Effects of GABAB receptor antagonist, agonists and allosteric positive modulator on the cocaine-induced self-administration and drug discrimination

Available online at www.sciencedirect.com

European Journal of Pharmacology 574 (2007) 148 – 157 www.elsevier.com/locate/ejphar

Effects of GABAB receptor antagonist, agonists and allosteric positive modulator on the cocaine-induced self-administration and drug discrimination Małgorzata Filip ⁎, Małgorzata Frankowska, Edmund Przegaliński Laboratory of Drug Addiction Pharmacology, Department of Pharmacology, Institute of Pharmacology, Polish Academy of Sciences, Kraków, Poland Received 2 March 2007; received in revised form 23 July 2007; accepted 23 July 2007 Available online 31 July 2007

Abstract Preclinical and clinical findings indicate that a GABAB receptor agonist baclofen decreases cocaine use. The present study investigated the effects of the GABAB receptor antagonist (2S)-(+)-5,5-dimethyl-2-morpholineacetic acid (SCH 50911), the agonists baclofen and 3-aminopropyl (methyl)phoshinic acid (SKF 97541) and the allosteric positive modulator 3,5-bis(1,1-dimethylethyl-4-hydroxy-β,β-dimethylbenzenepropanol (CGP 7930) in cocaine-and food-maintained responding under a fixed ratio 5 schedule of reinforcement in male Wistar rats. The effects of the GABAB receptor ligands on cocaine (10 mg/kg)-induced discriminative stimulus in a two-lever, water-reinforced fixed ratio 20 task and on basal locomotor activity were also assessed. Baclofen (2.5–5 mg/kg), SKF 97541 (0.1–0.3 mg/kg) and CGP 7930 (30–100 mg/kg) decreased the cocaine (0.5 mg/kg/injection)-maintained responding; SCH 50911 (3–10 mg/kg) was inactive in this respect. Baclofen (5 mg/kg) and SKF 97541 (0.3 mg/kg), but not CGP 7930 or SCH 50911 attenuated the food-maintained responding. The inhibitory effects of the GABAB receptor agonists and the modulator were blocked by SCH 50911. SKF 97541 (0.1 mg/kg) or CGP 9730 (30–100 mg/kg) did not produce a significant shift in the cocaine (1.25–10 mg/kg) dose–response curve in a drug discrimination procedure, while baclofen (1.5 mg/kg) or SCH 50911 (10 mg/kg) attenuated the effects of separate doses of cocaine. Baclofen (5 mg/kg) and CGP 7930 (100 mg/kg) significantly reduced basal horizontal activity. We found that pharmacological stimulation of GABAB receptors by direct agonists or allosteric positive modulation reduces cocaine reinforcement while this property of cocaine is not related to tonic activation of GABAB receptors. The GABAB receptor stimulation-induced reduction of cocaine reinforcement was separated from its discriminative stimulus effects. Moreover, a dissociation between effects of direct GABAB receptor agonists and a GABAB allosteric positive modulator on cocaine vs. food-maintained responding was demonstrated. © 2007 Elsevier B.V. All rights reserved. Keywords: GABAB receptors; Cocaine discrimination; Cocaine self-administration; Food self-administration; Locomotor activity; (Rats)

1. Introduction The major inhibitory neurotransmitter in the central nervous system, γ-aminobutyric acid (GABA), activates three classes of receptors: ligand-gated chloride channels GABAA and GABAC receptors, and G-protein-coupled metabotropic GABAB receptors. Recent attention has been given to GABAB receptors since

⁎ Corresponding author. Laboratory of Drug Addiction Pharmacology, Department of Pharmacology, Institute of Pharmacology, Polish Academy of Sciences, 31-343 Kraków, 12 Smętna, Poland. Tel.: +48 12 6623293; fax: +48 12 6374500. E-mail address: [email protected] (M. Filip). 0014-2999/$ - see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.ejphar.2007.07.048

drugs that activate those receptors maybe useful as putative “anti-additive” therapies. The functional GABAB receptors are formation of heterodimeric assemblies between GABAB(1) and GABAB(2) proteins (e.g. Jones et al., 1998; Kaupmann et al., 1998). It has been proposed two binding sites within GABAB receptors, both agonist and antagonist sites that differ in binding affinity (nanomolar and micromolar, respectively). GABAB receptors are directly activated by GABA, (R)-(−)-baclofen, or their derivatives. By presynaptic increase in potassium and decrease in calcium conductances, and by the adenylate cyclase inhibition in postsynaptic systems, these agonists inhibit neurotransmitters release and neuronal excitability, respectively. Antagonists of GABAB receptors being analogues of GABA interact with receptors located either presynaptically, to enhance

M. Filip et al. / European Journal of Pharmacology 574 (2007) 148–157

the release of several neurotransmitters, or postsynaptically to block late inhibitory postsynaptic potentials. Apart from direct regulation, an allosteric modulation of GABAB receptors has been recently demonstrated (Galvez et al. 2000; Urwyler et al. 2001). Such allosteric positive modulators have no intrinsic efficacy on their own and act only in the presence of the endogenous agonist GABA (Urwyler et al. 2001). As mentioned above, growing evidence suggests that GABAB receptor agonists could be promising pharmacotherapies for cocaine addiction. In fact, baclofen, a GABAB receptor agonist licensed as an anti-spastic drug, has yielded positive findings in clinical trials where it reduced cocaine use in heavy cocaine addicts (Shoptaw et al., 2003) and decreased limbic activation during cue-induced cocaine craving (Brebner et al., 2002). In contrast, baclofen does not alter subjective (i.e. “high”, “stimulated”) effects of cocaine (Haney et al., 2006). In laboratory animals, acute pretreatment with baclofen decreases the acquisition (Campbell et al., 1999), maintenance (Barrett et al., 2005; Brebner et al., 2000a; Di Ciano and Everitt, 2003; Roberts et al., 1996) and reinstatement of cocaine-reinforced responding (Campbell et al., 1999; Di Ciano and Everitt, 2003) as well as attenuates cocaine lowering brain reward thresholds, using the intracranial self-stimulation paradigm (Slattery et al., 2005). The inhibitory properties of baclofen on cocainemaintained responding have also been noted following repeated (3 days) administrated (Shoaib et al., 1998). Another GABAB receptor agonist 3-amino-2[S]-hydroxypropyl)-methylphosphinic acid (CGP 44532) and GABA B positive allosteric modulators also inhibits the maintenance of cocaine selfadministration in rats (Brebner et al., 1999; Smith et al., 2004) and baboons (Weerts et al., 2005). Interestingly, recent studies raise the question whether the GABAB receptor agonist-induced decrease in cocaine self-injection might be a pharmacologically non-specific effect since they also induced the depression of food-maintained responding in rats (Barrett et al., 2005; Munzar et al., 2000) and baboons (Weerts et al., 2005). However, other studies in rats separate the lower vs. higher doses of GABAB receptor agonists to reduce responding for cocaine and for food, respectively (Brebner et al., 1999; Roberts and Andrews, 1997; Shoaib et al., 1998). Similarly to clinical reports in which baclofen did not alter the subjective effects of cocaine, baclofen does not produce changes in discriminative stimulus effects of cocaine in rats (Barrett et al., 2005; Munzar et al., 2000) or rhesus monkeys (Negus et al., 2000) trained in a two-lever, food-reinforced task. To extend knowledge on the GABAB receptor ligands and modulators in the behavioral effects of cocaine and support the hypothesis that direct or indirect GABAB receptor stimulation may alter the behavioral effects of cocaine, in the current study we investigated the effects of the new selective and highly potent GABAB receptor antagonist (2S)-(+)-5,5-dimethyl-2morpholineacetic acid (SCH 50911; Froestl et al., 1995b) and the agonist 3-aminopropyl(methyl)phoshinic acid (SKF 97541; Froestl et al., 1995a) and as well as the GABAB receptor allosteric positive modulator 3,5-bis(1,1-dimethylethyl-4-hydroxy-β,β-dimethylbenzenepropanol (CGP 7930; Urwyler et al., 2001) in the self-administration and drug discrimination

