Involvement of the Wnt signaling pathway in experimental and human osteoarthritisProminent role of Wnt-induced signaling protein 1.код для вставкиСкачать
ARTHRITIS & RHEUMATISM Vol. 60, No. 2, February 2009, pp 501–512 DOI 10.1002/art.24247 © 2009, American College of Rheumatology Involvement of the Wnt Signaling Pathway in Experimental and Human Osteoarthritis Prominent Role of Wnt-Induced Signaling Protein 1 Arjen B. Blom,1 Sarah M. Brockbank,2 Peter L. van Lent,1 Henk M. van Beuningen,1 Jeroen Geurts,1 Nozomi Takahashi,1 Peter M. van der Kraan,1 Fons A. van de Loo,1 B. Wim Schreurs,1 Kristen Clements,2 Peter Newham,2 and Wim B. van den Berg1 Objective. Wnt signaling pathway proteins are involved in embryonic development of cartilage and bone, and, interestingly, developmental processes appear to be recapitulated in osteoarthritic (OA) cartilage. The present study was undertaken to characterize the expression pattern of Wnt and Fz genes during experimental OA and to determine the function of selected genes in experimental and human OA. Methods. Longitudinal expression analysis was performed in 2 models of OA. Levels of messenger RNA for genes from the Wnt/␤-catenin pathway were determined in synovium and cartilage, and the results were validated using immunohistochemistry. Effects of selected genes were assessed in vitro using recombinant protein, and in vivo by adenoviral overexpression. Results. Wnt-induced signaling protein 1 (WISP-1) expression was strongly increased in the synovium and cartilage of mice with experimental OA. Wnt-16 and Wnt-2B were also markedly up-regulated during the course of disease. Interestingly, increased WISP-1 expression was also found in human OA cartilage and synovium. Stimulation of macrophages and chondrocytes with recombinant WISP-1 resulted in interleukin-1–independent induction of several matrix metalloproteinases (MMPs) and aggrecanase. Adenoviral overexpression of WISP-1 in murine knee joints induced MMP and aggrecanase expression and resulted in cartilage damage. Conclusion. This study included a comprehensive characterization of Wnt and Frizzled gene expression in experimental and human OA articular joint tissue. The data demonstrate, for the first time, that WISP-1 expression is a feature of experimental and human OA and that WISP-1 regulates chondrocyte and macrophage MMP and aggrecanase expression and is capable of inducing articular cartilage damage in models of OA. Supported in part by ZonMW (program Alternatieven voor dierproeven, project no. 11-400-0082), the Dutch Arthritis Association (grant NR 08-1-309), and AstraZeneca. Dr. Takahashi’s work was supported by IOP Genomics (grant IGE02032). 1 Arjen B. Blom, PhD, Peter L. van Lent, PhD, Henk M. van Beuningen, PhD, Jeroen Geurts, MSc, Nozomi Takahashi, PhD, Peter M. van der Kraan, PhD, Fons A. van de Loo, PhD, B. Wim Schreurs, PhD, Wim B. van den Berg, PhD: Radboud University Nijmegen Medical Center, Nijmegen, The Netherlands; 2Sarah M. Brockbank, PhD, Kristen Clements, PhD, Peter Newham, PhD: AstraZeneca, Macclesfield, UK. Drs. Brockbank, Clements, and Newham own stock or stock options in AstraZeneca. Address correspondence and reprint requests to Arjen B. Blom, PhD, Experimental Rheumatology and Advanced Therapeutics, Radboud University Nijmegen Medical Center, Geert Grooteplein 26-28, 6525 GA Nijmegen, The Netherlands. E-mail: a.blom@reuma. umcn.nl. Submitted for publication May 14, 2008; accepted in revised form October 13, 2008. Osteoarthritis (OA) results in the destruction of cartilage and bone, ultimately leading to loss of joint function. The cause of the disease is largely unknown, although obesity, genetic factors, and injury have all been associated with increased risk of OA (1,2). Although it is likely that in most cases the initial events leading to OA occur within the cartilage or subchondral bone (3,4), the synovial tissue of many OA patients shows a changed morphology, with a marked inflammatory phenotype (5,6). Loss of articular cartilage extracellular matrix is thought to be mediated by matrix metalloproteinases (MMPs) and aggrecanases (ADAMTS-4 and ADAMTS-5), which are likely the most important groups of enzymes in the breakdown of extracellular 501 502 cartilage matrix (7). Specifically, MMP-3 (8), MMP-13 (9,10), ADAMTS-4, and ADAMTS-5 (11,12) have been implicated in OA pathology. The inflammatory cytokine interleukin-1 (IL-1) potently up-regulates MMP expression in synovial cells and chondrocytes (13,14) and may drive cartilage damage in OA; however, the role of IL-1 in disease pathology remains unclear (13). Recently, it has been hypothesized that OA entails a recapitulation of limb development, a process characterized by chondrocyte hypertrophy and matrix calcification (15,16). Interestingly, MMP-13 is an early marker of chondrocyte terminal differentiation, an event that is of key importance in limb development (17). Wnt proteins form a family of secreted glycoproteins that bind to Frizzled (Fz) receptors and regulate embryonic development, body patterning, and tissue morphogenesis. Moreover, several Wnt family members, such as Wnt-4, Wnt-14, and Wnt-16, have been shown to be crucial for development of avian and murine limbs, by controlling osteogenesis and chondrogenesis (18–20). Wnt signaling also induces both MMP expression and cell differentiation, and thus directly regulates tissue remodeling and also regulates chondrocyte phenotype, a process critical for bone growth and articular cartilage formation. Recent genetic data suggest that abnormal Wnt signaling may contribute to the pathogenesis of OA: an association between the development of hip OA and a polymorphism in the FrzB gene has been demonstrated (21–23). FrzB appears to function as a soluble Fz receptor and thus can inhibit Wnt/Fz signaling. Interestingly, the FrzB mutation associated with OA exhibits decreased affinity for Wnt molecules, suggesting a compromised ability to suppress Wnt signaling. In addition, functional deletion of FrzB in a murine model of OA results in exacerbation of disease (24). These data, together with the association of a polymorphism in a Wnt coreceptor, low-density lipoprotein receptor– related protein 5, with OA (25), strongly implicate the involvement of the Wnt signaling pathway in the etiology or progression of