Yiguanjian decoction inhibits macrophage M1 polarization and attenuates hepatic fibrosis induced by CCl4/2-AAF

Abstract Context Our previous studies indicated that Yiguanjian decoction (YGJ) has an anti-hepatic-fibrosis effect and could regulate macrophage status. Objective To elucidate the mechanism of YGJ in regulating macrophages. Materials and methods Liver cirrhosis was induced by CCl4 for 12 weeks combined with 2-acetylaminofluorene (2-AAF) for the last 4 weeks in male Wistar rats. YGJ (3.56 mg/kg) orally administered in the last 4 weeks, and SORA (1 mg/kg) as control. In vitro, RAW264.7 cells were treated with lipopolysaccharides (LPSs) to induce macrophage polarization to the M1 phenotype, and they were co-cultured with WB-F344 cells and allocated to M group, YGJ group (2 μg/mL) and WIF-1 group (1 μg/mL) with untreated cells as control. The differentiation direction of WB-F344 cell line was observed in the presence or absence of YGJ. Pathology, fibrosis-related cytokines, macrophage polarization-related components, and Wnt signalling pathway components were detected. Results In vivo, the expression levels of α-SMA, Col (1), OV6, SOX9, EpCAM and M1 macrophage-related components (STAT1, IRF3, IRF5, IRF8, SOCS3) significantly decreased in the YGJ group compared with those in the 2-AAF/CCl4 group (p < 0.01 or 0.05). In vitro, the expression levels of M1 macrophage-related components, including STAT1, NF-κB, IRF3, IRF5, and SOCS3, in RAW264.7 cells decreased significantly in the YGJ group compared with those in the M group (p < 0.05 or p < 0.01). The expression levels of Wnt3A, FZD5, LRP-5/-6, and β-catenin significantly increased in the YGJ group compared with those in the M group (p < 0.05 or p < 0.01). In addition, the expression levels of Wnt-4/-5A/-5B, and FZD2 significantly decreased in the YGJ group compared with those in the M group (p < 0.05 or p < 0.01). Conclusion This study suggests that the anti-cirrhosis effect of YGJ is associated with its ability to inhibit macrophage M1-polarization, which provides a scientific basis for the clinical application of YGJ.


Introduction
Liver cirrhosis has emerged as a major contributor to the global health burden (Mokdad et al. 2014;Wang et al. 2016). Kupffer cells (KCs), also known as macrophages of the liver, regulate fibrosis development by modulating pro-and anti-inflammatory pathways (Chen et al. 2013;You et al. 2013). Distinct macrophage subsets express different types of chemokines and surface markers and exhibit diverse functions (Anderson and Mosser 2002;Gordon 2003;Rauh et al. 2005). In the liver, M1 macrophages are induced by pro-inflammatory mediators and produce an abundance of pro-inflammatory cytokines, such as lipopolysaccharides (LPSs). By contrast, M2 macrophages are involved in blocking the inflammatory response and promoting tissue repair (Mosser 2003;Moestrup and Møller 2004;Arnold et al. 2007;Hume 2015). Macrophages also regulate stem cell differentiation by secreting Wnt ligands (Carpino et al. 2016).

Materials and methods
Materials YGJ ingredients (18 g of Rehmanniae radix, 10 g of Glehniae radix, 10 g of Ophiopogonis radix, 12 g of Lycii fructus, 10 g of Radix Angelicae sinensis, and 4.5 g of Toosendan fructus) were provided by Shanghai Huayu Herbs Co. Ltd., All herbs were powdered and further extracted by 10 L 20% ethanol twice. Afterward, the extracted liquid was filtered and was subsequently evaporated through rotary vaporation under decompressing, and was vacuum dehydrated to afford extract of YGJ. The process was conducted in Shuguang Hospital affiliated to Shanghai University of Traditional Chinese Medicine (TCM), and the solution was stored at À20 C. Chemical analysis of YGJ and its fractions by UHPLC-QOrbitrap HRMS. The results have been shown in the Supplementary materials. Sorafinib (SORA), purchased from Bayer (Leverkusen, Germany), was used as a positive control drug.

