Blockade of C5aR1 alleviates liver inflammation and fibrosis in a mouse model of NASH by regulating TLR4 signaling and macrophage polarization

C5^−/− (B10.D2-Hc⁰H2^dH2-T18^c/0SnJ) mice and their haplotype (B10.D2-Hc¹H2^dH2-T18^c/nSnJ) were purchased from the Jackson Laboratory (Bar Harbor, ME, USA). C5aR1^−/− mice with a C57BL/6 background were purchased from GemPharmatech (Jiangsu, China). All mice were maintained in a specific pathogen-free facility at Guangxi Medical University. A Western diet (WD) and chemically induced murine NASH model as described previously [22] were used. Briefly, male mice were fed normal chow diet/corn oil (ND + Oil), WD/corn oil (WD + Oil) or WD combined with low weekly dose of intraperitoneal carbon tetrachloride (CCl₄) for 12 weeks. Male C5^−/− and their haplotype, C5aR1^−/− mice and C57BL/6 wild-type mice, aged 8–12 weeks, were fed a WD/CCl₄ (WD + CCl₄) for 12 weeks. Corn oil or CCl₄ were intraperitoneally injected into male mice once a week. All animal experiments were approved by the Animal Care and Use Committee of Guangxi Medical University, Guangxi, China.

Fresh liver tissues were frozen, and sliced into 8 μm thick sections. The sections were stained with an Oil Red O staining. Liver biopsy specimens were also fixed with 10% neutral formalin and embedded in paraffin, and then tissue sections (5 μm thick) were stained with hematoxylin–eosin (H&E) and Sirius Red. Histopathological changes in the liver biopsies were assessed using a NanoZoomer S60 digital slide scanner (Hamamatsu, Japan).

Paraffin-embedded sections (4 μm thick) were processed for immune-histochemical staining. Antigen retrieval was performed by pressure cooking for 5 min in citrate buffer (pH 6), followed by peroxidase and serum blocking steps. The sections were incubated with goat anti-C3d antibody (R&D Systems, 1:100 dilution), anti-F4/80 (CST, 1:500 dilution) or anti-α-SMA (Abcam, 1:500 dilution) for 2 h at room temperature, followed by antibody detection with an anti-goat ImmPRESS kit (Vector Laboratories). The images were collected using the NanoZoomer S60 digital slide scanner.

Following killing, retro-orbital blood of the experimental mice was collected under isoflurane anesthesia to obtain serum for analysis. Serum alanine aminotransferase (ALT) and aspartate aminotransferase (AST) levels were measured with an autoanalyzer (ANTECH Diagnostics, Los Angeles, CA, USA). Serum C5a level was measured using commercially available ELISA kits (CUSABIO, Wuhan, China), according to the manufacturer’s instructions.

The levels of TG, MDA, and GSH in liver tissue homogenates from each group were measured using the corresponding kits (Catalog# A110-1, A003-1 and A006-2, respectively) from Nanjing Jiancheng Bioengineering Institute (Nanjing, China) according to the manufacturer’s protocols.

Proteins of hepatic samples or cells were analyzed using standard western blotting techniques. The antibodies of anti-α-SMA (ab32575, Abcam), anti-TGF-β1 (SAB4502954, Sigma), anti-P65 (10,745–1-AP, Proteintech), anti-p-P65 (AF2006, Affinity), anti-iNOS (13120S, CST), anti-CD86 (19,589, CST), anti-CD163 (16,646–1-AP, Proteintech), anti-CD206 (24,595, CST), anti-TLR4 (19,811–1-AP, Proteintech), anti-NLRP3 (A5652, ABclonal), anti-AKT (10,176–2-AP, Proteintech), anti-p-AKT (4060S, CST), anti-HO-1 (43,966, CST), anti-SOD1 (10,269–1-AP, Proteintecch), anti-MDA (MDA11-S, ADI), or anti-CASP1 (24232S, CST) were used as primary antibodies. Anti-GAPDH (10,494–1-AP, Proteintech) or anti-TUBULIN (#2144, CST) was used to normalize the signals. Bands were quantified by ImageJ software.

Total RNA was extracted from 0.1 g frozen hepatic tissues according to the TRIzol reagent protocol (No. 15596–026, Invitrogen). Next, 2 μg total RNA was reverse transcribed to complementary DNA (cDNA) using the RevertAid First Strand cDNA Synthesis Kit (No.K1622, Thermo Scientific) according to the manufacturer’s protocol. Relative mRNA levels of genes were measured by qRT-PCR, using a SYBR Green PCR master mix (No.1725125, Bio-Rad). All experiments were performed in triplicate. The qRT-PCR primers used in this study were shown in supplementary Table 1. GAPDH was used to normalize the signals of target gene in the same sample.

