Interleukin-20 antibody is a potential therapeutic agent for experimental arthritis.код для вставкиСкачать
ARTHRITIS & RHEUMATISM Vol. 62, No. 11, November 2010, pp 3311–3321 DOI 10.1002/art.27689 © 2010, American College of Rheumatology Interleukin-20 Antibody Is a Potential Therapeutic Agent for Experimental Arthritis Yu-Hsiang Hsu and Ming-Shi Chang Conclusion. Selectively blocking IL-20 inhibited inflammation and bone loss in rats with CIA. Treatment with 7E combined with etanercept protected rats from CIA better than treatment with etanercept alone. Our findings provide evidence that IL-20 is a novel target and that 7E may be a potential therapeutic agent for RA. Objective. Interleukin-20 (IL-20) is a proinflammatory cytokine involved in the pathogenesis of rheumatoid arthritis (RA). We investigated whether anti– IL-20 antibody treatment would modulate the severity of the disease in a collagen-induced arthritis (CIA) rat model. Methods. We generated a CIA model by immunizing rats with bovine type II collagen. Rats with CIA were treated subcutaneously with anti–IL-20 antibody 7E, with the tumor necrosis factor (TNF) blocker etanercept, or with 7E in combination with etanercept. Arthritis severity was determined according to the hind paw thickness, arthritis severity score, degree of cartilage damage, bone mineral density, and cytokine production, which were evaluated using radiologic scans, microfocal computed tomography, and enzyme-linked immunosorbent assay. To analyze gene regulation by IL-20, rat synovial fibroblasts (SFs) were isolated and analyzed for the expression of RANKL, IL-17, and TNF␣. We also used real-time quantitative polymerase chain reaction analysis and flow cytometry to determine IL-20– regulated RANKL in mouse osteoblastic MC3T3-E1 cells and Th17 cells. Results. In vivo, treatment with 7E alone or in combination with etanercept significantly reduced the severity of arthritis by decreasing the hind paw thickness and swelling, preventing cartilage damage and bone loss, and reducing the expression of IL-20, IL-1␤, IL-6, RANKL, and matrix metalloproteinases (MMPs) in synovial tissue. In vitro, IL-20 induced TNF␣ expression in SFs from rats with CIA. IL-20 markedly induced RANKL production in SFs, osteoblasts, and Th17 cells. Rheumatoid arthritis (RA) is a chronic inflammatory disease characterized by severe synovitis, leukocyte infiltration, synovial membrane hyperplasia, and cartilage destruction (1). Synovial fibroblasts (SFs) play a major role in the pathogenesis of RA through the synthesis of various proteases, superoxide, chemokines, and cytokines, such as tumor necrosis factor ␣ (TNF␣), interleukin-1␤ (IL-1␤), and IL-20 (2–4). Various molecules, such as macrophage colony-stimulating factor and RANKL, are expressed in the inflamed synovium to stimulate osteoclastogenesis and bone resorption (5,6). Patients with RA manifest irreversible bone destruction and poor functional outcomes (7). Therefore, preventing bone destruction is important for any antiarthritis therapy. The bovine type II collagen–induced arthritis (CIA) model shares many pathophysiologic properties with human RA (8). In both diseases, local synovial and immune cells, such as lymphocytes and macrophages, produce a complex array of cytokines and other soluble mediators in a pathogenic inflammatory cascade. These inflammatory responses are associated with the severity of cartilage destruction and bone erosion during the progression of RA (9,10). CIA has been extensively studied to elucidate the pathologic mechanisms relevant to human RA and to identify potential therapeutic targets. TNF␣ promotes inflammation and destruction in endothelial cells, SFs, osteoclasts, and cartilage from patients with RA (2,11–14). Because TNF␣ is produced primarily by synovial macrophages, macrophages have been identified as key players in therapeutic interven- Yu-Hsiang Hsu, MS, Ming-Shi Chang, PhD: National Cheng Kung University, Tainan, Taiwan. Address correspondence and reprint requests to Ming-Shi Chang, PhD, Department of Biochemistry and Molecular Biology, National Cheng Kung University, College of Medicine, 138 Sheng-Li Road, Tainan 70428, Taiwan. E-mail: email@example.com. Submitted for publication March 9, 2010; accepted in revised form July 27, 2010. 3311 3312 HSU AND CHANG tions targeting TNF␣ (15). Two drugs that target TNF␣, namely, etanercept and infliximab, ameliorate the symptoms and joint destruction of RA. However, therapy against TNF is clinically effective in only 40–70% of patients and has only short-term therapeutic effects (15–18). TNF␣ targeting only suppresses, but does not cure, RA. Therefore, it is necessary to find novel targets for which to develop new therapeutic agents for RA. Secretion of IL-17 by Th17 cells is crucial in the pathogenesis of RA (19–21). IL-17 levels are elevated in the synovial fluid of RA patients, and RA synovial membranes cultured in vitro have been shown to spontaneously produce IL-17 (22,23). IL-17 induces the production of proinflammatory cytokines, such as TNF␣, IL-1␤, and IL-6, from joint cells, such as SFs, macrophages, and chondrocytes. IL-17 also acts on osteoblasts and induces RANKL expression, which is crucial for osteoclastogenesis because it promotes cartilage destruction and bone erosion (24). IL-20, a member of IL-10 family (consisting of IL-10, IL-19, IL-20, IL-22, IL-24, IL-26, IL-28, and IL-29 ) has two receptor complexes: IL-20R1/IL-20R2 and IL-22R1/IL-20R2 (26). IL-20 targets keratinocytes, endothelial cells, SFs, renal epithelial cells, and mesangial cells (3,26–29). IL-20 is a proinflammatory cytokine that provokes potent