A Novel Role for Interleukin-18 in Adhesion Molecule Induction through NF (cid:1) B and Phosphatidylinositol (PI) 3-Kinase-dependent Signal Transduction Pathways*

Interleukin-18 (IL-18) is a novel proinflammatory cytokine found in serum and joints of patients with rheumatoid arthritis (RA). We studied a novel role for IL-18 in mediating cell adhesion, a vital component of the inflammation found in RA and other inflammatory diseases. We examined the expression of cellular cell adhesion molecules E-selectin, vascular cell adhesion mole-cule-1 (VCAM-1), and intercellular adhesion molecule-1 (ICAM-1) on endothelial cells and RA synovial fibroblasts using flow cytometry. Adhesion of the monocyte-like cell line HL-60 to endothelial cells was determined by immunofluorescence. IL-18 significantly enhanced ICAM-1 and VCAM-1 expression on endothelial cells and RA synovial fibroblasts. In addition, IL-18 induced E-selectin expression on endothelial cells and promoted the adhesion of HL-60 cells to IL-18-stimulated endothelial cells. Neutralizing anti-VCAM-1 and anti-E-selectin could completely inhibit HL-60 adherence to endothelial cells. IL-18-induced adhesion molecule expression appears to be mediated through nuclear factor (cid:1) B (NF (cid:1) B) and phosphatidyl-inositol

Adhesion molecules have been classified based on structure into three major groups: selectins, integrins, and the immunoglobulin (Ig) supergene family. The Ig supergene family contains diverse proteins that share the same immunoglobulin amino acid domains including adhesion proteins, such as intercellular adhesion molecule-1 (ICAM-1) 1 and vascular adhesion molecule-1 (VCAM-1), as well as proteins not involved in adhesion, such as the T cell receptor or major histocompatibility complex proteins HLA-DR (1).
Rheumatoid arthritis (RA) is a chronic progressive rheumatic disease characterized by the proliferation of the synovial membrane, which leads to the degradation of articular cartilage and subchondral bone. Another characteristic of RA synovial tissue is the exuberant leukocyte infiltration often present around the newly formed blood vessels (2). These inflammatory cells in the RA synovium are derived from the peripheral blood, and there is clear evidence that adhesion molecules expressed on endothelial cells mediate the migration of the leukocytes into the synovial membrane (3). An important example of this process is the localized endothelial expression of VCAM-1 and the selective recruitment of mononuclear leukocytes through its integrin counterreceptor, very late antigen-4 (VLA-4) (4). Adhesion molecules are also involved in interactions between leukocytes and RA synovial fibroblasts (4). During this cell/cell contact, adhesion molecules act as costimulators resulting in the activation of transcription factors and the production of cytokines, metalloproteinases, or other effector molecules (5).
Interleukin-18 (IL-18) is a novel cytokine that has been classified in the IL-1 family in virtue of structural similarity to IL-1␤. IL-18 acts as a strong inducer and coinducer of interferon ␥ (IFN␥) production in T cells and natural killer cells (6). Recent studies suggest that IL-18 also has some proinflammatory activities independent of IFN␥; IL-18 induces the synthesis of tumor necrosis factor ␣ (TNF-␣), granulocyte-macrophage colony stimulating factor, nitric oxide, and chemokines by T cells and natural killer cells (7). Others and we showed that IL-18 operates through nuclear factor B (NFB) in natural killer cells, T lymphocytes, and RA synovial fibroblasts (8 -10). The signaling cascade leading to NFB activation involves several intermediary molecules that finally lead to the activation of inhibitory B (IB). Activated IB releases NFB, which migrates into the nucleus and induces gene transcription (8,11). IL-18 has been found in synovial tissue, and enhanced levels of IL-18 were measured in the joint and in the serum of RA patients (12). The role of IL-18 in the pathogenesis of RA remains poorly understood. However, recent evidence suggests that IL-18 enhances the infiltration of inflammatory cells into the synovial tissue (13,14). Therefore, we examined the ability of IL-18 to induce endothelial and RA synovial fibroblast adhesion molecule expression and the mechanism of this expression. Here we demonstrate that IL-18 up-regulates ICAM-1 and VCAM-1 either on RA synovial fibroblasts or human dermal microvascular endothelial cells (HMVECs), and we identify IL-18 as a novel inducer of E-selectin on HMVECs. Furthermore, IL-18 promotes endothelial cell-leukocyte adhesion and appears to act in this system via VCAM-1 and E-selectin. We provide evidence that IL-18 induces RA synovial fibroblast VCAM-1 expression through NFB. Finally, we report an alternative pathway involving PI 3-kinase, which influences the level of IL-18-induced VCAM-1 in RA synovial fibroblasts.
