Augmentation of serotonin release by sustained exposure to MDMA and methamphetamine in rat organotypic mesencephalic slice cultures containing raphe serotonergic neurons

Authors


Address correspondence and reprint requests to Takayuki Nakagawa, Department of Molecular Pharmacology, Graduate School of Pharmaceutical Sciences, Kyoto University, 46-29 Yoshida-Shimoadachi-cho, Sakyo-ku, Kyoto 606-8501, Japan. E-mail: tnakaga@pharm.kyoto-u.ac.jp

Abstract

Several lines of evidence suggest the involvement of the raphe-serotonergic neurons in addiction to psychostimulants and some recreational drugs. In this study, we established rat organotypic mesencephalic slice cultures containing the raphe nuclei and examined the effects of sustained exposure to 3,4-methylenedioxymethamphetamine (MDMA) and methamphetamine (METH). Immunostaining for tryptophan hydroxylase (TPH) studies revealed that serotonergic neurons were abundant in the slice cultures. Sustained exposure to MDMA and METH (1–1000 μM) for 4 days had little effect on the serotonin tissue content, [3H]citalopram binding, or expression/phosphorylation of TPH. Treatment with MDMA or METH for 30 min increased serotonin release in a concentration-dependent manner. Slice cultures were exposed to MDMA for 4 days following a 1-day withdrawal period and then challenged with MDMA (10 μM). Sustained MDMA exposure augmented MDMA-induced serotonin release in a concentration-dependent manner, indicating serotonergic sensitization. Similar serotonergic sensitization was observed for METH. The development of MDMA-induced serotonergic sensitization was attenuated by the NMDA receptor antagonist, MK-801 (10 μM). These results suggest that in mesencephalic slice cultures sustained MDMA or METH exposure induces serotonergic sensitization through activation of NMDA receptors without serotonergic neurotoxicity. The in vitro model system could help to elucidate the mechanisms underlying drug addiction.

Abbreviations used
5,7-DHT

5,7-dihydroxytryptamine

5-HIAA

5-hydroxyindolacetic acid

5-HT

serotonin

KRB

Krebs–Ringer buffer

MDMA

3,4-methylenedioxymethamphetamine

METH

methamphetamine

PBS

phosphate-buffered saline

SERT

serotonin transporter

TPH

tryptophan hydroxylase

3,4-Methylenedioxymethamphetamine (MDMA, ‘ecstasy’) and methamphetamine (METH) are popular drugs of abuse throughout the world. In rodents, repeated intermittent exposure to MDMA and METH is known to cause progressive augmentation of the locomotor response (Spanos and Yamamoto 1989; Itzha et al. 2004). This phenomenon termed behavioral sensitization is thought to underlie certain aspects of drug addiction (Robinson and Berridge 1993). It is well established that the behavioral effects of psychostimulants depend on their ability to elevate extracellular dopamine levels in the mesocorticolimbic dopaminergic system that originates in the ventral tegmental area (for review, see Robinson and Berridge 1993; Nestler 2005). The mesocorticolimbic dopaminergic neurons could be regulated by the excitatory glutamatergic system, which project from limbic and cortical areas to the ventral tegmental area and the nucleus accumbens. A body of evidence suggests that neuroadaptive changes in these neural circuits play a critical role in behavioral sensitization (for review, see Wolf 1998; Vanderschuren and Kalivas 2000).

