Protein kinase C delta induces apoptosis of vascular smooth muscle cells through induction of the tumor suppressor p53 by both p38-dependent and p38-independent mechanisms.

Apoptotic death of vascular smooth muscle cells (SMCs) is a prominent feature of blood vessel remodeling. In the present study, we examined the novel PKC isoform protein kinase C delta (PKCdelta) and its role in vascular SMC apoptosis. In A10 SMCs, overexpression of PKCdelta was sufficient to induce apoptosis, whereas inhibition of PKCdelta diminished H2O2-induced apoptosis. Moreover, evidence is provided that the tumor suppressor p53 is an essential mediator of PKCdelta-induced apoptosis in SMCs. Activation of PKCdelta led to accumulation as well as phosphorylation of p53 in SMCs; this induction correlated with apoptosis. Furthermore, blocking p53 induction with small interference RNA or targeted gene deletion prevented PKCdelta-induced apoptosis, whereas restoring p53 expression rescued the ability of PKCdelta to induce apoptosis in p53 null SMCs. We also establish that PKCdelta regulates p53 at both transcriptional and post-translational levels. Specifically, the transcriptional regulation required p38 MAPK, whereas the post-translational modification, at least for serine 46, did not involve MAPK. Additionally, PKCdelta, p38 MAPK, and p53 co-associate in cells under conditions favoring apoptosis. Together, our data suggest that SMC apoptosis proceeds through a pathway that involves PKCdelta, the intermediary p38 MAPK, and the downstream target tumor suppressor p53.


Apoptosis of vascular smooth muscle cells (SMCs)
is a well established component of the remodeling that occurs during normal development of the circulatory system (1,2) as well as during the course of neointimal formation after intervention for atherosclerosis (3)(4)(5). Because increased total cellularity is a prominent feature of an occluding neointima, the balance between proliferation and apoptosis during vessel healing appears central to this pathologic process (6,7). Indeed, accumulating evidence suggests that abnormal SMC apoptosis leads to neointimal hyperplasia (8 -11). However, despite the importance of SMC apoptosis, the precise molecular mechanism underlying the regulation of apoptotic pathways in SMCs remains largely undetermined.
Apoptosis is a multistage, genetically controlled process of selective cell deletion. Protein kinases regulate the early stages of apoptosis by phosphorylating key proteins (12,13), whereas caspases, a family of cysteine proteases, are the main effectors whose activation results in the characteristic morphological changes associated with programmed cell death such as membrane blebbing, chromatin condensation, and DNA fragmentation (14,15).
Members of the protein kinase C (PKC) family are activated by diverse stimuli and participate in multiple cellular processes such as growth, differentiation, and apoptosis (16). The novel PKC isoform, protein kinase C delta (PKC␦), has been shown to be associated with the response to DNA damage and other apoptotic stimuli in specific cell types (17)(18)(19)(20). The critical role of PKC␦ in vascular SMC apoptosis and pathogenesis of a neointimal lesion has been recently demonstrated using PKC␦ "knock-out" mice. The PKC␦ null mice developed exacerbated vein graft intimal lesions that contain fewer apoptotic vascular cells compared with the wild-type mice (21). Furthermore, aortic SMCs isolated from PKC␦ null mice are resistant to apoptotic stimuli including H 2 O 2 . However, the mechanism by which PKC␦ mediates SMC apoptosis remains to be defined.
The tumor suppressor p53 is the master regulator of cell cycle arrest and apoptosis. In particular, an important role of p53 in the pathogenesis of vascular diseases is suggested by decreased p53 levels in human restenotic (22) and atherosclerotic lesions (23). The importance of p53 is also confirmed in various animal models. Adenovirus-mediated gene transfer of p53 to rat carotid arteries inhibited neointimal formation following balloon injury (24), whereas target deletion of p53 led to larger intimal lesions in a mouse vein graft model (25).
