This paper is dedicated to the memory of Dr R. Samanin, who died on June 5, 2001.
Address correspondence and reprint requests to Dr R. W. Invernizzi, Istituto di Ricerche Farmacologiche ‘Mario Negri’, Via Eritrea 62, 20157 Milan, Italy. E-mail: email@example.com
The present study investigated whether 5-HT2C receptors in the ventrotegmental area and prefrontal cortex regulate basal and stimulus-evoked dopamine release in the prefrontal cortex. Using the in vivo microdialysis technique in conscious rats, we studied the effect of a selective 5-HT2C receptor agonist, Ro60–0175, on basal and immobilization stress-induced dopamine release in the prefrontal cortex. Ro60–0175 intraperitoneally (2.5 mg/kg) and into the ventrotegmental area (10 µg/0.5 µL) completely antagonized the effect of stress on extracellular dopamine without altering basal levels. Infusion of 10 µm Ro60–0175 through the cortical probe had no significant effect on basal and stress-induced dopamine release. SB242084 (10 mg/kg), a selective antagonist of 5-HT2C receptors, significantly increased basal extracellular dopamine and completely prevented the effect of intraperitoneal and intraventrotegmental Ro60–0175 on the stress-induced rise of extracellular dopamine, but had no effect itself in stressed rats. The results show that Ro60–0175 suppresses cortical dopamine release induced by immobilization stress through the stimulation of 5-HT2C receptors in the ventrotegmental area. While confirming that endogenous 5-HT acting on 5-HT2C receptors tonically inhibit basal dopamine release in the prefrontal cortex, the present findings suggest that the stimulation of 5-HT2C receptors with an exogenous agonist preferentially inhibit stimulated release.
In previous studies on cortical DA release selective 5-HT2C receptor agonists and antagonists were administered systemically (Millan et al. 1998; Gobert et al. 2000), and consequently the part that 5-HT2C receptors of the VTA and other brain regions play in controlling DA transmission is not clear.
In the present study we examined the roles of 5-HT2C receptors of the VTA and PFC in controlling cortical DA transmission by studying the effect of a selective 5-HT2C receptor agonist, Ro60–0175 (Martin et al. 1998), on DA release after its application to the VTA or PFC.
Male Sprague–Dawley rats (300–350 g; CD-COBS; Charles River Como, Italy) were used. They were housed at constant temperature (21 ± 1°C) and relative humidity (60 ± 5%%) under a regular light–dark schedule (light 07:00–19:00 h) with food and water freely available.
Procedures involving animals and their care were conducted in conformity with the institutional guidelines that are in compliance with national (DLn 116, GU, suppl. 40, 18 February 1992; Circolare No. 8, GU, 14 July 1994) and international laws and policies (EEC Council Directive 86/609, OJ L 358,1, December 12, 1987; Guide for the Care & Use of Laboratory Animals, US National Research Council 1996).
Rats were anaesthetized with 3.5 mL/kg equithesin intraperitoneally (i.p.) (composition: 1.2 g sodium pentobarbital, 5.3 g chloral hydrate, 2.7 g MgSO4, 49.5 mL propylene glycol, 12.5 mL ethanol and 58 mL distilled water) and placed on a stereotaxic apparatus (mod. 900; David Kopf, Tujunga, CA, USA). A dialysis probe of the concentric type, made of a copolymer of acrylonitrile-sodium methallyl sulphonate (AN 69, Hospal SpA; 0.31-mm outer diameter, with more than 44 kDa molecular weight cut-off), 4-mm long, was lowered slowly into the rat PFC at the following stereotaxic coordinates: AP = +4.3 mm, L = ±0.7 mm and V = −4.8 mm from the bregma and dura surface with the incisor bar set at − 3.3 mm according to the atlas of Paxinos and Watson (1982). Twenty hours after probe implantation each rat was placed in a Plexiglas cage where the inlet cannula was connected by polythene tubing (Portex Ltd, Hythe, UK) to a 2.5-mL syringe mounted on a CMA/100 microinjection pump (CMA Microdialysis, Stockholm, Sweden) containing artificial cerebrospinal fluid (aCSF; composition in mm NaCl 145, KCl 3, CaCl2 1.26, MgCl2 1, buffered at pH 7.4 with 2 mm sodium phosphate buffer). The probe was perfused with aCSF at a constant flow-rate of 2 µL/min. After 30 min washout, perfusate was collected every 20 min and immediately assayed by HPLC with electrochemical detection as previously described (Invernizzi et al. 1992).
