Introduction

Top Abstract Introduction Methods Results Discussion References

Earlier studies indicate that EtOH has a fluidizing influence on membrane lipids (Chin et al., 1979; Goldstein and Chin, 1981) that could, in turn, alter the function of proteins imbedded in the membrane. More recent data, however, indicate that EtOH has interactions with specific neurotransmitter mechanisms (Deitrich et al., 1989; Shefner, 1990). A number of behavioral studies suggest that some effects of EtOH are mediated through GABA mechanisms (Becker and Anton, 1990; Ferko, 1990; Hinko and Rozanov, 1990; Liljequist and Engel, 1982; Martz et al., 1983), and neurochemical studies show that EtOH potentiates chloride flux through the GABA_A/Cl channel in brain synaptoneurosomes (Allan and Harris, 1986, 1987; Suzdak et al., 1986b) as well as in cultured spinal neurons (Ticku et al., 1986). Furthermore, GABA antagonists have been reported to block EtOH effects on chloride flux in both brain synaptoneurosomes (Suzdak et al., 1986a) and cultured spinal neurons (Mehta and Ticku, 1988) in vitro and to reverse EtOH-induced inhibitions of cerebellar Purkinje neuron firing in vivo (Freund et al., 1993; Palmer and Hoffer, 1990).

Electrophysiological studies of EtOH interactions with GABA effects, however, have not consistently provided evidence for an EtOH-induced enhancement of GABA-induced responses. For example, a number of electrophysiological studies in the central nervous system have found that EtOH either does not alter or it antagonizes the GABA effects on the majority of neurons sampled in hippocampus, ventral tegmental area, locus coeruleus, lateral septum or cerebellum (Bloom and Siggins, 1987; Carlen et al., 1982; Freund et al., 1993; Harris and Sinclair, 1984; Mancillas et al., 1986; Shefner, 1990; Siggins et al., 1987; Whiting et al., 1990). In contrast to these data, EtOH has been reported to facilitate GABA-mediated responses in hippocampus (Aguayo, 1990; Weiner et al., 1994), cerebellum (Lee et al., 1995; Lin et al., 1991, 1993), neocortex (Nestoros, 1980; Reynolds et al., 1992; Soldo et al., 1994), septum (Givens and Breese, 1990; Soldo et al., 1994), substantia nigra pars reticulata and inferior colliculus (Criswell et al., 1993), spinal cord (Celentano et al., 1988; Reynolds et al., 1992) and Xenopus laevis oocytes, which express mouse brain mRNA for GABA_A receptor/Cl channels (Wafford et al., 1990). In some cases, the discrepancies may be due to methodological differences in, for example, recording techniques, drug applications or biological preparations. In other cases, a lack of interaction may have been concluded when the EtOH sensitivity of a given animal model for GABA interactions was lower than that which was tested. Not only did the doses of GABA and EtOH that were used differ among studies, but also we previously found that the neuronal sensitivity to EtOH varied among animal species as well as among strains and lines within a species (Palmer et al., 1987, 1992; Spuhler et al., 1982); it even varied among brain regions in a particular animal model (Palmer et al., 1986). Furthermore, Aguayo et al. (1994) recently reported that GABA responses in hippocampal neurons are more sensitive to the potentiating effects of EtOH in C57 mice than they were in Sprague-Dawley rats.

Much of the observed difference in EtOH sensitivity of GABA responses, however, may be related to the post-translational regulation of GABA_A mechanisms. Thus, EtOH has been reported to potentiate the electrophysiological effects of GABA in some brain areas and not in others in the same study (Criswell et al., 1993; Soldo et al., 1994), and there is some evidence that variations in GABA_A receptor subunit composition involving protein kinase phosphorylation sites (Kofuji et al., 1991; Sikela et al., 1991; Whiting et al., 1990) might mediate these differences in EtOH sensitivity (Wafford et al., 1990; Wafford and Whiting, 1992). Indeed, PKA/PKC antagonists have been reported to block the EtOH potentiation of GABA mechanisms in hippocampus and in X. laevis oocytes expressing GABA_A receptors (Wafford and Whiting, 1992; Weiner et al., 1994), and we recently reported that 8-Br-cAMP, a PKA activator, sensitizes GABA-induced depressions in the cerebellum to the potentiating effects of EtOH (Freund and Palmer, 1996). Thus, both PKC and PKA mechanisms have been implicated in the actions of EtOH in various brain areas.

