Emodin Alleviates Liver Fibrosis of Mice by Reducing Infiltration of Gr1hi Monocytes

Emodin, as a major active component of Rheum palmatum L. and Polygonum cuspidatum, has been reported to have antifibrotic effect. However, the mechanism of emodin on antifibrotic effect for liver fibrosis was still obscure. In the present study, we aimed to investigate whether emodin can alleviate carbon tetrachloride- (CCl4-) induced liver fibrosis through reducing infiltration of Gr1hi monocytes. Liver fibrosis was induced by intraperitoneal CCl4 injection in mice. Mice in the emodin group received emodin treatment by gavage. Pretreatment with emodin significantly protected mice from liver inflammation and fibrosis revealed by the decreased elevation of serum alanine aminotransferase (ALT) and aspartate aminotransferase (AST), as well as reduced hepatic necrosis and fibrosis by analysis of hematoxylin-eosin (HE) staining, Masson staining, α-smooth muscle actin (α-SMA), and collagen-I immunohistochemistry staining. Further, compared to CCl4 group, mice in the emodin group showed significantly less intrahepatic infiltration of Gr1hi monocytes. Moreover, emodin significantly inhibited hepatic expression of interleukin-1β (IL-1β), tumor necrosis factor-α (TNF-α), transforming growth factor-β1 (TGF-β1), granulin (GRN), monocyte chemoattractant protein 1 (MCP-1), and chemokine ligand 7 (CCL7), which was in line with the decreased numbers of intrahepatic Gr1hi monocytes. In conclusion, emodin can alleviate the degree of liver fibrosis by reducing infiltration of Gr1hi monocytes. These results suggest that emodin is a promising candidate to prevent and treat liver fibrosis.


Introduction
Liver fibrosis, characterized by excessive deposition of extracellular matrix (ECM), is a pathological change after chronic liver injury including viral infection, toxic damage, metabolic disorders, and alcohol abuse [1]. However, there is still a lack of effective drugs for the treatment of liver fibrosis [2].
Understanding the mechanisms of liver fibrosis is important for prevention and control of liver disease progression.
Hepatic macrophages comprise resident Kupffer cells (KFs) and infiltrating monocyte-derived cells [13]. During the development of liver fibrosis, macrophage pool of the liver can be rapidly expanded by infiltrating phagocytes that mainly originate from circulating monocytes [14][15][16]. In mice, two major populations of circulating monocytes exist: Gr1 hi (Ly-6C hi ) monocytes and Gr1 lo (Ly-6C lo ) monocytes. In acute and chronic liver injury of mice, Gr1 hi monocytes are massively recruited into the liver [8,11]. As vigorous secretors of proinflammatory and profibrotic factors including TNFand TGF-1, they drive inflammation and activate HSC thereby triggering a cascade of events leading to liver fibrosis [10,17].
In the past two thousand years, Rheum palmatum L. and Polygonum cuspidatum has been used as Traditional Chinese Medicine to treat liver diseases, which has achieved significant curative effects [23]. As the main active ingredient of Rheum palmatum L. and Polygonum cuspidatum, emodin has been demonstrated to have anti-inflammatory, antitumor, and antifibrotic effects [24,25]. Further study suggested that emodin has certain antifibrotic effect on liver fibrosis [26,27]. However, the mechanism of emodin on the antifibrotic effect for liver fibrosis is still obscure. Therefore, in the present study we investigated whether emodin could alleviate the degree of carbon tetrachloride-(CCl 4 -) induced liver fibrosis by inhibiting infiltration of Gr1 hi monocytes.

