Flavor Chemistry and Fidelity in Processed Large Yellow Croaker Products: Multi‐Pathway Mechanisms, Evidence‐Graded Process Guidance, and Transferable Principles for Marine Fish

Résumé fourni par la source

Marine fish products are vulnerable to flavor drift during salting, drying, heating, storage, and reheating because lipid oxidation, ATP/protein degradation, Maillard/Strecker chemistry, and microbial metabolism occur concurrently. This review uses large yellow croaker (Larimichthys crocea; LYC), a commercially important n-3 polyunsaturated fatty acid (PUFA)-rich marine fish, as an LYC-centered case system rather than a universally representative model for marine fish. By integrating direct LYC evidence with comparative seafood literature, this review summarizes transferable pathway-level principles while emphasizing the need for species-specific calibration before process transfer to other fish matrices and product formats. Cross-pathway interactions among lipid oxidation, proteolysis/ATP degradation, Maillard/Strecker reactions, microbial metabolism, and flavor-matrix interactions are discussed together with reported operational parameter ranges and evidence-graded process guidance where direct LYC data are available. The review further maps volatile and nonvolatile flavor drivers to sensory outcomes and summarizes current characterization and modeling approaches, including metabolomics, molecular sensomics, molecular dynamics simulations, and gas chromatography-mass spectrometry (GC-MS)/GC-ion mobility spectrometry (IMS) coupled with chemometrics. Processing is highlighted as a double-edged sword: Appropriate pretreatment and thermal regimes can enhance umami and aroma formation, whereas poorly controlled conditions accelerate oxidation, protein carbonylation, bitterness precursor accumulation, and warmed-over flavor development. Emerging technologies and preservation strategies are critically evaluated, particularly where validation in LYC systems remains limited. Finally, this review proposes a staged roadmap toward predictive flavor control based on standardized datasets, kinetic modeling, online sensing, and pilot-scale digital twin frameworks.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.