Molecular Docking of Natural Product-Derived Compounds: Estimation of Selectivity on Cyclo-oxygenase-2 นิพนธ์ต้นฉบับ Original

Objective: To investigate the binding modes and molecular selectivity onCOX-1 and COX-2 enzymes of 12 natural product-derived compoundsreported as anti-inflammatory agents using in silico prediction. Method:AutoDock 4.2 was employed to determine the free energy of binding (ΔGb)and inhibition constants (Ki). The evaluated inhibition constant (Ki) fromdocking result was used to estimate the calculated selectivity index (ratio ofCOX-2 Ki to COX-1 Ki of each compound). Results: γ-Mangostin gainedthe lowest calculated selectivity index (0.0269) comparable to rofecoxib(0.0407). The calculated selectivity indices of gingerol, [8]-paradol,isorhapontigenin and rutaecarpine were in a range of 0.2-0.5 that could bedefined as preferential COX-2 inhibitors. Conclusion: Binding modes andmolecular selectivity of 12 natural product-derived compounds reported asanti-inflammatory agents were determined. This information from theinhibitor-enzyme interactions and calculated selectivity indices could beuseful in designing NSAIDs with new scaffold and favorable safety profile.Keywords: docking, COX-2 selective inhibitors, natural product-derivedcompounds, Autodock