Emodin Protects Against Concanavalin A-Induced Hepatitis in Mice Through Inhibiting Activation of the p38 MAPK-NF-κB Signaling Pathway

Fulminant hepatitis is a devastating inflammatory disease of the liver that is widespread in China and mainly develops from chronic or acute hepatitis B virus (HBV) infection. HBV-related end-stage liver disease and hepatocellular carcinoma (HCC) are responsible for over 0.5-1 million deaths per year [1,2,3,4], while acute fulminant hepatitis B causes an additional 40,000 deaths each year [5]. Fulminant hepatitis usually begins with a sudden onset and progresses rapidly, leaving little time for effective treatment, and it is associated with a significant mortality rate. Therefore, it is crucial to develop strategies for its prevention.

T cell-mediated immune responses play important roles in the development and progression of autoimmune and viral hepatitis [6,7,8]. Concanavalin A (Con A)-induced hepatic injury is a well-characterized murine model of T cell-mediated hepatic damage with a pathophysiology similar to those of human viral and autoimmune hepatitis [9,10]. This model has been widely used to study the etiopathogenesis, pathogenesis, and clinical treatment of immunological hepatitis in humans [11]. Activated CD4 ⁺ T cells and Kupffer cells play key roles in hepatocyte damage in this model. These cells infiltrate into the liver parenchyma and induce the secretion of pro-inflammatory cytokines, such as tumor necrosis factor (TNF)-α, interferon (IFN)-γ, interleukin (IL)-1, IL-6 and IL-4 [12,13,14]. Alanine aminotransferase (ALT) and aspartate aminotransferase (AST) become elevated as a result of hepatocyte necrosis following the intravenous administration of Con A [15].

Emodin (1,3,8-trihydroxy-6-methyl-anthraquinone), which is an anthraquinone derivative from the Chinese herb Radix et Rhizoma Rhei, has been reported to possess a variety of biological properties, such as anti-inflammatory [16], anti-viral [17], anti-tumor [18], and anti-oxidant activities [19]. Emodin was found to inhibit inflammatory cytokine production in HMGB1-induced inflammatory responses in vitro and in vivo [20] and to suppress inflammatory responses in TNF-α-induced aortic smooth muscle cells [21]. Emodin also has a protective effect on cholestatic hepatitis [22]. However, little is known regarding its effects on fulminant hepatitis. In the present study, we assessed the effects of emodin on the prevention of liver injury by establishing a mouse model of fulminant hepatitis induced by the intravenous injection of Con A, and the underlying molecular mechanisms were also investigated.

Balb/c mice (6-8 weeks old, male) were obtained from the Experimental Animal Center of the Chinese Science Academy (Shanghai, China), and all mice were housed under pathogen-free conditions. All procedures were performed according to the guidelines for the Care and Use of Laboratory Animals and approved by the ethics committee of the Zhejiang University School of Medicine.

Mice were randomly divided into control (vehicle), emodin (50 mg/kg body weight), Con A (20 mg/kg body weight), and emodin plus Con A (20 mg/kg body weight Con A plus 1.5625 mg/kg, 3.125 mg/kg, 6.25 mg/kg, 12.5 mg/kg, 25 mg/kg, and 50 mg/kg body weight emodin, respectively) groups, and each group contained 5 mice. Con A and emodin (Sigma Chemical Co., St. Louis, MO, USA) were prepared with pyrogen-free saline and sodium carboxymethyl cellulose (CMC-Na; Sigma Chemical Co., St. Louis, MO, USA), respectively. Emodin was administered orally, and Con A was given through intravenous injection at 2 h after emodin administration. In place of emodin, CMC-Na was used in the Con A group, and saline (vehicle) was used in place of Con A in the emodin group, and the normal control mice were treated with a vehicle. The mice were sacrificed 10 h after Con A injection.

Mice were randomly divided into control (vehicle), emodin (25 mg/kg body weight), Con A (20 mg/kg body weight), and emodin plus Con A (20 mg/kg body weight Con A plus 25 mg/kg body weight emodin) groups, and each group contained 5 mice. In the emodin plus Con A group, emodin was administered orally at 2 h before and 30 min, 60 min, 90 min, and 2 h after Con A exposure, respectively. In place of emodin, CMC-Na was used in the Con A group, and saline (vehicle) was used in place of Con A in the emodin group, and the normal control mice were treated with a vehicle. The mice were sacrificed 10 h after Con A injection.

