Identification of inhibitors against p90 ribosomal S6 kinase 2 (RSK2) through structure-based virtual screening with the inhibitor-constrained refined homology model.

P90 ribosomal S6 kinase 2 (RSK2), which was shown to be overexpressed in human cancers, is a serine/threonine kinase and a potential target for cancer treatment. RSK2 comprises two terminal kinase domains (NTKD and CTKD) that can be inhibited by binding with different types of inhibitors at the ATP binding sites. In the absence of a crystal structure of RSK2, we constructed a model for the 3D structure of the RSK2 NTKD by homology modeling and stepwise constrained refinement with the reported inhibitors using a molecular docking method. Structure-based virtual screening was subsequently performed against a library containing commercially available compounds using the refined model. This resulted in the identification of seven novel RSK2 inhibitors with IC₅₀ values ranging from 2.4 to 14.45 μM.