|Home | About | Journals | Submit | Contact Us | Français|
To examine the mechanism of regulation of interleukin-18 (IL-18) bioactivity by IL-18 binding protein (IL-18BP) induction.
Levels of IL-18 and IL-18BPa expression were determined by enzyme-linked immunosorbent assays (ELISA) in osteoarthritis (OA) and rheumatoid arthritis (RA) synovial fluids, followed by free IL-18 calculation. IL-18 and IL-18BPa synthesis in RA synovial fibroblasts treated with pro- and anti-inflammatory cytokines were assessed by qRT-PCR and ELISA, respectively, followed by IL-18 bioactivity determination using KG-1 cells. Chemical signaling inhibitors and antisense oligonucleotides were used for validation of the signal transduction pathways involved in IL-18BPa/IL-18 regulation. TNF-α-induced caspase-1 activity was determined by a colorimetric assay.
IL-18BPa was lower in RA synovial fluid than in OA synovial fluid (n=8; P < 0.05) and free IL-18 was higher in RA synovial fluid than in OA synovial fluid. TNF-α induced RA synovial fibroblast IL-18BPa and IL-18 in a time dependent manner (P < 0.05). Evaluation of signaling pathways suggested that TNF-α induced IL-18 production through extracellular signal-regulated kinases (ERK)1/2, protein kinase C (PKC)δ, and Src pathways, whereas IL-18BPa synthesis was mediated through nuclear factor kappa-light-chain-enhancer of activated B cells (NFκB), PKC, Src, and c-Jun N-terminal kinases (JNK) pathways. Furthermore, addition of exogenous IL-18BPa-Fc reduced the RA synovial fibroblast phosphorylation of ERK1/2 induced by TNF-α.
These results suggest that IL-18BPa reduces IL-18 bioactivity induced by TNF-α, by regulating the ERK1/2 pathway in RA synovial fibroblasts. Targeting IL-18 bioactivity by induction or addition of IL-18BPa may provide another therapeutic option in the management of RA.
Rheumatoid arthritis (RA) and osteoarthritis (OA) are two common chronic joint disorders whose etiology remains unknown. The RA synovium is characterized by angiogenesis, or new blood vessel growth, and leukocyte infiltration that lead to tissue invasion and joint destruction (1). OA is considered mainly a noninflammatory disease, in which mild to moderate inflammatory changes at certain stages of the disease correlate with disease progression (2). However, in both diseases, proinflammatory cytokines play an important role in their pathophysiology (3).
Interleukin-1 (IL-1) family members play a key part in the pathogenesis of both RA and OA (4). Among this family, IL-18 plays an important role in inducing the T helper-1 immune response through the induction of interferon-gamma (IFN-γ) in T cells and natural killer cells (5), and has both a local and systemic effect on angiogenesis (6, 7). IL-18 plays an important role in the pathophysiology of RA and OA (8, 9). Various sources of IL-18 have been identified including Kupffer cells, dendritic cells, keratinocytes, articular chondrocytes, osteoblasts, and synovial fibroblasts (8, 10-12).
IL-18 is produced as a precursor molecule (pro-IL-18), then is processed by IL-1β-converting enzyme (ICE, caspase-1) to obtain the mature form of IL-18 which is biologically active (5). The importance of IL-18 has also been shown in an animal model of arthritis (13). To control some of the potentially deleterious properties of IL-18, IL-18 binding protein (IL-18BP) has been identified as a specific endogenous inhibitor of IL-18 bioactivity (14, 15). Four isoforms of IL-18BP are described in humans as isoforms a, b, c, and d, which are produced as a result of an alternative splicing. IL-18BPa is the major splicing variant with the highest binding affinity of 400 pM for IL-18 (14). Furthermore, the potentially beneficial role of IL-18BP therapy has been demonstrated, in which the administration of IL-18BP led to the resolution of rodent arthritis (15, 16).