Neuronal sensitivity for EtOH interactions with GABA mechanisms may be determined not only by the nature of the EtOH interaction with the GABA_A/Cl channel on a given neuron but also by the influence of neuromodulators on the responsiveness of the GABA mechanism. Thus, although we previously reported that bicuculline, a GABA_A antagonist, blocks the higher-dose depressant effects of EtOH on Purkinje neurons (Freund et al., 1993), we and others have reported that local EtOH applications do not potentiate GABA-induced depressions of the majority of cerebellar Purkinje neurons (Bloom et al., 1984; Freund et al., 1993; Harris and Sinclair, 1984; Lee et al., 1995). However, we (Lin et al., 1991) and others (Lee et al., 1995) did find that systemic EtOH will potentiate the facilitation of GABA-induced inhibitions of cerebellar Purkinje neuron firing through the local applications of the neuromodulator norepinephrine. We also found that subdepressant local applications of EtOH will potentiate these inhibitory effects of GABA on Purkinje neurons if the GABA response is first sensitized to this EtOH effect by catecholamine modulation (Lin et al., 1991, 1993) and that this catecholamine sensitization of GABA responses to the potentiating effects of EtOH is mediated by a beta adrenergic mechanism (Lin et al., 1993). Furthermore, the activity of endogenous beta adrenergic mechanisms can similarly sensitize the interaction between EtOH and GABA in the absence of exogenous catecholamine application, and this effect can be blocked by timolol, a -adrenergic antagonist (Lin et al., 1994). Because this EtOH interaction with GABA effects in cerebellum is not routinely evident in the absence of beta adrenergic receptor stimulation, we concluded that activation of this catecholamine mechanism is required for its expression.

One well known consequence of beta adrenergic receptor stimulation is an increase in adenylate cyclase activity, resulting in elevation of intracellular levels of cAMP (Bloom, 1975; Skolnick et al., 1976) and, thus, stimulation of PKA. Indeed, the cAMP/PKA second-messenger system has been shown to mediate the modulatory influence of beta adrenergic receptor activation on GABA responses on cerebellar Purkinje neurons (Cheun and Yeh, 1992, Freund and Palmer, 1996; Hoffer et al., 1972; Llano and Gerschenfeld, 1993: Sessler et al., 1989), and previous studies showing an EtOH potentiation of GABA effects in the cerebellum indicate that beta adrenergic activation is involved in this EtOH interaction (Lee et al., 1995; Lin et al., 1991, 1993). Although other regulatory mechanisms may also influence GABA_A responsiveness to EtOH in the cerebellum, in the present study we specifically investigated the role of the cAMP/PKA second-messenger regulatory mechanism in the observed beta adrenergic influence on EtOH interactions with GABA mechanisms on Purkinje neurons.

