T cells regulate the expression of matrix metalloproteinase in human osteoblasts via a dual mitogen-activated protein kinase mechanism.код для вставкиСкачать
ARTHRITIS & RHEUMATISM Vol. 48, No. 4, April 2003, pp 993–1001 DOI 10.1002/art.10872 © 2003, American College of Rheumatology T Cells Regulate the Expression of Matrix Metalloproteinase in Human Osteoblasts via a Dual Mitogen-Activated Protein Kinase Mechanism Leonard Rifas1 and Sophia Arackal2 Objective. To investigate the role of T cell induction of matrix metalloproteinase 13 (MMP-13) production by human osteoblasts in order to better understand the process of bone loss in rheumatoid arthritis (RA). Methods. Activated T cell–conditioned medium (ACTTCM) was used to mimic the physiologic conditions of inflammation. MMP-13 production by human osteoblasts was assessed using a specific enzyme-linked immunosorbent assay. Specific inhibitors of the p38 mitogen-activated protein (MAP) kinase and the extracellular signal–regulated kinase 1/2 (ERK-1/2) MAP kinase signaling pathways were used to assess their roles in T cell–mediated MMP-13 production. Finally, recombinant cytokines representative of the major components in ACTTCM were assessed for their ability to induce MMP-13. Results. ACTTCM powerfully induced MMP-13 in human osteoblasts. Inhibition of p38 activity abolished, while inhibition of ERK-1/2 activity enhanced, MMP-13 production. We next investigated physiologic levels of the T cell cytokines tumor necrosis factor ␣ (TNF ␣ ), transforming growth factor ␤ (TGF ␤ ), interferon-␥ (IFN␥), and interleukin-17 (IL-17) for their roles in MMP-13 induction. Although individual cytokines had no significant effect, the combination of TNF␣, TGF␤, IFN␥, and IL-17 resulted in a dramatic p38-dependent induction of MMP-13 identical to that produced by ACTTCM. Conclusion. These studies demonstrate for the first time that human osteoblasts produce MMP-13. The results also show that under conditions of chronic inflammation, multiple T cell cytokines synergize to induce high levels of MMP-13 via a mechanism that is dependent on activated p38 MAP kinase and is suppressed by activated ERK-1/2. Selective inhibition of p38 activity may offer a target for pharmacologic inhibition of bone loss in RA. Osteoporosis occurs frequently in patients with rheumatoid arthritis (RA) and is observed in 2 characteristic patterns: juxtaarticular bone loss, occurring around inflamed joints, and generalized bone loss with reduced bone mass (for review, see refs. 1–3). It has been proposed that T cells are continuously involved in the pathogenesis of RA, from its initiation phase through to the chronic stage (4). Recent studies have defined a role for T cells in the severe joint inflammation and bone and cartilage destruction that are the hallmarks of RA (5,6). Elevated T cell populations have been identified in the synovial cavity of patients with RA but not in those with osteoarthritis (7–9), and elevated levels of cytokines such as tumor necrosis factor ␣ (TNF␣) (10), transforming growth factor ␤ (TGF␤) (11), interferon-␥ (IFN␥) (10), and interleukin-17 (IL-17) (12) have been described as well. Furthermore, the combination of TNF␣ blockade with IL-1 and IL-17 blockade is more effective than is TNF␣ blockade alone for controlling synovial inflammation and cartilage degradation in RA (6), suggesting that disease progression is regulated by the synergistic action of multiple cytokines. However, the process by which T cells induce bone matrix degradation has yet to be identified. Bone resorption is complex, requiring not only the recruitment of osteoclasts to the site of resorption, but also the removal of collagen from the bone surface, through the action of matrix metalloproteinase 13 Supported by NIH grants AR-32087 and AR-46370. 1 Leonard Rifas, MS: Washington University School of Medicine, St. Louis, Missouri; 2Sophia Arackal, BS: Barnes-Jewish Hospital, St. Louis, Missouri. Address correspondence and reprint requests to Leonard Rifas, MS, Washington University School of Medicine, Barnes-Jewish Hospital North, 216 South Kings Highway, St. Louis, MO 63110. E-mail: firstname.lastname@example.org. Submitted for publication June 18, 2002; accepted in revised form December 19, 2002. 993 994 (MMP-13; collagenase 3), to allow for osteoclast attachment (13,14). MMP-13 is secreted by cytokinestimulated cartilage cells (15) and cleaves fibrillar collagens (with a preference for type II collagen over type I collagen and type III collagen), and displays ⬎40-fold stronger gelatinase activity compared with MMP-1 (interstitial collagenase) and MMP-8 (neutrophil collagenase) (16). Human MMP-13 is present in the pannus of rheumatoid synovium, suggesting a role in both bone remodeling and inflammation-mediated bone erosion (17). MMP-13 expression is under the control of mitogen-activated protein (MAP) kinases (18), serinethreonine kinases that are activated by phosphorylation of a Thr-Xxx-Tyr motif in the activation loop of the molecule by dual-specificity MAP kinase kinases (MEKs). MEKs are themselves activated by MAP kinase kinase kinases (MEKKs) (19,20). At present, the 3 major MAP kinase cascades that are known are as follows: the extracellular signal– related kinase 1/2 (ERK-1/2; also known as p44/42 MAP kinase), the c-Jun N-terminal kinase (JNK) or stressactivated protein kinases (SAPKs) 1 and 2, and the p38 group of protein kinases. The ERK-1/2 pathway (Raf13 MEK-1/23 ERK-1/2) is strongly activated by growth factor receptor occupancy but is indirectly activated by cytokines and environmental stress through secondary protein kinases (e.g., Src family tyrosine kinases). In contrast, the main stimuli for the JNK (MEKK-1– 33 MEK-4/SAPK/ERK kinase 1, MEK-73 JNK/SAPK) and p38 (MEKK3 MEK-3/63p38) pathways