Functional Dichotomy of Protein Kinase C (PKC) in Tumor Necrosis Factor-α (TNF-α) Signal Transduction in L929 Cells

Tumor necrosis factor-α (TNF-α) is capable of inducing a variety of biologic responses through multiple signaling pathways. Because of the potential role of protein kinase C (PKC) in apoptosis, we examined the effects and mechanisms of TNF-α on PKC regulation, specifically on PKCα. In L929 murine fibroblasts, TNF-α (0.5– 5 nm) caused potent inhibition of PKCα activity and induced translocation of PKCα from the cytosol to the membrane. Treatment of cells with TNF-α also induced dephosphorylation of PKCα as detected by a mobility shift on SDS-polyacrylamide gel and inhibition of PKC phosphorylation as probed by anti-phospho-PKC antibodies. Since PKC is activated directly by diacylglycerol and inactivated indirectly by ceramide, we next examined the roles of these lipid mediators in the regulation of PKCα. Addition of TNF-α led to accumulation of both ceramide and diacylglycerol. Fumonisin B1, an inhibitor of ceramide synthase, and glutathione, an inhibitor of neutral sphingomyelinase, both reversed the effect of TNF-α on PKCα activity, suggesting that ceramide production is necessary for the action of TNF-α. The diacylglycerol mimic phorbol 12-myristate 13-acetate was sufficient to cause translocation of PKCα, but not the mobility shift. Okadaic acid at 2 nm, a potent protein phosphatase inhibitor, blocked the effects of TNF-α on PKCα activity, but not on PKCα translocation, thus demonstrating that dephosphorylation and translocation are independent processes. These results demonstrate that PKCα acts as a downstream target for TNF-α and that different lipid-mediated pathways in TNF-α signaling lead to opposing signals in the regulation of PKCα activity.

Tumor necrosis factor-␣ (TNF-␣), 1 a pleiotropic cytokine, mediates multiple biologic responses in different cell types through binding to either 55-or 75-kDa membrane receptors (1,2). The biologic properties of TNF-␣ include inhibition of cell growth, induction of differentiation and apoptosis, modulation of gene transcription, and activation of protein phosphorylation cascades (3). In many cell systems, TNF-␣ receptor-mediated cellular responses are preceded by the elevation of intracellular ceramide levels through hydrolysis of sphingomyelin by sphingomyelinase (4,5). Although the elevation of intracellular ceramide levels has been proposed to mediate at least some of the effects of TNF-␣ on cell growth and differentiation (6), the signaling mechanisms involved in transduction of biologic effects of TNF-␣ are not fully understood.
Ceramide-activated protein phosphatase (CAPP) is one of the cellular targets of ceramide (7,8). CAPP belongs to the PP2A heterotrimeric subfamily of the serine/threonine protein phosphatases and is potently inhibited by okadaic acid, with an IC 50 of 1-10 nM (7,8). Activation of CAPP plays an important role in the induction of c-Myc down-regulation by ceramide in myeloid leukemia cells (9). The antiproliferative effect of ceramide and CAPP activity have also been detected in Saccharomyces cerevisiae (10), and a role of yeast CAPP in mediating growth suppressor effects of ceramide has been documented (11). We previously demonstrated that inhibition of PKC␣ activity by ceramide in Molt-4 leukemia cells was also through activation of a protein phosphatase of the PP2A subfamily, consistent with a role for CAPP (12). Other putative targets of ceramide include PKC (13,14) and ceramide-activated protein kinase, which is a kinase suppressor of Ras (5,15) and which may function to transduce the effects of ceramide on Ras (15).
PKC is a family of serine/threonine kinases that plays an important role in modulating a variety of biologic responses ranging from regulation of cell growth to cell death. The involvement of PKC in signal transduction of TNF-␣ has been reported (16). It was shown that TNF-␣ activates PKC and induces PKC translocation in Jurkat, K562, and U937 cell lines. However, the specific subtypes of PKC activated by TNF-␣ and the underlying mechanisms for their role in TNF-␣ signaling remain to be elucidated (16). A recent study has also demonstrated that an atypical PKC isozyme, PKC, may act as a molecular switch in transducing mitogenic and growth inhibitory signals of TNF-␣ (14). More recently, studies have pointed to specific effects of the PKC␣ isozyme in regulating apoptosis. One study has shown that inactivation of PKC␣ is sufficient to induce apoptosis (17), whereas another study showed that activation of PKC␣ protects from apoptosis (18). Therefore, given the emerging role of PKC␣ in apoptosis, the role of ceramide in TNF-␣ signaling, and the effects of ceramide on PKC␣, we investigated the relationship of these regulators and the role of ceramide in this pathway.
