Drs. Kyoung-Woon Kim and Cho contributed equally to this work.
Up-regulation of stromal cell–derived factor 1 (CXCL12) production in rheumatoid synovial fibroblasts through interactions with T lymphocytes: Role of interleukin-17 and CD40L–CD40 interaction
Article first published online: 28 MAR 2007
Copyright © 2007 by the American College of Rheumatology
Arthritis & Rheumatism
Volume 56, Issue 4, pages 1076–1086, April 2007
How to Cite
Kim, K.-W., Cho, M.-L., Kim, H.-R., Ju, J.-H., Park, M.-K., Oh, H.-J., Kim, J.-S., Park, S.-H., Lee, S.-H. and Kim, H.-Y. (2007), Up-regulation of stromal cell–derived factor 1 (CXCL12) production in rheumatoid synovial fibroblasts through interactions with T lymphocytes: Role of interleukin-17 and CD40L–CD40 interaction. Arthritis & Rheumatism, 56: 1076–1086. doi: 10.1002/art.22439
- Issue published online: 28 MAR 2007
- Article first published online: 28 MAR 2007
- Manuscript Accepted: 6 DEC 2006
- Manuscript Received: 18 APR 2006
- Korea Science and Engineering Foundation through the Rheumatism Research Center at Catholic University of Korea. Grant Number: R11-2002-07003-0
Stromal cell–derived factor 1 (SDF-1) is a potent chemoattractant for memory T cells in inflamed rheumatoid arthritis (RA) synovium. This study was undertaken to investigate the effect of interleukin-17 (IL-17) and CD40–CD40L interaction on SDF-1 production in RA fibroblast-like synoviocytes (FLS).
Synovial fluid (SF) and serum levels of SDF-1 in RA patients were measured by enzyme-linked immunosorbent assay (ELISA). The SDF-1 produced by cultured RA FLS was evaluated by real-time polymerase chain reaction and ELISA after FLS were treated with IL-17 and inhibitors of intracellular signal molecules. The SDF-1 level was also determined after FLS were cocultured with T cells in the presence and absence of IL-17.
Concentrations of SDF-1 in the sera and SF were higher in RA patients than in osteoarthritis patients, although the increase in the serum levels did not reach statistical significance. The production of SDF-1 in RA FLS was enhanced by IL-17 stimulation. This effect of IL-17 was blocked by inhibitors of phosphatidylinositol 3-kinase (PI 3-kinase), NF-κB, and activator protein 1 (AP-1). When FLS were cocultured with T cells, SDF-1 production was up-regulated, especially in the presence of IL-17, but FLS were inhibited by neutralizing anti–IL-17 and anti-CD40L antibodies. Addition of RA SF to cultured RA FLS significantly up-regulated SDF-1 messenger RNA expression, which was hampered by pretreatment with anti–IL-17 antibody.
SDF-1 is overproduced in RA FLS, and IL-17 could up-regulate the expression of SDF-1 in RA FLS via pathways mediated by PI 3-kinase, NF-κB, and AP-1. Our findings suggest that inhibition of the interaction between IL-17 from T cells and SDF-1 in FLS may provide a new therapeutic approach in RA.
Rheumatoid arthritis (RA) is a systemic inflammatory disease characterized by synovial inflammation and progressive joint destruction. Although the definite etiology remains unknown, the complex and delicate networks of various inflammatory cells, cytokines, and chemokines may play a central role in the pathogenesis of RA. RA synovium is characterized by the infiltration of T cells and monocytes as well as the proliferation of synoviocytes. These cells costimulate each other, resulting in a vicious circle of synovial inflammation.
The predominant T cell subset in the sublining of RA synovium is CD4+ cells. The majority of these CD4+ T cells are mature memory CD45RO+ T cells (1). The migration of T cells, B cells, and monocyte/macrophages into the inflamed synovium is promoted by a chemokine, stromal cell–derived factor 1 (SDF-1; CXCL12). SDF-1 is produced by bone marrow stromal cells, fibroblast-like synoviocytes (FLS), macrophages, and endothelial cells in RA. The concentration of SDF-1 is elevated in plasma and synovial fluid (SF) specimens from patients with RA (2, 3), and overexpression of SDF-1 is also observed in RA synovial tissue (4, 5).
