Protein Kinase Cδ Mediates Ethanol-induced Up-regulation of L-type Calcium Channels*

Brief ethanol exposure inhibits L-type, voltage-gated calcium channels in neural cells, whereas chronic exposure increases the number of functional channels. In PC12 cells, this adaptive response is mediated by protein kinase C (PKC), but the PKC isozyme responsible is unknown. Since chronic ethanol exposure increases expression of PKCδ and PKCε, we investigated the role these isozymes play in up-regulation of L-type channels by ethanol. Incubation with the PKC inhibitor GF 109203X or expression of a PKCδ fragment that inhibits phorbol ester-induced PKCδ translocation largely prevented ethanol-induced increases in dihydropyridine binding and K+-stimulated 45Ca2+uptake. A corresponding PKCε fragment had no effect on this response. These findings indicate that PKCδ mediates up-regulation of L-type channels by ethanol. Remaining responses to ethanol in cells expressing the PKCδ fragment were not inhibited by GF 109203X, indicating that PKCδ-independent mechanisms also contribute. PKCδ overexpression increased binding sites for dihydropyridine and L-channel antagonists, but did not increase K+-stimulated45Ca2+ uptake, possibly because of homeostatic responses that maintain base-line levels of channel function. Since L-type channels modulate drinking behavior and contribute to neuronal hyperexcitability during alcohol withdrawal, these findings suggest an important role for PKCδ in alcohol consumption and dependence.

Understanding biochemical mechanisms that underlie alcohol tolerance and dependence may lead to new treatments for alcoholism. In nonalcoholic persons, intoxication develops at blood alcohol levels of 10 -35 mM, and acute tolerance develops rapidly, so that after a few hours, an individual can appear sober at alcohol levels that previously caused intoxication (1). The mechanism for acute tolerance may involve activation of Fyn kinase and reduced sensitivity of tyrosine-phosphorylated N-methyl-D-aspartate receptors to inhibition by alcohol (2). Chronic tolerance is characteristic of alcoholism, and its magnitude in alcoholics can be quite striking. For example, blood alcohol concentrations above 100 mM produce coma in a nonalcoholic person, whereas human alcoholics may appear sober or only mildly intoxicated with blood levels of 100 -150 mM (3,4).
The ability of ethanol to alter the function of neuronal voltage-dependent calcium channels appears to contribute to chronic tolerance. In several neural preparations, ethanol inhibits voltage-dependent calcium influx and calcium currents (5)(6)(7)(8)(9)(10)(11)(12)(13). Chronic exposure results in the development of tolerance to the inhibitory actions of ethanol on calcium channels (6,7). The mechanisms underlying this adaptive response have been studied most in the neural cell line PC12. In PC12 cells, prolonged exposure to 25-200 mM ethanol for 2-6 days produces a reversible concentration-and time-dependent increase in K ϩ -evoked 45 Ca 2ϩ uptake (8,9,14) and L-type calcium currents (14) measured in the absence of ethanol. This is associated with a corresponding increase in the number of binding sites for dihydropyridine Ca 2ϩ channel antagonists (8,9), suggesting that cells adapt to chronic ethanol exposure by increasing expression of functional L-type calcium channels. Similar increases in dihydropyridine binding have been detected in ethanol-treated NG108-15 neuroblastoma ϫ glioma cells (15) and in brain membranes from ethanol-treated rats (16). Up-regulation of L-type channels could promote alcohol consumption since L-channel antagonists reduce consumption in animals (17)(18)(19)(20). Increases in L-type calcium channels may also contribute to the intense neuronal hyperexcitability observed during alcohol withdrawal (21). A role for L-type channels in alcohol withdrawal syndromes is supported by evidence that L-channel antagonists reduce tremors, seizures, and mortality in alcohol-dependent rodents deprived of ethanol (22)(23)(24). Moreover, ethanol-induced increases in binding sites for Lchannel antagonists are greater in mice bred for severe alcohol withdrawal seizures than in mice bred for minor signs of alcohol withdrawal (25). These findings suggest that L-channel up-regulation plays an important role in alcohol dependence.