OA. To explore the role of the Wnt/Fz axis in OA, we profiled a panel of genes from the Wnt signaling pathway by longitudinal expression analysis in 2 separate experimental models of OA. Furthermore, we analyzed the expression of key genes in both experimental and human OA by quantitative polymerase chain reaction (PCR) and immunohistochemistry, and assayed the potential of selected gene(s) to induce cartilage breakdown in vitro and in vivo. BLOM ET AL MATERIALS AND METHODS Animals and models of OA. For all experiments with induced OA, C57BL/6 mice (Janvier, Le Genest St. Isle, France) were used. Spontaneous OA was studied in the STR/Ort strain (originally obtained from Charles River [Sulzfeld, Germany] and bred at our facilities). CBA mice (Charles River) were used as controls for the STR/Ort strain. IL-1␣/␤– deficient mice were originally kindly provided by Dr. Y. Iwakura (Tokyo, Japan) and were bred at our facilities. Mice were fed a standard diet and tap water ad libitum. All animal experiments were approved by the local ethics commission. Instability OA was induced by intraarticular injection of 1 unit of collagenase on day 0 and day 2. OA pathology develops gradually through day 42, at which time full-blown OA is manifested. Knee joints were isolated on day 7, day 21, and day 42 and processed for histologic analysis. Synovium and cartilage samples were also obtained and processed for PCR analysis. OA develops spontaneously in STR/Ort mice. The first pathologic changes are observed within 8 weeks after birth. Tissue from STR/Ort and CBA mice was isolated at ages 4 weeks, 8 weeks, and 16 weeks. Histologic analysis of murine knee joints. Isolated knee joints were processed for histologic and immunohistochemistry analysis. Tissue was fixed in 4% buffered formalin and then decalcified in formic acid and embedded in paraffin. Six representative sections of each joint from various depths were mounted on slides. Standard histologic analyses with Safranin O and hematoxylin and eosin staining were conducted, and studies to detect Wnt-induced signaling protein 1 (WISP-1) and ␤-catenin were performed. Briefly, sections were deparaffinized and subsequently incubated in buffered citrate (pH 6.0). Sections were incubated overnight with either polyclonal rabbit anti–WISP-1 or goat anti–␤-catenin (Santa Cruz Biotechnology, Santa Cruz, CA) at a concentration of 2 g/ml. Thereafter, sections were incubated with biotinylated polyclonal anti-rabbit IgG and biotinylated ant-goat IgG (Vectastain Elite kit; Vector, Burlingame, CA), respectively. For staining of the neoepitope VDIPEN the procedure was essentially the same as for WISP-1, but citrate buffer was replaced with chondroitinase ABC and staining was enhanced using a nickel-enhancement step and orange G counterstaining; antiVDIPEN was a kind gift from Dr. J. S. Mort (Montreal, Quebec, Canada). Staining for the neoepitope NITEGE using an anti-NITEGE antibody (Acris, Hiddenhausen, Germany) was performed by a process similar to that used for ␤-catenin staining, with the addition of a 30-minute hyaluronidase treatment after the treatment with citrate buffer. Human cartilage and synovium. Permission of patients and the local ethics commission was obtained prior to harvesting of study samples. Human OA synovium and cartilage were obtained from patients undergoing total knee or hip joint replacement surgery. Control synovium from patients with acute knee joint trauma was obtained at the time of arthroscopic examination. Control cartilage was obtained postmortem from subjects with no history of OA. Human synovial fibroblasts were isolated by enzymatic digestion of synovial tissue. Synovial tissue was placed in cell culture medium at ambient temperature and subjected to enzymatic digestion within 2 hours. Samples were finely WISP-1 IN EXPERIMENTAL AND HUMAN OA 503 Table 1. Expression patterns of Wnt signaling pathway genes in synovium and cartilage during collagenase-induced OA, and during spontaneous OA in STR/Ort mice* Synovium Collagenase-induced OA Cartilage OA in STR/Ort mice Collagenase-induced OA OA in STR/Ort mice Gene Week 1 Week 3 Week 6 Week 4 Week 8 Week 16 Week 1 Week 3 Week 6 Week 4 Week 8 Week 16 Axin2 Dkk-3 Dishevelled 1 FrzB Indian hedgehog LRP-5 LRP-6 sFRP-1 sFRP-2 Siah-2 WISP-1 Fz-3 Fz-4 Fz-5 Fz-6 Fz-7 Fz-8 Fz-10 Wnt-2 Wnt-2B Wnt-3 Wnt-3A Wnt-4 Wnt-5A Wnt-5B Wnt-6 Wnt-7A Wnt-7B Wnt-8A Wnt-8B Wnt-9A Wnt-9B Wnt-10A Wnt-10B Wnt-11 Wnt-16 ⫽ 1 13 2 2.0 ⫽ – ⫽ ⫽ 1 14 1 6.5 ⫽ 1 37 ⫽ ⫽ 1 2.1 1 3.5 2 2.3 1 2.3 ⫽ ⫽ 1 82 – – 2 2.0 ⫽ 1 2.5 1 4.0 – – – – ⫽ – ⫽ ⫽ ⫽ 1 111 ⫽ 1 8.7 ⫽ ⫽ – ⫽ ⫽ 1 3.2 1 4.9 ⫽ 1 15 ⫽ ⫽ ⫽ ⫽ ⫽ 1 2.0 ⫽ ⫽ 1 8.6 – – ⫽ ⫽ ⫽ ⫽ – – – – ⫽ – ⫽ ⫽ ⫽ 1 181 ⫽ ⫽ ⫽ ⫽ – ⫽ ⫽ 1 2.7 1 2.4 ⫽ 1 9.1 ⫽ ⫽ ⫽ ⫽ ⫽ 1 1.2 ⫽ ⫽ ⫽ – – ⫽ ⫽ ⫽ ⫽ – – – – ⫽ – ⫽ ⫽ ⫽ 1 104 ⫽ ⫽ 2 1.5 2 1.9 ⫽ ⫽ ⫽ ⫽ ⫽ ⫽ ⫽ ⫽ 2 1.6 2 2.2 ⫽ ⫽ ⫽ ⫽ ⫽ ⫽ ⫽ – ⫽ ⫽ ⫽ 2 1.7 ⫽ 1 4.3 ⫽ – ⫽ 1 12 ⫽ ⫽ 1 1.4 ⫽ 2 1.7 ⫽ ⫽ 2 2.4 2 3.2 ⫽ ⫽ ⫽ ⫽ ⫽ ⫽ ⫽ ⫽ 2 1.3 ⫽ ⫽ ⫽ ⫽ ⫽ 1 1.5 1 3.4 – ⫽ ⫽ ⫽ ⫽ ⫽ 1 7.8 ⫽ – ⫽ ⫽ ⫽ ⫽ 1 1.4 ⫽ 2 1.5 1 2.0 2 1.2 2 2.2 ⫽ ⫽ ⫽ ⫽ 1 1.8 ⫽ 1 2.6 ⫽ ⫽ ⫽ 1 1.8 ⫽ ⫽ 2 4.1 ⫽ 1 2.8 1 4.2 – ⫽ ⫽ ⫽ ⫽ ⫽ ⫽ ⫽ – ⫽ ⫽ ⫽ ⫽ 1 1.5 1 5.6 ⫽ ⫽ ⫽ 2 4.3 ⫽ ⫽ ⫽ 1 4.2 ⫽ ⫽ 1 2.5 ⫽ ⫽ 2 2.5 1 2.0 2 2.0 2 2.1 ⫽ – ⫽ – – ⫽ ⫽ – ⫽ – ⫽ – – ⫽ – ⫽ ⫽ ⫽ ⫽ ⫽ ⫽ ⫽ ⫽ ⫽ ⫽ ⫽ 1 2.1 ⫽ ⫽ 1 2.8 ⫽ ⫽ ⫽ ⫽ ⫽ ⫽ ⫽ – ⫽ – – ⫽ ⫽ – ⫽ – ⫽ – – ⫽ – ⫽ ⫽ ⫽ ⫽ ⫽ 1 2.8 ⫽ ⫽ ⫽ ⫽ ⫽ ⫽ ⫽ ⫽ ⫽ ⫽ ⫽ ⫽ ⫽ ⫽ ⫽ ⫽ – ⫽ – – ⫽ ⫽ – ⫽ – ⫽ – – ⫽ – ⫽ ⫽ ⫽ ⫽ ⫽ ⫽ ⫽ ⫽ ⫽ ⫽ ⫽ ⫽ ⫽ ⫽ ⫽ ⫽ ⫽ ⫽ ⫽ ⫽ ⫽ ⫽ – ⫽ – – ⫽ ⫽ ⫽ ⫽ – ⫽ – – ⫽ – ⫽ ⫽ ⫽ ⫽ ⫽ ⫽ 2 1.9 ⫽ ⫽ ⫽ ⫽ ⫽ ⫽ ⫽ ⫽ ⫽ ⫽ ⫽ 1 1.7 ⫽ ⫽ 2 3.7 – ⫽ – – ⫽ ⫽ ⫽ ⫽ – ⫽ – – ⫽ – 2 2.2 ⫽ ⫽ ⫽ ⫽ 1 1.9 ⫽ ⫽ ⫽ ⫽ 1 2.3 ⫽ ⫽ ⫽ 1 1.8 ⫽ 1 2.3 ⫽ 11.9 ⫽ ⫽ ⫽ – ⫽ – – ⫽ ⫽ ⫽ 1 1.7 – ⫽ – – 1 3.4 – ⫽ 1 1.8 ⫽ 1 2.1 * Values are the fold increase (1) or decrease (2) in mice with collagenase-induced osteoarthritis (OA) compared with naive mice, and in STR/Ort mice (in which OA spontaneously develops) compared with CBA mice. Equal signs represent genes that were not up- or down-regulated; dashes represent genes that were not detected. minced and digested for 30 minutes at 37oC in 10 ml phosphate buffered saline containing 0.1% trypsin (Sigma, St. Louis, MO). Subsequently, samples were digested in 20 ml 0.1% collagenase P (Roche Diagnostics, Indianapolis, IN) in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal calf serum (FCS; Gibco, Breda, The Netherlands), 1 mM pyruvate, and 40 g/ml gentamicin, for 2 hours at 37°C