Cell Culture-Fibroblasts were isolated from synovium obtained from patients meeting the American College of Rheumatology criteria for RA who had undergone total joint replacement surgery (17). Fresh synovial tissues were minced and digested in solution of dispase, collagenase, and DNase. Synovial fibroblasts were cultured in RPMI 1640 supplemented with 10% FBS and 1% P/S. The cells were used at passage 5 or older, at which time they were a homogeneous population of fibroblasts. HMVECs were purchased from BioWhittaker. HMVECs were cultured in endothelial cell growth medium-2 for microvascular cells (BioWhittaker). HMVECs were used between passage 3 and 12. Immortalized human dermal endothelial cells (HMEC-1) were a generous gift of Dr. Edmund W. Ades of the Centers for Disease Control and Dr. Thomas Lawley of Emory University (Atlanta, GA). HMEC-1 were grown in endothelial cell basal medium supplemented with epidermal growth factor (10 ng/ml, Becton Dickinson), hydrocortisone (1 g/ml, Sigma), and 5% FBS. The promyelocytic leukemia HL-60 cells (HL-60) were cultured in RPMI 1640 supplemented with 10% FBS and 1% P/S. Synovial fibroblasts, HMVECs, and HMEC-1 were grown in 175-mm tissue culture flasks (Falcon) in their respective media at 37°C in a humidified atmosphere with 5% CO 2 . Upon reaching confluence, the cells were passaged by brief trypsinization as previously described (18).
Flow Cytometry-RA synovial fibroblasts or HMVECs were plated onto 6-cm Petri dishes (Falcon) at 1 ϫ 10 5 cells/ml and allowed to adhere overnight at 37°C in an incubator gassed with 5% CO 2. Cells were stimulated with IL-18 (5 nM) for 8 h, 16 h for VCAM-1, 4 h for E-selectin, and 24 h for ICAM-1 and HLA-DR. Cells were harvested with a cell scraper and transferred to fluorescence-activated cell sorting tubes (Becton Dickinson). Cells were then treated with mouse anti-VCAM-1, mouse anti-ICAM-1, mouse anti-E-selectin, mouse anti-HLA-DR, or isotype-matched control (5 g/ml) as primary antibody. Cells were then incubated for 30 min with a phycoerythrin-conjugated goat anti-mouse antibody (1.5 g/ml). Samples were washed twice with PBS, 1% FBS, and then fixed in 1% paraformaldehyde. Samples were assayed using an Epics XL-MCL flow cytometer (Beckman Coulter). Prior to data acquisition, the phycoerythrin channel was standardized using fluores-cent beads (Rainbow Beads, Spherotech). For each condition (no stimulation or stimulation with IL-18 or IFN␥), an isotype-matched control was used, and the value was subtracted from the test result.
Cell Adhesion Assay-96-well plates (Dynex Technologies) were placed under UV for 30 min and coated with 0.02% gelatin (Sigma). HMEC-1 were plated into the 96-well plate at a concentration of 5 ϫ 10 4 cells/well and incubated overnight at 37°C. Cells were stimulated with 0 or 5 nM IL-18 or 1.15 nM TNF-␣ for 6 h at 37°C/5% CO 2 . HL-60 were washed twice with PBS and adjusted to 5 ϫ 10 6 cells/ml in RPMI 1640, 1% P/S. HL-60 were incubated with 5 m calcein-acetoxymethyl ester (calcAM, Molecular Probes, Eugene, Oregon) for 30 min at 37°C. Cal-cAM is a fluorescent dye that is incorporated into the living cells and thereby fixed in the cells. HL-60 were then washed twice with prewarmed RPMI 1640, 1% P/S to remove unincorporated dye and then adjusted to 2.5 ϫ 10 6 cells/ml. Labeled HL-60 (25 ϫ 10 4 /100 l) were added to each well. Cells were incubated for 1 h at 37°C and then were washed very carefully four or five times with prewarmed PBS. Fluorescence was determined by measurement with a fluorescent plate reader (SpectraMAX Gemini, Molecular Devices) set to 494 nm for excitation and 517 nm for emission. Adhesion was automatically expressed in relative fluorescence units. For better comparisons of the differentially treated groups and to avoid the use of relative fluorescence units, the adhesion of HL-60 to unstimulated HMEC-1 was chosen as a reference. The adhesion index was therefore defined as a ratio of adhesion of HL-60 to stimulated HMEC-1 (in relative fluorescence units) to adhesion of HL-60 to unstimulated HMEC-1 (in relative fluorescence units). For experiments in which the adhesion to HMEC-1 was blocked, cells were treated with monoclonal antibodies to VCAM-1, E-selectin, or isotype mouse-matched control (2.5 g/ml) for 1 h at 37°C and 5% CO 2 . HL-60 were labeled with calcAM and added to the HMEC-1 as outlined above.
