discriminative stimulus effects in rats

(1)

Role of opioidergic mechanisms and GABA uptake inhibition in the heroin-induced

discriminative stimulus effects in rats

Wojciech Solecki¹, Tomasz Krówka¹, Ma³gorzata Filip², Ryszard Przew³ocki^1,3

Department of Molecular Neuropharmacology, Department of Pharmacology, Institute of Pharmacology, Polish Academy of Sciences, Smêtna 12, PL 31-343 Kraków, Poland

!Institute of Applied Psychology, Jagiellonian University, Józefa 19, PL 31-056 Kraków, Poland Correspondence:Ryszard Przew³ocki, e-mail: nfprzewl@cyf-kr.edu.pl

Abstract:

The present study was designed to investigate the involvement of opioidergic component as well as to study GABAergic mechanisms in the expression of heroin discrimination. Male Wistar rats were trained to discriminate heroin (0.5 mg/kg,ip) from saline (ip) in a two-choice water reinforced fixed ratio (FR) 20 drug discrimination paradigm. In substitution tests, heroin (0.0625–0.5 mg/kg) and morphine (0.5–2 mg/kg,ip) evoked a dose-dependent generalization for the drug lever-responding, while muscimol (1 mg/kg,ip, a GABA)receptor agonist) produced a weak partial substitution (ca. 48% heroin-lever responding). Neither tiagabine (2.5 mg/kg, ip; a GABA reuptake inhibitor), vigabatrin (75–150 mg/kg, ip; an irreversible inhibitor of GABA transaminase), nor baclofen (0.5 mg/kg,ip; a GABA*receptor agonist) substituted for heroin. In combination studies, the stimulus effects produced by heroin (0.5 mg/kg) or morphine (2 mg/kg) were dose-dependently blocked by opioid receptor antagonists naltrexone (0.1–1 mg/kg,ip), and naloxone (0.5–1 mg/kg, ip). The peripherally-acting naloxone methiodide at a dose of 1 mg/kg, ip did not alter, while at a dose of 10 mg/kg that penetrates across the blood-brain barrier, it reduced the stimulus effects of heroin or morphine. Pretreatment with tiagabine (2.5–5 mg/kg) produced a rightward shift of a heroin dose-response curve, while vigabatrin (75–300 mg/kg), baclofen (0.5–2.5 mg/kg) or muscimol (0.5–2 mg/kg) given prior to heroin (0.0625–0.5 mg/kg) failed to alter heroin discrimination. Our findings confirm previous studies on the significance of µ-opioidergic mechanisms in the expression of heroin discrimination and add the observation that selective inhibition of GABA reuptake, but not inhibition of GABA transaminase or direct stimulation of GABA₎and GABA_*receptors, can decrease the overall effects of heroin.

Key words:

heroin, opioid receptors, GABA, discriminative stimulus

Introduction

Heroin is one of the most rapidly acting and abused of the opiates, and its dependence presents serious medi- cal, social and criminal challenge in the today’s world [10]. Unfortunately, high abuse liability of heroin is not matched by effective pharmacological therapy

[23]. Despite the well-characterized subjective effects that heroin produces in humans and the recognized relevance of these effects for heroin addiction, com- paratively less has been reported about the mechanism(s) underlying heroin reinforcement in a drug discrimination model in rodents [35].

Some previous observations have shown that opioidergic system mediates the discriminative stimulus

Pharmacological Reports 2005, 57, 744754 ISSN 1734-1140

(2)

effects of heroin in animals [4, 35], with a crucial role of µ-opioid receptors in this behavior [3, 6, 14, 30, 37, 39].

Apart from the established µ-opioidergic mechanism, recent studies point to a significance of dopaminergic andg-aminobutyric acid (GABA)ergic neurotransmission in the mediation of heroin discrimination [14, 28]. Thus, several lines of evidence supports the hypothesis that the mesocorticolimbic dopamine system, which originates in the ventral tegmental area (VTA) and projects to the nucleus accumbens (NAcc) and to the medial prefrontal cortex, may have a role in mediating morphine and heroin discrimination [38, 41] and reinforcement in the self-administration model [2, 48]. In fact, indirect dopamine agonist cocaine and methamphetamine have been shown to share discriminative stimulus effects with µ-opioid receptor agonists under certain conditions [24, 33, 37]

and Corrigall and Coen [14] have found an involvement of dopamine D₁ receptors in mediating heroin discriminability. However, other findings have indicated that pharmacological elevation of interstitial dopamine levels does not mimic the discriminative stimulus properties of heroin [35], as well as that the destruction of presynaptic dopamine terminals in the NAcc [32] or pharmacological blockade of dopamine receptors by systemic or intra-NAcc administration of antagonists do not alter heroin self-administration [for review see 40] what raised the possibility that the behavioral effects of heroin and other opiates were not mediated directly and exclusively through the dopaminergic neurotransmission.

