Ethanol-induced Phosphorylation and Potentiation of the Activity of Type 7 Adenylyl Cyclase

Ethanol can enhance Gsα-stimulated adenylyl cyclase (AC) activity. Of the nine isoforms of AC, type 7 (AC7) is the most sensitive to ethanol. The potentiation of AC7 by ethanol is dependent on protein kinase C (PKC). We designed studies to determine which PKC isotype(s) are involved in the potentiation of Gαs-activated AC7 activity by ethanol and to investigate the direct phosphorylation of AC7 by PKC. AC7 was phosphorylated in vitro by the catalytic subunits of PKCs. The addition of ethanol to AC7-transfected HEK 293 cells increased the endogenous phosphorylation of AC7, as indicated by a decreased “back-phosphorylation” of AC7 by PKCin vitro. The potentiation of Gαs-stimulated AC7 activity by either phorbol 12,13-dibutyrate or ethanol, in HEL cells endogenously expressing AC7, was not through the Ca2+-sensitive conventional PKCs. However, the potentiation of AC7 activity by ethanol or phorbol 12,13-dibutyrate was found to be reduced by the selective inhibitor of PKCδ (rottlerin), a PKCδ-specific inhibitory peptide (δV1-1), and the expression of the dominant negative form of PKCδ. Immunoprecipitation data indicated that PKCδ could bind and directly phosphorylate AC7. The results indicate that the potentiation of AC7 activity by ethanol involves phosphorylation of AC7 that is mediated by PKCδ.

Intracellular signaling via cAMP generates downstream effects that range from changes in the function of ion channels to changes in intracellular energy metabolism to changes in gene transcription (for review see Ref. 1). It is therefore not surprising that the generation of intracellular cAMP is a tightly regulated process that involves the ␣ and ␤␥ subunits of G-proteins, intracellular Ca 2ϩ acting independently or in concert with calmodulin, and phosphorylation events that are postulated to involve protein kinase A and protein kinase C (PKC) 1 (2,3). The current evidence indicates that each of the nine adenylyl cyclase (AC) isoforms differs in its repertoire of regulatory controls (2,3), and all of the isoforms can be regulated coincidentally by multiple signals to modulate the production of cAMP.
It was therefore not entirely unexpected that the activity of the various isoforms of AC were also found to be differentially sensitive to the effect of ethanol (4). We and others have shown that ethanol acutely potentiates G␣ s -activated AC activity (5,6), and studies with HEK 293 cells transfected with the various isoforms of AC have demonstrated a broad range of AC sensitivity to ethanol. Some isoforms were found to be insensitive to ethanol (types 1, 3, 8a, and 8b), others were moderately sensitive (types 2, 5, 6, 8c, and 9), and Type 7 AC was at least two to three times more sensitive to the stimulatory actions of ethanol in comparison with all other tested ACs (4). Agonist-stimulated AC7 activity in transfected HEK 293 cells can be significantly potentiated by 10 -20 mM ethanol (which corresponds to tissue concentrations of 46 -92 mg/100 ml ethanol), and the potentiation by ethanol of AC7 activity is not mediated through the inhibition of phosphodiesterase activity or an adenosine receptor-mediated event (7). Ethanol has been shown to further potentiate AC activity when AC is activated through a variety of different membrane receptor systems, including dopamine D1a, ␤-adrenergic, and prostaglandin receptors (4,5,8). These data indicate that the ability of ethanol to potentiate cAMP accumulation is independent of transmitter receptors and is dependent primarily on the AC isoform present in the cell. cAMP signaling has been shown to be important in the behavioral effects of ethanol through studies with Drosophila (9) and mice (10,11). Mutations in the components of the cAMP-generating or degrading systems in Drosophila significantly altered the responses of the fly to the intoxicating (incoordinating) effects of ethanol (9). A similar relationship between ethanol-induced incoordination and changes in cAMP generating systems was found in both mice and rats (10,12). Activation of cAMP generation in the brains of mice, by the use of forskolin, altered the development of tolerance to the sedative effects of ethanol (11). Clinical studies with humans have also linked the cAMP signaling system to behavioral manifestations of physical dependence on ethanol. Alcohol-dependent individuals, even when abstinent for substantive periods of time, have been shown to have lower G-protein-activated AC activity in their platelets than control subjects (13). More recently it was found that low platelet AC activity was characteristic of individuals classified as family history positive for alcoholism, regardless of whether the subjects were themselves diagnosed alcoholics (14,15). Because ethanol has been demonstrated to enhance receptor/G-protein-coupled generation of cAMP in platelets, the depressed AC activity in the platelets of individuals at risk to develop alcoholism may parallel what is occurring in the brain and may contribute to the addiction process (16).
We have recently shown that in human erythroleukemia (HEL) platelet precursor cells, which contain a preponderance of AC7 mRNA (17), both ethanol and phorbol esters (PDBu and phorbol 12-myristate 13-acetate) could significantly potentiate AC activity generated in response to the activation of the prostanoid receptor by the agonist PGE 1 (8). This potentiation of AC activity by either ethanol or PDBu could be diminished by PKC inhibitors such as staurosporine and chelerythrine (8). In the current work, we report on a series of studies that determine the PKC isotype(s) involved in the enhancement of G␣ sactivated AC7 activity by ethanol, and we investigate the phosphorylation of AC7 by PKC.

EXPERIMENTAL PROCEDURES
Materials- [2-3 H]Adenine was obtained from Amersham Biosciences. 3-Isobutyl-1-methylxanthine, phorbol 12,13-dibutyrate (PDBu), staurosporine, chelerythrine, rottlerin, thapsigargin, and Gö 6976 were purchased from Calbiochem (La Jolla, CA). Anti-PKC monoclonal antibodies were obtained from Transduction Laboratories (Lexington, KY). Dr. W.-J. Tang (University of Chicago) kindly provided the anti-AC II family antibody, C6C. PGE 1 was obtained from the Cayman Chemical Company (Ann Arbor, MI). Cicaprost was a generous gift from Schering AG (Berlin, Germany). All of the other products were purchased from Sigma. Concentrated solutions of drugs, including PGE 1 and cicaprost, were prepared in Me 2 SO, and the final concentration of Me 2 SO in the assay mixtures was never greater than 0.8%. Control assays were always performed containing the appropriate amounts of Me 2 SO. Dr. Daria Mochly-Rosen (18) kindly provided the following PKC-derived inhibitory peptides: ␦V1-1 (␦PKC 8 -16 , AFN-SYELGS), ⑀V1-2 (⑀PKC 14 -21 , EAVSLKPT), and a control nonsense octapeptide (LSETKPAV). These peptides were conjugated to a Drosophila antennapedia peptide (RQIKIWFQNRRMKWKK) to make them more cell permeant. Catalytic subunits of PKC purified from rat brain were prepared by trypsin digestion by Dr. Michael D. Browning (University of Colorado Health Sciences Center, Denver, CO). Dr. Trevor Biden (Garvan Institute of Medical Research, Sydney, Australia) kindly provided the replication deficient adenovirus (Ad5 DL312) vectors carrying either the wild-type or dominant negative PKC␦. Fulllength recombinant human PKC␦ and PKC⑀ were purchased from Calbiochem (La Jolla, CA).