SDF-1 promotes CD4+ memory T cell recruitment and accumulation (6) and monocyte migration into synovium (7) and induces pseudoemperipolesis (e.g., migration into synovium) of both T and B cells through a vascular cell adhesion molecule 1–dependent mechanism (8). In addition, SDF-1 can promote joint destruction by inducing angiogenesis through immobilization of endothelial cells on heparin sulfate molecules and by inducing angiogenesis via activation of chondrocytes and osteoclasts through matrix metalloproteinase 3 (MMP-3) and MMP-9, respectively (5). Inhibition of SDF-1 decreases the development of collagen-induced arthritis (CIA) (9), and the inhibitor of CXCR4, which is an SDF-1 receptor, also reduces the incidence and clinical severity of CIA (10). SDF-1 potentially plays an important role in inflammatory response, neovascularization, and joint destruction in the pathogenesis of RA.
SDF-1 is generally known to be mainly a stimulating chemokine, and investigators have focused on its role in the stimulation and induction of other molecules. Therefore, the regulation and induction mechanism of SDF-1 remains unknown. It has been demonstrated that hypoxia enhances the expression of SDF-1 messenger RNA (mRNA) in synovial fibroblasts, but cytokines such as transforming growth factor β, interleukin-1β (IL-1β), and tumor necrosis factor (TNF) do not affect SDF-1 production (11).
IL-17, a major T cell–derived proinflammatory cytokine in RA synovium, stimulates FLS to produce inflammatory cytokines and chemokines (12). Previously, we observed reciprocal activation of antigen-stimulated T cells and FLS from RA patients, where IL-17 production from T cells increased upon coculture with FLS (13). Synovial tissue and FLS express IL-17 receptors (14, 15), and FLS have the potential to respond to IL-17 produced by activated T cells. CD40L, a member of the TNF superfamily, is a 30–33-kd type II transmembrane protein expressed on activated T cells, mast cells, basophils, and eosinophils (16, 17). It has been reported that stimulation with CD40L-expressing cells or purified recombinant CD40L induces the secretion of proinflammatory cytokines such as IL-1, IL-6, IL-8, and TNFα from monocytes, dendritic cells, epithelial cells, and fibroblasts, and augments the expression of adhesion molecules and metalloproteinases (18–22).
In this study, we hypothesized that the interaction between the T cell–derived cytokine IL-17 and FLS may affect the production of SDF-1 by FLS through some distinct pathway. We assessed whether IL-17 (a representative cytokine from activated T cells) and physical interplay between FLS and T cells through CD40–CD40L interaction have a role in the regulation of SDF-1 production. In addition, we attempted to identify which signaling pathways are involved in the reciprocal effect of synovial SDF-1 on T cell migration into the inflamed synovium.
PATIENTS AND METHODS
SF and sera were obtained from 20 RA patients fulfilling the 1987 revised criteria of the American College of Rheumatology (formerly, the American Rheumatism Association) (23). Twenty age- and sex-matched patients with osteoarthritis (OA) were studied as controls. Informed consent was obtained from all patients, and the experimental protocol was approved by the Catholic University of Korea Human Research Ethics Committee. Synovial tissue specimens were isolated from 8 additional RA patients with a mean ± SEM age of 63.4 ± 4.6 years (range 38–76 years) who were undergoing total knee replacement surgery.
Recombinant IL-17, CD40L, macrophage migration inhibitory factor (MIF), and anti–IL-17 neutralizing antibody were purchased from R&D Systems (Minneapolis, MN), recombinant TNFα and IL-1β from Endogen (Cambridge, MA), and concanavalin A (Con A) from Sigma (St. Louis, MO). Pyrrolidine dithiocarbamate (PDTC), curcumin, and parthenolide were obtained from Sigma. LY294002, wortmannin, PD98059, JNK inhibitor, SB203580, and SP600125 were obtained from Calbiochem (Schwalbach, Germany).
Isolation of FLS.
