p38 Kinase Regulates Nitric Oxide-induced Apoptosis of Articular Chondrocytes by Accumulating p53 via NFκB-dependent Transcription and Stabilization by Serine 15 Phosphorylation*

Nitric oxide (NO) during primary culture of articular chondrocytes causes apoptosis via p38 mitogen-activated protein kinase in association with elevation of p53 protein level, caspase-3 activation, and differentiation status. In this study, we characterized the molecular mechanism by which p38 kinase induces apoptosis through activation of p53. We report here that NO-induced activation of p38 kinase leads to activation of NFκB, which in turn induces transcription of the p53 gene. Activated p38 kinase also physically associates and phosphorylates the serine 15 residue of p53, which results in accumulation of p53 protein during NO-induced apoptosis. Ectopic expression of wild-type p53 enhanced NO-induced apoptosis, whereas expression of a dominant negative p53 blocked it, indicating that p53 plays an essential role in NO-induced apoptosis of chondrocytes. The increased accumulation of p53 caused expression of Bax, a pro-apoptotic member of the Bcl-2 family that is known to cause apoptosis via release of cytochrome c and caspase activation. These results suggest that NO-activated p38 kinase activates p53 function in two different ways, transcriptional activation by NFκB and direct phosphorylation of p53 protein, leading to apoptosis of articular chondrocytes.

Nitric oxide (NO) production via inducible NO synthase in chondrocytes plays a central role in degenerative diseases of cartilage, for example osteoarthritis and rheumatoid arthritis (1,2). Although NO-induced cartilage destruction can be caused by various ways, the increased apoptotic cell death of articular chondrocytes appears to be an important contributor (3)(4)(5)(6). However, the signaling pathway leading to NO-induced apoptosis of chondrocyte is poorly understood. In other cell types, it has been shown that subtypes of mitogen-activated protein kinase play an important role in NO-induced apoptosis. For example, NO-induced apoptosis of neuronal progenitor cells is mediated by p38 kinase (7), whereas extracellular sig-nal-regulated kinase and c-Jun N-terminal kinase but not p38 kinase are involved in NO-induced apoptosis of cardiomyocytes (8). Our recent study demonstrated that apoptosis of chondrocyte apoptosis caused by direct production of NO with the NO donor sodium nitroprusside (SNP) 1 is regulated by opposite functions of two mitogen-activated protein kinase subtypes, extracellular signal-regulated kinase 1/2 and p38 kinase, in association with the elevation of p53 protein level, caspase-3 activation, and differentiation status (9). SNP treatment stimulated activation of both extracellular signal-regulated kinase 1/2 and p38 kinase. The activated extracellular signal-regulated kinase 1/2 plays a role as an inducing signal for dedifferentiation and an inhibitory signal for NO-induced apoptosis, whereas p38 kinase functions as a signal for the maintenance of the differentiated status and as an inducing signal for apoptosis of chondrocytes. NO production is less pro-apoptotic in chondrocytes that are dedifferentiated by a serial monolayer culture that is associated with decreased potential of NO to activate p38 kinase in dedifferentiated cells compared with that in differentiated chondrocytes.
Apoptotic death of NO-treated chondrocytes is due to the ability of p38 kinase to stimulate caspase-3 activity, an inducer of apoptosis, and to stimulate expression and/or accumulation of p53, a known signaling molecule that acts upstream of caspase-3 (9). The signaling pathways leading to caspase activation during apoptosis involves the release of cytochrome c and other apoptogenic factors from injured mitochondria. The release is mediated by the translocation of cytosolic Bax, a pro-apoptotic member of the Bcl-2 family, to mitochondria in response to various apoptotic stimuli (10 -13). Several studies indicate that p53 regulates the function of Bax and mitochondrial integrity (14 -16). p53 has a short half-life, and the proapoptotic function of p53 is achieved by increased expression at the transcriptional level and by post-translational stabilization of the protein by escaping from ubiquitin-dependent degradation (17,18). Phosphorylation of p53 at multiple sites is the main post-translational modification that is regulated by several different protein kinases depending on types of cells and extracellular stimuli. The protein kinases include ataxia telangiectasia-mutated kinase (19,20) and the major subtypes of mitogen-activated protein kinase, i.e. extracellular signal-regulated kinases 1 and 2 (21,22), p38 kinase (22)(23)(24)(25)(26), and c-Jun NH 2 -terminal kinase (27).
