Global public health is confronted with significant challenges due to the prevalence of cancer and the emergence of treatment resistance. This work focuses on the identification of cyclin-dependent kinase 2 (CDK2) through a systematic computational approach to discover novel cancer therapeutics. A ligand-based pharmacophore model was initially developed using a training set of seven potent CDK2 inhibitors. The obtained most robust model was characterized by three features: one donor (|Don|) and two acceptors (|Acc|). Screening this model against the ZINC database resulted in identifying 108 hits, which underwent further molecular docking studies. The docking results indicated binding affinity, with energy values ranging from -6.59 kcal mol⁻¹ to -7.40 kcal mol⁻¹ compared to the standard Roscovitine. The top 10 compounds (Z1-Z10) selected from the docking data were further screened for ADMET profiling, ensuring their compliance with pharmacokinetic and toxicological criteria. The top 3 compounds (Z1-Z3) chosen from the docking were subjected to Density Functional Theory (DFT) studies. They revealed significant variations in electronic properties, providing insights into the reactivity, stability, and polarity of these compounds. Molecular dynamics simulations confirmed the stability of the ligand-protein complexes, with acceptable RMSD and RMSF values. Specifically, compound Z1 demonstrated stability, around 2.4 Å, and maintained throughout the 100 ns simulation period with minimal conformational changes, stable RMSD, and consistent protein-ligand interactions.
Keywords: Cancer; Cyclin-dependent kinase; Molecular docking; Molecular dynamics; Pharmacophore modelling.
Copyright © 2024 Elsevier Ltd. All rights reserved.