Ethanol Regulates Calcium Channel Subunits by Protein Kinase C δ-dependent and -independent Mechanisms*

Chronic exposure to ethanol increases the number of functional L-type voltage-gated calcium channels in neural cells. In PC12 cells, this adaptive response is mediated by protein kinase C δ (PKCδ), but the mechanisms by which this occurs are not known. Since expression of several different calcium channel subunits can increase the abundance of functional L-type channels, we sought to identify which subunits are regulated by ethanol. Incubation of PC12 cells with 120–150 mm ethanol for 6 days increased levels of α1C, α2, and β1b subunit immunoreactivity in cell membranes and selectively increased the abundance of mRNA encoding the α1C-1 splice variant of α1C. In cells expressing a fragment of PKCδ (δV1) that selectively inhibits PKCδ, there was no increase in membrane-associated α1C, α2, and β1b immunoreactivity following chronic ethanol exposure. However, ethanol still increased levels of α1C-1 mRNA in these cells. These results indicate that ethanol increases the abundance of L-type channels by at least two mechanisms; one involves increases in mRNA encoding a splice variant of α1Cand the other is post-transcriptional, rate-limiting, and requires PKCδ.

Voltage-gated calcium channels mediate calcium entry into neurons and regulate neurotransmitter release, firing patterns, gene expression, and differentiation (1,2). L-type channels are a subfamily of voltage-gated calcium channels that are activated by high voltage, inactivate slowly, and are blocked by dihydropyridines (DHPs). 1 Acute ethanol exposure inhibits the function of L-type channels in several neuronal preparations (3)(4)(5). In contrast, in the neural crest-derived cell line PC12, chronic ethanol exposure produces a reversible concentrationand time-dependent increase in K ϩ -evoked 45 Ca 2ϩ uptake and depolarization-evoked calcium currents through L-type channels (6 -8). Ethanol-induced increases in L-type channel function are associated with corresponding increases in the density of binding sites for DHP calcium channel antagonists (6,7), indicating that chronic exposure to ethanol increases the number of functional L-type channels. Similar increases in DHP binding have been detected in brain membranes from ethanol-dependent rodents (9,10). Up-regulation of L-type calcium channels appears to contribute to intense neuronal hyperexcitability observed during alcohol withdrawal since L channel antagonists reduce tremors, seizures, and mortality in alcoholdependent rodents deprived of ethanol (11,12). Ethanol-induced increases in L-type channels may also promote alcohol consumption since L channel antagonists reduce ethanol selfadministration in animals (13)(14)(15)(16).
Neuronal voltage-gated calcium channels are multimeric complexes of at least three types of subunits as follows: ␣ 1 , ␣ 2 ␦, and ␤ (2). Diversity within the ␣ 1 subunit family is responsible for the major pharmacological and physiological features that distinguish the different classes of calcium channels. ␣ 1 subunits contain the calcium pore and binding sites for selective channel antagonists. They are comprised of four homologous repeats (I-IV) each containing six transmembrane segments (S1-S6). Four L-type channel ␣ 1 genes have been cloned thus far as follows: ␣ 1S from skeletal muscle (17), ␣ 1C from heart and brain (18), ␣ 1D from neural and endocrine tissues (19), and ␣ 1F from retina (20,21). In brain, ␣ 1D and ␣ 1C are localized to neuronal cell bodies and proximal dendrites (22). Two splice variants of the rat ␣ 1C subunit have been identified, ␣ 1C-1 and ␣ 1C-2 , and are differentially expressed in rat brain (18). The ␣ 1C-2 protein differs from ␣ 1C-1 by having a 3-amino acid (aa) insert in the cytoplasmic loop between domains II and III and a 28-aa substitution in the S3 segment in domain IV. In the human ␣ 1C gene, this alternatively spliced IV-S3 transmembrane segment is encoded by homologous alternative exons 31 and 32 (23). It is not known if ␣ 1C-1 and ␣ 1C-2 differ in function.
Protein kinase C (PKC) is a multigene family of phospholipiddependent, serine-threonine kinases that regulate cell growth and differentiation, neurotransmitter release, receptor regulation, ion channel modulation, and gene expression (24). Twelve PKC isozymes, encoded by 11 genes, have been identified (␣, ␤I, ␤II, ␥, ␦, ⑀, , , , , , and ) and differ in structure, requirements for activation, and patterns of expression (24 -27). We recently found that ethanol-induced increases in L-type channels can be blocked by an inhibitor of PKC␦ (28). In this study we examined the abundance of specific calcium channel subunits and their mRNAs to explore further the mechanisms by which PKC␦ mediates up-regulation of L-type channels by ethanol.

