Phosphorylation of Protein Kinase Cδ on Distinct Tyrosine Residues Regulates Specific Cellular Functions*

Protein kinase Cδ (PKCδ) inhibits proliferation and decreases expression of the differentiation marker glutamine synthetase (GS) in C6 glioma cells. Here, we report that distinct, specific tyrosine residues on PKCδ are involved in these two responses. Transfection of cells with PKCδ mutated at tyrosine 155 to phenylalanine caused enhanced proliferation in response to 12-phorbol 12-myristate 13-acetate, whereas GS expression resembled that for the PKCδ wild-type transfectant. Conversely, transfection with PKCδ mutated at tyrosine 187 to phenylalanine resulted in increased expression of GS, whereas the rate of proliferation resembled that of the PKCδ wild-type transfectant. The tyrosine phosphorylation of PKCδ and the decrease in GS expression induced by platelet-derived growth factor (PDGF) were abolished by the Src kinase inhibitors PP1 and PP2. In response to PDGF, Fyn associated with PKCδ via tyrosine 187. Finally, overexpression of dominant negative Fyn abrogated the decrease in GS expression and reduced the tyrosine phosphorylation of PKCδ induced by PDGF. We conclude that the tyrosine phosphorylation of PKCδ and its association with tyrosine kinases may be an important point of divergence in PKC signaling.

domain and a C-terminal catalytic domain with serine-threonine kinase activity (10,11). Both domains contain conserved (C) regions of extended sequence homology and variable (V) regions. In the classical PKC isoforms the regulatory domain contains a Ca 2ϩ -binding domain, and in both the classical and novel PKC isoforms it contains a pair of highly conserved zinc fingers (C1 domains) that bind phorbol esters and a pseudosubstrate region (12)(13)(14). PKC chimeras have been used to study the role of the regulatory and catalytic domains of different PKC isoforms.
PKC␦ is a widely expressed member of the novel PKCs (15). This isoform has been associated with the proliferation of various cells in a cell type-specific manner. For example, PKC␦ inhibited the proliferation of smooth muscle cells (16) and glial cells (17) and caused cell arrest at the G 2 /M phase of the cell cycle in Chinese hamster ovary cells (18). In contrast, in breast cancer cells PKC␦ has been shown to increase tumorigenicity (19). PKC␦ has also been reported to play a role in cell differentiation. Thus, PKC␦ has been shown to undergo translocation and activation during differentiation of keratinocytes (20) and overexpression of this isoform induced squamous (21) and myeloid cell differentiation (22).
Recent studies suggest that PKC␦ associates with different tyrosine kinases and that this association can induce the tyrosine phosphorylation of PKC␦ itself and can affect the activity of both the tyrosine kinases and PKC (15). PKC␦ has been shown to be tyrosine phosphorylated in response to various stimuli such as PMA, EGF, PDGF (23)(24)(25), ligands for the IgE receptor (26,27), ATP, and H 2 O 2 (28). The phosphorylation site(s) and the role of tyrosine phosphorylation of PKC␦ in its activity and in its function are just beginning to be understood. Two specific sites in the regulatory domain, tyrosines 187 and 52, have been reported so far to be phosphorylated in response to PDGF and Fc⑀RI, respectively (29), and tyrosine 311 has been shown to be phosphorylated by Src (30). In contrast, tyrosine residues in the catalytic domain of PKC␦ have been reported to be phosphorylated in response to H 2 O 2 (28).
In a recent study (23), we found that in C6 glioma cells tyrosine phosphorylation of PKC␦ in the regulatory domain mediated the inhibitory effect of this isoform on the expression of the astrocytic marker, glutamine synthetase (GS). In the present study, we found that tyrosine phosphorylation of PKC␦ also plays a role in the inhibitory effect of PKC␦ on cell proliferation and have identified different tyrosine residues that are involved in the selective effects of PKC␦ on cell proliferation and GS expression.

EXPERIMENTAL PROCEDURES
Materials-PDGF and an affinity-purified polyclonal anti-PKC⑀ antibody against a polypeptide corresponding to amino acids 726 -737 of PKC⑀ were purchased from Life Technologies, Inc. Monoclonal anti-PKC and anti-GS antibodies were obtained from Transduction Laboratories (Lexington, KY). Polyclonal anti-PKC antibodies and anti-Src, -Fyn, and -Lyn antibodies were from Santa Cruz (Santa Cruz, CA). PMA was from Alexis Co. (San Diego, CA). Leupeptin, aprotinin, phenylmethylsulfonyl fluoride, and sodium vanadate were obtained from Sigma.
