Repeated exposure to MDMA triggers long-term plasticity of noradrenergic and serotonergic neurons

Abstract

3,4-Methylenedioxymethamphetamine (MDMA or ‘ecstasy’) is a psychostimulant drug, widely used recreationally among young people in Europe and North America. Although its neurotoxicity has been extensively described, little is known about its ability to strengthen neural circuits when administered in a manner that reproduces human abuse (i.e. repeated exposure to a low dose). C57BL/6J mice were repeatedly injected with MDMA (10 mg kg−1, intraperitoneally) and studied after a 4-day or a 1-month withdrawal. We show, using in vivo microdialysis and locomotor activity monitoring, that repeated injections of MDMA induce a long-term sensitization of noradrenergic and serotonergic neurons, which correlates with behavioral sensitization. The development of this phenomenon, which lasts for at least 1 month after withdrawal, requires repeated stimulation of α1B-adrenergic and 5-hydroxytryptamine (5-HT)2A receptors. Moreover, behavioral and neuroendocrine assays indicate that hyper-reactivity of noradrenergic and serotonergic networks is associated with a persistent desensitization of somatodendritic α2A-adrenergic and 5-HT1A autoreceptor function. Finally, molecular analysis including radiolabeling, western blot and quantitative reverse transcription-polymerase chain reaction reveals that mice repeatedly treated with MDMA exhibit normal α2A-adrenergic and 5-HT1A receptor binding, but a long-lasting downregulation of Gαi proteins expression in both locus coeruleus and dorsal raphe nucleus. Altogether, our results show that repeated MDMA exposure causes strong neural and behavioral adaptations and that inhibitory feedback mediated by α2A-adrenergic and 5-HT1A autoreceptors has an important role in the physiopathology of addictive behaviors.

Introduction

3,4-Methylenedioxymethamphetamine (MDMA), commonly known as ‘ecstasy’, is a substituted amphetamine with psychostimulant and hallucinogenic properties. This illicit drug is extensively consumed by teenagers and young people in clubs and rave parties, despite the increasing evidence of its putative neurotoxicity1, 2, 3 and its adverse effects on mental health. Indeed, chronic use of MDMA has been associated with many psychiatric disorders including anxiety, depression or psychosis,4, 5, 6 and may lead to addiction in vulnerable individuals.7

In rodents, MDMA induces locomotor hyperactivity and repeated injections result in behavioral sensitization.8, 9, 10 This long-lasting phenomenon is thought to have a critical role in the development of compulsive drug seeking and drug taking, as well as in cue-induced relapse.11, 12

Studies of the brain circuits underlying addictive behaviors have focused on the mesolimbic dopaminergic system,13, 14 as it was shown that all drugs of abuse, including MDMA, increase extracellular dopamine levels in the nucleus accumbens of rodents.15, 16, 17, 18 However, several in vitro studies indicate that MDMA shows higher affinity for both norepinephrine and serotonin transporters than for dopamine transporter.19, 20, 21

Moreover, growing evidence suggests that the involvement of both noradrenergic and serotonergic systems in MDMA-induced addictive behaviors may have been seriously underestimated. The pharmacological blockade of either α1-adrenergic or 5-hydroxytryptamine (5-HT)2A receptors reduces the hyperlocomotor effects of MDMA and prevents the development of behavioral sensitization.22, 23 Also, self-administration of MDMA is abolished in mice lacking the serotonin transporter (serotonin transporter KO mice),24 whereas mice lacking 5-HT2A receptors (5-HT2A KO mice) are insensitive to MDMA-induced reinforcement and cue-induced reinstatement of MDMA-seeking behaviors.25

Recently, we identified a new kind of neural plasticity that may be involved in the long-term behavioral effects of drugs of abuse. Indeed, we showed that repeated administration of amphetamine, cocaine, morphine, ethanol or nicotine+monoamine oxidase inhibitor induces long-lasting sensitization of noradrenergic and serotonergic neurons in C57BL/6J mice.26, 27, 28, 29 Molecular mechanisms underlying this phenomenon remain however unknown.

The main purpose of this study was to investigate the long-term effects of a repeated MDMA exposure on noradrenergic and serotonergic transmissions. The reactivity of noradrenergic and serotonergic neurons was investigated by in vivo microdialysis in the prefrontal cortex (PFC) and behavioral effects were recorded in parallel. The mechanisms underlying MDMA-induced changes were then investigated at the molecular, physiological and behavioral levels.

Materials and methods

Subjects

Animals were 2–3 months old (26–32 g) C57BL/6J male mice (Charles River, L'Arbresle, France). They were housed eight per cage and maintained on a 12 h light/dark cycle with food and water available ad libitum. Seven hundred mice were used in this study (300 mice for the microdialysis experiments, 256 mice for the behavioral experiments, 80 mice for the neuroendocrine experiments and 64 mice for the molecular experiments). Each mouse was naive at the beginning of each experiment and was not used repeatedly for the different tests. Animal experimentation was conducted in accordance with the guidelines for care and use of experimental animals of the European Economic Community (86/809).

Drugs

3,4-Methylenedioxymethamphetamine (MDMA) hydrochloride, D-amphetamine sulfate, p-chloroamphetamine (PCA) hydrochloride, prazosin hydrochloride, WAY-100635 (N-[2-[4-(2-[O-methyl-3H]methoxyphenyl)-1-piperazinyl]ethyl]-N-2-pyridinyl) cyclohexane carboxamide trihydrochloride) hydrochloride and efaroxan hydrochloride were purchased from Sigma-Aldrich (L’Isle d’Abeau-Chesne, France). Dexmedetomidine hydrochloride was purchased from Tocris Bioscience (Bristol, UK). SR46349B hemifumarate was a generous gift from Sanofi-Aventis (Paris, France) and F13640 was kindly supplied by Pierre Fabre Laboratories (Castres, France). All drugs were dissolved in saline (0.9% NaCl) except prazosin, which was dissolved and sonicated in water (50% of final volume), and then completed with saline and SR46349B, which was dissolved and sonicated in saline plus lactic acid (0.1%), and then finally neutralized with 10 M NaOH. Doses are expressed as salts. MDMA was given at 10 mg kg−1.10 D-amphetamine was given at 2 mg kg−1 and PCA at 7 mg kg−1.26 Doses of prazosin (1 mg kg−1) and SR46349B (1 mg kg−1) were identical to those used in previously reported experiments.26 Efaroxan was given at 2.5 mg kg−1 and WAY-100635 was given at 1 mg kg−1. The dose of F13640 given was between 0.1 and 0.6 mg kg−1 (ref. 30) and of dexmedetomidine was between 0.1 and 0.5 mg kg−1.31 In systemic experiments, drugs were injected intraperitoneally (0.1 ml per mouse). In local experiments, drugs were dissolved in artificial cerebrospinal fluid (see below) at a concentration of 1–1000 μM and infused incrementally in the PFC, the LC or the dorsal raphe nucleus (DRN) through the dialysis probe.

Repeated treatment

Mice were given four consecutive daily injections of saline or MDMA in the cylindrical compartment used for microdialysis or in that used to monitor locomotor activity. After a 4-day or a 1-month withdrawal period, all experiments were performed as described below (Supplementary Figure 1a). To test the effects of prazosin and SR46349B on the development of neurochemical and behavioral sensitization to MDMA, as well as on the MDMA-induced desensitization of α2A-adrenergic and 5-HT1A autoreceptors function, mice were given a pretreatment (saline or prazosin plus SR46349B) every day 30 min before the injection of MDMA.

In vivo microdialysis

Surgery

Mice were anesthetized with sodium pentobarbital (60 mg kg−1; Sanofi Santé Animale, Libourne, France) and placed in a stereotaxic frame (Kopf Instruments, Tujunga, CA, USA). An unilateral permanent cannula (CMA/7; CMA Microdialysis, Solna, Sweden) was placed at the edge of the PFC, the LC or the DRN and secured on the skull with screws and dental cement. The coordinates for the guide cannula tip were as follows: PFC—anteroposterior, +2.6 relative to bregma, mediolateral, +0.5 and dorsoventral, 0 mm from dura; LC—anteroposterior, −5.4 relative to bregma, mediolateral, +0.75 and dorsoventral, −2.4 mm from dura; and DRN—anteroposterior, −4.36 relative to bregma, mediolateral, 0, and dorsoventral, −1.5 mm from dura, according to the atlas of Paxinos and Franklin32 (Supplementary Figure 1b). After surgery, mice were allowed to recover for at least 4 days.

Monitoring of cortical extracellular monoamines levels

On the day of the experiment, the microdialysis probe was inserted into the PFC (membrane length, 2 mm; diameter, 0.24 mm; cutoff, 6000 Da; CMA/7; CMA Microdialysis), the LC or the DRN (membrane length, 1 mm). Artificial cerebrospinal fluid (in mM: 147 NaCl, 3.5 KCl, 1 CaCl2, 1.2 MgCl2, 1 NaH2PO4, 25 NaHCO3, pH 7.6) was perfused through the probe at a rate of 1 μl min−1 by a CMA/100 microinjection pump and samples from the PFC were collected in a refrigerated computer-controlled fraction collector (CMA/170). Adequate steady-state monoamines levels in perfusate samples were reached 140 min after probe insertion. Cortical samples (20 μl every 20 min) were collected for 100 min, to determine basal extracellular values. Then, MDMA, amphetamine, PCA, dexmedetomidine or F13640 were injected either intraperitoneally or infused locally into the PFC, the LC or the DRN through the dialysis probe and cortical samples were collected for 200 min. In local experiments, the concentration of dexmedetomidine and F13640 was gradually increased every 40 min (i.e. every two samples).

