Modelling of novel bornoel analogs as Influenza A Virus inhibitors through genetic function approximation, comparative molecular fields, molecular docking, and ADMET/Pharmacokinetic studies

Abstract

Influenza A Virus (IAV) is a human respiratory pathogen prone to mutations and genome re-assortment leading to global pandemics. In this study, we applied the molecular modelling strategies such as, two-dimensional (2D), three-dimensional (3D)-quantitative structure–activity relationship (QSAR), and molecular docking simulation on a novel series of borneol compounds as influenza inhibitors. The best developed 2D-QSAR models, MLR (Q² ​= ​0.8735, R² _(train) ​= ​0.9096) and ANN [3-2-1] (Q² ​= ​0.8987, R²_(train) ​= ​0.9171) revealed good and acceptable statistical validation metrics for the inhibitory activity predictions. The 3D-QSAR models were generated using the comparative molecular field analysis (CoMFA) and comparative molecular similarity indices analysis (CoMSIA), which showed CoMFA_S ​+ ​E (Q² ​= ​0.559, R²_(train) ​= ​0.939) and CoMSIA_S ​+ ​E (Q² ​= ​0.577, R²_(train) ​= ​0.941) as the best-observed models in accordance with the model acceptability standards. In addition, the contour maps generated from the CoMFA and CoMSIA models illustrates the steric and electrostatic molecular field relationships with the inhibitory effects of the studied molecules. Moreover, the binding modes of the active ligands were studied through molecular docking simulation with the Human Hemagglutinin (HA) receptor of influenza A virus (A/Puerto Rico/8/34(H1N1)). The studied compounds revealed the formation of H-bonds, CH-bonds, and hydrophobic interactions with the active amino acid residues such as Asn 543, Asn 614, Asn 617, Leu 618, Ser 540, Lys 539, and Lys 621 in the HA binding cavity. The prediction of drug-likeness and ADMET properties of the compounds revealed their good bioavailability and pharmacokinetic profiling. This study may provide a valuable in-silico guideline for discovering novel potent influenza inhibitors.