Serotonin antagonists fail to alter MDMA self-administration in rats

Schenk and colleagues situate the study in the apparent paradox that MDMA (±3,4-methylenedioxymethamphetamine) preferentially increases synaptic serotonin (5-HT) yet nonetheless produces dependence-like behaviour in some human users. Earlier work has established a prominent role for dopaminergic mechanisms in the reinforcing and reinstating effects of many psychostimulants, including evidence that dopamine (DA) antagonists attenuate MDMA self-administration and reinstatement in rodents, but the contribution of specific 5-HT receptor subtypes to MDMA self-administration and relapse-like drug-seeking has not been adequately determined. The present study therefore tested whether pharmacological manipulation of selected 5-HT receptor subtypes alters maintenance of MDMA self-administration and reinstatement of extinguished drug-taking in rats. Specifically, the investigators examined the effects of antagonists at 5-HT1A (WAY 100635), 5-HT1B (GR 127935) and 5-HT2A (ketanserin), and also probed whether selective 5-HT agonists (8-OH-DPAT and RU 24969) modify reinstatement induced by a drug-paired cue. The work aims to clarify whether these 5-HT mechanisms could be targets for reducing MDMA taking or cue- or drug-primed relapse-like behaviour.

Adult male Sprague-Dawley rats were used. Animals were group-housed until reaching approximately 300 g, then singly housed prior to experiments. Intravenous jugular catheters were implanted under ketamine/xylazine anaesthesia with standard post-operative care; catheter patency was checked weekly with pentobarbital. Self-administration sessions occurred in operant chambers with two levers; active-lever presses produced an infusion (100 μl over 12 s) and illumination of a cue light, whereas inactive-lever presses were recorded but had no programmed consequence. Acquisition began with experimenter-delivered infusions to clear lines, then daily 2 h FR1 sessions with MDMA at 1.0 mg/kg per infusion until rats had self-administered 90 infusions or 25 sessions. Approximately 50% of rats failed this initial acquisition criterion and were not tested further; 45 rats met it. For those that progressed, the MDMA dose was reduced to 0.5 mg/kg/infusion until an additional 520 infusions were taken (cumulative exposure ≥350 mg/kg), then the schedule was increased to FR2 (minimum 5 days) and FR5 (at least 7 days) prior to testing. Separate groups were trained to self-administer cocaine (0.5 mg/kg/infusion; n = 16) using a comparable protocol. Three antagonists were tested for effects on ongoing MDMA self-administration (0.5 mg/kg/infusion) in separate groups: WAY 100635 (0, 0.1, 0.3, 1.0 mg/kg, subcutaneous, n = 5), GR 127935 (0, 1.0, 3.0 mg/kg, sc, n = 5) and ketanserin (0, 1.0, 3.0 mg/kg, intraperitoneal, n = 4). Doses were administered in random order with at least two intervening days of baseline self-administration; GR 127935 was given 30 min before sessions, WAY 100635 15 min before, and ketanserin immediately before. Reinstatement testing used an extended protocol with three phases: baseline FR5 self-administration, extinction (vehicle substitution and omission of the light cue until responding fell to <20% of baseline for two consecutive days), and reinstatement (6 h session with presentation of vehicle plus either a priming injection or cue reintroduction). To assess drug-primed reinstatement, separate groups received an injection of MDMA (10.0 mg/kg, ip) after pretreatment with the antagonists (MDMA-trained groups: WAY n = 8, GR n = 7, ketanserin n = 8). Parallel cocaine-trained groups tested antagonist effects on cocaine-primed reinstatement (cocaine 10.0 mg/kg, ip; WAY n = 6, GR n = 5, ketanserin n = 5). To test cue-induced reinstatement, small groups received either the 5-HT1A agonist 8-OH-DPAT (0, 0.1, 0.3, 1.0 mg/kg, sc; n = 4) or the 5-HT1B/1A agonist RU 24969 (0, 0.3, 1.0, 3.0 mg/kg, sc; n = 4), administered 15 min prior to the reinstatement session in which the drug-paired light was reintroduced and vehicle infusions were delivered. Locomotor activity tests were performed to confirm behavioural relevance of antagonist doses and to inspect time-course: rats received antagonist pretreatments (WAY 0 or 1.0 mg/kg; GR 0 or 3.0 mg/kg; ketanserin 0 or 3.0 mg/kg) with MDMA (10.0 mg/kg, ip) given at intervals appropriate to each antagonist’s pretreatment time; activity was recorded for 30 min pre‑ and 60 min post-MDMA. Data were analysed with repeated measures ANOVA (lever and dose as factors) in SPSS v20, with significance at p < 0.05.

