IL-20 and IL-20R1 antibodies protect against liver fibrosis

Authors

  • Yi-Shu Chiu,

    1. Institute of Clinical Pharmacy and Pharmaceutical Sciences
    Search for more papers by this author
  • Chi-Chen Wei,

    1. Department of Biochemistry and Molecular Biology, College of Medicine, National Cheng Kung University, Tainan, Taiwan
    2. Medicine and Cosmetics Testing Group, Verification Technology Department, Plastic Industry Development Center, Taiwan
    Search for more papers by this author
  • Yih-Jyh Lin,

    1. Department of Surgery, National Cheng Kung University College of Medicine and Hospital, Tainan, Taiwan
    Search for more papers by this author
  • Yu-Hsiang Hsu,

    1. Department of Biochemistry and Molecular Biology, College of Medicine, National Cheng Kung University, Tainan, Taiwan
    Search for more papers by this author
  • Ming-Shi Chang

    Corresponding author
    1. Institute of Clinical Pharmacy and Pharmaceutical Sciences
    2. Department of Biochemistry and Molecular Biology, College of Medicine, National Cheng Kung University, Tainan, Taiwan
    • Address reprint requests to: Prof. Ming-Shi Chang, Department of Biochemistry and Molecular Biology, National Cheng Kung University, College of Medicine, 138 Sheng-Li Road, Tainan 704, Taiwan. E-mail: mschang@mail.ncku.edu.tw; fax: +886-6-274-1694.

    Search for more papers by this author

  • Potential conflict of interest: Nothing to report.

Abstract

Interleukin (IL)-20 is a proinflammatory cytokine of the IL-10 family and involved in rheumatoid arthritis, atherosclerosis, stroke, and osteoporosis. However, the pathophysiological roles of IL-20 in liver injury have not been extensively studied. We explored the involvement of IL-20 in liver injury and the therapeutic potential of IL-20 antagonists for treating liver fibrosis. Compared with normal liver tissue from healthy individuals, the amount of IL-20 was much higher in hepatocytes and hepatic stellate cells in liver biopsies from patients with fibrosis, cirrhosis, and hepatocellular carcinoma. Carbon tetrachloride (CCl4) treatment induced IL-20 that further up-regulated the expression of transforming growth factor (TGF)-β1 and p21WAF1 and resulted in cell cycle arrest in the Clone-9 rat hepatocyte cell line. IL-20 activated quiescent rat hepatic stellate cells (HSCs) and up-regulated TGF-β1 expression. IL-20 also increased TGF-β1, tumor necrosis factor (TNF)-α, and type I collagen (Col-I) expression, and promoted the proliferation and migration of activated HSCs. Serum IL-20 was significantly elevated in mice with short-term and long-term CCl4-induced liver injury. In mice with short-term liver injury, anti-IL-20 monoclonal antibody (7E) and anti-IL-20 receptor (IL-20R1) monoclonal antibody (51D) attenuated hepatocyte damage caused by CCl4, TGF-β1, and chemokine production. In mice with long-term liver injury, 7E and 51D inhibited CCl4-induced cell damage, TGF-β1 production, liver fibrosis, HSC activation, and extracellular matrix accumulation, which was caused by the reduced expression of tissue inhibitors of metalloproteinases as well as increased metalloproteinase expression and Col-I production. IL-20R1-deficient mice were protected from short-term and long-term liver injury. Conclusion: We identified a pivotal role of IL-20 in liver injury and showed that 7E and 51D may be therapeutic for liver fibrosis. (Hepatology 2014;60:1003–1014)

Abbreviations
51D

anti-IL-20R1 monoclonal antibody

7E

anti-IL-20 monoclonal antibody

α-SMA

alpha-smooth-muscle actin

AST

aspartate aminotransferase

ALT

alanine aminotransferase

CCl4

carbon tetrachloride

CDK

cyclin-dependent kinase

ConA

concanavalin A

ECM

extracellular matrix

HSC

hepatic stellate cell

IL

interleukin

KC

keratinocyte chemoattractant CXCL1

MCP-1

monocyte chemotactic protein-1

MIP-2β

macrophage inflammatory protein 2 beta

MMPs

matrix metalloproteinases

p21WAF1

cyclin-dependent kinase (CDK) inhibitor

PCNA

proliferating cell nuclear antigen

qRT-PCR

quantitative real-time polymerase chain reaction

TGF-β1

transforming growth factor-beta 1

TIMPs

tissue inhibitors of metalloproteinases

TNF-α

tumor necrosis factor-alpha

Liver disease is one of the important worldwide causes of morbidity and mortality. Hepatic viral infections, injury, exposure to drugs or toxic compounds, and autoimmune processes have been implicated in the development of liver diseases. Hepatic responses to injury have various phases and involve various hepatic cell types. The initial event is liver epithelial-cell stress, which causes necrosis or apoptosis,[1-3] and then the activation of hepatic stellate cells (HSCs). Subsequently, an inflammatory response is induced, which leads either to tissue regeneration and repair in an acute phase, or to fibrogenesis, cirrhosis, and, finally, hepatocellular carcinoma when the injury is prolonged.[4]

