Uralenol, Glycyrol, and Abyssinone II as potent inhibitors of fibroblast growth factor receptor 2 from anti-cancer plants: A deep learning and molecular dynamics approach

Abstract

Fibroblast Growth Factor Receptor 2 (FGFR2) plays a critical role in cellular proliferation and differentiation, and its dysregulation is associated with multiple cancers. This study integrates molecular docking, deep learning, pharmacokinetic profiling, and molecular dynamics (MD) simulations to identify potential FGFR2 inhibitors from a library of 1,350 phytochemicals derived from 51 anti-cancer medicinal plants that were traditionally used for anticancer purposes. Initial screening through AutoDock Vina revealed several top candidates with high binding affinities to FGFR2. The top three compounds, uralenol, glycyrol, and abyssinone II, underwent further evaluation via deep learning models, which predicted the potential efficacy of the pIC50 (negative logarithm of the half-maximal inhibitory concentration) values. The ADME/T (absorption, distribution, metabolism, excretion, and toxicity) analysis confirmed favorable pharmacokinetic profiles and low toxicity risks. MD simulations validated the stability and compactness of protein–ligand complexes, with principal component analysis (PCA) and free energy landscape analyses confirming these interactions’ conformational stability and thermodynamic favorability. These findings suggest that uralenol, glycyrol, and abyssinone II are potential FGFR2 inhibitors and need further experimental validation for potential therapeutic use in cancer treatment.