Significance of glutamate and dopamine neurons in the ventral pallidum in the expression of behavioral sensitization to amphetamine

Life Sciences 68 (2001) 973–983

Significance of glutamate and dopamine neurons in the ventral pallidum in the expression of behavioral sensitization to amphetamine Jin-Chung Chen*, Kai-Wen Liang, Yi-Kung Huang, Cheng-Sheng Liang, Yao-Chang Chiang Laboratory of Neuropharmacology, Department of Pharmacology, Chang-Gung University, Tao-Yuan, Taiwan, R.O.C. Received 18 November 1999; accepted 12 July 2000

Abstract To explore the significance of ventral pallidum (VP) during the amphetamine sensitization, we first investigated if there are neurochemical alterations in the VP during amphetamine withdrawal period. Chronic amphetamine-treated (5 mg/kg 3 14 days) rats displayed an apparent locomotion sensitization as compared with saline controls when challenged with 2 mg/kg amphetamine at withdrawal days 10–14. A microdialysis analysis revealed that output of the dopamine metabolites, 3,4-dihydroxyphenylacetic acid and homovanillic acid, in the VP of amphetamine-sensitized rats increased approximately two-fold as compared to controls at both pre- and post-amphetamine challenge period. On the other hand, the in vivo glutamate output in the VP increased upon amphetamine challenge in the behaviorally sensitized rats, but not in the controls. To evaluate if drug manipulation in the VP would affect the behavioral sensitization, we treated both groups of rats with NMDA receptor antagonist, MK-801 (5 mg/ml for 5 days; bilateral) in the VP during withdrawal days 6–10. Animals were challenged with 2 mg/kg amphetamine at withdrawal day 11. The behavioral profile exhibited that MK-801 pre-treatment significantly blocked the locomotion hyperactivity in amphetamine-sensitized rats. Taken together, the current results suggest that the excitatory amino acid in the VP plays a significant role during the expression of behavioral sensitization to amphetamine. © 2001 Elsevier Science Inc. All rights reserved. Keywords: Amphetamine; Behavioral sensitization; Ventral pallidum; Dopamine; Glutamate

Introduction Animals that receive repetitive injections of amphetamine develop behavioral sensitization, a phenomenon that has been described as a behavioral augmentation to the subsequent * Corresponding author: Department of Pharmacology, Chang-Gung University, 259 Wen-Hwa 1st Rd., TaoYuan, Taiwan, R.O.C. Tel.: 1886-3-3283016 x 5282; fax: 1886-3-3283031. E-mail address: [email protected] (J.-C. Chen) 0024-3205/01/$ – see front matter © 2001 Elsevier Science Inc. All rights reserved. PII: S 0 0 2 4 - 3 2 0 5 ( 0 0 )0 0 9 9 5 -4

974

J.-C. Chen et al. / Life Sciences 68 (2001) 973–983

amphetamine challenge [29,32]. This behavioral pattern has been proposed to simulate the human psychosis that results from the compulsive use of psychostimulants, such as morphine, amphetamine and cocaine [31]. It is well known that the mesolimbocortical dopaminergic pathway plays a crucial role during the development of amphetamine sensitization [25]. The mesolimbocortical dopaminergic system originates in the ventral tegmental area (VTA) whose nerve terminals project mainly to the nucleus accumbens (NAc), but also reach the prefrontal cortex and other forebrain regions. According to previous evidence [3,15,26], it has been suggested that the neurons in the VTA are crucial for the induction of amphetamine sensitization, while the NAc (shell and core) is essential during the behavioral expression. Neurons in the NAc receive not only the dopaminergic innervation, but the excitatory amino acid inputs from the amygdala, prefrontal cortex and hippocampus. From the NAc, neurons send out fibers mainly to the ventral pallidum [VP] and also to the VTA and substantia nigra via GABAergic projections [6,21,33]. In contrast to the functionally well-characterized NAc, the physiological role of the VP and its underlying biochemistry are less understood. Some evidence suggests that neurons in the VP are involved in the acquisition of amphetamine conditioned place preference [7,8], cocaine self-administration [10] and in noveltyinduced behavioral activation [9]. It is, thus, proposed that the VP, which mediates both the motor activity and the rewarding effect of psychostimulants, serves as a major output relay of the VTA-NAc projection [28]. The administration of the GABAA receptor agonist muscimol into the VP suppressed the locomotor activity produced by local injections of amphetamine, enkephalin or dopamine into the NAc [2,22,24]. However, GABAA receptor blockade did not affect the motor activation caused by direct administration of dopamine into the VP [33]. Moreover, the injection of the GABAB receptor agonist baclofen into the VTA abolished the locomotor activity elicited by the excitatory amino acid agonist AMPA from the VP [13]. These results imply that signals mediated by dopamine, GABA and excitatory amino acids converge at the level of the VP for fine-tuning the intensity of locomotor activity initiated from the VTA-NAc axis. Considering that neurons in the NAc displayed profound changes during behavioral sensitization to amphetamine [5,19,26], we suspected the chronic amphetamine treatment would have significant impacts on the function(s) and/or neurochemistry of efferent neurons in the VP. The purpose of this study was to (1) characterize the significance of dopamine and amino acid transmitters in the VP during amphetamine sensitization and (2) test if pharmacological manipulations of the neurons in the VP could modulate the behavioral sensitization at ‘expression phase’ (late withdrawal period). The results suggest that the VP, similar to its participation in locomotor activity and conditioned place preference, plays a significant role in behavioral sensitization to amphetamine.

