In order to understand the differences and changes of metabolites in large yellow croaker roe sauce during rapid fermentation, samples were collected over a fermentation period of 30 days and analyzed for metabolite composition by gas chromatography-time-of-flight mass spectrometry (GC-TOF-MS). Multivariate statistical analysis was performed on the obtained data, and the differential metabolites were selected based on statistically significant difference at P ≤ 0.05 and the variable importance in the projection (VIP) of the first principal component (PC1) ≥ 1 in the orthogonal partial least squares-discriminant analysis and were analyzed by hierarchical agglomerative clustering. The results showed that a total of 46 differential metabolites were identified, including 10 sugars, 9 alcohols, 7 of amino acids, 6 organic acids, 4 esters and 3 aldehydes, 1 fatty acid, and 6 other compounds. These metabolites were involved in the metabolic pathways of valine, leucine and isoleucine biosynthesis, protein digestion and absorption, galactose metabolism, ammonia acyl-tRNA biosynthesis, glycine, serine and threonine metabolism, glucosinolate biosynthesis, citric acid cycle, amino acid biosynthesis, shikimic acid-based biosynthesis of alkaloids, inositol phosphate metabolism, and biosynthesis of unsaturated fatty acids. Fifteen significantly differential metabolites were identified included amino acid metabolites involved in multiple metabolic pathways. The results of this study provide a theoretical basis for the industrial production of large yellow croaker roe sauce.

Intake of marine phospholipids can ameliorate the gut microbiota dysbiosis, thereby improving the metabolic disturbance of the host. It is unclear whether consuming marine phospholipids affects the gut microbiota community structure in a normal-diet host. The aim of this research was to compare the gut microbiota community structure between normal-diet mice consuming and not consuming large yellow croaker roe phospholipids (LYCRPLs) by 16S rRNA amplicon sequencing technology. Results indicated that LYCRPLs could regulate the relative abundance of some certain bacteria of normal-diet mice, i.e., increasing the abundance of some beneficial genera including Lachnospiraceae_NK4A136_group and Lactobacillus, and decreasing harmful bacteria such as GCA-900066575 and Romboutsia. Simultaneously, intake of LYCRPLs was predicted to reduce the bacterial biosynthesis of carbohydrate and lipids. This research revealed the regulation of LYCRPLs on the gut microbiota community structure in normal-diet mice, and provides a valuable reference for LYCRPLs as a functional ingredient in the food industry.