Dr. Schaefer owns stock or stock options in Berlex Biosciences.
Central role of the MEK/ERK MAP kinase pathway in a mouse model of rheumatoid arthritis: Potential proinflammatory mechanisms
Article first published online: 28 SEP 2007
Copyright © 2007 by the American College of Rheumatology
Arthritis & Rheumatism
Volume 56, Issue 10, pages 3347–3357, October 2007
How to Cite
Thiel, M. J., Schaefer, C. J., Lesch, M. E., Mobley, J. L., Dudley, D. T., Tecle, H., Barrett, S. D., Schrier, D. J. and Flory, C. M. (2007), Central role of the MEK/ERK MAP kinase pathway in a mouse model of rheumatoid arthritis: Potential proinflammatory mechanisms. Arthritis & Rheumatism, 56: 3347–3357. doi: 10.1002/art.22869
- Issue published online: 28 SEP 2007
- Article first published online: 28 SEP 2007
- Manuscript Accepted: 1 JUN 2007
- Manuscript Received: 3 AUG 2006
To evaluate the role of the MEK/ERK MAP kinase pathway in murine collagen-induced arthritis (CIA) using the selective MEK inhibitor PD184352. We examined the effects of the inhibitor in cytokine-stimulated synovial fibroblasts and in cytokine-induced arthritis in rabbits to investigate its antiinflammatory mechanisms.
Murine CIA was used to assess the effects of the selective MEK inhibitor on paw edema, clinical scores, weight loss, histopathologic features, and joint levels of p-ERK. Western blotting and immunohistochemistry techniques were used to assess p-ERK in human and rabbit synovial fibroblasts and synovial tissue from rheumatoid arthritis (RA) patients. Interleukin-1α (IL-1α)–stimulated stromelysin production in rabbit synovial fibroblasts was assessed by enzyme-linked immunosorbent assay. A rabbit IL-1α–induced arthritis model was used to assess the effects of the inhibitor on IL-1α–induced MEK activity, stromelysin production, and cartilage degradation.
In the CIA model, PD184352 inhibited paw edema and clinical arthritis scores in a dose-dependent manner. Disease-induced weight loss and histopathologic changes were also significantly improved by treatment. Inhibition of disease-induced p-ERK levels in the joints was seen with the inhibitor. Levels of p-ERK in the synovium were higher in RA patients than in normal individuals. PD184352 reduced IL-1α–induced p-ERK levels in human RA synovial fibroblasts. The production of p-ERK and stromelysin was also inhibited in IL-1α–stimulated rabbit synovial fibroblasts. We observed IL-1α–induced p-ERK in the synovial lining, subsynovial vasculature, and articular chondrocytes. IL-1α–induced stromelysin production and proteoglycan loss from the articular cartilage were reduced by PD184352.
These data demonstrate the inhibition of murine CIA by PD184352, support the hypothesis that antiinflammatory activity contributes to the mechanism of action of the inhibitor, and suggest that a selective inhibitor may effectively treat RA and other inflammatory disorders.
Rheumatoid arthritis (RA) is a chronic inflammatory, systemic autoimmune disease of uncertain etiology (1). While it is has been hypothesized that an infection or other environmental insult, in the presence of an appropriate genetic background, leads to the breakdown of tolerance and initiation of the autoimmune response, neither the “arthritogenic” immunogen nor the targeted self antigen has been clearly defined (2). Chronic progressive RA leads to extensive synovitis and destruction of both articular cartilage and subchondral bone, eventually causing loss of function of the affected synovial joints. Moreover, systemic manifestations of the disease can significantly shorten the lifespan of patients with RA (1).
Standard approaches to treating RA have historically been broadly immunosuppressive or have had other dose-limiting toxicities that strongly influence their efficacy and tolerability (3). Newer anti–tumor necrosis factor α (anti-TNFα) biologic agents have so far held a significant edge in terms of both efficacy and safety, but specific safety concerns and the need for parenteral administration are a continuing problem. Furthermore, there remains a significant percentage of patients with disease that is resistant to anti-TNFα therapy (4, 5). The selective targeting of specific B cell or T cell populations or functions has also shown promise in recent studies (6, 7). Of significant current interest as potential therapeutic targets in RA are members of the MAP kinase signal transduction cascades.
