d1, but not d2, receptor blockade within the infralimbic and … · d1, but not d2, receptor...

D1, but not D2, receptor blockade within the infralimbicand medial orbitofrontal cortex impairs cocaine seekingin a region-specific manner

Caitlin V. Cosme1, Andrea L. Gutman1, Wensday R. Worth1 & Ryan T. LaLumiere1,2

Department of Psychological and Brain Sciences, University of Iowa, Iowa City, IA, USA1 and Interdisciplinary Graduate Program in Neuroscience, University of Iowa,Iowa City, IA, USA2

ABSTRACT

Evidence suggests that the infralimbic cortex (IL), a subregion of the ventromedial prefrontal cortex (vmPFC),suppresses cocaine-seeking behavior in a self-administration paradigm, whereas the more anterior vmPFC subregion,the medial orbitofrontal cortex (mOFC), has received very little attention in this regard. Despite the established dopa-minergic innervation of the vmPFC, whether dopamine receptor blockade in each subregion alters the reinstatementof cocaine seeking is unclear. To address this issue, male Sprague–Dawley rats underwent 2weeks of cocaine self-administration, followed by extinction training and reinstatement testing. Immediately prior to each reinstatementtest, rats received microinjections of the D1 receptor antagonist SCH 23390, the D2 receptor antagonist sulpiride ortheir respective vehicles. D1 receptor blockade in the IL reduced cued reinstatement but had no effect on cocaine primeand cue+ cocaine-prime reinstatement, whereas D2 receptor blockade in the IL had no effect on reinstatement. For themOFC, however, D1 receptor blockade reduced cocaine seeking in all reinstatement types, whereas blocking D2receptors in the mOFC had no effect on any form of cocaine seeking. These findings suggest different roles for D1receptors in the IL versus the mOFC in regulating cocaine-seeking behavior. Moreover, even as previous work indicatesthat IL inactivation does not affect reinstatement but, rather, induces cocaine seeking during extinction, the presentfindings suggest that dopamine receptor activation in the IL is necessary for cocaine seeking under somecircumstances.

Keywords cocaine prime, rat, SCH 23390, self-administration, sulpiride, ventromedial prefrontal cortex.

Correspondence to: Caitlin V. Cosme, E11 Seashore Hall, Department of Psychological and Brain Sciences, Iowa City, IA 52242, USA. E-mail:[email protected]

Increasing evidence has pointed to a critical role for theventromedial prefrontal cortex (vmPFC), and especiallythe infralimbic (IL) region of the vmPFC, in regulatingdrug seeking (LaLumiere et al. 2012; Peters et al. 2008;for a larger review of PFC involvement in drug seeking,see Van den Oever et al. 2010). Prior work indicates thatIL inactivation induces cocaine seeking during an extinc-tion session, whereas IL activation reduces cocaine seek-ing during cocaine prime or cued reinstatement(LaLumiere et al. 2012; Peters et al. 2008). Together withother findings (LaLumiere et al. 2010), these resultssuggest that IL activity suppresses cocaine seeking follow-ing self-administration and extinction training and isinvolved in the consolidation of extinction learning forcocaine-seeking behavior.

However, prior studies have not investigated the role ofIL dopamine receptors in cocaine seeking, despite knowndopaminergic projections to the vmPFC (Hitchcott et al.2007; Smiley et al. 1992; Van Eden et al. 1987). Previouswork examining this region during non-drug-relatedmotivated behaviors indicates that intra-IL infusions ofdopamine or a D1 receptor antagonist produce a shift fromhabitual to goal-oriented instrumental behaviors(Hitchcott et al. 2007). Additionally, D1 receptor antago-nism and D2 receptor agonism within the IL reduce com-pulsive sucrose seeking (Barker et al. 2013). Given thataddiction is conceptualized as a shift from goal-orientedbehaviors to a habitual response (Koob & Volkow 2010),such findings suggest that dopamine activity within theIL may be critical for the expression of relapse behaviors.

ORIGINAL ARTICLE doi:10.1111/adb.12442

© 2016 Society for the Study of Addiction Addiction Biology, 23, 16–27

bs_bs_banner

Indeed, addressing this issue may be of importance to re-lated controversial questions raised about the role of theIL in inhibiting drug seeking in general, as previous stud-ies have shown blocking 5-HT2A receptors in the vmPFCdecreases both cue and cocaine-prime reinstatement(Pockros et al. 2011). Moreover, IL/vmPFC activity drives,rather than inhibits, the reinstatement of heroin seeking(Bossert et al. 2011; Rogers et al. 2008), leading Peterset al. (2013) to suggest that dopamine activity withinthe IL may account for these discrepant results regardingthe ability of the IL to regulate drug seeking but note thatno data exist to address this question.

In addition to the IL, the vmPFC also includes a moreanterior subregion known as the medial orbitofrontal cor-tex (mOFC). Previous research has found that identicalmanipulations in different vmPFC regions along therostrocaudal axis produce distinct behavioral effects(Smith & Berridge 2005). Indeed, IL activation appearsto suppress feeding behavior that is induced via glutamatereceptor blockade in the nucleus accumbens shell,whereas mOFC activation enhances this feeding behavior(Richard & Berridge 2013). However, akin to prior workon the IL, mOFC inactivation using a gamma-aminobutyric acid (GABAA/B) agonist cocktail has noeffect on the reinstatement of cocaine-seeking behavior(Fuchs et al. 2004). Nevertheless, previous research onthe nucleus accumbens shell has found conflicting resultsby using GABA receptor activation versus dopaminereceptor manipulations in terms of the reinstatement ofcocaine seeking (Anderson et al. 2006; McFarland &Kalivas 2001). Given that the mOFC receives dopaminer-gic innervation from the ventral tegmental area(Swanson 1982) and maintains efferent projections tothe amygdala and nucleus accumbens (Brog et al. 1993;Ishikawa & Nakamura 2003; Malkusz et al. 2015), weconsidered the possibility that dopaminergic manipula-tions in the mOFC would alter reinstatement. As D1receptor blockade in the lateral OFC reduces context-induced cocaine seeking (Lasseter et al. 2014), we hypoth-esized that blocking these receptors in the medial region ofthe OFC would similarly result in a disruption of cocaineseeking. To examine these issues, the current studyinvestigated the effects of D1 and D2 receptor blockadein the IL and mOFC during the reinstatement of cocaineseeking.

