Role of GABAA receptors in alcohol use disorders suggested by chronic intermittent ethanol (CIE) rodent model

Abstract

GABAergic inhibitory transmission is involved in the acute and chronic effects of ethanol on the brain and behavior. One-dose ethanol exposure induces transient plastic changes in GABA_A receptor subunit levels, composition, and regional and subcellular localization. Rapid down-regulation of early responder δ subunit-containing GABA_A receptor subtypes mediating ethanol-sensitive tonic inhibitory currents in critical neuronal circuits corresponds to rapid tolerance to ethanol's behavioral responses. Slightly slower, α1 subunit-containing GABA_A receptor subtypes mediating ethanol-insensitive synaptic inhibition are down-regulated, corresponding to tolerance to additional ethanol behaviors plus cross-tolerance to other GABAergic drugs including benzodiazepines, anesthetics, and neurosteroids, especially sedative-hypnotic effects. Compensatory up-regulation of synaptically localized α4 and α2 subunit-containing GABA_A receptor subtypes, mediating ethanol-sensitive synaptic inhibitory currents follow, but exhibit altered physio-pharmacology, seizure susceptibility, hyperexcitability, anxiety, and tolerance to GABAergic positive allosteric modulators, corresponding to heightened alcohol withdrawal syndrome. All these changes (behavioral, physiological, and biochemical) induced by ethanol administration are transient and return to normal in a few days. After chronic intermittent ethanol (CIE) treatment the same changes are observed but they become persistent after 30 or more doses, lasting for at least 120 days in the rat, and probably for life. We conclude that the ethanol-induced changes in GABA_A receptors represent aberrant plasticity contributing critically to ethanol dependence and increased voluntary consumption. We suggest that the craving, drug-seeking, and increased consumption in the rat model are tied to ethanol-induced plastic changes in GABA_A receptors, importantly the development of ethanol-sensitive synaptic GABA_A receptor-mediating inhibitory currents that participate in maintained positive reward actions of ethanol on critical neuronal circuits. These probably disinhibit nerve endings of inhibitory GABAergic neurons on dopamine reward circuit cells, and limbic system circuits mediating anxiolysis in hippocampus and amygdala. We further suggest that the GABA_A receptors contributing to alcohol dependence in the rat and presumably in human alcohol use disorders (AUD) are the ethanol-induced up-regulated subtypes containing α4 and most importantly α2 subunits. These mediate critical aspects of the positive reinforcement of ethanol in the dependent chronic user while alleviating heightened withdrawal symptoms experienced whenever ethanol is absent. The speculative conclusions based on firm observations are readily testable.

Keywords: Chronic intermittent ethanol; GABAA receptors; Inhibitory synaptic plasticity; Rodent model of alcoholism.

Fig. 1

Time course of behavioral state and PTZ seizure threshold in rats given EtOH by gavage. a. Cartoon representation of behavioral state over time after administration of EtOH by oral intubation (gavage) in rat. EtOH exhibits maximum absorption into the brain by ~2 h, accompanied by behavioral depression. As the EtOH leaves the brain, activity (arbitrary units, amplitude depends on dose) returns to normal. Before the EtOH is even eliminated, the behavioral activity returns to normal and overshoots to produce a rebound hyperexcitability (withdrawal), then returns to normal by 24 h (blue diamonds). CIE after 5 doses (pink squares), reduces initial depression (tolerance) and slows return to normal with heightened severity of rebound hyperexcitability. After 60 doses (open triangles) in rats (30 in mice) the heightened withdrawal does not return to normal and stays elevated for at least 40–120 days, possibly for life [109]. This is the CIE ‘kindled’ state. b. Effect of CIE on PTZ seizure threshold: persistent decrease after cessation of EtOH treatment. EtOH, 5.0 g/kg/48 h, was given by oral intubation; PTZ seizure threshold was measured 18 h after EtOH. CIV rats tested at the same times as the CIE rats showed no significant changes in PTZ seizures. Horizontal bars indicate mean PTZ seizure threshold. ** p < 0.01. Reproduced from Kokka et al. (1993) [109] with permission. * p < 0.05

Fig. 2

Plastic changes in GABA_AR subunits and currents in rat hippocampal formation induced by CIE. A. EtOH-enhanced mIPSCs observed in hippocampal slices from CIE vs. CIV. Top left of A, recordings from CIV and CIE, including exposure to various concentrations of EtOH in the recording chamber. Top right of a, averaged mIPSC from each period response to EtOH applications during the recordings (left of a). Bottom of a, Summary of mIPSC area and tonic current for EtOH vs. pre-EtOH application. Redrawn from Liang et al., [81]. b. Upper: Summary of Western blot analyses of hippocampal GABA_AR subunit peptides after CIE compared with CIV. Data are presented as percent changes from control peptide levels mean ± SEM. (n = 10 ~ 12 rats). ** p < 0.01, t-test. b Lower: GABA_AR subunit mRNA levels assayed by PCR, normalized to the unchanged reference gene GADPH. Data are expressed as percentage of CIV group (control) mean ± SEM, ** p < 0.01, t-test. c. Post-embedding immunogold labeling reveals a change in α4 but not in δ subunit location from perisynaptic to synaptic sites in the molecular layer of the DG after CIE. In CIV (top and middle of c), colloidal gold labeling of the α4 subunit (arrows) was present on or near the plasma membrane of dendrites that contacted axon terminals (T). Gold particles were found predominantly at the outer edges of symmetric synapses (arrows) but not at the center of these synapses (arrowheads). After CIE (bottom of c), labeling for α4 was found mainly in the center of symmetric synapses (arrows). d. Quantitative analysis showed that perisynaptic labeling was found at 93% of α4-labeled synapses (open bar) in CIV (n = 3). In CIE (n = 3), perisynaptic labeling was observed at 22% (open bar) of labeled synapses, but synaptic labeling was evident at 78% of labeled synapses (black bar). * p < 0.001 vs. CIV. In contrast to the α4 labeling, δ subunit labeling (arrow) in CIE was present at perisynaptic locations but not within the synaptic contact (arrowhead). Figs. a, c, and d are reproduced from Liang et al. [81] with permission. Figs. b are redrawn from Cagetti et al. [131]