Synoviocytes were isolated by enzymatic digestion of synovial tissue specimens obtained from patients with RA and patients with OA undergoing total joint replacement surgery. The tissue samples were minced into 2–3-mm pieces and treated for 4 hours with 4 mg/ml of type I collagenase (Worthington, Freehold, NJ) in Dulbecco's modified Eagle's medium (DMEM) at 37°C in 5% CO2. Dissociated cells were then centrifuged at 500g, resuspended in DMEM supplemented with 10% fetal calf serum (FCS), 2 mML-glutamine, 100 units/ml of penicillin, and 100 μg/ml of streptomycin, and plated in 75-cm2 flasks. After overnight culture, the nonadherent cells were removed, and the adherent cells were cultivated in DMEM supplemented with 20% FCS. The cultures were kept at 37°C in 5% CO2, and the medium was replaced every 3 days. When the cells approached confluence, they were passaged after dilution (1:3) with fresh medium.
Synoviocytes from passages 4–8 were used in each experiment. The cells were morphologically homogeneous and exhibited the appearance of synovial fibroblasts, with typical bipolar configuration under inverse microscopy. The purity of cells (1 × 104) was tested by flow cytometric analysis using phycoerythrin (PE)–conjugated anti-CD14 (PharMingen, San Diego, CA) and fluorescein isothiocyanate–conjugated anti-CD3 or anti–Thy-1 (CD90) monoclonal antibodies (PharMingen). At passage 4, most cells (>95%) expressed the surface markers for fibroblasts (CD90+), whereas 3.5% of the cells were CD14+, and <1% of the cells were CD3+.
CD4+ T cell isolation by magnetic-activated cell sorting (MACS).
Anti-CD4 microbeads were used according to the recommendations of the manufacturer (Miltenyi Biotec, Sunnyvale, CA) (24). Peripheral blood mononuclear cells were resuspended in 80 μl FCS staining buffer. Anti-CD4 microbeads (20 μl) were added and incubated for 15 minutes at 6–12°C. Saturating amounts of fluorochrome-conjugated antibodies were added, and cells were incubated for an additional 10 minutes. Cells were diluted in 2.5 ml 2% FCS staining buffer, pelleted, resuspended in 500 μl FCS, and magnetically separated, usually on an AutoMACS magnet (Miltenyi Biotec, Bergisch Gladbach, Germany) fitted with a MACS mass spectrophotometry column. Flow-through and two 1-ml washes were collected as the negative fraction. Enriched cells were collected in two 0.5-ml aliquots from the column after removal from the magnet. Alternatively, cells stained with PE-conjugated anti-CD4 were washed, magnetically labeled with anti-PE microbeads (20 μl added to an 80-μl cell suspension for 15 minutes at 6–12°C), and magnetically separated as described above. The purity of cells was assessed by flow cytometric analysis of stained cells on a FACSVantage sorter. Most of the isolated cells (>97%) exhibited the CD4 T cell marker.
Coculture of RA T cells and FLS.
The FLS were seeded in 24-well plates at 5 × 104 cells/well in 1 ml serum-free DMEM/insulin–transferrin–selenium A (Life Technologies, Gaithersburg, MD). Subsequently, CD4+ T cells (5 × 105 cells/well) were added to the FLS monolayers, and the culture plates were incubated at a ratio of FLS to T cells of 1:10 for 48 hours alone or with 10–100 ng/ml of IL-17. The culture supernatants were collected and stored at −20°C until assayed. All cultures were set up in triplicate. In some experiments, neutralizing monoclonal antibodies to IL-17, CD40L, interferon-γ (all from R&D Systems), or isotype-matched mouse IgG1 were added to the cocultures for 48 hours.
Concentrations of SDF-1 determined by sandwich enzyme-linked immunosorbent assay (ELISA).
Concentrations of SDF-1 in sera and SF were measured by sandwich ELISA, as follows. Antibody to human SDF-1 (4 μg/ml; R&D Systems) was added to a 96-well plate (Nunc, Roskilde, Denmark) and incubated overnight at 4°C. After treatment with blocking solution (phosphate buffered saline [PBS] containing 1% bovine serum albumin and 0.05% Tween 20) for 2 hours at room temperature, test samples and the standard recombinant SDF-1 (R&D Systems) were added to the 96-well plate and incubated at room temperature for 2 hours.
