Abstract

The accuracy of protein-ligand molecular docking is strongly limited by the protein conformational changes taking place upon binding. Ensemble docking addresses this “flexibility issue” by using various conformations of the protein to implicitly account for structural deformations induced by ligand binding. While homology-modelling and/or molecular dynamics simulations are commonly used to generate receptor conformations for ensemble docking runs, the sampling of holo-like structures (prone to host ligands and often experimentally not available) remains challenging. This, in turns, affects the ability of molecular docking to reproduce the native structure of protein-ligand complexes. To address this limitation, we recently introduced EDES - Ensemble Docking with Enhanced-sampling of pocket Shape -, a computational approach based on metadynamics simulations that allows to successfully generate holo-like conformations of proteins by only exploiting their apo structure. In this work, we further developed the original protocol in order to handle multiple - allosteric - binding sites in extremely flexible proteins. We applied our method to a very challenging target, namely the enzyme adenylate kinase (AK), which undergoes very large conformational changes upon ligand binding. Our protocol generated a significant fraction of structures featuring a low RMSD from the experimental geometry of the complex between AK and an inhibitor, outperforming previous attempts to sample such conformations. Importantly, the accurate generation of holo-like structures was not limited to the backbone of the protein but involved the fine geometry of the two binding sites located at the interfaces between the three quasi-rigid domains building up this enzyme. The sampled conformations were used in ensemble docking calculations with three programs implementing different search algorithms and scoring functions. Native-like poses of substrates and inhibitors of adenylate kinase could be obtained among the top-ranked ones.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call