In fact, opiate reinforcement may be linked to dopamine neurotransmission indirectly via GABAergic disinhibitory mechanism [20, 28, 36]. In support of the above hypothesis, several opiates,via µ-opioid re- ceptor, inhibit GABAergic interneurons in the VTA leading to a decrease in GABA release which subse- quently disinhibits dopamine neurons and evokes dopamine release in the NAcc [1, 16, 20, 29, 42, 51]. In line with those neurochemical reports, heroin self- administration was reduced by the direct intra-VTA or intra-ventral pallidum administration of the irreversible inhibitor of GABA transaminase vigabatrin, that elevates GABA level in the synaptic cleft [11, 27], the latter effect being attenuated by the GABA_Breceptor antagonist 2-hydroxysaclofen in rats [50]. Similarly, the conditioned place preference induced by morphine was significantly attenuated by the intra-VTA

administration of the selective GABA_Bagonist baclofen [43]. Interestingly, to date, there have been no reported results on the role of GABA neurotransmission in the discriminative stimulus effects of heroin.

In the present study, we tested the hypothesis that enhancement of GABA transmission would alter the heroin-evoked discriminative stimulus effect. To this end, we used tiagabine (a GABA reuptake inhibitor;

[27]) and vigabatrin (an irreversible inhibitor of GABA transaminase; [27]). Since in combination studies tiagabine reduced the interoceptive effects of heroin, we studied the effects of the direct GABA_Are- ceptor agonist (muscimol) or the direct GABA_B receptor agonist; baclofen. Moreover, to extend previous studies we used µ-opioid receptor agonist (morphine; [38]) and antagonists (naloxone, naltrexone;

[34]) and tested if these drugs function as a positive control and alter heroin discrimination.

Materials and Methods

Animals

Male Wistar rats (Institute of Pharmacology, Kraków, Poland) weighting 280–300 g at the beginning of the experiment were used. The rats were housed in groups of 2/cage (38 × 25 × 15 × cm) in a colony room maintained at 21 ± 1°C on 12-h light-dark cycle (the lights on at 12.00 h). Rodent chow was available ad libitum; the amount of water that an animal re- ceived was restricted to that given during daily training sessions, after test sessions (15 min) and on week- ends (36 h). All experiments were conducted during light phase of the light-dark cycle, with the approval of the Bioethics Commission.

Drugs

The following drugs were used: baclofen (Tocris, UK), heroin, morphine hydrochloride (Polfa Kutno, Poland), muscimol, naloxone hydrochloride, naloxone methiodide, naltrexone hydrochloride (Sigma, St.

Louis, USA), tiagabine (Sanofi-Biocom, USA), and vigabatrin (Marion Merrell Dow, USA). All the drugs were administered intraperitoneally (ip) in a volume of 1 ml/kg and dissolved in a 0.9% NaCl, except for tiagabine and vigabatrin which were suspended in

(3)

a 1% Tween 80 solution (Sigma-Aldrich, St. Louis, USA). The doses refer to the weight of the respective salts. Heroin and saline were given at 15 min, naloxone and naloxone methiodide at 20 min, morphine, baclofen, muscimol and tiagabine at 30 min, naltrexone 45 min and vigabatrin at 240 min before the tests.

In combination tests with heroin or morphine, naloxone hydrochloride or naloxone methiodide were administrated 5 min prior to the given drug.

Apparatus

Two lever operant chambers (MedAssociates, St. Al- bans, USA) were used. Each chamber was equipped with a water-filled dispenser mounted equidistantly between two response levers on one wall and was housed in a light and soundproof cubicle (MedAssoci- ates). Illumination came from a 28V house light, Ven- tilation and masking noise were supplied with a blower. A computer with MedState software was used to program and record all the experimental events.

Discrimination procedure

Standard two lever, water-reinforced drug discrimination procedures were utilized [17, 18]. Two groups of rats were trained to discriminate heroin (0.5 mg/kg, ip) from saline (1 ml/kg. ip). The training dose of her- oin was chosen on the basis of the literature [36]. In the initial phase, only the stimulus-appropriate (drug or saline) lever was present in a chamber during 15-min daily sessions conducted on Mondays through Fridays. Training began under a fixed ratio 1 (FR 1) schedule of water reinforcement and the FR require- ment was incremented until all the animals were responding under FR 20 schedule for each experimental condition. For half of the rats, left-lever responses were reinforced with heroin administration, whereas right lever responses were reinforced with saline administration; the conditions were reversed for the re- maining rats. During that phase of training, heroin and saline were administered irregularly with the restric- tion that neither condition prevailed for more than two consecutive sessions. After the responding was stabi-

Fig. 1.Substitution studies with heroin and morphine in rats trained to discriminate heroin (0.5 mg/kg) from saline. Symbols show the mean percentage of heroin-lever responding (± SEM; closed symbols) and the mean number of responses/min (± SEM; open symbols). Performance is shown after injection of saline (1 ml/kg; squares, left), heroin (0.06250.5 mg/kg; circles, center) or morphine (0.52 mg/kg; triangles, right). All the data points are the mean of data from 5 to 8/8 rats completing the FR 20. Asterisks stand for a significant difference between performance during test session and preceding heroin maintenance session; p < 0.05

(4)

lized, discrimination training was initiated and both levers were presented simultaneously during 15-min training session. The animals were made to respond on the stimulus-correct lever in order to obtain water reinforcement. That phase of training lasted until the performance of all the trained rats met the criterion defined as mean accuracies of at least 80% correct for 10 consecutive sessions.

