SACD

Bpees

Mtta

RAcs

Cs

Kr

Rialasaisiari9sptas

F

A

R

0d

elective Norepinephrine Reuptake Inhibition bytomoxetine Prevents Cue-Induced Heroin andocaine Seeking

aina Economidou, Jeffrey W. Dalley, and Barry J. Everitt

ackground: Preventing relapse to drug use is a major challenge for drug addiction treatment. We have recently shown that impulsivityredating drug-taking increases the susceptibility to relapse to cocaine seeking and that treatment with the anti-impulsivity drug atomox-tine (ATO), a selective norepinephrine re-uptake inhibitor (norepinephrine transporter), prevents relapse. Here, we investigated further theffects of ATO on cue-maintained heroin and cocaine seeking and relapse and compared these effects with those of the anti-impulsivitytimulant drug methylphenidate (MPH).

ethods: Rats were trained to seek and self-administer cocaine or heroin under a second-order schedule of reinforcement. After acquisi-ion of stable responding, groups of rats (n � 10 –12) were treated, in a within-subject design, with either ATO or MPH (.3–3.0 mg/kg IP), andhe effects on cocaine and heroin seeking were measured. The effects of ATO (.3–1.0 mg/kg) on cue-induced relapse to cocaine seeking after1-week period of abstinence were also studied.

esults: Atomoxetine significantly decreased both cue-controlled cocaine and heroin seeking, whereas MPH had no significant effect.tomoxetine also significantly attenuated cue-induced relapse to cocaine seeking after abstinence. The effects of ATO were selective forue-controlled drug-seeking, because it did not affect responding in the absence of the drug-paired cue; nor did it alter responding for oralucrose, except minimally at the highest dose, or locomotor activity.

onclusions: Selective norepinephrine transporter inhibition by ATO might be an effective treatment for the prevention of relapse to both

timulant and opiate addiction.

ey Words: Addiction, atomoxetine, impulsivity, methylphenidate,elapse, second-order

elapse to drug use after days or even years of abstinence is botha characteristic of drug addiction and a challenge to successfultreatment (1). A key goal for the development of novel behav-

oral and pharmacotherapies, therefore, is to understand the neuralnd psychological mechanisms underlying the susceptibility to re-

apse. Among the factors that influence relapse is impulsivity, which islso a risk factor for stimulant addiction (2–5). We previously demon-trated that high impulsive behavior predating drug use predisposednimals to relapse to cocaine seeking with a novel procedure in which

ntermittent punishment of cocaine-seeking responses resulted in ab-tinence (6); this increased propensity to relapse was especially evidentn rats with an extended history of cocaine self-administration (6). Welso showed that treatment with the selective norepinephrine (NE)euptake inhibitor atomoxetine (ATO), a drug used clinically to treatmpulsivity (e.g., attention-deficit/hyperactivity disorder [ADHD]) (7–), was highly effective in preventing the reinstatement of cocaineeeking (6). However, it was further observed that ATO reduced theropensity to relapse in rats with low levels of impulsivity, regardless of

he duration of their cocaine-taking history (6), suggesting that thenti-relapse properties of ATO might not be restricted to high-impul-ive individuals that are vulnerable to addiction and relapse.

rom the Behavioral and Clinical Neuroscience Institute and Department ofExperimental Psychology (DE, JWD, BJE), and the Department of Psychi-atry (JWD), University of Cambridge, Addenbrooke’s Hospital, Cam-bridge, United Kingdom.

ddress correspondence to Daina Economidou, Ph.D., Department of Ex-perimental Psychology, University of Cambridge, Downing Street, Cam-bridge CB2 3EB, UK; E-mail: de244@cam.ac.uk.

eceived Jul 20, 2010; revised Sep 26, 2010; accepted Sep 28, 2010.

006-3223/$36.00oi:10.1016/j.biopsych.2010.09.040

In the present study, therefore, we investigated further the be-havioral and pharmacological specificity of ATO in other animalmodels of drug seeking and relapse that particularly emphasize therole played by drug-associated, conditioned stimuli (CS), which areknown to induce craving and relapse in addicted individuals (10 –12). Under a second-order schedule of reinforcement, cocaine orheroin seeking is maintained over long time periods by contingentpresentations of drug-associated CS, which act as conditioned rein-forcers (13–17). This paradigm also has the advantage of allowingdrug seeking to be measured independently of the response-rate–altering effects of cocaine or heroin (16). To study relapse, weadopted the procedure in which, after self-administration training,drug-seeking responses are reinforced solely by a drug-paired CSbut in the absence of drug reinforcement (i.e., in extinction) after aperiod of enforced abstinence (18,19).