Protein kinase C (PKC) 1 is a family of phospholipid-dependent serine/threonine kinases involved in cell growth and differentiation, neurotransmitter release and receptor regulation, ion channel modulation, and gene expression (26). Eleven PKC isozymes have been identified (␣, ␤I, ␤II, ␥, ␦, ⑀, , , , , and ), and they differ in structure and requirements for activation by diacylglycerol and calcium (26 -28). In PC12 cells, we found that up-regulation of L-type channels by ethanol is inhibited by the kinase inhibitors sphingosine and polymyxin B (29). The effect of sphingosine is reversed by phorbol esters that activate all PKC isozymes except PKC and PKC (26), suggesting that ethanol-induced up-regulation of L-type channels requires activation of a phorbol ester-sensitive PKC. We also found that chronic ethanol exposure increases total PKC activity, high affinity phorbol ester binding, and PKC-mediated phosphorylation in PC12 cells (30). This is associated with a selective increase in immunoreactivity (30) and mRNA (31) for two PKC isozymes, PKC␦ and PKC⑀. Taken together, these findings suggest that chronic exposure to ethanol up-regulates functional L-type channels through a mechanism that involves ethanol-induced increases in expression of PKC␦ or PKC⑀.
In this paper, we examined whether PKC␦ or PKC⑀ mediates ethanol-induced increases in L-type channels by using stably transfected PC12 cell lines that express the ␦V1 or ⑀V1 fragment, which are derived from the first variable domains of PKC␦ and PKC⑀, respectively. These fragments selectively inhibit phorbol ester-induced translocation of the corresponding isozyme, and the ⑀V1 fragment specifically prevents phorbol ester-mediated inhibition of contraction in cultured cardiac myocytes (32) and enhancement of nerve growth factor-induced neurite outgrowth and mitogen-activated protein kinase activation by ethanol or phorbol esters in PC12 cells (33). We found that chronic exposure to ethanol increased K ϩ -stimulated 45 Ca 2ϩ uptake and dihydropyridine binding in PC12 cells, vector-transfected cells, and cells expressing ⑀V1, but not in cells expressing ␦V1. We also found that overexpression of PKC␦ increased dihydropyridine binding, but did not enhance K ϩstimulated 45 Ca 2ϩ uptake. These results demonstrate that PKC␦ mediates increases in L-type calcium channels induced by chronic exposure to ethanol. These results are the first to demonstrate a functional role for PKC␦ in neural cells.

EXPERIMENTAL PROCEDURES
Cell Culture-In these experiments, we used PC12 cells obtained from Dr. John A. Wagner (Cornell University, New York). Cells were grown at 37°C in plastic tissue culture flasks in Dulbecco's modified Eagle's medium supplemented with 5% fetal calf serum, 10% horse serum, 50 units/ml penicillin, 50 g/ml streptomycin, and 2 mM glutamine in a humidified atmosphere of 90% air and 10% CO 2 . Cells were cultured with ethanol in tightly capped tissue culture flasks or in six-well plates wrapped in Parafilm, and the medium was changed daily as described previously (8,29). Parallel control samples were cultured in a similar manner without ethanol. Stably transfected cell lines that overexpress PKC␦ or PKC⑀ have been described previously (34). Cell lines that express Flag epitope-tagged ␦V1 or ⑀V1 were created and analyzed by reverse transcriptase-polymerase chain reaction and Western analysis for the Flag epitope tag as described (33).
Phorbol Ester-stimulated Translocation of PKC␦-Cells (4 ϫ 10 6 ) were plated on 100-mm 2 plastic tissue culture plates. After 2 days, the plates were rinsed with 10 ml of Dulbecco's modified Eagle's medium and incubated with or without 30 nM phorbol 12-myristate 13-acetate (PMA) for 2 min at 37°C. Cells were rinsed twice with Ca 2ϩ -and Mg 2ϩ -free phosphate-buffered saline and scraped into ice-cold buffer 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 in a Teflon-glass homogenizer, and sucrose was added to a final concentration of 250 mM. Samples were then centrifuged at 150,000 ϫ g for 1 h, and the supernatant was saved as the cytosolic fraction. The pellet was dispersed by sonication, and samples of supernatant and pellet derived from 100 g of crude homogenate were analyzed by Western analysis as described (33).
