Introduction

Top Abstract Introduction Methods Results Discussion References

Determination of the neural mechanisms underlying the acute effects of alcohol (ethanol) and identification of the genetic factors mediating these effects are primary goals of current alcohol research. In part, this interest derives from recent evidence suggesting that sensitivity to alcohol's acute effects in humans may be predictive of later propensity to develop alcoholism (Pollock et al., 1986; Schuckit, 1988, 1994). Locomotor stimulation is one of many acute alcohol effects that are studied to gain insight into alcohol's mechanisms of action. In addition, locomotor activation in mice is a putative model of alcohol-induced euphoria in humans and may be particularly relevant for understanding alcohol's addictive properties (Lukas and Mendelson, 1988; Wise and Bozarth, 1987).

Selectively bred mouse and rat lines have been useful preparations for studying alcohol genetics and alcohol pharmacology. The FAST and SLOW mouse lines were created by bidirectional selective breeding for extreme sensitivity and insensitivity to the locomotor stimulant effects of a moderate dose of ethanol (EtOH; Crabbe et al., 1987; Phillips et al., 1991; Shen et al., 1995b). The activity selection project, responses of the FAST and SLOW lines across 36 generations of selection, and results of studies designed to identify correlated responses to selection have been recently documented (Shen et al., 1995b; Shen et al., 1996). A strength of this selection project is that replicated lines were produced. When differences found in one set of lines (e.g., FAST-1 vs. SLOW-1) are also found in the other set (e.g., FAST-2 vs. SLOW-2), strong evidence of a genetically correlated response to selection is obtained (Crabbe et al., 1990). Moderate evidence is provided by a difference found in only one set of replicated lines.

Previous research exploring mechanisms of EtOH stimulant sensitivity in FAST and SLOW mice examined glutamate (Daniell and Phillips, 1994; Shen and Phillips, 1998) and dopamine systems (Shen et al., 1995a). The noncompetitive NMDA antagonist, MK-801, stimulated the locomotor activity of FAST mice to a greater degree than that of SLOW mice and, in general, potentiated the expression of EtOH's locomotor depressant properties (Shen and Phillips, 1998). Ethanol decreased L-glutamate-stimulated increases in intracellular free calcium concentrations in brain microsac preparations from SLOW mice, but not in microsacs from FAST mice (Daniell and Phillips, 1994). Whereas dopaminergic agonists and antagonists had similar effects on the locomotion of FAST and SLOW lines in the absence of EtOH treatment, the stimulant response of FAST mice to EtOH was decreased by dopamine receptor antagonists (Shen et al., 1995a) and enhanced by dopamine agonists (Phillips and Shen, 1996).

Ethanol is thought to produce at least some of its effects via interactions with the GABA_A receptor (reviewed by Deitrich et al., 1989) although the precise site or sites of interaction are not known. The GABA_A receptor is a macromolecular complex that contains several specific binding sites for GABA, picrotoxin, benzodiazepines, anesthetic steroids and barbiturates. Studies during the course of selection of the FAST and SLOW lines determined them to be differentially sensitive to the locomotor stimulant effects of several alcohols and two barbiturates (pentobarbital and phenobarbital), with FAST mice exhibiting larger stimulant responses in each case (Phillips et al., 1992). These results suggested to us that GABA_A receptor modulated systems might play an important role in determining the differences in EtOH sensitivity between these sets of selected lines.

In the current work, two approaches were used to examine differences in sensitivity of GABA systems between the FAST and SLOW lines. Chemically induced convulsions were utilized as a simple means of identifying sensitivity differences, and the effects of GABAergic drugs on locomotion in the absence and presence of EtOH were assessed. In the latter studies, drugs interacting with GABA_B receptors were also investigated to follow up interesting reports of GABA_B receptor involvement in EtOH's stimulant effects (Allan and Harris, 1989; Cott et al., 1976; Humeniuk et al., 1993).

The convulsant effects of seven drugs were tested. Since the focus of this particular work was primarily on GABAergic systems, the drugs were chosen for actions at three sites on the GABA_A receptor subtype. To address the specificity of GABA mediation of line differences in convulsant sensitivity, the effects of two drugs thought to act through non-GABAergic mechanisms were also evaluated. Previous work demonstrated a positive relationship between convulsant sensitivity and the locomotor stimulant effects of EtOH in selectively bred Long-Sleep (LS) and Short-Sleep (SS) mice (Phillips et al., 1989a). Thus, we hypothesized that FAST mice would be more sensitive to GABAergic convulsants than SLOW mice, but predicted that there would be no systematic line differences in sensitivity to non-GABAergic convulsants.

