Ligand-based pharmacophore modelling, in silico virtual screening, molecular docking and molecular dynamic simulation study to identify novel Francisella tularensis ParE inhibitors

Abstract

ParE, a subunit of topoisomerase IV, is involved in the management of DNA topology and validated to be an attractive target for the development of antibacterial agents. Availability of the crystal structure of Francisella tularensis in combination with one of the pyrrolo [2,3-d] pyrimidine-2-thiolinhibitor facilitated us to emphasize the combined computational approach for discovering the presumed binding mode of selected inhibitors into the binding pocket of ParE (pdb. 4HY1). In the current study, pharmacophore modelling and 3D-QSAR studies were performed using 33 reported F. tularensis ParE inhibitors having pKi ranging from 5.06 to 9.00. The developed five featured pharmacophore model, DHHRR_1 was statistically validated with different parameters like Q2 (0.66), R2 (0.99) and F value (682) at four-component partial least squares factor. Enrichment analysis was performed to validate the generated pharmacophore model. Extra-precision molecular docking, binding free energy calculation using PRIME MM-GBSA were performed for the selected inhibitors. Induced fit docking was performed for the highest active inhibitor 16. The highest-ranked induced fit docked complex 16/4HY1 was used to run a 50 ns molecular dynamic simulation to validate the stability. Further, in silico High Throughput Virtual Screening was performed using 22 lakhs chemical database molecules to identify the potential virtual hits and predicted activity was found for the virtual hits. These results provide promising strategies for the development of novel molecules with better inhibitory activity against F. tularensis ParE.