Histamine is required for H₃ receptor-mediated alcohol reward inhibition, but not for alcohol consumption or stimulation

Abstract

Background and purpose: Conflicting data have been published on whether histamine is inhibitory to the rewarding effects of abused drugs. The purpose of this study was to clarify the role of neuronal histamine and, in particular, H₃ receptors in alcohol dependence-related behaviours, which represent the addictive effects of alcohol.

Experimental approach: Alcohol-induced conditioned place preference (alcohol-CPP) was used to measure alcohol reward. Alcohol-induced locomotor stimulation, alcohol consumption and kinetics were also assessed. mRNA levels were quantified using radioactive in situ hybridization.

Key results: Low doses of H₃ receptor antagonists, JNJ-10181457 and JNJ-39220675, inhibited alcohol reward in wild-type (WT) mice. However, these H₃ receptor antagonists did not inhibit alcohol reward in histidine decarboxylase knock-out (HDC KO) mice and a lack of histamine did not alter alcohol consumption. Thus H₃ receptor antagonists inhibited alcohol reward in a histamine-dependent manner. Furthermore, WT and HDC KO mice were similarly stimulated by alcohol. The expression levels of dopamine D₁ and D₂ receptors, STEP61 and DARPP-32 mRNA in striatal subregions were unaltered in HDC KO mice. No differences were seen in alcohol kinetics in HDC KO compared to WT control animals. In addition, JNJ-39220675 had no effect on alcohol kinetics in WT mice.

Conclusions and implications: These data suggest that histamine is required for the H₃ receptor-mediated inhibition of alcohol-CPP and support the hypothesis that the brain histaminergic system has an inhibitory role in alcohol reward. Increasing neuronal histamine release via H₃ receptor blockade could therefore be a novel way of treating alcohol dependence.

Keywords: H3 receptor antagonist; alcohol; dopamine; histamine; reward.

Figure 1

Only low doses of H₃ receptor antagonists JNJ-10181457 and JNJ-39220675 inhibit alcohol-induced conditioned place preference (alcohol-CPP) in DBA/2J male mice. Mice develop alcohol-CPP (Contr), which is inhibited by a pretreatment with JNJ-10181457 (5 mg·kg⁻¹, i.p.) or JNJ-39220675 (0.3 mg·kg⁻¹) (A). Higher doses of JNJ-10181457 (10 mg·kg⁻¹) and JNJ-39220675 (3 mg·kg⁻¹ and 10 mg·kg⁻¹) had no effect on alcohol-CPP. Columns indicate the subgroup that received alcohol paired with the metal floor and the subgroup that received alcohol paired with the plastic floor. Place preference is confirmed by the significant difference between the two subgroups of each conditioning group. n = 8–10 per subgroup **P > 0.01, *P > 0.05, ns P > 0.05, one-way anova. Data are expressed as mean time spent (s·min⁻¹ ± SEM) on the metal cue side. H₃ receptor antagonist JNJ-39220675 (10 mg·kg⁻¹, i.p.) has no effect on plasma alcohol concentration at any measured time point (10, 20 100 and 150 min after 2 g·kg⁻¹ alcohol injection) (B). n = 3–5 per group. P > 0.05, two-way anova. Data are expressed as mean ± SEM. JNJ-39220675 does not alter alcohol-induced locomotor stimulation (C). Pretreatment with saline or JNJ-39220675 (0.3, 3 or 10 mg·kg⁻¹, i.p.) was given after a 90-min habituation period and 30 min before alcohol injection (1.0 g·kg⁻¹, i.p.). Alcohol induces significant locomotor activation regardless of the JNJ-39220675 pretreatment; n = 13 per group. ***P < 0.0001, two-way RM anova. Data are expressed as mean ± SEM.

Figure 2

Alcohol-induced conditioned place preference in histidine decarboxylase knock-out (HDC KO) mice. HDC KO mice develop alcohol-CPP (Contr) which was unaffected by H₃ receptor antagonist pretreatments (Cipr – ciproxifan 3 mg·kg⁻¹, JNJ-10181457 1 or 5 mg·kg⁻¹, JNJ-39220675 0.3 or 10 mg·kg⁻¹) (A). Data are expressed as mean time spent (s·min⁻¹ ± SEM) on the metal cue side. Columns indicate the subgroup that received alcohol paired with the metal floor and the subgroup that received alcohol paired with the plastic floor; n = 5–8 per subgroup. Place preference is confirmed by the significant difference between the two subgroups of each conditioning group ***P < 0.001, **P < 0.01 and *P < 0.05, one-way anova. Alcohol concentration in the mouse plasma in HDC KO and wild-type (WT) (129/Sv) control mice (B). Blood was collected 10, 20 100 and 150 min after the alcohol injection (2 g·kg⁻¹, i.p) No differences were detected between the genotypes; n = 3–5. P > 0.05, two-way anova. Data are expressed as mean ± SEM.

Figure 3

Alcohol consumption of histidine decarboxylase knock-out (HDC KO) and wild-type (WT) mice in two-bottle choice and in the drinking in the dark experiments. In the two-bottle choice test, no differences were observed between male HDC KO and WT mice (in 129/Sv background strain) in total alcohol consumption (A) or water-alcohol preference ratio (B). P > 0.05, two-way repeated measures (RM) anova. n = 13–15 per genotype. In the drinking in the dark no differences were observed between male HDC KO and WT mice (in C57BL/6J background strain) in total alcohol (20%, v v^-1, n = 7–8) (C) or sucrose (3%, w v^-1, n = 5–7) (D) consumption. In female mice, no difference was observed in alcohol (20%, v v^-1, n = 4–5) consumption (E) but HDC KO mice consumed less sucrose (3%, w v^-1).P = 0.017, two-way RM anova, n = 4–6. (F). All data are expressed as mean ± SEM.

Figure 4

H₃ receptor antagonists do not alter alcohol-induced locomotor stimulation in wild-type (WT; 129/Sv) or in histidine decarboxylase knock-out (HDC KO) mice. Pretreatment (saline, ciproxifan 3 mg·kg⁻¹, JNJ-39220675 0.3 or 10 mg·kg⁻¹, i.p.) was given after a 60–90 min habituation period and alcohol (1.5 g·kg⁻¹, i.p.) was injected 30 min after the pretreatment. Alcohol induces stimulation in both genotypes. Alcohol-induced stimulation is not affected by the pretreatment with ciproxifan in WT (A) or in HDC KO (B) mice; n = 12–14 per group. Alcohol-induced stimulation is also not affected by the pretreatment with JNJ-39220675 in WT (C) or in HDC KO (D) mice; n = 8–9 per group. Two-way repeated measures anova, no interaction or treatment effect, significant time effect in all groups (P < 0.0001) indicating similar alcohol-induced stimulation regardless of pretreatments with H₃ receptor antagonist. All data are expressed as mean ± SEM.