MAP kinases are proline-directed serine/threonine protein kinases that become activated in response to extracellular stimuli, such as mitogens, growth factors, cytokines, and stress. Nuclear translocation of activated MAP kinases facilitates the modulation of gene transcription via the induction and/or transactivation of transcription factors, including activator protein 1, which plays a central role in the control of cellular activation, proliferation, and the production of cytokines, adhesion molecules, and metalloproteinases (8, 9). MAP kinases are therefore pivotal components of signal transduction cascades that culminate in the enhanced expression of mediators of inflammation that are central to the pathophysiology of RA and other inflammatory diseases (10, 11). The 3 most well-characterized mammalian MAP kinase pathways include the ERK pathway, the JNK/SAPK pathway, and the p38 pathway. Multiple isoforms of each MAP kinase exist and are activated by a series of upstream kinases. In the case of the ERK pathway, ERK-1 and ERK-2 (also known as p44 and p42 MAP kinases, respectively) are activated by the MAP kinase kinases MEK-1 and MEK-2 (also known as MAP kinase/ERK kinases 1 and 2). MEKs are dual-specificity kinases that phosphorylate critical tyrosine and threonine residues of ERK-1 and ERK-2 (12), thereby activating the ERKs.
It is well established that the JNK and p38 pathways are activated by inflammatory cytokines and cellular stressors, such as osmotic shock and exposure to ultraviolet irradiation, and that the p38 pathway at least plays a critical role in regulating inflammatory processes in vivo (13, 14). Although it is known to be activated by a wide variety of stimuli, the ERK pathway is frequently characterized as being important primarily in mitogen- and growth factor–induced effects on the cell cycle (8). The role of MEK in cellular proliferation and apoptosis is indeed well validated, and this knowledge led to the initial investigations into the use of MEK (or upstream Raf) inhibitors for treating cancers (15, 16). A common generalization, however, is that JNK and p38 are the main drivers behind the pathophysiology of inflammation-mediated diseases, while proliferative, neoplastic responses are the domain of the MEK/ERK pathway (8, 9). Others have argued that the stimulating signals for, and proinflammatory consequences of, activating the different pathways are much less distinct (17).
The MEK/ERK cascade has been shown to play a role in the production of metalloproteinases by a variety of cell types (18, 19) and in affecting lymphocyte activation and differentiation (20–22). The MEK/ERK pathway, however, also enhances the production of a variety of proinflammatory cytokines, such as interleukin-1β (IL-1β), TNFα, and IL-6 (23–25), which either have clinical precedence as targets in the treatment of RA or are currently under investigation for this indication (5). Moreover, all 3 of the MAP kinase pathways are activated in RA synovium (26). There is certainly clear evidence for the role of MEK in cellular proliferation, and given that RA pannus has been described as having neoplastic-like qualities (2), there is also good rationale for antiproliferative strategies in the treatment of RA. These data suggest, however, that more broad-reaching effects of enhanced MEK activity on numerous inflammatory and/or immune pathways could play a major role in the development of the synovitis, pain, and tissue destruction characteristic of RA.
The assessment of the role of MEK in inflammation and immunity has been hampered by the lack of selective and bioavailable chemical tools suitable for in vivo investigations. In the present study, we assessed the effects of a potent and specific MEK inhibitor in an experimental animal model of RA. Collagen-induced arthritis (CIA) in mice is a well-characterized animal model of arthritis that shares many of the pathologic, immunologic, and genetic features of RA in humans (27). A progressive inflammatory arthritis develops in immunized mice that is characterized clinically by edema and eventual joint ankylosis and histologically by synovitis, pannus formation, and cartilage and bone erosions. We also assessed the status of ERK phosphorylation in joint tissues from immunized mice, as well as the effect of MEK inhibitors on the levels of p-ERK. MEK inhibition in this model resulted in significant, dose-dependent effects on both the clinical and histologic features of the arthritic response. To further investigate the hypothesis that antiinflammatory activities of the MEK inhibitor could contribute to its therapeutic activity, we also assessed its effects on inflammatory cytokine–stimulated synovial fibroblasts in vitro and on cytokine-induced arthritis in vivo. Our findings suggest that selective inhibitors of MEK may have therapeutic benefit in RA and other chronic inflammatory diseases.
