Role of serotonin via 5-HT2B receptors in the reinforcing effects of MDMA in mice

Abstract

The amphetamine derivative 3,4-methylenedioxymethamphetamine (MDMA, ecstasy) reverses dopamine and serotonin transporters to produce efflux of dopamine and serotonin, respectively, in regions of the brain that have been implicated in reward. However, the role of serotonin/dopamine interactions in the behavioral effects of MDMA remains unclear. We previously showed that MDMA-induced locomotion, serotonin and dopamine release are 5-HT(2B) receptor-dependent. The aim of the present study was to determine the contribution of serotonin and 5-HT(2B) receptors to the reinforcing properties of MDMA.We show here that 5-HT(2B) (-/-) mice do not exhibit behavioral sensitization or conditioned place preference following MDMA (10 mg/kg) injections. In addition, MDMA-induced reinstatement of conditioned place preference after extinction and locomotor sensitization development are each abolished by a 5-HT(2B) receptor antagonist (RS127445) in wild type mice. Accordingly, MDMA-induced dopamine D1 receptor-dependent phosphorylation of extracellular regulated kinase in nucleus accumbens is abolished in mice lacking functional 5-HT(2B) receptors. Nevertheless, high doses (30 mg/kg) of MDMA induce dopamine-dependent but serotonin and 5-HT(2B) receptor-independent behavioral effects.These results underpin the importance of 5-HT(2B) receptors in the reinforcing properties of MDMA and illustrate the importance of dose-dependent effects of MDMA on serotonin/dopamine interactions.

Figure 1. Effect of 5-HT2B receptor inhibition on locomotor activity and behavioral sensitization after MDMA injection.

Locomotor sensitization after MDMA two-injection protocol in WT and 5-HT_2B^−/− mice (A–D): MDMA (10 mg/kg i.p.) increases locomotor activity after the first injection (1^st) in WT mice but not in 5-HT_2B ^−/− mice A) or RS127445-treated WT mice B). The stimulant effect of a challenge dose of MDMA (10 mg/kg) 7 days later (2^nd) was significantly enhanced compared to the first injection in WT mice, while it had no effect in 5-HT_2B ^−/− mice A) or RS127445 pre-treated WT mice B). To evaluate the role of 5-HT_2B receptors in the development of locomotor sensitization, WT mice were treated with RS127445 30 min before the first injection of MDMA. The increased responsiveness to the challenge injection of MDMA in absence of RS127445 was totally abolished C). MDMA (30 mg/kg i.p.) induces locomotor activity after the first injection in WT and 5-HT_2B ^−/− mice compared to saline injection D). The stimulant effect of a challenge dose of MDMA (30 mg/kg) was significantly enhanced compared to the first injection in both WT and 5-HT_2B ^−/− mice D). Data (means±SEM, n = 8−14 per group) were analyzed by two-way ANOVA with genotype (A–D) or RS127445 pre-treatment (B–C) and MDMA treatment as main factors. A significant interaction was observed for the locomotor activity in figure A) F (2,66) = 12.86 p<0.01, and B) F (2,54) = 11.49 p<0.01, as well as a main effect of genotype F (2, 66) = 15.68, p<0.001 (A) or RS127445 pre-treatment F (2, 54) = 10.24, p<0.05 (B) and of MDMA treatment F (1,66) = 9.26, p<0.001 (A), and F (1,54) = 17.04, p<0.001 (B). No significant interaction was observed for the locomotor activity in figure C) F (2,66) = 3.57, ns, whereas a main effect of RS127445 pre-treatment at the 1^st MDMA injection, F (2,66) = 28.56, p<0.001, and of MDMA treatment F (1,66) = 6.64, p<0.01 were detected. Neither a significant interaction, F (2,50) = 0.25, ns, nor a main effect of genotype F (1,50) = 0.9, ns, was observed for the locomotor activity in figure D), whereas a main effect of MDMA treatment, F (2,50) = 82.72, p<0.001, was detected. Bonferroni tests were used for post-hoc comparisons. The null hypothesis was rejected at the p<0.05 level; *p<0.05; **p<0.01; ***p<0.001 compared to saline-treated mice. °p<0.05; °°°p<0.001 compared to MDMA 1^st injection. Locomotor sensitization after repeated MDMA injection in WT and 5-HT_2B^−/− mice (E): MDMA (10 mg/kg i.p.) increases locomotor activity after the first injection (day 1; d1) in WT mice but not in 5-HT_2B ^−/− mice compare to saline injection. Repeated MDMA injection during the following days (Day 2 to 5, d2–d5) increases locomotor activity only in WT mice. The stimulant effect of a challenge dose of MDMA (10 mg/kg) 5 days later (day 10; d10) was significantly enhanced compared to the first injection in WT mice, while it had no effect in 5-HT_2B ^−/− mice. Data (means±SEM, n = 8 per group) were analyzed by two-way ANOVA with genotype and MDMA treatment as main factors. Bonferroni tests were used for post-hoc comparisons. The null hypothesis was rejected at the p<0.05 level; **p<0.01; ***p<0.001 compared to saline-treated mice. °°°p<0.001 compared to MDMA day 1 injection.