Effects of GABAergic drugs on locomotion and in altering the effects of EtOH on locomotor activity were studied as more directly relevant to the selection phenotype. Based on our previous results with barbiturates and alcohols, we predicted that FAST mice would be more sensitive to any locomotor stimulant effects and SLOW mice would be more sensitive to the locomotor depressant effects of the GABA_A receptor drugs. GABA_B receptor involvement was tested by determining the effects of the agonist, baclofen, on locomotor activity of FAST and SLOW mice. Based on previous reports in the literature, we predicted that FAST and SLOW mice would be differentially sensitive to baclofen's locomotor effects and that baclofen administration would decrease EtOH-stimulated locomotion in FAST mice. The involvement of a particular neurotransmitter system in mediating locomotor activation by EtOH is more strongly supported when results are similar among 2 or more sets of independently derived genetic models. Thus, when our expectations were met in FAST and SLOW mice, we further tested our hypothesis by assessing baclofen's effects in a pair of inbred strains with extreme sensitivity and insensitivity to the stimulant effects of EtOH (Dudek et al., 1991; Phillips and Crabbe, 1991).

	Methods

Top Abstract Introduction Methods Results Discussion References

Animals. Selective breeding of FAST and SLOW mice has been described in detail (Shen et al., 1995b). Briefly, two replicate sets of FAST and SLOW mice were concurrently bred for high and low (or negative) magnitudes of locomotor activation, respectively, in response to 2.0 g/kg EtOH. After 36 generations of selective breeding, the limits of selection appeared to have been reached and mice were placed on a relaxed breeding scheme (no selection pressure) on which they are currently maintained. Inbreeding is minimized by avoiding mating of animals with common grandparents. FAST and SLOW mice from generations S₃₆G₃₉-S₃₆G₄₉ and generations S₂₂-S₃₆G₃₇ were used in studies of sensitivity to chemical convulsants and studies of locomotor activity, respectively. Male mice were used in all convulsant studies. Males and females were used in locomotor activity studies, depending on availability, but only one sex was used in any one experiment. There was one exception; males and females were tested for locomotor responses to picrotoxin in the absence of EtOH due to low availability of either sex. We have not observed systematic sex differences in the selection response (Shen et al., 1995b), nor have we observed changes in selection response due to relaxed breeding (Shen et al., 1996). FAST and SLOW mice were produced and raised in the breeding facilities at the Portland VA Medical Center. They were reared by dam and sire until 21 ± 2 days of age, when they were separated into isosexual groups of 2 to 4 mice per cage. C57BL/6J and DBA/2J mice were purchased from Jackson Laboratory (Bar Harbor, ME) and allowed to acclimate in our facilities for 1 to 2 weeks before activity testing. Animals were housed in cages made of clear polycarbonate (28 × 18 × 13 cm) with corn cob bedding that was changed twice weekly. Cages were never changed on a test day. Animals were 7-19 weeks old at time of testing, and were offered unlimited water and rodent block food at all times except during testing. Ambient temperature in breeding, colony and testing rooms was a constant 21 ± 2°C, and the light cycle in all rooms was 12-hr light/12-hr dark, with full illumination beginning at 6 a.m.

Convulsant behavior. Timed tail vein infusions were used to assess convulsant sensitivity. All testing was performed between 9 a.m. and 12 noon. Mice were moved to a testing room 45 to 60 min before testing. Just before testing, each mouse was weighed, placed in a clear acrylic plastic restraining tube, and its tail was immersed in a warm water bath (40-45°C) for 30 to 45 sec to dilate the tail veins. A butterfly needle (27 g, 3/8 in.; Abbott Laboratories) was inserted into the lateral tail vein at approximately the midlength point of the tail. Proper insertion was verified by the appearance of blood in the infusion tubing. Infusion began immediately and latencies to the occurrence of myoclonus, face and forelimb clonus, running/bouncing clonus and tonic hind limb extension (THE) seizures were recorded to the nearest second. Some drugs produced only a subset of these convulsion signs. Drug was delivered by a calibrated Sage Instruments infusion pump (model 355). Infusion rates (0.25-0.5 ml/min) and drug concentrations (0.01-5 mg/ml) were chosen from prior work (Kosobud and Crabbe, 1990) to keep the infusion time under three min and the infusion volume under 1.0 ml for each drug. In those cases when the endpoint convulsion was not terminal, mice were immediately euthanized by cervical dislocation.