MATERIALS AND METHODS
Female DBA/1LacJ mice ages 8–12 weeks were obtained from The Jackson Laboratory (Bar Harbor, ME). Male New Zealand white rabbits weighing 2–3 kg were obtained from Harlan (Indianapolis, IN). All procedures were carried out in compliance with the National Institutes of Health Guidelines for the Care and Use of Laboratory Animals, under a protocol approved by the Pfizer Global Research and Development Animal Use Committee.
MEK inhibitor PD184352.
PD184352 is a potent and selective, orally bioavailable, non-ATP, nonsubstrate, competitive inhibitor of MEK-1 and MEK-2, which has been described in detail elsewhere (15). The compound was suspended in 0.5% hydroxypropyl methylcellulose/0.2% Tween 80, which was prepared fresh immediately before oral gavage. In the CIA model, dosing with PD184352 began on day 27 after immunization and continued once daily through day 42 at doses of 10, 30, and 100 mg/kg. In the rabbit knee IL-1α–induced arthritis model, PD184352 was dosed 1 hour before injection of IL-1α.
Induction of murine CIA.
Bovine type II collagen (Marie Griffiths, University of Utah, Salt Lake City, Utah) was diluted with 0.01N acetic acid to a concentration of 2 mg/ml and emulsified with an equal volume of Freund's complete adjuvant (Difco, Detroit, MI) supplemented with 1 mg/ml of Mycobacterium tuberculosis H37Ra (Difco). DBA/1LacJ mice were immunized with 100 μl of emulsion (100 μg of collagen) administered intradermally at the base of the tail. On day 28 after immunization, the mice were given 50 μg of lipopolysaccharide (LPS; Sigma-Aldrich, St. Louis, MO) in 100 μl of saline administered intraperitoneally.
On days 27, 31, 34, 37, and 41 after immunization, mice were examined for the development of arthritis, which is characterized by erythema and edema of the ankle or wrist. Measurements of ankle and wrist joints for edema were taken using a constant-tension caliper (Dyer, Lancaster, PA). A clinical score of 0–3 was assigned for each limb, according to the following scoring system (28): 0 = normal, 1 = erythema and edema, 2 = joint distortion, and 3 = joint ankylosis. Changes in body weight were also monitored. The mice were euthanized on day 42.
Determination of p-ERK levels by Western blot analysis.
Frozen joints were pulverized and then homogenized at 100 mg of tissue per 1 ml of lysis buffer (70 mM NaCl, 50 mM β-glycerol phosphate, 10 mM HEPES, 1% Triton X-100, 100 mM Na3VO4, 100 mM phenylmethylsulfonyl fluoride, and 1 mg/ml of leupeptin). Protein concentrations in the lysates were determined using a bicinchoninic acid protein assay kit (Pierce, Rockford, IL). Fifty micrograms of protein mixed with Tris glycine sample buffer was loaded onto a 10% Tris glycine gel (Invitrogen, Carlsbad, CA) and run at 125V until the dye front reached the bottom of the gel. The gel was then transferred to nitrocellulose paper (Invitrogen) using a 20% methanol buffer (Invitrogen) for 1 hour 15 minutes at 215 mA. The membrane was washed with Tris buffered saline (TBS)/0.1% Tween 20 for 5 minutes.