The test sessions were initiated after all rats achieved the criterion but the training sessions were run on the intervening days to maintain discrimination accuracy. The rats were required to maintain accuracies of at least 80% correctness for training sessions which preceded the test. During test sessions, the animals were placed in the chambers and upon completion of 20 responses on either lever, a single reinforcement was delivered and the house lights were turned off. The session lasted 15-min if an animal did not complete 20 responses on either lever.

In substitution (generalization) tests, the rats were examined for lever responses after various doses of training drug, morphine, naloxone, naloxone methiodide, naltrexone, baclofen, muscimol, tiagabine and vigabatrin. In combination tests, doses of naloxone, naloxone methiodide, naltrexone as well as baclofen, muscimol, tiagabine and vigabatrin were administered prior to the given drug (heroin or morphine) only at doses that did not affect performance of animals in substitution tests.

Data analysis

During test sessions, performance was expressed as the percentage of heroin appropriate responses to total responses before the delivery of the first reinforcement. Response rate (response per minute) were also evaluated during test sessions and were calculated as the total number of responses before completion of 20 responses on either lever divided by the number of minutes necessary to compete the FR 20. Only the data from animals which completed the FR 20 during test sessions were analyzed. A drug was considered to fully substitute for heroin if at least 80% of responses occurred on the drug appropriate lever, while a complete antagonism was claimed if about 20% of correct responses occurred.

Student’st-test for repeated measures was used to compare the percentage of drug lever responding and response rate during test session and the previous drug training session (substitution tests) or the corresponding dose of heroin given alone (combination tests). ED₅₀ and ID₅₀ values were calculated using Litchfield and Wilcoxon methodology [25]. The ED₅₀ values were compared using two-way analysis of variance (ANOVA) for repeated measures. All com- parisons were made with an alpha set at 0.05.

Results

Heroin – saline discrimination

The training dose of heroin (0.5 mg/kg) produced ca.

95% heroin-lever responding whereas saline engendered

< 10% heroin-lever responding. Response rates during the heroin or saline maintenance sessions did not differ.

Tab. 1.Substitution studies in rats trained to discriminate heroin (0.5 mg/kg) from saline

Drug Dose

(mg/kg)

% Heroin-lever responses (means ± SEM)

Responses/min (means ± SEM)

Saline 3.43 ± 2.65* 40.2 ± 5.4

Naloxone 1 18.18 ± 15.47* 44.5 ± 3.68 Naloxone methiodide 10 0.96 ± 0.95* 42 ± 6.67 Naltrexone 1 17.4 ± 13.75* 35.76 ± 4.42 Baclofen 0.5 24.68 ± 10.56* 60.36 ± 11.42

1.25 – 0/0

2.5 – 0/0

Muscimol 0.5 32.53 ± 19.83* 40.1 ± 7.68 1 48.58 ± 17.62* 55.2 ± 8.88 2 45.74 ± 18.75* 28.32 ± 5.71 Tiagabine 2.5 4.62 ± 4.2* 35.04 ± 5.26

5 – 0/0

Vigabatrin 75 13.1 ± 12.42* 38.18 ± 6.94 150 5 ± 4.99* 39.48 ± 5.26

300 – 0/0

The values are mean percentage of heroin-lever responding (± SEM) and the mean number of responses/min (± SEM). All the data points are the mean of data from 5 to 8/8 rats completing the FR 20. Aster- isks stand for a significant difference between performance during test session and preceding heroin maintenance session; p < 0.05;

0/0 responses/min indicate behavioral disruption

(5)

Substitution studies

Opioid ligands

The drug-lever responding after 0.0625, 0.125 and 0.25 mg/kg of heroin was significantly different from the preceding heroin maintenance session (p < 0.05).

The ED₅₀for heroin-lever responding was 0.14 mg/kg (95% CL 0.12–0.16 mg/kg). Response rate for all the tested doses of heroin did not differ from those ob- tained during the preceding heroin maintenance session (Fig. 1).

The µ-opioid receptor agonist morphine (0.5–2 mg/kg) fully substituted for heroin; nearly 80% heroin-lever responding was found at a dose of 2 mg/kg of morphine. The drug-lever responding at 0.5, 0.75 and 1 mg/kg of morphine was significantly different from the preceding heroin maintenance session (p < 0.05).

The ED₅₀ for morphine was 1.25 mg/kg (95% Cl 1.14–1.37 mg/kg). None of the doses of morphine altered response rates (Fig. 1).