MATERIALS AND METHODS

Subjects

Male Sprague–Dawley rats (250–275 g at time of arrival;Charles River Laboratories; n=98) were single-housed ona 12-hour reverse light cycle (and kept at constanttemperature) with food and water ad libitum in an

Association for Assessment and Accreditation of Labora-tory Animal Care-approved vivarium. All animals wereallowed to acclimate to the vivarium for a minimum of5 days before undergoing surgery. All procedures werein accordance with the National Institutes of HealthGuide for the Care and Use of Laboratory Animals andapproved by the University of Iowa Institutional AnimalCare and Use Committee.

Surgery

The rats were anaesthetized by using ketamine (100mg/kg, i.m.) and xylazine (6mg/kg, i.m.). Additionally,ketorolac was given (3mg/kg, i.p.) as an analgesic onboth the day of surgery as well as the day immediatelyfollowing surgery. Catheter implantation was performed,as described previously (Cosme et al. 2015). The rats werethen placed in a small animal stereotaxic instrument(Kopf Instruments, Tujunga, CA, USA). Jewelers’ screwswere affixed to the skull surface to serve as anchors.Double-barreled cannulae with a 1.2-mm center-to-center distance (Plastics One, Roanoke, VA, USA) wereimplanted, aimed at either the IL or mOFC and thensecured by using dental cement. Coordinates were asfollows: IL: 2.8mm anterior to bregma and 2.1mm ven-tral from the skull surface (cannula aimed 3mm aboveIL); and mOFC: 3.5mm anterior to bregma and 2.8mmventral from skull surface (cannula aimed 3mm abovemOFC). These coordinates were chosen based on previ-ous work (LaLumiere et al. 2010) and refined in ourlaboratory.

After surgery, the animals received 3ml of subcutane-ous saline and a topical application of the anestheticbupivacaine to the cranial and chest incision sites.Dummy injectors were placed in each cannula alongwith a tight fitting cap. The rats were then returned totheir home cages, where they were allowed to recoverfor 5–7 days. During recovery, catheters were flusheddaily with 0.1ml of heparinized saline (100 USP) toensure catheter patency.

Cocaine self-administration and extinction

Self-administration was conducted in standard operantchambers (Med Associates, Fairfield, VT, USA) with tworetractable levers, a house light, a cue light and a tone-generator (4500Hz). The rats had all food removed24 hours prior to a single 15-hour overnight food train-ing session. During this session, a press on the active leverresulted in a single food pellet (45mg) on a fixed-ratio 1schedule. Following the single food training session, therats were fed 10 g of rat chow and began self-administration training the following day. All ratsreceived ~20 g of rat chow daily after each cocaine self-

D1 blockade and reinstatement 17


administration, extinction or reinstatement session. Afterfood training, the rats’ catheters were assessed forpatency by using 0.1ml of sodium brevital (10mg/ml).

One day after food training, cocaine self-administration began. The rats were placed in operantchambers for 2 hours/day for a minimum of 12 days,during which active lever presses produced an infusionof cocaine (50 μl of infusion of 150μg of cocaine dis-solved in sterile saline, given over 2.18 seconds; cocainewas kindly provided by the National Institute on DrugAbuse) along with a 5-second light and tone cue on afixed-ratio 1 schedule. A 20-second timeout periodfollowed each infusion. Inactive lever presses had no con-sequence. The animals would begin extinction training ifthey received a minimum of 15 infusions of cocaine perday for at least 10 days, including the last 3 days of self-administration, and they demonstrated discriminationbetween the active and inactive lever. During extinctiontraining, active lever presses did not produce cocaineinfusions or the light and tone cues. Each rat underwentextinction for a minimum of 7 days and enteredreinstatement testing when it had 25 or fewer activelever presses for at least 2 consecutive days immediatelyprior to the reinstatement session. The final 2 days ofextinction training prior to each reinstatement testserved as the extinction baseline.

Microinjections

Immediately prior to each reinstatement test, the ratsreceived intra-IL or intra-mOFC microinjections.Microinjectors (with 3-mm projections beyond the endof the respective cannula) were connected to PE20tubing, which was attached to 10-μl Hamilton syringescontrolled by an infusion pump. Microinjections (0.3μl/side) were given over 1min, and injectors were left inplace for an additional 2minutes to permit drug diffusion.Following the microinjection, the rats were immediatelyplaced into the operant chamber for the appropriatereinstatement test. Microinjected drugs consisted of theD1 receptor antagonist SCH 23390 (0.1μg/side) andthe D2 receptor antagonist sulpiride (30 ng/side), eachdissolved in artificial cerebral spinal fluid as the vehicle.Doses were chosen based on prior research (Lalumiereet al. 2004). Although some previous studies haveinfused higher concentrations of sulpiride (1–3μg/side)into the mPFC (Granon et al. 2000; Winter et al. 2009),others have shown that smaller concentrations ofsulpiride are equally as effective (Druzin et al. 2000),and, in some cases, lower doses of sulpiride in the mPFCand nucleus accumbens have proven to be more effectiveat revealing behavioral differences (Pakdel & Rashidy-Pour 2007; Setlow & McGaugh 1998).

Reinstatement testing

During each reinstatement test (2 hours), active leverpresses never produced a cocaine infusion. Following eachreinstatement test, the rats had their lever pressingre-extinguished to baseline levels for a minimum of 3 daysby using the same criteria described previously. Each ratunderwent three reinstatement tests (cued, cocaine primeand cue+ cocaine prime) and completed each test twice,once following avehiclemicroinjection and once followinga drug microinjection in a counterbalanced design,resulting in a total of six reinstatement sessions per rat.For cued reinstatement, active lever presses produced thelight and tone cues that were previously paired with thecocaine infusion during self-administration. Cocaine-prime reinstatement consisted of an injection of cocaine(10mg/kg, i.p.), immediately before the reinstatementsession. Cued+ cocaine-prime reinstatement involved aninjection of cocaine (10mg/kg, i.p.) immediately prior toa cued reinstatement session. As cue + cocaine-primereinstatement typically produces higher levels of cocaine-seeking behavior than cue or cocaine-prime reinstatementalone, we included this test in our experimental design toprovide us with the optimal opportunity to observe behav-ioral differences following dopamine receptor blockade.

Experiment 1

Prior to reinstatement testing, the rats received intra-ILmicroinjections of the D1 receptor antagonist SCH 23390(n=11). All rats underwent reinstatement testing in thefollowing order: cued, cocaine prime and cued+ cocaineprime, as described in the preceding texts. Due to cloggedcannula and illness, three rats were removed from theexperiment after the cued reinstatement, and one morewas removed following the cocaine prime test.

