Discovery of Plant-Derived Natural Compounds as Novel GABA Aminotransferase Inhibitors: Structure-Based Discovery, Experimental Validation, and Molecular Dynamics Analysis

<b>Background/Objectives:</b> γ-Aminobutyric acid aminotransferase (GABA-AT) is a key enzyme responsible for GABA catabolism and represents a validated therapeutic target for epilepsy. Although existing GABA-AT inhibitors such as vigabatrin are clinically effective, their long-term use is limited by safety concerns, highlighting the need for alternative inhibitors with improved profiles. In this study, we employed an integrated natural product-oriented discovery strategy to identify novel GABA-AT inhibitors from plant-derived compounds. <b>Methods</b>: A library of 1006 plant-derived compounds collected from seven medicinal plants traditionally associated with sedative or anxiolytic effects was subjected to primary virtual screening using GNINA. Top-ranked candidates were further refined through secondary precision docking using aglycone forms to account for biologically relevant metabolic conversion. Detailed interaction analyses and molecular dynamics simulations were performed to assess binding stability and energetic favorability. <b>Results:</b> Based on computational prioritization, quercetin, salvianolic acid A, and scutellarein were selected for experimental validation. Cell-based GABA-AT activity assays in HepG2 cells demonstrated that quercetin and salvianolic acid A significantly inhibited intracellular GABA-AT activity, exhibiting comparable or greater efficacy than vigabatrin, while scutellarein showed moderate inhibition. The observed cellular inhibitory effects were consistent with predicted binding modes and dynamic stability observed in in silico analyses. <b>Conclusions:</b> Collectively, this study highlights the utility of an aglycone-focused, structure-based screening strategy for natural product drug discovery and identifies plant-derived aglycones as promising GABA-AT inhibitor candidates for further pharmacological development.