In this study, we show that TNF-␣ leads to inhibition of PKC␣ activity and induces translocation of PKC␣ in L929 cells. Inhibition of PKC␣ activity by TNF-␣ is dependent on the generation of ceramide. We also provide evidence that TNF-␣ causes dephosphorylation of PKC␣ through activation of a protein phosphatase consistent with CAPP.

Materials
L929 murine fibroblasts have been described previously (19,20). Dulbecco's modified Eagle's medium was purchased from Life Technologies, Inc. Fetal bovine serum was supplied from Summit Biotechnology Inc. (Fort Collins, CO). TNF-␣ was a gift from Dr. Phil Pekiala (East Carolina University). D-erythro-C 6 -ceramide was a kind gift from Dr. Alicja Bielawska (Duke University). PKC isozyme-specific antibodies have been prepared and characterized previously (21

Methods
Cell Culture-Cells were maintained in Dulbecco's modified Eagle's medium supplemented with 5% fetal bovine serum and 60 mg/liter kanamycin sulfate in 5% CO 2 at 37°C in a humidified incubator. For experiments, cells were normally seeded at 2 ϫ 10 5 cells/plate and allowed to grow to 50 -70% confluence before treatment.
Ceramide Measurements-Cells were harvested and lipids were extracted as described (22) and then assayed for ceramide by the diacylglycerol kinase method (23), normalized to total cellular phospholipid (24), and represented as picomoles of ceramide/nmol of lipid phosphorus.
Western Blotting-Following the indicated times of different treatments, L929 cells were harvested on ice. Proteins were fractionated on 7.5% gels by SDS-PAGE as described by Laemmli (25). Following SDS-PAGE, proteins were electrophoretically transferred to a nitrocellulose membrane (26). The nitrocellulose membrane was blocked in 5% dried milk in 1ϫ phosphate-buffered saline containing 0.1% Tween 20, incubated with anti-retinoblastoma antibody, and then developed using the ECL detection system. Immunoprecipitation-Following different treatments, cells were lysed on ice in lysis buffer (150 mM NaCl, 50 mM Tris-HCl, pH 7.4, 0.5% Nonidet P-40, 5 mM NaF, 10 g/ml leupeptin, 10 g/ml aprotinin, 10 g/ml trypsin-chymotrypsin inhibitor, 5 g/ml pepstatin A, 1 mM dithiothreitol, 1 mM Na 3 VO 4 , and 1 mM phenylmethylsulfonyl fluoride). The lysates were incubated with the appropriate antibody at 4°C, followed by protein A-Sepharose adsorption. The beads were washed and used for in vitro kinase assay (see below) or for Western blot analysis.
In Vitro Kinase Assay-The protein A-Sepharose beads were resuspended in reaction buffer (20 mM Tris-HCl, pH 7.4, 10 mM MgCl 2 , 10 mM unlabeled ATP, 0.2 mg/ml histone H1, and 2.5 Ci of [␥-32 P]ATP). The reactions were carried out at 30°C for 10 min and terminated by the addition of an equal volume of 2ϫ sample buffer. Proteins were separated by SDS-PAGE and visualized by autoradiography.
PKC Translocation-L929 cells were treated with 0.1 M PMA for 20 min at 37°C and harvested on ice. The cells were washed once with cold phosphate-buffered saline and resuspended in homogenization buffer (20 mM Tris-HCl, pH 7.5, 10 mM EGTA, pH 7.4, 2 mM EDTA, pH 7.4, 250 mM sucrose, 1 mM phenylmethylsulfonyl fluoride, and 0.2% (w/v) leupeptin). The cells were then sonicated twice for 10 s each and centrifuged in a TL-100.3 rotor at 40,000 rpm for 40 min at 4°C. The supernatant (cytosol) was removed and solubilized with an equal volume of 2ϫ sample buffer. The membrane pellet was resuspended in an equal volume of homogenization buffer and mixed 1:1 with 2ϫ sample buffer. All fractions were boiled for 5 min and analyzed by Western blot analysis.
Data Analysis-Data obtained from Western blotting or in vitro kinase assay were analyzed by Alphaimager (Alpha Innotech Corp., San Leandro, CA), normalized to the total amounts of protein, and represented as levels of kinase activity or levels of protein.