	Methods

Top Abstract Introduction Methods Results Discussion References

Male Sprague-Dawley rats weighing 200-450 g were anesthetized with 1.25 g/kg urethane and placed in a stereotaxic frame. Body temperature was monitored by a rectal thermistor probe and maintained at 37°C by a heating pad. The skull and other superficial tissues over the cerebellar vermis were removed, and the dura was opened to expose the underlying brain. The exposed brain was covered with 2% agar, and the cisterna was opened to reduce brain pulsations. Five-barrel micropipettes, constructed as previously described (Palmer, 1982), were then stereotaxically lowered into the fifth or sixth vermal lobules of the cerebellum. One 3 or 5 M NaCl-filled barrel was used to record spontaneous Purkinje neuron firing rates, identified by their characteristic discharge pattern (Eccles et al., 1967). A second barrel filled with 3 M NaCl was connected to a current neutralization circuit to minimize tip potentials and electronic artifacts associated with microiontophoresis (Geller and Woodward, 1972). Other barrels of each micropipette were used to apply drugs locally into the microenvironment of each cell from which recordings were taken. Using methods that we have previously described in detail, GABA (0.5 M, pH 5; Sigma Chemical, St. Louis, MO), ()-Iso (0.25 M, pH 4; Sigma and 8-Br-cAMP (0.125 M, pH 4.75; Sigma) were applied locally from micropipettes by microiontophoresis. Sp-cAMP (1 mM, pH 7; Calbiochem, San Diego, CA), Rp-cAMP (1 mM, pH 7; Calbiochem), PDA (100 µM in 0.05% dimethylsulfoxide, pH 7.6; LC Laboratories, Woburn, MA), M-LR (100 mM, pH 6.7; Calbiochem) and OkA (100 µM in 0.05% dimethylsulfoxide, pH 6.5; LC Laboratories) were applied locally by micropressure ejection. EtOH (750 mM, pH 7) was applied by electro-osmosis, a variant of microiontophoresis (Palmer and Hoffer, 1980; Stone, 1985). The microiontophoretically applied drugs were dissolved in distilled water, and pressure ejected and electro-osmotically applied drugs were dissolved in 0.9% saline. Drug concentrations are typically diluted 10- to 1000-fold as they diffuse into the tissue with pressure ejection or microiontophoresis (Gerhardt and Palmer, 1987). Ejection pressure was regulated with a pneumatic valve, and iontophoretic current was controlled by operational amplifiers; the timing of drug applications was controlled with a crystal clock circuit (Smith and Hoffer, 1978). We used previously described controls for local anesthesia and pH effects (Hoffer et al., 1971), as well as for artifactual responses to iontophoretically applied and pressure-ejected drugs (Palmer, 1982; Palmer et al., 1986; Palmer and Hoffer, 1980). In addition, 30 mg/kg theophylline (10 mg/ml) was administered intraperitoneally.

Action potentials from single Purkinje neurons were filtered, amplified and isolated by a window discriminator. The firing rates were integrated over 1-sec time intervals and displayed as ratemeter records on a strip-chart recorder. These data were then digitized and analyzed by computer to determine percent responses to local drug applications as previously described (Palmer and Hoffer, 1980). Some data were collected and analyzed digitally using DataWave Technologies (Longmont, CO) Experimenter's WorkBench software on a personal computer. Each neuron was required to exhibit a stable firing rate during predrug and postdrug periods, and drug antagonist responses were acceptable only if they were repeatable and reversible. For tabular data, neurons were assigned to categories (potentiation, antagonism, or no effect) according to whether they increased or decreased the indicated response by >10%. Statistical significance was determined using one-way ANOVA, followed by Tukey-Kramer posthoc analyses if the ANOVA was significant.

	Results

Top Abstract Introduction Methods Results Discussion References

Brief, repeated microiontophoretic applications of GABA reliably produced dose-dependent depressions of cerebellar Purkinje neuron firing. For this study, the GABA dose was adjusted to cause a 15% to 25% depression of firing for each neuron (fig. 1, top), and the GABA application was maintained at that dose for the remainder of the experiment. Although local applications of EtOH alone only potentiated the unmodulated GABA effect on <30% of the neurons studied (table 1), prolonged applications of the membrane-permeable cAMP analogue Sp-cAMP significantly potentiated (modulated) these GABA inhibitions (fig. 1, second panel) of Purkinje neuron firing on 6 of 11 neurons tested. The continuous application of EtOH further potentiated the Sp-cAMP-modulated GABA response (fig. 1, third panel) on seven cells, including one in which Sp-cAMP did not modulate GABA responses (table 1). This EtOH effect was statistically significant among those 7 cells (P < .0001; fig. 2), and GABA responses returned to control after cessation of EtOH and Sp-cAMP applications (fig. 1, fourth panel). EtOH antagonized Sp-cAMP-treated GABA responses on the remaining four cells, which were also unresponsive to Sp-cAMP (table 1); however, this latter effect was not statistically significant.