Construction of the T7-tagged AC7-A BamHI site was introduced upstream of the initiation codon of the human AC7 cDNA (19) by in vitro mutagenesis using an oligonucleotide, CGTGCCAAGGATCCG-GAGGATGCCAG. The mutation was confirmed by DNA sequencing. A 3.6-kb fragment containing the human AC7 coding sequence was prepared from pBlueScript II SK-containing the AC7 cDNA by digestion with BamHI and XbaI. The fragment was inserted into a pcDNA3.1 His (Invitrogen) mammalian expression vector after digestion of this vector with the restriction enzymes BamHI and XbaI. The constructed vector provides an N-terminal fusion of hexahistidine tag and T7 epitope with AC7 under the control of the cytomegalovirus promoter.
The recombinant baculovirus for AC7 was constructed as follows. cDNA for human AC7 (19) was subcloned into the BamHI and BglII sites of an Invitrogen pBlueBacHis2 vector (frame C). Sf9 cells were cotransfected with the AC7 plasmid and wild-type viral DNA, and positive recombinant viral plaques were isolated. When expressed in the insect cells, this vector generated a fusion protein including AC7 that contains N-terminal (His) 6 , T7, and Xpress™ epitopes. The fusion junction was verified by automated sequencing. The control vector was generated from the T7-AC7 vector by creating a stop codon in the multiple cloning site of the vector and the control vector produced only the fusion tag.
Cell Culture of HEL and HEK 293 Cells and Expression of Tagged AC7 in HEK 293 Cells-HEL cells (American Type Culture Collection, Manassas, VA) were grown in suspension culture in RPMI 1640 medium (Invitrogen) containing 10% heat-inactivated, charcoal-stripped fetal calf serum (Gemini-Bio-Products, Calabasas, CA). HEL cells were maintained at 37°C (5% CO 2 ) and used for experiments at a density range of between 0.2 and 0.4 ϫ 10 6 cell/ml. HEK 293 cells were grown in 20 mM HEPES-minimal essential medium containing 10% fetal bovine serum at 37°C (5% CO 2 ). pcDNA3.1-T7.His-AC7 and the control pcDNA3.1-T7.His vector were transfected in HEK 293 cells using Effectene reagent (Qiagen, Valencia, CA) while maintaining the cells in the regular growth medium. After incubation with the transfection reagents for 48 h, the cells were maintained in serum containing minimal essential medium for another 24 h prior to harvesting the cells.
Expression of Tagged AC7 in Sf9 Cells-AC7 was expressed in Sf9 cells using the baculovirus expression system. The Sf9 cells were cultured in a spinner vessel containing Grace's insect cell medium (Invitrogen) and 10% fetal bovine serum at 27°C (Tissue Culture Core/University of Colorado Cancer Center). Sf9 cells (300 ϫ 10 6 ) were then infected with the positive recombinant baculovirus for 60 min in serum-free Grace's insect medium followed by an additional 48 h incubation in Grace's insect medium containing 10% fetal bovine serum. AC7-expressing Sf9 cells were identified based on Western blot analysis of total Sf9 cell protein using a monoclonal anti-T7 antibody (Novagen, Madison, WI). No signals in the molecular mass range of full-length AC7 were detected by the anti-T7 antibody in solubilized protein preparations from control vector-infected Sf9 cells.
HEK 293, HEL, and Sf9 Cell Membrane Preparations-For the "back-phosphorylation" experiments, T7-tagged AC7-transfected HEK 293 or control transfected HEK 293 cells were pretreated with drugs or other reagents prior to harvesting as indicated in the figure legends and the text. For harvesting, the cells were resuspended in ice-cold membrane lysis buffer containing 20 mM HEPES, pH 7.4, 2 mM Na 2 EDTA, 150 mM NaCl, 0.2 mM Na 3 VO 4 , 10 mM ␤-glycerophosphate, 2 mM NaF, and Calbiochem Protease Inhibitor Mixture Set III (l mM AEBSF, 0.8 M aprotinin, 50 M bestatin, 15 M E-64, 20 M leupeptin, and 10 M pepstatin A). The cell suspension was sonicated briefly on ice and centrifuged at 500 ϫ g for 10 min at 4°C. The resulting supernatant was then centrifuged at 90,000 ϫ g for 45 min at 4°C, and the pellet was solubilized as described below.
For the isolation of HEL or HEK 293 cell membranes in experiments not involving back-phosphorylation, a minimum of 15 ϫ 10 6 cells were washed twice with ice-cold phosphate-buffered saline. The pelleted cells were resuspended at 5-6 ϫ 10 6 cells/ml of lysis buffer containing 50 mM Tris, pH 7.6, 2 mM MgCl 2 , 0.1 mM Na 3 VO 4 , 10 mM ␤-glycerophosphate, 1 mM benzamidine, 10 mM dithiothreitol, and Calbiochem Protease Inhibitor Mixture Set III. The cell suspension was drawn through a 22-gauge syringe 10 times, while the preparation was kept on ice and then centrifuged at 500 ϫ g for 3 min at 4°C. The resulting supernatant was then centrifuged at 90,000 ϫ g for 60 min at 4°C. Both the pellet (membrane protein) and supernatant (cytosolic protein) were saved. For immunoprecipitation (phosphorylation) experiments and Western blot analysis, the membrane protein was solubilized in 1.0% SDS. The samples were then incubated for 20 min at 80°C and sonicated briefly. The protein concentration was determined using the BCA method (Pierce).
