Non-steroidal anti-inflammatory drugs inhibit nitric oxide-induced apoptosis and dedifferentiation of articular chondrocytes independent of cyclooxygenase activity.

Nitric oxide (NO) causes apoptosis and dedifferentiation of articular chondrocytes by the modulation of extracellular signal-regulated kinase (ERK), p38 kinase, and protein kinase C (PKC) alpha and -zeta. In this study, we investigated the effects and mechanisms of non-steroidal anti-inflammatory drugs (NSAIDs), such as indomethacin, ketoprofen, ibuprofen, sulindac sulfide, and flurbiprofen, in NO-induced apoptosis and dedifferentiation of articular chondrocytes. We found that all of the examined NSAIDs inhibited apoptosis and dedifferentiation. NO production in chondrocytes caused activation of ERK-1/2 and p38 kinase, which oppositely regulate apoptosis and dedifferentiation. NO production also caused inhibition of PKCalpha and -zeta independent of and dependent on, respectively, p38 kinase, which is required for apoptosis and dedifferentiation. Among the signaling molecules modulated by NO, NSAIDs blocked NO-induced activation of p38 kinase, potentiated ERK activation, and blocked inhibition of PKCalpha and -zeta. NSAIDs also inhibited some of the apoptotic signaling that is downstream of p38 kinase and PKC, such as NFkappaB activation, p53 accumulation, and caspase-3 activation. The inhibitory effects of NSAIDs on apoptosis and dedifferentiation were independent of the inhibition of cyclooxygenase (COX)-2 and prostaglandin E(2) (PGE(2)) production, as evidenced by the observation that specific inhibition of COX-2 activity and PGE(2) production or exogenous PGE(2) did not affect NO-induced apoptosis and dedifferentiation. Taken together, our results indicate that NSAIDs block NO-induced apoptosis and dedifferentiation of articular chondrocytes by the modulation of ERK, p38 kinase, and PKCalpha and -zeta in a manner independent of their ability to inhibit COX-2 and PGE(2) production.

Chondrocytes are a unique cell type in which the differentiated phenotype is reversible. The phenotype of chondrocytes is regulated by the balance between anabolic and catabolic reactions of molecules, which are involved in the maintenance of cartilage homeostasis (1). Differentiated chondrocytes both in vivo and in vitro dedifferentiate into fibroblastic cells upon exposure to interleukin-1␤ (2,3), retinoic acid (4), or nitric oxide (NO) 1 (5). Although the molecular mechanism is not yet clear, dedifferentiation of articular chondrocytes is believed to play a role in the pathophysiology of arthritis. In addition to dedifferentiation, increased apoptotic death of chondrocytes was observed in arthritic cartilage, and apoptosis is closely related to cartilage destruction (6,7), indicating that chondrocyte apoptosis plays an important role in the pathogenesis of arthritis.
NO is generally believed to be an important mediator of the dedifferentiation and apoptosis of articular chondrocytes in arthritic cartilage (5,8,9). NO is produced in chondrocytes by the action of proinflammatory cytokines, such as interleukin-1␤. NO production in chondrocytes causes activation of matrix metalloproteinases (10), decreased production of interleukin-1 receptor antagonists (11), inhibition of proteoglycan synthesis and type II collagen expression (12,13), and apoptosis of chondrocytes (6,7,14). Indeed, inhibition of NO production protects against damage of cartilage and chondrocytes in a number of experimental models. For instance, in experimentally induced osteoarthritis in a range of animal species, a significant correlation was observed between the level of NO and the prevalence of apoptotic cells in cartilage tissue (7). Moreover, inhibition of NO resulted in reduced articular cartilage damage and apoptotic cell death (11,15). In our previous studies, we have shown that direct production of NO by treating chondrocytes with an NO donor, sodium nitroprusside (SNP), induces apoptosis and dedifferentiation of primary culture articular chondrocytes (16 -18). NO-induced apoptosis and dedifferentiation of articular chondrocytes were regulated by opposite functions of mitogen-activated protein (MAP) kinase subtypes, extracellular signal-regulated protein kinase (ERK), and p38 kinase (16). NO-induced activation of ERK-1/2 induces dedifferentiation with the inhibitory effects on apoptosis, whereas activation of p38 kinase induces apoptosis and is responsible for the maintenance of differentiated phenotype. In addition to MAP kinase signaling, NO production caused the inhibition of protein kinase C (PKC) ␣ and -activities (17). The inhibition of PKC␣ activity is caused by inhibition of its expression, which is independent of MAP kinase signaling. In contrast, PKC activ-ity is blocked as a result of p38 kinase activation, and inhibition of PKC activity is followed by proteolytic cleavage by caspase-3. We also found that p38 kinase induces NO-induced apoptosis by accumulating p53 via NFB-dependent transcription and stabilization by serine 15 phosphorylation (18).
