Mycobacterial enoyl acyl carrier protein (ACP) reductase is an attractive target for focused design of novel antitubercular agents. Structural information available on enoyl-ACP reductase in complex with different ligands was used to generate receptor-based pharmacophore model in Discovery Studio (DS). In parallel, pharmacophore models were also generated using ligand-based approach (HypoGen module in DS). Statistically significant models were generated (r2 = 0.85) which were found to be predictive as indicated from internal and external cross-validations. The model was used as a query tool to search Zinc and Maybridge databases to identify lead compounds and predict their activity in silico. Database searching retrieved many potential lead compounds having better estimated IC50 values than the training set compounds. These compounds were then evaluated for their drug-likeliness and pharmacokinetic properties using DS. Few selected compounds were then docked into the crystal structure of enoyl-ACP reductase using Dock 6.5. Most compounds were found to have high score values, which was found to be consistent with the results from pharmacophore mapping. Additionally, molecular docking provided useful insights into the nature of binding of the identified hit molecules. In summary, we show a useful strategy employing ligand- and structure-based approaches (pharmacophore modeling coupled with molecular docking) to identify new enoyl- ACP reductase inhibitors for antimycobacterial chemotherapy.
Keywords: Mycobacterium tuberculosis, enoyl acyl carrier protein reductase, pharmacophore modeling, molecular docking, binding interactions.