Identification of Phosphatidylinositol 3-kinase δ (PI3Kδ) Inhibitor: Pharmacophore-based Virtual Screening and Molecular Dynamics Simulation

Phosphatidylinositol 3-kinase δ (PI3Kδ) is a validated drug target for the treatment of cancer. The present study aims to search for new inhibitors of PI3Kδ by employing pharmacophore modelling using LigandScout Advanced 4.3 software. The three hydrogen bond acceptors and two hydrophobic features were proposed as a pharmacophore model using LASW1976 structure. The model was then validated using the Area Under Curve (AUC) of Receiver Operating Characteristic (ROC) and GH score. It was used to screen new molecules in the ZINC database, which resulted in 599 hits. All 599 hits were then docked into PI3Kδ protein, and five best hits were submitted to 50 ns molecular dynamics simulations. Each hit complexed with PI3Kδ underwent minor conformational changes as indicated by the values of Root Mean Square Deviation (RMSD) and Root Mean Square Fluctuation (RMSF). Furthermore, prediction of the binding free energy using Molecular Mechanics-Poisson Boltzmann Surface Area (MM-PBSA) method showed that five hits, i.e., Lig25/ZINC253496376, Lig682/ZINC98047241, Lig449/ZINC85878047, Lig554/ZINC253389510, and Lig199/ZINC12638303, had lower binding energy compared to LASW1976. This result indicated their potentials as new inhibitors of PI3Kδ.