Oral administration of Blautia wexlerae ameliorates obesity and type 2 diabetes via metabolic remodeling of the gut microbiota

Abstract

The gut microbiome is an important determinant in various diseases. Here we perform a cross-sectional study of Japanese adults and identify the Blautia genus, especially B. wexlerae, as a commensal bacterium that is inversely correlated with obesity and type 2 diabetes mellitus. Oral administration of B. wexlerae to mice induce metabolic changes and anti-inflammatory effects that decrease both high-fat diet-induced obesity and diabetes. The beneficial effects of B. wexlerae are correlated with unique amino-acid metabolism to produce S-adenosylmethionine, acetylcholine, and L-ornithine and carbohydrate metabolism resulting in the accumulation of amylopectin and production of succinate, lactate, and acetate, with simultaneous modification of the gut bacterial composition. These findings reveal unique regulatory pathways of host and microbial metabolism that may provide novel strategies in preventive and therapeutic approaches for metabolic disorders.

Fig. 1. Intestinal bacterial genera associated with body mass index (BMI) and type 2 diabetes mellitus (T2DM) in Japanese adults.

A BMI-related bacterial genera, which were selected and ranked according to R² score from single regression analysis (P < 0.05) among 16 genera that were identified through multiple-regression analysis by forward selection (Supplementary Table 3) by using the data of the 217 participants (Supplementary Table 1). B T2DM-related bacterial genera, which are selected and ranked according to R² score from single regression analysis (P < 0.05) among 22 genera that were identified through multiple logistic regression analysis by forward selection (Supplementary Table 4) by using the data of 192 participants (comprising 147 nonDM subjects and 45 T2DM patients and excluding 25 patients with Type 1 diabetes) (Supplementary Table 1). C Intestinal genera associated with both BMI and T2DM. D Odds ratios for Blautia abundance in the development of obesity (BMI ≥ 25) and T2DM. E Estimation of Blautia species according to BlastN analysis of representative OTU sequences.

Fig. 2. High-fat diet (HFD)-induced obesity and diabetes were ameliorated by oral administration of Blautia wexlerae to mice.

A Mice were maintained on standard chow (control diet, CD) or HFD without or with oral administration of B. wexlerae (Bw) three times each week and were weighed weekly. Data are representative of at least three independent experiments (n = 5, mean ± 1 SD). *P = 0.0105; **P = 0.0005 (two-way ANOVA). B Photographs of representative mice. C Photographs of representative mice and weight of epididymal adipose tissue (eAT). Data are combined from two independent experiments (n = 10, mean). **P = 0.0039 (one-way ANOVA). D HOMA-IR, an indicator of insulin resistance, calculated as ‘glucose (mg/dl) × insulin (µU/ml)/405’. Data are combined from two independent experiments without hemolytic samples (n = 7–10, mean). **P < 0.01 (one-way ANOVA). E Blood insulin was monitored through intraperitoneal glucose tolerance testing (IPGTT). Data are combined from two independent experiments (n = 10, mean ± 1 SD). **P < 0.01 (one-way ANOVA). F Blood glucose was monitored by using IPGTT. Data are combined from two independent experiments (n = 10, mean ± 1 SD). **P < 0.01 (one-way ANOVA). G Representative immunohistologic analysis of eAT. Macrophages, lipid droplets, and nuclei were visualized by using F4/80 monoclonal antibody (red), BODIPY (green), and DAPI (blue) staining, respectively. Scale bar, 100 µm. H Number of macrophages in the eAT. Data are combined from two independent experiments (n = 10, mean). *P = 0.0140; **P = 0.0001 (one-way ANOVA). I Gene expression of Tnfα, an inflammatory cytokine, and S100a8, a chemokine for recruiting macrophages, in the eAT mature adipocyte fraction (MAF). Data are combined from two independent experiments (n = 10, mean ± 1 SD). *P < 0.05; **P < 0.01 (one-way ANOVA). CD, lean mice fed a standard chow diet (CD-fed mice); HFD, obese mice fed a high-fat diet (HFD-fed mice); HFD + Bw, HFD-fed mice orally supplemented with B. wexlerae.

