Roles of the Infralimbic Cortex and Nucleus Accumbens Shell

European Journal of Neuroscience

European Journal of Neuroscience, Vol. 35, pp. 614–622, 2012 doi:10.1111/j.1460-9568.2012.07991.x

BEHAVIORAL NEUROSCIENCE

Neural circuit competition in cocaine-seeking: roles of the infralimbic cortex and nucleus accumbens shell

Ryan T. LaLumiere,1 Kyle C. Smith2 and Peter W. Kalivas2 1Department of Psychology, University of Iowa, Iowa City, IA 52242, USA 2Neurobiology of Addiction Research Center, Department of Neurosciences, Medical University of South Carolina, Charleston, SC, USA

Keywords: basolateral amygdala, prelimbic cortex, rat, reinstatement, ventral tegmental area

Abstract Following cocaine self-administration and extinction training, activity in the infralimbic cortex (IL) suppresses cocaine-seeking behavior. IL inactivation induces cocaine-seeking whereas activation suppresses cocaine-reinstated drug-seeking. We asked how the suppression of cocaine-seeking induced by IL activation integrates with the circuitry promoting reinstated cocaine-seeking. Following cocaine self-administration and extinction training, rats underwent cue-induced reinstatement. In order to activate IL projections, microinjections of PEPA, a positive allosteric modulator of AMPA receptors, were made into the IL in combination with microinjections into a variety of nuclei known to regulate cocaine-seeking. Intra-IL PEPA administration suppressed cue-induced reinstatement without affecting locomotor activity. The suppression of cocaine-seeking was reversed by activating dopamine neurons in the ventral tegmental area with microinjections of the l-opioid receptor agonist DAMGO, and was partially reversed by dopamine microinjections into the prelimbic cortex or basolateral amygdala. Previous evidence suggests that the nucleus accumbens shell both promotes and suppresses cocaine-seeking. The suppression of cue-induced cocaine seeking by PEPA in the IL was reversed by intra-shell microinjections of either dopamine or the AMPA receptor antagonist CNQX, suggesting that the accumbens shell bidirectionally regulates cocaine-seeking depending on whether dopamine input is mimicked or glutamate input is inhibited. Together, these ﬁndings indicate that the IL acts ‘upstream’ from structures promoting cocaine-seeking, including from the mesolimbic dopamine projections to the prelimbic cortex and basolateral amygdala, and that the accumbens shell may be a crucial point of integration between the circuits that promote (ventral tegmental area) and inhibit (IL) reinstated cocaine-seeking.

Introduction Inactivation studies reveal that reinstatement of cocaine-seeking seeking without any other reinstatement trigger. Moreover, recent requires activity in a variety of structures including the ventral findings suggest that the IL is involved in the consolidation of tegmental area (VTA), the basolateral amygdala (BLA), the dorsal memories for the extinction of cocaine-seeking behavior, confirming a portion of the medial prefrontal cortex (prelimbic cortex, PL) and the role for this structure in the extinction of cocaine-seeking (LaLumiere nucleus accumbens core (NAcore; McFarland & Kalivas, 2001; et al., 2010). McLaughlin & See, 2003; Di Ciano & Everitt, 2004). Reinstated drug- Cocaine-seeking induced by IL inactivation is reversed by concur- seeking is a widely employed model of drug relapse (Shaham et al., rent inactivation of the PL or BLA, suggesting that, during the 2003), and typically involves drug self-administration followed by expression of extinction, activity in the IL actively competes with the extinction training, after which drug-seeking is reinstated. As extinc- circuitry underlying cocaine-seeking (Peters et al., 2008). In support, tion involves new learning rather than erasure of the original memory intra-IL microinjections of AMPA suppress cocaine-reinstated drug- (Bouton, 2002), extinction learning and its expression involves a seeking, though it is not known whether such activation suppresses neural circuit that suppresses drug-seeking. Akin to findings with the cue-induced reinstatement. Although the NAshell is also involved in extinction of fear conditioning (Quirk et al., 2006), recent work the suppression of cocaine-seeking behavior (Di Ciano et al., 2008; identified the ventral portion of the medial prefrontal cortex (infra- Peters et al., 2008), other work suggests that the NAshell promotes limbic cortex; IL), as well as the IL’s efferent target, the nucleus cocaine-seeking behavior. NAshell blockade of either D1 or D2 accumbens shell (NAshell), as critically involved in the suppression of receptors or AMPA receptors prevents cocaine-induced reinstatement reinstated cocaine-seeking (Peters et al., 2008). Inactivation of the IL (Anderson et al., 2003, 2006; Famous et al., 2008). One possible or NAshell immediately prior to an extinction session induces cocaine- explanation for the discrepant results is that the NAshell is a site of competition between the IL and dopaminergic afferents in regulating cocaine-seeking. Thus, depending on the activated inputs, the NAshell Correspondence: Dr Ryan T. LaLumiere, as above. can bidirectionally determine whether rats engage in cocaine-seeking. E-mail: [email protected] Therefore, the present experiments examined whether IL activation Received 18 August 2011, revised 25 November 2011, accepted 7 December 2011 suppresses cue-induced cocaine-seeking and whether this could be