K ϩ -stimulated 45 Ca 2ϩ Uptake-PC12 cells were plated onto poly-Dlysine-coated six-well tissue culture plates at a density of 0.8 -1.6 ϫ 10 6 cells/well. After 24 h, cells were cultured for another 1-6 days in the presence or absence of ethanol. On the day of assay, cells were rinsed twice with 1 ml of 5 mM KCl buffer (85 mM NaCl, 5 mM KCl, 45 mM choline chloride, 2 mM CaCl 2 , 5 mM glucose, and 25 mM HEPES, pH 7.4) at room temperature, and incubated in the same buffer for 25 min. Cells were then incubated at 25°C in 5 or 50 mM KCl buffer containing 0.75 Ci of 45 Ca 2ϩ . The composition of the 50 mM KCl buffer was identical to that of the 5 mM KCl buffer except that choline chloride was replaced by KCl. After incubation for 2.5 min, cells were washed four times with 2.5 ml of ice-cold 5 mM KCl buffer and incubated overnight with 1 ml of 1 M NaOH. Radioactivity in neutralized samples was measured by liquid scintillation counting, and protein levels were determined by the method of Lowry et al. (35). K ϩ -stimulated uptake is defined as the difference between uptake in 50 and 5 mM KCl buffers and represents 85 Ϯ 1% (n ϭ 10) of total uptake in 50 mM KCl buffer.
(ϩ)-[ 3 H]PN200-110 Binding-Binding of (ϩ)-[ 3 H]PN200-110 to whole cells was measured as described previously (36). Specific binding was determined by subtracting binding in the presence of 1 M nimo-dipine from binding in its absence and represents 51 Ϯ 2% (n ϭ 10) of total binding at 50 pM radioligand. 45 Ca 2ϩ Uptake-Previous work suggested that up-regulation of L-type channels by ethanol requires activation of PKC since it is prevented by the nonselective PKC inhibitors sphingosine and polymyxin B (29). Since we had used a different clone of PC12 cells for these earlier studies, we needed to determine whether our current PC12 cell line responds similarly to ethanol. In control cells, 45 Ca 2ϩ uptake was 3.6 Ϯ 0.21 nmol of Ca 2ϩ /mg of protein/2.5 min (n ϭ 103) and was inhibited 92 Ϯ 2% by the L-type channel antagonist nimodipine (1 M). Exposure to 25-150 mM ethanol for 1-6 days increased K ϩstimulated 45 Ca 2ϩ uptake assayed in the absence of ethanol ( Fig. 1 A and B). This is similar to the increase we observed previously with another PC12 cell line (8), except that a maximal response was achieved at 6 days in those cells instead of at 5 days as observed in the current cell line. To examine whether the increase in K ϩ -stimulated 45 Ca 2ϩ uptake was PKC-dependent, we cultured cells in the presence of 1 M GF 109203X, which is a relatively selective PKC inhibitor that inhibits all phorbol ester-sensitive PKC isozymes except PKC (37,38). As shown in Fig. 1A, GF 109203X prevented increases in K ϩstimulated 45 Ca 2ϩ uptake induced by treatment with 150 mM ethanol for 6 days. Thus, ethanol appears to increase K ϩstimulated 45 Ca 2ϩ uptake in these cells by a PKC-dependent mechanism.

GF 109203X Prevents Ethanol-induced Increases in K
Characterization of V1-expressing Cell Lines-To investigate whether PKC␦ or PKC⑀ is required for up-regulation of L-type calcium channels by ethanol, we created PC12 cell lines that stably express V1 fragments derived from the first variable domains of PKC␦ (␦V1) and PKC⑀ (⑀V1). These fragments inhibit phorbol ester-induced translocation and activation of their corresponding PKC isozyme (32,33). Characterization of ⑀V1-expressing PC12 cell lines (V1⑀1 and V1⑀2) was described recently (33). Expression of the ␦V1 fragment in V1␦2, V1␦3, and V1␦4 cells was confirmed by reverse transcriptase-polymerase chain reaction ( Fig. 2A) and Western analysis (Fig. 2B). Treatment with 30 nM PMA stimulated translocation of both PKC␦ and PKC⑀ to the particulate fraction in the parent PC12 cell line and in C cells transfected with vector alone (Fig. 2, C  and D). However, in cell lines expressing the ␦V1 fragment, translocation of PKC␦ was inhibited, whereas translocation of PKC⑀ was not (Fig. 2, C and D). In contrast, PMA-induced translocation of PKC⑀ is selectively inhibited in ⑀V1-expressing lines (33). Thus, expression of ␦V1 or ⑀V1 fragments selectively inhibits phorbol ester-stimulated translocation of the corresponding PKC isozyme in these cells.

