Complement Inhibition Alleviates Cholestatic Liver Injury Through Mediating Macrophage Infiltration and Function in Mice

Abstract

Background and aims: Cholestatic liver injury (CLI), which is associated with inflammatory reactions and oxidative stress, is a serious risk factor for postoperative complications. Complement system is involved in a wide range of liver disorders, including cholestasis. The present study assessed the role of complement in CLI and the therapeutic effect of the site-targeted complement inhibitor CR2-Crry in CLI.

Methods: Wild-type and complement gene deficient mice underwent common bile duct ligation (BDL) to induce CLI or a sham operation, followed by treatment with CR2-Crry or GdCl3. The roles of complement in CLI and the potential therapeutic effects of CR2-Crry were investigated by biochemical analysis, flow cytometry, immunohistochemistry, ELISA, and quantitative RT-PCR.

Results: C3 deficiency and CR2-Crry significantly reduced liver injuries in mice with CLI, and also markedly decreasing the numbers of neutrophils and macrophages in the liver. C3 deficiency and CR2-Crry also significantly reduced neutrophil expression of Mac-1 and liver expression of VCAM-1. More importantly, C3 deficiency and CR2-Crry significantly inhibited M1 macrophage polarization in these mice. Intravenous injection of GdCl3 inhibited macrophage infiltration and activation in the liver. However, the liver injury increased significantly. BDL significantly increased the level of lipopolysaccharide (LPS) in portal blood, but not in peripheral blood. GdCl3 significantly increased LPS in peripheral blood, suggesting that macrophages clear portal blood LPS. Oral administration of ampicillin to in GdCl3 treated mice reduced LPS levels in portal blood and alleviated liver damage. In contrast, intraperitoneal injection LPS increased portal blood LPS and reversed the protective effect of ampicillin. Interestingly, C3 deficiency did not affect the clearance of LPS.

Conclusions: Complement is involved in CLI, perhaps mediating the infiltration and activation of neutrophils and macrophage M1 polarization in the liver. C3 deficiency and CR2-Crry significantly alleviated CLI. Inhibition of complement could preserve the protective function of macrophages in clearing LPS, suggesting that complement inhibition could be useful in treating CLI.

Keywords: CR2-Crry; cholestatic liver injury; complement system; macrophage; neutrophil.

Figure 1

Complement activation in CLI and alleviation of CLI by C3 deficiency or CR2-Crry in mice. Wild-type or C3 deficient mice were subjected to BDL or sham operations. WT-BDL mice were injected i.p. with PBS or CR2-Crry. The mice were sacrificed after 3 days and the samples were collected. Formalin-fixed, paraffin-embedded liver sections were stained with HE or immunostained with antibodies to C3d and MAC. Liver function and C3 activation were measured in serum samples. (A) HE staining to assess liver necrosis (scale bar, 100 μm). (B–E) Serum concentrations of (B) total bilirubin, (C) ALT, (D) AST, and (E) C3a. (F) Immunostaining of liver samples for C3d (scale bar, 100 μm) and MAC (scale bar, 50 μm). Results are expressed as mean ± SD for 6 samples per group. *p < 0.05, **p < 0.01; NS, not significant.

Figure 2

Contribution of the alternative pathway to the over-activation of the complement system in mice with CLI. Wild-type mice and mice deficient in the genes encoding complement components C4 and Factor B were subjected to BDL or sham operations. The mice were sacrificed after 3 days and the samples were collected. (A) Immunostaining for C1q in fresh frozen liver sections (scale bar, 100 μm). (B, C) Serum C3a concentrations in mice deficient in genes encoding (B) C4 and (C) Factor B (n = 5–8 for each sham operated group; n = 6–9 for each BDL group). Results are expressed as mean ± SD. *p < 0.05; NS, not significant.