ª 2012 The Authors. European Journal of Neuroscience ª 2012 Federation of European Neuroscience Societies and Blackwell Publishing Ltd Competition over cocaine-seeking 615 reversed by increased dopaminergic tone in mesolimbic dopamine Prior to drug self-administration, all rats were food-deprived for 24 h terminal fields important in reinstated cocaine-seeking. Additionally, and then underwent a single 15-h food-training session in which the we evaluated whether the NAshell could bidirectionally regulate rats were trained to press the active lever for a single food pellet cocaine-seeking. (45 mg; Noyes, Lancaster, NH, USA) on a fixed-ratio 1 (FR1) schedule. Following the food training, the rats were given a limited quantity of food (20 g) immediately after every self-administration session, and this food restriction was maintained through extinction Materials and methods and reinstatement testing. One day after the food training, the rats Subjects began cocaine self-administration and each session lasted 2 h or until Male Sprague–Dawley rats (300 g at the time of surgery; Charles the rats had taken a maximum of 200 infusions. The self-administra- River, n = 135) were used in this study. They were individually tion program was an FR1 schedule with a 20-s timeout to prevent housed, maintained in a temperature-controlled environment (22 C) overdose. Each active lever press produced a 0.05 mL infusion of on a 12-h light–dark cycle (lights on at 07.00 h) with food and water 200 lg of cocaine (dissolved in 0.9% sterile saline; cocaine-HCl ad libitum, and given 6–7 days to acclimatize to the vivarium before kindly provided by NIDA). Concurrent with the drug infusion, a cue undergoing surgery. Behavioral procedures began 5–6 days after tone (2900 Hz) and cue light immediately above the active lever surgery. All methods used were in compliance with NIH guidelines for turned on for 5 s. Rats underwent self-administration 6 days per week, care of laboratory animals and were approved by the Medical excluding Sundays, for at least 2 weeks (i.e. 12 days). In order to University of South Carolina Institutional Animal Care and Use move into extinction, rats were required to take at least 10 days of at Committee or the University of Iowa Institutional Animal Care and least 10 infusions of cocaine, including the last 3 days of self- Use Committee. administration, as well as discrimination between the active and inactive lever. If rats had not met these criteria by 4 weeks (i.e. 24 days), they were excluded from the study. Surgery After self-administration, rats began extinction training. Active lever presses produced no drug infusion or light ⁄ tone cues. For all The rats were anesthetized with ketamine HCl (87.5 mg ⁄ kg, i.m.) and experiments, rats underwent at least 10 days of extinction and, in xylazine (5 mg ⁄ kg, i.m.). Ketorolac (3 mg ⁄ kg, i.p.) was administered order to undergo reinstatement, rats had to have 25 or fewer active prior to surgery to provide analgesia and again 1 day after surgery. For lever presses for at least two consecutive days immediately prior to the catheter implantation, a 13-cm piece of Silastic tubing (0.51 mm inner reinstatement session. The 2 days of extinction prior to reinstatement diameter, 0.94 mm outer diameter; Dow Corning) with a silicone ball testing served as the extinction baseline. Rats were excluded if they placed 3 cm from the end was threaded under the skin from the back did not reach extinction criterion within 40 days. For cue-induced to the ventral side of the rat. The catheter was inserted into the jugular reinstatement, the session program was identical to the self-adminis- vein and secured with sutures. The other end was externalized through tration session program (i.e. cue tone and light turned on with an a small hole in the skin between the shoulder blades. The externalized active lever press). In one experiment, rats also underwent cocaine- end was connected to a 22-gauge guide cannula (22 ga, C313G; induce reinstatement, in which they received a single i.p. injection of Plastics One, Inc., Wallingford, CT, USA) that was secured in the cocaine (10 mg ⁄ kg) immediately prior to the reinstatement session. middle of a harness (Strategic Applications, Inc., USA) that looped Rats did not receive intravenous drug infusions during either around the forelegs of the rat and thus remained securely attached to reinstatement. Rats’ lever-pressing was re-extinguished between the rat. Rats were given daily infusions of cefazolin and heparin successive reinstatement sessions with the same criteria described (0.1 mL of 100 U ⁄ mL) for 1 week after surgeries and then infusions above and a minimum of 3 days of extinction between reinstatements. of heparin alone throughout the remainder of the self-administration. The rats were then placed in a small animal stereotaxic instrument (Kopf Instruments, Tujunga, CA, USA). Four surgical screws were implanted into the skull as anchors, and guide cannulae were Microinjection procedures implanted and secured with dental cement. The nose bar was Microinjectors (33-gauge) were connected to PE20 tubing, which was maintained at -3.5 mm relative to the interaural line. All AP and attached to 10-lL Hamilton syringes controlled by an infusion pump. ML coordinates were calculated from Bregma, whereas DV coordi- For the BLA, VTA and NAshell, the microinjectors extended 2 mm nates were calculated from skull surface. The surgical coordinates beyond the end of the cannula and into the structure of interest. For the were as follows: PL ⁄ IL (double-barreled cannulae) – AP, +3.1 mm; PL and IL, the microinjectors extended 1 and 3 mm, respectively, ) ) ) DV, 2.6 mm; BLA – AP, 2.8 mm; ML, ±5.0 mm; DV, 6.5 mm; beyond the end of the cannula. Rats were not anesthetized during ) ) VTA – AP, 5.2 mm; ML (10 angle), ±2.2 mm; DV, 7.1 mm; microinjections. All microinjections (0.3 lL) occurred over 60 s, and ) NAshell – AP, +1.7 mm; ML (17 angle), ±3.0 mm; DV, 5.8 mm. the injectors were left in place for an additional 60 s to allow for After the surgery, the rats were retained in an incubation chamber until diffusion from the site of the injection. Following the microinjection, they recovered from the anesthesia. Obdurators were placed in all rats then underwent their appropriate reinstatement testing. The cannulae and maintained through the reinstatement testing. The rats following drugs were used in the microinjections – the positive were then returned to their home cages and checked on the days modulator of AMPA receptors 2-[2,6-Difluoro-4-[[2-[(phenylsulfo- following surgery to ensure that their wounds healed. nyl)amino]ethyl]thio]phenoxy]acetamide (PEPA; 0.075 nmol ⁄ side), CNQX (1 nmol ⁄ side), dopamine (25 nmol ⁄ side for PL, 75 nmol ⁄ side for BLA, 100 nmol ⁄ side for NAshell), and [D-Ala2, NMe-Phe4, Gly- Self-administration, extinction and reinstatement procedures ol5]-enkephalin (DAMGO; 0.1 nmol ⁄ side). All drugs were dissolved All self-administration experiments occurred in standard operant in artificial cerebrospinal fluid (aCSF) except PEPA, which was chambers (Med Associates, Fairfield, VT, USA) with two retractable dissolved initially in 50% DMSO ⁄ 50% aCSF and was then diluted to levers, a house light, and a cue light and tone-generator (2900 Hz). 10% DMSO ⁄ 90% aCSF immediately prior to use. Doses of all four

ª 2012 The Authors. European Journal of Neuroscience ª 2012 Federation of European Neuroscience Societies and Blackwell Publishing Ltd European Journal of Neuroscience, 35, 614–622 616 R. T. LaLumiere et al. drugs were chosen based on prior work (Hooks & Kalivas, 1994; between extinction baseline and reinstatement and between the McFarland & Kalivas, 2001; LaLumiere et al., 2005; Zushida et al., reinstatements of the group receiving PEPA into the IL and vehicle 2007; LaLumiere & Kalivas, 2008). into the other structure and the group receiving PEPA into the IL and a drug into the other structure. The locomotor activity data were analyzed using a two-way repeated-measures anova. Experiments In all experiments, rats underwent two cue-induced reinstatements in a counterbalanced manner with respect to microinjection. In the first Results experiments, rats received microinjections of PEPA or vehicle into the A total of 89 rats were included in the final data, while another 46 IL immediately prior to the cue-induced reinstatement session. In all were excluded due to failure to acquire self-administration (17 rats), subsequent experiments, rats received PEPA or vehicle into the IL as failure to extinguish lever pressing (three rats), incorrect cannula well as microinjections into an additional structure. In the second and placements (15 rats), more inactive lever pressing than active lever third experiments, rats received dopamine into the PL or BLA, pressing at the end of self-administration (two rats), death prior to the respectively. In the fourth experiment, rats received the l-opioid end of the experiment (seven rats) and clogged cannulas (two rats). receptor agonist DAMGO into the VTA. In the fifth and sixth Figure 1A shows active and inactive lever presses during self- experiments, rats received dopamine or CNQX into the NAshell. In administration from all animals included in the final data. Within each addition, a subset of rats (n = 11) was used in the locomotor activity experiment there was no significant difference between microinjection test to determine whether PEPA administration into the IL affected treatment groups in average active lever pressing during the last locomotor activity. 3 days of self-administration (data not shown).

