Elucidation of molecular interactions between selected phenylpropanoids and myofibrillar protein: Conformational remodeling and machine learning analysis.

Phenylpropanoid flavor compounds (PFCs) are major contributors to the characteristic herbal and anise-like aromas of stewed beef with spices (SBS). This study investigated the binding interactions between five PFCs and myofibrillar proteins (MPs) under varying heat and concentration conditions. Heat treatment partially unfolded MPs, thereby exposing thiol groups and hydrophobic regions that enhanced PFC binding. In unheated MPs, the addition of PFCs increased α-helix content by promoting hydrogen bond formation driven by hydrophobic interactions; however, prolonged heating attenuated these effects due to protein aggregation that restricted access to binding sites. The distinct binding behaviors among the five PFCs were successfully characterized using machine learning-based clustering models. K-means clustering analysis revealed distinct binding behaviors of eugenol (EG) and anethole (AT) toward MPs, compared with methyleugenol (ME), estragole (ET), and anisaldehyde (AL). The two clusters were consistent with the binding energy of molecular docking. Disruption agent tests and molecular docking revealed hydrophobic interactions as the main driving force for PFCs-MP binding. Structural differences among PFCs, such as hydroxyl group positions, affected their hydrogen bonding and steric accessibility. Overall, thermally induced protein conformational changes play a crucial role in flavor retention, thereby informing flavor management during meat product processing.