Figure 3

C3 deficiency or CR2-Crry significantly inhibits the activation of neutrophils and decreases the expression of VCAM-1 and Mac-1 in mice with CLI. Wild-type and C3 deficient mice were subjected to BDL or sham operations. WT-BDL mice were injected i.p. with PBS or CR2-Crry and sacrificed 3 days later. Formalin-fixed, paraffin-embedded liver sections were stained for specific esterase or immunostained for MPO, 3-chlorotyrosine and VCAM-1. Anticoagulated blood was subjected to flow cytometry for Mac-1 detection. (A) Specific esterase staining for neutrophils and immunohistochemical staining for VCAM-1 in liver parenchyma. (B) Expression of Mac-1 on neutrophils. (C) Assessment of neutrophils in the liver portal area. (D) Immunohistochemical staining for MPO and 3-chlorotyrosine. Scale bars for all, 100 μm; n = 5–6 per group. Black arrows, positive staining in the liver parenchyma; red arrows, positive staining in the portal area. Results are expressed as mean ± SD. *p < 0.05, **p < 0.01; NS, not significant.

Figure 4

C3 deficiency or CR2-Crry significantly inhibits macrophage infiltration and M1 polarization in mice with CLI. Wild-type and C3 deficient mice were subjected to BDL or sham operations. WT-BDL mice were injected i.p. with PBS or CR2-Crry and sacrificed 3 days later. (A) F4/80 immunostaining (n = 6 per group; scale bars, 100 μm). (B) Immunofluorescence test for M1 macrophages (n = 3 per group; scale bars, 50 μm). (C) Flow cytometry analysis for M1 macrophages (n = 6 per group). (D–G) qRT-PCR measurements of (D) TNF-α, (E) IL-6, (F) iNOS and (G) IL-1β mRNA levels (n = 5-6 per group). Results are expressed as mean ± SD. *p < 0.05, **p < 0.01.

Figure 5

GdCl3 inhibition of macrophages exacerbates CLI in mice. Wild-type and C3 deficient mice were subjected to BDL or sham operations. The mice were injected i.v. with PBS or GdCl3 and sacrificed after 3 days. (A, B) F4/80 immunostaining (scale bars: 100 μm; n = 6 per group). (C, D) Serum concentrations of (C) ALT and (D) AST (n = 6 per group). (E) Assessment of liver necrosis (n = 6 per group). (F) HE staining (red boxes, central venous areas; blue boxes, portal areas; scale bars, 500 μm for upper images, 100 μm for lower images). Results are expressed as mean ± SD. *p < 0.05, **p < 0.01; NS, not significant.

Figure 6

Macrophage clearance of LPS from portal blood. Wild-type mice were subjected to BDL or sham operations and treated with GdCl3, ampicillin and/or LPS, as indicated. The mice were sacrificed 3 days later and portal blood and peripheral blood samples were collected separately. (A) Concentrations of LPS in portal and peripheral blood. (B, C) Effects of ampicillin and GdCl3 on liver function. (D) Effects of exogenous LPS injection on LPS concentration in portal blood. (E, F) Effects of LPS injection on liver function (n = 6 per group). Results are expressed as mean ± SD. *p < 0.05, **p < 0.01; NS, not significant.

Figure 7

C3 deficiency does not affect phagocytosis and clearance of LPS in mice with CLI. Wild-type (injected i.v. with GdCl3 or PBS) and C3 deficient mice were subjected to BDL or sham operations or injected i.p with LPS. The mice were injected i.v. with India ink 3 days later immediately before sacrifice. Formalin-fixed, paraffin-embedded liver sections were immunostained for F4/80, and LPS was measured in serum samples. (A) Ability of macrophages to phagocytize carbon particles (scale bars, 100 μm; red circles, F4/80 positive cells containing carbon particles; n = 5 per group). (B) LPS concentrations in portal blood and peripheral blood (n =6 per group). Results are expressed as mean ± SD. ^#p < 0.01 compared with the WT-LPS or C3^−/−-LPS group; **p < 0.01, compared with the WT-BDL or C3^−/−-BDL group; NS, not significant.