NSAIDs such as aspirin and indomethacin have been used to relieve pain and inflammation in arthritic cartilage. NSAIDs exert their effects primarily by the inhibition of cyclooxygenase (COX), a key enzyme that converts arachidonic acid to prostaglandin (PG) (19). In addition to the alleviation of inflammation, some NSAIDs also modulate homeostasis of chondrocyte and cartilage such as matrix molecule synthesis. For example, several NSAIDs, such as sodium salicylate, inhibit proteoglycan synthesis but others, such as nimesulide, induce cartilage matrix synthesis, whereas others including piroxicam have no effect on matrix synthesis (20,21). In addition, some NSAIDs, such as nimesulide and ibuprofen, have a protective effect in staurosporine-induced apoptosis of chondrocytes (22). Several lines of evidence suggest that some of the effects of NSAIDs are independent of the inhibition of COX (23). Indeed, it has been shown that NSAIDs modulate COX-independent signaling pathways, such as Ras (24), NFB (25), activator protein-1 (26), ERK (26), p38 kinase (27), and others. Because the role of NSAIDs in the maintenance of homeostasis and apoptosis of articular chondrocytes is not clearly understood, although the action of NSAIDs in inflammation is clear, we investigated the function of various NSAIDs in NO-induced dedifferentiation and apoptosis and characterized the molecular mechanism of NSAIDs action in articular chondrocytes.

EXPERIMENTAL PROCEDURES
Isolation of Rabbit Articular Chondrocytes and Culture Conditions-Articular chondrocytes were isolated from knee joint cartilage slices of 2-week-old New Zealand White rabbits by enzymatic digestion as described previously (28). Briefly, cartilage slices were dissociated enzymatically for 6 h in 0.2% collagenase type II (381 units/mg solid; Sigma) in Dulbecco's modified Eagle's medium. After collecting individual cells by brief centrifugation, the cells were resuspended in culture medium supplemented with 10% (v/v) fetal bovine calf serum, 50 g/ml streptomycin, and 50 units/ml penicillin, and then plated on culture dishes at a density of 5 ϫ 10 4 cells/cm 2 . The medium was replaced every 2 days after plating, and cells reached confluence at ϳ5 days of culture. Cells from day 4 cultures were treated with various concentrations of the indicated pharmacological reagents for 1 h before SNP treatment. These reagents included various NSAIDs, PD98059 (Calbiochem) to inhibit MAP kinase kinase-1/-2 (29), SB203580 (Calbiochem) to inhibit p38 kinase (30), N-benzyloxycarbonyl-Asp-Glu-Val-Asp-fluoromethyl ketone (Bachem, Heidelberg, Germany) to inhibit caspase-3 (31), or SN50 (Biomol, Plymouth Meeting, PA) to inhibit nuclear translocation of NFB (32). In some experiments, where indicated, chondrocytes from day 3 cultures were infected with either control adenovirus or adenovirus containing wild-type rabbit PKC␣ or mouse PKC that was inserted into a cosmid cassette, pAxCAwt. Infected cells were cultured in complete medium for 24 h and were then treated with 1.5 mM SNP for an additional 24 h.
Cartilage Explants Culture-Rabbit joint cartilage explants (ϳ125 mm 3 ) were cultured in Dulbecco's modified Eagle's medium in the absence or presence of various pharmacological reagents as indicated in each experiment. The cartilage explants were fixed in 4% paraformaldehyde for 24 h at 4°C, dehydrated with graded ethanol, embedded in paraffin, and sectioned into 4-m slices as described previously (33). The sections were stained by standard procedures using Alcian blue to determine differentiation status of chondrocytes. Apoptotic cells were determined by the procedure described below.