Materials and methods Animals Male Sprague-Dawley rats weighing 250–300 gm (Academic Sinica, Taipei, Taiwan) were used for the experiments. After arrival, they were acclimatized to a room with controlled ambient temperature (22 6 28C), humidity (50 6 10%) and a 12-h day-night cycle (lights on, 07.00–19.00 h) for at least 7 days before experimentation. The animals in most of the groups

J.-C. Chen et al. / Life Sciences 68 (2001) 973–983

975

studied were housed three to four per cage. For the microdialysis experiment, in order to avoid damage to the guide cannula, animals were caged individually after the implantation. The rats were given food (Western Lab Product 7001, Orange, CA) and water ad libitum. Drug treatment and behavioral assessment Rats were randomly assigned to either the chronic saline or chronic amphetamine groups. They received either saline (0.9% NaCl solution, i.p.) or amphetamine (5.0 mg/kg in saline, i.p.) once daily (09.00–10.00 h) for 14 consecutive days. To examine the behavioral sensitization to amphetamine during the withdrawal period, both groups of rats were challenged with either 2 mg/kg amphetamine or saline at withdrawal days 10–14. Locomotor activity was monitored with an electronic Animal Activity Cage (Model 7400, Ugo Basile, Italy) in an acoustically insulated room [5]. Each animal was accommodated to the cage (28 3 32 3 40 cm) for a period of 60 min prior to drug administration. Immediately after the amphetamine injection, the occurrence of horizontal locomotion was recorded every 10 min for a total behavioral session of 2 h (10.00–12.00 h). VP microdialysis For brain microdialysis, chronic amphetamine- or saline-treated animals were anesthetized with ketamine / xylazine (90 and 6 mg/kg i.p., respectively) and surgically implanted with a guide cannula at withdrawal day 3–5. The guide cannula was made from a 23G needle with a 30G stainless steel tubing stuck inside as a stylet. The whole setting was implanted at the rostral tip of the VP (mm: AP-0.3, ML12.5, reference to the bregma; DV17.2 from the skull) [27] with the help of a rat stereotaxic instrument (Model 900; David Kopf, Tujunga, CA). Animals were allowed at least 7 days for recovery. The perfusion was routinely performed during withdrawal days 10–14. On the day of the experiment, a homemade dialysis probe was inserted. The probe (Polymicro Technologies Inc.) was constructed as described by Wolf et al. [41] with 1 mm of exposed dialysis membrane (Spectra/Por@ hollow fiber, molecular weight cut-off: 13,000; Spectrum, Houston, TX) protruding 2–3 mm from the guide cannula. The dialysis solution (Krebs-Ringer phosphate buffer: 5 mM KCl, 120 mM NaCl, 1.2 mM CaCl2, 1.2 mM MgCl2, and 0.1 mM phosphate buffer, pH 7.4) was advanced through the probe via a PE-10 tube driven by a microsyringe pump (CMA 102; Sweden) at a flow rate of 1 ml/min. After 60 min of equilibration, two 20-min dialysis samples were collected to define the basal efflux. Afterwards, animals were systemically injected with 2 mg/kg amphetamine and samples were collected every 20 min for a continuous period of 2 h. The samples were acidified with a solution of HClO4 to a final concentration of 0.1 N and subjected immediately to an high-performance liquid chromatography (HPLC) system for the analysis of catecholamines and amino acid transmitters as described below. Following each experiment, the location of the probe was verified histologically in a series of 50-mm sections (752M Vibroslice; Campton Instruments, Sileby, UK). MK-801 microinjection After 14 days of treatment with saline or amphetamine, another set of animals was surgically implanted with a guide cannula into both sides of the VP as described above. Animals

