The design of TOPK inhibitors using similarity search, molecular docking, and MD simulations

Abstract

Cancer is still a major cause of death worldwide. Unfortunately, the majority of current anticancer treatments suffer many limitations, mainly emergence of resistance and lack of selectivity which necessitate the search for new therapeutics. The TOPK enzyme emerges as a promising target due to its overexpression in many cancer types while being rarely detected in normal tissues. Therefore, targeting TOPK would affect the malignant activity of cancerous cells while sparing normal ones. Further, its vital role in cell division, particularly in cytokinesis, adds to its safety to normal non-multiplying cells. In this study, a combined ligand and structure-based approach was utilized to identify potential TOPK inhibitors. Previously, we identified TOPK inhibitors using a structure-based approach following the construction of a 3D homology model of the TOPK enzyme. Herein, the most active identified inhibitor (lead) was used as a search query to conduct similarity search against PubChem and ChemBridge databases. Retrieved hits were filtered using drug-like filters, docked into the ATP binding site of the enzyme, and finally, the binding free energies of all docked poses were calculated. Based on the computational scores, eight hits were selected as potential TOPK inhibitors. The predicted ADMET descriptors of the eight selected hits were generally favorable. Further, MD simulations of the top scoring hit were conducted to investigate its binding dynamics compared to the lead compound and OTS964 which agreed with the docking results and propose the selected hits as potential TOPK inhibitors. Yet, biochemical testing is still needed to validate these results. Communicated by Ramaswamy H. Sarma