Berberine Directly Affects the Gut Microbiota to Promote Intestinal Farnesoid X Receptor Activation

Abstract

Intestinal bacteria play an important role in bile acid metabolism and in the regulation of multiple host metabolic pathways (e.g., lipid and glucose homeostasis) through modulation of intestinal farnesoid X receptor (FXR) activity. Here, we examined the effect of berberine (BBR), a natural plant alkaloid, on intestinal bacteria using in vitro and in vivo models. In vivo, the metabolomic response and changes in mouse intestinal bacterial communities treated with BBR (100 mg/kg) for 5 days were assessed using NMR- and mass spectrometry-based metabolomics coupled with multivariate data analysis. Short-term BBR exposure altered intestinal bacteria by reducing Clostridium cluster XIVa and IV and their bile salt hydrolase (BSH) activity, which resulted in the accumulation of taurocholic acid (TCA). The accumulation of TCA was associated with activation of intestinal FXR, which can mediate bile acid, lipid, and glucose metabolism. In vitro, isolated mouse cecal bacteria were incubated with three doses of BBR (0.1, 1, and 10 mg/ml) for 4 hours in an anaerobic chamber. NMR-based metabolomics combined with flow cytometry was used to evaluate the direct physiologic and metabolic effect of BBR on the bacteria. In vitro, BBR exposure not only altered bacterial physiology but also changed bacterial community composition and function, especially reducing BSH-expressing bacteria like Clostridium spp. These data suggest that BBR directly affects bacteria to alter bile acid metabolism and activate FXR signaling. These data provide new insights into the link between intestinal bacteria, nuclear receptor signaling, and xenobiotics.

Fig. 1.

Scheme for determining the effects of BBR in vivo (A) and vitro (B). LC, liquid chromatography; MS, mass spectrometry; RT, real time.

Fig. 2.

Quantification of bile acids in the feces (A), liver (B), ileum (C), and total (D) from mice after 5 days vehicle or BBR treatment. (E) qPCR analysis of mRNA encoding bile acid synthesis in the liver from mice after 5 days vehicle or BBR treatment. Values are medians and interquartile ranges (n = 6 per group). *P < 0.05; **P < 0.01. Hyocholic acid, ursocholanic acid, lithocholenic acid, isolithocholenic acid, allolithocholenic acid, hyodexycholic acid, isodexycholic acid, tauro-ursocholanic acid, 3,7,12 tauro-dehydrocholic acid, TαMCA, TLCA, and THCA were measured that under detection limit in this study. BA, bile acid; CA, cholic acid; CDCA, chenodeoxycholic acid; DCA, deoxycholic acid; HCA, hyocholic acid; LCA, lithocholic acid; MCA, muricholic acid; T, taurine-conjugated species; TαMCA, tauro-α-muricholic acid; TDCA, taurodeoxycholic acid; THCA, taurohyocholic acid; THDCA, taurohyodeoxycholic acid; TLCA, taurolithocholic acid; TUCA, tauroursocholanic acid; UDCA, ursodeoxycholic acid.

Fig. 3.

(A) qPCR analysis of Fgf15, Fxr, and Shp mRNAs in the ileum and Fxr and Shp mRNA expression in the liver from mice after 5 days of vehicle or BBR treatment. (B) qPCR analysis of mRNA encoding bile acid conjugation and transporters in the liver and ileum from mice after 5 days of vehicle or BBR treatment. Values are medians and interquartile ranges (n = 6 per group). *P < 0.05. Baat, bile acid-CoA: amino acid N-acyltransferase; Bacs, bile acid-CoA synthetase; Cdo, cysteine dioxygenase; Csad, cysteine sulfinic acid decarboxylase; Hnf4α1, hepatic nuclear factor 4α1; Ibabp, ileal bile acid-binding protein; Ibat, ileal bile acid transporter; Mrp, multidrug resistance-associated protein; Ntcp, Na+/taurocholate cotransporter Oatp, organic anion transporting protein; Ost, organic solute transporter.

Fig. 4.

Luciferase assays of the activation of FXR by 100 μM TCA, TCDCA, TUDCA, TβMCA, and BBR. Values are means ± S.D. (n = 3 per group). *P < 0.05, ****P < 0.0001. RLU, relative light unit.

Fig. 5.

(A) qPCR analysis of Firmicutes/Bacteroidetes, Clostridium XIVa, and Clostridium IV in the cecal contents of mice after 5 days of vehicle or BBR treatment. (B) Fecal BSH enzyme activity after BBR 5 days of vehicle or BBR treatment. (C) Orthogonal projection to latent structure-discriminant analysis scores plot (left) and coefficient plot (right) derived from ¹H NMR spectra of urine samples from mice after 5 days of vehicle (squares) or BBR (circles) treatment. The model was evaluated with cross-validated analysis of variance (P = 0.021). (D) Relative abundance of urine bacterial metabolites measured by ¹H NMR data from mice after 0, 1, 3, and 5 days of BBR treatment. Values are medians and interquartile ranges (n = 6 per group). *P < 0.05, **P < 0.01, ***P < 0.001. DMA, dimethylamine; PAG, phenylacetylglycine; TMA, trimethylamine.

Fig. 6.

Flow cytometric analyses of proportions of SYBR Green I, LNA, and HNA (A) and PI, DiBAC₄, and CFDA (B) cells from cecal bacteria with vehicle or three doses of BBR treatment. Values are means ± S.D. (n = 5 per group). *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. LNA, low nucleic acid; HNA, high nucleic acid.

Fig. 7.

qPCR analysis of Firmicutes/Bacteroidetes, Clostridium XIVa, and Clostridium IV (A) and bacterially produced butyryl-CoA:acetate CoA-transferase (but) (B) from cecal bacteria with vehicle or three doses BBR treatment. (C) Acetate, propionate, butyrate, and glucose levels in the cecal bacteria with vehicle or three doses of BBR treatment by ¹H NMR analysis. Values are means ± S.D. (n = 5 per group, flow cytometry; n = 6 per group, qPCR and NMR). *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001.

Fig. 8.

Remodeling of the gut microbiota by BBR leads to a change in the composition of bile acids and associated activation of FXR signaling. The metabolite, bacteria, enzyme, or mRNA in red or blue represents a higher or lower level in the intestines obtained from BBR-treated mice compared with vehicle.