View larger version (67K): [in this window] [in a new window]   Fig. 1.   Sp-cAMP modulates GABA responses and sensitizes those responses to the potentiative effects of EtOH on a representative cerebellar Purkinje neuron. Ratemeter records illustrating the firing rate (Y-axis) of a cerebellar Purkinje neuron over time (X-axis). The application of GABA by microiontophoresis, indicated by the solid bars above the records, caused a small, but repeatable inhibition of neuronal firing (top). Continuous application of Sp-cAMP by pressure ejection (striped line) resulted in modulated (potentiated) GABA responses (second panel), and the local application of EtOH by electro-osmosis (dashed line) during Sp-cAMP administration produced large potentiations of GABA responses (third panel). Termination of both Sp-cAMP and EtOH allowed GABA responses to return to control levels (bottom). Calibration bars indicate the firing rate in Hz (vertical) and the time in sec (horizontal).

View larger version (15K): [in this window] [in a new window]   Fig. 2.   Effect of various modulators of the cAMP/PKA/phosphatase system on GABA-EtOH interactions in cerebellar Purkinje neurons. When no modulator is present, GABA responses are slightly potentiated by EtOH in some cells, but over all the cells tested, this effect is not significant (see None; n = 8). In the presence of effective doses of various modulators (8-Br-cAMP, n = 8; Sp-cAMP, n = 7; M-LR, n = 5) and OkA (n = 5), the continuous local application of EtOH produced large and significant potentiations of GABA responses. Data are the mean ± S.E.M. for the cells that showed GABA responses that could be modulated. Statistical significance was determined by two-tailed Student's t test.

As in our previous report (Freund and Palmer, 1996), the membrane-permeable cAMP analogue 8-Br-cAMP produced effects very similar to those of Sp-cAMP. 8-Br-cAMP significantly potentiated (modulated) GABA inhibitions of Purkinje neuron firing on 12 of 13 neurons tested, and the continuous application of EtOH further potentiated 8-Br-cAMP-modulated GABA responses on 8 of the 12 modulated cells (table 1). This EtOH effect was significant (P < .0001; fig. 2), and the GABA-induced depressions recovered to control levels after cessation of the applications of EtOH and 8-Br-cAMP. Neither the modulatory actions of 8-Br-cAMP nor the potentiative actions of EtOH were blocked by the adenosine antagonist theophylline (30 mg/kg i.p.). In addition, the GABA effect on one Purkinje neuron was unaltered by both 8-Br-cAMP and EtOH (table 1), and EtOH significantly antagonized the GABA responses on four of the 12 8-Br-cAMP-modulated cells (P < .005; table 1).

The role of PKA in the beta adrenergic sensitization of GABA responses to the potentiating effects of EtOH was tested with Rp-cAMP, a membrane-permeable cAMP analogue that inhibits PKA activity (Parker-Botelho et al., 1988; Rothermel et al., 1983). We found that Iso-modulated GABA responses were significantly potentiated by continuous local applications of EtOH from micropipettes (P < .01; fig. 3, +EtOH). Rp-cAMP significantly antagonized the EtOH potentiation of Iso-modulated GABA depressions (fig. 3, +Rp-cAMP) for all five neurons on which it was tested (P < .01), and the EtOH potentiation of GABA responses returned (fig. 3, Rp-cAMP) after cessation of the Rp-cAMP application (P < .05).

View larger version (29K): [in this window] [in a new window]   Fig. 3.   The PKA inhibitor Rp-cAMP reverses the potentiating effects of EtOH and Iso on GABA responses in cerebellar Purkinje neurons. Small GABA inhibitions (GABA, open bar) were modulated by Iso (+Iso) and further potentiated by local application of EtOH (+EtOH). While maintaining application of both Iso and EtOH, application of Rp-cAMP (+Rp-cAMP) significantly reversed the potentiative effects of EtOH and Iso to control GABA response levels. Termination of Rp-cAMP (Rp-cAMP), while maintaining application of both Iso and EtOH, resulted in a recovery of the potentiation of the GABA responses. Data represent the mean ± S.E.M. for five neurons. Statistical significance was determined by one-way ANOVA followed by Tukey-Kramer posthoc analysis, but for clarity, the only statistical comparison indicated here is that of the antagonism of the Iso-modulated, EtOH-potentiated GABA response by Rp-cAMP.