In addition to p53, mounting evidence indicates that a nuclear factor B (NFB) regulates apoptosis by protecting cells from apoptosis in most cases. However, NFB also has pro-apoptotic function, depending on cell type and extracellular stimuli (28). NFB exists in a latent form in the cytoplasm of unstimulated cells bound to an inhibitory protein, IB. Upon stimulation of cells, IB is rapidly phosphorylated at serine residues, leading to ubiquitin-dependent degradation by the 26 S proteasome. The released NFB dimer from IB then translocates to the nucleus and activates target genes by binding to the promoter/enhancer region (29). The known target genes of NFB include both apoptosis-protective genes such as Bcl-2 (30) and pro-apoptotic p53 (31), suggesting that the cell typeand extracellular stimuli-dependent effects of NFB on apoptosis may be due to its specific effects on the expression of apoptosis-regulating genes.
The functional relationship between NFB and p53 during the regulation of apoptosis is controversial, especially when NFB functions as a pro-apoptotic signal. For instance, genotoxic agents stimulate transcription of p53 via activation of NFB (31), whereas induction of p53 causes activation of NFB, that correlates with the ability of p53 to induce apoptosis (32). Thus, NFB appears to function as either an upstream or downstream signaling molecule of p53 during the apoptotic process. Moreover, there is no available evidence indicating the involvement of NFB and its relationship with p53 in NO-induced apoptosis. We therefore investigated the role and underlying molecular mechanism of NFB and p53 in NO-induced and p38 kinase-mediated apoptosis of rabbit articular chondrocytes by using SNP as a NO donor. The investigation additionally focused on the characterization of the downstream signal of p53. We report here that NO-induced activation of p38 kinase activates NFB, which led to increased transcriptional expression p53. p38 kinase also associated with and phosphorylated the serine 15 residue of p53, which caused accumulation of p53 by stabilization of the protein. The increased expression and accumulation of p53 caused apoptosis by inducing expression and activation of pro-apoptotic Bax.

Culture of Rabbit Articular Chondrocytes and Experimental
Condition-Rabbit articular chondrocytes were released from cartilage slices by enzymatic digestion, as previously described (9). To summarize, cartilage slices were dissociated enzymatically for 6 h in 0.2% collagenase type II (381 units/mg solid, Sigma). Individual cells were suspended in Dulbecco's modified Eagle's medium supplemented with 10% (v/v) fetal bovine calf serum, 50 g/ml streptomycin, and 50 units/ml penicillin, after which they were plated on culture dishes at a density of 5 ϫ 10 4 cells/cm 2 . The cells from day 3 cultures were treated with various concentrations of SNP for the indicated period and reached confluence in ϳ5 days. In some experiments, the cells were pretreated with various inhibitors or activators for 30 min unless otherwise indicated, including 50 g/ml of SN-50 (Biomol, Plymouth Meeting, PA) to inhibit nuclear translocation of NFB (33) and SB203580 (Calbiochem) to inhibit p38 kinase (34).
