ARTHRITIS & RHEUMATISM Vol. 60, No. 12, December 2009, pp 3534–3545 DOI 10.1002/art.27251 © 2009, American College of Rheumatology REVIEW Mediators of Structural Remodeling in Peripheral Spondylarthritis Bernard Vandooren,1 Nataliya Yeremenko,2 Troy Noordenbos,2 Johannes Bras,2 Paul P. Tak,2 and Dominique Baeten2 Introduction RA cannot be fully explained by differential synovial expression of mediators of tissue destruction. In contrast, increasing evidence suggests that the distinct type of remodeling may relate to bone and cartilage anabolism rather than to degradation as such, reflecting sustained activity of tissue repair responses in SpA. In contrast with bone and cartilage degradation, these tissue repair responses have not yet been shown to be influenced by TNF blockade. Integration of experimental models and translational data as well as prospective imaging and biomarker studies are now needed to determine how to block ankylosis in SpA. This may also lead to new strategies for promoting tissue repair in destructive diseases such as RA. The inflamed peripheral joint is relatively protected from erosive damage in nonpsoriatic spondylarthritis (SpA) as compared with rheumatoid arthritis (RA), with psoriatic arthritis (PsA) showing a mixed pattern of destruction and remodeling. The cellular and molecular mechanisms of cartilage and bone destruction have been extensively studied in experimental models and in human RA. In contrast, the mechanisms and role of new tissue formation are only starting to emerge from studies in animal models, and translation to the arthritis in humans remains challenging. Here, we review the recent insights into the mechanisms of structural remodeling in peripheral SpA. The inflammatory milieu in the joints of patients with SpA and the joints of patients with RA appears to be clearly different, with a relative overexpression of the proinflammatory cytokines tumor necrosis factor ␣ (TNF␣) and interleukin-1␤ (IL-1␤) in the latter. Even if these cytokines can drive destructive mechanisms, the differences in structural remodeling between SpA and Structural remodeling in inflammatory arthritis Destruction of connective tissues is a key feature of inflammatory arthritis in humans, since it determines in large part the long-term functionality of the joint and, thus, the quality of life of the patients. Joint destruction has been investigated in depth in RA, where it is characterized by osteoporosis of paraarticular bone, progressive loss of articular cartilage, and the development of lytic bone lesions at the synovium–bone interface. This pronounced cartilage and bone degradation and damage is not associated with clear signs of new bone formation or repair and can be halted by TNF antagonists, even when this treatment fails to completely halt the inflammatory component of the disease (1). It is of major interest, however, that other types of chronic joint inflammation lead to completely different patterns of joint remodeling. The prototypical example is spondylarthritis, which, besides modest cartilage and bone degradation, mainly shows signs of extensive new endochondral bone formation. This process affects not only the sacroiliac joints and the spine, leading to This publication reflects only the authors’ views. The European Community is not liable for any use that may be made of the information herein. Supported by The Netherlands Organization for Scientific Research (NWO), the Dutch Arthritis Association, and the European Commission Sixth Framework Programme (project AutoCure). Dr. Vandooren’s work was supported by a Research Fellowship from the Fund for Scientific Research, Flanders (FWO Vlaanderen). 1 Bernard Vandooren, MD, PhD: Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands, and Ghent University Hospital, Ghent, Belgium; 2Nataliya Yeremenko, PhD, Troy Noordenbos, Johannes Bras, MD, Paul P. Tak, MD, PhD, Dominique Baeten, MD, PhD: Academic Medical Center, University of Amsterdam, Amsterdam, The Netherlands. Address correspondence and reprint requests to Dominique Baeten, MD, PhD, Department of Clinical Immunology and Rheumatology, F4-105, Academic Medical Center, University of Amsterdam, Meibergdreef 9, 1105 AZ Amsterdam, The Netherlands. E-mail: D.L. Baeten@amc.uva.nl. Submitted for publication February 1, 2009; accepted in revised form August 31, 2009. 3534 MEDIATORS OF STRUCTURAL REMODELING IN PERIPHERAL SpA 3535 regulated in different arthritic conditions. Since the Holy Grail of treatment is to maintain and/or restore the structural balance in the different types of arthritis, the challenge is now to identify the cellular and molecular mechanisms that underlie these differences. Mediators of bone remodeling Figure 1. Characteristics of peripheral joint remodeling in spondylarthritis (SpA). Shown are radiographs and drawings of a normal distal interphalangeal joint (top) and an ankylosed joint of a patient with psoriatic SpA (bottom). axial ankylosis, but also the peripheral joints (Figure 1). Although TNF blockade also strongly suppresses signs and symptoms in SpA, clinical trials have thus far failed to show any inhibitory effects on new bone formation and ankylosis (2,3). In terms of joint remodeling, PsA is a distinct subset of SpA: it shares the features of new bone formation, but in contrast to nonpsoriatic SpA, erosive disease is fairly common in PsA and can be even much more aggressive than in RA. Again, TNF blockade appears to block erosive disease (4–6), but there is no evidence that it affects new bone formation and ankylosis. These remarkable clinical differences teach us 3 important lessons about pathogenesis. First, whereas it is clear that osteoclast-mediated bone erosions are crucial in RA, nonpsoriatic and psoriatic SpA emphasize that the structural outcome is not determined by destruction alone, but rather, by the more complex balance between destruction and repair. Second, TNF blockade has a marked effect on the progression of erosions but has not yet been demonstrated to influence new bone formation in established SpA, suggesting that the mechanisms of destruction versus repair may be either uncoupled or differentially regulated. And third, the fact that ankylosis results from endochondral bone formation in SpA raises the question concerning the extent to which not only bone, but also cartilage, metabolism is differentially The cellular and molecular mechanisms of bone and cartilage remodeling have been extensively studied and described in vitro as well as in different experimental models. Although it is beyond the scope of this review to give an extensive discussion of the biology of bone and cartilage, a number of these pathways provide essential information for our understanding of structural damage in human inflammatory