Interleukin-17 (IL-17) is a T cell–derived cytokine produced by activated T cells, predominantly activated CD4+,CD45RO+ memory T cells (1, 2). This cytokine may play a role in T cell–triggered inflammation by stimulating stromal cells to secrete various cytokines and growth factors associated with inflammation (1–4). A pathogenic role for IL-17 was found in organ allograft rejection (5), and increased IL-17 expression was detected in several diseases, such as systemic sclerosis (6), nephrotic syndrome (7), systemic lupus erythematosus (8), and rheumatoid arthritis (RA) (9, 10). In contrast with the restricted expression of IL-17, the IL-17 receptor is ubiquitously expressed in virtually all cells and tissues. It is a type I transmembrane protein that has no sequence similarity with any other known cytokine receptor (3). Binding of IL-17 to its unique receptor results in activation of the adaptor molecule tumor necrosis factor (TNF) receptor–associated factor 6, which is required for IL-17 signaling (11).
RA is considered a systemic Th1-associated inflammatory joint disease that is characterized by chronic synovitis and destruction of cartilage and bone. T cells represent a large proportion of the inflammatory cells invading the synovial tissue. Since the etiology of RA is still unknown, regulating the cytokine imbalance might represent an effective way to control this disease. The proinflammatory cytokines TNFα and IL-1β play a crucial role in the pathology of arthritis, driving enhanced production of cytokines, chemokines, and degradative enzymes (12). In vivo studies have shown that neutralizing TNFα or IL-1β controls chronic inflammation and cartilage degradation, respectively (13–15). Consistent with this, clinical studies revealed efficacy after blocking of TNFα or IL-1β. However, a subset of patients did not respond to these inhibitors, and none of the treatments cured the disease. Therefore, it is tempting to speculate that cytokines or factors other than IL-1β and TNFα also participate in the proinflammatory cytokine cascade.
T cell cytokine IL-17 is spontaneously produced by RA synovial membrane cultures (9), and high levels have been detected in the synovial fluid of patients with RA (9, 10). IL-17 can stimulate the production of IL-1β and TNFα from macrophages (4) and triggers human synoviocytes to produce IL-6, IL-8, granulocyte–macrophage colony-stimulating factor, and prostaglandin E2 (2, 16), suggesting that IL-17 could be an upstream mediator in the pathogenesis of arthritis. Early neutralization of endogenous IL-17 prior to the development of arthritis in the experimental arthritis model suppresses the onset of disease (17, 18). Furthermore, IL-17 may be involved in tissue destruction. IL-17 has biologic activities similar to those of IL-1β, and additive/synergistic effects with IL-1β and TNFα have been reported (19). In vitro, IL-17 suppresses matrix synthesis by articular chondrocytes through enhancement of nitric oxide (NO) production (20, 21). In addition, in vitro studies suggested a role for IL-17 in bone erosion by induction of receptor activator of NF-κB ligand (RANKL) expression (22). Recently, we showed that IL-17 promotes bone erosion in murine collagen-induced arthritis (CIA) through loss of the RANKL/osteoprotegerin (OPG) balance (23). These observations indicate that IL-17 may promote joint inflammation as well as tissue destruction during the initial phase of arthritis. However, the role of T cell IL-17 during the effector phase of arthritis has still not been identified.
In the present study, we demonstrated the therapeutic effect of anti–IL-17 antibody treatment in CIA, implying that the T cell cytokine IL-17 not only plays a role in the early stage of arthritis, but also has a function in propagating and prolonging the arthritis. Furthermore, fewer synovial IL-1β–positive and RANKL-positive cells were found after treatment with neutralizing endogenous IL-17. This suggests that IL-17 might be a novel target for the treatment of destructive arthritis and implies that neutralization of this T cell factor during the effector phase of arthritis has therapeutic potential. Our data suggest that anti–IL-17 cytokine therapy is an interesting new approach that may contribute to the prevention of joint destruction and could provide an important additional strategy to the current anti-TNF and anti–IL-1β therapy for RA.
- Top of page
- MATERIALS AND METHODS
This is the first study to demonstrate that neutralizing T cell IL-17 during the effector phase of CIA has therapeutic potential. Radiologic and histologic analyses revealed significant protection from joint damage when endogenous IL-17 was neutralized after arthritis expression. This protective effect was associated with down-regulation of IL-1β and RANKL in the synovium. Furthermore, neutralizing endogenous IL-17 at a later stage of CIA still slowed the progression of the disease.