Biochemistry

Dialysate samples (20 μl) were injected every 30 min through a rheodyne valve in the mobile phase circuit with a refrigerated automatic injector (Triathlon; Spark Holland, Emmen, The Netherlands). High-performance liquid chromatography was performed with a reverse-phase column (80 × 4.6 mm; 3 μm particle size; HR-80; ESA, Chelmsford, MA, USA). The mobile phase (for NE analysis: 0.1 M NaH2PO4, 0.1 mM EDTA, 2.75 mM octane sulfonic acid, 0.25 mM triethylamine, 3% methanol, pH 2.9; for 5-HT analysis: 0.1 M NaH2PO4, 0.1 mM EDTA, 2.75 mM octane sulfonic acid, 0.25 mM triethylamine, 15% methanol, 5% acetonitrile, pH 2.9; and for DA analysis: 0.1 M NaH2PO4, 0.1 mM EDTA, 2.75 mM octane sulfonic acid, 0.25 mM triethylamine, 6% methanol, pH 2.9) was delivered at 0.7 ml min−1 by an ESA-580 pump. An ESA coulometric detector (Coulochem II 5100A, with a 5014B analytical cell; Eurosep Instruments, Cergy, France) was used for electrochemical detection. The conditioning electrode was set at –0.175 mV and the detecting electrode was set at +0.175 mV.

Histology

At the end of the experiment, a blue dye was infused through the microdialysis probe. Then, the brain was quickly removed and immediately frozen at −30 °C using isopentane cooled by dry ice and serial coronal slices were cut on a microtome to ensure the accurate probe implantation.

Locomotor activity

Mice were introduced into a circular corridor (4.5 cm width, 17 cm external diameter) crossed by four infrared beams (1.5 cm above the base) placed at every 90° (Imetronic, Pessac, France). Locomotor activity was scored when animals interrupted two successive beams and thus had traveled one-quarter of the circular corridor. Spontaneous activity was recorded for 120 min (habituation to the experimental procedure), and then mice were injected with amphetamine or PCA, and locomotor responses were recorded for an additional 200 min.

Dexmedetomidine-induced sedation

Mice were injected with dexmedetomidine (0.1–0.5 mg kg−1, intraperitoenally) and gently rolled onto their backs 30 min later. The hypnotic response to dexmedetomidine was defined as the loss of the mouse’s righting reflex (LORR). The sleep time was measured as the time from the mouse’s inability to right itself when placed on its back until the time when it spontaneously and completely reverted to the prone position.

F13640-induced hypothermia

Body temperature was measured by inserting a 2-cm-long 2-mm-diameter thermistor probe into the rectum of mice that had been gently handled for 20 s. Measurements were made every 10 min for 30 min (basal temperature). Then, the response to a systemic injection of F13640 (0.1–0.6 mg kg−1, intraperitoneally) was assessed every 10 min until the temperature had returned to the basal value. For each animal, the hypothermic response to F13640 was calculated as the difference between basal body temperature (calculated as the average temperature during the period before injection) and the minimal body temperature after injection.

Quantitative autoradiography

Mice were killed by decapitation

The brain was quickly removed and immediately frozen at −30 °C using isopentane cooled by dry ice. Sections (16 μm thick) were cut at −20 °C in a cryostat (Leica CM3050 S, Leica Microsystems, Wetzlar, Germany), and stored at −80 °C for <2 weeks until used for radiolabeling. Locus coeruleus (LC) sections were located between −5.80 and −5.34 mm from bregma, raphe sections between −4.84 and −4.36 mm and PFC sections between +2.22 and +2.80 mm, according to the atlas of Paxinos and Franklin.32

α2A-Adrenergic receptor labeling

Sections were preincubated for 20 min at room temperature in 170 mM Tris-HCl (pH 7.4) containing 20 mM MgCl2 and then incubated for 90 min with the same buffer supplemented with 1.1 nM [125I]-para-iodoclonidine (2200 Ci mmol−1) in a humidified chamber. After two 5-min washes in cold buffer, the sections were quickly dipped in ice-cold distilled water, dried in cold air and exposed for 36 h to [125I]Hyperfilm (GE Healthcare, Little Chalfont, UK) at −80 °C. Nonspecific binding was determined by the inclusion of 10 μM clonidine in the incubation medium.

5-HT1A receptor labeling

Sections were preincubated for 30 min at 20 °C in 50 mM Tris-HCl, pH 7.4, and then incubated for 1 h at 20 °C in the same buffer supplemented with 0.8 nM [3H]WAY-100635 (81 Ci mmol−1). After two 5-min washes in cold buffer, the sections were quickly dipped in ice-cold distilled water, dried in cold air and apposed for 6 weeks to [3H]Hyperfilm (GE Healthcare). Nonspecific binding was estimated on adjacent sections processed through the same steps except that a saturating concentration (10 μm) of 5-HT was added to the incubation medium.

Labeling quantification

Autoradiographic results were analyzed and quantified with MCID software (Imaging Research, Brock University, St Catherines, ON, Canada). Optical density within structures was measured on 10–18 sections for each structure in each animal. Values for specific binding were obtained by subtracting nonspecific binding from total binding.

Immunodetection of Gαi/o proteins

Preparation of homogenates

Brains were obtained as described above and serial coronal sections (400 μm thick for LC, 500 μm for DRN and 1000 μm for PFC) were cut at −18 °C in a cryostat (Leica CM3050 S). Ten milligram samples of LC and DRN and 50 mg samples of PFC (corresponding to eight mice in each group) were punched from the slices according to the atlas of Paxinos and Franklin32 (Supplementary Figure 1c). The punched samples of LC, DRN and PFC were homogenized by sonication, (1 g/10 V) in ice-cold Tris-HCl buffer (50 mM Tris, 100 mM NaCl, 2 mM MgCl2, 1 mM ethylenediaminetetraacetic acid, 1 mM ethylene glycol tetraacetic acid, 0.5 mM dithiothreitol, 0.32 M sucrose, pH 7.3) supplemented with Protease Inhibitor Mixture tablet (Roche Diagnostics, Basel, Switzerland).

Western blot analysis

The protein concentration in all extracted samples was determined by the method of Bradford and bovine serum albumin was used as a protein standard. Protein samples (35 μg) were boiled for 5 min at 95 °C in NuPAGE LDS Sample Buffer (Invitrogen, Carlsbad, CA, USA) containing 2.5% β-mercaptoethanol, and then run on a 10% NuPAGE SDS-PAGE Gel Electrophoresis System and transferred to PVDF membrane (Immobilon; Millipore, Billerica, MA, USA). Membranes were then blocked 1 h at room temperature in Odyssey Blocking Buffer (LI-COR Biosciences, Lincoln, NE, USA), and further incubated overnight at 4 °C with mouse monoclonal antibodies against Gαi1 (1:500, ab19932) or Gαi2 (1:500, ab78193) or rabbit polyclonal antibodies against Gαi3 (1:300, sc-262), or Gαo (1:200, sc-387), plus Tween-20 (0.1%). Gαi1 and Gαi2 antibodies were purchased from Abcam (Cambridge, UK), and Gαi3 and Gαo antibodies were purchased from Santa Cruz Biotech (Heidelberg, Germany). Membranes were then incubated for 1 h at room temperature in Odyssey Blocking Buffer plus Tween-20 (0.1%) with a goat anti-mouse or a goat anti-rabbit IR-Dye 680/800-coupled secondary antibody (1:5000; LI-COR Biosciences). Immunoreactivity was detected with the Odyssey Infrared Imaging System (LI-COR Biosciences). Each Gα protein subtype migrated at approximately 40 kDa. β-Actin (A544J; Sigma-Aldrich) was immunoprobed as loading control. Tyrosine hydroxylase (LI-COR Biosciences) and tryptophan hydroxylase (LI-COR Biosciences) were immunoprobed as markers of tissue specificity. MCID software (Imaging Research) was used for analysis and quantification of immunostainning results.

Isolation of total mRNA and quantification analysis by quantitative RT-PCR of Gαi/o mRNA

mRNA extraction

Samples of LC and DRN (1 mg tissue per structure per mouse) were dissected as described above (Supplementary Figure 1c) and total cellular mRNA was purified using RNeasy Micro Kit from Qiagen (Courtabeauf, France).

Quantitative RT-PCR of Gαi/o mRNA

After extraction, 200 ng of mRNA were reverse transcribed. Quantitative reverse transcription-polymerase chain reaction (qRT-PCR) analysis was performed with SYBR green detection of amplified products in a light cycler instrument (Roche Applied Science, Penzberg, Germany). Primers used for Gαi/o were as follows: Gαi1 (forward, 5′-IndexTermGTCACTGCCCTTACCCTGAA-3′; reverse, 5′-IndexTermAATTGATCCAAAGGCAGGTG-3); Gαi2 (forward, 5′-IndexTermCATCCGTTCGGGTTTTCTAA-3′; reverse, 5′-IndexTermACATTTGGACTTGGGTCAGG-3′); Gαi3 (forward, 5′-IndexTermGCATGACAGGACCAAGGAAT-3′; reverse, 5′-IndexTermGAAGGCAGCACCTCAACTTC-3′); and Gαo (forward, 5′- IndexTermTCGTGATTTTCTCCCCTTTG -3′; reverse 5′- IndexTermTTTTTCTCAATCGCCTTGCT-3′). The relative quantification for a given gene was corrected to the β-actin mRNA values.

