Protein Kinase C δ (PKCδ) Inhibits the Expression of Glutamine Synthetase in Glial Cells via the PKCδ Regulatory Domain and Its Tyrosine Phosphorylation*

Protein kinase C (PKC) plays an important role in the proliferation and differentiation of glial cells. In a recent study we found that overexpression of PKCδ reduced the expression of the astrocytic marker glutamine synthetase (GS). In this study we explored the mechanisms involved in the inhibitory effect of PKCδ on the expression of glutamine synthetase. Using PKC chimeras we first examined the role of the catalytic and regulatory domains of PKCδ on the expression of glutamine synthetase. We found that cells stably transfected with chimeras between the regulatory domain of PKCδ and the catalytic domains of PKCα or ε inhibited the expression of GS, similar to the inhibition exerted by overexpression of PKCδ itself. In contrast, no significant effects were observed in cells transfected with the reciprocal PKC chimeras between the regulatory domains of PKCα or ε and the catalytic domain of PKCδ. PKCδ has been shown to undergo tyrosine phosphorylation in response to various activators. Tyrosine phosphorylation of PKCδ in response to phorbol 12-myristate 13-acetate and platelet-derived growth factor occurred only in chimeras which contained the PKCδ regulatory domain. Cells transfected with a PKCδ mutant (PKCδ5), in which the five putative tyrosine phosphorylation sites were mutated to phenylalanine, showed markedly diminished tyrosine phosphorylation in response to phorbol 12-myristate 13-acetate and platelet-derived growth factor and normal levels of GS. Our results indicate that the regulatory domain of PKCδ mediates the inhibitory effect of this isoform on the expression of GS. Phosphorylation of PKCδ on tyrosine residues in the regulatory domain is implicated in this inhibitory effect.

receptor for the potent tumor-promoting phorbol esters, which can substitute for DAG in PKC activation (5). The "classical" cPKCs ␣, ␤1, ␤2, and ␥ are Ca 2ϩdependent and PMA-responsive. The "novel" nPKCs ␦, ⑀, , and are Ca 2ϩ -independent but PMA-responsive, whereas the atypical PKC isoforms and do not depend on Ca 2ϩ or respond to PMA.
PKC␦, a member of the novel PKCs, has been associated with proliferation and differentiation of various cell types (6,7). For example, overexpression of PKC␦ inhibited the proliferation of fibroblasts and induced monocytic differentiation of the myeloid progenitor cell line 32D (8). PKC␦ has also been shown to be downstream in the signaling pathway of the PDGF receptor (9). PKC␦ has been reported to undergo tyrosine phosphorylation in response to various stimuli such as PMA, epidermal growth factor, and PDGF (10 -12). The phosphorylation site(s) and the role of tyrosine phosphorylation of PKC␦ in its activity and in its functional role are just beginning to be defined.
All PKC isoforms consist of an N-terminal regulatory domain and a C-terminal catalytic domain with serine-threonine kinase activity (1). In the classical PKC isoforms the regulatory domain contains a Ca 2ϩ -binding domain, and in both the classical and novel PKC isoforms the regulatory domain contains a pseudosubstrate region near the N terminus that is thought to inhibit the activity of the catalytic domain (13)(14)(15). The regulatory domain also contains a pair of highly conserved zinc fingers that bind phorbol esters (16). One of the approaches for dissecting the role of the regulatory and catalytic domains of different PKC isoforms is the use of PKC chimeras between the regulatory and catalytic domains of the different isoforms.
Glutamine synthetase (GS) is an important enzyme in the conversion of the excitatory amino acid glutamate to glutamine (17). GS has been used as an astrocytic marker, and its regulation has been studied especially with regard to cAMP (18) and glucocorticoids (19). A recent study suggested an important role for GS as a mediator of interactions between neurons and glial cells under pathological conditions (20). Thus, understanding the factors or mechanisms involved in the regulation of this enzyme has significant implications in understanding pathological processes in the central nervous system.