The mitogen-activated protein kinase (MAPK) p38 has been shown to be activated by cellular stress, UV light radiation, growth factor withdrawal, and pro-inflammatory cytokines (26 -29). Upon activation, p38 phosphorylates various transcription factors but, of particular note, has been demonstrated to phosphorylate the tumor suppressor p53 (30). p38 MAPK is also implicated in both pro-apoptotic and anti-apoptotic signaling pathways (12,31) However, its activity is likely cell type-specific, with most studies focusing predominantly on inflammatory cells (32).
In this report we scrutinized the precise molecular mechanism of PKC␦-induced vascular SMC apoptosis and provide evidence that PKC␦ activation leads to accumulation/modification of p53, which is essential for the induction of apoptosis. Moreover, the role of p38 as an intermediate in PKC␦-induced p53 accumulation and apoptosis has also been demonstrated. In total, these studies provide an explicit link between PKC␦, p38 activation, and p53 modulation in the process of SMC apoptosis. These findings have implications for the development of improved approaches for the prevention and management of restenosis.

EXPERIMENTAL PROCEDURES
General Materials-Phorbol 12-myristate 13-acetate (PMA) was purchased from Biomol (Plymouth Meeting, Pennsylvania), and dimethyl sulfoxide (Me 2 SO), rottlerin, SB20358, and H 2 O 2 , along with other chemicals not specified, were purchased from Sigma. Dulbecco's modified Eagle's medium and cell culture reagents were from Invitrogen.
Antibodies-The rabbit polyclonal antibody to PKC␦ and the mouse monoclonal antibody to ␤-actin were obtained from Santa Cruz Biotechnology (Santa Cruz, California). Polyclonal rabbit antibodies to cleaved casapase-3 and phospho-p38 were obtained from Cell Signaling Technology Inc. (Beverly, Massachusetts). Biotinylated p53 antibody and antibody to phosphorylated p53 were obtained from R&D Systems (Minneapolis, Minnesota).
SMC Culture-Rat aortic A10 SMCs, obtained from the American Tissue Culture Collection, were grown as recommended. Mouse aortic SMCs were isolated from the thoracic aorta of p53Ϫ/Ϫ male mice (33) (Jackson Laboratories, Bar Harbor, Maine) based on a protocol described by Clowes et al. (34) and maintained in Dulbecco's modified Eagle's medium containing 10% fetal bovine serum at 37°C with 5% CO 2 .
Construction of Adenoviral Vectors and Infection-A recombinant adenoviral vector was constructed to express PKC␦. Briefly, a DNA fragment containing the desired sequence was generated by PCR using the human cDNA as a template. Following DNA sequencing, the PCR product was then cloned into an E1-and E3-deficient adenoviral vector (pEasy). Adenoviruses were propagated in HEK 293 cells and purified by CsCl density gradient centrifugation. A recombinant adenovirus encoding for the wild-type p53 protein was a generous gift from Enrico Ascher (24,35).
Apoptosis Assay-DNA fragmentation was determined using the Cell Death Detection ELISA system (Roche Applied Science), an assay based on a quantitative sandwich-enzyme-immunoassay principle using mouse monoclonal antibodies directed against DNA and histones. Activation of caspase 3 was quantified by Western blotting using an antibody specific for cleaved caspase-3.
Immunoblotting-Protein extracts were resolved by electrophoresis as described previously (36). Equal amounts of protein extracts were separated by SDS-PAGE and transferred to a polyvinylidene difluoride membrane and blotted with antibodies. Labeled proteins are visualized with an ECL system (Amersham Biosciences).
Transient Transfection and Luciferase Activity Assay-The P1-p53 promoter luciferase reporter was a generous gift from David Reisman (37). Transient transfections and luciferase assays were performed as described previously (38). Briefly, plasmid DNA was introduced into A10 SMCs by using SuperFect (Qiagen, Valencia, CA). After transfection, cells were recovered in media containing 10% fetal bovine serum for 48 h, followed by PMA (1 M) for 12 h. In all experiments, 50 ng of the renilla luciferase construct (pRL-CMV) was co-transfected as an internal control.
