Portal de Periódicos da CAPES

Comparison of nafadotride, CNQX, and haloperidol on acquisition versus expression of amphetamine-conditioned place preference in rats

2011; Lippincott Williams & Wilkins; Volume: 23; Issue: 1 Linguagem: Inglês

10.1097/fbp.0b013e32834ecb32

1473-5849

Tomek J Banasikowski, Lindsey S. MacLeod, Richard J Beninger,

Memory and Neural Mechanisms

Introduction Reward-related incentive learning refers to the acquisition by neutral stimuli paired with reward of an increased ability to elicit approach and other responses, and takes place when an animal encounters a biologically important (rewarding) stimulus (Bolles, 1972). It has been suggested that conditioned place preference (CPP) can be understood as involving incentive learning (Beninger, 1983). In this view, the side of the apparatus paired with the rewarding properties of the drug acquires an increased ability to elicit approach and other responses during conditioning, which manifests as CPP during testing. Although dopamine (DA) and glutamate have been previously implicated in incentive learning, their receptor subtypes may be differentially involved in acquisition versus expression (Beninger and Gerdjikov, 2004). For example, the D2 receptor-preferring antagonist pimozide blocked the acquisition of conditioned activity based on amphetamine at a dose that failed to block expression (Beninger and Hahn, 1983). In contrast, the D3 receptor-preferring partial agonist BP 897 that would act as a partial antagonist by preventing the full activation of D3 receptors by DA (Gyertyan et al., 2007), blocked the expression of a CPP based on amphetamine at a dose that failed to affect acquisition (Aujla and Beninger, 2005). We showed recently that the D3 receptor-preferring antagonist ABT-127 blocked the expression of conditioned activity based on cocaine at a dose that failed to block acquisition, and that the D2 receptor-preferring antagonist haloperidol blocked acquisition at a dose that failed to block expression, a double dissociation (Banasikowski et al., 2010). Our results with ABT-127 support behavioral findings reported with more selective D3-receptor antagonist like SB 277011A on cocaine-conditioned activity (Le Foll et al., 2002). Although further studies are needed, results suggest that D2 receptors may play a more important role in the acquisition and that D3 receptors may play a more important role in the expression of incentive learning. Fewer studies have systematically evaluated the role of N-methyl-D-aspartic acid (NMDA) and α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA)/kainate receptors in the acquisition and expression of incentive learning based on conditioning with amphetamine (see Beninger and Gerdjikov, 2004). Papp et al. (2002) found that the NMDA-receptor antagonist 1-aminocyclopropanecarboxylic acid blocked acquisition but not expression of CPP based on amphetamine. Further, Cervo and Samanin (1995) found that the noncompetitive NMDA-receptor antagonist MK 801 blocked the acquisition but not expression of CPP based on cocaine. In contrast, the AMPA/kainate receptor antagonist 6,7-dinitroquinoxaline-2,3-dione (DNQX) blocked expression but not acquisition of CPP based on cocaine (Cervo and Samanin, 1995). These findings suggest that NMDA receptors may play a more important role in acquisition and possibly that AMPA/kainate receptors may play a more important role in expression of incentive learning. In these series of experiments, we tested whether the DA D3 receptor-preferring antagonist nafadotride or the AMPA/kainate receptor antagonist 6-cyano-7-nitroquinoxaline-2,3-dione disodium salt (CNQX) would block the expression of CPP based on amphetamine at doses that fail to block acquisition, and that the D2 receptor-preferring antagonist haloperidol would block acquisition of CPP based on amphetamine at doses that fail to block expression. Smith-Roe and Kelley (2000) showed a synergistic role of DA D1-like receptors and NMDA receptors in incentive learning assessed in the acquisition of lever-pressing for food. Both D3 and AMPA/kainate receptors have been implicated in the expression of incentive learning. Therefore, we also tested the hypothesis that coadministration of subthreshold doses of nafadotride and CNQX will synergize to block expression. Methods Subjects Two hundred and eighty-six experimentally naive male albino Wistar rats (Charles River, St. Constant, Quebec, Canada) weighing 175–200 g on arrival were housed in pairs in clear polycarbonate cages. Rats were kept in a temperature-controlled colony room (21±1°C; humidity 40–70%) on a reversed 12 h light/dark schedule, with lights off at 07:00 h. Unlimited access to food (Lab Diet # 5001; PMI Nutrition International Inc., Brentwood, Missouri, USA) and