Inhibition of PKC␣ Activity by Ceramide-
We had previously demonstrated that ceramide inhibits PKC␣ activity in Molt-4 cells (12). To investigate a possible role for ceramide in L929 cells, we determined the ability of exogenous ceramide to inhibit PKC␣ activity. L929 cells were treated with various concentrations of a synthetic and cell-permeable analog (C 6 -ceramide) for 14 h. PKC activity was determined using an immunoprecipitation/kinase assay as described previously (12). The lysates were immunoprecipitated with anti-PKC␣ antibody, and the immunocomplexes were collected. The calcium-and lipid-dependent PKC␣ activity was then determined in the absence and presence of calcium and lipid activators using histone H1 as an exogenous substrate. Fig. 1 shows that immunocomplexes contained a basal level of histone H1 kinase activity as defined by the presence of 10 mM EGTA. The histone H1 kinase activity was significantly enhanced in the presence of the lipid activators phosphatidylserine/diC 18:1 . The maximal effects of ceramide on PKC␣ were detected at a concentration of 40 M (Fig. 1A, upper panel). The structurally similar lipid molecule dihydro-C 6 -ceramide did not have an appreciable effect on PKC␣ activity (Fig. 1A, upper panel). Densitometric analysis of kinase activity is shown in Fig. 1A (lower panel). In addition, the alteration of PKC␣ activity by ceramide was not due to the change in PKC␣ protein levels. Immunoprecipitation/Western blot analysis (Fig. 1b, upper panel) and densitometric analysis (lower panel) demonstrated that the amounts of protein in control and treated cells were similar. The intensity of the Coomassie Brilliant Blue staining indicated that the amounts of protein loaded in each lane were approximately equal (data not shown). These results are consistent with our previous finding that ceramide specifically inhibits PKC␣ activity in Molt-4 cells (12). To determine whether the effect of ceramide on PKC is isozyme-specific, we examined PKC␤II, another major calcium-dependent isozyme present in significant amounts in L929 cells. It was found that PKC␤II was also inhibited by C 6 -ceramide (data not shown). For the purpose of this study, we focused on PKC␣ activity.
Inhibition of PKC␣ Activity by TNF-␣-Since ceramide mediates some of the TNF-␣ effects on cellular growth regulation, we wondered whether TNF-␣ can affect PKC␣ activity. L929 cells were treated with various concentrations of TNF-␣ for 12-14 h. Treatment of cells with TNF-␣ (0.5-5 nM) led to a remarkable decrease in both basal and lipid-activated PKC␣ H1 kinase activities as compared with the control (Fig. 2A). This change in PKC activity was not due to the alteration of protein levels as revealed by immunoprecipitation/Western blot analysis (Fig. 2B). Therefore, TNF-␣ exerts an inhibitory effect on PKC␣, and the diminution of PKC␣ activity is not due to alteration of protein levels.
Induction of Translocation of PKC␣ from Cytosol to Membrane by TNF-␣-To examine whether the inhibitory effect of TNF-␣ on PKC␣ is dependent on enzyme translocation, L929 cells were treated with TNF-␣, followed by PMA stimulation. The membrane and cytosolic fractions were prepared and analyzed after 30 min of stimulation with PMA (0.1 M) by Western blot analysis. The addition of PMA led to PKC␣ translocation in control and TNF-␣-treated cells (Fig. 3A). TNF-␣, however, induced PKC␣ translocation on its own in these cells (Fig. 3A). Interestingly, ceramide did not affect PKC␣ translocation on its own or in response to PMA (Fig. 3B), which is consistent with our studies in Molt-4 cells (12).
Effects of TNF-␣ on Dephosphorylation of PKC␣-In an attempt to determine whether TNF-␣ treatment affected PKC␣ levels, Western blot analysis was carried out using extracts prepared from cells treated with vehicle or 5 nM TNF-␣ using anti-PKC␣ antibody. TNF-␣ treatment did not cause a major change in the levels of PKC (Fig. 4). Interestingly, TNF-␣ treatment resulted in an appearance of a doublet of PKC␣ on SDS-PAGE in L929 cells (Fig. 4) as detected by Western blotting. The specific shift in the electrophoretic mobility of PKC␣ is indicated by the lower band of the doublet. This result suggests that TNF-␣ may lead to dephosphorylation of PKC␣ through activation of protein phosphatases since a similar shift has been detected in PKC treated with the protein phosphatase PP2A (27,28).