Liver injury is characterized by persistent hepatocyte damage and death.[1-3] Transforming growth factor (TGF)-β1 inhibits the growth of hepatocytes by arresting the cell cycle at the G1/S phase.[5] TGF-β1-induced growth arrest in hepatocytes is partly mediated by expression of the cyclin-dependent kinase (CDK) inhibitor p21WAF1.[6]

In response to liver injury, quiescent HSCs are activated and develop a myofibroblast-like phenotype that proliferates and expresses intermediate filament alpha-smooth-muscle actin (α-SMA) and profibrotic genes.[7] HSC activation, a key event in the pathogenesis of liver fibrosis, is mediated by various cytokines released from the damaged hepatocytes or from activated Kupffer cells. The events subsequent to HSC activation, including the increased production of extracellular matrix (ECM) components and increased proliferation, are crucial for the hepatic fibrogenesis cascade.

A complex inflammatory response triggered by liver injury is executed by a variety of cells. The process is initiated by the mutual activation of HSCs and Kupffer cells that together provide a cytokine/chemokine milieu that induces the infiltration of monocytes. Cytokines and chemokines from activated macrophages, in turn, stimulate lymphocytes, thus creating a persistent inflammatory response.[7, 8]

Liver fibrosis is the pathologic result of ongoing chronic liver injury and is characterized by impaired hepatocyte proliferation[9]; activated HSCs are responsible for wound closure and fibrosis in persistent liver injury.[9-11] TGF-β1 is required for liver fibrosis, which up-regulates the production of the major ECM constituents[12] and alters the balance between matrix metalloproteinases (MMPs) and the tissue inhibitors of metalloproteinases (TIMPs).[9-11]

Interleukin (IL)-20 is a member of the IL-10 family, which includes IL-10, -19, -20, -22, -24, and -26. IL-20 signals through two types of receptor (R) complexes: IL-20R1/IL-20R2 and IL-22R1/IL-20R2.[13] IL-20 targets keratinocytes,[14] endothelial cells,[15, 16] synovial fibroblasts,[17, 18] renal epithelial and mesenchymal cells,[19-21] osteoblasts and osteoclasts,[22] and several types of tumor cells, especially in squamous cell carcinoma of the skin, tongue, esophagus, and lung.[23] IL-20 is involved in inflammation, angiogenesis, arteriogenesis, chemotaxis, and osteoclastogenesis, all of which are important for the pathogenesis of psoriasis,[14] atherosclerosis,[15] rheumatoid arthritis,[17, 18] ischemic disorders,[24] and osteoporosis.[22]

IL-20 is recognized as a dominant mediator and may be a potential therapeutic target in many inflammatory diseases. The role of IL-20 in liver disease has not been established. In this study, we explored the detrimental effects of IL-20 on liver injury based on analysis from human specimens and a murine model of carbon tetrachloride (CCl4)-induced liver injury. We hypothesized that antibodies against IL-20 and IL-20R1 have therapeutic potential to treat liver injury and to prevent liver fibrosis.

Materials and Methods

Full Methods

Detailed methods and associated references are described in the Supporting Methods.

Clinical Specimens

The protocol of clinical study conformed to the ethical guidelines of the 1975 Declaration of Helsinki and was approved by the Ethics Committee of National Cheng Kung University Hospital. Liver biopsies were collected from 66 patients. Written informed consent was obtained from each patient. Three healthy specimens were purchased from SuperBioChips Laboratories (Seoul, Korea). Information about the clinical specimens is listed in Supporting Tables 1 and 2.

Cell Culture

A healthy cell line, rat hepatocyte Clone-9 cells, was purchased from the American Type Culture Collection (Manassas, VA). Primary rat HSCs and hepatocytes were isolated from male Sprague-Dawley rats.

Antibodies Preparation

Anti-IL-20 monoclonal antibody (7E) was prepared as previously described.[14] Anti-IL-20 receptor (IL-20R1) monoclonal antibody (51D) was generated with standard protocols.

IL-20-Induced Cell Responses and the Neutralizing Abilities of 7E and 51D

For a functional assay of IL-20, Clone-9 cells, rat hepatocytes, and rat HSCs were treated with IL-20 (200 ng/mL), 7E (2 μg/mL), IL-20 (200 ng/mL) plus 7E (2 μg/mL), 51D (2 μg/mL), IL-20 (200 ng/mL) plus 51D (2 μg/mL) for various time periods and analyzed in different assays.

Polymerase Chain Reaction (PCR) and Quantitative Reverse-Transcribed (q-RT)-PCR

The total RNA of cells and liver tissue samples was isolated (Invitrogen, Carlsbad, CA) and reverse transcription was done (PrimeScript RT-PCR kit; ClonTech, Palo Alto, CA).

