Rheumatoid arthritis (RA) is a chronic inflammatory disease marked by synovial hyperplasia and joint destruction. The synovial intimal lining contributes to the damage by releasing prodigious amounts of proteases that degrade cartilage and bone. Signaling pathways, especially the mitogen-activated protein kinases (MAPKs), that regulate production of these enzymes have been defined and represent potential therapeutic targets (1–3). The MAPKs are serine/threonine kinases that mediate signal transduction and orchestrate cellular responses to environmental stress. In mammalian cells, 3 main MAPK pathways have been identified, including extracellular signal–regulated kinase (ERK), p38 MAPK, and c-Jun N-terminal kinase (JNK). Among this family of signaling kinases, JNK is thought to be especially important for matrix metalloproteinase (MMP) expression in RA by virtue of its ability to phosphorylate c-Jun and activate the transcription factor activator protein 1 (AP-1) (4, 5). Multiple MAPK pathways can be simultaneously activated, and their relative balance is determined by upstream kinase cascades known as MAPK kinases (MAPKKs or MKKs) (6).
Our previous studies demonstrated that interleukin-1 (IL-1) is a potent inducer of JNK phosphorylation and MMP gene expression in cultured synoviocytes (7). Furthermore, a selective JNK inhibitor, SP600125, blocks MMP production and bone destruction in rat adjuvant-induced arthritis (AIA) (8). These results suggest that JNK blockade might be effective in RA, and several compounds are currently being developed (9, 10). An alternative approach might be to target the upstream kinases that regulate JNK, and thereby permit activation of JNK under some circumstances but modulate its activity in cytokine-activated synoviocytes. JNK is activated by phosphorylation of tyrosine and threonine residue in the TPY (Thr-Pro-Tyr) motif by 2 dual-specific MKKs, MKK-4 and MKK-7, respectively, and these 2 kinases act synergistically to activate JNK (11). Either one could potentially serve as a therapeutic target in RA.
A variety of regulatory functions have been defined for MKK-4 and MKK-7, and distinct cellular stresses or cytokines can differentially activate MKK-4 or MKK-7 (12). However, little is known about the pharmacology of MAPK kinase inhibition in primary human cells. We have previously shown that MKK-4 and MKK-7 are phosphorylated in RA synovial tissue (13), and that both can be activated to form a signaling complex in cultured fibroblast-like synoviocytes (FLS) (13). In the present study, we used MKK-4 and MKK-7 small interfering RNA (siRNA) to dissect their respective roles in synoviocyte JNK signaling and gene expression. Our data demonstrate that MKK-7 is the pivotal kinase that regulates JNK in IL-1–stimulated FLS.
- Top of page
- MATERIALS AND METHODS
Joint destruction in RA is mediated, in part, by MMPs that degrade the articular extracellular matrix (17). The production of these enzymes is regulated by cytokines, such as IL-1β, that activate a broad array of cell signaling mechanisms. MAPKs are especially important, and 3 families (ERK, JNK, and p38) contribute by phosphorylating key transcription factors that are required for MMP transcription, such as AP-1 (4, 5). JNK, in particular, plays a pivotal role in cytokine-mediated c-Jun activation, AP-1 induction, and MMP gene expression in FLS (7). Previous studies indicated that JNK-2 is the most abundant isoform in human FLS (7), although cultured synoviocytes from either JNK-1– or JNK-2–knockout mice have defective MMP-3 production (8). The JNK inhibitor SP600125, which inhibits all 3 JNK isoforms, also blocks MMP expression and AP-1 activation in FLS and prevents joint destruction in rat AIA (8). These data suggest that the JNK cascade is involved in the regulation of MMPs and represents a potential therapeutic target in RA (18).
The MAPKs are regulated by upstream kinases called MAPK kinases, 2 of which (MKK-4 and MKK-7) are responsible for phosphorylating JNK. The present study was designed to evaluate the relative contribution of these 2 upstream kinases on cytokine-driven JNK activation and AP-1–driven gene expression in primary human FLS. We have previously shown that both MKKs are phosphorylated in rheumatoid synovial lining and are constitutively expressed in cultured FLS (13). MKK-4 and MKK-7 also form a stable complex with JNK in FLS that can be activated by IL-1 to induce MMP expression (13). Recent advances that permit high transfection efficiency enabled us to dissect this pathway with siRNA knockdown technology (19). Using MKK-4 and MKK-7 siRNA, we assessed their individual roles in JNK activation and JNK-mediated gene transcription.