Convulsant drugs. The drugs used were picrotoxin (Sigma), t-butyl-bicyclo-2,2,2-phosphorothionate (TBPS; Research Biochemicals) and pentylenetetrazol (PTZ; Sigma), all thought to act at the picrotoxin GABA_A site, bicuculline (Research Biochemicals), which binds to the GABA site, methyl-6,7-dimethoxy-4-ethyl--carboline-3-carboxylate (DMCM; Research Biochemicals), which is an inverse agonist at the benzodiazepine receptor site, kainic acid (Sigma), an excitatory amino acid agonist, and strychnine sulfate (Sigma), thought to produce convulsions via a glycine receptor associated chloride channel. All drugs were prepared on the day of the experiment. PTZ, picrotoxin, DMCM and strychnine sulfate were dissolved in saline. Bicuculline was dissolved in a few µl of concentrated hydrochloric acid (37%), diluted with saline, and adjusted to pH 7.26-7.41 with 5 M NaOH. Kainic acid was dissolved in distilled water, and the pH adjusted with 5 M NaOH. TBPS was suspended in 2% DMSO in distilled water without pH adjustment, and was vortexed just before each infusion.

Locomotor activity testing. Locomotor activity was assessed using Omnitech Digiscan Activity Monitors (Columbus, OH), interfaced with an IBM-compatible computer. Animals were placed in a clear acrylic plastic test box (40 cm × 40 cm) set inside an activity monitor that had 8 photocell beams and detectors evenly spaced along each of its 4 sides, approximately 2 cm above the floor. The test box and activity monitor were housed inside a sound-attenuating chamber made of opaque black acrylic plastic. The chamber was equipped with a small ventilation fan mounted on the rear right wall that also provided masking noise. Locomotor activity was assessed with white fluorescent lights on during studies with GABA_A drugs and with lights off during GABA_B experiments, solely because this was consistent with other ongoing work that these studies originally accompanied. Since we are not attempting to compare results across drugs, we do not believe that this environmental difference impacts interpretation of the current results. Ethanol's stimulant effects can be equally well measured in the light (Shen et al., 1995a) or the dark (Dudek et al., 1991). Sound attenuation was made possible with foam lining on the walls, ceiling, and door of the testing chamber. Data were automatically recorded in 5-min time samples.

Drugs used in locomotor activity studies. Diazepam was a generous gift from Ed Gallaher. Muscimol, an agonist for the GABA site on the GABA_A receptor, and picrotoxin were purchased from Sigma. Bicuculline, an antagonist at the GABA site, and the GABA_B receptor agonist, baclofen, were obtained from Research Biochemicals. Midazolam, a short-acting benzodiazepine, was obtained from Hoffman-LaRoche. CGP-35348 was a generous gift from Ciba-Geigy. Diazepam was dissolved in 10% -cyclodextrin (Research Biochemicals) in saline; all other drugs were dissolved in saline. Ethanol was obtained from Pharmco Products and was diluted to 20% (v/v) in saline. All drug solutions were prepared on the day of the experiment.

Determination of blood ethanol concentrations. In experiments involving EtOH treatment, a 20 µl blood sample was obtained from the retro-orbital sinus of each EtOH-treated mouse immediately following the conclusion of the activity test. Each blood sample was immediately placed in a microcentrifuge tube containing 50 µl ice-cold ZnSO₄ and further processed as previously described (Shen and Phillips, 1998). Analysis of BEC was performed using gas chromatography with flame ionization detection (Hewlett-Packard 5890).

Data analyses. For the convulsion studies, latency to each convulsion sign was converted to dose (mg/kg) for each animal. Distance traveled (cm) was the dependent measure for the locomotor activity studies. Analysis of variance (ANOVA), performed using the CRUNCH4 statistical package, was used to detect significant differences between genotypes, replicates, and drug treatments as appropriate for each study. Data for both replicates are presented in the figures when the line × replicate interaction was statistically significant; otherwise data are shown collapsed on replicate. Significant two-way interactions were further characterized by simple effects analysis followed by Tukey post-hoc mean comparisons when appropriate. A P < .05 was considered to indicate a significant difference in all statistical tests.