Animals and the rat model of liver fibrosis
Male Wistar rats (aged 7-8 weeks and weighing 160-180 g, n ¼ 32) were obtained from the Shanghai Experimental Animal Centre of the Chinese Academy of Sciences (Shanghai, China). The animals were maintained in a constant-temperature environment and supplied with standard laboratory chow and water ad libitum. All experimental protocols were approved by the Animal Research Committee of Shanghai University of TCM (No. 20170132).
The rat fibrotic model of progressive cirrhosis was induced by CCl 4 combined with 2-acetylaminofluorene (2-AAF) . Rats were subcutaneously injected with 50% CCl 4 -olive oil solution (2 mL/kg) twice a week for 8 weeks. From the ninth week, the concentration of CCl 4 -olive oil solution was adjusted to 30% (2 mL/kg), and all model rats were treated with 2-AAF (10 mg/kg/d) via oral administration once a day and continuously administered for 4 weeks to maintain the progress of hepatic fibrosis. On the first day of week 9, the fibrotic rats were randomly divided into groups (n ¼ 8): 2-AAF/CCl 4 , 2-AAF/CCl 4 þ YGJ (YGJ), and 2-AAF/CCl 4 þ SORA (SORA). YGJ and SORA were orally administered at dosages of 3.56 and 1.0 mg/kg, respectively, once per day for 4 weeks. Naive, untreated rats (N; n ¼ 8) received an equal amount of subcutaneously administered olive oil and orally administered volume of solvent. At the end of the animal experiment, all rats were euthanized with pentobarbital sodium at a dose of 60 mg/kg, and blood and hepatic tissue samples were obtained.

Serum chemistry
Serum alanine aminotransferase (ALT) and aspartate aminotransferase (AST) were detected by the experimental centre of Shanghai University of TCM.

Histochemical and immunohistochemical analyses
Paraformaldehyde-fixed (4%) specimens were cut into 4 lm sections and stained with haematoxylin and eosin (H&E) or Sirius Red.
Immunostaining was performed according to a previously published method (Zhang et al. 2016). Briefly, sections were deparaffinized, washed, and preincubated in blocking solution, followed by incubation with anti (1:2000) secondary antibodies (Jackson ImmunoResearch, West Grove, PA, USA) were used with counterstaining. Images were acquired using a laser scanning confocal microscope (FV10i, Olympus, Japan).
RNA preparation and quantitative real-time reverse transcription polymerase chain reaction (RT-PCR) Total RNA was extracted from the liver tissue using a total RNA purification kit (Lot. 250800; TOYOBO, Osaka, Japan). RNA was reverse transcribed to cDNA, and gene expression was measured using a SYBR Green Real-time PCR Master Mix (Lot. 411900) (TOYOBO) and the ViiA 7 Real-Time PCR System (ABI, American). Primers and oligonucleotide probes were designed using Primer Express (Takara Chemical) and are listed in Table 1. The mRNA expression of TNF-a, TGF-b-1, a-SMA, Col (1), Hep, CD68, CD163, OV6, SOX9, EpCAM, CK19, STAT1, NF-jB, IRF3, IRF5, IRF8, SOSC3, Wnt-4/Wnt-5A/Wnt-5B/Wnt-11, b-catenin, FZD-2/FZD-3/FZD-6, and GAPDH was quantified using quantitative RT-PCR performed on five samples (in vivo) or on three samples (in vitro) in both the experimental and control groups. Individual gene expression was normalized to GAPDH expression levels. The parameters for the One-Step SYBR RT-PCR (Perfect Real Time) were as follows: an initial step of 15 min at 42 C, 2 min at 95 C, 40 amplification cycles of denaturation at 95 C for 15 s, and annealing and extension at 60 C for 1 min.

Gene
Primer sequence Gene Primer sequence a-SMA

Statistical analysis
All data in the experiment are expressed as mean ± standard deviation (SD). SPSS 17.0 was used for statistical analysis, and comparisons between two groups were made using an unpaired t-test. For comparisons between multiple groups, one-way ANOVA was used. p < 0.05 was considered statistically significant.