Western Blot Analysis-After various experimental treatments, cytoplasmic and nuclear extracts from synovial fibroblasts were prepared as previously described (20). The concentration of protein in each extract was determined using a bicinchoninic acid assay (Pierce) using bovine serum albumin as the standard. Cell extracts were separated on standard SDS-polyacrylamide gel electrophoresis according to the method of Laemmli (19) and transferred to nitrocellulose membranes using a semi-dry transblotting apparatus (Bio-Rad, Hercules, CA). Nitrocellulose membranes were blocked with 5% nonfat milk in Trisbuffered saline Tween buffer (TBST; 20 mM Tris, 137 mM NaCl, pH 7.6, with 0.1% Tween 20) for 60 min at room temperature. Blots were incubated with anti-RelA at 1:1000 in TBST containing 5% nonfat milk for 14 h at 4°C. Blots were washed three times for 30 min at room temperature and then incubated in horseradish peroxidase using the enhanced chemiluminescence (ECL) reagents (Amersham Pharmacia Biotech) as per the manufacturer's instructions. Blots were scanned and analyzed for the measurement of the band intensities with the UN-SCAN-IT version 5.1 software (Silk Scientific, Orem, Utah).
Statistical Analysis-Data were expressed as the mean Ϯ S.E. Group means were compared with a Student's t test. p values Ͻ 0.05 were considered statistically significant.  IL-18 Induces the Adhesion of HL-60 to HMEC-1-The functional significance of the adhesion molecules induced by IL-18 was evaluated with an adhesion assay in which HMEC-1 cultured in 96-well plates were prestimulated with IL-18 or positive control TNF-␣ and coincubated with the monocyte precursor cell line HL-60. These cells are known to express high amounts of very late antigen (VLA)-4, lymphocyte functionassociated antigen-1, and Lewis X (20,21), three known ligands for VCAM-1, ICAM-1, and E-selectin, respectively. HMEC-1, which are SV 40-transformed normal HMVECs, were employed in place of HMVECs because they grow rapidly and provide a ready source of endothelial cells. Furthermore, HMEC-1 retain the morphologic, phenotypic and functional characteristics of HMVECs (22). To allow comparison of the different experimental conditions at each incubation time point we used an adhesion index. Following stimulation with IL-18 (5 or 10 nM) for 4 and 6 h, the adhesion index was increased between 1.6-to 2-fold as compared with nonstimulated HMEC-1 (Fig. 3). This effect completely declined after 24 h stimulation with IL-18. Interestingly, we observed similar kinetics when HMEC-1 were stimulated with TNF-␣ (1.15 nM). Indeed, the adhesion index peaked at 4 and 6 h and returned to the baseline at 24 h.

IL-18 Up-regulates ICAM-1 and VCAM-1 Expression on RA
Adhesion Blockade by Monoclonal Antibodies to VCAM-1 and to E-selectin-To investigate the role of VCAM-1 and E-selectin expression in IL-18-stimulated HMEC-1 adhesion, HMEC-1 stimulated with IL-18 were incubated with monoclonal anti-VCAM-1 (monoclonal antibodies 4B9 and GH12), anti-E-selectin (monoclonal antibodies BB11) or isotype-matched control (2.5 g/ml) antibodies (Fig. 4). Although clones 4B9 and GH12 detect two different epitopes of VCAM-1, they both block VLA-4/VCAM-1 binding. Pretreatment of IL-18-stimulated HMEC-1 with anti-VCAM-1 or anti-E-selectin antibodies completely blocked the adhesion of HL-60 to HMEC-1 as assessed by the adhesion index, which is equal to 1. Interestingly, there was also a significant decrease of the adhesion index when TNF-␣stimulated HMEC-1 were preincubated with the same antibodies (data not shown).