A cannula made of 23-gauge stainless steel tubing, 14-mm long, was implanted 2 mm above the VTA ipsilateral to the probe, at the following stereotaxic coordinates: AP = +3.7 mm, L = ±0.8 mm and V = +4.0 mm from the interaural line with the incisor bar set at −3.3 mm (Paxinos and Watson 1982). Drug or vehicle was delivered to the VTA at a rate of 0.5 µL/min by a 10-µL Hamilton syringe and a CMA/100 infusion pump connected by polythene tubing to the injection unit (30-gauge stainless steel tubing) terminating 2 mm below the tip of the cannula. The injection cannula was left in place for another 1 min before withdrawal to allow diffusion from the tip and prevent reflux of the solution.
The immobilization procedure involves the following steps. The rats were taken from the microdialysis cage and their limbs inserted into four holes made in a plastic board at distances corresponding to the limbs. They were then immobilized by taping their limbs under the board, which was taped to the edge of the microdialysis cage. The fixation procedure was carried out within 2 min and immobilization was applied for 2 h. Extracellular DA was measured 1 h before, during and for 1 h after immobilization was terminated. Samples of dialysate were collected at 20-min intervals.
Animals were deeply anaesthetized with chloral hydrate and decapitated. Their brains were removed, frozen and sliced in 40-µm sections for examination of probe and cannula tracks. Data were included in the results if no signs of non-specific tissue damage were seen and the probes and cannula were positioned in the region between ±0.5 mm AP, ±0.3 mm V and ±0.3 mm L for the prefrontal cortex and between ±0.3 mm AP, V and L for the VTA.
Drugs were injected once the extracellular DA concentrations were stable (at least three consecutive samples differing by less than 15% from the mean basal value). Ro60–0175 [(S)-2-(chloro-5-fluoroindol-1-yl)-methylethylamine)fumarate; Hoffman–La Roche, Basel, Switzerland] or vehicle was injected i.p. (2.5 mg/kg), into the VTA (1 and 10 µg/0.5 µL) or perfused through the probe in the PFC (0.1, 1 and 10 µm) of unstressed rats.
Stressed rats received Ro60–0175 i.p. (2.5 mg/kg) or into the VTA (10 µg/0.5 µL) 20 min before immobilization. The effect of Ro60–0175 on the stress-induced rise of extracellular DA in the PFC was also studied in rats continuously perfused with the drug in the PFC during immobilization.
To assess the specificity of Ro60–0175 for 5-HT2C receptors, one group of rats was pretreated with 10 mg/kg i.p. SB242084 [(6-chloro-5-methyl-1-[2-(2-methylpyridyl-3-oxy)-pyrid-5-ylcarbamoyl] indoline) SmithKline Beecham, Harlow, UK] 5 min before the i.p. or intra-VTA injection of Ro60–0175 or vehicle.
Vehicle consisted of aCSF solution containing 50% dimethylsulfoxide (DMSO) and 1% 4 m tartaric acid. Ro60–0175 and SB242084 were dissolved in vehicle and the pH of the solutions was adjusted to > 5.0 with minimal quantities of 5 m NaOH. Control rats received an equal volume of vehicle (adjusted to pH > 5.0) i.p. or in the VTA. When Ro60–0175 was infused through the dialysis probe, the compound was dissolved as described to obtain a 1 mm solution, and further diluted with aCSF to obtain 0.1, 1.0 and 10 µm solutions.