NO Assay and Determination of Caspase-3 Activity and Apoptosis-NO production was measured by estimating nitrite using the Griess reagent as previously described (9). Activation of caspase-3 was determined by measuring the absorbance at 405 nm after cleavage of the synthetic substrate acetyl-DEVD-p-nitroanilide as described previously (9). Apoptotic cell death in this study was quantified by a flow cytometric assay based on the number of cells with fragmented DNA (9). Briefly, cells were harvested by centrifugation and fixed in 80% ethanol that had been precooled to Ϫ20°C. The cells were re-suspended in phosphate-buffered saline containing 50 g/ml propidium iodide, 0.1% Nonidet P-40, and 100 g/ml RNase A (Sigma) and incubated for 1 h. The number of cells with fragmented DNA was then quantified using 1-2 ϫ 10 4 cells on a FACSort flow cytometer (BD PharMingen).
Immunoprecipitation and Kinase Assay-Cell lysates were prepared in a lysis buffer containing 20 mM Tris-HCl, pH 7.5, 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1% Triton X-100, 2.5 mM sodium pyrophosphate, 1 mM ␤-glycerol phosphate, and inhibitors of proteases (10 g/ml leupeptin, 10 g/ml pepstatin A, 10 g/ml aprotinin, and 1 mM 4-(2-aminoethyl) benzenesulfonyl fluoride) and phosphatases (1 mM NaF and 1 mM Na 3 VO 4 ). Samples were precipitated with antibody against p38 kinase (Santa Cruz Biotechnology Inc.) or phosphorylated p38 kinase (New England Biolabs, Beverly, MA), and immune complexes were collected on protein A-Sepharose beads (Pierce). The immune complex was used for Western blot analysis or p38 kinase assay as previously described (9). For p38 kinase assay, the beads were re-suspended in 20 l of kinase reaction buffer containing 25 mM Tris-HCl, pH 7.5, 5 mM ␤-glycerol phosphate, 2 mM dithiothreitol, 0.1 mM sodium orthovanadate, 10 mM MgCl 2 , [␥-32 P]ATP, and 1 g of ATF-2 fusion protein as a substrate (New England Biolabs). The kinase reaction was performed for 30 min at 30°C, and ATF-2 phosphorylation was determined by autoradiography. To determine direct phosphorylation of p53 by p38 kinase, cells were treated with SNP for 12 h, and active p38 kinase was precipitated with phospho-specific anti-p38 kinase antibody (New England Biolabs). Immune complex kinase assays were then performed as described above using glutathione S-transferase fusion protein of p53. After electrophoresis and transfer of proteins to a nitrocellulose membrane, phosphorylation of p53 was determined by Western blot analysis using phosphorylation site-specific anti-p53 monoclonal antibody from Cell Signaling Technology (Beverly, MA).
Extraction of Nuclear Protein and Electrophoretic Mobility Shift Assay-Chondrocytes were treated with various concentrations of SNP for the indicated time period. Nuclear extracts were then prepared as described by Schreiber et al. (35). Briefly, cell pellets were suspended in hypotonic buffer containing 10 mM Hepes, pH 7.9, 10 mM KCl, 0.1 mM EGTA, 0.1 mM EDTA, and inhibitors of proteases and phosphatases as described above and lysed by the addition of 0.07% Nonidet P-40. The nuclei were pelleted by centrifugation and re-suspended in extraction buffer containing 20 mM Hepes, pH 7.9, 25% glycerol, 400 mM NaCl, 1 mM EDTA, 1 mM EGTA and inhibitors of proteases and phosphatases. Nuclear protein was collected by centrifugation at 10,000 ϫ g for 10 min at 4°C. The nuclear extracts (10 g) were assayed for NFB DNA binding activity as described by Rupec and Baeuerle (36) using doublestranded oligonucleotide 5Ј-AGTTGAGGGGACTTTCCCAGGC-3Ј (the binding site is underlined: Promega, Madison, WI). Binding reactions were performed in 20 mM Hepes, pH 7.9, 80 mM NaCl, 0.1 mM EDTA, 1 mM dithiothreitol, 8% glycerol, and 2 g of poly(dI-dC) (Amersham Biosciences) for 20 min, and samples were analyzed by electrophoresis in a 6% nondenaturing polyacrylamide gel.