Statistics

Statistical analysis was performed using Graph Pad Prism 5.0 software (San Diego, CA, USA). Data from microdialysis and locomotor activity experiments are described as a function of time. Data from microdialysis are expressed as percentages of the respective mean basal value. Extracellular monoamine levels and locomotor activity data obtained after MDMA, amphetamine, PCA, dexmedetomidine or F13640 injection were analyzed for 120 min and compared with a two-way analysis of variance (repeated-measures). Dose-dependent effects of dexmedetomidine (sleep time) and F13640 (hypothermia) were compared with a two-way analysis of variance, followed by a Student’s t-test. Differences in the amounts of α2A-AR, 5-HT1AR, Gαi/o protein and mRNA were assessed using an unpaired Student’s t-test. Pharmacological treatments correspond to independent groups of animals. Each group contained eight mice except microdialysis and qRT-PCR experiments, for which five to six animals were used. Western blot experiments were performed three times with n=8 mice per group. The threshold for significant differences was set at P<0.05.

Results

Acute MDMA increases norepinephrine, serotonin and dopamine levels in the mouse PFC

Acute injection of MDMA (10 mg kg−1, intraperitoneally) induced a 4.2-fold increase in extracellular NE levels in the PFC of mice (F(1,5)=26.70, P<0.0001) (Figure 1a). The maximum effect was observed 20 min after drug injection and NE levels remained significantly higher than those of control mice for at least 80 min. MDMA also induced a fivefold increase in cortical extracellular 5-HT levels (F(1,5)=47.13, P<0.0001) (Figure 1b) and a 2.5-fold increase in cortical extracellular DA levels (F(1,5)=19.39, P<0.005) (Figure 1c).

Figure 1
figure1

Acute 3,4-methylenedioxymethamphetamine (MDMA) increases extracellular norepinephrine (NE), 5-hydroxytryptamine (5-HT) and dopamine (DA) levels in the mouse prefrontal cortex (PFC). Mice were acutely injected with either saline or MDMA (10 mg kg−1, intraperitoneally) and cortical extracellular NE (a), 5-HT (b) and DA levels (c) were monitored by in vivo microdialysis. Cortical extracellular monoamines levels are expressed as a percentage of the respective mean basal value±s.e.m. (absolute baseline values: saline, 0.69±0.06 pg NE/20 min, 0.93±0.08 pg 5-HT/20 min and 1.2±0.2 pg DA/20 min; MDMA, 0.72±0.05 pg NE/20 min, 1.12±0.16 pg 5-HT/20 min and 0.9±0.25 pg DA/20 min); n=5–6 mice per group. There was no significant difference in basal cortical extracellular monoamines levels between the different conditions (P>0.05).

PowerPoint slide

Repeated MDMA administration induces a long-lasting sensitization of noradrenergic and serotonergic neurons, which is correlated to behavioral sensitization

The reactivity of noradrenergic neurons was tested using D-amphetamine as an NE releaser, whereas that of serotonergic neurons was assessed with p-chloroamphetamine, an amphetamine analog that releases 5-HT without modifying extracellular NE levels in mice.26 As expected, acute injection of amphetamine (2 mg kg−1, intraperitoneally) induced an increase in cortical extracellular NE levels (F(1,5)=276.7, P<0.0001) (Figure 2a) and locomotor hyperactivity (F(1,23)=16.89, P<0.005) (Figure 2c). Similarly, PCA (7 mg kg−1, intraperitoneally) induced an increase in cortical extracellular 5-HT levels (F(1,5)=23.03, P<0.0001) (Figure 2b) and locomotor hyperactivity, but with slightly different magnitude and kinetics (F(1,23)=23.79, P<0.0005) (Figure 2d). In mice repeatedly injected with MDMA, the amphetamine-induced release of NE and the PCA-induced release of 5-HT were, respectively, 121.1% and 84.9% higher than in mice repeatedly injected with saline, 4 days after withdrawal (F(1,5)=21.66, P<0.005 and F(1,5)=20.02, P<0.005, respectively) (Figures 2a and b). Moreover, mice repeatedly injected with MDMA exhibited a greater locomotor response to amphetamine (+349.5%) or PCA (+242.5%), 4 days after withdrawal, than mice repeatedly injected with saline (F(1,23)=56.89, P<0.0001 and F(1,23)=36.60, P<0.0001, respectively) (Figures 2c and d).

Figure 2
figure2

Repeated 3,4-methylenedioxymethamphetamine (MDMA) induces a long-lasting sensitization of noradrenergic and serotonergic neurons, which is correlated to behavioral sensitization. Mice were repeatedly injected with either saline or MDMA (10 mg kg−1, intraperitoneally). To investigate the role of α1B-adrenergic and 5-hydroxytryptamine (5-HT)2A receptors in the action of MDMA, mice were pretreated with either saline or SR46349B (SR, 1 mg kg−1, intraperitoneally) plus prazosin (Pz, 1 mg kg−1, intraperitoneally) 30 min before each injection of the drug. Four days or 1 month after the last injection, amphetamine (2 mg kg−1, intraperitoneally) or p-chloroamphetamine (PCA) (7 mg kg−1, intraperitoneally) was administered and cortical extracellular norepinephrine (NE) (a) or 5-HT (b) levels as well as locomotor responses (c and d) were monitored. Cortical extracellular monoamines levels are expressed as a percentage of the respective mean basal value±s.e.m. (absolute baseline values: acute, 0.67±0.08 pg NE/20 min and 0.81±0.07 pg 5-HT/20 min; repeated saline, 0.73±0.11 pg NE/20 min and 1.04±0.07 pg 5-HT/20 min; repeated MDMA 4-d, 0.75±0.13 pg NE/20 min and 0.99±0.09 pg 5-HT/20 min; repeated MDMA 1-m, 0.77±0.08 pg NE/20 min and 0.91±0,13 pg 5-HT/20 min; repeated MDMA+SR/Pz, 0.68±0.09 pg NE/20 min and 0.82±0.16 pg 5-HT/20 min). There was no significant difference in basal cortical extracellular monoamines levels between the different conditions (P>0.05). Locomotor activity is expressed as one-fourth turn per 5 min±s.e.m. Each group included at least five mice for microdialysis experiments and eight mice for behavioral experiments. Locomotor activity following acute saline injection was almost zero and is not shown for clarity. It was verified that repeated (SR+Pz) did not modify the amphetamine and PCA-induced neurochemical and behavioral responses on the test day (data not shown for clarity). 4-d, 4 days after withdrawal; 1-m, 1 month after withdrawal.

PowerPoint slide

Strikingly, the amphetamine-induced release of NE and the PCA-induced release of 5-HT were still higher than control values 1 month following repeated MDMA exposure (+106.5%, F(1,5)=19.91, P<0.005 and +76.5%, F(1,5)=23.26, P<0.0005 for NE and 5-HT, respectively) (Figures 2a and b). Accordingly, behavioral responses to amphetamine and PCA were still significantly different after 1 month withdrawal when compared with those of control animals (+302.3%, F(1,23)=34.00, P<0.0001 and +268.22%, F(1,23)=32.70, P<0.0001, for amphetamine and PCA, respectively) (Figures 2c and d).

To investigate whether the increased reactivity of noradrenergic and serotonergic neurons induced by repeated injections of MDMA was related to the stimulation of α1B-adrenergic and 5-HT2A receptors, as observed with other drugs of abuse,26, 27, 28 mice were pretreated with antagonists of these two receptors, prazosin and SR46349B, respectively, 30 min before each injection of MDMA. We found that mice pretreated with prazosin (1 mg kg−1, intraperitoneally) and SR46349B (1 mg kg−1, intraperitoneally) exhibited an amphetamine-induced release of NE and a PCA-induced release of 5-HT identical to those of animals repeatedly injected with saline (F(1,5)=0.144, P>0.05 and F(1,5)=0.416, P>0.05, respectively) (Figures 2a and b). Moreover, these mice exhibited locomotor responses to amphetamine or PCA identical to those observed in mice repeatedly injected with saline (F(1,23)=0.023, P>0.05 and F(1,23)=0.439, P>0.05, respectively) (Figures 2c and d).

Repeated MDMA fails to sensitize mesocortical dopaminergic neurons

To investigate the role of the mesocortical dopaminergic system in MDMA-induced behavioral sensitization, we also measured cortical extracellular DA levels following repeated MDMA exposure. We found that amphetamine increased extracellular DA levels in the mouse PFC (F(1,5)=39.85, P<0.0005) (Supplementary Figure 2). However, this effect was not significantly different between mice repeatedly injected with MDMA and those repeatedly injected with saline, after a 4-day withdrawal (F(1,5)=0.5747, P>0.05).