Mechanisms of TNF-␣ Action on PKC␣ Activity-To determine the mechanisms by which TNF-␣ regulates PKC␣ activity, we first examined the effects of TNF-␣ on the generation of ceramide in L929 cells. L929 cells were treated with various concentrations of TNF-␣ for 12-14 h. TNF-␣ treatment of L929 cells resulted in an increasing level of ceramide with an increasing amount of TNF-␣ (Fig. 5). An ϳ2-fold increase in ceramide level was detected at a TNF-␣ concentration as low as 0.5 nM, whereas a Ͼ3-fold increase in ceramide level was seen with 5 nM TNF-␣ treatment. These results are consistent with a previous study reported by Jayadev et al. (29). Interestingly, in parallel to ceramide increase, diacylglycerol levels also increased 2-fold when cells were treated with 0.5 nM TNF-␣. However, a lesser increase in diacylglycerol levels was detected when treated with 5 nM TNF-␣, a concentration that led to a substantial degree of cell death as reported before (29). In an attempt to determine whether ceramide production is necessary for the observed effect of TNF-␣ on PKC␣, we examined the effect of fumonisin B 1 , an inhibitor of sphingosine N-acyltransferase or ceramide synthase (30), on PKC activity upon TNF-␣ treatment. Fig. 5B shows that fumonisin B 1 significantly blocked the effect of TNF-␣ on PKC activity. To substantiate our result, we evaluated the effect of GSH, an inhibitor of neutral sphingomyelinase (31), on PKC activity following treatment with TNF-␣. Fig. 5C demonstrates that GSH reversed the effect of TNF-␣ on PKC activity. These results suggest that the effects of TNF-␣ on PKC activity were in part mediated by the generation of ceramide through either de novo synthesis of sphingolipids or sphingolipid signaling via neutral sphingomyelinase.
Previous studies have demonstrated that ceramide activates a cytosolic protein phosphatase (CAPP) in vitro (7). The mechanisms underlying the cellular action by CAPP remain to be elucidated. To determine the role of CAPP in TNF-␣ signal transduction through ceramide, we evaluated the ability of the phosphatase inhibitor okadaic acid to block the effect of TNF-␣ on inhibiting PKC activity. L929 cells were treated with TNF-␣ in the absence and presence of a low concentration of okadaic acid (2 nM) that is specific to PP2A. PKC␣ activity was measured using histone H1 as the substrate as described under "Experimental Procedures." Fig. 6A shows that okadaic acid partially reversed the effect of TNF-␣ on PKC␣ inhibition. In addition, okadaic acid also partially reversed the effect of C 6ceramide on PKC␣ inhibition (Fig. 6B), which is consistent with previous studies on Molt-4 cells (12). These data demonstrate that okadaic acid blocked the effects of TNF-␣ and ceramide on the inhibition of PKC␣ activity and suggest that activation of CAPP may represent one of the mediators regulating TNF-␣ signaling.
It has been shown that dephosphorylation of PKC by the  protein phosphatases PP1 and PP2A leads to specific shifts in the mobility of PKC␣ on SDS-PAGE (27). Since PKC is regulated in vivo by phosphorylation, we next determined the effects of TNF-␣ on the phosphorylation state of PKC. We utilized two different antibodies either detecting activated phosphorylation at Ser 660 (32) or recognizing the phosphorylated activa-tion loop of PKCs (p500) (33). L929 cells were treated with TNF-␣, C 6 -ceramide, or dihydro-C 6 -ceramide as described under "Experimental Procedures." Cell lysates were immunoprecipitated with anti-PKC␣ antibody, followed by Western blot analyses using anti-p500 antibodies, which revealed that treatment of either TNF or C 6 -ceramide diminished phosphorylation of PKC, whereas dihydro-C6 ceramide did not have an appreciable effect as detected by anti-p500 antibody (Fig. 7, A  and B). A similar result was obtained when the anti-Ser 660 antibody was used in Western blot analyses (Fig. 7, C and D).
To assess the effects of either TNF-␣ or C 6 -ceramide on the total PKC␣ protein, we performed Western blot analyses using anti-PKC␣ antibody. Treatment of TNF-␣ resulted in the appearance of a doublet of PKC␣ on SDS-PAGE as detected before (Figs. 4 and 7E). Treatment of C 6 -ceramide did not affect the levels of PKC␣ protein (Fig. 7f). These results demonstrate that alteration of PKC phosphorylation is a crucial event in TNF/ ceramide signal transduction.