Materials and Methods
.. Mice. Male C57BL/6 mice, 7-8 weeks old, weighing 25 ± 2 g, purchased from Beijing Vital River Experimental Animals Technology (Beijing, China), were used in this study. Mice were housed under laminar airflow hoods in a specific pathogen-free room with a 12 h light and 12 h dark schedule and fed autoclaved chow and water at a controlled temperature of 22 ± 2 ∘ C, 50-60% relative humidity.
.. Experimental Protocol. Mice were randomly assigned to the control group, CCl 4 group, and emodin group. For induction of liver fibrosis, mice in the emodin group and CCl 4 group were injected intraperitoneally with 0.6 ml/kg dose of CCl 4 (CCl 4 : olive oil = 1 : 4, 3 l/g CCl 4 oil) twice weekly for 4 weeks. Mice in the control group were injected with equal volume of olive oil as control. Emodin was dissolved in 0.25% sodium carboxymethyl cellulose (CMC-Na) aqueous solution to prepare the appropriate concentration. Emodin was given by gavage at a dose of 20 mg/kg/d in the emodin group. This dose of emodin is the optimal dose proved by the previous study [27]. For the control group and CCl 4 group, mice were administrated orally with the same volume of CMC-Na aqueous solution.
After 48 h of the last CCl 4 administration, mice were sacrificed after anesthesia. Blood samples were collected for ALT and AST detection. The liver tissues of mice were collected.
.. Liver Function Detection. Blood samples were centrifuged at 4000 rpm/min for 15 min to get serum. ALT and AST levels were determined with clinical testing technique through biochemical instrument (Beckman Coulter, Inc., Brea, CA, USA).
.. Flow Cytometry. Isolation of liver-infiltrating lymphocytes was performed by an automated, mechanical disaggregation system (Medimachine, Becton Dickinson, USA). The liver samples of mice were washed with PBS twice and then cut into small pieces of 3-4 mm 3 . Five pieces were immediately put in a disposable disaggregator Medicon with 50 m separator mesh (Becton Dickinson, USA) and then 1 ml PBS was added and processed in the Medimachine System for 1 min. Disaggregated cells were removed and pressed through 70 m cell strainers to obtain single cell suspensions. Cells were used immediately for flow cytometric analysis. Related antibodies were listed as follows: CD45 (557235, BD Pharmingen, USA), CD11b (557397, BD Pharmingen, USA), Gr1/Ly6C (560595, BD Pharmingen, USA), Ly6G (551460, BD Pharmingen, USA), and F4/80 (25-4801, eBioscience, USA). Flow cytometric analysis was performed on a FACS Aria II (BD Bioscience, USA).
.. Histopathology. Mouse liver tissues were fixed in paraformaldehyde for 48 h, then dehydrated, and finally embedded with paraffin. Paraffin-embedded liver samples were sectioned to 3 m thin slices, which were performed with hematoxylin-eosin (HE) staining and Masson staining according to standard protocols.

.. Histological and Immunohistochemical Evaluation.
Sections with HE staining were examined under light microscopy by an experienced pathologist in a blinded fashion for liver steatosis, necrosis, and leukocyte infiltration. For Masson staining and immunohistochemical staining of -SMA and collagen-I, five ×100 magnification images of Masson staining and immunohistochemical staining were captured in each section. Positive staining area of Masson staining (blue), -SMA, or collagen-I immunohistochemical staining (brown) per high-power field (HPF) of each image was converted into pixels by Image-ProPlus software (Media Cybernetics, USA). The percentage of positive area was expressed as a fraction of the total number of pixels, averaged across the 5 different regions per section.
Absolute counts of CD45 + cells (leucocytes), F4/80 + cells (macrophages), and CD11b + cells (monocytes) per HPF of stained liver sections were manually assessed in 5 different fields per section. All quantification was carried out blinded and without prior knowledge of sections or groups.
.. Real-Time Gene Expression Analysis. Total RNA was extracted from 50 mg frozen liver tissues using TRIzol according to the manufacturer's instructions (Life Technologies, USA) and then reverse-transcribed to get cDNA by using the PrimeScript RT Master Mix kit (Takara, China) according to the relevant experimental manual. Real-time PCR was performed on Step One Plus Real-Time PCR Systems (Life Technologies, USA) using SYBR Premix Ex Taq kit (Takara, China). All primers and PCR product sizes of this study are listed in Table 1. -Actin was used as an internal control [28].
.. Statistical Analysis. All data are expressed as the mean ± standard error of the mean (SEM). Statistical analysis was performed using one-way analysis of variance (ANOVA) test by SPSS 22.0 software. < 0.05 was considered to be statistically significant.