water was maintained. All animals were treated in accordance with the guidelines and principles set by the Canadian Council on Animal Care and this experimental protocol was approved by the Queen’s University Animal Care Committee. Apparatus The CPP apparatus used is described in detail by Brockwell et al. (1996). CPP was monitored using four testing boxes each constructed with wooden sides and a hinged Plexiglas cover perforated for ventilation. Each box was contained within an outer plywood cabinet, which was insulated with sound-attenuating styrofoam and ventilated with a small fan. Individual boxes consisted of two rectangular chambers (38×27×34 cm high) connected by a tunnel (8×8×8 cm) that could be blocked to prevent movement between chambers by inserting two Plexiglas guillotine doors. All testing boxes had, on the right, a floor consisting of parallel wire bars that were spaced 1.0 cm apart and ran perpendicular to the tunnel and, on the left, a wire-mesh floor with openings of 1.0 cm2. In two of the testing boxes, the right chamber had brown urethane-sealed and Plexiglas-covered wooden walls and the left chamber had Plexiglas-covered walls with alternating 1-cm-wide black and white vertical stripes. In the other two boxes, the left chamber had the plain walls and the right chamber had the striped walls (see Brockwell et al., 1996). Each box was illuminated indirectly by a single 7.5 W incandescent light bulb placed 33 cm above the tunnel. Each box was outfitted with six pairs of infrared sensors: two located 5 cm above the floor of each chamber and two located 3 cm above the floor of the tunnel. Sensors were spaced so that they trisected the chambers and tunnel, with the beams from the sensors in the chamber running parallel to the ones in the tunnel. Beam breaks in each chamber and tunnel were used to measure activity levels during conditioning and time spent in each location during pre-exposure and test. Beam breaks were recorded from the sensors by a 6809 microcontroller (Motorola, Schaumburg, Illinois, USA) with custom-made software; data were then transferred to a personal computer. Behavioral procedure All experimental sessions occurred during the dark phase (07:00–19:00 h). The unbiased CPP experimental protocol consisted of three phases: three pre-exposure sessions, eight conditioning sessions, and one testing session. Animals in all groups received three 15-min pre-exposure trials over three consecutive days, during which no drug was administered and the tunnel doors were open to allow free movement between chambers. The conditioning phase began the next day and consisted of eight 30-min sessions with tunnel doors closed; four drug sessions in one chamber alternated daily with four vehicle sessions in the other. Rats were randomly assigned to conditioning chambers in a counterbalanced manner. Half of the rats were confined to the left side of the box on drug days and to the right on vehicle days. The other half of the rats were confined to the right side on drug days and left side on vehicle days, so that each chamber was drug-paired for some rats and vehicle-paired for others. The test phase took place 24 h following the last conditioning session. It consisted of a single 15-min session with the tunnel doors opened. Experiment 1 This experiment examined the effects of nafadotride on acquisition (given during the conditioning phase) and expression (given during the test phase) of CPP based on amphetamine. Experiment 1A investigated the effect of nafadotride on expression of amphetamine CPP. During the conditioning phase, rats (n=60) were injected intraperitoneally with amphetamine (2.0 mg/kg) on days 1, 3, 5, 7 and saline on days 2, 4, 6, 8 and were placed immediately into a conditioning chamber for 30 min. An additional control group (n=10) received saline on both sides during conditioning. On the test day, animals that were conditioned with amphetamine were pretreated with vehicle (n=12) or nafadotride (0.01, 0.1, 0.5, or 1.0 mg/kg; n=12/group) 30 min before placement into the CPP chambers. Saline controls were pretreated with saline 30 min before the test. On the basis of the findings in experiment 1A, we pretreated rats (n=24) in experiment 1B with nafadotride doses of 0.5 or 1.0 mg/kg (n=12/group) during the amphetamine conditioning days 30 min before placement into the CPP chambers. On the test day, rats received only vehicle injections and were placed into the CPP apparatus for a 15-min session. Experiment 2 We examined the effect of CNQX on acquisition and expression of CPP based on amphetamine. Experiment 2A investigated the effects of CNQX during the test phase. During the conditioning phase, rats (n=60) were injected intraperitoneally with amphetamine (2.0 mg/kg) on