Data (expressed as fmol/40 µL) were analysed by mnova for repeated measures (split-plot) with treatment as the between-subject factor and time as the within-subject factor. Post-hoc comparisons were made by Tukey–Kramer's test. Statistical analysis was performed using the SmnovaVmnova 5.0 (SAS Institute Inc., Cary, NC, USA) statistical package for a Macintosh computer.
The mean basal extracellular concentration of DA (not corrected for probe recovery) in the prefrontal cortex was 13.8 ± 0.6 fmol/40 µL (n = 74). No significant differences in basal extracellular DA were found between the different groups of rats (F4,28 = 1.4, p > 0.05).
Effect of Ro60–0175 and SB242084 on basal extracellular DA in the prefrontal cortex
The injection of vehicle i.p. or into the VTA had no significant effect on basal extracellular DA in the prefrontal cortex of unstressed rats (Fig. 1). The extracellular concentrations of DA in rats given 2.5 mg/kg Ro60–0175 dissolved in aCSF/DMSO/tartaric acid or phosphate-buffered saline were not significantly different from vehicle-treated rats (F2,8 = 0.3, p > 0.05; Fig. 1). Similarly, intra-VTA injection of 10 µg/0.5 µL Ro60–0175 had no effect on basal extracellular DA (F1,6 = 0.04, p > 0.05; Fig. 1).
The infusion of Ro60–0175 (0.1–10 µmnova) through the cortical probe had no significant effect on basal extracellular DA in the PFC of unstressed rats (F3,6 = 0.7, p > 0.05; Fig. 1). SB242084 (10 mg/kg) significantly increased extracellular DA (F3,6 = 4.1, p < 0.01; Fig. 1), reaching 186% of basal values at 40 min. The effect was significant from 40 to 100 min after injection.
Effect of systemic Ro60–0175 on the immobilization-induced increase of extracellular DA in the PFC: antagonism by SB242084
Immobilization stress significantly increased extracellular DA, by about 250%, in the PFC of rats given vehicle i.p. (Fig. 2). The effect was maximal during the first hour and DA returned to baseline in about 100 min, although the rats were still immobilized. A significant increase of extracellular DA was also observed 20–60 min after the end of immobilization. Ro60–0175 (2.5 mg/kg) completely prevented the rise in extracellular DA induced by stress and that observed after the rats were released (F3,17 = 4.4, p < 0.05; Fig 2). Consistent with the effect of SB242084 on extracellular DA in unstressed rats (see Fig. 1), 10 mg/kg SB242084 tended to raise extracellular DA 20 min after injection (Figs 2 and 4) but the effect was not significant. SB242084 (10 mg/kg), which had no significant effect by itself on extracellular DA in stressed rats, completely antagonized the inhibitory effect of 2.5 mg/kg Ro60–0175 on the stress-induced rise (F30,170 = 2.0, p < 0.01; Fig. 2).
Effect of intra-VTA and intracortical infusion of Ro60–0175 on the immobilization-induced increase of extracellular DA in the PFC
Stress significantly increased extracellular DA (205–210% of basal levels, at peak) from 20 to 80 min after the onset of immobilization in rats injected with 0.5 µL vehicle into the VTA. A significant increase of extracellular DA was also seen 20 and 40 min after the rats were set free (Fig. 3). Ro60–0175 (1 and 10 µg/0.5 µL) dose-dependently inhibited the effect of immobilization on cortical DA release throughout the period of immobilization and after the rats were released (F2,15 = 7.2, p < 0.01; Fig. 3). Post-hoc analysis showed that only the higher dose significantly antagonized the immobilization-induced rise of extracellular DA.
The inhibitory effect of 10 µg/0.5 µL Ro60–0175 into the VTA on the immobilization-induced rise of extracellular DA was completely prevented by 10 mg/kg SB242084 (F30,170 = 2.5, p < 0.01; Fig. 4). In contrast, 10 µmnova Ro60–0175 through the cortical probe had no effect on the increase of extracellular DA caused by stress (F9,45 = 1.8, p > 0.05; Fig. 5).