Determination of Caspase-3 Activity-Activation of caspase-3 was determined by measuring the absorbance of the cleaved synthetic substrate of caspase-3, Ac-Asp-Glu-Val-Asp-chromophore p-nitroaniline, as described previously (16). Briefly, chondrocytes were lysed on ice for 10 min in the cell lysis buffer provided in the Clontech A ApoAlert CPP32 colorimetric assay kit. The lysates were reacted with 50 M Ac-Asp-Glu-Val-Asp-chromophore p-nitroaniline in a reaction buffer (0.1 M HEPES, 20% glycerol, 10 mM DTT, and protease inhibitors, pH 7.4). The mixtures were maintained at 37°C for 1 h in a water bath and subse-quently analyzed in an enzyme-linked immunosorbent assay reader. The enzyme activity was calculated from a standard curve prepared using p-nitroanaline. The relative levels of pNA were normalized against the protein concentration of each extract.
Determination of Apoptosis-We have shown previously that NOinduced death of primary culture of articular chondrocytes is caused by apoptosis, as demonstrated by DNA fragmentation and terminal deoxynucleotidyl transferase dUTP nick-end labeling (TUNEL) (16). In this study, apoptotic cells were quantified by counting surviving cells using an MTT assay kit (Roche Molecular Biochemicals) according to the manufacturer's protocol. Apoptosis of articular chondrocytes in cartilage explant culture was determined by TUNEL assay according to the manufacturer's protocol (Roche Molecular Biochemicals).
NFB Luciferase Assay-NFB activity was determined by a reporter gene assay. Briefly, chondrocytes were transfected with plasmid containing luciferase and three tandem repeats of serum response element or a control vector. Transfection of the expression vector was performed by using LipofectAMINE Plus as described previously (18). The transfected cells were cultured in complete medium for 24 h and used to determine luciferase activity using an assay kit from Promega. Luciferase activity was normalized against ␤-galactosidase activity. PGE 2 Assay-PGE 2 production in articular chondrocytes was determined by measuring the levels of cellular and secreted PGE 2 by using a PGE 2 assay kit (Amersham Biosciences). Briefly, chondrocytes were plated in standard 96-well microtiter plates at a density of 2 ϫ 10 4 cells/well. After treatment with various pharmacological reagents, as indicated in each experiment, the total cell lysate was used to quantify the amount of PGE 2 according to the manufacturer's protocol. PGE 2 levels were calculated against a standard curve of PGE 2 .
Western Blot Analysis-Whole-cell lysates were prepared by extracting proteins using a buffer containing 50 mM Tris-HCl, pH 7.4, 150 mM NaCl, 1% Nonidet P-40, and 0.1% SDS supplemented with protease inhibitors and phosphatase inhibitors as described above. The proteins were size-fractionated by SDS-polyacrylamide gel electrophoresis and transferred to a nitrocellulose membrane. Proteins were detected using the following antibodies: type II collagen from Chemicon (Temecula, CA); rabbit anti-p53 polyclonal antibody and phosphorylation-specific antibody for ERK from New England Biolabs (Beverly, MA); PKC␣, PKC, and ERK-1/2 from BD Transduction Laboratories; and p38 kinase from Santa Cruz Biotechnology Inc. Blots were developed using a peroxidase-conjugated secondary antibody and an enhanced chemiluminescence system.

NSAIDs Block NO-induced Apoptosis of Primary Culture
Articular Chondrocytes-Direct production of NO by SNP treatment causes dedifferentiation and apoptosis of articular chondrocytes (16 -18). To examine the effects of NSAIDs on articular chondrocyte function, we first examined the effects of NSAIDs on NO-induced apoptotic cell death. Consistent with our previous reports, SNP treatment in rabbit articular chondrocytes caused apoptosis in a dose-dependent manner, which was demonstrated by the reduction of cell viability determined by MTT assay (Fig. 1A). The reduction of cell viability was detectable as early as 3 h after SNP treatment (data not shown). The effects of NSAIDs on the survival of articular chondrocytes were examined by pretreating chondrocytes with various concentrations of NSAIDs for 1 h before exposure to 1.5 mM SNP for 18 h. As shown in Fig. 1B, all of the examined NSAIDs, including indomethacin, ketoprofen, ibuprofen, sulindac sulfide, and flurbiprofen, blocked chondrocyte apoptosis in a dose-dependent manner. These results indicate that NSAIDs block NO-induced apoptosis of articular chondrocytes.