NFB Activation Induced by IL-18 Is Inhibited by PDTC-We previously showed that IL-18 increases the translocation of NFB p65 to the nucleus in RA synovial fibroblasts by Western blotting (10). This translocation occurred at 1 h and was sustained for at least 4 h. Treatment of RA synovial fibroblasts with PDTC (300 M) for 1, 2, 4, 6, or 24 h prior to the addition of 5 nM IL-18 inhibited the translocation of NFB into the nucleus after 2 h of incubation with PDTC (Fig. 5).
PDTC and NAC Inhibit IL-18-induced VCAM-1 Expression on RA Synovial Fibroblasts-The effect of NFB inhibition on VCAM-1 expression was evaluated by flow cytometry. RA synovial fibroblasts were incubated for 12 h with PDTC (300 M) prior to the addition of IL-18 (5 nM) for 8 h. As shown in Fig. 6A, PDTC inhibited IL-18-induced VCAM-1 expression by 50%. To confirm this result, we tested NAC, an additional antioxidant which is structurally distinct from PDTC. The pH of NAC was adjusted to 7.3 prior to the incubation with RA synovial fibroblasts because the acidity of NAC was toxic to the cells. Prior to IL-18 stimulation, NAC was added at a final concentration of 25 M for 2 h. As shown in Fig. 6B, NAC effectively inhibited IL-18 activation of VCAM-1 by 60%. The concentration of PDTC and NAC used in all experiments had no cytotoxic effect during 24 h of observation, since the viability of the cell using propidium iodide uptake as an indicator was greater than 95% (data not shown). Antioxidant PDTC and NAC partially inhibit VCAM-1 expression, suggesting that there are probably other intracellular pathways that mediate VCAM-1 induction.

Inhibition of PI 3-Kinase Down-regulates IL-18-induced VCAM-1 Expression-
To determine possible alternative pathways involved in IL-18-induced VCAM-1 expression, we therefore tested different specific signaling inhibitors including a PI 3-kinase inhibitor (LY294002), a src kinase inhibitor (PP2), a mitogen-activated protein kinase (MAPK) p38 inhibitor (SB203580), and a Erk1/Erk2 inhibitor (PD98059). RA synovial fibroblasts were pretreated with the specific signaling inhibitors at 10 M or Me 2 SO vehicle control for 1 h and then stimulated with IL-18 (5 nM) for 12 h. Flow cytometry was performed to measure cell surface VCAM-1 expression on RA synovial fibroblasts. Data presented in Fig. 7 indicate that PI 3-kinase is involved in IL-18 signaling since LY294002 significantly down-regulates VCAM-1 expression on IL-18-stimulated RA synovial fibroblasts. This effect was not a toxic effect of LY294002 since the viability of the cells evaluated by trypan blue exclusion was greater than 90%.