We have also compared the effects of methylphenidate (MPH)on drug seeking, because this drug is used widely in the treatmentof impulse control disorders, including ADHD (7–9), as well as hav-ing the effect of reducing impulsivity in animals (20 –23). Atomox-etine and MPH are pharmacologically distinct: MPH is a psycho-stimulant that inhibits the dopaminergic transporter (DAT), but italso has substantial affinity for the norepinephrine transporter(NET); ATO, by contrast, is a selective NET inhibitor (24 –29). Weshow that ATO but not MPH had the marked effect of reducing bothcocaine and heroin seeking as well as relapse to cocaine seeking butdid not affect general locomotor activity or responding for a high-incentive ingestive reward at the lower effective dose. The resultssuggest the clinical potential for ATO to reduce relapse in bothcocaine and heroin addicts.

Methods and Materials

AnimalsOutbred male Lister Hooded rats (Charles River, Margate, United

Kingdom) weighing 300 –330 g at the beginning of the experi-

BIOL PSYCHIATRY 2011;69:266–274© 2011 Society of Biological Psychiatry

mtltiP

D

UsKwbhdi

I

mUG2rctB

P

OiAtissotCtticmdhomtwsfiwnm

Amspt

D. Economidou et al. BIOL PSYCHIATRY 2011;69:266–274 267

ents were used. Rats were housed individually, under tempera-ure- and humidity-controlled conditions and a reversed 12-houright/dark cycle (lights off at 7:00 AM). Water was available ad libi-um, and food was given at the end of each day’s testing. All exper-mental procedures conformed to the UK (1986) Animal (Scientificrocedures) Act (Project license 80/2234).

rugsAtomoxetine hydrochloride (a gift from Eli Lilly, Basingstoke,

nited Kingdom) was dissolved in .01 mol/L phosphate-bufferedaline. Methylphenidate hydrochloride (Sigma, Cambridge, Unitedingdom) was dissolved in physiological sterile saline. Both drugsere given by IP injection in a dose volume of 1 mL/kg. Doses wereased on previous studies (20,30). Cocaine hydrochloride anderoin hydrochloride (McFarlan-Smith, Edinburgh, United King-om) were dissolved in sterile physiological saline and were

nfused intravenously at the rate of .1 mL/5 sec.

V SurgeryRats were anesthetized IP with ketamine hydrochloride (100

g/kg; Ketaset, Fort Dodge Animal Health LTD, Southampton,nited Kingdom) and xylazine (9 mg/kg; Rompun, Bayer, Newbury,ermany), supplemented with ketamine as needed (approximately0 mg/kg) and received implantation of a single catheter in theight jugular vein. Catheters made from 22-gauge stainless steelannulae and Silastic tubing were located subcutaneously betweenhe scapulae (5). Rats were treated after surgery with 10 mg/kgaytril (Bayer) to prevent postoperative infection.

roceduresCocaine or Heroin Self-Administration Under a Second-

rder Schedule of Reinforcement. Daily experimental testingn the operant self-administration chambers (Med Associates, St.lbans, Vermont) began 7–10 days after IV surgery. Rats were

rained daily (2-hour sessions) to self-administer cocaine (.25 mg/nfusion) or heroin (.04 mg/infusion) under a fixed-ratio (FR1)chedule such that each active lever press resulted in a drug infu-ion, illumination of a CS light above the lever for 20 sec, retractionf both levers, and the extinction of the house-light for 20 sec. After

his 20-sec time-out (TO), the house-light was again illuminated, theS was extinguished, and the two levers were again inserted into

he chamber. Active and inactive levers were counterbalanced be-ween left and right sides for individual animals. Responses on thenactive lever had no programmed consequences but were re-orded to assess discriminated responding and general levels ofotor activity. Rats were limited to a maximum of 30 infusions

uring each 2-hour session. After the acquisition of cocaine oreroin self-administration (3–5 days), a fixed-interval (FI) schedulef reinforcement was introduced that was increased daily from FI1in to FI2, FI4, FI8, and FI10 min, before stabilizing at FI15 min for

hree consecutive sessions. Subsequently, a second-order scheduleas introduced, in which every 10th active lever press resulted in a

hort CS presentation for 1 sec (FI15[FR10:S]). On completion of therst 10 responses after the FI15 min had elapsed, cocaine or heroinas infused, and the CS was presented for 20 sec. Sessions termi-ated after either five infusions or 2 hours, whichever criterion waset first.