inflammation, angiogenesis, arteriogenesis, chemotaxis, and apoptosis, and is involved in the pathogenesis of psoriasis, atherosclerosis, RA, and acute and chronic renal diseases (3,27,28,30–33). We previously (3) showed that IL-20 not only induced RA SFs to secrete monocyte chemoattractant protein 1 (MCP-1), IL-6, and IL-8, but it also promoted neutrophil chemotaxis. Moreover, electroporating soluble IL-20R1 into rats with CIA significantly reduced the severity of their arthritis, which indicated that IL-20 could be the target of a therapeutic drug. Aside from IL-20, IL-19 and IL-24 also act via the same IL-20 receptor complex, IL20R1/IL-20R2 (34). Thus, reducing the severity of arthritis by blocking the function of IL-20 by inhibiting IL-20R1 might also affect the function of IL-19 and IL-24. We therefore explored whether an alternative specific antagonist of IL-20, IL-20 monoclonal antibody (mAb) 7E, has a therapeutic effect on RA. We also examined whether the combination of etanercept and 7E synergistically ameliorates the severity of CIA. MATERIALS AND METHODS Assessing the severity of CIA. A CIA rat model was generated as previously described (3). The onset of CIA occurred between days 11 and 13 after the first immunization. The severity of the arthritis in each hind paw was monitored and scored as previously described (3). A score higher than 3 was considered to represent severe swelling in rats with CIA. Hind paw thickness was measured with Vernier calipers. Using radiographic equipment outfitted with a direct digital imaging system, we radiographed the ankle joints of rats after they had been given a general anesthetic (pentobarbital; SigmaAldrich). Bone destruction in the ankle joint was scored on a scale of 0–3, where 0 ⫽ no swelling or bone damage, 1 ⫽ mild bone damage, 2 ⫽ moderate bone erosion, and 3 ⫽ severe bone erosion. The severity of arthritis in each rat was determined independently and blindly by 3 individual observers, and the averages of their scores were calculated. Treatment. For the first experiment, the rats were divided into 4 groups (n ⫽ 5 rats per group). Three groups were comprised of rats with CIA, and 1 group consisted of healthy controls. The rats with CIA were given a subcutaneous injection of 3 mg/kg of one of the following treatments 3 times a week: phosphate buffered saline (PBS), 7E, or etanercept (Enebrel; Wyeth). For the second experiment, the rats were divided into 5 CIA groups and 1 healthy control group (n ⫽ 9 rats per group). The CIA groups were given one of the following treatments 3 times a week: PBS, murine IgG (Chemicon), etanercept (6 mg/kg subcutaneously), 7E (6 mg/kg subcutaneously), or 7E plus etanercept (3 mg/kg of each protein subcutaneously). The treatment was started on day 11 after the initial immunization with bovine type II collagen. The dosages and anti-TNF␣ treatments in the positive controls were based on previous protocols known to be effective on arthritic rats (35). Microfocal computed tomography (micro-CT). Micro-CT analyses of the metaphysis of the tibia were done on a system (SkyScan 1076 micro-CT-40) equipped with a highresolution, low-dose x-ray scanner. The x-ray tube was operated with photon energy of 48 kV, current of 200 A, and exposure time of 1,180 msec through a 0.5-mm–thick filter. After standardized reconstruction of the scanned images, the data sets for each tibia sample were resampled with software (CTAn; SkyScan) to orient each sample in the same manner. Consistent conditions, such as thresholds, were applied throughout all analyses. Bone mineral density, a 3-dimensional bone characteristic parameter, was analyzed in 50 consecutive slices. The results were calculated as a percentage compared with the values from PBS-treated controls. Measuring mediators of inflammation in synovial tissue. To measure cytokine levels in synovial tissue surrounding the knee joints, fresh synovial tissue from the knee joints of healthy rats and rats with CIA was isolated by an orthopedic surgeon. Histologic analysis confirmed the healthy and arthritic synovium. Synovial tissue was placed in PBS solution, homogenized, and then centrifuged, and the supernatant was stored at –80°C for analysis. The levels of TNF␣, IL-1␤ (R&D Systems), and IL-20 (PeproTech Asia/CytoLab) were evaluated using sandwich enzyme-linked immunosorbent assay (ELISA) according to the manufacturer’s instructions. Isolating and culturing rat SFs. Freshly isolated synovial tissue from rats with CIA and from healthy rats was collected, and histologic analysis confirmed the healthy and arthritic synovium. The synovial tissue was then finely minced into 2–3-mm pieces and digested with Dispase (Roche) for 45 minutes at 37°C in Dulbecco’s modified Eagle’s medium (DMEM). Isolated SFs were cultured in DMEM containing 10% fetal bovine serum IL-20 ANTIBODY AS A POTENTIAL RA THERAPY (FBS). All in vitro experiments were performed with primary SF cultures between passages 2 and 6. Analyzing RANKL and TNF␣ expression in rat SFs. SFs were plated for 1 day in DMEM with 10% FBS. The cells were incubated for 0, 2, 6, and 8 hours with 200 ng/ml of IL-20 to analyze RANKL expression, and were incubated for 3, 6, and 9 hours to analyze TNF␣ expression. Total RNA was isolated (Invitrogen). Reverse transcription was performed with reverse transcriptase according to the manufacturer’s protocol (Clontech). Expression of RANKL and TNF␣ was then amplified on a thermocycler (LC 480; Roche Diagnostics), with SYBR Green (Roche Diagnostics) as the interaction agent. Quantification analysis of messenger RNA (mRNA) was normalized with ␤-actin, which was used as the housekeeping gene. Relative