The membrane was blocked for 1 hour at room temperature in 1× Western blocking reagent (Roche, Indianapolis, IN) and then incubated overnight at 4°C with phospho-p44/42 MAP kinase (Thr202/Tyr204) antibody (Cell Signaling Technology, Beverly, MA) diluted 1:1,000 in 15 ml of 0.5× Western blocking reagent. The blot was washed 3 times for 5 minutes each in TBS/Tween 20 and then incubated with goat anti-rabbit horseradish peroxidase–conjugated antibody (Cell Signaling Technology) diluted 1:2,000 in 15 ml of 0.5× Western blocking reagent. Then, the membrane was again washed 3 times for 5 minutes each in TBS/Tween 20, detected for 1 minute with enhanced chemiluminescence substrate (Amersham Pharmacia Biotech, Little Chalfont, UK), and exposed to chemiluminescent film.
Formalin-fixed front and back paws were sent to BolderPath (Boulder, CO) for decalcification, paraffin embedding, sectioning, and toluidine blue staining. Histopathologic features of CIA were scored according to the severity of synovitis, pannus formation, and cartilage and bone erosions (28). An overall score was assigned based on these criteria and using the following scale: 0 = normal, 1 = minor changes, 2 = moderate inflammatory disease, 3 = major inflammatory disease, 4 = destructive, erosive arthritis, and 5 = destructive, erosive arthritis with major bone remodeling.
Assessment of p-ERK in human synovium.
Human synovial tissue samples (normal and RA) from LifeSpan Biosciences (Seattle, WA) were sectioned, stained for p-ERK, and imaged by LifeSpan Biosciences.
Assessment of p-ERK in human RA synovial fibroblasts.
Synovial fibroblasts from RA patients were a gift from Dr. Leslie Crofford (University of Michigan, Ann Arbor, MI). The cells were placed in T25 flasks containing Dulbecco's modified Eagle's medium supplemented with 20% fetal bovine serum (FBS), penicillin (100 units/ml), and streptomycin (100 μg/ml), and cultured at 37°C in an atmosphere of 5% CO2 until confluent. All cell culture reagents were purchased form Gibco (Grand Island, NY) except where indicated otherwise. Cells were treated with PD184352 at various concentrations for 30 minutes prior to stimulation with 100 units/ml of recombinant human IL-1α. After 30 minutes, cells were lysed with the same lysis buffer that was used for the frozen mouse joints and run in a Western blot to analyze p-ERK levels.
Preparation of rabbit synovial fibroblasts.
New Zealand white rabbits were euthanized, and synovium was immediately removed by incising the quadriceps femoris tendon and retracting the patella. The synovium with the infrapatellar fat pad body was cut away from the patellar ligament and placed in sterile phosphate buffered saline (PBS). The synovium was finely minced with a sterile scalpel and placed in 6 ml of PBS containing 4 mg of type I collagenase per ml (BD Biosciences, San Jose, CA) for 3 hours at 37°C. The cells were washed and then cultured in α-minimum essential medium supplemented with 8% FBS, 10 mM HEPES, 100 units/ml of penicillin, 100 μg/ml of streptomycin, and 50 μg/ml of gentamicin. After reaching confluence, cells were treated with PD184352 and stimulated with IL-1α similar to the treatment of human RA synovial fibroblasts. The media were frozen for measurement of stromelysin, and the cells were lysed for Western blot analysis of p-ERK levels.
Preparation of the rabbit IL-1α–induced arthritis model.
Rabbit knees were shaved and disinfected. Then, 25 ng of recombinant human IL-1α (R&D Systems, Minneapolis, MN) in 0.5 ml of saline was injected intraarticularly into 1 knee through the suprapatellar ligament of the stifle joint space, using sterile technique. The contralateral joint received an equal volume of vehicle. After 24 hours, the animals were euthanized, and the joints were lavaged with 1 ml of sterile saline injected into the stifle joint space through the suprapatellar ligament. The leg was flexed 10 times, and the lavage fluid was withdrawn using a 20-gauge needle. The fluid was frozen for measurement of stromelysin. Synovial tissue was fixed for immunofluorescence, and the articular cartilage was removed from the femoral condyles with a scalpel, weighed, and either fixed for immunostaining or frozen for proteoglycan analysis.
Enzyme-linked immunosorbent assay (ELISA) for stromelysin.
