Discriminative stimulus properties of the selective ... - Springer Link

Psychopharmacology (2001) 154:213–218 DOI 10.1007/s002130100895

R A P I D C O M M U N I C AT I O N

Anne Dekeyne · Alain Gobert · Loretta Iob Laetitia Cistarelli · Christophe Melon Mark J. Millan

Discriminative stimulus properties of the selective norepinephrine reuptake inhibitor, reboxetine, in rats Received: 15 March 2001 / Accepted: 21 July 2001 / Published online: 11 September 2001 © Springer-Verlag 2001

Abstract Rationale: Although drug discrimination procedures have proven difficult to apply to antidepressant agents, we recently characterized discriminative stimulus properties of the selective serotonin (5-HT) reuptake inhibitor, citalopram, in rats. However, discriminative stimulus properties of selective norepinephrine (NE) reuptake inhibitors remain to be evaluated. Objective: We determined the potential discriminative stimulus properties of the highly selective NE reuptake inhibitor and antidepressant, reboxetine. Methods: Employing a twolever discrimination procedure, rats were trained to discriminate reboxetine (2.5 mg/kg, IP) from saline. In parallel, the influence of reboxetine (2.5 mg/kg) upon dialysate levels of monoamines in frontal cortex and dorsal hippocampus of freely moving rats was determined. Results: After 54±10 training sessions, reboxetine elicited robust stimulus recognition, fully generalizing to itself with an ED50 of 1.2 mg/kg. Two further NE reuptake inhibitors, desipramine (5.3) and maprotiline (1.8), as well as the 5-HT/NE reuptake inhibitor, venlafaxine (1.0), likewise generalized. In contrast, the 5-HT reuptake inhibitors, paroxetine, citalopram and sertraline, and the DA reuptake inhibitors, GBR12935 and bupropion, did not show significant generalization. Reboxetine markedly increased dialysate levels of NE, but not 5-HT, in frontal cortex and hippocampus. Dopamine (DA) levels were also (though less markedly) enhanced in frontal cortex. Conclusion: In parallel with an elevation in extracellular levels of NE, the selective NE reuptake inhibitor, reboxetine, elicits a specific discriminative stimulus in rats.

A. Dekeyne (✉) · A. Gobert · L. Iob · L. Cistarelli · C. Melon M.J. Millan Institut de Recherches Servier, Centre de Recherches de Croissy, Psychopharmacology Department, 125 Chemin de Ronde, 78290 Croissy-sur-Seine, Paris, France e-mail: [email protected] Tel.: +33-1-55722438, Fax: +33-1-55722470

Keywords Drug discrimination · Dialysis · Norepinephrine transporter · Norepinephrine reuptake inhibitor · Depression · Antidepressant

Introduction While drug discrimination procedures have proven invaluable in the characterization of anxiolytics and antipsychotics, comparatively little information is available for antidepressant agents (Goudie and Leathley 1993). Generalized toxic properties of tricyclics, and their perturbation of behaviour, severely compromise their utilization (Schechter 1983; Zhang and Barrett 1991), while discriminative stimulus (DS) properties of the “atypical” antidepressant, mianserin, reflect blockade of muscarinic receptors, an action related to side-effects rather than therapeutic activity (Kelley et al. 1995). Nevertheless, it was recently demonstrated that the selective serotonin (5-HT) reuptake inhibitor (SSRI), citalopram, elicits a DS in rats (Marona-Lewicka and Nichols 1998; Millan et al. 1999a, 1999b). Notably, other SSRIs, such as paroxetine and sertraline, as well as the mixed 5-HT/ norepinephrine (NE) reuptake inhibitor, venlafaxine (Schweizer et al. 1997), potently and fully generalize to citalopram (Millan et al. 1999b). Like SSRIs, drugs preferentially interfering with neuronal reuptake of NE versus 5-HT and dopamine (DA), such as desipramine, possess antidepressant properties (Burke and Preskorn 1995; Owens et al. 1997), and it would similarly appear of importance to characterize their DS properties. However, data in this respect are limited. Although desipramine generated a DS in rats, general toxicity (training could only be completed for approximately half the population commenced) precluded further characterization (Shearman et al. 1978). Further, desipramine interacts with many sites other than NE transporters (Owens et al. 1997). With regard to a further NE reuptake inhibitor, nisoxetine, data are similarly limited inasmuch as cross-generalization was demonstrated with amphetamine but no generalization

