The primary aim of this study was to investigate, using an experimental rabbit model of osteoarthritis (OA), the effect of a selective mitogen-activated protein kinase kinase 1/2 (MEK-1/2) inhibitor, PD 198306, on the development of structural changes. Additional aims were to assess the effects of the inhibitor on levels of phosphorylated extracellular signal–regulated kinase 1/2 (phospho–ERK-1/2) and matrix metalloproteinase 1 (MMP-1; collagenase 1) in OA chondrocytes.
After surgical sectioning of the anterior cruciate ligament of the right knee joint, rabbits with OA were separated into 3 experimental groups: oral treatment with placebo or with PD 198306 at a therapeutic concentration of 10 mg/kg/day or 30 mg/kg/day. Each treatment started immediately after surgery. The animals were killed 8 weeks after surgery. Macroscopic and histologic studies were performed on the cartilage and synovial membrane. The levels of phospho–ERK-1/2 and MMP-1 in OA cartilage chondrocytes were evaluated by immunohistochemistry. Normal, untreated rabbits were used as controls.
OA rabbits treated with the highest dosage of MEK-1/2 inhibitor showed decreases in the surface area (size) of cartilage macroscopic lesions (P < 0.002) and in osteophyte width on the lateral condyles (P = 0.05). Histologically, the severity of synovial inflammation (villous hyperplasia) was also reduced (P < 0.02). In cartilage from placebo-treated OA rabbits, a significantly higher percentage of chondrocytes in the superficial layer stained positive for phospho–ERK-1/2 and MMP-1 compared with normal controls. Rabbits treated with the highest dosage of PD 198306 demonstrated a significant and dose-dependent reduction in the level of phospho–ERK-1/2 and a lower level of MMP-1.
This study demonstrates that, in vivo, PD 198306, a selective inhibitor of MEK-1/2, can partially decrease the development of some of the structural changes in experimental OA. This effect was associated with a reduction in the level of phospho–ERK-1/2 in OA chondrocytes, which probably explains the action of the drug.
The degradation of osteoarthritic (OA) cartilage is related to a complex interaction of mechanical and biochemical factors (1–3). Among the latter, a number of catabolic factors, including proinflammatory cytokines, nitric oxide (NO), and proteases, have been demonstrated to play major roles (1, 4).
OA cells, at both the cartilage and synovial membrane levels, are actively involved in producing an excess amount of catabolic factors (1). The increase in the metabolic activity of these cells is related to the stimulation of proinflammatory cytokines, such as interleukin-1 (IL-1), tumor necrosis factor α (TNFα), IL-6, and leukemia inhibitory factor (1, 4–6). After binding to a specific membrane-bound receptor, these cytokines initiate the activation of a number of intracellular signaling pathways that lead to the synthesis of transcription factors that will induce the expression of multiple genes, including those from catabolic factors (7).
The protein kinase pathways are believed to be of key importance among the different signaling systems activated by proinflammatory cytokines (7–12). These pathways comprise a number of signaling cascades. The three most predominant cascades culminate in the activation of the extracellular signal–regulated kinase 1/2 (ERK-1/2), c-Jun N-terminal kinase (JNK), and p38 families of mitogen-activated protein kinases (MAPKs). These cascades have been demonstrated in several studies to be essential to the synthesis of a number of catabolic factors responsible for inducing the structural changes seen in OA (7).
ERK-1/2 is activated by MAPK kinase 1/2 (MEK-1/2) as part of the MAPK pathway (7, 13). In this pathway, the Raf kinases phosphorylate and activate MEK-1/2, which in turn phosphorylates and activates ERK-1/2. Activated ERK-1/2 can then translocate in the nucleus and activate transduction factors by phosphorylation, thus altering specific gene expression. In addition, ERK-1/2 has a number of cytosolic substrates that can influence gene expression directly or indirectly (13).
The MAPK pathway has also been demonstrated to be a key factor in the induction of matrix metalloproteinases (MMPs) by cytokines (14), production of cyclooxygenase 2 (COX-2) by chondrocytes, and chondrocyte apoptosis (15). The JNK and p38 signaling cascades are implicated in the synthesis of MMP by chondrocytes (7, 14). Moreover, p38 has been shown to be involved in the expression of the inducible NO synthase (iNOS) (16) and proinflammatory cytokines (17). A recent study has demonstrated that an inhibitor of p38 very effectively reduced the synthesis in vitro of cytokines and iNOS (16). In another recent study, treatment with a specific p38 inhibitor in a rat model of inflammatory arthritis was found to reduce the progression of structural damage while simultaneously reducing the synthesis of proinflammatory cytokines (18).