To determine whether activation of protein phosphatase and dephosphorylation of PKC␣ are responsible for the translocation process, we treated L929 cells with TNF-␣ in the absence and presence of okadaic acid (2-5 nM), followed by PMA stimulation. The membrane and cytosolic fractions were collected as described under "Experimental Procedures." Clearly, PMA led to PKC␣ translocation in control and TNF-␣-treated cells as seen before (Fig. 8). The addition of okadaic acid did not affect the induction of PKC␣ translocation by TNF-␣. Furthermore, C 6 -ceramide had a minimal effect on the translocation of PKC␣ in the absence or presence of okadaic acid (data not shown). These results demonstrate that the dephosphorylation and translocation processes are two separate events and that translocation of PKC␣ by TNF-␣ is independent of ceramide. DISCUSSION This study demonstrates conflicting signals on PKC␣ activity by TNF-␣. Specifically, TNF-␣ inactivates cellular PKC␣, and TNF-␣ is also capable of inducing PKC␣ translocation from the cytosol to the membrane. In addition, this study on the mechanisms of TNF-␣ action illustrates the following. 1) TNF-␣ causes a concentration-dependent generation of ceramide and diacylglycerol; 2) activation of protein phosphatases (such as CAPP) appears to mediate some of the TNF effect on the alteration of PKC␣ activity; and 3) TNF-␣-induced dephosphorylation and translocation of PKC␣ are independent processes. PKC activation is often initiated by direct translocation from the cytosol to the membrane, where it is activated by diacylglycerol. One potential mechanism by which TNF-␣ could exert its inhibitory effect on PKC␣ is through inhibition of its translocation. Our data demonstrate that this as not the case, as inhibition of PKC␣ activity by TNF-␣ is independent of enzyme translocation (Fig. 3). More important, TNF-␣ induced PKC␣ translocation on its own and independent of ceramide and of activation of okadaic acid-inhibitable phosphatases ( Fig. 8 and  data not shown). These results suggest that the interactions between different signaling pathways may initiate a conflict in cell signaling. PKC is likely to be a downstream target of TNF-␣ signal transduction, and inhibition of PKC␣ activity may represent the end point of a series of different cellular signals. In fact, we found that the levels of diacylglycerol were increased concomitantly with ceramide following TNF-␣ treatment (Fig. 5A). Thus, activation of diacylglycerol by TNF-␣ may lead to the induction of PKC␣ translocation. These results are in agreement with a previous study demonstrating cell typespecific activation and translocation of PKC by TNF-␣ (16). On the other hand, generation of ceramide by TNF-␣ may tilt the balance between PKC/diacylglycerol and ceramide/phosphatase interactions and lead to inhibition of PKC␣ activity. Recently, PKC has been linked to the regulation of both cell growth and cell death. Numerous studies have demonstrated its involvement in stimulation of cell growth and transformation (34 -38). Accumulating evidence also points to its negative regulation of cell growth and the initiation of apoptosis (39 -42). In the prostate epithelial cell line LNCaP, activation of PKC␣ isozyme following treatment with 12-O-tetradecanoylphorbol-13-acetate led to apoptosis (43). A few other studies have also demonstrated that PKC activators inhibit DNA synthesis in certain cell types (44 -47). These observations suggest that PKC isozymes play specific, differential roles in signal transduction. The differences among individual PKC isozymes with respect to their tissue distribution, substrate specificity, and mode of activation constitute the multifunctions of PKC under different circumstances. In support of this notion, it has been reported that C 2 -ceramide is able to induce translocation of PKC␦ and PKC⑀ to the cytosol, leading to apoptosis (48).
Another recent study has also demonstrated that ionizing radiation induces PKC␣ translocation from the membrane to the cytosol in the TF-1 human erythroleukemia cell line (18).