976

J.-C. Chen et al. / Life Sciences 68 (2001) 973–983

were allowed 4 days for recovery. From withdrawal days 6 to 10, five daily injections of MK801 (Sigma Chemical Co., St. Louis, MO; 5 mg/1 ml over 2 min) or an equal volume of the vehicle (0.1 M phosphate buffered saline, pH 7.4) was administrated bilaterally into the VP through a 30-gauge stainless steel cannula with the help of the syringe pump (CMA 102). After each injection, the cannula was left in place for another 2 min to insure the local delivery of the drug. On withdrawal day 11, animals were examined for their motor response to a challenge with 2 mg/kg amphetamine. The procedures for amphetamine challenge and behavioral assessment were similar to those described above. Locomotor activity was recorded every 10 min for a total 2-h session. The location of injection site in each animal was verified histologically in a series of 50-mm sections after each experiment. HPLC analysis The HPLC system (Beckman 125, System Gold) was composed of a reverse-phase C-18 column (5 mm ODS; 3 3 10 cm; Beckman Instruments, Fullerton, CA), two gradient pumps, 406 interface and connected with an electrochemical (LC-4C; BAS Inc., West Lafayette, IN) or a fluorescence detector (Model 121: Gilson Company, Worthington, OH). The concentrations of 3,4-dihydroxyphenylacetic acid (DOPAC) and homovanillic acid (HVA) were determined with the electrochemical detector with the mobile phase consisting of 5 mM NaH2PO4, 30 mM citric acid, 0.1 mM EDTA, 0.02% sodium octylsulfate, and 9% methanol (pH 3.3–3.5). For the analysis of glutamate and aspartate, the perchloric acid-treated samples (final concentration: 0.1N) were reacted first with o-phthaldialdehyde (1:1, v/v) in the dark for 2 min prior to fluorescence detection. A binary gradient buffer system consisting of buffer A (40 mM sodium acetate and 10% methanol, pH 5.7) and buffer B (20% buffer A with 80% methanol, pH 6.7) was modified from Kendrick. [17]. To quantify the sample peaks, each chemical was compared with the external standards which were prepared freshly and injected every five sample runs. Statistics The behavioral and neurochemical data were analyzed by a two-way analysis of variance (ANOVA) followed by the post-hoc Dunnett’s multiple comparison test. The value P,0.05 was considered significant. Results Validation of behavioral sensitization to chronic amphetamine treatment Repetitive administration of 5 mg/kg amphetamine for 14 consecutive days resulted in a broad spectrum of behaviors. Typically, a rapid onset of the focused stereotypes (sniffing, head bobbing and licking) appeared first followed by an enhanced locomotor activity. The intensity and duration of the stereotypy increased with time after injection, reached maximal levels at days 5–7, and then remained high until day 14 (data not shown). On the other hand, the intensified locomotor activity decreased along with amphetamine injection time due to interference by the prolonged stereotypes. When animals were challenged with 2 mg/kg amphetamine at withdrawal days 10–14, the drug-induced locomotor activity was significantly

J.-C. Chen et al. / Life Sciences 68 (2001) 973–983

977

Table 1 Behavioral responses of control and sensitized rats to amphetamine (2 mg/kg) or saline challenge Groups Chronic saline 1 saline Chronic saline 1 AMPH Chronic AMPH 1 saline Chronic AMPH 1 AMPH

Locomotor activity (counts / 2 h) 344 6 54 2043 6 211** 629 6 107 3654 6 407**,1

Number 6 6 8 8

1

Animals were chronically treated with saline or amphetamine (5 mg/kg) for 14 consecutive days and challenged with 2 mg/kg amphetamine or saline at withdrawal days 10–14. 2 Asterisks (**) indicate significance (p,0.01) compared to the chronic saline 1 saline or chronic AMPH 1 saline group; 1 indicates significance (p,0.05) compared to the chronic saline 1 AMPH group.