Locomotor activity PEPA microinjections into the IL suppressed cue-induced Following completion of the reinstatement studies, a subset of rats cocaine-seeking underwent locomotor activity testing in order to determine whether Figure 1B shows that intra-IL microinjections of PEPA immediately intra-IL microinjections of PEPA affect locomotor activity. The prior to a cue-induced reinstatement session significantly reduced behavioral chambers contained 16 photobeams (AccuScan Instru- active lever-pressing compared with that found in vehicle-control rats. ments). Rats were given microinjections of PEPA or its vehicle into A one-way anova identified a significant main effect (F2,23 = 15.87, the IL immediately prior to being placed in a chamber for a 2-h P < 0.001). Post hoc tests revealed that rats receiving vehicle had locomotor activity test. significantly more active lever presses during the cue-induced reinstatement compared to the extinction baseline (P < 0.0001). Those rats receiving PEPA had significantly more active lever presses Histology and statistics compared to the extinction baseline but significantly fewer active lever Rats were overdosed with sodium pentobarbital (100 mg ⁄ kg, i.p.) and presses compared to those of the vehicle-treated rats (P < 0.05). Thus, intracardially perfused with 0.9% saline. The brains were removed and intra-IL microinjections of PEPA reduced, but did not completely stored in 10% formalin for at least 24 h. Coronal sections (100 lm abolish, cue-induced cocaine-seeking. There were no significant thick) were made using a vibratome. Sections were mounted on gel- differences in active lever presses between rats receiving PEPA as coated slides and then stained with Cresyl violet. Sites of injector the first or second microinjection (n = 4 for each; mean ± SEM – first, needles were verified with a light microscope, according to the atlas of 23.25 ± 6.61; and second, 32.25 ± 1.97). Similarly, when data were Paxinos & Watson (2005). Lever-pressing data were analyzed using pooled for all rats receiving PEPA alone across all experiments, there one-way anova. For the first experiment, Bonferroni post hoc tests were no significant differences in active lever presses between rats were conducted, while the remaining experiments used Bonferroni receiving PEPA as the first or second microinjection (n = 19 for each; adjustments to the overall alpha level of 0.05 for planned comparisons mean ± SEM – first, 23.84 ± 4.60; and second, 16.37 ± 2.9). For rats

150 100 6000 A Active lever B C (n = 89) Vehicle (n = 5) Inactive lever PEPA (n = 6) 100 4000 50 * + 50 2000 Lever presses Active lever presses

–12–11–10 –9 –8 –7 –6 –5 –4 –3 –2 –1 Horizontal activity counts Self-administration days prior to extinction Time (min)

Fig. 1. Suppression of cue-induced cocaine-seeking by intra-IL microinjections of PEPA. (A) Active and inactive lever presses during last 12 days of cocaine self- administration. (B) Active lever presses (mean ± SEM) during extinction and cue-induce reinstatement. Intra-IL microinjections of PEPA immediately prior to a cue- induced reinstatement session suppressed cocaine-seeking behavior. (C) Horizontal activity counts (mean ± SEM) measured in an open-ﬁeld test. Intra-IL microinjections of PEPA immediately prior to the open-ﬁeld test did not affect the rats’ locomotor activity. *P < 0.0001 compared to extinction baseline; +P < 0.05 compared to extinction baseline and to vehicle group.

ª 2012 The Authors. European Journal of Neuroscience ª 2012 Federation of European Neuroscience Societies and Blackwell Publishing Ltd European Journal of Neuroscience, 35, 614–622 Competition over cocaine-seeking 617 shown in Fig. 1B, there were no significant differences among the comparisons revealed significant differences in active lever pressing groups for inactive lever presses during reinstatement (data not between the extinction baseline and all groups except the group shown). PEPA microinjections into the IL alone before being placed receiving PEPA into the IL and vehicle into the VTA (P < 0.05). into an activity chamber did not affect the rats’ horizontal activity Active lever pressing in rats receiving PEPA was significantly lower compared to vehicle-control rats (Fig. 1C). A two-way repeated- than that found in rats receiving PEPA into the IL and DAMGO into measures anova revealed there was a significant effect of time the VTA (P < 0.05), indicating that DAMGO microinjections into the (F11,99 = 26.82, P < 0.001) but no significant effect of drug VTA reversed the suppression induced by PEPA microinjections into (F1,99 = 0.2836, P > 0.05) or interaction between the drug and time the IL. When microinjected alone into the VTA to a subset of rats (F11,99 = 0.3507, P > 0.05). (n = 5) immediately prior to an extinction session with no cues, this We then examined whether the suppression of cocaine-seeking was dose of DAMGO did not significantly increase active lever presses. reversed by microinjections of dopamine into the PL (Fig. 2A) or into There were no significant differences in inactive lever presses during the BLA (Fig. 2B). For the PL experiment, a one-way anova reinstatement (data not shown). revealed a significant main effect (F4,50 = 4.478, P < 0.01). Planned comparisons found that all groups significantly reinstated compared to their extinction baseline except those rats receiving PEPA into the IL The NAshell regulated cue-induced cocaine-seeking and vehicle into the PL (P < 0.05). However, active lever pressing for The NAshell receives glutamatergic input from the IL and dopami- rats receiving PEPA into the IL did not significantly differ from those nergic input from the VTA (Fallon & Moore, 1978; Sesack et al., receiving PEPA into the IL and dopamine into the PL. For the BLA 1989), and previous evidence suggests that the NAshell can bidirec- experiment shown in Fig. 2B, a one-way anova revealed a significant tionally regulate reward-seeking (Schmidt et al., 2006; Peters et al., main effect (F4,56 = 6.295, P < 0.001). Planned comparisons found 2009). We therefore investigated whether stimulating dopamine that all groups significantly reinstated compared to their extinction receptors or blocking AMPA in the NAshell might differentially baseline except those rats receiving PEPA into the IL and vehicle into affect the capacity of PEPA activation of glutamatergic afferents to the BLA (P < 0.05). As with the PL, active lever pressing for rats inhibit cue-reinstated cocaine seeking. In the first experiment, intra- receiving PEPA into the IL did not significantly differ from those NAshell microinjections of dopamine reversed the suppression of cue- receiving PEPA into the IL and dopamine into the BLA. While it has induced cocaine-seeking by PEPA (Fig. 4A). A one-way anova previously been shown that dopamine alone into the PL reinstates revealed a significant main effect (F4,41 = 5.666, P < 0.01). Planned cocaine-seeking (McFarland & Kalivas, 2001), the effect of dopamine comparisons revealed significant differences in active lever presses in the BLA is not known, although microinjection of a D1 antagonist between the extinction baseline and all groups except those rats into the BLA has been shown to inhibit cue-induced reinstatement receiving PEPA into the IL and vehicle into the NAshell. Active lever (See et al., 2001). Figure 2B shows that dopamine given alone into pressing in rats receiving PEPA was significantly lower than that the BLA immediately before an extinction session did not significantly found in rats receiving PEPA into the IL and dopamine into the reinstate cocaine seeking. Inactive lever pressing during the reinstate- NAshell (P < 0.05). A one-way anova also revealed a significant ment trials did not differ between treatment groups (data not shown). main effect for dopamine microinjected alone into the NAshell These findings suggest that the IL’s ability to suppress cue-induced immediately prior to an extinction session (F2,37 = 8.833, P < 0.001). cocaine-seeking is reversed by mimicking activation of the mesolim- Planned comparisons revealed that those rats receiving dopamine into bic dopaminergic afferents to the BLA and PL. the NAshell had significantly more active lever presses compared to Therefore, in the next experiment, illustrated in Fig. 3, we their extinction baseline (P < 0.05). Vehicle given alone into the determined whether broad activation of the mesocorticolimbic dopa- NAshell had no effect. Figure 4B shows the inactive lever presses for mine projection by VTA dopamine neurons by microinjecting the the experiment. A one-way anova found a significant effect for the l-opioid receptor agonist DAMGO into the VTA could reverse the inactive lever presses during cue-induced reinstatement suppression of cocaine-seeking (Kalivas, 1993). A one-way anova (F4,41 = 3.215, P < 0.05). Planned comparisons revealed that those revealed a significant main effect (F4,32 = 4.562, P < 0.01). Planned rats receiving dopamine into the NAshell and vehicle into the IL had