The primary goal of the present study was to examine, using a rabbit experimental model of OA, the effect of a specific inhibitor of MEK-1/2, PD 198306, on the development of structural changes in articular tissues. We also examined the effect of the inhibitor on the phosphorylation level of ERK-1/2, as well as on some of the major pathophysiologic pathways of OA.
PD 198306 is a potent, selective, non-ATP competitive inhibitor of MEK-1/2. It inhibits the isolated enzyme at a concentration of 8 nM and inhibits MEK activity in synovial fibroblasts at concentrations of 30–100 nM, depending on the species. The compound is highly selective for MEK and has a 50% inhibition concentration of >1 μM for ERK, >4 μM for c-Src, >10 μM for phosphatidylinositol 3-kinase γ, and >4 μM for cyclin-dependent kinases. PD 198306 has a bioavailability of 62% when taken orally and is active in several animal models of rheumatoid arthritis, including rat streptococcal cell wall–induced arthritis (median effective dose [ED50] 11.2 mg/kg) and rat adjuvant arthritis (ED50 6.6 mg/kg) according to an internal file of Pfizer Global Research & Development (Ann Arbor, MI). Figure 1 shows the chemical structure of PD 198306.
MATERIALS AND METHODS
Forty-six white New Zealand male rabbits (ages 16–18 weeks, weighing 2.5–3 kg) were used in this study. Sectioning of the anterior cruciate ligament (ACL) of the right knee was performed on all rabbits through a medial parapatellar incision, as previously described (19, 20). Briefly, rabbits were anesthetized with ketamine (30 mg/kg; Wyeth-Ayerst, Montreal, Quebec, Canada), xylazine (5 mg/kg; Bayer, Etobicoke, Ontario, Canada), and intramuscular acepromazine maleate (1 mg; Wyeth-Ayerst). Under sterile conditions, an anteromedial incision of the right knee was performed. The subcutaneous tissue and retinaculum were incised and retracted, along with the articular capsule. The medial compartment was visualized, and the ACL was sectioned with a scalpel. The capsule, medial retinaculum, and skin were sutured. All rabbits were housed in regular individual cages, fed ad libitum, and allowed to exercise.
The rabbits were randomly distributed into three experimental groups. Group 1 (n = 20) consisted of OA rabbits that received placebo treatment (equivalent volume of vehicle only). Groups 2 and 3 (n = 10 per group) consisted of OA rabbits given PD 198306 at a dosage of 10 mg/kg/day or 30 mg/kg/day, respectively. PD 198306 was administered once daily at 8:00 AM as a liquid solution, by oral gavage into the stomach. The drug was dissolved in a 3-ml volume of 1.25% hydroxypropyl cellulose and 0.05% sodium lauryl sulfate, which was prepared fresh daily before administration. Treatment was initiated the day after surgery and continued for 8 weeks, at which time the rabbits were killed. The animal care personnel were blinded to the treatment groups. Four rabbits did not complete the study. One animal from group 1 had a wound infection. The other 3 rabbits, 1 in group 2 and 2 in group 3, had traumatic injuries during the course of the study. Six additional unoperated normal rabbits were used as controls and were killed at the same time as the animals in the experimental groups. The study protocol was approved by the Institutional Review Board.
Drug dosage and pharmacokinetic study.
Dosage levels for this study were determined by comparing plasma levels of PD 198306 from a rabbit pharmacokinetic study with efficacious plasma levels in rat models of arthritis. In the pharmacokinetic study, plasma samples were taken 1, 2, 4, 8, and 24 hours after a single oral dose of the compound. Exposures of PD 198306 were also determined during the present study by taking plasma samples between 1 hour and 24 hours after the last dose.