greater in the chronic amphetamine-treated groups than in the chronic saline-treated controls, indicating clearly a behavioral sensitization (Table 1). In vivo output of dopamine metabolites, aspartate and glutamate in the VP of control and amphetamine-sensitized rats In order to evaluate the extracellular levels of neurotransmitters in the VP during behavioral sensitization, the VP of control and amphetamine-treated rats were dialyzed with KrebsRinger phosphate buffer at withdrawal days 10–14 followed by systemic challenge with 2 mg/kg amphetamine. In both groups, the locomotor activity began to increase at approximately 30 min after amphetamine injection and remained elevated until 90 min. The motor activity gradually returned to normal at approximately 2 hrs after drug treatment. As illustrated in the Fig. 1, the outputs of DOPAC and HVA, two acidic metabolites of dopamine, did

Fig. 1. In vivo outputs of (A) DOPAC and (B) HVA from the VP of amphetamine-treated rats and saline controls at withdrawal days 10–14 (n 5 7,7). After two samples at baseline, animals were challenged with 2 mg/kg amphetamine (arrowhead). Every 20-min, dialysis samples were collected for a total period of 2 h. Two-way ANOVA (treatment 3 time, with repeated measures over time) indicated there was a significant treatment 3 time interaction for both chemicals (F(7, 96) 5 27.13 for DOPAC and 15.64 for HVA; P,0.01). However, there was no time-effect after amphetamine injection in either group (F 5 0.87 and 1.33 in the control group; 1.56 and 0.97 in sensitized group, DOPAC and HVA respectively). *, significant difference between control and amphetamine group at that time point (P,0.05, Dunnett’s multiple comparison).

978

J.-C. Chen et al. / Life Sciences 68 (2001) 973–983

not respond to the amphetamine challenge in either the amphetamine-sensitized or control rats. Nevertheless, the outputs of DOPAC and HVA in amphetamine-sensitized rats displayed an approximately two-fold increase as compared to the saline controls at both pre- and postamphetamine challenge period. The in vivo dopamine outputs in both groups of the animals were below the detection limit (25 pg) throughout the experiment. On the other hand, there were no difference in basal aspartate (Fig. 2a) or glutamate (Fig. 2b) output between amphetamine- and saline-treated rats. The output of glutamate, however, increased significantly at 60 min (one-way ANOVA, P,0.05) in the behaviorally sensitized rats after amphetamine injection (Fig. 2b). The outputs of glutamate in control animals did not respond to the amphetamine challenge within the same time course. Comparing the individual post-drug intervals, there were significant difference between amphetamine-pretreated animals and controls at 60 and 80 min after amphetamine challenge. The response of aspartate, although there seemed to be a tendency of increase to the drug challenge in the sensitized rats, the values did not reach statistical significance in each post-drug periods in both groups (Fig. 2a). The locations of the dialysis probes in these animals are illustrated in Fig. 3. The effect of chronic MK-801 treatment in the VP on behavioral sensitization to amphetamine Figure 4 illustrates the effect of chronic MK-801 (5 mg/1 ml for 5 days) pretreatment into the VP bilaterally during the withdrawal period on behavioral profiles of the control and amphetamine-sensitized rats. Locomotor activity was recorded after a challenge with 2 mg/kg amphetamine at withdrawal day 11. The results showed that chronic MK-801 treatment did not affect the acute amphetamine-induced locomotor activity in control animals (Fig. 4a). However, the treatment significantly suppressed the locomotor activation in amphetaminesensitized rats (post-amphetamine 20–60 and 80–100 min). It is noteworthy that there was no

Fig. 2. In vivo outputs of (A) aspartate and (B) glutamate from the VP of amphetamine-treated rats and saline controls at withdrawal days 10–14 (n 5 10,10). After three samples at baseline, animals were challenged with 2 mg/kg amphetamine (arrowhead). Every 20-min, dialysis samples were collected for a total period of 140 min. Two-way ANOVA (treatment 3 time, with repeated measures over time) indicated there was a significant effect of drug treatment (F(1, 146) 5 4.33, P 50.011) and treatment 3 time interaction for glutamate (F(9, 146) 5 2.66, P 5 0.038). *, significant difference between control and amphetamine group at that time point (P,0.05, Dunnett’s multiple comparison). For aspartate, no main effects of treatment, time and treatment 3 time interaction were observed.

