Protein kinase Cδ (PKCδ) regulates cell apoptosis and survival in diverse cellular systems. PKCδ translocates to different subcellular sites in response to apoptotic stimuli; however, the role of its subcellular localization in its proapoptotic and antiapoptotic functions is just beginning to be understood. Here, we used a PKCδ constitutively active mutant targeted to the cytosol, nucleus, mitochondria, and endoplasmic reticulum (ER) and examined whether the subcellular localization of PKCδ affects its apoptotic and survival functions. PKCδ-Cyto, PKCδ-Mito, and PKCδ-Nuc induced cell apoptosis, whereas no apoptosis was observed with the PKCδ-ER. PKCδ-Cyto and PKCδ-Mito underwent cleavage, whereas no cleavage was observed in the PKCδ-Nuc and PKCδ-ER. Similarly, caspase-3 activity was increased in cells overexpressing PKCδ-Cyto and PKCδ-Mito. In contrast to the apoptotic effects of the PKCδ-Cyto, PKCδ-Mito, and PKCδ-Nuc, the PKCδ-ER protected the cells from tumor necrosis factor–related apoptosis-inducing ligand–induced and etoposide-induced apoptosis. Moreover, overexpression of a PKCδ kinase-dead mutant targeted to the ER abrogated the protective effect of the endogenous PKCδ and increased tumor necrosis factor–related apoptosis-inducing ligand–induced apoptosis. The localization of PKCδ differentially affected the activation of downstream signaling pathways. PKCδ-Cyto increased the phosphorylation of p38 and decreased the phosphorylation of AKT and the expression of X-linked inhibitor of apoptosis protein, whereas PKCδ-Nuc increased c-Jun NH2-terminal kinase phosphorylation. Moreover, p38 phosphorylation and the decrease in X-linked inhibitor of apoptosis protein expression played a role in the apoptotic effect of PKCδ-Cyto, whereas c-Jun NH2-terminal kinase activation mediated the apoptotic effect of PKCδ-Nuc. Our results indicate that the subcellular localization of PKCδ plays important roles in its proapoptotic and antiapoptotic functions and in the activation of downstream signaling pathways. (Mol Cancer Res 2007;5(6):627–39)
The ubiquitously expressed isoform protein kinase Cδ (PKCδ) has been shown to regulate cell apoptosis and survival in various cells depending on the specific cellular system and apoptotic stimuli (1-3). Most studies report a proapoptotic function of PKCδ in response to various stimuli, such as H2O2 (4), ceramide (5), tumor necrosis factor-α (6), and the DNA-damaging agents UV radiation (7), cisplatin (8), and etoposide (9). Different apoptotic stimuli induce caspase-dependent cleavage of PKCδ, which result in the generation of a constitutively active catalytic fragment (10, 11). The cleavage of PKCδ has been implicated in its apoptotic function (12, 13), and the expression of the catalytic fragment has been shown to induce cell apoptosis in different cellular systems (14, 15). The apoptotic function of PKCδ has been associated with the activation of multiple signaling proteins, such as c-Jun NH2-terminal kinase (JNK; ref. 16), p38 (17), AKT (18), p73 (19), DNA-PK (20), scramblase 3 (21), and lamin B (22).
In addition to its apoptotic functions, PKCδ has been also reported to exert antiapoptotic effects. Thus, PKCδ protects macrophages from apoptosis induced by nitric oxide (23) and exerts antiapoptotic effects on glioma cells treated with tumor necrosis factor–related apoptosis-inducing ligand (TRAIL; ref. 24) or infected with a virulent strain of Sindbis virus (25). Similarly, PKCδ promotes survival and chemotherapeutic drug resistance of non–small cell lung cancer cells (26).
One of the factors that may contribute to the diverse effects of PKCδ on cell apoptosis is its different subcellular localization. Indeed, various apoptotic stimuli induce translocation of PKCδ to distinct subcellular sites. Thus, PKCδ translocates to the Golgi in response to ceramide and IFN-γ (5, 27) to the mitochondria in response to oxidative stress, UV radiation, and phorbol 12-myristate 13-acetate (28, 29) and to the nucleus in response to etoposide, γ-irradiation, and cytosine arabinoside (9, 30). Translocation of PKCδ to the Golgi, mitochondria, and nucleus in response to these apoptotic stimuli has been associated with proapoptotic effects of this isoform (5, 9, 28). In addition, PKCδ has been shown to translocate to the endoplasmic reticulum (ER) in cells infected with Sindbis virus and in glioma cells treated with TRAIL, where PKCδ exerts antiapoptotic effects (24, 25). Although these studies suggest different functions of PKCδ in the various subcellular sites, the role of the translocation of PKCδ in its proapoptotic and antiapoptotic effects and in the activation of downstream signaling pathways is largely not defined.