Fig. 3

EtOH-induced plasticity of GABA_AR subunits and currents in rat after single-dose EtOH, CIE, and two-pulse EtOH. a: Summary of changes in mIPSCs, and b: inhibitory tonic currents after single-dose EtOH vs. pre-EtOH application (redrawn from Liang et al. [65]). A single dose EtOH induces loss of EtOH-sensitive tonic current and gain of EtOH-sensitive mIPSCs. Mean ± SEM are shown as % of vehicle-treated controls (red dashed line, n = 4–6. * p < 0.05). c: Biochemical analysis of GABA_AR subunit plasticity in rat DG within 24 h after single-dose EtOH compared with the changes induced by CIE, 40-d withdrawal. Surface protein levels of GABA_AR subunits measured using protein cross-linking and Western blotting. Mean ± SEM as % of vehicle-treated controls (red dashed line, n = 4–6. * p < 0.05). The α2 and γ1 subunits cell surface expression are up-regulated by both one-dose EtOH and CIE, γ1 total peptide is up-regulated, but not α2; and the heteropentameric subunit partnerships up-regulated are α4βγ2 and α2β1γ1. d, Upper panel: The protocol of double-dose EtOH experiment. d, Lower panel: Averaged mIPSC from each time point response to EtOH applications during the recordings. e: Summary of acute EtOH-induced changes in tonic current and mIPSCs (n = 5). f: Quantification of surface levels of GABA_AR (n = 4–6) by Western blots for GABA_AR α4 and γ1 after cross-linking in slices. g: Anxiety assayed by EPM (n = 6). The duration time rats stayed in arms (% of total 5 min). e,f,g: all bars are compared to the control (E0 value for that parameter): * p , 0.05; † p < 0.05. In e, the control level (dashed red line, at 100%) applies only to mIPSCs; in f, the red line refers to control (100%) for both subunits; in g, the dashed red line corresponds to the E0 point for either open or closed arms. c,d,e,f,g: from Lindemeyer et al., [30] with permission

Fig. 4

Hippocampal cells mIPSC kinetics patterns for GABA_AR subtypes in CIE rats and α4KO mice. A: mIPSC sample traces of CIE- vs. CIV-treated rats and α4KO and WT mice in hippocampal DG cells. B: Averaged mIPSC shape patterns detected by DataView revealed 3–4 relatively abundant distinct templates. In CIV, mIPSC patterns ‘a’, ‘c’, and ‘d’ were detected. Pattern ‘a’ is a standard shape, typical rise and decay kinetics; patterns ‘c’ and ‘c’ are slow rise-slow decay patterns correlated in abundance (not shown here) with α2 subunit subtypes. Three patterns of mIPSCs were also detected in CIE, but the ‘a’ pattern was not seen in CIE, and replaced by the slower decay pattern ‘b’. See text for interpretation that ‘a’ is mainly α1 and ‘b’ is mainly α4 subunit subtypes (as in Liang et al., 2006). Patterns of mIPSCs in WT and α4KO mice are similar to CIV rats, with peaks ‘a’, ‘c’, and ‘d’. However, the abundance of pattern ‘d’ was increased in CIE relative to CIV and in the α4ko mouse relative to WT. Since the CIE but not CIV, and a4KO mouse but not WT exhibited EtOH-enhanced mIPSCs, we examined recordings of these four animal groups with 50 mM EtOH (E50, dashed line) compared to without EtOH (E0, solid line) in the recording chamber. Peak pattern ‘a’ was not significantly enhanced by EtOH, but ‘b’, ‘c’, and ‘d’ were enhanced. Peak ‘b’ in CIE correlates with up-regulated α4, and is not seen in the α4KO mouse. Peak ‘d’ is up-regulated in both CIE rat and α4KO mouse, as is the α2 subunit surface expression, and peak ‘d’ has slow kinetics consistent with the α2 subunit subtypes. Its increase in abundance correlates with the increased average stimulation by EtOH in the recording chamber for both CIE and α4KO. Reproduced from Lindemeyer et al. [30] with permission

Fig. 5

A Reasonable Hypothesis of GABA_AR Subunit Plasticity Induced Within Two Days by One Dose of EtOH. Administration of EtOH to rats leads to changes of physio-pharmacological properties in GABAergic ionotropic receptor-mediated inhibitory synaptic transmission in hippocampus. The text at right of figure explains the time course of EtOH-induced plasticity, and how these same changes become persistent after CIE treatment. Reproduced from Lindemeyer et al. [30] with permission

Fig. 6

Two-Bottle Choice Assessment of EtOH Drinking by GABA_AR Wild Type and α2KO and α4KO Mice. a. EtOH preference assayed by voluntary access to EtOH (15%) in the 2 BC. (Δ, WT [C57/BL/6]; Ο, α4KO (G Homanics); and ם, α2KO (U Rudolph), n = 6–8). After the 3rd week, the EtOH intake in the α4KO group became significantly higher than that in the WT group. In contrast, the α2KO group did not show EtOH preference. b. Anxiety assay after 3 weeks of 2 BC measured by EPM (n = 6 ~ 8). The α4KO EPM data show reduced anxiety relative to wild type, while the α2KO exhibit more anxiety