are cytokines and environmental stress, whereas mitogenic growth factors have no or little effect (21). In order to examine the role of T cell cytokines in the process of inflammation-induced bone degradation (such as occurs in inflammatory diseases such as RA), we used activated T cell–conditioned medium (ACTTCM) to examine, in vitro, the interactions and intracellular signaling pathways of the cytokine repertoire produced by activated T cells that leads to MMP-13 production. Furthermore, because MAP kinase pathways have been shown to regulate the transcription factors necessary for MMP-13 gene regulation, we also studied these signaling events to determine their contribution to the production of MMP-13 in normal human osteoblasts. We show that during chronic inflammation, activated T cells produce 4 major cytokines, TNF␣, TGF␤, IFN␥, and IL-17, which alone do not affect MMP-13, but in combination synergistically and potently stimulate production of MMP-13 in normal human osteoblasts. We also demonstrate that MMP-13 expression is regulated by the p38 MAP kinase pathway. RIFAS AND ARACKAL MATERIALS AND METHODS Antibodies. Mouse anti-human CD3 monoclonal antibody (clone HIT3a) and mouse anti-human CD28 monoclonal antibody (clone 28.2) were obtained from PharMingen (San Diego, CA), as sterile, azide-free, and low-endotoxin preparations. Recombinant cytokines. Recombinant human TNF␣, TGF␤1, IFN␥, and IL-17 were obtained from R&D Systems (Minneapolis, MN). MAP kinase inhibitors. The specific MEK-1/2 inhibitor, PD98059, and the specific p38 MAP kinase inhibitor, SB203580, were purchased from Calbiochem (San Diego, CA) and were prepared as stock solutions in DMSO (SigmaAldrich, St. Louis, MO) at 40 mM and 20 mM, respectively. Isolation of T cells. Peripheral blood mononuclear cells (PBMCs) were obtained in the form of buffy coats from the American Red Cross (St. Louis, MO) and were further purified by separation on Histopaque-1077 (1.077 gm/ml) lymphocyte separation medium (Sigma-Aldrich), as previously described, with some modification (22). Briefly, the buffy coat preparations were diluted 1:1 in phosphate buffered saline (PBS). Twenty milliliters of diluted buffy coat were overlayered onto 15 ml of Histopaque and centrifuged at 400g for 30 minutes at room temperature. Mononuclear cells were recovered from the interface, washed twice with PBS (Ca⫹2-, Mg⫹2-free; Sigma-Aldrich) by centrifugation at 300g for 5 minutes, and then T cells were isolated by positive selection from the PBMCs using CD4⫹ Dynabeads (Dynal, Lake Success, NY ) according to the manufacturer’s instructions. This process results in a population of T cells enriched ⬎95%. T cell cultures. T cells were cultured at 1 ⫻ 106 cells/ml in AIM-V serum-free medium (Gibco, Grand Island, NY). T cells were activated by the addition of mouse anti-human CD3 monoclonal antibody (1 g/ml) and mouse anti-human CD28 monoclonal antibody (5 g/ml). After a 72-hour incubation period, the ACTTCM was harvested and frozen at ⫺80°C until use in the experimental protocols. Preparation of human osteoblast cultures. Human rib specimens from 10 different donors were obtained from the Missouri Transplantation Services (St. Louis, MO) as donor tissue. Use of these tissues (and buffy coats) was approved by the Human Studies Committee, Washington University Medical Center. Human osteoblast cultures were prepared from rib specimens, as previously described (22). The cells have the characteristics of osteoblasts, as previously described (22–24). Induction of MMP-13 in human osteoblasts. Cells (2 ⫻ 104 cells/well) were seeded into 48-well tissue culture plates in ␣-minimum essential medium (␣-MEM; Mediatech, Herndon, VA) containing 10% heat-inactivated fetal bovine serum (HIFBS), and were incubated for 4 days. The confluent cells were washed with PBS and then were incubated for 24 hours in either ␣-MEM containing 0.2% HIFBS or AIM-V medium. Cytokines or ACTTCM was added, and the conditioned media were collected after the appropriate incubation times, as noted in the figure legends. To examine the regulation of MMP-13 by MAP kinases, DMSO (vehicle control), PD98059, or SB203580 was added to the cultures, at the specified concentrations, 1 hour before addition of test agents. The DMSO concentration was 0.1% in all cases. Media were collected after the indicated times and were frozen at ⫺80°C until assayed. T CELLS AND HUMAN OSTEOBLAST MMP-13 PRODUCTION Protein assays. After collection of the conditioned media, the cell layers were washed 3 times with Tris buffered saline (50 mM Tris HCl, pH 7.4, 150 mM NaCl), then solubilized in 0.1% sodium dodecyl sulfate (SDS). Protein assays were performed using a DC Protein Assay Kit (Bio-Rad Laboratories, Hercules, CA). Enzyme-linked immunosorbent assays (ELISAs). Human IL-1␤, TNF␣, and MMP-13 were assayed using specific ELISA kits obtained from Amersham Biosciences (Piscataway, NJ). TGF␤ was assayed using an ELISA kit from Promega (Madison, WI). IFN␥ was assayed using an ELISA kit obtained from R&D Systems. Whole cell lysates and Western blotting. Human osteoblasts (1 ⫻ 106 cells) were subcultured into 100-mm tissue culture plates in ␣-MEM containing 10% HIFBS and incubated for 4 days. The cell layers were then rinsed with PBS and incubated for 24 hours in ␣-MEM containing 0.2% HIFBS. After a further 24 hours of incubation, the medium was changed, and the cells were preincubated with either DMSO or SB203580 (20 M) for 1 hour before adding either medium alone or 25% ACTTCM. After 24 hours of incubation, the cell layers were rapidly rinsed with PBS, extracted with boiling lysis buffer (1% SDS, 1.0 mM sodium orthovanadate, 10 mM Tris HCl, pH 7.4), and then the cell layers were scraped. The lysed cells were then passed several times