	Results

Top Abstract Introduction Methods Results Discussion References

Convulsant Sensitivity

Data for all seizure signs produced by each convulsant, except DMCM, are presented in figure 1; however, some convulsants produced different subsets of all possible seizure signs. There were significant line differences in response to all of the convulsants with GABA_A receptor complex sites of action, with the exception of DMCM. In some instances the line difference emerged in only one of the replicate pairs of lines, and was present for some, but not for other, seizure types. In these cases, ANOVA grouped on line and replicate revealed significant line × replicate interactions. Sensitivity differences between selected lines and replicates, as well as F-values for significant main effects of line and line × replicate interactions, are summarized in table 1. FAST mice were significantly more sensitive to the convulsant effects of picrotoxin, bicuculline, TBPS, and PTZ. The dose of strychnine required to produce some seizure signs was also lower in FAST than in SLOW mice, indicating greater sensitivity of the FAST line. This line difference was apparent in only the second replicate pair of lines. In contrast to all other drugs, SLOW mice of one replicate were significantly more sensitive to THE seizures induced by kainic acid. Finally, DMCM consistently produced only face and forelimb seizures, and there was no difference in sensitivity between FAST and SLOW mice. The mean DMCM doses required to produce face and forelimb seizures were 1.21 ± 0.08 mg/kg and 1.32 ± 0.05 mg/kg (mean ± S.E.) in FAST and SLOW mice, respectively.

View larger version (35K): [in this window] [in a new window]   Fig. 1.   Sensitivity of FAST and SLOW mice to seizures induced by timed tail-vein infusion of convulsant drugs. Shown is the average dose (±S.E.) in mg/kg required to induce the seizure sign indicated on the x-axis. Replicate lines (REP1, REP2) are shown separately when significant differences were statistically supported. Mice were from generations S36G39-S36G49; n = 10-17 per line per replicate for each drug.

Locomotor Activity

Effects of GABA_A drugs. Of all the GABA_A drugs tested, only the benzodiazepine receptor agonists, diazepam and midazolam, produced significant line differences in distance traveled. Cumulative distance traveled during 15 min of activity testing is presented for midazolam and diazepam in figures 2 and 3, respectively. In addition, examination of time-course data indicated that locomotor stimulant effects could be seen during the first 5 min after injection. Since this is the time period upon which selective breeding of FAST and SLOW mice was based, 5-min data are shown as well (see figs. 2 and 3 insets). Midazolam significantly increased the locomotor activity of FAST mice at 5 min (P < .05 for 2.5 mg/kg; P < .01 for 5-10 mg/kg) and during the first 15 min (5 mg/kg, P < .05) of the test, but had no significant effect on SLOW mice at either time point. In contrast, although line differences were consistently observed, the pattern of response to diazepam was dependent on the time frame analyzed. At 5 min post-injection, FAST mice were significantly activated (P < .05 for 4 and 16 mg/kg; P < .01 for 8 mg/kg) and SLOW mice were significantly depressed by diazepam (8 mg/kg, P < .05). Significant locomotor depression was still observed for SLOW mice when cumulative 15-min data were analyzed (P < .01 for all doses); however, the activity of FAST mice was not significantly altered by any diazepam dose because the stimulation observed early in the activity test was masked by subsequent mild locomotor depression. Doses of diazepam tested in these studies differed between FAST and SLOW mice because preliminary studies suggested robust line differences in responsiveness to diazepam, leading to an extension of the doses used to test FAST mice. To conserve animals, two lower diazepam doses were excluded from the dose range tested in FAST mice. Locomotor activity of FAST and SLOW mice differed significantly at both time points for common doses tested (P < .001), but did not differ in baseline activity.

View larger version (18K): [in this window] [in a new window]   Fig. 2.   Differential locomotor responses of S23 FAST and SLOW male mice to midazolam. Mice were injected with midazolam and immediately tested for 15 min with fluorescent lights on in the activity test chambers. Data for distance traveled are shown collapsed on replicate. Inset shows midazolam effects on distance traveled during the first 5 min of the activity test. At 5 min, 2.5, 5 and 10 mg/kg significantly (P < .01) increased locomotor activity of FAST mice. At 15 min, FAST mice were significantly activated by 5 mg/kg (P < .05). No significant effects of midazolam were detected in SLOW mice. S.E. larger than symbol size are shown; n = 18-22 per treatment group.