FIG. 2. Characterization of stably transfected PC12 cells ex-
pressing Flag epitope-tagged ␦V1. A, reverse transcriptase-polymerase chain reaction products were analyzed on 1.2% agarose gels to demonstrate expression of ␦V1-Flag mRNA in clones V1␦2, V1␦3, and V1␦4. No product was found in PC12 cells or C cells transfected with vector alone. kb, kilobase; Std, DNA size markers. B, lysates of V1␦2, V1␦3, and V1␦4 cells were analyzed on Western blots to detect the presence of 17-kDa Flag immunoreactivity that comigrates with purified Flag-tagged ␦V1 expressed in bacteria (␦-Flag). C, cells were treated with 30 nM PMA for 2 min to induce PKC translocation and then fractionated into cytosolic (Cy) and particulate (P) cell fractions. PKC␦ and PKC⑀ immunoreactivity in cell fractions was detected by Western analysis. The data shown are from a representative experiment. D, PKC isozyme-specific immunoreactivity on Western blots of cytosolic and particulate fractions was quantified by scanning densitometry and is expressed as a percentage of total immunoreactivity. PKC isozyme translocation was then calculated as the increase in isozyme-specific immunoreactivity found in the particulate fraction following treatment with PMA. The data shown are the means Ϯ S.E. from four to seven experiments. *, p Ͻ 0.01 compared with PC12 or C cells (ANOVA and Newman-Keuls test). 45 Ca 2؉ uptake and dihydropyridine binding are inhibited in cells expressing ␦V1. Cells were treated with 150 mM ethanol for 6 days. A, depolarization-induced 45 Ca 2ϩ uptake was measured in PC12 cells, vector-transfected cells (C), and cells expressing the ␦V1 fragment (V1␦2, V1␦3, or V1␦4) or the ⑀V1 fragment (V1⑀1 or V1⑀2). The data shown are the means Ϯ S.E. (n ϭ 5-24) and are expressed as the percent above 45 Ca 2ϩ uptake measured in parallel control cells cultured without ethanol. *, p Ͻ 0.05 compared with PC12 or C cells (ANOVA and Newman-Keuls test). B, binding of (ϩ)-[ 3 H]PN200-110 was measured in depolarized cells, and the results are expressed as the percent above or below specific binding measured in parallel control cells cultured without ethanol. The data shown are the means Ϯ S.E. (n ϭ 6 -9). *, p Ͻ 0.05 compared with PC12 or C cells (ANOVA and Newman-Keuls test).
Overexpression of PKC␦-To examine whether increases in PKC␦ are sufficient to increase L-channel density, we examined (ϩ)-[ 3 H]PN200-110 binding in PC12 cell lines ␦1 and ␦2, which overexpress PKC␦ (34). Compared with nontransfected PC12 cells, binding of 50 pM (ϩ)-[ 3 H]PN200-110 was increased in ␦1 and ␦2 cells, but not in C cells transfected with vector alone (Fig. 4). To determine whether the increase in binding in PKC␦-overexpressing cells was due to a change in binding affinity or binding site number, binding was measured in ␦2 cells at increasing concentrations of (ϩ)-[ 3 H]PN200-110 (from 10 to 280 pM). Fig. 5 shows a representative experiment. Scatchard analysis yielded similar values for the equilibrium dissociation constant (K D ) in PC12 (84 Ϯ 4 pM) and ␦2 (102 Ϯ 14 pM) cells (p ϭ 0.27; n ϭ 3). In contrast, the maximal number of binding sites (B max ) in ␦2 cells (16.8 Ϯ 1.9 fmol/mg) was greater (p Ͻ 0.028; n ϭ 3) than the B max in PC12 cells (10.0 Ϯ 0.7 fmol/mg). These findings are consistent with an increase in L-channel density rather than an increase in binding affinity in cells that overexpress PKC␦. However, despite the increase in dihydropyridine binding, K ϩ -stimulated 45 Ca 2ϩ uptake was not increased in ␦1 or ␦2 cells (Fig. 4).