Recently, we reported that overexpression of PKC␦ in the C6 glial cell line reduced the expression of the astrocytic marker GS, whereas overexpression of PKC␣ and PKC⑀ did not significantly affect the expression of this protein (21). We have now characterized the inhibitory effect of PKC␦ on GS expression using chimeras between the regulatory and catalytic domains of PKC␦, ␣, and ⑀. We found that the regulatory domain of PKC␦ mediates its inhibitory effect on GS expression. Furthermore, tyrosine phosphorylation in the regulatory domain of PKC␦ is implicated in this inhibition. anti-glutamine synthetase antibody, monoclonal anti-GFAP antibody, and anti-PKC␣, ␤, ␥, ␦, , , , and were  obtained from Transduction Laboratories (Lexington, KY). Protein A/G-Sepharose was purchased from Santa Cruz Biotechnology Inc. (Santa Cruz, CA). An affinity purified polyclonal anti-PKC⑀ antibody against a polypeptide corresponding to amino acids 726 -737 of PKC⑀ was purchased from Life Technologies, Inc. Leupeptin, aprotinin, phenylmethylsulfonyl fluoride, and sodium vanadate were obtained from Sigma. Polyclonal rabbit antibody against GFAP was purchased from DAKO Corp. (Carpinteria, CA).

Materials-Monoclonal
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. (22). 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 mammalian ⑀-epitope tagging mammalian expression vector MTH described in detail by Olah et al. (23). The vector was attached to the end of the C-terminal 12-amino acid tag, originally derived from the Cterminal sequence of PKC⑀. The ⑀-tag or the overexpression did not affect the localization, translocation, or phorbol ester responsiveness of the wild type constructs relative to the respective endogenous isoforms (22,24).
Site-directed Mutagenesis of PKC␦-Mouse PKC␦ was subcloned into the pGEM-T vector (Promega, Madison, WI) as described previously (25). 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, 155, and 565 into phenylalanine was performed as described (25). The additional mutations of tyrosine residues at positions 64 and 187 to phenylalanine were introduced into the triple tyrosine-phenylalanine mutant of PKC␦. The following oligonucleotides served as primers for selection and mutagenesis. The selection primer CGCCTGCAGGTC-GACCATATGGGAGAG was used to change back the formerly introduced ClaI restriction site (25) in the pGEM-T vector in position 75 into the underlined SalI restriction site to facilitate selection. The primer ACGCCCACATCTTCGAAGGCCGTGTTATC was used to mutate the tyrosine residue at site 64 into phenylalanine (bold letters) and to create a BstBI restriction site (underlined) to facilitate the detection of mutant plasmids. The primer CTCAACAAGCAAGGCTTTAAATGCAGGCAA-TGC was used to mutate the tyrosine residue at site 187 into phenylalanine (bold letters) and to create a DraI restriction site (underlined) to facilitate the detection of the mutant plasmids. The mutations were confirmed by direct sequencing (Paragon Biotech Inc., Baltimore, MD). PKC␦ and PKC␦5 were subcloned into a metallothionein promoterdriven 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 (23).
C6 Glial Cultures and Cell Transfection -C6 cells of early passage (C6 -30 cells) were used in this study. Conditions for growth and transfection were as described previously (21). Experiments were routinely carried out on pools of transfected cells, but all the results were confirmed on two different individual clones for each of the different isoforms and chimeras.
Preparation of Cell Homogenates and Immunoblot Analysis-Preparation of lysates from cells and the analysis of GS and of PKC by Western blotting were performed as described previously (21,26).
[ 3 H]PDBu Binding-[ 3 H]PDBu binding was measured using the polyethylene glycol precipitation assay (27). Briefly, cell lysates (4 -60 g of protein/assay) were incubated with 20 nM [ 3 H]PDBu in the presence of phosphatidylserine. Nonspecific binding, determined in the presence of 30 M nonradioactive PDBu, was subtracted to give specific binding. Data represent triplicate determinations in each experiment.
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 (27).