Experiment 2

Prior to reinstatement testing, the rats received intra-ILmicroinjections of the D2 receptor antagonist sulpiride(n=14). All rats underwent reinstatement testing inthe following order: cued, cocaine prime and cued+ co-ocaine prime, as described in the preceding texts.

Experiment 3

Prior to reinstatement testing, the rats received intra-mOFC microinjections of the D1 receptor antagonistSCH 23390 (n=12). The rats underwent reinstatementtesting in the following order: cued, cocaine prime andcued+ cocaine prime, except for a subset of rats that, un-intentionally, received exposure to the cues during extinc-tion training. Therefore, the data for this subset (n=5) forthe cued reinstatement were not included in the analysis

18 Caitlin V. Cosme et al.


due to concern about the robustness of the reinstatementlevels. The data for this subset for the cued+ cocaine-prime reinstatement were included, as the levels of this re-instatement were robust and the results were not differentfrom those observed with the rest of the rats in this exper-iment. Due to clogged cannula and illness, two rats wereremoved from the experiment after the cued reinstate-ment and four more were removed following the cocaineprime test.

Experiment 4

Prior to reinstatement testing, the rats received intra-mOFC microinjections of the D2 receptor antagonistsulpiride (n=6). All rats underwent reinstatement test-ing in the following order: cued, cocaine prime andcued+ cocaine prime, as previously described. Due to alack of significant cocaine-prime reinstatement, an addi-tional group of rats underwent only cocaine-prime rein-statement with no prior reinstatement tests (n=6).

Histological analysis

To verify cannula placement, the rats were overdosedwith sodium pentobarbital (100mg/kg, i.p.) and intra-cardially perfused by using phosphate-buffered saline.Brains were then placed in 3.7 percent formaldehyde fora minimum of 24 hours before being sectioned. Coronalslices were 75μm thick and were mounted directly ontoa gelatin-coated slide. Sections were stained by usingCresyl violet and analyzed for the correct termination siteof the microinjectors. Any rat with an injector termina-tion point outside the borders of the IL or mOFC was ex-cluded from further analysis.

Data analysis

Two-way analyses of variance (ANOVA) were used toanalyze reinstatement lever pressing data with both com-parisons as within-subjects (extinction versus reinstate-ment; vehicle versus drug). Post-hoc analysis wascompleted by using Holm–Sidak’s multiple comparisontests. P-values of less than 0.05 were considered signifi-cant. All measures were expressed as mean±SEM. Eachgroup’s n is indicated in the figure.

RESULTS

Out of a total of 96 rats used in the present experiments,51 rats were included in the final data. The rats were ex-cluded due to misplaced (18 rats) or unverifiable (15 rats)microinjection termination locations, failure to acquireself-administration (1 rat), loss of cannula patency (7rats) and loss of catheter patency (4 rats). Figure 1ashows the number of active and inactive lever pressesand cocaine infusions over the final 12 days of cocaineself-administration. Active lever pressing was signifi-cantly higher than inactive lever pressing over the last3 days of self-administration [t(50) =9.55, P<0.05].Figure 1b and c shows the termination site of themicroinjector tips in the IL and mOFC, respectively.

Experiment 1

D1 receptor blockade in the IL attenuates cued reinstatement

D1 receptor blockade within the IL significantly reducedactive lever presses for cued reinstatement (Fig. 2a). Atwo-way repeated-measures ANOVA of active lever

Figure 1 Self-administrationdata and histological representa-tions. (a) Number of active andinactive lever presses and cocaineinfusions for the last 12 days ofcocaine self-administration for allrats included in the final analysis.(b and c) Diagrams showing thetermination of needle tracks formicroinjections aimed at theinfralimbic cortex (IL) and medialorbitofrontal cortex (mOFC), re-spectively. Figures are adaptedfrom Paxinos & Watson (2007),and A/P coordinates (in mm)are given relative to Bregma



presses indicated a significant effect of reinstatement[F(1,10) = 7.78, P<0.05], a significant effect of microin-jection [F(1,10) = 7.23, P<0.05] and a significantinteraction [F(1,10) = 7.017, P<0.05]. Post hoc testsrevealed that, although active lever pressing during rein-statement with vehicle treatment was significantlyhigher than the extinction baseline (P<0.05), activelever pressing during reinstatement following SCH treat-ment was no different from extinction baseline and wassignificantly lower than what was observed with thevehicle treatment (P<0.05). Intra-IL D1 receptorblockade had no effect on active lever pressing duringcocaine-prime reinstatement (Fig. 2b). A two-wayrepeated measures ANOVA of active lever presses indi-cated a significant effect of reinstatement [F(1,7)= 17.37, P<0.01], no effect of microinjection [F(1,7)= 2.28, P>0.05)] and no interaction [F(1,7) = 0.96,P>0.05]. Post hoc tests revealed that active lever press-ing during reinstatement was significantly higher thanextinction for both treatments (P<0.05). Intra-IL D1receptor blockade had no effect on cue+ cocaine-primereinstatement (Fig. 2c). A two-way repeated measuresANOVA of active lever presses indicated a significanteffect of reinstatement [F(1,6) = 27.53, P<0.01], noeffect of microinjection [F(1,6) = 1.07, P>0.05] and nointeraction [F(1,6) = 1.48, P>0.05]. Post hoc testsrevealed that active lever pressing during reinstatement

was significantly higher than extinction for both treat-ments (P<0.05). There were no significant differencesin inactive lever presses across any type of reinstatement(Fig. 2d–f).