.. Emodin Attenuated CCl -Caused Liver Inflammation.
The protective effect of emodin on CCl 4 -induced liver injury was identified by liver function and HE staining. As shown in Figure 1(a), the serum ALT and AST increased significantly in mice of CCl 4 group as compared with the control group. Emodin can significantly reduce the increased liver enzymes. The liver tissues of mice in the CCl 4 group showed liver steatosis, necrosis, and leukocyte infiltration which were significantly relieved in the emodin group (Figure 1(b)). These results demonstrated that emodin can significantly reduce CCl 4 -induced liver inflammation in vivo.
.. Emodin Alleviated Liver Fibrosis Induced by CCl . Masson staining was used to observe the collagen fibers of liver tissues. As shown in Figure 2, fiber staining only appeared on the central venous wall, part of portal area, and interlobular septa in hepatic lobules of control mice. However, compared to the control group, the collagen fibers in CCl 4 group were significantly higher. The initial formation of interlobular septa and significantly increased portal fiber deposition were observed. The liver tissues of mice in the emodin group also had fiber deposition, mainly distributed at the portal area, fibrous septum, and central vein. But the density of * * * * * * * * * * * collagen deposition was significantly less than that of CCl 4 group. Moreover, the liver tissues of mice in the control group had only a small amount of -SMA and collagen-I expression ( Figure 2). However, the hepatic expression of -SMA and collagen-I significantly increased in the CCl 4 group (Figure 2). Emodin treatment can significantly reduce the expression of -SMA and collagen-I induced by CCl 4 (Figure 2). These results collectively demonstrated that emodin can significantly reduce CCl 4 -induced liver fibrosis in vivo.

.. Emodin Reduced Gr hi Monocyte Infiltration.
After CCl 4 challenge for 4 weeks, infiltrations of leukocytes, monocytes, and macrophages increased significantly in the mouse liver tissues of CCl 4 group. However, compared to CCl 4 group, mouse liver tissues of emodin group had greatly reduced infiltrations of leukocytes, monocytes, and macrophages (Figures 3(a) and 3(b)). The proportion of Gr1 hi monocyte (CD45 + Ly6G − CD11b + F4/80 + Gr1 hi ) subset increased significantly in the liver tissues of CCl 4 group as compared with the control group. However, the proportion of Gr1 hi monocyte subset was significantly lower in the emodin group compared to the CCl 4 group (Figure 4(b)). These results suggested that emodin can significantly reduce infiltration of monocytederived macrophages, especially Gr1 hi monocyte in chronic liver injury.

.. Emodin Inhibited the Expression of MCP-and CCL.
MCP-1 and CCL7 are the prime monocyte chemotactic factor [8,[18][19][20]. Mice in CCl 4 group have strongly increased hepatic expression of MCP-1 and CCL7, which may be responsible for increasing numbers of intrahepatic Gr1 hi monocytes ( Figure 7). However, it is obvious that emodin effectively inhibited the hepatic expression of MCP-1 and CCL7 (Figure 7), which may explain why emodin could reduce the infiltration of monocytes.