PD 198306 plasma levels were analyzed by a liquid/liquid extraction procedure using 650 μl of methyl t-butyl ether (95%/5% ethanol), 75 μl of 0.1% acetic acid, and a 75-μl sample. Samples were shaken 15–20 minutes and were then centrifuged at 4,000 revolutions per minute for 15 minutes. Ether (550 μl) was transferred via a 96-well multipipette (Tomtec, Hamden, CT) to a 96-well block (1 ml/well), then evaporated under N2. The samples were reconstituted with 100 μl of a 60:40 solution of acetonitrile (ACN; Fisher Scientific, Nepean, Ontario, Canada) to 0.1% acetic acid and then assessed by liquid chromatography/mass spectrometry on a Betasil (Thermo Electron Corporation, Waltham, MA) phenyl 2.1 × 10 × 3–μm column (mobile phase 50:50 ACN:0.1% formic acid, flow rate 0.25 ml/minute, ionization electrospray positive-ion mode, and injection volume 2 μl).
Macroscopic grading of lesions.
Immediately after the treatments were discontinued, the rabbits were killed and the right and left knees were dissected. Each knee was examined for gross morphologic changes, including the presence of osteophyte formation and cartilage lesions, as previously described (21, 22). The schedule for killing the animals was based on the randomization protocol, and examinations were performed by two independent observers who were blinded to the treatment group. The degree of osteophyte formation was graded by measuring the maximum width (in mm) of the spur on the medial and lateral femoral condyles using a Digimatic digital caliper (Mitutoyo, Kawasaki, Japan). These 2 values recorded for each rabbit were considered separately for the purpose of statistical analysis. The cartilage changes on the medial and lateral femoral condyles and tibial plateaus were graded separately, using a dissecting microscope (Stereozoom; Bausch & Lomb, Rochester, NY).
The depth (grade) of erosion was graded on a scale of 0–4, where 0 = normal-appearing surface, 1 = minimal fibrillation or a slight yellowish discoloration of the surface, 2 = erosion extending into superficial or middle layers, 3 = erosion extending into the deep layers, and 4 = erosion extending to the subchondral bone. The surface area (size) of the articular surface changes was measured and expressed in mm2. A macroscopic total joint score was also obtained by adding the mean scores of cartilage lesions from the medial and lateral condyles together with those from the medial and lateral plateaus.
Histologic evaluation was performed on sagittal sections of cartilage from the lesion areas of each femoral condyle and tibial plateau of the right knees in OA rabbits and from equivalent sites in normal rabbits as described previously (21, 22). Specimens were dissected, fixed in 10% neutral buffered formalin, and embedded in paraffin for histologic evaluation. Serial sections (5 μm) were stained with Safranin O. The severity of the OA lesions was graded on a scale of 0–14 by two independent observers under blinded conditions, using the histologic/histochemical scale of Mankin et al (23). This scale evaluates the severity of OA lesions based on the loss of staining with Safranin O (scale of 0–4), cellular changes (scale of 0–3), invasion of the tidemark by blood vessels (scale of 0–1), and structural changes (scale of 0–6, where 0 = normal cartilage structure and 6 = erosion of the cartilage down to the subchondral bone). The scoring was based on the most severe histologic changes within each cartilage section.
Representative specimens of synovial membrane from the medial and lateral compartments of the knee were also dissected from the underlying tissues. The specimens were fixed in 10% buffered formalin, embedded in paraffin, sectioned (5-μm sections), and stained with hematoxylin and eosin. Two synovial membrane specimens were examined for each compartment, and the highest score from each compartment was recorded. The average was calculated and considered as a unit for the whole knee. The severity of synovitis was graded on a scale of 0–10 (24) by two independent observers under blinded conditions, adding the scores for 3 histologic criteria: synovial lining cell hyperplasia (scale of 0–2), villous hyperplasia (scale of 0–3), and degree of cellular infiltration by mononuclear and polymorphonuclear cells (scale of 0–5).
Cartilage specimens from condyles and plateaus were processed for immunohistochemical analysis as previously described (25, 26). Briefly, specimens were fixed with TissuFix #2 (Laboratoires Gilles Chaput, Montreal, Quebec, Canada) for 24 hours and were then embedded in paraffin. Sections (5 μm) of paraffin-embedded specimens were placed on slides (Superfrost Plus; Fisher Scientific), deparaffinized in toluene, hydrated in a series of graded dilutions of ethanol, and preincubated with chondroitinase ABC (0.25 units/ml) in phosphate buffered saline (PBS) for 60 minutes at 37°C, citrate buffer (0.01 mM, pH 6.0) for 20 minutes, and Triton X-100 (0.3%) for 30 minutes. The specimens were washed in PBS and then in 2% H2O2/PBS for 30 minutes. Slides were further incubated with Universal Blocking Solution with Avidin Blocking Solution (0.2 ml/ml; Vector, Burlingame, CA) in PBS for 15 minutes, blotted, and overlaid with a mouse monoclonal antibody against MMP-1 (100 μg/ml, dilution 1:1,500; Calbiochem-Novabiochem, San Diego, CA) or a mouse monoclonal antibody against phosphorylated ERK-1/2 (phospho–ERK-1/2) (40 μg/ml, dilution 1:25; Cell Signaling Technology, Beverly, MA) for 18 hours at 4°C in a humidified chamber.