Transient Transfection and NFB Luciferase Assay-For the luciferase assay, chondrocytes were transfected with plasmid containing luciferase and three tandem repeats of serum response element or a control vector. In experiments where indicated, chondrocytes were cotransfected with dominant negative forms of IKK␣, IKK␤, or IB␣ (37), dominant negative p38 kinase (38), wild-type p53, dominant negative p53 (p53 273 ), or p53 mutant in which serine 6,9,15,20,33,37, and threonine 18 were mutated to alanine (39). Transfection of the expression vector was performed as described previously (40). The expression vector was introduced to cells using LipofectAMINE PLUS (Invitrogen) using the procedure recommended by the manufacturer. The transfected cells were cultured in complete medium for 24 h and used for further assay as indicated in each experiment. Luciferase gene expression and activity were analyzed using an assay kit purchased from Promega and normalized by ␤-galactosidase activity.
Pulse-Chase Analysis-Pulse-chase analysis of p53 protein was performed to determine p53 stability. Briefly, chondrocytes, treated with 1 mM SNP for 24 h were preincubated for 20 min in methionine-free medium, pulse-labeled for 30 min with 50 Ci/ml [ 35 S]methionine, washed twice, and incubated in medium containing excess unlabeled methionine for 0, 1, 2, or 4 h. At each indicated time point, the cells were lysed, and p53 protein was immunoprecipitated as described above.
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 antibodies purchased from the following sources. Mouse anti-p53 monoclonal antibody was from New England Biolabs, phosphorylation sitespecific anti-p53 polyclonal or monoclonal antibodies was from Cell Signaling Technology, anti-IB and anti-Bax antibody were from Transduction Laboratories (Lexington, KY), and mouse monoclonal p21 antibody was from Santa Cruz Biotechnology Inc. (Santa Cruz, CA). The blots were developed using a peroxidase-conjugated secondary antibody and enhanced chemiluminescence system.

RESULTS
p38 Kinase Stimulates NFB Activity, Which Functions as a Pro-apoptotic Signal in Articular Chondrocytes-Treatment of chondrocytes with SNP (1 mM) leads to activation of the NFB transcription factor as determined by the increased DNA binding activity of NFB (Fig. 1A), transcriptional activation of the NFB-responsive promoter (Fig. 1B), and degradation of IB (Fig. 1C). SNP-induced activation of NFB was blocked by the addition of 50 g/ml SN-50, an inhibitory peptide for nuclear translocation of the activated NFB. Because NFB can either protect or stimulate apoptosis, depending on cell type and extracellular stimuli, we examined whether NFB activation is essential for NO-induced apoptosis. As shown in Fig. 2A, inhibition of NFB activity with the SN-50 peptide significantly blocked SNP-induced caspase-3 activation and apoptosis. SNPinduced NFB activation (Fig. 2B) and apoptosis (Fig. 2C) were also blocked by the transient expression of dominant negative forms of IB kinases (IKK␣ and IKK␤) or IB␣. Therefore, the above results clearly indicate that SNP-induced activation of NFB functions as a pro-apoptotic signal in articular chondrocytes.
The results of a recent study by our group (9) indicate that NO production in chondrocytes activates p38 kinase, which signals the induction of apoptosis. Therefore, in the present study we examined the role of p38 kinase in NFB activation. Consistent with our previous observation, SNP treatment caused transient activation of p38 kinase in a dose-dependent manner (Fig. 3A). Blocking p38 kinase activation by treatment with SB203580 or ectopic expression of dominant negative p38 kinase inhibited NO-induced caspase-3 activation and apoptosis (Fig. 3B). Treatment of cells with SN-50 peptide before SNP treatment, a condition that inhibits NFB activation (Fig. 1, A and B) did not affect SNP-induced p38 kinase activation (Fig.  3C). However, inhibition of p38 kinase with SB203580 or expression of dominant negative p38 kinase significantly blocked SNPinduced IB degradation (Fig. 3C) and NFB transcriptional activity (Fig. 3D). Therefore, activation of p38 kinase appears to lead to NFB activation in SNP-treated articular chondrocytes.
