The glucagon-like peptide-1 receptor as a potential treatment target in alcohol use disorder: evidence from human genetic association studies and a mouse model of alcohol dependence

Abstract

The hormone glucagon-like peptide-1 (GLP-1) regulates appetite and food intake. GLP-1 receptor (GLP-1R) activation also attenuates the reinforcing properties of alcohol in rodents. The present translational study is based on four human genetic association studies and one preclinical study providing data that support the hypothesis that GLP-1R may have a role in the pathophysiology of alcohol use disorder (AUD). Case-control analysis (N = 908) was performed on a sample of individuals enrolled in the National Institute on Alcohol Abuse and Alcoholism (NIAAA) intramural research program. The Study of Addiction: Genetics and Environment (SAGE) sample (N = 3803) was used for confirmation purposes. Post hoc analyses were carried out on data from a human laboratory study of intravenous alcohol self-administration (IV-ASA; N = 81) in social drinkers and from a functional magnetic resonance imaging study in alcohol-dependent individuals (N = 22) subjected to a Monetary Incentive Delay task. In the preclinical study, a GLP-1R agonist was evaluated in a mouse model of alcohol dependence to demonstrate the role of GLP-1R for alcohol consumption. The previously reported functional allele 168Ser (rs6923761) was nominally associated with AUD (P = 0.004) in the NIAAA sample, which was partially replicated in males of the SAGE sample (P = 0.033). The 168 Ser/Ser genotype was further associated with increased alcohol administration and breath alcohol measures in the IV-ASA experiment and with higher BOLD response in the right globus pallidus when receiving notification of outcome for high monetary reward. Finally, GLP-1R agonism significantly reduced alcohol consumption in a mouse model of alcohol dependence. These convergent findings suggest that the GLP-1R may be an attractive target for personalized pharmacotherapy treatment of AUD.

Figures

Figure 1

Comparison of risk allele frequencies between non-smoking controls, non-smoking subjects with alcohol use disorder (AUD) and smoking subjects with AUD in the LCTS study using Pearson χ2 test. Total n for the three groups were 125–130, 121–128, 202–215 for the Caucasian sample and 26–29, 47–50, 137–147 for the African American sample, respectively. *P<0.05, **P<0.01, ***P<0.005. LCTS, Laboratory of Clinical and Translational Studies.

Figure 2

Statistical maps for fMRI analysis contrasting the genotype groups 168Gly/Gly × 168Ser/Gly+168Ser/Ser in the rs6923761 SNP. Statistical maps (top) and ROI results (bottom) for notification of high reward (left) and its net difference with low reward (right) contrasting (168Gly/Gly)−(168Ser/Gly+168Ser/Ser) are shown; ROI localization is displayed in yellow at the axial view in the bottom. fMRI, functional magnetic resonance imaging; SNP, single-nucleotide polymorphism.

Figure 3

GLP-1R agonism in a mouse model of alcohol dependence. (a) AC3174 treatment #1 (alcohol intake during test cycle 5): EtOH mice consumed significantly more alcohol than CTL mice (P< 0.001). Analysis of variance (ANOVA) failed to indicate a main effect of treatment or group × treatment interaction. (b) AC3174 treatment #2 (alcohol intake during test cycle 6): there was a significant main effect of group (P<0.001), with EtOH mice consuming a greater amount of alcohol than CTL mice. ANOVA failed to indicate a main effect of treatment or group × treatment interaction. (c) AC3174 treatment #3 (alcohol intake during test cycle 7): ANOVA indicated a significant main effect of group (P<0.00001) and a significant group × treatment interaction (P<0.025). Post hoc comparisons indicated that, as expected, EtOH mice injected with vehicle consumed more alcohol than CTL mice. In addition, all doses of AC3174 significantly reduced drinking compared with the vehicle condition in EtOH mice, while AC3174 treatment did not significantly alter alcohol intake in nondependent CTL mice. Further, AC3174 treatment abolished the difference in alcohol intake between EtOH and CTL conditions. (d) Placebo (washout) test #1 (alcohol intake during test cycle 8): all mice were treated with saline (drug-washout test) to substantiate the apparent efficacy of AC3174 to reduce escalated alcohol drinking in dependent mice. ANOVA revealed a significant main effect of group (P<0.00001) and a significant group × treatment interaction (P<0.01). Post hoc comparisons supported the expected greater alcohol intake in EtOH compared with CTL mice that continued to receive vehicle. A similar profile of results was obtained in mice that received the lowest AC3174 dose (0.03 μg kg−1) in the previous test cycle. However, EtOH mice that received 0.10 or 0.30 μg kg−1 AC3174 doses in the previous test period continued to consume significantly less alcohol compared with mice that previously received vehicle, and their lower level of intake was similar to that exhibited by the corresponding CTL groups. (e) Placebo (washout) test #2 (alcohol intake during test cycle 9): ANOVA indicated a significant main effect of group (P<0.001), but no effect of treatment or an interaction between group × treatment during this second washout test period. These results indicate that after a second week of placebo (saline) treatment, elevated drinking in EtOH compared with CTL mice was restored in all the test groups. *P<0.05, significantly different from corresponding CTL group; ^P<0.05, significantly different from corresponding vehicle group.