Generation of PKC Chimeras-The PKC chimeras were generated by exchanging the regulatory and catalytic domains of PKC␣, -␦, and -⑀ as described by Acs et al. (31). PKC␣/␦ refers to the chimera with the PKC␣ regulatory domain and the PKC␦ catalytic domain, and PKC␦/␣ refers to the reciprocal chimera. The PKC cDNAs were subcloned into the metallothionein promoter-driven eukaryotic expression vector (MTH). The vector sequence encodes a C-terminal PKC⑀-derived 12-amino acid tag (⑀MTH) that is added to the expressed proteins (44). The expression of these chimeras and their activities in C6 cells were described recently (23).
Site-directed Mutagenesis of PKC␦-Mouse PKC␦ was cloned into the pGEM-T vector (Promega, Madison, WI) as described previously (23). This plasmid served as our "master" vector for the site-directed mutagenesis, using the Transformer Site-Directed Mutagenesis Kit from CLONTECH (Palo Alto, CA). Conversion of tyrosine residues at sites 52, 64, 155, 187, and 565 into phenylalanine was performed as described previously (23). PKC␦ and the PKC␦ mutants were subcloned into the metallothionein promoter-driven eukaryotic expression vector (⑀MTH).
Construction of PKC␦-GFP Fusion Protein-cDNAs encoding the murine PKC␦ and the various PKC␦ mutants were fused into the Nterminal-enhanced GFP vector pEGFP-N1 (CLONTECH, Palo Alto, CA). The original pEGFP-N1 vector was modified by the insertion of an MluI site in the plasmid polylinker. The restriction site was created by ligating a phosphorylated linker containing the MluI site into pEGFP-N1 digested with SmaI. The construct was verified by sequencing. The clones containing the GFP-PKC␦or GFP-fused to the different PKC␦ mutants were constructed by the excision of PKC␦ or the specific mutants from MTH-PKC plasmids by digestion with XhoI and MluI. The inserts were then ligated into the modified GFP vector using the same restriction sites. DNA sequencing of the GFP-PKC constructs confirmed the intended reading frame.
C6 Glial Cultures and Cell Transfection-C6 cells of late passages (50 -60), that exhibit an astrocytic phenotype, were used in this study. Cells of these passages showed somewhat smaller response to overexpression of PKC␦ as compared with the C6 cells of passage 30, which exhibit progenitor properties (17,23). For the current studies we chose cells of late passages, because we wanted to focus on the effects of PKC␦ on the expression of GS and not on general aspects of cell differentiation. Cells (1 ϫ 10 5 cells/ml) were seeded on tissue culture dishes (10 cm) and were grown in medium consisting of Dulbecco's modified Eagle's medium containing 10% heat-inactivated fetal calf serum, 2 mM glutamine, penicillin (50 units/ml), and streptomycin (0.05 mg/ml). The cells were transfected either with the empty vectors, the different PKC␦ expression vectors, or a Fyn dominant negative mutant in pSG5 (kindly provided by Alan P. Saltiel) using LipofectAMINE (Life Technologies, Inc.) as described previously (17). Experiments were routinely carried out on a clone of the transfected cells, but all the results were confirmed on one pool and two additional individual clones.
For overexpression of the GFP-PKC␦ fusion proteins, C6 cells were seeded onto 40-mm round glass coverslips at a density of 5 ϫ 10 4 cells/coverslip. Twenty-four hours later, cells were transfected with the different GFP-PKC␦ constructs using LipofectAMINE Plus reagent according to the manufacturer's instructions. All experiments were performed 48 h post-transfection.
Preparation of Cell Homogenates-Cells were washed and resuspended in serum-free medium. The plates were placed on ice, scraped with a rubber policeman, and centrifuged at 1,400 rpm for 10 min. The supernatants were aspirated, and the cell pellets were resuspended in 100 l of lysis buffer (25 mM Tris-HCl, pH 7.4, 50 mM NaCl, 0.5% sodium deoxycholate, 2% Nonidet P-40, 0.2% SDS, 1 mM phenylmethylsulfonyl fluoride, 50 g/ml aprotinin, 50 M leupeptin, 0.5 mM Na 3 VO 4 ) on ice for 15 min. The cell lysates were centrifuged for 15 min at 14,000 rpm in an Eppendorf microcentrifuge, supernatants were removed, and 2ϫ sample buffer was added.