J.-C. Chen et al. / Life Sciences 68 (2001) 973–983

979

Fig. 3. Schematic diagram illustrates the needle tracks and probe placements in the VP for rats used in microdialysis study. The coronal sections were adapted from Paxinos and Watson’s rat atlas [31] and represent the levels corresponding to bregma 12.0 mm to 24.0 mm (left to right). Solid line: amphetamine-sensitized rats; dash line: saline controls.

apparent interference by stereotypes in behaviorally sensitized rats during the testing session. The site of injection in these experimental animals is illustrated in Fig. 5. Discussion The results of this study suggest that the VP, like its counterpart the NAc, plays a significant role in the behavioral sensitization to amphetamine. In the VP of amphetamine-sensitized rats, not only did the levels of dopamine metabolites increased but there appears to be an activation of excitatory amino acid upon amphetamine challenge. When amphetamine-sensitized rats were treated chronically with MK-801 in the VP during the withdrawal periods, the loco-

Fig. 4. Locomotor activity induced by 2 mg/kg amphetamine challenge in (A) controls (n 5 5,5) and (B) amphetamine-sensitized rats (n 5 6,6) with or without MK-801 pretreatment during withdrawal periods (see text for details). The behavior was recorded every 10 min for 2 h after amphetamine injection. A two-way ANOVA indicated a significant drug effect for the amphetamine-treated rats (F(11, 120) 5 0.19, P 5 0.0007), while no difference was observed in controls. *, significant difference between control and amphetamine 1 MK801 group (P,0.05, Dunnett’s multiple comparison).

980

J.-C. Chen et al. / Life Sciences 68 (2001) 973–983

Fig. 5. Schematic diagram illustrates cannula tip locations in both sides of the VP for amphetamine-sensitized rats used in MK-801 study. The coronal sections were adapted from Paxinos and Watson’s rat atlas [31] and represent the levels corresponding to bregma 12.0 mm to 24.0 mm (top to bottom). d, MK-801 treatment; s, saline treatment. The tip locations and MK-801 damage area in the control rats exhibit similar patterns compared to amphetamine-sensitized rats and were not shown.

motion sensitization to amphetamine was completely blocked. These findings suggest that the activities of dopaminergic and excitatory neurons in the VP have profound changes after chronic amphetamine treatment. Numerous studies have examined the neural substrates mediating the behavioral sensitization to amphetamine, particularly at the level of the VTA and the NAc [11,19,25,36]. The neural events underlying the induction of behavioral augmentation take place in the VTA while neurons in the NAc play a critical role during the expression of amphetamine sensitization [8,15,36]. It was reported that NAc neurons could be sensitized at early withdrawal times (days 1–3) as well as at late withdrawal periods (days 10–14 or longer). Sensitization of NAc neurons after long abstinent time indicates that an enduring change occurred in the NAc due to chronic amphetamine treatment [40]. This behavioral sensitization, however, could not be simply explained by the occurrence of enhanced dopamine release in the NAc since dopamine was increased only at late withdrawal days (i.e., . days 10–14, but not at days 1–3) upon amphetamine challenge [40]. Previously, we found that D2 dopamine receptors were down-regulated in the NAc during late amphetamine withdrawal times [5]. Considering that the majority of D2 receptors in the NAc are located on GABAergic projection neurons that send fibers mainly to the VP [6,9,12], we speculate that the alteration of accumbal D2 receptors would have a significant impact on neural activity in the VP. The projection from the NAc to the VP is known to be essential for the expression of motor behaviors elicited by environmental or pharmacological stimuli [18,22,37,38]. Several neurotransmitters, such as dopamine, glutamate and GABA have been identified in the VP which were responsible for the motor activation. Bilateral injections of GABA agonist, dopamine antagonist or excitatory amino acid antagonist into the VP attenuated the locomotorstimulating effect of amphetamine [13,22,33]. The hypermotility induced by glutamate in the