NSAIDs Block NO-induced Dedifferentiation of Primary Culture Articular Chondrocytes-We also examined the effects of NSAIDs on NO-induced dedifferentiation of articular chondrocytes. NO production caused loss of the differentiated phenotype of articular chondrocytes, as demonstrated by the reduction of sulfated proteoglycan accumulation and type II collagen expression (16). Pretreatment of chondrocytes with indomethacin blocked the NO-induced decrease in sulfated proteoglycan accumulation as determined by Alcian blue staining ( Fig. 2A,  top) and type II collagen expression as determined by Western blot analysis ( Fig. 2A, bottom). Treatment of chondrocytes with indomethacin alone did not affect the accumulation of sulfated proteoglycan or type II collagen expression ( Fig. 2A). Other NSAIDs examined in this study (i.e. ketoprofen, ibuprofen, sulindac sulfide, and flurbiprofen), with effects similar to those of indomethacin, also blocked the NO-induced decrease in proteoglycan accumulation and type II collagen expression in a dose-dependent manner (Fig. 2B). These results indicate that NSAIDs inhibit not only NO-induced apoptosis but also dedifferentiation of primary culture articular chondrocytes.

NSAIDs Block NO-induced Apoptosis and Dedifferentiation of Articular Chondrocytes during Cartilage Explants Culture-
Because it is possible that the responses of chondrocytes in monolayer culture may differ from their responses in a 3-dimensional natural matrix, we examined whether NO production causes apoptosis and dedifferentiation of chondrocytes during cartilage explants culture and whether NSAIDs also block NO-induced apoptosis and dedifferentiation. Similar to the effects on chondrocytes cultured on plastic, NO production caused apoptotic cell death and inhibition of sulfated proteoglycan synthesis in chondrocyte in cartilage explants as demonstrated by TUNEL assay and Alcian blue staining, respectively (Fig. 3). In addition, pretreatment of chondrocytes with indomethacin blocked the NO-induced apoptosis and decrease in sulfated proteoglycan accumulation (Fig. 3), indicating that indomethacin blocks NO-induced apoptosis and dedifferentiation of chondrocytes cultured either on plastic or in a threedimensional natural matrix.
Modulation of NO-induced Activation of ERK and p38 Kinase by Indomethacin-We next investigated the molecular mechanism of NSAID modulation of apoptosis and dedifferen-tiation. We used indomethacin as an NSAID and first examined the possible modulation of NO-induced activation of the MAP kinase subtypes ERK-1/2 and p38 kinase, because these MAP kinase subtypes have opposite effects on NO-induced apoptosis and dedifferentiation (16). SNP treatment caused transient activation of ERK-1/2 and p38 kinase as determined by Western blot analysis and in vitro kinase assay, respectively (Fig. 4A, top). Treatment of chondrocytes with indomethacin before SNP treatment slightly but consistently enhanced NOinduced ERK-1/2 activation (Fig. 4A, middle). In contrast, NOinduced activation of p38 kinase was blocked by indomethacin treatment in a dose-dependent manner (Fig. 4A, bottom). entiation (Fig. 4, B and C). The present results are consistent with our previous observation (16) and indicate that NO-induced ERK activation causes dedifferentiation and inhibits apoptosis, whereas activation of p38 kinase induces apoptosis and inhibits dedifferentiation. Therefore, the inhibitory effects of indomethacin on NO-induced apoptosis and dedifferentiation with the potentiation and inhibition of ERK and p38 kinase, respectively, suggest that inhibition of apoptosis by indomethacin is caused by blockade of p38 kinase and potentiation of ERK. In contrast, the above results also suggest that modulation of MAP kinase subtypes by indomethacin is not responsible for the inhibition of NO-induced dedifferentiation, because the potentiation of ERK and inhibition of p38 kinase is a condition that enhances dedifferentiation.