arthritis. Bone remodeling is a coupled process involving bone formation and resorption. This process is regulated by local and systemic factors that drive osteoblast and osteoclast differentiation and function. Osteoblasts differentiate from mesenchymal precursors and deposit bone matrix constituents, such as type I collagen and osteocalcin. Although the regulation of osteoblast differentiation and function is a complex process, a number of factors appear to be of pivotal importance (Figure 2). For example, the osteoblast-specific transcription factor runt-related transcription factor 2 (RUNX-2) integrates the actions of different anabolic factors, such as transforming growth factor ␤ (TGF␤) and insulin-like growth factor 1 (IGF-1), and hormones on osteoblasts and thereby controls the transcription of osteocalcin and other osteoblast-specific genes (7). This pathway is further influenced by WNT signaling: upon binding to Frizzled receptors, WNT indirectly inhibits the phosphorylation of ␤-catenin by glycogen synthase kinase 3␤ (GSK-3␤). Since nonphosphorylated ␤-catenin is less susceptible to ubiquitin-mediated degradation, this leads to accumulation of ␤-catenin, nuclear ingress, and ultimately, the transcription of factors involved in osteoblast differentiation and activation. Accordingly, mutations in members of the WNT family, ablation of the WNT coreceptors low-density lipoprotein receptor–related protein 5 (LRP-5) or LRP-6, or use of WNT inhibitors, such as sclerostin or the Dkk family members, reduce osteoblastogenesis and bone formation (8). Apart from synthesizing bone matrix, osteoblasts also regulate the differentiation and activity of osteoclasts (Figure 3). Osteoclasts are multinuclear cells of hematopoietic origin that are specialized in resorbing bone (9). Early osteoclast precursor cells in the bone marrow express the c-Fms receptor and proliferate 3536 VANDOOREN ET AL Figure 2. Schematic representation of the steps in the differentiation of mesenchymal stem cells toward either chondrocytes or osteocytes. The boxes indicate the main differentiation factors for each step. The main matrix products and transcription factors are indicated in the cells. BMP ⫽ bone morphogenetic protein; TGF␤ ⫽ transforming growth factor ␤; Coll-II ⫽ type II collagen; MIA ⫽ melanoma inhibitory activity; IGF-1 ⫽ insulin-like growth factor 1; FGF ⫽ fibroblast growth factor; RUNX-2 ⫽ runt-related transcription factor 2; PTH ⫽ parathyroid hormone; 1,25(OH) 2 D 3 ⫽ 1,25-dihydroxyvitamin D3; BSP ⫽ bone sialoprotein. under the influence of macrophage colony-stimulating factor (M-CSF). Later on, they also start to express c-Fos and RANK. Binding of RANKL, which is expressed by osteoblasts, is absolutely required for the full maturation and activity of osteoclasts. This binding can be regulated by osteoprotegerin (OPG), the decoy receptor for RANKL. In pathologic conditions, proinflammatory mediators can profoundly alter the balance between bone resorption by osteoclasts and new bone formation by osteoblasts. This has been best studied for TNF, which on the one hand, directly (10,11) and indirectly (through up-regulation of M-CSF and RANKL expression by stromal cells) (12,13) promotes osteoclast formation and activation and, on the other hand, suppresses WNT signaling by up-regulating inhibitors such as Dkk-1 (14). The relevance of TNF for the bone phenotype in vivo has been demonstrated both experimentally in the TNF-transgenic mouse model (15,16) and clinically in RA patients treated with TNF blockers (17). Mediators of cartilage remodeling Articular cartilage is built up by 4 different layers of chondrocytes that have different gene expression profiles. The superficial layer consists of 2 or 3 flattened cell layers, which express lubricin as well as type II collagen, aggrecan core protein, tenascin, and SOX9. The intermediate zone expresses no lubricin, but higher levels of cartilage intermediate-layer protein. Chondrocytes in the radial zone express markers of chondrocyte hypertrophy, such as type X collagen and alkaline phosphatase. In the mineralized cartilage, the cells start to express osteoblast factors, such as RUNX-2 and osterix. Chondrogenesis is crucially dependent on the transcription factor SOX9, because no cartilage is formed in the absence of SOX9 (18) and SOX9 directly regulates the expression of cartilage-specific genes, such as type II collagen (19). The expression and activity function of SOX9 is promoted by anabolic cytokines, such as TGF␤ (20), and impaired by proinflammatory cytokines, such as IL-1␤ (21). A clear distinction should be made here between articular cartilage and growth plate and/or developmental cartilage. Although the same transcription factors govern the behavior of these cells, the articular chondrocyte is a stable cell type that is regulated by a very strict balance of signaling pathways under physiologic conditions. In contrast, the biologic status of the devel- MEDIATORS OF STRUCTURAL REMODELING IN PERIPHERAL SpA 3537 Figure 3. Mediators of osteoclast development from blood-borne myeloid precursors. Monocytes enter the inflamed synovial tissue from a blood vessel and differentiate into osteoclasts under the influence of mediators produced by resident cells. M-CSF ⫽ macrophage colony-stimulating factor; TNF␣ ⫽ tumor necrosis factor ␣. opmental chondrocyte is influenced by changing signals, leading to hypertrophy and, eventually, apoptosis of these cells. This occurs typically in endochondral ossification, a complex and finely tuned process involving different steps. The first step is the condensation of mesenchymal stem cells, which differentiate into chondrocytes. Subsequently, these cells proliferate, deposit typical cartilage matrix, become hypertrophic, and release factors that promote the calcification of the matrix. This triggers osteoclasts to resorb and invade the calcified matrix, paving the path for in-growing blood vessels. Subsequently, poorly differentiated mesenchymal cells that were initially located at the periphery of the process attach to the excavated calcified matrix and differentiate into osteoblasts. These osteoblasts finally secrete bone matrix that replaces the calcified cartilage matrix. In pathologic conditions, both processes (the stable metabolism of articular chondrocytes and the new tissue formation by developmental chondrocytes) could be disturbed. Whereas the latter is an emerging field of particular interest in tissue remodeling in SpA (see below), the modulation of cartilage metabolism by inflammatory conditions has already been extensively studied in different arthritic conditions (Figure 4). Prototypically, high levels of IL-1␤ will strongly suppress cartilage anabolism and, thus, the production of normal matrix constituents (11,22). Moreover, IL-1␤ and other inflammatory mediators induce the production of a series of enzymes that cleave cartilage extracellular matrix constituents, such as type II collagen and aggrecan. These matrix-degrading enzymes, including cathepsins, aggrecanases, and matrix metalloproteinases (MMPs), are produced by chondrocytes themselves or by neighboring stromal cells (23,24), with the fibroblastlike synoviocyte (FLS) as prototype. More recently, it was demonstrated that the ability of FLS to degrade cartilage is also dependent on the expression of cadherin 11, a molecule that mediates homophilic adhesion between FLS and is critical for the maintenance of the normal synovial microarchitecture (25). Upon induction of arthritis, cadherin 11–deficient mice are protected from cartilage loss but not bone erosions (26), indicating a specific role of this molecule in cartilage damage (Figure 4). Kiener et al (27) recently demonstrated the abundant expression of cadherin 11 at the cartilage–pannus junction and the crucial role of cadherin 11 expression on FLS for in vitro invasive behavior in the Matrigel invasion assay. These data indicate that that this adhesion molecule is important not only for cell–cell adhesion between FLS, but also for the adherence (and subsequent invasion) of FLS to cartilage. Taken together, the suppression of new matrix 3538 VANDOOREN ET AL Figure 4. Mediators of cartilage degradation in chronic arthritis. Cartilage matrix is degraded by mediators that are released from the inflamed synovium, as well as by synovial fluid neutrophils (top). Pannus tissue extends over the cartilage surface and degrades the matrix by producing proteolytic enzymes (bottom). MMP ⫽ matrix metalloproteinase; IL-1 ⫽ interleukin-1; Coll-II ⫽ type II collagen; MIA ⫽ melanoma inhibitory activity. formation by IL-1␤ and the degradation of existing matrix by proteases contribute to the loss of articular cartilage. Inflammatory milieu in peripheral spondylarthritis As mentioned above, proinflammatory cytokines, such as TNF␣ and IL-1␤, play a crucial role in driving bone and cartilage damage. Therefore, the differential structural phenotype in RA, nonpsoriatic SpA, and PsA may relate to differences in the type or degree of local inflammation in these diseases. Indeed, there are clear differences in the synovial histopathology between SpA and RA synovitis, indicating that, despite similar levels of overall inflammation, the type of inflammation is different in the two diseases (28–34). Of particular interest in the context of structural damage, however, is that PsA resembles nonpsoriatic SpA rather than RA (35). In both diseases, the synovial lining layer can MEDIATORS OF STRUCTURAL REMODELING IN PERIPHERAL SpA become hyperplastic and form a macroscopic and microscopic pannus that adheres to cartilage (28,32). The difference between the two diseases appears to be quantitative, since the degree of intimal lining layer hyperplasia is slightly higher in RA than in SpA. At the cellular level, an intriguing finding of most, but not all, studies is the preferential infiltration by macrophages expressing the alternative activation marker CD163 in SpA (30,36,37). This specific macrophage subset appeared to correlate well with global disease activity as well as with the response to treatment in SpA (38,39). The subclassification of macrophages according to the expression of phenotype markers such as CD163 refers to the increasing evidence that macrophages are not a homogenous population but can be divided into specific, although overlapping, subsets according to their polarization requirements, phenotype, and function (40). Of importance in the context of structural damage, classically activated macrophages (M1) are the main source of proinflammatory cytokines, such as TNF␣ and IL-1␤, whereas alternatively activated macrophages (M2) expressing CD163 have been implicated in immune regulation, phagocytosis, and tissue remodeling. This also has functional implications for synovitis, since synovial fluid levels of M1-derived cytokines, such as IL-1␤ and TNF␣, are indeed significantly lower in SpA than in RA synovitis despite a similar overall degree of tissue inflammation (41). These data are consistent with an earlier observation on TNF described by Cañete et al (42). Consistent with the histopathologic findings, we have also shown that the M1-derived cytokine profile in PsA had a greater resemblance to that in SpA than to that in RA (41). Bone damage in peripheral spondylarthritis The question raised by these studies is whether the different inflammatory milieus, especially the lower levels of TNF␣ and IL-1␤ in SpA, affect the presence and activity of key mediators of bone and cartilage remodeling. Since bone destruction is mediated exclusively by osteoclasts, several studies have compared the potential of peripheral blood–derived myeloid cells to differentiate into osteoclasts in different types of inflammatory arthritis. When incubated in vitro with M-CSF and RANKL, RA peripheral blood–derived cells were more prone to differentiate to osteoclasts than were SpA peripheral blood–derived cells (43). Using a different in vitro assay without the addition of exogenous factors (called spontaneous osteoclastogenesis), peripheral 3539 blood cells from patients with PsA showed increased osteoclastogenic potential as compared with those from healthy controls (44,45). Although the relevance of these assays for in vivo bone destruction remains unclear, the increased ex vivo osteoclastogenic potential appears to be related to the inflammatory milieu, since it was reversed by in vivo treatment with TNF blockers (43,44). We recently reassessed the mechanisms of spontaneous osteoclastogenesis in order to explore whether the different bone phenotypes of SpA, PsA, and RA patients was related to either intrinsic differences in myeloid cells or differences in T cell activation (46). T cell depletion abrogated spontaneous osteoclastogenesis in fresh peripheral blood mononuclear cells, but these cells were completely redundant in the enhanced osteoclastogenesis observed in cryopreserved cells. Under these conditions, osteoclastogenesis resulted from a direct activation of CD14-positive osteoclast precursors without clear differences between SpA, PsA, and RA patients, indicating that there is no obvious intrinsic difference in the osteoclastogenic potential of peripheral blood monocytes between the different pathologic conditions. These data raised questions about the translational relevance of the in vitro osteoclast differentiation assays and emphasized the importance of in situ analysis at the site of bone destruction. Because the RANK/ RANKL pathway is critical for bone remodeling under both physiologic and inflammatory conditions, we also