ARTHRITIS & RHEUMATOLOGY Vol. 66, No. 9, September 2014, pp 2391–2402 DOI 10.1002/art.38701 © 2014, American College of Rheumatology
Key Role of Suppressor of Cytokine Signaling 3 in Regulating gp130 Cytokine–Induced Signaling and Limiting Chondrocyte Responses During Murine Inflammatory Arthritis Xiao Liu,1 Ben A. Croker,1 Ian K. Campbell,1 Stephanie J. Gauci,2 Warren S. Alexander,1 Brett A. Tonkin,3 Nicole C. Walsh,3 Edmond M. Linossi,1 Sandra E. Nicholson,1 Kate E. Lawlor,1 and Ian P. Wicks1 Socs3 driven by the Col2a1 promoter in vitro (Socs3⌬/⌬col2) and from mice during CD4ⴙ T cell–dependent inflammatory monarthritis. Bone erosions in the murine joints were analyzed by micro–computed tomography. Results. On chondrocytes from WT mice, gp130 and the oncostatin M (OSM) receptor were strongly expressed, whereas the transmembrane interleukin-6 (IL-6) receptor was expressed at much lower levels. Compared to other gp130 cytokines, OSM was the most potent activator of the JAK/STAT pathway and of SOCS-3 induction. Treatment of Socs3⌬/⌬col2 mouse cartilage explants and chondrocytes with gp130 cytokines prolonged JAK/STAT signaling, enhanced cartilage degradation, increased the expression of Adamts4, Adamts5, and RANKL, and elevated the production of IL-6, granulocyte colony-stimulating factor, CXCL1, and CCL2. Socs3⌬/⌬col2 mice developed exacerbated inflammation and joint damage in response to gp130 cytokine injections, and these histopathologic features were also observed in mice with inflammatory monarthritis. Conclusion. The results of this study highlight a key role for SOCS-3 in regulating chondrocyte responses during inflammatory arthritis. Within the gp130 cytokine family, OSM is a potent stimulus of chondrocyte responses, while IL-6 probably signals via trans-signaling. The gp130 cytokine–driven production of RANKL in chondrocytes may link chondrocyte activation and bone remodeling during inflammatory arthritis. Thus, these findings suggest that the inhibition of OSM might reduce the development and severity of structural joint damage during inflammatory arthritis.
Objective. To examine the impact of the gp130 cytokine family on murine articular cartilage and to explore a potential regulatory role of suppressor of cytokine signaling 3 (SOCS-3) in murine chondrocytes. Methods. In wild-type (WT) mouse chondrocytes, baseline receptor expression levels and gp130 cytokine– induced JAK/STAT signaling were determined by flow cytometry, and expression of SOCS-3 was assessed by quantitative polymerase chain reaction. The role of endogenous SOCS-3 was examined in cartilage explants and chondrocytes from mice with conditional deletion of
Supported by the Reid Charitable Trusts, the Arthritis Foundation of Australia, the National Health and Medical Research Council of Australia (grants 1023407 and 1016647 to Dr. Wicks, Fellowship 1058344 to Dr. Alexander, and Independent Research Institutes Infrastructure Support Scheme 361646), and the Victorian State Government (Operational Infrastructure Grant). 1 Xiao Liu, PhD, Ben A. Croker, PhD (current address: Boston Children’s Hospital, Boston, Massachusetts), Ian K. Campbell, PhD (current address: CSL Ltd., Parkville, Victoria, Australia), Warren S. Alexander, PhD, Edmond M. Linossi, BSc, Sandra E. Nicholson, PhD, Kate E. Lawlor, PhD, Ian P. Wicks, MB BS, FRACP, PhD: The Walter and Eliza Hall Institute of Medical Research, Parkville, Victoria, Australia and University of Melbourne, Parkville, Victoria, Australia; 2Stephanie J. Gauci, PhD: Murdoch Children’s Research Institute and Royal Children’s Hospital, Parkville, Victoria, Australia; 3Brett A. Tonkin, SBMS (Hons), Nicole C. Walsh, PhD: St Vincent’s Institute of Medical Research, Fitzroy, Victoria, Australia. Dr. Alexander holds patent PCT/AU97/00729, which pertains to the use of SOCS proteins in experimental models of arthritis. Address correspondence to Ian P. Wicks, MB BS, FRACP, PhD, Reid Rheumatology Laboratory, Inflammation Division, The Walter and Eliza Hall Institute of Medical Research, 1G Royal Parade, Parkville, Victoria 3052, Australia. E-mail:
[email protected]. Submitted for publication December 18, 2013; accepted in revised form May 6, 2014.
2391
2392
Articular cartilage degradation is a key feature of inflammatory joint diseases such as rheumatoid arthritis (RA). Chondrocytes are the sole cell type present in cartilage and are responsible for producing the complex extracellular matrix (ECM) that is remodeled by catabolic enzymes of the matrix metalloproteinase and ADAMTS families throughout life. Recent therapeutic developments have highlighted the importance of the interleukin-6 (IL-6) family of cytokines (hereafter referred to as gp130 cytokines) and downstream signaling pathways in inflammatory arthritis (1,2). However, in contrast to other cell types, relatively little is known about how chondrocytes regulate the responses to gp130 cytokines. These cytokines bind to ligand-specific receptors on the cell surface and then engage a common signaling gp130 -subunit to exert biologic activity. Activation of the ligand– receptor–gp130 complex leads to the phosphorylation of cytoplasmic tyrosine residues on gp130 by JAK-1, JAK-2, and Tyk-2 kinases. C-terminal–phosphorylated tyrosine residues on gp130 are required for the activation of latent STAT-1, STAT-3, and STAT-5 transcription factors. The suppressor of cytokine signaling (SOCS) family of proteins regulates the intensity and duration of signal transduction in response to a wide range of cytokines and hormones (3). Within this family, SOCS-3 is the key regulator of cellular responses downstream of the gp130 receptor (4). In murine models of experimental arthritis, the inhibition of gp130 cytokines via neutralizing agents (5,6) or the use of gene-knockout mice (7–9) have provided evidence to demonstrate that this family of cytokines has both proinflammatory and catabolic roles. Previous studies have addressed the potential role of SOCS-3 in joint disease. In Socs3⫺/⫺ mice, conditional deletion of the SOCS-3 gene is embryonically lethal, due to excessive leukemia inhibitory factor (LIF) signaling in trophoblasts (4). Therefore, conditional gene targeting is required to assess the role of SOCS-3 in a tissue-specific manner. Previously, we showed that deletion of Socs3 in hematopoietic and endothelial cells using the Vav-Cre system exacerbated acute inflammatory monarthritis in mice, with increased generation of Th17 cells (10). Furthermore, mice expressing a truncated SOCS-3 protein lacking the SOCS box also had exacerbated acute inflammatory arthritis (11). In this study, we investigated whether SOCS-3 plays a specific role in modulating gp130 cytokine signaling in chondrocytes. We found that, in murine articular cartilage and chondrocytes, IL-6 mediates its