days 1, 3, 5, 7 and saline on days 2, 4, 6, 8 and placed immediately into a conditioning chamber for 30 min. On the test day, animals were pretreated with vehicle (n=12) or CNQX (0.01, 0.05, 0.1, or 1.0 mg/kg; n=12/group) 20 min before placement into the CPP chambers. On the basis of the findings in experiment 2A, we pretreated rats (n=48) in experiment 2B with CNQX doses of 0.05, 0.1, or 1.0 mg/kg (n=12, 12, 24) during the amphetamine conditioning days 20 min before placement into CPP chambers. On the test day, rats received only vehicle injections and were placed into the CPP apparatus for a 15-min session. Experiment 3 This experiment examined the possible synergistic effects of subthreshold doses of CNQX and nafadotride during the expression of CPP based on amphetamine. We used doses that previously did not block CPP expression in experiments 1A and 2A. On the 15 min test day, rats (n=12) received injections of nafadotride (0.1 mg/kg) 30 min before being placed into the apparatus and CNQX (0.01 mg/kg) 20 min before. Experiment 4 This experiment examined the effect of haloperidol on acquisition and expression of CPP based on amphetamine. In experiment 4A, we investigated the effects of haloperidol during the test phase. Rats (n=36) were conditioned with amphetamine and received haloperidol (0.05, 0.1, or 0.15 mg/kg, n=12/group) 1 h before placement into the CPP chambers on the test day. In experiment 4B, rats (n=36) were pretreated with haloperidol (0.05, 0.1, or 0.15 mg/kg, n=12/group) 1 h before placement into the CPP chambers for 30 min on days 1, 3, 5, and 7. Immediately before placement into the chamber rats received an injection of amphetamine (2.0 mg/kg) on days 1, 3, 5, 7 and saline on days 2, 4, 6, 8. On the test day, rats received only vehicle injections and were placed into the CPP apparatus for a 15-min session. Drugs Nafadotride (Tocris, Ellisville, Missouri, USA) and haloperidol (Sigma-Aldrich, Oakville, Ontario, Canada) were dissolved in 20% water solution of dimethyl sulfoxide (Sigma-Aldrich). CNQX (Tocris) was dissolved in distilled water and agitated for approximately 5 min over low heat. D-amphetamine sulfate (USP, Rockville, Maryland, USA) was dissolved in saline. All drugs were prepared fresh each day. Statistical analysis To evaluate the possibility of a side bias, we analyzed the times spent on the two sides during the pre-exposure phase using a paired-sample t-test in each experimental group. Two-variable (group×phase) mixed design analyses of variance (ANOVA) followed by tests of simple effects were used to analyze locomotor activity and place preference. The first analyses compared the saline control and the amphetamine-alone groups from experiments 1A and 2A. The results revealed that the two amphetamine-alone groups did not differ and they were combined for further analyses. Separate ANOVAs for each drug (including groups that received the drug in acquisition and those that received it in expression) for each of locomotor activity and place preference analyses included the saline and the combined amphetamine-alone groups. Significant interactions were further analyzed using tests of simple main effects. An α level of 0.05 was used for all analyses. All statistical tests were calculated using PASW 18.0 for Windows (IBM, Armonk, New York, USA). Results During the pre-exposure phase (averaged across the three sessions), there was no significant difference between the times spent on the two sides in any of the experiments (see Table 1). Thus, the procedure was unbiased.Table 1: Mean (±SEM) time (s) spent in the to-be-drug-paired and to-be-vehicle-paired side of the conditioned place preference apparatus, averaged over the three pre-exposure sessions for each group, showing that the procedure was unbiasedActivity during conditioning sessions With the exception of the saline control group (Fig. 1a) and some of the groups treated with haloperidol during acquisition (Fig. 1g), all groups showed greater activity during averaged amphetamine sessions than during averaged vehicle sessions of the conditioning phase (Fig. 1). For the saline control and the two amphetamine-alone groups, one from experiment 1A and one from experiment 2A (Fig. 1a), ANOVA revealed a significant interaction of group by drug [F(2,31)=6.98, P<0.01]. Simple effects of drug were significant in each of the amphetamine-alone groups [F(1,20)=110.79, P<0.001, F(1,20)=24.91, P<0.001, respectively] but not the saline control. The two amphetamine-alone groups were combined for further analyses.Fig. 1: Mean (±SEM) activity counts/30 min averaged over four conditioning sessions with saline (vehicle) or four conditioning sessions with amphetamine (amph, 2.0 mg/kg). Note that the doses shown on the x-axes of (b), (d), and (f) do not indicate drugs given during the conditioning phase; these values indicate the doses that were given during the test phase and are included here to identify the groups. (a) Groups that received nafadotride (NAF 0 mg/kg) or 6-cyano-7-nitroquinoxaline-2,3-dione disodium salt (CNQX; 0 mg/kg) in the test [amph/NAF (0) and amph/CNQX (0) groups], the group that received subthreshold doses of NAF (0.1) plus CNQX (0.01) in the test [amph/NAF (0.1)+CNQX (0.01)] and the control group that received saline on both sides during conditioning (saline). (b) Groups receiving NAF in the test phase (expression); (c) groups receiving NAF during conditioning (acquisition); (d) groups receiving CNQX in the test phase; (e) groups receiving CNQX during conditioning; (f) groups receiving haloperidol in the test phase; (g) groups receiving haloperidol during conditioning. *Significantly (P<0.05) different from saline by a test of simple effects of phase following observation of a significant phase by group interaction, in an analysis of variance that included the saline group and the combined data of the amph/NAF (0) and amph/CNQX (0) groups shown in (a).ANOVA for each drug, including the saline and combined amphetamine-alone groups, revealed a significant interaction of group×drug in every case: nafadotride [F(7,98) =10.02, P<0.001], CNQX [F(8,121)=7.19, P<0.001], subthreshold nafadotride plus CNQX [F(2,43)=25.74, P<0.001], and haloperidol [F(7,98)=16.11, P<0.001] (Fig. 1). Tests of simple effects of drug versus vehicle for each group revealed a significant effect in every case except for the saline group and haloperidol doses of 0.05 and 0.15. Thus, the interactions reflect the locomotor stimulant effect of amphetamine compared with saline during pairing days in experiments 1A and B, 2A and B, 3 and 4A, and the locomotor attenuating effects of haloperidol (0.05 and 0.15) during conditioning in experiment 4B. Conditioned place preference The change in time (s) spent in the drug-paired side from the average of three pre-exposures to test was used to evaluate place conditioning for each group. For the saline and two amphetamine-alone groups (Fig. 2a) ANOVA revealed a significant group×phase interaction [F(2,31)=4.54, P<0.05]. Simple effects of phase were significant in each of the amphetamine-alone groups [F(1,11)=8.08, P<0.05, F(1,11)=9.08, P<0.05, from experiment 1A and 2A, respectively] but not the saline controls. The two amphetamine-alone groups were combined for further analyses.Fig. 2: Mean (±SEM) time (s) spent in the drug-paired [amphetamine (amph) 2.0 mg/kg] side during pre-exposure and test for the groups that received nafadotride (NAF 0 mg/kg) or 6-cyano-7-nitroquinoxaline-2,3-dione disodium salt (CNQX) (0 mg/kg) in the test [amph/NAF (0) and amph/CNQX (0) groups in (a)], the group that received subthreshold doses of NAF (0.1) plus CNQX (0.01) in the test [amph/NAF (0.1)+CNQX (0.01) in (a)] and the control group that received saline on both sides during conditioning [saline in (a)]. (b) Groups receiving NAF in the test phase (expression); (c) groups receiving NAF during conditioning (acquisition); (d) groups receiving CNQX in the test phase; (e) groups receiving CNQX during conditioning; (f) groups receiving haloperidol in the test phase; (g) groups receiving haloperidol during conditioning. *Significantly (P<0.05) different from pre-exposure by a test of simple effects of phase following observation of a significant phase by group interaction in an analysis of variance that included the saline group and the combined data of the amph/NAF (0) and amph/CNQX (0) groups shown in (a).ANOVA for each drug, including the saline and combined amphetamine-alone groups, revealed a significant group×phase interaction in every case: nafadotride [F(7,98)=2.53, P<0.05], CNQX [F(8,121)=2.67, P<0.05], subthreshold nafadotride plus CNQX [F(2,43)=6.25, P<0.01], and haloperidol [F(7,98)=2.21, P<0.05] (Fig. 2b–g). These interactions indicate that the phase effect differed among groups, revealing that not all groups showed a significant CPP. Simple effects of phase for each group were therefore analyzed. For experiment 1A, nafadotride during expression (Fig. 2b), significant increases in time spent in the drug-paired chamber from pre-exposure to test were observed for the groups treated with doses of 0.01 [F(1,11)=18.10, P<0.01] and 0.1 [F(1,11)=4.87, P<0.05] but not 0.5 and 1.0 mg/kg. In experiment 1B, groups treated with nafadotride doses of 0.5 [F(1,11)=11.36, P<0.01] or 1.0 mg/kg [F(1,11)=15.43, P<0.01] during acquisition showed a significant increase in time spent on the drug-paired side (Fig. 2c). Thus, nafadotride doses of 0.5 and 1.0 mg/kg that blocked the expression of CPP when given during the test failed to block the