Nicotinic acetylcholine receptors containing the α4 subunit modulate alcohol reward

Abstract

Background: Nicotine and alcohol are the two most co-abused drugs in the world, suggesting a common mechanism of action might underlie their rewarding properties. Although nicotine elicits reward by activating ventral tegmental area dopaminergic (DAergic) neurons via high-affinity neuronal nicotinic acetylcholine receptors (nAChRs), the mechanism by which alcohol activates these neurons is unclear.

Methods: Because most high-affinity nAChRs expressed in ventral tegmental area DAergic neurons contain the α4 subunit, we measured ethanol-induced activation of DAergic neurons in midbrain slices from two complementary mouse models, an α4 knock-out (KO) mouse line and a knock-in line (Leu9'Ala) expressing α4 subunit-containing nAChRs hypersensitive to agonist compared with wild-type (WT). Activation of DAergic neurons by ethanol was analyzed with both biophysical and immunohistochemical approaches in midbrain slices. The ability of alcohol to condition a place preference in each mouse model was also measured.

Results: At intoxicating concentrations, ethanol activation of DAergic neurons was significantly reduced in α4 KO mice compared with WT. Conversely, in Leu9'Ala mice, DAergic neurons were activated by low ethanol concentrations that did not increase activity of WT neurons. In addition, alcohol potentiated the response to ACh in DAergic neurons, an effect reduced in α4 KO mice. Rewarding alcohol doses failed to condition a place preference in α4 KO mice, paralleling alcohol effects on DAergic neuron activity, whereas a sub-rewarding alcohol dose was sufficient to condition a place preference in Leu9'Ala mice.

Conclusions: Together, these data indicate that nAChRs containing the α4 subunit modulate alcohol reward.

Figure 1

Ethanol activation of VTA DAergic neurons. A) Representative action potential firing frequency histogram from a VTA DAergic neuron before, during, and after 5-min bath application of 100 mM ethanol (EtOH) alone (n = 10) or in the presence of B) 10 μM mecamylamine (MEC, n = 7) or C) 100 nM MLA (n = 5). Action potentials were recorded in cell-attached mode. Representative action potential traces (top of each panel, a, b, c) are shown from the corresponding times on the histograms. D. Average time course of ethanol responses in each condition is shown (top panel). Each data point represents the average 1-min firing frequency normalized to baseline for each recording. The bar over the averaged frequency plot represents the duration of ethanol application. Recording times between groups were aligned based on time of ethanol application to facilitate comparison. Antagonists were applied at times indicated in the individual histograms as indicated in B and C. (Bottom panel) Fold-change in average firing frequency at baseline (1 min. prior to alcohol application, dotted line) compared to 5 min of ethanol application for each condition. ^### p < 0.001, ^#### p < 0.0001, baseline frequency (1 min prior to drug application) compared to 5 min alcohol exposure, paired t-tests. ** p < 0.01, Students t-test (effect of ethanol or ethanol + MLA on AP frequency compared to ethanol + MEC). n = 6-10 neurons/condition.

Figure 2

Functional α4* nAChR expression modulates DAergic neuron activation by ethanol. Representative action potential firing frequency histogram from a VTA DAergic neuron before, during, and after a 5-min bath application of 100 mM ethanol in sagittal midbrain slices from A) α4 KO and B) Leu9’Ala mice. Representative action potential traces (top of each panel, a, b, c) are shown from the corresponding times on the histograms. C) Time course of the effects of ethanol on average normalized frequency for each genotype are shown (n = 8-10 neurons/genotype). Ethanol (100 mM) was applied at the times indicated by the bar. D. Fold-change in average firing frequency in response to 100 mM ethanol in WT (n = 8) and α4 KO mice (n = 15), 20 mM in WT (n = 5) and Leu9’Ala mice (n = 8), and 100 mM ethanol in WT (n = 10) and Leu9’Ala mice (n = 13). ^#p < 0.05, ^##p < 0.01, ^###p < 0.001, ^####p < 0.0001, as in 2D. * p < 0.05, ** p < 0.01 response to alcohol compared between genotypes, One-way ANOVA, Bonferroni post-test.