Animal Models of Liver Injury

In CCl4-induced liver injury model, C57BL/6J, IL-20R1+/+, and IL-20R1−/− (IL-20R1-deficient) male mice used in this study were between 10 and 12 weeks old (n = 5 per experimental group). Short-term liver injury was induced using a single intraperitoneal (i.p.) injection of CCl4 (1 mL/kg of body weight [1:5 v/v in mineral oil]); long-term liver injury was induced using two injections a week for 8 or 12 weeks. Animal models of concanavalin A (ConA)-induced and high-fat diet-induced liver injury are described in the Supporting Methods

Antibody Treatments

For short-term CCl4-induced liver injury, C57BL/6J mice were injected (i.p.) with 3 mg/kg of 7E once per day or with 6 mg/kg of 51D once per day for 3 days. The first 7E injection was given 1 hour after CCl4 treatment and the first 51D injection was given 1 hour before CCl4 treatment. For long-term CCl4 and high-fat diet-induced liver injury, the mice were injected with 3 mg/kg of 7E twice a week after CCl4 treatment or with 6 mg/kg of 51D twice a week before CCl4 treatment. Details of antibody treatment in ConA-induced and high-fat diet-induced liver injury are given in the Supporting Methods.

Statistical Analysis

All data are means ± standard error of the mean (SEM). Prism 5.0 (GraphPad Software) was used for the statistical analysis. A one-way analysis of variance (ANOVA) test (Kruskal-Wallis test) was used to compare the data between groups. Post-hoc comparisons were done using Dunn's multiple comparison tests. P < 0.05 was considered significant.

Results

IL-20 Was Highly Expressed in Injured Liver Tissue

To investigate the clinical implications and the underlying mechanism of IL-20 in liver diseases, we used immunohistochemical staining to analyze IL-20 expression in liver biopsies from patients with various liver diseases. IL-20 protein was more highly expressed in the HSCs (black arrows) and hepatocytes (white arrows) of patients with liver fibrosis (Fig. 1B), liver cirrhosis (Fig. 1C), and hepatocellular carcinoma (Fig. 1D) than in the liver tissue of healthy controls (Fig. 1A). Immunofluorescence staining showed costaining of IL-20 and α-SMA (Supporting Fig. 1), which indicated that IL-20 was expressed in liver HSCs of patients with cirrhosis. We also analyzed the mRNA expression of IL-20 in noncirrhotic and cirrhotic liver tissue using qRT-PCR, which showed that IL-20 expression was significantly higher in cirrhotic than in noncirrhotic liver tissue (Fig. 1E). These results indicated that IL-20 was associated with liver cirrhosis.

Figure 1.

Increased IL-20 expression in the liver of patients with liver diseases. Paraffin sections of liver biopsies were obtained from (A) healthy volunteers (n = 3) and patients with (B) liver fibrosis (n = 3), (C) cirrhosis (n = 3), and (D) hepatocellular carcinoma (n = 3). IL-20 was detected using immunohistochemical (IHC) staining with 7E. The reaction was detected using AEC chromogen stain (red), and the nuclei were counterstained with hematoxylin (blue). Black arrows: HSCs; white arrows: hepatocytes. (E) qRT-PCR was used to analyze IL-20 expression in 21 noncirrhotic and 19 cirrhotic liver tissue samples. Data are means ± SEM. *P < 0.05 compared with the noncirrhotic group. Data are representative of three independent experiments.

TGF-β1 Expression, Cell Cycle Arrest, and p21WAF1 Expression Were Up-Regulated in IL-20-Treated Clone-9 Cells

To explore how up-regulated IL-20 in the liver is involved in the pathogenesis of liver injury, we analyzed the expression of IL-20 and its receptors in Clone-9 cells, a normal rat hepatocyte cell line. The results of qRT-PCR (Supporting Fig. 2A) and immunocytochemical staining (Supporting Fig. 2B) showed that IL-20 and its receptors were expressed in Clone-9 cells. Hepatocytes were therefore identified as an IL-20 target that can respond to IL-20 in an autocrine or a paracrine manner.

We treated Clone-9 cells with CCl4 to establish a liver injury model in vitro and then studied the involvement of IL-20 in cytotoxicity. IL-20 expression was up-regulated in Clone-9 cells after 4 hours of CCl4 treatment (Fig. 2A). Because TGF-β1 is important for regulating a variety of processes in liver fibrosis, we analyzed the effect of CCl4-up-regulated IL-20 expression on TGF-β1 production in hepatocytes and the inhibitory effects of 7E and 51D. 7E has previously shown specificity and neutralization activity in vitro and in vivo.[22, 24, 27, 30, 31] 51D not only showed specificity against the extracellular domain of human (h)IL-20R1,[25] but also cross-reacted with mouse (m)IL-20R1 and rat (r)IL-20R1 (Supporting Fig. 3). TGF-β1 mRNA expression was induced in IL-20-treated cells, the activity of which was significantly inhibited in cells treated with 7E and 51D (Fig. 2B).