Experiment 2

D2 receptor blockade in the IL has no effect on cocaine-seekingbehavior

Intra-IL D2 receptor blockade had no effect on activelever pressing during cued reinstatement (Fig. 3a). Atwo-way repeated measures ANOVA indicated a signifi-cant effect of reinstatement [F(1,13) = 13.60, P<0.05],no effect of microinjections [F(1,13) = 1.37, P>0.05]and no interaction [F(1,13) = 1.13, P>0.05]. Post hoctests revealed that active lever pressing during reinstate-ment was significantly higher than extinction for bothtreatments (vehicle versus sulpiride; P<0.05). Intra-ILD2 receptor blockade had no effect on active lever press-ing during cocaine-prime reinstatement (Fig. 3b). A two-way repeated measures ANOVA indicated a significanteffect of reinstatement [F(1,12) = 12.54, P<0.05], noeffect of microinjections [F(1,12) = 0.06, P>0.05] andno significant interaction [F(1,12) = 0.23, P>0.05]. Posthoc tests revealed that active lever pressing during rein-statement was significantly higher than extinction for

Figure 2 Intra-IL D1 receptorblockade via SCH 23390 (SCH) me-diates cued reinstatement (note thatthe y-axis maximum is either 100 or300 across the different experimentsto accommodate the large differ-ences in values). (a–c) The active le-ver presses and (d–f) the inactivelever presses for each reinstatementtest for experiment 1 are shown. (a)Intra-IL microinjections of the D1 re-ceptor antagonist SCH significantlyreduced active lever pressing duringcued reinstatement compared withvehicle controls (n= 11). (b) Intra-ILmicroinjections of SCH had no effecton active lever presses during co-caine-prime reinstatement (n= 8).(c) Intra-IL microinjections of SCHhad no effect on active lever pressingduring cued + cocaine-prime rein-statement (n= 7). (d–f) There wereno effects of reinstatement or micro-injections on inactive lever pressesfor any type of reinstatement test.*P< 0.05 compared with extinctionbaseline. #P< 0.05 compared withvehicle-control group. EXT, extinc-tion baseline



both treatments (P<0.05). Intra-IL D2 receptor block-ade had no effect on active lever pressing during cued+ cocaine-prime reinstatement (Fig. 3c). A two-wayrepeated measures ANOVA revealed a significant effectof reinstatement [F(1,11) = 17.18, P<0.05) but no effectof microinjections [F(1,11) = 3.12, P>0.05] or significantinteraction [F(1,11) = 3.90, P>0.05]. Post hoc tests re-vealed that active lever pressing during reinstatementwas significantly higher than extinction for both treat-ments (P<0.05). There were no significant differencesin inactive lever presses across any type of reinstatement(Fig. 3d–f).

Experiment 3

D1 receptor blockade in the mOFC reduces cocaine-seekingbehavior

Intra-mOFC D1 receptor blockade significantly reducedactive lever pressing during cued reinstatement (Fig. 4a).A two-way repeated measures ANOVA indicated a signifi-cant effect of reinstatement [F(1,8) = 24.00, P<0.05], asignificant effect of microinjections [F(1,8) = 5.12,P<0.05] and a significant interaction [F(1,8) = 7.51,P<0.05]. A post-hoc analysis revealed that, althoughvehicle-treated animals had significantly increased activelever pressing during reinstatement compared with theextinction baseline, active lever pressing during reinstate-ment following SCH treatment did not significantly differfrom extinction baseline and was significantly lower thanthat observed with vehicle treatment (P<0.05). Intra-mOFC D1 receptor blockade also resulted in a significant

reduction of active lever pressing during cocaine-prime re-instatement (Fig. 4b). A two-way repeated measuresANOVA indicated a significant effect of reinstatement[F(1,11) = 20.72, P<0.05], a significant effect of microin-jections [F(1,11) = 8.91, P<0.05] and a significant inter-action [F(1,11) = 10.07, P<0.05]. Post hoc analysisrevealed that the vehicle-treated rats demonstrated signif-icantly increased active lever pressing during reinstate-ment compared with extinction baseline (P<0.05). Incontrast, the SCH-treated rats showed no difference in ac-tive lever presses during reinstatement compared with ex-tinction baseline and had significantly fewer active leverpresses compared with the vehicle-treated group(P<0.01). Intra-mOFCD1 receptor blockade significantlyattenuated active lever pressing during cued+ cocaine-prime reinstatement (Fig. 4c). A two-way repeated mea-sures ANOVA revealed a significant effect of reinstatement[F(1,7) = 11.87, P<0.05], a significant effect of microin-jections [F(1,7) = 9.90, P<0.05] and a significant interac-tion [F(1,7) = 14.78, P<0.05]. Post hoc analysis showedthat although both the vehicle- and SCH-treated groupsdemonstrated significantly higher lever pressing duringreinstatement compared with extinction baseline, theSCH-treated animals had significantly fewer active leverpresses compared with the vehicle-treated group(P<0.05).

A two-way repeated measures ANOVA of inactive le-ver presses during cued reinstatement (Fig. 4d) revealeda significant effect of reinstatement [F(1,8) = 6.24,P<0.05], no effect of microinjections [F(1,8) = 2.04,P>0.05] and a significant interaction [F(1,8) = 8.86,

Figure 3 Intra-IL D2 receptorblockade via sulpiride does not altercued or cocaine-prime reinstate-ment. (a–c) The active lever pressesand (d-f) inactive lever presses foreach reinstatement test for experi-ment 2 are shown. (a) Intra-IL micro-injections of the D2 receptorantagonist sulpiride had no effect oncued reinstatement (n= 14). (b)Intra-IL microinjections of sulpiridehad no effect on cocaine-prime rein-statement (n= 13). (c) Intra-IL mi-croinjections of sulpiride had noeffect on cued + cocaine-prime rein-statement (n= 12). (d–f) There wereno effects of reinstatement or micro-injections on inactive lever pressesfor any type of reinstatement.*P< 0.05 compared with extinctionbaseline. EXT, extinction baseline



P<0.05]. Post hoc analysis indicated that the SCH-treated animals had significantly higher inactive leverpresses during extinction baseline when compared withboth the extinction baseline for the vehicle group andthe SCH-treated animals during reinstatement testing.However, analyses of inactive lever pressing for the co-caine prime and cued+ cocaine-prime reinstatementshowed no significant differences (Fig. 4e and f). Becausethe statistically significant increase in inactive leverpressing shown in Fig. 4d occurred during the extinctionbaseline when no manipulations were given, becausethere was no significant effect in the active lever pressesduring the same extinction baseline, and because therewere no other statistically significant effects of inactivelever pressing either in the baseline or reinstatement testsin this experiment or across all the experiments, it isdifficult to conclude that the observed effect is anythingbut a random statistical artifact.