through a 22-gauge syringe needle, boiled for 5 minutes, followed by centrifugation at 15,000g in order to remove insoluble material. Aliquots of whole cell extracts containing 30 g of protein were diluted 3:1 (volume/volume) with 4⫻ SDS–polyacrylamide gel electrophoresis (SDSPAGE) sample buffer, boiled again for 5 minutes, then separated by 10% SDS-PAGE and blotted onto polyvinylidene difluoride membranes (Immunoblot; Bio-Rad), using a semidry transfer method. After transfer, the membranes were dried overnight, rewetted in methanol, washed 2 times in Tris buffered saline– Tween (TBST; 50 mM Tris HCl, 150 mM NaCl, 0.1% Tween 20, pH 7.4) then probed with the specified primary antibodies at 1:1,000 dilution in Superblock (Pierce, Rockford, IL) containing 0.1% Tween 20 (SBT). After 3 washes with TBST, the membranes were incubated with goat anti-rabbit IgG conjugated with horseradish peroxidase (1:50,000) in SBT for 1 hour. Proteins were visualized using SuperSignal chemiluminescence substrate (Pierce) and exposure to Hyperfilm MP (Amersham Biosciences). The blots were stripped with Restore Western blotting stripper (Pierce), according to the manufacturer’s protocol, and were successively reprobed using different antibodies. Whole cell lysate preparation for kinase assays. Human osteoblasts (4 ⫻ 105 cells) were subcultured into 60-mm tissue culture plates in ␣-MEM containing 10% HIFBS and incubated for 4 days. The medium was changed to ␣-MEM containing 0.2% HIFBS and incubated for 24 hours. The medium was again changed, and the cells were preincubated with either vehicle control or SB203580 (20 M) for 1 hour before adding either medium alone or 25% ACTTCM. After 24 hours of incubation, the cell layers were rapidly rinsed with PBS, then the cells were lysed in a cold lysis buffer (10 mM Tris HCl, pH 7.4, 1.0% Triton X-100, 0.5% Nonidet P40, 150 mM NaCl, 20 mM sodium fluoride, 0.2 mM sodium orthovanadate, 1.0 mM EDTA, 1.0 mM EGTA, 0.2 mM phenylmethylsulfonyl fluoride, 4 mM 4-(2-aminoethyl)benzenesulfonylfluoride, 3.2 995 M aprotinin, 84 M leupeptin, 0.14 mM bestatin, 60 M pepstatin, 56 M E-64) in the culture dish for 30 minutes at 4°C. The lysed cells were then scraped off the dish and the lysate passed several times through a 26-gauge needle to disperse any large aggregates. The lysate was centrifuged at 16,000g for 30 minutes at 4°C. The supernatant (total cell lysate) was collected for kinase assays. Aliquots were assayed for protein as described above. Immunoprecipitation for kinase assays. Cell lysates (200 l containing 200 g total protein) were added to 20 l immobilized phospho–p38 MAP kinase (Thr180/Tyr182) monoclonal antibody. Each mixture was incubated with gentle rocking overnight at 4°C. The samples were then microcentrifuged for 30 seconds, and the pellet was washed twice with 500 l of ice cold 1⫻ lysis buffer (20 mM Tris, pH 7.5, 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1% Triton X-100, 2.5 mM sodium pyrophosphate, 1 mM ␤-glycerolphosphate, 1 mM sodium orthovanadate, 2.1 M leupeptin) and kept on ice. Finally, the pellet was washed twice with 500 l of ice cold 1⫻ kinase buffer (25 mM Tris, pH 7.5, 5 mM ␤-glycerolphosphate, 2 mM dithiothreitol, 0.1 mM sodium orthovanadate, 10 mM MgCl2) at 4°C. Kinase assays. For the p38 MAP kinase assay, the pellet was suspended in 50 l 1⫻ kinase buffer supplemented with 200 M adenosine triphosphate and 2 g activating transcription factor 2 (ATF-2) fusion protein and incubated for 30 minutes at 30°C. The reaction was terminated with 25 l of 3⫻ SDS-PAGE sample buffer, then samples were boiled for 5 minutes, vortexed, and microcentrifuged for 2 minutes. Thirty microliters of each sample was subjected to SDS-PAGE and analyzed by Western blotting using phospho–ATF-2 antibody, according to the manufacturer’s instructions. Statistical analysis. Group mean values were compared by analysis of variance (ANOVA). Subsequent multiple comparisons were performed using Fisher’s protected least significance difference test. The 5% significance level was used. RESULTS Human osteoblasts produce MMP-13 in response to T cell–derived inflammatory cytokines. T cell–derived cytokines are known to participate in the destruction of articular cartilage by inducing MMP-13 in chondrocytes (15,25,26). We observed that although MMP-13 is not produced constitutively in human osteoblasts, upon their exposure to ACTTCM for a 72-hour period, abundant MMP-13 was produced in a dosedependent manner (Figure 1). Because MMP-13 induction by 25% ACTTCM was not significantly different from that by 50% ACTTCM, 25% ACTTCM was chosen for the experiments described below. A time course analysis of MMP-13 production after exposure to ACTTCM (Figure 2) revealed that MMP-13 was not expressed in significant quantities until 24 hours (mean ⫾ SEM 16.83 ⫾ 0.81 ng/mg protein), but there was a dramatic, 44-fold linear increase in MMP-13 production at 48 hours (mean ⫾ SEM 131.24 ⫾ 6.23 ng/mg protein), 996 Figure 1. Dose-dependent induction of matrix metalloproteinase 13 (MMP-13) in human osteoblasts by activated T cell–conditioned medium (ACTTCM). Human osteoblasts were cultured in ␣-minimum essential medium (␣-MEM) containing 10% heat-inactivated fetal bovine serum (HIFBS) until confluent (5 days). The medium was changed to ␣-MEM containing 0.2% HIFBS and incubated for 24 hours. The medium was changed again, the ACTTCM was added at the indicated concentrations, and cells were incubated for 72 hours. Medium was collected and assayed for MMP-13 by a specific enzymelinked immunosorbent assay. Proteins were assayed using the Bio-Rad DC protein assay kit. Values are the mean and SEM of 3 independent cultures. P ⬍ 0.01 for all comparisons, by analysis of variance. and a 71-fold increase at 72 hours (mean ⫾ SEM 