Figure 3

Alcohol potentiates the response of DAergic neurons to ACh. A) Representative action potential firing frequency histogram from a VTA DAergic neuron before, during, and after 10-min bath application of 300 μM ACh. Representative action potential traces (top of each panel, a, b, c) are shown from the corresponding times on the histograms. Representative action potential firing frequency histogram from a VTA DAergic neuron before, during, and after 10-min bath co-application of 300 μM ACh and 50 mM ethanol in the absence (B), or presence (C) of 1 μM DHβE. D) Representative action potential firing frequency histogram from an α4 KO VTA DAergic neuron before, during, and after 10 min bath co-application of 300 μM ACh and 50 mM ethanol. E) Time course of the effects of ethanol on average normalized frequency under each condition are shown (n = 6-12 neurons/genotype). ACh ± ethanol was applied at the times indicated by the bar. Recording times between groups were aligned based on time of ethanol application to facilitate comparison. F) Fold-change in average DAergic neuron firing frequency in response to 300 μM ACh alone (n = 6), in the presence of 50 (n = 12) or 100 mM (n = 5) ethanol, in the presence of 50 mM ethanol and DHβE (n = 6), or in the presence of 50 mM ethanol in α4 KO slices (n = 10). ^#p < 0.05 as in 1D. * p < 0.05 response to alcohol compared between treatments/genotypes.

Figure 4

Ethanol-induced c-Fos expression in VTA TH(+) neurons is dependent on expression and activation of α4* nAChRs. Representative photomicrographs illustrating midbrain sections of the posterior VTA from A) WT mice and B) α4 KO mice injected with 2 g/kg ethanol. Sections were immunolabeled for TH (red) and c-Fos (green). White boxes delineate slice regions that are magnified in the adjacent photomicrographs. White arrowheads point to neurons that are TH (+), c-Fos (+). Scale bar = 100 μm. Merged images are shown. C) Number of TH (+) c-Fos (+) neurons per slice taken from mice given an i.p. injection of 2 g/kg ethanol. Forty-eight slices/treatment/mouse were analyzed, n = 3 mice/treatment. D. Representative photomicrographs illustrating midbrain sections of the posterior VTA from WT mice and E) Leu9’Ala mice injected with 0.5 g/kg ethanol. Sections were immunolabeled for TH and c-Fos as in panels A and B. F) Average number of TH (+), c-Fos (+) neurons/slice calculated from mice given an i.p. injection of 0.5 g/kg or 2 g/kg ethanol. Forty-eight slices/treatment/mouse were analyzed, n = 3 mice/treatment. One-way ANOVA and Bonferroni post-test comparing saline to ethanol treatments in WT, α4 KO, or Leu9’Ala mice was used, ^##p<0.01, ^###p<0.001. Two-way ANOVA and Bonferroni post-test comparing treatments in WT and α4 KO mice was also used, ** p < 0.01, ***p<0.001.

Figure 5

α4* nAChR expression modulates alcohol reward. A. Average difference score (test – baseline) in ethanol-paired (black bars) and saline-paired (white bars) chambers in Leu9’Ala, WT, and α4 KO mice in response to the alcohol doses indicated. Because each line has been back-crossed at least ten generations to the C57Bl/6J strain and no differences in alcohol responses between α4 KO and Leu9’Ala WT littermates were detected, WT mice were combined. n = 9 – 22 mice/dose. * p < 0.05, ** p < 0.01, *** p < 0.001 compared to saline. B. Locomotor activity (ambulation) in WT, Leu9’Ala, and α4 KO mice. Each data point represents the average total locomotor activity over 5-min (n=5-10 mice/genotype). C. Blood ethanol concentration at 30-min intervals after an acute, 2.0 g/kg i.p. injection of alcohol in WT, α4 KO, and Leu9’ala mice (n = 3-4 mice/genotype).