The non-selective opioid receptor antagonists naloxone (1 mg/kg), naltrexone (1 mg/kg) and peripherally acting naloxone methiodide (10 mg/kg), evoked

< 20% heroin-lever responding (lack of substitution;

Tab. 1), and they did not alter response rates (Tab. 1).

GABAergic ligands

The GABA_B receptor agonist baclofen (0.5 mg/kg) did not substitute for heroin, while its higher doses (1.25–2.5 mg/kg) induced behavioral disruption (0 responses/min during test sessions; Tab. 1).

The GABA_Areceptor agonist muscimol (0.5–2 mg/kg) produced a weak partial substitution (ca. 48% heroin- lever responding after 1 mg/kg of muscimol). None of the doses of muscimol altered response rates (Tab. 1).

The GABA reuptake inhibitor tiagabine (2.5 mg/kg) did not substitute for heroin, while its higher dose (5 mg/kg) induced behavioral disruption (0 responses/min during test sessions; Tab. 1).

Fig. 2.Antagonism studies with naloxone in rats trained to discriminate heroin (0.5 mg/kg) from saline. Symbols show the mean percentage of heroin-lever responding (± SEM; closed symbols) and the mean number of responses/min (± SEM; open symbols). Performance is shown after injection of heroin alone (0.5 mg/kg; circles, left) or heroin preceded by injection of naloxone (0.11 mg/kg; triangles, center left), morphine alone (2 mg/kg; hexagons, center right) or morphine preceded by injection of naloxone (0.11 mg/kg; diamonds, right). All the data points are the mean of data from 5 to 8/8 rats completing the FR 20. Asterisks stand for a significant difference between performance during test session and preceding heroin maintenance session or for a significant difference between performance during test session and morphine (2 mg/kg) alone; p < 0.05

(6)

The irreversible inhibitor of GABA transaminase vigabatrin (75–150 mg/kg) failed to generalize for the stimulus effects of heroin and to alter animals’ response rate. The higher dose of vigabatrin (300 mg/kg) completely disrupted rats’ behavior (0 responses/min during test sessions; Tab. 1).

Combination studies

Opioid ligands

Pretreatment with opioid receptor antagonist naloxone (0.1–1 mg/kg) caused a dose-dependent attenuation of the heroin (0.5 mg/kg) discrimination, and morphine (2 mg/kg) substitution for heroin (Fig.

2). The statistically significant (p < 0.05) reduction of the heroin and morphine discriminative stimulus effects was found following 0.5–1 mg/kg of naloxone.

The ID₅₀ for naloxone given prior to heroin was

0.31 mg/kg (95% CL 0.1–0.95 mg/kg), and did not differ from that calculated for naloxone given prior to morphine (ID₅₀= 0.36 mg/kg, 95% Cl 0.33–0.4 mg/kg).

The response rates after any dose of naloxone in combination with either heroin or morphine were not altered (Fig. 2).

Given before heroin (0.5 mg/kg), the opioid receptor antagonist naltrexone (0.1–1 mg/kg) dose-dependently reduced the discriminative stimulus effects of the opioid agonist (p < 0.05). The ID₅₀ for naltrexone given prior to heroin was 0.11 mg/kg (95% Cl 0.08–0.15 mg/kg). The response rates after any dose of naltrexone in combination with heroin were not altered (data not shown).

The peripherally-acting opioid receptor antagonist naloxone methiodide at doses of 1–5 mg/kg did not alter the discriminative stimulus effects of heroin (0.5 mg/kg) or morphine (2 mg/kg), while at a dose of 10 mg/kg that penetrates across the blood-brain barrier, it reduced heroin discrimination or morphine sub-

Fig. 3.Antagonism studies with naloxone methiodide in rats trained to discriminate heroin (0.5 mg/kg) from saline. Symbols show the mean percentage of heroin-lever responding (± SEM; closed symbols) and the mean number of responses/min (± SEM; open symbols). Perform- ance is shown after injection of heroin alone (0.5 mg/kg; circles, left) or heroin preceded by injection of naloxone methiodide (110 mg/kg; circles, center left) and injection of morphine alone (2 mg/kg; hexagons, center right) or morphine preceded by injection of naloxone methiodide (110 mg/kg; hexagons, right). All the data points are the mean of data from 5 to 8/8 rats completing the FR 20. Asterisks stand for a significant difference between performance during test session and preceding heroin maintenance session or for a significant difference between performance during test session and morphine (2 mg/kg) alone; p < 0.05

(7)

stitution (p < 0.05; Fig. 3). The ID₅₀for naloxone methiodide given prior to heroin was 4.9 mg/kg (95% Cl 4.1–5.87 mg/kg), and did not differ from that calculated for naloxone methiodide given prior to morphine (ID₅₀= 4.88 mg/kg, 95% Cl 4.14–5.76 mg/kg).

The response rates after any dose of naloxone methiodide in combination with either heroin or morphine were not altered (Fig. 3).

