Fig 1.
Rifaximin attenuated liver fibrosis.
A-B. (left) H&E and staining of Sirius Red, Collagen III in liver tissue in CCl4 (A) and BDL mice (B) treated with or without rifaximin. (right) Quantification of positive area of staining using ImageJ (n = 5 mice per group). Scale bar: 400 μm. C. Serum PIIINP concentration in CCl4 (left) and BDL (right) mice treated with or without rifaximin, as measured by ELISA (n = 5 mice per group). D. Hydroxyproline content in liver tissues from CCl4 and BDL mice treated with or without rifaximin (n = 5 mice per group). E. Western blotting images and quantification of α-SMA in liver from CCl4 and BDL mice treated with or without rifaximin. GAPDH was used as a reference protein. (n = 4-5 mice per group). F. The relative mRNA expression of α-SMA in liver tissue from CCl4 and BDL mice treated with or without rifaximin (n = 5 mice per group). G. Spleen/body weight ratio in CCl4 (left) and BDL (right) mice treated with or without rifaximin (n = 5 mice per group). *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 (one-way ANOVA with Tukey’s post-hoc test). H&E, hematoxylin and eosin; PIIINP, procollagen III N-terminal propeptide; ELISA, the enzyme-linked immunosorbent assay; HYP, hydroxyproline.
Fig 2.
Rifaximin alleviated LSEC dysfunction.
A. Immunocytochemistry staining of CD31 and LYVE-1 in murine primary LSECs. Scale bar: 100 μm. B. The relative mRNA expression of eNOS in primary LSECs isolated from control (sham), CCl4 (BDL), and rifaximin treated CCl4 (BDL) mice. C-D. Western blotting images and quantification of eNOS and p-eNOS in primary LSECs isolated from CCl4 (C) and BDL mice(D). GAPDH was used as a reference protein. E-F. (left)The staining of LYVE1 and CD31 in liver tissue in CCl4 (E) and BDL mice(F) treated with or without rifaximin. (right) Quantification of positive area of staining using ImageJ (n = 5 mice per group). Scale bar: 50 μm. G. The relative mRNA expression of LYVE1 and VWF in primary LSECs isolated from control (sham), CCl4 (BDL), and rifaximin treated CCl4 (BDL) mice. All data are presented as mean ± SD, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 (one-way ANOVA with Tukey’s post-hoc test). LSECs, liver sinusoidal endothelial cells.
Fig 3.
Rifaximin induced alterations in the gut microbiota of mice.
A. Richness of the gut microbiota in mice, assessed using the Ace and Chao indices. (n = 5 mice per group). B. The β-diversity of gut microbiota in control and CCl4 mice treated with or without rifaximin, as indicated by PCoA and NMDS analysis. (n = 5 mice per group). C. The component of gut microbiota at the genus level in control and CCl4 mice treated with or without rifaximin. (n = 5 mice per group). D-E. Significant differences in the proportion of gut microbiota at the genus level (D) and species level (E) among the three groups. (n = 5 mice per group). *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 (one-way ANOVA with Tukey’s post-hoc test).
Fig 4.
LPS induced LSEC dysfunction by inhibiting eNOS mRNA expression.
A. Serum LPS concentration in CCl4 and BDL mice treated with or without rifaximin, as measured by ELISA (n = 5 mice per group). B. NO concentration in the culture medium of LSECs treated with LPS in different concentrations (100, 200, and 400 ng/mL). C. Western blotting images and quantification of eNOS and p-eNOS in primary LSECs treated with LPS (100 ng/mL) for 24h or 48h. GAPDH was used as a reference protein. D. Western blotting images of eNOS and p-eNOS in primary LSECs treated with LPS at different concentrations (100, 200, and 400 ng/mL) for 48h. GAPDH was used as a reference protein. E. The mRNA expression of eNOS in primary LSECs treated with LPS at different time points (24h or 48h) or at different concentrations (100, 200, and 400 ng/mL). All data are presented as mean ± SD, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 (one-way ANOVA with Tukey’s post-hoc test in A, B, E(right); unpaired Student’s t-test in C,E(left)). NO, nitric oxide; LSECs, liver sinusoidal endothelial cells; LPS, lipopolysaccharide; ELISA, the enzyme-linked immunosorbent assay.
Fig 5.
LPS-induced LSEC dysfunction relied on TLR4 signaling.
A. Western blotting images of eNOS, p-eNOS, TLR4, MyD88, p-NF-κB, and NF-κB in primary LSECs incubated with or without TAK-242 for 2 h, followed by LPS treatment for 48 h. B. Immunocytochemistry staining of NF-κB in primary LSECs incubated with or without TAK-242 for 2 h, followed by LPS treatment for 48h. Scale bar:30 μm. C. The mRNA expression of eNOS in primary LSECs incubated with or without TAK-242 for 2h, followed by LPS treatment for 48h. All data are presented as mean ± SD, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 (one-way ANOVA with Tukey’s post-hoc test). D. Western blotting images of TLR4, p-NF-κB, and NF-κB in primary LSECs isolated from CCl4 and BDL mice. GAPDH was used as a reference protein. E. The graphical abstract of this study. The intestinal microbiota dysbiosis allows more LPS to enter the portal circulation, which exacerbates LSEC dysfunction and liver fibrosis. LPS aggravates LSEC dysfunction by reducing eNOS expression which relies on TLR4 signaling. Intestinal decontamination with rifaximin improves LSEC function and alleviates liver fibrosis, a process linked to the reconstruction of the gut microbiome and a reduction in gut-derived LPS. LSECs, liver sinusoidal endothelial cells; LPS, lipopolysaccharide.