Immunoblot Analysis-Lysates (20 g of protein) were resolved by SDS-PAGE (10%) and were transferred to nitrocellulose membranes. The membranes were blocked with 5% dry milk in phosphate-buffered saline and subsequently stained with the primary antibody. Specific reactive bands were detected using a goat anti-rabbit or goat antimouse IgG conjugated to horseradish peroxidase (Bio-Rad), and the immunoreactive bands were visualized by the ECL Western blotting detection kit (Amersham Pharmacia Biotech).
Immunoprecipitation-Immunoprecipitation was performed as described previously (23). Briefly, C6 cells overexpressing PKC␦ or the PKC␦ mutants were serum-starved overnight and treated for different periods of time with PMA (10 nM) or PDGF (100 ng/ml). The samples were preabsorbed with 25 l of protein A/G-Sepharose (50%) for 10 min, and immunoprecipitation was performed using 4 g/ml antibody for 1 h at 4°C and then incubated with 30 l of A/G-Sepharose for an additional hour. Following washes, the pellets were resuspended in 25 l of SDS sample buffer and boiled for 5 min. The entire supernatants were subjected to Western blotting. Membranes were incubated with horseradish peroxidase-conjugated secondary antibodies for 1 h at room temperature. The membranes were washed and visualized by the ECL system.
PKC Kinase Assay-PKC activity was assayed by measuring the incorporation of 32 P from [␥-32 P]ATP into substrate in the presence of 100 g/ml phosphatidylserine and 1 M PMA as described previously (23).
Cell Proliferation Assay-Cells overexpressing the wild-type PKC␦ or the PKC␦ mutants were seeded in triplicate and incubated in the absence or presence of ZnCl 2 (20 M) for 24 h followed by treatment with PMA (30 nM) for an additional 48 h. Cells were pulsed with 0.5 Ci of Confocal Microscopy-Confocal fluorescent images were collected with a Bio-Rad MRC 1024 confocal scan head (Bio-Rad) mounted on a Nikon microscope with a 60ϫ planapochromat lens. Excitation at 488 nm was generated by a krypton-argon gas laser with a 522/32 emission filter for green fluorescence. For kinetics of GFP-PKC␦ translocation in living cells, cells plated on a 40-mm-round coverslip were enclosed in a Bioptechs Focht Chamber System (Bioptechs, Butler, PA). The chamber was inverted and attached to the microscope stage with a custom stage adapter; a temperature controller set at 37°C was connected, and medium was perfused through the chamber with a Lambda microperfusion pump. Sequential images of the same cell were collected at various time points using LaserSharp Software.
Statistical Analysis-The results are presented as the mean values Ϯ S.E. All data were analyzed using a paired Student's t test to determine the level of difference between the treatments.

The Regulatory Domain of PKC␦ and Its Tyrosine Phosphorylation Mediate the Decrease in Cell Proliferation Induced by
PMA-In a recent study we demonstrated that the regulatory domain of PKC␦ mediated its inhibitory effect on the expression of the astrocytic marker GS (23). To characterize the effect of PKC␦ on C6 cell proliferation we first examined the relative contributions of the regulatory and catalytic domains of this isoform. For these studies, we used chimeras between the regulatory and catalytic domains of PKC␣, -␦, and -⑀, combined at the highly conserved hinge region. The expression and activity of C6 cells overexpressing the different chimeras were already described previously (23).
Cells were pretreated for 24 h with ZnCl 2 , followed by PMA treatment (20 nM) for an additional 48 h. This concentration of PMA induced prolonged activation of PKC␦ without marked down-regulation. Cells overexpressing PKC␦/␦ exhibited a lower rate of cell proliferation than control cells. In contrast, cells overexpressing PKC␣/␣ and ⑀/⑀ exhibited a higher rate of cell proliferation than controls both in the presence and absence of PMA. Similar to cells expressing PKC␦/␦, cells overexpressing the chimeras containing the regulatory domain of PKC␦, namely ␦/␣ and ␦/⑀, also showed a reduced level of cell proliferation, whereas cells expressing chimeras containing the catalytic domain of PKC␦ together with the regulatory domain of PKC␣ or -⑀ exhibited an increased level of proliferation similar to that observed with cells expressing PKC␣/␣ or PKC⑀/⑀ (Fig. 1A). Untreated cells overexpressing PKC␦/␦ or chimeras containing the regulatory domain of PKC␦ also showed decreased cell proliferation as compared with control vector cells, but the magnitude of this decrease was much lower than that observed in PMA-treated cells (Fig. 1A).
