An Integrated Computational Workflow for Discovering Alkaloid-Derived Ligands of Cyclin-Dependent Kinase 2

Abstract

Background/Objectives: Cyclin-dependent kinase 2 (CDK2) is a key regulator of cell-cycle progression and a potential anticancer target. This study aimed to identify alkaloid-derived CDK2 ligands using an integrated computational workflow and to obtain preliminary evidence of their effects on cancer-cell viability. Methods: Molecular docking with mVina and fast pulling of ligand (FPL) simulations were benchmarked using 20 experimentally characterized CDK2 inhibitors. A library of 2692 PubChem-derived alkaloids was screened, followed by ADMET evaluation, 100 ns molecular dynamics simulations, and FPL-based relative-affinity re-ranking. The three prioritized compounds were evaluated in HepG2 and HGC-27 cells using an MTT assay after 48 h of exposure. Results: Docking and FPL showed correlations with experimental affinity data of RDock = 0.549 ± 0.180 and RW = −0.676 ± 0.119, respectively. CID 636885, CID 46184320, and CID 101691758 were prioritized for detailed evaluation. All three compounds reduced cell viability, with lower IC50 values observed in HepG2 cells than in HGC-27 cells. CID 101691758 exhibited the highest growth-inhibitory activity among the tested compounds, with IC50 values of 15.37 ± 0.46 µg mL−1 in HepG2 cells and 52.64 ± 1.33 µg mL−1 in HGC-27 cells. Conclusions: The workflow identified three preliminary alkaloid hits, with CID 101691758 showing the most favorable combined computational and cell-viability profile. However, the MTT assay does not establish direct CDK2 inhibition or kinase selectivity. Biochemical CDK2 inhibition, target-engagement, and kinase-panel studies are therefore required.