Structure-based pharmacophore clustering of multi-conformation proteins: Application to identify novel and diverse CypD inhibitors

Objectives: Cyclophilin D (CypD) is a chief regulatory protein of the necroptosis pathway involved in various neurological disorders, and ablation/inhibition of this protein confers neuroprotection. Current in silico drug design strategies employ multiple structures of a protein target since they enable the identification of diverse inhibitor molecules. However, structure-based drug design against a protein target becomes challenging if it contains numerous known structures with varying ligand interactions. Considering all these structures for virtual screeing of database compounds would be inappropriate in view of the computational resources that might be demanded. Therefore, identifying appropriate structures with varied binding site conformations is of utmost importance in order to identify inhibitors with diverse scaffolds. Method: In the present study, clustering of a large number of CypD structures was carried out by comparing the pharmacophores derived from their binding sites. A representative structure from each cluster was adopted to build an ensemble pharmacophore that was further employed in dual ensemble screening of database compounds. Results: Two compounds that exhibited better docking scores, compared to the already reported CypD inhibitors, formed stable complexes and desirable interactions with the protein during molecular dynamics (MD) simulation.