We reported that the expression of the PKC␦5 mutant (in which tyrosines 52, 64, 155, 187, and 565 were mutated to phenylalanine) in C6 cells abolished the decrease in the expression of the astrocytic marker GS induced by PMA or PDGF. Expression of the PKC␦5 mutant also resulted in a lower tyrosine phosphorylation of PKC␦ in response to these treatments (23). Using cells expressing PKC␦5 we found that in response to PMA (20 nM) these cells displayed enhanced proliferation as compared with vector control cells, contrasting with cells overexpressing PKC␦ WT which exhibited a significantly lower proliferative response (Fig. 1B). Similar results were obtained when the proliferation of C6 cells expressing PKC␦ and PKC␦5 was followed over the course of 5 days (data not shown).
Overexpression of PKC␦ Mutants-To identify the specific tyrosines that are involved in the inhibitory effect of PKC␦ on cell proliferation and GS expression, we examined the role of tyrosines 52, 155,and 187, which were already reported to be phosphorylated in other systems (25,29). C6 cells were stably transfected with PKC␦ mutants in which one of these three tyrosines was mutated to phenylalanine. Fig. 2A illustrates a representative Western blot of C6 cells overexpressing PKC␦, PKC␦5, PKC␦Y155F, PKC␦Y187F, and PKC␦Y52F mutants, and the vector control. Using the previously described tagging system (44), we were able to detect specifically the transfected PKC isoforms using an antibody against the PKC⑀ epitope tag on the constructs. The levels of the overexpressed PKC␦ WT and the PKC␦ mutants in the transfected cells we used were 8 -10-fold higher than the endogenous PKC␦ as determined using an anti-PKC␦ specific antibody (data not shown). Cells overexpressing PKC␦ or the PKC␦ mutants expressed similar levels of PKC␣, -␤, -␥, -⑀, -, -, -, and -(data not shown), thus excluding the possibility that the different effects of PKC␦, and the PKC␦ mutants were indirectly mediated by changes in the expression of the other PKC isoforms. To establish that the overexpressed PKC␦ mutants were functionally active, we measured kinase activity on cell lysates. All PKC␦ mutants expressed higher kinase activity as compared with the control vector cells (Fig. 2B).
Tyrosine Phosphorylation of the PKC␦ Mutants by PMA and PDGF-The degree of tyrosine phosphorylation of the various PKC␦ mutants was examined in response to PMA and PDGF. As illustrated in Fig. 3, a small basal level of tyrosine phosphorylation was observed in untreated cells. PMA and PDGF induced marked tyrosine phosphorylation of PKC␦ WT but, as already described, no significant degree of enhanced tyrosine phosphorylation of PKC␦5 (p Ͼ 0.07, n ϭ 5). The degree of tyrosine phosphorylation of PKC␦Y155F and PKC␦Y187F in response to PMA was intermediate between that of PKC␦ WT and that of the PKC␦5 mutant, suggesting that both tyrosines 155 and 187 are being phosphorylated in response to PMA. In contrast, cells overexpressing PKC␦Y52F exhibited a level of tyrosine phosphorylation similar to that of PKC␦ WT, suggesting that this tyrosine may not be phosphorylated in response to PMA. As already described, PDGF also induced tyrosine phosphorylation of PKC␦ and this effect was abolished in the PKC␦5 cells. We found that treatment of cells overexpressing PKC␦Y187F with PDGF resulted in a very low tyrosine phosphorylation of PKC␦. In contrast, PDGF-stimulated cells overexpressing PKC␦Y155F or PKC␦Y52F exhibited similar levels of tyrosine phosphorylation to those observed in cells overexpressing PKC␦ WT.