Fig. 3. Blautia wexlerae-derived metabolites showed anti-inflammatory and anti-adipogenesis properties in adipocytes as well as alteration of mitochondrial metabolism.

A 3T3L1 pre-adipocytes were differentiated into mature adipocytes in the absence (none) or presence of the cultured supernatant of B. wexlerae at a final concentration of 10%. Gene expression of Pparγ (a transcription factor used as a marker of adipocyte differentiation), S100a8 (a chemokine for recruiting macrophages), and Nrf1 (a transcriptional factor used as a marker of mitochondrial biogenesis) was measured in 3T3L1 pre-adipocytes and adipocytes. Data are representative of two independent experiments without samples below the detection limit (n = 3–4, mean ± 1 SD). *P < 0.05; **P < 0.01; n.s. not significant (one-way ANOVA). B Mitochondrial mass was measured by flow cytometry analysis using Mitogreen in 3T3L1 adipocytes treated without or with culture supernatant (sup.) of B. wexlerae at a final concentration of 1% or 10%. Data are representative of two independent experiments (n = 4, mean ± 1 SD). **P = 0.0012 (one-way ANOVA). C By using an XF24 extracellular flux analyzer, the oxygen consumption rate (OCR) was measured in 3T3L1 adipocytes treated without or with the cultured supernatant of B. wexlerae at the concentration of 10%. Data are combined from two independent experiments (n = 14, mean ± 1 SD). D Lipid accumulation was assessed by using oil red O staining in 3T3L1 adipocytes treated without or with culture supernatant of B. wexlerae at a final concentration of 1% or 10%. Data are representative of two independent experiments (n = 4, mean ± 1 SD). *P = 0.0316 (one-way ANOVA). E Gene expression of Nrf1 in the eAT MAF of mice. Data are combined from two independent experiments without samples below the detection limit (n = 9–10, mean ± 1 SD). *P = 0.0421; **P < 0.0001 (one-way ANOVA). F Representative metabolites of glycolysis (lactate) and the TCA cycle (citrate, isocitrate, and succinate) in the eAT MAF of mice were measured by using liquid chromatography–tandem mass spectroscopy (LC–MS/MS). Data are combined from two independent experiments (n = 8, mean ± 1 SD). *P < 0.05; **P < 0.01 (one-way ANOVA). G Acetyl-l-carnitine, a constituent of the inner mitochondrial membrane, in the eAT MAF of mice was measured by using LC–MS/MS. Data are combined from two independent experiments without samples below the detection limit (n = 5–6, mean ± 1 SD). *P = 0.0231 (one-way ANOVA). CD CD-fed mice; HFD HFD-fed mice, HFD + Bw HFD-fed mice supplemented with B. wexlerae.

Fig. 4. Blautia wexlerae showed unique characteristics in amino acid metabolism, such as production of S-adenosylmethionine, acetylcholine, and l-ornithine.

A Volcano plot showing LC–MS/MS analysis of B. wexlerae culture supernatant. Red and blue dots indicate metabolites increased and decreased, respectively, by more than fourfold in B. wexlerae culture supernatant compared with fresh medium (n = 4 biologically independent samples). Statistical significance was evaluated by using two-tailed unpaired t-test. B Quantitative measurement by LC–MS/MS of S-adenosylmethionine, acetylcholine, and l-ornithine in fresh medium (none) and culture supernatants of B. wexlerae (Bw) and major intestinal bacteria including Bacteroides vulgatus (Bv), Prevotella copri (Pc), and Faecalibacterium prausnitzii (Fp) (n = 3, mean ± 1 SD). ND not detected. **P < 0.01 (one-way ANOVA in comparison with none group). C Lipid accumulation was measured by oil red O staining in 3T3L1 adipocytes treated without (none) or with S-adenosylmethionine, acetylcholine, and l-ornithine at the concentration of 1 or 100 µM (n = 3–4, mean ± 1 SD). **P < 0.01 (one-way ANOVA). D Gene expression of S100a8, a chemokine for recruiting macrophages, in 3T3L1 adipocytes treated without (none) or with S-adenosylmethionine, acetylcholine, and l-ornithine at 100 µM (n = 4, mean ± 1 SD). *P = 0.0295 (one-way ANOVA in comparison with none group). Data are representative of two independent experiments (A–D).

