Structure-based identification of CaMKIIα-interacting MUPP1 PDZ domains and rational design of peptide ligands to target such interaction in human fertilization.

Abstract

The recognition and association between Ca(2+)/calmodulin-activated protein kinase II-α (CaMKIIα) and multi-PDZ domain protein 1 (MUPP1) plays an important role in sperm acrosome reaction and human fertilization, which is mediated by the binding of CaMKIIα's C-terminal tail to one or more PDZ domains of the scaffolding protein MUPP1. In this study, we attempt to identify the CaMKIIα-interacting MUPP1 PDZ domains and to design peptide ligands that can potently target and then competitively disrupt such interaction. Here, a synthetic biology approach was proposed to systematically characterize the structural basis, energetic property, dynamic behavior and biological implication underlying the intermolecular interactions between the C-terminal peptide of CaMKIIα and all the 13 PDZ domains of MUPP1. These domains can be grouped into four clusters in terms of their sequence, structure and physiochemical profile; different clusters appear to recognize different classes of PDZ-binding motifs. The cluster 3 includes two members, i.e. MUPP1 PDZ 5 and 11 domains, which were suggested to bind class II motif Φ-X-Φ(-COOH) of the C-terminal peptide SGAPSV(-COOH) of CaMKIIα. Subsequently, the two domains were experimentally measured as the moderate- and high-affinity binders of the peptide by using fluorescence titration (dissociation constants K d=25.2±4.6 and 0.47±0.08µM for peptide binding to PDZ 5 and 11, respectively), which was in line with theoretical prediction (binding free energies ΔG total=-7.6 and -9.2kcal/mol for peptide binding to PDZ 5 and 11, respectively). A systematic mutation of SGAPSV(-COOH) residues suggested few favorable amino acids at different residue positions of the peptide, which were then combined to generate a number of potent peptide mutants for PDZ 11 domain. Consequently, two peptides (SIAPNV(-COOH) and SIVMNV(-COOH)) were identified to have considerably improved affinity with K d increase by ~tenfold relative to wild type peptide. Thus, the two peptides are considered as promising lead entities to develop therapeutic molecular agents with high efficacy and specificity to target CaMKIIα-MUPP1 interaction. Other five designed peptides (SILPSV(-COOH), SGLPNV(-COOH), SIVMSV(-COOH), SIVPNV(-COOH) and SIAMNV(-COOH)) possessed comparable affinity with the wild type, and they may be further optimized to obtain higher potency.