149

procedures. Baclofen (a GABAB receptor agonist) was also used as a positive control. The above drugs were administered during the maintenance of cocaine or food self-administration or in substitution/combination studies in a cocaine discrimination paradigm. Finally, we also examined the effects of the above GABAB receptor agents on spontaneous locomotor activity. 2. Materials and methods 2.1. Animals Male Wistar rats (280–300 g) delivered by a licensed breeder (T. Górzkowska, Warsaw, Poland) were housed individually (self-administration procedures), or 2/cage (drug discrimination), or 8/cage (locomotor activity studies) in standard plastic rodent cages in a colony room maintained at 20 ± 1 °C and at 40–50% humidity under a 12-h light-dark cycle (lights on at 06:00). Animals had free access to food (Labofeed pellets) and water during the 7-day habituation period. Then, rats used in locomotor activity studies had free access to water, while those used in the cocaine self-administration procedures were maintained on limited water during initial training sessions (see below), or the amount of water that an animal received was restricted to that given during daily training sessions (5–6 ml/ rat/session), after test sessions (15 min) and on weekends (36 h) in drug discrimination procedures (see below) while animals used in the food self-administration procedures were maintained on limited food intake (see below). All experiments were conducted during the light phase of the light-dark cycle (between 08:00–15:00) and were carried out in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and with approval of the Bioethics Commission as compliant with the Polish Law (21 August 1997). The animals were experimentally naïve. 2.2. Drugs 3-Aminopropyl(methyl)phoshinic acid (SKF 97541; Tocris, UK), (R)-baclofen (Tocris, UK), cocaine hydrochloride (National Institute on Drug Abuse, RTI International, USA), (2S)(+)-5,5-dimethyl-2-morpholineacetic acid (SCH 50911; Tocris, UK) and 3,5-bis(1,1-dimethylethyl-4-hydroxy-β,β-dimethylbenzenepropanol (CGP 7930; Tocris, UK) were used. Baclofen, cocaine, SCH 50911 and SKF 97541 were dissolved in sterile 0.9% NaCl, while CGP 7930 was diluted in 2 drops (20 μl/drop) of ethanol and then in 1% Tween (Sigma Aldrich, USA). Cocaine was given either i.v. (0.05 ml/injection) or i.p. (1 ml/ kg); other drugs were injected i.p. in a volume of 1 ml/kg, except from 100 mg/kg of CGP 7930 which was administered in a volume of 3 ml/kg. Baclofen, CGP 7930, SCH 50911 or SKF 97541 were administered 30, 30, 45 or 30 min, respectively, before cocaine or saline. The dose-range and pretreatment intervals of the GABAB receptor ligands were chosen based on their functional in vivo activity at GABAB receptors (Carai et al., 2004a,b; Froestl et al., 1995a,b; Slattery et al., 2005; Smith et al., 2004).

150

M. Filip et al. / European Journal of Pharmacology 574 (2007) 148–157

2.3. Cocaine self-administration Rats were trained to press the lever of a standard operant conditioning chambers (Med-Associates, USA) under a fixed ratio 5 schedule of water reinforcement. Two days following lever-press training and free access to water, the rats were chronically implanted with a silastic catheter in the external right jugular vein, as described previously (Filip, 2005). Catheters were flushed every day with 0.1 ml of saline solution containing heparin (70 U/ml, Biochemie, Austria) and 0.1 ml of solution of cephazolin (10 mg/ml; Biochemie GmbH, Austria), Catheter patency was tested periodically with the ultrashortacting barbiturate anesthetic methohexital (10 mg/kg, i.v.; loss of consciousness within 5 s). After a 10-day recovery period, all animals were water deprived for 18 h and trained to lever press to fixed ratio 5 schedule of water reinforcement over a 2-h session. Then, subjects were given access to cocaine during 2-h daily sessions performed 6 days/week (maintenance) and from that time they were given ad libitum water. The house light was illuminated throughout each session. Each completion of five presses on the “active” lever complex (fixed ratio 5 schedule) resulted in a 5-s infusion of cocaine (0.5 mg/kg per 0.1 ml) and a 5-s presentation of a stimulus complex (activation of the white stimulus light directly above the “active” lever and the tone generator, 2000 Hz; 15 dB above ambient noise levels). Following each injection, there was a 20-s time-out period during which responding was recorded but had no programmed consequences. Response on the “inactive” lever never resulted in cocaine delivery. Acquisition of the conditioned operant response lasted a minimum of 10 days until subjects met the following criteria: minimum requirement of 30 reinforcements with an average of 6 days and active lever presses with an average of 6 consecutive days and a standard deviation within those 6 days of b10% of the average; this criterion was selected based on our prior experiments (Filip, 2005). Once stable rates of responding were established, the separate groups of rats (n = 7–8 rats/group) were pretreated with either vehicle, baclofen (1.25–5 mg/kg), CGP 7930 (10– 100 mg/kg), SCH 50911 (3–10 mg/kg), or SKF 97451 (0.03– 0.3 mg/kg) before the test sessions. Additionally, to test if the inhibitory effects of baclofen (2.5–5 mg/kg), CGP 7930 (30 mg/kg), or SKF 97541 (0.1–0.3 mg/kg) on cocaine (0.5 mg/kg)-maintained behavior were dependent on GABAB receptors, separate groups of animals (n = 6–7 rats/group) received either vehicle or a fixed dose of SCH 50911 (10 mg/ kg) before the above GABAB receptor agonists or the allosteric positive modulator. The order of injections was counterbalanced according to a Latin square design, and test sessions were separated by at least two-three baseline days of a dose unit cocaine self-administration. 2.4. Food self-administration Food self-administration was conducted in a manner parallel to cocaine self-administration. Food-restricted rats (20 g/rat/ day) were trained to press the lever of a standard operant