EXPERIMENTAL PROCEDURES
Materials-Radioisotopes and nucleotides were purchased from Amersham Pharmacia Biotech. Restriction endonucleases and modifying enzymes were purchased from Promega (Madison, WI). JM109 (Promega) and XL-1 Blue (Stratagene, La Jolla, CA) bacteria were used. All other reagents were analytical grade and were from Sigma or Life Technologies, Inc.
Cell Culture-PC12 cells (J. Wagner, Cornell University) were cultured at 37°C 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 120 -150 mM ethanol in tightly capped flasks, and the medium was changed daily as in prior work (28). Control samples were cultured in parallel without ethanol.
Western Blotting-For detection of ␣ 2 and ␤ subunit immunoreactivity, cells were washed with PBS, solubilized in 1% digitonin, and homogenized in Buffer A containing 10 mM HEPES, pH 7.4, 0.3 M sucrose, 10 mM EDTA, 10 mM EGTA, and protease inhibitors (1 mM phenylmethylsulfonyl fluoride, 0.75 mM benzamide, 0.7 g/ml pepstatin A, 1 g/ml aprotinin, 1 g/ml leupeptin, 0.01 mg/ml lysozyme, and 8 mg/ml calpain 1 ϩ 2). Cell membranes were collected by ultracentrifugation at 135,000 ϫ g for 30 min at 4°C and were resuspended in solubilization buffer containing 300 mM KCl, 150 mM NaCl, 10 mM sodium phosphate, pH 7.4, 10 mM EGTA, 10 mM EDTA, and the protease inhibitors used in Buffer A. Unsolubilized material was removed by centrifugation at 175,000 ϫ g for 45 min at 4°C. For detection of ␣ 1C and neuronal ␥ (stargazin (29)) subunit immunoreactivity, cells were treated as described (30). Samples (80, 160, and 320 g) of crude membrane pellet or digitonin-solubilized membranes were separated on SDS-polyacrylamide gels. Proteins were then transferred to nitrocellulose Hybond C Extra membranes (Amersham Pharmacia Biotech) for 2 h at 100 V, and membranes were incubated in PBS containing 0.1% Tween 20 and 5% nonfat dried milk for 1 h at room temperature. After incubation in primary antibody in the same buffer for 2 h at room temperature, or overnight at 4°C, the blots were washed three times in PBS containing 0.1% Tween 20 for 10 min and then incubated with peroxidase-conjugated goat anti-rabbit IgG (Roche Molecular Biochemicals, 1:1000 dilution) for 1 h in PBS, 0.1% Tween 20 at room temperature. Calcium channel subunit immunoreactivity was detected by chemiluminescence with ECL reagent (Amersham Pharmacia Biotech). Immunoreactive bands were quantified by densitometric scanning, and linear regression analysis of protein concentrations and corresponding density values was used to determine the slope of the regression line for each condition. Slopes for ethanol-treated samples were divided by slopes calculated for paired control samples to calculate the percentage increase in immunoreactivity induced by ethanol. Polyclonal antibodies against ␣ 2 , ␤ 1b , ␤ 2 , ␤ 3 , and neuronal ␥ (stargazin) subunits were obtained from K. Campbell (University of Iowa) and were used at dilutions of 1:500 to 1:1000. Polyclonal antibody CARD I against ␣ 1C (1:1000) was a gift from M. Hosey (Northwestern University). Antibody against ␣ 1D (1:600) was purchased from Alomone Labs (Jerusalem, Israel).