Biotinylated SDF-1 polyclonal antibody to human cytokines (400 ng/ml; R&D Systems) was added after washing 4 times with PBS containing Tween 20, and the reactions were allowed to proceed for 2 hours at room temperature. After further washing, 2,000-fold diluted streptavidin–alkaline phosphate (Sigma) was added, and the reactions were again allowed to proceed for 2 hours. Fifty microliters of diluted avidin–peroxidase (1:2,000 in diluent) was added after 4 additional washes. After incubation for 2 hours at room temperature, 50 μl tetramethylbenzidine substrate solution (Kirkegaard & Perry, Guildford, UK) was added to each well and incubated for 20–30 minutes.
Initially, the reaction produced a blue color that was monitored by absorbance at 595 nm with a microplate reader (MRX Revelation; Dynex Technologies, Chantilly, VA). When the desired intensity was reached (optical density [OD] <0.8), sulfuric acid (2.0 moles/liter) was added to each 50-μl well to stop the color-generating reaction. An automated microplate reader (Vmax; Molecular Devices, Palo Alto, CA) set at 450 nm was used to measure OD. The limit of sensitivity for SDF-1 was 15.6 pg/ml. Recombinant human cytokines diluted in the culture medium were used as a calibration standard, ranging from 10 pg/ml to 2,000 pg/ml. A standard curve was drawn by plotting OD versus the log of the concentration of recombinant cytokine.
Expression of SDF-1 mRNA determined by reverse transcriptase–polymerase chain reaction (RT-PCR).
FLS were incubated with various concentrations of IL-17 in the presence or absence of various signal inhibitors (LY294002, wortmannin, SB203580, PD98059, JNK inhibitor, curcumin, SP600125, PDTC, and parthenolide). After 12 hours of incubation, mRNA was extracted using RNAzol B, according to the recommendations of the manufacturer (Biotecx, Houston, TX). Reverse transcription of 2 μg total mRNA was carried out at 42°C using the Superscript reverse transcription system (Takara, Shiga, Japan). PCR amplification of complementary DNA aliquots was performed by adding 2.5 mM dNTPs, 2.5 units Taq DNA polymerase (Takara), and 0.25 μM sense and antisense primers. The reaction took place in 25 μl of PCR buffer, consisting of 1.5 mM MgCl2, 50 mM KCl, and 10 mM Tris HCl (pH 8.3).
The following primers were used for each molecule: for SDF-1, 5′-ATG-AAC-GCC-AAG-GTC-GTG-GTC-3′ (sense) and 5′-TGG-CTG-TTG-TGC-TTA-CTT-GTT-T-3′ (antisense); for GAPDH, 5′-CGA-TGC-TGG-GCG-TGA-GTA-C-3′ (sense) and 5′-CGT-TCA-GCT-CAG-GGA-TGA-CC-3′ (antisense). Reactions were processed in a DNA thermal cycler (PerkinElmer Cetus, Wellesley, MA) through 25 cycles of 30 seconds of denaturation at 94°C, 1 minute of annealing at 56°C, followed by 30 seconds of elongation at 72°C. PCR products were run on a 1.5% agarose gel and stained with ethidium bromide. Results were expressed as the ratio of SDF-1 product to GAPDH product.
Expression of SDF-1, IL-17, and CD40L mRNA determined by real-time PCR with SYBR Green I.
Real-time PCRs were performed in 20-μl final volumes in capillary tubes in a LightCycler instrument (Roche Diagnostics, Mannheim, Germany). The following primers were used for each molecule: for SDF-1, 5′-ATG-AAC-GCC-AAG-GTC-GTG-GTC-3′ (sense) and 5′-TGG-CTG-TTG-TGC-TTA-CTT-GTT-T-3′ (antisense); for IL-17, 5′-TGG-AGG-CCA-TAG-TGA-AGG-3′ (sense) and 5′-GGC-CAC-ATG-GTG-GAC-AAT-3′ (antisense); for CD40L, 5′-CCA-GGT-GCT-TCG-GTG-TTG-3′ (sense) and 5′-GAC-GTG-AAG-CCA-GTG-CCA-T-3′ (antisense); for β-actin, 5′-GGA-CTT-CGA-GCA-AGA-GAT-GG-3′ (sense) and 5′-TGT-GTT-GGC-GAT-CAG-GTC-TTT-G-3′ (antisense).