Figure 2.

IL-20 induced TGF-β1 and p21WAF1 expression and cell cycle arrest in Clone-9 cells. (A) Clone-9 cells were incubated with CCl4 (10 μM) for the indicated times. IL-20 expression was analyzed using qRT-PCR. (B) TGF-β1 mRNA expression, (C) cell cycle distribution, and (D) p21WAF1 expression of Clone-9 cells treated with IL-20 and 7E or 51D were analyzed using qRT-PCR, flow cytometry, and western blotting. Data are means ± SEM. *P < 0.05 compared with untreated cells; **P < 0.05 compared with IL-20-treated cells. Data are representative of three independent experiments.

The percentage (77.4%) of hepatocytes arrested in the G0/G1 phase after 24 hours of IL-20 treatment was significantly higher in the IL-20-treated group. IL-20-induced G0/G1-phase cell cycle arrest was partially inhibited in the IL-20+7E-treated group (63%) and in the IL-20+51D group (66.1%) (Fig. 2C). Furthermore, cell cycle inhibitor p21WAF1 expression was higher in IL-20-only-treated Clone-9 cells, but mostly inhibited in cells treated with 7E and 51D with IL-20 (Fig. 2D).

IL-20 showed the same effects on primary rat hepatocytes as on Clone-9 cells (Supporting Fig. 4). Isolated rat T cells were used as controls. The T cells were unaffected by IL-20 treatment (data not shown). These results suggested that hepatotoxin-induced IL-20 increased TGF-β1 production caused by the up-regulation of p21WAF1 and prevented the growth of hepatocytes, which promoted the progression of liver fibrosis.

Primary Rat HSCs Were Activated, Proliferated, and Then Migrated After IL-20 Treatment

Activated HSCs that express α-SMA are responsible for liver fibrosis because they deposit excessive ECM and increase the contractility of fibroblasts mediated by TGF-β1.[9-11] Therefore, we first determined whether IL-20 was involved in activating HSCs. Primary rat HSCs cultured for less than 48 hours were the “quiescent” phenotype. In contrast, primary HSCs cultured for 10 to 14 days were the “activated” phenotype.[27] qRT-PCR was used to analyze the gene expression of IL-20 and α-SMA of the HSCs cultured for 1 day (p0), 7 days (p1), and 20 days (p3). IL-20 expression was 100% higher on p3 than on p0; α-SMA mRNA expression was 100% higher on p1 than on p0, but 800% higher on p3 than on p0 during the activation of quiescent HSCs (Fig. 3A). qRT-PCR analysis also showed that TGF-β1 mRNA and α-SMA expression was significantly induced in quiescent HSCs treated with IL-20, but this activity was inhibited in cells treated with 7E and 51D (Fig. 3B).

Figure 3.

The effects of IL-20 on quiescent and activated HSCs. (A) Quiescent HSCs were freshly isolated and subcultured for 1 day (p0), 7 days (p1), or 20 days (p3) to analyze the expression levels of IL-20 and α-SMA using qRT-PCR. *P < 0.05 compared with p0. (B) Quiescent HSCs were exposed to IL-20 with or without 7E and 51D. qRT-PCR was used to analyze the expression of TGF-β1 and α-SMA. (C) TGF-β1, TNF-α, and Col-I mRNA expression in activated HSCs treated with IL-20 and 7E or 51D was analyzed using qRT-PCR. (D) Light microscopy was used to capture micrographs of migrated cells. The migration of activated HSCs treated with IL-20 and 7E or 51D was analyzed using a modified Boyden chamber. (E) The number of cells was counted. The results are expressed as a mean of 15 randomly selected fields. (F) The proliferation of activated HSCs treated with IL-20 and 7E or 51D was analyzed using an MTT assay. Data are means ± SEM. *P < 0.05 compared with untreated cells; **P < 0.05 compared with IL-20-treated cells. Data are representative of three independent experiments.

RT-PCR and immunofluorescence staining were used to confirm the expression of IL-20 and its receptors on activated HSCs that had been cultured for 20 days (p3). The results of RT-PCR (Supporting Fig. 5A) and immunofluorescence staining (Supporting Fig. 5B) showed that IL-20 and its receptors were expressed in activated HSCs. qRT-PCR showed that the mRNA expression of the inflammatory cytokine TNF-α, TGF-β1, and the ECM component Col-I was significantly higher in activated HSCs treated with IL-20 only but inhibited in cells treated with 7E and 51D (Fig. 3C).

Because the proliferation and migration of activated HSCs are critical to liver fibrosis, we analyzed the proliferation and migration of IL-20-treated activated HSCs using a modified Boyden chamber and a 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay. Migration (Fig. 3D,E) and proliferation (Fig. 3F) were significantly higher in IL-20-treated activated HSCs but inhibited in cells treated with 7E and 51D.

