Molecular docking, molecular dynamics simulation, and structure-based 3D-QSAR studies on estrogenic activity of hydroxylated polychlorinated biphenyls

Abstract Hydroxylated polychlorinated biphenyls (HO-PCBs), major metabolites of PCBs, have been reported to present agonist or antagonist interactions with estrogen receptor α (ERα) and induce ER-mediated responses. In this work, a multistep framework combining molecular docking, molecular dynamics (MD) simulations, and structure-based three-dimensional quantitative structure–activity relationship (3D-QSAR) studies were performed to explore the influence of structural features on the estrogenic activities of HO-PCBs, and to investigate the molecular mechanism of ERα–ligand interactions. The CoMSIA (comparative molecular similarity indices analysis) model was developed from the conformations obtained from molecular docking. The model exhibited statistically significant results as the cross-validated correlation coefficient q 2 was 0.648, the non-cross-validated correlation coefficient r 2 was 0.968, and the external predictive correlation coefficient r pred 2 was 0.625. The key amino acid residues were identified by molecular docking, and the detailed binding modes of the compounds with different activities were determined by MD simulations. The binding free energies correlated well with the experimental activity. An energetic analysis, MM-GBSA energy decomposition, revealed that the van der Waals interaction was the major driving force for the binding of compounds to ERα. The hydrogen bond interactions between the ligands and residue His524 help to stabilize the conformation of ligands at the binding pocket. These results are expected to be beneficial to predict estrogenic activities of other HO-PCB congeners and helpful for understanding the binding mechanism of HO-PCBs and ERα.