Antisense Oligo-An antisense oligonucleotide specific to rat PKC␦ and its scrambled control were obtained from Biognostik (Göttingen, Germany). 200 M oligo was introduced to the media of cultured A10 SMCs, seeded 18 h earlier at equal density. The efficacy of oligo uptake by SMCs was Ͼ90%, determined by using a fluorescein isothiocyanatelabeled oligo provided by the manufacturer.
Quantitative Reverse Transcription PCR-As per our laboratory's protocol (36), total RNA was isolated using RNA Aqueous (Ambion, Austin, Texas) and reverse-transcribed using a reverse transcriptase kit and probes from Applied Biosystems (Foster City, Calif). Quantification of mRNA was performed using the ABI Prism7700 (Applied Biosystems).
Immunoprecipitation-Immunoprecipitation was carried out as described previously by our laboratory (39). Briefly, SMCs were lysed in Nonidet P-40 buffer. Total protein concentration was determined by a modification of the method of Lowry, and the protein amount of each sample was then equalized. Following primary antibody incubation and centrifugation, pellets were washed three times with Nonidet P-40 buffer and one time with 50 mM Tris. The final pellet was re-suspended in 30 l of sample buffer and heated to 95°C for 3 min. Samples were then subjected to SDS-PAGE.
Statistical Analysis-Values were expressed as a fold increase (means Ϯ S.E.). Unpaired Student's t test was used to evaluate the statistical differences. Values of p Ͻ 0.05 were considered significant. All experiments were repeated at least three times.  (Fig. 1A). Alternatively, we inhibited PKC␦ expression by treating cells with an antisense oligonucleotide. Compared with the control oligo, PKC␦ antisense oligo produced a significant reduction in the level of endogenous PKC␦ (Fig. 1B). More importantly, the PKC␦ antisense oligo-treated cells became resistant to H 2 O 2 treatment (Fig. 1B). Taken together, these data confirmed that PKC␦ is a necessary component of the apoptotic pathway in vascular SMCs.

PKC␦ Is Necessary for Smooth Muscle Cell Apoptosis-We
Overexpression of PKC␦ Induces Apoptosis and Accumulation of the Tumor Suppressor p53-We next evaluated whether overexpression of PKC␦ would be sufficient to induce vascular SMC apoptosis. To this end, we employed an adenovirus encoding full-length wild-type PKC␦ (AdPKC␦), which led to a marked increase in cellular levels of PKC␦ (Fig. 2B). Additionally, AdPKC␦ induced a small but significant elevation in the level of fragmented DNA and cleaved caspase-3 (Fig. 2, A and  B). To facilitate the activation of the ectopically expressed PKC␦, we treated A10 SMCs with 1-5 M of PMA for 12 h. At these concentrations, PMA alone did not induce apoptosis as indicated by the lack of fragmented DNA as well as the absence of activated caspase-3. However, the PKC activator PMA, in combination with overexpression of PKC␦, resulted in an increase in DNA fragmentation by Ͼ3-fold ( Fig.  2A) and cleaved caspase-3 by Ͼ300% (Fig. 2B). These data establish that overexpression of PKC␦ is sufficient to induce SMC apoptosis. Because p53 has been implicated in SMC apoptosis, we investigated whether the overexpression of PKC␦ and its dramatic increase in SMC apoptosis were associated with an induction of p53. As shown in Fig. 2B, overexpression of PKC␦ in A10 SMCs significantly increased p53 levels. In parallel to the induction of apoptosis, the ability of AdPKC␦ to induce p53 expression was further enhanced by PMA (Fig. 2B). Next, we tested whether the PKC␦-induced p53 accumulation leads to enhanced p53dependent gene transcription by using a luciferase reporter gene containing a p53-specific enhancer element (40,41). Co-transfection of a PKC␦ expression vector significantly increased p53 reporter activity. More importantly, the addition of the PKC activator PMA facilitated the effect of PKC␦ on the p53 reporter, which is consistent with the ability of PKC␦ to induce p53 expression and SMC apoptosis (Fig. 2C).