In this study, we explored the role of the subcellular localization of PKCδ in its proapoptotic and antiapoptotic functions. Specifically, we asked whether the localization of PKCδ in the ER provides antiapoptotic signals as opposed to proapoptotic signals in the mitochondria and nucleus and how the localization of PKCδ in the distinct subcellular sites affects the activation of downstream signaling pathways. For these experiments, we targeted PKCδ to the cytoplasm, mitochondria, nucleus, and ER using the pShooter vectors, which have been widely used to examine the role of the subcellular localization of various proteins in their specific cellular functions (31-33). We found that overexpression of a constitutively active form of PKCδ (PKCδ-CA) in the cytosol, mitochondria, or nucleus resulted in cell apoptosis. In contrast, expression of PKCδ-CA in the ER had no apoptotic effect but it rather protected glioma and HeLa cells against TRAIL- and etoposide-induced apoptosis. In addition, we found that the subcellular localization of PKCδ had an important role in the induction of distinct apoptosis-related signaling pathways.
Targeted Localization of PKCδCA in Distinct Subcellular Sites of the A172 Glioma Cells
PKCδ has been associated with the regulation of both cell apoptosis and survival (3, 6, 8, 9, 24, 30). One of the factors that may contribute to the diversity of PKCδ effects is its ability to localize to different subcellular sites. To examine the role of PKCδ localization in its proapoptotic and antiapoptotic functions, we used a PKCδ constitutively active mutant that was targeted to the cytosol (PKCδ-Cyto), mitochondria (PKCδ-Mito), nucleus (PKCδ-Nuc), and ER (PKCδ-ER) using the pShooter vectors (31-33). A172 cells were transiently transfected with the different pShooter vectors, and the localization of PKCδ was examined using anti-Myc antibody (Fig. 1A). In parallel, A172 cells were transfected with the corresponding pShooter vectors encoding green fluorescent protein (GFP), which have been used as positive controls (Fig. 1B). Figure 1 shows immunostaining of A172 cells transfected with the different pShooter vectors using anti-Myc antibody and A172-expressing GFP pShooter vectors. The expression of PKCδ was selectively localized to the cytoplasm, nucleus, mitochondria, and ER in the PKCδ-Cyto, PKCδ-Nuc, PKCδ-Mito, and PKCδ-ER, respectively (Fig. 1A). The localization of PKCδ in the different subcellular sites was similar to that of the corresponding GFP pShooter vectors (Fig. 1B).
The localization of the different PKCδ shooter vectors in their respective subcellular sites was also shown using the ER marker anti-calnexin, the nuclear marker 7-aminoactinomycin D (7-AAD), and the mitochondria marker MitoTracker Orange. Figure 1C shows that anti-calnexin, 7-AAD, and MitoTracker Orange exhibited similar patterns to that of PKCδ-ER, PKCδ-Nuc, and PKCδ-Mito, respectively. Moreover, merged images clearly showed colocalization of the green fluorescence of PKCδ-ER, PKCδ-Nuc, and PKCδ-Mito with the red fluorescence of calnexin, 7-AAD, and MitoTracker Orange, respectively. In contrast, the PKCδ-Cyto did not colocalize with any of the organelle markers that were used in this study (data not shown).
Overexpression of PKCδ-CA in the Cytoplasm, Mitochondria, and Nucleus, but not in the ER, Induces Cell Apoptosis
We next examined the effect of the organelle-targeted PKCδ on cell apoptosis. For these experiments, we used both the A172 and the HeLa cells. Following transfection with the various PKCδ pShooter vectors, the expression of PKCδ was determined using anti-Myc antibody (Fig. 2A) and cell apoptosis was determined using propidium iodide staining and fluorescence-activated cell sorting (FACS) analysis and by anti–histone ELISA. As shown in Fig. 2B, transfection of A172 cells with the PKCδ-Cyto, PKCδ-Mito, and PKCδ-Nuc induced a large degree of cell apoptosis in the cells. The most significant effect was induced by the PKCδ-Nuc and PKCδ-Mito, whereas a smaller effect was observed with the PKCδ-Cyto. The apoptotic effect was first observed after 24 h of the transfection, and by 48 h thereafter, the majority of the cells were apoptotic (data not shown). In contrast, cell apoptosis was not induced by the PKCδ-ER by 24 h (Fig. 2B) or 48 h after transfection (data not shown). Overexpression of the A172 cells with the PKCδ-ER also did not induce major morphologic changes in the cells, whereas cells overexpressing the PKCδ-Cyto, PKCδ-Mito, and PKCδ-Nuc exhibited apoptotic morphology (Fig. 2C). Similar results were obtained using anti–histone ELISA (Fig. 2D).
We also examined the effect of the different PKCδ pShooter vectors in the HeLa cells and found similar results. Thus, PKCδ-Cyto, PKCδ-Mito, and PKCδ-Nuc induced cell apoptosis, whereas no apoptosis was observed in cells overexpressing PKCδ-ER (Fig. 2E).