Membranes from T7-AC7-expressing Sf9 cells or control vector-expressing Sf9 cells were harvested by centrifugation (2000 ϫ g, for 5 min at 4°C), suspended in 20 ml (per 100 ml of culture pellet) of 20 mM HEPES, pH 7.8, 500 mM NaCl, 5 mM EDTA, 1 mM EGTA, 2 mM dithiothreitol, and protease inhibitors (as described above). The samples were freeze-thawed twice (from liquid nitrogen into a 42°C water bath), and the DNA was sheared by passing the preparation through an 18-gauge needle four times. The cell debris was removed by centrifugation at 500 ϫ g for 10 min at 4°C. The supernatant was then centrifuged at 100,000 ϫ g for 40 min at 4°C, and the membrane pellet was resuspended in 20 mM HEPES, pH 7.8, 200 mM sucrose, 1 mM dithiothreitol plus protease inhibitors, as described above. The suspension was then recentrifuged at 100,000 ϫ g for 40 min at 4°C, and the final pellet was solubilized in 1.0% SDS and stored at Ϫ80°C.
Immunoprecipitation, Phosphorylation, and Back-phosphorylation of AC7-For the immunoprecipitation of T7-tagged AC7, all of the steps were carried out at 4°C, except when noted. In each sample, 150 -200 g of solubilized total Sf9 or HEK 293 cell membrane protein (AC7 and control vector-transfected) in 1% SDS was diluted with the IP buffer (40 mM Tris-HCl, pH 7.8, 100 mM NaCl, 5 mM Na 2 EDTA, 0.4 mM MgCl 2 , 2 mM methionine, 10 mM NaF, 1 mM Na 3 VO 4 , 20 mM sodium pyrophosphate, 25 mM Na 2 -␤-glycerophosphate, and protease inhibitors as described above, and 1.2% Nonidet P-40) to a final SDS concentration of 0.08 -0.12% (a minimum 10-fold Nonidet P-40/SDS ratio). The samples were incubated for 5 min with washed (IP buffer) protein A-agarose (PAA) beads (ImmunoPure immobilized protein A; Pierce) at a ratio of 10 -20%:80 -90% beads to solubilized protein in IP buffer, respectively, to remove material that bound nonspecifically to the beads. The beads were pelleted by centrifugation (10,000 ϫ g), and the supernatants were collected and shaken with 6 g of monoclonal anti-T7 antibody (Novagen) for 2 h at room temperature and then incubated with washed (IP buffer) PAA beads for 1 h. The antigen/antibody/PAA conjugates were pelleted (10,000 ϫ g) and washed once with IP buffer. The beads were then washed twice at 22°C with phosphorylation buffer (50 mM Tris-HCl, pH 8.0, 5 mM MgCl 2 , 1 mM EGTA, 0.1% Triton X-100) and incubated for 2-5 min at 37°C in 40 l of phosphorylation buffer including 0.3 M [␥-32 P]ATP (6000 Ci/mmol; PerkinElmer Life Sciences) and 250 nM of the constitutively active PKC catalytic subunit purified from rat brain (20). The reaction was terminated with 10 l of 200 mM Na 2 EDTA, pH 8.0. The beads were pelleted and washed two times with IP buffer. The samples were then reduced and alkylated to produce a tight band of AC during the PAGE. The sample was reduced by incubation in 40 mM dithiothreitol at 80°C for 15 min. After a 15-min incubation at room temperature with 65 mM N-ethylmaleimide, loading dye was added, and the sample was briefly boiled before the PAGE. The precast 8% Tris-glycine polyacrylamide gels were from Novex/Invitrogen (Carlsbad, CA). After the electrophoresis, the gel was dried, and Kodak X-Omat Blue film was used for the autoradiography.
For the phosphorylation by full-length recombinant PKC, T7-tagged AC7-transfected HEK 293 cells or control transfected HEK 293 cells were lysed in ice-cold membrane lysis buffer and a Triton X-100-insoluble plasma membrane fraction (representing a membrane microdomain thought to have characteristics similar to lipid rafts and enriched in signaling molecules) was isolated. This membrane fraction was isolated as described above under "HEK 293, HEL, and Sf9 Cell Membrane Preparations" with the following additional modification; in the final step the non-nuclear cell lysate was centrifuged at 90,000 ϫ g in the presence of 0.5% Triton X-100 for 60 min at 4°C. This membrane pellet was solubilized in 1% SDS, and the T7-tagged AC7 was immunoprecipitated with an anti-T7 monoclonal antibody with the addition of PAA beads (as described above). The immunoprecipitates from T7-AC7 and control transfected HEK 293 cells were then incubated for 30 min at 30°C with recombinant PKC␦ (0.8 specific activity units) or PKC⑀ (1.3 specific activity units) and 10 Ci of [␥-32 P]ATP (3000 Ci/ mmol from PerkinElmer Life Sciences; final ATP concentration equaled 40 M) in phosphorylation buffer (40 mM Tris-HCl, pH 7.4, 10 mM MgCl 2 , 0.2 mM CaCl 2 , 1 mM dithiothreitol, 25 mM ␤-glycerol phosphate, 1 mM Na 3 VO 4 , 2 g/ml phosphatidylserine, 0.2 g/ml diolein, 10 M PDBu, and 0.02% Triton X-100). The reaction mixture was pelleted to separate the T7-AC7/anti-T7 antibody/PAA immunocomplex from the supernatant containing PKC and [␥-32 P]ATP. The pellet was washed with 40 mM Tris-HCl, pH 8.0, and the supernatant was precipitated with trichloroacetic acid. Aliquots from both the pelleted fraction and the supernatant were dithiothreitol-and N-ethylmaleimide-treated, boiled in gel loading buffer, and loaded onto an 8% Tris-glycine gel for electrophoretic separation. The gel was dried and then exposed to x-ray autoradiography film.
PKC Binding Assay-Recombinant purified PKC␦ or PKC⑀ (2 g) was 32 P-labeled through autophosphorylation for 30 min at 30°C in the presence of 10 Ci of [␥-32 P]ATP (3000 Ci/mmol from PerkinElmer Life Sciences; final ATP concentration equaled 40 M) in phosphorylation buffer. After this reaction, the 32 P-labeled PKC was trichloroacetic acid-precipitated (10%) on ice, and the free 32 P was removed with three washes in 95% ice-cold ethanol. 32 P-Labeled PKC was resuspended at 100 ng/l in 50% glycerol buffer (100 mM NaCl, 2 mM EGTA, 2 mM EDTA, 0.05% Triton X-100, and 5 mM TCEP adjusted to pH 7.8 with HEPES).