p53 Is Necessary for PKC␦-induced Apoptosis-To confirm the significance of p53 in PKC␦-induced apoptosis, we designed a specific p53 small interference RNA (siRNA) to block p53 translation. 72 h following the administration of this p53 siRNA (50 nM), levels of p53 in A10 SMCs were decreased by 44 Ϯ 2.2% (Fig. 3A). Next, we examined whether the p53 siRNA affects apoptosis. 48 h following infection with AdPKC␦ or AdNull, A10 cells were incubated with p53 siRNA (50 nM for 72 h) prior to PMA treatment (1 M for 12 h). Cell apoptosis was then assessed using ELISA for DNA fragmentation and Western blot analysis for cleaved caspase-3. Inhibition of p53 with the specific siRNA led to a significant decrease in both PKC␦-induced cleaved caspase-3 (Fig. 3A) and DNA fragmentation (Fig. 3B), suggesting that p53 is necessary for PKC␦-induced SMC apoptosis. To confirm these findings using siRNA, we isolated SMCs from the thoracic aorta of p53 null mice and tested their ability to undergo apoptosis following overexpression of PKC␦ and activation with PMA. Interestingly, AdPKC␦ failed to induce apoptosis in p53 null SMCs (Fig. 3C). Next, we attempted to rescue apoptosis by restoring p53 expression using an adenovirus encoding wild type p53 (35). Overexpression of p53 alone did not induce apoptosis, which is consistent with reports in HCT116 colon carcinoma cells (42). However, the expression of p53 restored completely the ability of PKC␦ to induce apoptosis (Fig. 3C). These results provide further confirmation of the requirement of p53 for PKC␦-induced apoptosis of vascular SMCs.
PKC␦ Increases p53 Transcription-We next explored the mechanism by which PKC␦ regulates p53. We began by examining the effect of PKC␦ overexpression on the level of p53 mRNA. A10 SMCs were infected with AdNull or AdPKC␦ and then treated with the PKC activator PMA (1 M for 6h). Total RNA was isolated from control or PKC␦/PMA-treated cells, and p53 was quantified using TaqMan real  Apoptosis was determined by ELISA-measured DNA fragmentation. B, A10 cells were treated exactly as described for panel A. Cell lysate was analyzed with anti-PKC␦, p53, and cleaved caspase-3 antibody. C, A10 cells that were transiently transfected with a luciferase reporter gene under the control of p53-specific enhancer elements and either a PKC␦ expression or control plasmid. Following infection, cells were treated with PMA (1 M for 12 h). Reporter activity is expressed as a ratio of firefly luciferase to renilla luciferase. *, p Ͻ 0.05; **, p Ͻ 0.01 as compared with non-PMA treated control; n ϭ 3.
time reverse transcription PCR analysis. As shown in Fig. 4A, overexpressed PKC␦ elicited a significant increase in the level of p53 mRNA.
To determine whether the increase in p53 mRNA is secondary to increased promoter activity, we employed a luciferase construct containing the proximal portion of the human p53 promoter (37). We cotransfected A10 cells with a PKC␦ expression or control vector and the luciferase reporter. This experiment demonstrated an increase in p53 promoter activity in response to PKC␦ overexpression (Fig. 4B), indi-cating that PKC␦ regulates p53 transcription by up-regulating promoter activity in vascular SMCs.