Short-Term CCl4-Induced Liver Injury Was Attenuated in 7E- and 51D-Treated Mice

C57BL/6J mice were injected (i.p.) with CCl4 to induce short-term liver injury for 3 days. They were then analyzed to confirm both the involvement of IL-20 in hepatocyte damage and the protective effects of 7E and 51D against short-term liver injury. Serum levels of IL-20 were significantly higher in CCl4-treated mice than in the mineral oil-treated control group mice 72 hours after CCl4 treatment (Fig. 4A).

Figure 4.

The protective effects of 7E and 51D on short-term CCl4-induced liver injury. C57BL/6J mice were treated with mineral oil and CCl4 (Ctrl) with or without mIgG, 7E, or 51D. The mice were killed 72 hours after CCl4 treatment. (A) Serum levels of IL-20 in mice 72 hours after CCl4 treatment were analyzed using ELISA. Data are means ± SEM. *P < 0.05 compared with mineral oil treatment. (B) Serum AST and ALT in mice were measured 24, 48, and 72 hours after CCl4 treatment. (C) The TGF-β1 mRNA expression in liver tissue was analyzed using qRT-PCR. (D) Serum levels of TGF-β1 of mice were analyzed using ELISA 72 hours after treatment. (E) Positive IHC staining of PCNA mAb represents proliferating hepatocytes in the liver tissue of mice. PCNA-positive cells in more than 12 mm2 tissue sections were counted for each mouse. Data are means ± SEM. *P < 0.05 compared with CCl4 treatment (Ctrl). Data are representative of three independent experiments.

To confirm the protective effects of 7E and 51D on mice with short-term CCl4-induced liver injury, the mice were treated (i.p.) with 7E and 51D after CCl4 injection. Serum levels of aspartate aminotransferase (AST) and alanine amino transferase (ALT) were measured 24, 48, and 72 hours after the antibody treatment. Serum levels of ALT and AST peaked at 24 hours. The levels of ALT and AST were significantly lower in the 7E- and 51D-treated mice than in the mIgG-treated and control groups (Fig. 4B), which indicated markedly less hepatic injury. TGF-β1 expression in liver tissue was significantly lower in both 7E- and 51D-treated groups (Fig. 4C). However, serum levels of TGF-β1 were significantly (P < 0.05) lower in 7E- but not in 51D-treated mice (Fig. 4D).

Immunohistochemical staining with proliferating cell nuclear antigen (PCNA) was used to analyze hepatocyte proliferation in liver sections from the mice 72 hours after they had been treated with CCl4 and 7E or 51D. PCNA-positive cells were more numerous in the liver sections of mice treated with 7E and 51D after CCl4 than in the liver sections of controls and mIgG-treated mice (Fig. 4E).

Furthermore, we also established the mice model of ConA-induced immune-mediated liver injury to clarify the specific effects of blocking IL-20 during liver recovery. 7E treatment provided partial protective effects on ConA-induced immune-mediated liver damage (Supporting Fig. 6).

Studies analyzing the acute effect of CCl4 on the liver have also found an early production of chemical mediators, which might be essential for the infiltration of inflammatory cells into the parenchyma and the activation of resident macrophages.[28, 29] Therefore, we also used qRT-PCR to determine the expression of the chemokines MCP-1, keratinocyte chemoattractant CXCL1 (KC), and macrophage inflammatory protein 2 beta (MIP-2β) in liver tissue. The expression of MCP-1 and KC in mice treated with short-term CCl4 was significantly attenuated in 7E- and 51D-treated mice; however, the expression of MIP-2β was only significantly attenuated in 7E-treated mice (Supporting Fig. 7).

Long-Term CCl4-Induced Liver Injury and Metabolic Liver Injury Was Attenuated in 7E- and 51D-Treated Mice

To determine the effect of 7E and 51D on long-term CCl4-induced liver injury, after 8 weeks of CCl4 treatment the mice were killed and analyzed. Serum levels of IL-20 were significantly higher in CCl4-treated mice than in the mineral oil-treated control groups (Fig. 5A).

Figure 5.

The protective effects of 7E and 51D on long-term CCl4-induced liver injury. C57BL/6J mice were treated with mineral oil or with CCl4 alone to induce liver injury (Ctrl), or with mIgG, 7E, or 51D. The mice were killed 8 weeks after treatment. (A) Serum levels of IL-20 in mice 8 weeks after CCl4 treatment were analyzed using ELISA. Data are means ± SEM. *P < 0.05, compared with mineral oil administration. (B) Serum AST and ALT levels in mice were measured 8 weeks after CCl4 treatment. (C) The TGF-β1 mRNA expression in the liver tissue of mice was analyzed using qRT-PCR 8 weeks after CCl4-induced liver injury. Serum TGF-β1 levels of mice were analyzed using ELISA 8 weeks after treatment. (D) H&E and Sirius red staining of liver sections from mice were used to assess the effects of CCl4 with or without 7E and 51D. (E) Relative collagen content (per g of liver) was measured using hydroxyproline biochemical determinations. Data are means ± SEM. *P < 0.05 compared with CCl4 treatment (Ctrl). Data are representative of three independent experiments.