in 5% CO2. The cell suspension was filtered over a sterile 0.7-m filter (BD Falcon, Bedford, MA). Fibroblasts were purified by negative selection using CD14 magnetic-activated cell sorting (26) and maintained in DMEM supplemented with 10% FCS, 1 mM pyruvate, and 80 g/ml gentamicin. Cells from passages 3–6 were used for experiments. Human articular chondrocytes were obtained by overnight incubation of 1-mm2 cartilage slices with 2 mg/ml colla- genase P in DMEM supplemented with 10% FCS and 40 g/ml gentamicin at 37oC. After resuspension and filtering over a 0.7-m filter, cells were cultured in a 24-well plate at a concentration of 2 ⫻ 105 cells/ml. TaqMan quantitative PCR analysis using low-density arrays and conventional quantitative PCR. To determine whether expression of genes was altered during experimental OA, quantitative PCR analysis was performed, either by a microfluid technique using low-density TaqMan arrays (ABI Prism low-density arrays) (LDA; Applied Biosystems, Foster City, CA) or by regular TaqMan analysis. A bespoke LDA of 48 genes was used to perform quantitative PCR analysis. Marker genes ADAMTS4, BMP4, COL10A1, COL1A2, COL2A1, IL1B, IL6, IL18, MMP13, and TGFB3 were added to confirm whether the kinetics of the experiment were 504 BLOM ET AL representative of the OA models; 18S and HPRT were used for normalization. Results were expressed as fold change, by comparing OA knee joints with normal joints after correction for housekeeping genes, using the threshold cycle (Ct) method and the 2⌬⌬Ct formula. Comparison of the regulation of the above-mentioned marker genes indicated that in the current experiments, regulations were representative of those observed during collagenase-induced OA and OA in STR/Ort mice in previous experiments (data not shown). Stimulation of macrophages and chondrocytes by recombinant WISP-1. RAW 264.7 cells (as a model for synovial macrophages; American Type Culture Collection, Rockville, MD), H4 chondrocytes (27), and freshly isolated human cells (fibroblasts, chondrocytes) were tested by in vitro stimulation with recombinant human WISP-1 (PeproTech, Rocky Hill, NJ). After stimulation of cells for 24 hours with WISP-1 at 10, 50, 100, 500, or 1,000 ng/ml, RNA was isolated and used for quantitative PCR analysis. In some cultures polymyxin B sulfate (Sigma, St. Louis, MO) was added to the medium at a concentration of 10 g/ml. Macrophages, human fibroblasts, and human chondrocytes were cultured in DMEM with Glutamax, 10% FCS, and gentamicin. The murine chondrocyte cell line H4 was generated in our laboratory and cultured in DMEM/Ham’s F-12 with 10% FCS and gentamicin. Generation of adenoviral WISP-1 vector. An adenoviral vector to induce overexpression of WISP-1 was generated using a previously described method (28). Briefly, the coding sequence of WISP-1 was cloned with the primers 5⬘TTTTGTCGACACCGCCATGAGGTGGCTCCTGCCC-3⬘ (forward) and 5⬘-TTTTCTAGACTAATTGGCAATCTCTTC-3⬘ (reverse). These primers contained restriction sites for cloning into an adenoviral vector. WISP-1 was cloned from murine OA synovial tissue complementary DNA. Viral vectors were E1A, B, and E3 deleted, and produced according to the method described by Chartier et al (29). For in vivo overexpression, 107 plaque-forming units of the WISP-1 adenovirus or of an empty control vector (Ad5del70-3) was injected intraarticularly, and total knee joints, synovial tissue, and cartilage were isolated after 2 days or 7 days for analysis. RESULTS Up-regulation of ␤-catenin in experimental OA. To investigate whether Wnt signaling might be relevant in experimental OA, we tested for the accumulation of ␤-catenin, as a reporter of Wnt/Fz signaling in the knee joints of naive and osteoarthritic mice, using immunohistochemical staining (results available online at http:// www.rheumaresearch.nl/BlomAnRFigureS1.pdf). Staining for ␤-catenin in naive mouse cartilage was present and was restricted to the deeper cartilage chondrocyte layers. However, 21 days after intraarticular collagenase injection, ␤-catenin staining intensity was increased, and greater numbers of superficial chondrocytes were positive. Staining for ␤-catenin was widespread in the synovium of naive mice, but was strongly increased 21 days after induction of OA. Regulation of several Wnt- and Frizzled-related genes during experimental OA. Since intracellular ␤-catenin accumulation could represent evidence of Wnt-dependent signaling, we used an LDA method to determine the kinetics of Wnt/Fz family gene expression in 2 murine models of OA: intraarticular collagenase– induced OA and spontaneous OA in STR/Ort mice. Gene expression was determined in synovium and cartilage of naive knee joints and compared with expression in tissue from knee joints 1, 3, and 6 weeks after induction of OA by administration of collagenase. Similarly, the expression of Wnt family genes was measured in the cartilage and synovium of 4-, 8-, and 16-week-old STR/Ort mice. In the collagenase-induced OA model we found that many Wnt/Fz family genes were up- or down-regulated in OA synovium compared with synovium from naive mice (Table 1). However, none of the Wnt genes was regulated at the messenger RNA (mRNA) level in the cartilage in this model. Expression of WISP-1 mRNA is shown in Figures 1A and B. WISP-1 was up-regulated 37-fold in the synovium in the collagenase-induced OA model after 1 week, and 9-fold after 6 weeks. In the cartilage, maximal up-regulation of WISP-1 was found on day 21 (3-fold). In the STR/Ort mouse model of spontaneous OA, Wnt genes were similarly up-regulated in the synovium, although the magnitude of gene expression change was smaller than in the collagenase-induced OA model (Table 1). In contrast with the findings in the cartilage of mice with collagenase-induced OA, in STR/ Ort mouse cartilage some of the Wnt genes were regulated as well. Wnt-16 expression in synovium was up-regulated 181-fold and 5.6-fold in mice with collagenase-induced OA and STR/Ort mice, respectively, and Wnt-2B 82-fold and 2.8-fold, respectively (Table 1). Consistent with findings in the collagenase-induced OA model, WISP-1 was found to be significantly upregulated (⬃2-fold) in both the synovium and the cartilage of the STR/Ort mice at age 16 weeks. Increased levels of WISP-1 protein in murine OA knee joints. Our data