GABAergic ligands

Pretreatment with the GABA reuptake inhibitor tiagabine (2.5 mg/kg) produced a rightward shift of heroin (0.0625–0.5 mg/kg) dose-response curve (Fig. 4).

This rightward shift was significant as indicated by nonoverlapping 95% CIs of ED₅₀ values for vehicle and tiagabine pretreatments [vehicle + heroin ED₅₀ values with 95% CIs were 0.14 (0.12–0.16), and for tiagabine + heroin ED₅₀ values with CIs 95% were 0.27 (0.24–0.3)], as well as by two-way ANOVA for repeated measures (F(9,66) = 13.86, p < 0.005). Ad- ministration of a fixed dose of tiagabine (2.5 mg/kg) in combination with 0.125, 0.25 or 0.5 mg/kg of her-

oin significantly decreased the heroin-lever responding as compared to a dose of heroin tested alone (p < 0.05; Fig. 4). The response rates after tiagabine (2.5 mg/kg) in combination with any dose of heroin were not altered (Fig. 4). The irreversible inhibitor of GABA transaminase vigabatrin, at doses of 75 or 150 mg/kg, did not alter the discriminative stimulus effects of heroin (0.125–0.5 mg/kg) or the animals’ response rates (Fig. 4). The GABA_Areceptor agonist muscimol (0.5–1 mg/kg) failed to alter the discriminative stimulus effects of heroin (0.125–0.5 mg/kg) or the animals’ response rates (Fig. 5). The GABA_B receptor agonist baclofen (0.5 mg/kg) did not change the discriminative stimulus effects of heroin (0.125–0.5 mg/kg) or the animals’ response rates (Fig. 5).

Discussion

In support of previous studies [14, 24, 30], we report that heroin can be used as a stimulus cue in rats.

Moreover, pharmacological analyses using µ-opioid

Fig. 4.Combination studies with tiagabine or vigabatrin in rats trained to discriminate heroin (0.5 mg/kg) from saline. Symbols show the mean percentage of heroin-lever responding ± SEM; closed symbols) and the mean number of responses/min (± SEM; open symbols). Perform- ance is shown after injection of heroin alone (0.5 mg/kg; circles, left), heroin (0.1250.5 mg/kg; circles) preceded by injection of tiagabine (2.5 mg/kg; triangles, center) or vigabatrin (75 mg/kg; triangles, 150 mg/kg; squares, right). All the data points are the mean of data from 5 to 8/8 rats completing the FR 20. Asterisks stand for a significant difference between performance during test session and a corresponding dose of heroin alone; p < 0.05

(8)

receptor agonist and antagonists as well as different GABAergic drugs provided an evidence that heroin discriminative stimulus properties depended on the selective µ-opioid receptors and could be attenuated by the inhibition of GABA uptake.

Morphine substitution as well as naloxone and naltrexone antagonism studies corroborated previous findings that heroin’s discriminative stimulus effects are primarily mediatedvia the µ-opioid peptide receptor [5, 44, 46]. Interestingly, in our study we found that naloxone methiodide (1–5 mg/kg), a derivative of naloxone that crosses the blood-brain barrier with limitation [19], did not alter the heroin discriminative stimulus effects of heroin or the morphine substitution for heroin. However, at a dose of 10 mg/kg, naloxone methiodide potently reduced the stimulus effects of heroin and morphine raising the possibility that at such high dose the antagonist might reach the central nervous system. The reduction of heroin discrimination seems to be specific since morphine, naloxone, naloxone methiodide or naltrexone did not alter the animals’ response rates.

We found that enhancement of GABAergic transmission by GABA uptake inhibition potently reduced

the heroin discriminative stimulus effects. In fact, the GABA uptake inhibitor tiagabine given separately (2.5 mg/kg) shifted rightward the dose-response curve for heroin and increased its ED₅₀value. The observed reduction in heroin discrimination did not result from disruption in responding caused by sedation or loco- motor depression since we have not observed differ- ences in response rates after tiagabine pretreatment.

The similar antagonistic interaction between drugs that enhance GABA neurotransmission and heroin was demonstrated in the self-administration model in which another GABA uptake inhibitor NipA or the irreversible inhibitor of GABA transaminase vigabatrin consistently blocked heroin self-administration following systemic or intra-VTA application [50].

The pharmacological mechanism of tiagabine de- pends on elevation of GABA level in the synaptic cleft and antagonism of opioid-induced excitatory effects on VTA dopamine cells in the VTA. In fact, activation of opioid receptors located within mesolimbic structures such as VTA and NAcc mediate the discriminative stimulus effects of opiates [13, 38, 41]

and are critically involved in opiate self-administration [47]. Opioid-induced excitatory effects on VTA DA

Fig. 5.Combination studies with muscimol and baclofen in rats trained to discriminate heroin (0.5 mg/kg) from saline. Symbols show the mean percentage of heroin-lever responding (± SEM; closed symbols) and the mean number of responses/min (± SEM; open symbols). Perform- ance is shown after injection of heroin alone (0.5 mg/kg; circles, left), heroin (0.06250.5 mg/kg, circles) preceded by injection of muscimol (0.5 mg/kg; triangles, 1 mg/kg; squares, center) or heroin (0.1250.5 mg/kg, circles) preceded by injection of baclofen (0.5 mg/kg; triangles, right). All the data points are the mean of data from 5 to 8/8 rats completing the FR 20. There are no significant difference between performance during test session and corresponding dose of heroin alone

(9)

cells are mediated, at least in part, by inhibiting GABAergic interneurons on which opioid receptors are located [16, 20, 42, 47, 49].

