Flavor-Enhancing Pentapeptide AGPNY from Tomato Proteins: Potential Dual-Targeting of Umami Receptors T1R1/T1R3 and GRM1 through Computational-Experimental Synergy

Abstract

This study establishes a multidimensional screening-verification framework for the precise identification of umami peptides from tomato (Solanum lycopersicum) proteins and the atomic-level elucidation of their dual-targeting mechanisms. By integrating bioinformatic virtual hydrolysis, ADMET-based safety profiling, and machine learning (Mlp4Umami) predictions, a 639-member peptide library was systematically prioritized. Molecular docking and 200 ns molecular dynamics (MD) simulations were employed to evaluate the binding landscapes of candidates with umami receptors T1R1/T1R3 and the metabotropic glutamate receptor 1 (GRM1). Experimental validation via electronic tongue and sensory evaluation identified a pentapeptide, AGPNY, as a potent flavor enhancer with a remarkably low threshold of 0.0625 mg/mL─approximately 4.8-fold more potent than monosodium glutamate (MSG). MM/GBSA binding free energy analysis and per-residue decomposition revealed that electrostatic interactions are the primary driving force for binding. Specifically, AGPNY is anchored within receptor pockets through an intricate hydrogen-bonding and salt-bridge network with key residues, notably Asp147 and Arg277 in T1R1. This integrated “computational prediction–molecular simulation–experimental validation” paradigm provides a high-precision methodological framework for the discovery of plant-derived flavor enhancers and offers significant implications for the rational design of functional food additives.