A 100 Extinction B 100 Vehicle in PL (n = 9) * Extinction DA (25 nmol) in PL (n = 8) 80 80 * Vehicle in BLA * * * DA (75 nmol) in BLA 60 * 60 40 40

20 20 Active lever presses Active lever presses

Vehicle in IL PEPA (0.075 Vehicle in IL PEPA (0.075 BLA injections nmol) in IL (n = 11) nmol) in IL (n = 9) alone (n = 6)

Fig. 2. Partial reversal of PEPA-induced suppression of cue-induced cocaine-seeking by dopaminergic activation of the PL or BLA. (A) Active lever presses (mean ± SEM) during extinction and cue-induced reinstatement for the IL-PL experiments. Intra-IL microinjections of PEPA suppressed cue-induced reinstatement, whereas rats receiving PEPA into the IL and dopamine into the PL showed significant reinstatement. Rats were counterbalanced across IL microinjections. (B) Active lever presses (mean ± SEM) during extinction and cue-induced reinstatement for the IL-BLA experiments. Intra-IL microinjections of PEPA suppressed cue- induced reinstatement, whereas rats receiving PEPA into the IL and dopamine into the PL showed significant reinstatement. For the IL-BLA experiment, intra-BLA microinjections of dopamine alone immediately prior to an extinction session did not significantly affect cocaine-seeking behavior. Rats were counterbalanced across BLA micrionjections. *P < 0.05 compared to extinction baseline.

150 Extinction lever presses (F2,37 = 5.419, P < 0.01), but planned comparisons did Vehicle in VTA (n = 5) * not reveal any significant differences between the inactive lever DAMGO (0.1 nmol) in VTA (n = 6) presses of the extinction baseline and those of either microinjection 100 group. Because dopamine microinjections into the NAshell appeared * to increase active and inactive lever presses, we created a difference * score for both active and inactive lever presses, in which we subtracted 50 the lever presses for rats receiving PEPA into the IL and vehicle into the shell from the lever presses for rats receiving PEPA into the IL and + dopamine into the shell. These difference scores showed no significant Active lever presses difference between the increase in active lever presses (mean ± SEM – 58.7 ± 29.83) compared with the increase in inactive lever presses Vehicle in IL PEPA (0.075 VTA injection nmol) in IL alone (n = 5) (36.5 ± 21.31). Figure 4C shows that intra-NAshell microinjections of the AMPA Fig. 3. Reversal of PEPA-induced suppression of cue-induced cocaine- receptor antagonist CNQX reversed the suppression of cue-induced seeking by VTA disinhibition, as shown by active lever presses (mean ± SEM) cocaine-seeking. A one-way anova revealed a significant main effect during extinction and cue-induced reinstatement. Intra-IL microinjections of (F4,59 = 3.576, P < 0.05). Planned comparisons revealed significant PEPA suppressed cue-induced reinstatement, an effect that was reversed by differences in active lever presses between the extinction baseline and intra-VTA microinjections of the l-opioid receptor agonist DAMGO. Intra- VTA microinjections of DAMGO did not significantly induce reinstatement all groups except those rats receiving PEPA into the IL and vehicle when given alone prior to an extinction session. Rats were counterbalanced into the NAshell (P < 0.05). Active lever pressing in rats receiving across IL microinjections. *P < 0.05 compared to extinction baseline; PEPA was not significantly different than that found in rats receiving +P < 0.05 compared to PEPA-DAMGO group. PEPA into the IL and CNQX into the NAshell but, as the data in the PEPA-alone group were non-normal in their distribution, a non- parametric analysis (Mann–Whitney U-test) was employed. A Mann- significantly more inactive lever presses compared to their extinction Whitney U-test revealed a significant difference between the two baseline (P < 0.05), while inactive lever presses for those receiving groups (P < 0.05). Because previous work indicated that AMPA dopamine into the NAshell and PEPA into the IL showed a trend receptor blockade in the NAshell prevents cocaine-prime-induced toward an increase compared to the extinction baseline (P < 0.08). For reinstatement (Famous et al., 2008), the rats then underwent an microinjections given into the NAshell alone prior to an extinction additional experiment in which an injection of cocaine (10 mg ⁄ kg, session, a one-way anova revealed a significant effect for inactive i.p.) was administered immediately prior to the extinction session. A

ABExtinction 100 Vehicle in NAshell * 100 Dopamine (20 μg) in NAshell * * # * 50 50 * + Active lever presses Inactive lever presses

Vehicle in PEPA (0.075 NAshell Vehicle in PEPA (0.075 NAshell IL (n = 8) nmol) in IL injections IL nmol) in IL injections (n = 6) alone (n = 12) alone C 100 Extinction Vehicle in NAshell CNQX (1 nmol) in NAshell

50 * * * + * Active lever presses

Vehicle in PEPA (0.075 Cocaine prime IL (n = 10) nmol) in IL w/ NAshell (n = 10) injections alone (n = 16)

Fig. 4. Regulation of cue-induce cocaine-seeking behavior by the NAshell. (A and B) Active and inactive lever presses (mean ± SEM), respectively, during extinction and cue-induced reinstatement for rats receiving dopamine microinjections into the NAshell. Intra-IL microinjections of PEPA suppressed cue-induced reinstatement, an effect that was reversed by dopamine microinjections into the NAshell. Intra-NAshell microinjections of dopamine alone prior to an extinction session induced a signiﬁcant increase in active lever presses. Rats were counterbalanced across NAshell microinjections. (C) Active lever presses (mean ± SEM) during extinction and cue- and cocaine-prime-induced reinstatement for rats receiving intra-NAshell microinjections of the AMPA receptor antagonist CNQX. Intra- NAshell microinjections of PEPA suppressed cue-induced reinstatement, an effect that was reversed by intra-NAshell microinjections of CNQX. Rats then underwent a cocaine-prime-induced reinstatement session in which CNQX or vehicle was administered into the NAshell prior to the session. CNQX administration did not prevent the cocaine-prime-induced reinstatement. Rats were counterbalanced across NAshell microinjections. *P < 0.05 compared to extinction baseline; #P < 0.08 compared to extinction baseline; +P < 0.05 compared to PEPA-dopamine or PEPA-CNQX group.