To determine whether PKA is unique among kinase mechanisms in sensitizing GABA responses to the potentiating effects of EtOH, we tested whether the phorbol ester PDA, a PKC activator, would also influence the interactions between EtOH and GABA mechanisms on Purkinje neurons. For these experiments, GABA was intermittently applied before and during continuous local application of PDA by pressure ejection. This PKC activator facilitated the depressant effects of GABA on all five cells studied, and subsequent EtOH administration potentiated these modulated GABA responses.

If the observed sensitization of GABA responses to EtOH involves the activation of protein kinases and subsequent protein phosphorylation, then endogenous phosphatase activity would be expected to reverse this effect. Furthermore, phosphatase inhibitors would be expected to facilitate any endogenous regulation of the GABA_A receptor complex by cAMP or other second messengers, which would result in the activation of these kinase mechanisms. We found that prolonged pressure applications of two membrane-permeable phosphatase inhibitors, M-LR and OkA, positively modulated GABA inhibitions of Purkinje neuron firing and that the continuous application of EtOH further potentiated these modulated GABA responses (fig. 4). These effects were similar to those observed with 8-Br-cAMP and Sp-cAMP (fig. 2). M-LR modulated GABA responses on five of six neurons tested, and EtOH potentiated the GABA responses (P < .001; fig. 2) on all five neurons that exhibited modulation (table 1). OkA modulated GABA responses in five of six neurons tested, and EtOH potentiated this modulated GABA response (P < .0001; fig. 2) on the same five neurons (table 1).

View larger version (87K): [in this window] [in a new window]   Fig. 4.   Two ratemeter records showing that the phosphatase inhibitors M-LR (A) and OkA (B) modulated GABA responses and sensitized those responses to potentiation by EtOH on separate representative neurons. Sufficient GABA was locally applied (solid bars above records) by microiontophoresis to produce a small inhibition of the firing of Purkinje neurons (top). These GABA-induced depressions were modulated (potentiated) by the continuous local pressure ejection of M-LR (A.) and OkA (B; middle, hatched bars). The continuous electro-osmotic application of EtOH (dashed bar) potentiated the modulation of GABA inhibitions caused by the phosphatase inhibitors. Calibration bars, firing rate (in Hz) (vertical) and time (in sec) (horizontal).

PKA-mediated phosphorylation might either directly or indirectly alter the function of the GABA_A receptor complex on Purkinje neurons. There are at least two ways in which this regulatory mechanism could facilitate the EtOH potentiation of GABA effects on these cells: 1) PKA might render the GABA_A mechanism more responsive to EtOH-induced facilitation of agonist activation, and 2) EtOH might activate adenylate cyclase, resulting in increased levels of cAMP; by this mechanism, we predict that the influences of EtOH on the regulation of GABA_A receptor function by PKA would be additive with ongoing PKA activity. If cAMP and EtOH potentiate GABA responses on Purkinje neurons through this same PKA mechanism, then the effects of EtOH should become nonadditive with maximally effective doses of cAMP. We tested this hypothesis by applying increasing doses of 8-Br-cAMP until further increases did not result in further modulation of GABA responses. We found that 8-Br-cAMP caused a significant (P < .005) dose-dependent potentiation of threshold GABA-induced depressions of Purkinje neuron activity (fig. 5, ). Although EtOH did potentiate GABA responses that were submaximally modulated by 8-Br-cAMP (fig. 5; P < .05), EtOH did not potentiate GABA inhibitions that were maximally modulated by 8-Br-cAMP. Furthermore, the greatest potentiation of GABA responses, when caused by EtOH, was not greater than those caused by maximally effective doses of 8-Br-cAMP. Although these data suggest that EtOH potentiates GABA responses through a cAMP mechanism in these neurons, EtOH alone was not effective for potentiating GABA-induced depressions in the absence of 8-Br-cAMP applications.

