An inhibitory fragment derived from protein kinase Cepsilon prevents enhancement of nerve growth factor responses by ethanol and phorbol esters.

We have studied nerve growth factor (NGF)-induced differentiation of PC12 cells to identify PKC isozymes important for neuronal differentiation. Previous work showed that tumor-promoting phorbol esters and ethanol enhance NGF-induced mitogen-activated protein (MAP) kinase activation and neurite outgrowth by a PKC-dependent mechanism. Ethanol also increases expression of PKCdelta and PKCepsilon, suggesting that one these isozymes regulates responses to NGF. To examine this possibility, we established PC12 cell lines that express a fragment encoding the first variable domain of PKCepsilon (amino acids 2-144), which acts as an isozyme-specific inhibitor of PKCepsilon in cardiac myocytes. Phorbol ester-stimulated translocation of PKCepsilon was markedly reduced in these PC12 cell lines. In addition, phorbol ester and ethanol did not enhance NGF-induced MAP kinase activation or neurite outgrowth in these cells. In contrast, phorbol ester and ethanol increased neurite outgrowth and MAP kinase phosphorylation in cells expressing a fragment derived from the first variable domain of PKCdelta. These results demonstrate that PKCepsilon mediates enhancement of NGF-induced signaling and neurite outgrowth by phorbol esters and ethanol in PC12 cells.

Protein kinase C (PKC) 1 is a multigene family of phospholipiddependent, serine-threonine kinases that plays a central role in cell growth and differentiation. Molecular cloning studies have identified 10 isozymes encoded by 9 different mRNAs (1,2). Based on sequence homology and biochemical properties, the PKC gene family has been divided into three groups: "conventional" PKCs (␣, ␤I, ␤II, and ␥) regulated by calcium and diacylglycerols or phorbol esters; "novel" PKCs (⑀, ␦, , and ), which are calcium-independent but diacylglycerol-and phorbol ester-sensitive; and "atypical" PKCs (, and /), which are insensitive to calcium, diacylglycerol, and PMA. In addition, two related phospholipid-dependent kinases, PKC and protein kinase D, share sequence homology in their regulatory domains to novel PKCs and may constitute a new subgroup (3,4).
Recent evidence suggests that PKC⑀ plays a role in neural differentiation and plasticity. PKC⑀ is expressed predominantly in the nervous system and is particularly abundant in the hippocampus, olfactory tubercle, and layers I and II of cerebral cortex (19). Within immunoreactive neurons, it is localized to the Golgi apparatus and to axons and presynaptic nerve terminals (19). PKC⑀ is activated by growth factors that stimulate neural differentiation such as insulin (20) and NGF (21). In addition, in developing chick brain, it is the major isozyme found in nondividing, differentiating neurons (22).
Further evidence for involvement of PKC⑀ in neural differentiation has come from studies with PC12 cells. PC12 cells are derived from neural crest and, when treated with NGF or fibroblast growth factors, undergo dramatic biochemical and morphological differentiation, developing several characteristics of mature sympathetic neurons (23). PKC-activating phorbol esters enhance NGF-induced activation of ERK1 and ERK2 mitogen-activated protein (MAP) kinases and neurite outgrowth in PC12 cells, suggesting that PKC modulates responses to NGF (11,12,24). Studies with ethanol-treated PC12 cells helped direct us toward the PKC isozyme responsible for this effect. Like phorbol esters, ethanol increases NGF-induced MAP kinase activation and neurite outgrowth through a PKCdependent mechanism (11,24). Ethanol promotes PKC-mediated phosphorylation in PC12 cells by increasing levels of messenger RNA and protein for two PKC isozymes, PKC␦ and PKC⑀ (25,26). Recently, we found that overexpression of PKC⑀, but not of PKC␦, enhances NGF-induced MAP kinase activation and neurite outgrowth (27). These findings establish PKC⑀ as a positive modulator of neurite growth. They also suggest that PKC⑀ mediates the neurite-promoting effect of ethanol and phorbol esters in PC12 cells. However, proof of this hypothesis requires studies with PKC isozyme-specific inhibitors or cells lacking specific PKC isozymes.