indicated that a panel of Wnt/Fz genes is regulated during the progression of experimental OA; thus, Wnt/Fz signaling via ␤-catenin may play a role in OA pathology. We also found that a small subset of genes (WISP-1, Dkk-3, and Fz-6) were significantly up-regulated in both the synovium and the cartilage in both experimental models of OA; of these, WISP-1 was the most highly regulated gene (Figure 1; further results available online at www.rheumaresearch.nl/ BlomAnRFigureS2.pdf). WISP-1, a known Wnt/␤catenin–regulated gene, is a member of the CCN family WISP-1 IN EXPERIMENTAL AND HUMAN OA 505 Figure 1. Expression of Wnt-induced signaling protein 1 (WISP-1) in synovium and cartilage in 2 separate models of osteoarthritis (OA) and in humans with OA. A and B, WISP-1 expression in synovium (A) and cartilage (B) of mice with collagenase-induced OA. Strong up-regulation of WISP-1 was found in the synovium throughout the course of disease. WISP-1 was also significantly up-regulated in cartilage on day 7 and day 21. C and D, WISP-1 expression in synovium (C) and cartilage (D) of STR/Ort mice compared with CBA controls. WISP-1 was up-regulated in both the synovium and the cartilage of 16-week-old mice. Threshold cycle (Ct) was determined, and results in A–D are expressed as the ⌬⌬Ct (OA joint – naive joint after correction for housekeeping gene). Each point represents the expression in pooled samples of synovium or cartilage from 4 mice; bars show the mean and 95% confidence interval. E, Relative expression (2⌬Ct) of WISP-1 mRNA in synovial biopsy specimens from 5 patients with OA and 5 patients with trauma (controls). WISP-1 expression was compared by microarray analysis, and was found to be significantly higher in the OA patients. Values are the mean and SEM. F, Relative expression (2⌬Ct) of WISP-1 mRNA in cartilage from 5 patients with OA and 5 postmortem control specimens. Bars show the means. of growth factors; the biologic properties of CCN proteins include stimulation of cell proliferation, migration, adhesion, and extracellular matrix formation. We therefore focused on the role of WISP-1 as a prototypical Wnt/␤-catenin target gene in experimental and human OA. To confirm that the increased expression of WISP-1 mRNA in synovium and cartilage had consequences regarding expression at the protein level, we assessed the presence of WISP-1 protein in knee joints of mice after the onset of collagenase-induced OA, using immunohistochemical methods. Synovial WISP-1 staining increased strongly over time, both in the synovial lining and in the subsynovium (Figures 2A, C, and E). At 2 different locations in the joint, i.e., near the patellofemoral junction and near the tibiofemoral junction, this expression of WISP-1 increased rapidly up to day 7 after induction of OA, and slowly subsided thereafter. WISP-1 protein expression in the cartilage after induction of OA was scored and was found to increase up to day 21; thereafter, expression diminished in conjunction with the loss of chondrocytes from the damaged cartilage (Figures 2B, D, and F). WISP-1 expression was similar in all layers of cartilage, and this even distribution did not change notably during the development of OA. Increased WISP-1 mRNA levels in human OA tissue. We next examined WISP-1 mRNA expression in 5 human OA synovial tissue specimens and compared this with expression in synovial tissue from 5 patients with acute trauma (controls). Expression of WISP-1 in human OA synovium was 2.7-fold higher than in control synovium (P ⬍ 0.01) (Figure 1E). In addition, expression of WISP-1 in the cartilage of OA patients (n ⫽ 12) was compared with that in control cartilage (n ⫽ 11). WISP-1 expression was ⬎2-fold higher in OA cartilage compared with control cartilage (Figure 1F). Immunohistochemical detection of WISP-1 in human OA cartilage. In order to determine whether WISP-1 protein levels were also increased in OA cartilage, immunohistochemical staining was performed. Cartilage samples from 3 OA patients, obtained during joint replacement surgery, were compared with 3 control cartilage samples. WISP-1 was undetectable in 2 control 506 BLOM ET AL Figure 2. Immunohistochemical detection of WISP-1 in synovium (A, C, and E) and cartilage (B, D, and F) of mice with collagenase-induced OA. Synovium from naive mice (A) stained less strongly than synovium from mice 42 days after OA induction (C) (arrows). JS ⫽ joint space. Calculation of the percentage of WISP-1–positive cells in the synovium of mice with collagenase-induced OA (E) showed that in both the synovial lining and the subsynovium, WISP-1 expression increased rapidly after OA induction, and slowly subsided after day 7. Expression of WISP-1 was relatively low in cartilage from naive mice (B), but increased until day 21 in cartilage from mice with collagenase-induced OA (D). Scoring of WISP-1 staining in the cartilage of mice with collagenase-induced OA (F) showed that by day 42, staining had decreased to below pretreatment levels, due to the loss of cellularity of cartilage. A–D show representative results (original magnification ⫻ 200). Values in E and F are the mean ⫾ SEM. See Figure 1 for other definitions. Figure 3. Results of immunohistochemistry analysis for WISP-1 in knee cartilage obtained postmortem from 3 donors with no history of musculoskeletal disorders (A–C) and in cartilage obtained during total knee joint replacement in 3 patients with OA (D–F). OA cartilage showed significantly higher expression of WISP-1 compared with control cartilage. Insets in D and F demonstrate strong staining for WISP-1 in chondrocyte clusters. (Original magnification ⫻ 200; ⫻ 400 in insets.) See Figure 1 for definitions. WISP-1 IN EXPERIMENTAL AND HUMAN OA 507 Figure 4. In vitro effect of human Wnt-induced signaling protein 1 (WISP-1) on murine macrophage and chondrocyte cell lines and on human synovial cells and chondrocytes. A and B, Reactivity with WISP-1 was assayed by stimulation of the macrophage cell line RAW 264.7 (A) and the murine chondrocyte cell line H4 (B) with human WISP-1 protein for 24 hours. In macrophages, WISP-1 stimulation resulted in an up-regulation of matrix metalloproteinases (MMPs) except for MMP-2, which was not expressed. This up-regulation appeared to be independent of interleukin-1 (IL-1), since up-regulation of that