Overexpression of PKC␦ Results in the Phosphorylation of p53-Because phosphorylation is an important element of p53 regulation, we next investigated the possibility that PKC␦ may affect p53 through phosphorylation. To assess PKC␦-induced p53 phosphorylation, we utilized a specific antibody to p53 phosphorylated on serine residue 46, shown by others to provide p53 with greater affinity to promoters of . C, SMCs were harvested from the thoracic aorta of p53Ϫ/Ϫ male mice. 24 h following seeding at equal densities in 10% fetal bovine serum media, mouse aortic SMCs were infected with equal quantities of AdNull, AdPKC␦, or adenovirus (Ad) p53 (30,000 total viral particles per cell). Following PMA treatment (1 M for 12 h), apoptosis was evaluated though ELISA-measured DNA fragmentation. *, p Ͻ 0.05 as compared with non-treated control; n ϭ 3.
apoptosis-related genes (43). PKC␦ substantially increased the level of phospho-p53 (at serine 46), which was barely detectable in cells infected with the empty viral vector (Fig. 5A). To further confirm the role of PKC␦-induced p53 phosphorylation in apoptosis, we evaluated the phosphorylation status of p53 in A10 cells treated with H 2 O 2 , a stimulus for both apoptosis and p53 phosphorylation (44). Indeed, Western blot analysis of A10 cell lysates demonstrated a significant increase in p53 phosphorylation at serine 46 following treatment with H 2 O 2 (200 M for 4 h). Importantly, pre-incubation with the PKC␦ chemical inhibitor rottlerin (1 M for 1 h) dramatically diminished the ability of H 2 O 2 to induce p53 phosphorylation (Fig. 5B). These data demonstrate that PKC␦ is responsible, directly or indirectly, for p53 phosphorylation. Lastly, we searched for the mechanism that enables the interaction between PKC␦ and p53 by examining the potential physical association between the two proteins. Lysates from A10 SMCs infected with AdPKC␦ were immunoprecipitated for p53 followed by Western blotting for PKC␦. As a negative control, the same cell lysate was immunoprecipitated with normal rabbit IgG. We found a prominent p53 band in the PKC␦ immunoprecipitate. In the converse experiment, PKC␦ was detected in the p53 immunocomplex (Fig. 5C). These findings suggest that PKC␦ is physically associated with p53 in conditions favoring apoptosis.
p38 MAPK Is Necessary for PKC␦-induced Up-regulation of p53-Next, we explored the molecular mechanisms underlying the p53 induction by evaluating the stress-activated kinase p38 in response to PKC␦ overexpression. Overexpression of PKC␦ in A10 SMCs activated p38 as demonstrated by the marked increase in p38 phosphorylation (Fig. 6A). To investigate a possible role of p38 in PKC␦-induced p53 expression, we inhibited p38 using the p38 chemical inhibitor SB20358 (20 M for 1 h) in PKC␦-overexpressing cells. We found that p38 inhibition resulted in a large decrease in the ability of PKC␦ to up-regulate p53 protein levels (Fig. 6B). Next, we tested the effect of SB20358 on p53 promoter activity. As shown above, p53 promoter activity was increased in response to PKC␦ overexpression. However, inhibition of p38 by SB20358 diminished the ability of PKC␦ to up-regulate p53 transcription (Fig. 6C). These data suggest that p38 is the intermediate responsible for the induction of p53 transcription in PKC␦-overexpressing SMCs.
p38 MAPK Is Not Necessary for PKC␦-induced p53 Phosphorylation-Because investigations in non-vascular cell types have demonstrated that p38 MAPK is a potential p53 serine 46 kinase and that p53-mediated apoptosis is dependent on this event, we explored whether these findings apply to SMCs. Chemical inhibition of p38 failed to block PKC␦-induced p53 phosphorylation but did decrease total p53 protein levels as demonstrated above (Fig. 7A). Despite the lack of influence of p38 MAPK on serine 46 phosphorylation, we found that both proteins were co-associated with PKC␦ by a co-immunoprecipitation assays (Fig. 7B).