YGJ inhibits the activation of hepatic stellate cells and the progression of liver fibrosis
Sirius Red staining showed that the collagen in the bridge connecting the central veins and neighbouring portal areas was continually increased, and pseudonodules were formed in the 2-AAF/CCl 4 group; they were markedly attenuated in the YGJ group compared with those in the 2-AAF/CCl 4 group ( Figure  1(a)). In addition, the positive area of Sirius Red staining was increased significantly in the 2-AAF/CCl 4 group compared with that in the N group (p < 0.01) but was decreased significantly in the YGJ group compared with that in the 2-AAF/CCl 4 group (p < 0.01) (Figure 1(d)).
Immunostaining showed that a-SMA and Col (1) expression was markedly increased in the 2-AAF/CCl 4 group, while these pathological changes were clearly alleviated in the YGJ group (Figure 1(b,c)). Immunoblot analysis showed that the protein expression levels of a-SMA and Col (1) were significantly increased in the 2-AAF/CCl 4 group compared with those in the N group (p < 0.01) but were significantly decreased in the YGJ group compared with those in the 2-AAF/CCl 4 group (p < 0.01) (Figure 1(e,f)). The mRNA expression levels of a-SMA, Col (1), and TGF-b1 were increased significantly in the 2-AAF/CCl 4 group compared with those in the N group (p < 0.01) but were decreased significantly in the YGJ group compared with those in the 2-AAF/CCl 4 group (p < 0.01) (Figure 1(g)).
YGJ inhibits the inflammation response and bile duct reaction in the liver H&E staining showed extensive hepatocyte steatosis and hepatic lobular structural disorders in the 2-AAF/CCl 4 group. However, these pathological changes were suppressed by YGJ treatment (Figure 2(a)). Immunostaining showed that CD68 (a marker of M1 macrophage) expression was markedly increased, but CD163 (a marker of M2 macrophage) expression was clearly decreased in the 2-AAF/CCl 4 group. By contrast, CD68 expression was markedly decreased and CD163 was clearly increased after YGJ treatment (Figure 2(b,c)). The mRNA expression levels of CD68 and TNF-a (secreted by M1 macrophages) were increased significantly in the 2-AAF/CCl 4 group compared with those in the N group (p < 0.01). However, the expression levels were reduced significantly in the YGJ group compared with those in the 2-AAF/CCl 4 group (p < 0.01) (Figure 2(g)). Conversely, the mRNA expression of CD163 was decreased significantly in the 2-AAF/CCl 4 group compared with that in the N group (p < 0.01) but was increased significantly in the YGJ group compared with that in the 2-AAF/CCl 4 group (p < 0.01) (Figure 2(g)).
Immunostaining showed that CK19 expression was markedly increased in the 2-AAF/CCl 4 group compared with that in the N group but was decreased in the YGJ group compared with that in the 2-AAF/CCl 4 group (Figure 2(d)). Consistent with immunostaining, immunoblot analysis also showed that the protein expression level of CK19 was significantly increased in the 2-AAF/CCl 4 group compared with that in the N group (p < 0.01). However, it was significantly decreased in the YGJ group compared with that in the 2-AAF/CCl 4 group (p < 0.01) (Figure 2(e,f)).
In addition, serum liver function assay showed that in the 2-AAF/CCl 4 group, ALT and AST activation was increased significantly compared with that in the N group (p < 0.01). However, it was decreased significantly in the YGJ group compared with that in the 2-AAF/CCl 4 group (p < 0.01) (Figure 2(h)). Figures 3(a,b) show that the protein levels of M1 macrophagerelated components, including STAT1, IRF3, IRF5, IRF8, and SOCS3, were increased significantly in the 2-AAF/CCl 4 group compared with those in the N group (p < 0.05 or p < 0.01). However, they were decreased significantly in the YGJ group compared with those in the 2-AAF/CCl 4 group (p < 0.05 or p < 0.01), suggesting that YGJ inhibits macrophage polarization into M1 phenotypes.