chambers (Med-Associates, USA) under a fixed ratio 5 schedule of reinforcement (each completion of a fixed ratio 5 schedule on the “active” lever resulted in a delivery of the portion of sweetened milk (0.1 ml)) in daily 2-h sessions. Following each reward, there was a 20-s time-out period during which responding was recorded but had no programmed consequences. Response on the “inactive” lever never resulted in food delivery. The house light was on throughout each session. Rats remained in maintenance training until lever pressing stabilized (the number of active lever presses varied by 10% or less over the course of 3 consecutive maintenance days). Once stable rates of responding were established, the separate groups of rats (n = 7 rats/group) were pretreated with either vehicle, baclofen (1.25–5 mg/kg), CGP 7930 (10–100 mg/kg), SCH 50911 (3– 10 mg/kg) or SKF 97451 (0.03–0.3 mg/kg) before the test sessions. Additionally, to test if the inhibitory effects of baclofen (5 mg/kg), or SKF 97541 (0.3 mg/kg) on foodmaintained behavior were dependent on GABAB receptors, separate groups of animals (n = 6 rats/group) received either vehicle or a fixed dose of SCH 50911 (10 mg/kg) before the above GABAB receptor agonists or the allosteric positive modulator. The order of injections was counterbalanced according to a Latin square design, and test sessions were separated by at least two-three baseline days of food selfadministration. 2.5. Cocaine discrimination Rats with restricted access to water during the daily training sessions (5–6 ml/rat/session), after test sessions (15 min), and over the weekends were trained to discriminate cocaine (10 mg/ kg; i.p.) from 0.9% NaCl (i.p.), as described previously (Filip et al., 2006). Briefly, cocaine or saline was administered 15 min before daily (Monday-Friday) sessions (15 min), in two-lever standard operant conditioning chambers equipped with a waterfilled dispenser mounted equidistant between two response levers on one wall (Med-Associates; USA), under a fixed ratio 20 schedule of continuous water reinforcement (0.1 ml per each completed fixed ratio 20) and depending on the treatment left or right lever became active, until animals met the criteria (an individual mean accuracy of at least 80% of correct responses, before the first reinforcer during 10 consecutive sessions). Test sessions were conducted once or twice a week to separate groups of rats tested with a different GABAB receptor ligand (n = 7–8 rats/group). Cocaine and saline sessions intervened between the test sessions to maintain discrimination accuracy. Only the rats that met 80% performance criterion during the preceding cocaine and saline sessions were used in the tests. After completion of 20 responses to either lever, or after the session time elapsed, a single reinforcer was delivered and the animals were removed from the chamber. In substitution tests, rats were tested with different doses of cocaine (1.25–10 mg/ kg), baclofen (1.25–5 mg/kg), CGP 7930 (30–100 mg/kg), SCH 50911 (10 mg/kg), or SKF 97541 (0.1–0.3 mg/kg). In combination tests, a fixed dose of a GABAB receptor ligand or the allosteric positive modulator was given before different doses of cocaine.

M. Filip et al. / European Journal of Pharmacology 574 (2007) 148–157

151

cages surrounded with a 15 × 15 array of photocell beams located 3 cm from the floor surface (Columbus Instruments, Columbus, USA). Interruptions of these photobeams resulted in horizontal activity defined as distance traveled and expressed in cm. Separate records of horizontal activity were made by the control software (Columbus Instruments) for subsequent statistical evaluation. Before locomotor activity was recorded, rats were injected with baclofen (1.25–5 mg/kg), CGP 7930 (10–100 mg/kg), SCH 50911 (3–10 mg/kg), or SKF 97451 (0.03–0.3 mg/kg) in their home cages and at the appropriate time were transferred to the experimental cages. Locomotor activity recording started immediately following rat placement and lasted 2 h. N = 7–8 rats/group. 2.7. Statistical analyses

Fig. 1. Effects of baclofen, SKF 97541, CGP 7930 and SCH 50911 on cocaine selfadministration in rats responding under a fixed ratio 5 schedule of reinforcement. Number of the active (cross bars), inactive (white bars) lever presses during cocaine self-administration (mean of the last 3 maintenance days ± S.E.M.). Selfadministration active lever responses resulted in a delivery of a cocaine infusion (0.5 mg/kg per injection) and simultaneous presentation of a light+ tone stimulus complex. Baclofen (BAC; 1.25–5 mg/kg), SKF 97541 (SKF; 0.03–0.3 mg/kg), CGP 7930 (CGP; 10–100 mg/kg) or SCH 50911 (SCH; 3–10 mg/kg) were given to 4 independent groups of rats. ⁎P b 0.05, ⁎⁎P b 0.01, ⁎⁎⁎P b 0.001 vs. corresponding vehicle (VEH) group.

2.6. Locomotor activity The spontaneous locomotor activity of rats was recorded for each animal as described previously (Przegaliński et al., 2001). Briefly, the locomotor activity was measured in Opto-Varimex

During maintenance of cocaine self-administration, the number of responses on the active and inactive lever (including time-out responding) for each group pretreated with a GABAB receptor ligand were analyzed by separate one-way analyses of variance (ANOVAs) for repeated measures. Post hoc Dunnett's test was used to analyze differences between group means. During maintenance of food self-administration, the number of responses on the active and inactive lever were analyzed by separate oneway ANOVAs for repeated measures; post hoc Dunnett's test was used to analyze differences between group means. In drug discrimination studies, accuracy was defined as a ratio of correct responses to the total number of responses before delivery of the first reinforcer. During test sessions, performance was expressed as the percentage of drug-lever responses to total responses upon completion of an fixed ratio 20 on either lever. Response rates (responses per s), regarded as a measure of behavioral disruption, were calculated as the total number of responses to either lever before completion of the first fixed ratio 20, divided by the number of seconds required to complete the fixed ratio. The data from animals that completed the fixed ratio 20 during the test sessions were used for analysis of percentage of

Fig. 2. Combination studies with SCH 50911 injected before baclofen, SKF 97541 or CGP 7930 on responding maintained by cocaine (0.5 mg/kg/injection) under a fixed ratio 5 schedule of reinforcement. A fixed dose of SCH 50911 (SCH; 10 mg/kg) was given before a fixed dose of baclofen (BAC; 2.5 or 5 mg/kg), SKF 97541 (SKF; 0.1 or 0.3 mg/kg) or CGP 7930 (CGP; 30 mg/kg) to rats. ⁎P b 0.01, ⁎⁎P b 0.001 vs. corresponding vehicle (VEH) + vehicle group; #P b 0.01 vs. corresponding vehicle + baclofen group; ^P b 0.01 vs. corresponding vehicle + SKF 97541 group; +P b 0.05 vs. vehicle + CGP 7930 group.