To generate a probe for ␣ 1C-1 , two oligonucleotide primers were constructed from the rat ␣ 1C-1 sequence (GenBank TM accession number M67516) to flank the mismatch between ␣ 1C-1 and ␣ 1C-2 at the IV S3 domain as follows: JD1802 (CCCAAGCTTG GGTTTGACAA TGTTCT-GGCA GCC), upstream of the mismatch with a HindIII site incorporated into the sequence, and JD1803 (CGCGGATCCG CGTGACGATG AGGAAGTCAA AAAC), downstream and spanning the mismatch region, with a BamHI site incorporated into the sequence. The first strand of ␣ 1C-1 cDNA was synthesized using the SuperScript Choice System for DNA synthesis kit (Life Technologies, Inc.). Total RNA (12 g), isolated from PC12 cells, was heated to 70°C for 10 min with 100 pmol of oligo(dT) primer (Life Technologies, Inc.) and then chilled on ice before 1ϫ first strand buffer (Life Technologies, Inc.), 10 mM dithiothreitol, and 0.5 mM dNTP (dATP, dCTP, dGTP, dTTP) mix were added. The reaction was then heated for 2 min at 42°C, and then 1 l of Super-Script II reverse transcriptase (Life Technologies, Inc.) was added. After incubation for 1 h at 42°C, 2 l of the reaction product, containing reverse-transcribed cDNA, was added to a PCR mixture containing 1ϫ polymerase chain reaction (PCR) buffer (Perkin-Elmer), 0.5 mM dNTP mix, 2.5 units of Amplitaq (Perkin-Elmer), 100 pmol of JD1802, and 100 pmol of JD1803. The reaction mixture was heated to 94°C for 4 min and then subjected to 30 amplification cycles. Each cycle consisted of 53°C for 45 s, 72°C for 2 min, and 94°C for 10 min. Finally, the mixture was incubated at 53°C for 45 s and 2°C for 2 min and then placed on ice.
PCR products were digested with BamHI and HindIII and separated on a 1.2% agarose gel. The resultant fragments were excised and gel-purified using a QIAEX II Gel Extraction kit (Qiagen, Chatsworth, CA). Purified fragments were subcloned into pBluescript II SK(ϩ) (Stratagene, La Jolla, CA), and positive colonies were sequenced. The sequence was identical to the predicted rat ␣ 1C-1 sequence (nucleotides 4046 -4628). A HindIII-linearized plasmid was used as a template with T3 RNA polymerase to generate a [␣-32 P]CTP-labeled 582-bp cRNA probe.
Ribonuclease Protection Assay-Total RNA was extracted from PC12 cells using the RNA STAT-60 method (Tel-Test, Friendswood, TX) and quantified by absorbance at 260 nm. Ribonuclease protection assays (RPAs) were performed as described previously (31). Briefly, total RNA (20 g) was dissolved in 30 l of hybridization solution containing 60,000 cpm of a 32 P-labeled calcium channel subunit cRNA probe and 10,000 cpm of a 32 P-labeled GAPDH cRNA probe. The cRNA probes were allowed to anneal to the endogenous RNA at 45°C overnight. The next day, digestion was performed at 37°C for 30 min using an RNase solution containing a final concentration of 30 g/ml RNase A (Ambion) and 800 units of RNase T1. The RNA:RNA hybrids were separated on a 5% polyacrylamide, 8 M urea sequencing gel. The gel was dried, and mRNA fragments were visualized and densities calculated using a Storm 860 PhosphorImager and ImageQuant software (Molecular Dynamics). Century Template RNA markers (Ambion) were used to determine molecular weights. Results were expressed relative to parallel control samples cultured without ethanol.
Slot Blot Analysis-Total RNA (1, 2.5, 5, and 10 g) was prepared for analysis in sample buffer (66% formamide, 10% formaldehyde, and 1ϫ SSC). Samples were heated to 68°C for 15 min before an equal volume of ice-cold 20ϫ SSC was added. BA-85 nitrocellulose membranes were loaded into a slot blot apparatus (Schleicher & Schuell) and rinsed with 500 l/well of 10ϫ SSC. After the samples were loaded, the apparatus was washed with 1 ml of 10ϫ SSC under vacuum. RNA was cross-linked to damp membranes under UV light.