View larger version (12K): [in this window] [in a new window]   Fig. 5.   Dose-response data for 8-Br-cAMP potentiation of GABA inhibition with and without EtOH. Threshold-depressant doses of GABA were intermittently applied from micropipettes to cerebellar Purkinje neurons. The continuous application of various doses of 8-Br-cAMP caused a significant (P < .005, one-way ANOVA) dose-dependent potentiation of GABA responses (). EtOH was applied in the absence of 8-Br-cAMP and in the presence of submaximally and maximally modulating doses of 8-Br-cAMP (). EtOH only potentiated the GABA inhibition at doses of 8-Br-cAMP that were submaximal for potentiating GABA inhibitions (P < .05, two-tailed t test). 8-Br-cAMP doses were normalized to the maximum potentiating dose, and GABA responses are given as the percent increase over control. Data are mean ± S.E.M. for five neurons.

	Discussion

Top Abstract Introduction Methods Results Discussion References

In the present study, we found that EtOH facilitates the potentiation of GABA responses by both PKA activation and inhibition of the phosphatase-mediated termination of endogenous protein kinase activity on cerebellar Purkinje neurons. Furthermore, the antagonism of PKA activity in these cells caused a reversal of the EtOH facilitation of Iso-modulated GABA effects. In confirmation of our preliminary report (Freund and Palmer, 1996), local applications of 8-Br-cAMP from micropipettes sensitized GABA responses to the potentiating effects of EtOH in a manner similar to that caused by Iso application. In the present study, we found that this effect was mimicked by the membrane-permeable PKA agonist Sp-cAMP. Previous reports suggest that locally applied cAMP analogues can also activate adenosine receptor mechanisms extracellularly (Dunwiddie et al., 1992; Dunwiddie and Hoffer, 1980), but the effect of adenosine locally applied to cerebellar Purkinje neurons is a theophylline-sensitive depression of spontaneous activity (Dunwiddie et al., 1984). We, however, did not observe neuronal inhibitions at the doses of either 8-Br-cAMP or Sp-cAMP that were applied, and we found both here and in our earlier study of 8-Br-cAMP (Freund and Palmer, 1996) that the GABA interaction of this membrane-permeable cAMP analogue was not blocked by the adenosine antagonist theophylline. Furthermore, the observed GABA interaction of 8-Br-cAMP and Sp-cAMP is probably not extracellularly mediated because our previous data also indicated that this effect was not mimicked by similar applications of membrane-impermeable cAMP (Freund and Palmer, 1996). These data suggest that cAMP is an intracellular second messenger that can regulate the observed EtOH interaction with GABA responses.

Our data suggest that the sensitization of GABA responses to EtOH actions occurs only at doses of the cAMP analogues that also cause a small potentiation of GABA-induced depressions before the EtOH application. We and others previously found that beta adrenergic receptor activation in the cerebellum results in facilitation of GABA-induced depressions (Freund and Palmer, 1996; Lin et al., 1993; Moises et al., 1979, Sessler et al., 1989; Waterhouse et al., 1982), and this beta adrenergic effect is also mediated by a cAMP second-messenger mechanism (Cheun and Yeh, 1992; Sessler et al., 1989). Thus, the cAMP/PKA mechanism mediating the beta adrenergic modulation of GABA-induced depressions in the cerebellum is likely the same as the mechanism that sensitizes GABA responses to EtOH actions in this brain area. If this is true and the influence of Iso on EtOH interactions with GABA in this brain area is either mediated or regulated by the activation of PKA, as is suggested by our present finding that similar effects are caused by the PKA agonist Sp-cAMP, then we would expect the membrane-permeable PKA inhibitor Rp-cAMP to antagonize the beta adrenergic sensitization of GABA-induced depressions to potentiation by EtOH. Indeed, we found that local applications of Rp-cAMP reversed the EtOH potentiation of Iso-modulated GABA responses. Although we also found that PKC activation can regulate the interactions of EtOH with GABA-induced depressions, these data clearly implicate the cAMP-mediated PKA activation in the observed beta adrenergic influence on EtOH interactions with GABA mechanisms.