Experiment 4

D2 receptor blockade in the mOFC has no effect oncocaine-seeking behavior

Intra-mOFC D2 receptor blockade had no effect on activelever presses during cued reinstatement (Fig. 5a). A two-

way repeated measures ANOVA of active lever pressesduring cued reinstatement revealed a significant effectof reinstatement [F(1,5) = 10.55, P<0.05], no effect ofmicroinjection [F(1,5) = 0.65, P>0.05] and no interac-tion [F(1,5) = 0.89, P>0.05]. Post hoc tests revealed thatactive lever pressing during reinstatement was signifi-cantly higher than extinction for both treatments.Figure 5b shows the active lever presses during cocaine-prime reinstatement following intra-mOFC D2 receptorblockade. A two-way repeated measures ANOVA of activelever pressing during cocaine-prime reinstatement re-vealed no effect of reinstatement [F(1,5) = 5.87,P>0.05], no effect of microinjection [F(1,5) = 0.17,P>0.05] and no interaction [F(1,5) = 0.10, P>0.05].Intra-mOFC D2 receptor blockade had no effect on activelever pressing during cued+ cocaine-prime reinstatement(Fig. 5c). A two-way repeated measures ANOVA of activelever presses revealed a significant effect of reinstatement[F(1,5) = 15.01, P<0.05], no effect of microinjections[F(1,5) = 0.20, P>0.05] and no significant interaction[F(1,5) = 0.31, P>0.05]. Post hoc tests revealed that ac-tive lever pressing during reinstatement was significantlyhigher than extinction for both treatments (P<0.05).

Because the cocaine-prime reinstatement did notproduce significantly higher lever pressing compared

Figure 4 Intra-mOFC D1 receptorblockade via SCH 23390 (SCH)mediates cocaine-seeking behaviors.(a–c) The active lever presses and(d–f) the inactive lever presses foreach reinstatement test for experi-ment 3 are shown. (a) Intra-mOFCmicroinjections of SCH significantlyreduced active lever pressing duringcued reinstatement compared withvehicle controls (n= 9). (b) Intra-mOFC microinjections of SCH signif-icantly reduced active lever pressingduring cocaine-prime reinstatement(n= 12). (c) Intra-mOFC microinjec-tions of SCH significantly reduced ac-tive lever pressing during cued+ cocaine-prime reinstatement(n= 8). (d) Intra-mOFC microinjec-tions significantly reduced inactive le-ver presses compared withextinction baseline, although the ex-tinction baseline itself was elevated.(e and f) There were no effects ofreinstatement or microinjections oninactive lever presses for cocaineprime or cued + cocaine-prime rein-statement. *P< 0.05 compared withextinction baseline. #P< 0.05 com-pared with vehicle-control group.EXT, extinction baseline



with the extinction baseline, we conducted a separateexperiment in which the rats underwent only cocaine-prime reinstatement with the goal of achieving morerobust levels of this form of reinstatement. Figure 5dshows the active lever presses for those rats that onlyunderwent the cocaine-prime reinstatement. A two-wayrepeated measures ANOVA of active lever pressing duringthis cocaine-prime reinstatement test revealed no effect ofmicroinjections [F(1,5) = 0.72, P>0.05], no interaction[F(1,5) = 0.06, P>0.05] and only a marginal effect ofreinstatement [F(1,5) = 2.96, P<0.15]. Post hoc testsrevealed that active lever pressing during reinstatementwas marginally higher than extinction for bothtreatments (P<0.09) and was not different betweentreatment groups. There were no significant differencesin inactive lever presses across any type of reinstatement.

DISCUSSION

In contrast to prior work using GABA receptor activation(Peters et al. 2008), the current findings indicate that D1receptor activation in the IL and mOFC is involved in thereinstatement of cocaine seeking, though to differentextents. Blocking D1 receptors in the IL reduced cuedreinstatement but did not affect cocaine prime or cued

+ cocaine-prime reinstatement. A similar blockade inthe mOFC reduced cocaine seeking for all forms ofreinstatement tested. In contrast, the present resultssuggest that D2 receptor activation in either structureplayed little role in the reinstatement of cocaine seekingas blocking D2 receptors in the IL and mOFC had noeffect on reinstatement. Although it is possible that thesulpiride dose used in the present experiments was toolow to produce any effects on drug seeking, sulpiride dosesgiven in the mPFC within this range have revealed behav-ioral differences in the past (Druzin et al. 2000; Pakdel &Rashidy-Pour 2007). Nonetheless, we cannot rule outthe possibility that higher doses may be effective at alter-ing drug seeking, although non-specific ‘off-target’ effectsmay become a concern at higher doses. These findingssuggest that activation of dopamine receptors within thevmPFC, particularly the D1 receptors, is involved in thereinstatement of cocaine seeking, although the preciserole appears to depend on an interaction of the type ofreinstatement and the vmPFC subregion.

Infralimbic results

Intra-IL administration of the D1 receptor antagonist re-duced cue-induced cocaine seeking but did not alter

Figure 5 Intra-mOFC D2 receptor blockade via sulpiride has no effect on cocaine-seeking behaviors. (a–d) The active lever presses and (e–h)inactive lever presses for each reinstatement test for experiment 4 are shown. (a) Intra-mOFC microinjections of sulpiride had no effect on activelever pressing during cued reinstatement compared with vehicle controls (n= 6). (b) Although intra-mOFC microinjections of sulpiride had noeffect on active lever pressing during cocaine-prime reinstatement, cocaine-prime reinstatement levels were not significantly above the extinctionbaseline (n= 6). (c) Intra-mOFC microinjections of sulpiride had no effect on active lever pressing during cued + cocaine-prime reinstatement(n= 6). (d) Due to the low levels of cocaine-prime reinstatement, a separate group of rats underwent only a cocaine-prime reinstatement(n= 6). Intra-mOFC microinjections of sulpiride did not affect active lever pressing when only a cocaine-prime reinstatement test was conducted.(e–h) There were no effects of reinstatement or microinjections on inactive lever presses for any type of reinstatement. *P< 0.05 comparedwith extinction baseline. @P< 0.09 compared with extinction baseline. EXT, extinction baseline



reinstatement when a cocaine prime was given. D2 re-ceptor blockade in the IL had no effect on any form of re-instatement. The results with IL dopamine receptorblockade show partial consistency with prior work, asIL inactivation via GABA receptor activation has no effecton cocaine-prime reinstatement (McFarland & Kalivas2001) and blocking 5HT2A receptors in the IL decreasescued and cocaine-prime reinstatement (Pockros et al.2011). Moreover, several studies, including work fromour laboratory, suggest that the IL inhibits cocaine seek-ing following self-administration and extinction and isalso involved in the consolidation of the extinction ofcocaine-seeking behavior (LaLumiere et al. 2010;LaLumiere et al. 2012; Peters et al. 2008). However, thosestudies also suggest that IL inactivation and activationhave no effect and reduce, respectively, cued reinstate-ment of cocaine seeking.