Rabbit synovial fibroblast medium and rabbit knee lavage fluid were analyzed for stromelysin levels using a stromelysin ELISA kit according to the manufacturer's instructions (Amersham Biosciences, Piscataway, NJ).
Dimethylmethylene blue (DMB) assay.
The glycosaminoglycan content in articular cartilage was determined with the DMB assay. Papain in buffer was vortexed with the frozen cartilage and incubated for 7 hours at 65°C. The incubated samples were diluted with assay buffer, and DMB dye was added. The absorbance at 540 nm was read immediately. A standard curve was prepared using chondroitin sulfate C derived from shark cartilage.
Rabbit synovium and cartilage samples were fixed in 10% neutral buffered formalin, embedded in paraffin, and sectioned into 6-μm sections. The sections were deparaffinized, rehydrated, and stained for 1 hour with a monoclonal antibody to phospho-p44/42 MAP kinase (Sigma-Aldrich). Fluorescence detection was performed using a peroxidase-labeled secondary antibody and the Renaissance Tyramide Signal Amplification System (NEN, Boston, MA). The sections were photographed with a Zeiss Axiophot fluorescence microscope.
Effect of PD184352 on clinical parameters of arthritis.
Mice were dosed daily for 16 days with vehicle or with 10, 30, or 100 mg/kg of PD184352, beginning on day 27 after immunization, 1 day prior to LPS injection. A statistically significant, dose-dependent effect of PD184352 on both paw edema and clinical scores was seen. The inhibition of paw edema, as calculated by the area under the curve, was 45%, 79%, and >100% for doses of 10, 30, and 100 mg/kg, respectively (Figure 1A). The clinical arthritis score was inhibited by 31%, 58%, and 87%, respectively (Figure 1B). The 30-mg/kg and 100-mg/kg doses of PD184352 also significantly inhibited weight loss as compared with vehicle treatment (Figure 1C).
Kinetics of ERK phosphorylation in the joints of mice with CIA.
The kinetics of p-ERK production in whole joint extracts was determined by Western blot analysis (Figure 2A). Low basal levels of p-ERK were seen in paws from mice with CIA on the day of immunization, with a slow increase through day 25 after immunization. On day 28, the level of p-ERK sharply increased, with the highest levels observed on day 32 after immunization; these changes correlated closely with the development of peak edema in these mice. After day 35, the p-ERK levels decreased toward baseline levels.
Presence of p-ERK in the joints of mice with CIA treated with PD184352.
To determine whether PD184352 diminished p-ERK levels in arthritic mice, whole joint extracts were again used for Western blot analysis. The p-ERK levels were measured 2, 7, and 24 hours after the last dose of PD184352 at 10, 30, and 100 mg/kg (Figure 2B). The phosphorylation of ERKs 1 and 2 was inhibited in a dose-dependent manner at all time points assessed.
Effect of PD184352 on the histopathology of murine CIA.
Histologic parameters, including synovitis, pannus formation, and cartilage and bone erosions, were scored on a scale of 0–5 as described in Materials and Methods. An overall histology score was also assigned. PD184352 inhibited all of the parameters in a dose-dependent manner (Figure 3A). Moderate effects on the overall score were seen at 10 mg/kg, and significant inhibition was seen at 30 mg/kg and 100 mg/kg (51% and 76% inhibition, respectively). Joints from normal mice, vehicle-treated arthritic mice, and PD184352-treated mice showed histologic changes due to treatment (Figure 3B). There was severe synovitis and extensive pannus formation, along with advanced erosions, in the vehicle-treated mice, which is typical in this arthritis model. In contrast, the histopathologic features of the joint sections from the PD184352-treated mice displayed minimal synovitis and pannus formation and no erosions, which is consistent with the reduced clinical arthritis scores in these groups of mice.
Enhanced p-ERK levels in synovial tissue from RA patients.
Immunohistochemical staining for p-ERK demonstrated enhanced levels of MEK activity in synovial tissues from RA patients as compared with those in normal human synovium (Figure 4A).
Production of p-ERK in human RA synovial fibroblasts.