multiples of change in mRNA expression were determined by calculating 2–⌬⌬Ct. Detecting IL-6, RANKL, and matrix metalloproteinases (MMPs) in synovial tissue. Total RNA was isolated. Reverse transcription was performed with reverse transcriptase according to the manufacturer’s protocol. An analysis of IL-6, RANKL, and MMPs 1, 2, 3, 9, and 13 was then conducted with gene-specific primers and a thermocycler. Quantification of mRNA was normalized with ␤-actin, which was used as the housekeeping gene. Relative multiples of change in mRNA expression were determined by calculating 2–⌬⌬Ct. Analyzing RANKL and IL-17 expression in MC3T3-E1 cells. Cells were plated for 1 day in DMEM with 10% FBS before stimulation. Cells were incubated for 0, 2, 6, and 8 hours with IL-20 (200 ng/ml) to analyze RANKL and IL-17 transcripts. Total RNA was isolated, and reverse transcription was performed with reverse transcriptase. Equal quantities of complementary DNA and primers specific for RANKL and IL-17 were used in a polymerase chain reaction (PCR) to amplify the transcripts. To assay RANKL production, cells were stimulated with IL-20 (200 ng/ml) or both IL-20 (200 ng/ml) and IL-17 (50 ng/ml). Stimulated cells were trypsinized and then stained with phycoerythrin-conjugated antibody against RANKL (eBioscience) for flow cytometric analysis. The mean fluorescence index was quantitated using WinMDI 3.2 software (obtained from Joseph Trotter, The Scripps Research Institute, La Jolla, CA). Th17 cell differentiation and flow cytometric analysis. CD4⫹ T cells from the spleen were purified by negative selection using a cocktail of antibodies and magnetic beads (BD Biosciences). The purity of the CD4⫹ T cells was ⬎90%. CD4⫹ T cells were stimulated for 3 days with a plate-bound anti-CD3 mAb and anti-CD28 mAb (1 g/ml each; BD Biosciences) in the presence of IL-2 (100 units/ml), IL-6 (20 ng/ml), transforming growth factor ␤ (TGF␤; 5 ng/ml), and IL-23 (10 ng/ml) (eBioscience) with 10 g/ml each of anti–interferon-␥ (anti-IFN␥) and anti–IL-4 mAb (BD Biosciences) to differentiate the Th17 cells. Six hours before they were harvested, the T cells were treated with IL-20 (100 ng/ml), phorbol myristate acetate (25 ng/ml), and ionomycin (1 g/ml) for restimulation, and then brefeldin A (10 g/ml) was added for intracellular staining. For surface staining, cells were stained with anti-CD4 and anti-RANKL (eBioscience) and kept on ice for 30 minutes. Intracellular staining for IL-17 (BD PharMingen) was measured according to the manufacturer’s protocols. Cell staining was detected with a flow cytometer and analyzed with CellQuest software (Becton Dickinson). 3313 Figure 1. The effects of the anti–interleukin-20 monoclonal antibody 7E and etanercept on changes in hind paw thickness in rats with collageninduced arthritis (CIA). Arthritis was induced in 3 groups of rats (n ⫽ 5 per group) by intradermal injection of bovine type II collagen on day 0. Booster doses were given on day 8. A fourth group of untreated rats served as healthy controls (n ⫽ 5). A, Rats with CIA were injected subcutaneously with 7E or etanercept (3 mg/kg) 3 times a week throughout the study. Rats with CIA treated with phosphate buffered saline (PBS) served as negative controls. The arrow indicates the initial date of treatment. Hind paw thickness as an indictor of disease activity was measured on the indicated days. Similar antiarthritic effects were observed for 7E and etanercept treatment after the induction of CIA. Values are the mean and SEM. ⴱ ⫽ P ⬍ 0.05 versus PBS-treated controls. B, Hind paw thickness in rats injected as described in A was evaluated on day 25. Data are shown as box plots. Each box represents the 25th to 75th percentiles. Lines outside the boxes represent the 10th and the 90th percentiles. Lines inside the boxes represent the median. Statistical analysis. Commercial statistical software (Prism 5.0; GraphPad Software) was used for the statistical analyses. A one-way analysis of variance (ANOVA) nonparametric test (Kruskal-Wallis test) was used to compare the data between groups. Post hoc comparisons were performed using Dunn’s multiple comparison test. The results of hind paw 3314 HSU AND CHANG thickness measurements are expressed as the mean ⫾ SEM. Severity scores are expressed as box plots (with medians, 25th to 75th percentiles, and 10th and the 90th percentiles). Other results are expressed as the mean ⫾ SD. P values less than 0.05 were considered significant. RESULTS Figure 2. Effects of combination therapy with anti–interleukin-20 monoclonal antibody 7E and etanercept on rats with collagen-induced arthritis (CIA). Arthritis was induced in 5 groups of rats (n ⫽ 9 per group). A sixth group of untreated rats served as healthy controls (n ⫽ 9). A, Rats with CIA were injected subcutaneously with phosphate buffered saline (PBS), murine IgG (mIgG), etanercept (6 mg/kg), 7E (6 mg/kg), or 7E plus etanercept (3 mg/kg each) 3 times a week throughout the study. Hind paw thickness as an indicator of disease activity was measured on the indicated days. The arrow indicates the initial date of treatment. Values are the mean and SEM. ⴱ ⫽ P ⬍ 0.05 versus murine IgG controls; # ⫽ P ⬍ 0.05 versus treatment with 7E or etanercept alone. B, Arthritis severity scores were determined according to the degree of joint swelling and erythema in the hind paws of rats with CIA on day 29. Data are shown as box plots. Each box represents the 25th to 75th percentiles. Lines outside the boxes represent the 10th and the 90th percentiles. Lines inside the boxes represent the median. C, The incidence of severe swelling of the hind paws of rats with CIA (severity score ⬎3) was determined as the percentage of rats with severe swelling (values inside the boxes). Results are from a representative experiment. All experiments were repeated 3 times. Amelioration of arthritis severity by anti–IL-20 mAb 7E in a rat model of CIA. We previously showed (3) that IL-20 was pivotal in the pathogenesis of RA and that electroporation of soluble IL-20R1 significantly reduced the severity of CIA in rats. Therefore, we wanted to study whether 7E, an alternative IL-20 antagonist, could be a therapeutic drug for RA. 7E has shown specificity and neutralization activity in vitro and in vivo (3,27,36). We tested the therapeutic efficacy of 7E in our rat model of experimental CIA. Hind paw thickness was significantly (P ⬍ 0.05) lower in rats with CIA treated with 7E (3 mg/kg) than in control rats treated with PBS (Figure 1A). The median thickness of the hind paws on day 25 was 0.53 (25th to 75th percentiles 0.52–0.54) in the healthy controls and 1.05 (25th to 75th percentiles 1.02–1.13) in the PBS-treated CIA controls, whereas in the 7E-treated group, the median hind paw thickness was 0.84 (25th to 75th percentiles 0.72–0.93) and in the etanercept-treated group, the median hind paw thickness was 0.86 (25th to 75th percentiles 0.78–0.91). These findings indicated that CIA severity had been ameliorated by treatment with 7E and etanercept, respectively. Furthermore, 7E and etanercept were similarly effective in reducing the hind paw thickness of rats with CIA. Synergistic activity of 7E in combination with etanercept in the reduction of arthritis severity in rats with CIA. We found that 7E was as potent as etanercept in reducing the severity of arthritis in rats with CIA (Figure 1). To test whether the combination of 7E and etanercept was more efficacious, we compared the therapeutic efficacy of single (6 mg/kg) and combined (3 mg/kg of each protein) treatments in rats with CIA. Treatment with 7E in combination with etanercept significantly (P ⬍ 0.05) arrested the development and progression of inflammation and reduced hind paw thickness in rats with CIA (Figure 2A). The median severity scores in the 6 groups were as follows: 0.2 (25th to 75th percentiles 0.0–0.4) in healthy controls, 4.2 (25th to 75th percentiles 3.9–4.5) in PBS-treated controls, 4.0 (25th to 75th percentiles 3.5– 4.2) in murine IgG–treated controls, 2.0 (25th to 75th percentiles 0.5–3.1) in 7E-treated rats, 2.1 (25th to 75th percentiles 0.7–3.6) in etanercept-treated rats, and 0.9 IL-20 ANTIBODY AS A POTENTIAL RA THERAPY (25th to 75th percentiles 0.0–2.2) in rats treated with the combination of 7E and etanercept. The arthritis severity scores were significantly different between the etanercept group and the 7E plus etanercept group (P ⬍ 0.05) (Figure 2B). The differences in the incidence of severe hind paw swelling (defined as a severity score ⬎3) in the etanercept group and the 7E plus etanercept group were statistically significant (P ⬍ 0.05) (Figure 2C). The results further demonstrated that combined treatment was more efficacious than the etanercept treatment alone. Radiologic analysis of the bones and joints of rats with CIA. Twenty-five days after the initial immunization, the severity of bone damage was examined radiologically in healthy rats and in rats with CIA under general anesthesia. Joints in the hind paws were severely damaged in the PBS-treated and murine IgG–treated control rats. The thickness of the hind paws and the bone damage in these 2 groups significantly and consistently increased from day 11 until the end of the study (Figure 3A). In contrast, the severity of local ankle bone damage was relatively mild in the groups treated with 7E, etanercept, and 7E plus etanercept (Figure 3A). The differences in the severity of bone damage in the ankle joints between the control groups and the treatment groups were statistically significant (P ⬍ 0.01 to P ⬍ 0.05) (Figure 3B). The results further confirmed that 7E was as potent as etanercept in reducing the severity of arthritis in the hind paws of rats with CIA. Protection against bone destruction and increase in bone density in rats with CIA treated with 7E alone and with 7E in combination with etanercept. Bone destruction and loss of bone density accompany the progression of CIA. To further test whether 7E protects against bone destruction, we conducted micro-CT analyses of the tibias from rats with CIA. The tibias from the PBS-treated and murine IgG–treated control groups showed more prominent bone damage than did the intact joints in the healthy controls (Figure 4A). Treatment with 7E clearly protected rats with CIA against bone loss as compared with rats in the murine IgG control group (Figure 4A). Bone mineral density measurements showed that 7E treatment in rats with CIA significantly inhibited bone loss as compared with the murine IgG control group (P ⬍ 0.05). Although micro-CT analysis showed that neutralizing TNF␣ decreased the bone loss in the tibia, when the data were analyzed by ANOVA, they showed no significant difference (Figure 4B). The combination of 7E plus etanercept was indeed more effective than etanercept alone (P ⬍ 0.05) (Figure 4B), but not more effective than 7E alone. The micro-CT results supported the radiologic 3315 data from the ankle joints. These results provided evidence that 7E protected rats with CIA not only by reducing the severity of the arthritis, but also by decreasing the degree of bone loss. Inhibition of the production of cytokines, RANKL, and MMPs in synovial tissue following treatment with 7E. TNF␣, IL-1␤, and IL-6 are key mediators of the multiple articular manifestations of RA (37). In