The reasons underlying this discrepancy regardingthe IL remain unclear, although to some degree, theyparallel the discrepancies with the mOFC results. How-ever, such conflicting findings are not unprecedented.Previous work investigating the nucleus accumbens shellhas found that inactivation via GABA receptor activationdoes not alter the reinstatement of cocaine seeking and,akin to the IL, induces cocaine seeking during an extinc-tion session (McFarland & Kalivas 2001; Peters et al.2008). In contrast, dopamine receptor blockade withinthe accumbens shell impairs the reinstatement of cocaineseeking (Anderson et al. 2006). Together with the pres-ent findings, this work suggests that conclusions drawnbased on one type of pharmacological manipulationmay not always predict the outcomes for a similar set ofstudies by using a different pharmacological manipula-tion. Previous work indicates that dopamine in the ac-cumbens shell appears to over-ride excitatory inputsfrom the IL that would normally suppress cocaine seeking(LaLumiere et al. 2012). A similar process may occurwithin the IL itself. More speculatively, this process mayalso involve different neuronal ensembles. Indeed, recentfindings suggest that a small minority of IL neurons is re-sponsible for the suppression of alcohol seeking (Pfarret al. 2015). However, there may also be IL neurons thatpromote cocaine seeking (e.g. Koya et al. 2009), and it ispossible that D1 receptor blockade on these specific neu-rons may inhibit the neuronal ensembles that promotecocaine seeking, resulting in a decrease in the reinstate-ment of cocaine seeking observed in the present study.These neuronal ensembles that drive cocaine seeking,however, may typically be masked by the pre-potent in-hibitory drive within the structure with regard to cocaineseeking. Unfortunately, it is not clear whether this wouldexplain the discrepancy for the mOFC, as prior work hasnot identified a role for this region in inhibiting cocaineseeking.

Alternatively, the actions of dopamine within the IL(and possibly mOFC) may have subtler effects on the typeof behavior in which an animal is engaged and may ex-plain the differences between the effects of D1 receptorblockade in the IL during cued and cocaine-prime rein-statement. Past studies have suggested that IL dopamineis critically involved in how the IL influences goal-seekingversus habit-based behavior (Hitchcott et al. 2007). Cuedreinstatement takes advantage of the ability of the cues toserve as conditioned reinforcers that increase cocaine-seeking behavior even when the reinforcement of cocaineitself is removed. Blockade of dopamine receptors in the ILmay, therefore, reduce any reinforcing properties of thecues, thus preventing the goal-seeking behavior. In con-trast, cocaine-prime reinstatement involves a single co-caine injection given prior to the reinstatement sessionwith no reinforcer given during the session itself. Thus,cocaine-prime reinstatement likely does not involve thesame type of goal-seeking behavior as cued reinstatementand, therefore, may not depend on IL dopamine recep-tors. Nonetheless, it may be a combination of type of rein-statement, and therefore goal-seeking behavior versushabitual behavior, and the selective activation of specificensembles within the IL that are responsible for the cur-rent results. Additionally, and perhaps in concert withthe speculation in the preceding texts, the lack of a rolefor IL dopamine receptors during cocaine-prime rein-statement may be due to the direct actions of cocaineon PL dopamine, thereby directly activating the PL-nucleus accumbens core pathway that is necessary forreinstatement (Stefanik et al. 2013). Indeed, prior studiesindicate that PL inactivation, dopamine receptor block-ade in the PL and optical inhibition of the PL inputs tothe accumbens core attenuate cocaine-prime reinstate-ment (McFarland & Kalivas 2001; Stefanik et al. 2013).

Medial orbitofrontal results

Blocking D1 receptors within the mOFC impaired allforms of reinstatement tested (cued, cocaine prime andcued+ cocaine-prime reinstatement), whereas D2 recep-tor blockade in the mOFC did not alter any form of rein-statement. Previous work indicates that mOFCactivation potentiates the feeding behavior that is in-duced by AMPA receptor blockade in the nucleus accum-bens shell which lies downstream from the mOFC(Hoover & Vertes 2011; Richard & Berridge 2013). Giventhat prior work indicates that inactivation of the accum-bens shell induces cocaine-seeking behavior (Peters et al.2008), it may be possible that activation of the mOFCwould potentiate cocaine seeking in the same way thatit potentiates feeding. Thus, the present results indicatinga role for the mOFC in promoting cocaine seeking may beconsistent with those examining feeding. Nonetheless,



prior studies directly examining the role of the mOFC incocaine seeking found no effect of mOFC inactivation oncued reinstatement or of mOFC lesions on cocaine-primereinstatement (Fuchs et al. 2004), making the presentstudy the first, to our knowledge, to show a role for thisregion in the reinstatement of cocaine seeking. Indeed,although much work has focused on the lateral OFCand medial PFC in general (PL and IL specifically) in drugseeking and reward-related behavior, there has been lessattention directed toward examining the role of themOFC in such behaviors. It is worth noting, however,that our results are consistent with previous studies indi-cating that D1 receptor blockade within the lateral OFCdisrupts cocaine-seeking behavior (Lasseter et al. 2009),suggesting that these two distinct regions may be actingsimilarly to influence reinstatement. Moreover, relativelyfew studies consider differences along the rostrocaudalaxis within the mPFC, which has likely obscured func-tional differences among these subregions. Thus, with arelative paucity of studies in the literature, it is difficultto reconcile the current results by using dopamine recep-tor blockade with those examining lesions and inactiva-tion of the mOFC. Nonetheless, the present study, alongwith previous work, provides a critical basis for futurework to examine this vmPFC subregion as well asconsider functional differences along the rostrocaudalaxis during motivated behaviors, including drug seekingand feeding.