NFB-dependent Transcriptional Activation of the p53 Genes during NO-induced Apoptosis-Our previous study showed that p38 kinase-mediated apoptosis of SNP-treated chondrocytes is associated with an increase in p53 protein level (9). To investigate the functional relationship between NFB activation and p53 protein level, the effects of NFB inhibition on p53 protein level were first examined. The SNP-induced increase in p53 protein level was abrogated by the inhibition of p38 kinase with SB203580 or expression of dominant negative p38 kinase. Inhibition of NFB with SN-50 peptide or by forced expression of dominant negative IKK␣, IKK␤, or IB␣ also abrogated the SNP-induced increase in p53 protein level (Fig. 4A). In contrast, however, ectopic expression of wild-type p53 or dominant negative p53 did not affect SNP-induced IB degradation (Fig.  4B), indicating that NFB activity is necessary to increase the p53 protein level in response to NO. We next examined the role of p53 in caspase activation and apoptosis to determine whether the increase in p53 protein level is directly involved in NO-induced apoptosis. Forced expression of a dominant negative mutant form of p53 (p53 273 ) blocked SNP-induced caspase-3 activation and apoptosis of chondrocytes (Fig. 4, C and D). In contrast, ectopic expression of wild-type p53 significantly enhanced SNP-induced caspase activity and apoptosis (Fig. 4, C and D), indicating an essential role for p53 in NOinduced chondrocyte apoptosis.
To elucidate the mechanism leading to the increase in p53 protein level, we first examined whether p53 proteins are accumulated during NO-induced apoptosis by transcriptional activation of the p53 gene. Consistent with the increased protein level, mRNA transcript levels of p53, as determined by reverse transcription-PCR, were also increased in a time-dependent manner in cells treated with SNP (Fig. 5A). The enhanced transcription of p53 was reduced by the inhibition of NFB (with SN-50 peptide) or p38 kinase (with SB203580 or ectopic expression of a dominant negative p38 kinase) (Fig. 5B), indicating that NFB-mediated enhanced transcription of p53 leads to accumulation of p53 protein during NO-induced apoptosis.
p38 Kinase Stabilizes p53 by Direct Phosphorylation at Serine 15-In addition to transcriptional regulation, p53 activity is more commonly regulated by post-translational stabilization of protein that blocks ubiquitin-dependent degradation by the 26 S proteasome (29). Therefore, pulse-chase analyses were used to determine whether NO production causes an alteration in the p53 half-life. The results, shown in Fig. 6A, clearly indicate a dramatically increased half-life of p53 in SNP-treated cells, suggesting that post-translational stabilization of p53 also contributes to the increased p53 protein level upon production of NO. p38 Kinase-mediated p53 Regulation during Chondrocyte Apoptosis Because phosphorylation of p53 at multiple sites commonly leads to its stabilization, we determined which site(s) of p53 is phosphorylated in response to SNP treatment by Western blot analysis using phosphorylation site-specific antibodies. Among the phosphorylation sites tested, SNP treatment caused significantly increased phosphorylation at serine 15 (Fig. 6B). The kinetics of dose-and time-dependent phosphorylation on serine 15 (Fig. 6C) is similar to that of the increase in p53 protein level (9). Serine 15 phosphorylation of p53 was reduced to the control level when p38 kinase was blocked with SB203580 or by the expression of dominant negative p38 kinase (Fig. 6D). Inhibition of NFB activity with SN-50 also abrogated phosphorylation of serine 15, probably because of low available level of p53 protein.