Each slide was washed 3 times in PBS (pH 7.4) and stained using the avidin–biotin complex method (Vectastain ABC Kit; Dako Diagnostics Canada, Mississauga, Ontario, Canada). This method entails incubation in the presence of the biotin-conjugated secondary antibody for 60 minutes at room temperature, followed by the addition of the avidin–biotin–peroxidase complex for 45 minutes. All incubations were carried out in a humidified chamber, and color was developed with 3,3′-diaminobenzidine (Dako Diagnostics Canada) containing H2O2.
To determine the specificity of staining, 3 control procedures were used, according to the same experimental protocol: 1) use of adsorbed immune serum (1 hour at 37°C) with a 20-fold molar excess of recombinant or purified antigen, 2) omission of the primary antibody, and 3) substitution of the primary antibody with an autologous preimmune serum. The antigens used in our study were purified human MMP-1 (Calbiochem-Novabiochem) and the phospho–Thr 202/Thr 204 peptide (Cell Signaling Technology). Several sections were made from each block of cartilage, and slides from each specimen were processed for immunohistochemical analysis. Each section was examined under a light microscope (Leitz Orthoplan; Wild Leitz, Ville St. Laurent, Quebec, Canada) and photographed with Ektachrome 64 ASA film (Eastman Kodak, Rochester, NY).
Three sections from each femoral condyle and tibial plateau specimen were examined using a Leitz Diaplan microscope (40×; Wild Leitz), and each section was scored separately. The number of chondrocytes staining positive in the upper zone (superficial and upper intermediate layers of cartilage) for MMP-1 or phospho–ERK-1/2 was estimated as previously described (25–27). The analysis was performed as previously described (28) only in the superficial layers of cartilage, since the majority of OA chondrocytes staining positive for MMP-1 (26, 28) or phospho–ERK-1/2 are located in these layers. Briefly, each section was divided into 3 different areas at the superficial layers of cartilage. For each specimen, it was ensured before the evaluation that an intact cartilage surface at the margin of the lesion could be detected and used as a marker to validate the morphometric analysis.
The cell count scores were determined separately for the lateral and medial sides of the condyles and plateaus. The total number of chondrocytes and the number of chondrocytes staining positive were then quantitated separately for the superficial zone. The final results were expressed as the percentage of positive chondrocytes. The maximum score for each cartilage specimen was 100%. Each slide was evaluated by two independent observers under blinded conditions; variation between the observers' findings was <5%.
All data were expressed as the mean ± SEM or as the median and range and analyzed with the Mann-Whitney U test when appropriate. P values less than 0.05 were considered significant.
At a dosage of 10 mg/kg/day, PD 198306 gave a maximum concentration (Cmax) and an area under the curve (AUC; 0–24 hours) of 3.7 μg/ml and 24.2 μg/ml, respectively. At 30 mg/kg/day, the Cmax and AUC (0–24 hours) values were 5.9 μg/ml and 48.7 μg/ml, respectively.
Macroscopic and microscopic findings.
Osteophytes. In placebo-treated OA rabbits, osteophytes were present on both sides of the condyles in all animals, with mean ± SEM widths were 2.2 ± 0.3 mm and 1.4 ± 0.2 mm for the medial and lateral condyles, respectively. In OA rabbits treated with PD 198306 at 10 mg/kg/day, the incidence of osteophytes was 94% and the medial and lateral condyle widths were 2.1 ± 0.3 mm and 1.1 ± 0.2 mm, respectively. In OA rabbits treated with 30 mg/kg/day, the incidence was 69% and the medial and lateral condyle widths were 2.4 ± 0.4 mm and 0.5 ± 0.3 mm. The osteophytes on the lateral condyles of rabbits treated with the highest dosage of the drug were found to be smaller (P = 0.05) compared with those of the placebo-treated group.