YGJ inhibits the macrophage M1 polarization and the activation of non-canonical wnt signalling pathway
Given that macrophages are closely related to the activation of the Wnt signalling pathway, we further examined the mRNA expression levels of key molecules in this pathway. The mRNA Figure 3. YGJ inhibits the activation of M1 macrophages and non-canonical Wnt signalling pathways and inhibits the differentiation of hepatic progenitor cells into myofibroblasts in vivo. (a) STAT1, IRF3, IRF5, IRF8, and SOCS3 protein bands were depicted in the immunoblot images, and (b) the densitometric quantification of the protein bands presented as a histogram (n ¼ 5 per group). (c) The mRNA expressions of Wnt-4, -5 A, -5B, FZD-2, -3, and -6 in liver were measured by RT-PCR and normalized to GAPDH mRNA (n ¼ 5 per group). (d) Wnt5A, Wnt5B, and FZD2 protein bands were depicted in the immunoblot images, and (e) the densitometric quantification of the protein bands presented as a histogram (n ¼ 5 per group). (f) OV6, SOX9, EpCAM, CK19, and Hep mRNA expressions were measured by RT-PCR and normalized to GAPDH mRNA (n ¼ 5 per group). (g) EpCAM protein bands were depicted in the immunoblot images, and (h) the densitometric quantification of the protein bands presented as a histogram (n ¼ 5 per group). Ã p < 0.05 and ÃÃ p < 0.01. N: untreated group (control); 2-AAF/CCl 4 : 2-acetylaminofluorene/carbon tetrachloride-treated group; YGJ: 2-AAF/CCl 4 þ Yiguanjian decoction-treated group; SORA: 2-AAF/CCl 4 þ sorafenib-treated group. levels of non-canonical Wnt-pathway components, including Wnt-4/-5A/-5B, and FZD-2/-3/-6, were increased significantly in the 2-AAF/CCl 4 group compared with those in the N group (p < 0.01). The mRNA expression levels of Wnt-4/-5A/-5B and FZD-2/-3 were decreased significantly in the YGJ group compared with those in the 2-AAF/CCl 4 group (p < 0.05 or p < 0.01) (Figure 3(c)). The protein expression levels of Wnt-5A/-5B and FZD2 were increased significantly in the 2-AAF/CCl 4 group compared with those in the N group (p < 0.01) but were significantly decreased in the YGJ group compared with those in the 2-AAF/CCl 4 group (p < 0.01) (Figure 3(d,e)). These results suggested that YGJ can inhibit the activation of non-canonical Wnt signalling in vivo.
As we previously reported in 2-AAF/CCl 4 models, hepatic progenitor cells are activated and differentiated into myofibroblasts . In the present study, the mRNA expression levels of the hepatic progenitor cell markers OV6, SOX9, EpCAM, and the cholangiocyte marker CK19 were increased significantly in the 2-AAF/CCl 4 group compared with those in the N group (p < 0.01); they were reduced significantly after YGJ treatment (p < 0.05 or p < 0.01). The protein expression level of EpCAM was increased significantly in the 2-AAF/ CCl 4 group compared with that in the N group (p < 0.01) but was significantly decreased in the YGJ group compared with that in the 2-AAF/CCl 4 group (p < 0.01) (Figures 3(g,h)). The mRNA expression of Hep (a hepatocyte marker) was decreased significantly in the 2-AAF/CCl 4 group compared with that in the N group (p < 0.01), but no significant change was noted after YGJ treatment (p > 0.05) (Figure 3(f)). This result suggests that YGJ can effectively inhibit the activation of hepatic progenitor cells and bile duct reaction, but the effects on liver regeneration may be limited in the 2-AAF/CCl 4 model.