Reaction mixtures contained 2 μl of LightCycler FastStart DNA mastermix for SYBR Green I (Roche Diagnostics), 0.5 μM each primer, 4 mM MgCl2, and 2 μl of template DNA. All capillaries were sealed, centrifuged at 500g for 5 seconds, and then amplified in a LightCycler instrument, with activation of polymerase (95°C for 10 minutes), followed by 45 cycles of 10 seconds at 95°C, 10 seconds at 60°C, and 10 seconds at 72°C. The temperature transition rate was 20°C/second for all steps. Double-stranded PCR product was measured during the 72°C extension step by detection of fluorescence associated with the binding of SYBR Green I to the product. Fluorescence curves were analyzed with LightCycler software, version 3.0. For quantification analysis of SDF-1, IL-17, or CD40L mRNA, LightCycler was used.
Relative expression levels of samples were calculated by normalizing SDF-1, IL-17, or CD40L levels to the endogenously expressed housekeeping gene (β-actin). Melting curve analysis was performed immediately after the amplification protocol under the following conditions: 0 second (hold time on reaching temperatures) at 95°C, 15 seconds at 65°C, and 0 second (hold time) at 95°C. The temperature change rate was 20°C/second, except in the final step, in which it was 0.1°C/second. The melt peak generated represented the specific amplified product. The crossing point was defined as the maximum of the second derivative from the fluorescence curve. Negative controls, which contained all the elements of the reaction mixture except template DNA, were also included. All samples were processed in duplicate.
Cell viability measured by trypan blue dye exclusion.
Trypan blue dye exclusion was performed as previously described (25) to evaluate the potential for direct cytotoxic effects of the chemical inhibitors on the cultured cells. Following 24 hours of incubation, the cells were harvested, and the percentage of cell viability was expressed using the formula 100 × (number of viable cells/number of both viable and dead cells).
Data are expressed as the mean ± SEM. Statistical analysis was performed using the Mann-Whitney U test for independent samples and Wilcoxon's signed rank test for related samples. P values less than 0.05 were considered significant.
SDF-1 concentrations in RA sera and SF.
Serum and SF levels of SDF-1 in 20 RA patients and 20 OA patients were measured by sandwich ELISA. The 20 patients with RA included 3 men and 17 women, with a mean ± SEM age of 50.4 ± 1.5 years (range 23–77 years) and a mean ± SEM disease duration of 71.5 ± 8.2 months (range 3–240 months). The mean ± SEM erythrocyte sedimentation rate was 40.2 ± 3.8 mm/hour, and the mean ± SEM C-reactive protein level was 2.6 ± 0.4 mg/dl in the patients with RA. Rheumatoid factor was positive in 24 of 28 patients (85.7%), at a mean ± SEM titer of 110.8 ± 27.7 IU/ml (range 6.7–1,060 IU/ml).
In RA patients, the mean ± SEM concentration of SDF-1 in SF was 1,001.5 ± 103.9 pg/ml. In patients with OA, the concentration was 779.2 ± 30.8 pg/ml. Statistical analysis revealed that RA patients had higher concentrations compared with OA patients (P = 0.04). The serum concentrations of SDF-1 tended to be higher in RA patients than in OA patients, although the difference was not statistically significant (749.3 ± 103.6 pg/ml versus 516.4 ± 31.4 pg/ml; P = 0.08) (Figure 1).
Regulation of RA FLS production of SDF-1 by various cytokines.
To determine the cytokines that cause SDF-1 overproduction in RA, FLS were stimulated with various inflammatory cytokines. RA FLS were isolated and cultured in the presence of 50 ng/ml IL-17, 10 ng/ml IL-1β, 10 ng/ml TNFα, 10 ng/ml CD40L, or 0.1 μg/ml Con A, and the concentrations of SDF-1 in the culture supernatants were determined by ELISA. When FLS were stimulated with exogenous IL-17, IL-1β, MIF, or CD40L, as well as with Con A, the production of SDF-1 increased significantly. However, TNFα did not affect the production of SDF-1 in the FLS culture supernatants (Figure 2).
Dose-dependent IL-17 enhancement of SDF-1 production by RA FLS.