212.97 ⫾ 21.65 ng/mg protein), compared with the 8-hour time point. MMP-13 expression in human osteoblasts is regulated via p38 MAP kinase and ERK-1/2 MAP kinase. To assess the role of MAP kinases in the regulation of MMP-13, human osteoblasts were exposed to ACTTCM in the presence or absence of increasing doses of the specific ERK-1/2 pathway inhibitor PD98059 (0–40 M) and the specific p38 MAP kinase inhibitor SB203580 (0–20 M). Increasing concentrations of PD98059 resulted in production of an increased amount of MMP-13 (Figure 3A), demonstrating that the ERK pathway inhibited expression of the enzyme. In contrast, SB203580 (Figure 3B) inhibited MMP-13 secretion ⬎50% at 0.1 M, the lowest dose tested, and almost completely inhibited MMP-13 secretion at 20 M, demonstrating that activated p38 MAP kinase is a potent stimulator of MMP-13 production in human osteoblasts. To confirm that p38 MAP kinase is critical in MMP-13 production, human osteoblasts were treated with ACTTCM in the presence or absence of 20 M SB203580 over a time course of 0–96 hours, and RIFAS AND ARACKAL MMP-13 was assayed by a specific ELISA. During the 0–96-hour time course, ACTTCM increased MMP-13 protein production 51-fold, while only a 2.6-fold increase was observed in the presence of SB203580 (ACTTCM versus ACTTCM plus SB203580: at 8 hours, mean ⫾ SEM 9.76 ⫾ 0.52 versus 9.39 ⫾ 0.33 ng/mg protein; at 96 hours, mean ⫾ SEM 493.96 ⫾ 75.66 versus 24.63 ⫾ 1.00 ng/mg protein; P ⬍ 0.001 by ANOVA [n ⫽ 4 cultures]). These results demonstrate that p38 MAP kinase is essential in the process of MMP-13 expression. Activation of p38 depends on phosphorylation. Therefore, we investigated whether ACTTCM increases phosphorylation of p38, and whether SB203580 inhibits its kinase activity. To do so, we examined the phosphorylation state of p38 MAP kinase in the cells using Western blot analysis, and the kinase activity of activated p38 directly using ATF-2 as substrate. Our results (Figure 4) show that after 24 hours, ACTTCM increased p38 phosphorylation as determined by Western blot analysis. Surprisingly, SB203580, in the presence of ACTTCM, also enhanced p38 phosphorylation. To determine whether SB203580 inhibited the activity of ACTTCM-induced phospho-p38, we analyzed the kinase activity of p38 directly using ATF-2 as substrate. The results (Figure 4) show that ACTTCM induced p38 Figure 2. Effects of ACTTCM on MMP-13 production in human osteoblasts over time. Human osteoblasts were cultured as described in Figure 1. The medium was changed again, and the cells were stimulated with 25% ACTTCM. Medium was collected at 8, 16, 24, 48, and 72 hours and assayed for MMP-13 by a specific enzyme-linked immunosorbent assay. Proteins in the cell layers were assayed as described in Figure 1. Values are the mean ⫾ SEM of 3 independent cultures. ⴱ ⫽ P ⬍ 0.001 versus 8 hours, by analysis of variance. See Figure 1 for definitions. T CELLS AND HUMAN OSTEOBLAST MMP-13 PRODUCTION 997 kinase activity, as demonstrated by its ability to phosphorylate its substrate, ATF-2. However, the presence of SB203580 completely inhibited p38 kinase activity (Figure 4). Figure 4. Western blot analysis of the effect of SB203580 on the activation (phosphorylation) and kinase activity of p38 mitogenactivated protein (MAP) kinase in human osteoblasts treated with ACTTCM. Human osteoblasts were grown as described in Figure 1. Human primary osteoblasts were incubated for 1 hour with the vehicle (DMSO) or SB203580 (20 M) before stimulation with ACTTCM for 24 hours. p38 or its phosphorylated (activated) isoform (phospho-p38) was identified by immunoblotting, using specific polyclonal antibodies. To determine the activity of p38, cellular extracts were analyzed for their ability to phosphorylate the p38-specific substrate, activating transcription factor 2 (ATF-2). Phosphorylated ATF-2 (p-ATF-2) was identified by immunoblotting using specific polyclonal antibodies as described in Materials and Methods. See Figure 1 for other definitions. Figure 3. Regulation of MMP-13 expression in human osteoblasts by PD98059 and SB203580. Human osteoblasts were cultured until confluent (5 days) in ␣-MEM containing 10% HIFBS. The medium was changed to ␣-MEM containing 0.2% HIFBS and incubated for 24 hours. After the 24-hour incubation period, cells were rinsed with phosphate buffered saline, then ␣-MEM containing 0.2% HIFBS was added to control wells or medium containing PD98059 (0–40 M) or SB203580 (0–40 M). Cells were incubated for 1 hour with the inhibitors, then stimulated with 25% ACTTCM for 48 hours. Media from osteoblasts treated with either PD98059 (A) or SB203580 (B) were assayed for MMP-13 by enzyme-linked immunosorbent assay. Protein concentrations were determined as described in Figure 1. PD98059 up-regulated and SD203580 down-regulated expression of MMP-13. Values are the mean ⫾ SEM of 3 independent cultures. ⴱ ⫽ P ⬍ 0.01 as determined by analysis of variance. See Figure 1 for definitions. T cell–derived cytokines synergistically induce MMP-13 in human osteoblasts. Activated T cells produce numerous inflammatory cytokines, including TNF␣ (22), TGF␤ (27), IFN␥ (27,28), and IL-17 (28). The mean ⫾ SEM amount of each cytokine in 25% ACTTCM (as used in our experiments) was as follows: for TNF␣ 116.61 ⫾ 2.1 pg/ml, for TGF␤ 285.5 ⫾ 3.0 pg/ml, and for IFN␥ 5.88 ⫾ 0.6 ng/ml. Fossiez et al reported that under the same activating conditions as those used in our studies, IL-17 was secreted at levels of ⬃8 ng/ml (29). Therefore, 25% ACTTCM would contain ⬃2 ng/ml. Thus, we examined these factors individually and in combinations. Figure 5 shows that when the cytokines were tested either individually or in any combination of two, no significant induction of MMP-13 secretion was observed. However, when