The levels of liver injury were analyzed to confirm the protective effects of 7E and 51D. Serum levels of AST and ALT were significantly lower in 7E- and 51D-treated groups than in the control and mIgG-treated groups (Fig. 5B).

qRT-PCR and enzyme-linked immunosorbent assay (ELISA) results showed that TGF-β1 mRNA expression levels in liver tissue were significantly lower in the 7E- and 51D-treated groups, but TGF-β1 protein expression levels in serum were significantly lower only in the 7E-treated group (Fig. 5C).

The morphological changes of liver injury and fibrosis caused by long-term CCl4 treatment were visualized in liver sections stained with hematoxylin and eosin (H&E) (Fig. 5D; left) and Sirius red (Fig. 5D; right). 7E- and 51D-treated mice with long-term liver injury showed mild liver fibrosis compared with the control and mIgG-treated groups (Fig. 5D). To quantify the degree of fibrosis accurately, collagen deposition was biochemically determined as hydroxyproline content in liver samples. After 8 weeks, the liver hydroxyproline content, which reflects the amount of collagen, was significantly lower in the 7E- and 51D-treated groups than in the controls and mIgG-treated group (Fig. 5E).

qRT-PCR, used to analyze α-SMA mRNA expression to represent activation of quiescent HSCs during the process of liver fibrosis, showed that HSC activation was significantly lower in 7E- and 51-D-treated mice than in the controls and the mIgG-treated group (Supporting Fig. 8).

Activated HSCs overexpress tumor necrosis factor (TNF)-α, monocyte chemotactic protein (MCP)-1, Col-I, and TIMP-1 and -2. qRT-PCR showed that the expression of TNF-α, MCP-1, Col-I, TIMP-1, and TIMP-2 were significantly higher and the expression of MMP-2 and MMP-12 were significantly lower in the liver tissue of the control and mIgG group mice than in the 7E group mice after long-term CCl4 treatment (Supporting Fig. 8). Expression levels of MMP-12, TIMP-1, and TIMP-2, however, were not significantly different in the 51D group and in the control and mIgG groups (Supporting Fig. 8).

Different doses of 7E (1 mg/mL, 3 mg/mL, and 9 mg/mL) and 51D (2 mg/mL, 6 mg/mL, and 10 mg/mL) were used to treat mice with long-term CCl4-induced liver injury. The levels of serum ALT, AST, and TGF-β1 in CCl4-treated mice were dose-dependently lower after 7E and 51D treatment (Supporting Fig. 9A,B). H&E staining showed that mice of long-term liver injury and treated with all three doses of 7E or 51D had milder liver fibrosis than did the control and mIgG-treated groups (Supporting Fig. 9C).

Moreover, we evaluated the protective effects of 7E on high-fat diet mice. Mice treated with 7E had lower serum ALT and AST than did the mIgG control group (Supporting Fig. 10A). Liver sections with H&E staining showed that liver injury was significantly alleviated in high-fat diet mice after 7E treatment (Supporting Fig. 10B).

Therapeutic Potential of 7E and 51D for Liver Injury

To analyze the therapeutic potential of 7E and 51D in long-term CCl4-induced liver injury, we used 7E or 51D treatment in mice from the 5th week to the 8th week after CCl4 treatment. Eight weeks after CCl4 treatment, the levels of serum ALT, AST, and TGF-β1 were significantly lower in the 7E- and 51D-treated mice than in the mIgG-treated mice (Supporting Fig. 11A,B). The results suggested the therapeutic potential of 7E and 51D for liver injury. H&E staining also showed that CCl4-induced liver fibrosis was inhibited in the groups treated with 7E and 51D compared with control and mIgG-treated groups (Supporting Fig. 11C).

IL-20R1−/− Mice Were Resistant to Short-Term CCl4-Induced Liver Injury

We have generated an IL-20R1-deficiency mice.[22] To further confirm the importance of IL-20 in liver injury, we analyzed if IL-20R1−/− mice were protected from CCl4-induced injury. IL-20R1+/+ and IL-20R1−/− mice were treated with CCl4 to induce short-term liver injury for 3 days. Serum ALT and AST levels (Fig. 6A), TGF-β1 mRNA levels in liver tissue (Fig. 6B), TGF-β1 protein levels in serum (Fig. 6C), and MCP-1, KC, and MIP-2β mRNA levels (Fig. 6E) were significantly lower, and the number of PCNA+ cells was significantly higher (Fig. 6D) in IL-20R1−/− mice.

Figure 6.

