INTRODUCTION

We have presented evidence that the cAMP signaling system is important for regulating the response to ethanol in a cell culture model of tolerance. We found that ethanol acts on a specific nucleoside transporter in NG108-15 neuroblastoma × glioma cells (1), inhibiting adenosine uptake (2). In cells treated chronically with ethanol, however, adenosine uptake becomes insensitive when rechallenged with acute ethanol, and adenosine uptake is no longer inhibited by ethanol (2). The loss of ethanol sensitivity is an example of ethanol tolerance at a cellular level. Using this model system, we have investigated the molecular mechanisms that regulate tolerance to ethanol.

We have shown that ethanol inhibition of adenosine uptake requires cAMP-dependent protein kinase (PKA)¹ activity; ethanol does not inhibit adenosine uptake in mutant cells lacking PKA activity (3). We also found that the ethanol sensitivity of transport in cells chronically exposed to ethanol can be restored by activating PKA (4). In addition, the inhibition of PKA in naive cells reproduces the ethanol-tolerant phenotype. This latter effect can be prevented by inhibiting protein phosphatase activity with okadaic acid (4), suggesting that protein phosphatase (1 or 2A) is also involved in regulating the ethanol sensitivity of adenosine uptake. Based on this earlier work, we concluded that ethanol sensitivity of adenosine uptake appears to be due to a balance of PKA and phosphatase activities.

Recent evidence also implicates protein kinase C (PKC) in cellular responses to ethanol. On acute exposure to ethanol, PKC activity increases in human lymphocytes (5) and epidermal keratinocytes (6). Moreover, chronic exposure to ethanol causes increased PKC activity in NG108-15 and PC12 cells (7). PKC also regulates the ethanol sensitivity of GABA_A receptors (8, 9, 10), NMDA receptors (11), AMPA/kainate receptors (12), and 5HT1c and M1 muscarinic receptors (13). We therefore investigated the role of PKC in the regulation of ethanol sensitivity of adenosine transport. We found that activation of PKC produces the tolerant phenotype in naive cells and that inhibition or down-regulation of PKC during chronic exposure to ethanol prevents the development of tolerance. Our results suggest that PKC activity is required for the development of tolerance of adenosine transport after chronic exposure to ethanol.

EXPERIMENTAL PROCEDURES

The a and b isomers of phorbol 12-myristate 13-acetate (PMA) (LC Services, Woburn, MA) were dissolved in dimethyl sulfoxide or 95% ethanol to a concentration of 10 mM, stored at 20 °C, and diluted in assay medium (Dulbecco's modified Eagle's medium: F-12 (Ham's), 3:1; 25 mM HEPES, pH 7.4) to 100 nM for use in acute ethanol experiments. PMA was diluted in defined medium for chronic ethanol experiments. Bradykinin (Calbiochem) was dissolved in water to a concentration of 5 mM and diluted in assay medium to a concentration of 50 mM. Chelerythrine (Calbiochem) was dissolved in dimethyl sulfoxide to a concentration of 13 mM, stored at 20 °C, and diluted in defined medium to 1 mM. Nitrobenzylthioinosine, [³H]adenosine (30 Ci/mmol), and non-radioactive adenosine were from Sigma, DuPont NEN, and Boehringer Mannheim, respectively. ¹²⁵I-protein A was purchased from Amersham Corp., and anti-PKC isozyme-specific antibodies were from Santa Cruz Biotechnology, Inc.

NG108-15 neuroblastoma × glioma hybrid cells (passage number 19-23) obtained from the cell culture facility at the University of California, San Francisco, were grown in 10% Nu-Serum and maintained for 2-3 days in complete defined medium as described previously (14). The cells were then seeded in six-well plates at a density of 4-6 × 10⁴ cells/well in complete defined medium and used for acute experiments on day 3. For most chronic ethanol experiments, cells were seeded as above and treated on day 3 for a further 48 h with or without 200 mM ethanol or other additions. The medium was replaced daily. The data in Fig. 3C for varying concentrations of ethanol were obtained after a 4-day exposure to ethanol. For the experiments investigating inhibition of PKC by chelerythrine, cells were incubated in the presence or absence of 1 µM chelerythrine for 30 min prior to exposure to 200 mM ethanol or defined medium alone with or without 1 µM chelerythrine for 48 h. For the PKC down-regulation experiments, cells were pretreated on day 3 with -PMA (100 nM) or vehicle for 24 h and then incubated in the presence or absence of ethanol with or without -PMA for an additional 48 h. For the uptake experiments using phorbol esters, bradykinin, or chelerythrine, cells incubated in assay medium containing carrier only were included as controls.