In the current study, we used specific inhibitors of PKC␦ and PKC⑀ to investigate whether PKC⑀ mediates enhancement of neurite outgrowth by phorbol esters and ethanol. To achieve this goal we used dominant-negative inhibitors based on the amino acid sequences for PKC␦ and PKC⑀. This approach is based on the observation that upon activation, PKC isozymes translocate to specific intracellular sites where they appear to bind anchoring proteins, termed RACKs (receptors for activated C-kinase) (28). One such protein that has been cloned is RACK1, which interacts with the C2 domain of conventional PKCs (29). The sites of interaction between RACK1 and the C2 domain of PKC␤ have been mapped, and short peptides derived from these domains inhibit translocation and activation of PKC␤ in cardiac myocytes and Xenopus oocytes (29 -31). Homology has been noted between the unique first variable region of PKC⑀ (⑀V1) and the C2 domain of conventional PKCs (32), suggesting that, similar to the C2 domain of conventional PKCs, ⑀V1 may contain a binding site for an PKC⑀-specific RACK. If that is the case, then expression of an ⑀V1 fragment should inhibit PKC⑀ translocation and function. Indeed, recent work has shown that an ⑀V1 fragment and a peptide corresponding to amino acids 14 -21 in this region prevent phorbol ester-induced PKC⑀ translocation and inhibition of contraction in cultured cardiac myocytes (33).
In this paper, we describe studies with PC12 cell lines that stably express the fragments ⑀V1 or ␦V1, which are derived from the first variable domains of PKC⑀ or PKC␦, respectively. We found that each fragment selectively inhibited phorbol ester-induced translocation of its corresponding isozyme, indicating that these fragments can function as isozyme-selective translocation inhibitors. NGF-induced MAP kinase phosphorylation and neurite outgrowth were not enhanced by phorbol esters or ethanol in cells expressing ⑀V1, but they were increased by these agents in cells expressing ␦V1 and in cells transfected with empty vector. These results demonstrate that PKC⑀ specifically mediates enhancement of MAP kinase activation and neurite growth by phorbol esters and ethanol in PC12 cells.
Cell Culture-PC12 cells, originally obtained from Dr. John A. Wagner (Cornell University, New York, NY), were cultured in plastic tissue culture flasks at 37°C in Dulbecco's modified Eagle's medium containing 10% heat-inactivated horse serum, 5% fetal calf serum, 2 mM glutamine, 50 units/ml penicillin, and 50 g/ml streptomycin, in a humidified atmosphere of 90% air and 10% CO 2 . For studies of neurite outgrowth, cells were plated at a density of 30 -40 ϫ 10 3 cells/well on 24-well plastic culture plates pretreated for 1 h with poly-L-ornithine (100 g/ml in 15 mM sodium borate, pH 8.4). In some experiments, cells were plated at a density of 20 ϫ 10 3 cells/well in 8-well glass chamber slides (Nunc, Naperville, IL) treated first with poly-L-ornithine and then laminin (30 g/ml) overnight. Cells were cultured in medium containing 50 ng/ml of NGF for 4 days, and neurites were measured as described previously (27). Ethanol-treated cultures were wrapped in Parafilm to prevent evaporation of ethanol, as in prior studies (24,25).
Immunofluorescence Microscopy-To detect PKC⑀ immunoreactivity, cells plated on glass chamber slides were incubated in PBS (137 mM NaCl, 2.7 mM KCl, 1.47 mM KH 2 PO 4 8 mM Na 2 HPO 4 , 0.5 mM MgCl 2 , 0.9 mM CaCl 2 , pH 7.2) containing 2% paraformaldehyde for 30 min and 4% paraformaldehyde for 30 min at 4°C. Cells were washed three times in PBS and incubated for 2 h in PBS containing 1% normal goat serum and 0.1% Triton X-100. Cells were incubated 48 h at 4°C in PBS containing 2 mg/ml of bovine serum albumin, 0.1% Triton X-100, and 2 g/ml of rabbit anti-PKC⑀ antibody (34) provided by Dr. Susan C. Kiley (W. Alton Jones Cell Science Center, Lake Placid, NY). Cells were washed three times in PBS, and immunoreactivity was detected using fluorescein-conjugated goat anti-rabbit IgG as described previously (27). Images were detected using a liquid-cooled CCD camera (Photometrics Ltd., Tucson, AZ) fitted with a Thompson 7883 chip (384 ϫ 576 pixels) attached to an Olympus IMT-2 inverted microscope equipped with a ϫ 60 1.3 numerical aperture objective. Exposure times were 0.5 s, and images were stored on an Apple Macintosh Quadra 950 computer. Fluorescence intensity in growth cones and cytoplasm was measured using the program BDS Image (Oncor Imaging Systems, Gaithersburg, MD).