Immunoprecipitation-Immunoprecipitation of PKC␦ was performed as described previously (22). Briefly, C6 cells overexpressing PKC␦ or PKC␦5 were serum-starved overnight and treated for 30 min with PMA (10 nM) or with 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 anti-PKC⑀ antibody and 30 l of protein A/G-Sepharose at 4°C. Following washes, the pellets were resuspended in 25 l of SDS sample buffer and boiled for 5 min. Before SDS-PAGE, samples were centrifuged again as described above, and all of each supernatant was subjected to Western blotting. Membranes were probed with anti-phosphotyrosine antibody or anti-PKC␦ antibody.

Differential Effects of PMA and Bryostatin 1 on GS and PKC␦ Expression-Recently
, we have reported that overexpression of PKC␦ reduced the level of GS expression in C6 cells, whereas it did not affect the expression of the astrocytic marker GFAP (21). To further examine the role of PKC␦ in this effect we utilized the two PKC activators PMA and bryostatin 1 because these two activators have been reported to differentially affect the activation and down-regulation of PKC␦ in other cell types (26,29). For these studies we used the C6 -30 cells because they express low levels of GS and GFAP and can be differentiated to either oligodendrocytes or astrocytes (30). Treatment of the cells with PMA (100 nM) for 24 h did not induce significant changes in the expression of GS, as described previously. In contrast, bryostatin 1 (100 nM) induced a marked decrease in the expression of GS (Fig. 1A), whereas it did not affect the expression of GFAP (data not shown). As illustrated in Fig tion by PMA and bryostatin 1 was observed for PKC␥, , and , which are weakly expressed in these cells (data not shown). As has been previously described for keratinocytes (31), co-treatment of the C6 cells with PMA and bryostatin 1 protected PKC␦ from down-regulation by PMA. Likewise, under these conditions GS expression was similar to that of bryostatin-treated cells (Fig. 1A). The results suggest an inverse relation between the expression of PKC␦ and GS. To further explore this point, we examined the effect of bryostatin on GS expression in cells already depleted of PKC␦. Cells were treated for 24 h with 100 nM PMA followed by treatment with bryostatin 1 for 48 h. In these cells, PKC␦ remained down-regulated, and, correspondingly, bryostatin 1 did not reduce the expression of GS (Fig. 1C).
Overexpression of PKC Chimeras-To further characterize the role of PKC␦ in the inhibition of GS expression, we 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 chimeras were engineered in a way that allowed similar construction of the wild type PKC isoforms, thus providing us with exact positive controls (22). The presence of the ⑀-tag on the different constructs allowed better detection on Western blots and provided a good epitope for immunoprecipitation. Cells were transfected with the different chimeras, PKC ␣/␦, ␦/␣, ␣/⑀, ⑀/␣, ␦/⑀, and ⑀/␦ and with the PKC wild type isoforms PKC␣/␣, ␦/␦, and ⑀/⑀. To examine the level of protein expression, we analyzed by Western blotting three pooled cultures and five different overexpressing clones for each of the chimeras as well as for the vector controls. Fig. 2 illustrates a representative Western blot of C6 cells overexpressing the different PKC chimeras and the vector control. Using the previously described tagging system (23), we were able to detect specifically the transfected PKC isoforms with the antibody against PKC⑀. The band corresponding to the endogenous PKC⑀ can also be seen. To establish that the overexpressed PKC chimeras were functionally active we measured specific [ 3 H]PDBu binding and kinase activity on cell lysates. All PKC chimeras exhibited increased [ 3 H]PDBu binding and kinase activity as compared with the control cells transfected with the empty vector (Table I). Moreover, binding and kinase activity were further enhanced in cells treated for 24 h with 20 M ZnCl 2, an inducer of the metallothionein expression vector (data not shown).
GS Levels in Cells Overexpressing PKC Chimeras-The effect of the PKC chimeras on the level of GS was examined. In cells overexpressing PKC␦/␦, the level of GS was low as compared with control cells and similar to that reported previously for cells overexpressing wild type PKC␦. In contrast, the level of GS was unchanged in cells overexpressing PKC␣/␣ and ⑀/⑀ (Fig.  3A) or in cells overexpressing PKC␣/⑀ and ⑀/␣ (data not shown).