Indomethacin Blocks NO-induced Inhibition of PKC␣ and -Activity-A previous study from our lab (17) indicated that inhibition of PKC␣ and -activities is required for NO-induced apoptosis and dedifferentiation of articular chondrocytes. We also demonstrated that inhibition of PKC is caused by activation of p38 kinase, whereas inhibition of PKC␣ is independent of MAP kinase signaling (17). We therefore examined whether indomethacin blocks NO-induced dedifferentiation and apoptosis by modulating PKC␣ and -signaling. Consistent with our previous results (17), expression and activity of PKC␣ anddecreased over time in SNP-treated chondrocytes (Fig. 5A). The NO-induced decrease in expression and activity of PKC␣ andwas completely blocked by the pretreatment of indomethacin in a dose-dependent manner (Fig. 5B). Moreover, ectopic expression of PKC␣ or -by adenovirus infection blocked NO-induced dedifferentiation (Fig. 5, C and D) and apoptosis (Fig. 5E). Therefore, the above results suggest that the inhibitory effects of indomethacin on NO-induced dedifferentiation are caused by the blockade of NO-induced inhibition of PKC␣ and -activity, whereas the inhibition of apoptosis by indomethacin is caused by its ability to modulate both MAP kinase subtypes and PKC␣ and -. Consistent with the inhibition of apoptosis, indomethacin also blocked signaling molecules downstream of p38 kinase during NO-induced apoptosis, such as activation of NFB, which was demonstrated by the inhibition of IB degradation (Fig. 6A) and NFB reporter gene assay (Fig. 6B), the accumulation of p53 (Fig. 6A), and the activation of caspases-3 (Fig. 6C).

NSAIDs Inhibit Apoptosis and Dedifferentiation Independent of the Expression of COX-2 and Production of PGE 2 -Because
NSAIDs block COX-2 activity and PGE 2 production and because PGE 2 is known to regulate differentiation of chondrocytes (34,35), we examined whether the inhibition of apoptosis and dedifferentiation by NSAIDs is caused by their ability to block COX-2 activity and PGE 2 production. NO-induced COX-2 expression (Fig. 7A, upper) and PGE 2 production (Fig. 7B) were blocked by all of the examined NSAIDs, and the effects of indomethacin on COX-2 expression were dose-dependent (Fig.  7A, bottom). Although a low concentration of indomethacin (i.e. 70 M) did not significantly block COX-2 expression (Fig. 7A), it completely blocked PGE 2 production (Fig. 7C), indicating that COX-2 was blocked at this concentration. Interestingly, although PGE 2 production was completely blocked at 70 M indomethacin, the effects on apoptosis and proteoglycan synthesis were observed at higher concentrations of indomethacin (Fig. 7C), suggesting that the ability of indomethacin to modulate apoptosis and dedifferentiation is independent of the inhibition of COX-2 and PGE 2 production.
This possibility was further examined by the specific inhibition of COX-2 with NS398. As shown in Fig. 8A, NS398 (5 M) completely blocked NO-induced PGE 2 production but did not affect apoptosis and accumulation of sulfated proteoglycans. In addition, treatment of chondrocytes with various concentrations of PGE 2 in the absence or presence of SNP did not affect cell viability, accumulation of sulfated proteoglycan (Fig. 8B), or type II collagen expression (Fig. 8D, top). The COX-2-and PGE 2 -independent regulation of apoptosis and dedifferentiation were further demonstrated by inhibiting ERK-1/2 and p38 kinase, which block NO-induced COX-2 expression (Fig. 8D, bottom) and PGE 2 production (Fig. 8C). Although inhibition of both MAP kinase subtypes blocked PGE 2 production, NO-induced apoptosis and dedifferentiation were conversely regulated by ERK-1/2 and p38 kinase (Fig. 8C). Taken together, these results indicate that the ability of indomethacin to block NO-induced apoptosis and dedifferentiation is independent of the inhibition of COX-2 activity and PGE 2 production. DISCUSSION NO production in articular chondrocytes plays a central role in the pathophysiology of arthritis (5,8,9,36). High levels of nitrite/nitrate are found in the synovial fluid and serum of arthritis patients (37), and it has been shown that NO causes loss of a differentiated phenotype and apoptosis of articular chondrocytes (6,7,(12)(13)(14). We have previously shown that direct production of NO causes apoptosis and dedifferentiation of articular chondrocytes by the activation of ERK-1/2 and p38 kinase and inhibition of PKC␣ and - (16 -18). In this study, we investigated the effects of various NSAIDs on NO-induced apoptosis and dedifferentiation of articular chondrocytes cultured on plastic or in a 3-dimensional natural matrix (i.e. explants culture) and found that all of the examined NSAIDs inhibit apoptosis and dedifferentiation in both conditions independent of the inhibition of COX-2 expression and PGE 2 production. As summarized in Fig. 9, we also demonstrated that the inhibitory effects of NSAIDs on apoptosis are caused by their ability to potentiate NO-induced ERK activation, to inhibit activation of p38 kinase, and to block inhibition of PKC␣ and -. In contrast, the inhibition of NO-induced dedifferentiation by NSAIDs is caused by the blockade of PKC␣ and -signaling but not the modulation of ERK-1/2 and p38 kinase signaling.