To detect PKC␦ immunoreactivity, cells were incubated with rabbit polyclonal antibody against PKC␦ (0.2 g/ml) from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA) as described previously (27). Immunofluorescence in 0.5-m optical sections was detected with a Bio-Rad MRC 1024 confocal laser-scanning microscope equipped with a Nikon ϫ 60, 1.4 numerical aperture oil immersion objective. Nuclei were stained by incubating cells in PBS containing 2 mg/ml of bovine serum albumin, 0.1% Triton X-100, and 0.2 M TOTO-3 (Molecular Probes, Eugene, OR) for 2 h at 25°C. After three washes in PBS, slides were dried, and coverslips were mounted with Vectashield (Vector, Burlingame, CA). TOTO-3 immunofluorescence was detected by confocal laser-scanning microscopy. Images were analyzed in BDS Image to measure the area of each nucleus at its widest diameter.
Generation of Cell Lines-The plasmid pDM27 (33), containing a Flag-epitope tag followed by the sequence encoding amino acids 2-144 of PKC⑀ was used to amplify a 480-base pair fragment containing a NotI site at the 5Ј-end and a XbaI site at the 3Ј-end. This amplified fragment, containing an ATG start codon, the Flag epitope sequence, and the PKC⑀ sequence, was subcloned into the NotI and XbaI sites of pRc/RSV (Invitrogen, San Diego, CA) to generate the plasmid pR⑀V-1. Another plasmid, pDM68 (33), containing the Flag epitope followed by the sequence encoding amino acids 2-144 of PKC␦ was used to amplify and subclone a homologous Flag-tagged PKC␦ sequence into pRc/RSV to generate the plasmid pR␦V-1. PC12 cells (10 7 ) were suspended in 0.5 ml of Ca 2ϩ -and Mg 2ϩ -free PBS containing 137 mM NaCl, 2.7 mM KCl, 1.47 mM KH 2 PO 4 , 8 mM Na 2 HPO 4 , pH 7.2, and were electroporated with 80 g of pR⑀V-1, pR␦V-1, or pRc/RSV as described previously (27). Geneticin was initially added at 400 g/ml to select clones and later was added at 200 g/ml to maintain cultures. For each vector, 46 clones were selected, expanded, and then examined for expression of Flagtagged PKC fragment mRNA using RT-PCR and for Flag immunoreactivity by Western analysis.
RT-PCR-Poly(A) mRNA was isolated from 10 6 cells using a Micro-Fast Track mRNA isolation kit (Invitrogen, San Diego, CA). Reverse transcription was carried out with 100 ng of mRNA using a Stratagene (La Jolla, CA) RT-PCR kit according to the manufacturer's protocol. Amplification of cDNA was achieved using the forward primer (5Ј-ACACTGGCGGCCGCATGGACTACAAGGACGACGAT-3Ј) and the reverse primers 5Ј-AGCGAGCTCTAGATCGTTCTTCATTGTCTTTA-3Ј for pR⑀V-1, and 5Ј-ACAGACCTCTAGAGCGGTTCATAGTTGGGAA-3Ј for pR␦V-1. These primers were designed to specifically amplify the Flag-V1 sequences and not the V1 sequences of endogenous PKC␦ and PKC⑀ in cells. Samples were heated to 94°C, and amplification was started by the addition of Taq DNA polymerase. The amplification cycle was as follows: annealing for 45 s at 48°C; elongation for 2 min at 72°C, and denaturation for 1 min at 94°C. Amplification was repeated for 30 cycles.