Tyrosine 155 Mediates the Inhibitory Effect of PKC␦ on Cell Proliferation, whereas Tyrosine 187 Mediates the Inhibitory Effect of PKC␦ on GS Expression-We then examined the degree of cell proliferation and GS expression in cells overexpressing the different PKC mutants. We found that, in re- sponse to PMA, cells overexpressing PKC␦Y155F displayed an increased cell proliferation over control vector expressing cells and significantly increased proliferation compared with cells overexpressing PKC␦ WT (Fig. 4A). Interestingly, the degree of cell proliferation in the PKC␦Y155F cells was similar to that obtained with the PKC␦5 overexpressing cells. In contrast, cells overexpressing the PKC␦Y187F or PKC␦Y52F mutant exhibited a reduced degree of cell proliferation as compared with vector control cells and were similar to cells overexpressing PKC␦ WT.
The effects of the PKC␦ mutants were also examined on the expression of GS in response to PDGF. Cells were pretreated for 24 h with ZnCl 2 and then with PDGF (100 ng/ml) for an additional 48 h. As described previously, cells overexpressing PKC␦ WT exhibited lower levels of GS expression upon PDGF treatments, whereas cells overexpressing PKC␦5 exhibited increased levels of GS compared with vector control cells. Cells overexpressing PKC␦Y155F or PKC␦Y52F exhibited lower GS expression in PDGF-treated cells similar to the level obtained in cells overexpressing PKC␦ WT. Opposite results were obtained in cells overexpressing PKC␦Y187F. These cells exhibited an increased expression of GS in response to stimulation with PDGF, similar to the results obtained in cells overexpressing PKC␦5 (Fig. 4B). Similar results with the mutants were obtained in cells treated with 20 nM PMA (data not shown).
Translocation and Degradation of PKC␦ and the PKC␦ Mutants-One possible explanation for the differential effect of the mutants on cell proliferation and GS expression is translocation to different cellular compartments following stimulation. We therefore examined the translocation of the different PKC␦ mutants in response to PMA and PDGF. For these experiments we used GFP-tagged PKC␦ WT or the different PKC␦ mutants. Cells were transiently transfected with the specific GFP-PKC␦ mutants, and the response of the cells to PMA or PDGF was monitored over a period of 30 min. Stimulation of the cells with 100 nM PMA induced initial translocation of PKC␦ to the plasma membrane followed by some translocation of PKC␦ to the perinuclear membrane (Fig. 5). PDGF also induced translocation of PKC␦, but the magnitude of this translocation was much lower than that induced by PMA. PDGF induced some membranal translocation of PKC␦ and distribution of PKC␦ around the perinuclear membrane (Fig. 5). A similar pattern of translocation was observed for PKC␦ and the PKC␦ mutants in response to PMA (Fig. 6) and PDGF (data not shown). Thus, PMA induced translocation of the PKC␦ mutants to both the plasma membrane and the perinuclear membrane (Fig. 6). The kinetics of the translocation as well as the response to lower concentrations of PMA were likewise similar in all of the mu-tants (data not shown). PMA or PDGF did not induce any changes in cells overexpressing GFP protein alone (data not shown).
Recent studies suggested that tyrosine phosphorylation of PKC␦ may play a role in its degradation (30). Since the inhibitory effects of PMA on GS expression and cell proliferation require a prolonged exposure of the cells to PMA, we examined whether the various PKC␦ mutants exhibit different degrees of degradation as compared with PKC␦ WT. For these experiments, cells overexpressing PKC␦ WT or the PKC␦ mutants were treated for 24 h with a range of concentrations of PMA, and the expression was examined by Western blot analysis using the anti-⑀ tag antibody. PMA induced a dose-dependent decrease in the expression of the exogenous PKC␦ WT. Similar results were obtained for PKC␦5 or the different PKC␦ mutants (data not shown).
Phosphorylation of PKC␦ by PDGF Is Inhibited by PP1 and PP2-To examine the role of Src-related kinases in the tyrosine phosphorylation of PKC␦, we employed the Src kinase inhibitors PP1 and PP2. Pretreatment of the cells with either PP1 or PP2 abolished the tyrosine phosphorylation of PKC␦ in response to PDGF (Fig. 7A). PP1 and PP2 also abrogated the inhibitory effect of PKC␦ on GS expression (Fig. 7B), providing further support that the tyrosine-phosphorylated form of PKC␦ is involved in the inhibition of GS expression. Since PP1 has also been reported to inhibit the kinase activity of the PDGFR␤ at a similar concentration range (30), our results suggest that either Src-related kinases or the PDGFR␤ are involved in the phosphorylation of PKC␦ in response to PDGF.