Our studies to assess MAPK kinase function in FLS demonstrate that MKK-7, but not MKK-4, contributes to IL-1–induced JNK phosphorylation, JNK function, AP-1 binding, AP-1–driven transcription, and MMP-3 production. Although MKK-4 is a component of the JNK signalsome in FLS and is readily phosphorylated by IL-1 stimulation (13), it makes no detectable contribution to JNK activation in this system. As a positive control, we also evaluated non–receptor-mediated cell activation by stimulating cells with anisomycin. MKK-4 and MKK-7 contributed equally to anisomycin-induced JNK activation, confirming that the MKK-4 knockdown was effective. Previous studies with murine embryonic fibroblasts (MEFs) also suggest that MKK-7 can contribute to JNK phosphorylation by IL-1 and TNFα, although MKK-4 is still required for full JNK activation (12). MKK-4 and MKK-7 contribute equally to JNK activation induced by environmental stress such as ultraviolet (UV) light (12). Stimulus-specific utilization of MAPK kinases has also been observed in the p38 pathway, where MKK-3 and MKK-6 function also depend on how cells are activated (19).
Because the JNK substrate c-Jun is a major component of the transcription factor AP-1, we further evaluated the consequences of MKK-4 and MKK-7 deficiency in primary FLS by assessing AP-1 binding and transcriptional activity. These studies confirmed that the events downstream of JNK in primary FLS are strictly MKK-7 dependent after IL-1 stimulation. Finally, MMP-3 expression, which requires JNK and AP-1, was also regulated by MKK-7 without a contribution from MKK-4. Of interest, we also examined the contribution of the 2 MKKs to TNFα-mediated JNK kinase function and MMP-3 expression. Although TNFα is a less potent inducer of MMP expression, the results were similar to those with IL-1 and suggest that the upstream signaling events leading to JNK activation by TNFα are also MKK-7 dependent. Taken together, our data suggest that MKK-7, but not MKK-4, could represent a therapeutic target to modulate JNK activity and expression of MMPs in RA.
The mechanism for the stimulus-dependent utilization of MKKs in JNK activation in FLS is not known, but the JNK signaling complex scaffold proteins could potentially participate. For instance, the JNK-interacting proteins bind JNK and MKK-7 but do not appear to interact with MKK-4 (20). Despite the differences in the functional outcomes of cytokine stimulation, IL-1 efficiently induces phosphorylation of both MKK-4 and MKK-7 in cultured FLS (13). Hence, the lack of MKK-4 activation cannot be the mechanism for stimulus-dependent JNK activation. One putative advantage of the 3-component JNK signalsome (MKK-4, MKK-7, and JNK) is that the 2 MAPK kinases can potentially activate distinct amino acids on JNK; MKK-4 and MKK-7 mainly phosphorylate the Tyr and Thr residues, respectively, in the TPY motif (11). Previous studies show impairment of JNK activation in MKK-7−/− embryonic stem (ES) cells accompanied by a loss of the Thr phosphorylation of JNK without a marked reduction in the Tyr phosphorylation. On the other hand, the Thr phosphorylation of JNK in MKK-4−/− ES cells is also attenuated, in addition to decreased Tyr phosphorylation. These results suggest that Tyr and Thr residues of JNK can be sequentially phosphorylated by MKK-4 and MKK-7 in stressed ES cells (21, 22). Thus, it is possible that targeting MKK-7 might be a more selective approach to inhibit cytokine-mediated functions of JNK, while maintaining certain MKK-4–mediated stress responses.
Another interesting feature that distinguishes MKK-4 from MKK-7 is the spectrum of substrates. For instance, MKK-7 functions only as a specific activator of JNK (23). Unlike other MKKs, MKK-4 can potentially activate either JNK or p38 when overexpressed in mammalian cells (23). In MEFs, MKK-4 is also required for the full activation of p38 after UV irradiation (24). In addition to our experiments with JNK, we showed that MKK-4 does not appear to contribute to IL-1–mediated activation of p38 (Figure 2) or p38-driven IL-6 production (results not shown) in FLS. However, MKK-4 can regulate some cellular functions that still are relevant to host defense and stress responses in FLS, as demonstrated by the anisomycin experiments. MKK-4 clearly plays some role in vivo, because the MKK-4 deletion is embryonic-lethal due to massive hepatic apoptosis (25, 26). Mutations of the MKK-4 gene in some carcinomas indicate that it might also function as a tumor suppressor or regulator of metastasis (27, 28). Furthermore, T lymphocytes deficient in MKK-4 show impaired IL-2 production following T cell receptor activation (29). Hence, the patterns of MKK-4 and MKK-7 activation can be stimulus and cell dependent; some environmental stresses, such as UV light, primarily activate MKK-4, while inflammatory cytokines tend to engage MKK-7 (12).
In conclusion, our data suggest that MKK-7 is the primary regulator of JNK activation in FLS after exposure to IL-1. This raises the possibility that selective MKK-7 inhibition might be an effective way to block cytokine-driven JNK responses but spare some stress-induced pathways. By suppressing only a subset of JNK functions that are relevant to cytokine-mediated diseases such as RA, it is possible that the benefit of JNK inhibition might be achieved with less impact on stress responses.