The mechanism by which IL-18BPa may control IL-18 bioactivity in RA synovial fibroblasts is not yet known. In the present study, we report that free IL-18 is higher in RA synovial fluid compared to OA synovial fluid due to a paucity of IL-18BPa expression in RA synovial fluid. TNF-α is a powerful inducer of IL-18BPa through Src, PKC, JNK2, and NFκB pathways. TNF-α also induces IL-18 and caspase-1 expression and activity in the same manner, but TNF-α induced IL-18 through Src, PKCδ, and ERK1/2 pathways. Exogenous IL-18BPa-Fc significantly reduced TNF-α induced phophorylation of ERK1/2. Our results indicate that in RA, in which TNF-α plays a key role, blocking the ERK1/2 pathway reduces IL-18 bioactivity by the reduction of IL-18 production and concomitant increase in IL-18BPa production.
TNF-α, IL-1β, IL-13, IL-17, IL-18, IL-18BPa-Fc, IgG-Fc, and mouse monoclonal anti-human IL-18 were purchased from R&D Systems (Minneapolis, MN). IFN-γ, IL-4, and IL-10 were purchased from PeproTech (Rocky Hill, NJ).
Fibroblasts were isolated from synovium obtained from RA patients who had undergone total joint replacement or synovectomy according to an institutional review board-approved protocol and processed as described previously (17, 18). RA synovial fibroblasts were grown in RPMI 1640 with 10% fetal bovine serum supplementation. All the experiments were performed in serum free media.
Synovial fluids from RA and OA patients were collected from patients with joint effusions, and then stored at −80°C until IL-18BPa determination. To avoid any possible confounding effects of rheumatoid factor on assays, rheumatoid factor was immunodepleted from synovial fluids using anti-IgM antibodies coupled to agarose beads (Sigma, St. Louis, MO) as previously described (19). RA synovial fibroblasts (2 × 105/well in 6-well plates) were stimulated with cytokines for 2-48 hours in serum free media. Upon termination, conditioned medium was collected and concentrated 10-fold using Amicon Ultra 10,000 mW concentrators from Millipore (Bedford, MA). IL-18 and IL-18BPa levels in synovial fluid and concentrated conditioned medium were determined using ELISA kits from Bender MedSystems (Burlingame, CA) and R&D Systems (Minneapolis, MN), respectively.
Free IL-18 was determined according to the law of mass action and by assessments of both total IL-18 and total IL-18BPa by ELISA. The calculation of free IL-18 was based on a 1:1 stoichiometry in the complex of IL-18 and IL-18BPa with the dissociation constant of 0.4 nM (14).
To study the effect of several proinflammatory and anti-inflammatory cytokines on IL-18BPa production, RA synovial fibroblasts were treated with IL-1β (10 ng/ml), TNF-α (20 ng/ml), IL-17 (50 ng/ml), IL-18 (180 ng/ml), IL-4 (50 ng/ml), IL-10 (50 ng/ml) or IL-13 (50 ng/ml) for 24 hours. IL-18BPa expression was assessed at the mRNA level and at the protein level.
Following the manufacturer's protocol, RNA was isolated using RNAeasy mini RNA isolation kits in conjunction with QIAshredders (Qiagen Inc., Valencia, CA), as described previously (17). Following the isolation, RNA was quantified and checked for purity by spectrophotometry (Nanodrop Technologies, Wilmington, DE). Complementary DNA (cDNA) was then prepared using a Reverse-IT MAX first strand synthesis kit (Abgene Inc., Rochester, NY), as described previously (17). Quantitative polymerase-chain reaction (qPCR) was performed using Platinum SYBR Green qPCR SuperMix-UDG (Invitrogen, Carlsbad, CA) following the manufacturer's protocol, using specific primer sequences for human IL-18 (Forward, 5’-GCTTGAATCTAAATTATCAGTC-3’; Reverse, 5’-GAAGATTCAAATTGCATCTTAT-3’) (20), I L-18BPa (Forward, 5’-ACCTCCCAGGCCGACTG-3’; Reverse, 5’-CCTTGCACAGCTGCGTACC-3’) (21), caspase-1 (Forward, 5’-CAAGGGTGCTGAACAAGG-3’; Reverse, 5’-GGGCATAGCTGGGTTGTC-3’) (22) and β-actin (Forward, 3’-GTCAGGCAGCTCGTAGCTCT-5’; Reverse, 5’-GCCATGTACGTTGCTATCCA-3’). Diluted cDNA was mixed with Platinum SYBR Green qPCR SuperMix-UDG, forward and reverse primers specific for each gene (0.2 μM final concentrations), and incubated at the following cycles: 50°C for 2 minutes, 95°C for 2 minutes, and 40 cycles of 95°C for 30 seconds, 55°C for 30 seconds, 68°C for 30 seconds using an Eppendorf Mastercycler ep realpex thermal cycler (Eppendorf, Hamburg, Germany). All samples were run in duplicate and analyzed using Eppendorf software. For quantification, the relative abundance of each gene was normalized to β-Actin.