Stimulation of 5-HT2C receptors by the selective agonist Ro60–0175 completely prevented the stress-induced increase of extracellular DA without affecting basal extracellular concentrations, suggesting that the control exerted by the 5-HT2C receptor agonist on cortical DA release appears to be dependent on the activity of mesocortical dopaminergic neurones.
In contrast with the present results, a previous study found that 2.5 mg/kg Ro60–0175 reduced basal extracellular DA in the PFC (Gobert et al. 2000). The reasons for the different effects of Ro60–0175 in the two studies are not clear. The use of different vehicles (aCSF containing DMSO and tartaric acid, or saline) is probably not important because the failure of 2.5 mg/kg Ro60–0175 to reduce extracellular DA in the PFC was confirmed using saline as vehicle. The discrepancy may be attributable to other methodological factors perhaps reflecting the different ionic composition of the perfusion solution (Moghaddam and Bunney 1989), time between probe insertion and perfusion (Westerink and De Vries 1988), rat strain or route of administration of Ro60–0175.
The finding that intra-VTA and intracortical Ro60–0175 had no effect on extracellular DA confirms the results with systemic Ro60–0175 and indicates that stimulating 5-HT2C receptors in these two regions has no effect on cortical DA release under basal conditions.
Previous studies have shown that the effect of several serotonergic agonists and antagonists on DA transmission may be different under resting conditions or in response to stressful and pharmacological stimuli. Specifically, stimulation of 5-HT1A receptors with the selective receptor agonist 8-hydroxy-2-(di-n-propylamino)tetralin increased basal but reduced the footshock-induced increase in DA metabolism in the PFC (Rasmusson et al. 1994). Similarly, 5-HT3 and 5-HT4 receptor antagonists reduced, respectively, morphine-stimulated DA release in the nucleus accumbens (Imperato and Angelucci 1989; Pozzi et al. 1995) and haloperidol-induced increase of DA release in the striatum (Lucas et al. 2001), but had no effect on basal release.
The present results show that the stimulation of 5-HT2C receptors had no effect on basal extracellular DA in the PFC, whereas blockade of 5-HT2C receptors increased extracellular DA. Although confirming that endogenous 5-HT tone on 5-HT2C receptors maintains low cortical extracellular DA concentrations in basal conditions, these findings do not support the inhibitory effect of Ro60–0175 on basal DA release observed by Gobert et al. (2000). They do in fact show an inhibitory effect on stress-induced release of DA in the PFC, suggesting that the inhibitory tone exerted by 5-HT2C receptors on basal DA release might be reduced during stress. Acute stress increases extracellular 5-HT in different brain regions (Rueter et al. 1997) and previous studies have shown that desensitization of 5-HT2C receptors occurs rapidly in vitro in response to 5-HT exposure (Berg et al. 2001), and occurs in vivo after a sustained increase in serotonergic tone (Kennett et al. 1994).
We found that extracellular DA in the PFC was increased to a similar extent by stress regardless of whether 5-HT2C receptors were blocked by systemic SB242084. The maximal increase and duration of the effect of stress on extracellular DA (including the rebound increase of extracellular DA after the rats were set free) were similar in rats given vehicle or SB242084 as pretreatment.
The failure of SB242084 to enhance stress-induced DA release in the PFC seems surprising in view of the fact that 5-HT2C receptor antagonists enhanced the effect of phencyclidine (Hutson et al. 2000) and low doses of haloperidol (Lucas et al. 2000) on extracellular DA in subcortical regions. A short duration of action is unlikely to account for the failure of SB242084 to enhance stress-induced DA release in the PFC because SB242084 antagonized the inhibitory effect of Ro60–0175 on the stress-induced increase of extracellular DA over the entire observation period. Consistent with this, a previous study showed that SB242084 antagonized m-CPP-induced hypolocomotion for at least 8 h (see Discussion in Kennett et al. 1997).
