Osteoarthritis (OA) causes pain and dysfunction and is the leading cause of disability in elderly people in industrialized countries (1). It results in breakdown of articular cartilage with concomitant changes in the underlying bone, development of osteophytes, and some degree of synovial inflammation (2). The extracellular matrix of cartilage is destroyed and the phenotype of chondrocytes altered due to changes in their pattern of gene expression. They lose their differentiated phenotype and undergo focal cell death and degeneration (3). Several epidemiologic studies have shown a positive association between obesity and hip and knee OA, highlighting the key role of mechanical loading on cartilage metabolism (4). An association between obesity and OA in non–weight-bearing joints, such as the hand, has also been described (5, 6). Moreover, the loss of body fat seems to be more important than the loss of body weight in improving the symptoms of OA (7). The increased fat mass in obesity may therefore alter the metabolism of articular tissues such as cartilage.
Obesity is a chronic metabolic disease that is becoming increasingly common in industrialized countries (8). Fat cells secrete a variety of proteins with the functional and structural properties of cytokines; these are termed “adipokines.” Adiponectin, leptin, and resistin are the most abundant adipokines produced by adipose tissue, and production of leptin and resistin is increased in obese individuals (8). Adipokines have autocrine, paracrine, and endocrine effects and may be an important link between the immune response and metabolism, predisposing individuals to increased risk of disease (9, 10). For example, the increased level of circulating leptin in obese patients is positively correlated with increased concentrations of interleukin-1 (IL-1) receptor antagonist, IL-6, and tumor necrosis factor α (TNFα) in serum, and with type 2 diabetes and cardiovascular disease (11, 12). Interestingly, plasma levels of leptin, adiponectin, and resistin are increased in arthropathies, such as rheumatoid arthritis (RA) (13, 14) and OA (15). Presle and colleagues demonstrated that adipokines were expressed by tissue from various OA-affected joints and that levels of adipokines in serum did not correspond to levels in synovial fluid (15). Therefore, it appears that the joint cavity is the site at which each adipokine's expression is individually regulated.
One recently described adipokine is visfatin. It is secreted by mature adipocytes, and its plasma concentration markedly increases in parallel with the amount of visceral fat. Visfatin exerts insulin-mimetic effects in vivo and in vitro (16). This 52-kd protein binds to and activates insulin receptor, with a similar binding equilibrium dissociation constant. Mutation of the extracellular α-subunit of insulin receptor abrogates binding of insulin, but not of visfatin. Therefore, visfatin activates insulin receptor in a manner distinct from its activation by insulin, but the two proteins trigger the same signaling pathways in adipocytes (16).
Visfatin was previously identified as a secreted growth factor for early B lymphocytes (pre–B cell colony-enhancing factor [PBEF]). Several tissues produce visfatin/PBEF, including skeletal muscle, liver, and bone marrow (17). It is pleiotropic and regulates both inflammatory and immune responses. Its synthesis is regulated by cytokines such as TNFα, IL-1β, and IL-6, by lipopolysaccharide (LPS), and by dexamethasone (18–20). Visfatin is overproduced in colorectal cancer and is increased in acute lung injury, where it is a potential biomarker (21). It is produced by LPS-induced neutrophils and inhibits their apoptosis via a caspase 3– and caspase 8–mediated mechanism (18). Moreover, visfatin was recently found in foam cell macrophages within unstable atherosclerotic lesions, where it was involved in plaque destabilization (22). Finally, plasma concentrations of visfatin are increased in patients with RA (13). However, the effect of visfatin on cartilage physiology remains completely unknown.
Visfatin is also called Nampt because of its nicotinamide phosphoribosyltransferase (NAmPRTase) activity. Visfatin/Nampt has been implicated in the synthesis of NAD, an essential cofactor for cell metabolism. Rongvaux and colleagues showed that during polyclonal immune responses, visfatin/PBEF production is increased in lymphocytes, and could stimulate their proliferation (23). Visfatin is present in both the cytoplasm and the nucleus in rat and mouse cell lines, where its concentration depends upon the cell cycle phase (24). Finally, Nampt appears to be a longevity gene which adds stress-resistant life to human smooth muscle cells (25). Therefore, visfatin is a multifunctional protein whose intracellular and extracellular effects are dissimilar.