IL-17 shares many properties with IL-1β and TNFα, and this T cell–derived cytokine has been shown to be present in the synovium of RA patients (9, 10). IL-17 induces the production of proinflammatory mediators, such as IL-1β and TNFα, from several joint cells including synovial fibroblasts, macrophages, and chondrocytes. In addition, IL-17 induces RANKL expression (22, 23), which is a novel cytokine crucial for osteoclastogenesis (28). Moreover, IL-17, IL-1β, TNFα, and RANKL activate the common transcription factor NF-κB in a variety of cells. IL-17 can synergize with these cytokines (IL-1β, TNFα, and RANKL), but probably has direct activity as well (17, 29). Since a subset of RA patients does not respond to neutralizing IL-1β or TNFα, other factors, such as IL-17, may participate in the proinflammatory cytokine cascade. In the present study, we showed the involvement of IL-17 in joint inflammation and prolongation of arthritis.
IL-17 has dual effects on cartilage. In vitro, it inhibits chondrocyte metabolism in intact articular cartilage of mice and results in proteoglycan breakdown (19, 20, 30, 31). Furthermore, in vitro studies show the induction of metalloproteinases by IL-17 in synoviocytes and chondrocytes (32–34). Interestingly, the effects of IL-17 on matrix degradation and synthesis were not dependent on IL-1β production by chondrocytes, and IL-1 receptor antagonist did not block IL-17–induced matrix release nor did it prevent the inhibition of matrix synthesis in vitro using porcine articular cartilage explants (31). Moreover, we recently identified an IL-1β–independent role of IL-17 in the pathogenesis of arthritis in vivo (17). The downstream signaling pathways for IL-17 and IL-1β seem to be distinct, and differential activation of activator protein 1 members by IL-17 and IL-1β has been described (33). IL-1β is by far the more catabolic cytokine in experimental arthritis compared with TNFα; however, IL-17 synergizes with TNFα to induce cartilage destruction in vitro (35). This underscores the potential of IL-17 to act additively or even synergistically with IL-1β/TNF, but IL-17 may have direct catabolic effects as well. In the present study, we found a clear reduction of synovial IL-1β expression after neutralizing endogenous IL-17 in CIA, indicating that IL-17 is an upstream mediator of IL-1β. Furthermore, we showed IL-17 to be a novel target for the treatment of cartilage destruction in experimental arthritis.
IL-17 seems to be a potent stimulator of osteoclastogenesis (22, 23). In the present study, we found reduced multinucleated cells after neutralizing endogenous IL-17, indicating that anti–IL-17 antibody treatment prevents the formation of osteoclast-like cells. Recently, it was shown that overexpression of IL-17 in the knee joints of CIA mice results in elevated expression of RANKL in the synovium and loss of the RANKL/OPG balance, leading to enhanced bone resorption (23). Of interest, systemic OPG treatment inhibits the local IL-17–induced bone resorption in the knee joints of CIA mice, suggesting that IL-17–induced bone erosion is at least partly mediated by RANKL (23). The observation that neutralizing IL-17 results in fewer RANKL-positive cells in the synovium further implies a relation between IL-17 and RANKL expression. Furthermore, it underscores the role of IL-17 in enhancement of the joint destruction process and makes IL-17 an attractive target for the treatment of destructive arthritis. Osteoclasts are potent bone-resorbing cells and play a crucial role in joint destruction (36). RANKL and TNFα contribute to osteoclast formation, while several other cytokines are responsible for osteoclast survival and/or activation. Neutralizing the RANKL/RANK pathway by administration of OPG prevents bone destruction (37–39). However, this kind of treatment is not antiinflammatory or chondroprotective, as shown in the present study with anti–IL-17, suggesting anti–IL-17 as a more appropriate therapy for destructive arthritis.