To better characterize the effects of IL-17 on SDF-1 production by RA FLS, we performed dose-response studies on IL-17–induced SDF-1 production. After FLS was cultured with IL-17, the production of SDF-1 in culture supernatants was determined by ELISA, and the expression of SDF-1 mRNA by RA FLS was evaluated by real-time PCR. As shown in Figure 3, IL-17 treatment increased the production of SDF-1 in FLS culture supernatants and the expression of SDF-1 mRNA in RA FLS in a dose-dependent manner.
Signal pathways involving IL-17–induced SDF-1 production in RA FLS.
To determine the signal transduction pathways mediating the production of SDF-1 by IL-17, we used 100 μM PDTC and 10 μM parthenolide (inhibitors of NF-κB activation), 20 μM LY294002 and 100 nM wortmannin (inhibitors of phosphatidylinositol 3-kinase [PI 3-kinase] activation), and 10 μM SB203580 (an inhibitor of p38 MAPK). Curcumin (10 μM) and 1 μM SP600125 were tested as antagonists of activator protein 1 (AP-1), and 1 μM PD98059 as an inhibitor of MEK-1/2.
RA FLS were preincubated for 1 hour with the inhibitors, and stimulated with 10 ng/ml of IL-17 for 12 hours. The production of SDF-1 in the culture supernatants was determined by ELISA. SDF-1 production was decreased after inhibition of PI 3-kinase, NF-κB, and AP-1 (Figure 4A). In contrast, disruption of p38 MAPK and JNK activities showed no effect on IL-17–induced SDF-1 production. At the experimental concentrations used, the chemical inhibitors exhibited no cytotoxic effects on FLS (Figure 4B).
For determination of the inhibitory effects at the level of transcription, RA FLS were preincubated for 1 hour with LY294002, SP600125, or parthenolide, and cultured in the presence of 10 ng/ml of IL-17 for 12 hours. The expression of SDF-1 mRNA was determined by RT-PCR. SDF-1 mRNA expression was completely blocked when PI 3-kinase, NF-κB, and AP-1 were inhibited (Figure 5).
CD4+ T cell augmentation of SDF-1 production by RA FLS.
To determine the effect of T cells on SDF-1 production by FLS, we isolated and cultured FLS with or without CD4+ T cells for 24–48 hours. The SDF-1 concentration in the culture supernatants was determined using ELISA. The production of SDF-1 by CD4+ T cells was minimal, whereas RA FLS produced moderate basal levels of SDF-1 (mean ± SEM 273.7 ± 24.8 pg/ml). However, when RA FLS were cocultured with CD4+ T cells, SDF-1 production was augmented to 492.7 ± 49.6 pg/ml (P = 0.05 versus FLS cells cultured without T cells).
Since increased SDF-1 production was observed in the T cell–FLS coculture system, we hypothesized that T cell–derived cytokines and/or physical interaction between T cells and FLS may contribute to the augmentation of SDF-1 production by RA FLS. The addition of a neutralizing anti–IL-17 monoclonal antibody to cocultures of FLS and CD4+ T cells decreased SDF-1 production significantly (to 361.0 ± 46.3 pg/ml, versus 492.7 ± 49.6 pg/ml; P = 0.04) (Figure 6A).
To further define the nature of interaction between RA T cells and FLS, we tested the effect of inserting a barrier in the coculture chamber. Disrupting direct cell contact by placing the 2 cell types in chambers separated by a barrier reversed the induction effect on SDF-1 production, and SDF-1 levels returned almost, but not completely, to the levels produced by FLS alone.
In addition, pretreatment with anti-CD40L antibody, which blocks the CD40L–CD40 interaction between T cells and FLS, also significantly decreased SDF-1 production by RA FLS. Combined treatment with anti–IL-17 antibody and anti-CD40L antibody showed additive effects on the decreased production of SDF-1. A similar tendency was observed in OA FLS, but the difference was not as evident as with RA FLS (Figure 6A).
To investigate the role of exogenous IL-17 in the FLS–T cell coculture system, we added IL-17 in various concentrations to FLS–T cell cocultures. SDF-1 production was augmented in a dose-dependent manner when FLS was cocultured with CD4+ T cells in the presence of recombinant IL-17 (Figure 6B), suggesting a direct effect of IL-17 on the production of SDF-1 by RA FLS.