Hepatic sections were obtained from the mice. The sections were fixed with 10% neutral-buffered formalin, embedded in paraffin and cut into 3-5 μm slices. After deparaffinization and rehydration, the slices were stained with hematoxylin and eosin (H&E staining). All specimens were histologically assessed by two experienced pathologists. Five visual fields (10× magnification) randomly selected from each section were used for the image analysis. The area of necrotic liver tissue was measured using the Image-Pro Plus 5.0 software (Media Cybernetics, Inc., Bethesda, MD, USA). The necrosis rate of the hepatocytes was calculated according to the necrotic areas divided by the liver area of the image.

The blood was collected from the retro-orbital sinus in mice following exposure to Con A and/or emodin. Levels of serum ALT and AST were measured using the Automatic Chemical Analyzer 7600-100 (Hitachi, Ltd., Tokyo, Japan).

Total RNA was extracted from tissues using TRIzol reagent (Invitrogen Corp., Carlsbad, CA) following the manufacturer's instructions. For the analysis, the total RNA (1 μg) was reverse-transcribed using the PrimeScript™ RT Reagent Kit with the gDNA Eraser (Code no. RR047A, Takara). The gene expression analysis of the mouse livers was performed by qRT-PCR with SYBR Premix EX TaqTM II (Code no. RR820A, Takara) using the ABI PRISM 7900 sequence detector (Applied Biosystems, Foster City, CA, USA). The total amplification reaction volume of 20 µL contained 2× SYBR® Premix Ex Taq™ II, 0.4 μmol/L primers, and 1 μL of template cDNA. Thermal cycling was carried out for 30 s at 95 °C, followed by 40 cycles of 5 s at 95 °C, and 30 s at 60 °C. Each PCR assay was performed in triplicate, and the changes in mRNA levels were normalized by the levels of the control gene mRNA (β-actin). The activation of cells was determined by measuring RNA of TNF-α in RAW264.7 cells and IL-2 in EL-4 cells, respectively. RAW264.7 cells (2 × 10⁵/ml) or EL4 cells (2 × 10⁶/ml) were treated with or without emodin for 2 h and then were stimulated by Con A. After 24 h incubation, cells were collected and relative quantitative real time PCR was performed. The primers purchased from Sangon Biotech (Shanghai) are listed in Table 1.

The paraffin-embedded liver sections from both the normal and Con A-treated mice were deparaffinized, rehydrated, and treated with an antigen-repairing solution for the immunohistochemistry analysis. Thereafter, the sections were incubated with a blocking solution (5% BSA in PBS), followed by incubation with fluorescein isothiocyanate (FITC)-conjugated F4/80 or phycoerythrin (PE)-conjugated CD4 antibodies (BD Biosciences, San Diego, CA, USA) overnight at 4 ℃. After washing with PBS 3 times, the sections were covered with coverslips and observed using a confocal microscope (Olympus Inc., Center Valley, PA, USA). The numbers of positive cells in each section were counted in five randomly selected fields in each group, and the mean number of immunoreactive cells was calculated for each case. Five mice in each group were analyzed.

Murine macrophage-like RAW264.7 cells and EL4 murine T-lymphoma cells were obtained from the American Type Culture Collection (ATCC, Rockville, MD). RAW264.7 cells were pre-cultured in DMEM medium (Gibco BRI, Grand Island, NY) supplemented with 10% fetal bovine serum (FBS). EL4 cells were maintained in RPMI 1640 medium (Gibco BRI, Grand Island, NY) supplemented with 10% FBS, 2 mM glutamine, 100 μg/ml of penicillin and 100 μg/ml of streptomycin. Cells were pretreated with emodin for 2h and then were stimulated by Con A for 24 h.

The cytotoxicity of Con A and emodin to cells was evaluated by MTT method (Beyotime Biotechnology). Briefly, RAW264.7 cells (2 × 10⁵/ml) or EL4 cells (2 × 10⁶/ml) were cultured in six replicates in 96-well plates in a volume of 200μl. Cells were incubated alone (control) or in presence of increasing concentrations of Con A or emodin (Sigma Chemical Co., St. Louis, MO, USA) for 24 h. Then the cells were incubated with a solution containing 0.5 mg MTT/mL phosphate-buffered saline at 37℃ for 4 h. The MTT solution was removed and the cells were overlaid with 150 μL/well DMSO for 10 min at 37℃. The OD value was measured using a Bio-Rad microplate reader at 490 nm with DMSO as blank.