addition, RANKL is considered an important mediator of bone erosion in RA (38). An imbalance between excessive levels of MMPs facilitates joint destruction, as has been demonstrated in RA cartilage extracts (39). Treatment with 7E significantly reduced the severity of CIA in vivo in rats. To confirm whether the amelioration of disease severity was correlated with the inhibition of proinflammatory cytokine, RANKL, and MMP production, we isolated the knee synovial tissue from rats with CIA and used ELISA and real-time quantitative PCR to analyze the expression of cytokines, RANKL, and MMPs 1, 2, 3, 9, and 13. Treatment with 7E did not significantly reduce TNF␣ expression in synovial tissue as compared with controls (Figure 5A), nor was there any significant difference in the expression of MMP-2 and MMP-9 in synovial tissue after 7E or etanercept treatment (data not shown). However, synovial expression of IL-1␤, IL-20, IL-6, RANKL, MMP-1, MMP-3, and MMP-13 (Figures 5B–H) in rats with CIA treated with 7E was significantly lower than in the murine IgG–treated control rats (P ⬍ 0.01 to P ⬍ 0.05). Furthermore, 7E in combination with etanercept inhibited the expression of IL-6 and RANKL in synovial tissue more so than did 7E alone (P ⬍ 0.05). Therefore, 7E treatment reduced not only the local production of IL-20, but also the production of IL-6, IL-1␤, RANKL, MMP-1, MMP-3, and MMP-13. Taken together, these results suggested that IL-20 might combine with IL-6 and IL-1␤ to trigger synovial inflammation. IL-20 was also involved in joint destruction by up-regulating the expression of MMPs 1, 3, and 13. IL-20-induced expression of TNF ␣ and RANKL in SFs. RANKL is a member of the TNF cytokine family. RANKL expression is regulated by a number of bone resorption–inducing factors (vitamin D3, IL-1, IL-6, and TNF␣) (40) and is critical in osteoclastogenesis. Our in vivo data demonstrated that 7E protected against bone loss and downregulated RANKL, which indicated that IL-20 might be involved in osteoclastogenesis. We therefore investigated whether IL-20 alters RANKL expression in vitro. We found that RANKL levels were higher in synovial tissue from rats with CIA than in that from healthy controls (Figure 6A). We hypothesized that 3316 HSU AND CHANG Figure 3. Radiography of the hind paws of rats with collagen-induced arthritis (CIA). Rats with CIA were injected subcutaneously with phosphate buffered saline (PBS), murine IgG (mIgG), anti–interleukin-20 monoclonal antibody 7E, etanercept (6 mg/kg), or 7E plus etanercept (3 mg/kg each) 3 times a week throughout the study. A sixth group of untreated rats served as healthy controls. A, On day 25, rats were anesthetized, and radiographs were obtained. As compared with the hind paws of the healthy controls, the hind paws of rats with CIA were significantly swollen, with severe damage in the PBS-treated and murine IgG–treated controls. Representative radiographs as shown. B, The degree of joint swelling and bone erosion noted on the radiographs of the hind paws of the rats was scored on a scale of 0–3 as described in Materials and Methods. Values are the mean and SD of 9 rats per group. activated SFs may be a major source of RANKL and TNF␣ induced in response to IL-20 in the inflamed synovium. Therefore, we isolated primary cultures of SFs from rats with CIA and from healthy rats to analyze whether IL-20 up-regulated the gene expression of RANKL and TNF␣ in SFs in vitro. IL-20 (200 ng/ml) induced TNF␣ and RANKL expression in primary cultured SFs derived from rats with CIA Figure 4. Microfocal computed tomography (micro-CT) of the knee joints of rats with collagen-induced arthritis (CIA). A, On day 29 after initial immunization, the tibial metaphysis was harvested from rats with CIA that had been injected subcutaneously with phosphate buffered saline (PBS), murine IgG (mIgG), anti–interleukin-20 monoclonal antibody 7E, etanercept (6 mg/kg), or 7E plus etanercept (3 mg/kg each) 3 times a week throughout the study, as well as from untreated healthy control rats. Representative micro-CT images are shown. B, Tibial bone mineral density was calculated as the percentage of the density in PBS-treated controls. Values are the mean and SD. IL-20 ANTIBODY AS A POTENTIAL RA THERAPY Figure 5. Inhibition of cytokine, RANKL, and matrix metalloproteinase (MMP) production in rats with collagen-induced arthritis (CIA) by treatment with anti–interleukin-20 monoclonal antibody 7E. Rats with CIA were injected subcutaneously with phosphate buffered saline (PBS), murine IgG (mIgG), 7E, etanercept (6 mg/kg), or 7E plus etanercept (3 mg/kg each) 3 times a week throughout the study. Synovial tissue was collected and homogenized on day 29, and total cytoplasmic protein was isolated. Production of the cytokines tumor necrosis factor ␣ (TNF␣) (A), interleukin-20 (IL-20) (B), and IL-1␤ (C) was measured by enzyme-linked immunosorbent assay. The expression of IL-6 (D), RANKL (E), MMP-1 (F), MMP-3 (G), and MMP-13 (H) in synovial tissue was determined by real-time quantitative polymerase chain reaction analysis. Results are from a representative experiment. All experiments were performed 3 times, and the results were similar. Values are the mean and SD. 3317 3318 HSU AND CHANG Figure 6. Functions of interleukin-20 (IL-20) in synovial fibroblasts (SFs), osteoblastic MC3T3-E1 cells, and Th17 cells. A, Total RNAs were isolated from the synovial tissue of healthy control rats and rats with collagen-induced arthritis (CIA) and analyzed by reverse transcription– polymerase chain reaction (RT-PCR) using primers specific for RANKL. B, IL-20 induction of tumor necrosis factor ␣ (TNF␣) in SFs from rats