Western Analysis for Flag and Phospho-MAP Kinase Immunoreactivity-To detect expression of Flag-tagged peptides, cells were cultured on poly-L-ornithine-coated, 100-mm tissue culture dishes at a density of 6 ϫ 10 6 cells/dish. Medium was removed, and cells were rinsed twice at 4°C with buffer A (120 mM NaCl, 5 mM KCl, 1.4 mM CaCl 2 , 0.8 mM MgSO 4 , 1 mM NaH 2 PO 4 , 10 mM glucose, 25 mM HEPES, pH 7.4). Cells were scraped into 1 ml of buffer A containing 40 g/ml leupeptin, 40 g/ml soybean trypsin inhibitor, and 1 mM phenylmethylsulfonyl fluoride and then were frozen on dry ice. Concentrated 5 ϫ sample buffer was added to 400-l frozen samples to yield a final solution containing 62.5 mM Tris-HCl (pH 6.8), 2% SDS, 10% glycerol, 5% 2-mercaptoethanol, and 12.5 mg/ml bromphenol blue. Samples were heated at 90°C for 10 min, passed five times through a 26-gauge needle, and centrifuged at 10,000 ϫ g for 10 min. Samples (80 g/lane) were separated by SDS-polyacrylamide gel electrophoresis using 14% gels. Proteins were electrophoretically transferred for 2 h at 4°C to Hybond-C extra membranes (Amersham Corp.). Membranes were blocked for 1 h with 3% nonfat dry milk dissolved in Tris-buffered saline (TBS; 20 mM Tris HCl, pH 7.4, 137 mM NaCl). Blots were then incubated with anti-Flag M2 antibody (10 g/ml) for 2 h at 25°C. Blots were washed three times with TBS containing 0.05% Tween-20 (TBS-T) for two minutes and then were incubated with goat anti-mouse IgG-peroxidase-conjugated antibody (1:1000 dilution) in blocking solution overnight at 25°C. Blots were washed three times for 15 min in TBS-T and once with TBS. Immunoreactive bands were detected with the ECL kit from Amersham.
Activation of ERK1 and ERK2 MAP kinases was assayed with a phospho-specific 42/44-kDa MAP kinase rabbit polyclonal antibody raised against a phosphotyrosine peptide corresponding to residues 196 -209 of human ERK1. The antibody detects phosphorylation of ERK1 at tyrosine 204, which is required for ERK1 activation (35). The antibody also detects phosphorylation of the corresponding activating tyrosine of ERK2. Blots from 11% polyacrylamide gels were washed for 5 min with 25 ml of buffer B containing 58 mM Na 2 HPO 4 , 17 mM NaH 2 PO 4 , and 68 mM NaCl, pH 7.4. Membranes were blocked in buffer B containing 0.1% Tween 20 and 5% milk (blocking buffer) for 1 h. Blots were incubated overnight at 4°C with 1 g/ml of anti-phospho-MAP kinase antibody in buffer B containing 0.05% Tween 20 and 5% bovine serum albumin (incubation buffer). They were then washed three times for 5 min with 15 ml of blocking buffer and incubated with goat antirabbit alkaline phosphatase-conjugated antibody (1:2000 dilution) in incubation buffer for 1 h at 25°C. Blots were finally washed three times with 15 ml of blocking buffer, and immunoreactive bands were detected using the Western-Light chemiluminescent detection system from Tropix, Inc. (Bedford, MA).
PKC Translocation-Cells (3-4 ϫ 10 6 ) were plated on 100-mm plastic tissue culture plates. After 48 h, cultures were rinsed at 37°C twice with 10 ml of medium and incubated with or without 30 nM PMA for 2 min. Cells were rapidly rinsed twice at 4°C with Ca 2ϩ -and Mg 2ϩ -free PBS and then were scraped into 1 ml of ice-cold buffer C containing 20 mM Tris HCl, pH 7.5, 2 mM EDTA, 10 mM EGTA, 40 g/ml leupeptin, 40 g/ml aprotinin, 20 g/ml soybean trypsin inhibitor, and 1 mM phenylmethylsulfonyl fluoride. Cells were homogenized at 4°C with 10 strokes of a Teflon-glass homogenizer. Sucrose was added to a final concentration of 250 mM, and the sample was homogenized with 10 additional strokes. An aliquot of 800 l was centrifuged at 150,000 ϫ g for 1 h, and the supernatant was frozen on dry ice. The pellet was dispersed in 800 l of buffer C by sonication in a Branson Sonifier 450 for 2 s at a setting of six. Samples of supernatant and pellet suspension derived from 100 g of crude homogenate were separated by SDS-polyacrylamide gel electrophoresis using 10% gels and analyzed for PKC␦ and PKC⑀ immunoreactivity by Western analysis as described previously (27).