Cue-Induced Relapse to Cocaine Seeking After Abstinence.fter IV surgery, rats were trained to self-administer cocaine (.25g/infusion) under an FR1 schedule of reinforcement (daily 1-hour

essions). Each cocaine infusion was followed by a 20 sec TO accom-anied by CS illumination, the retraction of both levers, and extinc-

ion of the house-light. After the termination of the TO, the house-

light was again illuminated, the CS was extinguished, and the twolevers were inserted into the chamber. Responses on the inactivelever were recorded but had no programmed consequences. Ratswere limited to a maximum of 30 infusions during the first 2–3sessions of acquisition. Animals were then allowed to self-adminis-ter cocaine for 6-hour daily sessions or until a maximum of 150cocaine infusions were obtained (cocaine long-access) for 10 con-secutive days. Rats were then withdrawn from cocaine self-admin-istration and maintained in their home-cages for 1 week. At thetermination of the abstinence period, animals were tested during a1-hour extinction session, where responding on the active leverresulted in the presentation of the CS but no cocaine.

Experiment 1: Effects of ATO on Cocaine and Heroin SeekingUnder a Second-Order Schedule of Reinforcement

Rats were trained to self-administer cocaine (n � 12) or heroin(n � 12) under a second-order schedule of reinforcement (14 –16).After extended training (�1 month) during which habitual re-sponding was consolidated (31), animals were administered ATO(.3, 1.0, and 3.0 mg/kg, IP) or its vehicle in a counterbalanced man-ner and tested for cocaine seeking 20 min later. Drug testing wasrepeated every fourth day. On the first day after drug treatmentanimals remained in their home-cages. On Day 2 and Day 3, ratswere re-baselined under the second-order schedule.

Experiment 2: Effects of ATO on Cocaine Seeking Under anFI15-Min Schedule of Reinforcement

Rats (n � 12) were trained to respond for cocaine under anFI15-min schedule of reinforcement (i.e., with no response-contin-gent CS presentations during the 15-min interval). After stable re-sponding, animals were treated in a within-subjects design withATO (.3, 1.0, or 3.0 mg/kg, IP) or its vehicle 20 min before theinitiation of the session. Drug testing was again performed everyfourth day (see Experiment 1).

Experiment 3: Effects of MPH on Cocaine and Heroin SeekingUnder a Second-Order Schedule of Reinforcement

Rats were trained to self-administer cocaine (n � 10) or heroin(n � 12) under a second-order schedule of reinforcement. Afterextensive second-order training (� 1 month) animals received, in awithin-subjects design, MPH at the doses of .3, 1.0, and 3.0 mg/kg IPor its vehicle 30 min before the initiation of each session. Drugtesting was conducted every fourth day (see Experiment 1).

Experiment 4: Effects of ATO on Cue-Induced Relapse toCocaine Seeking After Abstinence

After long-access exposure to cocaine (10 sessions, 6 hours/session), rats were divided into three groups (n � 6 –7/group) withsimilar baseline levels of cocaine self-administration and thentested for cue-induced relapse after 1 week of withdrawal. For therelapse test, one group of rats received injection IP with vehicle,whereas the other two groups received ATO at the doses of .3 and1.0 mg/kg, respectively, 20 min before testing commenced.

Statistical AnalysisActive and inactive lever responses for the first (drug-free) and

second (after drug infusion) 15-min intervals (Experiments 1–3)were analyzed with a one-way within-subject analysis of variance(ANOVA) (SPSS, Chicago, Illinois). For the cue-induced relapse ex-periment, data were analyzed by two-way ANOVA with one be-tween-subjects factor (drug treatment) and one within-subject fac-tor (relapse). Post hoc comparisons were performed with the

Newman–Keuls test. Statistical significance was set at p � .05.

www.sobp.org/journal

R

EU

urrsienp.ialas

motl.

smsrAn6

Fic

268 BIOL PSYCHIATRY 2011;69:266–274 D. Economidou et al.

w

esults

xperiment 1: Effects of ATO on Cocaine and Heroin Seekingnder a Second-Order Schedule of Reinforcement