Similar to cells expressing PKC␦/␦, cells overexpressing the chimeras containing the regulatory domain of PKC␦, namely ␦/␣ and ␦/⑀, also showed a reduced level of GS (Fig. 3A). Cells expressing PKC ␦/␣ showed a greater reduction in the level of GS than cells expressing PKC␦/⑀. The effect of the regulatory domain of PKC␦ was specific for GS because no differences were observed in the expression of the other astrocytic marker, GFAP (data not shown). In contrast to the chimeras containing the regulatory domain of PKC␦, chimeras containing just the catalytic domain of PKC␦ did not have a significant effect on GS expression. Thus, cells expressing PKC␣/␦ and ⑀/␦ expressed   GS levels similar to those of cells expressing PKC␣/␣ or PKC⑀/⑀ (Fig. 3A).
Treatment of the cells with 10 nM PMA, which induces activation of PKC without the marked down-regulation that occurs at 100 nM PMA, enhanced the response observed in untreated cells (Fig. 3A). Thus, PMA further reduced the expression of GS in cells overexpressing PKC ␦/␦ or the chimeras ␦/␣ and ␦/⑀, which contain the ␦ regulatory domain. Similar to what was observed in the vector control cells, transfectants expressing PKC ␣/␣ or ⑀/⑀ or chimeras that contain the ␦ catalytic domain showed a smaller decrease in GS expression presumably mediated by the endogenous PKC␦. Similar results were obtained in transfectants expressing PKC ␣/⑀ and ⑀/␣ (data not shown).
We also examined the expression of GS in cells treated for 24 h with 20 M ZnCl 2 . Under these conditions, cells overexpressing PKC␦/␦ or chimeras containing the regulatory domain of PKC␦ showed a larger decrease in GS expression as compared with PMA-treated cells, whereas cells expressing chimeras that contain the ␦ catalytic domain had levels of GS similar to those of the control vector cells (Fig. 3B).
Tyrosine Phosphorylation of PKC␦ Occurs in the Regulatory Domain-The translocation and activation of PKC␦ in response to PMA and physiological stimuli such as epidermal growth factor and PDGF have been shown to be associated with tyrosine phosphorylation of this isoform (10 -12). However, the sites of phosphorylation and the role of this process in the activity and function of PKC␦ are still unclear.
We first examined which domain is involved in the tyrosine phosphorylation of PKC␦. Cells overexpressing PKC␦/␦ or the various chimeras containing either the regulatory or the catalytic domains of PKC␦ were stimulated with PMA and PDGF. Following immunoprecipitation of PKC␦ and Western blotting, membranes were probed with anti-phosphotyrosine antibody or with anti-PKC␦ antibody. As can be seen in Fig. 4, PKC␦/␦, PKC␦/␣, and PKC␦/⑀ underwent tyrosine phosphorylation in response to PMA. In contrast, no tyrosine phosphorylation was observed in cells overexpressing PKC␣/␦ or PKC⑀/␦, thus suggesting that tyrosine phosphorylation in response to these stimuli occurred on the regulatory domain of the enzyme. (25) and in vivo (12) suggested five putative phosphorylation sites in PKC␦, one in the catalytic domain (Tyr 565 ) and four in the regulatory domain (Tyr 52 , Tyr 64 , Tyr 155 , and Tyr 187 ). Because four of the putative phosphorylation sites in PKC␦ are located in the regulatory domain, we examined the role of tyrosine phosphorylation of PKC␦ in its inhibitory effect on GS expression.