To study the signaling mechanism of IL-18BPa production by TNF-α, RA synovial fibroblasts were incubated with MAPK inhibitors (ERK1/2, PD98059; p38, SB202190; and JNK2, SP600125), PKC inhibitors (PKCα/β, Gö6976; and PKCδ, Rottlerin), a Src inhibitor (PP2), an NFκB inhibitor (PDTC), or a JAK2 inhibitor (AG-490) for 1 hour, followed by stimulation with TNF-α (20 ng/ml) for 48 hours. Supernatants were processed for estimation of IL-18BPa production. The concentration of the inhibitors used in the study was based on our previous publications (17, 23). Cells were pretreated for 1 hour with 10 μM of each inhibitor (except for PDTC, 200 μM) before stimulation with TNF-α (20 ng/ml). The concentration of the inhibitors was based on our previous publications (17, 23). All inhibitors were purchased from Calbiochem (San Diego, CA).
RA synovial fibroblasts (2 × 106/well) were treated with TNF-α (20 ng/ml), for 8, 24, and 48 hours in serum free RPMI 1640. Cells were washed and then lysed with the lysis buffer from the caspase-1 activity assay kit. Cell lysates were centrifuged, and the supernatant was used as the cell extract. Caspase-1 activity in cell extract was determined using a colorimetric caspase-1 activity assay kit (R&D Systems).
The biologic activity of IL-18 was measured by using human myelomonocytic KG-1 cells, as previously described (24). KG-1 cells (3 × 106 cells/ml; 100 μl) with or without mouse monoclonal anti-IL-18 antibody at 1 μg/ml (R&D Systems), were dispensed into the wells of 96-well microtiter plates (Falcon, Becton Dicinson, Franklin Lakes, NJ). Then, 100 μl of samples or recombinant human IL-18 standards were added to each well. The plates were incubated and culture supernatants were harvested 24 hours later. IFN-γ concentration in this media was determined by ELISA (Invitrogen). IL-18 bioactivity was determined by the difference in IFN-γ levels between those cultures with and without mouse monoclonal anti-IL-18 antibody.
To study the effect of IL-18BPa on TNF-α-induced ERK1/2, RA synovial fibroblasts were incubated with or without TNF-α (20 ng/ml) after 1 hour of preincubation with IgG-Fc or IL-18BPa-Fc (25 and 50 ng/ml) in serum free RPMI 1640 for 20 or 30 minutes. Cells were lysed in cell lysis buffer containing 1 mM phenylmethylsulfonyl fluoride (PMSF), and protease inhibitors (Thermo Scientific, Rockford, IL) as previously decribed (18). Protein was measured using a BCA protein assay kit (Pierce, Rockford, IL). Equal amounts of protein (15 μg) were loaded and separated by SDS-polyacrylamide gel electrophoresis and transferred onto nitrocellulose membranes (Bio-Rad, Richmond, CA). Nitrocellulose membranes were blocked with 5% nonfat milk in Tris buffered saline-Tween 20 (TBST) for 60 minutes at room temperature. Blots were incubated overnight at 4°C with optimally diluted specific primary antibody in TBST containing 5% nonfat milk. Phosphorylation state-specific antibody to ERK1/2, JNK2, PKCδ, c-Jun, or NFκB (Cell Signaling Technology, Beverly, MA) were used as the primary antibody. Blots were washed 3 times and incubated in horseradish peroxidaseconjugated antibody (1:1,000 dilutions) for 1 hour at room temperature. Protein bands were detected using enhanced chemiluminescence (GE Healthcare, Piscataway, NJ) in accordance with the manufacturer's instructions. After stripping, blots were probed again for β-actin using a rabbit polyclonal anti-β-actin antibody. Blots were scanned and analyzed for band intensities using UN-SCAN-IT version 5.1 software (Silk Scientific, Orem, UT).