Inhibition of PI 3-Kinase and NFB Almost Completely
Inhibits VCAM-1 Expression-To determine whether PI 3-kinase and NFB act sequentially in the same pathway or through separate pathways, we studied the effect of simultaneous addition of the PI 3-kinase inhibitor (LY294002) with the NFB inhibitor (PDTC). RA synovial fibroblasts were pretreated with PDTC (300 M) for 14 h and LY294002 (10 M) for 1 h and then stimulated for 8 h with IL-18 (5 nM). VCAM-1 expression on RA synovial fibroblasts was detected by flow cytometry. The viability of the cells evaluated by trypan blue was greater than 80%. When applied together, PDTC and LY294002 decreased VCAM-1 expression on IL-18-stimulated RA synovial fibroblasts to the level of that found on nonstimulated cells (Fig. 8). DISCUSSION Adhesion molecules have been extensively studied in RA. Immunostaining of RA synovial tissue revealed that ICAM-1 and VCAM-1 are expressed on synovial fibroblasts and endo- thelial cells, whereas E-selectin is expressed only on endothelial cells. Further up-regulation of these adhesion molecules occurs in vitro following exposure to inflammatory cytokines TNF-␣ and IL-1␤ (23). The novel proinflammatory cytokine IL-18 has been demonstrated in synovial fluid, synovial tissue, and serum of patients with RA, but its role in the pathogenesis of the disease remains unclear. Here, we report that IL-18 up-regulates ICAM-1 and VCAM-1 expression on endothelial cells and RA synovial fibroblasts, and we extend the known effects of IL-18 to include its capacity to induce endothelial cell E-selectin expression. Indeed, the ability of IL-18 to induce E-selectin had never been described to date in any cell type. In fact, very few studies exist on the role of IL-18 in relation to cellular adhesion molecules. In myeloid KG-1 cells, IL-18 increases expression of ICAM-1 and mediates heterotypic aggregation between the KG-1 myeloid cell line and the Peer T cell line (24). This interaction was blocked by anti-ICAM-1. In melanoma cells (B16M), IL-18 up-regulates VCAM-1 expression as well as the adhesion of B16M to the hepatic sinusoidal endothelium (25). B16M adherence to hepatic sinusoidal endothelium was blocked by anti-VCAM-1 antibody. In this report, we show that IL-18 stimulates endothelial cells to increase binding to promyelocytic leukemia HL-60. The adhesion of HL-60 to HMEC-1 was maximal after 4 and 6 h of stimulation with IL-18. These kinetics suggested that VCAM-1 and/or Eselectin might be involved in this process. Hence, we tested the effect of monoclonal antibodies to VCAM-1 and E-selectin on HL-60 adherence to IL-18 stimulated HMEC-1. Blocking VCAM-1 or E-selectin completely reduced the adhesion of HL-60 to IL-18-stimulated endothelial cells.
The ability of IL-18 to promote cell adhesion is particularly interesting since binding of leukocytes to endothelial cells is the first step in the emigration of leukocytes into perivascular space, an important early development of chronic inflammation. This, in combination with the coordinated generation of chemokines by IL-1-stimulated RA synovial fibroblasts, as we previously described (10), provides an important mechanism promoting the migration of leukocytes into and within the synovium. Data obtained from animal models tend to support this hypothesis. Indeed, mice immunized with type II collagen in incomplete Freund's adjuvant and treated with IL-18 exhibit an extensive inflammatory infiltrate consisting of mononuclear cells (13). Furthermore, IL-18-defective mice develop a less severe collagen-induced arthritis than wild-type mice with a lower histological cellular infiltration in the joints (14). As compared with wild-type mice, cells isolated from arthritic IL-18-deficient mice also produce significantly lower amounts of TNF-␣ and IFN␥. Since IL-18 has the ability to induce TNF-␣ and IFN␥ in vitro, it is possible that IL-18 exerts its proinflammatory effects, including adhesion molecule expression, via these cytokines. However, in vitro, the neutralization of TNF-␣ and IFN␥ did not block the IL-18 induced up-regula- tion of ICAM-1 in KG1 cells and VCAM-1 in B16M cells. In our system, RA synovial fibroblasts and endothelial cells are not known to produce either TNF-␣ or IFN␥ (26 -28). Moreover, the ability of IL-18 to activate NFB, as we reported previously (11), strongly suggests that IL-18 induces RA synovial fibroblast and endothelial cell adhesion molecule expression by itself. Indeed, NFB is known to regulate the transcription of ICAM-1, VCAM-1, and E-selectin (29 -31).
