Gut Bifidobacterium pseudocatenulatum protects against fat deposition by enhancing secondary bile acid biosynthesis

Abstract

Gut microbiome is crucial for lipid metabolism in humans and animals. However, how specific gut microbiota and their associated metabolites impact fat deposition remains unclear. In this study, we demonstrated that the colonic microbiome of lean and obese pigs differentially contributes to fat deposition, as evidenced by colonic microbiota transplantation experiments. Notably, the higher abundance of Bifidobacterium pseudocatenulatum was significantly associated with lower backfat thickness in lean pigs. Microbial-derived lithocholic acid (LCA) species were also significantly enriched in lean pigs and positively correlated with the abundance of B. pseudocatenulatum. In a high-fat diet (HFD)-fed mice model, administration of live B. pseudocatenulatum decreased fat deposition and enhances colonic secondary bile acid biosynthesis. Importantly, pharmacological inhibition of the bile salt hydrolase (BSH), which mediates secondary bile acid biosynthesis, impaired the anti-fat deposition effect of B. pseudocatenulatum in antibiotic-pretreated, HFD-fed mice. Furthermore, dietary LCA also decreased fat deposition in HFD-fed rats and obese pig models. These findings provide mechanistic insights into the anti-fat deposition role of B. pseudocatenulatum and identify BSH as a potential target for preventing excessive fat deposition in humans and animals.

Keywords: Bifidobacterium pseudocatenulatum; bile salt hydrolase; fat deposition; lithocholic acid; secondary bile acid biosynthesis.

Figure 1

Distinct microbial profile associated with backfat deposition in lean and obese pigs. (A) Backfat thickness. Data are shown as mean ± SEMs. n = 40 per group, *p < 0.05 (Student's t test). (B) Representative hematoxylin and eosin staining images of subcutaneous adipose tissue (100×). Data are shown as mean ± SEMs. n = 40 per group, *p < 0.05 (Student's t test). (C) Colonic microbial α diversity based on observed operational taxonomic units. Data are shown as mean ± SEMs, n = 40 per group, *p < 0.05 (Student's t test). (D) Colonic microbial principal coordinates analysis based on Bray–Curtis distances. (E) The association between gut microbiome and backfat thickness (Mantel test). (F) Top 20 differential pathways based on phylogenetic investigation of communities by reconstruction of unobserved states 2.0 analysis. DLY group, Duroc × Landrace × Yorkshire pigs; NX group, Ningxiang pigs; SEM, standard error of the mean.