Targeting PKCδ to Distinct Subcellular Sites Results in Differential Cleavage of PKCδ and Activation of Caspase-3
The induction of cell apoptosis by PKCδ has been associated with cleavage of this isoform and with the generation of a constitutively active catalytic fragment (4, 9, 13, 30); however, the role of PKCδ localization in its cleavage is not well characterized. We therefore examined the cleavage of PKCδ in cells overexpressing the different PKCδ pShooter vectors. The expression of the ectopic PKCδ was detected using Western blot analysis and an anti-PKCδ antibody that preferentially recognizes the rat PKCδ and not the human kinase. As presented in Fig. 3A, overexpression of the PKCδ-CA in the cytoplasm and in the mitochondria resulted in cleavage of PKCδ and in the accumulation of a 40-kDa catalytic fragment, whereas no significant cleavage was observed in PKCδ-Nuc and PKCδ-ER. Similar results were obtained using an anti-Myc antibody that recognizes the COOH-terminal Myc tag (Fig. 2A).
The cleavage of PKCδ is mostly mediated by caspase-3, which is activated by PKCδ (6, 9, 12). Therefore, we examined the activity of caspase-3 in A172 cells expressing the different PKCδ pShooter vectors. Figure 3B shows that overexpression of PKCδ-Cyto and PKCδ-Mito significantly increased the activity of caspase-3 compared with control vector–expressing cells, whereas only a slight increase was observed in the PKCδ-Nuc–overexpressing cells. In contrast, no increase in the activation of caspase-3 was observed in cells overexpressing the PKCδ-ER.
To examine the role of caspase-3 activity in the cleavage of PKCδ in the different subcellular sites and in the apoptosis induced by the PKCδ-Cyto and the PKCδ-Mito, we used the caspase-3 inhibitor DEVD-FMK (10 μmol/L). Treatment of the cells with DEVD-FMK following the transfection of the cells with the different pShooter vectors decreased the cleavage of the PKCδ-Cyto and the PKCδ-Mito (Fig. 3C). In addition, treatment of the cells with the caspase-3 inhibitor markedly reduced the apoptosis induced by the PKCδ-Cyto and PKCδ-Mito and only moderately decreased the apoptotic effect of PKCδ-Nuc (Fig. 3D).
PKCδ has been reported to activate the mitochondrial pathway, which eventually leads to the activation of caspase-3 (14, 29). To examine the role of PKCδ localization in the activation of the mitochondrial pathway, we examined its effect on cytochrome c release and on caspase-9 activation. As presented in Fig. 3E, PKCδ-Mito significantly increased the release of cytochrome c from the mitochondria to the cytosol, whereas a smaller effect was observed with the PKCδ-Cyto and no significant effects were observed with the PKCδ-Nuc and PKCδ-ER. Similarly, the PKCδ-Mito significantly increased the activation of caspase-9, whereas smaller increase was obtained in the PKCδ-Cyto–overexpressing cells (Fig. 3F).
Targeting of PKCδ to the ER Protects Glioma and HeLa Cells against Cell Apoptosis
Recent studies in our laboratory showed that PKCδ translocated to the ER in response to TRAIL treatment (24) and Sindbis virus neurovirulent infection (25) and that in both cases PKCδ acted as a survival kinase. As we have found that PKCδ-ER does not induce apoptosis in either the A172 or HeLa cells, we considered the possibility that expression of PKCδ in the ER might actually protect glioma cells against cell apoptosis. We first examined whether the PKCδ-ER protected the cells against TRAIL-induced apoptosis. A172 cells were transfected with control vector, PKCδ-Nuc, and PKCδ-ER, and after 24 h, the cells were treated with TRAIL (100 ng/mL) for an additional 5 h. As presented in Fig. 4A, TRAIL induced ∼45% cell apoptosis in the control vector A172 cells and increased the apoptotic effect of the PKCδ-Nuc (data not shown). In contrast, cells overexpressing the PKCδ-ER exhibited lower levels of cell apoptosis and only 20% of the cells underwent cell apoptosis (Fig. 4A). Similar results were obtained with HeLa cells; in these cells, overexpression of PKCδ-ER reduced cell apoptosis by 68% compared with control vector cells (Fig. 4B). TRAIL did not induce the cleavage of PKCδ-ER in either A172 or HeLa cells (data not shown).
We also examined the antiapoptotic effect of the PKCδ-ER in etoposide-treated cells. Etoposide induced a 22.5% and 45.6% cell apoptosis in A172 and HeLa cells, respectively. Overexpression of the PKCδ-ER in these cells significantly reduced the apoptotic effect of etoposide. Thus, the PKCδ-ER almost completely abolished the apoptotic effect of etoposide in A172 cells and reduced HeLa cell apoptosis by 53% (Fig. 4C). In addition to its protective effects against the apoptotic effects of TRAIL and etoposide, the PKCδ-ER partially protected the A172 cells against thapsigargin- and tunicamycin-induced cell death (Fig. 4D). In these experiments, cells were treated with thapsigargin (2 μg/mL) and tunicamycin (4 μmol/L) for 36 h and analyzed for cell apoptosis using propidium iodide staining and FACS analysis.