Plasmids containing cDNA inserts encoding ␣ 2 ␦ (50 ng; a gift from A. Schwartz, University of Cincinnati) and GAPDH (50 ng; CLONTECH, Palo Alto, CA) were used as templates for random primer labeling using the Life Technologies, Inc., Radprime DNA labeling system. Excess radioactivity was removed by centrifugation through TE-10 columns (CLONTECH) at 700 ϫ g for 7 min. Membranes were placed in prehybridization solution (60% formamide, 3ϫ SSC, 5ϫ Denhardt's solution, 0.82 mM sodium pyrophosphate, 82 g/ml salmon sperm DNA, 1% SDS) for 4 h at 42°C. 32 P-Labeled DNA (3-5 ϫ 10 6 cpm/ml) was boiled for 10 min in the presence of salmon sperm DNA (0.1 mg/ml) and placed immediately on ice. Cooled 32 P-labeled probes were added to pre-hybridization solution with the addition of 5% w/v dextran sulfate and 1% SDS. After an overnight incubation at 42°C, membranes were washed at 25°C three times in 1ϫ SSC, 0.1% SDS, 5 min for each wash. Radioactivity bound to the blots was quantified using a Storm 860 PhosphorImager and ImageQuant software. Levels of ␣ 2 ␦ and GAPDH mRNA increased linearly over the concentrations of samples examined. Linear regression analysis of total mRNA and corresponding hybridization signals for ␣ 2 ␦ and GAPDH mRNA were used to calculate the slope of the regression lines for each condition. Corresponding slopes calculated for ␣ 2 ␦ and GAPDH mRNA were divided to normalize for abundance of GAPDH mRNA. Normalized slopes for ethanol-treated and paired control samples were then compared to calculate the percentage increase in ␣ 2 ␦ mRNA induced by ethanol.

RESULTS
Chronic Exposure to Ethanol Increases Levels of ␣ 1C , ␣ 2 , and ␤ 1b Polypeptides in Cell Membranes-To determine which Ltype calcium channel subunits are regulated by ethanol, we first identified which subunits are expressed in PC12 cells. Western analysis of PC12 cell membranes with subunit-specific antibodies demonstrated immunoreactive bands of appropriate molecular mass (32,33) for ␣ 2 (140 kDa), ␤ 1b (72 kDa), and ␤ 3 (58 kDa) subunits (Fig. 1A). We also found an 86-kDa immunoreactive band using an anti-␤ 2 antibody (Fig. 1A). In a prior report (33), this antibody labeled a 74-kDa protein in Western blots of PC12 cell membranes but recognized 87-, 74-, and 70-kDa protein bands in Western blots of rat cardiac micro-some membranes. A 225-kDa band was detected using the anti-␣ 1C antibody, which recognizes a band of 240 kDa in ␣ 1C -expressing Sf9 cells (34). Although we found ␣ 1D -like immunoreactivity in brain tissue, we were unable to detect immunoreactive bands of the appropriate molecular mass for this subunit in our line of PC12 cells (data not shown). This is similar to what has been previously reported by Liu and colleagues (33), who also found that PC12 cells do not express ␤ 4 subunit immunoreactivity. We also did not detect immunoreactivity for the neuronal ␥ (stargazin) subunit in PC12 cells (data not shown).
One mechanism by which ethanol could increase the density of functional L-type calcium channels in PC12 cells is by increasing the expression of ␣ 1 subunits, which form the calcium pore. However, ␣ 2 and especially ␤ subunits promote the assembly and targeting of channel complexes to the cell membrane and together with ␣ 1C or ␣ 1D increase DHP binding and enhance L-type channel function (19,(35)(36)(37)(38)(39). Therefore, ethanol-induced increases in abundance of ␣ 2 or ␤ subunits might also increase the density of functional L-type channels in PC12 cells.
To investigate which calcium channel subunits were increased by ethanol, we examined subunit immunoreactivity in membranes isolated from PC12 cells treated with 150 mM ethanol for 0 -6 days. Ethanol evoked a time-dependent increase in ␣ 1C immunoreactivity, with levels peaking at 6 days of ethanol exposure (Fig. 1, A and B). Ethanol also increased levels of immunoreactivity for ␣ 2 and ␤ 1b without altering levels of immunoreactivity for ␤ 2 or ␤ 3 subunits (Fig. 1, A and  C). Ethanol did not induce the appearance of ␣ 1D -like immunoreactivity (data not shown). These findings suggest that in PC12 cells, chronic ethanol exposure increases the abundance of L-type calcium channels composed of ␣ 1C , ␣ 2 , and ␤ 1b subunits.