Figure 2

The gut microbiome of lean and obese pigs contributes to the variations of lipid deposition. (A) Schematic of the colonic microbiota transplantation experiment. C57BL/6 male mice pretreated with antibiotic cocktail for 2 weeks, and received colonic microbiota suspension of Ningxiang (NX) or Duroc × Landrace × Yorkshire (DLY) pigs for 1 week. Mice were fed with chow–diet (CD)/high‐fat diet (HFD) for 9 weeks. (B) Body weight change; DLY‐CD: Mice received colonic microbiome from DLY pigs and fed with chow diet; NX‐CD: Mice received colonic microbiome from NX pigs and fed with chow diet. Data are shown as mean ± SEMs, n = 8 per group, *p < 0.05 (repeated measures analysis of variance). (C) Accumulative food intake. Data are shown as mean ± SEMs, n = 4 per group, *p < 0.05 (Student's t test). (D) Organ weight; epididymal white adipose tissue (eWAT); perirenal white adipose tissue (pWAT); subcutaneous white adipose tissue (sWAT); white adipose tissue (WAT). Data are shown as mean ± SEMs, n = 8 per group, *p < 0.05 (Student's t test). (E) Representative hematoxylin and eosin (H&E)‐stained image of eWAT, and quantification of the mean adipocytes area. Data are shown as mean ± SEMs, n = 4 per group, *p < 0.05 (Student's t test), Scale bar = 50 μm. (F) Representative H&E‐stained image of sWAT and quantification of the mean adipocytes area. Data are shown as mean ± SEMs, n = 4 per group, *p < 0.05 (Student's t test), Scale bar = 50 μm. (G) Body weight change; DLY‐HFD: Mice received colonic microbiome from DLY pigs and fed with HFD; NX‐CD: Mice received colonic microbiome from NX pigs and fed with HFD. Data are shown as mean ± SEMs, n = 8 per group, *p < 0.05 (repeated measures analysis of variance). (H) Accumulative food intake. Data are shown as mean ± SEMs, n = 4 per group, *p < 0.05 (Student's t test). (I) Adipose tissue weight. epididymal white adipose tissue (eWAT), perirenal white adipose tissue (pWAT), subcutaneous white adipose tissue (sWAT), white adipose tissue (WAT). Data are shown as mean ± SEMs, n = 8 per group, *p < 0.05 (Student's t test). (J) Representative H&E‐stained image of eWAT, and quantification of the mean adipocytes area. Data are shown as mean ± SEMs, n = 4 per group, *p < 0.05 (Student's t test), Scale bar = 50 μm. (K) Representative H&E‐stained image of sWAT, and quantification of the mean adipocytes area. Data are shown as mean ± SEMs, n = 4 per group, * p < 0.05 (Student's t test), Scale bar = 50 μm. DLY‐CD: Mice received colonic microbiome from DLY pigs and fed with chow diet; NX‐CD: Mice received colonic microbiome from NX pigs and fed with chow diet; DLY‐HFD: Mice received colonic microbiome from DLY pigs and fed with HFD; NX‐CD: Mice received colonic microbiome from NX pigs and fed with HFD. SEM, standard error of the mean.

Figure 3

Colonic‐specific microbiome mediates variations in secondary bile acid metabolism between lean and obese pigs. (A) Volcano plot illustrating differential bacterial species between Duroc × Landrace × Yorkshire (DLY) and Ningxiang (NX) pigs (false discovery rate [FDR] < 0.01, Log₂ fold change > 2 or Log₂ fold change < −2), n = 25. None indicates species with no significant differences; DLY indicates species enriched in DLY pigs; NX indicates species enriched in NX pigs. (B) Twelve bacterial species were identified as bacterial biomarkers to predict backfat thickness by random forest analysis, n = 25. (C) Two‐tailed spearman correlation coefficient analysis of bacterial species, and backfat thickness (Top 20). + means positively correlation; − means negatively correlation. (D) Venn diagram of differential analysis, random forest, and Spearman analysis. (E) Principal component analysis score plot of colonic metabolomics, n = 8. (F) Volcano plot showing differential metabolites between DLY and NX pigs (FDR < 0.05, Log₂ fold change > 1.32 or Log₂ fold change < −1.32), n = 8. None indicates metabolites with no significant differences; DLY indicates metabolites enriched in DLY pigs; NX indicates metabolites enriched in NX pigs.