To further study the role of PKCδ in the ER, we used a PKCδ kinase-dead mutant targeted to the ER to interfere with endogenous PKCδ that localizes in this subcellular site. For these experiments, we used the apoptotic stimulus TRAIL, which induces translocation of PKCδ to the ER (24). As a control in this experiment, we used the PKCδKD-Nuc. Figure 4E and F shows the overexpression of the PKCδKD-ER and PKCδKD-Nuc using Western blot analysis (Fig. 4E) and their localization in the ER and nucleus, respectively, using anti-Myc antibody staining and confocal microscopy (Fig. 4F). Figure 4G shows that the PKCδKD-ER slightly increased the apoptosis of the A172 cells but significantly enhanced the apoptotic effect of TRAIL. In contrast, overexpression of a PKCδKD mutant that was targeted to the nucleus did not alter the apoptotic response of the cells to TRAIL. These results suggest that the localization of PKCδ in the ER exerts an antiapoptotic effect against TRAIL-induced apoptosis and that inhibition of the endogenous PKCδ in the ER increases the apoptotic effect of TRAIL.
PKCδ-Cyto and PKCδ-Nuc Induce the Activation of Different Signaling Pathways
The proapoptotic and antiapoptotic effects of PKCδ are mediated by activating various downstream signaling pathways (3). To identify signaling pathways that are induced by the compartmentalized PKCδ, we examined the expression and phosphorylation of various apoptosis-related proteins in cells overexpressing the various PKCδ pShooter vectors.
As presented in Fig. 5, the overexpression of the PKCδ-Cyto selectively increased the phosphorylation of p38 and decreased the phosphorylation of AKT and the expression of X-linked inhibitor of apoptosis protein (XIAP), whereas the expression of PKCδ-Nuc induced the phosphorylation of JNK. In contrast, we did not detect changes in the phosphorylation of signal transducers and activators of transcription 1 or in the expression of Bax and Bcl2 in cells overexpressing the different PKCδ pShooter vectors.
To further study the role of p38 in the apoptosis induced by the PKCδ-Cyto, we first examined the phosphorylation of the upstream effector of p38, MKK3. We found that overexpression of PKCδ-Cyto significantly increased the phosphorylation of MKK3/MKK6 compared with the control vector cells (Fig. 6A). To examine the role of p38 in the apoptotic effect of PKCδ-Cyto, we first used the p38 inhibitor SB203580. We found that SB203580 (10 μmol/L) almost completely inhibited the phosphorylation of p38 induced by the PKCδ-Cyto (Fig. 6B), whereas it only moderately reduced the apoptosis of these cells (Fig. 6C). Similar results were obtained with silencing of p38 using specific small interfering RNA (siRNA) duplexes. Thus, transfection of the A172 cells with p38 siRNA duplexes significantly reduced p38 expression (Fig. 6D), whereas it only moderately abrogated the apoptotic effect of PKCδ-Cyto (Fig. 6E), suggesting that the phosphorylation of p38 only partially contributed to the apoptosis induced by the PKCδ-Cyto. In contrast, overexpression of XIAP (Fig. 6F), which was significantly decreased in the PKCδ-Cyto–overexpressing cells, significantly abolished the apoptotic effect of the PKCδ-Cyto (Fig. 6G).
PKCδ-Nuc did not affect the phosphorylation of p38 but rather increased the activation of JNK. To delineate the mechanisms by which JNK is activated in the PKCδ-Nuc–overexpressing cells, we examined the phosphorylation of MKK4 and MKK7 in these cells. As presented in Fig. 7A, overexpression of PKCδ-Nuc increased the phosphorylation of MKK7, whereas no increase was observed in the phosphorylation of MKK4. To examine the role of JNK in the apoptotic effect of PKCδ-Nuc, we used the JNK inhibitor SP600125. For these experiments, the A172 cells were transfected with the control vector and PKCδ-Nuc and treated with the pharmacologic inhibitor 3 h after transfection. Treatment of the PKCδ-Nuc–overexpressing cells with the JNK inhibitor SP600125 (20 μmol/L) significantly abolished the increased phosphorylation of JNK that was observed in these cells (Fig. 7B). In parallel, SP600125 significantly reduced the apoptotic effect of PKCδ-Nuc as was observed by the morphology of the cells (Fig. 7C) and by propidium iodide staining and FACS analysis (Fig. 7D). Thus, cell apoptosis was decreased from 57.8% in the absence of the JNK inhibitor to 18.3% in its presence. Similar results were obtained with JNK inhibitor III (data not shown).
The role of JNK in the apoptosis induced by the PKCδ-Nuc was specific because the SP600125 did not inhibit the apoptosis induced by the PKCδ-Cyto. Similarly, the p38 inhibitor SB203580 did not alter the apoptosis induced by the PKCδ-Nuc (Fig. 7E).
PKCδ has been reported to play a major role in the regulation of cell apoptosis and survival in response to various stimuli (1-3). Although most studies report a proapoptotic function of PKCδ, there are other studies that show an antiapoptotic effect of this isoform (24, 25). One of the main factors that affect PKC activity and functions and contributes to the diverse effect of PKC is its subcellular localization (34). Indeed, PKCδ undergoes translocation to distinct subcellular sites in response to various apoptotic stimuli (5, 9, 27-29); however, the role of the subcellular localization of PKCδ in its apoptotic functions is just beginning to be understood. In this study, we targeted PKCδ to the cytoplasm, ER, nucleus, and mitochondria and examined the role of the localization of PKCδ in its proapoptotic and antiapoptotic effects.