p38 MAPK Is Necessary for PKC␦-induced SMC Apoptosis-After establishing a potential direct interaction between PKC␦ and p38 and the necessity of p38 for PKC␦-induced up-regulation of p53, we investigated the role of p38 MAPK in SMC apoptosis using the p38 inhibitor SB20358. Interestingly, a treatment with this inhibitor (20 M for 1 h) prior to PMA activation(1 M for 12 h) decreased apoptosis by Ͼ50% as quantified by both cleaved caspase-3 (Fig. 8A) and DNA fragmentation   for 12 h). Cell lysates were immunoprecipitated with an anti-p53 or anti-PKC␦ antibody. As a negative control, cell lysates were also immunoprecipitated with normal IgG. Immunoprecipitate was analyzed for PKC␦ or p53 via Western blot (WB) analysis. (Fig. 8B). Thus, p38 MAPK contributes in part to PKC␦-induced SMC apoptosis.

DISCUSSION
PKC␦, a member of the novel PKC subfamily, can be pro-apoptotic or anti-apoptotic depending on cell types and stimuli. In agreement with a previous report (21), our results support the notion that PKC␦ is proapoptotic in vascular SMCs. Because molecular activation of PKC␦ alone was sufficient to result in caspase-3 activation and DNA fragmentation, we speculate that activation of PKC␦ is an early event leading to the onset of programmed cell death in SMCs.
The molecular mechanisms linking PKC␦ to the induction of apoptosis have been explored to some extent in non-smooth muscle cells. Several studies suggest that the presence of a positive regulatory loop between PKC␦ and caspase-3; however, exactly how PKC␦ might stimulate caspase-3 remains unclear (45, 46). Another important target of PKC␦ is the mitochondria. It was demonstrated in HeLa cells (47) and keratinocytes (48) that overexpression and activation of PKC␦ leads to a reduction in mitochondrial membrane potential and release of cyto-chrome c, which subsequently leads to activation of caspases and apoptosis. Additionally, several nuclear proteins, including DNA-dependent protein kinase (49), p73␤ (50), and lamin B (51) have been identified as PKC␦ targets/substrates. Activated PKC␦ associates with and phosphorylates these proteins; such interactions, at least in part, contribute to apoptosis. In the current study, we present evidence that the tumor suppressor p53 is a necessary mediator of PKC␦-induced apoptosis in . Total p53 protein levels were assessed via Western blot. C, A10 cells were co-transfected with a PKC␦ expression or control vector and the luciferase construct containing the p53 promoter. SB20358 (20 M for 1 h) was used to inhibit p38 activity prior to the addition of PMA (1 M for 12 h). p53 reporter activity was expressed as a ratio of firefly luciferase to renilla luciferase. *, p Ͻ 0.05 as compared with non-treated control. FIGURE 7. p38 co-associates with both PKC␦ and p53 but is not necessary for PKC␦induced p53 phosphorylation. A, A10 SMCs, infected with AdPKC␦ or AdNull, were preincubated with the p38 MAPK inhibitor SB20358 (20 M for 1 h) prior to stimulation with PMA (1 M for 12 h). Cell lysates were examined via Western blotting using a phospho-p53 antibody specific for serine residue 46. B, A10 SMCs were infected with AdPKC␦ and stimulated for PMA (1 M for 12 h). Cell lysates were immunoprecipitated with an anti-p38 antibody. As a negative control, cell lysates were also immunoprecipitated with normal IgG. The immunoprecipitate was analyzed for PKC␦ and p53 via Western blot (WB) analysis. vascular SMCs. We have shown that activation of PKC␦ led to accumulation as well as phosphorylation of p53 in SMCs; this induction correlated with SMC apoptosis. Moreover, blocking p53 induction with siRNA prevented apoptosis. Finally, targeted gene deletion of p53 prevented PKC␦-induced apoptosis, whereas restoring p53 expression through adenovirus-mediated p53 gene transfer rescued the ability of PKC␦ to induce apoptosis. To our knowledge, this is the first demonstration of the direct involvement of p53 in the regulation of SMC apoptosis by PKC␦.