Elevation of ceramide has been implicated to be responsible in part for the regulation of cellular functions in TNF-␣ signaling pathways. Neutral sphingomyelinase is considered a key enzyme in the regulation of hydrolysis of sphingomyelin to form ceramide in response to a variety of extracellular agents (4,5). Recent observations demonstrated that GSH inhibits neutral sphingomyelinase in vitro and provides a useful tool to further understand the regulatory mechanism of sphingolipid signaling (31). Besides activation of sphingomyelinases, another major pathway controlling cellular ceramide levels is through de novo synthesis. The realization of the importance of de novo synthesis has come from studies utilizing the myco-toxin fumonisin. Fumonisin B 1 , an inhibitor of ceramide synthase (30), has been widely used to block de novo synthesis of ceramide. Recent studies have also used fumonisin as a tool in ceramide-mediated apoptosis. Using fumonisin B 1 and GSH, we were able to show that both fumonisin B 1 and GSH reversed the effect of TNF-␣ on PKC activity (Fig. 5, B and C). These data suggest that ceramide may mediate some of the effects of TNF-␣ on cellular functions through either de novo biosynthesis or activation of neutral sphingomyelinase.
In addition to ceramide generation, TNF-␣ also leads to generation of arachidonic acid via activation of phospholipase A 2 (49). Phospholipase A 2 hydrolyzes phospholipids to generate free fatty acids and lysophospholipids. These fatty acids (including arachidonic acid) potentiate the diacylglycerol-dependent activation of PKC (50,51). Therefore, generation of free fatty acids through activation of phospholipase A 2 could induce translocation of PKC␣ and activate the enzyme. On the other hand, subsequent generation of ceramide through activation of phospholipase A 2 in TNF-␣ signal transduction (29) can lead to inhibition of PKC␣ activity. Thus, alteration of PKC activity reflects the existence of a complex array of cross-talk between signaling pathways. Tilting the balance of cell regulation by a variety of intracellular and extracellular agents including TNF-␣ will ultimately change cell growth processes.
Activation of PKC significantly influences cellular responses, including cell proliferation and differentiation (51). One important aspect of PKC activation relates to its phosphorylation state. Phosphorylation of PKC is a preliminary and necessary event for its subsequent activation. Initially, PKC␣ is synthesized as a dephosphorylated, membrane-bound, inactive protein of 74 kDa that is converted to an active and mature protein of 80 kDa by post-translational modification such as phospho- rylation (52,53). Recent studies have identified three functionally distinct phosphorylation sites of protein kinase C (32). Transphosphorylation of threonine 500 on the activation loop yields catalytic competency of PKC autophosphorylation. Autophosphorylation of threonine 641 at the carboxyl terminus maintains its competency of PKC autophosphorylation. The membrane-bound enzyme is autophosphorylated at position 660 at the carboxyl terminus. This autophosphorylation may play an important role in the regulation of the membrane affinity of the enzyme and its subcellular localization (32). Our results using anti-phospho-PKC antibodies demonstrated that treatment with TNF-␣ or C 6 -ceramide altered the phosphorylation state of PKC at p500 and Ser 660 (Fig. 7). Inhibition of PKC phosphorylation will ultimately lead to the inactivation of the enzyme. Another recent study (54) has shown that phosphorylation of PKC at Ser 657 is important in controlling the accumulation of phosphate at other sites on PKC␣ and in maintaining its phosphatase-resistant conformation. PKC␣ with C-terminal mutations at Ser 657 exhibited a doublet on SDS-PAGE. Dephosphorylation at Ser 657 in PKC␣ could cause hypersensitivity of PKC␣ toward phosphatases, which in turn leads to a further alteration of the enzyme activity. It is conceivable that activation of CAPP by ceramide through TNF-␣ signal transduction may lead to dephosphorylation of PKC␣ at multiple sites, including Thr 500 , Ser 660 , and Ser 657 . Following the dephosphorylation at Ser 657 , the PKC isozyme becomes vulnerable to protein phosphatases and ultimately becomes inactivated. Clearly, the regulation of phosphorylation of PKC␣ has an essential role in cell signaling.
Following sustained activation of PKC by exposure to the metabolic activator 12-O-tetradecanoylphorbol-13-acetate, which leads to dephosphorylation of PKC␣ in COS cells (28), PKC␣ migrated as a doublet characteristic of phosphorylated and dephosphorylated PKC␣ species. Indeed, dephosphorylation of PKC␣ by 12-O-tetradecanoylphorbol-13-acetate correlates with the membrane-associated heterotrimeric PP2A. Consistent with this finding, our data have also demonstrated an appearance of a doublet of PKC␣ on SDS-PAGE upon TNF-␣ treatment (Fig. 4). Moreover, both TNF-␣ and ceramide were able to block PKC phosphorylation as detected by anti-phospho-PKC antibodies (Fig. 7). These results suggest that generation of ceramide following exposure to TNF-␣ may lead to dephosphorylation of PKC␣ through activation of a protein phosphatase such as CAPP and ultimately inactivation of the enzyme.