During 2 h FR5 tests of MDMA self-administration (0.5 mg/kg/infusion) responding on the active lever was high and inactive-lever responding was low. Separate two-way repeated measures ANOVAs for each antagonist revealed a robust main effect of lever (WAY 100635: F(1,4) = 99.77; GR 127935: F(1,3) = 35.769; ketanserin: F(1,3) = 129.74), indicating discriminated responding, but there were no significant main effects of antagonist dose and no Dose × Lever interactions for any antagonist. In short, none of the three 5-HT antagonists altered maintenance of MDMA self-administration under these conditions. In reinstatement tests following extinction of MDMA self-administration, a priming injection of MDMA (10.0 mg/kg, ip) reinstated responding (0.0 antagonist dose condition). Pretreatment with WAY 100635, GR 127935 or ketanserin failed to reduce MDMA-primed reinstatement in the MDMA-trained groups; ANOVAs did not show significant effects of these antagonists on MDMA-produced drug-seeking. By contrast, some antagonists did alter cocaine-primed reinstatement in cocaine-trained rats. Baseline responding was high and fell with extinction, and cocaine (10.0 mg/kg, ip) reinstated responding. There was a significant Lever × WAY Dose interaction (F(1,5) = 7.393), with a significant reduction in active-lever responses after 1.0 mg/kg WAY 100635. Ketanserin produced a main effect of Dose (F(1,4) = 7.916) and active-lever responses were significantly decreased at both 1.0 and 3.0 mg/kg. GR 127935 produced no significant effects on cocaine reinstatement. Cue-induced reinstatement tests (reintroduction of the light stimulus previously paired with MDMA infusions) showed that reintroducing the light markedly increased responding relative to extinction; there was a Test Day × Lever interaction (F(1,7) = 8.954). The 5-HT1A agonist 8-OH-DPAT significantly decreased light cue–produced reinstatement, whereas the 5-HT1B/1A agonist RU 24969 did not produce a significant effect. The authors also report that the antagonist doses used attenuated MDMA-produced hyperactivity in locomotor assays, indicating pharmacological activity at the used doses.

Schenk and colleagues interpret the pattern of findings as evidence that antagonism at 5-HT1A, 5-HT1B or 5-HT2A receptors does not alter maintenance of MDMA self-administration nor MDMA-primed reinstatement under the extensive training and testing conditions used. They emphasise that a moderate MDMA self-administration dose was selected to allow upward or downward shifts in responding to be detected, and note that the antagonist doses were behaviourally active because they reduced MDMA-induced hyperactivity and, in some cases, attenuated cocaine-primed reinstatement—making inadequate dosing an unlikely explanation for the null effects on MDMA measures. The investigators acknowledge a potential pharmacokinetic caveat: the elimination half-life of some antagonists (notably WAY 100635) is relatively short compared with MDMA and, given the 6 h reinstatement sessions, a transient antagonist effect might have been missed; nevertheless, ketanserin and GR127935 have relatively long half-lives compared with MDMA and were still inactive, and prior work with other short‑half-life DA antagonists produced clear attenuation of MDMA reinstatement, arguing against a purely pharmacokinetic account. The authors situate their findings within a broader hypothesis that repeated MDMA exposure produces reductions in serotonergic markers (for example decreased 5-HT transporter density) together with enhanced dopaminergic responses in nucleus accumbens, and that these neuroadaptive changes shift the dependence-like control of behaviour toward DA mechanisms. Extended acquisition procedures used in this study produced substantial MDMA exposure (≥350 mg/kg), and the investigators suggest that deficits in 5-HT may be a prerequisite for the high, stable levels of MDMA self-administration and for reinstatement observed after extinction. They note procedural differences across laboratories—such as infusion duration, presence/absence of post-infusion time-out, and total exposure—that may account for variability in MDMA self-administration findings between groups. Although antagonists at the tested 5-HT receptors did not reduce MDMA taking or MDMA-primed reinstatement, the 5-HT1A agonist 8-OH-DPAT decreased cue-induced reinstatement produced by reintroduction of the drug-paired light. The authors propose that 5-HT1A agonists may suppress cue-driven drug-seeking, potentially via interactions that reduce psychostimulant-evoked DA overflow, and thus might warrant consideration as adjunctive agents to limit cue-triggered relapse. Overall, the present data do not support 5-HT1A, 5-HT1B or 5-HT2A antagonists as promising therapeutics for MDMA dependence, while indicating that certain serotonin agonists could influence cue-elicited drug-seeking. Limitations the authors acknowledge include possible pharmacokinetic mismatches for some antagonists, procedural variability across studies, and species/enantiomer differences reported in non-human primate work; they recommend further examination of these factors to clarify serotonergic contributions to MDMA reinforcement and relapse-like behaviour.