152

M. Filip et al. / European Journal of Pharmacology 574 (2007) 148–157

Fig. 3. Effects of baclofen, SKF 97541, CGP 7930 and SCH 50911 on food selfadministration in rats responding under a fixed ratio 5 schedule of reinforcement. Number of the active (cross bars), inactive (white bars) lever presses during food self-administration (mean of the last 3 maintenance days ± S.E.M.) are shown. Baclofen (BAC; 1.25–5 mg/kg), SKF 97541 (SKF; 0.03–0.3 mg/kg), CGP 7930 (CGP; 10–100 mg/kg) or SCH 50911 (SCH; 3–10 mg/kg) were given to 4 independent groups of rats. ⁎Pb 0.05, ⁎⁎Pb 0.01 vs. corresponding vehicle (VEH) group.

drug-lever responses, while the data from all the rats used in tests were included for analysis of rate of responding. Separate one-way ANOVAs for repeated measures were used to compare the percentage of drug-lever responding and response rates during the test sessions with the corresponding values of the preceding drug session (substitution tests). Post hoc Dunnett's test was used to analyze differences between group means. In combination experiments, a two-way ANOVA for repeated measures was used to analyze the effects of a GABAB receptor ligand (factor 1) and cocaine dose (factor 2). A priori planned pairwise comparisons to compare the effects of each dose of cocaine in the presence vs. absence of a GABAB receptor ligand were conducted using Bonferonni-corrected test for planned comparisons. For locomotor activity studies, a one-way ANOVA, followed by post-hoc Dunnett's test, was used to evaluate the treatment group effect.

requiring that the rate of active lever presses varied by less than 10%. The animals had self-administered 30–34 injections of cocaine with the daily mean cocaine intake between 15–17 mg/ kg. Rats responded significantly more frequently on the active lever than on the inactive lever (P b 0.01), independently of selfadministration test day. Baclofen (1.25–5 mg/kg) significantly altered the number of active lever presses (F(3,24) = 10.04, P b 0.001), but not inactive lever presses (F(3,24) = 0.9). Baclofen at doses of 2.5 and 5 mg/ kg inhibited the maintenance of cocaine (0.5 mg/kg/injection) self-administration (Fig. 1, left upper panel). SKF 97541 (0.03–0.3 mg/kg) significantly altered the number of active lever presses (F(3,28) = 4.8, P b 0.01), but not inactive lever presses (F(3,28) = 0.07). SKF 97541 at doses of 0.1 and 0.3 mg/kg inhibited the maintenance of cocaine (0.5 mg/kg/injection) self-administration (Fig. 1, right upper panel). CGP 7930 (10–100 mg/kg) significantly altered the number of active lever presses (F(3,25) = 6.9, P b 0.01), but not inactive lever presses (F(3,25) = 0.52). CGP 7930 at doses of 30 and 100 mg/kg inhibited the maintenance of cocaine (0.5 mg/kg/ injection) self-administration (Fig. 1, left bottom panel). SCH 50911 (3–10 mg/kg) neither changed the number of active (F(2,19) = 1.4) nor inactive (F(2,19) = 1.04) lever presses (Fig. 1, right bottom panel). A significant main effect of SCH 50911 on baclofen (2.5 mg/ kg)-and baclofen (5 mg/kg)-induced reduction of cocaine (0.5 mg/kg/injection) self-administration was observed (SCH 50911 + baclofen, 2.5 mg/kg: F(2,18) = 12.1, P b 0.001 for active lever presses; SCH 50911 + baclofen, 5 mg/kg: F(2,18) = 5.7, P b 0.01 for active lever presses). Administration of baclofen (2.5 and 5 mg/kg) significantly reduced the number of active lever presses (P b 0.01). Pretreatment with SCH 50911 (10 mg/ kg) significantly reduced the inhibitory effects of baclofen (2.5 mg/kg) on the active lever presses (P b 0.001). However, pretreatment with SCH 50911 (10 mg/kg) did not attenuate the

3. Results 3.1. Effects of the GABAB receptor ligands on cocaine (0.5 mg/ kg/injection) self-administration Rats showed stable responding on levers during the last 6 selfadministration maintenance sessions with an acquisition criterion

Fig. 4. Combination studies with SCH 50911 injected before baclofen or SKF 97541 on responding maintained by food under a fixed ratio 5 schedule of reinforcement. A fixed dose of SCH 50911 (SCH; 10 mg/kg) was given before a fixed dose of baclofen (BAC; 5 mg/kg) or SKF 97541 (SKF; 0.3 mg/kg) to rats. ⁎P b 0.01, ⁎⁎P b 0.001 vs. corresponding vehicle (VEH) + vehicle group; #P b 0.05 vs. vehicle + baclofen group; ^P b 0.01 vs. vehicle+ SKF 97541 group.

M. Filip et al. / European Journal of Pharmacology 574 (2007) 148–157

153

Fig. 5. Results of substitution and combination tests with baclofen and SKF 97541 in rats trained to discriminate cocaine (10 mg/kg) from saline. Data represent the mean percentage of cocaine-lever responding (±S.E.M.; upper panels) and the response rate in mean number of responses (±S.E.M.; lower panels). Performance is denoted after injection of vehicle (VEH; circle), baclofen (BAC; 1.25 mg/kg; reverse triangle), SKF 97541 (SKF; 0.1 mg/kg; diamond), vehicle + cocaine (COC; 1.25–10 mg/kg; circles), a fixed dose of baclofen (1.25 mg/kg) + cocaine (1.25–10 mg/kg; reversed triangles) or a fixed dose of SKF 97541 (0.1 mg/kg)+ cocaine (1.25–10 mg/kg; diamonds) during a test session. ⁎P b 0.01 vs. previous maintenance cocaine session; #P b 0.01 vs. vehicle at the same dose of cocaine.

inhibitory effects of baclofen (5 mg/kg) on the active lever presses (Fig. 2, left panel). A significant main effect of SCH 50911 on SKF 97541 (0.1 mg/kg)-and SKF 97541 (0.3 mg/kg)-induced reduction of

cocaine (0.5 mg/kg/injection) self-administration was observed (SCH 50911 + SKF 97541, 0.1 mg/kg: F(2,18) = 9.1, P b 0.01 for active lever presses; SCH 50911 + SKF 97541, 0.3 mg/kg: F(2,18) = 8.9, P b 0.01 for active lever presses). Administration

Fig. 6. Results of substitution and combination tests with CGP 7930 and SCH 50911 in rats trained to discriminate cocaine (10 mg/kg) from saline. Performance is denoted after injection of vehicle (VEH; circles), CGP 7930 (CGP; 30 mg/kg; reverse triangle), CGP 7930 (100 mg/kg; triangle), SCH 50911 (SCH; 10 mg/kg; diamond), vehicle+ cocaine (COC; 1.25–10 mg/kg; circles), a fixed dose of CGP 7930 (30 mg/kg) + cocaine (1.25–10 mg/kg; reversed triangles), a fixed dose of CGP 7930 (100 mg/kg)+ cocaine (2.5–10 mg/kg; triangles) or a fixed dose of SCH 50911 (10 mg/kg)+ cocaine (2.5–10 mg/kg; diamonds) during a test session. ⁎P b 0.01 vs. previous maintenance cocaine session; #P b 0.01 vs. vehicle at the same dose of cocaine. The cocaine dose–response data were replotted from Fig. 5. For more details see Fig. 5.