compared the expression of these molecules in the synovium of SpA and RA patients. We found no differences in the expression of RANK, RANKL, or the decoy receptor OPG between RA, SpA, and PsA patients (47). The partial discrepancy of these findings with those of previous studies suggesting higher OPG levels in SpA synovium is related to the poor specificity of the antibodies used in previous studies (48–50). Although the differences in structural remodeling between RA and SpA cannot be explained by differences in the synovial expression of RANK, RANKL, or OPG, an important caveat may be that the investigated synovium was obtained from a site away from the synovium–bone junction. Consequently, the expression of these factors in the proximity of the bone or within the bone itself remains to be investigated. Cartilage damage in peripheral spondylarthritis Our knowledge of articular cartilage damage in peripheral SpA is still very limited. Results of imaging studies suggest that cartilage integrity is better preserved 3540 in peripheral SpA than in RA (51), but this has not yet been unequivocally confirmed by analysis of cartilage degradation biomarkers in synovial fluid or by histopathologic assessments (52). As indicated in our previous study, synovial fluid levels of IL-1␤ and TNF were significantly lower in SpA than RA (41). Since these proinflammatory cytokines can directly stimulate synovial fibroblasts to develop an aggressive phenotype, we investigated the expression of cadherin 11 as a key factor for cartilage degradation by FLS (53). Synovial cadherin 11 expression was significantly correlated with the degree of tissue inflammation and was decreased upon in vivo treatment with glucocorticoids or upon TNF blockade. Accordingly, cadherin 11 expression by FLS was up-regulated by TNF␣ in vitro. Although these data indicate that cadherin 11 upregulation may form an important link between TNFdriven synovial inflammation and cartilage degradation, we did not find differences in cadherin 11 expression between RA and SpA. Another group of important mediators of cartilage destruction produced by activated FLS are the MMPs. Performing a systematic analysis of the synovial expression of MMP-1, MMP-3, and MMP-9 as well as their endogenous inhibitors tissue inhibitor of metalloproteinases 1 (TIMP-1) and TIMP-2, we found no differences between SpA and RA synovitis (54). Similar to cadherin 11, synovial MMP expression was downregulated by TNF blockade in vivo (54,55). Of interest, MMP-3 appeared to be the most sensitive marker of treatment response and reflected peripheral joint disease rather than axial disease in SpA. Although this concept has been confirmed by several studies (56,57), it remains unclear if MMP-3 expression directly drives structural damage or if it is merely an epiphenomenon of the ongoing inflammatory process. Prospective studies correlating cadherin 11 and MMP expression with levels of biomarkers and with the results of imaging of cartilage damage are warranted in order to address these issues. Bone repair in peripheral spondylarthritis Whereas most studies have focused on bone and cartilage destruction, the typical ankylosis observed in SpA emphasizes that bone and cartilage repair is at least as important for the structural outcome of inflammatory arthritis. Histologic studies have indicated that the ankylosis mainly results from chondroid metaplasia and subsequent endochondral bone formation (58). This suggests that the developmental chondrocytes discussed VANDOOREN ET AL above may play a crucial role in new tissue formation in SpA, but the underlying molecular mechanisms and triggers of this process remain elusive. The study of new bone formation in SpA has recently been revived by molecular studies in animal models. Lories et al (59) investigated the role of bone morphogenetic proteins (BMPs) in the structural remodeling of the joint in aging male DBA/1 mice. These mice spontaneously develop arthritis and enthesitis of the paw and ankle joints and, in contrast to TNF␣transgenic mice, also develop reactive endochondral bone formation that leads to bony ankylosis. BMP-2 and BMP-6 were demonstrated to play an important role in this process. Of interest, a clear up-regulation of the expression of BMPs 2 and 6, together with an activation of the Smad signaling pathway, was also observed in SpA synovium, but this was not different from that in RA (60). Another molecular pathway of major interest, the Wingless/WNT signaling pathway, was investigated by Diarra et al (14) in human TNF–transgenic mice. Deletion of the WNT inhibitor Dkk-1 not only reduced bone erosions, but more importantly, it allowed new bone formation despite ongoing inflammation. In contrast to the previous model, the new bone formation may be directly mediated here by osteoblasts rather than by developmental chondrocytes and endochondral bone formation. Also in humans, Dkk-1 is produced by FLS in response to TNF␣ in vitro and is expressed in the inflamed RA synovium, and it may therefore contribute to the suppression of bone repair in the RA joint. Although serum levels of Dkk-1 were reported to be decreased in SpA as compared with RA, the interpretation of these data is complicated by the differences in disease activity between both groups as well as the predominant axial disease in the SpA cohort (14). Therefore, the relevance of the Dkk-1/WNT pathway for new bone formation in human SpA remains to be assessed. Both experimental models discussed here point toward an intriguing relationship between inflammation and new bone formation. In the ankylosing enthesitis model, new bone formation could not be blocked by TNF inhibition (61). In TNF-transgenic mice, inflammation and inhibition of new bone formation could be completely uncoupled by targeting Dkk (14). In human SpA, there is now increasing evidence that TNF blockade is very effective in controlling inflammation, but it does not appear to have a major impact on the progression of axial ankylosis in established SpA (2,3). These data need to be confirmed in longer-term studies as well MEDIATORS OF STRUCTURAL REMODELING IN PERIPHERAL SpA as in studies of patients with early disease. Moreover, the extent to which TNF blockade affects new bone formation (either its inhibition or its promotion) in peripheral SpA remains to be investigated. The enthesis and new bone formation in peripheral SpA The experimental studies by Lories et al (59–61) mentioned above are not only relevant to the molecular mechanisms of new bone formation in SpA-like disease, but they also emphasize the importance of the anatomic context of such responses. This aspect was originally described by McGonagle and colleagues in an important magnetic resonance imaging study: focal inflammation in affected