The elucidation of the role of phosphoinositide-dependent kinase-1 in phosphoinositide (PI) 3-kinase signaling unveiled a key regulatory pathway in the regulation of protein kinase B/Akt (55). Phosphoinositide-dependent kinase-1 binds to phosphatidylinositol 3,4,5-triphosphate and phosphorylates and activates PI 3-kinase targets, protein kinase B/Akt and p70 S6K (56,57). Similarly, phosphoinositide-dependent kinase-1 regulates the phosphorylation of conventional PKC isozymes in vivo and in vitro (58). Phosphoinositide-dependent kinase-1 specifically phosphorylates the activation loop of both PKC␣ and PKC␤II, thus providing a critical step in the regulation of PKC function. Given the facts that agents that activate the PI 3-kinase/Akt1 pathway trigger an anti-apoptotic signal (59) and that ceramide regulates phospholipase D and PKC (12,60,61) and leads to activation of apoptotic signaling, it is conceivable that TNF-␣ and/or ceramide could inhibit PKC␣ through regulation of phosphoinositide-dependent kinase-1. In fact, several recent studies have reported that TNF-␣ and/or ceramide activates PI 3-kinase (62, 63), whereas Zundel and Giaccia (64) have shown the inhibition of PI 3-kinase/Akt kinase by ceramide. A few other studies have demonstrated the inhibition of Akt kinase activity by ceramide, but the effect of ceramide on PI 3-kinase activity was not significant (65,66). The differences between these studies may be due to the different experimental designs and the different model systems used in the studies. In attempt to delineate the mechanism by which ceramide regulates Akt kinase activity, Cuadrado and coworkers (66) found that down-regulation of the Akt kinase by ceramide is through dephosphorylation of Akt1 at Thr 308 and Ser 473 by activation of CAPP. These results are consistent with the finding that PP2A inactivates Akt kinase in vitro (67). However, the concentrations of okadaic acid used were relatively high and probably did not inhibit PP2A or CAPP specifically. Others did not observe a significant effect by okadaic acid (65,68). Thus, the involvement of PP2A or CAPP in the regulation of phosphorylation of Akt kinase needs further clarification. Nonetheless, inhibition of Akt kinase by ceramide through modulation of phosphorylation provides one of the key mechanisms of its regulation. Whether disruption of the phosphoinositide-dependent kinase-1/PI 3-kinase/Akt signaling pathway has a direct impact on the regulation of PKC␣ activity by TNF-␣ and/or ceramide in L929 cells through activation of CAPP remains to be determined.
The underlying mechanism by which TNF-␣ regulates PKC␣ activity is still poorly understood. Evidently, the regulation of phosphorylation and translocation of PKC has a significant impact on its activity and signaling pathways. It remains unclear how phosphorylation is related to the translocation process in cell signaling. Our current studies demonstrate that activation of a protein phosphatase such as CAPP and dephosphorylation of PKC␣ by TNF-␣ are independent of PKC␣ translocation activated by TNF-␣ (Fig. 8). Thus, it is highly likely that a combination of a complex intracellular environment and extracellular influences dictate the balance among different signaling pathways (including PKC and TNF-␣) that then direct the outcome of cellular responses. Studying the mechanisms by which PKC signaling pathways respond to TNF-␣ FIG. 7. Effects of TNF and C 6 -ceramide on PKC␣ phosphorylation. Cells were treated with TNF-␣ (A, C, and E), C 6 -ceramide (B, D, and F), or dihydro-C 6 -ceramide (DHC6; B, D, and E) as described under "Results." Lysates were immunoprecipitated with anti-PKC␣ antibody, and Western blot analyses were done using either anti-p500 (A and B, upper panels) or anti-Ser 660 (C and D, upper panels) antibody (Ab). Phospho-PKC is indicated on the right. Densitometric analyses were done as described under "Results" (A-D, lower panels). Total protein levels were analyzed by Western blotting using anti-PKC␣ antibody as described under "Experimental Procedures" (E and F, upper panels). Data analyses accompanying the total protein levels are shown in the lower panels. Results are representative of two to three different experiments. Densitometric analyses of the autoradiographs are expressed as means Ϯ S.D. of two or three experiments. C, control; T, TNF. exposure may add to our knowledge of the basic biology of cell survival.