In addition to the dopaminergic and glutamatergic systems, several lines of evidence suggest the involvement of the serotonergic system in drug addiction (for review, see Müller et al. 2007; Howell and Kimmel 2008). METH has an ability to interact with the serotonin transporter (SERT), and MDMA acts preferentially to SERT and norepinephrine transporter, rather than to the dopamine transporter (Han and Gu 2006). Unlike reuptake inhibitors such as the selective SERT inhibitors and cocaine, METH, and MDMA are substrate-type releasers, which are transported into the nerve terminals and cause rapid release of serotonin (5-HT) via a reverse transport-mediated and secretory vesicle-dependent mechanism (for review, see Howell and Kimmel 2008). Indeed, in vivo microdialysis studies have shown that systemic injection or local application of METH and MDMA increases the release of 5-HT from serotonergic terminals located in the nucleus accumbens, striatum, hippocampus, and prefrontal cortex (Müller et al. 2007). This acute 5-HT release is, at least in part, thought to underlie the behavioral and thermal responses to MDMA (Callaway et al. 1990; Cole and Sumnall 2003), although norepinephrine is also important (Sprague et al. 2005; Selken and Nichols 2007). Serotonergic neurons from the dorsal and medial raphe nuclei directly and indirectly innervate the mesocorticolimbic–dopaminergic neurons in the ventral tegmental area, nucleus accumbens, and prefrontal cortex (Imai et al. 1986; Hervéet al. 1987; Phelix and Broderick 1995) and modulate their activity (Blandina et al. 1989; Parsons and Justice 1993a; Prisco et al. 1994; Di Matteo et al. 1999; Bortolozzi et al. 2005; Díaz-Mataix et al. 2005), although the interaction between the serotonergic and dopaminergic systems is complicated and controversial because of a variety of 5-HT receptor subtypes and their localization. Indeed, 5-HT1A receptor agonists (Przegaliński and Filip 1997; Przegaliński et al. 2000; Ago et al. 2006), 5-HT1B receptor antagonists (Przegaliński et al. 2001), and 5-HT2A receptor antagonists (Moser et al. 1996) attenuate amphetamine-induced hyperactivity and reduce behavioral sensitization. Repeated treatments with MDMA cause not only sensitization of locomotion, but also sensitization of 5-HT syndrome behaviors and hyperthermia (Spanos and Yamamoto 1989; Dafters 1995). These effects are mediated, at least in part, through the indirect stimulation of 5-HT2A receptors (Herin et al. 2005; Ross et al. 2006). Furthermore, repeated treatments with amphetamines and other drugs of abuse augment the release of 5-HT as well as norepinephrine in the prefrontal cortex (Ago et al. 2006; Salomon et al. 2006; Lanteri et al. 2008). Taken together, these findings suggest that repeated administration of METH or MDMA alters the activity of the serotonergic and noradrenergic systems, and that this plays a role in their behavioral sensitization, although the mechanisms still remain unclear. In addition, the raphe-serotonergic neurons and glutamatergic neurons are reciprocally connected (Aghajanian and Wang 1977; Groenewegen and Uylings 2000; Roche et al. 2003), and the activity of serotonergic neurons and 5-HT release is modulated by glutamatergic neuron function (Pallotta et al. 1998; Celada et al. 2001; Gartside et al. 2007). For example, antipsychotic drugs partly exert their therapeutic effect through modulation interactions between serotonergic and glutamatergic neurons (Breese et al. 2002; Amargós-Bosch et al. 2007).

Many studies of drug addiction have been assessed by in vivo experiments in whole animals, because these addiction-related phenomena are thought to be because of long-term alterations in psychological behavior caused by plasticity in the underlying neural circuits. Nevertheless, in vitro model systems that retain neuronal and synaptic function for extended periods should provide a great deal of data, which should facilitate investigations of the neural plasticity underlying drug addiction. However, there have been no reports of in vitro studies that have successfully analyzed the alterations in serotonergic activity following long-term exposure to drugs of abuse. In the present study, we have established rat organotypic mesencephalic slice cultures containing raphe serotonergic neurons. These cultures make the serotonergic neurons available for analyses of the acute and chronic effects of MDMA and METH. Here we report that sustained exposure of the slice cultures to MDMA or METH-induced augmentation of their 5-HT releasing effects, that is, they induced serotonergic sensitization. Furthermore, we examined the involvement of NMDA glutamate receptors in the acute MDMA-induced 5-HT release and sustained MDMA-induced serotonergic sensitization.

Materials and methods

Preparation of rat organotypic mesencephalic slice cultures

All animals were handled in accordance with the ethical guidelines of the Kyoto University Animal Experimentation Committee and the Japanese Pharmacological Society. Organotypic slice cultures were prepared according to the methods described previously (Katayama et al. 2002; Maeda et al. 2004) with slight modifications. Briefly, Sprague–Dawley rats (Nihon SLC, Shizuoka, Japan) at postnatal day 2–3 were anesthetized with hypothermia, decapitated, and the brain was removed from the skull. Coronal sections (350 μm thick) were prepared under sterile conditions with a tissue chopper (Narishige, Tokyo, Japan) at the mesencephalic levels. Tissue samples were dissected for the mesencephalic slice including dorsal and median raphe nuclei, with the aid of the Atlas of the Developing Rat Brain (Paxinos et al. 1991). Slice cultures were transferred onto 30 mm Millicell-CM insert membranes (pore size 0.4 μm; Millipore, Bedford, MA, USA) in 6-well plates (Becton Dickinson, Tokyo, Japan). Four slices were cultivated on each insert membrane. Culture medium consisting of 50% minimal essential medium/HEPES, 25% Hank’s balanced salt solution, and 25% heat-inactivated horse serum (Invitrogen, Carlsbad, CA, USA) supplemented with 6.5 mg/mL glucose and 2 mM l-glutamine, 100 U/mL penicillin G potassium, and 100 μg/mL streptomycin sulfate (Gibco), was supplied at a volume of 0.7 mL per well such that the slices were maintained at the liquid/air interface. The culture medium was exchanged for fresh medium on the day following culture preparation, and on every second day thereafter. Slice cultures were maintained in a humidified atmosphere of 5% CO2 and 95% air at 37°C for 14–16 days before drug treatment.