Cartilage degradation in OA is due in part to increased release of catabolic mediators, such as IL-1β and prostaglandin E2 (PGE2) (2, 26). IL-1β concentrations are markedly elevated in the synovial fluid of patients with RA (27) and OA (28). PGE2 is a prostanoid derived from arachidonic acid, and its production in the setting of inflammation depends on the coordinated activities of cyclooxygenase 2 and microsomal PGE synthase 1 (mPGES-1). Collagen-induced arthritis has been shown to be significantly less severe, and cartilage better preserved, in mice lacking mPGES-1 than in control mice (29). The synthesis of mPGES-1 by chondrocytes from OA patients is also increased by IL-1β (30). In vivo, PGE2 is rapidly converted to an inactive metabolite by 15-hydroxy–PG dehydrogenase (15-PGDH) (31). Synthesis of 15-PGDH is subnormal in various diseases, including cancer (32) and inflammatory bowel disease (33). We recently reported that 15-PGDH is synthesized by mouse cartilage (34), but its regulation in this tissue has not been characterized.
Increased release of these catabolic mediators in OA cartilage triggers the expression of matrix metalloproteinases (MMPs) and ADAMTS (also called aggrecanases) (35). MMPs are zinc-dependent endopeptidases that can break down all kinds of extracellular matrix proteins and are probably involved in both normal turnover and breakdown under disease conditions. MMP-13 (collagenase 3) and MMP-3 (stromelysin 1) are implicated in degradation of the main constituents of cartilage matrix, type II collagen and aggrecans, whereas ADAMTS-4 and ADAMTS-5 are aggrecanases most efficient at degrading aggrecans (35). Their production is increased in experimental models of OA and in articular cartilage from OA patients (36, 37).
In the present study we investigated the role of visfatin in cartilage metabolism, focusing on inflammation and matrix degradation, and its possible implication in OA. We found that visfatin is produced by human OA chondrocytes and that this synthesis is increased by IL-1β. Visfatin also triggers PGE2 synthesis by increasing production of mPGES-1 and reducing expression of 15-PGDH. Moreover, it increases the synthesis and release of MMP-3, MMP-13, ADAMTS-4, and ADAMTS-5 by chondrocytes and decreases aggrecan production. Finally, inhibition of visfatin synthesis reduces IL-1β–induced PGE2 synthesis.
- Top of page
- MATERIALS AND METHODS
- AUTHOR CONTRIBUTIONS
Visfatin is a newly identified adipokine that is produced by adipose tissue, bone marrow, and skeletal muscle (16). It was originally discovered as a growth factor for B lymphocyte precursors, and is alternatively called PBEF (17). Visfatin/PBEF exerts actions in pathophysiologic processes such as colorectal cancer, acute lung injury, and atherosclerosis. In the present work, we studied its extracellular role in cartilage homeostasis. Our findings reveal that visfatin is produced by chondrocytes from patients with OA, and that its production is increased by IL-1β. Moreover, visfatin is a potent inducer of PGE2 release in both human and immature mouse articular chondrocytes, as a result of increased mPGES-1 and decreased 15-PGDH synthesis. Finally, visfatin triggers the synthesis and release of MMP-3, MMP-13, ADAMTS-4, and ADAMTS-5 by chondrocytes and reduces HMW aggregated PGs. This demonstration of the potent proinflammatory and prodegradative effects of visfatin suggests that it also contributes to the progression of OA.
Interestingly, blocking of visfatin synthesis reduces the effect of IL-1β on PGE2 release in chondrocytes. Our working hypothesis of the interactive effect of IL-1β and visfatin on PGE2 synthesis in cartilage is illustrated in Figure 6. Increased mPGES-1 synthesis and the subsequent release of PGE2 by chondrocytes stimulated with IL-1β has been well described (30). We showed in this study that IL-1β induced a dramatic decrease in 15-PGDH levels, enhancing this effect. IL-1β stimulated the synthesis of visfatin by chondrocytes, and visfatin increased the production of mPGES-1 and reduced that of 15-PGDH, leading to the release of excess PGE2 from articular chondrocytes. Therefore, after IL-1β–triggered release of visfatin into the medium, an autocrine/paracrine loop would allow visfatin to act on the chondrocytes themselves via its receptor, and to modify the production of mPGES-1 and 15-PGDH in the same way that IL-1β does. Interestingly, a similar mechanism was not found with regard to MMP-3 and MMP-13 expression. The receptors for visfatin are insulin receptor and insulin-like growth factor 1 (IGF-1) receptor. Insulin receptor has not yet been found in chondrocytes, and affinity of the IGF-1 receptor for visfatin is too low to trigger an intracellular signal (16). We are presently working to identify the signaling pathway of visfatin in chondrocytes.
