Gut mycobiome alterations in obesity in geographically different regions

Abstract

The gut fungi play important roles in human health and are involved in energy metabolism. This study aimed to examine gut mycobiome composition in obese subjects in two geographically different regions in China and to identify specific gut fungi associated with obesity. A total of 217 subjects from two regions with different urbanization levels [Hong Kong (HK): obese, n = 59; lean, n = 59; Kunming (KM): obese, n = 50; lean, n = 49. Mean body mass index (BMI) for obesity = 33.7] were recruited. We performed deep shotgun metagenomic sequencing on fecal samples to compare gut mycobiome composition and trophic functions in lean and obese subjects across these two regions. The gut mycobiome of obese subjects in both HK and KM were altered compared to those of lean subjects, characterized by a decrease in the relative abundance of Nakaseomyces, Schizosaccharomyces pombe, Candida dubliniensis and an increase in the abundance of Lanchanceathermotolerans, Saccharomyces paradox, Parastagonospora nodorum and Myceliophthorathermophila. Reduced fungal - bacterial and fungal - fungal correlations as well as increased negative fungal-bacterial correlations were observed in the gut of obese subjects. Furthermore, the anti-obesity effect of fungus S. pombe was further validated using a mouse model. Supplementing high-fat diet-induced obese mice with the fungus for 12 weeks led to a significant reduction in body weight gain (p < 0.001), and an improvement in lipid and glucose metabolism compared to mice without intervention. In conclusion, the gut mycobiome composition and functionalities of obese subjects were altered. These data shed light on the potential of utilizing fungus-based therapeutics for the treatment of obesity. S. pombe may serve as a potential fungal probiotic in the prevention of diet-induced obesity and future human trials are needed.

Keywords: Obesity; Schizosaccharomyces pombe; dietary habit; fungi dysbiosis; mycobiome.

Figure 1.

Fecal fungal community structure in Hong Kong and Kunming cohorts. (A) Variations in gut mycobiome composition at genus level in association with obesity across two regions. (B) The effect size of covariables on human gut mycobiome variation. *p < 0.05, ***p < 0.001. (C–E) Variations in the α-diversity of human gut mycobiome in association with obesity across two regions. Fecal fungal Shannon diversity (C) evenness (D) and Chao1 richness (E). HK, Hong Kong; KM, Kunming. Statistical significance was determined via t-tests; bars with numbers indicate statistical differences.

Figure 2.

Variations in gut mycobiome composition (a) and differences between the relative abundances of major fungal taxa (b) at the species level in association with obesity in Hong Kong and Kunming cohorts. Only dominant fungal species (mean relative abundance ≥1.0%) are presented. Control indicates blank group for assessing environmental contamination during sample collection and processing. Log transformation of relative abundance is shown in the boxplot. HK, Hong Kong; KM, Kunming. Statistical significance was determined by t-test with FDR correction; *p<0.05, **p<0.01.

Figure 3.

Putative fungal functional profiles in the fecal samples of lean and obese subjects from Hong Kong and Kunming cohorts. (a) Pie chart of the 12 guilds in the fecal samples. Lean in HK, lean subjects in Hong Kong; obesity in HK, obese subjects in Hong Kong; lean in KM, lean subjects in Kunming; obese in KM, obese subjects in Kunming. (b-e) relative abundance of animal endosymbiotic (b), epiphytic (c), plant pathogenic (d), plant saprotrophic (e) fungi in the fecal samples. All data are presented as the mean ± SEM. Bars with numbers indicate statistical differences, p-values are from t-test.

Figure 4.

Intra-kingdom co-occurrence networks between fungal taxa in the feces of lean and obese subjects from Hong Kong and Kunming cohorts. (a) Lean in HK; lean group from Hong Kong cohorts; (b) obese in HK; obese group from Hong Kong cohorts; (c) lean in KM; lean group from Kunming cohorts; (d) obese in KM; obese group from Kunming cohorts. Colored nodes represent operational taxonomic units (OTUs) assigned to major genera. Edges between nodes stands for either positive (blue) or negative (red) co-abundance relationships inferred from OTU abundance profiles using the SparCC method (pseudo p<0.05, correlation value ≤−0.1 or ≥0.1).

Figure 5.

Correlations between α diversity (Shannon diversity, evenness, and Chao1 richness) of fecal bacteria and fungi in lean and obese subjects from Hong Kong and Kunming cohorts. (a) Lean in HK; lean group from Hong Kong cohorts; (b) obese in HK; obese group from Hong Kong cohorts; (c) Lean in KM; Lean group from Kunming cohorts; (d) obesity in KM; obese group from Kunming cohorts. The asterisks inside circles indicate significant associations between mycobiome and bacteriome; *p<0.05, **p<0.01, ***p<0.001; their significance was corrected with FDR adjustment. Blue circles indicate positive correlations, and red circles indicate inverse correlations; color and size are intensified according to the correlation coefficient.

Figure 6.

Effects of S. pombe on obesity and serological parameters. (a) Schematic diagram, (b) body weight gain; (c-f) adipose change, adiposity index (c), subcutaneous adipose tissue weight (d), epididymal adipose tissue weight (e), perirenal adipose weight (f); (G-J) glucose metabolism changes, insulin tolerance test (ITT) (g), area under curve (AUC) of ITT (h), oral glucose tolerance test (OGTT) (i), AUC of OGTT (J). Data are shown as means ± SEM. p-values were corrected with FDR; *p<0.05, **p<0.01, ***p<0.001, ns indicates no significant difference. ND (n=5): group fed with normal diet; HF (n=5): group fed with high-fat diet; SE+ND (n=5): normal diet group treated with Saccharomyces cerevisiae; SE+HF (n=4): high-fat diet group treated with S. cerevisiae; SP+ND (n=5): normal diet group treated with S. pombe; SP+HF (n=5): high-fat diet group treated with S. pombe.