The liver tissues or cells were homogenized and centrifuged at 12,000 g for 10 min at 4 °C. The proteins were quantified by the Pierce BCA Protein Assay Kit (Thermo Fisher Scientific Inc., Rockford). Equivalent protein amounts (40 µg) were separated by 12% SDS-PAGE and transferred to polyvinylidene difluoride membranes (Millipore, Billerica, MA, USA), which were then blocked in TBST containing 5% defatted milk and incubated with primary antibodies specific for phosphorylated p38 MAPK (#4511, Cell Signaling Technology, USA), phosphorylated NF-κB p65 (#3036, Cell Signaling Technology, USA), total p38 MAPK (#9212, Cell Signaling Technology, USA), total NF-κB p65 (#ab7970, Abcam, USA), and β-actin (cytoplasmic protein marker, #4970, Cell Signaling Technology, USA ) at 4 ℃ overnight. The membranes were then incubated with horseradish peroxidase- conjugated anti-rabbit or anti-mouse immunoglobulin G (Southern Biotechnology Associates, Inc., Birmingham, AL, USA). Bound antibodies were visualized by enhanced chemiluminescence (Thermo Fisher Scientific Inc., Rockford) and exposed to X-ray film. The changes in protein levels were normalized by the levels of β-actin proteins. The densitometric analysis was performed using Quantity One v4.62 (Bio-Rad, Inc., Berkeley, CA, USA).

All data were processed by the SPSS 16.0 software and presented as the mean ± SE. Analysis of variance (ANOVA) and LSD tests were used for comparisons among the groups and between the paired data, respectively. When the data were not normally distributed, the Mann-Whitney U test and the one-way non-parametric ANOVA (Kruskal-Wallis test) were used to compare quantitative variables between two groups and among more than two groups, respectively. A p value of less than 0.05 was considered to be statistically significant.

Increased serum ALT and AST levels (P <0.01) and confluent necrotic foci were observed at 10 h after exposure to Con A (Fig. 1). Fig. 1a shows that emodin administered 2 h before Con A at concentrations of 1.5625 mg/kg, 3.125 mg/kg, 6.25 mg/kg, 12.5 mg/kg, 25 mg/kg, and 50 mg/kg led to reduced serum ALT and AST levels in the mice (P <0.05 and P <0.01, respectively). However, small and scattered necro-inflammatory foci with polymorphonuclear cell infiltration were observed when emodin was administered at concentrations of 1.5625 mg/kg, 3.125 mg/kg, 6.25 mg/kg, while only sparse polymorphonuclear leukocyte infiltration and almost no necrotic foci were observed when emodin was given at concentrations of 12.5 mg/kg, 25 mg/kg and 50 mg/kg (Fig. 1b, c). However, there were no significant differences in serum ALT levels or AST levels when emodin was given at concentrations of 12.5 mg/kg, 25 mg/kg and 50 mg/kg.

Increased serum ALT and AST levels (P <0.01) and large areas of necrosis were observed at 10 h after Con A administration (Fig. 2), which could be reversed by emodin administered 2 h before the Con A challenge (P <0.01). However, Fig. 2 also demonstrates that emodin was not able to alleviate Con A-induced liver injury when given at 30 min, 60 min, 90 min, and 2 h after the Con A challenge (P>0.05).

Fig. 3 shows that the levels of hepatic IFN-γ, TNF-α, IL-1β, IL-6, IL-12, and inducible nitric oxide synthase (iNOS) mRNA in Con A-induced mice were significantly higher than those in the normal mice (P <0.01). However, emodin ingestion dose-dependently restored hepatic IFN-γ, TNF-α, IL-1β, IL-6, IL-12, and iNOS mRNA levels in Con A-induced mice (P <0.05 or P <0.01).

As shown in Fig. 4, the Con A-induced mice showed higher mRNA levels of hepatic integrin alpha M (ITGAM), chemokine (CC motif) ligand 2 (CCL2), macrophage inflammatory protein 2 (MIP-2), and its receptor, chemokine (CXC motif) receptor 2 (CXCR2) compared with the normal mice (P <0.01). However, emodin administration dose-dependently restored hepatic ITGAM, CCL2, MIP-2 and CXCR2 mRNA levels in Con A-induced mice (P <0.05 or P <0.01).