Chronic Exposure to Ethanol Does Not Increase Levels of ␣ 1C , ␣ 2 , and ␤ 1b Polypeptides in Cells That Express an Inhibitor of PKC␦-To examine the role of PKC␦ in calcium channel regulation, we used PC12 cell lines that stably express the fragment ␦V1, which is derived from the first variable domain of PKC␦ and selectively inhibits phorbol ester-induced translocation of PKC␦ (40,41). No differences in basal levels of immunoreactivity for ␣ 1C , ␣ 2 , and ␤ 1b were observed in PC12 cells, vectortransfected cells, or V1␦2 or V1␦3 cells that express ␦V1 (data not shown). Treatment with 150 mM ethanol for 6 days increased membrane-associated immunoreactivity for ␣ 1C , ␣ 2 , and ␤ 1b by 60 -75% in PC12 cells and vector-transfected cells (Fig. 2). However no increase was observed in V1␦2 or V1␦3 cells. These results suggest that although PKC␦ does not reg- ulate basal levels of ␣ 1C , ␣ 2 , and ␤ 1b , it is required for ethanolinduced increases in these calcium channel subunits.
Ethanol Increases Levels of mRNA for a Splice Variant of ␣ 1C -Chronic exposure to ethanol alters the abundance of several proteins including tyrosine hydroxylase (42) phosducinlike protein (43), and the molecular chaperones Hsc 70 (44), GRP 78 (45), and GRP 94 (45) by increasing gene expression. Ethanol-induced increases in L-type channels might also occur at a transcriptional level since PKC␦ has been implicated in activation of AP1/Jun-regulated gene expression (46,47).
To investigate this possibility, we measured the abundance of calcium channel subunit mRNAs in PC12 cells after chronic ethanol exposure. Levels of total RNA per cell were not altered by exposure to 120 mM ethanol for 1-5 days (n ϭ 7, p ϭ 0.3754, ANOVA). Therefore we performed all mRNA studies using total RNA. Subsequent ribonuclease protection assay (RPA) analysis of GAPDH mRNA expression demonstrated that GAPDH mRNA levels were not altered by exposure to 120 mM ethanol for 1-5 days (n ϭ 4, p ϭ 0.8742, ANOVA). Therefore, we used GAPDH as an internal control to normalize all subsequent RPAs and slot blots for variation in sample loading. These studies revealed that ethanol did not increase levels of mRNA for ␣ 2 ␦ or ␤ 1b (Fig. 3).
Two alternatively spliced variants of rat ␣ 1C have been identified as ␣ 1C-1 and ␣ 1C-2 (18). The major sequence difference between these variants lies within the S3 segment of domain IV within a region that is 43% different at the nucleotide level (36 differences in 84 nucleotides). Therefore, to identify both splice variants simultaneously by RPA, we analyzed ␣ 1C mRNA using a riboprobe made from the domain IV S3 segment of ␣ 1C-2 that includes this region of mismatch at its 5Ј end (Fig. 4A). This riboprobe was predicted to protect a 498-bp fragment from ␣ 1C-2 mRNA. Mismatch within the 5Ј tail of the riboprobe was predicted to yield a fragment that is 84 bp shorter (approximately 414 bp) when complexed with ␣ 1C-1 mRNA. As expected, RPA analysis with this probe revealed two fragments, one approximately 500 bp and the other approximately 410 bp in size. Only the abundance of the shorter fragment was increased by chronic ethanol exposure (Fig. 4, B and C). Increases in this putative ␣ 1C-1 transcript were apparent within 1 day of ethanol exposure (p Ͻ 0.001, ANOVA, Newman Keuls) and persisted throughout the 6 days of treatment (p Ͻ 0.02, ANOVA, Newman Keuls). Upon removal of ethanol from the cultures, the abundance of this mRNA species declined rapidly, reaching base-line levels within 24 h. These findings suggest that chronic exposure to ethanol selectively increases the abundance of ␣ 1C-1 mRNA in PC12 cells.
To ensure that increases in the 410-bp fragment observed with the ␣ 1C-2 probe represent increases in ␣ 1C-1 mRNA, we repeated the RPA analysis with a probe made from ␣ 1C-1 cDNA. As predicted, this probe recognized a major fragment of approximately 582 bp, and exposure to 120 mM ethanol for 0 -5 days increased its abundance (data not shown). The magnitude of this increase was similar to that observed for the 410-bp fragment detected with the ␣ 1C-2 probe. This finding confirms our results in Figs. 4 and 5 suggesting that ethanol selectively increases levels of ␣ 1C-1 mRNA.