cytokine was minimal. In chondrocytes, MMP-3 was up-regulated by WISP-1 in a dose-dependent manner. C, Human primary synovial cells and chondrocytes were stimulated with 1.0 g/ml WISP-1. After WISP-1 treatment, levels of MMP-3 and MMP-9 were increased in both cell types. Results in A–C are expressed as the mean fold change compared with unstimulated cells. specimens (Figures 3B and C), whereas 1 control specimen exhibited very low levels of WISP-1 staining (Figure 3A). The tissue specimens obtained from all 3 OA patients demonstrated strong WISP-1 staining, with especially strong staining in the midzone of the cartilage and in chondrocyte clusters (Figures 3D–F). Staining was localized on the cell surface or in the pericellular space adjacent to the extracellular matrix. Up-regulation of matrix-degrading enzymes by WISP-1 on murine and human cells. To determine whether the increased expression of WISP-1 in both the synovium and the cartilage could result in cellular reactions consistent with OA pathology, we studied the effect of WISP-1 on murine and human cells. First, murine RAW 264.7 macrophages were stimulated with recombinant WISP-1 for 24 hours and the gene expression of a panel of catabolic enzymes (MMPs/ADAMTS genes) was monitored. WISP-1 caused a dose-dependent increase in expression of mRNA for ADAMTS-4, MMP-3, MMP-9, and MMP-13, but not ADAMTS-5 or MMP-2 (Figure 4A and data not shown). Since several MMP genes are responsive to IL-1␤, we also assayed IL-1␤ expression in response to WISP-1. The results showed that IL-1␤ expression was not up-regulated by WISP-1. Stimulation of murine H4 chondrocytes with WISP-1 resulted in a marked up-regulation of MMP-3 mRNA, whereas other MMPs were unaffected (Figure 4B). We next investigated the effect of WISP-1 on primary human cells (human synovial cell cultures [passage 6] and human chondrocytes [passage 1]). Synovial cells responded to WISP-1 (after 24-hour stimulation), with MMP-3 and MMP-9 expression increased 2-fold and 3-fold, respectively. Similarly, WISP-1 stimulation increased chondrocyte MMP-3 and MMP-9 levels by 5-fold and 2-fold, respectively (Figure 4C). No signifi- 508 BLOM ET AL Figure 5. In vivo effect of adenoviral overexpression of WISP-1 in naive joints. A and B, Quantitative polymerase chain reaction findings in synovium (A) and cartilage (B) from knee joints of mice that were transfected with adenoviral WISP-1. Results are presented as the mean fold change in the quantity of mRNA compared with that in knee joints from mice transfected with control virus, after correction for GAPDH expression. MMP expression was up-regulated in synovium and cartilage. C–F, Immunohistochemical analysis of the expression of NITEGE (C and D) and VDIPEN (E and F) in the knee joints of mice that were transfected with adenoviral WISP-1 (D and F) or control virus (C and E). G and H, NITEGE staining in the knee joint of a wild-type mouse (G) and an IL-1␣/␤⫺/⫺ mouse (H) following adenoviral overexpression of WISP-1. C–H show representative results (magnification ⫻ 200). See Figure 4 for definitions. WISP-1 IN EXPERIMENTAL AND HUMAN OA cant effect on MMP-2 or MMP-13 was detected. Coincubation of cells with WISP-1 and polymyxin B (as a control for lipopolysaccharide [LPS] contamination) did not significantly affect WISP-1–induced changes (data not shown). Induction of cartilage damage by adenoviral overexpression of WISP-1 in murine knee joints. Since we observed that WISP-1 was expressed in synovial and cartilage tissue in rodent OA models and in human OA and that WISP-1 was capable of inducing MMP expression, we explored the potential of WISP-1 to induce cartilage damage, by overexpressing WISP-1 in knee joints of naive mice. WISP-1 adenovirus or control adenovirus was delivered to the knee joints of naive mice. Two days after inoculation, WISP-1 was strongly up-regulated (700-fold versus control) in the synovium of WISP-1 adenovirus–infected mice (Figure 5A). In accordance with the in vitro findings, we observed increased expression of MMP-3 in the synovium (Figure 5A) and cartilage (Figure 5B) of WISP-1 adenovirus–treated mice (15-fold and 110-fold, respectively); stimulation of MMP-13 was also detected in both synovium and cartilage (15-fold induction in cartilage). Enhanced MMP-9 expression was found to be restricted to the synovium. Expression of ADAMTS-4 and, to a lesser extent, ADAMTS-5 was also increased in the synovium of WISP-1 adenovirus–inoculated animals (Figures 5A and B). Similar increases were not observed in mice inoculated with control adenovirus (data not shown). In addition to analyzing the expression of MMP and ADAMTS mRNA in the joint compartment, we examined the histologic features of the joints after WISP-1 adenovirus infection (Figures 5C–F). Some synovial changes were induced by WISP-1 overexpression, such as focal loss of cells and the presence of cells in the synovial cavity. However, no strong inflammatory response was observed. Markers for enzymatic cartilage damage were markedly increased in the cartilage of WISP-1–expressing mice only. Notably, staining for both the NITEGE and the VDIPEN neoepitopes, indicative of aggrecanase and MMP activity, respectively, was observed pericellularly and in the extracellular matrix (VDIPEN) following WISP-1 adenovirus injection (Figures 5D and F). NITEGE and VDIPEN neoepitope staining was absent in control adenovirus–treated mice (Figures 5C and E). Evidence that the effects of WISP-1 are not mediated by IL-1. Although WISP-1 did not appear to induce IL-1␤ release in in vitro cultures, we did observe some differences in MMP expression profiles following 509 in vivo overexpression of WISP-1. Since IL-1␤ is strongly associated with expression of proinflammatory and matrix-degrading genes, we explored the effects of WISP-1 overexpression in IL-1␣/␤–deficient mice. With the exception of MMP-13, which was slightly decreased (1.8-fold) in IL-1␣/␤–null mice, there were no significant differences in synovial or cartilage MMP or ADAMTS expression. In accordance with these observations, no reduction of cartilage damage, proteoglycan depletion as assessed by Safranin O staining, or NITEGE or VDIPEN neoepitope expression was observed following intraarticular injection of WISP-1 adenovirus into IL-1␣/ ␤–deficient mice (Figures 5G and