To identify time-dependent effects on the expression of IL-17 and CD40L mRNA in the FLS–T cell coculture system, T cells were cultured with or without FLS for 12–48 hours. IL-17 and CD40L mRNA expression were augmented at 24–48 hours (Figure 6C).
In addition, we tested whether IL-17 in RA SF specimens could induce SDF-1 production by RA FLS. Addition of RA SF (with a mean ± SEM IL-17 concentration of 790 ± 153 pg/ml) mimicked the effect of IL-17 treatment on SDF-1 production and mRNA expression. To further define the stimulatory effect of RA SF on SDF-1 production, we added neutralizing monoclonal antibodies to IL-17 to cultured FLS incubated with RA SF. The addition of anti–IL-17 antibody (10 μg/ml) abrogated the elevation of SDF-1 production induced by RA SF, whereas the equivalent concentration of isotype-matched control monoclonal antibody had no effect (Figure 6D). These findings support the notion that the IL-17 present at elevated levels in RA SF can directly stimulate SDF-1 production in RA FLS.
SDF-1 is a potent CXC chemokine which binds to its single receptor, CXCR4, and has a unique role in the regulation of cell retention or homing (26). Interaction between SDF-1 and CXCR4 plays an important role in many diseases, such as human immunodeficiency virus (27), cancer (28), autoimmune diabetes (29), and RA. Although the important roles of SDF-1 in the pathogenesis of RA have been demonstrated, the regulation mechanism of SDF-1 is not fully understood. Unlike other chemokines, SDF-1 is constitutively expressed in numerous normal tissue types (30). Therefore, it was thought that SDF-1 was not specifically regulated, and little attention has been paid to its regulation mechanism. The few previous studies of SDF-1 regulation that have been conducted have shown that anti-CD40 stimulation enhances the production of cultured RA FLS (6), hypoxia induces SDF-1 expression in RA FLS (11), and IL-1β and TNFα decrease the production and expression of SDF-1 in dermal and gingival fibroblasts (31).
In this study, we hypothesized that a specific molecule, such as a certain cytokine, is able to regulate the expression of SDF-1 in RA, and that the production and action of SDF-1 results from the interplay of T cells and FLS. T cells express CXCR4 (6) and respond to SDF-1 produced by FLS. It is possible that a specific molecule from T cells activated by SDF-1 may stimulate FLS. We hypothesized that IL-17, a T cell–derived proinflammatory cytokine, might stimulate FLS reciprocally. FLS express IL-17 receptors on their cell surface (14, 15) and have the potential to respond to stimulation by IL-17. IL-17 is mainly produced by CD4+,CD45RO+ memory T cells, and overexpression of IL-17 is observed in the SF and synovial tissue of RA patients (32, 33).
IL-17 plays a critical role in the inflammatory process in RA, and recently it has become a subject of special interest in RA pathogenesis (34, 35). It stimulates the production and expression of proinflammatory cytokines from monocyte/macrophages (34, 36) and IL-6 and IL-8 from RA synovial fibroblasts (12, 37). It also stimulates chemokines, such as CCL20 (38). Furthermore, IL-17 contributes to bone erosion and tissue destruction in RA. It induces chondrocytes and synovial fibroblasts to produce prostaglandin E2 (39) and up-regulates nitric oxide production (40). IL-17 induces IL-6, osteoclast differentiation factor, and RANKL production by T cells and osteoblasts (14). Taken together, these findings provide evidence that IL-17 regulates the inflammatory and destructive process in RA.
We found that IL-17 induced the production of SDF-1, as well as the expression of SDF-1 mRNA, in cultured FLS in a dose-dependent manner. This result implies that there is a reciprocal action between IL-17 from T cells and SDF-1 from FLS, in RA pathogenesis. T cells, which migrate into the inflamed synovium with the guidance of SDF-1 from FLS, produce IL-17, resulting in the induction of SDF-1 from FLS. SDF-1 overproduction induces the migration of T cells into the synovial tissue, resulting in the augmentation and acceleration of the inflammatory response and bone destruction in RA.
We used higher concentrations of IL-17 (1–50 ng/ml) than the actual level of IL-17 in SF (1–27 ng/ml) to stimulate RA FLS. Therefore, there was some discrepancy between the SF level of IL-17 and the concentration of the cytokine used in our experiments. This discrepancy can be rationalized by the fact that the cytokine environment in vivo is complicated and is influenced by other serum factors.
