Dopamine D2 receptors in the bed nucleus of the stria terminalis modulate alcohol-related behaviors

Abstract

Dysregulation of the dopamine (DA) system is a hallmark of substance abuse disorders, including alcohol use disorder (AUD). Of the DA receptor subtypes, the DA D2 receptors (D2Rs) play a key role in the reinforcing effects of alcohol. D2Rs are expressed in numerous brain regions associated with the regulation of appetitive behaviors. One such region is the bed nucleus of the stria terminalis (BNST), which has been linked to the development and maintenance of AUD. Recently, we identified alcohol withdrawal-related neuroadaptations in the periaqueductal gray/dorsal raphe to BNST DA circuit in male mice. However, the role of D2R-expressing BNST neurons in voluntary alcohol consumption is not well characterized. In this study, we used a CRISPR-Cas9-based viral approach, to selectively reduce the expression of D2Rs in BNST VGAT neurons and interrogated the impact of BNST D2Rs in alcohol-related behaviors. In male mice, reduced D2R expression potentiated the stimulatory effects of alcohol and increased voluntary consumption of 20% w/v alcohol in a two-bottle choice intermittent access paradigm. This effect was not specific to alcohol, as D2R deletion also increased sucrose intake in male mice. Interestingly, cell-specific deletion of BNST D2Rs in female mice did not alter alcohol-related behaviors but lowered the threshold for mechanical pain sensitivity. Collectively, our findings suggest a role for postsynaptic BNST D2Rs in the modulation of sex-specific behavioral responses to alcohol and sucrose.

Keywords: D2 receptor; alcohol; bed nucleus of the stria terminalis (BNST); dopamine; pain.

Figure 1.. Validation of CRISPR-Cas9 deletion of BNST D2Rs.

(A) Top: Surgical schematic of bilateral virus infusion in the BNST of VGAT-cre male and female mice. Bottom: Detailed schematic of Drd2 (D2 KO) and Rosa26 (control) CRISPR construct provided by the Zweifel lab. (B) Representative images of in situ hybridization for D2 (cyan) and VGAT (green) mRNA from control (top) and D2 KO (bottom) in the mouse BNST. Scale bar=200 μm. (C) Representative images at higher magnification showing Drd2 (cyan) labeling inside neurons labeled with DAPI (blue) from control (left) and D2 KO (right) mice. Scale bar=50 μm. (D) Compared to controls, D2 KO mice showed a significant reduction in Drd2 mRNA expression in the dBNST. (E) CRISPR knockout resulted in ~30% reduction in Drd2+ mRNA expression. (F) Representative viral infections in BNST in males (left) and females (right), as indexed by GFP expression. Scale bar=500 μm. (G) Representative map of viral spread from +0.62 mm to −0.22 mm (from bregma). ac: anterior commissure; str: striatum. Data expressed as Mean ± SEM. *p<0.05.

Figure 2.. Experimental Timeline.

Schematics of the experimental timeline showing the sequence followed for behavioral assays in all cohorts of mice.

Figure 3.. Impact of BNST D2R deletion on anxiety-like behavior and pain sensitivity.

Anxiety-like behavior was assessed in an open field assay. No differences were observed either in the total distance traveled (cm) (A) or percentage time spent in the center of the open field arena (B) following BNST D2R knockout. Hargreaves test was used to assess thermal sensitivity and average paw withdrawal latencies were compared (C). In both sexes, deletion of D2-Rs had no impact on withdrawal latency. Von Frey test was used to assess the contribution of BNST D2Rs in mechanical sensitivity (D-F). Paw withdrawal threshold was defined as the minimum force (g) filament that elicits a withdrawal reflex for ≥ 50% of the trials. (D) No effect of D2 knockout on withdrawal threshold in male mice. (E) D2 KO female mice had a significantly lower withdrawal threshold compared to female controls. (F) There was a main effect of sex on mechanical pain threshold and a non-significant trend for a main effect of deletion. Post-hoc analysis revealed a significantly lower threshold in D2 KO females compared to D2 KO males. Data expressed as Mean ± SEM.*p<0.05; $ main effect of sex.

Figure 4.. BNST^VGAT D2R knockout does not affect acoustic startle response.

Average startle amplitudes (10 trials/dB) were plotted in response to increasing sound intensities. No effect of reduced D2R expression on startle amplitude in males (A) and females (B). (C) Comparison of startle response at 120 dB between male and female mice. Overall, female mice displayed lower amplitude compared to male mice. No differences were observed between D2 KO and control mice. Data expressed as Mean ± SEM.*p<0.05; @ main effect of decibel; $ main effect of sex.

Figure 5.. Selective deletion of BNST^VGAT D2Rs promotes ethanol-induced locomotion in male mice.

(A) Schematic diagram representing ethanol-induced locomotor assay. Mice were habituated to the locomotor boxes for two days followed by IP injection of saline and 2g/kg ethanol in a counter-balanced, crossover design. Total distance traveled (cm) was plotted in 5-minute bins for 30 minutes. 2g/kg ethanol injection resulted in a transient increase in locomotion in both sexes (B-C). Deletion of D2Rs altered the total distance traveled in a sex-dependent manner. (D-F) Average distance traveled in the first 5 minutes post-drug injection was compared across groups. In both sexes, ethanol IP resulted in increased locomotion compared to saline. Compared to control males, D2 KO males had a stronger stimulatory response to ethanol (D). There was no effect of knockout on ethanol-induced locomotion in female mice (E). (F) Female control mice traveled greater distances following ethanol injection compared to male control mice. Data expressed as Mean ± SEM. *p<0.05; @ main effect of time; % main effect of virus; & main effect of ethanol; $ main effect of sex; # ethanol x time interaction; ^ ethanol x sex interaction; ! sex x virus interaction; ethanol x sex x virus interaction p=0.058.

Figure 6.. Deletion of BNST D2Rs increases alcohol consumption in males.

(A) Schematic diagram representing two-bottle choice intermittent access to 20% w/v ethanol in mice for 6 weeks. Average weekly alcohol intake (g/kg/24h) in male (B) and female (C) mice for 6 consecutive weeks. Deletion of BNST^VGAT D2Rs increased alcohol intake in males but not females. (D) Averages of weekly intake compared between sexes. Female mice consumed more alcohol compared to male mice and there was a trend toward a sex x virus interaction (p=0.10). (E-F) Deletion of BNST^VGAT D2Rs did not affect ethanol preference in both sexes, there was a non-significant trend in increased ethanol preference in male mice (p=0.11). (G) Cumulative alcohol intake plotted over 6 consecutive weeks. D2 KO males had a higher cumulative intake than controls. (H) No effect of knockout on cumulative intake in females. Data expressed as Mean ± SEM.*p<0.05; @ main effect of time; % main effect of virus; $ main effect of sex; # virus x time interaction.

Figure 7.. Knockout of BNST D2Rs increases sucrose intake in a sex-dependent manner.

Average sucrose intake (ml/kg/2h) in male (A) and female (B) mice. Deletion of BNST^VGAT D2Rs increased sucrose intake selectively in male mice. (C) Averages of sucrose intake compared between sexes. Female controls consumed more sucrose compared to male controls. (D-F) Deletion of BNST^VGAT D2Rs also increased sucrose preference in males but not females. Female control mice had higher sucrose preference compared to male controls. There was no effect of virus in female mice. Data expressed as Mean ± SEM.*p<0.05; % main effect of virus; $ main effect of sex; # virus x sex interaction.