H). DISCUSSION In a comprehensive study of Wnt/Fz gene expression in 2 experimental models of OA, we found that 1) a panel of Wnt and Wnt signaling–related genes, including WISP-1 (a Wnt-responsive gene), was up-regulated; 2) WISP-1 mRNA expression was significantly upregulated in murine OA and human OA articular joint tissue and this expression was mirrored at the histologic level; 3) recombinant WISP-1 induced the production of matrix-degrading enzymes in murine and human synovial cells and human chondrocytes; and 4) overexpression of WISP-1 in the articular joints of mice in vivo resulted in the induction of matrix-degrading enzyme expression and cartilage extracellular matrix damage, independently of IL-1␣ or IL-1␤. Although some alternative factors have been reported to induce ␤-catenin activation (30), its accumulation and translocation to the nucleus is considered indicative of activated canonical Wnt signaling (31,32). We found that ␤-catenin, along with a panel of Wnt/Fzrelated genes, was up-regulated in cartilage and synovium during experimental OA. Fz-6, a Wnt receptor that activates protein kinase C and probably noncanonical Wnt signaling (33), was up-regulated in all samples, but only at very low levels. To our knowledge, no role of Fz-6 in joint development or OA has been suggested previously. Dkk-3, an associated Wnt pathway member, was up-regulated in synovium and cartilage in both models; however, the expression change was again modest. Interestingly, Dkk-3 has been associated with OA disease progression in a previous study (34), and therefore merits further investigation. One of the most pronounced changes we observed was the induction of WISP-1 in both the synovium and the cartilage of mice with collagenase-induced OA and STR/Ort mice with spontaneous OA. Interest- 510 ingly, cartilage WISP-1 expression was not changed in streptococcal cell wall–induced arthritis, a model of inflammatory arthritis (results not shown). Recently, French et al have shown that WISP-1 inhibits the differentiation of precursor cells into chondrocytes and may also affect chondrocyte phenotype (35). Taken together, these findings may indicate that WISP-1 expression is specific to OA-like processes. We therefore explored the function of WISP-1 in more detail. WISP-1 is a member of the CCN family of growth factors to which connective tissue growth factor (CTGF), Cyr61, nov, WISP-2, and WISP-3 also belong. Experimental evidence suggests that CCN family members are involved in skeletogenesis and bone healing, and CTGF is also thought to be involved in OA pathology (36–38). Moreover, French and colleagues have shown that overexpression of WISP-1 in ATDC-5 cells completely prevented type II collagen expression and ATDC-5 cell differentiation into a chondrocytic phenotype under the influence of bone morphogenetic protein 2 or growth differentiation factor 5 (35). Our immunohistochemical analyses revealed WISP-1 expression in synovium and cartilage in both experimental models; especially intense staining was observed in the synovial lining layer and in the upper zones of cartilage, complementing the ␤-catenin staining pattern. These findings obtained clinical relevance when we demonstrated that WISP-1 was also increased in human OA cartilage and synovium specimens relative to control tissue. The finding that WISP-1 is expressed in chondrocyte clusters suggests an association of WISP-1 expression with a phenotype change of the cells; however, additional studies are needed before such correlations can be considered further. Interestingly, very low expression of WISP-1 was also observed in one of the postmortem control specimens. This same control sample did exhibit some superficial matrix damage, indicative of early-stage OA, perhaps suggesting that WISP-1 is expressed during the very early stages of OA. Homology between human and murine WISP-1 is ⬎80%, and studies using recombinant human WISP-1 to stimulate murine macrophages, synovial fibroblasts, or human chondrocytes demonstrated that WISP-1 induced several matrix-degrading enzymes, suggesting that it has the potential to cause cartilage damage. The lack of effect when WISP-1 was cocultured with polymyxin B confirmed that the observed effects were not caused by LPS contamination of the recombinant protein. In addition, WISP-1 did not induce IL-1, which is known to be regulated by (among other mediators) LPS (39). Overexpression of WISP-1 in vivo resulted in the BLOM ET AL induction of several MMPs, including MMP-3 and MMP-13, as well as the aggrecanases ADAMTS-4 and ADAMTS-5. In studies using macrophage depletion and MMP-3–deficient mice, we have previously shown that MMP-3 produced by macrophages in the synovium is a mediator of cartilage breakdown (40). Apart from its ability to directly cleave cartilage matrix components, MMP-3 activates proMMPs such as proMMP-13, further affecting extracellular matrix remodeling. Thus, WISP-1 may indirectly activate MMP-13 and facilitate cartilage matrix damage. However, increased MMP-13 activity may also reflect a phenotype change of the chondrocytes, since MMP-13 is a marker for terminal differentiation of chondrocytes (41). We anticipated that WISP-1 would induce a phenotypic change, since it has been demonstrated to drive progenitor cells toward osteogenesis and it is expressed in chondrocyte clusters in OA cartilage. However, apart from MMP-13 upregulation, we failed to detect other signs of chondrocyte phenotype change, such as down-regulation of SOX9 or type II collagen or up-regulation of RUNX-2 or types I and X collagen (data not shown). Since our studies were essentially short-term stimulation assays, additional longer-term overexpression experiments may be necessary to address this question. Recently an association between polymorphisms of the WISP-1 gene and occurrence of spinal OA was found (42), which indicates a role of this protein in OA. Those authors described the absence of a G allele associated with end plate sclerosis in the spine (odds ratio 2.91, P ⫽ 0.0069). Due to the study design, hip, knee, and spinal facet joint OA was not evaluated. For future studies, it will be interesting to determine the functional consequences of this polymorphism, in order to elucidate the mechanism behind the association. Expression profiling of some