Cells were preincubated at room temperature for 4 min with 1 ml/well assay medium (2) in the presence or absence of 200 mM ethanol. The medium was replaced with assay medium containing 0.3 µM [³H]adenosine (0.03 Ci/µmol) in the presence or absence of ethanol. The incubation was stopped at 1 min (uptake is linear for at least 1.5 min (2)) by rapidly aspirating the medium and washing the cells with 3 ml of ice-cold non-radioactive medium. The cells were dissolved overnight in 1 ml of 2 N NaOH and aliquots taken for measurement of protein (15) and radioactivity. Specific uptake was determined by subtracting nonspecific uptake in the presence of 10 nM nitrobenzylthioinosine from total uptake. Data are presented as percent inhibition of specific uptake by ethanol.

Cells were collected in 20 mM Tris-HCl (pH 7.5), 2 mM EDTA, 10 mM EGTA, 1 mM phenylmethylsulfonyl fluoride, 10 µg/ml each of leupeptin and aprotinin, and 0.1 mM sodium orthovanadate. 400-µl samples (1.6 mg of protein) were each mixed with 100 µl of 5 × sample buffer (16) and heated for 5 min at 90 °C. After centrifugation at 10,000 × g for 10 min (4 °C), samples were diluted to 20-50 µg of protein and subjected to SDS-polyacrylamide gel electrophoresis on 10% acrylamide gels. Proteins were transferred electrophoretically to polyvinylidene difluoride membranes that were then incubated overnight at 4 °C in a blocking solution containing TBS (20 mM Tris-HCl (pH 7.6), 150 mM NaCl), 0.1% Tween 20, and 5% nonfat dry milk. Blots were incubated for 2 h at room temperature with affinity-purified rabbit antibodies to PKC isozymes (0.5 mg/ml, diluted 1:200), washed 3 times in TBS containing 0.1% Tween 20 (TBS-T), and then incubated for 2 h with ¹²⁵I-protein A (10 µCi/100 ml), followed by an extensive washing with TBS-T. The immunoreactive bands were visualized and quantitated using a Molecular Imager System (Bio-Rad).

RESULTS

Acute ethanol inhibits adenosine uptake by approximately 40% (Ref. 2 and Fig. 1). After chronic exposure to ethanol, adenosine uptake becomes tolerant to acute rechallenge with ethanol (Ref. 2 and Fig. 1). If increased amount and/or activity of PKC contributes to chronic ethanol-induced tolerance in NG108-15 cells, then activating PKC in naive control cells (cells not exposed to ethanol) should also produce ethanol tolerance. We therefore incubated naive cells for 10 min in the absence and presence of the phorbol ester, -PMA (100 nM), which activates PKC, and then measured ethanol inhibition of adenosine uptake. Control cells showed a typical level of inhibition of adenosine uptake by acute ethanol (34 ± 6%, Fig. 1). By contrast, ethanol did not significantly inhibit adenosine uptake in cells pretreated with -PMA. This loss of ethanol sensitivity of adenosine transport was similar to the tolerance caused by chronic ethanol exposure (Fig. 1). -PMA had no further effect on the tolerance of adenosine uptake in cells treated chronically with ethanol (Fig. 1). In both naive cells and cells chronically exposed to ethanol, the absolute values of uptake in the absence of acute ethanol were not affected by -PMA treatment (not shown). The inactive phorbol ester -PMA had no effect on the ethanol sensitivity of adenosine uptake in control cells (Fig. 1), suggesting that the ethanol tolerance produced by -PMA was due to the activation of PKC.

We also determined whether the activation of PKC by an endogenous receptor pathway produces the tolerant phenotype in naive cells. In NG108-15 cells, bradykinin receptors activate phospholipase C (17), which consequently activates PKC. Therefore, cells were incubated with bradykinin (50 µM) for 10 min before measuring ethanol sensitivity of adenosine uptake. Cells treated with bradykinin showed the tolerant phenotype; acute ethanol did not inhibit adenosine uptake (Fig. 2). Similar to the effect of -PMA, incubation with bradykinin had no effect on the tolerant phenotype produced by chronic ethanol exposure (Fig. 2). These data suggest that activation of PKC induces tolerance to ethanol inhibition of adenosine transport in naive cells not previously exposed to ethanol.