To determine whether p38 MAPK and NF-κB are involved in the protective effects of emodin on liver impairment in Con A-induced mice, the hepatic p38 MAPK and NF-κB activity were examined. As shown in Fig. 5, although there were no significant differences in hepatic basic p38 MAPK and NF-κB protein levels in the control and Con A-induced groups, the hepatic phosphorylated levels of p38 MAPK and NF-κB were increased in Con A-induced mice (P <0.05 or P <0.01). However, emodin administration reversed the Con A- induced increase of hepatic phosphorylated p38 MAPK and NF-κB protein levels in the mice (P <0.05 or P <0.01).

The effects of emodin on the infiltration of CD4⁺ T cells and F4/80⁺ macrophages in the livers of Con A-induced mice were determined by immunofluorescence. Fig. 6 shows that the numbers of CD4⁺ T cells and F4/80⁺ macrophages increased significantly in the livers after Con A injection (P <0.05 or P <0.01) but could be returned to basal levels following emodin treatment (P <0.05).

As shown in Fig. 7, some toxicity was observed when Con A was given at 31.25 μg/ml in RAW264.7 cells, but no toxic effect were found in EL4 cells even when cells were exposed to Con A at 100 μg /ml. The viability of RAW264.7 cells and EL4 cells was not affected by 24 h incubation with emodin at concentrations of up to 12.5 μg/ml and 25 μg /ml, respectively (Fig. 7). Based on the results of MTT assay and a previous study [23], 15.63 μg/ml and 4 μg/ml of Con A were used in RAW264.7 and EL4 cells, respectively; and emodin at concentrations of 12.5 μg/ml and 25 μg /ml were used in RAW264.7 and EL4 cells in our next study, respectively.

Fig. 8 shows that the level of TNF-α induced by Con A in RAW 264.7 cells increased significantly, as compared with control group (P <0.01). Moreover, the level of TNF-α of Con A plus emodin group were obviously lower than those of Con A group (P <0.01). In Fig. 8, Con A-induced production of IL-2 in EL4 cells was significantly (P <0.01) suppressed by emodin. These in vitro results explained that emodin could reduce the inflammatory factors induced by Con A in RAW264.7 and EL4 cells.

Fig. 9 and 10 show that although there were no significant differences in unphosphorylated p38 MAPK and NF-κB protein levels in the control and Con A-induced groups, the phosphorylated levels of p38 MAPK and NF-κB were increased in Con A-induced cells (P <0.05 or P <0.01). However, emodin administration reversed the Con A- induced increase in phosphorylated p38 MAPK and NF-κB protein levels in the two cells (P <0.05 or P <0.01).

Emodin is the main effective component of rhubarb, and it has not been demonstrated to be associated with significant deleterious side effects. Dang SS et al. described the inhibitory effects of emodin in combination with astragalus polysaccharide on the replication of HBV in vitro [24]. However, little is known regarding the preventive role of emodin in fulminant hepatitis. In the present study, the mice with Con A- induced hepatitis showed high serum levels of ALT and AST concomitant with hepatic injury. However, treatment with emodin alone reduced the serum levels of ALT and AST and hepatic levels of IFN-γ, TNF-α, IL-1β, IL-6, IL-12, iNOS, ITGAM, CCL2, MIP-2 and CXCR2 and attenuated hepatic injury in the mice with Con A- induced hepatitis. Also, the results of our study demonstrated that the inhibition of these pro-inflammatory cytokines and chemokines may be associated with the suppression of the infiltration of CD4⁺ T cells and macrophages as well as the activation of p38 MAPK-NF-κB signaling pathway in the two cells.

Many studies have shown that immunoregulatory cytokines play a central role in the pathogenesis of Con A-induced hepatitis, and previous studies have reported the pro-inflammatory roles of the cytokines TNF-α, IL-1β, IFN-γ, IL-12, and IL-6 [12,13,14,25]. The up-regulation of pro-inflammatory cytokines, such as IFN-γ and TNF-α, by Con A injection was shown to directly induce hepatocellular apoptosis and necrosis [26,27]. iNOS is mainly present in macrophages and hepatocytes and can be induced by TNF-α or IFN-γ in many types of immune cells [28,29].

Chemokines are small heparin-binding proteins that direct the movement of circulating leukocytes to sites of inflammation or injury. Chemokines and their receptors are now known to play crucial roles in engendering the adaptive immune response and contributing to the pathogeneses of a variety of diseases [30]. ITGAM, which is also known as cluster of differentiation molecule 11B (CD11B), mediates the inflammatory response by regulating leukocyte adhesion and migration and has been implicated in several immune processes, such as phagocytosis, cell-mediated cytotoxicity, chemotaxis and cellular activation [31]. CCL2 is also referred to as monocyte chemotactic protein-1 (MCP-1). It recruits monocytes, memory T cells, and dendritic cells to sites of inflammation caused by tissue injury or infection [32,33]. The murine CXC chemokine MIP-2 is a functional analog of human IL-8. Similar to IL-8, it acts as a potent neutrophil chemoattractant.