214

studies were performed with additional NE reuptake inhibitors or with other antidepressant agents: moreover, this work was undertaken in mice (Snoddy and Tessel 1983, 1985). Finally, although it was shown that the tricyclic, NE reuptake inhibitor, (+)-oxaprotiline, generates a DS in rats, this drug, like desipramine, interacts with receptors other than NE reuptake sites and apart from generalization of desipramine and maprotiline, no further data have been reported (Filip et al. 1993). Recently, it was shown that the novel morpholine derivative and antidepressant, reboxetine, behaves as a highly selective and potent inhibitor of NE reuptake both in vitro and in vivo (Riva et al. 1989; Sacchetti et al. 1999; Versiani et al. 2000; Wong et al. 2000; Millan et al. 2001a, 2001b). It appears thus to be an ideal agent for a thorough characterization of DS properties associated with selective inhibition of NE reuptake. The purpose of the present studies was as follows: first, to determine whether reboxetine generates a DS in rats; second, to evaluate whether other NE reuptake inhibitors generalize to the reboxetine DS, as compared to drugs inhibiting 5-HT or DA reuptake; and third, in order to confirm the specificity of the reboxetine DS, to examine the actions of other classes of psychotropic agent: the neuroleptic, haloperidol, and the anxiolytic, diazepam. Finally, in an independent study, the influence of reboxetine upon extracellular levels of NE, 5-HT and DA in the frontal cortex (FCX) and dorsal hippocampus was evaluated. These structures were selected in view of their major implication in the actions of antidepressant agents and since, employing a highly sensitive coulometric detection technique, we have previously demonstrated that it is possible simultaneously to determine levels of 5-HT, NE and (for the FCX) DA in single dialysate samples of these structures (Gobert et al. 1998; Millan et al. 1999b, 2000, 2001b).

Materials and methods

forced with food according to a fixed ratio 10 schedule of reinforcement. Each 15-min daily session (5 days/week) started 15 min after injection. During reboxetine sessions, responses on only one lever were reinforced while, during saline sessions, responses on the other lever were reinforced. “Reboxetine” (R) or “saline” (S) sessions alternated as follows: RSSRS-SRRSSSRSRR-RSRSR-, etc. Correct responding was defined as no more than 13 presses on both levers to obtain the first reinforcement. The discrimination criterion was ten consecutive sessions with correct responding. Thereafter, generalization tests were conducted every Wednesday and Friday, whereas training sessions continued on the other days. Rats were tested only if they showed correct responding on the two preceding training sessions. During testing, responding on the selected lever, i.e. the lever for which ten (not necessarily consecutive) responses were recorded first, was reinforced for the remainder of the 15-min session. Test drugs were administered instead of reboxetine, 15 min before the session.

Dialysis studies The influence of reboxetine upon levels of NE versus 5-HT and DA in single dialysate samples of the FCX, and upon levels of NE versus 5-HT in the hippocampus (DA was undetectable in this structure), was determined as detailed previously (Millan et al. 1999b), employing HPLC plus coulometric detection in freelymoving rats implanted 1 week prior to testing with a guide cannula. Samples were taken every 20 min. Basal DA, NE and 5-HT levels were monitored over 1 h, then reboxetine (2.5 mg/kg, IP) injected, and samples taken for a further 3 h. Changes were expressed relative to basal values. Drugs Drug doses are in terms of the base. Drug discrimination training was undertaken by the IP route in order to avoid potential cutaneous toxicity. As in previous studies in this laboratory, test drugs were dissolved in sterile water and administered SC (to avoid hepatic first-pass metabolism), except maprotiline which (since it is poorly tolerated SC) was given in a suspension IP. Drug sources, salts and structures were as follows: bupropion HCl (Burroughs Wellcome Co., N.C., USA); GBR12935 diHCl (Research Biochemicals International, Natick, Mass., USA); desipramine HCl, haloperidol and maprotiline HCl (Sigma, Chesnes, France); diazepam (HoffmanLa Roche, Basel, Switzerland) and paroxetine HCl (Beecham Pharmaceuticals, Brentford, UK). Citalopram HBr, reboxetine methane sulfonate, sertraline HCl and venlafaxine HCl were synthesized by Servier chemists (G. Lavielle and J.-L. Péglion).