The first objective of this study was to find a cellular system that will allow the selective targeting of PKCδ to the mitochondria, ER, and nucleus in a way that will resemble as much as possible what happens with the endogenous PKCδ. The pShooter vectors seemed to be the best choice for targeting PKCδ because these vectors have been widely and successfully used to study the role of the subcellular localization of different proteins in their cellular functions (31-33). Thus, this system has been used to show the role of the mitochondrial localization of mutant superoxide dismutase 1 in cell death in a model of familial amyotrophic lateral sclerosis (32), to study the role of the intracellular localization of transglutaminase in cell death (31), and to examine the functional coupling of chromogranin with the InsP3R for calcium signaling (33). Importantly, targeting of PKCδ to the mitochondria using the appropriate pShooter vector induced phospholipid scramblase-dependent cell apoptosis similar to the endogenous PKCδ (21).
We found that the expression of PKCδ-Cyto, PKCδ-Mito, and PKCδ-Nuc induced cell apoptosis, whereas expression of PKCδ-ER had no apoptotic effect. The role of the nuclear and mitochondrial localization of PKCδ in its apoptotic function has been shown thus far indirectly by studies that reported an association between the translocation of PKCδ to the nucleus and mitochondria and its proapoptotic effects. Indeed, the role of the nuclear localization of PKCδ in its apoptotic effect has been shown in studies showing that PKCδ translocated to the nucleus in response to apoptotic stimuli, such as etoposide (9, 30), and that the overexpression of a PKCδ nuclear localization signal (NLS) mutant abrogates the apoptotic effect of this drug (35). Similarly, PKCδ has been shown to translocate to the mitochondria in response to UV radiation (14), phorbol 12-myristate 13-acetate (28), and oxidative stress (29) and to induce cell apoptosis in response to these stimuli. Our results support these studies and provide direct evidence that the localization of PKCδ-CA in the cytosol, mitochondria, and nucleus, but not in the ER, promotes cell apoptosis.
PKCδ undergoes cleavage in response to various apoptotic stimuli, such as etoposide (9, 30), cisplatin (13), UV radiation (14), and TRAIL (24). Although the cleavage of PKCδ and its catalytic fragment are mainly associated with its proapoptotic effects (9, 13, 14, 30), in some systems PKCδ cleavage does not play a role in its apoptotic effects (36, 37) or it is associated with PKCδ protective effects (24). We found that PKCδ-Cyto and PKCδ-Mito were cleaved to generate a 40-kDa catalytic fragment. In contrast, the PKCδ-Nuc and PKCδ-ER did not undergo significant cleavage, although the PKCδ-Nuc induced a large degree of cell apoptosis. Although most of the studies implicate the catalytic domain of the nuclear PKCδ as an important mediator of its apoptotic activity (9, 19, 20, 35), our results show that the apoptotic effect of PKCδ-Nuc was not dependent on its cleavage and catalytic fragment and that the localization of PKCδ in the nucleus was sufficient to induce cell apoptosis.
We found that PKCδ-Cyto and PKCδ-Mito induced activation of caspase-3, whereas only minor activation of caspase-3 was detected in the PKCδ-Nuc overexpressors. The cleavage of PKCδ is mainly mediated by caspase-3 (6, 9, 30), which in turn is phosphorylated and activated by PKCδ (9, 30, 38). Thus, the large increase in caspase-3 activation induced by PKCδ-Cyto and PKCδ-Mito is in accordance with their cleavage. Similarly, inhibition of caspase-3 significantly inhibited the apoptosis induced by PKCδ-Cyto and PKCδ-Mito, whereas it only slightly abrogated the apoptosis induced by the PKCδ-Nuc. These results suggest that PKCδ-Nuc induced cell apoptosis mostly in a caspase-3–independent manner. One of the reasons for the lack of the activation of caspase-3 and PKCδ cleavage in the PKCδ-Nuc–overexpressing cells may be the different subcellular localization of PKCδ relative to that of caspase-3. In a recent study, we reported that etoposide induced colocalization of caspase-3 and PKCδ in the nucleus and that caspase-3 induced cleavage of PKCδ in this site (9). However, the localization of caspase-3 in the nucleus was dependent on the activation of caspase-9 by a tyrosine phosphorylated PKCδ. Indeed, PKCδ-Nuc did not induce cytochrome c release from the mitochondria or the activation of caspase-9.