Figure 4

Bifidobacterium pseudocatenulatum attenuates excessive fat deposition and enhances secondary bile acid biosynthesis in high‐fat diet (HFD)‐fed mice. (A) Schematic of the B. pseudocatenulatum monocolonization experiment. C57BL/6 male mice were randomly divided into five groups: (1) negative control (NC) group: chow diet + vehicle; (2) HFD group: HFD + vehicle; (3) BPS group: HFD + B. pseudocatenulatum supernatant; (4) HK group: HFD + heat‐killed B. pseudocatenulatum; (5) BP group: HFD + live B. pseudocatenulatum. (B) The copies of B. pseudocatenulatum 16S rDNA in colonic contents. Median with interquartile range (Kruskal–Wallis test), n = 8 per group, *p < 0.05. (C) Body weight change. Median with interquartile range (friedman rank sum test), n = 13 per group, *p < 0.05. (D) Accumulative food intake. Median with range (Kruskal–Wallis test), n = 3 per group, *p < 0.05. (E) Adipose tissue weight. epididymal white adipose tissue (eWAT), perirenal white adipose tissue (pWAT), subcutaneous white adipose tissue (sWAT), white adipose tissue (WAT). Median with interquartile range (Kruskal–Wallis test), n = 8 per group, *p < 0.05. (F) Representative hematoxylin and eosin‐stained image of eWAT and sWAT, and quantification of the mean adipocytes area. n = 5 per group, *p < 0.05, Scale bar = 50 μm. (G) The mRNA expression of lipid metabolism in eWAT. Median with interquartile range (Kruskal–Wallis test), n = 8 per group, *p < 0.05. Takeda G protein‐coupled receptor 5 (Tgr5), Farnesoid X receptor (Fxr), Peroxisome proliferator‐activated receptor gamma (Pparγ), Peroxisome proliferator‐activated receptors α (Pparα), Carnitine palmitoyltransferase 1 beta (Cpt1β). (H) BA concentration in colonic contents. 3‐keto‐5β‐cholic acid (3‐K‐5β‐CA), 3β‐ursodeoxycholic acid (3‐β‐UDCA), cholic acid (CA), chenodeoxycholic acid (CDCA), deoxycholic acid (DCA), epilithocholic acid (epiLCA), hyodeoxycholic acid (HDCA), lithocholic acid (LCA), ursodeoxycholic acid (UDCA). Median with interquartile range (Kruskal–Wallis test), n = 8 per group, *p < 0.05. NC group: chow diet + vehicle; HFD group: HFD + vehicle; BPS group: HFD + B. pseudocatenulatum supernatant; HK group: HFD + heat‐killed B. pseudocatenulatum; BP group: HFD + live B. pseudocatenulatum. BA, bile acids.

Figure 5

Pharmacological inhibition of the bile salt hydrolase impairs the anti‐fat deposition effect of Bifidobacterium pseudocatenulatum in antibiotic‐pretreated high‐fat diet (HFD)‐fed mice. (A) Schematic of pharmacological inhibition of bile salt hydrolase experiment. C57BL/6 male mice were pretreated with HFD for 8 weeks, and antibiotic cocktail for 2 weeks. Mice were gavage with B. pseudocatenulatum or B. pseudocatenulatum with GR7 for 3 weeks. ABP group: HFD + antibiotics mixture (ABX) + B. pseudocatenulatum + Vehicle (n = 8); GR7 group: HFD + ABX + B. pseudocatenulatum + Gut‐restricted 7 (GR7, n = 7). (B) Colonic bile acids concentration. 3β‐ursodeoxycholic acid (3‐β‐UDCA), cholic acid (CA), deoxycholic acid (DCA), hyodeoxycholic acid (HDCA), lithocholic acid (LCA), taurocholic acid (TCA), ursodeoxycholic acid (UDCA). Data are shown as mean ± SEMs, ABP (n = 8), GR7 (n = 7), *p < 0.05 (Student's t test). (C) The copies of B. pseudocatenulatum 16S rDNA in colonic contents. Data are shown as mean ± SEMs, ABP (n = 8), GR7 (n = 7), *p < 0.05 (Student's t test). (D) Body weight change. Data are shown as mean ± SEMs, ABP (n = 8), GR7 (n = 7), *p < 0.05 (repeated measures analysis of variance). (E) Organ weight. epididymal white adipose tissue (eWAT), perirenal white adipose tissue (pWAT), subcutaneous white adipose tissue (sWAT), white adipose tissue (WAT). Data are shown as mean ± SEMs, ABP (n = 8), GR7 (n = 7), *p < 0.05 (Student's t test). (F) Representative hematoxylin and eosin (H&E) image of eWAT, and quantification of the mean adipocyte area. Data are shown as mean ± SEMs, ABP (n = 8), GR7 (n = 7), *p < 0.05 (Student's t test), Scale bar = 50 μm. (G) Representative H&E image of sWAT, and quantification of the mean adipocyte area. Data are shown as mean ± SEMs, ABP (n = 8), GR7 (n = 7), *p < 0.05 (Student's t test), Scale bar = 50 μm. (H) The mRNA expression of lipid metabolism in eWAT. Data are shown as mean ± SEMs, ABP (n = 8), GR7 (n = 7), *p < 0.05 (Student's t test). Farnesoid X receptor (Fxr), Peroxisome proliferator‐activated receptor gamma (Pparγ), Carnitine palmitoyltransferase 1 beta (Cpt1β), Takeda G protein‐coupled receptor 5 (Tgr5), Hormone‐sensitive lipase (Hsl), Sterol regulatory element‐binding protein 1 (Srebp1), Peroxisome proliferator‐activated receptors α (Pparα), Acetyl‐CoA carboxylase (Acc), Fatty acid synthase (Fasn). ABP group: HFD + ABX + live B. pseudocatenulatum + Vehicle; GR7 group: HFD + ABX + live B. pseudocatenulatum + GR7. SEM, standard error of the mean.

