Target-bound generated pharmacophore model to improve the pharmacophore-based virtual screening: Identification of G-protein human CCR2 receptors inhibitors as anti-inflammatory drugs

Abstract

Pharmacophore-based virtual screening is being widely used to discover new drug candidates. Building a pharmacophore model based on a known inhibitor that is unbound to the target could be misleading and result in mining for the wrong hits. Results presented herein confirm that pharmacophore models based on unbound and bound ligand confirmations produce significantly, structurally different compound libraries and, consequently, change the outcome of the virtual screening. To further verify our findings, molecular dynamics and extensive post-dynamic analysis are performed for the best retrieved hits from each approach; these are the unbound and bound ligand pharmacophore-generated libraries. In this report, the proposed target-bound pharmacophore model is used to discover potential G-protein coupled CCR2 receptor inhibitors as potential anti-inflammatory drugs. Herein, various molecular modeling approaches are adopted including homology modeling, molecular docking, lipid bilayer molecular dynamics simulations and per-residue interaction energy decomposition analysis. The current study highlights some critical aspects in the pharmacophore-based virtual screening as a powerful tool in the drug discovery and development machinery. © 2013 Biomedical Engineering Society.