knee joints was more frequently observed at perientheseal sites in SpA than in RA (62), which led them to propose the novel hypothesis that SpA is a primary entheseal pathologic condition (63). Despite the obvious interest of this novel conceptual framework, the independent followup studies performed to confirm these findings yielded contradictory results (64–69). In particular, the findings of these imaging studies called into question the specificity of this finding for SpA, since enthesitis was also commonly observed in other types of joint inflammation, including RA. Whether the enthesitis is primary in SpA and secondary in other types of joint inflammation remains to be determined in prospective, longitudinal studies of patients with early, untreated disease. In addition, the exact histologic substrate of these imaging abnormalities at the enthesis remains to be better defined, since the 2 studies on this topic yielded conflicting results (70,71). In the context of the present review, the main question is whether the new bone formation seen in SpA originates at this particular site of inflammation. It is well known that bony spurs can develop in the normal enthesis of healthy humans (72). This process of endochondral ossification affects the fibrocartilage of the enthesis, without the need for preceding microtears or any inflammatory reaction (73). Patients with early SpA have the same type of small spurs in the distal part of the enthesis as normal control subjects, but they display bone erosions at the proximal insertion (74). Moreover, large spurs, which are not observed in normal controls, are found at the distal part of the enthesis in longstanding SpA. These data indicate that, indeed, new bone formation can occur at the entheseal site but that this process is topographically and temporally uncoupled from bone erosions. This finding is consistent with experimental data showing that inhibition of osteoclasts 3541 does not prevent ankylosis in the DBA/1 mouse model of enthesitis (75). Cartilage repair in peripheral spondylarthritis As mentioned above, there are few indications that articular cartilage may be better preserved in SpA than in RA. It is tempting to speculate that triggers of chondroid hyperplasia and resulting endochondral bone formation that play a role in ankylosis in SpA may not only affect developmental chondrocytes, for example, at the enthesis, but may also modulate the anabolic/ catabolic balance of articular chondrocytes. Different studies assessed cartilage metabolism in SpA by measuring serum biomarkers of cartilage. Kim et al (76) reported that serum levels of the 846 epitope and C-propeptide of type II collagen (CPII), both of which are considered to be anabolic markers, but not C2C, a serum biomarker of type II collagen degradation, were elevated in patients with ankylosing spondylitis as compared with healthy controls. This skewing toward cartilage anabolism in SpA contrasted with the catabolic pattern in RA, but the use of serum biomarkers unfortunately does not allow the differentiation between articular and developmental chondrocytes. In an attempt to study the articular chondrocyte more directly by measuring synovial fluid melanoma inhibitory activity (MIA) as a biomarker of cartilage anabolism, we found that chondrocyte anabolism was more activated in the joints of SpA patients than in the joints of RA patients despite similar levels of anabolic growth factors IGF-1 and TGF␤3 (77). The fact that IGF-1 levels correlated strongly with MIA levels in SpA patients, but not RA patients, suggests that IGF-1 plays an important role in driving cartilage anabolism in SpA. These findings fit with the concept that the high levels of proinflammatory cytokines found in RA as compared with SpA may desensitize RA cartilage to the anabolic effects of IGF-1 (78). Even if cartilage anabolism is better preserved in SpA than in RA, the levels of cartilage anabolism biomarkers are further up-regulated by TNF blockade in SpA. Infliximab therapy up-regulates MIA levels, and etanercept treatment leads to a significant increase in serum levels of CPII and the 846 epitope (77,79,80). Overall, these biomarker data suggest that articular cartilage anabolism is promoted by TNF blockade. This may partly explain why restoration of the articular cartilage thickness is observed after TNF blockade in some SpA patients. Whether these processes also affect the developmental chondrocyte and thereby contribute 3542 to cartilage metaplasia, endochondral bone formation, and, eventually, pathologic ankylosis is not yet known. Functional implications The translational expression and biomarker studies described here have their obvious limitations, and a number of important issues remain to be addressed. First, expression studies do not allow us to directly address the functional implications of these molecular pathways. Whereas transgenic and knockout models have turned out to provide useful biologic information, the major challenge is to link this functional information to a clinically relevant model of SpA. Analysis of these pathways and interactions in the previously described ankylosing enthesitis and TNF-transgenic models as well as in the HLA–B27–transgenic rat model is therefore of prime importance (14,59,81,82). Second, translational expression studies in SpA have been conducted in patients with peripheral disease, since clinically relevant material can be obtained from inflamed knee and ankle joints. It is only recently that similar studies have been initiated for axial disease. Over the last couple of years, Appel and colleagues (83,84) have described a number of important features of bone, cartilage, and synovium in patients with axial SpA. The extent to which the mechanisms of bone and cartilage remodeling are similar or different in axial versus peripheral disease remains to be determined. Third, cross-sectional and short-term longitudinal studies are not able to define the exact order of events in the structural remodeling that occurs in SpA. Therefore, the question remains open whether progressive ankylosis depends on initial inflammatory damage of bone and cartilage or whether the processes of destruction and new tissue formation are completely uncoupled (75,85). Future directions Our classic approach to complex biologic questions consists of careful histologic and immunopathologic analyses, functional experiments in vitro and in experimental models, identification of novel targets, and finally, development and testing of specific drugs in patients with the disease. The limited access to clinically relevant human tissue samples, the paucity of appropriate experimental models, and the requirement that long-term longitudinal studies be performed are all major hurdles for our classic