Quantitative assessment of kinase selectivity based the water-ring network in protein binding sites using molecular dynamics simulations

Abstract Protein kinases are implicated in a variety of diseases, such as cancer and inflammation, and are thus an important target for the pharmaceutical industry. However, the design of selective protein kinase inhibitors can prove challenging owing to the presence of a highly conserved binding site for ATP in kinases. Here we describe a novel method for analyzing the water-ring network that predicts kinase selectivity based on a molecular dynamics approach. To quantify the water-ring network in the binding site, we established a method to calculate the dipole moments of the water-ring network at specific positions in the ATP binding pocket. For two kinase systems (ZAP-70/Chk1 and MAP3K5/PDPK1), we found that the orientation of dipole moments plays a critical role in the protein-ligand binding mechanism. This solvent-centric approach complements current theoretical methods that consider only the steric and electrostatic properties of protein surfaces.