YGJ can inhibit the differentiation of WB-F344 cells induced by M1 polarized macrophages
Immunofluorescence staining showed that the protein expression of a-SMA in WB-F344 cells was markedly increased in the M group, but this increase was suppressed by adding YGJ or WIF-1 (a blocker of non-canonical Wnt pathway) (Figure 4(a)) (The figures of WB-F344 co-cultured with inactivated macrophages and WB-F344 co-cultured with M1 macrophages were reused from a previous study, and permission from the publisher was obtained). The mRNA level of a-SMA in WB-F344 cells was increased significantly in the M group compared with that in the N group (p < 0.01) but was decreased significantly in the YGJ Figure 4. YGJ inhibits the differentiation of WB-F344 cells into myofibroblasts through inhibiting the activation of M1 macrophages. (a) a-SMA immunofluorescent staining (green) in WB-F344 cells (Â200) (the presented in vitro experiments were conducted in the same batch, the normal and model groups used the same pictures as published articles. The figures of WB-F344 co-cultured with inactivated macrophages and WB-F344 co-cultured with M1 macrophages were reused form a previous study [Ying Xu et al. 2018] and are reproduced with permission here). (b) a-SMA mRNA expression in WB-F344 cells was measured by RT-PCR and normalized to GAPDH mRNA (n ¼ 3 per group). (c) STAT1, NF-jB, IRF3, IRF5, IRF8, and SOCS3 mRNA expressions in RAW264.7 cells were measured by RT-PCR and normalized to GAPDH mRNA (n ¼ 3 per group); (d) STAT1, NF-jB, IRF3, IRF5, and SOCS3 protein bands were depicted in the immunoblot images, and (e) the densitometric quantification of the protein bands presented as a histogram (n ¼ 3 per group). Ã p < 0.05 and ÃÃ p < 0.01. N: WB-F344 cells co-cultured with inactivated RAW264.7 cells; M: WB-F344 cells co-cultured with LPS (100 ng/mL)-activated RAW264.7 cells (referred to as LPS-RAW264.7); YGJ: WB-F344 cells co-cultured with LPS-RAW264.7 treated with Yiguanjian decoction; WIF-1: WB-F344 cells co-cultured with LPS-RAW264.7 treated with Wnt inhibitory factor-1. and WIF-1 groups compared with that in the M group (p < 0.01) (Figure 4(b)).
Figure 4(c) shows that the mRNA expression levels of M1 macrophage-related components, including STAT1, NF-jB, IRF3, IRF5, and SOCS3, in RAW264.7 cells were increased significantly in the M group compared with that in the N group (p < 0.01); however, the above indicators were decreased significantly in the YGJ group compared with that in the M group (p < 0.05 or p < 0.01). In addition, the protein expression levels of STAT1, NF-jB, IRF3, IRF5, and SOCS3 in RAW264.7 cells were increased significantly in the M group compared with those in the N group (p < 0.01). However, they were significantly decreased in the YGJ group compared with those in the M group (p < 0.05 or p < 0.01) (Figure 4(d,e)). Thus, YGJ may suppress macrophage polarization into the M1 phenotype and inhibit the differentiation of WB-F344 cells into myofibroblasts in vitro.

YGJ inhibition of WB-F344 cell differentiation into myofibroblasts is related to the regulation of wnt signalling activation in vitro
Furthermore, the mRNA expression levels of canonical Wnt-signalling components Wnt-1/-3A, FZD5, LRP-5/-6, and b-catenin in WB-F344 cells were decreased significantly in the M group compared with those in the N group (P < 0.01), while the mRNA expression levels of Wnt 3 A, FZD5, LRP-5/-6, and b-catenin were significantly increased in the YGJ group compared with those in the M group (p < 0.05 or p < 0.01) ( Figure  5(a)). In addition, the protein expression levels of LRP5 and b-catenin were decreased significantly in the M group compared with those in the N group (p < 0.01) but were significantly increased in the YGJ group compared with that in the M group (p < 0.01) (Figure 5(b,c)).
However, the mRNA expression levels of non-canonical Wntsignalling components, including Wnt-4/-5A/-5B and FZD-2/-4/-6 in WB-F344 cells were increased significantly in the M group compared with those in the N group (p < 0.01). The mRNA levels of Wnt-4/-5A/-5B, and FZD 2 were significantly decreased in the YGJ group compared with those in the M group (p < 0.05 or p < 0.01) (Figure 5(d)). The protein expression levels of Wnt-5A/-5B, and FZD 2 were increased significantly in the M group compared with those in the N group (p < 0.01), while they were significantly decreased in the YGJ group compared with that in the M group (p < 0.01) ( Figure 5(e,f)). Thus, the mechanism of YGJ regulation of the Wnt signalling pathway in WB-F344 cells may be through suppression of macrophage polarization into M1 in vitro.