Wnt signaling molecules other than WISP-1 revealed other potentially interesting changes that warrant further study. Wnt-16 and Wnt-2B are 2 proteins that may be involved in OA pathology and were highly and significantly up-regulated in experimental OA synovium. Wnt-2B is involved in the generation of bone, and may therefore have a detrimental effect on chondrocytes, driving them to terminal differentiation, an event thought to be of importance in OA (43). Guo et al have implicated Wnt-16 as being involved in the formation of synovial joints during embryonic development, by inducing dedifferentiation of chondrocytes to form the synovial space (18). We were not able to demonstrate any regulation of Wnt proteins in the cartilage during collagenaseinduced OA. However, it is likely that Wnt signaling WISP-1 IN EXPERIMENTAL AND HUMAN OA does occur in the cartilage, as indicated by ␤-catenin accumulation. In addition, the prominent expression of WISP-1 in cartilage suggests that Wnt signaling is activated in this tissue. Possibly, Wnt molecules that are produced in the synovium diffuse into the cartilage. In both models of OA we identified many genes that were up-regulated, especially in the synovium. These genes may be involved in common OA-inducing mechanisms. However, several were differentially regulated in collagenase-induced OA compared with spontaneous OA in STR/Ort mice. These genes may be involved in processes that are specific to the individual models, and more detailed mechanistic understanding of the models is needed in order to consider the implications of these differences. IL-1 is a potent inducer of metalloproteinases and a strong inducer of cartilage damage during experimental arthritis (14,44). However, the involvement of IL-1 in OA is unclear, as has been found both in clinical studies and in experimental OA (45,46). We have identified WISP-1 as a mediator that is capable of inducing MMPs and ADAMTS-4 and -5, independently of IL-1. Due to the complex nature of the Wnt signaling pathway, the importance of balanced Wnt signaling for joint integrity, and the diversity of Wnt proteins that are regulated in OA, it is arguable that the modulation of several Wnt family members is necessary for achievement of therapeutic efficacy; this was evidenced by the FrzB genetic data. An alternative may be to modulate a key Wnt target gene such as WISP-1. Since several Wnt proteins have the potential to converge on WISP-1, suppression of WISP-1 function may represent an attractive therapeutic target for OA disease modification; however, further work is needed to determine whether there are any safety liabilities associated with WISP-1 modulation. In conclusion, we have demonstrated for the first time that WISP-1 is overexpressed during OA and is capable of inducing cartilage damage in vivo. Further studies are needed to determine whether WISP-1 is critically required for cartilage loss in OA. ACKNOWLEDGMENTS We are grateful to all patients involved in the study, and to the orthopedic surgeons at the King’s Mill Centre for Healthcare Services (Sutton in Ashfield, UK) for providing clinical material. We especially thank Dr. David Walsh (University of Nottingham, Nottingham, UK) and Dr. Deborah Wilson (Kings Mill Hospital, Sutton in Ashfield, UK) for supplying human cartilage tissue, and Dr. Maarten de Waal 511 Malefijt (Department of Orthopedics, Radboud University Nijmegen Medical Center) for supplying human synovial tissue. AUTHOR CONTRIBUTIONS Dr. Blom had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study design. Blom, Brockbank, van der Kraan, Newham, van den Berg. Acquisition of data. Blom, Brockbank, van Beuningen, Geurts, Takahashi, van de Loo, Schreurs, Clements. Analysis and interpretation of data. Blom, Brockbank, van Lent, van Beuningen, Takahashi, van der Kraan, van de Loo, Clements, Newham, van den Berg. Manuscript preparation. Blom, Brockbank, van Lent, van der Kraan, Newham, van den Berg. Statistical analysis. Blom, van der Kraan, Newham. ROLE OF THE STUDY SPONSOR AstraZeneca applied no restrictions to publication of the manuscript. REFERENCES 1. Sharma L, Kapoor D, Issa S. Epidemiology of osteoarthritis: an update. Curr Opin Rheumatol 2006;18:147–56. 2. Goldring MB, Goldring SR. Osteoarthritis. J Cell Physiol 2007; 213:626–34. 3. Yagi R, McBurney D, Laverty D, Weiner S, Horton WE Jr. Intrajoint comparisons of gene expression patterns in human osteoarthritis suggest a change in chondrocyte phenotype. J Orthop Res 2005;23:1128–38. 4. Lajeunesse D, Reboul P. Subchondral bone in osteoarthritis: a biologic link with articular cartilage leading to abnormal remodeling. Curr Opin Rheumatol 2003;15:628–33. 5. Altman RD, Gray R. Inflammation in osteoarthritis. Clin Rheum Dis 1985;11:353–65. 6. Benito MJ, Veale DJ, FitzGerald O, van den Berg WB, Bresnihan B. Synovial tissue inflammation in early and late osteoarthritis. Ann Rheum Dis 2005;64:1263–7. 7. Smith RL. Degradative enzymes in osteoarthritis. Front Biosci 1999;4:D704–12. 8. Okada Y, Shinmei M, Tanaka O, Naka K, Kimura A, Nakanishi I, et al. Localization of matrix metalloproteinase 3 (stromelysin) in osteoarthritic cartilage and synovium. Lab Invest 1992;66:680–90. 9. Murphy G, Nagase H. Reappraising metalloproteinases in rheumatoid arthritis and osteoarthritis: destruction or repair? Nat Clin Pract Rheumatol 2008;4:128–35. 10. Burrage PS, Mix KS, Brinckerhoff CE. Matrix metalloproteinases: role in arthritis. Front Biosci 2006;11:529–43. 11. Bondeson J, Wainwright S, Hughes C, Caterson B. The regulation of the ADAMTS4 and ADAMTS5 aggrecanases in osteoarthritis: a review. Clin Exp Rheumatol 2008;26:139–45. 12. Glasson SS, Askew R, Sheppard B, Carito B, Blanchet T, Ma HL, et al. Deletion of active ADAMTS5 prevents cartilage degradation in a murine model of osteoarthritis. Nature 2005;434:644–8. 13. Blom AB, van der Kraan PM, van den Berg WB. Cytokine targeting in osteoarthritis. Curr Drug Targets 2007;8:283–92. 14. Cawston T, Billington C, Cleaver C, Elliott S, Hui W, Koshy P, et al. The regulation of MMPs and TIMPs in cartilage turnover. Ann N Y Acad Sci 1999;878:120–9. 