Predictive analysis of gut microbial function
The functional pathway abundance of KEGG corresponding to the 16S rRNA sequencing data was predicted using picrust2. The results show that the KEGG metabolic pathways at the first level mainly consisted of the following functional categories: metabolism (78.6%), genetic information processing (8.0%), environmental information processing (6.5%), cellular processes (3.0%), human diseases (2.5%), and organic systems (1.4%).
In the differential analysis of KEGG metabolic pathways at the second level, the carbohydrate metabolism, replication and repair, and drug resistance pathways were significantly enhanced in the spring group (P <0.05). Amino acid metabolism, metabolism of terpenoids and polyketides, xenobiotic biodegradation and metabolism, cell motility, signal transduction, and lipid metabolism were significantly enhanced in the autumn group (P <0.05; Figure 6A).
In the enrichment process of the third-level metabolic pathways, glycolysis/gluconeogenesis, amino sugar and nucleotide sugar metabolism, and metabolic pathways such as fructose and mannose metabolism, galactose metabolism, and purine metabolism were significantly higher in the spring group (P <0.05). The two-component system and glycine, serine, and threonine metabolism were significantly enhanced in the autumn group (P <0.05; Figure 6B).
Fecal metabolic profiling ofT. roborowskii
The fecal metabolic profiles of T. roborowskii were acquired by LC-MS, and metabolism was detected with good intra-group reproducibility. Fecal metabolites were significantly separated in the first principal component (PC1: 82.9%) (Figure 7A). In the OPLS-DA model, the metabolic curves were clearly different between the two sample groups (Q2=0.99>0.9 indicates an excellent model), and the results show that there are obvious differences in fecal metabolites between the spring and autumn groups (Figure 7B).
The absolute values of log2FC (Fold Change, FC) were sorted to get the top 10 metabolites in each group. In the spring group, the levels of hexaethylene glycol, leucyl-tyrosine, scillipheosidin 3-[glucosy-(1->2)-rhamnoside], 1-O-caffeoyl-(b-D-glucose 6-O-sulfate), serinyl-gamma-glutamate, bradykinin hydroxyproline, phaseolic acid, TR-saponin B, corepoxylone, and ethylparaben were high. Methyl dihydrojasmonate, dodecanoic acid, trans-2-hexyl-1-cyclopropaneacetic acid, cochliophilin A, “PC(20:5(5Z,8Z,11Z,14Z,17Z)/24:0),” uridine diphosphate glucose, grevillol, benzyl hexanoate, trans-cinnamic acid, etc. were higher in the autumn group (Figure 8).
The differential metabolites were classified as carboxylic acids and derivatives, fatty acids, organooxygen compounds, steroids and steroid derivatives, glycerol glycerophospholipids, prenol lipids, and glycerolipids.
Differential metabolites were screened based on the criteria ofP =0.05, VIP=1, and FC=1. Finally, 1264 differential metabolites were detected in the two groups of samples, of which 668 were upregulated (higher in the autumn group), and 596 were downregulated (higher in the spring group). The relative contents of deoxycholic acid glycine conjugate and 26-methyl nigranoate were relatively higher in the spring group. Simulansine, 3-oxocholic acid, and asparaginyl-proline were the three metabolites with the highest relative contents in the autumn group (Figure 9).
KEGG functional classification of metabolites showed that metabolites participate mainly in metabolism, organismal systems, human diseases, environmental information processing, and other functional pathways. Among these, metabolic pathways, biosynthesis of secondary metabolites, and microbial metabolism in different environments are related to a variety of metabolites.
The KEGG enrichment network map of differential metabolites showed that lysine degradation, pantothenate and CoA biosynthesis, steroid hormone biosynthesis, carbon fixation pathways in prokaryotes, and other differential metabolites between different seasons had a close positive or negative relationship (Figure 10).