View larger version (18K): [in this window] [in a new window]   Fig. 3.   Differential locomotor responses of S24 and S25 FAST and SLOW female mice to diazepam. Mice were injected with diazepam and immediately tested for 30 min with fluorescent lights on in the activity test chambers. Distance traveled during the first 15 min is presented for comparison with midazolam results. Inset shows drug effects during the first 5 min of the activity test. At 5 min, 4 and 16 (P < .05), and 8 mg/kg (P < .01) significantly increased locomotor activity of FAST mice, and 8 mg/kg decreased the activity of SLOW mice (P < .05). At 15 min, all doses of diazepam decreased activity of SLOW mice (P < .01), but FAST mice were not significantly affected. S.E. larger than symbol size are shown; n = 19-21 per treatment group.

View larger version (16K): [in this window] [in a new window]   Fig. 4.   Lack of line sensitivity differences in locomotor response to the GABAA antagonist, bicuculline. S27-S28 FAST and SLOW male mice were injected with saline or bicuculline, and after a 15-min wait, were injected with saline and immediately tested for locomotor activity for 15 min. Fluorescent lights were on in the test chambers. This protocol was used to mimic testing conditions in a separate experiment designed to assess the effects of bicuculline on EtOH-treated animals. Doses of 2.5 and 3 mg/kg bicuculline significantly decreased locomotor activity of both selected lines (P < .01). Vertical bars are S.E.; n = 16 per line and treatment group.

View larger version (15K): [in this window] [in a new window]   Fig. 5.   Lack of differential locomotor responses in FAST and SLOW mice to the locomotor depressant effects of picrotoxin. Male and female FAST and SLOW mice (S23) were injected with saline or picrotoxin and immediately tested in activity monitors for 15 min with fluorescent lights on. FAST mice were generally more active than SLOW mice. Significant locomotor depression resulted from administration of 1 and 2 mg/kg picrotoxin for all mice (P < .01). Vertical bars are S.E.; n = 21-23 per line and treatment group.

View larger version (16K): [in this window] [in a new window]   Fig. 6.   Similar locomotor effects of the GABAA agonist, muscimol, in S22 FAST and SLOW female mice. Animals were injected with saline or muscimol, followed by a 5-min wait before placement in lighted activity chambers for 15 min. FAST mice were generally more active than SLOW mice. Significant locomotor depression resulted from administration of 2 mg/kg muscimol in all genotypes (P < .01). Vertical bars are S.E.; n = 19-20 per line and treatment group.

Effects of GABA_B drugs in FAST and SLOW mice. The GABA_B agonist baclofen had locomotor depressant effects on EtOH-naïve FAST and SLOW mice (fig. 7). For replicate 1 animals, significant locomotor depression was observed after administration of 2.5 and 5 mg/kg baclofen in SLOW-1 mice, and after only 5 mg/kg in FAST-1 mice (P < .01 for each dose, compared to appropriate saline control). In replicate 2, the locomotor activity of both FAST and SLOW mice was reduced to the same extent by 2.5 and 5 mg/kg baclofen (P < .01). Thus, a line sensitivity difference in response to baclofen was observed in replicate 1 mice only.

View larger version (19K): [in this window] [in a new window]   Fig. 7.   Effects of baclofen on locomotor responses of male FAST and SLOW mice. Animals from S33 were injected with baclofen 15 min before a second injection of saline, followed by testing in lighted activity chambers. Total distance traveled during the 20-min activity test is shown separately for the replicate lines. Replicate 1 SLOW mice were more sensitive to the locomotor depressant effects of baclofen (significant effects at 2.5 and 5 mg/kg, P < .01) than FAST mice whose locomotor activity was decreased by 5 mg/kg (P < .01). Baclofen (2.5 and 5 mg/kg, P < .01) decreased locomotor activity of replicate 2 FAST and SLOW mice to the same degree. S.E. larger than symbol size are shown; n = 10-11 per line, replicate and dose.

View larger version (16K): [in this window] [in a new window]   Fig. 8.   Dose-dependent reduction of EtOH-stimulated activity by the GABAB agonist, baclofen, in S33-S34 FAST male mice. Significant reductions and complete blockade were observed in animals given 2.5 and 5.0 mg/kg baclofen, respectively (P < .01). Mice were injected with saline or baclofen 15 min before injection of saline or EtOH (2.0 g/kg). A 20-min activity test in darkened activity chambers (fluorescent lights off) immediately followed the second injection. Vertical bars are S.E.; n = 20 per treatment group.