Miscellaneous Procedures-Protein concentrations were measured by the Bradford method (36) using bovine IgG standards. Results are expressed as mean Ϯ S.E. values, and differences between means were analyzed by ANOVA. Where p Ͻ 0.05, the significance of differences between means was evaluated by the Scheffe F-test or the Newman Keuls test.
To determine whether the ␦V1 and ⑀V1 fragments expressed by our PC12 clones act as PKC isozyme-selective inhibitors, we measured PMA-induced translocation of PKC␦ and PKC⑀ to the particulate fraction of these cells. We predicted that if ␦V1 contains a binding site for a PKC␦-specific RACK, it should inhibit translocation of PKC␦. Likewise, if ⑀V1 interacts with an PKC⑀-specific RACK, then expression of ⑀V1 should block phorbol ester-mediated translocation of PKC⑀. In the parent PC12 cell line and in cells transfected with the pRc/RSV vector alone (clone C1), 30 nM PMA stimulated translocation of PKC␦ and PKC⑀ to the particulate fraction (Fig. 2). In V1␦1 and V1␦2 cells, which express the ␦V1 peptide, 30 nM PMA stimulated translocation of PKC⑀, but translocation of PKC␦ was reduced in these clones (Fig. 2, A and B). In V1⑀1 and V1⑀2 cells, 30 nM PMA stimulated translocation of PKC␦, but translocation of PKC⑀ was reduced compared with control cells (Fig. 2, A and  C). These results demonstrate that ␦V1 and ⑀V1 fragments selectively inhibit PMA-induced translocation of their corresponding PKC isozymes.
PKC Immunoreactivity in Cells Expressing V1 Fragments-We attempted to examine cells by indirect immunofluorescence with anti-Flag antibody to determine the subcellular localization of expressed ␦V1 and ⑀V1 but were unsuccessful because of high background staining. We next examined the subcellular localization of endogenous PKC␦ and PKC⑀ to determine whether their localization was altered by expression of ␦V1 or ⑀V1. In undifferentiated cells, PKC␦ and PKC⑀ immunoreactivity was observed throughout the cytoplasm, as previously reported (27) and was not altered in clones expressing ␦V1 or ⑀V1 (data not shown). In PC12 and C1 cells treated with 50 ng/ml of NGF for 4 days, most PKC␦ immunoreactivity was observed in the cytosol, asymmetrically next to the nucleus (Fig. 3). Less intense staining was seen in a thin perinuclear band and within nuclei. In the majority of V1␦1 and V1␦2 cells expressing ␦V1, most PKC␦ immunoreactivity appeared in the perinuclear region (Fig. 3). In addition, nuclei appeared larger in these cells. This was analyzed further using confocal images of TOTO-3-stained nuclei at their widest diameter. Nuclei were of similar size in PC12 (82.8 Ϯ 1.5 m 2 ; n ϭ 164) and C1 (88.3 Ϯ 2.1 m 2 ; n ϭ 145) cells but were significantly larger in V1␦1 (106.4 Ϯ 3.8 m 2 ; n ϭ 88) and V1␦2 (100.6 Ϯ 2.8 m 2 ; n ϭ 172) cells (p Ͻ 0.05 compared with PC12 and C1 nuclei; ANOVA and Scheffe-F test).
Following treatment with NGF, PKC⑀ immunoreactivity was observed in growth cones, neurite shafts, and the cytoplasm of the cell soma in PC12 and C1 cells (Fig. 4), as noted previously (27). In V1⑀1 and V1⑀2 cells expressing ⑀V1, PKC⑀ immunoreactivity was reduced in growth cones and neurite shafts (Fig.  4). This was examined further by calculating the ratio of mean fluorescence intensity of each growth cone and of the cytoplasm at the base of its neurite shaft. In PC12 (0.73 Ϯ 0.02; n ϭ 71) and C1 (0.76 Ϯ 0.02; n ϭ 55) cells, this ratio was significantly greater (p Ͻ 0.05; ANOVA and Scheffe F-test) than ratios measured in V1⑀1 (0.49 Ϯ 0.01; n ϭ 83) and V1⑀2 (0.50 Ϯ 0.01; n ϭ 71) cells. These results demonstrate that ⑀V1 and ␦V1 fragments alter the localization of their corresponding PKC isozymes in cells undergoing NGF-induced differentiation. In addition, expression of ␦V1 is associated with an increase in nuclear size.