The PKA involvement in EtOH potentiations of GABA effects on Purkinje neurons implicates a phosphorylation mechanism. In support of this hypothesis, we found that the phosphatase I and IIa inhibitors OkA and M-LR also modulated GABA responses and sensitized these GABA inhibitions to the potentiative effects of EtOH in a manner similar to that of Iso, 8-Br-cAMP and Sp-cAMP. Because these inhibitors prevent dephosphorylation at sites phosphorylated by PKA (Agostinis et al., 1987; Cohen, 1989), as well as other kinases, it seems likely that the phosphorylation state of some protein or set of proteins is critical for the interaction of EtOH with GABA_A receptors on cerebellar Purkinje neurons. Although these experiments with phosphatase inhibitors do not directly implicate the cAMP/PKA second-messenger system in the EtOH potentiation of GABA effects on these cells, the resulting data are consistent with the involvement of a kinase mechanism. Furthermore, because these experiments involve antagonism of dephosphorylation mechanisms, the data suggest that there is some spontaneous, endogenous kinase activity in these neurons, which is insufficient for the expression of this EtOH interaction with GABA responses under conditions of baseline phosphatase activity.

The data presented here suggest that EtOH potentiates GABA_A mechanisms on cerebellar Purkinje neurons indirectly through a cAMP/PKA second-messenger system in these cells or that regulation of the GABA_A complex by this second-messenger system also regulates its responsiveness to the potentiating effects of EtOH. Previous studies from other laboratories have also implicated post-translational modification in the EtOH potentiation of GABA effects in hippocampal CA1 neurons (Aguayo and Pancetti, 1994; Weiner et al., 1994) and in X. laevis oocytes expressing mouse brain mRNA for GABA_A receptor/Cl channels (Wafford and Whiting, 1992). Indeed, it has been reported that the long isoform of the  subunit (_2L) of the GABA_A receptor contains a consensus sequence for phosphorylation by PKC (Kofuji et al., 1991; Sikela et al., 1991; Whiting et al., 1990) and that the  subunit can be phosphorylated by both PKA and PKC (Browning et al., 1990). The _2L GABA_A receptor subunit is localized in brain areas (Criswell et al., 1995; Zahniser et al., 1992) in which EtOH has been found to potentiate the depressant effects of GABA (Aguayo and Pancetti, 1994; Criswell et al., 1993; Soldo et al., 1994; Weiner et al., 1994). Not only do we report here that antagonism of PKA activity prevents EtOH potentiation of GABA-induced depressions in the cerebellum, but also previous reports indicate that antagonists of PKA and PKC cause similar effects in hippocampal CA1 neurons (Weiner et al., 1994) and in oocytes expressing GABA_A receptors (Wafford and Whiting, 1992). Furthermore, intracellular ATP appears to be required for the potentiation of GABA inhibiting postsynaptic currents by EtOH but not by diazepam in the hippocampus (Weiner et al., 1994). Thus, the EtOH potentiation of GABA mechanisms in these cells could require not only certain GABA_A subunit conformations but also the post-translational regulation of these subunits by specific protein kinase mechanisms.

Although the EtOH potentiation of cerebellar GABA effects in the present study might appear to involve regulation of GABA mechanisms through a cAMP/PKA second-messenger system, a previous report indicates that EtOH did not alter the phosphorylation of purified GABA_A receptor protein by either PKC or the catalytic subunit of PKA in a cell-free system (Machu et al., 1991). These data suggest that any EtOH interaction with the kinase regulation of GABA_A function is either a direct effect of EtOH on the GABA_A mechanism, the expression of which is regulated by phosphorylation of the GABA_A receptor complex, or an influence of EtOH on the activation of protein kinase activity that, although absent in a cell-free system, would regulate the function of the GABA_A mechanism. Consistent with the latter hypothesis, we found in the present study that EtOH cannot potentiate GABA responses that are maximally modulated by 8-Br-cAMP. Furthermore, our data indicate that the EtOH potentiation of GABA responses that are submaximally modulated by 8-Br-cAMP is no greater than the maximal modulation caused by the cAMP agonist. Thus, EtOH effects on GABA depressions in the cerebellum appear to be limited by the responsiveness of the GABA mechanism to cAMP. Together with our findings that GABA responses are not potentiated by EtOH in the absence of cAMP modulation and that Rp-cAMP blocks these effects, these data suggest that cAMP mechanisms mediate this effect of EtOH. Indeed, Tabakoff and others previously found that EtOH acts to increase Iso- or guanine nucleotide-stimulated adenylate cyclase activity (Bode and Molinoff, 1988; Hoffman and Tabakoff, 1990; Luthin and Tabakoff, 1984; Saito et al., 1985). Their data suggested that EtOH might alter the action of the stimulatory guanine nucleotide binding protein, G_S. The resulting elevation of intracellular cAMP levels would, in turn, activate PKA. These findings together with our previous observation that the application of EtOH alone does not routinely potentiate GABA-induced depressions on cerebellar Purkinje neurons (Freund et al., 1993) but EtOH does potentiate these GABA responses if they are first facilitated slightly by beta adrenergic stimulation (Lin et al., 1991, 1993) could reflect an EtOH interaction with the cAMP second-messenger pathway. If so, then EtOH-induced alterations of endogenous beta adrenergic activity, by itself, must be insufficient to stimulate significant PKA influences on GABA_A receptor function in most of these cells under our experimental conditions. However, endogenous protein kinase activity may be sufficiently enhanced by cAMP analogues or phosphatase inhibitors in the present experiment to permit facilitation of GABA-induced depressions by EtOH on these cells.