The Triton X-100-insoluble plasma membrane fraction (100 g) from control or AC7-transfected HEK cells was suspended in IP binding buffer A (20 mM HEPES, pH 7.8, 100 mM NaCl, 1 mM Na 2 EDTA, 0.4 mM MgCl 2 , 1 mM CaCl 2 , 2 mM methionine, 1 mM Na 3 VO 4 , 20 mM sodium pyrophosphate, 25 mM Na 2 -␤-glycerophosphate, 1.2% Nonidet P-40, 0.5 g/ml ␥-globulin free BSA, and protease inhibitors as described above). The membrane protein mixture was then added to washed PAA beads for 5 min at 4°C to remove the nonspecific binding proteins. The beads were discarded, and the supernatant was incubated overnight at 4°C with 4 g of anti-T7 antibody. The solution containing the T7-AC7⅐anti-T7 immunocomplex was added to washed PAA beads and incubated for 120 min at 4°C. The beads containing the immunocomplex were pelleted and washed once in IP binding buffer A to remove residual SDS. The beads were then resuspended in IP binding buffer A, and 100 ng of 32 P-PKC␦ or ⑀ was added and incubated at 4°C for 120 min. The beads containing the immunocomplex and bound 32 P-PKC were pelleted and washed twice in IP binding buffer A. The supernatant and washes were combined and counted along with the pellets. 32 P-PKC binding to PAA beads in the absence of T7 antibody or membrane protein was used for determination of nonspecific binding of 32 P-PKC and was subtracted as background from the pelleted samples.
Assay of Adenylyl Cyclase Activity in HEL Cells-For whole cell cAMP synthesis measurements, HEL cells were preloaded with 2 Ci/ml [2-3 H]adenine in HEL culture medium for 6 h at 37°C as previously described (4,22). At the end of the incubation, HEL cells were pelleted by centrifugation (200 ϫ g for 5 min), washed, and resuspended at a cell concentration of 1 ϫ 10 6 /ml in serum-free RPMI 1640, without phenol red, supplemented with 20 mM HEPES, pH 7.4 (assay buffer). Aliquots (0.4 ml) of the cell suspension were added to each well (24-well plates) and allowed to equilibrate for 30 min at 36°C before the start of an assay. In experiments that utilized the PKCderived inhibitory peptides, the assay buffer was modified slightly as described below to prevent premature reduction of the chimeric peptide (the PKC-derived/antennapedia peptides contain a Cys-Cys disulfide bond) prior to entry into the cells. Briefly, HEL cells were extensively washed three times (to remove exogenous glutathione) and resuspended at a cell concentration of 1 ϫ 10 6 /ml in glutathione-free, serumfree RPMI 1640 supplemented with 20 mM HEPES, pH 7.4, and 0.05% BSA. PKC-derived inhibitory peptides were added (final concentration of 2.0 M) to the cell suspension and incubated at 37°C for 2 h prior to the start of the experiment. cAMP formation was measured by monitoring the conversion of [ 3 H]ATP to [ 3 H]cAMP. The cells were treated with the phosphodiesterase inhibitor, 3-isobutyl-1-methylxanthine (400 M), for 10 min prior to the addition of agonist. After the addition of agonist (PGE 1 or cicaprost), cAMP formation was allowed to continue for 5 min before the reaction was terminated with trichloroacetic acid (final concentration of 10%). Other modulators of AC activity were added at concentrations and time points as described in the text and figure legends. ATP and cAMP were separated by sequential chromatography on Dowex 50 and neutral alumina columns and quantitated using a Beckman LS 6000TA liquid scintillation counter as previously described (4,22). The amount of [ 3 H]cAMP produced in the assay was calculated and expressed as a fraction of the available pool of [ 3 H]ATP as previously described (4,22).
Fluorimetric Measurement of Intracellular Ca 2ϩ Concentration, [Ca 2ϩ ] i -To measure [Ca 2ϩ ] i in HEL cell populations, 2 ϫ 10 6 HEL cells were washed and resuspended in serum free RPMI 1640 supplemented with 20 mM HEPES, pH 7.4, and 0.1% BSA. The extracellular Ca 2ϩ concentration in the RPMI 1640 assay buffer was 0.4 mM. The cells were then incubated with 6 M Fura-2/AM for 45 min at room temperature to minimize vesicle sequestration and efflux of Fura-2/AM. At the end of the loading period, the medium containing Fura-2/AM and other drugs, if present, was removed. The cells were pelleted and resuspended in assay buffer containing 0.1% BSA and allowed to equilibrate for 5-10 min. Cells at a density of 1 ϫ 10 6 cells/ml were placed in a thermostatted cuvette maintained at 37°C and kept in suspension with a magnetic stir bar. Fluorescence was measured using an SLM-Aminco dual wavelength spectrofluorometer (excitation at 345 and 380 nm; emission at 505 nm) with a time resolution of 0.5 s. The data were stored on a NEC 286 computer, using software from SLM-Aminco (Urbana, IL). Intracellular calcium concentrations were calculated from the relationship [Ca 2ϩ ] i ϭ K d ϫ (r Ϫ R min )/(R max Ϫ r) ϫ (380 min/380 max) using 224 nM as the K d for the calcium complex of Fura-2 at 37°C (23). The R max was determined in the presence of 0.1% Triton X-100, and the R min was determined in the presence of Ca 2ϩ free buffer containing 3 mM EGTA. All of the traces represent individual experiments, but all of the determinations were repeated at least on two separate occasions. The reagents were added in volumes of 2-20 l to give the final concentrations noted in the text and figure legends.
Adenovirus Propagation, Purification, and Infection of HEL Cells-The Ad5 DL312 replication deficient adenovirus was propagated in HEK 293 cells as follows. Thirty T150 mm flasks of HEK 293 cells suspended in 5 ml of serum-free Dulbecco's modified Eagle's medium were infected using a solution containing diluted virus at a multiplicity of infection of 10 -25 (based on a plaque formation assay). After 60 min, 20 ml of 2% fetal bovine serum/Dulbecco's modified Eagle's medium was added to the cells, and they were allowed to grow for an additional 48 h. The cells were pelleted and combined with serum-free RPMI medium (10 mM HEPES, pH 8.0; final volume of 5 ml). The cell suspension was frozen (liquid nitrogen) and then thawed three times to lyse the cells and release the adenovirus. Adenovirus particles were purified from HEK 293 cell debris as follows. The suspension containing virus was sonicated (low energy; Ͻ70% of microtip limit) for 5 min on ice. Sodium deoxycholate was added to the virus suspension (final concentration of 0.1%), and the solution was then sonicated for 5 min on ice. An equal volume of prechilled (Ϫ20°C) chloroform was added to the virus suspension followed by sonication for 10 min on ice at the interphase to form a viscous emulsion. The emulsion was separated by a 15-min centrifugation (setting #6 on IEC clinical centrifuge), and the upper aqueous phase was isolated from the lower chloroform phase. The recovered aqueous phase was added to an equal volume of prechilled (Ϫ20°C) chloroform. The mixture was sonicated (ϳ5 min on ice) at the interphase until viscous. The upper aqueous phase was again separated by centrifugation and then overlaid on a two-step CsCl gradient consisting of 1.25 and 1.4 g/ml CsCl. The virus was banded by centrifugation at 95,000 ϫ g for 2 h in a Beckman L-90K Ultrafuge. The white virion band was collected at the gradient interface (lower band; the top band is empty capsids) by sterile side puncture. The virion band was dialyzed twice for periods of 1 h each against 500 ml of virion dialysis buffer (10 mM HEPES, pH 8.0, 1 mM EDTA, in phosphate-buffered saline) in a 10-kDa molecular weight cut-off slide-A-lyzer (Pierce) and then dialyzed against virion dialysis buffer in 50% glycerol for 1 h and stored at Ϫ20°C.