Measurement of 5-HT release and tissue content

For measurement of 5-HT release, the culture medium was aspirated and substituted with Krebs–Ringer buffer [KRB (in mM): 146 NaCl, 2.7 KCl, 1 MgCl2, 1.2 CaCl2, 10 d-(+)-glucose, 15 HEPES, 5 HEPES-Na, 0.2 ascorbic acid; pH 7.4). Slice cultures were pre-incubated with 0.7 mL of KRB for 15 min, and then the KRB was immediately replaced with an equivalent volume of fresh KRB containing drugs. After a 30-min incubation period, the KRB was collected into vials and immediately analyzed. For measurement of tissue 5-HT content after the treatment with drugs, the slice cultures were collected and then homogenized and sonicated in 500 μL ice-cold 0.1 N HClO4 containing 10 mM Na2S2O5 and 1 mM EDTA and placed on ice for 15 min. In some experiments, the protein concentrations were measured using the Bradford protein assay (Bio-Rad, Hercules, CA, USA). The homogenates were centrifuged at 18 000 g for 15 min at 4°C, and the pH of the supernatants was adjusted to 3.0 with 1 M CH3COONa.

HPLC with an electrochemical detection system (Eicom, Kyoto, Japan) was used to measure the release and tissue content of 5-HT, and the basal release of 5-hydroxyindolacetic acid (5-HIAA), as further described in Appendix S1 in supporting information. Because the actual number of serotonergic neurons in the prepared slice culture varied, depending on the preparation conditions, such as rostrocaudal level of the slice, the level of 5-HT release and tissue content in each well also varied. According to our preliminary study, the basal tissue content and release of 5-HT correlated well with the basal level of 5-HIAA release. Therefore to minimize variations, the 5-HT concentration in the KRB (representing 5-HT release) and in the supernatants of tissue homogenate (representing tissue 5-HT content) of each well was divided and corrected by the corresponding basal 5-HIAA concentration in the KRB (basal 5-HIAA release), measured 1 day before starting the drug treatment. The values of 5-HT release/basal 5-HIAA release and tissue 5-HT content/basal 5-HIAA release from the control slice cultures in each experiment served as 100%, and the results are presented as the mean of the percentage ± SEM of the control slice culture.

Sustained exposure protocol

Slice cultures were incubated with culture medium containing phosphate-buffered saline (PBS) or drugs. Immediately prior to drug exposure (day 0), 2 days (day 2) and 4 days (day 4) after the initiation of the sustained exposure of drug, 5-HT release was measured by incubating cultures with KRB containing PBS or drugs, for 30 min. For 4-day sustained exposure, the medium was exchanged for fresh medium containing PBS or drugs without washout on day 2. After sustained exposure for 4 days, the medium in the slice cultures were replaced with normal culture medium and cultured for 1 day. Following this 1-day withdrawal (day 5), both PBS- and drug-treated slice cultures were challenged with the drugs at a concentration of 10 μM for 30 min and 5-HT release was measured again.

Immunohistochemistry

Slice cultures were fixed with 4% paraformaldehyde, and then permeabilized and blocked with 0.2% Triton 100-X and 10% fetal calf serum. The cultures were then incubated overnight at 4°C with anti-tryptophan hydroxylase (TPH) antibody (1 : 500, Chemicon International, Temecula, CA, USA), followed by appropriate secondary antibody. See more details in Appendix S1 in supporting information.

Western blot

Slice cultures were homogenized, sonicated, and diluted in sodium dodecyl sulfate–polyacrylamide gel electrophoresis loading buffer. The samples were resolved on sodium dodecyl sulfate–polyacrylamide gel, transferred to polyvinylidene fluoride membrane, and probed overnight with anti-TPH (1 : 3000, Chemicon International) or anti-phospho-TPH (1 : 500; Sigma, St Louis, MO, USA). Proteins were visualized by enhanced chemiluminescence assay. See more details in Appendix S1 in supporting information.

[3H]Citalopram binding assay

Slice cultures were homogenized and sonicated. The samples were pelleted by centrifugation, resuspended in ice-cold binding buffer, and then incubated with 4 nM [3H]citalopram (76.0 Ci/mmol; GE Healthcare UK Ltd., Buckinghamshire, UK) for 60 min at 25°C. Incubation was terminated followed by rapid filtration. The radioactivity on each filter was measured by liquid scintillation spectrometry. See more details in Appendix S1 in supporting information.

Statistical analysis

Data are presented as means ± SEM. Differences between two groups were compared using Student’s t-test. When there were more than two groups, data were compared by one-way or two-way anova followed by the Bonferroni post hoc test for multiple comparisons. Differences with p < 0.05 were considered significant.