Cartilage. In normal rabbits, the cartilage was white and shiny, and no lesion could be detected. In placebo-treated OA rabbits, macroscopic lesions were of a moderate degree and were equally severe on both the femoral condyles and tibial plateaus (Table 1 and Figure 2). Rabbits treated with the highest dosage of PD 198306 had a decrease in the severity of the macroscopic lesions (surface only) on the femoral condyles and tibial plateaus. Moreover, the total joint score was found to be significantly lower for the grade of lesions in animals treated with the highest dosage of PD 198306 (P < 0.02). Histology of cartilage from normal rabbits demonstrated a normal microscopic appearance (Figure 2). In placebo-treated OA rabbits, the histologic lesions were of moderate degree and were more severe on the condyles than on the plateaus (mean ± SEM scores of 6.2 ± 0.6 and 4.1 ± 0.6, respectively) (Figure 3). Rabbits treated with the lowest dosage of PD 198306 had a slight increase in the severity of lesions on the condyles and plateaus (scores of 7.8 ± 0.7 and 4.7 ± 0.8, respectively). However, these differences were not statistically significant. In rabbits treated with the highest dosage of the inhibitor (30 mg/kg/day), the severity scores of histologic lesions on condyles and plateaus were 5.3 ± 0.8 and 2.5 ± 0.5, respectively.
Table 1. Cartilage macroscopic lesions on femoral condyles and tibial plateaus*
Surface area, mm2
Depth, 0–4 scale
Surface area, mm2
Depth, 0–4 scale
Surface area, mm2
Depth, 0–4 scale
Values are the median (range). PD 198306–treated and placebo-treated rabbits with osteoarthritis (OA) received the drug or the vehicle only, respectively, by oral gavage once a day at 8:00 AM for 8 weeks after anterior cruciate ligament section of the right knee. All rabbits were killed and their tissue examined at 8 weeks. ND = no lesion detected.
P < 0.002 versus OA placebo-treated group, by Mann-Whitney U test.
P < 0.02 versus OA placebo-treated group, by Mann-Whitney U test.
Synovial membrane. Synovium from normal rabbits was thin and lustrous with no obvious abnormalities. Synovium from placebo-treated OA rabbits was hypertrophic and demonstrated a reddish-yellow discoloration. In both groups of rabbits treated with PD 198306, the synovium was thinner and showed a less intense discoloration.
Histology of synovium from normal rabbits showed no abnormalities. The synovium from placebo-treated OA rabbits showed a moderate degree of inflammation with synovial lining cell hypertrophia, villous hyperplasia, and a moderate mononuclear cell infiltration of the sublining tissue (Table 2 and Figure 4). The two groups of rabbits treated with the PD 198306 showed significant reductions in the severity of synovial inflammation. The reductions resulted primarily from a less marked degree of synovial hyperplasia (Table 2).
Table 2. Histologic/histochemical grading of synovial membrane*
Synovial lining, 0–2 scale
Villous hyperplasia, 0–3 scale
MNC infiltration, 0–4 scale
PMN infiltration, 0–1 scale
Total, 0–10 scale
Values are the median (range). MNC = mononuclear cell; PMN = polymorphonuclear cell (see Table 1 for other definitions and explanations).
P < 0.001 versus OA placebo-treated group, by Mann-Whitney U test.
P < 0.003 versus OA placebo-treated group, by Mann-Whitney U test.
P < 0.02 versus OA placebo-treated group, by Mann-Whitney U test.
Experiments performed with the antibody that specifically recognized phospho–ERK-1/2 (Table 3 and Figure 5) demonstrated that in cartilage specimens from normal rabbits, only a few cells within the superficial layers stained positive for the kinase. In specimens from placebo-treated OA rabbits, a large number of cells, mainly from the superficial upper and upper middle zones of cartilage, showed positive staining that was located largely in the chondrocyte nuclei. Specimens from rabbits treated with PD 198306 demonstrated significant decreases in the numbers of chondrocytes staining positive for phospho–ERK-1/2 at both dosages tested; however, the decrease was more pronounced in the group treated with the highest dosage of the drug. OA cartilage treated with immunoadsorbed serum demonstrated only background staining (Figure 5). For the morphometric analysis, the average total number of cells counted was 190 for the femoral condyles and 242 for the tibial plateaus.