Discussion
Hepatic macrophages, especially monocyte-derived macrophages, hold a central position in the pathogenesis of chronic liver injury and have been proposed as potential targets for liver fibrosis [13,22,29]. In acute and chronic liver injury, Gr1 hi monocytes will firstly be recruited into inflammatory region and release proinflammatory and profibrogenic cytokines which will aggravate liver inflammation and fibrosis of liver [8,17]. In the present study, we confirmed that, after repeated CCl 4 injections, Gr1 hi monocytes were markedly increased in the liver tissues of mice. Pharmacological inhibition of Gr1 hi hepatic monocytes may be capable of limiting chronic liver injury and fibrosis in vivo [30]. Emodin, as a major active component of Rheum palmatum L. and Polygonum cuspidatum, has been reported to have anti-inflammatory, antitumor, and antifibrotic effect [24,25]. Emodin can ameliorate hepatic steatosis through endoplasmic reticulum-stress sterol regulatory element-binding protein 1c pathway in liquid fructosefeeding rats [31]. Liu et al. also reported that emodin could ameliorate liver steatosis and decrease hepatic triglyceride in mice fed with ethanol [32]. In addition, emodin could also significantly inhibit the growth of HSC-T6 cells in vitro [33]. Emodin can also protect the rat liver from CCl 4induced liver fibrosis by inhibiting HSCs activation in vivo  [27]. In the present study, we also confirmed that emodin could reduce liver inflammation and fibrosis in mouse model induced by CCl 4 . Recent studies show that emodin can inhibit homologous lymphotoxins-induced monocytes migration [34]. On this basis, we used immunohistochemistry staining and flow cytometry to observe the changes of monocytes, especially Gr1 hi monocytes. We found that emodin significantly reduced the infiltration of Gr1 hi monocytes which is novel mechanism of the antifibrotic effect of emodin.
Gr1 hi monocytes can release many proinflammatory and profibrogenic cytokines, including TNF-, IL-1 , GRN, and TGF-1. TNF-and IL-1 are mainly from monocytes and macrophages in the acute and chronic liver injury [35] and may trigger the production of many other proinflammatory cytokines and induce hepatocyte death through the recruitment of neutrophils [36,37]. TGF-1 is considered to be the most potent profibrogenic cytokine [38,39]. TGF-1 can promote liver fibrosis through multiple pathways, for example, activating HSCs, stimulating collagen gene transcription, and suppressing the expression of matrix metalloproteinases [40]. TGF-1 mainly comes from monocyte-derived macrophages [29,40]. GRNs is a family of protein growth factors that are involved in cell proliferation. Macrophage-secreted granulin may activate resident HSCs into myofibroblasts resulting in a fibrotic microenvironment [7]. In our present study, we found that the hepatic expression of proinflammatory and profibrogenic cytokines IL-1 , IL-6, TNF-, TGF-1, and GRN was significantly increased following CCl 4 injection. However, emodin administration was able to decrease the hepatic expression of IL-1 , TNF-, GRN, and TGF-1 in the liver fibrosis model, which also explains why emodin can alleviate liver inflammation and fibrosis. CCL7 and MCP-1 are the major cytokines for the recruitment of Gr1 hi monocytes in the liver [41,42]. MCP-1 has been well studied in liver fibrosis. In acute and chronic liver injury, MCP-1 can induce liver fibrosis through recruitment of Gr1 hi monocytes [28]. In the present study, we also found that the hepatic expression of MCP-1 was significantly higher in the CCl 4 group as compared with the control group. In Balb/C mice fed on the methionine/choline-deficient diet, a lack of MCP-1 was associated with lower ALT levels and reduced infiltration of inflammatory cells, together with a lower degree of liver fibrosis [43]. Pharmacological inhibition of MCP-1 by mNOX-E36 can also significantly reduce the infiltration of Gr1 hi monocytes in the process of acute and chronic liver injury and can reduce the degree of liver fibrosis in vivo [21,30,44]. CCL7, previously known as monocyte chemotactic protein-3, belongs to the MCP subfamily of CCLs. CCL7 is expressed at multiple sites of inflammation and is produced by monocytes, fibroblasts, endothelial cells, and mast cells [45][46][47][48]. Compared to MCP-1-deficient mice, CCL7-deficient mice showed less infiltration of monocytes in the encephalitis model caused by West Nile virus infection, which indicates that CCL7 has a stronger chemotactic activity than that of MCP-1 [49]. Inhibition of CCL7 could significantly reduce the infiltration of monocytes/macrophages to lung tissues of mice infected with rhinovirus [50]. Our study also found that MCP-1 and CCL7 expression increased significantly during liver fibrosis, suggesting that MCP-1 and CCL7 may play important roles in liver fibrosis. After emodin treatment, the hepatic expression of MCP-1 and CCL7 was significantly reduced which was in line with the reduction of Gr1 hi monocytes infiltration. However, the related mechanisms regarding how emodin reduces the secretion of MCP-1 and CCL7 needs further investigation.
In conclusion, we found that emodin has inhibitory effects on liver fibrosis, which may be associated with reducing the Gr1 hi monocytes infiltration by the inhibition of MCP-1 and CCL7. These results suggest that emodin may be considered a promising candidate in the prevention and treatment of liver fibrosis. However, the clinical effect of emodin on liver fibrosis deserves further investigation.