View larger version (16K): [in this window] [in a new window]   Fig. 9.   Reversal of baclofen's inhibitory effects on EtOH-stimulated activity by the GABAB antagonist, CGP-35348. FAST male mice (S36G37) were injected with a single solution containing 4 mg/kg baclofen and saline or CGP-35348. EtOH (2.0 g/kg) was injected 15 min later, followed by a 20-min activity test in dark chambers. Saline control animals, shown to the left of the vertical dashed line, were tested in the same manner, but received two saline injections. Significant reversal was observed at both 50 (P < .05) and 100 (P < .01) mg/kg CGP-35348. Vertical bars are S.E., n = 20 mice per treatment group.

Effects of GABA_B drugs on inbred strains. Given our findings in FAST mice indicating GABA_B receptor mediation of EtOH-stimulated activity, we sought to strengthen evidence for this conclusion by testing a pair of inbred strains known to differ drastically in their sensitivity to this EtOH effect. Results from the relatively insensitive B6 strain and sensitive D2 strain confirmed such a role for GABA_B receptors (fig. 10). Because B6 mice have demonstrated a slight stimulant response to EtOH in some studies (Crabbe et al., 1982), they were tested with the more stimulated D2 strain. However, in the current study B6 mice were not activated by EtOH, and an assessment of baclofen's effects on EtOH stimulation in this strain was not possible. Interestingly, the combination of 2.5 mg/kg baclofen and EtOH appeared to potentiate locomotor depression in B6 mice, whereas this baclofen dose had a small enhancing effect on locomotion in non-EtOH treated mice. The highest baclofen dose significantly decreased the locomotor activity of both saline and EtOH-treated B6 mice (P < .01). On the other hand, consistent with results obtained in FAST mice, baclofen blocked the expression of EtOH-stimulated activity in D2 mice at doses that had no effect on baseline activity. Further analysis of a significant baclofen × EtOH interaction (F[4,90] = 10.4, P < .001) in D2 mice revealed that both 0.625 and 1.25 mg/kg baclofen significantly reduced EtOH-stimulated activity (P < .01). Higher doses significantly reduced the activity of EtOH-treated D2 mice (P < .01). The activity of saline-treated D2 mice was also reduced by 5 mg/kg baclofen (P < .01), but 2.5 mg/kg had a slight stimulant effect on saline-treated animals. The strains did not differ significantly in locomotor response to baclofen alone.

View larger version (18K): [in this window] [in a new window]   Fig. 10.   Inhibition of EtOH's stimulant effects in DBA/2J mice by baclofen. Male mice were injected with saline or baclofen 15 min before injection of saline or EtOH (1.5 g/kg). Locomotor testing took place in darkened test chambers for a period of 20 min. Baclofen (0.625-5 mg/kg; P < .01) significantly reduced EtOH-stimulated activity in D2 mice, and significantly reduced activity of EtOH-treated B6 mice (2.5 and 5.0 mg/kg, P < .01). Locomotor activity of saline-treated mice of both inbred strains was slightly enhanced by 2.5 mg/kg (P < .05) and significantly reduced by 5 mg/kg baclofen (P < .01). S.E. larger than symbol size are shown; n = 10 per treatment group.

View larger version (16K): [in this window] [in a new window]   Fig. 11.   Reversal of baclofen's inhibitory effect on EtOH-stimulated activity by CGP-35348 in male DBA/2J mice. Procedural details follow those outlined in figure 9 legend, except mice in this study received 1.5 mg/kg baclofen, and 1.5 g/kg EtOH. Significant reversal of baclofen's effects was observed at both 50 and 100 mg/kg CGP-35348 (P < .01). Vertical bars are S.E.; n = 10 per treatment group.

	Discussion

Top Abstract Introduction Methods Results Discussion References

The role of GABAergic systems, specifically GABA_A and GABA_B receptors, in mediating the stimulant actions of EtOH were investigated in FAST and SLOW selected mouse lines, and subsequently in B6 and D2 inbred strains (GABA_B studies only). Consistent with work in earlier generations of these mice (Phillips et al., 1992), differential sensitivities between FAST and SLOW mice were observed in response to some GABA_A ligands tested. However, the GABA_A receptor antagonists, bicuculline and picrotoxin, did not selectively reduce the EtOH stimulant response. On the other hand, FAST and SLOW mice differed in locomotor responses to baclofen, and EtOH's stimulant effects in FAST mice were blocked by administration of this GABA_B agonist. These GABA_B effects were replicated in D2 inbred mice that display high sensitivity to the stimulant actions of EtOH. Taken together, these data suggest a role for GABA systems in mediating or modulating locomotor stimulant actions of EtOH. However, precise mechanisms of action, and interactions of GABA systems with other neurotransmitter systems within the nervous system that are responsible for this behavioral phenomenon are yet to be elucidated.