Association of PKC␦ with Src-related Kinases-To examine the association of PKC␦ with Src-related kinases we performed co-immunoprecipitation of PKC␦ with Src, Fyn, and Lyn. We found that in unstimulated C6 cells PKC␦ was constitutively associated with p60 Src (Fig. 8A). Stimulation of the cells with either PMA or PDGF for 1-60 min did not induce further association of PKC␦ with Src (data not shown). In contrast, PDGF induced association of Fyn with PKC␦ and, to a lesser extent, association of Lyn (Fig. 8A). To further characterize the association of Fyn with PKC␦, we performed a kinetic study and found that PDGF induced association of Fyn following 1 min of treatment and that this association was decreased after 30 min (Fig. 8B).
Since the decrease in GS expression by PDGF appeared to involve phosphorylation of tyrosine 187 in PKC␦, we examined the role of this tyrosine residue in the association of Fyn and PKC␦. Stimulation of cells overexpressing PKC␦Y187F with PDGF (Fig. 8B) or PMA (data not shown) did not lead to association of the mutated PKC␦ with Fyn as determined by co-immunoprecipitation, indicating the tyrosine 187 is essential for the association. In contrast, the mutant PKC␦Y155F associated with Fyn similarly to PKC␦ WT (data not shown), suggesting the tyrosine 155 does not play a role in the association of PKC␦ and Fyn.
Fyn Is Involved in the Inhibitory Effect of PKC␦ on GS Expression-Since Fyn associates via tyrosine 187 with PKC␦ in response to PDGF treatment, we wanted to examine the role of Fyn in the tyrosine phosphorylation of PKC␦ and in the inhibitory effect of PKC␦ on GS expression. We stably transfected a Fyn dominant negative mutant in C6 cells and examined their response to PDGF. Cells overexpressing empty vector exhibited an increase in the tyrosine phosphorylation of PKC␦ following 10 min of PDGF treatment. In contrast, a decrease in the degree of the tyrosine phosphorylation of PKC␦ was observed in cells overexpressing the Fyn dominant negative mutant (Fig.  9A). We also examined the expression of GS in the different cells. PDGF decreased the expression of GS in cells overex- pressing control vector. In contrast, there was no significant decrease in the expression of GS in response to PDGF in cells overexpressing the Fyn dominant negative mutant (Fig. 9B). Interestingly, the inhibitory effect of the Fyn dominant negative on the decrease in GS expression induced by PDGF was more marked than the decrease in tyrosine phosphorylation of PKC␦, suggesting that Fyn is not the only tyrosine kinase phosphorylating PKC␦ in response to PDGF. Similar results with the Fyn dominant negative mutant were obtained in cells treated with 20 nM PMA (data not shown). In contrast, Fyn dominant negative did not abolish the decrease in cell proliferation induced by PMA, suggesting that the effect of Fyn dominant negative was specific to GS expression (data not shown).

DISCUSSION
In this study we explored the mechanisms involved in the inhibitory effects of PKC␦ on C6 cell proliferation and GS expression. We found that the regulatory domain of PKC␦ is responsible for the effects of this isoform on C6 cell proliferation. These results are similar to those we recently described regarding the regulation of GS expression by PKC␦ (23). Chimeras have been used to delineate the contributions of individual PKC domains to the specific functions of different PKC isoforms in a number of systems. Both the catalytic and the regulatory domains of PKC may determine isoform-specific functions depending on the specific system. For example, the catalytic domain of PKC␤ was found to confer isoform-specific function in the differentiation of erythroleukemia cells (14), and the catalytic domain of PKC␦ in reciprocal ␦and ⑀-chimeras mediated PMA-induced macrophage differentiation of mouse promyelocytes (32). In contrast, the regulatory domain of PKC⑀ enhanced cell growth and induced colonies in soft agar in NIH 3T3 cells (33). Recently, it has been reported that the regulatory domain of PKC␦ overexpressed by itself inhibited mammary tumor cell metastases (34).
