Discussion
IL-17 was first described as a T cell product with proinflammatory properties [5,22]. RA is characterized by hyperplasia of synovial lining cells and an intense infiltration by mononuclear cells [23]. Proinflammatory cytokines such as IL-1 and TNF-α are abundant in rheumatoid synovium, whereas the T cell-derived cytokines, especially IL-4 and interferon-γ, have often proved difficult to detect in RA synovium [24]. Although T cells may have a role in the augmentation of rheumatoid synovial inflammation, the lack of T cell-derived cytokines has limited its importance. In this respect, IL-17 is appealing because it has been described as a T cell-derived cytokine with proinflammatory properties.
In our studies, we tried to evaluate how IL-17 production is regulated in RA PBMC, and which signaling pathway it used. Levels of IL-17 were found to be higher in RA synovial fluid than in OA synovial fluid [15]. However, there are few data available on the agents that stimulate IL-17 production in RA, although the highest level of IL-17 production can be achieved by anti-CD3/anti-CD28 stimulation in healthy individuals [25]. In our experiments, PHA as mitogens, as well as anti-CD3/anti-CD28 for signaling through the T cell receptor, increased IL-17 production from RA PBMC in a dose-dependent manner. We found, by a cell proliferation assay (data not shown), that this upregulation of IL-17 might be due to increased cellular activity rather than to cellular proliferation.
IL-17 is produced mainly by activated CD4+ T cells, especially for Th1/Th0 cells, not the Th2 phenotype [26]. However, it can also be produced by CD8+ T cells via an IL-23 triggering mechanism in Gram-negative pulmonary infection [14]. In addition, IL-17 production was significantly augmented by T cells recognizing type II collagen in a collagen-induced arthritis model [27]. A complex interaction between cells in inflamed RA joints might produce a variety of proinflammatory cytokines and chemokines, which also activate other cells in the joints. For example, IL-17 stimulates rheumatoid synoviocytes to secrete several cytokines such as IL-6, IL-8 and tumor necrosis factor-stimulated gene 6 as well as prostaglandin E2 in vitro [12,28,29]. There are as yet few data available on the agents that stimulate IL-17 production in RA, although some cytokines (IL-15 and IL-23) have been known to regulate IL-17 production [13,14]. We therefore investigated the in vitro production of IL-17 in RA PBMC responding to a variety of cytokines/chemokines and mitogens as well as T cell receptor (TCR) ligation using anti-CD3/anti-CD28. Our studies demonstrated that IL-15 and MCP-1 as well as TCR ligation significantly increased the production of IL-17 in RA PBMC. Adding IL-15 or MCP-1 to TCR ligation augmented IL-17 production more markedly. In contrast, IL-1 and TNF-α, which are known to have proinflammatory properties and to be increased in RA joints, did not affect IL-17 production. Our data were consistent with a recent report that IL-15 triggered in vitro IL-17 production in PBMC, but TNF-α did not do so [13]. Although there were no data that MCP-1 directly induces T cell activation, it might exert effects indirectly on T cells through the activation of monocytes/macrophages in PBMC cultures. As reported for normal individuals [25], T cell activation through anti-CD3/anti-CD28 also increases IL-17 induction in RA PBMC.
Although the signaling pathway for the induction of cytokines/chemokines by IL-17 has been documented widely [8,30,31], no data have been available on how IL-17 production can be regulated by certain signaling pathways. By using signal transduction inhibitors, we therefore examined which signaling pathway was mainly involved in the induction of IL-17 in RA PBMC.
We identified that anti-CD3-induced IL-17 production in RA PBMC was significantly hampered by the PI3K inhibitor LY294002 and the NF-κB inhibitor PDTC to comparable levels of basal production without stimulation. We also found that anti-CD3-induced IL-17 production was downregulated by the addition of SB203580, a p38 MAPK inhibitor. It is interesting that a series of evidence supports crosstalk between NF-κB and p38. In myocytes, IκB kinase-β is activated by p38 [32], and the activated p38 can stimulate NF-κB by a mechanism involving histone acetylase p300/CREB-binding protein [33]. Our results revealed that p38 MAPK activation was not affected by LY294002, whereas NF-κB binding activity was decreased by LY294002, which provided the evidence for a p38 MAPK pathway independent of PI3K activation. The direct relationship between p38 and NF-κB for IL-17 production needs to be studied in future experiments.
The search for a downstream pathway of PI3K seemed to have a maximal response of Akt activation at 1 hour and a gradual loss of activity at 2 hours. The fact that Akt is phosphorylated upon anti-CD3 stimulation suggests the possible involvement of PI3K in the induction of IL-17 in RA. In view of the fact that NF-κB was also activated by anti-CD3/anti-CD28, IL-15 or mitogens in our experiments, it is most likely that the NF-κB pathway is also actively involved in the induction of IL-17 in RA PBMC. In contrast, the AP-1 signal transduction pathway, another important signaling pathway for cytokines/chemokines, was not activated in our experiments (data not shown). Although PI3K and its downstream kinase Akt in association with NF-κB have been reported to deliver activating signals in many cell types, the data on the signal inducing IL-17 are lacking. Our data clearly demonstrated that PI3K/Akt and resultant NF-κB activation could be an important arbitrator of the upregulation of IL-17 in RA, on the basis of our experiments showing simultaneous blocking of NF-κB binding activity in the IL-17 promoter by PDTC and LY294002. Considering its proinflammatory activities and successful induction of anti-IL-17 for ameliorating arthritis in animal models [2,6,34-36], understanding the IL-17 signaling pathway is an important element of developing new targeted therapies in RA.