Growth and Neurite Formation in Cell Lines Expressing ␦V1 or ⑀V1-The growth rates of V1␦-and V1⑀-expressing clones, parent PC12 cells, and C1 cells were similar before NGF treatment (data not shown). After culture in poly-L-ornithinetreated culture dishes with 50 ng/ml NGF for 4 days, the number of neurite-bearing cells and the length of neurites was similar in all cell lines (Tables I and II). However, in NGFtreated V1⑀1 and V1⑀2 cells, neither ethanol nor PMA (10 nM) increased neurite length or the percentage of cells that expressed neurites. In contrast, ethanol and PMA increased neurite length and the percentage of neurite-bearing cells in NGFtreated PC12, C1, V1␦1, and V1␦2 cultures. When cells were cultured on glass slides treated with poly-L-ornithine and coated with laminin, the results were qualitatively similar but more dramatic because NGF-induced neurite outgrowth was especially robust following treatment with ethanol or PMA in all but V1⑀1 and V1⑀2 cultures (Fig. 5, A and B). Therefore, expression of the ⑀V1 fragment, but not of the ␦V1 fragment, appears to prevent enhancement of neurite extension by PMA or ethanol in NGF-treated cells.

MAP Kinase Phosphorylation in PKC-transfected Cells-
NGF and basic fibroblast growth factor stimulate sustained activation of ERK1 and ERK2 MAP kinases in PC12 cells, which is important for their neuronal differentiation (37). Previous work has shown that treatment with either PMA or ethanol increases NGF-induced phosphorylation and activation of ERK1 and ERK2 (24). Since overexpression of PKC⑀ also enhances NGF-induced MAP kinase phosphorylation (27), we examined whether expression of ⑀V1 would prevent enhancement of MAP kinase activation by PMA or ethanol.
ERK1 and ERK2 are activated by dual phosphorylation on neighboring tyrosine and threonine residues (35). We measured activation of ERK1 and ERK2 by Western analysis using an anti-phospho-MAP kinase antibody that specifically detects phosphorylation of the activating tyrosine of each enzyme. As described previously in PC12 cells (24), NGF stimulated phosphorylation of ERK1 and ERK2 in C1 cells with a biphasic time course (Fig. 6A). Phosphorylation was maximal after 5-10 min (peak phase) and then declined to a lower level (plateau phase) that was maintained for at least 2 h. A similar pattern of phosphorylation was observed in V1␦1, V1␦2, V1⑀1, and V1⑀2 cells (Fig. 6). As previously observed in PC12 cells (24), cotreatment with 10 nM PMA or pretreatment with 100 mM ethanol for 6 days increased NGF-induced ERK phosphorylation. This was particularly evident during the plateau phase of ERK phosphorylation, which was elevated to levels achieved during the peak phase in C1 and V1␦2 cells (Fig. 6A). A similar increase in phosphorylation was also observed in V1␦1 cells (Fig. 6, B and C). In contrast, the plateau phase of ERK phosphorylation was not increased in V1⑀1 or V1⑀2 cells treated with ethanol or PMA (Fig. 6, A-C). Therefore, expression of the ⑀V1 fragment specifically inhibits PMA-or ethanol-induced ERK phosphorylation in NGF-treated cells. DISCUSSION The current results identify PKC⑀ as the PKC isozyme responsible for enhancement of NGF responses by phorbol esters and ethanol in PC12 cells. We previously found that phorbol esters enhance NGF-induced neurite outgrowth and MAP kinase activation in PC12 cells (11,24). Ethanol also increases these responses to NGF by a PKC-dependent mechanism (11,24). Recently, we found that overexpression of PKC⑀ also enhances responses to NGF (27), suggesting that PKC⑀ mediates the neurite-promoting effect of PMA and ethanol. In this paper, we investigated this issue directly, by creating PC12 cell lines that express peptides encoding V1 domains of PKC␦ or PKC⑀, which act as isozyme-selective translocation inhibitors (33). We found that expression of the ⑀V1 fragment selectively inhibited PMA-induced translocation of PKC⑀, whereas expression of the ␦V1 fragment specifically inhibited translocation of PKC␦. Cells expressing these peptides showed no alterations in cell growth. However, expression of ⑀V1 prevented enhancement of NGF-induced MAP kinase activation and neurite growth by PMA or ethanol. In contrast, expression of ␦V1 did not alter enhancement of NGF responses by these agents. These findings indicate that PKC⑀ mediates enhancement of neurite outgrowth and MAP kinase activation by PMA or ethanol in NGFtreated PC12 cells.