Studies of convulsant sensitivity in FAST and SLOW mice generally showed line differences in one replicate, but not the other, which we interpret as moderate evidence for genetically correlated selection responses (Crabbe et al., 1990). This type of result may indicate spurious genetic correlations arising from chance fixation of trait-irrelevant alleles, may indicate that the subset of genes common to both phenotypes is small in number or effect, or that some of the important alleles were lost in one or the other set of replicate lines. Our data consistently demonstrated a greater sensitivity of FAST mice to the convulsant effects of the drugs tested, with the exception of kainic acid and DMCM. These data are in agreement with our previous suggestion that FAST mice are generally more sensitive to central nervous system excitation, as supported by a more severe EtOH withdrawal syndrome in FAST compared to SLOW mice (Shen et al., 1996). However, given the kainic acid and DMCM results we would not conclude that FAST mice are more susceptible to convulsions in general. Our current results are also consistent with previous reports of an association between EtOH stimulant sensitivity and seizure susceptibility in selectively bred LS and SS mice (Bowers et al., 1991; Peris et al., 1989; Phillips et al., 1989a). However, data obtained in the LSXSS recombinant inbred strain panel did not support such an association (Wehner et al., 1991). If, as suggested above, genetic factors common to both phenotypes are small in number or effect, the ability to detect a genetic correlation may be maximized in selected lines, but more difficult to detect in inbred strains, leading to such inconsistencies.

Administration of benzodiazepines generally resulted in locomotor stimulation that differed between FAST and SLOW mice. Locomotor depression was elicited by the other GABA_A agents tested, bicuculline, picrotoxin and muscimol. It is interesting that these drugs, with seemingly opposite effects on chloride flux through the GABA channel, produced the same locomotor response. Contradictory effects such as these have been reported in the literature (see Phillips and Shen, 1996), and it is not yet entirely clear how or why they occur. However, GABAergic systems are ubiquitous throughout the brain and form connections with many other neurotransmitter systems that could possibly affect locomotor behavior in opposite directions. Furthermore, the subunit composition of GABA_A receptors may differ from one neuroanatomical region to the next, which may in turn affect the affinity of a particular ligand for the receptor. Thus, depending on the ligand, the receptor affinity and location, and the neural connections, it is conceivable that an agonist and an antagonist could produce the same locomotor response.

Differences in sensitivity to GABA_A ligands were sometimes dependent on the behavioral measure. For example, the only GABA drug tested that did not differentiate FAST and SLOW mice in convulsant sensitivity was DMCM, which acts as an inverse agonist at the benzodiazepine site. When locomotor activity in response to GABA_A ligands was evaluated, the two benzodiazepines tested were the only drugs that produced different locomotor responses between the lines. Conversely, FAST and SLOW mice differed in sensitivity to the convulsant effects, but not the locomotor effects, of bicuculline and picrotoxin. The reason for these discrepancies is unclear. One possible explanation is that the two phenotypes, though both mediated in part by GABA receptors, are ultimately mediated by different neural pathways. For example, doses of picrotoxin and bicuculline necessary to induce convulsions are much higher than doses that alter locomotor activity. High doses of these drugs may affect neural pathways different from those affected by low doses, perhaps by preferentially acting on GABA_A receptors with different subunit composition located in a distinct neuroanatomical location.

GABA_B agonist administration blocked EtOH-stimulation in two mouse genotypes with extreme sensitivity to this acute EtOH effect. Whereas FAST mice were specifically bred for this extreme stimulant response, this characteristic of D2 mice arose by chance during inbreeding. The similar results obtained in these independent genetic models provide more compelling evidence for GABA_B mediation of EtOH-stimulated activity than either model alone. In addition, the current results are consistent with effects observed in other mouse genotypes (Cott et al., 1976; Humeniuk et al., 1993), and with effects on other acute measures of EtOH action (Allan and Harris, 1989). Restoration of EtOH-stimulated activity by CGP-35348 in D2 and FAST mice demonstrated GABA_B receptor specificity of this effect. Taken together, these data strongly support a role for GABA_B systems in mediating or modulating EtOH-stimulated locomotor activity.