IL-20R1−/− mice were resistant to short-term CCl4-induced liver injury. IL-20R1+/+ and IL-20R1−/− mice were treated with CCl4 to induce short-term liver injury. (A) Serum AST and ALT in mice were measured 24, 48, and 72 hours after CCl4 treatment. (B) TGF-β1 mRNA expression in liver tissue was analyzed using qRT-PCR. (C) Serum levels of TGF-β1 in mice were analyzed using ELISA 72 hours after treatment. (D) Positive IHC staining of PCNA represents proliferating hepatocytes in the liver tissue of mice. PCNA-positive cells in more than 12 mm2 tissue sections were counted for each mouse. (E) MCP-1, KC, and MIP-2β mRNA expression in the liver tissue of mice was analyzed using qRT-PCR. Data are means ± SEM. *P < 0.05 compared with IL-20R1+/+ mice. Data are representative of three independent experiments.

IL-20R1−/− Mice Were Resistant to Long-Term CCl4-Induced Liver Injury

IL-20R1+/+ and IL-20R1−/− mice were treated with CCl4 for 12 weeks to induce long-term liver injury. In IL-20R1−/− mice, serum ALT and AST levels (Fig. 7A), TGF-β1 mRNA levels in liver tissue (Fig. 7B), and TGF-β1 protein levels in serum (Fig. 7C) were significantly lower. Histological impairment of the liver (Fig. 7D, left) and liver fibrosis (Fig. 7D, right) were significantly less noticeable. Hydroxyproline levels (Fig. 7E), α-SMA levels, Col-I, TNF-α, TIMP-3, and TIMP-4 levels were all significantly lower (Supporting Fig. 12), and MMP-2, MMP-12, and MMP-13 levels were all significantly higher (Supporting Fig. 12) in IL-20R1−/− mice.

Figure 7.

IL-20R1−/− mice were resistant to long-term CCl4-induced liver injury. IL-20R1+/+ or IL-20R1−/− mice were treated with CCl4 to induce long-term liver injury. (A) Serum AST and ALT in mice were measured 8 and 12 weeks after CCl4 treatment. (B) TGF-β1 mRNA expression in the liver tissue of mice 12 weeks after CCl4 induction was analyzed using qRT-PCR. (C) Serum levels of the TGF-β1 of mice were analyzed using ELISA 12 weeks after treatment. (D) H&E and Sirius red staining of liver sections from mice were used to assess the effects of CCl4. (E) Relative collagen content (per g of liver) was measured (D) H&E and Sirius red staining of liver sections from mice were used to assess the effects of CCl4. (E) Relative collagen content (per g of liver) was measured using hydroxyproline biochemical determinations. Data are means ± SEM. *P < 0.05 compared with IL-20R1+/+ mice. Data are representative of three independent experiments.

Discussion

We found that IL-20 was highly expressed in hepatocytes and HSCs in liver sections from patients with fibrosis, cirrhosis, and hepatocellular carcinoma. IL-20 was involved in liver injury by inducing TGF-β1- and p21WAF1-associated cell cycle arrest. IL-20 also inhibited hepatocyte growth by activating quiescent HSCs and by promoting the proliferation, migration, and secretion of inflammatory cytokines, chemokines, and the ECM deposition of activated HSCs. Moreover, serum IL-20 was highly expressed in CCl4-treated mice with short-term and long-term liver injury. Treatment with 7E or 51D, as well as the IL-20R1 gene deficiency in IL-20R1−/− mice, showed protective potential against hepatotoxin-induced liver injury in vivo.

IL-20 is highly expressed in hepatocytes and HSCs in the liver under pathological conditions. We focused on exploring the involvement of IL-20 in liver fibrosis in the present study and identified IL-20 as a pivotal factor in fibrogenesis. It is worthwhile to continue exploring the clinical implications of IL-20 in hepatocellular carcinoma.

Hepatocytes comprise almost 80% of the total liver mass and perform the majority of the liver's numerous functions. Hepatocytes are damaged when hepatotoxins, hepatitis viruses, and alcohol metabolites injure the liver.[9] Clone-9 cells, a healthy hepatocyte cell line, and primary rat hepatocytes were treated with CCl4 to mimic human liver injury in vitro.[30] CCl4 treatment increased TGF-β1 expression from hepatocytes. It also up-regulated IL-20 production, which, in turn, further increased TGF-β1 expression. The increased TGF-β1 not only inhibited the growth of hepatocytes by up-regulating p21WAF1 expression and arrested cells in the G0/G1 cell cycle phase, but also activated quiescent HSCs.[31, 32]

HSCs modulate the production of cytokines and chemokines, which regulate inflammation. Newly activated HSCs produce ECM and collagen. When quiescent HSCs were activated in our study, IL-20 expression increased over time in parallel with the increasing expression of α-SMA, a marker for activated HSCs. During the activation process, the endogenous IL-20 from quiescent HSCs triggered the activation of HSCs and increased the expression of TGF-β1, which, in turn, further activated HSCs. IL-20 not only activated quiescent HSCs by itself, but also further acted on activated HSCs to induce more TGF-β1 and TNF-α expression, which resulted in the activation of HSCs with fibrogenesis-promoting effects such as collagen production. All of these effects of IL-20 increased the proliferation and migration of activated HSCs. Both effects were involved in the pathogenesis of liver inflammation and fibrogenesis and led to the secretion and accumulation of large amounts of ECM.[9] Thus, IL-20 and TGF-β1 form an auto-amplification loop on HSC and involve together in the cascade of liver fibrosis.