Results

Assessment of functional serotonergic neurons in rat mesencephalic slice cultures containing the raphe nuclei

The mesencephalic slices were maintained for 14–16 days in culture. The slice cultures gradually flattened during cultivation. Immunostaining for TPH, the rate-limiting enzyme in the biosynthesis of 5-HT, revealed that numerous TPH-positive cell bodies and neurites could be clearly observed by both light and fluorescence microscopy (Fig. 1). TPH-positive cell bodies were localized mainly around two regions, identified as the dorsal and median raphe nuclei according to the Atlas of the Developing Rat Brain (Paxinos et al. 1991). TPH-positive neurites were widely distributed throughout the culture. The tissue contents of 5-HT and its derivate 5-HIAA in the mesencephalic slice culture were 419.0 ± 68.9 pmol/mg protein (n = 15) and 185.8 ± 44.9 pmol/mg protein (n = 11), respectively. The concentration of 5-HT and 5-HIAA released totally in the 0.7 mL KRB buffer exposed to the slice for 30 min was 0.96 ± 0.17 nM (n = 28) and 229.6 ± 36.0 nM (n = 28), respectively.

Figure 1.

 Immunohistochemical staining of tryptophan hydroxylase (TPH) in the mesencephalic slice cultures. (a) Representative photomicrograph showing TPH immunoreactivity in the whole slice culture by diaminobenzidine staining. The asterisk represents the cerebral aqueduct. Scale bar = 500 μm. Square frames indicate the regions corresponding to (a′) and (a″) (Scale bar = 100 μm). (b) Representative photomicrograph for TPH immunofluorescence. Scale bar = 50 μm. TPH-positive cell bodies and neurites were observed mainly in two regions thought to be the dorsal and median raphe nuclei.

To assess the functionality of the serotonergic neurons, the mesencephalic slice cultures were treated with a selective serotonergic neurotoxin, 5,7-dihydroxytryptamine (5,7-DHT) (200 μM) for 2 days, and then tissue 5-HT content was measured (Fig. S1a). The tissue 5-HT content was significantly decreased by the 5,7-DHT treatment compared with PBS-treated cultures. Co-treatment with citalopram (10 μM), a selective SERT inhibitor, significantly attenuated the 5,7-DHT-induced decrease in tissue 5-HT content, while citalopram alone had no effect. Similarly, the 5,7-DHT treatment significantly decreased the specific [3H]citalopram binding from 7.87 ± 0.64 pmol/mg protein to 1.66 ± 0.51 pmol/mg protein (p < 0.001). Next, the extracellular release of 5-HT was measured in the presence of citalopram and high K+ (50 mM) stimulation (Fig. S1b). The 5-HT release was slightly, but not significantly, increased by 30-min treatment with citalopram (10 μM). High K+ combined with citalopram dramatically and significantly increased the 5-HT release compared with PBS- and citalopram-treated slice cultures.

Effect of sustained exposure to MDMA and METH on tissue 5-HT content and [3H]citalopram binding

To assess the changes in the serotonergic nerve terminals after sustained exposure of the slice cultures to MDMA and METH, we examined the effects of the drugs upon tissue 5-HT content and [3H]citalopram binding. Slice cultures were exposed to MDMA (1–1000 μM) or METH (1–1000 μM) in the culture medium for 4 days. One day after the withdrawal period, the slices were collected without measuring extracellular 5-HT and 5-HIAA release. Sustained exposure of the slice cultures to MDMA had no effect on the tissue 5-HT content, while sustained METH exposure resulted in a slight decrease in tissue 5-HT content, in a concentration-dependent manner. A significant decrease was observed only at the highest concentration of METH (1000 μM) compared with the PBS-exposed cultures (Fig. 2a and b). On the other hand, the 4-day sustained exposure to MDMA tended to decrease the [3H]citalopram binding, although not significantly. Sustained METH exposure (1–1000 μM) had no effect on the [3H]citalopram binding (Fig. 2c and d).

Figure 2.

 Effects of sustained exposure to 3,4-methylenedioxymethamphetamine (MDMA) (a, c) and methamphetamine (METH) (b, d) on tissue serotonin (5-HT) content and [3H]citalopram binding. Slice cultures were exposed to phosphate-buffered saline (PBS) and MDMA (1–1000 μM) or METH (1–1000 μM) in the culture medium for 4 days (days 0–4). One day after the withdrawal period (day 5), the slices were collected without measuring extracellular 5-HT and 5-hydroxyindolacetic acid (5-HIAA) release (a, b). Tissue 5-HT contents were normalized and values shown represent means of the percentage ± SEM of control slice cultures, which had been exposed to PBS only for 4 days. *p < 0.05 compared with PBS-exposed control culture, n = 4–5. (c, d) [3H]citalopram binding. Values represent means of the percentage ± SEM of control slice cultures exposed to PBS for 4 days. The control [3H]citalopram binding was 7.87 ± 0.64 pmol/mg protein, n = 4–8.