J.-C. Chen et al. / Life Sciences 68 (2001) 973–983

981

VP was mediated not only by the NMDA receptors, but also by a-amino-3-hydroxy-5methyl-isoxazole-4-propionate (AMPA) and kainic acid receptors [34,37]. In addition, the VP also participates the process of conditioned place preference, heroin and cocaine selfadministration and novelty-induced behavioral activation [7,8,9,10]. Hence, it is in general believed that the structure of VTA, NAc and VP fabricate the basic neural circuitry underlying both drug-induced rewarding and locomotor response. The finding that DOPAC and HVA increased in the VP of amphetamine-sensitized rats suggests that there was an increase in the activity of dopaminergic neurons in the VTA. Since both the NAc and VP receive dopaminergic innervation derived from the A10 cell groups [18,35], the current results along with others [32,40], clearly indicate that a general activation of A10 dopaminergic neurons in the forebrain results from chronic amphetamine treatment. To our knowledge, this is the first evidence that indicates dopaminergic activity increases in the VP during amphetamine sensitization. The results seem to be able to provide a molecular basis for recent findings that showed the VP could participate in amphetamine-induced conditioned place preference [7,8]. Surprisingly, the dopaminergic neurons in the VP did not response to the systemic challenge with amphetamine. Since it is known that the release of dopamine enhances during amphetamine challenge in the NAc [1,30], the insensitivity of dopamine neurons in the VP to amphetamine treatment suggests that there might be a subset of the A10 cell group innervates the VP. Alternatively, the dopamine-containing terminals in the VP might receive excess inhibitory input(s) upon amphetamine challenge. Interestingly, the amphetamine challenge induced increases of glutamate output in behaviorally sensitized rats, but not in the controls. The results indicate a supersensitivity of the VP glutamate neuron during the expression of amphetamine sensitization. It has been reported that afferents to the VP containing excitatory amino acids arise from the subthalamus and basolateral amygdala [4,23]. To determine whether the evoked glutamate efflux observed in this experiment was derived from either brain region requires further experimentation. Alternatively, to characterize the functional significance of glutamate in the VP during amphetamine sensitization, we treated both control and behaviorally sensitized rats with MK-801 in the VP. Since MK-801 did not affect the acute amphetamine-induced locomotor activity in the saline-treated rats, the MK-801 treatment, under our experimental condition, did not affect the locomotor activity per se. However, the present results reveal that the blockade of glutamatergic transmission through NMDA receptors in the VP during the drug withdrawal period prevents the behavioral sensitization to amphetamine. Previously, it was reported that systemic co-administration with MK-801 prevents the development of sensitization to amphetamine [13,16,39]. Taken together, both findings suggest that the MK-801 treatment could intervene the behavioral sensitization through an NMDA receptor-dependent event at both induction and expression stages. Further, a delayed dopamine response to amphetamine challenge was noticed when this event was compared with behavioral activation time in sensitized animals [14,20], suggesting there is a critical time frame for neural rewiring during drug abstinence. We speculated that subchronic MK-801 treatment in the VP during drug withdrawal period might disintegrate the neural plasticity at this particular time period that would be important for the expression of behavioral sensitization. The evidence that blockade of NMDA receptors at post-NAc levels impedes the expression of amphetamine sensitization could provide us with an alternative means to intervene in drug addiction.