TNF␣, TGF␤, and IFN␥ were added in combination, a significant synergistic (⬃6-fold) induction of MMP-13 was observed. Furthermore, addition of IL-17 increased the synergistic activity to ⬃100fold over control levels, and the level of MMP-13 induction by this combination of 4 cytokines was identical to that achieved with induction by ACTTCM. When 998 Figure 5. Effect of T cell cytokines and SB203580 on matrix metalloproteinase 13 (MMP-13) production in human osteoblasts. Human osteoblasts were cultured as described in Figure 3, then incubated with 25% activated T cell–conditioned medium (ACTTCM) or recombinant human tumor necrosis factor ␣ (TNF␣) (120 pg/ml), transforming growth factor ␤ (TGF␤) (300 pg/ml), interleukin-17 (IL-17) (2 ng/ml), or interferon-␥ (IFN␥) (10 ng/ml), alone or in combinations as shown, for 48 hours at 37°C. After the incubation period, the conditioned media were collected and assayed for MMP-13 by enzyme-linked immunosorbent assay. Values are the mean and SEM of 3 independent cultures. a ⴝ P ⬍ 0.05 versus control; b ⫽ P ⬍ 0.05 versus cytokines tested in combinations of two; c ⫽ P ⬍ 0.05 versus cytokines tested in combinations of three; d ⫽ not significantly different; e ⫽ P ⬍ 0.05 versus all 4 cytokines combined, as determined by analysis of variance. the combination of all 4 cytokines was added to cells pretreated with SB203580, MMP-13 production was almost completely inhibited (Figure 5), demonstrating that these 4 cytokines are responsible for the MMP-13– inducing activity present in ACTTCM. DISCUSSION Under normal physiologic conditions, bone homeostasis is governed by the balance between the process of resorption of bone by osteoclasts and the formation of bone by osteoblasts. These 2 processes are coupled in space and time such that there is no net loss of bone. However, under conditions of chronic inflammation, such as occurs in RA, there is an unbridled release of cytokines from T cells, which activates osteoclastogenesis (30) and induces osteoblastic cytokine production (22,24). The result is an uncoupling of resorption and formation, a process that favors bone resorption and ultimately net bone loss. The major cytokines that have been identified as inducing increased osteoclast formation and activity are RIFAS AND ARACKAL IL-1, TNF␣, and receptor activator of nuclear factor B ligand (30). However, in order for osteoclasts to degrade bone, collagen covering the surface must first be degraded so that osteoclasts can attach. The primary enzyme that is responsible for cleaving collagen is collagenase. MMP-13 can degrade type I, type II, and type III collagen (with a preference for type II collagen) and also acts as a gelatinase to further degrade the initial cleavage products of collagenolysis to small peptides (17,31) that can interact with osteoclasts, resulting in their activation (13). Recently, MMP-13 has been detected in human osteoblasts, but only during fetal development (32). However, we did not detect MMP-13 in adult human osteoblasts in the absence of cytokine stimulation. Now, we report that in the adult osteoblast, inflammatory cytokines are major inducers of MMP-13. These results point to a role for the T cell in the osteopenia associated with RA, and perhaps other inflammatory autoimmune diseases as well, by inducing osteoblast MMP-13 locally. The result of this high secretion of MMP-13 would be the release of collagen peptides and denuding of the bone surface to allow activated osteoclasts to attach and resorb the calcified matrix. The activated T cell has only recently been identified as playing a major role in the pathophysiology of RA (4,33,34), with destruction of articular cartilage representing the major thrust of investigation (35). We now report that T cells produce a repertoire of cytokines that potently stimulate MMP-13 in osteoblasts as well. Our data demonstrate for the first time that activated T cells in the local milieu may play a major role in bone destruction in RA due to the release of multiple cytokines that, by themselves, have little effect, but as a result of synergistic activity have a profound effect. We demonstrate that 4 major cytokines present in ACTTCM—TNF␣, TGF␤, IFN␥, and IL-17—are responsible for modulating MMP-13 production in human osteoblasts. Reported measurements of these 4 cytokines in synovial fluid from RA patients include the following: for TNF␣ 157 pg/ml and for IFN␥ 17 pg/ml (10), for TGF␤ 2–20 ng/ml (11), and for IL-17 12–5,000 pg/ml (12,36). These data support our observation that only small amounts of each of these cytokines need be present to have a powerful effect. Individually, many of these cytokines have been examined for their role in the regulation of MMP-13 in osteoblasts or osteoarthritic or rheumatoid cartilage (37–40). Of interest is that in our studies, IFN␥ played a major role in the synergistic induction of MMP-13. This finding was surprising, because IFN␥ has been reported to be a potent inhibitor of T CELLS AND HUMAN OSTEOBLAST MMP-13 PRODUCTION TGF␤-induced MMP-13 (41). IL-17 has been shown to participate in the bone erosion associated with RA (42–44), primarily as a function of cytokine induction via interaction with TNF␣ (43,45). However, we now show for the first time that IL-17 may play yet a further role in bone resorption in RA by enhancing T cell cytokine induction of MMP-13 production in osteoblasts. This finding may open new avenues for the treatment of bone loss in inflammatory diseases such as RA. Our results show that stimulation of human osteoblasts with ACTTCM or inflammatory cytokines results in the coordinate activation of at least 2 separate MAP kinase pathways: the ERK-1/2 pathway and the p38 pathway. Of interest is that stimulation of MMP-13 by ACTTCM was potently up-regulated by