Punishment-resistant alcohol intake is mediated by the nucleus accumbens shell in female rats

Abstract

Alcohol use is widespread across many societies. While most people can control their alcohol use, a vulnerable sub-population develops alcohol use disorder, characterized by continued alcohol use despite negative consequences. We used a rat model of alcohol self-administration despite negative consequences to identify brain activity associated with this addiction-like behaviour. We and others have previously shown that response-contingent punishment of alcohol self-administration with mild footshock reliably identifies two sub-populations. One group substantially decreases alcohol self-administration in the face of punishment (punishment-sensitive, controlled) and another group continues alcohol self-administration despite negative consequences (punishment-resistant, addiction-like behaviour). In this study, we aimed to validate this model in females and identify associated brain regions. We trained Long-Evans outbred rats (n = 96) to self-administer 20% ethanol, and then introduced response-contingent footshock. We found that female rats consumed more alcohol in unpunished and punished sessions compared to male rats. In one group of rats (n = 24, m/f), we identified neuronal activity associated with punishment-resistant alcohol self-administration using the neurobiological marker of activity cFos. We found lower cFos expression in NAcSh associated with punishment-resistant alcohol self-administration. In another group of rats (n = 72, m/f), we used chemogenetic inhibition of NAcSh during punished alcohol self-administration. We found that chemogenetic NAcSh inhibition had no effect on unpunished alcohol self-administration but selectively increased punished alcohol self-administration in punishment-resistant female rats. These results indicate that more female rats develop punishment-resistant alcohol consumption, and that NAcSh hypofunction may underlie this phenotype.

Fig. 1. Behavioural characterization & sex differences in punished alcohol self-administration.

A Behavioural procedure. B Operant chamber setup. C Punishment procedure. D Suppression Ratio (SR = Punished responding / (Baseline + Punished responding)). Average SR at 0.25 mA was used to create a Punishment-Resistance Score which we used to classify rats as punishment-resistant (high SR, orange) or punishment-sensitive (low SR, green). E–H Alcohol intake across punished alcohol self-administration phases in males (n = 27, purple squares) and females (n = 36, pink triangles). *Difference between Sexes, p < 0.05. E Alcohol intake (g/kg) during home-cage alcohol access sessions. F Final average total active lever presses (left) and alcohol intake (g/kg) (right) during 30-min unpunished alcohol self-administration sessions. G Total active lever presses (left) and breakpoint (right) during progressive ratio test. H SR across punished alcohol self-administration sessions. I (Left) Density histograms of average 0.25 mA Suppression Ratios (SR) in males (purple) and females (pink). Highest 33% SR are punishment-resistant (n = 22, orange), lowest 33% are punishment-sensitive (n = 20, green). The punishment-resistant group was 73% female (pink), and the punishment-sensitive group was 65% male (purple). J–M Punished alcohol-self administration data classified as punishment-resistant versus punishment-sensitive (Males = squares, females = triangles). *Difference between SR categories, p < 0.05. J Progressive ratio test. K Total active lever presses for final FR3 average (Baseline) and every punished session. L Alcohol intake (g/kg) from unpunished Baseline and every punishment session. M Average footshocks received at 0.20 mA, 0.25 mA and 0.30 mA.

Fig. 2. Neuronal activity associated with punished alcohol self-administration.

A Behavioural Procedure. The test consisted of unpunished (n = 6 (3 m, 3 f), white), punishment-sensitive (n = 9 (7 m, 2 f), green) and punishment-resistant groups (n = 9 (2 m, 7 f), orange). B Alcohol self-administration during test session (male = square, female = triangle). Suppression ratio (SR) and footshocks are also shown for punishment-sensitive (green) and punishment-resistant (orange) groups. *Difference between Test Groups, p < 0.05. m, Male; f, Female. C Representative illustration of analyzed brain regions (top). Representative images of cFos-positive cells in a sub-region of the analysed NAcSh region of interest (bottom). D Significant correlations between total active lever presses and cFos (male = square, female = triangle). E Significant correlations between SR and cFos in punished alcohol self-administration (Male = square, Female = triangle). F cFos-positive cells/mm² (mean ± SEM) in analyzed brain regions by Test Group. *Difference between Test Groups, p < 0.05. OFC orbitofrontal cortex, PrL prelimbic cortex, IL infralimbic cortex, aIns anterior insula, NAcC nucleus accumbens core, NAcSh nucleus accumbens shell.

Fig. 3. Chemogenetic inhibition of NAcSh during punished alcohol self-administration.

A Experimental timeline. Effect of NAcSh inhibition on alcohol intake (g/kg) in (B) males (mCherry: n = 10, hM4Di: n = 11) and females (C, E) (mCherry: n = 14, hM4Di: n = 11) during alcohol self-administration, progressive ratio test, punished alcohol self-administration at 0.25 mA and punished alcohol self-administration at 0.30 mA. Data expressed as Punishment-Resistance Score (Punishment-sensitive = green, punishment-resistant = orange) medians and individual datapoints. Individual datapoints are coloured using the continuous variable of Punishment-Resistance Score (Low = green, mid = grey, high = orange) and sex is indicated by squares (males) and triangles (females). Separate graphs in each panel show mCherry (left) or hM4Di rats (right) and each graph compares the repeated measure factor Ligand (Saline (SAL) vs. Deschloroclozapine (DCZ)). *Virus-hM4Di x Ligand-DCZ x Punishment-Resistance Score interaction effect, p < 0.05. D Bilateral expression of hM4Di in NAcSh. E Significant model predictions from synthetic data to illustrate significant Virus-hM4Di x Ligand-DCZ x Punishment-Resistance Score interaction effect in females at 0.30 mA, p < 0.05.