PKCδ-ER did not induce cell apoptosis in either A172 or HeLa cells but rather protected the cells from different apoptotic stimuli. Interestingly, we recently reported that translocation of PKCδ to the ER was associated with antiapoptotic effects of this isoform. Thus, PKCδ translocated to the ER in response to Sindbis virus infection (25) and in response to TRAIL treatment (24), and in both cases, PKCδ protected the cells from the apoptosis induced by these stimuli. These results are further supported by our current findings that targeting PKCδ to the ER reduced cell apoptosis induced by TRAIL and etoposide in both A172 and HeLa cells. In addition to showing the protective effect of the overexpressed PKCδ-ER, we also showed that the localization of the endogenous PKCδ in the ER has similar protective effect. Thus, expression of a PKCδKD mutant that was targeted to the ER increased the apoptosis induced by TRAIL by abrogating the protective effect of the ER-localized endogenous PKCδ. Collectively, these results indicate that PKCδ that resides in the ER, either ectopically or endogenously, similarly exerts antiapoptotic effects.
In addition to its protective effects against cell apoptosis, the PKCδ-ER partially inhibited cell death induced by thapsigargin and tunicamycin, suggesting that localization of PKCδ in the ER had a protective effect also in the ER stress-induced apoptosis. The role of PKCδ in the ER and the mechanisms involved in its antiapoptotic effects are currently not understood. Localization in the ER has been described for other PKC isoforms, such as PKCα (39) and PKCη (40), but the functions of these PKC isoforms in the ER have not been reported. Importantly, there are several apoptosis-related proteins, such as caspase-12, Bcl2, Bax, Bik, and Bak (41, 42), which reside in the ER and play a role in the regulation of cell apoptosis. One possible PKCδ substrate in the ER is Bcl2, which has been shown to undergo phosphorylation by PKC (43) and to regulate the cross-talk between the ER and the mitochondria during cell apoptosis (44). Interestingly, a recent study reported that targeting of Bcl2 to the ER plays an important role in its protective effect (41). Studies exploring the role of Bcl2 in the antiapoptotic effects of PKCδ-ER and the identification of novel PKCδ substrates/partners in the ER are currently under way.
PKCδ has been shown to activate a large number of downstream signaling pathways in response to various apoptotic stimuli (3). Thus, the activation of signaling proteins, such as c-Abl (45), JNK (16), p38 (17), p73 (19), and DNA-PK (20), has been associated with the proapoptotic effect of PKCδ, whereas phosphorylation of AKT (24) and HSP25 (46) is associated with its antiapoptotic effects. We found that the localization of PKCδ affected its ability to activate specific downstream signaling pathways. Thus, the PKCδ-Cyto induced phosphorylation of p38 and dephosphorylation of AKT and decreased the expression of XIAP, whereas the PKCδ-Nuc induced phosphorylation of JNK.
PKCδ has been recently reported to increase the phosphorylation of p38 in various cell types (17, 47). However, the role of the subcellular localization of PKCδ in this effect was not reported. We found that the PKCδ-Cyto induced the phosphorylation of MKK3/MKK6, which lies upstream of p38 (48). Although PKCδ-Cyto activated the p38 pathway, it played only a partial role in the apoptosis induced by the PKCδ-Cyto because the p38 inhibitor SB203580 and silencing of p38 only moderately inhibited cell apoptosis, suggesting that additional signaling pathways mediate the apoptotic effect of the PKCδ-Cyto.
In addition to the phosphorylation of the proapoptotic protein p38, the PKCδ-Cyto induced dephosphorylation of AKT, which regulates cell survival in a variety of cellular systems (49). The survival effects of AKT are exerted by phosphorylating proteins, such as BAD, caspase-9, the forkhead transcription factors (49), or XIAP (50). Indeed, XIAP has been recently identified as a downstream target of AKT and as an important regulator of AKT survival effects. AKT phosphorylates XIAP on Ser87 and this phosphorylation stabilizes XIAP and prevents its ubiquitination and degradation (50). Our results indicate that the PKCδ-Cyto reduced the expression of XIAP and that overexpression of XIAP significantly abrogated the apoptotic effect of PKCδ-Cyto. Thus, the decrease in XIAP expression contributed to the apoptotic effect of PKCδ-Cyto in addition to the moderate effect of p38.
In contrast to the effect of the PKCδ-Cyto on the AKT and p38 signaling pathways, overexpression of PKCδ-Nuc selectively increased the phosphorylation of JNK, whereas it did not affect the phosphorylation of AKT, p38, and signal transducers and activators of transcription 1 or the expression of Bax, Bcl2, or XIAP. To further characterize the effect of PKCδ-Nuc on JNK activation, we examined the phosphorylation of two JNK kinases: MKK4 and MKK7. We found that the PKCδ-Nuc induced the phosphorylation of MKK7, whereas it did not affect the activation of MKK4. PKCδ has been previously shown to activate MKK7 (51); however, this is the first report showing the role of nuclear PKCδ in this effect. The JNK pathway played a major role in the apoptotic effect of PKCδ-Nuc because inhibition of JNK significantly decreased the apoptotic effect of the PKCδ-Nuc. JNK has been implicated in the regulation of cell apoptosis in response to various stimuli (51, 52) and has been shown to act downstream of PKCδ (16, 51, 53, 54). A role for the nuclear PKCδ in the activation of JNK is further supported by recent studies that showed that the nuclear translocation of PKCδ by etoposide preceded the activation of JNK in salivary gland acinar cells (30, 53). Thus, JNK may represent another nuclear PKCδ substrate in addition to the already well-characterized substrates, p73 (19) and DNA-PK (20).