The cellular effects of NSAIDs are exerted primarily by the inhibition of COX and PGE 2 production. PGE 2 is known to regulate differentiation (34,35)  production in chondrocytes causes COX-2 expression and PGE 2 production, it is possible that NSAIDs modulate NO-induced apoptosis and dedifferentiation by inhibiting COX-2. However, our current results clearly indicate that the inhibitory effects of NSAIDs on NO-induced apoptosis and dedifferentiation are independent of the inhibition of COX-2 and PGE 2 production. This conclusion is clearly demonstrated by the specific inhibition of COX-2 activity and by the observation that PGE 2 production or exogenous PGE 2 did not affect NO-induced apoptosis and dedifferentiation. In addition, the concentration of indomethacin that is required for the inhibition of NO-induced apoptosis and dedifferentiation is higher than that needed for the inhibition of PGE 2 synthesis. This is consistent with observations that the COX-independent actions of NSAIDs such as inhibition of cell cycle progression (39,40), induction of apoptosis (41)(42)(43), and inhibition of angiogenesis (44,45), require high concentrations of NSAIDs that are 100-to 1000-fold higher than those needed to inhibit prostaglandin synthesis.
Our results indicated that the inhibition of NO-induced apoptosis of chondrocytes by NSAIDs is related to the potentiation of ERK activation, blockade of p38 kinase activation, blockade of PKC␣ and -inhibition, and inhibition of downstream apoptotic signaling, including NFB, p563, and caspase-3, as summarized in Fig. 9. The modulation of MAP kinase subtypes by indomethacin (i.e. inhibition of p38 kinase and potentiation of ERK) seems to be involved in the inhibition of NO-induced apoptosis based on the observation that NO-induced activation of p38 kinase induces apoptosis, whereas ERK activation inhibits apoptosis. The inhibition of p38 kinase and apoptosis by indomethacin is consistent with the inhibition of apoptotic signaling molecules located downstream of p38 kinase, such as PKC, NFB, p53, and caspase-3 (Fig. 9). In contrast to the inhibition of apoptosis, the inhibitory effects of indomethacin on NO-induced dedifferentiation are attributable to its ability to block NO-induced inhibition of PKC␣ and -, but not its ability to modulate MAP kinase subtypes. This conclusion is based on the observations that NO-induced inhibition of PKC␣ and -is required for the induction of dedifferentiation as well as apoptosis and that the potentiation and inhibition of ERK and p38 kinase, respectively, are the signaling events leading to the potentiation of dedifferentiation (Fig. 9). Modulation of ERK and p38 kinase by indomethacin is independent of its ability to block COX activity, consistent with previously reported results. Indeed, many studies have indicated that COXindependent effects of NSAIDs are exerted by the modulation of ERK and p38 kinase. For example, NSAIDs such as aspirin and sodium salicylate exert their effects by the inhibition ERK- 1/2 (26,(45)(46)(47), whereas p38 kinase is activated by sodium salicylate in human fibroblasts and is associated with induction of apoptosis (27).