ª 2012 The Authors. European Journal of Neuroscience ª 2012 Federation of European Neuroscience Societies and Blackwell Publishing Ltd European Journal of Neuroscience, 35, 614–622 Competition over cocaine-seeking 619 one-way anova revealed a significant main effect of active lever was correct. Figure 5B–D shows the location of the microinjector tips presses (F2,47 = 3.900, P < 0.05). Planned comparisons revealed that in the BLA, the VTA and the NAshell, respectively. those rats receiving CNQX into the NAshell had significantly more active lever presses than the extinction baseline (P < 0.05), whereas there was only a trend toward a difference in active lever presses Discussion between the extinction baseline and vehicle-treated rats (P < 0.13). The present findings indicate that IL activation, via the AMPA The lack of a significant reinstatement in the vehicle-control rats may receptor potentiator PEPA, reduces cue-induced cocaine-seeking, have been due to the repeated reinstatements the rats had received consistent with previous cocaine-seeking studies (Peters et al., 2008; prior to this. There were no significant differences in inactive lever LaLumiere et al., 2010) and fear conditioning studies (Quirk et al., presses for any of the groups in Fig. 4C (data not shown). 2000; Morgan et al., 2003) that support a critical role for the IL in the consolidation and expression of extinction in these paradigms. The findings regarding the IL suggest that cue-induced reinstatement Histology ‘overrides’ the normal signaling in the IL that suppresses cocaine- Figure 5 shows the location of the microinjector tips for the animals seeking. Although previous experiments used AMPA microinjections whose data were used in the study. Figure 5A shows the location of 20 to activate the IL and suppress cocaine-prime reinstatement (Peters randomly selected microinjector tips in the IL. Data from rats whose et al., 2008), PEPA provides a significant advantage as it is an AMPA microinjectors terminated ventral to the IL in the dorsal peduncular receptor potentiator that reduces the rate of, or completely abolishes, cortex were included (n = 6), as evidence suggests that this structure is desensitization of the AMPA receptor (Sekiguchi et al., 1997) and also part of the ventromedial prefrontal cortex that regulates extinction therefore requires the presence of endogenous glutamate to produce learning (Peters et al., 2009). Indeed, those rats with microinjectors in effects. Accordingly, the present results suggest that PEPA adminis- the dorsal peduncular nucleus who received PEPA into the IL and tration into the IL enhances the activation of glutamatergic receptors vehicle into the other brain structure (n = 4) showed similar and, thereby, increases the activation of IL neurons and their output. reinstatement levels (21.75 ± 7.49), as compared with the values This idea is supported by the fact that the NAshell is a major recipient shown in the figures. Because the same cannulae were used for IL and of glutamatergic projections from the IL (Sesack et al., 1989), and PL microinjections, there are no separate tip locations for PL blocking AMPA receptors in the NAshell reversed the capacity of injections. Correct PL placement was assumed if the IL placement PEPA in the IL to inhibit cue-reinstated cocaine seeking.

3.72 mm

–2.40 mm

3.24 mm –2.76 mm

–3.12 mm 3.0 mm

–5.04 mm 1.44 mm

1.80 mm –5.28 mm

2.28 mm –5.52 mm

Fig. 5. Diagrams showing the microinjector tracks. Figure 5 shows the location of the microinjector tips for the animals whose data were used in the study. (A) Diagram showing the location of 20 randomly selected microinjector tips in the IL. Because the same cannulae were used for IL and PL microinjections, there are no separate tip locations for PL injections. Correct PL placement was assumed if the IL placement was correct. (B) Diagram showing the location of the microinjector tips in the BLA. (C) Diagram showing the location of the microinjector tips in the VTA. (D) Diagram showing the location of the microinjector tips in the NAshell. Figures adapted from Paxinos & Watson (2005) and A ⁄ P coordinates (in mm) are given relative to Bregma.

Previous findings indicate that reinstated cocaine-seeking depends seeking or opiate-conditioned place preference (Ovari & Leri, 2008; on dopaminergic activation of the BLA and PL, which is believed to Van den Oever et al., 2008, 2010; He et al., 2011). While the present facilitate activity in neurons projecting from these nuclei to the data do not directly address the apparently distinct role of the IL in nucleus accumbens core (NAcore) and thereby reinstate cocaine cocaine-seeking and fear conditioning vs. heroin-seeking, it is seeking (McFarland & Kalivas, 2001; See et al., 2001; McFarland interesting to consider that whereas cocaine is characterized as having et al., 2003; Di Ciano & Everitt, 2004). Our findings that indirectly aversive qualities in rats heroin is not (Ettenberg & Geist, 1993; stimulating VTA dopamine neurons with DAMGO or directly Ettenberg, 2004), and perhaps the aversive qualities of cocaine and stimulating dopamine receptors in the NAshell reversed the inhibition fear conditioning contribute to the involvement of the IL in extinction of reinstated cocaine seeking by PEPA in the IL suggests that the IL is and regulating motivated behavior. ‘upstream’ from activity in the circuit that promotes cocaine-seeking Apparent contradictions also exist in the literature describing the role and that downstream activation of the circuit promoting cocaine- of the NAshell in cocaine-seeking. In the present study we found that seeking is a prepotent regulator of behavior. In the PL and BLA, the AMPA receptor blockade of the NAshell restored cocaine-seeking that effects were more mixed. Although rats receiving PEPA into the IL was inhibited by PEPA activation of IL afferents to the NAshell. This is and dopamine into the PL and BLA showed significant reinstatement, consistent with previous studies showing that pharmacological inacti- they did not significantly differ from those rats receiving PEPA alone vation of the NAshell induces cocaine-seeking (Peters et al., 2008), in their active lever presses, suggesting that the reversal of the