Fasting and refeeding triggers specific changes in bile acid profiles and gut microbiota

Abstract

Background: Bile acids (BAs) are closely related to nutrient supply and modified by gut microbiota. Gut microbiota perturbations shape BA composition, which further affects host metabolism.

Methods: We investigated BA profiles in plasma, feces, and liver of mice fed ad libitum, fasted for 24 h, fasted for 24 h and then refed for 24 h using ultraperformance liquid chromatography coupled to tandem mass spectrometry. Gut microbiota was measured by 16S rRNA gene sequencing. Expressions of BA biosynthesis-related genes in the liver and BA reabsorption-related genes in the ileum were analyzed.

Findings: Compared with the controls, unconjugated primary BAs (PBAs) and unconjugated secondary BAs (SBAs) in plasma were decreased whereas conjugated SBAs in plasma, unconjugated PBAs, unconjugated SBAs and conjugated SBAs in feces, and unconjugated SBAs in liver were increased in the fasting mice. The expression of BA biosynthesis-related genes in the liver and BA reabsorption-related genes in the ileum were decreased in the fasting mice compared with the controls. Compared with the controls, Akkermansia, Parabacteroides, Muribaculum, Eubacterium_coprostanoligenes and Muribaculaceae were increased in the fasting mice whereas Lactobacillus and Bifidobacterium were decreased. All these changes in BAs and gut microbiota were recovered under refeeding. Akkermansia was negatively correlated with plasma levels of unconjugated PBAs, unconjugated SBAs and glucose, whereas it was positively correlated with plasma conjugated SBAs, fecal unconjugated PBAs, and fecal unconjugated SBAs.

Conclusions: We characterized the BA profiles, gut microbiota, and gene expression responsible for BA biosynthesis and intestinal reabsorption to explore their rapid changes in response to food availability. Our study highlighted the rapid effect of nutrient supply on BAs and gut microbiota.

Keywords: bile acid; biosynthesis; fasting; gut microbiota; reabsorption; refeeding; 再吸收; 复食; 生物合成; 禁食; 肠道菌群; 胆汁酸.

FIGURE 1

Fasting and refeeding modifies bile acid composition. The mice were divided into three groups. One group was fed ad libitum, one group was fasted for 24 h, and one group was fasted for 24 h and then refed for another 24 h (A). The levels of individual bile acids were measured in the blood plasma (B), feces (D), and liver tissue (F) applying ultraperformance liquid chromatography coupled to tandem mass spectrometry. The levels of subgroup bile acids were calculated in the plasma (C), feces (E), and liver tissue (G). Statistical significance between the experimental groups was evaluated using analysis of variance with Bonferroni correction for multiple testing; n = 12; *p < .05. **p < .01. Data are presented as mean ± SEM. CA, cholic acid; CDCA, chenodeoxycholic acid; Con‐, conjugated; DCA, deoxycholic acid; GCA, glycocholic acid; GCDCA, glycochenodeoxycholic acid; HCA, hyocholic acid; LCA, lithocholic acid; MCA, muricholic acid; T‐β‐MCA, tauro‐β‐muricholic acid; T‐α‐MCA, tauro‐α‐muricholic acid; T‐ω‐MCA, tauro‐ω‐muricholic acid; TCA, taurocholic acid; TCDCA, taurochenodeoxycholic acid; TDCA, taurodeoxycholic acid; THCA, taurohyocholic acid; TLCA, taurolithocholic acid; TUDCA, tauroursodeoxycholic acid; PBA, primary bile acid; SBA, secondary bile acid; UDCA, ursodeoxycholic acid; Uncon‐, unconjugated.