PKC␦-independent Regulation of L-type Channels by Ethanol-Since overexpression of PKC␦ increased dihydropyridine binding but not K ϩ -stimulated 45 Ca 2ϩ uptake, we considered whether enhancement of L-channel function by ethanol requires additional, PKC␦-independent mechanisms. Initial evidence for a PKC␦-independent mechanism was obtained by examining ␦V1-expressing cells treated with ethanol and PKC inhibitors. Although treatment with GF 109203X substantially inhibited ethanol-induced increases in K ϩ -stimulated 45 Ca 2ϩ uptake in PC12 cells (Fig. 1A), in V1␦4 cells, it did not prevent increases due to ethanol, which were 26 Ϯ 3% of control in the absence and 25 Ϯ 5% of control in the presence of GF 109203X (p ϭ 0.9; n ϭ 6). This result indicates that a PKC␦-independent mechanism activated by ethanol contributes to L-channel upregulation. To examine if such a PKC␦-independent mecha-nism is required to increase L-channel function in PKC␦-overexpressing cells, we treated ␦1 and ␦2 cells with 150 mM ethanol for 5 days. We predicted that this treatment would markedly increase 45 Ca 2ϩ uptake with little or no effect on dihydropyridine binding. Ethanol enhanced K ϩ -stimulated 45 Ca 2ϩ uptake in these cells (Fig. 6A) without increasing PKC␦ immunoreactivity (Fig. 6B), consistent with activation of a PKC␦-independent mechanism. However, the increase in 45 Ca 2ϩ uptake was modest and associated with a similar increase in dihydropyridine binding (Fig. 6A).

DISCUSSION
In this paper, we found that ethanol-induced increases in L-channel density and function were largely prevented by expression of ␦V1, a selective inhibitor of PKC␦ translocation. In contrast, expression of ⑀V1, which inhibits PKC⑀ translocation, did not prevent L-channel up-regulation. Expression of ␦V1 did not alter L-channel density or function in the absence of ethanol. These results indicate that PKC␦, but not PKC⑀, is important for ethanol-induced increases in functional L-type calcium channels in PC12 cells, but not for the basal activity of these channels.
Since ethanol increases the abundance of PKC␦ (30), we examined whether overexpression of PKC␦ would mimic the effect of ethanol on L-channel density and function. Although PKC␦ overexpression increased the number of dihydropyridine-binding sites in PC12 cells, it did not increase K ϩ -stimulated 45 Ca 2ϩ uptake. This could have occurred because PKC␦independent mechanisms are also required to increase channel function. Two results provided evidence for PKC␦-independent mechanisms that are activated by ethanol. In V1␦4 cells, PKC␦ translocation was completely blocked with only partial sup- pression of the response to ethanol, and the remaining ethanolinduced response was resistant to the PKC inhibitor GF 109203X. Moreover, in PKC␦-overexpressing cells, ethanol increased K ϩ -stimulated 45 Ca 2ϩ uptake and dihydropyridine binding without a further increase in PKC␦ immunoreactivity. These findings indicate that PKC␦-independent mechanisms contribute to up-regulation of L-type channels by ethanol.
If PKC␦-independent mechanisms activated by ethanol are required for increases in L-channel function, treatment of PKC␦-overexpressing cells with ethanol should have markedly increased 45 Ca 2ϩ uptake to levels commensurate with levels of dihydropyridine binding in these cells. However, this did not occur. PKC␦ overexpression alone increased dihydropyridine binding by ϳ1.6-fold (Fig. 4), and ethanol treatment by another 1.4-fold (Fig. 6), for a combined increase of 2.2-fold over binding in untreated PC12 cells. In contrast, PKC␦ overexpression and ethanol treatment together only increased 45 Ca 2ϩ uptake by 1.4-fold over uptake in untreated PC12 cells (Figs. 4 and 6). It appears that inactivity of PKC␦-independent mechanisms cannot explain why PKC␦ overexpression alone increased L-channel density, but not L-channel function.