CCl4-induced liver injury has been used as an experimental animal model of liver damage, liver fibrosis, and, eventually, cirrhosis. Serum IL-20 was elevated after short-term and long-term exposure to CCl4 in vivo, which was consistent with the result observed in hepatocytes exposed to CCl4 in vitro. Although 7E treatment significantly inhibited the CCl4-induced effects, which maintained hepatocyte proliferation and prevented leukocyte infiltration and local inflammation, 51D only partially inhibited the CCl4-induced effects. This may be attributed to the other functional receptor complex IL-22R1/IL-20R2 in CCl4-treated mice. We established animal models of not only CCl4-induced liver injury but also of ConA-induced immune-mediated liver injury and metabolic liver injury to clarify how inhibiting IL-20 specifically affected liver recovery. Inhibiting IL-20 may directly relieve the cell-cycle arrest of TGF-β1-mediated hepatocytes and reduce the adverse affects of inflammation.[15, 17-21] Our findings indicate that 7E and 51D have potential therapeutic value as treatments for hepatotoxin-induced, immune-mediated, and metabolic liver injury.

Hepatocellular injury usually leads to inflammation and activation of the immune system, and releases growth factors and cytokines that can stimulate ECM synthesis by activating quiescent HSCs.[3, 9] When fibrogenic activation occurs, HSCs as well as inflammatory cells release TGF-β1 and then respond to the overexpressed TGF-β1 with additional ECM deposition.[33] TGF-β1 strongly up-regulates the expression of TIMPs as well as the production and deposition of the major ECM constituents, and it down-regulates fibrinolytic MMPs. In the presence of chronic hepatic injury, an imbalance between fibrogenesis and fibrinolysis may lead to excess ECM deposition.[9-11] In mice with long-term CCl4-induced liver injury, 7E treatment not only provided protection against hepatic damage, TGF-β1 induction, collagen generation, HSC activation, inflammation, and chemotaxis, but it also increased the expression of MMPs and inhibited the expression of TIMPs, which resulted in less liver fibrosis. 51D, however, did not affect the induction of TGF-β1 or the inhibition of TIMPs.

We used short-term and long-term CCl4-induced liver injury on IL-20R1−/− mice to confirm IL-20-mediated profibrogenic effects. We found that IL-20R1−/− mice were resistant to short-term CCl4-induced cell damage and liver inflammation. An IL-20R1-gene deficiency also protected mice from long-term CCl4-induced liver injury. All IL-20-mediated effects on cell damage, overexpression of inflammatory cytokines, chemokines, ECM components, and fibrogenic mediators like MMPs, as well as HSC activation, were attenuated by impaired IL-20R1 signaling.

We hypothesize that the working model of IL-20 in the progression of liver injury is as follows. When hepatocytes and HSCs are stimulated by CCl4, they express a large amount of IL-20. IL-20 initially inhibits the growth of hepatocytes by inducing the expression of TGF-β1 and p21WAF1, which, in turn, arrests the cell cycle. Persistent cell-damage signaling and increased IL-20 activate additional quiescent HSCs, which increases the proliferation and migration of activated HSCs. Subsequently, the activated HSCs trigger an inflammatory response by producing cytokines and chemokines. Moreover, activated HSCs become myofibroblasts responsible for the excess deposition of ECM. Eventually, all of these fibrogenesis-promoting effects contribute to liver fibrosis and cirrhosis. 7E and 51D treatment in mice potently inhibited IL-20-induced fibrogenesis and an IL-20R1 gene deficiency protected mice from liver injury, which supports our hypothesis.

In conclusion, our findings showed that IL-20 was pivotal in the pathogenesis of liver injury. Currently, the only treatment for advanced fibrosis and cirrhosis is liver transplantation. However, the demand for organ grafts outweighs their availability,[34] which emphasizes the need for effective antifibrotic therapies. 7E and 51D effectively ameliorated both injury and inflammation, which decreased fibrosis formation. They not only prevented the primary hepatocellular injury in response to liver insult, but also inhibited the HSC activation cascade, which down-regulated HSC activation and the fibrogenic responses of HSCs, and increased ECM degradation by up-regulating the production of MMPs and down-regulating the production of TIMPs. Therefore, we conclude that 7E and 51D are potential therapeutics for liver injury.

Acknowledgment

We thank Dr. Shih-Lan Hsu (Department of Education and Research, Taichung Veterans General Hospital, Taichung, Taiwan) and Prof. Shu-Chu Shiesh (Department of Medical Laboratory Science and Biotechnology College of Medicine, National Cheng Kung University, Tainan, Taiwan) for assistance with rat primary cell isolation.

Ancillary