LIU ET AL
effects via trans-signaling and that oncostatin M (OSM) may be the dominant member of the gp130 cytokine family. Furthermore, our studies revealed that chondrocytes use SOCS-3 as a key negative regulator of gp130 signal transduction during inflammatory arthritis and that these cells can produce a surprising array of effector molecules, including cytokines, chemokines, and RANKL. MATERIALS AND METHODS Animals. Socs3fl/fl mice (12) were crossed onto the Col2a1Cre (13) and Rosa26 (R26R) (14) backgrounds to generate Socs3⌬/⌬col2 or R26Rgt/⫹Socs3⌬/⌬col2 conditional knockout mice. Socs3fl/fl and R26Rgt/⫹Socs3fl/fl littermate mice served as controls in all experiments, unless stated otherwise. Socs3⫺/⌬vav mice were generated as previously described (15). Wild-type (WT) C57BL/6J mice were obtained from the Walter and Eliza Hall Institute (WEHI) animal services (Kew, Victoria, Australia). For ex vivo chondrocyte cultures, cartilage explant cultures, and in vivo models, we used mice that were 5–6 days old, 2–3 weeks old, and 8–12 weeks old, respectively. All mice were fed standard rodent chow and water ad libitum, and were housed (6 mice/cage) in sawdust-lined cages. The WEHI Animal Ethics Committee (Parkville, Victoria, Australia) approved all animal procedures. Isolation of cartilage and organ culture. Femoral head cartilage was isolated from 2–3-week-old mice using forceps to shear the capital femoral epiphyseal cartilage from the underlying metaphyseal bone. The cartilage explants were weighed and cultured for 3 days in Dulbecco’s modified Eagle’s medium (DMEM) (containing 100 units/ml penicillin, 100 g/ml streptomycin, 2 mM L-glutamine, and 20 mM HEPES), supplemented with 10% fetal calf serum (FCS) at 37°C (in a humidified atmosphere of 5% CO2/95% air), prior to cytokine stimulation in serum-free DMEM for 3 days. For analysis of aggrecan content, the femoral heads were blotted dry and stained with toluidine blue for histologic analysis or were digested with papain. Aggrecan content in the conditioned medium and in papain-digested cartilage was determined by 1,9-dimethylmethylene blue (DMMB) assay (Polysciences), as described previously (16). Isolation of primary chondrocytes for stimulation with cytokines. Chondrocytes were harvested as described previously (17). Proximal tibial and distal femoral epiphyses from 5–6-day-old mice were harvested under a dissecting microscope. Pooled epiphyses were incubated with 0.25% trypsin for 1 hour at 37°C, prior to digestion with type II collagenase (2 mg/ml in DMEM–5% FCS; Worthington), followed by shaking overnight at 37°C. Cells were then washed with DMEM–10% FCS and counted. For stimulation assays, chondrocytes were plated at 1.5–2 ⫻ 105 cells/well in 96-well flat-bottomed plates containing DMEM with 10% FCS. Cells were stimulated with recombinant human IL-1 or IL-6/soluble IL-6 receptor (sIL-6R) (both from R&D Systems) at equimolar concentrations or with
SOCS-3 REGULATION OF gp130 CYTOKINES IN MURINE ARTHRITIS
murine OSM (R&D Systems) at 100 ng/ml or IL-11 or LIF (both from WEHI) at 100 ng/ml. Quantitative polymerase chain reaction (qPCR). Total RNA was prepared from cultured cells using the RNeasy Mini kit (Qiagen). Complementary DNA was prepared by reverse transcription. Quantitative PCR was performed on a Bio-Rad C100 Thermal Cycler instrument, and expression levels of Gapdh (Mm99999915_g1), Adamts4 (Mm00556068_m1), Adamts5 (Mm00478620_m1), osteoprotegerin (OPG) (Tnfrsf11b, Mm01205928_m1), RANKL (Tnfsf11, Mm00441906_m1), and Socs3 (Mm00545913_s1) transcripts were determined using TaqMan probes. Transcript expression was determined using the 2⫺⌬⌬Ct method (18), with values normalized against those for Gapdh. Evaluation of STAT phosphorylation in chondrocytes by flow cytometry. Chondrocytes were pulsed with cytokine for 30 minutes, fixed in Phosflow Lyse/Fix Buffer (BD Biosciences), and permeabilized with Phosflow Perm Buffer III (BD Biosciences). Thereafter, the chondrocytes were incubated with mouse anti-human phycoerythrin (PE)–conjugated pSTAT-3, PE-conjugated pSTAT-1, or Alexa 647–conjugated pSTAT-5 antibodies (all from BD Biosciences) or isotype control (PE-conjugated mouse IgG2a, PE-conjugated IgG2a, or Alexa 647–conjugated IgG1 [all from Caltag Laboratories]) overnight at 4°C. Stained cells were washed and analyzed by flow cytometry (LSRII; BD Biosciences). Flow cytometry analysis of gp130 receptor expression on chondrocytes. Chondrocytes were incubated with 2.4G2 antibody for 15 minutes on ice, and then stained with 2 g/ml of rat anti-mouse biotin-conjugated gp130, biotin-conjugated OSM receptor  (OSMR), biotin-conjugated IL-6R, or biotin-conjugated LIF receptor ␣ (LIFR␣) (all from R&D Systems), biotin-conjugated anti–IL-11 receptor (anti–IL-11R) (WEHI), or isotype control antibodies (biotin-conjugated IgG2a and biotin-conjugated IgG1 [both from BD Biosciences]) for 15 minutes on ice. Cells were then incubated with a PE-conjugated streptavidin secondary antibody (R&D Systems) for 15 minutes on ice, and then washed and analyzed by flow cytometry (LSRII; BD Biosciences). Receptor expression was quantified using quantiBRITE PE beads (BD Biosciences), in accordance with the manufacturer’s instructions. Data were analyzed using FlowJo software (Tree Star). Analysis of cytokine and chemokine production by multiplex bead array. Cytokines and chemokines were measured in conditioned medium using the Bio-Plex mouse cytokine 23-plex panel (Bio-Rad Laboratories), performed according to the manufacturer’s instructions. The assay was read on a Bio-Plex 200 instrument and analyzed using Bio-Plex Manager version 5.0 software. Injection of mouse knee joints with IL-1, OSM, IL-6/sIL-6R, IL-11, or LIF. For cytokine-induced joint inflammation, 8–12-week-old mice were anesthetized and injected with 125 ng (in 10 l) of IL-1, OSM, IL-6/sIL-6R, IL-11, or LIF intraarticularly into the knee joint. Injections were repeated each day for the subsequent 2 days. Control joints received the same volume of saline. Knee joints were excised and harvested on day 4 for histologic assessment.
