Computer-aided Discovery of Novel COX-2 Inhibitors for Anti-inflammatory Therapy Using Pharmacophore Modelling, Molecular Docking, ADMET, and Virtual Screening

Cyclooxygenase-2 (COX-2) is a central enzyme in inflammation and a validated target for anti-inflammatory drug development. However, many existing COX-2 inhibitors cause gastrointestinal and cardiovascular side effects, necessitating safer alternatives. This study employed a computer-aided drug discovery approach integrating pharmacophore modeling, molecular docking, and Absorption, Distribution, Metabolism, Excretion, and Toxicity (ADMET) analysis to identify natural COX-2 inhibitors from 72 sesquiterpene lactones. A validated pharmacophore model (AUC = 0.77) screened 33 potential inhibitors, and molecular docking revealed 13 compounds with strong binding affinities. Among them, Tanshinone IA showed the highest binding affinity (-10.6 kcal/mol), surpassing celecoxib (-9.5 kcal/mol). ADMET profiling indicated favorable pharmacokinetic properties, high gastrointestinal absorption, and low toxicity for most compounds. Overall, Tanshinone IA emerged as a promising, naturally derived COX-2 inhibitor with potential for safer anti-inflammatory therapy. These findings provide a computational foundation for future experimental validation and preclinical development of natural anti-inflammatory agents.