The current study revealed the overexpression of hepatic IFN-γ, TNF-α, IL-1β, IL-12, IL-6, and iNOS in Con A-induced mice. However, emodin ingestion decreased the expression of these cytokines, which paralleled improvements in serum ALT and AST levels and reduced hepatic injury. The increased production of ITGAM, CCL2, MIP-2, and its receptor, CXCR2, in the livers of Con A-induced mice was also found in the present study, which could be attenuated by emodin treatment. These observations indicate that emodin reduced serum ALT and AST overexpression and liver injury through its inhibition of hepatic pro-inflammatory cytokines and chemokines.

As an important member of the MAPK family, p38 MAPK is considered to be closely related to the regulation of inflammation, stress and apoptosis. Wang et al. analyzed the structures of emodin and the p38 MAPK protein using molecular designing software and found that some of the chemical groups of emodin bind to the dewatering groups of the p38 MAPK protein, suggesting the possibility that emodin directly binds to p38 MAPK and inhibits its activity [34]. The activated form of p38 has been associated with the activation of NF-κB [35,36,37]. Inactive NF-κB (mainly p65/p50 dimers) is sequestered in the cytoplasm. Once it becomes phosphorylated and activated, it translocated to the nucleus where it binds to the promoters of specific target genes and up-regulates the transcription of cytokines. NF-κB plays a central role in the regulation of diverse biological processes, including the immune response, development, cell growth and survival [38,39]. Previous studies have shown that NF-κB regulates the release of inflammatory cytokines in Con A-induced hepatitis, and the blockade of NF-κB activation can reduce the expression of these cytokines [40,41]. Thus, we hypothesized that emodin exerted its protective effects on Con A-induced hepatitis by inhibiting the activation of the p38 MAPK-NF-κB signaling pathway. The current study revealed that the activation of p38 MAPK and NF-κB protein were augmented in the livers of mice and that emodin treatment significantly down-regulated these activations. These data suggest that the p38 MAPK-NF-κB pathway is involved in Con A-induced injury.

The recruitment of inflammatory cells to the liver by chemokines and the subsequent production of cytokines are the main factors for the pathogenesis and development of Con A-induced hepatitis, which is primarily characterized as T cell-mediated hepatic damage. Mounting evidence suggests that CD4⁺ T cells are the predominant lymphocytes recruited to liver after the intravenous administration of Con A and that these cells play an important role in Con A-induced hepatitis because their depletion prevents liver damage following Con A injection [14]. Macrophages also play an important role in the induction of Con A-induced hepatitis, as confirmed by depletion treatment [42]. Thus, the suppressions of CD4⁺ T lymphocytes and macrophage infiltration are therapeutic targets for Con A-induced hepatitis. In our study, the numbers of infiltrating CD4⁺ T and F4/80⁺ cells in the livers of the mice with Con A- induced hepatitis increased markedly but could be reduced by pretreatment with emodin. Also, in vitro studies showed that the activation of p38 MAPK and NF-κB protein were augmented in Con A-induced CD4⁺ T cells and macrophages and that emodin treatment significantly down-regulated these activations. Therefore, CD4⁺ T cells and macrophages are therapeutic targets of emodin that act in the pathogenesis of Con A-induced hepatitis in mice and are the main cells involved in p38 MAPK and NF-κB phosphorylation.

In summary, the present study first demonstrated that the targeting of emodin to the hepatic infiltration of CD4⁺ T cells and macrophages and the activation of p38 MAPK-NF-κB pathway in the two cells prevented the Con A-induced elevation of ALT and AST levels and hepatic injury. The ability of emodin to inhibit activation of the p38 MAPK-NF-κB pathway resulted in the suppression of hepatic IFN-γ, TNF-α, IL-1β, IL-12, IL-6, iNOS, ITGAM, CCL2, and MIP-2, as well as the MIP-2 receptor, CXCR2. Emodin may have therapeutic applications in the prevention of fulminant hepatitis.

This work was supported by the State S&T Projects of 12th Five Year (2012ZX10002007) and the Natural Science Foundation of China (81000730).

The authors declare that they have no conflict of interest.

F. Chen contributes to this work equally.