Total RNA was extracted from flash-frozen liver tissues of C5aR1^−/− and WT NASH mice with TRIzol reagent (Invitrogen, Carlsbad, CA, USA). The quality of the RNA samples was evaluated with a NanoDrop 2000c spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA) and an Agilent’s Bioanalyzer. Sequencing libraries were generated by reverse transcription-polymerase chain reaction (RT-PCR) amplification. The PCR products were sequenced on a HiSeq 2500 sequencing system (RIBOBIO, Guangzhou, China). Transcriptional profiling data were deposited at Mendeley Data (https://​data.​mendeley.​com), which can easily be accessed at http://​dx.​doi.​org/​10.​17632/​j8v2c8z7zt.​1

RAW264.7 cells were kindly provided by Kunming Cell Bank of Typical Culture Preservation Committee of Chinese Academy of Sciences, Kunming, China. RAW264.7 cells were cultured in Dulbecco’s modified Eagle medium (C11995500BT, Gibico) supplemented with 10% fetal bovine serum. The cells were grown at 37 °C in an atmosphere of 5% CO₂. The RAW264.7 cells were cultured overnight to about 80% confluent and treated by the recombinant protein C5a (HY-P7695, MCE) with 50 ng/mL for 24 h.

The antagonist PMX-53 was purchased from Nanjing Peptide Industry (Nanjing, China) and diluted in saline. PMX-53 (1 mg/kg) was administered i.p. to the experimental group and saline administration served as the control for the last 4 weeks of the experimental protocol. Mice subject to NASH model were treated with PMX-53 every other day from week eight forward.

Data are shown as the mean ± standard deviation. Significant differences between groups were determined by ANOVA, with a Bonferroni correction for continuous variables and multiple groups. Two-tailed Student’s t test was used for the comparison of a normally distributed continuous variable between two groups. Nonparametric statistical analysis was performed using the Mann–Whitney test. Statistical analysis was performed using GraphPad Prism (Version 7.0.4) software. P values less than 0.05 were considered as statistically significant.

To investigate whether the complement system was activated in NASH, mice were subjected to a normal diet/corn oil (ND + Oil), Western diet/corn oil (WD + Oil) and WD combined with a low weekly dose of intraperitoneal CCl₄ (WD + CCl₄) as described previously [22], resulting in a rapid progression to NASH with fibrosis. The results of H&E and Sirius Red staining showed that hepatic fibrosis was not induced in mice treated with WD + Oil, but increased in mice treated with WD + CCl₄ compared to ND + Oil group (Fig. 1A). Oil Red O staining indicated that hepatic lipid droplets were accumulated in the groups of WD + Oil and WD + CCl₄ compared with ND + Oil group (Fig. 1A). Serum levels of alanine aminotransferase (ALT) and aspartate transaminase (AST) were slightly increased in the WD + Oil group relative to ND + Oil, and significantly raised in WD + CCl₄ group compared to ND + Oil and WD + Oil (Fig. 1B, C). Immunohistochemical staining revealed that extensive C3d was deposited in the WD + Oil and WD + CCl₄ group (Fig. 1D). Furthermore, the serum level of C5a was elevated in the WD + Oil compared with ND + Oil, and obviously increased in the WD + CCl₄ group compared with ND + Oil and WD + Oil groups (Fig. 1E). Taken together, our results demonstrate that complement level and liver injury are significantly elevated in the NASH mice.

Previous studies had shown that C5 is correlated to the progression of NAFLD [14, 15]. Thus, we sought to investigate the effects of C5 on the hepatic steatosis and inflammation in the NASH model. Compared to WT mice, the results of H&E and Oil Red O staining revealed that C5 deficiency reduced hepatic lipid droplet accumulation in the NASH mice (Fig. 2A). In addition, the level of triglyceride (TG) in liver tissue was decreased in the C5 deficient mice (Fig. 2B). The results of the qRT-PCR analysis showed that the mRNA levels of inflammation associated genes such as Tnf-α, IL-1β and F4/80 were downregulated in the liver of C5 deficient mice (Fig. 2C). Immunohistochemical analysis showed that hepatic F4/80 deposition was reduced in C5 deficient mice compared to WT mice (Fig. 2A). Moreover, serum levels of ALT and AST were decreased in the C5 deficient mice compared with WT mice (Fig. 2D and E). As we know, hepatic lipid accumulation leads to oxidative stress, which is correlated to the progression of inflammation. Therefore, we further examined the effect of C5 absence on the expression of oxidative stress markers. The liver homogenate level of MDA was decreased in the C5 deficient mice; in contrast, the level of GSH was increased (Fig. 2F and G). Moreover, western blot examination showed that the protein level of MDA was downregulated, in contrast, the protein levels of heme oxygenase 1 (HO-1) and superoxide dismutase 1 (SOD1) were upregulated in the liver of C5^−/− mice compared with WT mice (Fig. 2H and I). These results indicate that C5 deficiency decreases the oxidative stress and ameliorates hepatic steatosis and inflammation in NASH mice.