We and others previously found that cAMP/PKA mechanisms mediate the beta adrenergic facilitation of GABA-induced depressions of neurons in the cerebellum and neocortex (Cheun and Yeh, 1992, Freund and Palmer, 1996; Hoffer et al., 1972; Llano and Gerschenfeld, 1992: Sessler et al., 1989) as well as of ganglion cells in the retina (Veruki and Yeh, 1991). Even so, PKA activation has been reported to not facilitate but rather to inhibit the function of GABA_A receptors in spinal trigeminal neurons (Song and Huang, 1990) as well as in HEK 293 cells transfected with GABA_A receptor subunits and in sympathetic ganglion cells in culture (Moss et al., 1992). It is possible that neurons in some areas of the brain do not express GABA responses that are potentiated by cAMP mechanisms, such as that observed in the cerebellum. Thus, although norepinephrine has been reported to elevate cAMP levels in hippocampus (Palmer et al., 1973) and cAMP has been reported to mimic the beta adrenergic electrophysiological effects of norepinephrine in that brain area (Madison and Nicoll, 1986), neither we² nor these investigators have been able to demonstrate robust modulation of GABA-induced depressions by beta adrenergic agonists on hippocampal CA1 pyramidal neurons. Differences in the expression of the GABA_A receptor in various neuronal populations might relate to the differences in cAMP regulation of GABA_A receptor function in any given cell type, and this might form a basis for differences in the interaction of EtOH with GABA mechanisms in those cells through a cAMP second-messenger system.

In conclusion, we have found evidence that subdepressant applications of EtOH on cerebellar Purkinje neurons will facilitate GABA-induced depressions through interactions with a cAMP/PKA second-messenger mechanism that regulates GABA_A function. As is discussed above, the evidence suggests that beta adrenergic mechanisms influence cAMP activity in cerebellar Purkinje cells, and that, thus, behavioral conditions that activate central catecholamine synapses, such as stress and arousal, might be expected to facilitate these low-dose EtOH interactions with GABA in this brain region. However, other neurotransmitter systems may regulate adenylate cyclase activity as well (Peroutka, 1994), and the activation of those receptors might also influence EtOH interactions with GABA mechanisms. Furthermore, other second-messenger systems, such as PKC, may also influence EtOH interactions with GABA mechanisms in some neurons (Aguayo and Pancetti, 1994; Weiner et al., 1994), and there may be EtOH-sensitive interactions between the various second-messenger systems. Indeed, we found here that PKC activation with the phorbol ester PDA also sensitizes GABA responses to the potentiating effects of EtOH. Further experiments will be required to elucidate EtOH effects on GABA mechanisms through alterations of interactions between the different second-messenger systems. However, it is becoming evident that subdepressant doses of EtOH do not directly facilitate GABA actions on Purkinje neurons in the absence of kinase regulation of the GABA_A complex. Rather, the data in this report suggest that EtOH may alter the second-messenger regulation of this GABA mechanism.