alcohol-seeking (Millan et al., 2010), second-order schedules of inhibition was not as strong in these groups. Most probably, this cocaine-seeking (Di Ciano et al., 2008) and food-seeking (Stratford finding reflects the necessary involvement of both the PL and the BLA & Kelley, 1997; Floresco et al., 2008). However, another study found in cue-induced cocaine-seeking and the fact that both structures that AMPA receptor blockade in the NAshell prevents reinstatement by independently project to the NAcore. Although not experimentally a systemic cocaine priming injection (Famous et al., 2008). To tested, it is likely that activation of the NAcore AMPA receptors determine whether the present data resulted from differences between would override the suppression signal, as previous findings suggest cue- and cocaine-induced reinstatement, we conducted an additional that PL and ⁄ or BLA projections to the NAcore are the critical pathways experiment in which NAshell AMPA receptors were blocked during a mediating reinstatement (McFarland et al., 2003; Di Ciano & Everitt, cocaine prime-induced reinstatement and found a partial potentiation of 2004). The idea that the IL is upstream from these other structures in the the reinstatement, suggesting that AMPA receptors in the NAshell are circuitry regulating cocaine-seeking is also consistent with the finding not involved in promoting cocaine-seeking. It is not clear why the that cocaine-seeking induced by IL inactivation is reversed by present results do not agree with those of Famous et al. (2008), though simultaneous BLA or PL inactivation by microinjected GABA agonists it should be noted that Famous et al. used a larger volume (0.5 lL) for (Peters et al., 2008). That the cocaine-seeking circuit is downstream microinjection than the current study (0.3 lL), posing the possibility from the suppression circuit may explain why, after weeks of extinction that some of the drug may have diffused to nearby regions of the training, reinstatement is easily triggered by exposure to a variety of NAcore, and AMPA receptor blockade in the NAcore prevents cue- stimuli. Although the purpose of the present experiments was to induced reinstatement of cocaine-seeking (Backstrom & Hyytia, 2007). determine whether the cocaine-seeking circuit was downstream from A number of studies find that NAshell blockade of dopamine the suppression circuit, it is likely that there are also structures upstream receptors prevents cocaine prime-induced reinstatement (Anderson from the IL that signal critical information to the IL to initiate the et al., 2003, 2006; Bachtell et al., 2005). In the present study, intra- suppression signal. Future work will address this issue. NAshell microinjections of dopamine reversed the suppression of In apparent contrast with the present findings, Koya et al. (2009) cocaine-seeking by PEPA administration into the IL. Moreover, found that IL activation 1 day after self-administration induced dopamine alone induced cocaine-seeking, in agreement with previous cocaine-seeking whereas IL inactivation after 30 days of withdrawal studies that dopaminergic input to the NAshell promotes drug-seeking inhibited cocaine-seeking. However, in both cases in that study, the IL (Schmidt et al., 2006; Anderson et al., 2008). Recent studies indicate manipulations were performed on the first day of extinction, either that the NAshell plays a crucial role in suppressing inappropriate 1 day or 30 days after self-administration. Our previous work suggests behaviors that the animal has learned not to engage in, such as that prior extinction training is required for the IL to be involved in responding to a nonrewarded cue or responding in the absence of a suppressing cocaine-seeking (Peters et al., 2008). Thus, akin to what reward-related cue (Ambroggi et al., 2011; Ghazizadeh et al.,In has been observed for fear conditioning (see below), the IL may be press). Because inactive lever pressing or pressing on the active lever involved in the learning and expression of the most recent contin- after extinction training are similar types of inappropriate behavior, it gencies, and only after extensive extinction training and the associated is not surprising that dopamine microinjections into the NAshell also new learning does the IL begin to suppress cocaine-seeking. It is increased inactive lever pressing. As previous work indicates that critical to note that studies using inactivation and activation may not NAshell inactivation, via GABAergic agonists, increases inappropriate always produce opposite behavioral outcomes, especially as the behaviors, including cocaine-seeking after extinction training (Peters mechanisms of inactivation or activation may be rather different and et al., 2008) or responding during the nonrewarded cue (Ambroggi alter neural activity in different manners, and this may account for et al., 2011), it appears that NAshell inactivation via GABAergic some discrepancies in the literature. agonists produces similar effects to dopamine in the NAshell. This In contrast to studies with cocaine-seeking and fear conditioning, conclusion is in agreement with evidence that dopamine receptor the evidence regarding the role of the IL in heroin- and morphine- blockade in the NAshell prevents the increase in responding to the seeking behavior has been conflicting. Previous work found that PL or nonrewarded cue induced by IL inactivation (Ghazizadeh et al.,In IL inactivation prevents reinstatement of heroin-seeking (Rogers press) and that dopamine inhibits reward-related firing of accumbens et al., 2008; Bossert et al., 2011), whereas studies with cocaine- neurons (Carlezon & Thomas, 2009). If this hypothesis is correct it seeking have shown that PL, but not IL, inactivation prevents cocaine- would help to resolve the apparent contradiction in the literature, in seeking (McFarland & Kalivas, 2001). However, other work suggests which findings indicate that NAshell inactivation has no effect on that, under at least some circumstances, the medial prefrontal cortex in (McFarland & Kalivas, 2001), whereas dopamine receptor blockade general or the IL specifically is involved in the extinction of opiate- prevents (Anderson et al., 2006), reinstatement of cocaine-seeking.