154

M. Filip et al. / European Journal of Pharmacology 574 (2007) 148–157

of SKF 97541 (0.1 and 0.3 mg/kg) significantly reduced the number of active lever presses (P b 0.01); the effects being blocked by SCH 50911 (10 mg/kg) (Fig. 2, center panel). A significant main effect of SCH 50911 on CGP 7930 (30 mg/kg)-induced reduction of cocaine (0.5 mg/kg/injection) self-administration was observed (SCH 50911 + CGP 7930, 30 mg/kg: F(2,18) = 9.1, P b 0.01 for active lever presses). Administration of CGP 7930 (30 mg/kg) significantly reduced the number of active lever presses (P b 0.01); the effect being blocked by SCH 50911 (10 mg/kg) (Fig. 2, right panel). 3.2. Food self-administration Rats showed stable responding on levers during the last 3 self-administration maintenance sessions with an acquisition criterion requiring that the rate of active lever presses varied by less than 10%. Rats responded significantly more frequently on the active lever than on the inactive lever (P b 0.01), independently of food self-administration test day. Baclofen (1.25–5 mg/kg) significantly altered the number of active lever presses (F(3,24) = 14.7, P b 0.001), but not inactive lever presses (F(3,24) = 0.54). Baclofen at a dose of 5 mg/kg inhibited the maintenance of food self-administration (Fig. 3, left upper panel). SKF 97541 (0.03–0.3 mg/kg) significantly altered the number of active lever presses (F(3,24) = 3.9, P b 0.05), but not inactive lever presses (F(3,24) = 0.4). SKF 97541 at a dose of 0.3 mg/kg inhibited the maintenance of food self-administration (Fig. 3, right upper panel). CGP 7930 (10–100 mg/kg) neither changed the number of active (F(3,23) = 0.99) nor inactive (F(3,23) = 0.96) lever presses (Fig. 3, left bottom panel). SCH 50911 (3–10 mg/kg) neither changed the number of active (F(2,18) = 0.6) nor inactive (F(2,18) = 0.2) lever presses (Fig. 3, right bottom panel). A significant main effect of SCH 50911 on baclofen (5 mg/ kg)-induced reduction of food self-administration was observed (SCH 50911 + baclofen, 5 mg/kg: F(2,15) = 25.9, P b 0.001 for active lever presses). Administration of baclofen (5 mg/kg) significantly reduced the number of active lever presses (P b 0.001); the effect being blocked by SCH 50911 (10 mg/ kg) (Fig. 4, left panel). A significant main effect of SCH 50911 on SKF 97541 (0.3 mg/kg)-induced reduction of food self-administration was observed (SCH 50911 + SKF 97541, 0.3 mg/kg: F(2,15) = 12.5, P b 0.001 for active lever presses). Administration of SKF 97541 (0.3 mg/kg) significantly reduced the number of active lever presses (P b 0.001); the effect being blocked by SCH 50911 (10 mg/kg) (Fig. 4, right panel).

10% of drug-lever responding. The response rates after saline (0.54 responses/s) and all the test doses of cocaine (Figs. 5–6) did not differ from those recorded during the preceding cocaine session. Following administration of baclofen (1.25 mg/kg), SKF 97541 (0.1 mg/kg), CGP 7930 (30 or 100 mg/kg), SCH 50911 (10 mg/kg), or their vehicles no substitution (less than 15% of cocaine-lever responding) was found (Figs. 5–6). Baclofen (2.5–5 mg/kg) or SKF 97541 (0.3 mg/kg) produced a behavioral disruption (0 responses/s) in 5–6 out of 7 animals used in the test sessions (data not shown). There was no main effect of a baclofen × cocaine dose on cocaine-lever responding (F(3,42) = 2.04) and on response rates (F(3,42) = 0.2) interaction as illustrated by a two-way ANOVA for repeated measures. A priori comparisons indicated that the 1.25 mg/kg dose of baclofen significantly reduced cocaine-lever responding only at 2.5 and 10 mg/kg of cocaine, while 1.25 mg/ kg of baclofen reduced rate of responding following 1.25 and 5 mg/kg of cocaine (Fig. 5, left panel). There was no main effect of a SKF 97541 × cocaine dose on cocaine-lever responding (F(3,42) = 0.77) and on response rates (F(3,42) = 0.71) interaction as illustrated by a two-way ANOVA for repeated measures. A priori comparisons indicated that the 0.1 mg/kg dose of SKF 97541 significantly reduced rate of responding following 10 mg/kg of cocaine (Fig. 5, right panel). There was no main effect of a CGP 7930 (30 mg/kg) × cocaine dose on cocaine-lever responding (F(3,36) = 1.29) and on response rates (F(3,36) = 0.66) interaction as illustrated by a

3.3. Drug discrimination studies The acquisition of cocaine (10 mg/kg) vs. saline discrimination was reached in the average of 26 sessions (ranging from 19 to 29). Administration of cocaine (1.25–10 mg/kg) produced a dose-dependent increase in the cocaine-lever responding (Figs. 7–8), while administration of saline evoked less than

Fig. 7. Effects of baclofen (BAC; 1.25–5 mg/kg), SKF 97541 (SKF; 0.03–0.3 mg/kg), CGP 7930 (CGP; 10–100 mg/kg) and SCH 50911 (SCH; 3–10 mg/kg) on spontaneous locomotor activity (expressed as distance traveled) in rats. ⁎Pb 0.01 vs. corresponding vehicle (VEH) group.