Discussion
Our previous study showed that in the rat model of progressive cirrhosis induced by 2-AAF/CCl 4 , endogenous hepatic progenitor cells differentiated into myofibroblasts and promoted fibrosis. Blockade of the Wnt signalling pathway inhibited TGF-b1induced differentiation of hepatic progenitor cells into myofibroblasts in vitro .
In the present study, YGJ inhibited the liver inflammation response, activation of HSCs, and the progression of liver cirrhosis induced by 2-AAF/CCl 4 . The expression levels of OV6, SOX9, EpCAM, and CK19 were significantly decreased in the liver after YGJ treatment, suggesting that the anti-liver fibrosis effect of YGJ is not only related to inhibiting HSC activation but also to inhibiting liver progenitor cell activation and bile duct action, and these effects are closely related to macrophage polarization.

YGJ inhibits macrophage M1 polarization
Macrophages are keen sensors of tissue homeostasis and can rapidly switch between pro-and anti-inflammatory regulatory functions in response to different perturbations in their microenvironment (Xu et al. 2018). These different functional phenotypes of macrophages are generally classified into two main types: those conditioned by pro-inflammatory stimuli, such as interferon (IFN)-c, also referred to as classically activated or M1 macrophages, and those polarized by anti-inflammatory cytokines, such as interleukin (IL)-4 and IL-13, also known as alternatively activated or M2 macrophages (Mosser and Edwards 2008;Martinez et al. 2009;Murray et al. 2014). Meijer et al. (2000) reported that Kupffer cell depletion by liposome-encapsulated dichloromethylene-diphosphonate abolishes the hepatic synthesis of IL-6 and IL-10 mRNA and decreases the hepatic expression of TNF-a, hepatocyte growth factor, and TGF-b1 mRNA after partial hepatectomy. A recent study found that Nogo-B can promote macrophage polarization into the M1 phenotype and further promote the progression of alcoholic liver disease. By contrast, Nogo-B deletion can promote macrophage polarization into the M2 phenotype, thus promoting recovery from alcoholic liver disease (Park et al. 2017), suggesting that KCs are closely related to liver regeneration and damage repair.
In the present study, the expression levels of CD68 and TNFa mRNA were significantly increased, whereas that of CD163 mRNA was significantly decreased in the liver tissue. These results suggested an imbalance between pro-inflammatory and anti-inflammatory macrophages in this model. In addition, the protein levels of STAT1, IRF3, IRF5, IRF8, and SOCS3 (proteins associated with polarization into M1 macrophages) were significantly increased in the 2-AAF/CCl 4 group but were significantly decreased after YGJ treatment, suggesting that YGJ has a regulatory effect on the M1 macrophages of liver.