approach to overcome in deciphering the exact cellular and molecular mecha- VANDOOREN ET AL nisms of tissue remodeling in SpA. The development and validation of reliable biomarkers that reflect specific aspects of the pathophysiology of tissue remodeling may circumvent some of these hurdles, but such markers are not yet available. An alternative and innovative approach is, however, now available with the increasing number of targeted biologic and small-molecule therapies. These drugs, which have usually been developed and tested for efficacy and safety in other pathologic conditions, indeed allow us to probe specific cellular and molecular pathophysiologic mechanisms in vivo in our patients. This reverse approach, using a targeted therapy as a “human knock-down model” in which to delineate a clinically relevant pathway, rather than to identify molecular mechanisms in order to develop new therapies, may be particularly efficient for deciphering tissue remodeling in SpA. This should include early and aggressive intervention with existing inflammation-directed biologic agents (anticytokine antibodies, costimulation blockade, cell depletion) as well as the evaluation of therapies designed to specifically target connective tissue in other fields of medicine in order to clarify the extent to which inflammation and tissue remodeling are coupled or uncoupled in SpA. 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. Baeten 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. Vandooren, Yeremenko, Noordenbos, Bras, Tak, Baeten. Acquisition of data. Vandooren. Analysis and interpretation of data. Vandooren, Yeremenko, Noordenbos, Bras, Tak, Baeten. REFERENCES 1. Lipsky PE, van der Heijde DM, St. Clair EW, Furst DE, Breedveld FC, Kalden JR, et al, and the Anti-Tumor Necrosis Factor Trial in Rheumatoid Arthritis with Concomitant Therapy Study Group. Infliximab and methotrexate in the treatment of rheumatoid arthritis. N Engl J Med 2000;343:1594–602. 2. Van der Heijde D, Landewe R, Einstein S, Ory P, Vosse D, Ni L, et al. Radiographic progression of ankylosing spondylitis after up to two years of treatment with etanercept. Arthritis Rheum 2008;58:1324–31. 3. Van der Heijde D, Landewe R, Baraliakos X, Houben H, van Tubergen A, Williamson, P, et al. Radiographic findings following two years of infliximab therapy in patients with ankylosing spondylitis. Arthritis Rheum 2008;58:3063–70. 4. Mease PJ, Kivitz AJ, Burch FX, Siegel EL, Cohen SB, Ory P, et al. Etanercept treatment of psoriatic arthritis: safety, efficacy, and effect on disease progression. Arthritis Rheum 2004;50:2264–72. 5. Van der Heijde D, Kavanaugh A, Gladman DD, Antoni C, MEDIATORS OF STRUCTURAL REMODELING IN PERIPHERAL SpA 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. Krueger GG, Guzzo C, et al, Infliximab inhibits progression of radiographic damage in patients with active psoriatic arthritis through one year of treatment: results from the Induction and Maintenance Psoriatic Arthritis Clinical Trial 2. Arthritis Rheum 2007;56:2698–707. Gladman DD, Mease PJ, Ritchlin CT, Choy EH, Sharp JT, Ory PA, et al. Adalimumab for long-term treatment of psoriatic arthritis: forty-eight week data from the Adalimumab Effectiveness in Psoriatic Arthritis Trial. Arthritis Rheum 2007;56:476–88. Ducy P, Zhang R, Geoffroy V, Ridall AL, Karsenty G. Osf2/ Cbfa1: a transcriptional activator of osteoblast differentiation. Cell 1997;89:747–54. Krishnan V, Bryant HU, Macdougald OA. Regulation of bone mass by Wnt signaling. J Clin Invest 2006;116:1202–9. Boyle WJ, Simonet WS, Lacey DL. Osteoclast differentiation and activation. Nature 2003;423:337–42. Lam J, Takeshita S, Barker JE, Kanagawa O, Ross FP, Teitelbaum SL. TNF-␣ induces osteoclastogenesis by direct stimulation of macrophages exposed to permissive levels of RANK ligand. J Clin Invest 2000;106:1481–8. Zwerina J, Redlich K, Polzer K, Joosten L, Kronke G, Distler J, et al. TNF-induced structural joint damage is mediated by IL-1. Proc Natl Acad Sci U S A 2007;104:11742–7. Kitaura H, Zhou P, Kim HJ, Novack DV, Ross FP, Teitelbaum SL. M-CSF mediates TNF-induced inflammatory osteolysis. J Clin Invest 2005;115:3418–27. Wei S, Kitaura H, Zhou P, Ross FP, Teitelbaum SL. IL-1 mediates TNF-induced osteoclastogenesis. J Clin Invest 2005;115:282–90. Diarra D, Stolina M, Polzer K, Zwerina J, Ominsky MS, Dwyer D, et al. Dickkopf-1 is a master regulator of joint remodeling. Nat Med 2007;13:156–63. Redlich K, Hayer S, Ricci R, David JP, Tohidast-Akrad M, Kollias G, et al. Osteoclasts are essential for TNF-␣-mediated joint destruction. J Clin Invest 2002;110:1419–27. Zwerina J, Hayer S, Tohidast-Akrad M, Bergmeister H, Redlich K, Feige U, et al. Single and combined inhibition of tumor necrosis factor, interleukin-1, and RANKL pathways in tumor necrosis factor–induced arthritis: effects on synovial inflammation, bone erosion, and cartilage destruction. Arthritis Rheum 2004;50: 277–90. Finckh A, Simard JF, Duryea J, Liang MH, Huang J, Daneel S, et al. The effectiveness of anti–tumor necrosis factor therapy in preventing progressive radiographic joint damage in rheumatoid arthritis: a population-based study. Arthritis Rheum 2006;54:54–9. Bi W, Deng JM, Zhang Z, Behringer RR, de Crombrugghe B. Sox9 is required for cartilage formation. Nat Genet 1999;22:85–9. Bell DM, Leung KK, Wheatley SC, Ng LJ, Zhou S, Ling KW, et al. SOX9 directly regulates the type-II collagen gene. Nat Genet 1997;16:174–8. Furumatsu T, Tsuda M, Taniguchi N, Tajima Y, Asahara H. Smad3 induces chondrogenesis through the activation of SOX9 via CREB-binding protein/p300 recruitment. J Biol Chem 2005;280: 8343–50. Roman-Blas JA, Stokes DG, Jimenez SA. Modulation of TGF-␤ signaling by proinflammatory cytokines in articular chondrocytes. Osteoarthritis Cartilage 2007;15:1367–77. Joosten LA, Helsen MM, Saxne T, van De Loo FA, Heinegard D, van Den Berg WB. IL-1 ␣ ␤ blockade prevents cartilage and bone destruction in murine type II collagen-induced arthritis, whereas TNF-␣ blockade only ameliorates joint inflammation. J Immunol 1999;163:5049–55. Buttle DJ, Handley CJ, Ilic MZ, Saklatvala J, Murata M, Barrett AJ. Inhibition of cartilage proteoglycan release by a specific inactivator of cathepsin B and an inhibitor of matrix metalloproteinases: evidence for two converging pathways of chondrocytemediated proteoglycan degradation. Arthritis Rheum 1993;36: 1709–17. 3543 24. Stanton H, Rogerson FM, East CJ, Golub SB, Lawlor KE, Meeker CT, et al. ADAMTS5 is the major aggrecanase in mouse cartilage in vivo and in vitro. Nature 2005;434:648–52. 25. Valencia X, Higgins JM, Kiener HP, Lee DM, Podrebarac TA, Dascher CC, et al. Cadherin-11 provides specific cellular adhesion between fibroblast-like synoviocytes. J Exp Med 2004;200:1673–9. 