the ERKspecific inhibitor, PD98059, while the p38-specific inhibitor, SB203580, blocked production of MMP-13 over a 96-hour period. We used inhibitor concentrations that are specific, as described by other investigators who have analyzed the roles of both the p38 MAP kinase pathway (46–48) and the ERK MAP kinase pathway (49–51) in osteoblasts and other cell types. However, SB203580 has been reported to inhibit certain isoforms of JNK at levels higher than the reported specificity of this inhibitor for p38 (⬃1–3 M) (52). We did observe that concentrations of SB203580 as low as 0.1–1.0 M resulted in a 50–70% inhibition of MMP-13 production, which is well within the specificity of the inhibitor for p38. Furthermore, ACTTCM did not induce phosphoJNK over a time course of 0–48 hours (data not shown). Of interest was the fact that SB203580 inhibited production of MMP-13 by ACTTCM despite the fact that the phosphorylated form of p38 was enhanced. These observations are in agreement with those of other investigators who have shown that SB203580 induces enhanced phosphorylation of p38 in the presence of specific stimulators (53,54). Our data confirm these observations, because we observed phospho-p38 to be increased, rather than decreased, in the presence of ACTTCM and SB203580, demonstrating that this compound can inhibit the activity, but not the phosphorylation, of p38. We have substantiated this by the finding that ACTTCM induces p38 activity, as demonstrated by phosphorylation of its substrate ATF-2, while SB203580 abolished this activity. The importance of the MAP kinase pathway in regulating cytokine-induced MMP-13 has been demonstrated in multiple cell types, and the particular pathway that regulates MMP-13 expression is cell specific. For example, TNF␣ induction of MMP-13 production in transformed keratinocytes requires p38 MAP kinase but is not inhibited by ERK-1/2 (55). In human dermal 999 Figure 6. Signaling pathways mediating regulation of matrix metalloproteinase 13 (MMP-13) expression in human osteoblasts by activated T cell cytokines. Binding of cytokines to their respective receptors results in the synergistic activation of the extracellular signal–regulated kinase 1/2 (ERK-1/2) and p38 mitogen-activated protein (MAP) kinase pathways. Because cytokines are weak inducers of the ERK pathway, although strong activators of the p38 pathway, an imbalance of p38 over ERK-1/2 favors the production of MMP-13 gene activation, most likely through activation of activator protein 1 (AP-1) and Cbfa-1. Inhibition of ERK-1/2 by PD98059 further favors the production of MMP-13, while inhibition of p38 by SB203580 leads to loss of MMP-13 production. SB203580 is a specific inhibitor of p38 MAP kinase; PD98059 is a specific inhibitor of MAP kinase kinase 1 (MEK-1) and MEK-2. TGF␤ ⫽ transforming growth factor ␤; TNF␣ ⫽ tumor necrosis factor ␣; MEKK ⫽ MAP kinase kinase kinase. fibroblasts, induction of MMP-13 by collagen, through the integrins ␣1␤1 and ␣1␤2 is induced by p38 and inhibited by ERK-1/2, demonstrating that integrin signaling regulates MMP-13 via a dual MAP kinase mechanism (49). However, recent evidence points to diverging roles of MAP kinases with respect to specific MMP activation. In articular chondrocytes, IL-1 induction of MMP-13 requires p38 activity, JNK activity, and nuclear factor B (NF-B) translocation, while in chondrosarcoma cells MMP-1 induction by IL-1 is dependent on p38 and MEK (ERK pathway) but does not require JNK (56). Although the requirement for NF-B in MMP-13 expression is essential in rabbit chondrocytes or human 1000 RIFAS AND ARACKAL chondrosarcoma cells (56,57), its role in human osteoblast MMP-13 expression is yet to be defined. However, our data clearly show that in human osteoblasts, p38 MAP kinase is both necessary and sufficient to regulate MMP-13. However, further studies must be undertaken in order to determine whether other signaling pathways may play a regulatory role. Based on our collective data, we propose a model (Figure 6) in which the release of cytokines from activated T cells results in binding to their respective receptors, leading to activation of the ERK-1/2 and p38 MAP kinase pathways. Because cytokines are weak inducers of the ERK pathway while being strong activators of the p38 pathway, an imbalance of p38 over ERK-1/2 favors the production of MMP-13 gene activation, most likely through activation of activator protein 1 and Runx2/Cbfa-1 (58,59). In conclusion, our data support the role of T cells in the process of bone resorption observed in patients with RA. The net MMP-13 production from stimulated osteoblasts appears to be the result of the balance between the two MAP kinase pathways, suggesting that selective inhibition of active p38 may prove to be a pivotal point for the pharmacologic inhibition of bone loss in inflammatory disease such as RA. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. ACKNOWLEDGMENTS We wish to thank Ms Aurora Fausto and Ms Linda Halstead for preparation of the human osteoblast. The authors also wish to thank Drs. M. Neale Weitzmann and Yousef Abu-Amer for their critical reading of this manuscript. REFERENCES 1. Deodhar AA, Woolf AD. Bone mass measurement and bone metabolism in rheumatoid arthritis: a review. Br J Rheumatol 1996;35:309–22. 2. Dequeker J, Maenaut K, Verwilghen J, Westhovens R. Osteoporosis in rheumatoid arthritis. Clin Exp Rheumatol 1995;13 Suppl 12:S21–6. 3. Gough AK, Peel NF, Eastell R, Holder RL, Lilley J, Emery P. Excretion of pyridinium crosslinks correlates with disease activity and appendicular bone loss in early rheumatoid arthritis. Ann Rheum Dis 1994;53:14–7. 4. Panayi GS. T-cell-dependent pathways in rheumatoid arthritis. Curr Opin Rheumatol 1997;9:236–40. 