The MDMA-induced sensitization of noradrenergic and serotonergic neurons is not a consequence of adaptations at the axon terminal level in the PFC

We have shown that the development of behavioral and neurochemical sensitization to MDMA requires the repeated stimulation of α1B-adrenergic and 5-HT2A receptors. Nevertheless, the neurobiological changes allowing the long-term persistence of this phenomenon remain unknown. To test whether MDMA-induced sensitization of noradrenergic and serotonergic neurons is mediated by adaptations at the axon terminal level, mice repeatedly injected with MDMA were infused with amphetamine or PCA directly through the dialysis probe, 4 days after withdrawal. Figures 3a–c shows that local infusion of 1–1000 μM amphetamine by reverse microdialysis dose-dependently increased cortical extracellular NE levels (F(1,5)=242.43, P<0.0001); administration of 10 μM of amphetamine was required to reach NE levels equivalent to those observed after systemic injection of 2 mg kg−1. Similarly, local infusion of 1–1000 μM PCA dose-dependently increased cortical extracellular 5-HT levels (F(1,5)=239.9, P<0.0001) (Figures 3b–d). Nevertheless, the increase of extracellular monoamines levels obtained in these experimental conditions did not differ significantly between mice repeatedly injected with MDMA and those repeatedly injected with saline, whatever the dose tested (F(1,5)=0.1956, P>0.05 and F(1,5)=0.1452, P>0.05, for NE and 5-HT levels, respectively) (Figures 3a–d).

Figure 3
figure3

The 3,4-methylenedioxymethamphetamine (MDMA)-induced sensitization of noradrenergic and serotonergic neurons is not a consequence of adaptations at the axon terminal level in the prefrontal cortex (PFC). Mice were repeatedly injected with either saline or MDMA (10 mg kg−1, intraperitoneally). After a 4-day withdrawal, amphetamine (0.1–1000 μM) or p-chloroamphetamine (PCA) (0.1–1000 μM) was infused locally in the PFC by reverse microdialysis and cortical extracellular norepinephrine (NE) (a and c) and 5-hydroxytryptamine (5-HT) levels (b and d) were monitored. Cortical extracellular monoamines levels are expressed as a percentage of the respective mean basal value±s.e.m. (absolute baseline values: repeated saline, 0.62±0.05 pg NE/20 min and 0.85±0.09 pg 5-HT/20 min; repeated MDMA, 0.75±0.13 pg NE/20 min and 0.99±0.16 pg 5-HT/20 min); n=5–6 mice per group. There was no significant difference in basal cortical extracellular monoamines levels between the different conditions (P>0.05).

PowerPoint slide

The MDMA-induced sensitization of noradrenergic and serotonergic neurons is associated with sustained desensitization of somatodendritic α2A-adrenergic and 5-HT1A autoreceptor functions

The reactivity of brain monoamines systems is under the negative control of autoreceptors. Noradrenergic neurons that innervate the PFC are inhibited by α2A-adrenergic autoreceptors located in the LC nucleus,33, 34 whereas serotonergic neurons are inhibited by 5-HT1A autoreceptors located in the DRN and the median raphe nucleus.35, 36 At the axon terminal level, NE and 5-HT release is controlled by α2A-adrenergic and 5-HT1B autoreceptors, respectively.

To test whether repeated MDMA exposure impairs the sensitivity of α2A-adrenergic autoreceptors, we analyzed the ability of dexmedetomidine, a selective α2-adrenergic agonist, to decrease cortical extracellular NE levels in mice repeatedly injected with MDMA or saline. As expected, dexmedetomidine (0.5 mg kg−1, intraperitoneally) decreased cortical extracellular NE levels of control mice (−83±6%, F(1,5)=108.9, P<0.0001) (Figure 4a). We found that mice repeatedly injected with MDMA were significantly less sensitive to the effects of dexmedetomidine than those repeatedly injected with saline both 4 days (−39±7%, F(1,5)=19.26, P<0.005) and 1 month after withdrawal (−44±9%, F(1,5)=8.368, P<0.05) (Figure 4a). Interestingly, the MDMA-induced desensitization of dexmedetomidine effects is prevented when mice were pretreated with prazosin (1 mg kg−1, intraperitoneally) plus SR46349B (1 mg kg−1, intraperitoneally) (F(1,5)=0.03, P>0.05 when compared with the control group after a 4-day withdrawal) (Figure 4a).

Figure 4
figure4

The 3,4-methylenedioxymethamphetamine (MDMA)-induced sensitization of noradrenergic and serotonergic neurons is associated with enduring desensitization of somatodendritic α2A-adrenergic and 5-hydroxytryptamine (5-HT)1A autoreceptors functions. Mice were repeatedly injected with either saline or MDMA (10 mg kg−1, intraperitoneally) after being pretreated with either saline or SR46349B (SR, 1 mg kg−1, intraperitoneally) plus prazosin (Pz, 1 mg kg−1, intraperitoneally) 30 min before each injection of the drug. After a 4-day or 1-month withdrawal, they were injected with dexmedetomidine or F13640 either systemically or locally (in the locus coeruleus (LC), the dorsal raphe nucleus (DRN) or the prefrontal cortex (PFC)). Cortical extracellular norepinephrine (NE) (ac) and 5-HT (bd) levels were monitored by in vivo microdialysis. Cortical extracellular monoamines levels are expressed as a percentage of the respective mean basal value±s.e.m. (Systemic experiments: Absolute baseline values of NE—repeated saline 4-d, 0.82±0.12 pg NE/20 min; repeated MDMA 4-d, 0.73±0.22 pg NE/20 min; repeated MDMA 1-m, 0.62±0.05 pg NE/20 min; and repeated MDMA+SR+Pz, 0.71±0.15 pg NE/20 min. Absolute baseline values of 5-HT—repeated saline, 0.77±0.09 pg 5-HT/20 min; repeated MDMA 4-d, 0.97±0.14 pg 5-HT/20 min; repeated MDMA 1-m, 0.89±0.12 pg 5-HT/20 min; repeated MDMA+SR+Pz, 0.86±0.2 pg 5-HT/20 min. Local experiments: Absolute baseline values of NE—repeated saline test LC, 0.57±0.13 pg NE/20 min; repeated MDMA test LC, 0.63±0.18 pg NE/20 min; repeated saline test PFC, 0.68±0.06 pg NE/20 min; and repeated MDMA test PFC, 0.81±0.22 pg NE/20 min. Absolute baseline values of 5-HT—repeated saline test DRN, 0.88±0.25 pg 5-HT/20 min; repeated MDMA test DRN, 1.23±0.22 pg 5-HT/20 min; repeated saline test PFC, 0.79±0.15 pg 5-HT/20 min; and repeated MDMA test PFC, 1.05±0.17 pg 5-HT/20 min). There was no significant difference in basal cortical extracellular monoamines levels between the different conditions (P>0.05). It was verified that repeated (SR+Pz) did not modify the dexmedetomidine- and F13640-induced decrease of cortical extracellular monoamines levels on the test day (data not shown for clarity). (e) Sedative effects of dexmedetomidine measured as the loss of the righting reflex (loss of the mouse’s righting reflex (LORR) test) (0.5 mg kg−1, intraperitoneally). (g) Dose–response curve of dexmedetomidine-induced sedation (0.1–0.5 mg kg−1). (f) Time course of F13640-induced hypothermia (0.6 mg kg−1, intraperitoneally). (h) Dose–response curve of F13640-induced hypothermia; n=6 mice per group for the microdialysis experiments and n=8 mice per group for the LORR test and the hypothermia assay. Data represent mean±s.e.m. *P<0.05, **P<0.005 (unpaired Student’s t-test).

PowerPoint slide

To distinguish which subtype of autoreceptors (axon terminal versus somatodendritic) is involved in dexmedetomidine-induced decrease of cortical NE levels, we then analyzed the effects of this compound when infused locally either in the LC or the PFC through the microdialysis probe. We found that dexmedetomidine (1–1000 μM) dose-dependently decreased cortical extracellular NE levels when injected into the LC (F(1,10)=21.55, P<0.0001) or the PFC (F(1,10)=20.22, P<0.005) (Figure 4c). As shown when injected systematically, we found that mice repeatedly injected with MDMA were significantly less sensitive to the effects of dexmedetomidine than those repeatedly injected with saline after a 4-day withdrawal, when infused locally in the LC (F(1,10)=9.045, P<0.05). In contrast, the effects of dexmedetomidine were not significantly different between the two groups (saline versus MDMA), when infused locally in the PFC (F(1,10)=0.36, P>0.05).

We also analyzed dexmedetomidine-induced sedation in mice repeatedly injected with MDMA or saline as it has been well demonstrated that the hypnotic effects of α2-adrenergic receptors agonists are mediated, at least partly, by the activation of α2A-adrenergic receptors within the LC.31, 37, 38 As expected, dexmedetomidine (0.1–0.5 mg kg−1, intraperitoneally) dose-dependently increased mouse sleep time, measured as the period of loss of the righting reflex (296±27 min at 0.5 mg kg−1, P<0.0001) (Figures 4e–g). We found that mice repeatedly injected with MDMA were significantly less sensitive to the hypnotic effects of dexmedetomidine than those repeatedly injected with saline both 4 days (209±14 min at 0.5 mg kg−1; P<0.05) and 1 month after withdrawal (227±19 min at 0.5 mg kg−1; P<0.05) (Figures 4e–g).

To test the sensitivity of 5-HT1A autoreceptors after MDMA-induced sensitization, we analyzed the ability of F13640, a highly specific 5-HT1A agonist,30 to decrease extracellular cortical 5-HT levels in mice repeatedly injected with MDMA or saline. We found that F13640 (0.6 mg kg−1, intraperitoneally significantly decrease cortical extracellular 5-HT levels of control mice (−81±6%, F(1,5)=81.96, P<0.0001), but the effects observed in mice repeatedly injected with MDMA were significantly lower both 4 days (−30±3%, F(1,5)=32.26, P=0.0005) and 1 month (−27±12%, F(1,5)=18.98, P<0.005) after withdrawal (Figure 4b). Interestingly, the MDMA-induced desensitization of F13640 effects is also prevented when mice were pretreated with prazosin (1 mg kg−1, intraperitoneally) plus SR46349B (1 mg kg−1, intraperitoneally) (F(1,5)=0.643, P>0.05 when compared with the control group after a 4-day withdrawal) (Figure 4b).