For infection, HEL cells were plated at 0.1-0.2 ϫ 10 6 cells/ml in fresh RPMI 1640 culture medium 1 day prior to the addition of adenovirus. The cells were pelleted and resuspended in 5 ml of Opti-MEM, 20 mM HEPES, pH 7.4, and were infected at a 5 ϫ 10 4 virus particles/cell during a 3-h incubation at 37°C. The cells were then pelleted, washed, and resuspended in RPMI 1640 culture medium and allowed to continue growing for 2 days prior to the assay. In preliminary experiments, a replication deficient adenovirus carrying the green fluorescent protein was used to verify the ability of the virus to infect HEL cells (data not shown).
Statistics-All of the statistical analyses were performed using the Sigmastat program (Jandel Scientific Software). The data are presented as the means Ϯ S.E. unless otherwise noted. p values of Ͻ0.05 were taken as statistically significant.

RESULTS
To determine whether AC7 protein could be phosphorylated by PKC, an N-terminal tagged T7 and His epitope AC7 expression vector was constructed that would allow for efficient purification and identification of AC7 by immunoprecipitation. There were no PKC consensus sites within or created by the addition of the T7 tag to the AC7 protein. As seen in Fig. 1A, the catalytic subunits of rat brain PKC phosphorylated a protein (that corresponded in size to AC7) from the membranes of Sf9 cells that were infected with the T7-AC7 vector but not the T7 control vector. This protein was also enriched bys immunoprecipitation (IP) using the T7 monoclonal antibody (Fig. 1A). Identification of this protein from infected Sf9 cells, as AC7, was accomplished by Western blot analysis using the T7 monoclonal antibody to the N-terminal T7 tag and a goat anti-mouse horseradish peroxidase-conjugated secondary antibody (Fig.  1A). The protein identified as AC7 by Western blot analysis corresponded in size to the protein that was phosphorylated by the rat brain PKC mixture (Fig. 1A).
To determine whether AC7 could be phosphorylated by PKC within live cells, the back-phosphorylation of AC7 after various treatments of transfected cells was assessed. The backphosphorylation method allows for the quantitation of the remaining phosphorylation sites that can be phosphorylated by a particular kinase in vitro after endogenous phosphorylation within live cells has occurred (24). If a treatment of live cells increases the PKC-mediated phosphorylation of AC7, then the subsequent in vitro incorporation of 32 P, catalyzed by exogenous PKC, will be diminished because a number of PKC-sensitive sites will have already been phosphorylated within the live cell. As shown in the autoradiogram in Fig. 1B, less 32 P-labeled phosphate was incorporated into AC7 (identified by immunoblotting) during the in vitro PKC phosphorylation under conditions where live cells were pretreated with PGE 1 (10 M) and ethanol (100 or 200 mM) compared with assays in which cells were treated only with PGE 1 . The extent of back-phosphorylation of AC7 harvested from T7-AC7-transfected HEK 293 cells was assessed in three separate experiments (including the representative autoradiograph shown in Fig. 1B). The densitometric measurements of these experiments were combined to calculate the intracellular PKC-sensitive phosphorylation of AC7 after exposure to ethanol. Using the ratio of back-phosphorylation units (densitometric analysis of 32 P-incorporation) for AC7 from untreated HEK 293 cells divided by the back-phosphorylation units for AC7 from ethanol-treated cells, we determined that the addition of 100 mM ethanol produced a back-phosphorylation ratio of 1.44 Ϯ 0.31 and that the presence of 200 mM ethanol increased this ratio to 2.24 Ϯ 0.47 (Fig. 1B). Because we also wished to investigate the actions of PKC and ethanol on AC7 in nontransfected cell systems, we chose to use the HEL cells from which we have previously isolated AC7 mRNA (17). The Ca 2ϩ -sensitive adenylyl cyclases (cyclases 1, 5, 6, and 8) have been shown to be regulated by store-operated Ca 2ϩ influx (25). To induce store-operated Ca 2ϩ influx, HEL cells were depleted of intracellular Ca 2ϩ with 100 nM thapsigargin (15 min of pretreatment) in the absence of external Ca 2ϩ . External Ca 2ϩ (5 mM) was then added simultaneously with 10 M PGE 1 , and the cAMP accumulation was measured for 5 min. Fig. 2A shows that the cAMP accumulation in the presence of 10 M PGE 1 in normal calcium containing media (0.9 mM Ca 2ϩ ) did not differ from the cAMP accumulation under conditions where store-operated Ca 2ϩ influx was induced, indicating that HEL cells have little or no Ca 2ϩ -regulated AC activity. In contrast, a 30-min pretreatment with the non-isoform-selective PKC inhibitor chelerythrine, at 20 M, significantly reduced the PGE 1 -induced cAMP accumulation ( Fig. 2A). When HEL cells were pretreated for 30 min with increasing concentrations of chelerythrine prior to the addition of 10 M PGE 1 , both the ethanol (200 mM) and PDBu (100 nM) enhancement of PGE 1 -stimulated AC activity in HEL cells was inhibited in a dose-dependent fashion (Fig. 2B). The dose re-sponse of inhibition by chelerythrine did not demonstrate any preferential sensitivity between ethanol and PDBu effects, and, at 20 M, chelerythrine almost completely inhibited the potentiation by either ethanol or PDBu. This AC activity data, along with our previous results based on the measurements of AC mRNA in HEL cells (17) suggests that the predominant AC activity in HEL cells is an AC that is nonresponsive to changes in intracellular Ca 2ϩ but is sensitive to ethanol in a PKC-dependent manner.
