The study of dual COX-2/5-LOX inhibitors by using electronic-topological approach based on data on the ligand–receptor interactions

•Conformational and electronic structure of COX-2/5-LOX inhibitors were investigated.•Pharmacophores and anti-pharmacophores analyzed using ETM combined with Neural Networks.•A prediction system developed recognizing correctly 93% of compounds as active ones.•Peculiarities of the ligand–receptor binding obtained with docking and DFT.•New active compounds were modeled based on the developed prediction system.

Structural and electronic factors influencing selective inhibition of cyclooxygenase-2 and 5-lipoxygenase (COX-2/5-LOX) were studied by using Electronic-Topological Method combined with Neural Networks (ETM–NN), molecular docking, and Density Functional Theory (DFT) in a large set of molecules. The results of the ETM–NN calculations allowed for the selection of pharmacophoric molecular fragments, which could be taken as a basis for a system capable of predicting the COX-2/5-LOX inhibitory activity. For the more effective extraction of the pharmacophoric molecular fragments, docking of molecules into the active sites of the two enzymes was carried out to get data on the ligand–receptor interaction. To make an assessment of these interactions, stabilization energies were calculated by using Natural Bond Orbital (NBO) analysis. Docking and data on the electronic structures of active sites of enzymes helped to reveal effectively the peculiarities of the ligand–receptor binding. The system for the selective COX-2/5-LOX inhibitory activity prediction that has been developed as the result of the ETM–NN study recognized correctly 93% of compounds as highly active ones. Thus, this system can be successfully used for carrying out computer screening and synthesis of potent inhibitors of COX-2/5-LOX with diverse molecular skeletons.

Graphical abstractFigure optionsDownload full-size imageDownload high-quality image (212 K)Download as PowerPoint slide