Bidirectional relationship between alcohol intake and sensitivity to social defeat: association with Tacr1 and Avp expression

Abstract

While epidemiological studies show that alcohol abuse is often co-morbid with affective disorders, the causal direction of this association is unclear. We examined this relationship using mouse models including social defeat stress (SDS), social interaction (SI) and voluntary alcohol consumption. C57BL6/J mice exposed to SDS segregate into two subpopulations, those that express depressive-like phenotypes ('susceptible') and those that do not ('resilient'). First, we stratified SDS-exposed mice and measured their voluntary alcohol consumption. Next, we determined whether SI behavior in alcohol-naïve mice could predict alcohol intake. Finally, we assessed the effect of binge-like alcohol exposure on sensitivity to SDS. We quantified Tacr1 (neurokinin-1 receptor gene) and Avp (vasopressin peptide gene) mRNA in brain regions involved in depression, addiction and social behavior. We found that susceptible mice consumed more alcohol compared with resilient mice, suggesting that depression-like phenotypes associate with increased alcohol intake. Interestingly, we observed a negative correlation between SI and alcohol intake in stress- and alcohol-naïve mice, suggesting that individual differences in SI associate with alcohol preference. Finally, alcohol pre-treatment increased sensitivity to SDS, indicating that alcohol exposure alters sensitivity to social stress. Quantification of mRNA revealed that increased expression of Tacr1 and Avp generally associated with decreased SI and increased alcohol intake. C57BL6/J mice are an inbred strain; thus, it is likely that individual differences in behavior and gene expression are driven by epigenetic factors. Collectively, these results support a bidirectional relationship between alcohol exposure and susceptibility to stress that is associated with variations in neuropeptide expression.

Keywords: Alcohol; amygdala; neurokinin-1 receptor; nucleus accumbens; stress; vasopressin.

Figure 1. Experimental timeline and diagram for the Social Interaction Task

(A) Timeline for the first experiment examining the effects of social defeat on social interaction and drinking behaviors (B) Timeline for the second experiment examining how social interaction in stress-naïve mice affects drinking behaviors (C) Timeline for the third experiment examining the effects of prior-exposure to a binge drinking paradigm on susceptibility to a sub-threshold level of social defeat (D) Diagram depicting the protocol of the social interaction task used in all experiments.

Figure 2. SI, corner time, and alcohol consumption in SDS-exposed mice

(A) One way ANOVA revealed a main effect for SI score (F_(2,54)=9.1,p=0.007). Susceptible mice had lower SI scores compared to all other groups (p<0.05). Dotted line indicates threshold criteria for susceptibility. (B) Average time spent (s) in the corner during pretest and test phases (±SEM). There was a significant main effect of group (F_(1,54) = 4.2, P = .03). Posthoc Bonferroni tests were performed. (p<0.05 vs. all groups). (C) Average alcohol consumption (g/kg; ±SEM) over the last three days of two-bottle choice after animals had met criteria of baseline. One way ANOVA revealed a main effect for alcohol consumption after SDS exposure (F_(2,54)=9.1, p=0.0004). Susceptible mice consumed more alcohol than all other groups (p<0.05). N=7–20/group.

Figure 3. Baseline alcohol consumption correlates with post-SDS SI score

Control mice are represented as red individual data points (N = 24), susceptible mice are indicated by individual black data points (N=16), and resilient mice are indicated by grey individual data points (N=15). The x axis represents the average of the last three days of pre-defeat ethanol consumption when mice had met baseline criteria. The dotted line indicates the threshold used to separate stressed mice into resilient and susceptible subpopulations.

Figure 4. Initial SI score in alcohol naïve mice correlates with subsequent alcohol consumption

Social interaction tested at 8–10 weeks of age and alcohol access in two bottle choice paradigm began 2 weeks later. (N=19)

Figure 5. Tacr1 (A,B) and Avp (C,D) expression are negatively correlated with SI Ratio and positively correlated to alcohol consumption in the AMG

Alcohol naïve mice were prescreened at 8–10 weeks of age and then allowed access to alcohol 2 weeks later. DCT: delta cycle threshold; relative difference in cycle threshold compared to GAPDH expression (housekeeping gene). Note: lower DCT value indicates greater expression level. N=16.

Figure 6. Susceptible mice have increased Avp expression in the AMG and increased Tacr1 expression in the NAC compared to resilient mice

(A) T-test revealed that susceptible mice had increased expression of Avp in the AMG (t=2.914, p=0.0165) (B) T-test revealed that susceptible mice had increased expression of Tacr1 in the NAC (t=2.953, p=0.0105). DCT: delta cycle threshold; relative difference in cycle threshold compared to housekeeping gene. Note: lower DCT value indicates greater expression level. N=7–8/group.

Figure 7. Effect of chronic alcohol exposure on behavioral response to subthreshold social defeat

Mice were treated with 5 g/kg alcohol (20% in water, intragastric gavage) once per day for 10 days. T-test revealed that this pretreatment increased the number of mice that were classified as susceptible and induced lower SI scores compared to water treated controls (t=2.164, p=0.0374). N=17–20/group.