YGJ regulates the crosstalk between macrophages and hepatic progenitor cells mediated by Wnt signalling pathway
The proliferation and differentiation of hepatic progenitor cells are regulated by various signalling pathways, such as Wnt, Hedgehog, and Notch, especially the Wnt signalling pathway, which is involved in the regulation of all stem cell populations (Reya and Clevers 2005). Wnt ligands (e.g., Wnt1-11) are mainly derived from macrophages and play an important role in tissue regeneration and repair (Wynn et al. 2013). Saha et al. (2016) reported that Wnt ligands derived from the extracellular vesicles of macrophages protect intestinal stem cells and improve the survival rate of intestinal stem cells after radiation injury. They also found that Porcupine (PORCN) is indispensable in the secretion mechanism of Wnt core ligands. The PORCN inhibitor Wnt-C59 abolished the Wnt-secretion function of macrophages and weakened its protective effect on intestinal stem cells, providing direct evidence for Wnts derived from macrophages.
As an important part of the stem cell microenvironment, macrophages have received widespread attention in the field of liver regeneration in recent years (Aurora and Olson 2014). The differentiation of hepatic progenitor cells into hepatocytes is mainly regulated by the classical Wnt/b-catenin signalling pathway and may be related to Wnt3A release by M2 macrophages after phagocytosis of apoptotic hepatocyte fragments and activation of the Wnt/b-catenin signalling pathway in hepatic progenitor cells (Boulter et al. 2012). Wnt3A also promotes IL-4 or TGF-b1-induced macrophage polarization into the M2 phenotype (Feng et al. 2018). The differentiation of hepatic progenitor cells into myofibroblasts may be regulated by TNF-a release by M1 macrophages to activate non-canonical Wnt signalling pathways, such as Wnt5/FZD2 (Jiang et al. 2006). A clinical study has shown that pro-inflammatory macrophages dominate the macrophage population in children with non-alcoholic fatty liver disease, and docosahexaenoic acid treatment can polarize these macrophages into anti-inflammatory phenotypes and upregulate Wnt3A expression. Thus, b-catenin phosphorylation is promoted in hepatic progenitor cells, which then differentiate into hepatocytes (Carpino et al. 2016). Collectively, these findings suggested that Wnt signalling mediated the crosstalk between macrophages and hepatic progenitor cells.
Our previous study showed that the non-canonical Wnt signalling pathway is significantly activated in the rat model of cirrhosis induced by 2-AAF/CCl 4 . In vitro, TGF-b1 induced WB-F344 cells to differentiate into myofibroblasts, and the noncanonical Wnt signalling pathway is significantly activated. Wnt signalling blocker WIF-1 inhibits this pathological process (Xu et al. 2018). In the current study, we found that in the components of the non-canonical Wnt signalling pathway, including Wnt-4/-5A/-5B, FZD-2/-3 expression levels were significantly reduced after YGJ treatment. As such, YGJ may have a regulatory effect on the activation of Wnt signalling in liver fibrosis induced by 2-AAF/CCl 4 in rats.
We used LPS to polarize macrophages into the M1 phenotype in vitro and then co-cultured with WB-F344 cells and treated with YGJ or WIF-1. The expression of a-SMA mRNA in WB-F344 cells was significantly increased when the two cell types were co-cultured, and YGJ significantly reduced its expression. Thus, M1 macrophages may have promoted the differentiation of hepatic progenitor cells into myofibroblasts, while YGJ inhibited this pathological process.
The protein levels of molecules related to macrophage polarization to M1, such as NF-OEB, IRF3, IRF5, STAT1, and SOCS3, were increased significantly. In terms of the expression levels of non-canonical Wnt signalling-related molecules Wnt-4/-5A/-5B, FZD-2/-4/-6 were significantly increased in WB-F344 cells. This result indicated that M1 macrophages promoted the activation of the non-canonical Wnt signalling pathway, which promoted the differentiation of hepatic progenitor cells into myofibroblasts. WIF-1 had no significant effect on macrophage polarization but significantly decreased a-SMA mRNA expression in WB-F344 cells. YGJ significantly inhibited macrophage polarization into the M1 phenotype, significantly promoted the activation of canonical Wnt signalling, and inhibited the activation of noncanonical Wnt signalling. These effects of YGJ reduced the mRNA expression of a-SMA in WB-F344 cells in vivo. Taken together, these results suggest that Wnt signalling serves as a key 'regulator' in the cross-talk between macrophages and hepatic progenitor cells in rat liver fibrotic model induced by 2-AFF/ CCl 4 . Likewise, YGJ may regulate macrophage polarization and Wnt signalling pathway mediated cross-talk between macrophages and hepatic progenitor cells.

Conclusions
We found that in the rat liver fibrosis induced by 2-AAF/CCl 4 , YGJ inhibited macrophage M1 polarization, thereby reducing the release of non-canonical Wnt ligands. The activation of noncanonical Wnt signalling in hepatic progenitor cells was inhibited, and in turn, their differentiation into myofibroblasts was inhibited. This study not only provides further evidence for the anti-hepatic fibrosis effect of YGJ but also presents a scientific basis for the use of YGJ in the treatment of cirrhosis.