5. Kong YY, Feige U, Sarosi I, Bolon B, Tafuri A, Morony S, et al. Activated T cells regulate bone loss and joint destruction in adjuvant arthritis through osteoprotegerin ligand. Nature 1999; 402:304–9. 6. Chabaud M, Miossec P. The combination of tumor necrosis factor ␣ blockade with interleukin-1 and interleukin-17 blockade is more effective for controlling synovial inflammation and bone resorption in an ex vivo model. Arthritis Rheum 2001;44:1293–303. 7. Berner B, Akca D, Jung T, Muller GA, Reuss-Borst MA. Analysis of Th1 and Th2 cytokines expressing CD4⫹ and CD8⫹ T cells in 18. 19. 20. 21. 22. 23. 24. 25. rheumatoid arthritis by flow cytometry. J Rheumatol 2000;27: 1128–35. Isomaki P, Luukkainen R, Lassila O, Toivanen P, Punnonen J. Synovial fluid T cells from patients with rheumatoid arthritis are refractory to the T helper type 2 differentiation-inducing effects of interleukin-4. Immunology 1999;96:358–64. Kusaba M, Honda J, Fukuda T, Oizumi K. Analysis of type 1 and type 2 T cells in synovial fluid and peripheral blood of patients with rheumatoid arthritis. J Rheumatol 1998;25:1466–71. Steiner G, Tohidast-Akrad M, Witzmann G, Vesely M, StudnickaBenke A, Gal A, et al. Cytokine production by synovial T cells in rheumatoid arthritis. Rheumatology (Oxford) 1999;38:202–13. Buckley CD, Amft N, Bradfield PF, Pilling D, Ross E, ArenzanaSeisdedos F, et al. Persistent induction of the chemokine receptor CXCR4 by TGF-␤ 1 on synovial T cells contributes to their accumulation within the rheumatoid synovium. J Immunol 2000; 165:3423–9. Ziolkowska M, Koc A, Luszczykiewicz G, Ksiezopolska-Pietrzak K, Klimczak E, Chwalinska-Sadowska H, et al. High levels of IL-17 in rheumatoid arthritis patients: IL-15 triggers in vitro IL-17 production via cyclosporin A-sensitive mechanism. J Immunol 2000;164:2832–8. Holliday LS, Welgus HG, Fliszar CJ, Veith GM, Jeffrey JJ, Gluck SL. Initiation of osteoclast bone resorption by interstitial collagenase. J Biol Chem 1997;272:22053–8. Zhao W, Byrne MH, Boyce BF, Krane SM. Bone resorption induced by parathyroid hormone is strikingly diminished in collagenase-resistant mutant mice. J Clin Invest 1999;103:517–24. Reboul P, Pelletier JP, Tardif G, Cloutier JM, Martel-Pelletier J. The new collagenase, collagenase-3, is expressed and synthesized by human chondrocytes but not by synoviocytes: a role in osteoarthritis. J Clin Invest 1996;97:2011–9. Ravanti L, Kahari V. Matrix metalloproteinases in wound repair. Int J Mol Med 2000;6:391–407. Konttinen YT, Salo T, Hanemaaijer R, Valleala H, Sorsa T, Sutinen M, et al. Collagenase-3 (MMP-13) and its activators in rheumatoid arthritis: localization in the pannus-hard tissue junction and inhibition by alendronate. Matrix Biol 1999;18:401–12. Mengshol JA, Vincenti MP, Coon CI, Barchowsky A, Brinckerhoff CE. Interleukin-1 induction of collagenase 3 (matrix metalloproteinase 13) gene expression in chondrocytes requires p38, c-Jun N-terminal kinase, and nuclear factor B: differential regulation of collagenase 1 and collagenase 3. Arthritis Rheum 2000;43:801–11. Ono K, Han J. The p38 signal transduction pathway: activation and function. Cell Signal 2000;12:1–13. Zhou G, Bao ZQ, Dixon JE. Components of a new human protein kinase signal transduction pathway. J Biol Chem 1995;270: 12665–9. Widmann C, Gibson S, Jarpe MB, Johnson GL. Mitogen-activated protein kinase: conservation of a three-kinase module from yeast to human. Physiol Rev 1999;79:143–80. Rifas L, Avioli LV. A novel T cell cytokine stimulates interleukin-6 in human osteoblastic cells. J Bone Miner Res 1999;14:1096–103. Rifas L, Fausto A, Scott MJ, Avioli LV, Welgus HG. Expression of metalloproteinases and tissue inhibitors of metalloproteinases in human osteoblast-like cells: differentiation is associated with repression of metalloproteinase biosynthesis. Endocrinology 1994; 134:213–21. Rifas L, Kenney JS, Marcelli M, Pacifici R, Cheng SL, Dawson LL, et al. Production of interleukin-6 in human osteoblasts and human bone marrow stromal cells: evidence that induction by interleukin-1 and tumor necrosis factor-alpha is not regulated by ovarian steroids. Endocrinology 1995;136:4056–67. Borden P, Solymar D, Sucharczuk A, Lindman B, Cannon P, Heller RA. Cytokine control of interstitial collagenase and collagenase-3 gene expression in human chondrocytes. J Biol Chem 1996;271:23577–81. T CELLS AND HUMAN OSTEOBLAST MMP-13 PRODUCTION 26. Tardif G, Pelletier JP, Dupuis M, Geng C, Cloutier JM, MartelPelletier J. Collagenase 3 production by human osteoarthritic chondrocytes in response to growth factors and cytokines is a function of the physiologic state of the cells. Arthritis Rheum 1999;42:1147–58. 27. Abbas AK, Murphy KM, Sher A. Functional diversity of helper T lymphocytes. Nature 1996;383:787–93. 28. Yao Z, Painter SL, Fanslow WC, Ulrich D, Macduff BM, Spriggs MK, et al. Human IL-17: a novel cytokine derived from T cells. J Immunol 1995;155:5483–6. 29. Fossiez F, Djossou O, Chomarat P, Flores-Romo L, Ait-Yahia S, Maat C, et al. T cell interleukin-17 induces stromal cells to produce proinflammatory and hematopoietic cytokines. J Exp Med 1996;183:2593–603. 30. Rodan GA, Martin TJ. Therapeutic approaches to bone diseases. Science 2000;289:1508–14. 31. Knauper V, Lopez-Otin C, Smith B, Knight G, Murphy G. Biochemical characterization of human collagenase-3. J Biol Chem 1996;271:1544–50. 32. Johansson N, Saarialho-Kere U, Airola K, Herva R, Nissinen L, Westermarck J, et al. Collagenase-3 (MMP-13) is expressed by hypertrophic chondrocytes, periosteal cells, and osteoblasts during human fetal bone development. Dev Dyn 1997;208:387–97. 33. Goldring SR, Gravallese EM. Pathogenesis of bone erosions in rheumatoid arthritis. Curr Opin Rheumatol 2000;12:195–9. 34. Fang Q, Sun YY, Cai W, Dodge GR, Lotke PA, Williams WV. Cartilage-reactive T cells in rheumatoid synovium. Int Immunol 2000;12:659–69. 