In contrast to the dopamine signaling, IL signaling appears to Abbreviations suppress these inappropriate behaviors, as suggested by the present aCSF, artificial cerebrospinal fluid; BLA, basolateral amygdala (BLA); DAM- findings and previous work (Ishikawa et al., 2008; Ghazizadeh et al., GO, [D-Ala2, NMe-Phe4, Gly-ol5]-enkephalin; FR1, fixed-ratio 1; IL, infralim- In press). Recent evidence indicates that reinstatement of cocaine- bic cortex; NAcore, nucleus accumbens core; NAshell, nucleus accumbens shell; conditioned place preference reduces the strength of excitatory inputs PEPA, 2-[2,6-Difluoro-4-[[2-[(phenylsulfonyl)amino]ethyl]thio]phenoxy]acetamide; PL, prelimbic cortex; vmPFC, ventromedial prefrontal cortex; VTA, to the NAshell, suggesting that a mechanism within the NAshell ventral tegmental area. permits reinstatement triggers to reduce IL inputs that normally inhibit drug-seeking (Rothwell et al., 2011). Indeed, dopamine acts as a modulator of glutamatergic transmission onto medium spiny neurons References in the NAshell (Nicola et al., 2000). In countermanding PEPA Ambroggi, F., Ghazizadeh, A., Nicola, S.M. & Fields, H.L. (2011) Roles of activation of IL glutamatergic input to the NAshell, dopamine may be nucleus accumbens core and shell in incentive-cue responding and acting presynaptically on D1 receptors to inhibit glutamate or behavioral inhibition. J. Neurosci., 31, 6820–6830. Anderson, S.M., Bari, A.A. & Pierce, R.C. (2003) Administration of the postsynaptically via D2 to inhibit the efficacy of glutamate signaling D1-like dopamine receptor antagonist SCH-23390 into the medial nucleus (Nicola et al., 2000). Because dopaminergic inputs to the NAcore accumbens shell attenuates cocaine priming-induced reinstatement of drug- produce similar effects as those in the NAshell yet prefrontal seeking behavior in rats. Psychopharmacology (Berl), 168, 132–138. glutamatergic inputs to the NAcore induce, rather than inhibit, Anderson, S.M., Schmidt, H.D. & Pierce, R.C. (2006) Administration of the D2 cocaine-seeking, it is unlikely that this mechanism fully accounts for dopamine receptor antagonist sulpiride into the shell, but not the core, of the nucleus accumbens attenuates cocaine priming-induced reinstatement of the differences in the NAshell. As analysis of the NAcore and shell drug seeking. Neuropsychopharmacology, 31, 1452–1461. indicates that subpopulations of neurons, rather than the entire Anderson, S.M., Famous, K.R., Sadri-Vakili, G., Kumaresan, V., Schmidt, structure, guide motivated behavior (Taha et al., 2007), understanding H.D., Bass, C.E., Terwilliger, E.F., Cha, J.H. & Pierce, R.C. (2008) CaMKII: the interplay of glutamatergic and dopaminergic inputs in each a biochemical bridge linking accumbens dopamine and glutamate systems in cocaine seeking. Nat. Neurosci., 11, 344–353. structure with specific behavior will require future work focused on Bachtell, R.K., Whisler, K., Karanian, D. & Self, D.W. (2005) Effects of intra- the distinct neuronal ensembles within the NAshell and core, rather nucleus accumbens shell administration of dopamine agonists and antago- than individual neurons, activated by specific inputs. nists on cocaine-taking and cocaine-seeking behaviors in the rat. Psycho- The present findings are consistent with human studies on the pharmacology (Berl), 183, 41–53. ventromedial prefrontal cortex (vmPFC or subgenual anterior cingu- Backstrom, P. & Hyytia, P. (2007) Involvement of AMPA ⁄ kainate, NMDA, and mGlu5 receptors in the nucleus accumbens core in cue-induced late cortex, including the medial orbitofrontal cortex), which appears reinstatement of cocaine seeking in rats. Psychopharmacology (Berl), 192, to be functionally homologous to the IL (Phelps et al., 2004; Bechara, 571–580. 2005; Milad et al., 2009). Activity in the vmPFC correlates with Bechara, A. (2005) Decision making, impulse control and loss of willpower to retention of extinction for fear conditioning (Phelps et al., 2004; Milad resist drugs: a neurocognitive perspective. Nat. Neurosci., 8, 1458–1463. Boes, A.D., Bechara, A., Tranel, D., Anderson, S.W., Richman, L. & et al., 2009). Moreover, cue-induced cocaine craving is associated Nopoulos, P. (2009) Right ventromedial prefrontal cortex: a neuroanatom- with decreased vmPFC activity (Bonson et al., 2002), and greater ical correlate of impulse control in boys. Soc. Cogn. Affect Neurosci., 4,1– vmPFC activity correlates with longer durations of abstinence from 9. cocaine use (Brewer et al., 2008). Other evidence indicates that low Bonson, K.R., Grant, S.J., Contoreggi, C.S., Links, J.M., Metcalfe, J., Weyl, impulse control in humans correlates with reduced vmPFC volume H.L., Kurian, V., Ernst, M. & London, E.D. (2002) Neural systems and cue- induced cocaine craving. Neuropsychopharmacology, 26, 376–386. (Boes et al., 2009) and that vmPFC-damaged patients make poor Bossert, J.M., Stern, A.L., Theberge, F.R., Cifani, C., Koya, E., Hope, B.T. & choices in the Iowa Gambling Task, showing a preference for high Shaham, Y. (2011) Ventral medial prefrontal cortex neuronal ensem- immediate rewards with long-term negative consequences (Bechara, bles mediate context-induced relapse to heroin. Nat. Neurosci., 14, 420– 2005). Thus, the IL in rats or vmPFC in humans is probably involved 422. Bouton, M.E. (2002) Context, ambiguity, and unlearning: sources of relapse in learning to suppress inappropriate behaviors, including cocaine- after behavioral extinction. Biol. Psychiatry, 52, 976–986. seeking, and is consequently critical for extinction learning and Brewer, J.A., Worhunsky, P.D., Carroll, K.M., Rounsaville, B.J. & Potenza, expression. It is tempting to speculate that differences in vmPFC M.N. (2008) Pretreatment brain activation during stroop task is associated activity in humans may contribute to an inability to suppress cocaine- with outcomes in cocaine-dependent patients. Biol. Psychiatry, 64, 998– seeking behavior, rendering the vmPFC a potential pharmacothera- 1004. Carlezon, W.A. Jr & Thomas, M.J. (2009) Biological substrates of reward and peutic target for addiction treatment. In fact, PEPA appears to activate aversion: a nucleus accumbens activity hypothesis. Neuropharmacology, the medial prefrontal cortex relatively selectively over the BLA or 56(Suppl 1), 122–132. hippocampus due to the increased presence in the medial prefrontal Di Ciano, P. & Everitt, B.J. (2004) Direct interactions between the basolateral cortex of AMPA receptor subunits and splice variants that preferen- amygdala and nucleus accumbens core underlie cocaine-seeking behavior by J. Neurosci. 24 et al. rats. , , 7167–7173. tially bind PEPA (Zushida , 2007). Intriguingly, in mice, Di Ciano, P., Robbins, T.W. & Everitt, B.J. (2008) Differential effects of systemic PEPA administration, as well as intra-medial prefrontal nucleus accumbens core, shell, or dorsal striatal inactivations on the cortex microinjections of PEPA, enhance the rate of extinction of fear persistence, reacquisition, or reinstatement of responding for a drug-paired conditioning (Zushida et al., 2007; Yamada et al., 2009, 2011), conditioned reinforcer. Neuropsychopharmacology, 33, 1413–1425. suggesting that it may prove to be a fruitful pharmacotherapy for Ettenberg, A. (2004) Opponent process properties of self-administered cocaine. Neurosci. Biobehav. Rev., 27, 721–728. promoting the inhibition of unwanted behaviors through extinction Ettenberg, A. & Geist, T.D. (1993) Qualitative and quantitative differences in training. the operant runway behavior of rats working for cocaine and heroin reinforcement. Pharmacol. Biochem. Behav., 44, 191–198. Fallon, J.H. & Moore, R.Y. (1978) Catecholamine innervation of the basal forebrain. IV. Topography of the dopamine projection to the basal forebrain Acknowledgements and neostriatum. J. Comp. Neurol., 180, 545–580. This research was supported by NIH grants DA027055 (R.T.L.) and Famous, K.R., Kumaresan, V., Sadri-Vakili, G., Schmidt, H.D., Mierke, D.F., DA015369 (P.W.K.). We thank Rachel Miller, Kate Niehoff, Karen Corson, Cha, J.H. & Pierce, R.C. (2008) Phosphorylation-dependent trafficking of Ana Weiland, Mary Shaw, Dil Patel and Giana Ravita for their excellent GluR2-containing AMPA receptors in the nucleus accumbens plays a critical technical assistance. The authors report no conflicts of interest. role in the reinstatement of cocaine seeking. J. Neurosci., 28, 11061–11070.