In the light of the above hypothesis, it was surpris- ing to observe that another drug that increases GABA- ergic neurotransmission was inactive in our behavioral model. In fact, vigabatrin, an irreversible inhibitor of GABA transaminase, failed to reduce the discriminative stimulus effects of heroin. However, it should be underlined that the effective dose of vigabatrin (300 mg/kg) capable of reducing the effects of heroin in a conditioned place preference model [31]

was behaviorally disrupted in our instrumental model (rats did not respond). The lack of attenuation of heroin discrimination by lower doses of vigabatrin (75–150 mg/kg) might result from too low concentra- tion of GABA in the synaptic cleft to counteract the effects of heroin.

Being released into synaptic cleft, GABA activates specific receptor subtypes, and two of them GABA_A and GABA_Bare of special interest. Both GABA_Aand GABA_B receptor subtypes have been identified within the VTA [7]. In the latter brain structure, the majority of GABA_Areceptors are found on both dopamine and GABAergic neurons [7, 12, 22]. Systemic administration of GABA_A agonists significantly inhibits VTA GABAergic neurons and excites VTA dopamine neurons [21, 45]. On the other hand, GABA_B receptors are located on dopaminergic neurons in the VTA [26]. Stimulation of GABA_Breceptors leads to dopamine cell inhibition [15, 20] and reduces basal [22, 51] and morphine- or D-Ala2,N-Me-Phe4,Gly- d5]-enkephalin (DAMGO)-induced dopamine release in the VTA and NAcc [21, 22]. The behavioral conse- quence of such actions of the GABA_Breceptor agonists (e.g. baclofen) is attenuation of heroin self- administration [9, 48] or morphine-induced conditioned place preference [8, 43].

In contrast to the above behavioral findings, in our study, the GABA_Breceptor agonist baclofen failed to modulate the heroin-induced discriminative stimulus effects. The GABA_A agonist receptor muscimol in a dose-range of 0.5–2 mg/kg evoked a weak partial substitution (ca. 48% drug-lever responding) for heroin while in combination studies it produced a non- significant attenuation of the heroin (0.25–0.5 mg/kg) discrimination. Such profile of muscimol may result from simultaneous activation of GABA_Areceptors located on dopamine cells and on GABAergic cells.

Lack of inhibitory effects of baclofen may come from

an insufficient dose of this agonist (0.5 mg/kg) used in combination studies. However, the higher doses of baclofen (1.25–2.5 mg/kg), being effective toward the discriminative stimulus effects of cocaine in rats (Filip et al., unpublished data), disrupted instrumental responding in rats trained to recognize heroin from saline.

In conclusion, our results corroborate previous findings, that heroin’s discriminative stimulus effects are primarily mediatedvia the µ-opioid peptide receptor. Furthermore, they indicate that heroin discrimination is reduced by enhancing GABA transmissionvia GABA reuptake inhibition. The systemically administrated GABA transaminase inhibitor vigabatrin, GABA_A receptor agonist muscimol, or GABA_B receptor agonist baclofen (at doses that had no locomo- tor disruptive properties), were not effective in attenu- ating the overall effects of heroin.

Acknowledgment:

This research was supported by statutory funds from the Ministry of Scientific Research and Information Technology (Warszawa, Poland).

References:

1. Andrews N, Loomis S, Blake R, Ferrigan L, Singh L, McKnight AT: Effect of gabapentin-like compounds on development and maintenance of morphine-induced conditioned place preference. Psychopharmacology, 2001, 157, 381–387.

2. Bardo MT: Neuropharmacological mechanisms of drug reward: beyond dopamine in the nucleus accumbens.

Crit Rev Neurobiol, 1998, 12, 37–67.

3. Beardsley PM, Harris LS: Evaluation of the discriminative stimulus and reinforcing effects of dihydroetorphine.

Drug Alcohol Depend, 1997, 48, 77–84.

4. Bertalmio AJ, Medzihradsky F, Winger G, Woods JH:

Differential influence of N-dealkylation on the stimulus properties of some opioid agonists. J Pharmacol Exp Ther, 1992, 261, 278–284.

5. Bickel U, Schumacher OP, Kang YS, Voigt K: Poor permeability of morphine 3-glucuronide and morphine 6-glucuronide through the blood-brain barrier in the rat.

J Pharmacol Exp Ther, 1996, 278, 107–113.