26. Lee DM, Kiener HP, Agarwal SK, Noss EH, Watts GF, Chisaka O, et al. Cadherin-11 in synovial lining formation and pathology in arthritis. Science 2007;315:1006–10. 27. Kiener HP, Niederreiter B, Lee DM, Jimenez-Boj E, Smolen JS, Brenner MB. Cadherin 11 promotes invasive behavior of fibroblast-like synoviocytes. Arthritis Rheum 2009;60:1305–10. 28. Baeten D, Demetter P, Cuvelier C, Van Den Bosch F, Kruithof E, Van Damme N, et al. Comparative study of the synovial histology in rheumatoid arthritis, spondyloarthropathy, and osteoarthritis: influence of disease duration and activity. Ann Rheum Dis 2000; 59:945–53. 29. Baeten D, Peene I, Union A, Meheus L, Sebbag M, Serre G, et al. Specific presence of intracellular citrullinated proteins in rheumatoid arthritis synovium: relevance to antifilaggrin autoantibodies. Arthritis Rheum 2001;44:2255–62. 30. Baeten D, Demetter P, Cuvelier CA, Kruithof E, Van Damme N, De Vos M, et al. Macrophages expressing the scavenger receptor CD163: a link between immune alterations of the gut and synovial inflammation in spondyloarthropathy. J Pathol 2002;196:343–50. 31. Baeten D, Steenbakkers PG, Rijnders AM, Boots AM, Veys EM, De Keyser F. Detection of major histocompatibility complex/ human cartilage gp-39 complexes in rheumatoid arthritis synovitis as a specific and independent histologic marker. Arthritis Rheum 2004;50:444–51. 32. Baeten D, Kruithof E, De Rycke L, Vandooren B, Wyns B, Boullart L, et al. Diagnostic classification of spondylarthropathy and rheumatoid arthritis by synovial histopathology: a prospective study in 154 consecutive patients. Arthritis Rheum 2004;50: 2931–41. 33. De Rycke L, Nicholas AP, Cantaert T, Kruithof E, Echols JD, Vandekerckhove EM, et al. Synovial intracellular citrullinated proteins colocalizing with peptidyl arginine deiminase as pathophysiologically relevant antigenic determinants of rheumatoid arthritis–specific humoral autoimmunity. Arthritis Rheum 2005; 52:2323–30. 34. De Rycke L, Vandooren B, Kruithof E, De Keyser F, Veys EM, Baeten D. Tumor necrosis factor ␣ blockade treatment downmodulates the increased systemic and local expression of Toll-like receptor 2 and Toll-like receptor 4 in spondylarthropathy. Arthritis Rheum 2005;52:2146–58. 35. Kruithof E, Baeten D, De Rycke L, Vandooren B, Foell D, Roth J, et al. Synovial histopathology of psoriatic arthritis, both oligoand polyarticular, resembles spondyloarthropathy more than it does rheumatoid arthritis. Arthritis Res Ther 2005:7:R569–80. 36. Baeten D, Moller HJ, Delanghe J, Veys EM, Moestrup SK, De Keyser F. Association of CD163⫹ macrophages and local production of soluble CD163 with decreased lymphocyte activation in spondylarthropathy synovitis. Arthritis Rheum 2004;50:1611–23. 37. Van Kuijk AW, Reinders-Blankert P, Smeets TJ, Dijkmans BA, Tak PP. Detailed analysis of the cell infiltrate and the expression of mediators of synovial inflammation and joint destruction in the synovium of patients with psoriatic arthritis: implications for treatment. Ann Rheum Dis 2006;65:1551–7. 38. Baeten D, Kruithof E, De Rycke L, Boots AM, Mielants H, Veys EM, et al. Infiltration of the synovial membrane with macrophage subsets and polymorphonuclear cells reflects global disease activity in spondyloarthropathy. Arthritis Res Ther 2005;7:R359–69. 39. Kruithof E, De Rycke L, Vandooren B, De Keyser F, FitzGerald O, McInnes I, et al. Identification of synovial biomarkers of response to experimental treatment in early-phase clinical trials in spondylarthritis. Arthritis Rheum 2006;54:1795–804. 3544 40. Gordon S, Taylor PR. Monocyte and macrophage heterogeneity. Nat Rev Immunol 2005;5:953–64. 41. Vandooren B, Noordenbos T, Ambarus C, Krausz S, Cantaert T, Yeremenko N, et al. Absence of a classically activated macrophage cytokine signature in peripheral spondylarthritis, including psoriatic arthritis. Arthritis Rheum 2009;60:966–75. 42. Canete JD, Martinez SE, Farres J, Sanmarti R, Blay M, Gomez A, et al. Differential Th1/Th2 cytokine patterns in chronic arthritis: interferon ␥ is highly expressed in synovium of rheumatoid arthritis compared with seronegative spondyloarthropathies. Ann Rheum Dis 2000;59:263–8. 43. Gengenbacher M, Sebald HJ, Villiger PM, Hofstetter W, Seitz M. Infliximab inhibits bone resorption by circulating osteoclast precursor cells in patients with rheumatoid arthritis and ankylosing spondylitis. Ann Rheum Dis 2008;67:620–4. 44. Ritchlin CT, Haas-Smith SA, Li P, Hicks DG, Schwarz EM. Mechanisms of TNF-␣- and RANKL-mediated osteoclastogenesis and bone resorption in psoriatic arthritis. J Clin Invest 2003;111: 821–31. 45. Colucci S, Brunette G, Cantatore FP, Oranger A, Mori G, Quarta L, et al. Lymphocytes and synovial fluid fibroblasts support osteoclastogenesis through RANKL, TNF␣, and IL-7 in an in vitro model derived from psoriatic arthritis. J Pathol 2007;212:47–55. 46. Vandooren B, Melis L, Veys EM, Tak PP, Baeten D. In vitro spontaneous osteoclastogenesis of human peripheral blood mononuclear cells is not crucially dependent on T lymphocytes. Arthritis Rheum 2009;60:1020–5. 47. Vandooren B, Cantaert T, Noordenbos T, Tak PP, Baeten D. The abundant synovial expression of the RANK/RANKL/Osteoprotegerin system in peripheral spondylarthritis is partially disconnected from inflammation. Arthritis Rheum 2008;58:718–29. 48. Crotti TN, Smith MD, Weedon H, Ahern MJ, Findlay DM, Kraan M, et al. Receptor activator of NF-B ligand (RANKL) expression in synovial tissue from patients with rheumatoid arthritis, spondyloarthropathy, osteoarthritis, and from normal patients: semiquantitative and quantitative analysis. Ann Rheum Dis 2002;61: 1047–54. 49. Haynes DR, Barg E, Crotti TN, Holding C, Weedon H, Atkins GJ, et al. Osteoprotegerin expression in synovial tissue from patients with rheumatoid arthritis, spondyloarthropathies and osteoarthritis and normal controls. Rheumatology (Oxford) 2003;42:123–34. 50. Waterman EA, Cross NA, Lippitt JM, Cross SS, Rehman I, Holen I, et al. The antibody MAB8051 directed against osteoprotegerin detects carbonic anhydrase II: implications for association studies with human cancers. Int J Cancer 2007;121:1958–66. 51. Mundinger A, Ioannidou M, Meske S, Dinkel E, Beck A, Sigmund G. MRI of knee arthritis in rheumatoid arthritis and spondyloarthropathies. Rheumatol Int 1991;11:183–6. 52. Na KS, Kim TH, Inman RD. Biomarkers in spondyloarthritis. Curr Rheumatol Rep 2006;8:283–6. 53. Vandooren B, Cantaert T, ter Borg M, Noordenbos T, Kuhlman R, Gerlag D, et al. Tumor necrosis factor ␣ drives cadherin 11 expression in rheumatoid inflammation. Arthritis Rheum 2008;58: 3051–62. 54. Vandooren B, Kruithof E, Yu DT, Rihl M, Gu J, De Rycke L, et al. Involvement of matrix metalloproteinases and their inhibitors in peripheral synovitis and down-regulation by tumor necrosis factor ␣ blockade in spondylarthropathy. Arthritis Rheum 2004; 50:2942–53. 