Using the C6C anti-AC antibody, an immunoreactive band in the range of 110 kDa was identified in the membrane fraction of HEL cells that corresponded in size to an immunoreactive band from T7-tagged AC7-infected Sf9 cells (data not shown) and the T7-tagged AC7-transfected HEK 293 cells. In addition, both the anti-T7 antibody and the anti-AC antibody recognized the same band on immunoblots for both the Sf9 (data not shown) and HEK 293 cell-derived T7-tagged AC7 (see Fig. 8A). The C6C type II family anti-AC monoclonal antibody was produced to recognize a region in the C2 domain of the type II family of ACs: AC2, AC4, and AC7 (21). Because we have previously demonstrated that HEL cells lack mRNA for AC2 and 4 (17), one can conjecture that the major AC in HEL cells is AC7.
To differentiate which PKC isotypes might be present in HEL cells and thus better focus our search for the particular PKC involved in the potentiation by ethanol, we used PKC isotype-specific antibodies. The results from the Western blot analysis shown in Fig. 3 reveal that HEL cells express ␤, , ␦, , , and () PKCs. The ␣ and ␥ isotypes were not expressed at detectable levels in HEL cells, and only after very long exposures did a faint band in the molecular mass range for PKC⑀ appear (data not shown). We did not test for the presence of the PKC isotype.
PKC inhibitors and biomolecular reagents were next employed to differentiate which PKC isotype(s) might be involved in the potentiation of the endogenous AC7 in HEL cells by PDBu or ethanol. The Ca 2ϩ sensitivity of the conventional PKCs (␣, ␤ and ␥) is well documented, and their activation and catalytic activity can be greatly attenuated by preventing increases in Ca 2ϩ influx or by preventing Ca 2ϩ release from intracellular stores. Preincubation with 1 mM EGTA (to reduce free extracellular Ca 2ϩ to ϳ50 nM) had no significant effect on the potentiation of AC activity by either 200 mM ethanol or 100 nM PDBu in the presence of 5 M PGE 1 (Fig. 4A). Ethanol enhanced AC activity by 68 Ϯ 10 and 75 Ϯ 16% in the absence and presence, respectively, of 1 mM EGTA in the medium. Chelating extracellular Ca 2ϩ prevents Ca 2ϩ influx but does not exclude the possibility that PGE 1 , in addition to activating G s , could also be activating a G q -coupled receptor (26) leading to subsequent phosphatidylinositol 4,5-bisphosphate turnover and release of internally stored Ca 2ϩ . To eliminate the possi-

FIG. 3. Western blot analysis of the PKC isoforms endogenously expressed in HEL cells.
HEL cells were hypotonically lysed, and a whole cell lysate was obtained by solubilization in 1% SDS. PKC isoform specific monoclonal antibodies were used to probe for the presence of various PKC isoforms expressed endogenously in HEL cells as described in detail under "Experimental Procedures." The subsequent chemiluminescence exposures are shown. Commercially obtained (Transduction Laboratories, KY) control protein (C; 1-2.5 g) was loaded next to 5-10 g of HEL cell lysate (HL). bility that internally released Ca 2ϩ might be activating conventional PKCs, HEL cells were pretreated with 100 nM thapsigargin, a sarcoendoplasmic reticulum Ca 2ϩ -ATPase inhibitor that depletes inositol 1,4,5-triphosphate-sensitive Ca 2ϩ stores. In the presence of 1 mM external EGTA alone, 5 M PGE 1 transiently raised intracellular Ca 2ϩ to 309 Ϯ 22 nM from a basal level of 106 Ϯ 15 nM (Fig. 4B). However, the pretreatment of HEL cells with 100 nM thapsigargin and 1 mM EGTA was sufficient to prevent any PGE 1 -induced rise in [Ca 2ϩ ] i (Fig. 4B). As shown in Fig. 4A, depleting the intracellular Ca 2ϩ stores prior to PGE 1 stimulation had no effect on the potentiation of AC activity by either 200 mM ethanol or 100 nM PDBu. Ethanol enhanced AC activity by 68 Ϯ 10 and 60 Ϯ 22% in the absence and presence of thapsigargin and EGTA, respectively.
Gö 6976 has been shown to selectively inhibit the conventional PKCs and PKC (27). Rottlerin, on the other hand, has been shown to selectively inhibit PKC␦ and (28,29). Pretreating HEL cells for 30 min with increasing concentrations of Gö 6976 had little or no effect on the potentiation of PGE 1stimulated AC activity by either PDBu or ethanol (Fig. 5A). However, when HEL cells were pretreated for 30 min with rottlerin, the potentiation of AC activity by both ethanol and PDBu was potently inhibited in a dose-dependent fashion (Fig. 5B). At 2.5 M rottlerin, the potentiation by ethanol was reduced from 83 Ϯ 10 to 28 Ϯ 7%, and the potentiation by PDBu was reduced from 112 Ϯ 11 to 59 Ϯ 12%. In the absence of ethanol or PDBu, concentrations of rottlerin above 2.5 M significantly inhibited the PGE 1 -stimulated AC activity in HEL cells.
Peptides containing specific sequences within the N-termi-nal V1 regions of PKCs, which are unique and not conserved across the PKC family, have been recently used to competitively inhibit the binding of PKCs to scaffolding proteins called receptors for activated C kinases (RACK) (30). These PKCderived peptides have been shown to be PKC isotype-specific (31). Dr. Daria Mochly-Rosen provided us with an octapeptide (␦V1-1) from the V1 region of PKC␦ that has been shown to be a selective inhibitor of PKC␦ localization and does not crossreact with other PKCs (18,32,33). The PKC-derived peptides have been conjugated via a disulfide linkage to a 16-amino acid antennapedia signal peptide from Drosophila that enables the chimeric peptides to translocate across the plasma membrane and after reduction of the disulfide bond be trapped within the cell (34). When HEL cells were pretreated with either PKCderived inhibitory peptide or a control peptide at 2 M for 2 h prior to the stimulation by PGE 1 , the potentiation of AC activity by 100 mM ethanol was selectively attenuated by ϳ50% by the PKC-derived inhibitory peptide, ␦V1-1, which is selective for PKC␦ (18,32,33). In contrast neither the inhibitory peptide (⑀V1-2) selective for PKC⑀, which is another member of the novel PKC family, nor the control nonsense octapeptide had any effect (Fig. 6). The Student's t test was used for pairwise comparisons and indicated that the PKC␦ peptide significantly reduced the ethanol potentiation of PGE 1 -stimulated AC activity, from 68 Ϯ 16 to 36 Ϯ 3% (*, p Ͻ 0.05), whereas the PKC⑀ peptide, and control peptide had no significant effect on the potentiation by ethanol, 62 Ϯ 11 and 68 Ϯ 16%, respectively. The PKC␦ peptide also reduced the PDBu potentiation compared with control (p Ͻ 0.05), whereas the PKC⑀ peptide and the control nonsense octapeptide were without effect.
