Computer-aided design of selective COX2 inhibitors: molecular docking of structurally diverse cyclooxygenase-2 inhibitors using FlexX

Abstract

Motivation: Three-dimensional structures of pharmacologically important macromolecules offer a route to the discovery of new drugs. Understanding the macromolecule-ligand interactions and validation of method used for docking and virtual screening of chemical databases is crucial step in structure-based design. We therefore carried out molecular docking for a set of eighty two structurally diverse COX-1/COX-2 inhibitors including traditional NSAIDs and the recent developed coxibs using FlexX method to find out how good this method differentiate between the active and inactive compounds. Method: FlexX is one of the fast flexible docking method that uses an incremental construction algorithm to place ligands into an active site. The scoring function (empirical binding free energy) of the flexX used to estimate the free binding energy of the protein-ligand complex is called F_score. Results: Reproducibility of the experimental conformations of the bound ligands such as SC-558, indomethacin, flurbiprofen indicates the better performance of FlexX method. Good correlation between the standard FlexX score (F_score) and the COX-2 inhibitory activity (pIC₅₀) was observed. Simple linear regression analysis provided the correlation coefficient values of 0.731 and 0.670 for two classes of COX-2 inhibitors. Conclusions: Flexible docking of eighty two structurally diverse COX-2 inhibitors have been successfully carried out. Some false positives and false negatives were observed but considering the limitations of the available docking programs, the results are encouraging. The detailed analysis of the resulted COX-2-ligand complexes may improve our knowledge in understanding the binding interactions in detail. Thus, this study will be useful for the design of novel COX-2 inhibitors based on docking and the resulted bioactive conformations of the ligands will be useful in building structure-based 3-D QSAR model.