55. Kruithof E, De Rycke L, Roth J, Mielants H, Van den Bosch F, De Keyser F, et al. Immunomodulatory effects of etanercept on peripheral joint synovitis in the spondylarthropathies. Arthritis Rheum 2005;52:3898–909. 56. Kane D, Gogarty M, O’Leary J, Silva I, Bermingham N, Bresnihan B, et al. Reduction of synovial sublining layer inflammation and proinflammatory cytokine expression in psoriatic arthritis treated with methotrexate. Arthritis Rheum 2004;50:3286–95. VANDOOREN ET AL 57. Yang C, Gu J, Rihl M, Baeten D, Huang F, Zhao M, et al. Serum levels of matrix metalloproteinase 3 and macrophage colonystimulating factor 1 correlate with disease activity in ankylosing spondylitis. Arthritis Rheum 2004;51:691–9. 58. Francois RJ, Garder DL, Degrave EJ, Bywaters EG. Histopathologic evidence that sacroiliitis in ankylosing spondylitis is not merely enthesitis: systematic study of specimens from patients and control subjects. Arthritis Rheum 2000;43:2011–24. 59. Lories RJ, Derese I, Luyten FP. Modulation of bone morphogenetic protein signaling inhibits the onset and progression of ankylosing enthesitis. J Clin Invest 2005;115:1571–9. 60. Lories RJ, Derese I, Ceuppens JL, Luyten FP. Bone morphogenetic proteins 2 and 6, expressed in arthritic synovium, are regulated by proinflammatory cytokines and differentially modulate fibroblast-like synoviocyte apoptosis. Arthritis Rheum 2003; 48:2807–18. 61. Lories RJ, Derese I, De Bari C, Luyten FP. Evidence for uncoupling of inflammation and joint remodeling in a mouse model of spondylarthritis. Arthritis Rheum 2007;56:489–97. 62. McGonagle D, Gibbon W, O’Connor P, Green M, Pease C, Emery P. Characteristic magnetic resonance imaging entheseal changes of knee synovitis in spondylarthropathy. Arthritis Rheum 1998;41: 694–700. 63. McGonagle D, Gibbon W, Emery P. Classification of inflammatory arthritis by enthesitis. Lancet 1998;352:1137–40. 64. D’Agostino MA, Said-Nahal R, Hacquard-Bouder C, Brasseur JL, Dougados M, Breban M. Assessment of peripheral enthesitis in the spondylarthropathies by ultrasonography combined with power Doppler: a cross-sectional study. Arthritis Rheum 2003;48: 523–33. 65. Frediani B, Falsetti P, Storri L, Allegri A, Bisogno S, Baldi F, et al. Ultrasound and clinical evaluation of quadricipital tendon enthesitis in patients with psoriatic arthritis and rheumatoid arthritis. Clin Rheumatol 2002;21:294–8. 66. Genc H, Cakit BD, Tuncbilek I, Erdem HR. Ultrasonographic evaluation of tendons and entheseal sites in rheumatoid arthritis: comparison with ankylosing spondylitis and healthy subjects. Clin Rheumatol 2005;24:272–7. 67. Falsetti P, Frediani B, Fioravanti A, Acciai C, Baldi F, Filippou G, et al. Sonographic study of calcaneal entheses in erosive osteoarthritis, nodal osteoarthritis, rheumatoid arthritis and psoriatic arthritis. Scand J Rheumatol 2003;32:229–34. 68. Olivieri I, Salvarani C, Cantini F, Scarano E, Padula A, Niccoli L, et al. Fast spin echo–T2-weighted sequences with fat saturation in dactylitis of spondylarthritis: no evidence of entheseal involvement of the flexor digitorum tendons. Arthritis Rheum 2002;46:2964–7. 69. Marzo-Ortega H, Tanner SF, Rhodes LA, Tan AL, Conaghan PG, Hensor EM, et al. Magnetic resonance imaging in the assessment of metacarpophalangeal joint disease in early psoriatic and rheumatoid arthritis. Scand J Rheumatol 2009;38:79–83. 70. McGonagle D, Marzo-Ortega H, O’Connor P, Gibbon W, Hawkey P, Henshaw K, et al. Histological assessment of the early enthesitis lesion in spondyloarthropathy. Ann Rheum Dis 2002;61:534–7. 71. Laloux L, Voisin MC, Allain J, Martin N, Kerboull L, Chevalier X, et al. Immunohistological study of entheses in spondyloarthropathies: comparison in rheumatoid arthritis and osteoarthritis. Ann Rheum Dis 2001;60:316–21. 72. Benjamin M, Toumi H, Suzuki D, Hayashi K, McGonagle D. Evidence for a distinctive pattern of bone formation in enthesophytes. Ann Rheum Dis 2009;68:1003–10. 73. Benjamin M, Rufai A, Ralphs JR. The mechanism of formation of bony spurs (enthesophytes) in the Achilles tendon. Arthritis Rheum 2000;43:576–83. 74. McGonagle D, Wakefield RJ, Tan AL, D’Agostino MA, Toumi H, Hayashi K, et al. Distinct topography of erosion and new bone formation in Achilles tendon enthesitis: implications for under- MEDIATORS OF STRUCTURAL REMODELING IN PERIPHERAL SpA 75. 76. 77. 78. 79. 80. standing the link between inflammation and bone formation in spondylarthritis. Arthritis Rheum 2008;58:2694–9. Lories RJ, Derese I, Luyten FP. Inhibition of osteoclasts does not prevent joint ankylosis in a mouse model of spondyloarthritis. Rheumatology (Oxford) 2008;47:605–8. Kim TH, Stone M, Payne U, Zhang X, Ionescu M, Lobanok T, et al. Cartilage biomarkers in ankylosing spondylitis: relationship to clinical variables and treatment response. Arthritis Rheum 2005; 52:885–91. Vandooren B, Cantaert T, van Lierop MJ, Bos E, De Rycke L, Veys EM, et al. Melanoma inhibitory activity, a biomarker related to chondrocyte anabolism, is reversibly suppressed by proinflammatory cytokines in rheumatoid arthritis. Ann Rheum Dis 2009; 68:1044–50. 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. Briot K, Roux C, Gossec L, Charni N, Kolta S, Dougados M, et al. Effects of etanercept on serum biochemical markers of cartilage metabolism in patients with spondyloarthropathy. J Rheumatol 2008;35:310–4. Maksymowych WP, Poole AR, Hiebert L, Webb A, Ionescu M, 81. 82. 83. 84. 85. 3545 Lobanok T, et al. Etanercept exerts beneficial effects on articular cartilage biomarkers of degradation and turnover in patients with ankylosing spondylitis. J Rheumatol 2005;32:1911–7. Hammer RE, Maika SD, Richardson JA, Tang JP, Taurog JD. Spontaneous inflammatory disease in transgenic rats expressing HLA-B27 and human ␤2m: an animal model of HLA-B27associated human disorders. Cell 1990;30:1099–112. Tran TM, Dorris ML, Satumtira N, Richardson JA, Hammer RE, Shang J, et al. Additional human ␤2-microglobulin curbs HLA–B27 misfolding and promotes arthritis and spondylitis without colitis in male HLA–B27–transgenic rats. Arthritis Rheum 2006;54:1317–27. Appel H, Kuhne M, Spiekermann S, Ebhardt H, Grozdanovic Z, Kohler D, et al. Immunohistologic analysis of zygapophyseal joints in patients with ankylosing spondylitis. Arthritis Rheum 2006;54: 2845–51. Appel H, Kuhne M, Spiekermann S, Kohler D, Zacher J, Stein H, et al. Immunohistochemical analysis of hip arthritis in ankylosing spondylitis: evaluation of the bone–cartilage interface and subchondral bone marrow. Arthritis Rheum 2006;54:1805–13. Sieper J, Appel H, Braun J, Rudwaleit M. Critical appraisal of assessment of structural damage in ankylosing spondylitis: implications for treatment outcomes [review]. Arthritis Rheum 2008; 58:649–56.