To confirm that somatodendritic 5-HT1A autoreceptors are involved in F13640-induced decrease of cortical extracellular 5-HT levels, we then analyzed the effects of this compound when infused locally either in the DRN or the PFC through the microdialysis probe. We found that F13640 (1–1000 μM) dose-dependently decrease cortical extracellular 5-HT levels when infused locally in the DRN (F(1,10)=26.85, P<0.001) but not in the PFC (F(1,10)=0.012, P>0.05) (Figure 4d). As shown when injected systematically, we found that mice repeatedly injected with MDMA were significantly less sensitive to the effects of F13640 than those repeatedly injected with saline after a 4-day withdrawal, when infused locally in the DRN (F(1,10)=10.36, P<0.05).

We also analyzed the ability of F13640 to decrease mouse body temperature, an effect mediated by presynaptic 5-HT1A receptors.30, 39 As expected, F13640 (0.1–0.6 mg kg−1, intraperitoneally) dose-dependently decreased body temperature of the control mice, the maximum effect being observed at 0.6 mg kg−1 (–3.78±0.21 °C, F(1,7)=122.9, P<0.0001) (Figures 4f–h). We found that mice repeatedly injected with MDMA were significantly less sensitive to the hypothermic effects of F13640 than those repeatedly injected with saline, both 4 days and 1 month after withdrawal (–2.48±0.2 °C, F(1,7)=12.77, P<0.005 after 4 days and –2.27±0.28 °C, F(1,7)=15.17, P<0.005 after 1 month, at 0.6 mg kg−1) (Figures 4f–h).

Repeated MDMA does not alter α2A-adrenergic and 5-HT1A receptors binding, but lastingly decreases Gαi proteins expression in LC and DRN cells

We investigated whether the decreased effects of α2-adrenergic and 5-HT1A agonists are mediated by decreased expression of α2A-adrenergic and 5-HT1A receptors. We used quantitative autoradiography to measure the density of these two proteins in the brainstem nuclei and the PFC of mice repeatedly injected with MDMA. The labeling of α2-adrenergic receptors was assessed by using the selective radioligand [125I]-para-iodoclonidine (Figure 5a) and that of 5-HT1A receptors was assessed by using the selective radioligand [3H]WAY-100635 (Figure 5b). We found that mice repeatedly injected with MDMA exhibited [125I]-para-iodoclonidine binding within the LC and the PFC similar to that in mice repeatedly injected with saline, 4 days after withdrawal (P>0.05) (Figure 5c). Similarly, [3H]WAY-100635 binding within the DRN and the PFC did not differ between the MDMA and the control group (P>0.05) (Figure 5d).

Figure 5
figure5

Repeated 3,4-methylenedioxymethamphetamine (MDMA) does not alter α2A-adrenergic and 5-hydroxytryptamine (5-HT)1A receptors binding, but lastingly decreases Gi proteins’ expression in locus coeruleus (LC) and dorsal raphe nucleus (DRN) cells. Mice were repeatedly injected with either saline or MDMA (10 mg kg−1, intraperitoneally). Four days or 1 month after the last injection, α2A-adrenergic and 5-HT1A receptors density was analyzed by quantitative autoradiography and inhibitory G-protein expression was investigated at both transcriptional and protein levels by quantitative reverse transcription-polymerase chain reaction (qRT-PCR) and western blot. [125I]-para-iodoclonidine was used to label α2A-adrenergic receptors in coronal brain sections at the level of the prefrontal cortex (PFC) and of the LC (a). 5-HT1A receptors were labeled with [3H]WAY-100635 (N-[2-[4-(2-[O-methyl-3H]methoxyphenyl)-1-piperazinyl]ethyl]-N-2-pyridinyl) cyclohexane carboxamide trihydrochloride) in the PFC and the DRN (b). Relative optical density of α2A-adrenergic and 5-HT1A receptors is shown in panels c and d. Expression levels of Gαi1, Gαi2, Gαi3 and Gαo were analyzed by western blot in the LC (e) and the DRN (g). Relative expression of each protein is given as a percentage of β-actin signal and is shown in panels f and h. Relative Gαi/o mRNA levels are given as a ratio of β-actin expression and are shown in panels i and j. Radiolabeling experiments were performed with n=8 mice per group. Western blot experiments were performed three times with n=8 mice per group. qRT-PCR experiments were performed with n=6 mice per group. Data represent mean±s.e.m. *P<0.05, **P<0.005 (unpaired Student’s t-test).

PowerPoint slide

Alternatively, the long-term decrease of α2A-adrenergic receptor- and 5-HT1A receptor-mediated effects may be due to alterations in their downstream signaling pathways. Given that both α2A-adrenergic and 5-HT1A receptors are coupled to inhibitory G proteins,40, 41 we used western blotting and qRT-PCR to assess the expression levels of Gi/o-like proteins and mRNAs in the LC, the DRN and the PFC after a repeated MDMA exposure. We found that mice repeatedly injected with MDMA exhibited significant less Gαi1 and Gαi2 immunoreactivity within the LC (t=3.176, P<0.05 and t=3.744, P<0.05 for Gαi1 and Gαi2, respectively) (Figures 5e and f) and Gαi3 immunoreactivity within the DRN (t=3.042, P<0.05) than controls (Figures 5g and h). These differences were observed 4 days after withdrawal and were still significant after 1 month (t=2.874, P<0.05 and t=3.015, P<0.05 for Gαi1 and Gαi2, respectively, and t=2.956, P<0.05 for Gαi3). In contrast, repeated MDMA exposure did not affect Gαi immunoreactivity in the PFC (P>0.05) (Supplementary Figure 3). Despite the persistence of the MDMA-induced decrease in Gαi protein levels observed within the LC and the DRN, we found that this effect was not associated to any significant modifications in Gαi mRNA levels (P>0.05 for each Gαi/o mRNA transcript tested) (Figures 5i and j).

Pharmacological blockade of α2A-adrenergic and 5-HT1A receptors mimics neurochemical and behavioral sensitization

To verify that the alterations of α2-adrenergic autoreceptor- and 5-HT1A receptor-mediated inhibitory feedback may be involved in MDMA-induced neurochemical and behavioral sensitization, we tested if pharmacological blockade of each receptor could enhance the reactivity of noradrenergic and serotonergic neurons. We found that efaroxan (2.5 mg kg−1, intraperitoneally) potentiated the amphetamine-induced release of cortical NE (F(1,5)=8.101, P<0.05) (Supplementary Figure 4a) as well as the locomotor hyperactivity of mice (F(1,23)=8.998, P<0.01) (Supplementary Figure 4c). Similarly, WAY-100635 (1 mg kg−1, intraperitoneally) potentiated the PCA-induced release of cortical 5-HT (F(1,5)=14.18, P<0.005) (Supplementary Figure 4b) as well as its locomotor effects (F(1,23)=8.548, P<0.05) (Supplementary Figure 4d). We also verified that neither efaroxan nor WAY-100635 modified cortical extracellular monoamines levels (F(1,5)=0.8231, P>0.05 and F(1,5)=0.7499, P>0.05 for NE and 5-HT, respectively) or mice locomotor activity by itself (F(1,23)=1.587, P>0.05 and F(1,23)=3.345, P>0.05 for amphetamine and PCA, respectively).

Discussion

The first finding of this study is that acute injection of MDMA (10 mg kg−1, intraperitoneally) significantly increases extracellular NE levels in the mouse prefrontal cortex. This effect is consistent with its pharmacological profile described in vitro and indicate that this compound does indeed induce the release of NE in vivo, although it is usually considered to be a DA/5-HT releaser.42, 43, 44 This finding suggests that compounds targeting the noradrenergic transmission may be useful for treating symptoms of acute ecstasy intoxication in humans, as proposed by recent clinical studies.45, 46 More generally, it highlights the putative role of NE in the mechanisms of action of drugs of abuse47, 48, 49, 50 and51 for review.

The second finding of this study is that repeated administration of MDMA elicits a long-lasting sensitization of noradrenergic and serotonergic systems. We found that the reactivity of noradrenergic and serotonergic neurons was higher than control values 4 days after withdrawal, and remained similarly high for at least 1 month; this effect was strongly correlated with the behavioral response.

