Prediction of allosteric key residues and their role in protein folding

Abstract

Allostery, an aspect of protein dynamics, is crucial in regulation of protein activity. It is believed that allostery is maintained through communication pathways of key residues residing within distant allosteric sites. Prediction of these key residues therefore, would be a milestone in understanding protein allostery. In this study, prediction of functional residues is carried out by a newly proposed Monte Carlo (MC) path generation method, where the protein structure is considered as a network of amino acid residues and inter-residue interactions are described by a potential function. Study of the effect of the type of the potential function used, is carried out with four different potential functions, among which atomistic potential function is found to be the best to describe the interactions. Three different approaches of MC path generation are studied; 1) generating paths between two residues (BTR), 2) generating paths with specific number of steps (PSNS) and analyzing residue frequencies, and 3) generating infinite step paths (ISP) and calculating network parameters such as closeness, betweenness and clustering coefficient. In studying Shaker potassium channel and HIV-1 protease systems using these approaches, paths are generated in ensembles rather than obtaining a single shortest path. MC path generation is also applied to study the communication within and between different monomers of a structure and different structures of a protein. Combined information from these approaches reveals a list of functionally important residues, such as catalytic, binding and allosterically important sites. The role of these proposed residues in protein folding is studied through trajectories of protein folding simulations, algorithm of which is based on robotic motion planning. Through the folding trajectories, residue contacts are analyzed and residues that form initial contacts and conduct folding are identified. Interestingly these residues are noticed to be among those that display high closeness and betweenness values in pathway analysis carried out for the native state of the proteins. Overall, this study suggests that communication pathways are evolutionarily conserved and MC path generation is an effective method for prediction of residues that are important in both allostery and protein folding..