2393
Methylated bovine serum albumin (mBSA)/IL-1– induced monarthritis. Monarthritis was induced in mice using intraarticular injections of mBSA/IL-1, as previously described (19). Briefly, 8–12-week-old mice were anesthetized and 10 l of 20 mg/ml mBSA (Sigma-Aldrich) was injected intraarticularly into the knee joint. Control joints received the same volume of normal saline. The rear footpad of each mouse was then injected subcutaneously with 20 l of 12.5 g/ml recombinant human IL-1 (R&D Systems), with injections repeated on the next 2 days. Mice were harvested on day 7 (the peak of disease) and the knee joints were excised for micro–computed tomography (micro-CT) analysis and histologic assessment. Histologic assessment of arthritis. Harvested knee joints were fixed in 4% paraformaldehyde and demineralized. Frontal sections (4 m) were then stained with hematoxylin and eosin or toluidine blue to assess the histopathologic features of the joints. Histologic sections were assessed in a manner as previously described (19). The sections were graded for the level of severity on a scale from 0 (normal) to 5 (severe) for the following components: joint exudate, synovitis, pannus formation, cartilage destruction, or bone erosion. Images were obtained using a Zeiss Axioplan 2 microscope (with 5⫻/0.15 and 10⫻/0.30 dry objectives) fitted with a Zeiss AxioCam HRc camera. Images were captured using Zeiss AxioVision version 3.1 and scale bars were added using NIH ImageJ software version 1.447q. Micro-CT. Ex vivo micro-CT imaging of the mouse tibiae was performed using a SkyScan 1076 scanner (Bruker micro-CT). The settings for data acquisition were as follows: voxel resolution of 9 m, aluminum filter of 0.5 mm, voltage of 48 kV, current of 100 A, exposure time of 2,400 msec, rotation of 0.5° across 180°, and frame averaging of 1. Reconstruction was performed using NRecon (version 1.6.3.1) with the following parameters: dynamic image range 0.0–0.12, ring artifact of 6, 5% pixel defect mask, and 35% beam-hardening correction. Three-dimensional models of the knee joint were generated from raw data using CTVox (version 2.4.0; Bruker micro-CT) and pseudo-colored according to density. Detection of SOCS-3 expression in cartilage by Western blotting. Articular cartilage was harvested from the joints of Socs3fl/fl and Socs3⌬/⌬col2 mice with mBSA/IL-1–induced inflammatory arthritis. Pooled cartilage samples were digested with type II collagenase (2 mg/ml in DMEM–5% FCS), followed by shaking overnight at 37°C. Protein lysates were prepared, and 60 g of protein was separated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis and transferred to nitrocellulose membranes, as described previously (20). Membranes were blocked overnight in 10% skim milk, incubated with primary rabbit anti–SOCS-3 antibody (IBL) for 2 hours, and then incubated with secondary peroxidaseconjugated donkey anti-rabbit immunoglobulin (GE Healthcare) or horseradish peroxidase–conjugated actin (Santa Cruz Biotechnology). Signals were visualized using the enhanced chemiluminescence system (Millipore). Statistical analysis. Results in all experiments represent the findings from at least 3 independent experiments with up to 3 biologic replicates, unless stated otherwise. Student’s unpaired 2-tailed t-tests were used to compare the extent of aggrecan release, levels of gene expression assessed by qPCR,
2394
LIU ET AL
and levels of cytokine/chemokine expression assessed by BioPlex assay. Multiple comparisons were analyzed by two-way analysis of variance, followed by Tukey’s test. The MannWhitney 2-tailed test was used to analyze histologic scores in arthritis models. P values less than 0.05 were considered significant.
RESULTS Expression of Socs3 and gp130 cytokine receptors and activation of STAT-3 and STAT-1 in mouse primary chondrocytes in response to stimulation with gp130 cytokines. Flow cytometry analysis of membrane-bound receptors on epiphyseal chondrocytes from WT mice provided evidence of gp130 receptor expression, including the expression of OSMR, IL-6R, IL-11R, and LIFR, as well as the gp130 subunit itself (Figure 1A). Semiquantitative assessment of the receptor number per cell showed that, compared to chondrocyte expression of OSMR and LIFR, the gp130 subunit was expressed at higher levels, while the membrane-bound IL-6R and IL-11R were expressed at lower levels. Of the ligandbinding gp130 receptors, OSMR appeared to be the most strongly expressed. To assess the biochemical consequences of gp130 signaling, JAK/STAT activation was investigated in isolated chondrocytes by intracellular staining and flow cytometry. Phosphorylation of STAT-3 and STAT-1 was maximal in WT mouse chondrocytes at 30 minutes after stimulation with gp130 cytokines in vitro (Figure 1B) and declined to basal levels over 1–2 hours. In contrast, the IL-1–induced phosphorylation of STAT-3 increased at 4 hours after stimulation, in keeping with the notion of indirect activation by IL-1. These findings demonstrate that gp130 cytokines activate the JAK/ STAT pathway in chondrocytes and that IL-6–induced responses are likely to be mediated via IL-6 transsignaling. SOCS-3 is a critical regulator of IL-6–mediated JAK/STAT signaling in hepatocytes, macrophages, and T cells (4). To investigate whether SOCS-3 is expressed in chondrocytes, mouse primary chondrocytes were stimulated in vitro with gp130 cytokines. Up-regulation of Socs3 transcript expression was confirmed by qPCR; IL-1, which does not signal via gp130, was used as a negative control. Socs3 messenger RNA (mRNA) expression was detected in the chondrocytes as early as 30 minutes after stimulation with gp130 cytokines, and maximal levels of Socs3 mRNA were observed between 1 and 2 hours after stimulation (Figure 1C). These findings correlated with the observed declines in pSTAT
Figure 1. Stimulation of mouse primary chondrocytes with gp130 cytokines induces Socs3 mRNA expression and phosphorylation of STAT-1 and STAT-3. A, Mouse primary epiphyseal chondrocytes were analyzed by flow cytometry for receptor expression of gp130, oncostatin M (OSMR), interleukin-6 (IL-6R), IL-11 (IL-11R), and leukemia inhibitory factor (LIFR) (open histograms). Shaded histograms represent the signal from appropriately labeled isotype control antibodies. Quantification of receptor expression per cell was done using the QuantiBRITE PE system. Bars show the mean ⫾ SEM relative expression levels in 3 replicates/ group. B, Staining for pSTAT-3 and pSTAT-1 signaling in wild-type mouse chondrocytes was performed on cells left unstimulated (Unstim) or stimulated with gp130 cytokines or IL-1. Bars show the mean ⫾ SEM fluorescence intensity of 2 replicates/group. C, Quantitative polymerase chain reaction was performed to analyze Socs3 mRNA expression in mouse epiphyseal chondrocytes left unstimulated or stimulated with gp130 cytokines or IL-1. Bars show the mean ⫾ SEM of 2 replicates/ group. Results are representative of at least 3 independent experiments. sIL-6R ⫽ soluble IL-6 receptor.
SOCS-3 REGULATION OF gp130 CYTOKINES IN MURINE ARTHRITIS
Figure 2. Deletion of suppressor of cytokine signaling 3 (SOCS-3) prolongs STAT-3, STAT-1, and STAT-5 signaling in mouse primary chondrocytes. A and B, Flow cytometry was used to analyze pSTAT-3 staining (A) and pSTAT-1 staining (B) in primary chondrocytes from Socs3fl/fl and Socs3⌬/⌬col2 mice after stimulation with the indicated cytokines. C, Phosphorylation of STAT-5 was assessed in oncostatin M (OSM)–stimulated chondrocytes from Socs3fl/fl and Socs3⌬/⌬col2 mice. Open histograms represent STAT signaling activity, and shaded histograms represent the signal from appropriately labeled isotype control antibodies. Results are representative of at least 3 independent experiments (n ⫽ 2 replicates/group). See Figure 1 for other definitions.