Psychostimulants are known to trigger neuroplasticity in ventral tegmental area dopaminergic neurons,52, 53, 54 although no temporal correlation has ever been established between the reactivity of mesolimbic system and the amplitude of the locomotor response. Indeed, the drug-induced increase of extracellular DA levels in the nucleus accumbens is diminished in the early days of withdrawal,55, 56 possibly due to adaptations in the uptake process and only studies conducted several weeks after withdrawal were able to detect a sensitization of DA release,9, 57, 58 suggesting that this phenomenon may not be primarily involved in the development of behavioral sensitization. Regarding the dopaminergic mesocortical pathway, we show here that MDMA-induced behavioral sensitization is not associated with any increase in the reactivity of DA fibers in the PFC at least in the early stages of its development. Interestingly, several studies suggest that MDMA could elicit behavioral sensitization independently of its ability to stimulate dopaminergic transmission. For example, although pharmacological blockade of D1 receptors can block the expression of behavioral sensitization to MDMA, it fails to prevent its induction,59, 60 as shown with cocaine61 and amphetamine.27

In mice, MDMA is known to induce selective and nearly irreversible loss of brain dopaminergic neurons. For example, it decreases DA tissue concentration, dopamine transporter labeling and tyrosine hydroxylase activity in many brain areas.62, 63 It is unlikely, however, that the long-term sensitization of noradrenergic and serotonergic neurons described here is related to the neurotoxicity of MDMA because no MDMA-induced loss of noradrenergic or serotonergic functions has been reported in this species, even at very high dose.64, 65 Moreover, the dose used in this study was substantially lower than the dose potentially toxic in mice. To our knowledge, this is a first time that a non-toxic dose of MDMA has been shown to induce such potent and enduring adaptations in mouse brain that may be relevant to the recreational abuse of this drug.

The third finding of this study is that the MDMA-induced sensitization of noradrenergic and serotonergic neurons is associated with a desensitization of their respective inhibitory feedback. Local perfusion with amphetamine or PCA in the PFC of mice repeatedly injected with MDMA failed to reproduce the substantial potentiation of NE and 5-HT release observed when these two releasers were injected systemically. These findings strongly suggest that the expression of noradrenergic and serotonergic neurons sensitization is not a consequence of adaptations at the nerve terminal level such as increased synthesis, increased storage and release ability and/or upregulation of uptake systems; rather, they suggest adaptations at the somatodendritic level.

Accordingly, we found that mice repeatedly injected with MDMA display a persistent downregulation of somatodendritic α2A-adrenergic and 5-HT1A autoreceptors function. This phenomenon is associated with a selective decrease in the expression of Gαi1 and Gαi2 proteins in the LC and of Gαi3 in the DRN but not in the PFC. Finally, we demonstrated that the blockade of α2-adrenergic receptors potentiated the amphetamine-induced increase in cortical NE, whereas that of 5-HT1A receptors potentiated the PCA-induced increase in cortical 5-HT, as well as the locomotor hyperactivity induced by both compounds. It is worth noting that the control of locomotor activity and anesthetic-sparing (i.e. the ability of a compound to potentiate the sedative/hypnotic effects of an aesthetic drug and to reduce its effective dose) by α2A-adrenergic receptors requires specific coupling with Gαi2 subtype in the LC.31, 41 Moreover, 5-HT1A receptors bind specifically to Gαi3 protein subtype in the raphe nuclei.39, 40 Altogether, these findings support the hypothesis that the hypersensitivity observed in MDMA-treated mice is a consequence of the impairment of somatodendritic α2A-adrenergic autoreceptor- and 5-HT1A autoreceptor-mediated inhibitory feedback. The findings of our local microdialysis experiments suggest that the function of terminal autoreceptors was not significantly altered by repeated administration of MDMA. These data may point to differential control of α2A-adrenergic and 5-HT1A receptors at terminal versus somatodentritic level that may be related to different Gi/o proteins coupling, which has been previously hypothesized.40

The mechanism that leads to the downregulation of Gi proteins in noradrenergic and serotonergic neurons remains, however, unknown. Chronic psychostimulants have been shown to downregulate Gi expression in important brain structures such as the ventral tegmental area and nucleus accumbens,66 but the mechanism involved in this process has not been described. Giving that MDMA does not significantly alter the levels of Gi mRNA, we can assume that such regulations occur at the post-translational level and may affect the proteins stability and their targeting to the plasma membrane as described by Giguère et al.67 It cannot be excluded, however, that because of the limits of the experimental procedure and the small number of NE and 5-HT neurons in each structure, a decrease of Gi mRNA may occur but be too small to be statistically detected.

We found that pharmacological blockade of α2-adrenergic and 5-HT1A receptors before each MDMA injection did not prevent the desensitization of their mediated effects nor the behavioral sensitization (data not shown), suggesting that this phenomenon is not due to their overactivation by the repeated release of NE and 5-HT, but is a consequence of a much more complex process. Interestingly, we found instead that the desensitization of α2A-adrenergic and 5-HT1A autoreceptors function and the consecutive neurochemical and behavioral sensitization were prevented by pretreatment with antagonists of α1-adrenergic and 5-HT2A receptors, as shown with other drugs of abuse.26, 27, 28 Several anatomical and functional studies have demonstrated that α1B-adrenergic and 5-HT2A receptors are critically involved in regulating the reactivity of noradrenergic and serotonergic systems at the subcortical level. In the brainstem, noradrenergic fibers arising from the LC control DRN firing through the activation of α1B-adrenergic receptors located on serotonergic neurons.68, 69 Conversely, the DRN exerts a tonic modulation of the LC via 5-HT2A receptors expressed by both noradrenergic and GABAergic cells.70, 71 Therefore, the repeated stimulation of α1B-adrenergic and 5-HT2A receptors may alter the expression of proteins involved in regulating neuronal excitability, such as inhibitory G proteins, possibly through the activation of the Gq/11 signaling pathway, thereby leading to a decrease of α2A-adrenergic autoreceptor- and 5-HT1A autoreceptor-mediated inhibition.

Interestingly, electrophysiological studies have shown that chronic stress increases LC neurons excitability and that these effects are mediated at least in part by a desensitization of α2A-adrenergic autoreceptors.72, 73 Chronic stress also downregulates 5-HT1A autoreceptors within the DRN.74, 75 It is not known how long such adaptations may persist after recovery but our findings are nevertheless consistent with the putative role of stress in the physiopathology of addiction. Indeed, it should be recorded that stress cross-sensitizes with the behavioral effects of most drugs of abuse, including MDMA.76, 77, 78 Therefore, the sensitization of noradrenergic and/or serotonergic neurons by chronic stressful conditions may explain how the environment could facilitate the development of compulsive drug seeking. By activating both nuclei through the hypothalamic-pituitary-adrenal axis,79, 80, 81 acute stressful stimuli may also promote relapse after withdrawal in the absence of any drug stimulation.

Conclusion

In humans, the abuse potential of ecstasy is still controversial.7 Here we showed that only four injections of a subtoxic dose of MDMA in mice are able to trigger the same long-term neural modifications as compounds with high potential of abuse such as amphetamine, cocaine, morphine, ethanol and nicotine+monoamine oxidase inhibitor , suggesting that ecstasy’s addictivity may have been underestimated. Moreover, we demonstrated that this drug-induced long-term plasticity is a consequence of a sustained impairment of inhibitory feedback mediated by α2A-adrenergic and 5-HT1A autoreceptors in the LC and the DRN, respectively. These latter findings provide potentially new targets for treatment of addiction.

References

  1. 1

    Ricaurte G, Bryan G, Strauss L, Seiden L, Schuster C . Hallucinogenic amphetamine selectively destroys brain serotonin nerve terminals. Science 1985; 229: 986–988.

  2. 2

    McCann UD, Szabo Z, Scheffel U, Dannals RF, Ricaurte GA . Positron emission tomographic evidence of toxic effect of MDMA (‘Ecstasy’) on brain serotonin neurons in human beings. Lancet 1998; 352: 1433–1437.

  3. 3

    Urban NB, Girgis RR, Talbot PS, Kegeles LS, Xu X, Frankle WG et al. Sustained recreational use of ecstasy is associated with altered pre and postsynaptic markers of serotonin transmission in neocortical areas: a PET study with [C]DASB and [C]MDL 100907. Neuropsychopharmacology 2012; 37: 1465–1473.

  4. 4

    Schifano F . Chronic atypical psychosis associated with MDMA (‘ecstasy’) abuse. Lancet 1991; 338: 1335.

  5. 5

    Benazzi F, Mazzoli M . Psychiatric illness associated with ‘ecstasy’. Lancet 1991; 338: 1520.

  6. 6

    Scott RM, Hides L, Allen JS, Burke R, Lubman DI . Depressive and anxiety symptomatology in ecstasy users: the relative contribution of genes, trauma, life stress and drug use. Psychopharmacology (Berl) 2010; 209: 25–36.

  7. 7

    Nutt D, King LA, Saulsbury W, Blakemore C . Development of a rational scale to assess the harm of drugs of potential misuse. Lancet 2007; 369: 1047–1053, review.

  8. 8

    Gold LH, Koob GF . MDMA produces stimulant-like conditioned locomotor activity. Psychopharmacology (Berl) 1989; 99: 352–356.

  9. 9

    Kalivas PW, Duffy P, White SR . MDMA elicits behavioral and neurochemical sensitization in rats. Neuropsychopharmacology 1998; 18: 469–479.

  10. 10

    Doly S, Bertran-Gonzalez J, Callebert J, Bruneau A, Banas SM, Belmer A et al. Role of serotonin via 5-HT2B receptors in the reinforcing effects of MDMA in mice. PLoS One 2009; 4: e7952.

  11. 11

    Robinson TE, Berridge KC . The neural basis of drug craving: an incentive-sensitization theory of addiction. Brain Res Brain Res Rev 1993; 18: 247–291.

  12. 12

    Robinson TE, Berridge KC . The psychology and neurobiology of addiction: an incentive-sensitization view. Addiction 2000; 95 (Suppl 2): S91–117.

  13. 13

    Nestler EJ, Aghajanian GK . Molecular and cellular basis of addiction. Science 1997; 278: 58–63, review.