Since activation of PKC and chronic exposure to ethanol produced the same tolerant phenotype, it was possible that ethanol increased the amount of PKC in NG108-15 cells, leading to tolerance of adenosine transport. We therefore determined whether chronic exposure to ethanol alters the amount of PKC in NG108-15 cells grown in defined medium. PKC is a family of closely related isozymes divided into three subfamilies (18): the Ca²⁺-dependent or conventional isozymes (PKC , ₁, ₂, ), the Ca²⁺-independent or novel isozymes (, , , , and µ), and phorbol ester-insensitive or atypical subfamilies (PKC  and /). Using isozyme-specific antibodies against the phorbol ester-sensitive PKC isozymes, we found that , ₁, ₂, , and  PKC are expressed in these cells (Fig. 3A, control).  PKC was not detectable. Antibodies against , , and µ were not available to us. Chronic exposure to 200 mM ethanol for 48 h caused a significant increase in the levels of PKC , , and  when compared with naive control cells (Fig. 3, A and B). There was no significant change in the amount of the ₁ and ₂ isozymes (Fig. 3, A and B). These experiments were carried out at a relatively high concentration of ethanol to allow us to use short exposure times (2 days). To determine whether lower concentrations of ethanol also increase  and PKC levels, NG108-15 cells were incubated in varying concentrations of ethanol for 4 days. The data in Fig. 3C indicate that the amounts of  and  PKC were significantly increased at 25 mM ethanol and that the increase was dependent on the concentration of ethanol.

If the increased amount of , , or  PKC produced by chronic ethanol exposure contributes to the development of tolerance, then inhibiting PKC activity during ethanol treatment should prevent the development of the chronic ethanol phenotype. We used the specific PKC inhibitor, chelerythrine (19), to test this hypothesis. Cells were pretreated for 30 min either with or without 1 µM chelerythrine in the medium. The cells were then incubated in the presence or absence of ethanol, with or without 1 µM chelerythrine, for 48 h, and ethanol inhibition of adenosine uptake was measured. Cells exposed to ethanol alone for 48 h show tolerance when rechallenged with ethanol (Fig. 4). In contrast, chelerythrine prevented the development of tolerance; ethanol inhibited adenosine uptake in cells incubated with chelerythrine and ethanol for 48 h just as in untreated cells (Fig. 4). Chelerythrine had no effect on adenosine uptake in the absence or presence of ethanol in naive control cells. However, chelerythrine completely blocked the chronic ethanol-induced increase in the amounts of  and  PKC (Fig. 3, D and E).

The functional consequences of reduced PKC activity can also be determined by down-regulating PKC. Therefore, the PMA-sensitive PKC isozymes were down-regulated with -PMA (20) prior to and during exposure to ethanol. Cells were pretreated for 30 min with medium alone or medium containing 100 nM -PMA and then incubated in the presence or absence of ethanol, with or without 100 nM -PMA for 48 h. The , ₁, ₂, , and  isozymes were virtually absent after -PMA treatment (Fig. 3A). Chronic ethanol exposure did not produce the tolerant phenotype in these down-regulated cells; ethanol inhibition of adenosine uptake was similar in down-regulated ethanol-exposed cells and naive control cells (Fig. 5). Down-regulation of PKC had no effect on ethanol sensitivity of adenosine uptake in naive cells. Therefore, reduced PKC activity produced either by PKC inhibition (with chelerythrine) or down-regulation (by -PMA) during chronic ethanol exposure prevented the development of the ethanol tolerant phenotype. These data support the hypothesis that an ethanol-induced increase in PKC amount or activity is required for tolerance of adenosine transport to develop during chronic exposure to ethanol.