Animals Male Wistar rats (180–200 g body weight upon arrival for drug discrimination and 200–220 g for dialysis studies; Iffa-Credo, l’Arbresle, France) were housed individually in sawdust-lined standard polycarbonate cages with free access to water and, for drug discrimination studies, with restricted access to chow (10–11 g per day) in order to maintain their weight at 80% of freefeeding values. They were kept under a 12-h/12-h light-dark cycle with lights on at 0700 hours. Laboratory temperature was 21±1.0°C and humidity, 60±5%. All animal use procedures conformed to international European ethical standards (86/609-CEE) and the French National Committee (décret 87/848) for the care and use of laboratory animals. Drug discrimination procedure As described previously (Millan et al. 1999b), rats were trained to discriminate reboxetine (2.5 mg/kg, IP) from saline in operant conditioning chambers equipped with two levers. They were rein-

Data analysis In the drug discrimination study, to evaluate the influence of reboxetine upon training, the mean total number of responses (i.e. presses on both levers)±SEM was calculated for the five last “reboxetine” versus “saline” sessions before the first test session. Data recorded during a test session were lever selection and the total number of responses. Lever selection data were expressed as the percentage of rats selecting the drug lever and were compared by a Fisher Exact Probability Test to control values of the preceding saline training session (0% drug lever selection). Effective dose50s (ED50s) plus 95% confidence limits (95% CL) were calculated to estimate drug potency. For description of drug actions, “full” generalization corresponds to 80% or greater “reboxetine” lever selection. The total numbers of responses in the presence of drug were compared by a paired t-test to that obtained during the preceding saline training session. In the dialysis studies, data were analyzed by ANOVA with sampling time as the repeated withinsubject factor.

215 Table 1 Actions of drugs which failed to show significant generalization to reboxetine. SSRI selective serotonin reuptake inhibitors; DARI dopamine reuptake inhibitor; BZD benzodiazepine; n number of rats selecting a lever; N number of tested rats. The total numbers of responses were not significantly modified in any case (P>0.05; paired t-test), as compared to the preceding control (saline) session

Drug

Clas

Citalopram

SSRI

Paroxetine

SSRI

Sertraline

SSRI

GBR12935

DARI

Bupropion

DARI

Diazepam

BZD

Haloperidol

Neuroleptic

Dose (mg/kg, SC) 0.63 2.5 0.16 0.63 0.16 0.63 0.63 10.0 2.5 10.0 0.16 0.63 0.01 0.04 0.16

% Lever selection 40 33 40 0 40 20 33 20 20 20 0 20 0 0 –

Total number of Responses ± SEMs Control session

Test session

1987±246 1626±207 1872±213 1698±179 1712±149 2035±170 1637±170 1786±206 1861±203 1954±249 1725±135 1713±167 1812±177 1788±104 1857±105

1631±118 1656±175 1886±214 1681±165 1877±150 1960±182 1737±127 1824±226 1944±166 2078±125 1828±84 870±264 1671±184 1262±311 –

n/N

5/5 6/6 5/5 5/5 5/5 5/5 6/6 5/5 5/5 5/5 5/5 5/5 5/5 5/5 0/3

Fig. 1 Percent correct responding during acquisition of a discriminative stimulus with reboxetine (2.5 mg/kg, IP)

Results As shown in Fig. 1, acquisition of a DS with reboxetine (2.5 mg/kg, IP) was successful. The mean±SEM number of training sessions required to achieve the discrimination criterion was 54±10 (n=9). The total number of responses during training was slightly but significantly lower (P0.05 in Fisher Exact Probability test), and they did

Fig. 2 Dose-dependent generalization of reboxetine, desipramine, maprotiline and venlafaxine to the discriminative stimulus elicited by reboxetine. Upper panels: percentage of rats selecting the “reboxetine” lever and lower panel: means±SEMs of total number of responses. n=5–9 per value. In the upper panel, asterisks indicate significance of differences to control values (0% “reboxetine” lever selection) in Fisher’s Exact Probability test. *P