FIGURE 2

Bile acid biosynthesis and intestinal reabsorption under the fasting‐refeeding cycle. Bile acid (BA) synthesis pathways in liver (A). The enzymes required for the synthesis of cholic acid (CA) and chenodeoxycholic acid (CDCA). The classic pathway of  BA synthesis is initiated by the rate‐limiting enzyme cholesterol 7α‐hydroxylase (CYP7A1) to specifically hydroxylate cholesterol at the 7α position, forming 7α‐hydroxycholesterol. 7α‐hydroxycholesterol is converted to 7α‐hydroxy‐4‐cholesten‐3‐one (C4) by 3β‐hydroxy‐∆⁵‐C₂₇‐steroid dehydrogenase (HSD3B7). Sterol 12α‐hydroxylase (CYP8B1) catalyzes 12α‐hydroxylation of C4 to 7α, 12α‐dihydroxy‐4‐cholesten‐3‐one, which is then converted to 5β‐cholestan‐3α, 7α, 12α‐triol by aldo‐keto reductase family 1 member D1 (AKR1D1). CYP8B1 is required for CA synthesis. Without CYP8B1, C4 is converted to 5β‐cholestan‐3α, 7α‐diol for CDCA synthesis. Mitochondrial sterol 27‐hydroxylase (CYP27A1) catalyzes the steroid side chain oxidation of 5β‐cholestan‐3α, 7α, 12α‐triol, which is subsequently converted to CA. The alternative BAs pathway in the liver is initiated by CYP27A1, which converts cholesterol to 27‐hydroxycholesterol and then to 3β‐hydroxy‐5‐cholestenoic acid. A nonspecific oxysterol 7α‐hydroxylase (CYP7B1) hydroxylates cholestenoic acid to 7α‐hydroxy‐3‐oxo‐4‐cholestanoic acid. In mouse liver, most CDCA is converted to α‐muricholic acid (α MCA) by a species‐specific sterol 6β‐hydroxylase (CYP2C70), and then the 7α‐OH group in α MCA is isomerized to a 7β‐OH group to form β MCA. MCAs are the primary BAs synthesized in mouse liver. The 7α‐OH group in CDCA can be isomerized to 7β‐OH to form ursodeoxycholic acid (UDCA). Primary unconjugated BAs are subsequently conjugated to glycine or taurine by BA: CoA synthetase (BACS) and BA‐CoA: amino acid N‐acyltransferase (BAAT) (A). Real‐time polymerase chain reaction (PCR) results of BA‐ biosynthesis‐related genes in liver tissue (CYP27A1, CYP7A1, CYP7B1, CYP8B1, CYP2C70, HSD3B1, AKR1D1, BAAT, and BACS) (B). Real‐time PCR results of intestinal reabsorption‐related genes in ileum tissue [organic solute transporter‐α (OST‐α), OST‐β and apical sodium dependent BA transporter (ASBT)] (C). A one‐way analysis of variance followed by Bonferroni's post hoc test was used to evaluate the significance of differences between the indicated groups; n = 12 (B and C). *p < .05, **p < .01.

FIGURE 3

Gut microbiota features of fasting and refeeding. Principal coordinate analysis (PCoA) based on Bray–Curtis dissimilarity at the operational taxonomic units (OTU) level (A). α‐diversity (Shannon index) at the OTU level (B). Venn diagram for each OTU to compare the richness shared among three groups (C). The linear discriminant analysis effect size (LEfSe) results on the hierarchy induced by the taxa. Only significantly changed taxa names with the corresponding color are shown. The figure presents results with a linear determinant analysis (LDA) score more than 2 and a p value less than .05 (D). The relative abundance of gut microbiota at genus level. Genera that took up <1% of the microbiota were labeled together as “others” (E). Relative abundances of top 15 microbial genera that showed significant differences among three groups. A one‐way analysis of variance followed by Bonferroni's post hoc test was used to evaluate the significance of differences between the indicated groups (F). *p < .05, **p < .01, ***p < .001.

FIGURE 4

Associations of bile acid and glucose with gut microbiota. Spearman's correlations of the bile acids concentration in plasma (A), feces (C), and liver tissue (E) with the relative abundance of top 20 microbial genera. The gradient colors represent the correlation coefficients, with red color being positive and blue color indicating negative. *p < .05, **p < .01, ***p < .001 (Spearman's correlation after the post hoc correction using the false discovery rate method). Redundancy analysis/canonical correlation (RDA/CCA) analysis of 16S rRNA gene sequencing data (symbols), bile acids (arrows) and glucose (arrow) in plasma (B), feces (D), and liver tissues (F). The values of axes 1 and 2 are the percentages explained by the corresponding axis. *p < .05, **p < .01, ***p < .001. PBA, primary bile acid; SBA, secondary bile acid; Con‐, conjugated; Uncon‐, unconjugated.