Effects of Prevotella copri on insulin, gut microbiota and bile acids

Abstract

Obesity is becoming a major global health problem in children that can cause diseases such as type 2 diabetes and metabolic disorders, which are closely related to the gut microbiota. However, the underlying mechanism remains unclear. In this study, a significant positive correlation was observed between Prevotella copri (P. copri) and obesity in children (p = 0.003). Next, the effect of P. copri on obesity was explored by using fecal microbiota transplantation (FMT) experiment. Transplantation of P. copri. increased serum levels of fasting blood glucose (p < 0.01), insulin (p < 0.01) and interleukin-1β (IL-1β) (p < 0.05) in high-fat diet (HFD)-induced obese mice, but not in normal mice. Characterization of the gut microbiota indicated that P. copri reduced the relative abundance of the Akkermansia genus in mice (p < 0.01). Further analysis on bile acids (BAs) revealed that P. copri increased the primary BAs and ursodeoxycholic acid (UDCA) in HFD-induced mice (p < 0.05). This study demonstrated for the first time that P. copri has a significant positive correlation with obesity in children, and can increase fasting blood glucose and insulin levels in HFD-fed obese mice, which are related to the abundance of Akkermansia genus and bile acids.

Keywords: Obesity; bile acids; gut microbiota; insulin; prevotella copri.

Figure 1.

Difference in gut microbiota between obesity and normal children. PCA, Chao1 index, Shannon index, observed specification index, and Simpson index of the gut microbiota in obesity and normal weight children (a). Composition of gut microbiota at the phylum (b), family (e), and genus (g) levels in obesity and normal weight children. OTU clustering and LDA analysis of gut microbiota in obesity and normal weight children, with the length of the bar graph indicating the impact of each OTU (d). The proportion of Firmicutes and Bacteroidetes in the gut microbiota of obesity and normal weight children (c). The difference in intestinal microflora between children with obesity and normal weight children at the family level (f) and the genus level (h). Obese, children with obesity (n = 87); Normal weight, normal weight children(n = 107). Data are shown as the means ± SEMs. ANOSIM: R = 0.135, p = 0.049, *p < 0.05.

Figure 2.

Correlation analysis between different bacteria and BMI. P. copri was positively correlated with BMI (a); Paraacteroides was negatively correlated with BMI (b); Bifidobacterium was negatively correlated with BMI (c); Oscillospira was negatively correlated with BMI (d). p values were obtained after Pearson’s correlation test.

Figure 3.

Effects of P. copri on body weight and insulin sensitivity in mice. Experimental design (a); growth curve of mice (b); final body weight (c); representative image of abdominal fat section (d); abdominal fat rate (e), fasting blood glucose (f), fasting insulin (g) and HOMA-IR index (h). Data are shown as the means±SEMs (n = 10). *p < .05, **p < .01.

Figure 4.

Modulation of P. copri transplantation on intestinal microflora in mice. Cluster analysis by using PCA (a). Shannon index, Chao 1 index, Simpson index and Ace index (b). Effect of Prevotella on the relative abundance of the main microbes at the phylum level (c) and Verrucomicrobiota phylum (d). Effect of P. copri on the relative abundance of the main microbes at the family level (e). Microorganisms with significant differences at the family level and genus level, with the left indicating microorganisms with differences at the family level and the right indicating micro elevation with differences at the genus level (f), Akkermansia genus (g). Data are shown as the means±SEMs (n = 6). *p < .05, **p < .01.

Figure 5.

Modulation of P. copri on intestinal SCFAs and bile acids in mice. The composition of BAs (a). The content of total BAs, primary BAs and secondary BAs (b). The content of UDCA (c). The composition of SCFAs (d). Content of acetic acid (e). Content of butanoic acid (f). Data are shown as the means±SEMs (n = 6). *p < .05, **p < .01.