Involvement of sphingosine 1-phosphate (SIP)/S1P3 signaling in cholestasis-induced liver fibrosis

Abstract

Bioactive sphingosine 1-phosphate (S1P) and S1P receptors (S1PRs) have been implicated in many critical cellular events, including inflammation, cancer, and angiogenesis. However, the role of S1P/S1PR signaling in the pathogenesis of liver fibrosis has not been well documented. In this study, we found that S1P levels and S1P(3) receptor expression in liver tissue were markedly up-regulated in a mouse model of cholestasis-induced liver fibrosis. In addition, the S1P(3) receptor was also expressed in green fluorescent protein transgenic bone marrow (BM)-derived cells found in the damaged liver of transplanted chimeric mice that underwent bile duct ligation. Silencing of S1P(3) expression significantly inhibited S1P-induced BM cell migration in vitro. Furthermore, a selective S1P(3) receptor antagonist, suramin, markedly reduced the number of BM-derived cells during cholestasis. Interestingly, suramin administration clearly ameliorated bile duct ligation-induced hepatic fibrosis, as demonstrated by attenuated deposition of collagen type I and III, reduced smooth muscle alpha-actin expression, and decreased total hydroxyproline content. In conclusion, our data suggest that S1P/S1P(3) signaling plays an important role in cholestasis-induced liver fibrosis through mediating the homing of BM cells. Modulation of S1PR activity may therefore represent a new antifibrotic strategy.

Figure 1

Activation of the S1P system after BDL-induced liver injury. Mice were subjected to sham or BDL operation and were sacrificed after 2 weeks (n = 10 per group). S1P levels in liver tissue (A), serum (B), and BM (F) from sham- and BDL-operated mice, were measured by high-performance liquid chromatography. Relative hepatic levels of SphK1 (C), S1P phosphatase (D), and S1P lyase mRNA (E) were measured by real-time RT-PCR. Data are presented as the mean ± SEM. *P < 0.05, compared with sham-operated mice.

Figure 2

Expression of S1P receptors in liver tissue after BDL. Two weeks after BDL, liver tissue was collected. Real-time RT-PCR and Western blot analysis for expression of S1P receptors in sham- or BDL-operated livers were performed (n = 10 per group). A: S1P_1-3 receptor mRNA expression. B: Western blot analysis of S1P₃ receptor protein level. Typical autoradiograms are shown. After quantification of the signals, results were normalized relative to glyceraldehyde-3-phosphate dehydrogenase (GAPDH) expression. Data are presented as the mean ± SEM. *P < 0.05, compared with sham-operated mice.

Figure 3

Expression of S1P₃ receptor in BM-derived cells. A and B: Two weeks (2W) after BDL, immunostaining for α-SMA in sham- or BDL-operated livers representative of 10 independent samples. Insets: Respective negative control staining. C: Confocal microscopy image of livers from mice that had received BM transplants (n = 5). D–F: Representative images of immunofluorescence analysis by confocal microscopy to track the expression of S1P₃ receptor (red) in BM-derived cells (green). Arrows indicate the cells positive for both S1P₃ receptor and EGFP. DAPI, 4,6-diamidino-2-phenylindole.

Figure 4

S1P-induced migration of BM cells via the S1P₃ receptor. A: Knockdown of S1P₃ mRNA in BM cells by S1P₃-siRNA transfection, measured by real-time RT-PCR. B: Effect of silencing S1P₃ expression on BM cell migration in response to S1P. The migration index was defined as the number of cells in the lower chamber under the tested condition divided by the number of cells in the control. Data are presented as the mean ± SEM *P < 0.05 compared with control siRNA. BSA, bovine serum albumin.

Figure 5

The effect of suramin on the homing of BM cells in mice during cholestasis. Mice (n = 6 per group) received sham or BDL operation with or without one-time suramin administration (20 mg/kg b.wt. i.p.). Three days later, mice were sacrificed and liver tissue was harvested. A and B: Representative H&E-stained liver sections (original magnification, ×200). Necrotic areas are indicated by arrows. C: Quantification of necrotic areas. At least 10 view fields per animal were included. Mice (n = 5 per group) were lethally irradiated and received transplants of EGFP-positive whole BM cells. Then liver fibrosis was induced by BDL over 2 weeks, simultaneously with or without suramin administration (20 mg/kg, twice per week, i.p.). Shown are representative fluorescent images of liver sections with (E and G) or without suramin administration (D and F). Scale bars = 40 μm. H: The average number of BM-derived cells was quantified by analyzing at least 10 random fields per animal, with Image-Pro Plus software. The quantitative result is the percentage of the number of BM-derived cells in BDL mice without suramin administration. Data are presented as the mean ± SEM. *P < 0.05, compared with control.

Figure 6

Antifibrotic effect of suramin on BDL-induced liver fibrosis in mice. After lethal irradiation and transplantation of EGFP-positive BM cells, mice (n = 5 per group) underwent sham or BDL operation with or without suramin administration. A: Expression of procollagen α1(I) (Col α1(I)), procollagen α1(III) (Col α1(III)), and α-SMA mRNA levels in liver, measured by real-time RT-PCR. B: Hydroxyproline content in the liver. C: Representative pictures of Sirius red staining (original magnification, ×100) in the BDL- or sham-operated livers. Inset: Sirius red staining for sham-operated livers. D: Quantitative analysis of liver fibrosis. Ten randomly selected fields were quantitated for each mouse using the Leica QWin V3 software. Data are presented as the mean ± SEM. *P < 0.05. compared with BDL-operated mice without suramin administration.