Of special interest in the present study was the result of experiments in which RA FLS were cocultured with CD4+ T cells. SDF-1 concentrations are negligible in the culture supernatants of CD4+ T cells; this means that T cells are not the source of SDF-1 in RA synovium. SDF-1 production increased spontaneously in cultured FLS over time, and its production was augmented when FLS were cocultured with CD4+ T cells. These findings indicate that there may be some synergistic interactions between FLS and T cells that affect the production of SDF-1 in RA synovium.
Certain molecules or pathways are thought to be involved in the reciprocal effects of these 2 types of RA cells, and in this study we presumed IL-17 was a potential candidate. For evaluation of the effect of IL-17 on the production of SDF-1 in cocultures of FLS with CD4+ T cells, we used anti–IL-17 monoclonal antibody. When FLS were cultured with CD4+ T cells in the presence of anti–IL-17 antibody, the production of SDF-1 was inhibited significantly. Similar results were obtained when cells were pretreated with anti-CD40L antibody. Moreover, when FLS were cultured with CD4+ T cells in the presence of recombinant human IL-17, the production of SDF-1 was enhanced in a dose-dependent manner.
On the basis of these findings, it can be proposed that CD4+ T cells may play an important role in the production of SDF-1 by FLS, through physical interaction with FLS and through the T cell–derived cytokine IL-17. To simulate the in vivo condition, RA SF was added to cultured RA FLS, and this also significantly up-regulated the expression of SDF-1 mRNA. This effect was mediated mainly by the IL-17 present in RA SF, as confirmed in experiments using anti–IL-17.
In early stages of inflammatory cell migration into inflamed synovium, FLS may initiate the migration of immune cells, but as the inflammatory response progresses, T cells may maintain and augment their migration through IL-17–induced FLS production of SDF-1. However, the role of IL-17 in SDF-1 induction should be confirmed in studies with genetically defined strains of mice and models of experimental synovitis in future investigations. The role of monocytes and B cells in the induction of SDF-1 and augmentation of their migration is also an attractive candidate for future experiments.
Although the SDF-1/CXCR4–mediated signaling pathways have been documented widely (41), there are no published studies of the mechanism of regulation of SDF-1 production by specific signaling pathways. Therefore, we used inhibitors of signal transduction molecules to evaluate which signaling pathway was primarily involved in the induction of SDF-1 in RA FLS. We demonstrated that IL-17–induced SDF-1 production in RA FLS was significantly hampered by inhibitors of NF-κB (PDTC and parthenolide), PI 3-kinase (LY294002 and wortmannin), and AP-1 (curcumin and SP600125).
The direct relationship between these molecules needs to be investigated in future studies. Our data demonstrated that PI 3-kinase/Akt could be the upstream arbitrator of IL-17–mediated up-regulation of SDF-1 in RA. We postulate that subsequent activation of NF-κB and AP-1 may also have contributed to the increased binding of inflammatory transcription factors to the promoter of SDF-1 in IL-17–stimulated synovial fibroblasts.
In summary, SDF-1 is overproduced in RA, and IL-17 could regulate the expression of SDF-1 in RA FLS via pathways mediated by PI 3-kinase, NF-κB, and AP-1. SDF-1 production in RA FLS was enhanced by coculture with CD4+ T cells through IL-17 itself and through CD40–CD40L interaction. These results suggest that inhibition of the interaction between IL-17 in T cells and SDF-1 in FLS may provide a new molecular target for future treatment of RA.
Dr. Lee 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 design. Lee, Ho-Youn Kim.
Acquisition of data. Kyoung-Woon Kim, Mi-Kyung Park, Oh.
Analysis and interpretation of data. Cho, Ju, Joon-Seok Kim.
Manuscript preparation. Hae-Rim Kim, Cho.
Statistical analysis. Sung-Hwan Park.
- 4CXCL12 chemokine up-regulates bone resorption and MMP-9 release by human osteoclasts: CXCL12 levels are increased in synovial and bone tissue of rheumatoid arthritis patients. J Cell Physiol 2004; 199: 244–51., , , , , .