Interestingly, PKCδ-Mito did not induce any significant changes in the expression or phosphorylation of the different signaling proteins that were examined in this study, whereas it induced the release of cytochrome c from the mitochondria and the activation of caspase-9. Thus, targeting of PKCδ to the mitochondria seems to activate the mitochondrial pathway probably via phosphorylation of mitochondrial-associated proteins, such as scramblase 3, which has been recently identified as a substrate of PKCδ in the mitochondria (21).
Various studies have shown the importance of the subcellular localization of signaling proteins in their biological functions. Indeed, the targeting of Bcl2 to the mitochondria induces cell apoptosis, whereas its targeting to the ER promotes its protective function (55). In addition, the localization of extracellular signal-regulated kinase 2 determines its protective effects against different apoptotic stimuli (56). With regard to PKC, numerous studies showed that the subcellular localization of PKC has major roles in the activity, functions, and diverse effects of the different PKC isoform effects (34, 57, 58). Less is known, however, about the role of the subcellular localization of a specific isoform in its diverse functions in a given cellular system. Our results show that the subcellular localization of PKCδ has an important role in its proapoptotic and antiapoptotic functions; proapoptotic effects when localized in the cytosol, mitochondria, and nucleus and antiapoptotic effect in the ER. Moreover, the localization of PKCδ differentially activates different downstream signaling pathways. Thus, the results of this study have important implications for our understanding of the role of PKCδ localization in its different effects and may provide the basis for the identification of novel PKC substrates/partners in the different subcellular sites and for the development of inhibitors that can selectively block distinct functions of PKCδ in a specific cellular system.
Materials and Methods
An affinity-purified polyclonal anti-PKCδ antibody (C-17) was purchased from Santa Cruz Biotechnology, and anti-Myc antibody was from Upstate USA, Inc. Anti-calnexin was from Stressgen, and the nuclear marker 7-AAD and the mitochondria marker MitoTracker Orange were from Molecular Probes, Invitrogen. Human TRAIL was from PeproTech, and anti–phosphorylated AKT, AKT, p38, phosphorylated p38, phosphorylated MKK3/MKK6, MKK3, phosphorylated JNK, JNK, phosphorylated MKK7, MKK7, phosphorylated MKK4, MKK4, signal transducers and activators of transcription 1, phosphorylated signal transducers and activators of transcription 1, XIAP, active caspase-3, Bax, and Bcl2 antibodies were obtained from Cell Signaling Technology. The caspase-3 inhibitor DEVD-FMK, the p38 mitogen-activated protein kinase inhibitor SB202190, and the JNK inhibitors SP600125 and JNK inhibitor III (HIV-TAT47-57-gaba-c-Jund33-57) were from Calbiochem. Etoposide, leupeptin, aprotinin, phenylmethylsulfonyl fluoride, and sodium vanadate were obtained from Sigma Chemical Co.
The glioma cell lines A172 and HeLa were obtained from the American Type Culture Collection. Cells were grown on tissue culture dishes in medium consisting of DMEM containing 10% heat-inactivated FCS, 2 mmol/L glutamine, 50 units/mL penicillin, and 0.05 mg/mL streptomycin. Medium was changed every 3 to 4 days and cultures were passaged using 0.25% trypsin.
The constitutive active PKCδ mutant contains double point mutations of alanine to arginine at positions 144 and 145 in the inhibitory pseudosubstrate sequences within the regulatory domain as described (59). The pShooter vectors pCMV/myc/cyto, pCMV/myc/ER, pCMV/myc/mito, and pCMV/myc/nuc were purchased from Invitrogen. The inserts, rat PKCδ-CA (constitutively active, DR144/145A) or PKCδKD (kinase-dead, K376A; ref. 9), were digested from the PKCδ-CA-EGFP-N1 and PKCδKD plasmids using the XhoI and MluI sites and subcloned into the XhoI/NotI site of the pShooter vectors. The non–organelle-targeted vector was designated PKCδ-Cyto, the mitochondria-targeted vector with the mitochondria signal was designated PKCδ-Mito, the ER-targeted vectors that contain the ER retention signal were designated PKCδ-ER and PKCδKD-ER, and the nucleus-targeted vectors that contain the nuclear localization signal were designated PKCδ-Nuc and PKCδKD-Nuc. Confirmation of proper ligation was done by DNA sequencing. The pCDNA3-XIAP-Myc plasmid (Addgene plasmid 11833) was kindly provided by Dr. Guy Salvesen (University of California, San Diego, CA).
A172 cells were transfected either with the control vector or with the different PKCδ pShooter vectors by electroporation using the Nucleofector device, protocol number U29 (Amaxa Biosystems). Transfection efficiency using nucleofection was between 70% and 90%.
The attachment of the cells was not significantly altered by the electroporation, and the attachment of the cells overexpressing the different pShooter vectors was similar to that of the control vector.
HeLa cells were transfected using LipofectAMINE (Invitrogen) according to the manufacturer's instructions.