A previous study reported that C5 has a causal role in liver fibrogenesis [16], but its role in the hepatic fibrosis of chemically induced NASH remains unclear. To explore the effects of C5 on the hepatic fibrosis in the NASH model, wild-type (WT) and C5^−/− mice were fed WD and injected with low-dose CCl₄ for 12 weeks. Then, analysis of hepatic fibrosis was performed by Sirius Red staining and immunohistochemical staining of α-smooth muscle actin (α-SMA). We found that compared to the WT mice, C5 deficiency resulted in a reduction of liver fibrosis in the NASH model (Fig. 3A). Immunohistochemical staining analysis showed that C5 deficient mice had decreased staining of α-SMA compared to WT (Fig. 3A). In addition, staining of Ki67 showed that loss of C5 inhibited cell proliferation (Fig. 3A). Again, C5 deficiency diminished the expression of several pro-fibrotic associated genes, such as α-Sma and Tgf-β1 (Fig. 3B). We further examined the protein level of α-SMA, a marker of activated hepatic stellate cells (HSCs), and TGF-β1 in the liver tissue using western blot. The protein levels of α-SMA and TGF-β1 were clearly decreased in the C5 deficient mice (Fig. 3C–F). These data demonstrate that C5 deficiency protects against the progression of liver fibrosis in NASH mice.

While C5a/C5aR1 interactions have been identified as an important driver of inflammation and fibrosis, their function in NASH progression in mice has not been completely elucidated. Therefore, C5aR1^−/− mice were maintained on the NASH model of WD + CCl₄ to functionally validate the effect of C5aR1 on the development of NASH. Histopathology and Sirius Red staining revealed that C5aR1 deficiency alleviated the fibrosis level in NASH mice (Fig. 4A). Moreover, the expression of pro-fibrotic markers, such as α-Sma, Col1a1, and Tgf-β1 was diminished in the liver tissue of C5aR1 deleted mice (Fig. 4B). The decreased protein level of α-SMA was further confirmed by western blot (Fig. 4C). Again, the mRNA levels of inflammation associated genes Tnf-α, IL-6 and IL-1β were downregulated in C5aR1^−/− mice compared with WT mice (Fig. 4D). Immunohistochemical staining showed that the deposition of F4/80 was clearly reduced in the liver of C5aR1 deficient mice (Fig. 4E), which is consistent with the qRT-PCR result of F4/80 (Fig. 4D). In addition, Oil Red O staining analysis showed that hepatic lipid droplets were reduced in C5aR1^−/− mice compared to WT mice (Fig. 4A). C5aR1 deletion markedly decreased the expression of lipogenesis genes such as Srebf1, Acc and Fasn in liver (Fig. 4F). Taken together, we conclude that loss of C5aR1 inhibits the progression of NASH, including resistance to steatosis, inflammation, and fibrosis.

To explore the putative mechanisms of C5aR1 in NASH, transcriptional profiling of liver tissues from NASH mice was performed. There were more than 2500 differentially expressed genes in the liver tissues of C5aR1^−/− and WT mice (Fig. 5A). KEGG pathway analysis revealed that several pathways such as Toll-like receptor signaling pathway, NFκB signaling pathway, TNF signaling pathway, and NOD-like receptor signaling pathway associated with NASH was enriched between C5aR1^−/− and WT mice (Fig. 5B). We hypothesized that C5aR1 deficiency slowed the development of NASH by regulating these signaling pathways. Toll-like receptor signaling pathway associated genes were further analyzed and we found that some genes such as TLR4, CD86, and Tnf were significantly decreased in C5aR1 deleted mice (Fig. 5C). Firstly, we verified the expression of some key genes related to these pathways. The result of qRT-PCR showed that the mRNA level of TLR4 was decreased in C5aR1 deficient NASH mice (Fig. 5D), but not affected in mice fed a normal diet (Supplementary Fig. 1). Western blot confirmed that the protein level of TLR4 was downregulated in the C5aR1^−/− mice compared to WT mice (Fig. 5E). As we know, Toll-like receptor signaling pathway regulates the downstream pathway NFκB signaling pathway. Therefore, the expression levels of some key genes associated with NFκB pathway were also assessed by western blot. As expected, western blot analysis showed that the ratio of p-NFκB P65/ NFκB P65 was decreased in the C5aR1 deficient mice (Fig. 5E). NFκB signaling pathway, which mediates the expression of TNFα and IL-6, is a crucial mediator of inflammation. Our above data showed that C5aR1 deletion exactly reduced the expression of TNFα and IL-6 (Fig. 4D). Previous studies showed that TLR4 signaling is involved in mediating inflammasome activation [23, 24]. Therefore, the effect of C5aR1 on the NLRP3 signaling was assessed. Our result showed that the mRNA and protein level of NLRP3 were reduced in C5aR1 deficient mice (Fig. 5F and G). The downstream factors of NLRP3 signaling, CASP1 and IL-1β, were decreased by the C5aR1 deletion (Fig. 5G, and Fig. 4D). In addition, RAW264.7 cells were treated with the recombinant protein C5a and we found that C5a treatment significantly increased the expression of TLR4, NLRP3 and CASP1 (Fig. 5H). Taken together, these results demonstrate that the C5a–C5aR1 axis regulates the TLR4 and NLRP3 signaling pathway in NASH.