M. Filip et al. / European Journal of Pharmacology 574 (2007) 148–157

two-way ANOVA for repeated measures. There was also no main effect of a CGP 7930 (100 mg/kg) × cocaine dose on cocainelever responding (F(2,20)= 2.9) and on response rates (F(2,20)= 1.4) interaction as illustrated by a two-way ANOVA for repeated measures. A priori comparisons indicated that the 30 and 100 mg/kg doses of CGP 7930 significantly reduced rate of responding following 10 mg/kg of cocaine (Fig. 6, left panel). There was no main effect of a SCH 50911 × cocaine dose on cocaine-lever responding (F(2,26) = 1.54) and on response rates (F(2,26) = 0.001) interaction as illustrated by a two-way ANOVA for repeated measures. A priori comparisons indicated that the 10 mg/kg dose of SCH 50911 significantly reduced cocaine-lever responding only at 5 mg/kg of cocaine (Fig. 6, right panel). 3.4. Locomotor activity Treatment with baclofen (1.25–5 mg/kg) significantly altered the rats' basal locomotor activity (F(3,24) = 3.8, P b 0.05). Baclofen at a dose of 5 mg/kg inhibited the spontaneous locomotor activity expressed as distance traveled (Fig. 7, left upper panel). Treatment with CGP 7930 (10–100 mg/kg) significantly altered the rats' basal locomotor activity (F(3,25) = 3.2, P b 0.05). CGP 7930 at a dose of 100 mg/kg inhibited the spontaneous locomotor activity expressed as distance traveled (Fig. 7, left bottom panel). Neither SKF 97541 (0.03–0.3 mg/kg; F(3,25) = 0.69) nor SCH 50911 (3–10 mg/kg; F(2,16) = 0.82) altered the rats' spontaneous locomotor activation (Fig. 7). 4. Discussion The present findings indicate that tonic activation of GABAB receptors is no required to the reinforcing effects of cocaine, since the selective antagonist SCH 50911 (which displays high affinity and in the in vivo efficacy IC50 = 1.1 μM for GABAB receptors; Bolser et al., 1995) did not alter cocaine selfadministration under a fixed ratio 5 schedule of reinforcement in rats. Our findings are supported by an earlier observation of Brebner et al. (2002) who found that the another potent and selective GABAB receptor antagonist 3-{1-(S)-[[3-(cyclohexylmethyl)-hydroxyphosphinyl-2-(S)-hydroxypropyl]amino] ethyl}-benzoic acid (CGP 56433A) failed to alter cocaine selfadministration. Moreover, we also observed that GABAB receptor blockade by SCH 50911 did not alter the expression of the discriminative stimulus effects of cocaine. On the other hand, we extend recent data showing that pharmacological stimulation of GABAB receptors by the selective and very potent in vivo GABAB receptor agonist SKF 97541 (Froestl et al., 1995a) at doses between 0.1–0.3 mg/kg during the maintenance of cocaine self-administration significantly reduced the cocaine intake under the 0.5 mg/kg unit dose. The inhibitory responses of SKF 97541 on active lever presses and cocaine reinforcers are shared with other direct GABAB receptor agonists baclofen (Barrett et al., 2005; Brebner et al., 2000a; Di Ciano and Everitt, 2003; Roberts et al., 1996; present

155

study), CGP 44532 (Brebner et al., 1999) as well as with the allosteric positive modulators CGP 7930 (Smith et al., 2004; present study) and N,N’-dicyclopentyl-2-methylsulfanyl-5-nitropyrimidine-4,6-diamine (GS 39783; Smith et al., 2004). Moreover, when using cocaine self-administration reinforced under a progressive ratio schedule, baclofen (Brebner et al., 2000b) or CGP 7930 decreased break points (Smith et al., 2004) what indicates a reduction of the reinforcing efficacy of cocaine. As found in this study, the inhibitory effects of GABAB receptor agonists and the allosteric positive modulator were blocked by the GABAB receptor antagonist SCH 50911 what supports a role for GABA B receptor stimulation in reduction of cocaine reinforcement. The extent to which these inhibitory effects of the GABAB receptor agonists or allosteric positive modulator are attributable to specific responses was clarified by using the dose-range of these drugs – as tested in the cocaine self-administration study – on food-maintained responding. As shown in this paper the specificity of baclofen (5 mg/kg) and SKF 97541 (0.3 mg/kg) in cocaine-maintained responding may be questionable since the drugs decreased the maintenance of food self-administration and their selectivity cocaine vs. food appear to be only 3-fold. Supporting our findings, the GABAB receptor agonists baclofen and CGP 44532 have been demonstrated to reduce foodmaintained behavior at the same doses that reduced cocaine selfadministration under a fixed ratio 5 or 10 schedule of reinforcement in rats (Barrett et al., 2005) or baboons (Weerts et al., 2005), respectively, while baclofen at doses N2.5 mg/kg produced “statistically significant, but relatively modest effect on break points when food was used as a reinforcer” (Roberts et al., 1996). These data were collected in the experimental conditions where only one reinforcer (either cocaine or food) was available during the sessions. In contrast to the above findings, baclofen or CGP 44532 at doses that reduced cocaine self-administration were without significant effects on food-maintained responding on a discrete trial procedure (Brebner et al., 1999; Roberts and Andrews, 1997) or a multiple schedule of drug and food (Shoaib et al., 1998) in rats. In those studies cocaine and food were available in the same experimental session and the baseline rates of responding differ between food and drug. Availability of cocaine and food under independent and concurrent schedules seems to be a major procedural source of the difference in findings that causes problems in estimating the drug potency of two consequent events where food-maintained behavior might occur at a higher baseline than for cocaine, thus food intake was more difficult to suppress. Another noticeable distinction in the results from the present study and those of Roberts and Andrews (1997) and Shoaib et al. (1998) was the difference between the liquid (sweetened milk) and food reinforcers, respectively. We observed a suppression of liquid-maintained responding following baclofen or SKF 97541 treatments while with food reinforcers there was no significant reduction (Brebner et al., 1999; Roberts and Andrews, 1997; Shoaib et al., 1998). Since, in the latter studies, food presentation schedules (either during experimental sessions as the only source of food (Roberts and Andrews, 1997), or during sessions and for a short time following them (Brebner