6. Bowen CA, Fisher BD, Mello NK, Negus SS: Antago- nism of the antinociceptive and discriminative stimulus effects of heroin and morphine by 3-methoxynaltrexone and naltrexone in rhesus monkeys. J Pharmacol Exp Ther, 2002, 302, 264–273.

7. Bowery NG, Hudson AL, Price GW: GABA)and GABA_*receptor site distribution in the rat central nervous system. Neuroscience, 1987, 20, 365–383.

(10)

8. Brebner K, Childress AR, Roberts DCS: A potential role for GABA_*agonists in the treatment of psychostimulant addiction. Alcohol Alcohol, 2002, 37, 478–484.

9. Brebner K, Froestl W, Roberts DSC: The GABA*antagonist CGP56433A attenuates the effect of baclofen on cocaine but not heroin self-administration in the rat. Psy- chopharmacology, 2002, 160, 49–55.

10. Brown R: Heroin dependence. WMJ, 2004, 103, 20–6.

11. Caille S, Parsons LH: Intravenous heroin self-

administration decreases GABA efflux in the ventral pallidum: anin vivo microdialysis study in rats. Eur J Neu- rosci, 2004, 20, 593–596.

12. Churchill L, Dilts RP, Kalivas PW: Autoradiographic localization of GABA₎receptors within the ventral tegmental area. Neurochem Res, 1992, 17, 101–106.

13. Cook CD, Beardsley PM: Modulation of the discriminative stimulus effects of mu opioid agonists in rats: II. Ef- fects of dopamine D2/D3 agonists. Behav Pharmacol, 2004, 15, 75–83.

14. Corrigall WA, Coen KM: Selective Dand D dopamine antagonists decrease response rates of food-maintained behavior and reduce the discriminative stimulus produced by heroin. Pharmacol Biochem Behav, 1990, 35, 351–355.

15. Di Ciano P, Everitt BJ: The GABA*receptor agonist baclofen attenuates cocaine and heroin-seeking behavior by rats. Neuropsychopharmacology, 2003, 28, 510–518.

16. Dilts RP, Kalivas PW: Autoradiographic localization of mu-opioid and neurotensin receptors within the mesolimbic dopamine system. Brain Res, 1989, 488, 311–327.

17. Filip M, Cunningham KA: Hyperlocomotive and discriminative stimulus effects of cocaine are under the control of serotonin(2C) (5-HT(2C)) receptors in rat prefrontal cortex. J Pharmacol Exp Ther, 2003, 306, 734–743.

18. Filip M, Cunningham KA: Serotonin 5-HT(2C) receptors in nucleus accumbens regulate expression of the hyperlocomotive and discriminative stimulus effects of cocaine.

Pharmacol Biochem Behav, 2002, 71, 745–756.

19. Furst S, Riba P, Friedmann T, Timar J, Al-Khrasani M, Obara I, Makuch W et al.: Peripheral versus central antinociceptive actions of 6-amino acid-substituted de- rivatives of 14-O-methyloxymorphone in acute and in- flammatory pain in the rat. J Pharmacol Exp Ther, 2005, 312, 609–618.

20. Johnson SW, North RA: Opioids excite dopamine neurons by hyperpolarization of local interneurons. J Neuro- sci, 1992, 12, 483–488.

21. Kalivas PW, Duffy P, Eberhardt H: Modulation of A10 dopamine neurons by gamma-aminobutyric acid agonists. J Pharmacol Exp Ther, 1990, 253, 858–866.

22. Klitenick MA, DeWitte P, Kalivas PW: Regulation of so- matodendritic dopamine release in the ventral tegmental area by opioids and GABA: anin vivo microdialysis study. J Neurosci, 1992, 12, 2623–2632.

23. Kreek MJ, Bart G, Lilly C, LaForge KS, Nielsen DA:

Pharmacogenetics and human molecular genetics of opiate and cocaine addictions and their treatments. Pharma- col Rev, 2005, 57, 1–26.

24. Lamas X, Negus SS, Gatch MB, Mello NK: Effects of heroin/cocaine combinations in rats trained to discrimi-

nate heroin or cocaine from saline. Pharmacol Biochem Behav, 1998, 60, 357–364.

25. Litchfield JT, Wilcoxon F: A simplified method of evaluation dose-effect experiments. J Pharmacol Exp Ther, 1949, 96, 99–113.

26. Margreta-Mitrovic M, Mitrovic I, Riley RC, Jan LY, Bas- baum AI: Immunohistochemical localization of GABAB receptors in the rat central nervous system. J Comp Neu- rol, 1999, 405, 299–321.

27. Meldrum BS: Update on the mechanism of action of an- tiepileptic drugs. Epilepsia, 1996, 37, 4–11.

28. Morgan D, Cook CD, Picker MJ: Sensitivity to the discriminative stimulus and antinociceptive effects of mu opioids: role of strain of rat, stimulus intensity, and in- trinsic efficacy at the mu opioid receptor. J Pharmacol Exp Ther, 1999, 289, 965–975.