Overexpression of a PKC␦ Mutant That Lacks Five Tyrosine Residues-Studies in vitro
For these experiments we utilized a PKC␦ mutant in which the five putative tyrosine phosphorylation sites were mutated to phenylalanine (␦5 mutant). Cells were transfected with PKC␦ or PKC␦5, and the level of expression was determined using Western blot analysis. Using the previously described tagging system (23), we were able to detect specifically the transfected PKC isoforms with the antibody against PKC⑀. Fig.  5A illustrates a representative Western blot of C6 cells overexpressing PKC␦, PKC␦5, and the vector control. The levels of the overexpressed PKC␦ and PKC␦5 in the transfected cells were about 7-9-fold higher than the endogenous PKC␦ as determined using an anti-PKC␦ specific antibody (data not shown). The membranes were probed with anti-PKC⑀ antibody that recognizes the ⑀-tag (A). C6 cells stably transfected with either PKC␦ or the PKC␦5 were treated with PMA (10 nM) or PDGF (100 ng/ml). Immunoprecipitation of PKC␦ was performed following 30 min of treatment, and membranes were stained with anti-phosphotyrosine (anti-PTyr) and anti-PKC␦ antibodies (B). After 48 h of treatment, cells were harvested, and the level of GS was determined using Western blotting (C). Results are from one representative experiment; similar results were obtained in each of five additional experiments. cells, whereas cells overexpressing PKC␦5 showed an increase of only 87 Ϯ 11% over control cells. The transfected PKC␦ and ␦5 displayed similar localization in untreated cells and translocated similarly to the membrane in response to PMA (data not shown).
Tyrosine Phosphorylation of PKC␦ Plays a Role in the Inhibitory Effect on GS Expression-We first examined the degree of tyrosine phosphorylation of PKC␦ and PKC␦5 in response to PMA and PDGF. Cells were treated with either compound for 30 min, PKC␦ was immunoprecipitated using the ⑀-tag specific antibody, and membranes were stained with anti-phosphotyrosine antibody. As illustrated in Fig. 5B, PMA and PDGF induced marked tyrosine phosphorylation of PKC␦ within 5-30 min. In contrast, these compounds induced negligible tyrosine phosphorylation of the PKC␦5 mutant. We then examined the level of GS in cells that were overexpressing PKC␦ or PKC␦5 and were either untreated or treated with PMA (10 nM) or PDGF (100 ng/ml). PMA (10 nM) activates PKC without the marked down-regulation that occurs at a concentration of 100 nM. We found that cells overexpressing the PKC␦5 mutant displayed a slightly higher level of GS compared with the control vector cells and markedly higher levels compared with cells overexpressing PKC␦ wild type. Treatment of cells overexpressing PKC␦ with PMA (10 nM) or PDGF (100 ng/ml) induced a marked decrease in GS expression. In contrast, similar treatments of cells expressing the ␦5 mutant did not exert a significant effect on GS expression (Fig. 5C).

DISCUSSION
Protein kinase C comprises a family of eleven closely related isoforms that are involved in the differentiation and proliferation of various cell types (1)(2)(3). In a recent study, we have shown differential roles of specific PKC isoforms in the proliferation and differentiation of glial cells (21). Specifically, PKC␣ and PKC␦ inhibited astrocytic differentiation by reducing the expression of the astrocytic markers GFAP and GS, respectively. In the present study we have explored the mechanisms involved in the inhibitory effect of PKC␦ on GS expression.
We found that the regulatory domain of PKC␦ is responsible for the inhibitory effect of this isoform on the expression of GS. Chimeras have been used to dissect the contribution of individual PKC domains to the specific functions of different PKC isoforms, and both the catalytic and the regulatory domains of PKC have been implicated in determining isoform-specific functions. For example, the catalytic domain of PKC␤ was found to confer isoform-specific function in the differentiation of erythroleukemia cells (32), and the catalytic domain of PKC␦ in reciprocal ␦and ⑀-chimeras mediated PMA-induced macrophage differentiation of mouse promyelocytes (33). In contrast, the regulatory domain of PKC⑀ enhanced cell growth and induced colonies in soft agar in NIH 3T3 cells, and the catalytic domain played a role in determining the saturation density of cell growth (34). Thus, it appears that either the regulatory or the catalytic domain can confer isoform-specific effects and that this depends on the cell type, the specific PKC isoform and the specific function studied.