35. Feldmann M, Brennan FM, Maini RN. Role of cytokines in rheumatoid arthritis. Annu Rev Immunol 1996;14:397–440. 36. 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. 37. Rydziel S, Varghese S, Canalis E. Transforming growth factor beta1 inhibits collagenase 3 expression by transcriptional and post-transcriptional mechanisms in osteoblast cultures. J Cell Physiol 1997;170:145–52. 38. Uchida M, Shima M, Shimoaka T, Fujieda A, Obara K, Suzuki H, et al. Regulation of matrix metalloproteinases (MMPs) and tissue inhibitors of metalloproteinases (TIMPs) by bone resorptive factors in osteoblastic cells. J Cell Physiol 2000;185:207–14. 39. Shlopov BV, Smith GN Jr, Cole AA, Hasty KA. Differential patterns of response to doxycycline and transforming growth factor ␤1 in the down-regulation of collagenases in osteoarthritic and normal human chondrocytes. Arthritis Rheum 1999;42:719–27. 40. Moldovan F, Pelletier JP, Hambor J, Cloutier JM, Martel-Pelletier J. Collagenase-3 (matrix metalloprotease 13) is preferentially localized in the deep layer of human arthritic cartilage in situ: in vitro mimicking effect by transforming growth factor ␤. Arthritis Rheum 1997;40:1653–61. 41. Ala-aho R, Johansson N, Grenman R, Fusenig NE, Lopez-Otin C, Kahari VM. Inhibition of collagenase-3 (MMP-13) expression in transformed human keratinocytes by interferon-gamma is associated with activation of extracellular signal-regulated kinase-1,2 and STAT1. Oncogene 2000;19:248–57. 42. Albanesi C, Scarponi C, Cavani A, Federici M, Nasorri F, Girolomoni G. Interleukin-17 is produced by both Th1 and Th2 lymphocytes, and modulates interferon-gamma- and interleukin-4-induced activation of human keratinocytes. J Invest Dermatol 2000;115:81–7. 43. Chabaud M, Miossec P. The combination of tumor necrosis factor ␣ blockade with interleukin-1 and interleukin-17 blockade is more effective for controlling synovial inflammation and bone resorption in an ex vivo model. Arthritis Rheum 2001;44:1293–303. 44. Van Bezooijen RL, Farih-Sips HC, Papapoulos SE, Lowik CW. 1001 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. Interleukin-17: a new bone acting cytokine in vitro. J Bone Miner Res 1999;14:1513–21. Chabaud M, Lubberts E, Joosten L, van Den BW, Miossec P. IL-17 derived from juxta-articular bone and synovium contributes to joint degradation in rheumatoid arthritis. Arthritis Res 2001;3: 168–77. Miwa M, Kozawa O, Tokuda H, Uematsu T. Mitogen-activated protein (MAP) kinases are involved in interleukin-1 (IL-1)-induced IL-6 synthesis in osteoblasts: modulation not of p38 MAP kinase, but of p42/p44 MAP kinase by IL-1-activated protein kinase C. Endocrinology 1999;140:5120–5. Kozawa O, Tokuda H, Matsuno H, Uematsu T. Involvement of p38 mitogen-activated protein kinase in basic fibroblast growth factor-induced interleukin-6 synthesis in osteoblasts. J Cell Biochem 1999;74:479–85. Tokuda H, Kozawa O, Uematsu T. Basic fibroblast growth factor stimulates vascular endothelial growth factor release in osteoblasts: divergent regulation by p42/p44 mitogen-activated protein kinase and p38 mitogen-activated protein kinase. J Bone Miner Res 2000;15:2371–9. Ravanti L, Heino J, Lopez-Otin C, Kahari VM. Induction of collagenase-3 (MMP-13) expression in human skin fibroblasts by three-dimensional collagen is mediated by p38 mitogen-activated protein kinase. J Biol Chem 1999;274:2446–55. Chaudhary LR, Avioli LV. Extracellular-signal regulated kinase signaling pathway mediates downregulation of type I procollagen gene expression by FGF-2, PDGF-BB, and okadaic acid in osteoblastic cells. J Cell Biochem 2000;76:354–9. Xiao G, Jiang D, Thomas P, Benson MD, Guan K, Karsenty G, et al. MAPK pathways activate and phosphorylate the osteoblastspecific transcription factor, Cbfa1. J Biol Chem 2000;275:4453–9. Whitmarsh AJ, Yang SH, Su MS, Sharrocks AD, Davis RJ. Role of p38 and JNK mitogen-activated protein kinases in the activation of ternary complex factors. Mol Cell Biol 1997;17:2360–71. Wang X, Rao J, Studzinski GP. Inhibition of p38 MAP kinase activity up-regulates multiple MAP kinase pathways and potentiates 1,25-dihydroxy vitamin D(3)-induced dedifferentiation of human leukemia. Birkenkamp KU, Tuyt LM, Lummen C, Wierenga AT, Kruijer W, Vellenga E. The p38 MAP kinase inhibitor SB203580 enhances nuclear factor-B transcriptional activity by a non-specific effect upon the ERK pathway. Br J Pharmacol 2000;131:99–107. Johansson N, Ala-aho R, Uitto V, Grenman R, Fusenig NE, Lopez-Otin C, et al. Expression of collagenase-3 (MMP-13) and collagenase-1 (MMP-1) by transformed keratinocytes is dependent on the activity of p38 mitogen-activated protein kinase. J Cell Sci 2000;113 Pt 2:227–35. Mengshol JA, Vincenti MP, Coon CI, Barchowsky A, Brinckerhoff CE. Interleukin-1 induction of collagenase 3 (matrix metalloproteinase 13) gene expression in chondrocytes requires p38, c-Jun N-terminal kinase, and nuclear factor B: differential regulation of collagenase 1 and collagenase 3. Arthritis Rheum 2000;43:801–11. Bondeson J, Brennan F, Foxwell B, Feldmann M. Effective adenoviral transfer of IB␣ into human fibroblasts and chondrosarcoma cells reveals that the induction of matrix metalloproteinases and proinflammatory cytokines is nuclear factor B-dependent. J Rheumatol 2000;27:2078–89. D’Alonzo RC, Selvamurugan N, Karsenty G, Partridge NC. Physical interaction of the activator protein-1 factors c-Fos and c-Jun with Cbfa1 for collagenase-3 promoter activation. J Biol Chem 2002;277:816–22. Hess J, Porte D, Munz C, Angel P. AP-1 and Cbfa/runt physically interact and regulate parathyroid hormone-dependent MMP13 expression in osteoblasts through a new osteoblast-specific element 2/AP-1 composite element. J Biol Chem 2001;276:20029–38.