Synovial fluid from RA patients contained significantly lower levels of IL-18BPa than in OA patients (Mean ± SEM, 3995 ± 660 pg/ml vs. 6765 ± 761 pg/ml; n = 8 per group; P < 0.05; Figure 1A). In the same fluids, IL-18 levels were similar in RA and OA synovial fluids (237 ± 33 pg/ml vs. 201 ± 26 pg/ml; n = 8 per group; NS; Figure 1B). However, free IL-18 calculated according to the dissociation constant, was higher in RA than in OA synovial fluids (63 ± 14 pg/ml vs. 27 ± 4 pg/ml; n = 8 per group; P < 0.05). This suggests that the high amount of free IL-18 in RA synovial fluid is the result of decreased IL-18BPa expression.
TNF-α (20 ng/ml) stimulation resulted in a 24-fold induction in IL-18BPa transcription in RA synovial fibroblasts (Figure 2A; P < 0.05). Other proinflammatory cytokines such as IL-1β (10 ng/ml), IL-17 (50 ng/ml), or IL-18 (180 ng/ml) showed no regulation of IL-18BPa in RA synovial fibroblasts (Figure 2A). Among the battery of “anti-inflammatory” cytokines used for stimulation, including IL-4 (50 ng/ml), IL-10 (50 ng/ml), or IL-13 (50 ng/ml), only IL-4 downregulated IL-18BPa expression at the transcriptional level. We then confirmed the IL-18BPa induction by TNF-α at the protein level (Figure 2B; P < 0.05). However, IL-4 has no effect on IL-18BPa production at the protein level. Overall cytokines profiled only TNF-α induced IL-18BPa in RA synovial fibroblasts.
TNF-α (20 ng/ml) stimulation of RA synovial fibroblasts was performed at the following times: 2, 4, 8, 24, and 48 hours. IL-18BPa was induced by TNF-α in a time dependent manner with a maximal effect at 48 hours at the mRNA and protein levels (P < 0.05; Figure 2C&D). Accordingly, we then used the 48 hour time point for further studies.
To identify the signaling events that are critical for TNF-α-induced IL-18BPa, RA synovial fibroblasts were incubated with chemical signaling inhibitors for 1 hour followed by stimulation with TNF-α (20 ng/ml) for 48 hours. The results of this study showed that the inhibitors of JNK2 (SP600125), PKCα/β (Gö6976), PKCδ (Rottlerin), Src (PP2) and NFκB (PDTC) pathways significantly inhibited TNF-α-induced IL-18BPa transcription (by 78%, 88%, 56%, 72%, and 49% respectively from the levels observed with TNF-α stimulation alone) in RA synovial fibroblasts (Figure 3; P < 0.05). In contrast, inhibitors of ERK1/2 (PD98059), p38 (SB202190), and JAK2 (AG-490) did not significantly inhibit TNF-α-induced-IL-18BPa transcription. These results indicate that TNF-α induction of IL-18BPa in RA synovial fibroblasts occurs via JNK2, PKC, and NFκB, but not via the ERK1/2 pathway.
To determine if the production of IL-18BPa has any regulatory effect on free IL-18 induced by TNF-α, we examined the effect of TNF-α on IL-18 expression at the mRNA level. Our study showed that TNF-α also induced IL-18 expression in a time dependent manner (Figure 4A). Furthermore, the ratio IL-18/IL-18BPa showed no significant variation in the range of 0.5 to 1.6 over time (Figure 4B), suggesting a similar induction of IL-18 and IL-18BP by TNF-α. As pro-IL-18 needs to be cleaved by caspase-1 to be activated, we then investigated TNF-α-induced caspase-1 expression at the mRNA and the functional level. We also observed that TNF-α induced caspase-1 at both the transcriptional (Figure 4C) and the functional level (Figure 4D). Since TNF-α induces both IL-18 and caspase-1 at the transcriptional level and induces functional caspase-1, TNF-α may be able to induce bioactive IL-18 in RA synovial fibroblasts.