Figure 6. Hypothetical role of visfatin in synthesis of PGE2, matrix metalloproteinase 3 (MMP-3), MMP-13, ADAMTS-4, and ADAMTS-5 by articular chondrocytes. In this model, expression of visfatin (mRNA and protein) in chondrocytes is increased by IL-1β (after binding to its receptor [IL-1R]). Visfatin then exerts action on the chondrocytes, in an autocrine manner. Extracellular visfatin, after binding to its receptor (which remains to be identified), increases mPGES-1 mRNA levels and reduces 15-PGDH mRNA levels, and subsequently triggers PGE2 release. In addition, extracellular visfatin triggers the synthesis and release of MMP-3, MMP-13, ADAMTS-4, and ADAMTS-5 by chondrocytes. Therefore, visfatin appears to act as a mediator of inflammation and cartilage breakdown. See Figure 2 for other definitions.
Download figure to PowerPoint
New concepts to explain the pathophysiology of OA have recently emerged. While OA is usually considered to be cartilage driven, other tissues, such as bone, muscle, and juxtaarticular adipose tissue, may also be involved. Therefore, OA could be a systemic disorder in which altered lipid homeostasis is a predisposing factor (42). Recent studies on the link between obesity and OA reveal potential roles of adipokines, such as leptin, adiponectin, and resistin, in cartilage degradation (15, 43, 44). Presle and colleagues recently reported that concentrations of adipokines in the serum do not reflect adipokine concentrations in the joint space and suggested that OA is associated with a specific local dysregulation of adipokines (15). The present study shows that human OA chondrocytes produce visfatin and that its production is increased by IL-1β, enforcing this hypothesis. Moreover, the visfatin gene is activated by hypoxia in a human breast cancer cell line, involving functional hypoxia-responsive element sites located within the proximal promoter region (45). Visfatin could be a critical factor in cartilage degradation, given that mature cartilage is an avascular tissue.
Visfatin is a highly conserved protein, and the death of mice lacking visfatin during embryogenesis (16) highlights its physiologic importance. Visfatin has separate extracellular and intracellular roles, and its implication in pathophysiology remains unclear. Two extracellular functions of visfatin have been identified. First, it acts as an insulin analog in glucose homeostasis (16). Second, it may be a mediator of late-stage inflammation: even though it does not have a signal peptide for its secretion and is unlike other known cytokines (46), its synthesis is increased in severely infected tissue (18, 46). Our data show that visfatin levels in chondrocytes increase in response to IL-1β and act in an autocrine/paracrine manner to trigger PGE2 synthesis. The human visfatin gene has various regulatory elements in the 5′-upstream region, such as binding sites for NF-κB and activator protein 1, the two major signaling pathways activated by IL-1β (46). Moreover, visfatin is an antiapoptotic mediator in neutrophils in both experimental inflammation and clinical sepsis (18). Apoptotic death of articular chondrocytes has been implicated in the pathogenesis of OA. PGE2 may sensitize chondrocytes to the cell death induced by nitric oxide (47) and can directly trigger apoptosis of bovine chondrocytes via cAMP signaling (48). Further experiments are needed to decipher the role of visfatin in chondrocyte apoptosis.
Visfatin has key intracellular roles, as evidenced by the implication of its NAmPRTase activity in cell proliferation and differentiation, demonstrated in B cells (23). Therefore, visfatin may act on chondrocyte differentiation in immature mouse articular chondrocytes. However, we did not observe any modification of the main marker of chondrocyte differentiation, i.e., the ratio of type II collagen mRNA to type I collagen mRNA. Thus, although we cannot definitively rule out a role of visfatin in the chondrocyte differentiation process, our results do not support such a role.
Several lines of evidence demonstrate that adipokines, released by white adipose tissue, participate in a wide variety of physiologic and pathophysiologic processes, including immunity and inflammation. Therefore, adipokines seem to link metabolic disorders to inflammatory and/or autoimmune conditions (49). It has been suggested that changes in lipid metabolism are involved in the diverse physiologic changes in generalized OA (50). However, there has been little reported experimental evidence to support this. We believe that the present results are the first demonstration of a catabolic role of visfatin, a newly discovered adipokine, in OA. Further studies are needed for more precise elucidation of the proinflammatory activities of visfatin in osteoarthritic joints and the contribution of each tissue to its synthesis. Broader knowledge of the deleterious action of visfatin in OA may lead to its emergence as a novel target for therapy.