Because the above observation suggests that p38 kinase is responsible for p53 phosphorylation, the role of p38 kinase was further analyzed. As shown in Fig. 7A, immunoprecipitation of p38 kinase or phosphorylated p38 kinase caused co-precipitation of serine 15-phosphorylated p53, indicating a physical association between p38 kinase and p53. To examine whether p38 kinase can directly phosphorylate serine 15 of p53, active p38 kinase was precipitated using phosphorylation-specific antibody, and an in vitro kinase assay was performed using a glutathione S-transferase fusion protein of p53. Western blot analysis indicated that p38 kinase phosphorylates serine 15 of p53 in vitro (Fig. 7B). Thus, activated p38 kinase in SNPtreated cell appears to associate and directly phosphorylate p53. Consistent with this observation, p53 phosphorylation was not affected by the addition of wortmannin (Fig. 7C), which inhibits ataxia telangiectasia-mutated kinase (41), a well known upstream kinase of p53. The increased accumulation of p53 and potentiation of apoptosis and caspase activity in cells transfected with wild-type p53 was reduced by the co-transfection of a dominant negative p38 kinase (Fig. 4, B-D) or inhibition of p38 kinase with SB203580 (data not shown), indicating the importance of p53 phosphorylation by p38 kinase for stabilization of p53 protein. Finally, ectopic expression of p53 mutant in which serine 6,9,15,20,33, and 37 and threonine 18 are mutated to alanine (39) reduced NO-induced apoptosis (Fig. 7D) and accumulation of p53 protein (Fig. 7E). This further supports the importance of serine 15 phosphorylation in the stabilization of p53 protein and subsequent induction of apoptosis.
p53 Induces Transcriptional Expression of Pro-apoptotic Bax-p53 functions as a transcription factor to regulate target genes involved in various processes including apoptosis. To determine the transcriptional activity of accumulated p53, the expression of p21, that is a well known target gene of p53, was examined. As shown in Fig. 8A, expression level of p21 was increased in SNP-treated cells, and the increase in p21 was blocked by the inhibition of p38 kinase with SB203580 or NFB with SN-50 or expression of dominant negative forms IKK␣, IKK␤, or IB␣, indicating that p53 induces p21 during NOinduced apoptosis.

p38 Kinase-mediated p53 Regulation during Chondrocyte Apoptosis
To elucidate the pro-apoptotic role of p53 in SNP-treated cells, we next investigated whether p53 regulates expression and/or activity of Bax, a pro-apoptotic member of the Bcl-2 family. Bax level was low in control cells and was dramatically increased by NO production. The kinetics of Bax expression were similar to those of the expression/accumulation and serine 15 phosphorylation of p53 (Fig. 8B, upper panel). The increase in Bax expression was blocked by the inhibition of p38 kinase with SB203580 or NFB with SN-50 or expression of dominant negative form of IKK␣, IKK␤, or IB␣ (Fig. 8B,  middle and lower panels). DISCUSSION Differentiated chondrocyte, which is the only cell type found in normal mature cartilage, is responsible for the maintenance of homeostasis and integrity of cartilage. The homeostasis is destroyed in degenerative diseases such as osteoarthritis and rheumatoid arthritis (42). The generation of NO in chondrocytes is primarily caused by inflammatory cytokines and is one of the leading causes of the destruction of cartilage homeostasis, including apoptotic death of chondrocytes (43). Our recent study (9) provides evidence that NO-induced activation of p38 kinase functions as an inducing signal for apoptosis of chon-drocytes. In this study, we further characterized the molecular mechanism of p38 kinase-mediated apoptosis and were able to show that NO-induced activation of p38 kinase activates NFB, leading to the increased transcriptional expression of p53, and also phosphorylates the serine 15 residue of p53, resulting in stabilization and accumulation of p53. Further experiments showed that p53 causes apoptosis by inducing expression and translocation of pro-apoptotic Bax. Our working hypothesis for the signaling pathway leading to NO-induced apoptosis of articular chondrocytes is depicted in Fig. 9.