  14. 14

    Nestler EJ . Is there a common molecular pathway for addiction? Nat Neurosci 2005; 8: 1445–1449, review.

  15. 15

    Di Chiara G, Imperato A . Drugs abused by humans preferentially increase synaptic dopamine concentrations in the mesolimbic system of freely moving rats. Proc Natl Acad Sci USA 1988; 85: 5274–5278.

  16. 16

    Pontieri FE, Tanda G, Orzi F, Di Chiara G . Effects of nicotine on the nucleus accumbens and similarity to those of addictive drugs. Nature 1996; 382: 255–257.

  17. 17

    Tanda G, Pontieri FE, Di Chiara G . Cannabinoid and heroin activation of mesolimbic dopamine transmission by a common mu1 opioid receptor mechanism. Science 1997; 276: 2048–2050.

  18. 18

    Cadoni C, Solinas M, Pisanu A, Zernig G, Acquas E, Di Chiara G . Effect of 3,4-methylendioxymethamphetamine (MDMA, ‘ecstasy’) on dopamine transmission in the nucleus accumbens shell and core. Brain Res 2005; 1055: 143–148.

  19. 19

    Rothman RB, Baumann MH, Dersch CM, Romero DV, Rice KC, Carroll FI et al. Amphetamine-type central nervous system stimulants release norepinephrine more potently than they release dopamine and serotonin. Synapse 2001; 39: 32–41.

  20. 20

    Han DD, Gu HH . Comparison of the monoamine transporters from human and mouse in their sensitivities to psychostimulant drugs. BMC Pharmacol 2006; 6: 6.

  21. 21

    Verrico CD, Miller GM, Madras BK . MDMA (Ecstasy) and human dopamine, norepinephrine, and serotonin transporters: implications for MDMA-induced neurotoxicity and treatment. Psychopharmacology (Berl) 2007; 189: 489–503.

  22. 22

    Selken J, Nichols DE . Alpha1-adrenergic receptors mediate the locomotor response to systemic administration of (+/−)-3,4-methylenedioxymethamphetamine (MDMA) in rats. Pharmacol Biochem Behav 2007; 86: 622–630.

  23. 23

    Herin DV, Liu S, Ullrich T, Rice KC, Cunningham KA . Role of the serotonin 5-HT2A receptor in the hyperlocomotive and hyperthermic effects of (+)-3,4-methylenedioxymethamphetamine. Psychopharmacology (Berl) 2005; 178: 505–513.

  24. 24

    Trigo JM, Renoir T, Lanfumey L, Hamon M, Lesch KP, Robledo P et al. 3,4-methylenedioxymethamphetamine self-administration is abolished in serotonin transporter knockout mice. Biol Psychiatry 2007; 62: 669–679.

  25. 25

    Orejarena MJ, Lanfumey L, Maldonado R, Robledo P . Involvement of 5-HT2A receptors in MDMA reinforcement and cue-induced reinstatement of MDMA-seeking behavior. Int J Neuropsychopharmacol 2010; 14: 1–14.

  26. 26

    Salomon L, Lanteri C, Glowinski J, Tassin JP . Behavioral sensitization to amphetamine results from an uncoupling between noradrenergic and serotonergic neurons. Proc Natl Acad Sci USA 2006; 103: 7476–7481.

  27. 27

    Lanteri C, Salomon L, Torrens Y, Glowinski J, Tassin JP . Drugs of abuse specifically sensitize noradrenergic and serotonergic neurons via a non-dopaminergic mechanism. Neuropsychopharmacology 2008; 33: 1724–1734.

  28. 28

    Lanteri C, Hernández Vallejo SJ, Salomon L, Doucet EL, Godeheu G, Torrens Y et al. Inhibition of monoamine oxidases desensitizes 5-HT1A autoreceptors and allows nicotine to induce a neurochemical and behavioral sensitization. J Neurosci 2009; 29: 987–997.

  29. 29

    Tassin JP . Uncoupling between noradrenergic and serotonergic neurons as a molecular basis of stable changes in behavior induced by repeated drugs of abuse. Biochem Pharmacol 2008; 75: 85–97.

  30. 30

    Colpaert FC, Tarayre JP, Koek W, Pauwels PJ, Bardin L, Xu XJ et al. Large-amplitude 5-HT1A receptor activation: a new mechanism of profound, central analgesia. Neuropharmacology 2002; 43: 945–958.

  31. 31

    Lakhlani PP, MacMillan LB, Guo TZ, McCool BA, Lovinger DM, Maze M et al. Substitution of a mutant alpha2a-adrenergic receptor via ‘hit and run’ gene targeting reveals the role of this subtype in sedative, analgesic, and anesthetic-sparing responses in vivo. Proc Natl Acad Sci USA 1997; 94: 9950–9955.

  32. 32

    Paxinos G, Franklin KBJ . The Mouse Brain in Stereotaxic Coordinates 2nd edn. Academic Press: New York, 1997.

  33. 33

    Aghajanian GK, VanderMaelen CP . Alpha 2-adrenoceptor-mediated hyperpolarization of locus coeruleus neurons: intracellular studies in vivo. Science 1980; 215: 1394–1396.

  34. 34

    Young WS 3rd, Kuhar MJ . Noradrenergic alpha 1 and alpha 2 receptors: light microscopic autoradiographic localization. Proc Natl Acad Sci USA 1980; 77: 1696–1700.

  35. 35

    Gozlan H, El Mestikawy S, Pichat L, Glowinski J, Hamon M . Identification of presynaptic serotonin autoreceptors using a new ligand: 3H-PAT. Nature 1983; 305: 140–142.

  36. 36

    Haddjeri N, Lavoie N, Blier P . Electrophysiological evidence for the tonic activation of 5-HT(1A) autoreceptors in the rat dorsal raphe nucleus. Neuropsychopharmacology 2004; 29: 1800–1806.

  37. 37

    Doze VA, Chen B-X, Maze M . Dexmedetomidine produces a hypnotic-anesthetic action in rats via activation of central, alpha sub 2-adrenoceptors. Anesthesiology 1989; 71: 75–79.

  38. 38

    Correa-Sales C, Rabin B, Maze M . A hypnotic response to dexmedetomidine, an alpha sub 2-agonist, is mediated in the locus coeruleus in rats. Anesthesiology 1992; 76: 948–952.

  39. 39

    Bortolozzi A, Castañé A, Semakova J, Santana N, Alvarado G, Cortés R et al. Selective siRNA-mediated suppression of 5-HT1A autoreceptors evokes strong anti-depressant-like effects. Mol Psychiatry 2012; 17: 612–623.

  40. 40

    Mannoury la Cour C, El Mestikawy S, Hanoun N, Hamon M, Lanfumey L . Regional differences in the coupling of 5-hydroxytryptamine-1A receptors to G proteins in the rat brain. Mol Pharmacol 2006; 70: 1013–1021.

  41. 41

    Albarrán-Juárez J, Gilsbach R, Piekorz RP, Pexa K, Beetz N, Schneider J et al. Modulation of alpha2-adrenoceptor functions by heterotrimeric Galphai protein isoforms. J Pharmacol Exp Ther 2009; 331: 35–44.

  42. 42

    Nash JF, Brodkin J . Microdialysis studies on 3,4-methylenedioxymethamphetamine-induced dopamine release: effect of dopamine uptake inhibitors. J Pharmacol Exp Ther 1991; 259: 820–825.

  43. 43

    Gudelsky GA, Nash JF . Carrier-mediated release of serotonin by 3,4-methylenedioxymethamphetamine: implications for serotonin–dopamine interactions. J Neurochem 1996; 66: 243–249.

  44. 44

    Baumann MH, Clark RD, Rothman RB . Locomotor stimulation produced by 3,4-methylenedioxymethamphetamine (MDMA) is correlated with dialysate levels of serotonin and dopamine in rat brain. Pharmacol Biochem Behav 2008; 90: 208–217.

  45. 45

    Hysek CM, Simmler LD, Ineichen M, Grouzmann E, Hoener MC, Brenneisen R et al. The norepinephrine transporter inhibitor reboxetine reduces stimulant effects of MDMA (‘ecstasy’) in humans. Clin Pharmacol Ther 2011; 90: 246–255.

  46. 46

    Hysek CM, Simmler LD, Nicola VG, Vischer N, Donzelli M, Krähenbühl S et al. Duloxetine inhibits effects of MDMA (‘ecstasy’) in vitro and in humans in a randomized placebo-controlled laboratory study. PLoS One 2012; 7: e36476.

  47. 47

    Drouin C, Darracq L, Trovero F, Blanc G, Glowinski J, Cotecchia S et al. Alpha1b-adrenergic receptors control locomotor and rewarding effects of psychostimulants and opiates. J Neurosci 2002; 22: 2873–2884.

  48. 48

    Ventura R, Cabib S, Alcaro A, Orsini C, Puglisi-Allegra S . Norepinephrine in the prefrontal cortex is critical for amphetamine-induced reward and mesoaccumbens dopamine release. J Neurosci 2003; 23: 1879–1885.

  49. 49

    Ventura R, Alcaro A, Puglisi-Allegra S . Prefrontal cortical norepinephrine release is critical for morphine-induced reward, reinstatement and dopamine release in the nucleus accumbens. Cereb Cortex 2003; 15: 1877–1886.