2395
signaling. Of note, stimulation of primary chondrocytes with OSM induced greater levels of Socs3 mRNA expression than did stimulation with the other gp130 cytokines. The pattern of induction of Socs3 mRNA expression by IL-6/sIL-6R showed similar kinetics as that by IL-11, whereas stimulation with LIF induced a rapid increase in Socs3 mRNA, followed by a rapid decline. IL-1 induced Socs3 mRNA expression at a later time point, consistent with the activation of a secondary cytokine pathway (Figure 1C). These results establish that SOCS-3 is inducible by gp130 cytokines in chondrocytes. Conditional deletion of Socs3 in chondrocytes associated with prolonged JAK/STAT activation. To investigate whether gp130 signaling is regulated by SOCS-3 in chondrocytes, mice with a conditional Socs3 allele (Socs3fl/fl) were crossed with a transgenic mouse expressing Cre recombinase under the control of the type II collagen (Col2a1) promoter (13). This promoter is reported to have activity in cartilaginous and some noncartilaginous elements during embryonic development (13), including ocular structures (21), renal tubules, pancreas, lungs (22), and synovial cells (23). To confirm the deletion of Socs3 in Socs3⌬/⌬col2 mice, we crossed this strain with -galactosidase reporter (R26R) mice (14). Staining for -galactosidase activity congt/⫹ firmed the expression of Cre recombinase in R26R Socs3⌬/⌬col2 mouse femoral head chondrocytes (results not shown). We also confirmed that IL-6/sIL-6R induced Socs3 mRNA expression in chondrocytes from both WT and Socs3fl/fl mice, but not in chondrocytes from Socs3⌬/⌬col2 mice (results not shown). Socs3⌬/⌬col2 mice were healthy and fertile but were smaller than their littermate controls. Skeletal radiographs and histologic assessments confirmed that the joint architecture of Socs3⌬/⌬col2 mice was normal. Further analysis of the skeletal phenotype will be the subject of a more detailed future report. To determine the effects of SOCS-3 on the regulation of cytokine responses in chondrocytes, we assessed the relative expression of gp130 cytokine receptors on Socs3⌬/⌬col2 mouse chondrocytes. We found that the levels of all of the gp130 cytokine receptors remained unchanged in Socs3⌬/⌬col2 mice compared to those in WT control mice (results not shown). We next evaluated whether SOCS-3 deficiency had any consequences for JAK/STAT signaling in chondrocytes. After stimulation of the chondrocytes with OSM, IL-6/sIL-6R, or LIF, Socs3⌬/⌬col2 mouse chondrocytes displayed sustained activation of STAT-3 for up to 2 hours, but
2396
IL-11–mediated phosphorylation of STAT-3 declined more rapidly (Figure 2A). The kinetics of pSTAT-1 signaling in SOCS-3–deficient mouse chondrocytes differed from those of pSTAT-3. Although pSTAT-1 was still detectable at 2 hours after stimulation with OSM, the IL-6/sIL-6R– and LIF-induced pSTAT-1 levels returned to baseline levels over 1–2 hours, and stimulation with IL-11 did not prolong the activation of pSTAT-1 (Figure 2B). Phosphorylated ERK signaling was not affected by the deletion of Socs3 in mouse chondrocytes (results not shown). Similar to our findings in WT mouse chondrocytes, OSM was the most potent inducer of pSTAT-3 and pSTAT-1 in Socs3⌬/⌬col2 mouse chondrocytes, and uniquely induced the activation of pSTAT-5 (Figure 2C). These results are in keeping with the observation that the Box-2 region of OSMR can recruit JAK-2 to activate STAT-5 (24). Therefore, while chondrocytes express receptors for each of the gp130 cytokines, the kinetics of SOCS-3 and JAK/STAT activation vary in response to different members of this cytokine family. Prolonged JAK/STAT signaling in the absence of SOCS-3 in chondrocytes and exacerbation of cartilage damage in vitro. Proteoglycans are macromolecular components of the cartilage ECM, and depletion of aggrecan, the most abundant proteoglycan, is a characteristic feature of inflammatory and degenerative joint diseases. To evaluate whether sustained JAK/STAT signaling in SOCS-3–deficient mouse chondrocytes potentiates cartilage degradation, femoral head cartilage explants from Socs3⌬/⌬col2 and Socs3fl/fl mice were cultured in the presence of gp130 cytokines. Aggrecan release was determined using the DMMB assay (16). All gp130 cytokines stimulated the release of aggrecan (measured as the percentage of glycosaminoglycan release) from Socs3fl/fl mouse cartilage (Figures 3A and B). In keeping with these findings, gp130 cytokines also strongly induced the aggrecanases, Adamts4 and Adamts5, with OSM and IL-6/sIL-6R inducing higher levels as compared to those induced by IL-11 or LIF (Figures 3C and D). Deletion of Socs3 resulted in a striking increase in aggrecan release in response to gp130 cytokine stimulation (Figures 3A and B). Expression analysis of Adamts4 and Adamts5 in stimulated Socs3⌬/⌬col2 mouse chondrocytes revealed marked increases in the levels of both transcripts (Figures 3C and D). IL-1 (a known activator of cartilage catabolism) also induced greater aggrecan release and aggrecanase expression in Socs3⌬/⌬col2 mouse chondrocytes (Figures 3A–D), which
LIU ET AL
Figure 3. Suppressor of cytokine signaling 3 (SOCS-3) regulates aggrecanase activity in mouse primary chondrocytes. A, Femoral head cartilage explants from Socs3fl/fl and Socs3⌬/⌬col2 mice were left unstimulated (Unstim) or were stimulated for 3 days with IL-1 or gp130 cytokines, followed by staining with toluidine blue for histologic analysis. Representative results are shown. Bar ⫽ 100 m. B, Quantification of aggrecan release was performed using the 1,9dimethylmethylene blue dye-binding assay in Socs3fl/fl and Socs3⌬/⌬col2 mouse cartilage explants left unstimulated or stimulated with IL-1 or gp130 cytokines. Bars show the mean ⫾ SEM aggrecan loss (% glycosaminoglycan [GAG] release) in 4 explants/group, calculated as (aggrecan released into the medium divided by [total aggrecan present in the medium ⫹ extracts ⫹ papain digests]) ⫻ 100. C and D, Expression of mRNA for Adamts4 (C) and Adamts5 (D) was determined in primary chondrocytes from Socs3fl/fl and Socs3⌬/⌬col2 mice. The chondrocytes were left unstimulated or were stimulated for 4 hours with IL-1 or gp130 cytokines. Bars show the mean ⫾ SEM of 4 replicates/group. Results are representative of 2–3 independent experiments. ⴱ ⫽ P ⬍ 0.05; ⴱⴱⴱ ⫽ P ⬍ 0.0001, versus Socs3fl/fl mouse chondrocytes within each condition. See Figure 1 for other definitions.
is probably attributable to the secondary induction of gp130 cytokines in addition to SOCS-3–independent effects. These findings demonstrate that SOCS-3 normally plays a powerful regulatory role in restraining the induction of ADAMTS-4 and ADAMTS-5 in response to gp130 cytokines.
