposite orientations dependent on target sequence. Furthermore, recent work has identified diverse modes of binding for SH3 domains and a wide variety of sequence motifs that are recognized by various domains. Specificity is often attributed to the RT and nSrc loops near the peptide-binding cleft in this domain family, particularly for Class I binding, which is defined as RT and nSrc loop interactions with the N-terminus of the ligand. Here, we used the Src and Abl SH3 domains as a model to further investigate the role of the RT and nSrc loops in SH3 specificity. We created chimeric domains with the loop sequences swapped between these SH3 domains, and used fluorescence anisotropy assays to test how relative binding affinities were affected for Src SH3- and Abl SH3-specific ligands. We also used Alphafold - Multimer to model our SH3:peptide complexes. We identified a position that contributes to the nSrc loop conformation, the amino acid immediately C-terminal to a highly conserved Trp that creates a hydrophobic pocket critical for SH3 ligand recognition. We defined this as the WX motif, where X = Trp for Src and Cys for Abl. The importance of this position for orienting the ligand is supported by analyses of previously deposited SH3 structures, multiple sequence alignment of SH3 domains in the human proteome, and our biochemical data of mutant Src and Abl SH3 domains. Overall, our work uses experimental approaches and structural modeling to better understand SH3 specificity determinants.