Figure 2. Effect of 5-HT2B receptor inhibition on MDMA (30 mg/kg) - evoked increase in NAcc 5-HT and DA levels as measured by in vivo microdialysis in awake mice.

Effect of MDMA (30 mg/kg i.p.) or saline injection on (A) 5-HT, (B) DA concentrations in dialysates from the NAcc of WT or RS127445 (0.5 mg/kg i.p.)-WT pre-treated mice. Effect of 5-HT2B receptor genetic ablation on MDMA (30 mg/kg)-induced changes in (C) 5-HT or (D) DA levels in NAcc. MDMA or saline solutions were injected 35 minutes after test began (arrow). Basal 5-HT and DA extracellular levels are reported at lower scale in E) and F) respectively for WT and 5-HT2B-/- mice. Data (means±SEM, n = 4−5 per group) were analyzed by two-way ANOVA (repeated measures) with RS-administration or genotype and time as main factors. A significant interaction was observed after 5-HT2B receptor pharmacological inhibition (RS127445) (A–B) for 5-HT levels F (36,180) = 429.5 p<0.001 as well as for DA levels F (36,180) = 82.43, p<0.001. A main effect of RS-treatment was observed for 5-HT levels F (2, 180) = 3677, p<0.001 and for DA levels F (2, 180) = 449.7, p<0.001, whereas a main effect of time was also detected for 5-HT levels F (18, 180) = 498.1, p<0.001 and for DA levels F (18, 180) = 166.5, p<0.001. For 5-HT2B-/- mice (C–D), a significant interaction was not observed for 5-HT levels F (18,126) = 1.33, ns, but it was the case for DA levels F (18,126) = 53.98, p<0.001. Neither a main effect of genotype F (1, 126) = 0.36, ns, nor of time F (18, 126) = 1.08, ns, was observed for 5-HT levels. On the contrary, a main effect of genotype F (1, 126) = 363.3, p<0.001, as well as of time F (18, 126) = 53.55, p<0.001 was detected for DA levels. Bonferroni tests were used for post-hoc comparisons. The null hypothesis was rejected at the p<0.05 level; *p<0.001.

Figure 3. Effect of 5-HT2B receptor inhibition on MDMA-induced conditioned place preference in mice.

A) Repeated i.p. injection of MDMA 10 mg/kg induced CPP in WT mice compared to saline injection, but this effect was absent in 5-HT_2B ^−/− mice. However, repeated injection of a high dose of MDMA (30 mg/kg) induced CPP in WT as well as in 5-HT_2B ^−/− mice. Data (means±SEM, n = 8 per group) were analyzed by two-way ANOVA with genotype and treatment as main factors, revealing a significant interaction, F(2, 41) = 3.93, p<0.05, a main effect of treatment F(2, 41) = 11.67, p<0,001 but no effect of genotype F(1, 41) = 0.27, ns. Bonferroni tests were used for post-hoc comparisons. In all cases, p<0.05 was considered statistically significant; **p<0.01 compared to saline-treated mice. B) After an initial extinction of the CPP (12 days) in WT mice, MDMA (10 mg/kg) re-exposure induced reinstatement of CPP. MDMA-induced reinstatement was not observed in RS127445-pretreated mice. Data (means±SEM, n = 10−20 per group) were analyzed by one-way ANOVA. Dunnetts tests were used for post-hoc comparisons. The null hypothesis was rejected at the p<0.05 level; *p<0.05 **p<0.01 compared to saline-treated mice.

