Receptor and metabolic insights on the ability of caffeine to prevent alcohol-induced stimulation of mesolimbic dopamine transmission

Abstract

The consumption of alcohol and caffeine affects the lives of billions of individuals worldwide. Although recent evidence indicates that caffeine impairs the reinforcing properties of alcohol, a characterization of its effects on alcohol-stimulated mesolimbic dopamine (DA) function was lacking. Acting as the pro-drug of salsolinol, alcohol excites DA neurons in the posterior ventral tegmental area (pVTA) and increases DA release in the nucleus accumbens shell (AcbSh). Here we show that caffeine, via antagonistic activity on A_2A adenosine receptors (A_2AR), prevents alcohol-dependent activation of mesolimbic DA function as assessed, in-vivo, by brain microdialysis of AcbSh DA and, in-vitro, by electrophysiological recordings of pVTA DA neuronal firing. Accordingly, while the A₁R antagonist DPCPX fails to prevent the effects of alcohol on DA function, both caffeine and the A_2AR antagonist SCH 58261 prevent alcohol-dependent pVTA generation of salsolinol and increase in AcbSh DA in-vivo, as well as alcohol-dependent excitation of pVTA DA neurons in-vitro. However, caffeine also prevents direct salsolinol- and morphine-stimulated DA function, suggesting that it can exert these inhibitory effects also independently from affecting alcohol-induced salsolinol formation or bioavailability. Finally, untargeted metabolomics of the pVTA showcases that caffeine antagonizes alcohol-mediated effects on molecules (e.g. phosphatidylcholines, fatty amides, carnitines) involved in lipid signaling and energy metabolism, which could represent an additional salsolinol-independent mechanism of caffeine in impairing alcohol-mediated stimulation of mesolimbic DA transmission. In conclusion, the outcomes of this study strengthen the potential of caffeine, as well as of A_2AR antagonists, for future development of preventive/therapeutic strategies for alcohol use disorder.

Fig. 1. Effects of caffeine, DPCPX, and SCH 58261 on alcohol-induced pVTA salsolinol formation and AcbSh DA increase, and effects of caffeine on salsolinol bioavailability and AcbSh DA increase during pVTA perfusion with salsolinol.

A Schematic representation of dual probe in vivo brain microdialysis procedures, sampled areas and neurotransmitters recorded. Effects of i.p. administration of Caf (3 mg/kg) (B), Caf (15 mg/kg) (C), DPCPX or SCH (D) and of pVTA perfusion with Caf (E) on pVTA SALS formation and AcbSh DA enhancement induced by i.g. EtOH, and (F) effects of i.p. administration of Caf (15 mg/kg) on pVTA SALS concentration and ipsilateral AcbSh DA transmission during pVTA perfusion with SALS. Horizontal bars depict the duration and content of the pVTA perfusion along the experiments. Vertical arrows indicate the last pVTA or AcbSh microdialysis sample before Veh, Caf, DPCPX or SCH and water or EtOH administrations. Filled symbols indicate samples representing p < 0.001 vs. basal; **p < 0.01 vs. Caf (3 mg/kg) + EtOH, vs. Caf (15 mg/kg) + EtOH, and vs. Caf (15 mg/kg) + SALS; *p < 0.05 vs. DPCPX + H₂O. Veh-H₂O (n = 4); Veh-EtOH (n = 6); Caf (3 mg/kg)-H₂O (n = 4); Caf (15 mg/kg)-H₂O (n = 4); Caf (3 mg/kg)-EtOH (n = 11); Caf (15 mg/kg)-EtOH (n = 12); DPCPX-H₂O (n = 3); SCH-H₂O (n = 3); DPCPX-EtOH (n = 5); SCH-EtOH (n = 6); Caf (10 μM)-H₂O (n = 3); Caf (10 μM)-EtOH (n = 8); Veh-SALS (n = 3); Caf (15 mg/kg)-SALS (n = 5). Veh Saline, Caf Caffeine, H₂O Water, EtOH Alcohol, SCH SCH 58261, SALS Salsolinol.

Fig. 2. Effects of alcohol, caffeine, salsolinol, and morphine on the firing rate of rat pVTA DA neurons.

A, B Representative traces of spontaneous firing recorded from single DA neurons before (baseline), during, and after (washout) bath application of 60 mM EtOH (A), 10 μM Caf (B), of 10 nM SALS (J), and of 1 μM Mor (N). Scale bar: 1 s. Graphs showing the effects of EtOH (C), Caf (D) and their combination (E), of the combination of 10 μM SCH (G) or 10 μM DPCPX (H) with 60 mM EtOH, of 10 nM SALS alone (K) and in association with 10 μM Caf (L), and of 1 μM Mor alone (O) and in association with 10 μM Caf (P) on the firing rate of DA neurons. Data are expressed as mean ± SEM. The bar graphs summarize the percentage of change from baseline produced by EtOH and Caf alone and by their combination (n = 33 neurons from 11 rats) (F), the effects of SCH and DPCPX on the stimulatory effect of EtOH (n = 36 neurons from 18 animals) (I), the effects of SALS and Caf when bath perfused alone or during their association (n = 22 neurons from 11 rats) (M), the effects of Mor alone and in combination with Caf (n = 22 neurons from 11 rats) (Q). One-way ANOVA, *p < 0.05 versus baseline; ^#p < 0.05 versus DPCPX alone. EtOH Alcohol, Caf Caffeine, SCH SCH58261, SALS Salsolinol, Mor Morphine.

Fig. 3. Effects of caffeine and alcohol on the biochemical profiles of rats pVTA.

A Unsupervised PCA of complete dataset highlighted pre-treatment effect on pVTA biochemical profiles (PERMANOVA, R² = 0.036 and p = 0.03). B Supervised PLS-DA models of complete dataset showed a stronger effect of pre-treatment over treatment. Classification error rate (CER) calculated with 5-fold cross validation and 999 permutations. VIP scores of pairwise PLS-DA models Saline-Water v Saline-Alcohol, Saline-Water v Caffeine-Alcohol and Saline-Alcohol v Caffeine-Alcohol (C) and Saline-Water v Caffeine-Water and Caffeine-Water v Caffeine-Alcohol (D) are plotted for molecules of interest. Stratified models performance and features with VIP > 1 are listed in Supplementary Tables 3–6. N = 9 per group. Veh Saline, Caf Caffeine, H₂O Water, EtOH Alcohol.