Development of potential CDK9 inhibitors through pharmacophore-based virtual screening, 3D-QSAR, molecular docking, MD simulation, and in vitro anticancer evaluation

Abstract Cyclin-dependent kinase 9 (CDK9) is a transcription-regulating serine/threonine kinase, and the dysregulation drives tumour initiation, thereby establishing CDK9 inhibition as a mechanistically validated and therapeutically attractive strategy for the treatment of diverse malignancies. In this study, a comprehensive computational strategy was utilized to identify novel CDK9 inhibitors. A pharmacophore-based virtual screening was implemented in combination with atom-based 3D-QSAR, molecular docking, binding free energies, in silico ADME, and MD simulation studies. A statistically validated five-point pharmacophore model (ADHRR) was developed and demonstrated strong predictive performance (R 2 = 0.98, Q 2 = 0.84). This optimized model was used to screen chemical databases for potential CDK9 inhibitors. Structural insights gained from the resulting hits guided the rational design of indole-based biphenyl amide hybrids (IBA’s). Seven analogues ( D1 - D7 ) exhibited strong binding affinities comparable to or greater than those of the screened hits and the reference CDK9 inhibitor 23 . Additionally, molecular dynamics simulations and DFT analysis confirmed the stability of both D3 and D6 complexes. Subsequent synthesis and biological evaluation against a panel of cancer cell lines identified compounds D3 and D6 as the most potent. Collectively, these results identify D3 and D6 as promising lead candidates for further CDK9-focused medicinal chemistry optimization and mechanistic studies.