PKC␦ has been shown to become tyrosine-phosphorylated in the regulatory domain in response to PMA and PDGF in C6 cells (23). Tyrosine phosphorylation of PKC␦ has been reported in response to EGF stimulation in keratinocytes (24), in response to carbachol, substance P and PMA stimulation in parotid acinar cells (35), in response to H 2 O 2 in CHO-K1 cells (28), and in response to activation of the IgE receptor in RBL-2H3 cells (26,27). Constitutive tyrosine phosphorylation was reported in Ras-transformed mouse keratinocytes (36). Our results suggest that Src-related kinases are involved in the tyrosine phosphorylation of PKC␦ in response to PDGF, since the Src kinase inhibitors PP1 and PP2 reduced significantly the phosphorylation induced by PDGF. Our results using cells overexpressing a Fyn dominant negative mutant further suggest that Fyn contributes to PKC␦ tyrosine phosphorylation in response to PDGF. Since PP1 has also been reported to inhibit the kinase activity of PDGFR␤ (3), we cannot exclude at this point that this receptor directly phosphorylates PKC␦ upon PDGF binding. Indeed, Li et al. (22) reported that the PDGF receptor phosphorylated PKC␦ in vitro.
The effect of tyrosine phosphorylation on the activity of PKC␦ or on its function differs, depending on the specific system. Thus, tyrosine phosphorylation of PKC␦ has been reported to reduce its activity in Ras-transformed cells and in response to activation of the EGF receptor (24,36). In contrast, tyrosine phosphorylation of PKC␦ by Fyn increased the kinase activity (37). Recently it was suggested that the tyrosine phosphorylation of PKC␦ in response to engagement of the IgE receptor leads to altered substrate specificity (26). A previous report has shown that mutation of PKC␦ at tyrosine 187 did not change kinase activity (25). In this study we did not further explore the kinase activity of the different PKC mutants using different substrates, since the nature of the endogenous substrates involved in the effects of PKC␦ on cell proliferation and GS expression have not yet been identified.
Our results using the PKC␦5 mutant suggest that tyrosine phosphorylation of PKC␦ plays a role in the inhibitory effect exerted by this isoform on cell proliferation. These results are similar to our recent findings, which showed a role for tyrosine phosphorylation of PKC␦ in the inhibitory effect of this isoform on the expression of the astrocytic marker GS (23). Thus, the PKC␦5 mutant appears to act in an opposite way to PKC␦ in the effect of this isoform on both GS expression and cell proliferation. Since tyrosine phosphorylation of PKC␦ in response to PMA occurs only in the regulatory domain of this isoform in the C6 cells (23), our results are consistent with the importance of the regulatory domain of PKC␦ in the inhibitory effect of this isoform on different cellular functions.
Although tyrosine phosphorylation in the regulatory domain of PKC␦ mediates the inhibitory effects of this isoform on both GS expression and cell proliferation, it appears that different tyrosine residues are involved in the different effects. Thus, tyrosine 155 is implicated as a phosphorylation site that is involved in the inhibitory effect of PKC␦ on cell proliferation but not on GS expression, whereas tyrosine 187 appears to be involved in the inhibitory effect of PKC␦ on GS expression. There have been a number of reports regarding the phosphorylation of specific tyrosines on PKC␦. For example, tyrosine  7. PP1 and PP2 inhibit the PDGF-induced tyrosine phosphorylation of PKC␦ and the decrease in GS expression. C6 cells were treated with PP1 or PP2 for 1 h following by stimulation with PDGF for an additional 15 min (A). Cells were then harvested and immunoprecipitation of PKC␦ was performed as described under "Experimental Procedures." Following SDS-PAGE, membranes were stained with anti-phosphotyrosine antibody (anti-Tyr(P)) or with anti-PKC␦ antibody. C6 cells were pretreated with PP1 or PP2 for 1 h and then with PDGF for 48 h (B). GS expression was determined using Western blot analysis. The results represent one of three separate experiments that gave similar results.