SOCS-3 REGULATION OF gp130 CYTOKINES IN MURINE ARTHRITIS
Association of SOCS-3 deficiency in murine chondrocytes with exacerbation of joint inflammation and cartilage damage in vivo. We confirmed that the expression of SOCS-3 protein was found in chondrocytes from Socs3fl/fl mice with mBSA/IL-1–induced inflammatory arthritis, but was not found in chondrocytes from the joints of naive Socs3fl/fl or diseased Socs3⌬/⌬col2 mice (Figure 4A). We have previously shown that hematopoietic cell–specific deletion of Socs3 (using Vav-Cre) exacerbated mBSA/IL-1–induced monarthritis in mice, an effect that was dependent on IL-6, IL-1, CD4⫹ T cells, macrophages, and neutrophils (7,10). To examine the effects of Socs3 deletion on chondrocytes in comparison to its effects on hematopoietic cells, we compared the severity of this model of arthritis between Socs3⫺/⌬vav mice and Socs3⌬/⌬col2 mice. All histopathologic features in the knee joints, including synovitis, pannus formation, cartilage degradation, and bone erosion, were exacerbated in both the Socs3⫺/⌬vav and Socs3⌬/⌬col2 mice, and the disease was at least as severe in Socs3⌬/⌬col2 mice (Figures 4B and C). Of particular interest, cartilage degradation and bone erosion were more prominent in Socs3⌬/⌬col2 mice. Taken together, our findings suggest a major, yet unrecognized, contribution of chondrocytes to joint inflammation and damage. In view of these findings, it was important to address the specificity of the Socs3 deletion, particularly because synovial fibroblasts (SFs) share the mesenchymal heritage of chondrocytes (25,26) and lacZ activity has recently been reported in the synovium of Col2-Cre/ R26R mice (23). Examination of Cre expression in SFs derived from R26Rgt/⫹Socs3⌬/⌬col2 reporter mice confirmed that the level of Col2a1-Cre expression was low. However, culture of SFs from Socs3fl/fl and Socs3⌬/⌬col2 mice with gp130 cytokines failed to show any alteration of Socs3 mRNA expression in Socs3-deleted mouse SFs, nor were there any differences in cytokine production from these cells (results not shown). We also detected Cre expression in a small percentage (mean ⫾ SEM 3 ⫾ 1%) of CD45⫹Gr-1⫹Ly-6G⫹ cells (i.e., probably neutrophils) in the bone marrow. To determine the significance of SOCS-3 expression in hematopoietic-derived cells compared to non–hematopoietic-derived cells, we generated reciprocal bone marrow chimeras (reconstitution efficiency mean ⫾ SEM 90 ⫾ 2%). In Socs3⌬/⌬col2 mice reconstituted with WT mouse bone marrow, mBSA/IL-1– induced arthritis was exacerbated. In contrast, there was no difference in the disease course or the severity of
2397
Figure 4. Joint inflammation is induced by gp130 cytokines, and suppressor of cytokine signaling 3 (SOCS-3) limits chondrocytemediated inflammatory arthritis. Acute inflammatory arthritis was induced in 8–12-week-old mice. A, Western blotting was performed to analyze the expression of SOCS-3 in chondrocytes from naive Socs3fl/fl mice and from Socs3fl/fl and Socs3⌬/⌬col2 mice with methylated bovine serum albumin (mBSA)/IL-1–induced arthritis. Embryonic stem (ES) cells cultured in LIF were used as a positive control for SOCS-3 protein expression; -actin was used as loading control. Endogenous SOCS-3 can be detected as a doublet at ⬃25 kd (see refs. 39 and 40 for more details). The asterisk indicates that a shorter exposure time was used for the positive control in the same Western blot. B, Processed knee joints from Socs3fl/fl, Socs3⫺/⌬vav, and Socs3⌬/⌬col2 mice with mBSA/IL-1–induced arthritis were graded for histopathologic features of arthritis. Bars show the mean ⫾ SEM of 16 joints/group. C, Knee cartilage sections from Socs3fl/fl, Socs3⫺/⌬vav, and Socs3⌬/⌬col2 mice were stained with hematoxylin and eosin (H&E) and toluidine blue (Tol B) for histologic analysis. Representative results are shown. Bars ⫽ 250 m (for H&E) and 100 m (for Tol B). D–I, IL-1 (D), OSM (E), IL-6/sIL-6R (F), IL-11 (G), LIF (H), or mBSA alone (I) was injected intraarticularly into the knee joints of Socs3fl/fl and Socs3⌬/⌬col2 mice on 3 consecutive days; joints were harvested on day 7. Bars show the mean ⫾ SEM histologic score in 8 joints/group. Results are representative of 2–3 independent experiments. ⴱ ⫽ P ⬍ 0.05; ⴱⴱ ⫽ P ⬍ 0.001; ⴱⴱⴱ ⫽ P ⬍ 0.0001, versus Socs3fl/fl mice or as indicated. See Figure 1 for other definitions.