Effect of acute and sustained exposure to MDMA on the expression and phosphorylation of TPH

The effects of sustained exposure to MDMA on the expression and phosphorylation of TPH were examined by immunohistochemistry and western blot. As shown in Fig. 3a, cultures exposed to MDMA for 4 days and immunostaining for TPH showed no apparent changes in fluorescence intensity, cell number, or morphology of TPH-immunoreactive cells. Furthermore, neither acute nor sustained exposure to MDMA (300 μM) affected the total expression and phosphorylation of TPH, as assessed by western blot (Fig. 3b).

Figure 3.

 Effect of acute and sustained treatment with 3,4-methylenedioxymethamphetamine (MDMA) on the expression and phosphorylation of tryptophan hydroxylase (TPH). (a) Representative photomicrograph showing TPH immunofluorescence in the slice cultures exposed to MDMA (300 μM) for 4 days. Scale bar = 50 μm. (b) The slice cultures were exposed to MDMA (300 μM) for 30 min or 4 days, and western blot analyses for total TPH and phosphorylated-TPH were performed. Actin was used as a control for sample loading. The upper panel shows representative blots. In the lower panel, the densities of specific TPH, phosphorylated-TPH and actin bands were measured. TPH levels were normalized against the corresponding actin levels and phosphorylated-TPH levels were normalized against the corresponding TPH levels. The value obtained for the phosphate-buffered saline-treated control slice cultures served as the 100% condition, and the results are presented as the mean of the percentage ± SEM, n = 4. PBS, phosphate-buffered saline.

Acute treatment with MDMA and METH induces release of 5-HT

In the mesencephalic slice cultures, we examined the effect of acute MDMA and METH treatment on 5-HT release (Fig. 4). Treatment with MDMA or METH (1–1000 μM) for 30 min increased the 5-HT level detected in the KRB in a concentration-dependent manner. MDMA and METH began to increase 5-HT release at a concentration of 100 μM, and significant increases were observed at concentrations of 300 and 1000 μM compared with PBS-treated slice cultures.

Figure 4.

 Effects of acute treatment with 3,4-methylenedioxymethamphetamine (MDMA) (a) and methamphetamine (METH) (b) on serotonin (5-HT) release in the mesencephalic slice cultures. Slice cultures were treated with MDMA or METH (1–1000 μM) in Krebs–Ringer buffer (KRB) for 30 min, and then the extracellular 5-HT release was determined and normalized. Values represent means of the percentage ± SEM of control slice cultures treated with phosphate-buffered saline (PBS) for 30 min. *p < 0.05, **p < 0.01 compared with PBS-treated culture, n = 5–9.

Sustained exposure to MDMA and METH induces serotonergic sensitization

We examined the effect of sustained exposure of the slice cultures to MDMA and METH on the 5-HT release. Slice cultures were exposed to MDMA or METH (300 μM) in normal culture medium. Immediately prior to exposure, 2 and 4 days after the initiation of sustained exposure of drug, the 5-HT levels in the KRB were measured. As described above, immediately prior to the initiation of the sustained drug exposure (day 0), treatment with MDMA or METH (300 μM) for 30 min significantly increased 5-HT release, compared with PBS-exposed cultures. Subsequently, 2 and 4 days after the sustained exposure to MDMA or METH (300 μM), re-treatment with MDMA or METH (300 μM), for 30 min enhanced 5-HT release, while the 5-HT release in the PBS-exposed cultures did not change when re-treated with PBS. The MDMA- or METH-induced 5-HT release observed on day 2 and day 4 were significantly higher than that observed on day 0 (Fig. 5a and c). After the 4-day exposure, the culture medium was replaced by normal culture medium. Following the withdrawal day (day 5), 5-HT release was measured following a challenge with MDMA or METH at the relatively low concentration of 10 μM for 30 min, to both PBS- and MDMA- or METH-exposed cultures. In the PBS-exposed cultures, 5-HT release induced by MDMA- or METH challenge was low, at almost the same level observed following acute treatment with 10 μM MDMA or METH. In contrast, in the cultures that received sustained exposure to MDMA or METH, 5-HT release induced by challenge with 10 μM MDMA or METH was significantly augmented, compared with that observed in the PBS-exposed cultures (Fig. 5b and d). The augmentation effects of MDMA or METH challenge upon 5-HT release on day 5 was dependent on the concentration of the MDMA or METH during sustained exposure (10–1000 μM). A significant augmentation of 5-HT release was observed following sustained exposure to MDMA at concentrations of 300 and 1000 μM and following sustained exposure to METH at a concentration of 1000 μM compared with PBS-exposed cultures (Fig. 6).