Table 3. Phosphorylated ERK-1/2, MMP-1, and TUNEL reaction levels in rabbit knee cartilage*
ERK-1/2–positive cells, %
MMP-1–positive cells, %
Values are the mean ± SEM. ERK-1/2 = extracellular signal–regulated kinase 1/2; MMP-1 = matrix metalloproteinase 1; OA = osteoarthritis (see Table 1 for explanations).
P < 0.0001 versus OA placebo-treated group, by Mann-Whitney U test.
P < 0.04 versus OA placebo-treated group, by Mann-Whitney U test.
P < 0.06 versus OA placebo-treated group, by Mann-Whitney U test.
P < 0.001 versus OA placebo-treated group, by Mann-Whitney U test.
Experiments performed using an antibody that specifically recognized MMP-1 showed that in normal cartilage, only a few cells in the superficial layers were positive for this protease. In contrast, in OA cartilage, a significantly larger number of chondrocytes located mainly in the superficial and upper intermediate layers stained positive for this protease (Table 3 and Figure 6). Rabbits treated with PD 198306 (30 mg/kg/day) showed a lower cell score for MMP-1, both on the femoral condyles and on the tibial plateaus. The differences between the groups were not statistically significant. OA cartilage treated with immunoadsorbed serum demonstrated only background staining (Figure 6). For the morphometric analysis, the average total number of cells counted was 276 for the femoral condyles and 265 for the tibial plateaus.
This study provides new and interesting findings about the role of MAPK in OA pathophysiology as well as the potential uses of MAPK and MEK-1/2 inhibitors for the treatment of the articular tissue structural changes in OA. This approach brings into perspective the possibility of pharmacologic intervention with regard to specific pathophysiologic targets.
The last few decades of research in the field of OA have allowed us not only to better characterize the morphologic changes taking place during the course of the disease, but also to discover a number of very important mechanisms responsible for such changes (1–4). For instance, the participation of synovial inflammation in the progression of cartilage changes at the clinical stage of the disease is becoming increasingly obvious (4). There are now a number of clinical studies that demonstrate a clear association between inflammation and disease progression (29–32). A number of factors synthesized within the inflamed synovium are responsible for stimulating the synthesis of a larger number of catabolic/anabolic factors, which in turn mediate the structural changes (1–4). Among the different synovial factors, cytokines, such as IL-1β and TNFα, are believed to play a prime role (6). The activation of cells by these cytokines is mediated by the binding to a specific cell membrane surface receptor, which triggers the activation of a number of complex intracellular signaling pathways (7, 33–36). Other inflammatory factors (e.g., NO) have also been demonstrated to activate intracellular signaling pathways (37). Among these, the protein kinase pathways are of major importance in the induction of the synthesis of transcription factors that mediate the up-regulation of a number of very important catabolic factors, such as MMP and NO (16, 38).
The findings of the present study suggest that PD 198306 could effectively reduce the phosphorylation of MEK-1/2 in situ, since this kinase is immediately upstream from ERK-1/2 and is the only kinase that has thus far been demonstrated to be able to phosphorylate ERK-1/2 (7, 13). The effect of PD 198306 was dose dependent and more pronounced in the highest-dosage treatment group, which also provides additional evidence of this inhibition. Since the MAPK pathway was involved in the synthesis of a number of catabolic factors participating in the development of the structural changes in OA, it is plausible that the action of PD 198306 was mediated through the inhibition of their synthesis. Moreover, the reduction in the severity of synovial inflammation, which was related to a reduction in the level of synovial hyperplasia, could probably have also contributed to this phenomenon. However, when considering the reduction in the histologic score of the synovial membrane, one must take into account that the synovial scoring system used in the study gave more weight to villous hyperplasia and cellular infiltration, which represented 8 of the 10 points (total score) in the system.
We have previously demonstrated in the OA experimental dog model that agents that could effectively reduce synovial inflammation (such as a selective iNOS inhibitor) could also simultaneously reduce the appearance/progression of structural changes in OA, including cartilage lesions and osteophyte formation (39). This phenomenon was associated with a concomitant reduction in the level of synthesis of IL-1β and inflammatory mediators such as prostaglandin E2 (PGE2) and NO (39, 40). This finding is most interesting, since the synthesis of PGE2, a well-known factor also capable of inducing bone remodeling (41), can be blocked by MEK-1/2 inhibition, which reduces COX-2 expression.