D1 V D2 Receptors in the mPFC

Although previous work indicates that systemic andintra-mPFC D2 receptor blockade alters drug-seekingbehaviors (Liu et al. 2010; Sun & Rebec 2005;Woolverton & Virus 1989), the present results suggestthat D2 receptor blockade in the IL and mOFC did notaffect cocaine seeking. Given that D1 and D2 receptorstypically have excitatory and inhibitory influences,respectively, via opposing actions on adenylyl cyclaseactivity and cAMP production (Gonon 1997; Paul et al.1992), differential localization of these receptors on ILand mOFC neurons may account for our results. Workusing in situ hybridization indicates that both D1 andD2 receptor mRNAs are found in efferent cortical popula-tions (Gaspar et al. 1995). However, the cortical popula-tions expressing D1 and D2 receptors appear to differsignificantly, depending on the receptor type expressed.Indeed, Gaspar et al. estimated that D2 receptor mRNA-containing neurons were 3–4 times less numerous thanthose containing D1 receptor mRNA in the PFC andnoted that, although 25 percent of corticothalamic neu-rons express D1 receptors, these projection neurons donot appear to express D2 receptors. Other work hassuggested that D1 and D2 receptors within the mPFC

are located in minimally overlapping neuronal popula-tions (Vincent et al. 1995). Thus, the present resultsmay reflect mPFC differences in D1 versus D2 receptorexpression level and/or localization on different neuronalpopulations or even on different parts of the neurons(dendrites versus axon terminals). Consistent with theimportance of the D1 receptor in particular in the mPFC,prior research indicates that mPFC functions such asworking memory and mental flexibility depends on dopa-mine activity at D1 receptors (Goldman-Rakic et al.2000; Okubo et al. 1997; Sawaguchi & Goldman-Rakic1991; Williams & Goldman-Rakic 1995).

CONCLUSION

Examining the role of D1 and D2 receptors in the IL andmOFC during the reinstatement of cocaine seeking, thepresent study found that D1 receptor blockade in themOFC reduced all forms of reinstatement tested but, inthe IL, only reduced cued reinstatement. D2 receptorblockade had no effect in either structure. Together withprevious work, showing a role for the IL in inhibiting co-caine seeking, the present study suggests that the actionsof dopamine within the IL may be quite different fromthose of the structure as a whole and introduces an addi-tional layer of complexity in our attempt to understandhow the IL regulates drug-seeking behavior. Moreover,the current results are among the first to identify themOFC as a region driving the reinstatement of cocaineseeking and also confirm the existence of critical func-tional differences along the rostrocaudal axis within thevmPFC. Future studies examining the vmPFC shouldcarefully consider such differences in drawing functionalconclusions about this region.

Acknowledgements

The authors would like to thank Victoria Ewald for herassistance in conducting the experiments. This researchwas supported by NIH grant DA034684 (RTL).

Disclosure/Conflict of Interest

The authors declare no conflict of interest.

Author Contributions

CVC and RTL designed the experiments. CVC, ALG andWRW conducted the experiments. CVC analyzed thedata. CVC and RTL wrote the manuscript, and ALGprovided manuscript revisions. All authors criticallyreviewed the content and approved the final version forpublication.



References

Anderson SM, Schmidt HD, Pierce RC (2006) Administration ofthe D2 dopamine receptor antagonist sulpiride into the shell,but not the core, of the nucleus accumbens attenuatescocaine priming-induced reinstatement of drug seeking.Neuropsychopharmacology 31:1452–1461.

Barker JM, Torregrossa MM, Taylor JR (2013) Bidirectionalmodulation of infralimbic dopamine D1 and D2 receptor activ-ity regulates flexible reward seeking. Front Neurosci 7:126.

Bossert JM, Stern AL, Theberge FR, Cifani C, Koya E, Hope BT,Shaham Y (2011) Ventral medial prefrontal cortex neuronalensembles mediate context-induced relapse to heroin. NatNeurosci 14:420–422.

Brog JS, Salyapongse A, Deutch AY, Zahm DS (1993) Thepatterns of afferent innervation of the core and shell in the“accumbens” part of the rat ventral striatum: immunohisto-chemical detection of retrogradely transported fluoro-gold. JComp Neurol 338:255–278.

Cosme CV, Gutman AL, LaLumiere RT (2015) The dorsalagranular insular cortex regulates the cued reinstatement ofcocaine-seeking, but not food-seeking, behavior in rats.Neuropsychopharmacology 40:2425–2433.

Druzin MY, Kurzina NP, Malinina EP, Kozlov AP (2000) The ef-fects of local application of D2 selective dopaminergic drugsinto the medial prefrontal cortex of rats in a delayed spatialchoice task. Behav Brain Res 109:99–111.

Fuchs RA, Evans KA, Parker MP, See RE (2004) Differentialinvolvement of orbitofrontal cortex subregions in conditionedcue-induced and cocaine-primed reinstatement of cocaineseeking in rats. J Neurosci 24:6600–6610.

Gaspar P, Bloch B, Le Moine C (1995) D1 and D2 receptor geneexpression in the rat frontal cortex: cellular localization indifferent classes of efferent neurons. Eur J Neurosci7:1050–1063.

Goldman-Rakic PS, Muly EC 3rd, Williams GV (2000) D(1)receptors in prefrontal cells and circuits. Brain Res Brain ResRev 31:295–301.

Gonon F (1997) Prolonged and extrasynaptic excitatory actionof dopamine mediated by D1 receptors in the rat striatumin vivo. J Neurosci 17:5972–5978.

Granon S, Passetti F, Thomas KL, Dalley JW, Everitt BJ, RobbinsTW (2000) Enhanced and impaired attentional performanceafter infusion of D1 dopaminergic receptor agents into ratprefrontal cortex. J Neurosci 20:1208–1215.

Hitchcott PK, Quinn JJ, Taylor JR (2007) Bidirectional modula-tion of goal-directed actions by prefrontal cortical dopamine.Cereb Cortex 17:2820–2827.

Hoover WB, Vertes RP (2011) Projections of the medial orbitaland ventral orbital cortex in the rat. J Comp Neurol519:3766–3801.

Ishikawa A, Nakamura S (2003) Convergence and interaction ofhippocampal and amygdalar projections within the prefrontalcortex in the rat. J Neurosci 23:9987–9995.

Koob GF, Volkow ND (2010) Neurocircuitry of addiction.Neuropsychopharmacology 35:217–238.

Koya E, Uejima JL, Wihbey KA, Bossert JM, Hope BT, Shaham Y(2009) Role of ventral medial prefrontal cortex in incubation ofcocaine craving. Neuropharmacology 56(Suppl 1):177–185.

Lalumiere RT, Nguyen LT, McGaugh JL (2004) Post-trainingintrabasolateral amygdala infusions of dopamine modulateconsolidation of inhibitory avoidance memory: involvementof noradrenergic and cholinergic systems. Eur J Neurosci20:2804–2810.

LaLumiere RT, Niehoff KE, Kalivas PW (2010) The infralimbiccortex regulates the consolidation of extinction after cocaineself-administration. Learn Mem 17:168–175.