RA synovial fibroblasts were used to study the production of p-ERK in a human system. The cells were treated with PD184352 at various concentrations for 30 minutes prior to a 30-minute stimulation with 100 units/ml of recombinant human IL-1α. PD184352 treatment inhibited in a dose-dependent manner the IL-1α–induced p-ERK levels in these cells (Figure 4B). Doses of 300 nM and 3,000 nM inhibited production by 69% and 86%, respectively.
Production of p-ERK and stromelysin in rabbit synovial fibroblasts.
Isolated synovial cells were cultured to confluence, treated with PD184352, and stimulated with 100 units/ml of recombinant human IL-1α. PD184352 treatment inhibited p-ERK levels in these cells in a dose-dependent manner, with 22%, 81%, and nearly 100% for doses of 10, 100, and 1,000 nM, respectively (Figure 5). Stromelysin production by the rabbit synovial fibroblasts was also inhibited in a dose-dependent manner by treatment with PD184352 ranging from 1 nM to 3,000 nM, resulting in 30% and 99% inhibition, respectively, at these 2 doses (Figure 5).
Effects of MEK inhibition on the rabbit IL-1α–induced arthritis model.
The rabbit IL-1α–induced arthritis model, a model of acute cytokine-induced cartilage degradation, was used to study the effects of MEK inhibition on p-ERK levels, stromelysin production, and proteoglycan loss. IL-1α injection into the knee joint enhanced the levels of p-ERK in the synovial lining, in endothelial cells of the subsynovial vasculature, and in articular chondrocytes (Figure 6A). Interestingly, the endothelial staining appeared earlier, at 30 minutes after IL-1α injection, while the synovial staining was not apparent until 90 minutes after the injection. Enhanced MEK activity in the articular chondrocytes was apparent at all time points examined. Oral dosing with PD184352 at 3, 10, and 30 mg/kg 1 hour before IL-1α injection into the knee joint inhibited in a dose-dependent manner the levels of stromelysin seen in the joint lavage fluid (Figure 6B) and protected against proteoglycan loss in the articular cartilage at 24 hours after stimulation with IL-1α (Figure 6C).
Numerous studies have demonstrated that inhibition of the p38 MAP kinase pathway has profound effects on inflammatory cytokine synthesis in vitro (most notably from a clinical standpoint, TNFα) and on the moderation of disease in vivo in animal models of both acute and chronic inflammation (14). Significant effects of potent, selective p38 inhibitors have been observed in models of pulmonary inflammation (29), rat and mouse CIA, and rat adjuvant-induced arthritis (30). These studies and others have amply demonstrated the significant therapeutic potential of p38 inhibitors in treating chronic inflammatory conditions such as RA, and indeed, a number of clinical trials have been initiated to explore the efficacy and safety of this therapeutic approach in RA.
The MAP kinase JNK-2 has also been implicated in the pathogenesis of RA (13). JNK, as well as p38, mediates cytokine-induced production of TNFα, and the inhibition of JNK activity was shown to be involved in glucocorticoid-induced inhibition of the translation of TNFα messenger RNA (31). Moreover, the MAP kinase kinases MKK-4 and MKK-7, which are upstream activators of JNK-2, are highly expressed in RA synovium (32) and are constitutively expressed in cultured RA synovial fibroblasts, where they form a stable functional “signalsome” complex with JNK-2 (33). Results of studies of JNK-2–knockout mice in the anti–type II collagen antibody–induced arthritis model were also mixed, with the JNK-2–deficient mice showing improvements in measures of cartilage damage, but significantly worse clinical scores than the wild-type mice (34).
