Gut microbiota depletion aggravates bile acid-induced liver pathology in mice with a human-like bile acid composition

Abstract

Cyp2c70-deficient mice have a human-like bile acid (BA) composition due to their inability to convert chenodeoxycholic acid (CDCA) into rodent-specific muricholic acids (MCAs). However, the hydrophobic BA composition in these animals is associated with liver pathology. Although Cyp2c70-ablation has been shown to alter gut microbiome composition, the impact of gut bacteria on liver pathology in Cyp2c70-/- mice remains to be established. Therefore, we treated young-adult male and female wild-type (WT) and Cyp2c70-/- mice with antibiotics (AB) with broad specificity to deplete the gut microbiota and assessed the consequences on BA metabolism and liver pathology. Female Cyp2c70-/- mice did not tolerate AB treatment, necessitating premature termination of the experiment. Male Cyp2c70-/- mice did tolerate AB but showed markedly augmented liver pathology after 6 weeks of treatment. Dramatic downregulation of hepatic Cyp8b1 expression (-99%) caused a reduction in the proportions of 12α-hydroxylated BAs in the circulating BA pools of AB-treated male Cyp2c70-/- mice. Interestingly, the resulting increased BA hydrophobicity strongly correlated with various indicators of liver pathology. Moreover, genetic inactivation of Cyp8b1 in livers of male Cyp2c70-/- mice increased liver pathology, while addition of ursodeoxycholic acid to the diet prevented weight loss and liver pathology in AB-treated female Cyp2c70-/- mice. In conclusion, depletion of gut microbiota in Cyp2c70-/- mice aggravates liver pathology at least in part by increasing the hydrophobicity of the circulating BA pool. These findings highlight that the potential implications of AB administration to cholestatic patients should be evaluated in a systematic manner.

Keywords: Cyp2c70; bile acids; gut microbiota; hepatobiliary disease; hydrophobicity; liver fibrosis.

Figure 1. Antibiotics treatment is not tolerated by female Cyp2c70^−/− mice but effectively prevents microbial bile acid deconjugation and synthesis of secondary bile acids in male mice

(A) Study design: male (n = 8–9 per group) and female (n = 4–6 per group) WT and Cyp2c70^−/− mice were fed a regular chow diet with or without AB. (B) Survival of female WT and Cyp2c70^−/− mice over time. (C) Unconjugated BAs in plasma of male mice. (D) Percentage of primary (Prim) and secondary (Sec) BAs as well as UDCA in plasma of male mice. Primary BAs: cholic acid, chenodeoxycholic acid, α/β-muricholic acid and conjugates. Secondary BAs: deoxycholic acid, lithocholic acid, hyo(deoxy)cholic acid, ω-muricholic acid and conjugates. Relative amounts of (E) total bacteria and (F) Clostridium XIVa in caecum as quantified by qPCR for 16S DNA. **P<0.01, ***P<0.001 according to Kruskal–Wallis H test followed by Conover post-hoc comparisons. AB, antibiotics; BA, bile acid.

Figure 2. Microbiota depletion induces profound alterations in bile acid 12α-hydroxylation in male Cyp2c70^−/− mice

(A) Total BAs in plasma. (B,C) Relative mRNA expression of hepatic BA transporters. (D) Total BA concentrations in gallbladder bile. (E) Relative hepatic mRNA expression of genes involved in BA synthesis. (F) Fecal total BA excretion. (G) BA profiles in gallbladder bile. (H) 12α-/non-12α-OH BA ratios in gallbladder bile. (I) Hydrophobicity index of biliary BAs. *P<0.05, **P<0.01, ***P<0.001 according to Kruskal–Wallis H-test followed by Conover post-hoc comparisons. AB, antibiotics; BA, bile acid.

Figure 3. Microbiota depletion aggravates liver pathology in male Cyp2c70^−/− mice

(A,B) Body weights and relative liver weights. (C) Plasma transaminase and (D) alkaline phosphatase (ALP) levels. (E) Representative images of liver sections stained with hematoxylin and eosin (H&E), Sirius Red, and anti-CK19 (bars represent 200 μm). (F,G) Relative hepatic mRNA expression of genes involved in inflammation and fibrosis. (H,I) Quantification of collagen and CK19 staining, expressed as percentage positively stained area per image field. *P<0.05, **P<0.01, ***P<0.001 according to Kruskal–Wallis H-test followed by Conover post hoc comparisons. AB, antibiotics; CK19, cytokeratin-19.

Figure 4. The hydrophobicity index of biliary bile acids is strongly associated with liver pathology, while Cyp8b1 mRNA expression levels negatively correlate with indicators of liver pathology

Spearman's rank-order correlation coefficients (r) were used to assess correlations between variables. Dots are displayed only for correlations with p<0.001. BW, body weight; HI, hydrophobicity index of biliary bile acids; LW, liver weight.

Figure 5. Ablation of hepatic Cyp8b1 expression promotes liver pathology in male Cyp2c70^−/− mice

(A) Study design: male Cyp2c70^−/−/L-cas9_tg mice (n = 6 per group) were injected with AAV-sgCTRL or AAV-sgCyp8b1 and maintained on a regular chow diet without additions for 8 weeks. (B) Relative hepatic Cyp8b1 mRNA expression. (C) 12α-/non-12α-OH BA ratios in gallbladder bile. (D) Hydrophobicity index of biliary BAs. (E,F) Body weights and relative liver weights. (G-I) Plasma transaminase and alkaline phosphatase (ALP) levels. (J) Representative images of liver sections stained with hematoxylin and eosin (H&E), Sirius Red, and anti-CK19 (bars represent 200 μm). (K,L) Relative hepatic mRNA expression of genes involved in inflammation and fibrosis. (M,N) Quantification of collagen and CK19 staining, expressed as percentage positively stained area per image field. *P<0.05, **P<0.01 as determined by Mann–Whitney U-test. BA, bile acid; CK19, cytokeratin-19; Cyp8b1-KD, Cyp8b1 knockdown.

Figure 6. Increased hydrophilicity of the bile acid pool prevents aggravation of liver pathology upon depletion of microbiota in female Cyp2c70^−/− mice

(A) Study design: female Cyp2c70^−/− mice were fed a UDCA-containing diet for 2 weeks prior to the onset of AB treatment, and were then fed a diet with AB only or with AB+UDCA for 6 additional weeks (n = 8 mice per group). (B) Survival of female Cyp2c70^−/− mice over time. (C) Hydrophobicity index of biliary BAs. (D) Total BAs in plasma. (E,F) Body weights and relative liver weights. (G–I) Plasma transaminase and alkaline phosphatase (ALP) levels. (J) Representative images of liver sections stained with hematoxylin and eosin (H&E), Sirius Red, and anti-CK19 (bars represent 200 μm). (K,L) Relative hepatic mRNA expression of genes involved in inflammation and fibrosis. (M,N) Quantification of collagen and CK19 staining, expressed as percentage positively stained area per image field. *P<0.05, **P<0.01 as determined by Mann–Whitney U-test. AB, antibiotics; BA, bile acid; CK19, cytokeratin-19.