FIG. 8. Association of Src-related kinases with PKC␦. C6 cells were treated with PDGF for 15 min, immunoprecipitation of PKC␦ was performed as described under "Experimental Procedures," and the membranes were stained with anti-Src, anti-Fyn, or anti-Lyn (A). Cells overexpressing PKC␦WT or PKC␦Y187F were treated with PDGF for different periods of time. PKC␦ or PKC␦Y187F was immunoprecipitated using anti-PKC⑀ antibody, and the association of Fyn with PKC␦ was measured (B). The results represent one of three separate experiments that gave similar results. phosphorylation in the catalytic domain of PKC␦ was described in response to treatment with H 2 O 2 (28). Tyrosine 311 was reported to be phosphorylated in response to Src (30), and engagement of the Fc⑀R1 resulted in tyrosine phosphorylation of tyrosine 52 in the regulatory domain (27,29). Interestingly, tyrosine 52 did not appear to be phosphorylated in response to either PMA or PDGF in the C6 cells. Our results of PDGFinduced phosphorylation of tyrosine 187 in PKC␦ are consistent with the results of Li et al. (25), who reported that PMA and PDGF induced phosphorylation of PKC␦ on tyrosine 187 in NIH 3T3 cells. The authors concluded that this phosphorylation site was not important for the monocytic differentiation of 32D cells by PMA. Thus, the monocytic differentiation of 32D cells resembles C6 cell proliferation and contrasts with GS expression in its control by tyrosine 187 phosphorylation of PKC␦.
One of the factors that could explain the different effects of the PKC␦ mutants is a distinct pattern of translocation. Translocation of PKC to specific cellular compartments could lead to different effects due to the phosphorylation of different substrates and to the association of PKC␦ with specific proteins present in these locations. One of the important factors that can determine the localization of PKC following its activation is association with RACKs (receptors for activated protein kinase Cs) (38). It is currently not clear to what extent tyrosine kinases can act as RACKs and affect the translocation of PKC isoforms. In a recent study, Ron et al. (39) suggested that Fyn, which is associated with PKC, might act as a RACK of this PKC isoform. We found that PMA induced initial translocation of PKC␦ to the plasma membrane followed by translocation to the perinuclear membrane. This pattern of translocation is similar to the translocation of PKC␦ reported in CHO-K1 cells (40). PDGF also induced membranal translocation of PKC␦, although to a lesser extent, with some accumulation around the perinuclear membrane. Stimulation of cells overexpressing PKC␦ WT or the different PKC␦ mutants with PMA or PDGF resulted in a similar pattern of translocation. Thus, the differential effects of the different PKC␦ mutants on cell proliferation and GS expression are probably not due to their different translocation following activation. At this point, however, we cannot exclude the possibility that the PKC␦ mutants undergo translocation to different membranal subdomains, causing their association with distinct signaling molecules.
PKC␦ has been reported to associate with different tyrosine kinases such as Src (27,30,41,42), Lyn (27), and c-Abl (43) in either a phosphorylation-dependent or -independent manner. The ability of PKC␦ to be tyrosine phosphorylated on more than one tyrosine suggests that PKC␦ can associate with different tyrosine kinases. We found that PKC␦ associated with Src in a phosphorylation-independent manner and with Fyn following stimulation of PDGF via tyrosine 187. Specifically, the effect of PKC␦ on cell proliferation and GS expression in C6 cells may be mediated by either different downstream PKC substrates or different tyrosine kinases that are associated with PKC␦. Indeed, our results suggest that Fyn is involved in the inhibitory effect of PDGF on GS expression, since overexpression of the Fyn dominant negative mutant abrogated the inhibitory effect of PDGF. It has been reported that the differential interaction of PKC␦ with tyrosine kinases may lead to changes in the activity and substrate recognition of PKC␦ and to changes in the activity of the associated tyrosine kinases (26,27). Thus, the differential phosphorylation of specific tyrosine residues may generate diversity in the effects of PKC␦ and positions this isoform as an important component in a complex bi-directional interaction between serine-threonine and tyrosine kinase signaling. The identities of the tyrosine kinases that are associated with PKC␦ via tyrosine 155 and are involved in the inhibitory effect of PKC␦ on cell proliferation are currently under investigation. FIG. 9. Effects of the Fyn dominant negative mutant on tyrosine phosphorylation of PKC␦ and GS expression by PDGF. For tyrosine phosphorylation of PKC␦, cells overexpressing Fyn dominant negative mutant or the control vector (Control) were treated with PDGF (100 ng/ml) for 10 min, immunoprecipitation of PKC␦ was performed as described under "Experimental Procedures," and the membranes were stained with anti-phosphotyrosine (A). C6 cells expressing a Fyn dominant negative mutant or the control vector (Control) were treated with PDGF for 24 h, and the expression of GS was determined using Western blot analysis (B). The results represent one of five separate experiments, which gave similar results.