156

M. Filip et al. / European Journal of Pharmacology 574 (2007) 148–157

et al., 1999)) may enhance the primary reinforcing properties of food as well as making these reinforcers essential to survival and which were resistant to GABAB receptor agonists. Further supporting such a statement Rodefer and Carroll (1996) have found that food consumption was less sensitive to pharmacological manipulation that liquid (saccharin) consumption. Interestingly, the inhibitory effects of the GABAB receptor allosteric positive modulator CGP 7930 on cocaine selfadministration seem to be related with a motivational decrease for the artificial reinforcer but not with a general suppression of appetitive behaviors since – in contrast to direct GABAB receptor agonists – no change in food self-administration was observed (present study). It should be added, however, that the inhibitory effects of baclofen (5 mg/kg) toward cocaine-and food-maintained responding and of CGP 7930 (100 mg/kg) toward cocaine-maintained responding are most probably linked with their locomotor suppressant actions. Supporting locomotor suppressing actions of CGP 7930, Carai et al. (2004b) found that this drug at doses 100–170 mg/kg potentiated the sedative/hypnotic effect of baclofen in DBA mice without significant loss of the righting reflex. The present data showing lack of significant effects of baclofen in rats trained to discriminate cocaine (10 mg/kg) from saline in a two-lever water-reinforced fixed ratio 20 task are similar to results of baclofen's pretreatment on cocaine discrimination using food-reinforced tasks and lower schedules of reinforcement (Barrett et al., 2005; Munzar et al., 2000). We extended these findings showing that the direct (SKF 97541 or SCH 50911) or indirect (CGS 7930) alteration of GABAB receptor function does not significantly alter the dose–response curve of cocaine. However, a priori comparisons indicated that a dose of baclofen (1.25 mg/kg) or SCH 50911 (10 mg/kg) significantly reduced drug-lever responding of particular doses of cocaine. These data showing that both the GABAB receptor agonist and the antagonist reduced the discriminative stimulus effects of cocaine may reflect differential effects on pre-versus post-synaptic GABAB receptors located at the same or at different synapses. It may be speculated that the antagonist blocked the presynaptic autoreceptors more potently than the postsynaptc receptors, and this way increased GABA release may lead to reduction of dopamine-dependent stimulus effects of cocaine. The same increase of GABA function can be seen by activating postsynaptic GABAB receptors by the agonists. Similar hypothesis is supported by the Macey et al. (2001) by showing that both GABAB receptor activation or blockade decrease brain stimulation reward in the rat. However, since the similar inhibitory pattern of responses toward the discriminative stimulus effects of cocaine was not observed for other GABAB receptor agents used in the present study (i.e. SKF 97543 or CGP 7930), the other factors probably mediated such an inhibition what needs further studies with more selective agonists and antagonists of GABAB receptors. The mechanism through which GABAB receptor agonists or allosteric positive modulators act to inhibit cocaine-maintained responding seems to be dependent on their involvement in the regulation of mesoaccumbal dopamine system, playing an important role in the reinforcing properties of cocaine (Wise,

1996). Thus, the mesoaccumbal dopamine system possess both the transcript and protein for GABAB receptors and GABAB receptors act as heteroreceptors present on dopamine and glutamatergic neurons in the ventral tegmental area (Bowery et al., 1987; Liang et al., 2000; Wirtshafter and Sheppard, 2001). Microdialysis studies have confirmed that baclofen-induced activation of GABAB receptors located in the ventral tegmental area decreases extracellular dopamine levels in the nucleus accumbens and prefrontal cortex (Westerink et al., 1997; Yoshida et al., 1994) or the cocaine-induced elevation in extracellular dopamine in the nucleus accumbens (Fadda et al., 2003). Additionally, behavior maintained by cocaine in self-administration procedures was attenuated by intra-ventral tegmental area or intra-nucleus accumbens administration of baclofen, with its stronger effect observed in the ventral tegmental area (Brebner et al., 2000b; Shoaib et al., 1998). In summary, cocaine reinforcement and discrimination do not depend on tonic activation of GABAB receptors while pharmacological stimulation of these receptors by direct agonists or allosteric positive modulators reduces cocaine reinforcement, but not the discriminative stimulus effects of this psychostimulant. The anti-reinforcing effects of direct GABAB receptor agonists cannot be totally dissociated from reduction in food intake and locomotor depression. On the other hand, dissociation in response of a GABAB receptor allosteric positive modulator, but not of direct GABAB receptor agonists, on cocaine vs. food-maintained responding was observed. Acknowledgements We are grateful to Dr. Andrew C. McCreary for the comments and language editing of the manuscript. Expert technical assistance was provided by Ewa Nowak and Karolina Wydra. The authors would like to kindly acknowledge donation of cocaine hydrochloride from the National Institute on Drug Abuse (RTI International, NC, USA). This study was supported by grants no. 2 PO5A 007 26 and N401 119 32/2421 from the Ministry of Science and Higher Education (Warszawa, Poland). Conflict of interest: none. References Barrett, A.C., Negus, S.S., Mello, N.K., Caine, S.B., 2005. Effect of GABA agonists and GABA-A receptor modulators on cocaine-and food-maintained responding and cocaine discrimination in rats. J. Pharmacol. Exp. Ther. 315, 858–871. Bolser, D.C., Blythin, D.J., Chapman, R.W., Egan, R.W., Hey, J.A., Rizzo, C., Kuo, S.C., Kreutner, W., 1995. The pharmacology of SCH 50911: a novel, orally-active GABA-beta receptor antagonist. J. Pharmacol. Exp. Ther. 274, 1393–1398. Bowery, N.G., Hudson, A.L., Price, G.W., 1987. GABAA and GABAB receptor site distribution in the rat central nervous system. Neuroscience 20, 365–383. Brebner, K., Childress, A.R., Roberts, D.C.S., 2002. A potential role for GABA (B) agonists in the treatment of psychostimulant addiction. Alcohol Alcoholism 37, 478–484. Brebner, K., Froestl, W., Andrews, M., Phelan, R., Roberts, D.C., 1999. The GABA(B) agonist CGP 44532 decreases cocaine self-administration in rats: demonstration using a progressive ratio and a discrete trials procedure. Neuropharmacology 38, 1797–1804.

