Inactivation of the lateral orbitofrontal cortex increases drinking in ethanol-dependent but not non-dependent mice

Abstract

Long-term consumption of ethanol affects cortical areas that are important for learning and memory, cognition, and decision-making. Deficits in cortical function may contribute to alcohol-abuse disorders by impeding an individual's ability to control drinking. Previous studies from this laboratory show that acute ethanol reduces activity of lateral orbitofrontal cortex (LOFC) neurons while chronic exposure impairs LOFC-dependent reversal learning and induces changes in LOFC excitability. Despite these findings, the role of LOFC neurons in ethanol consumption is unknown. To address this issue, we examined ethanol drinking in adult C57Bl/6J mice that received an excitotoxic lesion or viral injection of the inhibitory DREADD (designer receptor exclusively activated by designer drug) into the LOFC. No differences in ethanol consumption were observed between sham and lesioned mice during access to increasing concentrations of ethanol (3-40%) every other day for 7 weeks. Adulterating the ethanol solution with saccharin (0.2%) or quinine (0.06 mM) enhanced or inhibited, respectively, consumption of the 40% ethanol solution similarly in both groups. Using a chronic intermittent ethanol (CIE) vapor exposure model that produces dependence, we found no difference in baseline drinking between sham and lesioned mice prior to vapor treatments. CIE enhanced drinking in both groups as compared to air-treated animals and CIE treated lesioned mice showed an additional increase in ethanol drinking as compared to CIE sham controls. This effect persisted during the first week when quinine was added to the ethanol solution but consumption decreased to control levels in CIE lesioned mice in the following 2 weeks. In viral injected mice, baseline drinking was not altered by expression of the inhibitory DREADD receptor and repeated cycles of CIE exposure enhanced drinking in DREADD and virus control groups. Consistent with the lesion study, treatment with clozapine-N-oxide (CNO) further enhanced consumption only in CIE exposed DREADD mice with no change in air-treated mice. These results suggest that the LOFC is not critical for the initiation and maintenance of ethanol drinking in non-dependent mice, but may regulate the escalated drinking observed during dependence.

Keywords: Alcohol; DREADD; Excitotoxic lesion; Frontal cortex.

Figure 1

Effects of an excitotoxic lesion of the LOFC on ethanol consumption by mice on the intermittent access paradigm. (A) Schematic showing anatomical location of LOFC (left panel, shaded area, figure from Allen Brain Atlas) and examples of GFAP immunoreactivity from saline and NMDA injected animals. Positive GFAP staining is white. Right panel are atlas figures with shading indicating average extent of lesion (numbers are distance (mm) from bregma). (B) No effect of lesion on ethanol consumption (mean ± SEM, g/kg) during intermittent access to ethanol containing solutions. Animals were given 24 h access to increasing concentrations of ethanol every other day with water available every day. (C) No effect of lesion on ethanol preference for different concentrations of ethanol. (D) Sham and lesioned mice consumed more ethanol during access to the 40% ethanol solution as compared to the 20% ethanol solution (*p<0.05; ****p<0.0001). (E) Preference for the 40% ethanol solution was reduced as compared to that of the 20% solution (****p<0.0001). Number of animals per group: Sham (8); Lesion (11).

Figure 2

Effect of an excitotoxic lesion to the LOFC on the ability of sweet and bitter tastants to alter ethanol drinking. (A) No effect of lesion on consumption of 40% ethanol adulterated with either saccharin (0.2%) or quinine (0.06 mM). (B) No effect of lesion on preference for 40% ethanol adulterated with either saccharin (0.2%) or quinine (0.06 mM). (C) Sham and lesioned mice consumed greater amounts of 40% ethanol in the presence of saccharine as compared to ethanol alone or ethanol plus quinine (oneway Anova, **; p<0.01; ***p<0.001). (D) Sham and lesioned mice showed greater preference for 40% ethanol in the presence of saccharine as compared to ethanol alone or ethanol plus quinine (oneway Anova, **; p<0.01; ***p<0.001). Number of animals per group: Sham (8); Lesion (11).

Figure 3

Excitotoxic lesion to the LOFC enhances ethanol consumption in dependent but not non-dependent mice. Groups of mice were exposed to repeated cycles of chronic intermittent exposure (CIE) to ethanol vapor or air followed by weekly drinking sessions. (A) CIE exposure enhances ethanol drinking as compared to air-exposed mice (###; main effect of CIE on ethanol consumption, mixed Anova, p<0.0001). Lesioned mice exposed to two or more cycles of CIE exposure (Tests 2–4) consume more ethanol than non-lesioned CIE treated mice (*p<0.05; **p<0.01). Data represent mean (± SEM) daily ethanol consumption (g/kg) for each group during baseline, after each CIE exposure (Test 1–4) and during post-CIE periods when quinine was added to the ethanol solution. (B) No effect of LOFC lesion on water consumption during baseline drinking or following CIE exposures. Symbol (###, main effect of CIE on water consumption, mixed Anova, p<0.01). (C) No effect of LOFC lesion on ethanol preference during baseline drinking or following CIE exposures. Symbol (###, main effect of CIE on ethanol preference, mixed Anova, p<0.0001). (D) Lesioned CIE treated mice show reduced sensitivity to quinine adulteration of ethanol solution. Data represent mean (± SEM) percent inhibition of drinking by quinine during Test 4 relative to Test 3. Ethanol consumption data from panel A. Symbols (**, p<0.01, ***p<0.001; ****p<0.0001; paired t-test). (E) Atlas figures with shading indicate average extent of lesioned area (numbers indicate distance (mm) from bregma). Number of animals per group: Sham Air (9), Lesion Air (9), Sham CIE (9), Lesion CIE (10).

Figure 4

Effect of expression of the inhibitory hM4Di DREADD in LOFC on spike firing and ethanol consumption. (A) Representative traces showing action potentials from an LOFC neuron during sequential current injections before (baseline) and during application of CNO (5 μM). (B) Anatomical location of injection site for AAV virus in LOFC (shaded area in left panel, figure from Allen Brain Atlas) and fluorescent/bright field images (right panel) of hM4Di-mCherry expression two weeks after viral injection. Atlas figures with shading indicate average extent of lesioned area (numbers indicate distance (mm) from bregma). (C) Effect of CNO on action potential number. Columns show number of spikes during injection of current during the last 3 minutes of CNO exposure expressed as a percent of the baseline (mean ± SEM). Symbol (*): value significantly different from control (N=7, p<0.001, one-sample t-test). (D) Effect of the inhibitory hM4Di DREADD on ethanol consumption in CIE exposed mice. Following establishment of baseline drinking (15% ethanol v/v; 3 hr limited access), groups of virus injected mice (AAV; DREADD) were exposed to repeated cycles of air or ethanol vapor (CIE) interleaved with weekly assessments of drinking (Test 1–5). During Tests 3 and 4, all mice received an injection of CNO (10 mg/kg) 30 min prior to the introduction of the ethanol bottle. Symbols (###: main effect of CIE treatment on ethanol consumption, Mixed Anova, p<0.0001; *: Test 4; DREADD-CIE consumption significantly greater than AAV-CIE group, Mixed Anova, p<0.01), Test 5; DREADD-CIE consumption significantly greater than AAV-CIE group, Mixed Anova, p<0.05). Number of animals per group: AAV Air (8), DREADD Air (6), AAV CIE (9), DREADD CIE (10).