2398
acute monarthritis between WT mice and lethally irradiated WT mice reconstituted with Socs3⌬/⌬col2 mouse bone marrow (results not shown). These results suggest that although subpopulations of SFs and neutrophils in Socs3⌬/⌬col2 mice have Cre expression and possible deletion of Socs3, the functional consequences of this appear to be minimal. To more specifically examine the role of SOCS-3 in regulating chondrocyte function in vivo, gp130 cytokines were directly injected into the knee joints of Socs3⌬/⌬col2 mice and littermate control mice. Compared to the littermate controls, Socs3⌬/⌬col2 mice had marked exacerbation of inflammation and cartilage degradation in response to intraarticular injections of each gp130 cytokine, as well as to mBSA alone (Figures 4D–I). The exacerbation of disease in response to IL-1 was likely attributable to its direct effects and to the secondary induction of gp130 cytokines, as discussed above. These findings clearly show that deletion of Socs3 in chondrocytes has major pathologic consequences in vivo. Regulatory role of SOCS-3 in the production of cytokines, chemokines, and RANKL. Chondrocyte responses are traditionally considered to be limited to cartilage remodeling. However, our in vivo findings suggested that chondrocytes stimulated with gp130 cytokines might also drive inflammatory responses. To determine whether chondrocytes can produce proinflammatory mediators in response to gp130 cytokines, primary chondrocytes were stimulated with gp130 cytokines (or with IL-1 as a positive control), and the supernatants were examined by cytokine multiplex assays. OSM and, to a lesser extent, IL-6/sIL-6R stimulated the production of IL-6, granulocyte colonystimulating factor (G-CSF), CXCL1, and CCL2 from Socs3fl/fl mouse chondrocytes (Figure 5A). In contrast, these mediators were barely detectable in the conditioned medium from chondrocytes stimulated with IL-11 or LIF. Socs3⌬/⌬col2 mouse chondrocytes produced significantly more of these inflammatory mediators in response to gp130 cytokines (Figure 5A). In view of the pronounced bone erosion observed during inflammatory arthritis in Socs3⌬/⌬col2 mice, we examined whether mediators regulating bone resorption were altered in SOCS-3–deficient mouse chondrocytes. RANKL mediates bone resorption by promoting osteoclastogenesis and the activation of mature osteoclasts (27). RANKL was expressed in WT mouse chondrocytes, and the levels of RANKL were increased in response to IL-1, OSM, and IL-6/sIL-6R. This increase in the relative expression of RANKL was much greater
LIU ET AL
in Socs3⌬/⌬col2 mouse chondrocytes (Figure 5B). Of note, OSM and IL-6/sIL-6R induced a greater increase in RANKL expression than did IL-1. Levels of OPG, a decoy receptor for RANKL and an endogenous inhibitor of RANKL–RANK interactions (28,29), were also significantly increased after stimulation of WT mouse chondrocytes with OSM, whereas there was no significant increase in the OSM-induced levels of OPG in Socs3⌬/⌬col2 mouse chondrocytes (Figure 5B). The relative RANKL:OPG mRNA ratio was increased in chondrocytes after stimulation with OSM and IL-6/sIL-6R and was markedly increased in the absence of SOCS-3 (Figure 5B). Micro-CT imaging of the mouse knee joints during mBSA/IL-1–induced arthritis demonstrated extensive erosion of the cortical bone surface of the tibia and femur in Socs3⌬/⌬col2 mice compared to littermate controls (Figure 5C). Furthermore, these mice also showed reduced bone volume within the tibial epiphysis compared to the littermate controls, indicating that periarticular bone loss was increased (Figure 5D). These findings demonstrate that gp130 cytokines induce the production of proinflammatory cytokines and RANKL by chondrocytes, and highlight a regulatory role for SOCS-3 in a novel potential interaction between cartilage activation by gp130 cytokines and bone remodeling. DISCUSSION Recent therapeutic successes in patients with RA treated with tocilizumab, an anti–IL-6R antibody, or tofacitinib, a JAK-1/3 inhibitor, highlight the involvement of gp130 cytokines in joint inflammation. Cytokine-mediated cartilage damage is a major complication of inflammatory arthritis, but how chondrocytes regulate cellular responses to cytokines remains poorly understood. In this study, we demonstrate a key role for SOCS-3 in chondrocytes in mediating joint inflammation and joint damage. Genetic deletion of Socs3 in mouse primary chondrocytes prolonged the phosphorylation of STAT transcription factors and augmented responses to all gp130 cytokines. Intraarticular injections of each gp130 cytokine into the knee joints of WT mice caused synovial inflammation and cartilage damage. These effects were exacerbated in Socs3⌬/⌬col2 mice. Our findings showed that loss of SOCS-3 led to greater aggrecan release from cartilage explants and enhanced up-regulation of Adamts4 and Adamts5 expression in response to gp130 cytokines in vitro.
SOCS-3 REGULATION OF gp130 CYTOKINES IN MURINE ARTHRITIS
2399
Figure 5. Suppressor of cytokine signaling 3 (SOCS-3) regulates the production of cytokines and RANKL/osteoprotegerin (OPG) in chondrocytes to restrict bone erosion in mice. A, Primary chondrocytes from Socs3fl/fl and Socs3⌬/⌬col2 mice were cultured in the presence or absence of IL-1 or gp130 cytokines for 4 hours; secreted cytokines and chemokines in the conditioned medium were analyzed by multiplex bead array. Bars show the mean ⫾ SEM of 3 replicates/group. B, Epiphyseal chondrocytes from Socs3fl/fl and Socs3⌬/⌬col2 mice were stimulated with or without the indicated cytokines for 24 hours. The expression of mRNA for RANKL (Tnfsf11) and OPG (Tnfrsf11b) and the RANKL:OPG mRNA ratio were analyzed by quantitative polymerase chain reaction. Bars show the mean ⫾ SEM of 3 replicates/group. C, Representative micro–computed tomography (micro-CT) images show 3-dimensional reconstructions of the knee joints of Socs3fl/fl and Socs3⌬/⌬col2 mice injected with saline alone or injected with methylated bovine serum albumin (mBSA)/IL-1 to induce monarthritis. The results reveal increased bone erosion in Socs3⌬/⌬col2 mice with mBSA/IL-1–induced arthritis compared to their littermate controls. D, Representative micro-CT images show periarticular bone loss in mice with mBSA/IL-1–induced arthritis. The boxed area indicates marked loss of bone within the tibial epiphysis of Socs3⌬/⌬col2 mice. Results are representative of at least 3 independent experiments. ⴱ ⫽ P ⬍ 0.05; ⴱⴱ ⫽ P ⬍ 0.001; ⴱⴱⴱ ⫽ P ⬍ 0.0001, versus Socs3fl/fl mice or as indicated. G-CSF ⫽ granulocyte colony-stimulating factor (see Figure 1 for other definitions).
Increased cartilage degradation in Socs3⌬/⌬col2 mice during inflammatory arthritis is therefore likely to be a consequence of prolonged aggrecanase production in response to gp130 cytokines. Previous studies have demonstrated induction of aggrecanase mRNA and aggre-
can release by OSM and IL-6 in WT mouse chondrocytes (30,31), but to our knowledge, this is the first study to demonstrate that IL-11 and LIF also enhance aggrecanase activity. When stimulated with gp130 cytokines in vitro, SOCS-3–deficient mouse chondrocytes also
2400