Ethanol Increases ␣ 1C-1 mRNA Levels in Cells Expressing the ␦V1 Fragment of PKC␦-If PKC␦ mediates ethanol-induced increases in ␣ 1C-1 mRNA, then ␣ 1C-1 mRNA levels should not be altered in ethanol-treated cells that express ␦V1. To examine this possibility, we treated V1␦2 cells with 120 mM ethanol for 0 -5 days and found that ethanol exposure produced a time-dependent increase in ␣ 1C-1 mRNA abundance (Fig. 5, A and B). This increase was also observed in a second ␦V1-expressing cell line V1␦3 (Fig. 5C). In addition, increases in ␦V1-expresing cells were much greater than increases observed in PC12 or vector-transfected cells (Fig. 5C). Ethanol did not increase the abundance of ␣ 2 ␦ or ␤ 1b mRNA in these ␦V1-expressing cells (Fig. 5, D and E). These results indicate that ethanol selectively increases ␣ 1C-1 mRNA levels by a PKC␦-independent mechanism. DISCUSSION Among the several calcium channel subunits that can contribute to the formation of L-type calcium channels, our clone of PC12 cells appears to express only ␣ 1C , ␣ 2 , ␤ 1b , ␤ 2 , and ␤ 3 subunits. In previous work we found that exposure of PC12 cells to 150 mM ethanol for 6 days increases DHP binding and the function of L-type calcium channels by 55-85% (28). In the current study, we found that exposure to the same concentration of ethanol for the same amount of time increases membrane-associated immunoreactivity for ␣ 1C , ␣ 2 , and ␤ 1b calcium channel subunits by 60 -75% without altering immunoreactivity for ␤ 2 or ␤ 3 subunits. Increases in ␣ 1C immunoreactivity followed a time course similar to that observed previously for increases in K ϩ -stimulated 45 Ca 2ϩ uptake in PC12 cells (7,28). Since the magnitudes of ethanol-induced increases in DHP binding, L-type channel function, and ␣ 1C , ␣ 2 , and ␤ 1b subunit immunoreactivities are similar, ethanol-induced increases Ltype calcium channels are most likely due to increases in these calcium channel subunits.
Recently we found that increases in DHP binding and L-type channel function following chronic ethanol exposure are inhibited in cells that express the selective PKC␦ inhibitor, ␦V1 (28). In the current study, we found that ethanol-induced increases in ␣ 1C , ␣ 2 , and ␤ 1b subunits are also completely blocked in ␦V1-expressing cells. Therefore, our present results are consistent with our previous findings and together indicate that PKC␦ is required for ethanol-induced increases in the density of functional L-type channels in PC12 cells. Our finding that ␤ 1b , but not ␤ 2 or ␤ 3 , was selectively increased by ethanol, suggests that ethanol, via PKC␦, specifically recruits ␤ 1b subunits to newly formed L-type channel complexes. Exposure to ethanol did not alter ␤ 1b mRNA, indicating that ethanol regulates the abundance of ␤ 1b by post-transcriptional mechanisms. ␤ subunits contain consensus sites for PKC phosphorylation, and phosphorylation of ␤ subunits by PKC has been proposed to regulate L-type channel function (48,49). Further studies will be required to determine whether PKC␦ selectively phosphorylates and regulates ␤ 1b protein turnover or trafficking in ethanoltreated cells.
Ethanol exposure also selectively increased mRNA for ␣ 1C-1 without altering levels of mRNA for the alternative splice variant, ␣ 1C-2 , or for other channel subunits. This is the first report of ethanol regulating the abundance of a specific mRNA splice variant. The most striking difference between ␣ 1C-1 and ␣ 1C-2 is a 13-aa substitution within a 28-aa region corresponding to the S3 segment of transmembrane domain IV. Although most of the 13 substitutions within this region are conservative, one exception is the substitution of a proline in ␣ 1C-1 for an alanine in ␣ 1C-2 at the amino terminus of the S3 segment. Substitutions located within or near this region may regulate channel gating (50). Electrophysiological studies with expressed ␣ 1C-1 and ␣ 1C-2 subunits will be required to investigate this possibility.
Increases in ␣ 1C-1 were also observed in ␦V1-expressing cells, suggesting that ethanol increases levels of ␣ 1C-1 mRNA by PKC␦-independent mechanisms. Neither ␣ 2 ␦ nor ␤ 1b mRNA abundance was altered by ethanol treatment in these cell lines, indicating that the response is specific for ␣ 1C-1 . Increases in ␣ 1C-1 mRNA could be due to ethanol-induced changes in the splicing of ␣ 1C transcripts leading to greater production of ␣ 1C-1 mRNA. Alternatively, ethanol may act to decrease ␣ 1C-1 mRNA degradation. Why these effects of ethanol should be specific for ␣ 1C1 mRNA is unknown and requires further study.