Figure 4. Effect of 5-HT_2B receptor on MDMA-induced ERK activation.

24 hours after CPP training (Figure 3), the same WT and 5-HT_2B ^−/− mice were re-exposed to saline, MDMA 10 mg/kg or MDMA 30 mg/kg before being processed for immunohistochemistry. A) Single confocal sections showing p-ERK1/2 immunoreactivity in the NAcc Shell of WT and 5-HT_2B ^−/−mice 10 min after re-exposure. Scale bar: 40 µm. B, C) Quantification of p-ERK1/2 immunoreactive neurons in the NAcc Shell (B) and NAcc Core (C) of WT and 5-HT_2B ^−/− mice 10 min after re-exposure. Note the strength of ERK activation induced by re-exposure to both MDMA doses in WT animals. This activation was absent in 5-HT_2B ^−/− mice at 10 mg/kg MDMA, although it fully recovered at 30 mg/kg MDMA. Data (means±SEM; n = 4−10 mice per group) were analyzed using two-way ANOVA with genotype and treatment as main factors, revealing a significant interaction in NAcc Shell F(2, 33) = 21.27, p<0,001 and in NAcc Core, F(2, 33) = 4.77, p<0.05. A main effect of treatment was observed in NAcc Shell F(2, 33) = 45.29, p<0,001 and in NAcc Core, F(2, 33) = 21.85, p<0,001, but no effect of genotype neither in NAcc Shell F(1, 33) = 0.53, ns nor in NAcc Core, F(1, 33) = 0.21, ns. Bonferroni tests were used for post-hoc comparisons. The null hypothesis was rejected at the p<0.05 level; ** p<0.01, *** p<0.001 compared to saline-treated mice; °°° p<0.001, compared to MDMA 10 mg/kg treated mice.

Figure 5. MDMA reexposure after CPP induces strong ERK1/2 activation exclusively in striatonigral medium-sized spiny neurons (MSNs) of the NAcc shell.

drd2-EGFP BAC transgenic mice, in which EGFP expression is driven by the D2-receptor promoter, were trained for the same CPP protocol as in Figure 3 and re-exposed 24 hours after CPP training to saline, MDMA 10 mg/kg or MDMA 30 mg/kg. A) Single confocal sections showing p-ERK1/2 immunoreactivity (1–3, red) colocalized with D2-mediated EGFP (1′–3′, green) and DARPP-32 immunoreactivity (1″–3″, blue) in the NAcc Shell of drd2-EGFP mice 10 minutes after re-exposure to saline (1–1″), MDMA 10 mg/kg (2–2″) or MDMA 30 mg/kg (3–3″). The absolute segregation from D2-EGFP neurons in 1′–3′ and the complete colocalization with a fraction of DARPP-32-positive neurons in 1″–3″ reveal the identity of neurons expressing ERK1/2 activation, which are D1 receptor-containing striatonigral MSNs. Scale bar: 40 µm. B) p-ERK1/2 (red) + D2-EGFP (green) + DARPP-32 (blue) colocalized confocal image of the NAcc shell of an animal challenged with 10 mg/kg MDMA. Arrows point to neurons labeled only in blue, which are the small fraction of striatonigral MSNs that do not show ERK activation. Scale bar: 40 µm. C) Quantification of p-ERK1/2 immunoreactive neurons among EGFP-negative (−) or EGFP-positive (+) neurons in the NAcc Shell of drd2-EGFP mice 10 min after saline, MDMA 10 mg/kg or MDMA 30 mg/kg injections. D) Quantification of p-ERK1/2 immunoreactive neurons among EGFP-negative (−) or EGFP-positive (+) neurons in the NAcc Core of drd2-EGFP mice 10 min after saline, MDMA 10 mg/kg or MDMA 30 mg/kg injections. Data (means±SEM; n = 3 mice per group) were analyzed using two-way ANOVA with genotype and treatment as main factors, revealing a significant interaction F(2, 12) = 16.8, p<0,001 and a main effect of treatment F(2, 12) = 15.93, p<0,001 as well as of genotype F(1, 12) = 83.71, p<0.001. Bonferroni tests were used for post-hoc comparisons. The null hypothesis was rejected at the p<0.05 level; *** p<0.001 compared to saline-treated mice.