To examine the participation of NFB in IL-18-induced adhesion molecule expression, we tested the impact of inhibitors of NFB on RA synovial fibroblast VCAM-1 expression. These inhibitors, PDTC and NAC, are structurally different but prevent NFB activation through their antioxidant property. PDTC and NAC are believed to reversibly suppress the release of an inhibitor subunit, IB, from the latent cytoplasmic form of NFB in cells treated with inflammatory mediators (32). Therefore, there should be no migration of NFB into the nucleus. This is exactly what we observed in RA synovial fibroblasts, since the level of NFB present in the nucleus of IL-18 stimulated fibroblasts following 2 h of preincubation with PDTC is similar to the level of NFB present in nonstimulated cells. On the basis of this finding, we assessed the inhibitory effect of PDTC on IL-18-induced VCAM-1 expression. PDTC reduced the level of VCAM-1 detected on RA synovial fibroblasts by 50% following IL-18 or IL-1␤ stimulation. This result was confirmed using NAC. Inhibition of VCAM-1 expression is partial, whereas the concentrations of PDTC and NAC are optimal to block NFB activation during exposure to IL-18. Interestingly, in a previous study we showed that treatment of cells with antisense NFB p65 reduced IL-18-induced IL-8 production by about 50% (10). These results are consistent with Thomas et al. (33) who demonstrated that splenocytes extracted from mice deficient in IL-1R-associated kinase (IRAK), a serine-threonine kinase involved in NFB activation following stimulation with IL-18 (8), secrete one-half the amount of IFN␥ protein produced by wild-type cells. Hence, blocking NFB seems to partially inhibit the effect of IL-18. These observations strongly suggest that there are alternative pathways, which can compensate for the inactivation of NFB. This hypothesis is supported by Kalina et al. who demonstrated that IL-18 activates the transcription factor STAT3. The same authors reported that IL-18 may operate through the MAPK p38 since an inhibitor of this MAPK partially blocks the production of IFN␥ in the natural killer 92 cell line (34).
To explore alternative pathways, we tested different inhibitors of intracellular signaling intermediates on VCAM-1 expression. In RA synovial fibroblasts, surprisingly, MAPK inhibitors (SB203580 and PD98059) and the src kinase inhibitor (PP2) did not have any effect on VCAM-1 expression, whereas the PI 3-kinase inhibitor (LY294002) reduced VCAM-1 expression by 50%. PI 3-kinase is a serine threonine kinase, which can become activated by several pathways (35). The mechanism of activation of PI 3-kinase by IL-18 could be either direct or indirect. The indirect mechanism could involve a cytokine produced by IL-18-stimulated RA synovial fibroblasts, which may act in an autocrine fashion to induce VCAM-1 expression through PI 3-kinase activation. For example, growth factors are known to act through PI 3-kinase (35) and can modulate adhesion molecule expression in endothelial cells (36). Alternatively, IL-18 could directly activate PI 3-kinase following binding to its receptor. PI 3-kinase relays the signaling of G proteincoupled receptors (37). Hence, it is possible, although not described to date, that a G protein-coupled IL-18 receptor could exist. A similar hypothesis has been discussed for IL-1 to explain the modulating activity of G protein inhibitors and activators on IL-1␤-induced ICAM-1 expression (38). Another possibility might be that IL-18 could activate PI 3-kinase through a receptor tyrosine kinase pathway. Mechanistically, this pathway could involve the small GTPase Ras, or it might rely on a direct binding of PI 3-kinase to receptor tyrosine kinases (35). To activate transcription factors, PI 3-kinase has been demonstrated to operate through Akt (39) or small GTPases proteins such as Cdc42, Rac, and Raf-1 (37). Min and Pober demonstrated that these small GTPases initiate E-selectin transcription in human endothelial cells through a parallel pathway to NFB involving JNK-c-jun/ATF2 (40).
Likewise, we hypothesized that IL-18 might induce adhesion molecule expression through two parallel pathways. Experiments using PI 3-kinase and NFB inhibitors together strongly support this hypothesis since LY294002 in combination with PDTC virtually inhibited IL-18-induced VCAM-1 expression. These data suggest that IL-18 mediates VCAM-1 synthesis through two distinct pathways, one involving NFB and the other involving PI 3-kinase. To induce VCAM-1 expression, PI 3-kinase may use transcription factors other than NFB such as AP1 and STAT3 that are already known to be activated by PI 3-kinase (41,42). However, we can not exclude that PI 3-kinase may use NFB, at least in part, to induce VCAM-1 transcription. Indeed, it is possible that PI 3-kinase may initiate an activation of NFB that is not inhibited by PDTC. This theory is supported by Sizemore et al. who showed that PI 3-kinase could regulate NFB-dependent promoter expression independently of the IB degradation-NFB liberation pathway (43).
In conclusion, IL-18 induces the expression of functional adhesion molecules in two different cell types, endothelial cells and fibroblasts. IL-18-induced RA synovial fibroblast VCAM-1 expression is regulated by a NFB and PI 3-kinase suggesting that there might be two signaling pathways involved in IL-18induced adhesion molecule expression. Blocking IL-18 or its signaling pathways may point to a new therapeutic avenue for the treatment of inflammatory diseases like RA.