Our data are consistent with a recent study (33) in which a ␦V1 fragment, an ⑀V1 fragment, and an ⑀V1-derived peptide were introduced into cardiac myocytes by transient permeabilization. These fragments selectively inhibited PMA-induced a Significantly different compared with cells treated with NGF alone (ANOVA and Scheffe F-test).

FIG. 4. Indirect immunofluorescence of PKC⑀ in V1⑀-expressing cells. Parent PC12 cells (PC) and cells
transfected with the empty pRc/RSV vector (C1) or pR⑀V-1 (V1⑀1 and V1⑀2) were cultured for 4 days with 50 ng/ml NGF prior to fixation and immunostaining with antibody against PKC⑀. The arrows show prominent PKC⑀ immunoreactivity in growth cones of PC12 and C1 cells and reduced immunoreactivity in growth cones of V1⑀1 and V1⑀2 cells. Bar, 30 m.

TABLE I Percentage of cells expressing neurites after treatment with
NGF plus PMA or ethanol Cells were cultured on polyornithine-coated tissue culture dishes with or without 100 mM ethanol for 6 days and then with 50 ng/ml NGF in the continued presence or absence of ethanol for another 4 days. Additional cells were cultured with NGF and 10 nM PMA for 4 days without ethanol. The percentage of cells with neurites was calculated from 80 -120 cells/condition in 3-5 experiments.
a Significantly different compared with treatment with NGF alone (ANOVA and Scheffe F-test).
translocation of their corresponding PKC isozyme and not translocation of other isozymes concomitantly activated in these cells. Furthermore, the ⑀V1 fragment and the short peptide derived from it inhibited phorbol ester-or hormone-induced regulation of contraction rate, whereas the ␦V1 fragment or translocation inhibitors of PKC␤ did not. Together with the data presented here, these studies reinforce the concept that translocation of PKC is required for its function (28) and indicate that isozyme-selective inhibitors of PKC translocation can be used to determine the function of individual isozymes in a variety of cells.
The ability of the PKC␦ and -⑀ fragments to act as isozymeselective translocation inhibitors is also consistent with the hypothesis that each contains a binding site for a corresponding, isozyme-specific RACK (28). This hypothesis is further supported by the finding of structural homology between ⑀V1 (32) and the C2 domain of conventional PKCs, which contains a RACK1 binding site (29). Indeed, an PKC⑀-specific RACK, RACK2, that has recently been cloned, binds the ⑀V1 fragment in vitro. 2 Prominent PKC⑀ immunoreactivity was found in neurites and growth cones of NGF-treated parent PC12 and C1 cells. This was reduced in processes of cells expressing ⑀V1, suggesting that ⑀V1 displaces endogenous PKC⑀ from binding sites in neurites and growth cones. Localization of PKC⑀ to growth cones is consistent with a role for this isozyme in regulating neurite outgrowth (38). Immunoprecipitation-kinase assays indicate that NGF activates PKC⑀ in PC12 cells (21), suggesting that NGF-induced localization of PKC⑀ to neurites and growth cones may involve activation of this isozyme. This suggests that an PKC⑀-specific RACK may reside in growth cones and neurites.
In PC12 and C1 control cells, PKC␦ immunoreactivity was most prominent asymmetrically next to the nucleus. Expression of ␦V1 was associated with redistribution of PKC␦ immunoreactivity to the perinuclear region and with an increase in nuclear size. It is difficult to speculate on the physiologic significance of these changes, since the function of PKC␦ in these cells is not yet known. However, the findings clearly indicate that ␦V1 alters the localization of PKC␦ and produces a unique change in cell morphology.