with CIA was determined at the indicated time points by RT-PCR analysis (top) and by real-time quantitative RT-PCR analysis (bottom). Values are the mean and SD. C, IL-20 induction of RANKL in SFs from rats with CIA and healthy control rats was determined at the indicated time points by real-time quantitative RT-PCR analysis (top) and then quantified (bottom). Values are the mean and SD. D, IL-20 (200 ng/ml) induction of RANKL in MC3T3-E1 cells was determined at the indicated time points by RT-PCR analysis. E, IL-20 (200 ng/ml) induction of IL-17 in MC3T3-E1 cells was determined at the indicated time points by real-time quantitative RT-PCR analysis (top) and quantified (bottom). Values are the mean and SD. F, Induction of RANKL was measured in untreated MC3T3-E1 cells and in MC3T3-E1 cells treated for 16 hours with IL-20 (200 ng/ml), IL-17 (50 ng/ml), or both IL-20 and IL-27. Cells were stained with phycoerythrin-conjugated anti-RANKL antibody and analyzed by flow cytometry. Values are the mean and SD. G, RANKL expression in Th17 cells was determined in cells treated for 6 hours with IL-20 (100 ng/ml), during which time they were also restimulated with phorbol myristate acetate (PMA; 25 ng/ml) and ionomycin (1 g/ml), and then harvested for flow cytometry using anti-CD4, anti–IL-17, and anti-RANKL monoclonal antibody. Shaded histogram indicates cells treated with PMA and ionomycin; green histogram indicates cells treated with IL-20, PMA, and ionomycin. (Figures 6B and C). However, RANKL expression was not detected in SFs from healthy controls, even after 8 hours of IL-20 treatment (Figure 6C). Up-regulation of RANKL and IL-17 in MC3T3-E1 cells by IL-20. RANKL, RANK, and osteoprotegerin (OPG)—a biologically related family of molecules—coordinate to regulate osteoclast function (6) and trigger bone resorption, and thus, they control the status of bone mass. In the pathogenesis of RA, SFs and osteoblasts secrete RANKL, which activates the differentiation and proliferation of osteoclasts, which in turn, destroy bone and reduce bone mineral density. Therefore, based on the findings that 7E protected rats with CIA against bone loss and that IL-20 induced RANKL expression in SFs from rats with CIA, we also hypothesized that IL-20 was also involved in osteoclastogenesis by targeting osteoblasts as well as SFs to regulate the RANKL/RANK system. To test this hypothesis, we treated osteoblastic MC3T3-E1 cells with IL-20 and found that IL-20 up-regulated RANKL and IL-17 IL-20 ANTIBODY AS A POTENTIAL RA THERAPY transcripts (Figure 6D and E). Recent studies (19,24) showed that IL-17 acts directly on osteoblasts and promotes cartilage destruction and bone erosion. We detected higher levels of RANKL in osteoblasts after combined treatment with IL-20 and IL-17 (Figure 6F). These results suggested that IL-20 might be pivotal in inflammatory bone destruction because it induces osteoblasts to produce RANKL and IL-17. Promotion of the production of RANKL in Th17 cells by IL-20. Another study (41) showed that Th17 cells function as an osteoclastogenic helper T cell subset and that they link T cell activation and bone destruction in the pathogenesis of RA. To further investigate the role of IL-20 in Th17 cells, we stimulated CD4⫹ T cells with anti-CD3/CD28 mAb in the presence of IL-6, TGF␤, IL-23, anti-IFN␥ mAb, and anti–IL-4 mAb. RANKL expression on the surface of Th17 cells treated with IL-20 (100 ng/ml) was significant (Figure 6G). Thus, IL-20 acted not only on SFs and osteoblasts, but also on Th17 cells by inducing RANKL expression, which provided more evidence of its pivotal role in osteoclastogenesis. DISCUSSION In the present study, we found that using the specific IL-20 mAb 7E to neutralize IL-20 activity has therapeutic potential for RA. Treatment with 7E inhibited the expression of IL-1␤, IL-6, RANKL, MMP-1, MMP-3, and MMP-13, ameliorated the severity of CIA, and protected rats with CIA against arthritic bone destruction. In addition, IL-20 induced RANKL expression in osteoblasts and SFs, which suggested that IL-20 might be involved in inflammatory bone destruction through the regulation of osteoclastogenic activity. Taking all our findings together, we have found new evidence that 7E may be a potent therapeutic agent for RA. About 40% of RA patients are resistant to TNF␣ blockade, which suggests that some TNF␣-independent pathogenesis occurs in these patients and indicates a need for new anti-RA therapeutic agents. In the present study, 7E and etanercept were equally efficacious in reducing the severity of arthritis by inhibiting the inflammatory response and by preventing bone destruction in inflamed joints. The therapeutic efficacy of 7E combined with etanercept was better than etanercept alone, which indicated that IL-20 and TNF␣ synergistically promote synovial inflammation. Although micro-CT analysis showed that neutralization of TNF␣ decreased bone loss in the tibia, ANOVA showed no statistically significant difference. Even though TNF␣ is pivotal in a CIA model 3319 of cartilage and bone destruction, other critical factors may be involved. In addition, etanercept is a therapeutic agent for RA patients. Higher doses of human etanercept or a longer treatment time may be required to show its protective effect against bone loss in rats. Our previous study (42) showed that IL-20 in combination with IL-1␤ contributed to the inflammation of herniated intervertebral discs by up-regulating IL-8, MCP-1, and TNF␣. IL-1␤, as well as TNF␣, was found to be pivotal in a CIA model of cartilage and bone destruction (43). In the present study, 7E potently reduced IL-1␤ and IL-6 expression in synovial