M. Filip et al. / European Journal of Pharmacology 574 (2007) 148–157 Brebner, K., Phelan, R., Roberts, D.C., 2000a. Effect of baclofen on cocaine selfadministration in rats reinforced under fixed-ratio 1 and progressive-ratio schedules. Psychopharmacology 148, 314–321. Brebner, K., Phelan, R., Roberts, D.C., 2000b. Intra-VTA baclofen attenuates cocaine self-administration on a progressive ratio schedule of reinforcement. Pharmacol. Biochem. Behav. 66, 857–862. Campbell, U.C., Lac, S.T., Carroll, M.E., 1999. Effects of baclofen on maintenance and reinstatement of intravenous cocaine self-administration in rats. Psychopharmacology 143, 209–214. Carai, M.A., Colombo, G., Gessa, G.L., 2004a. Protection by the GABAB receptor antagonist, SCH 50911, of gamma-hydroxybutyric acid-induced mortality in mice. Eur. J. Pharmacol. 503, 77–80. Carai, M.A., Colombo, G., Froestl, W., Gessa, G.L., 2004b. In vivo effectiveness of CGP7930, a positive allosteric modulator of the GABAB receptor. Eur. J. Pharmacol. 504, 213–216. Di Ciano, P., Everitt, B.J., 2003. The GABA(B) receptor agonist baclofen attenuates cocaine and heroin-seeking behavior by rats. Neuropsychopharmacology 28, 510–518. Fadda, P., Scherma, M., Fresu, A., Collu, M., Fratta, W., 2003. Baclofen antagonizes nicotine-, cocaine-, and morphine-induced dopamine release in the nucleus accumbens of rat. Synapse 50, 1–6. Filip, M., 2005. Role of serotonin (5-HT)2 receptors in cocaine selfadministration and seeking behavior in rats. Pharmacol. Rep. 57, 35–46. Filip, M., Bubar, M.J., Cunningham, K.A., 2006. Contribution of serotonin (5-HT) 5-HT2 receptor subtypes to the discriminative stimulus effects of cocaine in rats. Psychopharmacology 183, 482–489. Froestl, W., Mickel, S.J., Hall, R.G., von Sprecher, G., Strub, D., Baumann, P.A., Brugger, F., Gentsch, C., Jaekel, J., Olpe, H.R., et al., 1995a. Phosphinic acid analogues of GABA. 1. New potent and selective GABAB agonists. J. Med. Chem. 38, 3297–3312. Froestl, W., Mickel, S.J., von Sprecher, G., Diel, P.J., Hall, R.G., Maier, L., Strub, D., Melillo, V., Baumann, P.A., Bernasconi, R., et al., 1995b. Phosphinic acid analogues of GABA. 2. Selective, orally active GABAB antagonists. J. Med. Chem. 38, 3313–3331. Galvez, T., Urwyler, S., Prezeau, L., Mosbacher, J., Joly, C., Malitschek, B., Heid, J., Brabet, I., Froestl, W., Bettler, B., Kaupmann, K., Pin, J.P., 2000. Ca(2+) requirement for high-affinity gamma-aminobutyric acid (GABA) binding at GABA(B) receptors: involvement of serine 269 of the GABA(B)R1 subunit. Mol. Pharmacol. 57, 419–426. Haney, M., Hart, C.L., Foltin, R.W., 2006. Effects of baclofen on cocaine selfadministration: opioid-and nonopioid-dependent volunteers. Neuropsychopharmacology 31, 1814–1821. Jones, K.A., Borowsky, B., Tamm, J.A., Craig, D.A., Durkin, M.M., Dai, M., Yao, W.J., Johnson, M., Gunwaldsen, C., Huang, L.Y., Tang, C., Shen, Q., Salon, J.A., Morse, K., Laz, T., Smith, K.E., Nagarathnam, D., Noble, S.A., Branchek, T.A., Gerald, C., 1998. GABA(B) receptors function as a heteromeric assembly of the subunits GABA(B)R1 and GABA(B)R2. Nature 396, 674–679. Kaupmann, K., Malitschek, B., Schuler, V., Heid, J., Froestl, W., Beck, P., Mosbacher, J., Bischoff, S., Kulik, A., Shigemoto, R., Karschin, A., Bettler, B., 1998. GABA(B)-receptor subtypes assemble into functional heteromeric complexes. Nature 396, 683–687. Liang, F., Hatanaka, Y., Saito, H., Yamamori, T., Hashikawa, T., 2000. Differential expression of gamma-aminobutyric acid type B receptor-1a and1b mRNA variants in GABA and non-GABAergic neurons of the rat brain. J. Comp. Neurol. 416, 475–495.

157

Macey, D.J., Froestl, W., Koob, G.F., Markou, A., 2001. Both GABAB receptor agonist and antagonists decreased brain stimulation reward in the rat. Neuropharmacology 40, 676–685. Munzar, P., Kutkat, S.W., Miller, C.R., Goldberg, S.R., 2000. Failure of baclofen to modulate discriminative-stimulus effects of cocaine or methamphetamine in rats. Eur. J. Pharmacol. 408, 169–174. Negus, S.S., Mello, N.K., Fivel, P.A., 2000. Effects of GABA agonists and GABA-A receptor modulators on cocaine discrimination in rhesus monkeys. Psychopharmacology 152, 398–407. Przegaliński, E., Filip, M., Papla, I., Siwanowicz, J., 2001. Effect of serotonin (5-HT)1B receptor ligands on cocaine sensitization in rats. Behav. Pharmacol. 12, 109–116. Roberts, D.C., Andrews, M.M., Vickers, G.J., 1996. Baclofen attenuates the reinforcing effects of cocaine in rats. Neuropsychopharmacology 15, 417–423. Roberts, D.C., Andrews, M.M., 1997. Baclofen suppression of cocaine selfadministration: demonstration using a discrete trials procedure. Psychopharmacology 131, 271–277. Rodefer, J.S., Carroll, M.E., 1996. Progressive ratio and behavioral economic evaluation of the reinforcing efficacy of orally delivered phencyclidine and ethanol in monkeys: effects of feeding conditions. Psychopharmacology 128, 265–273. Shoaib, M., Swanner, L.S., Beyer, C.E., Goldberg, S.R., Schindler, C.W., 1998. The GABAB agonist baclofen modifies cocaine self-administration in rats. Behav. Pharmacol. 9, 195–206. Shoptaw, S., Yang, X., Rotheram-Fuller, E.J., Hsieh, Y.C., Kintaudi, P.C., Charuvastra, V.C., Ling, W., 2003. Randomized placebo-controlled trial of baclofen for cocaine dependence: preliminary effects for individuals with chronic patterns of cocaine use. J. Clin. Psychiatry 64, 1440–1448. Slattery, D.A., Markou, A., Froestl, W., Cryan, J.F., 2005. The GABAB receptorpositive modulator GS39783 and the GABAB receptor agonist baclofen attenuate the reward-facilitating effects of cocaine: intracranial selfstimulation studies in the rat. Neuropsychopharmacology 30, 2065–2072. Smith, M.A., Yancey, D.L., Morgan, D., Liu, Y., Froestl, W., Roberts, D.C., 2004. Effects of positive allosteric modulators of the GABAB receptor on cocaine self-administration in rats. Psychopharmacology 173, 105–111. Urwyler, S., Mosbacher, J., Lingenhoehl, K., Heid, J., Hofstetter, K., Froestl, W., Bettler, B., Kaupmann, K., 2001. Positive allosteric modulation of native and recombinant gamma-aminobutyric acid(B) receptors by 2,6-Di-tertbutyl-4-(3-hydroxy-2,2-dimethyl-propyl)-phenol (CGP7930) and its aldehyde analog CGP13501. Mol. Pharmacol. 60, 963–971. Weerts, E.M., Froestl, W., Griffiths, R.R., 2005. Effects of GABAergic modulators on food and cocaine self-administration in baboons. Drug Alcohol Depend. 80, 369–376. Westerink, B.H., Kwint, H.F., de Vries, J.B., 1997. Eating-induced dopamine release from mesolimbic neurons is mediated by NMDA receptors in the ventral tegmental area: a dual-probe microdialysis study. J. Neurochem. 69, 662–668. Wirtshafter, D., Sheppard, A.C., 2001. Localization of GABA(B) receptors in midbrain monoamine containing neurons in the rat. Brain Res. Bull. 56, 1–5. Wise, R.A., 1996. Neurobiology of addiction. Curr. Opin. Neurobiol. 6, 243–251. Yoshida, M., Yokoo, H., Tanaka, T., Emoto, H., Tanaka, M., 1994. Opposite changes in the mesolimbic dopamine metabolism in the nerve terminal and cell body sites induced by locally infused baclofen in the rat. Brain Res. 636, 111–114.