We next determined if the same signaling events found to be critical for TNF-α induction of IL-18BPa are also required for TNF-α induction of IL-18. We found that the inhibitors of ERK1/2 (PD98059), PKCδ (Rottlerin), and Src (PP2) significantly inhibited TNF-α-induced IL-18 production (by 73%, 83%, and 75%, respectively) in RA synovial fibroblasts (Figure 5A; P < 0.05). In contrast, the inhibitors of NFκB (PDTC), p38 (SB202190), JNK2 (SP600125), JAK2 (AG-490), and PKCα/β (Gö6976) did not significantly inhibit TNF-α-induced IL-18. These results indicate that TNF-α induction of IL-18 by RA synovial fibroblasts occurs via ERK1/2, PKCδ and Src, but not via NFκB.
RA synovial fibroblasts were preincubated with or without PD98059 for 2 hours before TNF-α stimulation (20 ng/ml) for 48 hours. IL-18 bioactivity was then determined in the conditioned medium using KG-1 cells. TNF-α induced IL-18 bioactivity and this induction was reduced by 53% when the ERK1/2 pathway was blocked by chemical inhibitor (Figure 5B; P < 0.05; n = 7). These results indicate a crucial role of the ERK1/2 pathway in regulating TNF-α-induced-IL-18 bioactivity.
RA synovial fibroblasts were preincubated with IL-18BPa-Fc or IgG-Fc 1 hour prior to stimulation with TNF-α. Phospho ERK1/2 was then detected in cell lysates after 20 minutes. As expected, TNF-α induced ~11-fold phosphorylation of ERK1/2 (Figure 5C; P < 0.05; n = 3), while treatment with IL-18BPa-Fc at 25 ng/ml and 50 ng/ml reduced this phophorylation by 42% and 68%, respectively (P < 0.05 for 50 ng/ml; n = 3). Preincubation with IgG-Fc had no observable effect on ERK1/2 phosphorylation. These results suggest an important effect of IL-18BPa-Fc on TNF-α-induced ERK1/2 phosphorylation.
TNF-α induces phosphorylation of Src, ERK1/2, JNK2, and PKCδ in RA synovial fibroblasts in a time-dependent manner, with the maximal response at 15-30 minutes (Figures 6A). Western blotting was performed to determine which of the phosphorylated kinases might be upstream or downstream of the others. The phosphorylation of PKCδ and c-Jun were reduced by the JNK2 inhibitor (Figures 6B), demonstrating that PKCδ and c-Jun are downstream of JNK2.
The results from this inhibitor study showed that TNF-α enlists differential pathways in RA synovial fibroblasts for IL-18 and IL-18BPa expression. These signaling pathways are depicted in Figure 6C and the effect of IL-18BPa are represented in Figure 6D.
Recently higher levels of IL-18 were observed in the synovial fluid, synovial tissue, and sera of RA patients as compared to OA patients (8, 25, 26). Previous studies have shown that RA synovial tissues expressed more IL-18 mRNA than OA tissues, and spontaneously released larger amounts of IL-18 protein (25). Our study confirms that IL-18BPa levels in RA synovial fluid are lower than in OA synovial fluid (26). Furthermore, free or bioactive IL-18 was higher in RA synovial fluid than in OA synovial fluid and was related to a lower level of IL-18BPa (26). Since IL-18BPa regulates IL-18 bioactivity, IL-18BPa needs to be assessed for free IL-18 or IL-18 bioactivity, as suggested previously by the discordance between the IL-18 bioactivity and the protein level of IL-18 in RA synovial fluid. In fact, the IL-18 bioactivity was ~17-fold lower than its total protein level (25). Similar modulation by the natural inhibitor was recently reported with IL-1β and its natural inhibitor, IL-1Ra. Despite a higher level of IL-1Ra in RA synovial fluid than in OA synovial fluid, the ratio of IL-1Ra/IL-1 was higher in OA synovial fluid than in RA synovial fluid (27), and explained in part the mild to moderate inflammation seen in OA versus RA. Taken together, these data suggest that inhibition of IL-18 bioactivity can be attained by increasing IL-18BPa.