In the parent PC12 cell line, ␣ 1C-1 mRNA levels were increased after 1 day of ethanol exposure and remained nearly constant as long as ethanol was present. In contrast, in ␦V1expressing cells, ethanol induced a much greater rise in ␣ 1C-1 mRNA, which continued to increase over the 5 days of ethanol exposure. These results suggest that PKC␦ normally acts to limit ␣ 1C-1 mRNA abundance, possibly by regulating ␣ 1C mRNA splicing or by promoting degradation of ␣ 1C-1 mRNA. This could be due to a direct effect of PKC␦ on ␣ 1C mRNA processing or might occur indirectly, if PKC␦-mediated increases in calcium channel proteins evoke homeostatic mechanisms that decrease ␣ 1C-1 mRNA abundance.
Additional splice variants of human ␣ 1C have been identified (51). Alternative splicing of exons 21 and 22 of the human ␣ 1C gene produces splice variants in the S2 segment of transmembrane domain III, and these show differences in the voltage sensitivity of inhibition by DHP antagonists (52). Alternative splicing in the cytoplasmic tail alters the kinetics and the calcium dependence of channel inactivation (53). Therefore, FIG. 5. Ethanol exposure increases mRNA levels of an ␣ 1C-1 in PC12 and ␦V1-expressing cells. A, representative RPA of ␣ 1C and GAPDH mRNA using the ␣ 1C-2 cRNA 32 P-labeled probe. Undigested ␣ 1C and GAPDH mRNA controls (U␣ and UG), digested control (D), and V1␦2 mRNA treated for 0 -5 days with 120 mM ethanol. B, the abundance of mRNA for ␣ 1C-2 (E) and ␣ 1C-1 (q) after exposure to 120 mM ethanol for 0 -5 days in V1␦2 cells. Data are expressed as mean Ϯ S.E. values from three experiments. *, p Ͻ 0.046 compared with day 0 by ANOVA and Newman Keuls tests. C, the abundance of ␣ 1C-1 mRNA after exposure to 120 mM ethanol for 5 days, in PC12 cells (PC), vector transfected cells (C), and cells expressing the V1 fragment of PKC␦ (V1␦2 and V1␦3). Data are expressed relative to parallel control cells cultured without ethanol and are mean Ϯ S.E. values from three experiments. *, p Ͻ 0.023 compared with PC12 cells or vector-transfected cells (ANOVA and Newman Keuls tests). D, the abundance of ␣ 2 ␦ mRNA after exposure to 120 mM ethanol for 5 days in V1␦2 (⌬) and V1␦3 cells (OE) expressing the V1 fragment of PKC␦. Data are expressed relative to parallel control cells cultured without ethanol and are mean Ϯ S.E. values from three experiments. p Ͼ 0.632 by one-way ANOVA. E, the abundance of ␤ 1b mRNA after exposure to 120 mM ethanol for 5 days in V1␦2 (Ⅺ) and V1␦3 (f) cells. Data are expressed relative to parallel control cells cultured without ethanol and are mean Ϯ S.E. values from three experiments. p Ͼ 0.564 by one-way ANOVA.
alternative splicing of ␣ 1C transcripts can confer distinct functional characteristics to neuronal L-type channels. Ongoing studies will investigate whether similar splice variants can be identified in rat neural cells and whether ethanol regulates their abundance.
Our results provide additional evidence for PKC␦ as a regulator of L-type channel density and a mediator of cellular adaptation to ethanol. Our findings also indicate that PKC␦ acts via post-transcriptional mechanisms to increase the density of L-type channels. In addition, our results provide evidence for a PKC␦-independent mechanism leading to increases in ␣ 1C-1 mRNA that may also contribute to ethanol-induced up-regulation of L-type channels. However, PKC␦-dependent mechanisms appear to be essential and rate-limiting for increases in L-type channels, since inhibition of PKC␦ completely prevents ethanol-induced increases in DHP binding and L-type channel function (28). Since antagonists of L-type channels decrease alcohol self-administration (13,15,16) and reduce manifestations of alcohol withdrawal (11,12,54,55), PKC␦, through its actions on L-type channels, may play a key role in regulating alcohol consumption and dependence.