Floresco, S.B., McLaughlin, R.J. & Haluk, D.M. (2008) Opposing roles for the Peters, J., Kalivas, P.W. & Quirk, G.J. (2009) Extinction circuits for fear and nucleus accumbens core and shell in cue-induced reinstatement of food- addiction overlap in prefrontal cortex. Learn. Mem., 16, 279–288. seeking behavior. Neuroscience, 154, 877–884. Phelps, E.A., Delgado, M.R., Nearing, K.I. & LeDoux, J.E. (2004) Extinction Ghazizadeh, A., Ambroggi, F., Odean, N. & Fields, H.L. (In press) Prefrontal learning in humans: role of the amygdala and vmPFC. Neuron, 43, 897–905. cortex mediates extinction of responding by a dual neural mechanism in Quirk, G.J., Russo, G.K., Barron, J.L. & Lebron, K. (2000) The role accumbens shell. J. Neurosci. of ventromedial prefrontal cortex in the recovery of extinguished fear. He, Y.Y., Xue, Y.X., Wang, J.S., Fang, Q., Liu, J.F., Xue, L.F. & Lu, L. (2011) J. Neurosci., 20, 6225–6231. PKMzeta maintains drug reward and aversion memory in the basolateral Quirk, G.J., Garcia, R. & Gonzalez-Lima, F. (2006) Prefrontal mechanisms in amygdala and extinction memory in the infralimbic cortex. Neuropsycho- extinction of conditioned fear. Biol. Psychiatry, 60, 337–343. pharmacology, 36, 1972–1981. Rogers, J.L., Ghee, S. & See, R.E. (2008) The neural circuitry underlying Hooks, M.S. & Kalivas, P.W. (1994) Involvement of dopamine and excitatory reinstatement of heroin-seeking behavior in an animal model of relapse. amino acid transmission in novelty-induced motor activity. J. Pharmacol. Neuroscience, 151, 579–588. Exp. Ther., 269, 976–988. Rothwell, P.E., Kourrich, S. & Thomas, M.J. (2011) Synaptic adaptations in the Ishikawa, A., Ambroggi, F., Nicola, S.M. & Fields, H.L. (2008) Contributions nucleus accumbens caused by experiences linked to relapse. Biol. Psychiatry, of the amygdala and medial prefrontal cortex to incentive cue responding. 69, 1124–1126. Neuroscience, 155, 573–584. Schmidt, H.D., Anderson, S.M. & Pierce, R.C. (2006) Stimulation of D1-like Kalivas, P.W. (1993) Neurotransmitter regulation of dopamine neurons in the or D2 dopamine receptors in the shell, but not the core, of the nucleus ventral tegmental area. Brain Res. Brain Res. Rev., 18, 75–113. accumbens reinstates cocaine-seeking behaviour in the rat. Eur. J. Neurosci., Koya, E., Uejima, J.L., Wihbey, K.A., Bossert, J.M., Hope, B.T. & Shaham, Y. 23, 219–228. (2009) Role of ventral medial prefrontal cortex in incubation of cocaine See, R.E., Kruzich, P.J. & Grimm, J.W. (2001) Dopamine, but not glutamate, craving. Neuropharmacology, 56(Suppl. 1), 177–185. receptor blockade in the basolateral amygdala attenuates conditioned reward LaLumiere, R.T. & Kalivas, P.W. (2008) Glutamate release in the nucleus in a rat model of relapse to cocaine-seeking behavior. Psychopharmacology accumbens core is necessary for heroin seeking. J. Neurosci., 28, 3170–3177. (Berl), 154, 301–310. LaLumiere, R.T., Nawar, E.M. & McGaugh, J.L. (2005) Modulation of Sekiguchi, M., Fleck, M.W., Mayer, M.L., Takeo, J., Chiba, Y., Yamashita, S. memory consolidation by the basolateral amygdala or nucleus accumbens & Wada, K. (1997) A novel allosteric potentiator of AMPA receptors: 4–2- shell requires concurrent dopamine receptor activation in both brain regions. (phenylsulfonylamino)ethylthio–2,6-diﬂuoro-phenoxyaceta mide. J. Neuro- Learn. Mem., 12, 296–301. sci., 17, 5760–5771. LaLumiere, R.T., Niehoff, K.E. & Kalivas, P.W. (2010) The infralimbic cortex Sesack, S.R., Deutch, A.Y., Roth, R.H. & Bunney, B.S. (1989) Topographical regulates the consolidation of extinction after cocaine self-administration. Organization of the Efferent Projections of the Medial Prefrontal Cortex in Learn. Mem., 17, 168–175. the Rat - an Anterograde Tract-Tracing Study with Phaseolus-Vulgaris McFarland, K. & Kalivas, P.W. (2001) The circuitry mediating cocaine- Leucoagglutinin. J. Comp. Neurol., 290, 213–242. induced reinstatement of drug-seeking behavior. J. Neurosci., 21, 8655– Shaham, Y., Shalev, U., Lu, L., De Wit, H. & Stewart, J. (2003) The 8663. reinstatement model of drug relapse: history, methodology and major McFarland, K., Lapish, C.C. & Kalivas, P.W. (2003) Prefrontal glutamate ﬁndings. Psychopharmacology (Berl), 168, 3–20. release into the core of the nucleus accumbens mediates cocaine-induced Stratford, T.R. & Kelley, A.E. (1997) GABA in the nucleus accumbens shell reinstatement of drug-seeking behavior. J. Neurosci., 23, 3531–3537. participates in the central regulation of feeding behavior. J. Neurosci., 17, McLaughlin, J. & See, R.E. (2003) Selective inactivation of the dorsomedial 4434–4440. prefrontal cortex and the basolateral amygdala attenuates conditioned-cued Taha, S.A., Nicola, S.M. & Fields, H.L. (2007) Cue-evoked encoding of reinstatement of extinguished cocaine-seeking behavior in rats. Psychophar- movement planning and execution in the rat nucleus accumbens. J. Physiol., macology (Berl), 168, 57–65. 584, 801–818. Milad, M.R., Pitman, R.K., Ellis, C.B., Gold, A.L., Shin, L.M., Lasko, N.B., Van den Oever, M.C., Goriounova, N.A., Li, K.W., Van der Schors, R.C., Zeidan, M.A., Handwerger, K., Orr, S.P. & Rauch, S.L. (2009) Neurobi- Binnekade, R., Schoffelmeer, A.N., Mansvelder, H.D., Smit, A.B., Spijker, ological basis of failure to recall extinction memory in posttraumatic stress S. & De Vries, T.J. (2008) Prefrontal cortex AMPA receptor plasticity is disorder. Biol. Psychiatry, 66, 1075–1082. crucial for cue-induced relapse to heroin-seeking. Nat. Neurosci., 11, 1053– Millan, E.Z., Furlong, T.M. & McNally, G.P. (2010) Accumbens shell- 1058. hypothalamus interactions mediate extinction of alcohol seeking. J. Neuro- Van den Oever, M.C., Lubbers, B.R., Goriounova, N.A., Li, K.W., Van der sci., 30, 4626–4635. Schors, R.C., Loos, M., Riga, D., Wiskerke, J., Binnekade, R., Stegeman, Morgan, M.A., Schulkin, J. & LeDoux, J.E. (2003) Ventral medial prefrontal M., Schoffelmeer, A.N., Mansvelder, H.D., Smit, A.B., De Vries, T.J. & cortex and emotional perseveration: the memory for prior extinction training. Spijker, S. (2010) Extracellular matrix plasticity and GABAergic inhibition Behav. Brain Res., 146, 121–130. of prefrontal cortex pyramidal cells facilitates relapse to heroin seeking. Nicola, S.M., Surmeier, J. & Malenka, R.C. (2000) Dopaminergic modulation Neuropsychopharmacology, 35, 2120–2133. of neuronal excitability in the striatum and nucleus accumbens. Annu. Rev. Yamada, D., Zushida, K., Wada, K. & Sekiguchi, M. (2009) Pharmacological Neurosci., 23, 185–215. discrimination of extinction and reconsolidation of contextual fear memory Ovari, J. & Leri, F. (2008) Inactivation of the ventromedial prefrontal cortex by a potentiator of AMPA receptors. Neuropsychopharmacology, 34, 2574– mimics re-emergence of heroin seeking caused by heroin reconditioning. 2584. Neurosci. Lett., 444, 52–55. Yamada, D., Wada, K. & Sekiguchi, M. (2011) Facilitating actions of an Paxinos, G. & Watson, C. (2005) The Rat Brain in Stereotaxic Coordinates. AMPA receptor potentiator upon extinction of contextually conditioned fear Elsevier Academic Press, Amsterdam. response in stressed mice. Neurosci. Lett., 488, 242–246. Peters, J., LaLumiere, R.T. & Kalivas, P.W. (2008) Infralimbic prefrontal Zushida, K., Sakurai, M., Wada, K. & Sekiguchi, M. (2007) Facilitation of cortex is responsible for inhibiting cocaine seeking in extinguished rats. extinction learning for contextual fear memory by PEPA: a potentiator of J. Neurosci., 28, 6046–6053. AMPA receptors. J. Neurosci., 27, 158–166.

ª 2012 The Authors. European Journal of Neuroscience ª 2012 Federation of European Neuroscience Societies and Blackwell Publishing Ltd European Journal of Neuroscience, 35, 614–622