produced higher levels of other cytokines (including IL-6 and G-CSF) and chemokines (CCL1 and CXCL1) that could help localize inflammatory responses. Elevated production of the neutrophil chemokine CXCL1 and G-CSF may contribute to neutrophil infiltration during joint inflammation (32). Socs3⌬/⌬col2 mice also developed exacerbated T cell–dependent inflammatory monarthritis, with particularly marked bone destruction. Previous studies have demonstrated that SOCS-3 regulates not only gp130 receptor signaling, but also signaling via Toll-like receptors (33) and IL-1 signaling via tumor necrosis factor receptor–associated factor 6/transforming growth factor –activated kinase 1 (34) and insulin receptor substrate 1 signaling (35). Therefore, deletion of SOCS-3 in chondrocytes could have potentially exacerbated secondary signaling cascades. Nonetheless, these results support the notion that chondrocytes can play an active, proinflammatory role in inflammatory arthritis, which is normally restrained by SOCS-3. Periarticular bone loss and focal bone erosions are characteristic features of RA. Chondrocytes have been previously shown to express RANKL in a model of antigen-induced arthritis in rabbits (28). Adenoviral expression of OSM in murine joints induced RANKL staining in the superficial layer of articular cartilage (29). We found that, compared to their littermate controls, bone erosion in Socs3⌬/⌬col2 mice with inflammatory monarthritis and RANKL mRNA expression in Socs3⌬/⌬col2 mouse chondrocytes stimulated with gp130 cytokines were both dramatically increased. Interestingly, both OSM and IL-6/sIL-6R were more potent than IL-1 at increasing the crucial RANKL:OPG mRNA ratio. These findings raise the intriguing possibility that production of RANKL by gp130 cytokine– stimulated chondrocytes may contribute to activation of local osteoclasts, or promote osteoclastogenesis, during arthritis. Such a mechanism could potentially explain the presence of periarticular bone loss and juxtaarticular erosions in RA. Importantly, SOCS-3 may normally limit the production of RANKL by chondrocytes, and this effect could have therapeutic implications. The results of this study show that chondrocytes express receptors for all gp130 cytokines, and that OSM is a particularly potent cytokine for this cell type. OSM has previously been detected in the synovial fluid of RA patients, and the levels of OSM correlated with markers of joint inflammation and cartilage destruction (36). Intraarticular administration of adenovirus expressing OSM caused synovitis and increased IL-6 production
LIU ET AL
(37). Blockade of OSM abrogated joint inflammation and cartilage damage in collagen-induced arthritis (6). We found that OSM-induced STAT signaling was accompanied by increased aggrecan release, aggrecanase induction, cytokine production, and RANKL expression in murine cartilage explants and chondrocytes. Previously, the murine analog of OSM (mOSM) was believed to form only a heterodimeric-signaling complex with OSMR and gp130. However, recent evidence suggests that mOSM can induce differential effects when signaling through the OSMR or the LIFR in osteoblasts and osteocytes (38). In osteocytes, mOSM inhibited transcription of sclerostin, an inhibitor of mineralization, via the LIFR (38). In contrast, when acting through the OSMR, mOSM stimulated osteoclast formation in a RANKL-dependent interaction. In the present study, we compared the effects of OSM and LIF on catabolic responses in chondrocytes. In this system, the possibility that OSM signaling occurs via both the OSMR and LIFR cannot be ruled out, but the effect of OSM was much greater compared to that of LIF, implying that in chondrocytes, OSM predominantly signals via the OSMR. The potent OSM-induced responses observed in mouse primary chondrocytes may also be attributable to the ability of OSM to activate STAT-5, as well as STAT-1 and STAT-3. The induction of the STAT-5 pathway via the OSMR in chondrocytes might therefore induce a unique subset of genes, in addition to those shared between gp130 cytokines. Of course, the potency of OSM compared to other gp130 cytokines may also be a consequence of higher expression of the OSMR on chondrocytes relative to the expression of other gp130 receptors, which could amplify JAK/STAT signaling and drive an enhanced response. Our findings with regard to receptor expression are in keeping with this possibility, although definitive conclusions are limited by possible differences in the biotinylation efficiency and affinity of antireceptor antibodies. In this study, we used a number of strategies to comprehensively examine the contribution of gp130 cytokines and chondrocytes to cartilage metabolism and inflammatory arthritis. We provided in vitro and in vivo evidence that SOCS-3 is a critical regulator of gp130 cytokine signaling in murine chondrocytes, governing both the magnitude and the quality of chondrocyte responses. Chondrocytes can produce an array of catabolic and inflammatory mediators in response to gp130 cytokines, including the major osteoclast activator RANKL, which suggests the intriguing possibility that
SOCS-3 REGULATION OF gp130 CYTOKINES IN MURINE ARTHRITIS
there is potential cross-talk between chondrocytes and osteoclasts during bone remodeling. The response of chondrocytes to IL-6 is primarily through trans-signaling, which may have implications for the clinical use of anti–IL-6R antibody therapies. Within the gp130 family, OSM is likely to be the major mediator of chondrocyte-driven biologic processes, perhaps related to its ability to activate STAT-5, as well as STAT-1 and STAT-3, and to the high-level expression of its receptor. These findings, derived in a mesenchymal cell type, add to previous evidence highlighting the crucial regulatory role of SOCS-3 in hematopoietic cells. Finally, inhibition of OSM may be a particularly effective strategy for the preservation of cartilage, and perhaps bone, in inflammatory arthritis. ACKNOWLEDGMENTS We wish to thank J. E. Visvader for providing the Gt(Rosa26)Sor mice. We also thank C. B. Little, T. L. Putoczki, W. J. Martin, A. van Nieuwenhuijze, E. C. Stuart, and S. M. Chatfield for helpful discussions and technical assistance. AUTHOR CONTRIBUTIONS All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Wicks had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study conception and design. Liu, Croker, Campbell, Alexander, Lawlor, Wicks. Acquisition of data. Liu, Tonkin, Walsh, Linossi. Analysis and interpretation of data. Liu, Croker, Gauci, Alexander, Walsh, Linossi, Nicholson, Lawlor, Wicks.
8.
9.
10.
11.
12. 13. 14. 15.
16. 17.
18.
REFERENCES 1. Jones SA, Scheller J, Rose-John S. Therapeutic strategies for the clinical blockade of IL-6/gp130 signaling. J Clin Invest 2011;121: 3375–83. 2. Kishimoto T. IL-6: from its discovery to clinical applications. Int Immunol 2010;22:347–52. 3. Croker BA, Kiu H, Nicholson SE. SOCS regulation of the JAK/STAT signalling pathway. Semin Cell Dev Biol 2008;19: 414–22. 4. Babon JJ, Nicola NA. The biology and mechanism of action of suppressor of cytokine signaling 3. Growth Factors 2012;30: 207–19. 5. Wong PK, Campbell IK, Robb L, Wicks IP. Endogenous IL-11 is pro-inflammatory in acute methylated bovine serum albumin/ interleukin-1-induced (mBSA/IL-1) arthritis. Cytokine 2005;29: 72–6. 6. Plater-Zyberk C, Buckton J, Thompson S, Spaull J, Zanders E, Papworth J, et al. Amelioration of arthritis in two murine models using antibodies to oncostatin M. Arthritis Rheum 2001;44: 2697–702. 7. Lawlor KE, Wong PK, Campbell IK, van Rooijen N, Wicks IP.
19. 20.
21.
22.
23.