Our current study clearly indicates that NO-induced p38 kinase activation stimulates a transcription factor NFB. p38 kinase appears to activate NFB by phosphorylating IKK, an upstream kinase of IB, leading to release of NFB (29), because expression of dominant negative IKK␣ or -␤ inhibited p38 kinase-induced NFB activation (Fig. 2B). Our results are consistent with other reports, which indicate the potential role of p38 kinase in NFB activation in other experimental systems (44 -46). Blocking apoptosis by the inhibition of NFB activation (Fig. 2, A and C) indicates that the activated NFB functions as a pro-apoptotic signal in SNP-treated chondrocytes. Although NFB is known to protect cells from apoptosis in most cases, it is also known to contribute to apoptosis depending on cell types and extracellular stimuli (28). The proapoptotic effects of activated NFB may be due to transcriptional stimulation of pro-apoptotic genes such as p53 (31). We confirmed in this study that activated NFB in SNP-treated articular chondrocytes regulates expression of pro-apoptotic p53, as evidenced by the observation that inhibition of NFB activation abrogated an increase in p53 (Fig. 4A) and mRNA transcript (Fig. 5B). Because ectopic expression of p53 did not affect IB degradation (Fig. 4B), NFB appears to regulate p53 gene expression in articular chondrocytes (Fig. 9), although the possibility that SNP treatment increases stability of p53 mRNA transcript could not be eliminated.
Regulation of p53 function is complicated. The known mechanisms for regulation include increased rate of transcription and translation, a qualitative change of p53 from a latent to active form as a transcription factor, and post-translational stabilization of the protein. It is generally believed that the increase in p53 protein level occurs through stabilization and accumulation of p53 via down-regulation of Mdm2 expression (47) and alterations in the p53 protein (17,18). Increasing evidence indicates that phosphorylation of p53 at various sites by multiple stress-activated protein kinases, including ataxia telangiectasia-mutated kinase (19,20) and mitogen-activated  5. NFB regulates transcriptional expression of p53. A, chondrocytes were treated with 1 mM SNP for the indicated period. B, chondrocytes were pretreated for 30 min with vehicle alone or 50 g/ml SN50. Alternatively, the cells were transfected with dominant negative p38, cultured for 24 h in complete medium, and untreated or treated for 30 min with SN50. The cells were then incubated with 1 mM SNP for 24 h. After isolation of total RNA, reverse transcription-PCR was performed to detect p53 and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) transcripts. The data represent results of a typical experiment performed at least four times. MW, molecular weight.
protein kinase (22,23,25), increases p53 protein half-life due to the interference of Mdm2 function, resulting in inhibition of ubiquitination and degradation of p53 (48). Phosphorylation sites of interest include serines 15, 20, and 37 and threonine 18, which are located within or close to the Mdm2 binding site of p53. Among these sites, serine 15 has been studied more closely because it is the site phosphorylated by ataxia telangiectasia- p38 Kinase-mediated p53 Regulation during Chondrocyte Apoptosis mutated kinase (19,20) and also by extracellular signal-regulated kinase (22) and p38 kinase (22,26). In addition to serine 15, p38 kinase has been shown to phosphorylate serine 33 (25), 33 and 46 (23), and 389 (24) in various experimental systems. Because serine 33 and 46 are not present in rabbit p53, it is likely that serine 15 is a major target for the p38 kinase. Consistent with our current observation, it has been shown that NO production causes serine 15 phosphorylation of p53 in other cell types, although the signaling pathway leading to its phosphorylation has not been elucidated (49). Phosphorylation of p53 by p38 kinase leads to stabilization of p53 in most cases with few exceptions, where phosphorylation stimulates functional activity without stabilization (25). In this study, we demonstrated that NO-induced activated p38 kinase physically interacts and phosphorylates serine 15 of p53 as determined by co-immunoprecipitation and in vitro kinase assay (Fig. 7). Phosphorylation of serine 15 does not involve ataxia telangiectasia-mutated kinase because wortmannin, which is known to inhibit ataxia telangiectasia-mutated kinase (41), does not affect its phosphorylation (Fig. 7C). Because serine 15 phosphorylation blocks Mdm2-dependent degradation of p53 (48), its phosphorylation appears to be responsible for the stabilization of p53, as we observed in this study (Fig. 6A).