Figure 5.

 Effect of sustained exposure to 3,4-methylenedioxymethamphetamine (MDMA) (a, b) and methamphetamine (METH) (c, d) on serotonin (5-HT) release in the mesencephalic slice cultures. (a, c) Slice cultures were exposed to phosphate-buffered saline (PBS) and MDMA (a) or PBS and METH (b) at drug concentrations of 300 μM for 4 days. For 4-day sustained exposure, the medium was exchanged for fresh medium containing PBS or drugs without washout on day 2. Immediately prior to the experiment (day 0), 2 days (day 2) and 4 days (day 4) days after initiation of the sustained exposure of drug, the extracellular release of 5-HT into Krebs–Ringer buffer (KRB) was measured over a 30-minute period in the presence of PBS only (open column), MDMA (300 μM; closed column), or METH (300 μM; hatched column). (b, d) After the 4-day sustained exposure, the culture medium was replaced. One day after the withdrawal period (day 5), normalized 5-HT release was determined again following a challenge with MDMA (b) or METH (d) at a concentration of 10 μM in KRB for a period of 30 min in PBS-, MDMA-, or METH-exposed cultures. Values represent means of the percentage ± SEM of control slice cultures treated with PBS for 30 min. *p < 0.05, ***p < 0.001, n = 3–6.

Figure 6.

 Concentration-dependent augmentation of serotonin (5-HT) release by sustained exposure to 3,4-methylenedioxymethamphetamine (MDMA) (a) and methamphetamine (METH) (b). Slice cultures were exposed to phosphate-buffered saline (PBS) and MDMA (10–1000 μM) or PBS and METH (10–1000 μM) for 4 days (days 0–4). The medium was exchanged for fresh medium containing PBS or drugs without washout on day 2. One day after the withdrawal period (day 5), normalized 5-HT release was determined following a challenge with MDMA or METH at a concentration of 10 μM in Krebs–Ringer buffer (KRB), for a period of 30 min in PBS-, MDMA-, or METH-treated cultures. Values represent means of the percentage ± SEM of PBS-exposed control slice cultures challenged with 10 μM MDMA or METH for 30 min. **p < 0.01, ***p < 0.001, n = 3–6.

Effect of an NMDA glutamate receptor antagonist on acute 5-HT release and MDMA-induced development of serotonergic sensitization

Immediately prior to the initiation of sustained drug exposure (day 0), acute treatment with MK-801 (10 μM), an NMDA glutamate receptor antagonist, alone for 30 min had no effect on basal 5-HT level. Co-treatment with MK-801 for 30 min partially but significantly attenuated the 5-HT release induced by MDMA (300 μM), compared with MDMA only-treated cultures (Fig. 7a). Next, the cultures were exposed to MDMA (300 μM) in combination with or without MK-801 (10 μM) for 4 days and then incubated in normal cultured medium. One day after the withdrawal period (day 5), the augmentation of 5-HT release by the 10 μM MDMA challenge was significantly attenuated by sustained co-exposure to MK-801. In contrast, sustained exposure to MK-801 alone had no effect on 5-HT release following an MDMA challenge (Fig. 7b).

Figure 7.

 Effect of MK-801, an NMDA glutamate receptor antagonist on the serotonin (5-HT) release induced by acute 3,4-methylenedioxymethamphetamine (MDMA) treatment and the development of the 5-HT release augmentation induced by sustained MDMA exposure. (a) Immediately prior to the sustained exposure (day 0), slice cultures were treated with phosphate-buffered saline (PBS) or MDMA (300 μM) with (closed column) or without (open column) MK-801 (10 μM) for 30 min, and then of 5-HT into the Krebs–Ringer buffer (KRB) was measured. (b) The slice cultures were exposed to PBS or MDMA (300 μM) with or without MK-801 (10 μM) in the culture medium for 4 days (days 0–4). The medium was exchanged for fresh medium containing PBS or drugs without washout on day 2. One day after the withdrawal period (day 5), normalized 5-HT release was determined following a challenge with MDMA (10 μM) in KRB, for a period of 30 min. Values represent means of the percentage ± SEM of control slice cultures treated with PBS for 30 min. *p < 0.05, **p < 0.01, n = 3–6.

