Integrated Metagenomic and Metabolomic Analyses of the Effect of Astragalus Polysaccharides on Alleviating High-Fat Diet-Induced Metabolic Disorders

Abstract

Most herbal polysaccharides possess multiple benefits against metabolic disorders, such as non-alcoholic fatty liver disease (NAFLD) and obesity. However, the underlying mechanisms are largely unknown. Here, male C57BL/6J mice were fed with chow or high-fat diet (HFD) with or without Astragalus polysaccharides (APS) supplementation, and gut microbial profile and metabolite profile were studied by metagenomic sequencing and untargeted metabolomics, respectively. APS was effective in alleviating HFD-induced metabolic disorders, with the alteration of gut microbiota composition and function. A total of 188 species, which mainly from Bacteroidetes, Actinobacteria, Firmicutes, and Proteobacteria phyla, and 36 metabolites were markedly changed by HFD and revered by APS. Additionally, the altered glutathione metabolism and purine metabolism pathways were identified by both metagenomic function analysis and metabolite pathway enrichment analysis. Furthermore, the gut microbial alteration was associated with the changes of key intestinal metabolites. We found 31 and 20 species were correlated with purine metabolism and glutathione metabolism, respectively. Together, our results showed significant metagenomic and metabolomic changes after HFD feeding and APS intervention, revealed the potential correlation between gut microbial species and metabolites, and highlighted mechanisms of herb-derived polysaccharides by modulating gut microbiome and host metabolism underlying their benefits on metabolic disorders.

Keywords: Astragalus polysaccharides; hepatic steatosis; metabolic disorders; metabolomics; metagenomic.

Figure 1

APS attenuates hepatic steatosis in HFD-fed mice. Male C57BL/6J mice (4-week-old) were treated with chow-diet (Con) or high-fat diet (HFD) with or without APS supplementation (8% APS in HFD) for 14 weeks. (A) Body weight. (B) Liver weight. (C) Hepatic steatosis scores. (D) Representative photomicrographs of liver tissue with H&E staining (magnification, ×200, 50 μm). (E) Serum total cholesterol (TC) level. (F) Serum alanine transaminase (ALT) level. (G) Serum aspartate aminotransferase (AST) level. (H) Fasting blood glucose level. (I) Fasting blood insulin level. n = 5, **p < 0.01, ***p < 0.001.

Figure 2

APS reverses gut dysbiosis in HFD-fed mice. Cecum samples of Con, HFD, and APS groups were analyzed with metagenomics. (A) Shannon index. (B) Bray_curtis based PCoA analysis followed by Permutational Multivariate Analysis Of Variance (PERMANOVA, R²: 0.695, p-value: 0.005, p.adjust: 0.005). (C) Bray_curtis based distance matrix. (D) Multigroup difference analysis of the top 9 abundant phyla. (E) Venn diagram illustrating the overlap of species in intestinal microbiota among the samples and number of species in three groups. (F) Co-occurrence network deduced from 188 differential species significant changed in HFD group compared to Con group and restored in APS group. Red edges, Spearman's rank correlation coefficient > 0.8, p < 0.01; blue edges, Spearman's rank correlation coefficient < −0.8, p < 0.01. n = 3 per group. **p < 0.01.

Figure 3

APS regulates gut microbial function in HFD-fed mice. (A) Bray_curtis based PCoA analysis in KEGG orthologys (KOs) level followed by Permutational Multivariate Analysis Of Variance (PERMANOVA, R²: 0.745, p-value: 0.002, p.adjust: 0.002). (B) Multigroup difference analysis in KEGG pathway at level 1. *p < 0.05, **p < 0.01. (C) LDA Effect Size (LEfSe) Analysis of gut microbial function at level 3 of metabolism was profiled among three groups. Heatmap of the relative abundances of the metabolic pathways with the criteria of LDA > 2. The black dots mean significant difference (p < 0.05) between HFD and Con groups or between HFD and APS groups.

Figure 4

APS changes fecal metabolome in HFD-fed mice. The LC/MS-based untargeted metabolic profiling in positive and negative mode was performed on fecal samples (n = 5). The SIMCA positive-derived (A) and negative-derived (B) PCA among the Con, HFD, and APS groups. (C) Heatmap of differential metabolites identified in cecum samples between Con vs HFD, and HFD vs APS groups respectively with the double criteria of both VIP > 1 and p < 0.05. Each column represents an individual sample. (D) Bubble diagram of metabolic pathways enrichment based on the differential metabolites among three groups. One bubble represents one metabolic pathway. Top 5 pathways were listed on the right.

Figure 5

Spearman correlations between differential metabolites and differential bacterial species. Positive correlations indicated by red cubes and negative correlations indicated by blue cubes. A Spearman's correlation coefficient less than −0.7 or more than 0.7 with p < 0.01 was selected. The selected species were shown with its phylum and genus information.