Control (scrambled sequence) and p38 siRNA duplexes were synthesized and purified by Dharmacon. Transfection of siRNAs was done using Oligofectamine (Invitrogen) according to the manufacturer's instructions. Experiments were done 72 h after transfection.
Preparation of Cell Homogenates and Immunoblot Analysis
Cell pellets (106 cells/mL) were resuspended in 100 μL of lysis buffer [25 mmol/L Tris-HCl (pH 7.4), 50 mmol/L NaCl, 0.5% sodium deoxycholate, 2% NP40, 0.2% SDS, 1 mmol/L phenylmethylsulfonyl fluoride, 50 μg/mL aprotinin, 50 μmol/L leupeptin, 0.5 mmol/L Na3VO4] on ice for 15 min. Sample buffer (2×) was added and the samples were boiled for 5 min. Lysates (30 μg protein) were resolved by SDS-PAGE and transferred to nitrocellulose membranes. The membranes were blocked with 5% dry milk in PBS and subsequently probed with the primary antibody. Specific reactive bands were detected using a goat anti-rabbit or goat anti-mouse IgG conjugated to horseradish peroxidase (Bio-Rad), and the immunoreactive bands were visualized by the enhanced chemiluminescence Western blotting detection kit (Amersham).
Transfected cells were washed with PBS and fixed in 4% paraformaldehyde for 15 min. Subsequently, cells were washed in PBS and, after blocking with staining buffer (2% bovine serum albumin and 0.1% Triton X-100 in PBS) for 30 min at room temperature, incubated with an anti-Myc antibody. Following washes in PBS, cells were incubated with an anti-rabbit Alexa Fluor 488 antibody for additional 60 min and mounted in FluoroGuard antifade reagent. Cells were viewed and photographed using confocal microscopy with ×63 magnification at an excitation wavelength of 488 nm. For the visualization of the ER, cells were incubated with mouse anti-KDEL antibody followed by incubation with anti-mouse Alexa Fluor 546 antibody. Visualization of the nucleus and the mitochondria was done using labeling of the cells with the nuclear marker 7-AAD (10 mg/mL) and with the mitochondria marker MitoTracker Orange (50 nmol/L), respectively.
Cells transfected with the GFP pShooter vectors were fixed in 4% paraformaldehyde for 20 min and mounted in FluoroGuard antifade reagent.
Measurements of Cell Apoptosis
Cell apoptosis was measured using propidium iodide staining and analysis by flow cytometry as described previously (9, 24). Briefly, transfected cells (1 × 105/mL) were plated in six-well plates and treated with the indicated treatments for 24 h. Detached cells and trypsinized adherent cells were pooled, fixed in 70% ethanol for 1 h on ice, washed with PBS, and treated for 15 min with RNase (50 μmol/L) at room temperature. Cells were then stained with propidium iodide (5 μg/mL) and analyzed on a Becton Dickinson cell sorter.
Cell apoptosis was also measured using anti–histone ELISA (Cell Death Detection ELISA kit, Roche Applied Science). For these experiments, extracts of cells containing histone-associated DNA fragments were incubated in 96-well plates coated with anti-histone antibodies for 2 h. Plates were then washed and incubated with anti-DNA antibodies conjugated to peroxidase for an additional 2 h. Substrate solution was added and absorbance was measured at a wavelength of 405 nm.
Caspase-3 and Caspase-9 Activity Assays
Cells were lysed with caspase assay lysis buffer containing 0.5% NP40 and 5 mmol/L EDTA in 50 mmol/L Tris (pH 7.5). Following 20-min incubation on ice, the cells were centrifuged at 20,000 × g at 4°C. The supernatants were placed in a 96-well plate containing reaction mixture with DEVD-AMC or Ac-LEHD-AFC, fluorimetric substrates of caspase-3 and caspase-9, respectively. Following 60 min of incubation at 37°C, fluorescence was measured at a wavelength of 380 nm.
Cytochrome c Release
Cytochrome c release from the mitochondria was determined in the cytosolic fraction. Mitochondrial and cytosolic fractions were isolated using the ApoAlert Cell Fractionation kit (Clontech, BD Biosciences) as described previously (24). Briefly, cells were centrifuged at 600 × g for 5 min at 4°C. Cell pellets were resuspended with 0.8 mL of ice-cold fractionation buffer and incubated on ice for 10 min. Cells were then homogenized with an ice-cold Dounce homogenizer and centrifuged at 700 × g for 10 min. The supernatants were then centrifuged at 10,000 × g for 25 min at 4°C and the supernatants (cytosolic fraction) and pellets (mitochondrial fraction) were collected. Cytochrome c was identified in the cytosolic fraction by using a rabbit anti–cytochrome c antibody.
The results are presented as the mean values ± SE. Data were analyzed using ANOVA and a paired Student's t test to determine the level of significance between the different groups.
Grant support: NIH grant RO1CA109196 and William and Karen Davidson Fund, Hermelin Brain Tumor Center.
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Note: R. Gomel and C. Xiang contributed equally to this work.