Discussion

To analyze the mechanisms underlying drug addiction, an in vitro model system involving an organotypic brain slice culture, which retains neural and synaptic function enabling long-term analysis might be advantageous. In this study, we have established rat organotypic mesencephalic slice cultures containing raphe serotonergic neurons. Immunohistochemical analyses of the slice cultures revealed TPH-positive, probably serotonergic cell bodies and neurites that are abundantly located in the vicinity of the dorsal and medial raphe nuclei. In addition, neurochemical studies confirmed that the slice cultures have appropriate tissue contents of 5-HT and its metabolite, 5-HIAA, at levels even higher than those previously reported in the raphe nucleus of adult rats (5-HT, 130 pmol/mg protein; 5-HIAA, 110 pmol/mg protein; Cransac et al. 1996). Our organotypic mesencephalic slice cultures, which were maintained for 14–16 days, contained abundant raphe serotonergic neurons. It has been reported that TPH activity in the organotypic mesencephalic slice culture is barely detectable at the beginning of culture, increased gradually during the first week, reaching a maximum after 2 weeks in the culture (Jonakait et al. 1988). Although there have been a small number of studies on the raphe serotonergic neurons using organotypic slice cultures, most researchers have investigated the development of serotonergic projections, but not 5-HT release and re-uptake (Papp et al. 1995; Branchereau et al. 2002; Guthrie et al. 2005). The present results show that 5,7-DHT, which exerts a selective neurotoxic effect on the serotonergic neurons by uptake through SERT (Baumgarten and Lachenmayer 2004), dramatically decreases both the intracellular 5-HT content and specific [3H]citalopram binding. 5,7-DHT-induced serotonergic neurotoxicity is prevented by citalopram. These data suggest that the in vitro system using slice cultures might facilitate analyses of serotonergic neurotoxicity and also provide evidence for the existence of functional SERT in slice cultures. Furthermore, we show that 5-HT release was increased by stimulation with high K+ levels in the presence of citalopram, indicating that the slice cultures contain functional serotonergic neurons with the ability to release 5-HT in response to stimulation. Taken together, it is proposed that the organotypic mesencephalic slice cultures described here are suitable for the analysis of both acute and sustained effects of MDMA and METH on serotonergic neurons.

It is well recognized that MDMA and METH exert potent neurotoxic effects on serotonergic neurons. These effects include a long-lasting depletion of 5-HT and damage to serotonergic nerve terminals (Kita et al. 2003; Lyles and Cadet 2003). In addition, acute and chronic administration of MDMA and METH produce rapid and prolonged inhibition of TPH activity (Bakhit and Gibb 1981; Schmidt and Taylor 1987). However, the present results show that sustained MDMA and METH exposure have little effect on the tissue 5-HT content or [3H]citalopram binding, and neither acute nor sustained treatment with MDMA affects the expression and phosphorylation of TPH. These results suggest that MDMA and METH cause minimal serotonergic neurotoxicity and no changes in the amount of TPH in this in vitro model system. Consistent with our present findings, in vitro preparations such as brain slices or cell lines show no evidence of MDMA-induced serotonergic neurotoxicity (Schmidt and Taylor 1988; Hrometz et al. 2004). In addition, MDMA infused directly into the rat brain causes minimal long-term depletion of 5-HT (Breier et al. 2006). Inhibition of TPH activity is not thought to be caused directly by psychostimulants (Schmidt and Taylor 1987). Although the specific mechanisms underlying MDMA- and METH-induced serotonergic neurotoxicity are still controversial, a hypothesis that it is partly mediated by dopamine transported into the serotonergic terminals through SERT, has been proposed (Sprague et al. 1998; Hrometz et al. 2004). In the organotypic mesencephalic slice cultures, serotonergic and dopaminergic interconnection might occur to a lesser degree than observed in vivo, or alternatively, dopamine released from dopaminergic nerve terminals might diffuse and be diluted in the culture medium.

In the mesencephalic slice cultures, acute treatment with MDMA and METH increases 5-HT release, as expected. However, the concentrations required for the 5-HT release induced by acute treatment with MDMA or METH in slice cultures are higher than those reported in other in vitro systems, namely, cell lines expressing SERT (Johnson et al. 1998; Hilber et al. 2005; Pifl et al. 2005). In the preparation described here, because the level of 5-HT was measured in the absence of monoamine oxidase inhibitors, 5-HT might be vigorously degraded by monoamine oxidases before it can spill over into the KRB. In addition, unlike dissociated cultured cells, the slice culture is thick and its surface is covered by numerous glial cells, which might limit the diffusion into the KRB of any 5-HT released from nerve terminals. Nevertheless, the acute MDMA- and METH-induced release of 5-HT is concentration-dependent, suggesting that the present in vitro system is, at least, suitable for assessing the function of the serotonergic neurons.

