Modulation of the Gut Microbiota Structure with Probiotics and Isoflavone Alleviates Metabolic Disorder in Ovariectomized Mice

Abstract

The decrease in ovarian hormone secretion that occurs during menopause results in an increase in body weight and adipose tissue mass. Probiotics and soy isoflavones (SIFs) could affect the gut microbiota and exert anti-obesity effects. The objective of this study was to investigate the effects of probiotics and a diet containing SIF (SIF diet) on ovariectomized mice with menopausal obesity, including the gut microbiome. The results demonstrate that Bifidobacterium longum 15M1 can reverse menopausal obesity, whilst the combination of Lactobacillus plantarum 30M5 and a SIF diet was more effective in alleviating menopausal lipid metabolism disorder than either components alone. Probiotics and SIFs play different anti-obesity roles in menopausal mice. Furthermore, 30M5 alters the metabolites of the gut microbiota that increase the circulating estrogen level, upregulates the expression of estrogen receptor α in abdominal adipose tissue and improves the production of short-chain fatty acids (SCFAs). A SIF diet can significantly alter the structure of the fecal bacterial community and enrich the pathways related to SCFAs production. Moreover, 30M5 and a SIF diet acted synergistically to effectively resolve abnormal serum lipid levels in ovariectomized mice, and these effects appear to be associated with regulation of the diversity and structure of the intestinal microbiota to enhance SCFAs production and promote estrogen circulation.

Keywords: estrogen; isoflavones; menopause; microbiota; obesity; probiotics; short-chain fatty acids.

Figure 1

Animal trial design. OVX: Ovariectomy; Sham: Sham-operation.

Figure 2

Evolution of body weight during treatment. Data present as the means ± standard deviations (SD).

Figure 3

Effect of probiotics and a SIF diet on lipid metabolism and inflammation. (A) Glucose (Glu) and Total cholesterol (TC) levels in serum. (B) High-density lipoprotein cholesterol (HDL-C) and Low-density lipoprotein cholesterol (LDL-C) levels in serum. (C) Triglycerides (TG) and leptin in serum. (D) IL-6 level in abdominal fat. * p < 0.05, ** p < 0.01, *** p < 0.001 compared with the OVX group. Data present as the means ± SD. Significant difference between study groups were evaluated using ordinary one-way analysis of variance (ANOVA) and uncorrected Fisher’s LSD multiple comparison test.

Figure 4

L. plantarum 30M5 changes sex hormone levels and up-regulates estrogen receptors. (A) Severe atrophy of the uterus in OVX mice. (B) Estradiol in serum significantly increased by L. plantarum 30M5. (C) Follicle-stimulating hormone (FSH) in serum significantly decreased by ovariectomy. (D) The level of serum luteinizing hormone (LH). (E) The expression of Estrogen Receptor (ER) α in abdominal adipose tissue up-regulated by L. plantarum 30M5. * p < 0.05, ** p < 0.01, *** p < 0.001 compared with the OVX group. Data present as the means ± SD. Significant difference between study groups were evaluated using ordinary one-way analysis of variance (ANOVA) and uncorrected Fisher’s LSD multiple comparison test.

Figure 5

Diversity of intestinal microbiota reduced in ovariectomy and changed by a SIF diet treatment. (A) Alpha diversity in fecal samples. (B) Beta diversity in fecal samples was measured using principal coordinate analysis (PCoA) of unweighted UniFrac. (C)The radio of Firmicutes and Bacteroidetes. (D) The changed abundance at phylum level. (E) Differential abundance at family level between groups was shown with an OTU bubble plot. * p < 0.05, ** p < 0.01, *** p < 0.001 compared with the OVX group. Data present as the means ± SD. Significant difference between study groups were evaluated using ordinary one-way analysis of variance (ANOVA) and uncorrected Fisher’s LSD multiple comparison test.

Figure 6

Differential microbial analysis at genus level. (A) Phylogenetic tree depicting bacterial taxonomic hierarchy. The tax with log Linear Discriminant Analysis (LDA) scores above 3.00 and p score below 0.05 in Wilcoxon test and Kruskal–Wallis test. (B) Relative abundance of Lactococcus, Bifidobacterium, Romboutsia, [Eubacterium] brachy group, Catabacter, Dorea and Erysipelatoclostridium. * p < 0.05, ** p < 0.01, *** p < 0.001 compared with the OVX group. Data present as the means ± SD. Significant difference between study groups were evaluated using ordinary one-way analysis of variance (ANOVA) and uncorrected Fisher’s LSD multiple comparison test.

Figure 7

Heatmap of the differential pathways predicted by PICRUSt 2 (phylogenetic investigation of communities by reconstruction of unobserved states).

Figure 8

Correlation analysis of microbiota. (A) The correlation between sex hormones and intestinal microbiota was shown in network with OmicStudio tools. (B) Heatmap of changing microbial relationships.

Figure 9

Fecal Short-Chain Fatty Acids (SCFAs). (A) Acetate. (B) Propionate. (C) Butyrate. (D) Total SCFAs. * p < 0.05, ** p < 0.01, *** p < 0.001 compared with the OVX group. Data present as the means ± SD. Significant differences between study groups were evaluated using ordinary one-way analysis of variance (ANOVA) and uncorrected Fisher’s LSD multiple comparison test.