Amino Acid Regulation and Predictive Modeling of Phip and Pyrazines in Maillard Reaction Systems

Abstract

This study explored the regulatory roles of two amino acid types—basic (L-histidine) and neutral (L-proline)—in a model Maillard reaction system, focusing on their dual functions in mitigating the formation of the hazardous compound 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) while enhancing desirable pyrazine-derived flavor compounds. L-Histidine exhibited greater efficacy than L-proline in modulating both targets. At an addition level of 40 mg/L, L-histidine reduced the PhIP concentration by 18.75% to 58.69% and simultaneously increased the pyrazine content by 66.85% to 88.97%. A robust partial least squares (PLS) prediction model was established, which demonstrated strong predictive performance for both PhIP (R2c and R2p > 0.73) and pyrazine levels (R2c and R2p > 0.77). These findings provide a practical foundation for improving thermal food processing strategies, enabling the reduction of harmful byproducts while preserving or enhancing flavor quality—an approach that supports safer, more sustainable food production.

Keywords: L-Histidine, L-Proline, Prediction model, Basic amino acid