Unlike with p38, however, elucidating the true potential for JNK-2 inhibition in RA has been hampered by the lack of highly specific chemical tools suitable for in vivo studies. The purported JNK inhibitor SP600125 was shown to efficiently block IL-1–induced phospho-Jun accumulation and collagenase 3 expression in synovial fibroblasts, and in a rat model of antigen-induced arthritis, SP600125 showed a modest effect on paw swelling, but had dramatic effects on collagenase 3 expression in joint tissues and on joint damage as assessed by radiography (35). However, SP600125 is a rather weak (μM) inhibitor of both JNK-1 and JNK-2, and is not selective or is poorly selective for a wide variety of other kinases (36). Therefore, the potential for inhibitory effects of SP600125 on other pathways cannot be ruled out, especially in vivo. JNK remains an intriguing potential target for RA, but lacks the potent and selective tools necessary to support this role.
MEK activity is up-regulated in joint tissues from RA patients (ref. 26 and the present study), and it is up-regulated by a variety of proinflammatory cytokines in multiple cell types in vitro, including LPS stimulation of monocytes (23, 24), IL-1β and TNFα stimulation of human articular chondrocytes (37), IL-1β stimulation of human fibroblasts (18, 19), and TNFα stimulation of human monocyte-derived dendritic cells (38). TNFα was also shown to stimulate MEK activity in synovial tissue in vivo in the TNFα-transgenic mouse (39), an effect that was inhibited by blockade of TNFα. Elucidating the true role of the MEK/ERK pathway in vivo has been hampered by the lack of potent and selective, orally bioavailable inhibitors. MEK activation and its role in numerous aspects of cellular physiology, however, have been extensively investigated with the selective MEK inhibitors U0126 (40) and PD98059 (41).
U0126 was found to be highly effective at inhibiting MEK activity as well as IL-1β, IL-8, TNFα, and prostaglandin E2 (PGE2) production in LPS-stimulated monocytes (24). The LPS-induced increases in p38 and JNK activity were not affected by treatment with U0126, confirming its specificity against other MAP kinase pathways. In addition, U0126 blocked matrix metalloproteinase 2 (MMP-2) and MMP-9 secretion from, and the invasiveness of, a rat fibroblast cell line (42). U0126 inhibited collagenase 1 expression in human fibroblasts stimulated with basic calcium phosphate crystals (19) and PGE2 release from IL-1α–stimulated human bronchial epithelial cells (43). Although U0126 is not orally bioavailable, limited in vivo studies of this inhibitor have begun to delineate the true antiinflammatory potential of MEK inhibitors. In a mouse model of 12-O-tetradecanoylphorbol-13-acetate–induced ear edema, topical administration of U0126 effectively blocked both swelling and tissue ERK phosphorylation (44). When administered intravenously, U0126 blocked IL-1β production and focal ischemic injury in mice (45), as well as the inflammation associated with cisplatin-induced renal injury (46).
Numerous studies of PD98059 further support the pivotal role of MEK in cell activation and function. This selective inhibitor was shown to play a role in IL-17–induced nitric oxide synthesis in human articular chondrocytes (47), LPS- and interferon-γ–induced nitric oxide synthase and TNFα production in murine macrophages (48), LPS-induced TNFα in human monocytes (24), PGE2 release from IL-1β stimulated human bronchial epithelial cells (43), and okadaic acid–induced IL-6 production in human monocytes (25). In mice deficient for Tpl2, a serine/threonine protein kinase that phosphorylates and activates MEK, there is a defect in TNFα production in response to LPS (23). Furthermore, in macrophages, Tpl2 is required for LPS-induced MEK and ERK activation, but not for p38, JNK, or NF-κB activation. These data support a causal role of Tpl2-induced MEK activation in the elaboration of TNFα in response to LPS, a role that was further corroborated by the ability of PD98059 to inhibit LPS-induced TNFα secretion from macrophages derived from wild-type mice (23). PD98059 has also been to shown to inhibit the production of collagenases MMP-1 and MMP-3 (18, 19).
The MEK inhibitor PD184352 is equally selective as PD98059 but is significantly more potent, and it is readily bioavailable with oral dosing, which makes this reagent an ideal tool for elucidating the role of MEK in inflammatory conditions in vivo. The present study is the first to use this tool to investigate the role of the MEK/ERK pathway in mouse CIA. We observed a time-dependent up-regulation of p-ERK in joints from immunized mice that peaked at the time of maximum inflammation in this model. Treatment with PD184352 resulted in nearly complete inhibition of ERK phosphorylation, and a significant, dose-dependent inhibition of the edema, clinical arthritis, and histopathologic changes associated with this disease model.