tissue in vivo, which indicated that the active phase of IL-20 occurs earlier in the course of arthritic inflammation than do the active phases of IL-6 and IL-1␤. Our findings also suggested that treatment with 7E might totally inhibit other proinflammatory cytokines. In addition, the finding that 7E potently inhibited the expression of MMPs 1, 3, and 13 in synovial tissue in vivo also suggested that 7E may be a good therapeutic agent for cartilage damage. TNF␣, IL-1␤, and IL-17, which are secreted by SFs, osteoblasts, and activated T cells (44), regulate RANKL expression. RANKL stimulates osteoclast differentiation and increases bone resorption. In the present study, we confirmed that RANKL was upregulated in synovial tissue from rats with CIA. Furthermore, IL-20 induced the expression of TNF␣ in CIA SFs, which suggested that IL-20 acts upstream of TNF␣. We also found that IL-20 induced RANKL expression in SFs from rats with CIA, but not in those from healthy rats. The differential expression of RANKL in these 2 sources of SFs may be caused by a lack of IL-20R1 expression in the SFs of healthy rats (3). This result might also imply that IL-20–mediated RANKL induction occurs only during the progression of RA. We also found that, in addition to SFs, IL-20 induced RANKL expression in MC3T3-E1 osteoblasts. This is evidence that IL-20 is involved in regulating the osteo-immune system because it induces RANKL expression. It was recently reported (20) that Th17 is critical for the induction and progression of RA. Th17 involved in RA pathogenesis has been attributed to IL-17– stimulated osteoclastogenesis (22). IL-17 induces RANKL expression, one crucial mechanism for osteoclastogenesis and bone resorption. We showed that IL-20 also acted on Th17 cells and induced RANKL expression. Thus, IL-20 may directly induce RANKL, or else it may indirectly induce RANKL by up-regulating TNF␣ and IL-17. RANKL expression in osteoblasts, 3320 HSU AND CHANG SFs, and T cells is critical in the pathogenesis of RA. Therefore, IL-20 is an upstream activator of RANKL expression in these target cells and may be involved in bone destruction in RA patients. The potent inhibition of RANKL expression in synovial tissue of rats with CIA treated with 7E provided additional in vivo evidence. Furthermore, treatment with 7E in combination with etanercept inhibited RANKL more potently than did 7E or etanercept alone in rats with CIA, which suggested that IL-20, TNF␣, and RANKL closely regulated each other. The function of IL-20 in the RANK/RANKL axis and the direct involvement of IL-20 in osteoclastogenesis remain to be clarified. The pleiotropic cellular effects of IL-20 and other proinflammatory cytokines, such as TNF␣, IL-6, IL-17, and IL-1␤, may act cooperatively to mediate the initial inflammatory response and lead to the progression of cartilage damage and bone destruction in the pathogenesis of RA. Therefore, targeting a single cytokine may not be sufficient to treat complex inflammatory diseases such as RA. The combination of 7E with etanercept significantly ameliorated the progression of arthritis in rats with CIA. Therefore, this combined treatment may provide a novel therapeutic cocktail for RA. Extensive pharmacologic studies of this approach are required, however, to adequately evaluate the long-term use of IL-20 antibody for the treatment of chronic inflammatory diseases such as RA. In summary, the specific anti–IL-20 antibody 7E significantly decreased the severity of CIA, reduced the production of proinflammatory cytokines, and prevented in vivo bone destruction in rats with CIA. All of these results suggest its therapeutic potential in RA and in inflammation-related bone diseases. AUTHOR CONTRIBUTIONS Both authors were involved in drafting the article or revising it critically for important intellectual content, and both authors approved the final version to be published. Dr. Chang had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study conception and design. Hsu, Chang. Acquisition of data. Hsu, Chang. Analysis and interpretation of data. Hsu, Chang. REFERENCES 1. Feldmann M, Brennan FM, Maini RN. Rheumatoid arthritis. Cell 1996;85:307–10. 2. Feldmann M, Brennan FM, Williams RO, Cope AP, Gibbons DL, Katsikis PD, et al. Evaluation of the role of cytokines in autoimmune disease: the importance of TNF␣ in rheumatoid arthritis. Prog Growth Factor Res 1992;4:247–55. 3. Hsu YH, Li HH, Hsieh MY, Liu MF, Huang KY, Chin LS, et al. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. Function of interleukin-20 as a proinflammatory molecule in rheumatoid and experimental arthritis. Arthritis Rheum 2006;54: 2722–33. Brennan F, Beech J. Update on cytokines in rheumatoid arthritis. Curr Opin Rheumatol 2007;19:296–301. Kong YY, Yoshida H, Sarosi I, Tan HL, Timms E, Capparelli C, et al. OPGL is a key regulator of osteoclastogenesis, lymphocyte development and lymph-node organogenesis. Nature 1999;397: 315–23. Saidenberg Kermanac’h N, Bessis N, Cohen-Solal M, De Vernejoul MC, Boissier MC. Osteoprotegerin and inflammation. Eur Cytokine Netw 2002;13:144–53. Bresnihan B. Preventing joint damage as the best measure of biologic drug therapy. J Rheumatol Suppl 2002;65:39–43. Durie FH, Fava RA, Noelle RJ. Collagen-induced arthritis as a model of rheumatoid arthritis. Clin Immunol Immunopathol 1994;73:11–8. Lubberts E, van den Berg WB. Cytokines in the pathogenesis of rheumatoid arthritis and collagen-induced arthritis. Adv Exp Med Biol 2003;520:194–202. Luross JA, Williams NA. The genetic and immunopathological processes underlying collagen-induced arthritis. Immunology 2001;103:407–16. Kudo O, Fujikawa Y, Itonaga