That the increase of extracellular DA induced by stress is not additive with the increase of extracellular DA produced by SB242084 suggests that similar mechanisms may be involved in the two effects. SB242084 and acute stress increase the firing activity of DA neurones of the VTA (Mantz et al. 1989; Di Matteo et al. 1999; Gobert et al. 2000), and recent studies suggest that both acute stress and 5-HT2C receptors preferentially regulate impulse-flow dependent DA release (Keefe et al. 1993; Willins and Meltzer 1998). Thus, it may be speculated that stress activates the impulse-flow dependent release of DA in the PFC, and that the addition of SB242084 does not activate it further (or the additional activation is too little to alter extracellular DA as assessed by microdialysis). In support of this, a similar mechanism has already been shown to account for the failure of SB206553, a 5-HT2C/2B receptor antagonist (Kennett et al. 1996), to enhance the increase of extracellular DA induced by 0.1 and 1 mg/kg haloperidol, a condition in which impulse-flow dependent DA release is maximally stimulated (Lucas et al. 2000).
5-HT2C receptors are present in various regions of the brain including the VTA and PFC (Pompeiano et al. 1994). Thus, systemic administration of 5-HT2C receptor agonists or antagonists, as in most previous studies (Di Matteo et al. 1998; Millan et al. 1998; Di Giovanni et al. 1999; Gobert et al. 2000), does not clarify the neurochemical location of the relevant receptor. We found that direct application of 10 µg/0.5 µL Ro60–0175 into the VTA prevented the stress-induced increase of extracellular DA in the PFC, whereas infusion of Ro60–0175 (10 µmnova) through the cortical probe had no such effect. Thus, 5-HT2C receptors of the VTA play a major role in the inhibitory effect of Ro60–0175 on stress-induced DA release in the PFC. The fact that the effect of Ro60–0175 on the stress-induced increase of extracellular DA in the PFC was completely reversed by the selective 5-HT2C receptor antagonist SB242084 (Kennett et al. 1997) confirms an action of Ro60–0175 on the 5-HT2C receptor rather than non-specific drug effects.
Double labelling in situ hybridization studies found that the mRNA for 5-HT2C receptors was not expressed on dopaminergic neurones of the substantia nigra and VTA (Eberle-Wang et al. 1997). This suggests that Ro60–0175 has an indirect effect on the stress-induced increase of extracellular DA. Stimulation of 5-HT2C receptors excites GABAergic neurones of the VTA (Di Giovanni et al. 2001). In addition, the increase of cortical DA release evoked by acute handling stress was inhibited by infusion of the GABAB receptor agonist baclofen into the VTA (Enrico et al. 1998). These findings therefore suggest that the inhibitory effect of Ro60–0175 on stress-induced cortical DA release may be mediated by the excitation of GABAergic neurones, which in turn inhibit DA cells of the VTA.
In conclusion, this study found that Ro60–0175 suppresses cortical DA release evoked by immobilization stress, at least partly through the stimulation of 5-HT2C receptors in the VTA. In addition, the results confirm that the stimulation of 5-HT2C receptors by endogenous 5-HT tonically inhibits basal DA release in the PFC whereas activation of 5-HT2C receptors by an exogenous agonist preferentially inhibits stimulated DA release.
Stress-induced increases of DA release and turnover in the PFC have been associated with cognitive deficits (Arnsten and Goldman-Rakic 1998), which can be relieved by D1 receptor antagonists and drugs preventing the rise in DA turnover (Murphy et al. 1996a,b; Zarht et al. 1997). Thus, it would be interesting to examine whether Ro60–0175 helps counteract cognitive deficits induced by the activation of DA transmission in the PFC.
We thank Dr J.R. Martin from Hoffman-La Roche, Basel, Switzerland, and Dr T. Blackburn and Dr M. Wood from SKB, Harlow, UK for the generous gifts of Ro60–0175 and SB242084, M. Carli for helpful comments on the manuscript and J. Baggott for linguistic assistance.