Treatment with therapeutic concentrations of PD 198306 was also found to be capable of reducing the progression of cartilage damage. This effect was noted on the surface of the tibial plateaus and the depth of lesions in the total joint, both of which were significantly reduced. This protective effect was also associated with a better preservation of the cartilage histologic structure. The lower score on the Mankin scale observed in this study was in great part related to this particular finding. However, it should be noted that this scoring system gives a much greater emphasis to matrix changes, since 10 of the 14 points (total score) were for structural changes and loss of Safranin O staining. Moreover, these OA lesions were not always in the same topographic location; therefore, the sampling of cartilage specimens could also explain some of the variation observed in the study.
Concerning the role of cartilage catabolic factors, a number of biochemical factors that participate significantly in the degradation of cartilage macromolecules have been identified (1–4). Among these factors, the proteases, and more specifically MMP, have been demonstrated to be major factors in the catabolism of cartilage macromolecules (42). In the MMP family, collagenases are believed to be key players in the proteolysis of cartilage type II collagen (43).
Collagenase 1 (MMP-1) has been identified in OA cartilage (both in human and in experimental models) and shown to be synthesized in an increased amount by OA chondrocytes (25, 26, 43). Previous studies have demonstrated that there is a preferential synthesis of MMP-1 by chondrocytes located in the superficial layers in OA cartilage (25, 26). These reports are consistent with the findings of our study. The involvement of the excess synthesis and activity of MMP-1 and other collagenases in the structural changes in OA cartilage has been well established (6, 43). In the present study, we demonstrated that treatment with the inhibitor of MEK-1/2 was capable of reducing the number of chondrocytes that stained positive for MMP-1. However, it should be pointed out that the differences among the OA PD 198306–treated groups and the OA placebo-treated group did not reach statistical significance. It is of interest to note that the in situ level of synthesis of MMP-1 in OA cartilage chondrocytes, as assayed by immunohistochemistry, has been recently shown by Altman and Cheung to reflect the level of MMP activity in rabbit OA (44). The differences in the extent to which PD 198306 inhibits ERK-1/2 and MMP-1 raise a number of questions about the relationship between the effect of this drug on structural changes and its effect on MMP-1 synthesis.
A possible explanation for this finding could be that, since previous studies have demonstrated that MMP synthesis depends on the simultaneous activation of many of the protein kinase pathways, including the JNK, MEK-1/2, and p38 pathways (7, 11, 14), it is therefore understandable that the selective inhibition of a single protein kinase pathway could have induced only a partial inhibition of MMP synthesis. However, since the MEK-1/2/ERK-1/2 pathway is also involved in the synthesis of other catabolic factors by chondrocytes (including NO [27,38], which is important in inducing cartilage lesions), it is likely that the effect of PD 198306 was mediated through the inhibition of synthesis of several other catabolic factors. Further studies are currently under way to explore additional mechanisms of action of this drug.
Preliminary findings from this study provide information about potential mechanisms for mediating the structure-modifying effects of this MEK-1/2 inhibitor. Treatment with the inhibitor was found to reduce the severity of synovial inflammation, which was essentially related to a significant reduction in the villous hyperplasia. Therefore, the total number of cells (including synovial cells and infiltrating mononuclear cells) that could have potentially produced catabolic factors, such as cytokines and MMP, was reduced by treatment with PD 198306. However, caution must be taken not to overinterpret these results, and the synovial scoring system used in the present study must be taken into account. Moreover, a very important study by Firestein and Manning (7) recently showed that synovial fibroblasts from OA patients expressed the 3 main MAPK families. That study also showed that MEK-1/2 was readily phosphorylated by IL-1 and was responsible for increasing MMP-1 gene expression in these cells. Therefore, although this is very speculative, it is possible that the inhibition of MEK-1/2 in OA synovium could, by reducing cartilage catabolism, also contribute to a reduction in the level of synovial inflammation (1–4).
In summary, the present study has demonstrated that the activation of the MAPK pathway could probably contribute to the pathogenesis of OA. Therapeutic intervention with the goal of MEK-1/2 inhibition may have interesting potential for the development of agents for the treatment of OA.