  50. 50

    Olson VG, Heusner CL, Bland RJ, During MJ, Weinshenker D, Palmiter RD . Role of noradrenergic signaling by the nucleus tractus solitarius in mediating opiate reward. Science 2006; 311: 1017–1020.

  51. 51

    Sofuoglu M, Sewell RA . Norepinephrine and stimulant addiction. Addict Biol 2009; 14: 119–129,, review.

  52. 52

    Ungless MA, Whistler JL, Malenka RC, Bonci A . Single cocaine exposure in vivo induces long-term potentiation in dopamine neurons. Nature 2001; 411: 583–587.

  53. 53

    Zweifel LS, Argilli E, Bonci A, Palmiter RD . Role of NMDA receptors in dopamine neurons for plasticity and addictive behaviors. Neuron 2008; 59: 486–496.

  54. 54

    Mameli M, Halbout B, Creton C, Engblom D, Parkitna JR, Spanagel R et al. Cocaine-evoked synaptic plasticity: persistence in the VTA triggers adaptations in the NAc. Nat Neurosci 2009; 12: 1036–1041.

  55. 55

    Segal DS, Kuczenski R . In vivo microdialysis reveals a diminished amphetamine-induced DA response corresponding to behavioral sensitization produced by repeated amphetamine pretreatment. Brain Res 1992; 571: 330–337.

  56. 56

    Segal DS, Kuczenski R . Repeated cocaine administration induces behavioral sensitization and corresponding decreased extracellular dopamine responses in caudate and accumbens. Brain Res 1992; 577: 351–355.

  57. 57

    Kalivas PW, Duffy P . Time course of extracellular dopamine and behavioral sensitization to cocaine. I. Dopamine axon terminals. J Neurosci 1993; 13: 266–275.

  58. 58

    Paulson PE, Robinson TE . Amphetamine-induced time-dependent sensitization of dopamine neurotransmission in the dorsal and ventral striatum: a microdialysis study in behaving rats. Synapse 1995; 19: 56–65.

  59. 59

    Ramos M, Goñi-Allo B, Aguirre N . Studies on the role of dopamine D1 receptors in the development and expression of MDMA-induced behavioral sensitization in rats. Psychopharmacology (Berl) 2004; 177: 100–110.

  60. 60

    Ramos M, Goñi-Allo B, Aguirre N . Administration of SCH 23390 into the medial prefrontal cortex blocks the expression of MDMA-induced behavioral sensitization in rats: an effect mediated by 5-HT2C receptor stimulation and not by D1 receptor blockade. Neuropsychopharmacology 2005; 30: 2180–2191.

  61. 61

    White FJ, Joshi A, Koeltzow TE, Hu XT . Dopamine receptor antagonists fail to prevent induction of cocaine sensitization. Neuropsychopharmacology 1998; 18: 26–40.

  62. 62

    Granado N, O'Shea E, Bove J, Vila M, Colado MI, Moratalla R . Persistent MDMA-induced dopaminergic neurotoxicity in the striatum and substantia nigra of mice. J Neurochem 2008; 107: 1102–1112.

  63. 63

    O'Callaghan JP, Miller DB . Neurotoxicity profiles of substituted amphetamines in the C57BL/6J mouse. J Pharmacol Exp Ther 1994; 270: 741–751.

  64. 64

    Logan BJ, Laverty R, Sanderson WD, Yee YB . Differences between rats and mice in MDMA (methylenedioxymethylamphetamine) neurotoxicity. Eur J Pharmacol 1988; 152: 227–234.

  65. 65

    Stone DM, Hanson GR, Gibb JW . Differences in the central serotonergic effects of methylenedioxymethamphetamine (MDMA) in mice and rats. Neuropharmacology 1987; 26: 1657–1661.

  66. 66

    Nestler EJ, Terwilliger RZ, Walker JR, Sevarino KA, Duman RS . Chronic cocaine treatment decreases levels of the G protein subunits Gi alpha and Go alpha in discrete regions of rat brain. J Neurochem 1990; 55: 1079–1082.

  67. 67

    Giguère PM, Laroche G, Oestreich EA, Duncan JA, Siderovski DP . Regulation of the subcellular localization of the G-protein subunit regulator GPSM3 through direct association with 14-3-3 protein. J Biol Chem 2012; 287: 31270–31279.

  68. 68

    Baraban JM, Aghajanian GK . Suppression of firing activity of 5-HT neurons in the dorsal raphe by alpha-adrenoceptor antagonists. Neuropharmacology 1980; 19: 355–363.

  69. 69

    Bortolozzi A, Artigas F . Control of 5-hydroxytryptamine release in the dorsal raphe nucleus by the noradrenergic system in rat brain. Role of alpha-adrenoceptors. Neuropsychopharmacology 2003; 28: 421–434.

  70. 70

    Gorea E, Adrien J . Serotonergic regulation of noradrenergic coerulean neurons: electrophysiological evidence for the involvement of 5-HT2 receptors. Eur J Pharmacol 1988; 154: 285–291.

  71. 71

    Szabo ST, Blier P . Functional and pharmacological characterization of the modulatory role of serotonin on the firing activity of locus coeruleus norepinephrine neurons. Brain Res 2001; 922: 9–20.

  72. 72

    Jedema HP, Grace AA . Chronic exposure to cold stress alters electrophysiological properties of locus coeruleus neurons recorded in vitro. Neuropsychopharmacology 2003; 28: 63–72.

  73. 73

    Jedema HP, Gold SJ, Gonzalez-Burgos G, Sved AF, Tobe BJ, Wensel T et al. Chronic cold exposure increases RGS7 expression and decreases alpha(2)-autoreceptor-mediated inhibition of noradrenergic locus coeruleus neurons. Eur J Neurosci 2008; 27: 2433–2443.

  74. 74

    Lanfumey L, Pardon MC, Laaris N, Joubert C, Hanoun N, Hamon M et al. 5-HT1A autoreceptor desensitization by chronic ultramild stress in mice. NeuroReport 1999; 10: 3369–3374.

  75. 75

    Froger N, Palazzo E, Boni C, Hanoun N, Saurini F, Joubert C et al. Neurochemical and behavioral alterations in glucocorticoid receptor-impaired transgenic mice after chronic mild stress. J Neurosci 2004; 24: 2787–2796.

  76. 76

    Antelman SM, Eichler AJ, Black CA, Kocan D . Interchangeability of stress and amphetamine in sensitization. Science 1980; 207: 329–331.

  77. 77

    Deroche V, Marinelli M, Maccari S, Le Moal M, Simon H, Piazza PV . Stress-induced sensitization and glucocorticoids. I. Sensitization of dopamine-dependent locomotor effects of amphetamine and morphine depends on stress-induced corticosterone secretion. J Neurosci 1995; 15: 7181–7188.

  78. 78

    Rougé-Pont F, Marinelli M, Le Moal M, Simon H, Piazza P . Stress-induced sensitization and glucocorticoids. II. Sensitization of the increase in extracellular dopamine induced by cocaine depends on stress-induced corticosterone secretion. J Neurosci 1995; 15: 7189–7195.

  79. 79

    Abercrombie ED, Jacobs BL . Single-unit response of noradrenergic neurons in the locus coeruleus of freely moving cats. II. Adaptation to chronically presented stressful stimuli. J Neurosci 1987a; 7: 2844–2848.

  80. 80

    Abercrombie ED, Jacobs BL . Single-unit response of noradrenergic neurons in the locus coeruleus of freely moving cats. I. Acutely presented stressful and nonstressful stimuli. J Neurosci 1987b; 7: 2837–2843.

  81. 81

    Price ML, Curtis AL, Kirby LG, Valentino RJ, Lucki I . Effects of corticotropin-releasing factor on brain serotonergic activity. Neuropsychopharmacology 1998; 18: 492–502.

Download references

Acknowledgements

This work was supported by Université Pierre et Marie Curie (UPMC), Institut Cochin, Université Paris Descartes, Centre National de la Recherche Scientifique (CNRS) and Institut National de la Santé et de la Recherche Médicale (Inserm). We thank Drs Francis Colpaert (†) and Pierre Sokollof (Laboratoires Pierre Fabre, Castres, France) for kindly providing us F13640. We also thank Dr Marie-Pascale Martres and Caroline Chevarin for her help with radiolabeling experiments; Dr Marie Picot and Nicolas ‘boubou’ Bouveyron for their precious advices on western blot experiments; and Carole Jacq for her help with behavioral experiments. CL would like to thank Vanessa Houades for her support. This work is dedicated to the memory of Iderlinda Deus de Sousa.

Author information

Affiliations

Authors

Corresponding author

Correspondence to C Lanteri.

Ethics declarations

Competing interests

The authors declare no conflict of interest.

Additional information

Author contributions

CL and JPT designed the study. CL, ELD, SJHV, GG, ACB and LS performed research. CL, ELD, SJHV, ACB, LL and JPT analyzed the data. CL wrote the manuscript.

Supplementary Information accompanies the paper on the Molecular Psychiatry website

Supplementary information

PowerPoint slides

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Lanteri, C., Doucet, E., Hernández Vallejo, S. et al. Repeated exposure to MDMA triggers long-term plasticity of noradrenergic and serotonergic neurons. Mol Psychiatry 19, 823–833 (2014). https://doi.org/10.1038/mp.2013.97

Download citation

Keywords

  • α2A-adrenergic autoreceptors
  • behavioral sensitization
  • Gi proteins
  • 5-HT1A autoreceptors
  • MDMA
  • neuroplasticity

Further reading