Abstract
This study employed a comprehensive in silico pipeline to screen the Pan-African Natural Products Library (p-ANAPL) for potential COX-2 inhibitors with optimised pharmacokinetic and safety profiles. A rigorous ADMET filtering workflow was first applied to prioritize compounds with favourable drug-likeness and low toxicity risks. Subsequent structure-based virtual screening identified promising candidates based on their binding affinity and interaction patterns within the active site of murine COX-2 (PDB ID: 4PH9). The stability of the top-ranked complexes was confirmed through molecular dynamic simulations, which revealed low RMSD fluctuations and persistent interactions with key catalytic residues (Arg-120, Tyr-355). Binding free energy estimations via MM-GBSA and Density Functional Theory (DFT) calculations further corroborated the strong binding affinity and favourable electronic properties of these leads. The results indicate that two compounds of a library
152CSB2
and
Thix6g
exhibit significant potential as selective COX-2 inhibitors, with 152CSB2 showing as the most favourable, combining stable binding modes, favourable ADMET characteristics, and distinct electronic profiles that may underlie their enhanced activity. These compounds represent promising candidates for further experimental validation and development as novel anti-inflammatory agents.