LaLumiere RT, Smith KC, Kalivas PW (2012) Neural circuitcompetition in cocaine-seeking: roles of the infralimbic cortexand nucleus accumbens shell. Eur J Neurosci 35:614–622.

Lasseter HC, Ramirez DR, Xie X, Fuchs RA (2009) Involvementof the lateral orbitofrontal cortex in drug context-inducedreinstatement of cocaine-seeking behavior in rats. Eur JNeurosci 30:1370–1381.

Lasseter HC, Xie X, Arguello AA, Wells AM, Hodges MA, FuchsRA (2014) Contribution of a mesocorticolimbic subcircuit todrug context-induced reinstatement of cocaine-seekingbehavior in rats. Neuropsychopharmacology 39:660–669.

Liu X, Jernigen C, Gharib M, Booth S, Caggiula AR, Sved AF(2010) Effects of dopamine antagonists on drug cue-inducedreinstatement of nicotine-seeking behavior in rats. BehavPharmacol 21:153–160.

Malkusz DC, Yenko I, Rotella FM, Banakos T, Olsson K, Dindyal T,Vig V, Bodnar RJ (2015) Dopamine receptor signaling in themedial orbital frontal cortex and the acquisition and expres-sion of fructose-conditioned flavor preferences in rats. BrainRes 1596:116–125.

McFarland K, Kalivas PW (2001) The circuitry mediatingcocaine-induced reinstatement of drug-seeking behavior. JNeurosci 21:8655–8663.

Okubo Y, Suhara T, Suzuki K, Kobayashi K, Inoue O, Terasaki O,Someya Y, Sassa T, Sudo Y, Matsushima E, Iyo M, Tateno Y,Toru M (1997) Decreased prefrontal dopamine D1 receptorsin schizophrenia revealed by PET. Nature 385:634–636.

Pakdel R, Rashidy-Pour A (2007) Microinjections of the dopa-mine D2 receptor antagonist sulpiride into the medial prefron-tal cortex attenuate glucocorticoid-induced impairment oflong-term memory retrieval in rats. Neurobiol Learn Mem87:385–390.

Paul ML, Graybiel AM, David JC, Robertson HA (1992) D1-likeand D2-like dopamine receptors synergistically activaterotation and c-fos expression in the dopamine-depleted stria-tum in a rat model of Parkinson’s disease. J Neurosci12:3729–3742.

Paxinos G, Watson C (2007) The rat brain in stereotaxic coordi-nates. 6th ed. Amsterdam, Boston: Academic Press/Elsevier.

Peters J, LaLumiere RT, Kalivas PW (2008) Infralimbic prefrontalcortex is responsible for inhibiting cocaine seeking inextinguished rats. J Neurosci 28:6046–6053.

Peters J, Pattij T, De Vries TJ (2013) Targeting cocaine versusheroin memories: divergent roles within ventromedial prefron-tal cortex. Trends Pharmacol Sci 34:689–695.

Pfarr S, Meinhardt MW, Klee ML, Hansson AC, Vengeliene V,Schonig K, Bartsch D, Hope BT, Spanagel R, Sommer WH(2015) Losing control: excessive alcohol seeking after selectiveinactivation of cue-responsive neurons in the infralimbiccortex. J Neurosci 35:10750–10761.

Pockros LA, Pentkowski NS, Swinford SE, Neisewander JL (2011)Blockade of 5-HT2A receptors in the medial prefrontal cortexattenuates reinstatement of cue-elicited cocaine-seekingbehavior in rats. Psychopharmacology (Berl) 213:307–320.

Richard JM, Berridge KC (2013) Prefrontal cortex modulatesdesire and dread generated by nucleus accumbens glutamatedisruption. Biol Psychiatry 73:360–370.

Rogers JL, Ghee S, See RE (2008) The neural circuitry underly-ing reinstatement of heroin-seeking behavior in an animalmodel of relapse. Neuroscience 151:579–588.



Sawaguchi T, Goldman-Rakic PS (1991) D1 dopamine receptorsin prefrontal cortex: involvement in working memory. Science251:947–950.

Setlow B, McGaugh JL (1998) Sulpiride infused into the nucleusaccumbens posttraining impairs memory of spatial watermaze training. Behav Neurosci 112:603–610.

Smiley JF, Williams SM, Szigeti K, Goldman-Rakic PS (1992)Light and electron microscopic characterization ofdopamine-immunoreactive axons in human cerebral cortex.J Comp Neurol 321:325–335.

Smith KS, Berridge KC (2005) The ventral pallidum and hedonicreward: neurochemical maps of sucrose “liking” and foodintake. J Neurosci 25:8637–8649.

Stefanik MT, Moussawi K, Kupchik YM, Smith KC, Miller RL,Huff ML, Deisseroth K, Kalivas PW, LaLumiere RT (2013)Optogenetic inhibition of cocaine seeking in rats. Addict Biol18:50–53.

Sun W, Rebec GV (2005) The role of prefrontal cortex D1-likeand D2-like receptors in cocaine-seeking behavior in rats.Psychopharmacology (Berl) 177:315–323.

Swanson LW (1982) The projections of the ventral tegmentalarea and adjacent regions: a combined fluorescent retrograde

tracer and immunofluorescence study in the rat. Brain ResBull 9:321–353.

Van den Oever MC, Spijker S, Smit AB, De Vries TJ (2010)Prefrontal cortex plasticity mechanisms in drug seeking andrelapse. Neurosci Biobehav Rev 35:276–284.

Van Eden CG, Hoorneman EM, Buijs RM, Matthijssen MA,Geffard M, Uylings HB (1987) Immunocytochemical localiza-tion of dopamine in the prefrontal cortex of the rat at thelight and electron microscopical level. Neuroscience22:849–862.

Vincent SL, Khan Y, Benes FM (1995) Cellular colocalization ofdopamine D1 and D2 receptors in rat medial prefrontal cortex.Synapse 19:112–120.

Williams GV, Goldman-Rakic PS (1995) Modulation of memoryfields by dopamine D1 receptors in prefrontal cortex. Nature376:572–575.

Winter S, Dieckmann M, Schwabe K (2009) Dopamine in theprefrontal cortex regulates rats behavioral flexibility to chang-ing reward value. Behav Brain Res 198:206–213.

Woolverton WL, Virus RM (1989) The effects of a D1 and a D2dopamine antagonist on behavior maintained by cocaine orfood. Pharmacol Biochem Behav 32:691–697.



d1, but not d2, receptor blockade within the infralimbic and … · d1, but not d2, receptor...

Documents