Although our data support the involvement of GABA_B receptors in expression of EtOH-stimulated activity, we obtained only moderate evidence for differential alteration of GABA_B receptor systems during selective breeding of the FAST and SLOW lines, and no evidence for differences between B6 and D2 mice. Two of our previous studies are of relevance here. First, we have demonstrated the involvement of dopaminergic systems in mediating EtOH-stimulated activity, even in the absence of line differences in sensitivity to dopamine antagonists alone (Shen et al., 1995a). In another set of studies, MK-801 altered locomotor responses of all EtOH-treated genotypes, and differentially altered the locomotor activity of EtOH-naïve FAST and SLOW but not B6 and D2 mice (Shen and Phillips, 1998). Overall, these data sets have suggested to us that innate differences in receptor function (whether NMDA, DA, or GABA_B) may contribute, but are not necessary, for differential sensitivity to EtOH's stimulant effects.

Locomotor depressant effects of baclofen in FAST mice in the dose-response study were not replicated in the subsequent experiment that tested baclofen's effects in saline or EtOH-treated mice. One possible explanation for this discrepancy is that the dose-response study was conducted with lights on, while the EtOH study was conducted with lights off. It has previously been demonstrated that changes in lighting conditions affect expression of EtOH-stimulated activity (Crabbe et al., 1988) and it is likely that locomotor activity in response to other drugs is sensitive to such environmental changes.

The baclofen dose required to block EtOH-stimulated activity was much lower in D2 mice compared to FAST mice. This can be attributed to the lower dose of EtOH used, the slightly lower magnitude of stimulation seen in D2 mice, or may perhaps reflect differential sensitivity to baclofen among these genotypes. A lower dose of CGP-35348 was required to completely reverse the inhibition of EtOH-stimulated activity in D2 mice compared to FAST mice, probably due to the lower dose of baclofen required to block locomotor activation. B6 mice were not activated by acute EtOH administration in our studies, and doses that altered the activity of EtOH-treated D2 mice did not affect EtOH-treated B6 mice. However, the highest dose of baclofen tested produced a sharp decrease in locomotor activity in both saline and EtOH-treated groups of both inbred strains, demonstrating a locomotor depressant effect of baclofen at higher doses. There may have been potentiation of locomotor depression in B6 mice by one dose combination of baclofen and EtOH. This would not be surprising given that B6 mice are sometimes demonstrated to have an enhanced sensitivity to the locomotor depressant effects of EtOH (Phillips and Crabbe, 1991).

The possible involvement of GABA_A and GABA_B receptors in EtOH-stimulated activity is interesting because these receptors are thought to be crucial components of the neural circuitry that mediates locomotor activation in response to novelty, psychomotor stimulants, and other drugs of abuse (Amalric and Koob, 1993; Hooks and Kalivas, 1995; Kalivas et al., 1990). The mesoaccumbens-pallidal circuit comprises dopaminergic neurons that project from ventral tegmental area (VTA) to nucleus accumbens where they form connections with GABAergic neurons that in turn project to ventral pallidum/substantia innominata (VP/SI). Neuronal activity in the nucleus accumbens is modulated by glutamatergic projections, partly from hippocampus and prefrontal cortex. GABA_A receptors are thought to be present in VP/SI where they mediate output to the extrapyramidal system. GABA_B receptors are thought to modulate activity of the dopaminergic projections by actions at their cell bodies in VTA. The results reported here, along with previous reports of dopaminergic (Phillips and Shen, 1996; Shen et al., 1995a) and glutamatergic (Shen and Phillips, 1998) involvement in EtOH-stimulated activity of FAST mice, suggest that the mesoaccumbens-pallidal circuit may mediate EtOH's stimulant effects. This possibility is appealing in that it provides a common mechanism for the actions of several drugs of abuse, and may be related to the addictive potential of these drugs. In addition to acute locomotor effects, some components of the mesoaccumbens-pallidal circuit may be involved in the phenomenon of locomotor sensitization to psychostimulants (Pierce and Kalivas, 1997). Our immediate future plans include pursuing this possibility for EtOH-induced locomotor sensitization. In addition, we intend to investigate GABA_B receptor differences between FAST and SLOW mice using radioligand binding and to establish the locus of baclofen's effects on EtOH-stimulated activity in the central nervous system using central administration procedures.