Benefits of Huang Lian mediated by gut microbiota on HFD/STZ-induced type 2 diabetes mellitus in mice

Abstract

Background: Huang Lian (HL), one of the traditional Chinese medicines (TCMs) that contains multiple active components including berberine (BBR), has been used to treat symptoms associated with diabetes for thousands of years. Compared to the monomer of BBR, HL exerts a better glucose-lowering activity and plays different roles in regulating gut microbiota. However, it remains unclear what role the gut microbiota plays in the anti-diabetic activity of HL.

Methods: In this study, a type 2 diabetes mellitus (T2DM) mouse model was induced with a six-week high-fat diet (HFD) and a one-time injection of streptozotocin (STZ, 75 mg/kg). One group of these mice was administrated HL (50 mg/kg) through oral gavage two weeks after HFD feeding commenced and continued for four weeks; the other mice were given distilled water as disease control. Comprehensive analyses of physiological indices related to glycolipid metabolism, gut microbiota, untargeted metabolome, and hepatic genes expression, function prediction by PICRUSt2 were performed to identify potential mechanism.

Results: We found that HL, in addition to decreasing body fat accumulation, effectively improved insulin resistance by stimulating the hepatic insulin-mediated signaling pathway. In comparison with the control group, HL treatment constructed a distinct gut microbiota and bile acid (BA) profile. The HL-treated microbiota was dominated by bacteria belonging to Bacteroides and the Clostridium innocuum group, which were associated with BA metabolism. Based on the correlation analysis, the altered BAs were closely correlated with the improvement of T2DM-related markers.

Conclusion: These results indicated that the anti-diabetic activity of HL was achieved, at least partly, by regulating the structure of the gut microbiota and the composition of BAs.

Keywords: Huang Lian; bile acids; gut microbiota; microbial BA metabolism; type 2 diabetes mellitus.

Figure 1

HL treatment improved glucose homeostasis and insulin sensitivity. (A) Fasting blood glucose. (B) Blood glucose levels and the AUC of blood glucose during OGTT. (C) Fasting serum insulin. (D) Serum insulin levels and the AUC of serum insulin during OGTT. (E) Homeostasis model assessment of insulin resistance (HOMA-IR). (F) Insulin resistance index (the product of the AUC of blood glucose multiplied by the AUC of serum insulin). (G) The phosphorylation of IRS1 and Akt, as well the expression of Glut4 in the liver. The relative abundances of p-IRS1, p-Akt, and Glut4 were normalized by IRS1, Akt, and GADPH, respectively. Data are presented as the mean ± SEM. Significance was evaluated using one-way ANOVA followed by Tukey’s post-hoc test. *p < 0.05, ** p < 0.01, ***p < 0.001. For the curve of graph (B, D), *p < 0.05, **p < 0.01, ***p < 0.001 vs. NC group; # p < 0.05, ## p < 0.01 vs. DM group.

Figure 2

HL treatment blunted obesity and fat accumulation. (A) The body weight gain (% 1^st day body weight). (B) Perirenal, Retroperitoneal, and Epididymal mass (% body weight), respectively. (C) Red oil staining sections of epididymal fat (scale bar = 100 µm). (D) Representative H&E staining sections of epididymal fat (scale bar = 100 µm) and the mean adipocyte area. (E) Representative H&E staining sections of the liver (scale bar = 100 µm) and NAFLD activity score. Data are presented as the mean ± SEM. Significance was evaluated using one-way ANOVA followed by Tukey’s post-hoc test. *p < 0.05, ***p < 0.001.

Figure 3

HL treatment significantly shaped a unique structure of the gut microbiota. (A) α-diversity of the gut microbiota represented as Observed_ASVs and the Shannon index. Data was shown in box plot. And the line in the middle of box is plotted at the median, the inferior and superior limits of the box correspond to the 25^th and 75^th percentiles, the whiskers correspond to maximum and minimum. Significance was analyzed using one-way ANOVA followed by Tukey’s post-hoc test. (B) Principal coordinate analysis (PCoA) of the gut microbiota based on the Bray-Curtis distance. The box plots showed the changes of gut microbiota among three groups on PC1 or PC2 (the line in the middle of box is plotted at the median, the inferior and superior limits of the box correspond to the 25^th and 75^th percentiles, the whiskers correspond to maximum and minimum). Significance was analyzed using one-way ANOVA followed by Tukey’s post-hoc test. (C) Bray-Curtis distance from the DM group and DM-HL group to the NC group. Data are presented as the mean ± SEM. Significance was analyzed using a two-tailed Mann-Whitney test. **p < 0.01, ***p < 0.001. (D) Heatmap of 101 ASVs across all samples. The color of the block corresponds to the log₂-transformed value based on the absolute abundance of each sample. Left, several classifications of the change in ASV abundance among groups; frame colors indicate specific situations of change. Right, the changing direction of 101 ASVs through pairwise comparison of groups analyzed by the two-tailed Mann-Whitney test. Orange indicates that ASV abundance was higher in the latter group, and the grey indicates that ASV abundance was lower in the former group.

Figure 4

HL treatment altered the bile acid profile in the colon. (A) Principal component analysis (PCA) and orthogonal projection to latent structure-discriminant analysis (OPLS-DA) plot of the metabolome across all samples. (B) Enriched metabolite pathway in the DM-HL group compared to the DM group based on metabolic enrichment pathway analysis. (C) Enriched metabolite pathway in the DM-HL group compared to the NC group. (D) Metabolites involved in microbial bile acid metabolism in the gut. TCDCA, Taurochenodesoxycholic acid; TCA, Taurocholic acid; UDCA, Ursodeoxycholic acid; CDCA, Chenodeoxycholic acid; THCA, taurohyocholic acid; LCA, Lithocholic acid; isoLCA, isolithocholic acid; DCA, Deoxycholic acid. (E) Spearman’s correlation analysis between metabolites and T2DM-related parameters. For graph (D), data are presented as the mean ± SEM. Significance was evaluated by an unpaired t test with Welch’s correction. *p < 0.05, **p < 0.01, ***p < 0.001.

Figure 5

Alteration of BA composition was related to that in the gut microbiota. (A) The absolute abundance of the cbh gene predicted by PICRUSt2. (B) The absolute abundance of the baiB, baiCD, and baiH genes predicted by PICRUSt2. (C) The Spearman correlation between metabolites and microbes. ASVs belong to different classifications based on their inconsistent responses to HFD/STZ treatment and HL administration. The metabolites were involved in microbial bile acid metabolism in the gut. The color of the block indicates the correlation (red = positive, black = negative). The corrected p value was adjusted by FDR. *p < 0.05; **p < 0.01. For graph (A, B), data are presented as the mean ± SEM, and significance was evaluated by an unpaired t test with Welch’s correction. *p < 0.05, **p < 0.01.