15. Yang X, Chen L, Xu X, Li C, Huang C, Deng CX. TGF-␤/Smad3 signals repress chondrocyte hypertrophic differentiation and are 512 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. BLOM ET AL required for maintaining articular cartilage. J Cell Biol 2001;153: 35–46. Karpouzas GA, Terkeltaub RA. New developments in the pathogenesis of articular cartilage calcification. Curr Rheumatol Rep 1999;1:121–7. Tchetina EV, Kobayashi M, Yasuda T, Meijers T, Pidoux I, Poole AR. Chondrocyte hypertrophy can be induced by a cryptic sequence of type II collagen and is accompanied by the induction of MMP-13 and collagenase activity: implications for development and arthritis. Matrix Biol 2007;26:247–58. Guo X, Day TF, Jiang X, Garrett-Beal L, Topol L, Yang Y. Wnt/␤-catenin signaling is sufficient and necessary for synovial joint formation. Genes Dev 2004;18:2404–17. Mak KK, Chen MH, Day TF, Chuang PT, Yang Y. Wnt/␤-catenin signaling interacts differentially with Ihh signaling in controlling endochondral bone and synovial joint formation. Development 2006;133:3695–707. Goldring MB, Tsuchimochi K, Ijiri K. The control of chondrogenesis. J Cell Biochem 2006;97:33–44. Loughlin J, Dowling B, Chapman K, Marcelline L, Mustafa Z, Southam L, et al. Functional variants within the secreted frizzledrelated protein 3 gene are associated with hip osteoarthritis in females. Proc Natl Acad Sci U S A 2004;101:9757–62. Valdes AM, Loughlin J, Oene MV, Chapman K, Surdulescu GL, Doherty M, et al. Sex and ethnic differences in the association of ASPN, CALM1, COL2A1, COMP, and FRZB with genetic susceptibility to osteoarthritis of the knee. Arthritis Rheum 2007;56: 137–46. Min JL, Meulenbelt I, Riyazi N, Kloppenburg M, HouwingDuistermaat JJ, Seymour AB, et al. Association of the Frizzledrelated protein gene with symptomatic osteoarthritis at multiple sites. Arthritis Rheum 2005;52:1077–80. Lories RJ, Peeters J, Bakker A, Tylzanowski P, Derese I, Schrooten J, et al. Articular cartilage and biomechanical properties of the long bones in Frzb-knockout mice. Arthritis Rheum 2007;56: 4095–103. Urano T, Shiraki M, Narusawa K, Usui T, Sasaki N, Hosoi T, et al. Q89R polymorphism in the LDL receptor-related protein 5 gene is associated with spinal osteoarthritis in postmenopausal Japanese women. Spine 2007;32:25–9. Zimmermann T, Kunisch E, Pfeiffer R, Hirth A, Stahl HD, Sack U, et al. Isolation and characterization of rheumatoid arthritis synovial fibroblasts from primary culture: primary culture cells markedly differ from fourth-passage cells. Arthritis Res 2001;3: 72–6. Van Beuningen HM, Stoop R, Buma P, Takahashi N, van der Kraan PM, van den Berg WB. Phenotypic differences in murine chondrocyte cell lines derived from mature articular cartilage. Osteoarthritis Cartilage 2002;10:977–86. 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. Chartier C, Degryse E, Gantzer M, Dieterle A, Pavirani A, Mehtali M. Efficient generation of recombinant adenovirus vectors by homologous recombination in Escherichia coli. J Virol 1996;70:4805–10. Dominguez I, Green JB. Dorsal downregulation of GSK3␤ by a non-Wnt-like mechanism is an early molecular consequence of cortical rotation in early Xenopus embryos. Development 2000; 127:861–8. 31. Fuerer C, Nusse R, ten Berge D. Wnt signalling in development and disease: Max Delbrück Center for Molecular Medicine Meeting on Wnt Signaling in Development and Disease. EMBO Rep 2008;9:134–8. 32. Clevers H. Wnt/␤-catenin signaling in development and disease. Cell 2006;127:469–80. 33. Sheldahl LC, Park M, Malbon CC, Moon RT. Protein kinase C is differentially stimulated by Wnt and Frizzled homologs in a G-protein-dependent manner. Curr Biol 1999;9:695–8. 34. Meng J, Ma X, Ma D, Xu C. Microarray analysis of differential gene expression in temporomandibular joint condylar cartilage after experimentally induced osteoarthritis. Osteoarthritis Cartilage 2005;13:1115–25. 35. French DM, Kaul RJ, D’Souza AL, Crowley CW, Bao M, Frantz GD, et al. WISP-1 is an osteoblastic regulator expressed during skeletal development and fracture repair. Am J Pathol 2004;165: 855–67. 36. Blaney Davidson EN, Vitters EL, Mooren FM, Oliver N, van den Berg WB, van der Kraan PM. Connective tissue growth factor/ CCN2 overexpression in mouse synovial lining results in transient fibrosis and cartilage damage. Arthritis Rheum 2006;54:1653–61. 37. Fukunaga T, Yamashiro T, Oya S, Takeshita N, Takigawa M, Takano-Yamamoto T. Connective tissue growth factor mRNA expression pattern in cartilages is associated with their type I collagen expression. Bone 2003;33:911–8. 38. Canalis E. Nephroblastoma overexpressed (Nov) is a novel bone morphogenetic protein antagonist. Ann N Y Acad Sci 2007;1116: 50–8. 39. Pennanen N, Lapinjoki S, Palander A, Urtti A, Monkkonen J. Macrophage-like RAW 264 cell line and time-resolved fluoroimmunoassay (TRFIA) as tools in screening drug effects on cytokine secretion. Int J Immunopharmacol 1995;17:475–80. 40. Blom AB, van Lent PL, Libregts S, Holthuysen AE, van der Kraan PM, van Rooijen N, et al. Crucial role of macrophages in matrix metalloproteinase–mediated cartilage destruction during experimental osteoarthritis: involvement of matrix metalloproteinase 3. Arthritis Rheum 2007;56:147–57. 41. D’Angelo M, Yan Z, Nooreyazdan M, Pacifici M, Sarment DS, Billings PC, et al. MMP-13 is induced during chondrocyte hypertrophy. J Cell Biochem 2000;77:678–93. 42. Urano T, Narusawa K, Shiraki M, Usui T, Sasaki N, Hosoi T, et al. Association of a single nucleotide polymorphism in the WISP1 gene with spinal osteoarthritis in postmenopausal Japanese women. J Bone Miner Metab 2007;25:253–8. 43. Andrade AC, Nilsson O, Barnes KM, Baron J. Wnt gene expression in the post-natal growth plate: regulation with chondrocyte differentiation. Bone 2007;40:1361–9. 44. Van Meurs JB, van Lent PL, Singer II, Bayne EK, van de Loo FA, van den Berg WB. Interleukin-1 receptor antagonist prevents expression of the metalloproteinase-generated neoepitope VDIPEN in antigen-induced arthritis. Arthritis Rheum 1998;41: 647–56. 45. Clements KM, Price JS, Chambers MG, Visco DM, Poole AR, Mason RM. Gene deletion of either interleukin-1␤, interleukin1␤–converting enzyme, inducible nitric oxide synthase, or stromelysin 1 accelerates the development of knee osteoarthritis in mice after surgical transection of the medial collateral ligament and partial medial meniscectomy. Arthritis Rheum 2003;48:3452–63. 46. Iqbal I, Fleischmann R. Treatment of osteoarthritis with anakinra. Curr Rheumatol Rep 2007;9:31–5.