Overexpression of the dominant negative, catalytically inactive forms of PKC have previously been shown to be selective inhibitors of a targeted PKC isotype by competing with the endogenous PKC for cofactors, substrates, and cellular binding sites (35,36). The dominant negative construct for PKC␦ has a lysine mutated to a methionine within the ATP-binding site, making it catalytically inactive (36). HEL cells were infected with a replication deficient adenovirus (Ad5 DL312) carrying either the wild-type (WT) or dominant negative (DN) form of PKC␦ at two different adenovirus titers, 10 ϫ 10 3 or 50 ϫ 10 3 particles/cell. When normalized to ␤-actin levels, HEL cells infected with the DN-PKC␦ adenovirus demonstrated a 174% increase in PKC␦ immunoreactive protein, representing the expression of the dominant negative mutant PKC␦ in addition to the endogenous expression of wild-type PKC␦ (Fig. 7A).
When HEL cells were assayed for AC activity 20 h after viral infection, the potentiation of PGE 1 -stimulated AC activity by 100 mM ethanol was no longer statistically significant in cells infected with the DN-PKC␦ adenovirus, at either virus titer. At the highest virus titer of 50 ϫ 10 3 particles/cell, the potentiation of ethanol was reduced to 16 Ϯ 7% in cells pretreated with DN-PKC␦ as compared with 52 Ϯ 18% in cells infected with the WT-PKC␦ adenovirus (Fig. 7B). The potentiation of AC activity by ethanol in cells infected with the WT-PKC␦ was comparable with that of either HEL cells infected with a control adenovirus carrying lacZ or uninfected HEL cells under the same conditions, 52 Ϯ 18% versus 44 Ϯ 4% versus 41 Ϯ 6%, respectively. Two-way analysis of variance indicated a significant difference between DN and WT expressing cells (F(1,28) ϭ 5.748, p Ͻ 0.023) and between the treatments (ethanol versus no ethanol; F(2,28) ϭ 8.222, p Ͻ 0.002). There was no statistically significant interaction between virus exposure and treatments (F(2,28) ϭ 1.507, p Ͼ 0.239). The post-hoc Tukey-test also revealed a significant enhancement of PGE 1 -stimulated cAMP accumulation in the presence of ethanol for only WT-treated cells (*, p Ͻ 0.05 compared with PGE 1 stimulation alone).
Because the above described data demonstrated that the ethanol potentiation of AC activity in HEL cells was dependent on PKC␦, we next determined whether AC7 was a substrate for PKC␦ in vitro. Western blot analysis of the immunoprecipitated fractions of solubilized, AC7-transfected HEK 293 cell membranes confirmed that the T7-tagged AC7 was successfully expressed in these cells and that T7-tagged AC7 protein bands could be detected at ϳ110 kDa (Fig. 8A). No bands were detected in HEK 293 cells that were transfected with the control pcDNA3.1-T7.His in this molecular mass range (Fig. 8A, lanes  1 and 4). Cell membranes from PGE 1 -stimulated AC7-transfected HEK 293 cells were then prepared under conditions that included phosphatase inhibitors. The extracted T7-tagged AC7 was then subjected to immunoprecipitation and to phosphorylation by exogenously added full-length recombinant PKC␦ or PKC⑀. Shown in Fig. 8B is an autoradiogram of immunoprecipitated T7-tagged AC7 from the transfected HEK 293 cells.   8 demonstrates a preferential phosphorylation of AC7 by PKC␦. The arrow indicates 32 P incorporation within a protein band running at ϳ110 kDa that is unique to the T7-AC7 immunoprecipitate (lanes 2p and 5p). Longer exposures revealed a faint 110-kDa band in lane 5p. The darker bands that appear below AC7 are residual amounts of the autophosphorylated forms of PKC␦ (molecular mass ϭ 77 kDa) or ⑀ (molecular mass ϭ 88 kDa). No bands were evident, in the 110-kDa molecular mass range, at any length of exposure in the control lanes where either the T7-tagged AC7 (lanes 1p and 4p) or the anti-T7 antibody (lanes 3p and 6p) was not present. Parallel Western blots using the anti-T7 monoclonal antibody gave further evidence that the phosphorylated band at 110 kDa corresponded to AC7.
Prior sequence analysis of AC7 suggested that a possible PKC␦-binding site (see "Discussion") exists within the AC7 structure. Using 32 P-labeled PKC␦ or ⑀ as a probe, we were able to show an increased immunoprecipitation of 32 P-PKC␦ together with the T7-tagged AC7 when the anti-T7 antibody was used to immunoprecipitate the protein complex. In contrast, we could not detect an interaction between 32 P-PKC⑀ and the T7-tagged AC7 under our experimental conditions (Table I). DISCUSSION To date only AC2, AC5, and Ac6 have been shown to be directly phosphorylated by PKC (37)(38)(39)(40). The potentiation of G␣ s -stimulated AC2 activity and the enhancement of the basal activity of this enzyme by phorbol esters and PKC itself has been the most extensively investigated (37,38,41). PKC␣ was shown to markedly enhance the in vitro phosphorylation and the sensitivity of AC2 to G␣ s stimulation, whereas PKC␣ reduced the sensitivity of AC4 to G␣ s (38). Although AC2 has been shown to be phosphorylated on Thr 1047 after phorbol ester treatment (37) and this post-translational modification has been indicated to alter AC2 responsiveness to G s␣ (38), little evidence currently exists 1) demonstrating that another member of the type II AC family (i.e. AC7) is covalently modified by PKC-mediated phosphorylation, 2) demonstrating the resultant effects of such a phosphorylation on G s␣ -stimulated AC7 activity, and 3) as to whether a particular isoform(s) of PKC may dominate the phosphorylation of AC7 and mediate the stimulatory effects of ethanol or PDBu on AC7 activity in cells.