The MEK inhibitor Ro 09-2210 has been less well-characterized in terms of its specificity profile or antiinflammatory potential, but it has also been shown to inhibit LPS-induced TNF production in monocytes (49) and the proliferation and activation of T cells (50). Array BioPharma (Boulder, CO) is also developing potent and selective MEK inhibitors (16) and has reported significant efficacy data in animal models of inflammation and arthritis (51), further supporting the pivotal role of this kinase in inflammatory diseases.
While the precise mechanisms underlying the antiarthritic effects in this model are still not entirely clear, we have shown profound effects of MEK inhibition on inflammatory cytokine–induced synovial fibroblast activation in vitro and on stromelysin production and cartilage degradation in vivo. We know that these inhibitors also have significant effects on chondrocyte and endothelial cell activation and MMP and cytokine expression induced by LPS or inflammatory cytokines (Thiel MJ, et al: unpublished observations). We believe that a combination of broad antiinflammatory activities on multiple cell types (e.g., monocytes, synoviocytes, chondrocytes, and endothelial cells) and potentially antiproliferative effects, which reduce synovial pannus formation, are the major effector mechanisms in vivo. We are continuing to explore the mechanism(s) of action of these compounds both in vitro and in vivo.
The MEK inhibitor used in these studies gave a maximal antiarthritic effect that was superior to that of treatment with an anti-TNFα antibody (data not shown), and our data support the contention that the MEK inhibitors have the capacity to inhibit diverse inflammatory pathways. The profound antiarthritic effects of the MEK inhibitors, along with the apparent lack of immunosuppression, support the validity of MEK as a therapeutic approach for this indication, especially in patients who are minimally responsive or nonresponsive to anti-TNFα therapy. These findings support the use of MEK inhibitors as a novel treatment modality for RA and possibly other inflammatory and autoimmune diseases.
Ms Thiel 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 design. Thiel, Schaefer, Lesch, Schrier, Flory.
Acquisition of data. Thiel, Schaefer, Lesch, Mobley, Dudley.
Analysis and interpretation of data. Thiel, Schaefer, Lesch, Mobley, Dudley, Flory.
Manuscript preparation. Thiel, Schaefer, Lesch, Flory.
Statistical analysis. Thiel, Schaefer.
Selection of compounds, assessment ofpharmacokinetics/pharmacodynamics, structural-activity relationship. Tecle.
Discovery and preparation of PD184352. Barrett.
- 25Extracellular-regulated kinase 1/2, Jun N-terminal kinase, and c-Jun are involved in NF-κB-dependent IL-6 expression in human monocytes. J Immunol 1999; 163: 4893–902., , , , , , et al.
- 26Activation, differential localization, and regulation of the stress-activated protein kinases, extracellular signal– regulated kinase, c-Jun N-terminal kinase, and p38 mitogen-activated protein kinase, in synovial tissue and cells in rheumatoid arthritis. Arthritis Rheum 2000; 43: 2501–12., , , , , , et al.
- 38Extracellular signal-regulated kinase, stress-activated protein kinase/c-Jun N-terminal kinase, and p38mapk are involved in IL-10-mediated selective repression of TNF-α-induced activation and maturation of human peripheral blood monocyte-derived dendritic cells. J Immunol 1999; 162: 3865–72., , , , .
- 48Specific inhibitors of p38 and extracellular signal-regulated kinase mitogen-activated protein kinase pathways block inducible nitric oxide synthase and tumor necrosis factor accumulation in murine macrophages stimulated with lipopolysaccharide and interferon-γ. J Infect Dis 1999; 179: 939–44., , .
- 51Characterization of ARRY-438162, a potent MEK inhibitor in combination with methotrexate or ibuprofen in in vivo models of arthritis [abstract]. Arthritis Rheum 2006; 54 Suppl 9: S354., , , , , , et al.