NGF-induced activation of ERK1 and ERK2 involves phosphorylation and binding of phospholipase C␥ and the adapter protein Shc to the NGF receptor tyrosine kinase TrkA (39). Within minutes of NGF binding, phosphorylated Shc also forms a complex with another adapter protein, Grb2, and with the guanine nucleotide exchange factor mSOS (40), leading to activation of Ras (41) and sequential activation of B-Raf (42), MAP kinase kinase-1 (43,44), and ERK1 and ERK2 (45). This pathway appears essential for NGF-induced neurite outgrowth, since dominant negative inhibitors of Ras (46) or MAP kinase kinase-1 (47) block NGF-induced neurite outgrowth in PC12 cells. PKC⑀ could enhance ERK activation by increasing the activity of members of this pathway. For example, PKC␣ appears to phosphorylate and activate Raf-1 (48), suggesting that other PKC isozymes, such as PKC⑀, might also modulate Raf kinases. In addition, NGF stimulates binding of Shc to F-actin in PC12 cells (49). NGF also activates PKC⑀ (21), and upon activation, PKC⑀ binds F-actin in nerve terminals (50). This raises the intriguing possibility that Shc or proteins complexed with Shc interact with PKC⑀ when both Shc and PKC⑀ are anchored to actin. We do not yet know if the ⑀V1 fragment prevents binding of PKC⑀ to actin, since actin binds to PKC⑀ at a site between the first and second cysteine-rich regions of the C1 domain, which is outside of the ⑀V1 domain (50). Whether binding of PKC⑀ to actin is important for regulation of MAP kinases and neurite outgrowth also remains to be determined.
The major effect of phorbol esters or overexpressed PKC⑀ is to increase the late plateau phase of MAP kinase activation rather than the initial peak phase (24,27). These kinetics suggest that PKC⑀ may act by inhibiting dephosphorylation of ERK1 and ERK2 rather than by promoting ERK activation. MAP kinase phosphatase (MKP)-1, MKP-2, and hVH3 are dual specificity protein phosphatases that are expressed in the brain and dephosphorylate and inactivate ERKs (51,52). MKP-1 and MKP-2 mRNAs are constitutively expressed at low levels in PC12 cells and are increased by NGF (51). Inhibition of MKP-1 expression in PC12 cells does not accelerate the early phase of MAP kinase inactivation following stimulation with growth factors (53). Instead, protein phosphatase 2A and an unidentified protein-tyrosine phosphatase appear to mediate the rapid phase of inactivation in these cells (54). The role of dual specificity phosphatases in regulating the later plateau phase of NGF-induced MAP kinase activation is not known, but NGF induction of mRNA for MKP-1 and MKP-2 peaks at 1-2 h (51), suggesting that they may regulate this phase. If this is the case, then inhibition of these phosphatases by PKC⑀ may account for enhanced activation of ERK1 and ERK2.
ERK activity is also regulated by negative feedback inhibition. Phosphorylation of mSOS by MAP kinase promotes dissociation of mSOS from tyrosine-phosphorylated Shc and epidermal growth factor receptors (55,56). In addition, MAP kinase kinase-dependent phosphorylation of mSOS has been reported to cause dissociation of mSOS-Grb2 complexes, interrupting mSOS activation of Ras (57). Moreover, ERK1 phosphorylates MAP kinase kinase, and this phosphorylation ap- pears to reduce MAP kinase kinase activity (58). Therefore, inhibition of mSOS or MAP kinase kinase retrophosphorylation may be mechanisms by which PKC⑀ could enhance MAP kinase activation.
Together with studies in primary neurons (20,22,59), our findings suggest that PKC⑀ modulates neural differentiation. This may be a mechanism for enhancement of neurite growth by neurotransmitters that activate receptors coupled to phospholipase C and could be important for activity-dependent remodeling of synapses during normal development (60). Our studies also suggest that excessive activation of PKC⑀ may contribute to abnormal neurite growth observed in certain disease states. Chronic abuse of ethanol can damage the nervous system by disrupting the growth and remodeling of dendrites and axons. In certain brain regions, ethanol increases the growth of neural processes and terminals (61)(62)(63)(64)(65). This is particularly striking in the hippocampus, where expression of PKC⑀ is high (19). There prenatal exposure to ethanol causes marked overgrowth of dentate granule cell axons (mossy fibers) into the stratum pyramidale and stratum oriens of CA3, which may contribute to cognitive dysfunction (65). In addition, abnormal growth of mossy fibers into the supragranular layer of the dentate gyrus is found in some humans with temporal lobe epilepsy and in animals following stimuli that induce epilepsy (66). Future studies with transgenic and PKC⑀ mutant mice or with ⑀V1-derived inhibitory peptides may allow us to examine the role of PKC⑀ in normal development, alcohol-related neurologic disorders, and epileptogenesis.