In the present study, we used an anti–IL-17 antibody to neutralize endogenous IL-17 directly after onset and during a later stage of CIA. A single injection with anti–IL-17 antibody seems to be much more efficient than treatment with the soluble IL-17 receptor (sIL-17R):Fc fusion protein, as previously described (17, 18). Since the treatment protocols are not identical between our previous study and the present study, we also treated CIA mice with the sIL-17R:Fc fusion protein starting after the onset of CIA (4 injections on alternative days using the same dose as described in ref. 17). This treatment did not result in suppression of the arthritis score; however, radiographic analysis revealed a significant reduction of the degree of joint destruction. In addition, semiquantitative analysis of messenger RNA (mRNA) expression for IL-1β and RANKL using reverse transcriptase–polymerase chain reaction showed down-regulation of IL-1β and RANKL mRNA expression in the synovium after sIL-17R:Fc treatment (Lubberts E, et al: unpublished observations). It is known that despite the ability of IL-17 to signal at low concentrations, it shows a low affinity to its receptor, with Ka values between 2.107 and 2.108M−1 (40). Therefore, we speculate that the difference in affinity for IL-17 between the IL-17R:Fc and the anti-murine IL-17 antibodies may be an important reason for the higher efficacy using the anti–IL-17 antibody. Affinity studies must be performed to prove this hypothesis, and this is currently under investigation.
The role of T cell cytokines such as IL-17 in propagating and prolonging arthritis must be identified. T cells and their cytokines may play an important role in initiating the arthritis and during an early phase. However, during the later stage of the arthritis, T cell cytokines may be overruled by mediators produced by activated macrophages. It has been shown that IL-17 plays an inflammatory role in the initial phase of experimental arthritis (17, 18). The present study makes it clear that after the first clinical signs of arthritis, neutralizing endogenous IL-17 is still of therapeutic value. Systemic IL-6 levels were reduced and fewer synovial IL-1β–positive and RANKL-positive cells were detected, suggesting both IL-1β–dependent and IL-1β–independent mechanisms of action. Furthermore, even at a later stage of CIA, T cell IL-17 contributes to prolongation of the arthritis, since blocking endogenous IL-17 in this phase of CIA slowed the progression of the disease. This suggests that despite the abundant expression of macrophage mediators, which are partly produced independently of IL-17, T cell IL-17 plays a role in maintaining the inflammation.
We speculate that neutralizing IL-17 during this later stage of CIA has a suppressive effect on proinflammatory cytokine production. In addition, fewer additive/synergistic effects between IL-17 and other proinflammatory cytokines such as TNFα, IL-1β, and IL-6 can be expected. Previous blocking studies with anti–IL-1β and anti-TNFα performed in our laboratory have shown that TNFα plays an important role in early CIA, and IL-1β is important in early and established CIA (15). IL-17 is a potent inducer of IL-1β (17), and we hypothesize that neutralizing IL-17 results in a reduction of IL-1β expression in the synovium during this later stage of CIA. Since IL-1β expression is not completely dependent on IL-17, not all IL-1β will be blocked, and this IL-17–independent IL-1β production will contribute to the progression of CIA. Fewer additive/synergistic effects between IL-17 and IL-1β may also play a role in slowing the progression of the disease. TNFα is hardly detectable in later stages of CIA; however, synergistic effects between TNFα and IL-17 have been documented (35). Blocking of IL-17 using the soluble receptor for IL-17 further improves the neutralizing effect of TNFα blocking after the onset of CIA (Lubberts E, et al: unpublished observations).
We hypothesize that the mechanisms responsible for slowing the progression of the disease after neutralizing IL-17 during the later stage of CIA are suppression of proinflammatory cytokines such as IL-1β, TNFα, and IL-6, and elimination or reduction of the additive/synergistic effects between IL-17 and these proinflammatory cytokines. Studies are currently being done in our laboratory to further prove this hypothesis.
In summary, we have demonstrated the therapeutic potential of neutralizing T cell IL-17 during the effector phase of CIA. IL-17 seems to play a role in prolonging the arthritis process and may be considered to be an important target for the treatment of destructive arthritis. IL-17 induced key catabolic cytokines such as IL-1β and RANKL. Since it is known that this T cell factor can have synergistic effects with catabolic/inflammatory mediators (29), it is tempting to speculate that IL-17 levels can make the difference in whether an RA patient will respond to anti–TNFα/anti–IL-1β therapy. Our data strongly suggest that anti–IL-17 cytokine therapy is an interesting new antirheumatic approach that will contribute to the prevention of joint destruction. Furthermore, neutralizing IL-17 could provide an additional therapeutic strategy for RA, particularly in situations where elevated levels of IL-17 may attenuate the response of a patient to anti-TNFα/anti–IL-1β therapy.