2401
Acute CD4⫹ T lymphocyte–dependent interleukin-1–driven arthritis selectively requires interleukin-2 and interleukin-4, joint macrophages, granulocyte–macrophage colony-stimulating factor, interleukin-6, and leukemia inhibitory factor. Arthritis Rheum 2005;52:3749–54. Upadhyay A, Senyschyn D, Santos L, Gu R, Carroll GJ, Jazayeri JA. K/BxN serum transfer arthritis is delayed and less severe in leukaemia inhibitory factor (LIF)-deficient mice. Clin Exp Immunol 2012;169:71–8. Wong PK, Quinn JM, Sims NA, van Nieuwenhuijze A, Campbell IK, Wicks IP. Interleukin-6 modulates production of T lymphocyte–derived cytokines in antigen-induced arthritis and drives inflammation-induced osteoclastogenesis. Arthritis Rheum 2006;54:158–68. Wong PK, Egan PJ, Croker BA, O’Donnell K, Sims NA, Drake S, et al. SOCS-3 negatively regulates innate and adaptive immune mechanisms in acute IL-1-dependent inflammatory arthritis. J Clin Invest 2006;116:1571–81. Boyle K, Egan P, Rakar S, Willson TA, Wicks IP, Metcalf D, et al. The SOCS box of suppressor of cytokine signaling-3 contributes to the control of G-CSF responsiveness in vivo. Blood 2007;110: 1466–74. Croker BA, Krebs DL, Zhang JG, Wormald S, Willson TA, Stanley EG, et al. SOCS3 negatively regulates IL-6 signaling in vivo. Nat Immunol 2003;4:540–5. Sakai K, Hiripi L, Glumoff V, Brandau O, Eerola R, Vuorio E, et al. Stage- and tissue-specific expression of a Col2a1-Cre fusion gene in transgenic mice. Matrix Biol 2001;19:761–7. Soriano P. Generalized lacZ expression with the ROSA26 Cre reporter strain. Nat Genet 1999;21:70–1. Croker BA, Metcalf D, Robb L, Wei W, Mifsud S, DiRago L, et al. SOCS3 is a critical physiological negative regulator of G-CSF signaling and emergency granulopoiesis. Immunity 2004; 20:153–65. Farndale RW, Sayers CA, Barrett AJ. A direct spectrophotometric microassay for sulfated glycosaminoglycans in cartilage cultures. Connect Tissue Res 1982;9:247–8. Rogerson FM, Chung YM, Deutscher ME, Last K, Fosang AJ. Cytokine-induced increases in ADAMTS-4 messenger RNA expression do not lead to increased aggrecanase activity in ADAMTS-5–deficient mice [published erratum appears in Arthritis Rheum 2011;63:256]. Arthritis Rheum 2010;62:3365–73. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2⫺⌬⌬Ct method. Methods 2001;25:402–8. Lawlor KE, Campbell IK, O’Donnell K, Wu L, Wicks IP. Molecular and cellular mediators of interleukin-1–dependent acute inflammatory arthritis. Arthritis Rheum 2001;44:442–50. Linossi EM, Chandrashekaran IR, Kolesnik TB, Murphy JM, Webb AI, Willson TA, et al. Suppressor of cytokine signaling (SOCS) 5 utilises distinct domains for regulation of JAK1 and interaction with the adaptor protein Shc-1. PLoS One 2013;8: e70536. Savontaus M, Ihanamaki T, Perala M, Metsaranta M, SandbergLall M, Vuorio E. Expression of type II and IX collagen isoforms during normal and pathological cartilage and eye development. Histochem Cell Biol 1998;110:149–59. Kolpakova-Hart E, Nicolae C, Zhou J, Olsen BR. Col2-Cre recombinase is co-expressed with endogenous type II collagen in embryonic renal epithelium and drives development of polycystic kidney disease following inactivation of ciliary genes. Matrix Biol 2008;27:505–12. Fosang AJ, Golub SB, East CJ, Rogerson FM. Abundant LacZ activity in the absence of Cre expression in the normal and inflamed synovium of adult Col2a1-Cre;ROSA26RLacZ reporter mice. Osteoarthritis Cartilage 2013;21:401–4.
2402
24. Hintzen C, Evers C, Lippok BE, Volkmer R, Heinrich PC, Radtke S, et al. Box 2 region of the oncostatin M receptor determines specificity for recruitment of Janus kinases and STAT5 activation. J Biol Chem 2008;283:19465–77. 25. Hyde G, Dover S, Aszodi A, Wallis GA, Boot-Handford RP. Lineage tracing using matrilin-1 gene expression reveals that articular chondrocytes exist as the joint interzone forms. Dev Biol 2007;304:825–33. 26. Koyama E, Shibukawa Y, Nagayama M, Sugito H, Young B, Yuasa T, et al. A distinct cohort of progenitor cells participates in synovial joint and articular cartilage formation during mouse limb skeletogenesis. Dev Biol 2008;316:62–73. 27. Boyce BF, Xing L. Functions of RANKL/RANK/OPG in bone modeling and remodeling. Arch Biochem Biophys 2008;473: 139–46. 28. Martinez-Calatrava MJ, Prieto-Potin I, Roman-Blas JA, Tardio L, Largo R, Herrero-Beaumont G. RANKL synthesized by articular chondrocytes contributes to juxta-articular bone loss in chronic arthritis. Arthritis Res Ther 2012;14:R149. 29. Hui W, Cawston TE, Richards CD, Rowan AD. A model of inflammatory arthritis highlights a role for oncostatin M in proinflammatory cytokine-induced bone destruction via RANK/ RANKL. Arthritis Res Ther 2005;7:R57–64. 30. Durigova M, Roughley PJ, Mort JS. Mechanism of proteoglycan aggregate degradation in cartilage stimulated with oncostatin M. Osteoarthritis Cartilage 2008;16:98–104. 31. Flannery CR, Little CB, Hughes CE, Curtis CL, Caterson B, Jones SA. IL-6 and its soluble receptor augment aggrecanase-mediated proteoglycan catabolism in articular cartilage. Matrix Biol 2000; 19:549–53. 32. Lawlor KE, Campbell IK, Metcalf D, O’Donnell K, van Nieuwenhuijze A, Roberts AW, et al. Critical role for granulocyte colonystimulating factor in inflammatory arthritis. Proc Natl Acad Sci U S A 2004;101:11398–403.
LIU ET AL
33. Tomchuck SL, Henkle SL, Coffelt SB, Betancourt AM. Toll-like receptor 3 and suppressor of cytokine signaling proteins regulate CXCR4 and CXCR7 expression in bone marrow-derived human multipotent stromal cells. PLoS One 2012;7:e39592. 34. Frobose H, Ronn SG, Heding PE, Mendoza H, Cohen P, Mandrup-Poulsen T, et al. Suppressor of cytokine signaling-3 inhibits interleukin-1 signaling by targeting the TRAF-6/TAK1 complex. Mol Endocrinol 2006;20:1587–96. 35. Smeets RL, Veenbergen S, Arntz OJ, Bennink MB, Joosten LA, van den Berg WB, et al. A novel role for suppressor of cytokine signaling 3 in cartilage destruction via induction of chondrocyte desensitization toward insulin-like growth factor. Arthritis Rheum 2006;54:1518–28. 36. Manicourt DH, Poilvache P, van Egeren A, Devogelaer JP, Lenz ME, Thonar EJ. Synovial fluid levels of tumor necrosis factor ␣ and oncostatin M correlate with levels of markers of the degradation of crosslinked collagen and cartilage aggrecan in rheumatoid arthritis but not in osteoarthritis. Arthritis Rheum 2000;43:281–8. 37. Langdon C, Kerr C, Hassen M, Hara T, Arsenault AL, Richards CD. Murine oncostatin M stimulates mouse synovial fibroblasts in vitro and induces inflammation and destruction in mouse joints in vivo. Am J Pathol 2000;157:1187–96. 38. Walker EC, McGregor NE, Poulton IJ, Solano M, Pompolo S, Fernandes TJ, et al. Oncostatin M promotes bone formation independently of resorption when signaling through leukemia inhibitory factor receptor in mice. J Clin Invest 2010;120:582–92. 39. Yasukawa H, Ohishi M, Mori H, Murakami M, Chinen T, Aki D, et al. IL-6 induces an anti-inflammatory response in the absence of SOCS3 in macrophages. Nat Immunol 2003;4:551–6. 40. Sommer U, Schmid C, Sobota RM, Lehmann U, Stevenson NJ, Johnston JA, et al. Mechanisms of SOCS3 phosphorylation upon interleukin-6 stimulation: contributions of Src- and receptortyrosine kinases. J Biol Chem 2005;280:31478–88.
