A structure-based approach to explore novel COX-2 inhibitors using pharmacophore modelling, 3D-QSAR analysis, virtual screening and dynamics simulation study

Abstract

Anti-inflammatory medications are one of the most commonly preferred over-the-counter medications because inflammation and pain are closely related and caused by various disease conditions. Extensive research is mandatory to develop an effective anti-inflammatory therapy and good safety profile since both cyclooxygenase (COX-1 and COX-2) inhibition causes severe gastrointestinal irritation and cardiac toxicities. Only 190 molecules published in 2022 had selected for the investigation out of the many chemical architectures reported in the literature as COX-2 inhibitors. Using a four-point pharmacophore hypothesis AAAR2 (A-hydrogen bond acceptor and R- aromatic ring sites) generated from 138 molecules out of 190 with a site score of 0.84, a 3D atom-based QSAR model was built. With an outstanding correlation value of R2 0.92, fisher ratio of F 235.4, strong predictability of Q2 0.73, and Pearson-r 0.87, the 3D-QSAR model displays the highest and lowest efficient ligands reveals the favourable and unfavourable structural domains for the interaction toward COX-2. The selected 3D QSAR model is then utilised to screen compounds from bindingDB like a search engine. The molecules which passed the rule of five were chosen for further analysis after the screened molecules underwent an ADME analysis. Virtual screening (PDB ID-5KIR) was then used to identify 43 hit molecules, with the best docking scores falling between -11.07 and -9.55 kcal/mol, and these results were compared to the standard rofecoxib (docking score-9.479 kcal/mol). To evaluate the stable binding and interaction of the best molecule (Hit1), which is crucial for inhibiting COX-2, the molecular dynamics study was carried out for 100 ns.