Our current results clearly indicate that indomethacin blocked the NO-induced inhibition of PKC␣ and -activity that is required for the induction of apoptosis and dedifferentiation of SNP-treated chondrocytes (17). Because the activity of PKC is blocked as a result of p38 kinase activation (17), it is likely that the effects of indomethacin on PKC are caused by the inhibition of p38 kinase signaling. However, the possibility that indomethacin directly regulates PKC activity cannot be ruled out, although no evidence supports the direct action of NSAIDs in the regulation of PKC isoforms. Nevertheless, it is apparent that the blockade of the NO-induced inhibition of PKC by indomethacin inhibits both apoptosis and dedifferentiation of SNP-treated chondrocytes. In contrast to PKC, the inhibition of PKC␣ expression and activity is caused by the inhibition of its expression independent of ERK and p38 kinase signaling (17). Based on the observation that inhibition of PKC␣ activity is a prerequisite for the induction of apoptosis and dedifferentiation, the MAP kinase-independent inhibition of PKC␣ activity by indomethacin is also essential for the inhibitory effects of indomethacin on apoptosis and dedifferentiation. The mechanisms of indomethacin regulation of PKC␣ expression and activity remain to be determined, although it is possible that indomethacin regulates PKC␣ either directly or indirectly by modulating upstream signaling events. Nevertheless, because our present results indicated that indomethacin-induced inhibition of p38 kinase and potentiation of ERK is not directly involved in the inhibition of dedifferentiation as discussed above, we conclude that the blockade of the inhibition of PKC␣ and -activities by indomethacin plays a critical role in the inhibition of NO-induced dedifferentiation.
Consistent with the inhibition of apoptotic signaling mediators such as p38 kinase and PKC, indomethacin also blocked their downstream signaling molecules such as the activation of NFB, accumulation of p53, and activation of caspase-3. Indeed, it has been shown that some types of NSAIDs, including ibuprofen, sulindac, sulindac sulfide, and flurbiprofen, are able to inhibit NFB activation, whereas indomethacin, ketoprofen, and ketorolac are ineffective (23). Using a reporter gene assay and IB degradation, we found that ectopic expression of wild type PKC␣ or -by adenovirus infection in chondrocytes blocked the NO-induced activation of NFB (48), indicating that p38 kinase-dependent and -independent PKC and -␣ regulates NFB activation. We also found that NFB activation is required for COX-2 expression (48). This suggests that inhibition of PKC␣ and -is an upstream signaling event leading to NFB activation that causes COX-2 expression as summarized in Fig. 9. Given the inability of indomethacin to inhibit NFB (23), it is highly likely that the inhibitory effects of indomethacin on NFB activation observed in this study are caused by the inhibition of its upstream signaling molecules (i.e. p38 kinase and PKC) rather than by a direct action on NFB.
In summary, we found that various NSAIDs block apoptosis and dedifferentiation of articular chondrocytes caused by NO production in a manner independent of their ability to inhibit COX-2 and PGE 2 production. The inhibitory effects of NSAIDs on apoptosis are derived from their ability to block NO-induced activation of p38 kinase, to potentiate ERK activation, and to block inhibition of PKC␣ and -, whereas the inhibition of FIG. 9. Schematic summary of NSAIDs modulation of signaling pathway during NO-induced dedifferentiation and apoptosis of articular chondrocytes. NSAIDs inhibit NO-induced apoptosis by blocking NO-induced activation of p38 kinase, potentiating ERK activation, and blocking inhibition of PKC␣ and -, whereas the inhibition of NO-induced dedifferentiation by NSAIDs is caused by the blockade of PKC␣ and -signaling but not the modulation of ERK-1/2 and p38 kinase. The inhibitory effects of NSAIDs on apoptosis and dedifferentiation are independent of their ability to block COX-2 activity and PGE 2 production. NO-induced dedifferentiation by NSAIDs is caused by the blockade of PKC␣ and -signaling but not the modulation of ERK-1/2 and p38 kinase signaling. Additionally, the effects of NO production and NSAIDs treatment on chondrocytes derived from adult rabbit joint cartilage (from 4-month-old rabbits) or human osteoarthritic cartilage obtained from patients undergoing total knee arthroplasty are essentially same as in growth plate chondrocytes (from 2-week-old rabbits) (data not shown). Because NO production via inducible NO synthase in articular chondrocytes plays a central role in the pathophysiology of arthritis by causing inflammation, apoptosis, dedifferentiation, and the activation of matrix metalloproteinases, our results suggest that NSAIDs have protective effects on cartilage damage, not only by alleviating inflammation but also by inhibiting NO-induced apoptosis and dedifferentiation of articular chondrocytes.