It is possible that increases in channel density evoked by stable overexpression of PKC␦ activate homeostatic mechanisms that reduce channel function to maintain a set level of calcium signaling. These mechanisms could act by decreasing L-channel function through altered subunit composition or phosphorylation or by increasing calcium efflux through increased Na ϩ -Ca 2ϩ exchange and the action of membrane AT-Pases (39). Homeostatic mechanisms that regulate channel density are activated by treatments that alter L-channel activity. For example, prolonged treatment with drugs such as ethanol (8, 9, 14 -16), morphine (40), or nifedipine (41), which inhibit L-channel activity, increases L-channel density. In contrast, prolonged activation by exposure to depolarizing concentrations of KCl (42) or the L-channel agonist Bay K8644 (43) decreases L-channel density. It appears that L-channel function is highly regulated, and any alteration in channel activity or density leads to compensatory responses that serve to normalize function in the continued presence of the perturbing stimulus. Further studies will be needed to identify these compensatory mechanisms and to determine if they are activated in PKC␦-overexpressing cells.
One mechanism by which ethanol, acting via PKC␦, could increase L-channel density is by increasing expression of channel subunits. Neuronal high voltage-activated calcium channels are multimeric complexes of at least three types of subunits: ␣ 1 , ␣ 2 ␦, and ␤ (44). The major pharmacological and physiological features that distinguish different classes of voltage-gated channels are mainly due to ␣ 1 subunits, which contain the calcium pore and binding sites for selective calcium channel antagonists. There are five genes known to encode ␣ 1 subunits in brain (␣ 1A , ␣ 1B , ␣ 1C , ␣ 1D , and ␣ 1E ), and ␣ 1C and ␣ 1D are subunits of L-type channels (45)(46)(47)(48). PC12 cells express ␣ 1C (49). Transfected ␣ 1C can form functional L-type channels, and coexpression with ␣ 2 ␦ or ␤ subunits results in increased channel function and a corresponding increase in dihydropyridine binding (47, 50 -54). Thus, ethanol-induced increases in abundance of ␣ 1C , ␣ 2 ␦, or ␤ subunits could increase the number of functional L-type channels. This might occur at a transcriptional level since PKC␦ activates AP-1/Jun-regulated gene expression (55,56). Studies are currently underway to determine if chronic exposure to ethanol regulates expression of specific calcium channel subunits by a PKC␦-dependent mechanism.
PKC␦ is ubiquitously expressed and has been implicated in control of cell growth (56 -58), apoptosis (59), and exocytosis (60) in non-neuronal cells. Little is known about its role in neuronal cells. It is induced in rat brain through an N-methyl-D-aspartate receptor-dependent mechanism after transient focal ischemia (61), but its role in brain injury or repair is not known. PKC␦ binds to the growth-associated protein GAP-43 and appears to act as a GAP-43 kinase (62). However, it is not yet clear if PKC␦ specifically regulates functions such as neurite growth or neurotransmitter release, which appear to be modulated by GAP-43 (63,64). Our results provide the first evidence of a functional role for PKC␦ in neural cells. Our findings identify PKC␦ as a regulator of L-channel density and a mediator of cellular adaptation to ethanol. Since L-type channels modulate drinking behavior (17)(18)(19)(20) and contribute to manifestations of alcohol withdrawal (22)(23)(24), PKC␦ may play a key role in alcohol consumption and dependence. Ongoing studies will determine if inhibition of PKC␦ reduces ethanol consumption and the development of alcohol dependence in animals. FIG. 6. Ethanol treatment of PKC␦-overexpressing cells. Cells were treated with 150 mM ethanol for 5 days. A, K ϩ -stimulated 45 Ca 2ϩ uptake (stippled bars) and binding of 50 pM (ϩ)-[ 3 H]PN200-110 (black bars) were measured in PC12 cells and in ␦1 and ␦2 cells that overexpress PKC␦. The data shown are the means Ϯ S.E. from three to eight uptake experiments and four binding experiments and are expressed as the percent of uptake or binding above that measured in parallel control cultures of each cell line treated without drugs. Ethanol-induced increases in 45 Ca 2ϩ uptake and dihydropyridine binding in ␦1 and ␦2 cells were significantly less than in PC12 cells (p Ͻ 0.05; ANOVA and Newman-Keuls test). B, shown is the PKC␦ immunoreactivity on Western blots of total cell lysates (mean Ϯ S.E., n ϭ 2). Inset, Western blot of control (C) and ethanol-treated (E) cells from a representative experiment. The data shown are expressed as the percent of PKC␦ immunoreactivity above or below that measured in parallel control cultures of each cell line treated without ethanol.