Our data show that immunoprecipitated AC7, overexpressed in Sf9 cells, is phosphorylated in vitro by a mixture of PKCs isolated from rat brain. In addition, the back-phosphorylation experiments demonstrated an increase in a PKC-specific phosphorylation of AC7 within HEK 293 cells after exposure of the intact cells to ethanol. To elucidate which PKC isozyme(s) were mediating the effects of ethanol, we utilized pharmacologic and intracellularly expressed inhibitors of PKC.
A role for the involvement of the atypical PKCs, which are not sensitive to either Ca 2ϩ or diacylglycerol/phorbol esters, in the actions of ethanol was previously eliminated, because down-regulation of conventional and novel PKCs by prolonged treatment of HEL cells with phorbol esters blocked the effect of ethanol on HEL cell AC activity (8). The involvement of Ca 2ϩsensitive PKCs (conventional PKCs) was eliminated in the current work by showing that depletion of intracellular Ca 2ϩ and prevention of Ca 2ϩ influx into the cells did not alter the ability of ethanol to increase agonist-stimulated AC7 activity (Fig. 4). This conclusion was supported by the ineffectiveness of a selective inhibitor of the conventional PKCs, Gö 6976 (27), to modulate the potentiation of AC activity in HEL cells produced by ethanol or PDBu (Fig. 5A). On the other hand, rottlerin, which is a selective inhibitor of the novel PKC isoforms ␦ and (28,29), significantly reduced the potentiation of PGE 1 -stimulated AC activity in HEL cells by ethanol (Fig. 5B). Examination of the dose-response curves for rottlerin for the inhibition of the effects of PDBu and ethanol on AC7 activity revealed that higher concentrations of rottlerin were necessary to produce the initial inhibition of the actions of ethanol (Fig. 5B). The differential effects of rottlerin on PDBu and the actions of ethanol may well be due to the location of the pools of PKC␦ engaged by PDBU and by ethanol to carry out the post-translational modification of AC7. Ethanol may be acting on PKC␦, which is already in the vicinity of (bound to) AC7 (see below), whereas PDBu is activating and translocating an intracellular pool of PKC␦ to accomplish its effect on AC7 activity.
To further examine the involvement of PKC␦, additional molecular reagents were used. The dominant negative form of PKC␦ is catalytically inactive but can compete with the endogenous PKC␦ for substrates or cofactors (35,36). On the other hand, the N-terminal V1 fragment of PKC␦ and the related ␦V1-1 octapeptide can selectively block the proper localization and phorbol ester-induced translocation of PKC␦ (18,(31)(32)(33)42). We demonstrated that the effect of ethanol on agoniststimulated AC7 activity was blocked by either the dominant negative form of PKC␦ (Fig. 7) or a PKC␦-specific peptide (␦V1-1) derived from the N-terminal V1 region of PKC␦ (Fig. 6). In contrast, the corresponding PKC⑀-specific peptide (⑀V1-2) (18, 31) and the control nonsense octapeptide were without effect (Fig. 6).
Although it is has been thought that the PKC family shows little sequence specificity toward its substrates, recent studies have suggested that PKC⑀ tends to favor acidic residues downstream of the S/T phosphorylation site, whereas PKC␦ tends to favor hydrophobic residues (43). The suggestive substrate selectivity of PKC␦ over that of PKC⑀ for AC7 in Fig. 8 is not without precedence. Both the high affinity receptor for IgE and the elongation factor eEF-1␣ proteins have been shown to be selectively phosphorylated by PKC␦ compared with PKC⑀ (44,45).
Our experimental data in Table I also suggest a selective association between AC7 and PKC␦ under our experimental conditions. Examples of selective binding of PKC␦ to other proteins have been demonstrated (46 -48). Many of these PKCprotein associations do not fit the defined interaction of a PKC with a RACK protein. In cases where a RACK protein is involved, PKC binds only after PKC is activated and does not  32 P-labeled PKC␦ or PKC with AC7 Recombinant full-length PKC␦ or labeled with 32 P was mixed with an immunocomplex of protein A-agarose beads, anti-T7 antibody, and membrane protein from either control HEK 293 cells or T7-AC7-transfected HEK 293 cells. The amount of 32 P-PKC␦ or bound to the immunocomplex (pellet) or free in solution (supernatant) was quantitated by liquid scintillation, background subtracted, and converted into a percentage of the total 32  phosphorylate the RACK. Nevertheless, an interaction between the substrate protein, GAP-43, and the N-terminal V1 region of PKC␦ has been well demonstrated (47), as has the binding of the SRBC protein to PKC␦ (46). An examination of the AC7 sequence illustrates a possible binding site for PKC␦ that is not present in the sequences of the other ACs. When putative PKC binding sites on AKAP 79, AKAP 250 (gravin), and other PKC-binding proteins were aligned with full-length AC7, using the ClustalW 1.60 multiple sequence alignment algorithm set with gap penalties, in all cases, a contiguous sequence was best fit to the C 1b region of AC7. Furthermore, a region of the SRBC protein, which has been shown to bind to and be phosphorylated by PKC␦ (46), also aligned to the C1b domain of AC7. The following sequence from the C1b region of human AC7 491 GAARPFAHLNHRES-VSSGETHVPNGRRPKSVPQRHRRTPDRSMSPKGRSE 541 contained the greatest degree of alignment with the putative PKC␦-binding region of the SRBC protein. Two putative PKC phosphorylation sites unique to AC7 and conserved within AC7 across different species (mouse, cow, and human) are indicated by the residues italicized and underlined. Three additional putative PKC phosphorylation sites are located just outside this region.
Our demonstration of the involvement of PKC␦ in the phosphorylation of AC7 and modulation of ethanol-potentiated AC7 activity coincides with other studies that have implicated the novel PKCs in a number of both acute and chronic effects of ethanol. Chronic ethanol exposure was found to up-regulate the density and function of L-type Ca 2ϩ channels via a PKC␦dependent mechanism (42), whereas short term ethanol exposure of NG108 -15 cells was found to alter the subcellular localization and presumably the function of PKC␦ and ⑀ (49).
Whereas the exact mechanism or mechanisms by which ethanol utilizes PKC␦ to alter AC7 activity remains unknown, one possibility would be that ethanol is modulating the interaction between and colocalization of AC7 and PKC␦. Ethanol could promote a conformational change in AC7 that provides or enhances availability of a site(s) for PKC-mediated phosphorylation. This phosphorylation event could then act as an electrostatic switch for regulating enzymatic activity and/or regulating protein-protein (AC/G␣ s ) or protein-lipid interactions. The work presented here demonstrates a role for PKCinduced phosphorylation in mediating the effects of ethanol on AC7, and quite possibly this observation will be expanded to include other neuronal signaling proteins in the future.