982

J.-C. Chen et al. / Life Sciences 68 (2001) 973–983

In summary, we have demonstrated that there are significant neurochemical changes in the VP during behavioral sensitization to amphetamine. The neurotransmission of dopamine and excitatory amino acid in the VP of this particular physical stage apparently display a different pattern from that reported for the NAc and/or VTA. It appears that to develop a behavioral sensitization to amphetamine, the neural circuitry in the central nervous system must recruit a series of brain structures including the VP. Further experimentation would be required to identify the source of the glutamate and determine if other glutamate receptor subtype(s) (non-NMDA or metabotropic) in the VP are required for the development of amphetamine sensitization. Acknowledgment We thank Dr. Jerome L. Maderdrut (Tulane University) for his critical review of this manuscript. The research was supported in part by National Science Council (88-2314-B182-032) in Taiwan and also by Chang-Gung Memorial Hospital research grant (CMRP992). References 1. AKIMOTO K, HAMAMURA T, KAZAHAYA Y, AKIYAMA K, OTSUKI S. Brain Research 1990; 507: 344–6. 2. AUSTIN MC, KALIVAS PW. Japanese Journal of Pharmacology 1989; 50:487–90. 3. CADOR M, JIJOU Y, STINUS L. Neuroscience 1995; 65:385–95. 4. CARNES KM, FULLER TA, PRICE RJ. J. Comparative Neurology 1990; 302:824–52. 5. CHEN JC, SU HJ, HUANG LI, HSIEH MC. Life Science 1998; 64:343–54. 6. GRENEWEGEN HJ, BERENDSE HW, HABER SN. Neuroscience 1993; 57:113–42. 7. GONG W, NEILL D, JUSTICE JB. Brain Research 1997; 754:103–12. 8. HIROI N, WHITE NM. Neuroscience Letters 1993; 156:9–12. 9. HOOK MS, KALIVAS PW. Neuroscience 1995; 64:587–97. 10. HUBNER CB, KOOB GF. Brain Research 1989; 508:20–9. 11. HYMAN SE. Neuron 1996; 16:901–4. 12. INGLIS WL, DUNBAR JS, WINN P. Neuroscience 1994; 62:51–64. 13. JOHNSON K, CHURCHILL L, KLITENICK MA, HOOKS MS, KALIVAS PW. Journal of Pharmacology and Experimental Therapeutics 1996; 277:1122–31. 14. KALIVAS PW, DUFFY P. Journal of Neuroscience 193; 13:266–75. 15. KALIVAS PW, STEWART T. Brain Research Review 1991; 16:223–44. 16. KARLER R, CALDER LD, CHAUDHRY IA, TURKANIS SA. Life Sciences 1989; 45:599–606. 17. KENDRICK KM. Current Separation 1993; 12:1–5. 18. KLITENICK MA, DEUTCH AY, CHURCHILL L, KALIVAS PW. Neuroscience 1992; 50: 371–86. 19. KOOB F, SANNA PP, BLOOM FE. Neuron 1998; 21:467–76. 20. KURIBARA H. Pharmacology, Biochemistry and Behavior 1995; 52:759–63. 21. McGEORGE AJ, FUALL RLM. Neuroscience 1989 29:503–37. 22. MELE A, THOMAS DN, PERT A. Neuroscience 1998; 82:43–58. 23. NAPIER TC. Synapse 1992; 10:110–9. 24. NAPIER TC, POTTER PE. Neuropharmacology 1989; 28:757–60. 25. NESTLER EJ. Neuroscientist 1995; 1:212–20. 26. NESTLER EJ, AGHAJANIAN GK. Science 1997; 278:58–63. 27. PAXINOS G, WATSON C. The Rat Brain in Stereotaxic Coordinates, New York: Academic Press, 1986. 28. ROBERSTON GS, JIAN M. Neuroscience 1995; 64:1019–34.

J.-C. Chen et al. / Life Sciences 68 (2001) 973–983 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41.

983

ROBINSON TE, BECKER JB. Brain Research Review 1996; 11:157–98. ROBINSON TE, JURSON PA, BENNETT JA, BENTGEN TM. Brain Research 1988; 462: 211–22. SATO M, CHEN CC, AKIYAMA K, OTSUKI S. Biological Psychiatry 1983; 18:429–40. SEGAL DM, KUCZENSKI R. Journal of Pharmacology and Experimental Therapeutics 1997; 282:561–73. SHREVE PE, URETSKY NJ. Neuropharmacology 1988; 27:1271–7. SHREVE PE, URETSKY NJ. Pharmacology, Biochemistry and Behavior 1989; 34:101–6. SWANSON LW. Brain Research Bulletin 1982; 9:321–53. VEZINA P. Journal of Neuroscience 1996; 16:2411–20. WILLINS DL, WALLACE LJ, MILLER DD, URETSKY NJ. Journal of Pharmacology and Experimental Therapeutics 1992; 260:1145–51. WOLF ME. Progress in Neurobiology 1998; 54:679–720. WOLF ME, KHANSA MR. Brain Research 1991; 562:164–8. WOLF ME, WHITE FJ, NASSAR R, BROODERSON RJ, KHANSA R. Journal of Pharmacology and Experimental Therapeutics 1993; 264:249–55. WOLF ME, XUE CJ, WHITE FJ, DAHLIN SL. Brain Research 1994; 666:223–31.