acquisition of CPP when given during conditioning. For experiment 2A, CNQX during expression (Fig. 2d), a significant change in time spent on the drug-paired side from pre-exposure to test was observed for the group treated with 0.01 mg/kg [F(1,11)=21.96, P<0.01] but not 0.05, 0.1, or 1.0 mg/kg. In groups treated with CNQX during conditioning, significant CPP effects were found for 0.05 [F(1,11)=19.64, P<0.01], 0.1 [F(1,11)=6.48 P<0.05] but not for 1.0 mg/kg (Fig. 2e). Thus, CNQX doses of 0.05 and 0.1 mg/kg blocked the expression of CPP when given during the test but failed to block acquisition of CPP when given during conditioning. In experiment 3, we examined the possible synergistic effects of a subthreshold dose of nafadotride (0.1 mg/kg) and a subthreshold dose of CNQX (0.01 mg/kg) in the test (see Fig. 2a). We used a dose of nafadotride from experiment 1A and a dose of CNQX from experiment 2A that failed to block CPP when given in the test. A significant CPP effect was observed [F(1,11)=11.98, P<0.01]. Thus, a subthreshold dose of nafadotride (0.1 mg/kg) and a subthreshold dose of CNQX (0.01 mg/kg) given together failed to block the expression of CPP based on amphetamine. For experiment 4A, haloperidol during test (Fig. 2f), a significant change in time spent on the drug-paired side from pre-exposure to test was observed for the 0.05 [F(1,11)=8.73, P<0.05] and 0.10 mg/kg [F(1,11)=14.51, P<0.01] groups but not the 0.15 mg/kg group. When haloperidol was given during conditioning, significant changes in time spent on the drug-paired side from pre-exposure to test were observed for the groups treated with haloperidol 0.05 [F(1,11)=9.49, P<0.05] but not 0.10 or 0.15 mg/kg (Fig. 2g). Thus, a haloperidol dose of 0.10 mg/kg blocked the acquisition of CPP when given during conditioning but failed to block expression of CPP when given during the test. Discussion The results revealed that the D3 receptor-preferring antagonist nafadotride and the AMPA/kainate receptor antagonist CNQX blocked expression at doses that failed to block acquisition of CPP based on amphetamine. In contrast, the D2 receptor-preferring antagonist haloperidol blocked acquisition at a dose that failed to block expression of CPP based on amphetamine. The above results support our previous findings using cocaine-conditioned activity (Banasikowski et al., 2010). This suggests that D2 receptors play a more important role in the acquisition and that AMPA/kainate and D3 receptors play a more important role in the expression of incentive learning. Further, cotreatment with subthreshold doses of nafadotride and CNQX during the CPP test failed to block the expression of CPP. This suggests that the influence of D3 receptors and AMPA/kainate receptors is not additive in the expression of CPP. It should be noted that although nafadotride is not considered a selective, but a preferential, DA D3-receptor antagonist, the doses used in this study (<3.0 mg/kg) have previously been shown to act selectively on D3 receptors ex vivo (Levant and Vansell, 1997). Similarly, recent ex-vivo studies have shown that haloperidol is a D2-selective antagonist within the dose range used in this study (McCormick et al., 2010). The observation that nafadotride and CNQX blocked the expression of CPP when given during the test phase is unlikely to be attributable to the motor effects of these compounds, as previous studies have shown that at the doses used here they do not block motor activity (Maj et al., 1995; Sautel et al., 1995; Maldonado et al., 2007). The present study did not test nafadotride, CNQX, and haloperidol during conditioning to investigate if they had any aversive or rewarding effects when administered alone. However, others have shown that nafadotride lacks rewarding or aversive properties of its own (Chaperon and Thiebot 1996; Gyertyan and Gal, 2003), CNQX alone failed to produce CPP or conditioned place aversion (Maldonado et al., 2007) and that haloperidol lacks rewarding or aversive properties in the CPP paradigm (Spyraki et al., 1982; Gyertyan and Gal, 2003). The effects of CNQX and haloperidol on amphetamine-stimulated activity and change in time spent on the drug-paired side were dissociable. Thus, a CNQX dose of 1.0 mg/kg given with amphetamine during conditioning blocked the acquisition of CPP but not its stimulant effect. Haloperidol doses of 0.10 and 0.15 mg/kg given with amphetamine during conditioning blocked the acquisition of CPP but 0.10 failed to block the stimulant effect of amphetamine while 0.15 mg/kg did. In contrast, a haloperidol dose of 0.05 mg/kg blocked the stimulant effect of amphetamine but not the acquisition of CPP. We have frequently observed a dissociation of the effects of drug treatments on the stimulant versus incentive