RA synovial fibroblasts constitutively produced IL-18BPa. IFN-γ is an important regulator of IL-18BPa expression (28). As IFN-γ levels are low in the RA joint (29), we focused our studies on TNF-α, a cytokine known to be important in RA. To our knowledge, this is the first report addressing the mechanism of TNF-α-induced IL-18BPa synthesis. Few other cytokines have been described which upregulate IL-18BPa production, as IL-12 in RA synovial tissue cells or activated peripheral blood mononuclear cells (PBMCs) after 7 days of stimulation (28). IL-1Ra in combination with IFN-γ also induced IL-18BPa in a human epithelial cell line after 17 hours of stimulation (30). The combination of TNF-α, IL-1β and IFN-γ appeared to induce more IL-18BPa in RA synovial fibroblasts than IFN-γ alone after 48 hours of stimulation (21). We also confirmed that IL-18 has an effect on IL-18BPa production (28). Thus, the delay of TNF-α's effect on IL-18BPa production is consistent with previous studies using other stimuli. TNF-α transduces its signal by binding to TNF receptors, which can be divided into two subtypes 1 and 2 in RA synovial fibroblasts (31). Previously, we described a bioassay to assess the circulating TNF-α bioactivity from RA patients using the ability of RA synovial fibroblasts to produce IL-6 in response to TNF-α (32, 33). After 48 hours, TNF-α induces IL-6 production in a dose dependent manner (32), suggesting that at this time point all TNF-α likely (at 20 ng/ml) is bound and therefore consumed in the culture supernatant.
Despite the induction of IL-18BPa by TNF-α, TNF-α also induced IL-18 in RA synovial fibroblasts, as previously described in such different cell types as human adipocytes (34), rat intestinal epithelial cells (35) or rat cardiomyocytes (36). Besides, the ratio between IL-18/IL-18BPa remained equal over 48 hours despite few nonsignificant variations. Caspase-1 activity is necessary to cleave pro-IL-18 to its active IL-18 form (5). Previously, TNF-α was also described to induce caspase-1 transcription in a human lung carcinoma cell line (37). Here, we found, for the first time in RA synovial fibroblasts, that TNF-α induces caspase-1 transcription and activity. These data suggest that TNF-α-induced-IL-18 and caspase-1 in the same proportion. Therefore, IL-18 induced by TNF-α is bioactive as confirmed by KG-1 cells.
The signaling mechanisms involved in TNF-α-induced-IL-18BPa production in RA synovial fibroblasts have not been examined previously. In the current study, we found that TNF-α-induced-IL-18BPa production in RA synovial fibroblasts is dependent on JNK2, PKC, Src, and NFκB. TNF-α is known to activate JNK2, PKC, ERK1/2, and NFκB in RA synovial fibroblasts (18, 38-40). However, JAK2 and ERK1/2 inhibitors did not significantly inhibit TNF-α-induced-IL-18BPa production. Conversely, TNF-α-induced IL-18 secretion is dependent upon ERK1/2, PKCδ, and Src. So, in order to reduce IL-18 bioactivity, we showed that blocking ERK1/2 was able to alter the IL-18/IL-18BPa ratio by suppressing IL-18 and enhancing IL-18BPa expression and so reduced IL-18 bioactivity. Among the three most well-characterized mammalian MAP kinase pathways which include the ERK, JNK, and p38 pathways, the ERK pathway also enhances the production of a variety of proinflammatory cytokines, such as TNF-α (41). Moreover, the ERK pathway is a survival pathway which was found to be activated in the RA synovium (42, 43). Furthermore, TNF-α induces activation of the ERK pathway in human RA synovial fibroblast (39). TNF-α was also shown to stimulate ERK activity in synovial tissue in vivo in the TNF-α-transgenic mouse (44), an effect inhibited by blockade of TNF-α. Thus, these studies provided additional data suggesting that the ERK1/2 pathway is a central pathway for IL-18 activity induced by TNF-α.