Figure 6

Dietary supplementation with lithocholic acid (LCA) against fat deposition in high‐fat diet (HFD) fed rats and Ningxiang pigs. (A) Body weight change of rats. Data are shown as mean ± SEMs, n = 8 per group, *p < 0.05 (repeated measures analysis of variance). HFD group: rats fed an HFD; LCA group: rats fed an HFD supplemented with 0.3% LCA. (B) Accumulative food intake of rats. Data are shown as mean ± SEMs, n = 4 per group, *p < 0.05 (Student's t test). HFD group: rats fed an HFD; LCA group: rats fed an HFD supplemented with 0.3% LCA. (C) Adipose tissue weight of rats. epididymal white adipose tissue (eWAT), perirenal white adipose tissue (pWAT), subcutaneous white adipose tissue (sWAT), white adipose tissue (WAT). Data are shown as mean ± SEMs, n = 8 per group, *p < 0.05 (Student's t test). HFD group: rats fed an HFD; LCA group: rats fed an HFD supplemented with 0.3% LCA. (D) Representative hematoxylin and eosin (H&E)‐stained image of eWAT, and quantification of the mean adipocytes area in rats. Data are shown as mean ± SEMs, n = 8 per group, *p < 0.05 (Student's t test), Scale bar = 50 μm. HFD group: rats fed an HFD; LCA group: rats fed an HFD supplemented with 0.3% LCA. (E) The mRNA expression of lipid metabolism in eWAT of rats. Data are shown as mean ± SEMs, n = 8 per group. *p < 0.05 (Student's t test). HFD group: rats fed an HFD; LCA group: rats fed an HFD supplemented with 0.3% LCA. Diacylglycerol O‐acyltransferase 1 (Dgat1), Peroxisome proliferator‐activated receptor gamma (Pparγ), Farnesoid X receptor (Fxr), Takeda G protein‐coupled receptor 5 (Tgr5), Carnitine palmitoyltransferase 1 beta (Cpt1β), Hormone‐sensitive lipase (Hsl), Acetyl‐CoA carboxylase (Acc). (F) Serum triglycerides, and nonesterified fatty acid concentrations of rats. Data are shown as mean ± SEMs, n = 6 per group. *p < 0.05 (Student's t test). Negative control (NC) group: pigs fed a basal diet; LCA group: pigs fed a basal diet supplemented with 792 mg/kg LCA. (G) Backfat thickness of NX pigs. Data are shown as mean ± SEMs, n = 5 per group. *p < 0.05 (Student's t test). NC group: pigs fed a basal diet; LCA group: pigs fed a basal diet supplemented with 792 mg/kg LCA. (H) Representative H&E‐stained image of subcutaneous adipose tissue and quantification of the mean adipocytes area in NX pigs. Data are shown as mean ± SEMs, n = 3 per group. *p < 0.05 (Student's t test), Scale bar = 50 μm. NC group: pigs fed a basal diet; LCA group: pigs fed a basal diet supplemented with 792 mg/kg LCA. SEM, standard error of the mean.