The importance of p53 phosphorylation by p38 kinase for the stabilization of p53 protein was demonstrated by the observation that increased accumulation of p53 by the transfection of wild-type p53 was reduced by the co-transfection of dominant negative p38 kinase (Fig. 4) or by the inhibition of p38 kinase with SB203580 (data not shown). We also observed that ectopic expression of p53 mutant in which serine 6,9,15,20,33, and 37 and threonine 18 are mutated to alanine reduced NO-induced apoptosis and accumulation of p53 protein (Fig. 7), which further supports the importance of serine 15 phosphorylation in p53 protein stabilization. Although stabilization of p53 is important for transcriptional activation, our current results also suggest that transcriptional expression of p53 is also essential for p53 accumulation. This conclusion is based on the observation that inhibition of p53 expression after the inhibition of NFB activity blocked accumulation of p53 (Fig. 6D). Under this condition, serine 15 phosphorylation of p53 is also weak; this may be due to the low level of p53 available for phosphorylation by p38 kinase.
There is little evidence available that indicates a pro-apoptotic role of p53 in chondrocytes. For example, a study by Yatsugi et al. (5) suggests the possible involvement of p53 in chondrocyte apoptosis and cartilage destruction. Immunohistochemical analyses of p53 in the articular cartilage of patients with rheumatoid arthritis and osteoarthritis showed increased staining of the tumor suppressor protein in chondrocytes with apoptotic morphology. Although activation of p53 is believed to play an important role in NO-induced apoptosis of various cell types (50), some recent studies using p53Ϫ/Ϫ cells indicate that p53 is not required for NO-induced apoptosis in certain cell types including alveolar macrophages (51) and vascular smooth muscle cells (52). However, our recent observation that the potentiation of SNP induced apoptosis and caspase activation in p53-transfected cells and that inhibition of apoptosis by the ectopic expression of mutant p53 (Fig. 4) strongly suggests that increased p53 protein levels contribute to NO-induced apoptosis of articular chondrocytes.
To date, the mechanism of p53 regulation of apoptosis has not been clearly elucidated. However, the best understood function of p53 is as a transcription factor, an activity that contributes to apoptotic function (53). We confirmed in this study the transcriptional activity of p53 by examining the expression of p21 Waf1/Cip1 , a well known target gene of p53. Few genes are known to regulate p53-mediated apoptotic function, including Bax, although a large number of cellular genes that are transcriptionally regulated by p53 have been described. Among the genes, the function of Bax is best studied, and it clearly contributes to apoptosis (14). In response to apoptotic signals, Bax translocates to the mitochondria from the cytosol, where it causes a decline in mitochondrial membrane potential followed by cytochrome c release and caspase activation (10,12,13,15). Although the promoter sequence of Bax contains a putative NFB binding site, NFB does not directly regulate Bax expression (54). Our current study suggests that p53 directly regulates Bax expression. In addition, it is also likely that NFB-dependent transcriptional expression of Bax is mediated by p53. This idea is consistent with many other observations that indicate direct regulation of Bax expression by p53 (14,55). Although it is possible that the response of p53 is dependent on the coordinate expression of several gene products (56), it is likely that p53-mediated expression and redistribution of Bax causes SNP-induced apoptosis via release of cytochrome c and caspase activation as depicted in Fig. 9.
In summary, we demonstrate that NO-induced activation of p38 kinase stimulates NFB, which leads to the increased transcriptional expression of p53. p38 kinase also associates with and phosphorylates the serine 15 residue of p53, resulting in accumulation of p53 by stabilization of the protein. The increased expression and accumulation of p53 causes apoptosis by inducing expression and activation of pro-apoptotic Bax.