After finding that the ERK1/2 pathway was a key pathway controlling TNF-α-induced IL-18 bioactivity, we investigated the putative effect of IL-18BPa-Fc on ERK1/2 phosphorylation induced by TNF-α. We observed that IL-18BPa-Fc reduced activation of the ERK1/2 pathway induced by TNF-α. This data suggests a new feedback loop to control free IL-18, and to extend IL-18 bioactivity. IL-18BPa has been described to downregulate IFN-γ expression induced by IL-18, defining a different feedback loop (45, 46) in blood and in RA synovial fibroblasts (21). IL-18 induces TNF-α in RA synovial fibroblasts (8, 25) or human monocytes (47).
In terms of TNF-α signaling in RA synovial fibroblasts, we did not identify crosstalk between Src, ERK1/2, and NFκB. The finding that the JNK2 inhibitor reduced TNF-α-induced activation of PKCδ and NFκB in RA synovial fibroblasts suggests that JNK2 is upstream of PKCδ and NFκB.
As described herein, TNF-α activates PKCδ, ERK1/2, and JNK2 in RA synovial fibroblasts. These kinases are reported to regulate production of various proinflammatory mediators in RA synovial fibroblasts (23, 38, 42, 48, 49). With specific regard to the expression of IL-18 and IL-18BPa, we have demonstrated a critical role of the Src, ERK1/2, and PKCδ pathways for IL-18 and Src, PKC, JNK2, and NFκB pathways for IL-18BPa. In the present study, we found that PKCδ was regulated by JNK2, the observation that was consistent with the previous results using IL-18 as stimulus in RA synovial fibroblasts (23).
We propose the role of IL-18 in RA as follows: TNF-α induces both pro-IL-18 and caspase-1, which cleaves pro-IL-18 to active IL-18. However, TNF-α also induces IL-18BPa. Furthermore, IL-18 is known to induce TNF-α by RA synovial fibroblasts. This is thus a positive feedback loop that could explain the cytokine predominance in RA. The main known function of IL-18BPa is to reduce IL-18 bioactivity. We observed that blocking the ERK pathway reduces TNF-α-induced-IL-18 expression without interfering with TNF-α-induced-IL-18BPa expression. Therefore, blocking the ERK pathway reduced TNF-α-induced-IL-18 bioactivity. Furthermore, IL-18BPa itself is able to reduce TNF-α-induced-phophorylated ERK1/2. This suggests that there may be a negative feedback loop with synthesis of IL-18BPa downregulating TNF-α-induced-IL-18, in a manner similar to that described for IFN-γ. Thus, like other genes encoding cytokine inhibitors (soluble receptors, receptors antagonists and binding proteins), the cytokine itself or a related cytokine induces its own negative regulator in a feedback loop. Furthermore, this feedback loop was suggested by a previous in vivo study (50). Recombinant human IL-18BPa was able to reduce TNF-α level in colon homogenates in a mouse model of Crohn's disease, which is an inflammatory disorder in which IL-18 plays a crucial role (50). Overall, this study suggests that the presence of IL-18BPa, either by endogenous induction or exogenous addition, may contribute to the regulation of TNF-α-induced-IL-18 bioactivity in RA synovial fibroblasts, and suggests that IL-18BPa may be considered as a potential therapeutic strategy for RA.
The authors declare that they have no competing interests.
The authors thank the National Disease Research Interchange for providing RA synovial tissues and Dr. K. W. Janczak for providing the KG-1 cell line.
This study was supported by National Institutes of Health grants AI-40987 (A.E.K), AR-48267 (A.E.K.), AT-003633 (S.A.), AR-055741 (S.A.), AR-049907 (J.H.R), and AR-048310 (J.H.R); the Frederick G.L. Huetwell and William D. Robinson, M.D. Professorship in Rheumatology (A.E.K.); the Office of Research and Development, Medical Research Service, Department of Veterans Affairs (A.E.K.); the French Society of Rheumatology (H.M.), Lavoisier Foundation (H.M.), and Philippe Foundation (H.M.).