The major finding of the present study is that sustained exposure of the slice culture to MDMA and METH augments the 5-HT release induced by the challenge with a low concentration of MDMA or METH. This effect is concentration-dependent. Because the degradation and metabolism of METH or MDMA have not been observed during sustained exposure in this preparation, it is unclear whether the slice cultures were exposed to constant levels of METH and MDMA for the 4 days. Nevertheless, to our knowledge, this is the first demonstration showing that sustained MDMA and METH exposure can induce sensitization of serotonergic neurons in vitro. Consistent with our study, recent in vivo studies have reported that repeated administration of amphetamines induces augmentation of 5-HT release in the prefrontal cortex, and that this is accompanied by behavioral sensitization (Ago et al. 2006; Salomon et al. 2006) as well as other drugs of abuse (Parsons and Justice 1993b; Lanteri et al. 2008). In contrast, repeated administration of MDMA according to a neurotoxic regimen reduces the 5-HT release evoked by electrical stimulation in the prefrontal cortex (Gartside et al. 1996). However, our results from slice cultures demonstrate that the serotonergic sensitization induced by sustained exposure to MDMA or METH is not accompanied by serotonergic neurotoxicity or facilitation of 5-HT biosynthesis.

A growing body of evidence suggests that glutamate receptors play a critical role in the behavioral sensitization to psychostimulants (Wolf 1998; Vanderschuren and Kalivas 2000). Competitive and non-competitive NMDA receptor antagonists prevent the development of behavioral sensitization (Karler et al. 1989; Kuribara et al. 1992; Ohmori et al. 1994), and also dopaminergic sensitization (Wolf et al. 1994; Grönig et al. 2004), but there have been no reports investigating the effect of these antagonists on serotonergic sensitization. In the present study, we found that the development of MDMA-induced serotonergic sensitization is attenuated by co-exposure to MK-801. These data suggest that NMDA glutamate receptors play an important role in the neural plasticity underlying the development of MDMA-induced serotonergic sensitization. However, MK-801 also attenuats the 5-HT release induced by acute treatment with MDMA. Raphe serotonergic neurons receive glutamatergic innervation (Aghajanian and Wang 1977; Groenewegen and Uylings 2000; Roche et al. 2003), and the local application of glutamate and NMDA increases both the firing activity of serotonergic neurons and the release of 5-HT in the areas to which these neurons project (Pallotta et al. 1998; Celada et al. 2001; Gartside et al. 2007). The 5-HT release induced by acute MDMA treatment may be mediated, at least in part, by glutamatergic activation, although it is unknown whether and how MDMA directly or indirectly affects the glutamatergic neurons. It is possible that the mesencephalic slice cultures studied here contain a neural network constituting the link between the serotonergic and glutamatergic systems, which might contribute to the development of serotonergic sensitization.

The role of the raphe serotonergic neurons in behavioral sensitization to drugs of abuse still remains controversial. However, Ago et al. (2006) have reported that the selective 5-HT1A agonist osemozotan attenuates both the maintenance of METH-induced behavioral sensitization and serotonergic sensitization in the prefrontal cortex. In addition, the capacity of drugs to induce serotonergic sensitization corresponds closely to their capacity to induce development of behavioral sensitization (Salomon et al. 2006; Lanteri et al. 2008). These findings support the possibility that the serotonergic sensitization induced by MDMA and METH contributes to behavioral sensitization. However, the mechanisms underlying the development and expression of serotonergic sensitization are still unclear. Some agonists and/or antagonists for 5-HT receptor subtypes (5-HT1A and 5-HT2 receptors) might affect acute MDMA-induced 5-HT release and sustained MDMA-induced serotonergic sensitization. On the other hand, the serotonergic sensitization might be mediated by trans-synaptic function and caused by Ca2+-dependent release, rather than reverse transport-mediated release. We postulate that released 5-HT by MDMA may directly or indirectly activate glutamatergic neurons and/or suppress inhibitory GABAergic interneurons, which may lead to long-term increase in activity of the raphe serotonergic neurons. Further investigations are needed to elucidate the exact mechanisms.

In summary, we have established an in vitro model system, consisting of organotypic mesencephalic slice cultures, which is suitable for analysing the altered serotonergic activity observed following long-term exposure to drugs of abuse. Our results demonstrate that sustained exposure to MDMA or METH induced the augmentation of their 5-HT releasing effect, i.e., serotonergic sensitization, in the absence of any marked serotonergic neurotoxicity or facilitation of 5-HT biosynthesis. The development of serotonergic sensitization was mediated through activation of NMDA receptors. The in vitro model system, which reproduces serotonergic sensitization, could help to elucidate the molecular mechanisms underlying addiction to MDMA and METH, and serotonergic sensitization is thought to play, at least in part, a role in long-term alterations of psychological behavior following chronic treatment with drugs of abuse.

Acknowledgements

This work was supported in part by a Health Sciences Research Grant from the Ministry of Health, Labor, and Welfare and a Grant-in-Aid from the Ministry of Education, Culture, Sports, Science, and Technology of Japan and from Japan Society for the Promotion of Science. We thank Dr Tatsunori Iwamura (Gifu Pharmaceutical University) for providing MDMA.

Ancillary