A previous study reported that C5aR1 deficiency has an effect on macrophage phenotype in the kidney following renal ischemia/reperfusion insult [20]. Several lines of evidence have shown that the polarization of macrophage has an impact on the progression of NASH [25, 26]. Moreover, Orr et al. [27] reported that TLR4 deficiency promotes the alternative activation of adipose tissue macrophage. We therefore tested the function of C5aR1 in the phenotype changes of macrophages. Our results of western blot showed that C5aR1 loss decreased the expression of iNOS and CD86, but increased the expression of CD163 and CD206 (Fig. 5I). The results of qRT-PCR revealed that the hepatic level of iNOS was downregulated; in contrast, the expression level of CD163 was upregulated in C5aR1 deficient mice (Fig. 5J). These results demonstrate that C5aR1 deficiency promotes the differentiation of macrophage M2 phenotype in NASH mice.

KEGG pathway analysis showed that C5aR1 is also involved in mediating the signaling pathway of fatty acid metabolism (Fig. 5B). The fatty acid metabolism associated genes such Srebf1, Scd1 and Fasn was downregulated in C5aR1 deficient mice. Previous studies reported that AKT signaling pathway is involved in upregulating the expression of SREBP1c [28]. Several lines of evidence showed that C5a/C5aR1 signaling activates the AKT signaling pathways [29]. We speculated that C5aR1 participates in lipogenesis by regulating the AKT signaling pathway. Our result showed that C5aR1 deficiency reduced the ratio of p-AKT/AKT (Fig. 5K). As shown in Fig. 4F, the expression of Srebf1 and Fasn was decreased in the C5aR1 deficient mice. Taken together, C5aR1 deficiency prevents the progression of NASH by inhibiting the Toll-like receptor signaling pathway, promoting the differentiation of macrophage M2 phenotype and mediating the AKT signaling pathway.

Our data demonstrate that C5aR1 is a potential target for the intervention of NASH. We then tested the therapeutic effect of the C5aR1 antagonist, PMX-53, in the NASH mice by treating over the last 4 weeks. H&E and Sirius Red staining showed reduced hepatic fibrosis in the PMX-53-treated mice (Fig. 6A). Furthermore, the result of qRT-PCR showed that the pro-fibrotic genes such as Tgf-β1, α-Sma and Col1a1 were greatly lowered by treatment of PMX-53 (Fig. 6B). Similarly, the protein level of α-SMA was decreased by PMX-53 treatment (Fig. 6C). Immunohistochemistry analysis showed that the deposition of F4/80 in liver was decreased in PMX-53-treated mice, and the mRNA level of F4/80 got the similar result (Fig. 6D and E). At the same time, the hepatic mRNA levels of inflammation associated genes such as Tnf-α, IL-6, and IL-1β were downregulated by treatment with PMX-53 (Fig. 6E). Oil red staining displayed that PMX‑53 treatment reduced lipid droplet accumulation in the liver (Fig. 6A). The hepatic expression levels of lipogenesis associated genes were decreased in the PMX-53 treatment group compared to control group (Fig. 6F). Furthermore, the results of qRT-PCR showed that PMX-53 treatment decreased the mRNA levels of macrophage M1 marker such as iNOS and MCP1, contrarily, upregulated the expression of M2 marker such as MRC2 and CD163 (Fig. 6G and H). Overall, blockade of C5aR1 resulted in a reduction of hepatic fibrosis, inflammatory response, and steatosis in NASH mice.