Experimental evidence is now provided that PKC␦ regulates p53 at both transcriptional and post-translational levels, apparently mediated by separate signaling mechanisms. Specifically, the transcriptional regulation requires p38 MAPK, whereas the post-translational modification, at least for Ser-46, does not require MAPK. The importance of post-translational regulation of p53 by PKC␦ is demonstrated by our observation that ectopic expression of p53 alone was insufficient to induce apoptosis. Only when co-expressed with PKC␦ was p53 able to restore apoptosis of p53 null cells, presumably through a PKC␦-dependent phosphorylation. It was surprising that the p38 MAPK-specific inhibitor, SB20358, significantly inhibited PKC␦-induced accumulation of p53 but did not affect p53 phosphorylation at Ser-46, because direct interaction between p53 and p38 MAPK (52) has been suggested previously. Indeed, we showed that all three proteins, p53, p38 and PKC␦, co-associated in the immunoprecipitation complex isolated from SMC lysates. Our attempts to evaluate other serine residues within the p53 molecule are currently not possible due to a lack of specific phosphoantibodies. Therefore, it remains to be determined whether PKC␦ and p38 stimulate p53 phosphorylation at additional residues. With respect to Ser-46, our data suggest that it is p38-independent.
The potential role of PKC␦ in the regulation of p53 accumulation has been suggested previously. Using the selective PKC␦ inhibitor rottlerin, Niwa et al. (44)  We have now demonstrated that in vascular SMCs the overexpression of PKC␦ increased the accumulation of both the p53 protein and mRNA. In contrast, Johnson et al. (55) found that the PKC␦ inhibitor rottlerin increases p53 levels in cisplatin-treated HeLa cells, whereas the PKC activator phorbol 12,13-dibutyrate attenuates p53 levels in the same cell line. Therefore, it is likely that PKC␦, paralleling its dual functions in apoptosis, plays multiple roles in regulating p53 expression, which is dependent upon the cell type and stimuli employed.
Our findings that PMA alone did not induce apoptosis in control or AdNull-infected SMCs is seemingly surprising, because these cells express endogenous PKC␦, and PMA activates PKC␦. The lack of cell death associated with PMA treatment may be related to the effect of PMA on other PKC isotypes. We have shown previously that SMCs express at least eight isotypes of PKC, among which six can be activated by PMA (56). Some of these PMA-sensitive PKC isoforms, such as PKC␦, are pro-apoptotic, whereas others, such as PKC␣, have been demonstrated to be anti-apoptotic in non-SMCs (57)(58)(59). Our data show that overexpressing PKC␦ shifts the PMA response in favor of apoptosis. However, the outcome of simultaneously activating multiple PKC isotypes, as may occur in control SMCs treated with PMA, appears to result in no net effect on cell death.
The present study highlights another interesting finding that PKC␦ stimulates p53 gene expression through p38 MAPK. We showed that inhibition of p38 completely eliminated PKC␦-stimulated accumulation of p53 protein and mRNA. Moreover, analyses using a p53 promoter reporter demonstrated that PKC␦ up-regulated p53 promoter activity, also in a p38-dependent manner. Several transcription factors have been identified to bind and regulate the murine p53 promoter, including NFB (60). Interestingly, Kim et al. (61) showed in articular chondrocytes that NO-induced activation of p38 up-regulates p53 expression through NFB. Moreover, the NFB pathway has been shown to be activated by PKC␦ in several cell types (62)(63)(64). Future studies are mandatory to directly test the role of NFB in PKC␦-induced p38-dependent transcriptional regulation of p53.
In summary, our results demonstrate that PKC␦ plays a pivotal role in the signal transduction pathway leading to vascular SMC apoptosis. The p38-dependent-accumulation and independent-phosphorylation of p53 by PKC␦ contributes, at least in part, to SMC apoptosis. Given the critical role of apoptosis in intimal hyperplasia, it is possible that enhancement of PKC␦ activity at different stages after vascular intervention may provide a new strategy for the prevention and treatment of restenosis.