DISCUSSION

We have shown that a specific facilitative nucleoside transporter is a primary target for ethanol in NG108-15 cells (1). Acute ethanol inhibition of adenosine uptake via this transporter leads to activation of adenosine A₂ receptors and a cascade of intracellular signaling events leading to tolerance of adenosine transport on rechallenge with ethanol (2). We found that adenosine uptake is not inhibited by ethanol in mutant cells lacking PKA activity (3) and that the tolerance of adenosine transport induced by chronic ethanol exposure could be reversed by activating PKA (4). These results suggest that PKA activity is required for ethanol to inhibit adenosine uptake. Treatment with phorbol esters (Fig. 1), which activate PKC, or activation of bradykinin receptors (Fig. 2), which also leads to activation of PKC (21, 22), produces a tolerant phenotype similar to that produced by chronic exposure to ethanol. Conversely, ethanol tolerance is prevented in chronically exposed cells when PKC activity is inhibited (Fig. 4) or reduced by down-regulation (Fig. 5).

These data suggest that the ethanol sensitivity of adenosine transport is regulated by both PKA and PKC. PKA maintains the ethanol-sensitive phenotype; PKC activity is required for chronic ethanol-induced tolerance of adenosine transport. There is evidence in other systems that a balance of PKA and PKC activities regulates cellular functions. PKA can overcome PKC-dependent desensitization of the G-protein-activated K⁺ channel (23), and opposing effects of PKA and PKC on the glycine receptor have also been reported (24). Hoek and co-workers (25) have also shown that PKA and PKC have opposite effects on the regulation of basal and receptor-stimulated phospholipase C activity by ethanol in liver.

In naive NG108-15 cells, activation of PKC produces the tolerant phenotype of adenosine transport (Fig. 1), but ethanol-induced tolerance can be reversed by activating PKA. These observations are best explained by an indirect mechanism of action for PKC with protein phosphatase(s) 1 and/or 2A as intermediates (Fig. 6). We propose that PKC activates a phosphatase that dephosphorylates the PKA-phosphorylated site on the transporter or an associated regulatory protein producing the ethanol-insensitive phenotype. In support of this model, we found that inhibition of PKA activity produces an ethanol-tolerant phenotype only if protein phosphatase(s) 1 and/or 2A are active (4); in the presence of phosphatase inhibitors, inhibiting PKA activity had no effect on ethanol sensitivity of transport. Therefore, ethanol sensitivity of transport appears to be due to a balance between phosphorylation and dephosphorylation. In naive cells, phosphorylation predominates. Phosphatase activity predominates after inhibition of PKA activity or after chronic ethanol exposure. Phosphatase activity could be stimulated directly by PKC as described for the receptor-like protein-tyrosine phosphatase RPTP (26) or indirectly by phosphorylation and inactivation of a phosphatase inhibitor protein as described for -PMA activation of protein phosphatase 1 in L6 cells (27). Inhibition of PKC by chelerythrine or down-regulation of PKC by PMA in cells exposed chronically to ethanol would be expected to prevent activation of the phosphatase by PKC, leaving the PKA-phosphorylated site intact and resulting in an ethanol-sensitive phenotype.

PKC is a family of isozymes of which at least five are expressed in NG108-15 cells (Fig. 3A). Beckmann et al. (28) did not detect ₁ in NG108-15 cells. This may reflect the difference in culture medium in which the cells are grown. We found that the amounts of , , and  were increased after chronic exposure to ethanol (Fig. 3B). Although we have not measured PKC activity directly, an increase in PKC activity correlated with increased amounts of , , and  PKC after chronic exposure of PC12 cells to ethanol (7). Since chelerythrine inhibited both PKC activity and the ethanol-induced increase in amounts of  and  PKC, it remains to be determined whether increases in PKC amount or PKC activity, per se, are required for the development of tolerance of adenosine transport.

In addition to changes in the amount or activity of PKA, PKC, or possibly protein phosphatases 1 and 2A, ethanol might alter the intracellular localization of these enzymes, thus altering their function. Recent evidence suggests that the localization of intracellular signaling enzymes determines their specificity of action (29, 30, 31). PKA (32), protein phosphatases 1 and 2A (30), and specific isozymes of PKC (33) have unique locations within a number of cell lines including NG108-15 cells (28). Ethanol has been shown to cause translocation of PKC activity from cytosolic to membrane fractions in astroglial cells (34), human lymphocytes (5), and epidermal keratinocytes (6). It is therefore possible that ethanol may modulate the balance of PKA, PKC, and protein phosphatase activities by altering their localization in NG108-15 cells. Such changes may contribute to the development of ethanol tolerance.