I, Sabokbar A, Torisu T, Athanasou NA. Proinflammatory cytokine (TNF␣/IL-1␣) induction of human osteoclast formation. J Pathol 2002;198:220–7. Romas E, Gillespie MT, Martin TJ. Involvement of receptor activator of NFB ligand and tumor necrosis factor-␣ in bone destruction in rheumatoid arthritis. Bone 2002;30:340–6. Vasilopoulos Y, Gkretsi V, Armaka M, Aidinis V, Kollias G. Actin cytoskeleton dynamics linked to synovial fibroblast activation as a novel pathogenic principle in TNF-driven arthritis. Ann Rheum Dis 2007;66 Suppl 3:iii23–8. Bradley JR. TNF-mediated inflammatory disease. J Pathol 2008; 214:149–60. Pope RM. Apoptosis as a therapeutic tool in rheumatoid arthritis. Nat Rev Immunol 2002;2:527–35. Maini RN, Taylor PC. Anti-cytokine therapy for rheumatoid arthritis. Annu Rev Med 2000;51:207–29. Feldmann M, Brennan FM, Foxwell BM, Maini RN. The role of TNF␣ and IL-1 in rheumatoid arthritis. Curr Dir Autoimmun 2001;3:188–99. Saito H, Kojima T, Takahashi M, Horne WC, Baron R, Amagasa T, et al. A tumor necrosis factor receptor loop peptide mimic inhibits bone destruction to the same extent as anti–tumor necrosis factor monoclonal antibody in murine collagen-induced arthritis. Arthritis Rheum 2007;56:1164–74. McInnes IB, Schett G. Cytokines in the pathogenesis of rheumatoid arthritis. Nat Rev Immunol 2007;7:429–42. Lubberts E. IL-17/Th17 targeting: on the road to prevent chronic destructive arthritis? Cytokine 2008;41:84–91. Miossec P. Interleukin-17 in rheumatoid arthritis: if T cells were to contribute to inflammation and destruction through synergy [review]. Arthritis Rheum 2003;48:594–601. Kotake S, Udagawa N, Takahashi N, Matsuzaki K, Itoh K, Ishiyama S, et al. IL-17 in synovial fluids from patients with rheumatoid arthritis is a potent stimulator of osteoclastogenesis. J Clin Invest 1999;103:1345–52. Hwang SY, Kim HY. Expression of IL-17 homologs and their receptors in the synovial cells of rheumatoid arthritis patients. Mol Cells 2005;19:180–4. Stamp LK, James MJ, Cleland LG. Interleukin-17: the missing link between T-cell accumulation and effector cell actions in rheumatoid arthritis? Immunol Cell Biol 2004;82:1–9. Commins S, Steinke JW, Borish L. The extended IL-10 superfamily: IL-10, IL-19, IL-20, IL-22, IL-24, IL-26, IL-28, and IL-29. J Allergy Clin Immunol 2008;121:1108–11. IL-20 ANTIBODY AS A POTENTIAL RA THERAPY 26. Blumberg H, Conklin D, Xu WF, Grossmann A, Brender T, Carollo S, et al. Interleukin 20: discovery, receptor identification, and role in epidermal function. Cell 2001;104:9–19. 27. Hsieh MY, Chen WY, Jiang MJ, Cheng BC, Huang TY, Chang MS. Interleukin-20 promotes angiogenesis in a direct and indirect manner. Genes Immun 2006;7:234–42. 28. Li HH, Hsu YH, Wei CC, Lee PT, Chen WC, Chang MS. Interleukin-20 induced cell death in renal epithelial cells and was associated with acute renal failure. Genes Immun 2008;9:395–404. 29. Li HH, Cheng HH, Sun KH, Wei CC, Li CF, Chen WC, et al. Interleukin-20 targets renal mesangial cells and is associated with lupus nephritis. Clin Immunol 2008;129:277–85. 30. Wei CC, Chen WY, Wang YC, Chen PJ, Lee JY, Wong TW, et al. Detection of IL-20 and its receptors on psoriatic skin. Clin Immunol 2005;117:65–72. 31. Tritsaris K, Myren M, Ditlev SB, Hubschmann MV, van der Blom I, Hansen AJ, et al. IL-20 is an arteriogenic cytokine that remodels collateral networks and improves functions of ischemic hind limbs. Proc Natl Acad Sci U S A 2007;104:15364–9. 32. Chen WY, Cheng BC, Jiang MJ, Hsieh MY, Chang MS. IL-20 is expressed in atherosclerosis plaques and promotes atherosclerosis in apolipoprotein E-deficient mice. Arterioscler Thromb Vasc Biol 2006;26:2090–5. 33. Wei CC, Li HH, Hsu YH, Hsing CH, Sung JM, Chang MS. Interleukin-20 targets renal cells and is associated with chronic kidney disease. Biochem Biophys Res Commun 2008;374:448–53. 34. Parrish-Novak J, Xu W, Brender T, Yao L, Jones C, West J, et al. Interleukins 19, 20, and 24 signal through two distinct receptor complexes. Differences in receptor-ligand interactions mediate unique biological functions. J Biol Chem 2002;277:47517–23. 35. Mihara K, Almansa C, Smeets RL, Loomans EE, Dulos J, Vink 3321 36. 37. 38. 39. 40. 41. 42. 43. 44. PM, et al. A potent and selective p38 inhibitor protects against bone damage in murine collagen-induced arthritis: a comparison with neutralization of mouse TNF␣. Br J Pharmacol 2008;154: 153–64. Chen WY, Chang MS. IL-20 is regulated by hypoxia-inducible factor and up-regulated after experimental ischemic stroke. J Immunol 2009;182:5003–12. Rannou F, Francois M, Corvol MT, Berenbaum F. Cartilage breakdown in rheumatoid arthritis. Joint Bone Spine 2006;73: 29–36. Wada T, Nakashima T, Hiroshi N, Penninger JM. RANKL-RANK signaling in osteoclastogenesis and bone disease. Trends Mol Med 2006;12:17–25. Close DR. Matrix metalloproteinase inhibitors in rheumatic diseases. Ann Rheum Dis 2001;60 Suppl 3:iii62–7. Takayanagi H. Osteoimmunology: shared mechanisms and crosstalk between the immune and bone systems. Nat Rev Immunol 2007;7:292–304. Sato K, Suematsu A, Okamoto K, Yamaguchi A, Morishita Y, Kadono Y, et al. Th17 functions as an osteoclastogenic helper T cell subset that links T cell activation and bone destruction. J Exp Med 2006;203:2673–82. Huang KY, Lin RM, Chen WY, Lee CL, Yan JJ, Chang MS. IL-20 may contribute to the pathogenesis of human intervertebral disc herniation. Spine 2008;33:2034–40. Abramson SB, Amin A. Blocking the effects of IL-1 in rheumatoid arthritis protects bone and cartilage. Rheumatology (Oxford) 2002;41:972–80. Takayanagi H. Osteoimmunology and the effects of the immune system on bone. Nat Rev Rheumatol 2009;5:667–76.