29. Negus SS, Mello NK, Fivel PA: Effects of GABA agonists and GABA₎receptor modulators on cocaine discrimination in rhesus monkeys. Psychopharmacology, 2000, 152, 398–407.

30. Newman JL, Vann RE, May EL, Beardsley: Heroin discriminative stimulus effects of methadone, LAAM and other isomers of acetylmethadol in rats. Psychopharma- cology, 2002, 164, 108–114.

31. Paul M, Dewey SL, Gardner EL, Brodie JD, Ashby CR Jr: Gamma-vinyl GABA (GVG) blocks expression of the conditioned place preference response to heroin in rats.

Synapse, 2001, 41, 219–220.

32. Pettit HO, Ettenberg A, Bloom FE, Koob GF: Destruc- tion of dopamine in the nucleus accumbens selectively attenuates cocaine but not heroin self-administration in rats. Psychopharmacology, 1984, 84, 167–173.

33. Platt DM, Grech DM, Rowlett JK, Spealman RD: Dis- criminative stimulus effects of morphine in squirrel monkeys: stimulants, opioids, and stimulant-opioid combinations. J Pharmacol Exp Ther, 1999, 290, 1092–1100.

34. Platt DM, Rowlett JK, Izenwasser S, Spealman RD:

Opioid partial agonist effects of 3-o-methylnaltrexone in rhesus monkeys. J Pharmacol Exp Ther, 2004, 308, 1030–1039.

35. Platt DM, Rowlett JK, Spealman RD: Discriminative stimulus effects of intravenous heroin and its metabolites in rhesus monkeys: opioid and dopaminergic mechanisms. J Pharmacol Exp Ther, 2001, 299, 760–767.

36. Przew³ocki R: Opioid abuse and brain gene expression.

Eur J Pharmacol, 2004, 500, 331–349.

37. Rowlett JK, Spealman RD, Platt DM: Cocaine-like discriminative stimulus effects of heroin in squirrel monkeys: role of active metabolites and opioid receptor mechanisms. Psychopharmacology, 2000, 150, 191–199.

38. Shaham Y, Stewart J: Effects of restraint stress and intra-ventral tegmental area injection of morphine and methyl naltrexon on the discriminative stimulus effects of heroin in the rat. Pharmacol Biochem Behav, 1995, 51, 491–498.

39. Shannon HE, Holtzman SG: Further evaluation of the discriminative effects of morphine in the rat. J Pharma- col Exp Ther, 1977, 201, 55–66.

40. Shippenberg TS, Elmer GI: The neurobiology of opiate reinforcement. Crit Rev Neurobiol, 1998, 12, 267–303.

(11)

41. Shoaib M, Spanagel R: Mesolimbic sites mediate the discriminative stimulus effects of morphine. Eur J Pharma- col, 1994, 252, 69–75.

42. Spanagel R, Herz A, Shippenberg TS: The effects of opioid peptides on dopamine release in the nucleus accumbens: anin vivo microdialysis study. J Neurochem, 1990, 55, 1734–1740.

43. Tsuji M, Nakagawa Y, Ishibashi Y, Yoshii T, Takashima T, Shimada M, Suzuki T: Activation of ventral tegmental GABAB receptors inhibits morphine-induced place preference in rats. Eur J Pharmacol, 1996, 313, 169–173.

44. Umans JG, Inturrisi CE: Pharmacodynamics of subcuta- neously administered diacetylmorphine, 6-acetylmorphine and morphine in mice. J Pharmacol Exp Ther, 1981, 218, 409–415.

45. Waszczak BL, Walters JR: Intravenous GABA agonist administration stimulates firing of A10 dopaminergic neurons. Eur J Pharmacol, 1980, 66, 141–144.

46. Wu D, Kang YS, Bickel U, Pardridge WM: Blood-brain barrier permeability to morphine-6-glucuronide is mark- edly reduced compared with morphine. Drug Metab Dis- pos, 1997, 25, 768–771.

47. Xi ZX, Fuller SA, Stein EA: Dopamine release in the nucleus accumbens during heroin self-administration is modulated by kappa opioid receptors: anin vivo fast- cyclic voltammetry study. J Pharmacol Exp Ther, 1998, 284, 151–61.

48. Xi ZX, Stein EA: Baclofen inhibits heroin self- administration behavior and mesolimbic dopamine release. J Pharmacol Exp Ther, 1999, 290, 1369–1374.

49. Xi ZX, Stein EA: GABAergic mechanisms of opiate reinforcement. Alcohol Alcohol, 2002, 37, 485–494.

50. Xi ZX, Stein EA: Increased mesolimbic GABA concen- tration blocks heroin self-administration in the rat.

J Pharmacol Exp Ther, 2000, 294, 613–619.

51. Xi ZX, Stein EA: Nucleus accumbens dopamine release modulation by mesolimbic GABA₎receptors-anin vivo electrochemical study. Brain Res, 1998, 798, 156–165.

Received:

September 16, 2005; in revised form: November 4, 2005.