Fig. 5. Administration of B. wexlerae altered the intestinal environment, including gut bacterial composition and fecal short-chain fatty acid (SCFA) content, in mice.

A Raman spectroscopic analysis. Raman shift signals specific for protein and starch were plotted on bacterial cells of B. wexlerae (Bw) and major intestinal bacteria including Bacteroide vulgatus (Bv), Prevotella copri (Pc), and Faecalibacterium prausnitzii (Fp). a.u., arbitrary units. B Starch (amylose and amylopectin) contents in B. wexlerae (Bw) and major intestinal bacteria, including B. vulgatus (Bv), P. copri (Pc), and F. prausnitzii (Fp). Data are representative of two independent experiments (n = 4 biologically independent samples, mean ± 1 SD). C The concentrations of succinate, lactate, acetate, propionate, propionate, and butyrate in fresh medium (none) and B. wexlerae culture supernatant (Bw). Data are representative of two independent experiments (n = 4, mean ± 1 SD). * P = 0.0286 (two-tailed Mann–Whitney U test). D SCFA content in fecal samples from mice. Mice were maintained on CD or HFD for 8 weeks with or without oral administration of B. wexlerae three times each week, after which fecal SCFAs were measured by HPLC. Data are combined from three independent experiments (n = 15, mean ± 1 SD). *P < 0.05 (one-way ANOVA). E Principal coordinate analysis (PCoA) of fecal bacterial composition in mice according to the Bray–Curtis distance at genus level. Mice were maintained on CD or HFD for 8 weeks with or without oral administration of B. wexlerae three times each week, after which fecal bacterial composition was analyzed by 16S amplicon sequencing. Data are combined from two independent experiments (n = 10). (F) Differences in bacterial taxonomy were ranked according to the linear discriminant analysis (LDA) effect size between HFD-fed mice and HFD-fed mice supplemented with B. wexlerae. CD, CD-fed mice; HFD, HFD-fed mice, HFD + Bw, HFD-fed mice supplemented with B. wexlerae.

Fig. 6. Interaction between B. wexlerae and butyrate-producing bacteria.

A Heatmap showing correlated relationship among human intestinal bacteria (n = 217). Red and blue fonts indicate bacterial genera from Fig. 1 that were positively or inversely, respectively, related to BMI/T2DM. B Positive correlation between Blautia and Butyricicoccus in human fecal samples (n = 217) (Pearson, two-tailed P value). C Estimation of Butyricicoccus species according to BlastN analysis of representative OTU sequences. D The absorbance of Butyricicoccus faecihominis-cultured medium in which the organisms were grown in the absence or presence of B. wexlerae-cultured medium at a final concentration of 1% or 10% (n = 4, mean ± 1 SD). ****P < 0.0001 (one-way ANOVA). E The absorbance of B. faecihominis-cultured medium in which the organisms were grown in the absence (none) or presence of 10 mM succinate, 10 mM lactate, or 10 mM acetate (n = 4, mean ± 1 SD). **P = 0.0019 (one-way ANOVA). F The concentration of butyrate in B. faecihominis-cultured medium in which the organisms were grown in the absence (none) or presence of B. wexlerae culture supernatant at 10%, 10 mM succinate, 10 mM lactate, or 10 mM acetate (n = 3–4, mean ± 1 SD). ***P = 0.0010; ****P < 0.0001 (one-way ANOVA). Data are representative of two independent experiments (D–F).