PKC␦ has been shown to become tyrosine phosphorylated in response to various stimuli. In NIH 3T3 cells, PKC␦ is tyrosine phosphorylated upon PMA and PDGF stimulation (10,12), in response to carbachol, substance P, and PMA stimulation in parotid acinar cells (35) and through activation of the IgE receptor in RBL-2H3 cells (36). Tyrosine phosphorylation of PKC␦ was also observed in response to epidermal growth factor stimulation in keratinocytes (11) and in ras-transformed keratinocytes (37). It appears that the sites of tyrosine phosphorylation and the effect of this phosphorylation on the activity of PKC␦ or on its function depend somewhat on the specific sys-tem. PKC␦ contains 19 tyrosines located in both the regulatory and catalytic domains, and different tyrosines have been implicated as possible phosphorylation sites. Thus, Tyr 187 has been shown to be phosphorylated in 32D and NIH 3T3 cells in response to PMA or PDGF stimulation (12). However, a mutation in this tyrosine failed to alter either the activity of PKC␦ or the ability of PKC␦ to induce monocytic differentiation (12). In another study, Tyr 52 was suggested to mediate the tyrosine phosphorylation of PKC␦ upon stimulation of the IgE receptor in RBL-2H3 cells but not upon stimulation with PMA (25).
We found, similar to what has been recently described by Acs et al. in fibroblasts (34), that tyrosine phosphorylation of PKC␦ was confined to the regulatory domain of this isoform, suggesting that the tyrosines phosphorylated in response to PMA and PDGF are located on this domain. The role of the tyrosine phosphorylation of PKC␦ in its inhibitory effect on GS expression was examined using a PKC␦ mutant in which five putative sites of tyrosine phosphorylation were mutated to phenylalanine (␦5 mutant). This mutant was constructed based on in vitro assessment of the activity as tyrosine kinase substrates of the oligopeptides corresponding to the sequences around each of the 19 tyrosines in PKC␦ (25) as well as on in vivo studies (12). Our results indicate that the ␦5 mutant had a markedly diminished tyrosine phosphorylation in response to PMA and PDGF, indicating that at least some of these tyrosines are functional phosphorylation sites in this system. In addition, the cells overexpressing the PKC␦5 mutant showed normal levels of GS as compared with cells overexpressing PKC␦. These results suggest that tyrosine phosphorylation of PKC␦ plays a role in the inhibitory effect exerted by this isoform on GS expression. We found that although treatment of cells overexpressing PKC␦5 with PMA induced a similar pattern of translocation to that of PKC␦, the kinase activity of these cells was significantly lower, at least as evaluated using myelin basic protein as a substrate. It is currently not clear what is the effect of the tyrosine phosphorylation of PKC␦ on the activity of the enzyme toward its different substrates in this system. Different effects have been reported in different systems, suggesting changes in substrate specificity rather than simply in level of absolute activity (25,38).
The astrocytic glutamine synthetase is an important enzyme in the removal of the toxic substances NH 3 and glutamate and in the conversion of these substances into glutamine (39). Changes in GS expression have been associated with human pathology (40), and GS has been reported to protect against neuronal degeneration following trauma or ischemia (20). GS expression has been reported to be regulated by cAMP, glucocorticoids, and insulin (18,19,41). GS mRNA is regulated by the glucocorticoid response element, NF-B, and by contact with neuronal cultures (40,42,43). Because GS has also been used as an astrocytic marker, understanding of its regulation provides important information on astrocytic differentiation as well as on pathological conditions in the central nervous system. Our results imply a negative role for PKC␦ in GS expression, with a modulatory effect via tyrosine phosphorylation. A possible mechanism by which PKC␦ reduces the expression of GS may be through the induction of c-Jun, which has been implicated in negative regulation of GS transcription (42).
In conclusion, our results demonstrate that the regulatory but not the catalytic domain of PKC␦ determines the inhibitory effect of this isoform on the expression of GS. Our findings further emphasize a potentially important role of tyrosine phosphorylation of PKC␦ in this system. It is currently not clear which tyrosine kinases are involved in this process; however, tyrosine kinases of the Src family have been implicated in both the tyrosine phosphorylation of PKC␦ and the function of glial cells (28). Studies are currently underway to identify the specific tyrosine(s) involved in the phosphorylation of PKC␦ in this system and the tyrosine kinases that phosphorylate it at these sites.