learning properties of amphetamine (Beninger et al., 2003). Our findings are consistent with previous studies examining the role of DA receptors in incentive learning. Beninger and Hahn (1983) found that the D2 receptor-preferring antagonist pimozide blocked the acquisition of conditioned activity based on amphetamine at a dose that failed to block the expression. In contrast, the D3 receptor-preferring partial agonist BP 897 blocked the expression of CPP and the expression of conditioned activity based on amphetamine at doses that failed to affect acquisition (Aujla et al., 2002; Aujla and Beninger, 2005). We recently reported that the D3 receptor-preferring antagonist ABT-127 blocked the expression of conditioned activity based on cocaine at a dose that failed to block acquisition and that haloperidol blocked acquisition at a dose that failed to block expression, a double dissociation (Banasikowski et al., 2010). This significant rightward shift of the D2-preferring receptor antagonist curves from acquisition to expression has been reported previously (Hiroi and White, 1991). Perhaps the higher doses that affected expression also affected D3 receptors, making the results consistent with D3-receptor antagonist data. Thus at appropriate dose levels, DA D2 receptor-preferring antagonists block the acquisition but not the initial expression of incentive conditioning based on amphetamine (Hiroi and White, 1991; Beninger et al., 2003). Our finding that CNQX blocked expression at doses that did not block acquisition is in support of Cervo and Samanin (1995) but not Kaddis et al. (1995). Cervo and Samanin (1995) found that the more selective AMPA/kainate receptor antagonist DNQX, administered intracerebroventricularly, blocked the expression of CPP based on cocaine with no effect on acquisition. In contrast, Kaddis et al. (1995) reported that DNQX administered into the nucleus accumbens blocked both acquisition and expression of CPP based on cocaine. As Kaddis et al. (1995) examined only a single dose of DNQX during acquisition and expression of CPP, their data do not allow an assessment of possible differential sensitivity to a range of doses of DNQX. It should be noted that the selectivity of CNQX for AMPA/kainate receptors and their role in CPP have been previously questioned (Mead and Stephens, 1999). However, our findings that AMPA/kainate receptors are involved in CPP based on amphetamine are consistent with earlier reports using more selective AMPA/kainate receptor antagonists (Cervo and Samanin, 1995; Kaddis et al., 1995). Cotreatment with subthreshold doses of nafadotride and CNQX during test failed to block the expression of CPP. The lack of synergism might suggest a parallel process between D3 receptors and AMPA/kainate receptors. Previous studies have shown an increase in synaptic AMPA/kainate receptors following learning (Malinow and Malenka, 2002), and D3 receptor expression was found to increase in mice showing conditioned activity following cocaine-environment pairings (Le Foll et al., 2003). These observations suggest that the substrate of incentive learning may include AMPA/kainate and D3 receptors but the mechanisms underlying the expression of incentive conditioning and especially, how D3 receptors are involved, remain to be elucidated (see Beninger and Gerdjikov, 2004). In summary, the role played by D3 receptors is similar to that played by AMPA/kainate receptors in acquisition and expression of CPP based on amphetamine. Thus, AMPA/kainate and D3 receptors seem to be a part of the underlying change that takes place when reward-related incentive learning occurs. Furthermore, D3 receptors appear to play a substantially different role than that played by D2 receptors in incentive learning. Unlike D2 receptors, which appear to contribute more importantly to establishing incentive learning, D3 receptors appear to contribute more importantly to the expression of incentive learning. Acknowledgements This study was funded by a grant from the Natural Sciences and Engineering Research Council of Canada to R.J.B. Conflicts of interest L.S. MacLeod and T.J. Banasikowski declare that, except for income received from our primary employer, no financial support or compensation has been received from any individual or corporate entity over the past 3 years for research or professional service, and there are no personal financial holdings that could be perceived as constituting a potential conflict of interest. R.J. Beninger has been compensated as an expert witness by Kenyon and Kenyon LLP (New York, New York, USA) acting for defendants Teva Pharmaceutical Industries Ltd. and Teva Pharmaceuticals USA Inc., and has carried out research under contract for Institut de Recherche Pierre Fabre (Boulogne, France).