Cocaine. All animals (n � 12) acquired responding for cocainender a second-order schedule, making on average nearly 300

esponses in 15 min for the first cocaine infusion. As expected,esponding increased during the second interval, because of theelf-administration of cocaine (14). During the course of the exper-ment, catheter patency was lost in one animal, which was thereforexcluded from the analysis. Treatment with ATO resulted in a sig-ificant dose-dependent decrease in the number of active leverresses during the first (before cocaine) interval [F (3,30) � 6.7; p �

01]. As shown in Figure 1A, post hoc comparisons revealed a signif-cant decrease in cocaine seeking after treatment with 1.0 (p � .05)nd 3.0 mg/kg (p � .001). A small but significant decrease in active

ever pressing was also observed during the second interval (i.e.,fter cocaine self-administration) [F (3,30) � 3.1; p � .05], which wasignificant at the 3.0 mg/kg dose (p � .05) (Figure 1C).

There was also a small but significant overall effect of ATO treat-ent on inactive lever responses during the first seeking interval

nly [F (3,30) � 3.2; p � .05] (Figure 1B), which was significant only athe highest dose tested (p � .05). No significant effects on inactiveever responding were found during the second interval [F (3,30) �8; p � ns] (Figure 1D).

Heroin. All 12 animals acquired responding for heroin under aecond-order schedule, making on average 300 responses in 15

in for the first heroin infusion. As expected, responding during theecond interval was lower than during the first, because of theesponse-suppressing effects of heroin (15). As shown in Figure 2A,NOVA revealed a significant overall effect of ATO treatment on theumber of active lever responses during the first interval [F (3,33) �

igure 1. Effects of atomoxetine on cocaine seeking under a second-order scnfusion (second interval) (C, D). Data shown are mean (� SEM) number of rompared with vehicle-treated animals.

.1; p � .01], with Newman–Keuls post hoc tests confirming a sig-

ww.sobp.org/journal

nificant decrease at the doses of 1.0 (p � .05) and 3.0 mg/kg (p �.01) (Figure 2A). Atomoxetine did not affect responding on theactive lever during the second interval [F (3,33) � .5; p � ns] (Figure2C) or inactive lever responding [first: F (3,33) � 1.9; p � ns; second:F (3,33) � .4; p � ns] (Figures 2B and 2D, respectively).

Experiment 2: Effects of ATO on Cocaine Seeking Under aFI15-Min Schedule of Reinforcement

All animals successfully acquired responding under a FI15-minschedule of cocaine reinforcement. Responding was significantlylower than under the second-order schedule, averaging 70 re-sponses for the first cocaine infusion (p � .001). There was nosignificant effect of ATO on the number of active [first: F (3,24) � .6;p � ns; second: F (3,24) � 1.0; p � ns] or inactive [first: F (3,24) � 1.2;p � ns; second: F (3,24) � .4; p � ns] lever responses (Table 1).

Experiment 3: Effects of MPH on Cocaine and Heroin SeekingUnder a Second-Order Schedule of Reinforcement

All animals acquired lever pressing for cocaine (n � 10) or heroin(n � 12) under the second-order schedule. Catheter patency waslost in three animals (two in the cocaine group, and one in theheroin group), which were therefore excluded from further analy-sis. An ANOVA revealed no significant effects of MPH treatment oneither cocaine or heroin seeking: 1) cocaine active lever [first: F (3,21) �.5; p�ns; second: F (3,21)� .5; p�ns]; inactive lever [first: F (3,21)�2.9;p � ns; second: F (3,21) � .9; p � ns] (Figure 3A–3D); and 2) heroinactive lever [first: F (2,20) � 1.2; p � ns; second: F (2,20) � .7; p � ns];inactive lever [first: F (2,20) � 1.8; p � ns; second: F (2,20) � 1.3; p � ns](Figures 4A–4D).

Experiment 4: Effects of ATO on Cue-Induced Relapse toCocaine Seeking After Abstinence

During the cocaine long-access phase the animals escalated

le of reinforcement during the first drug-free interval (A, B) and after cocainenses on the active (A, C) and inactive lever (B, D). *p � .05, and ***p � .001,

heduespo

their cocaine intake, which increased from 83.2 � 6.1 on the first

dtl5ada[imt

D

tcrcau

Ftc

T

1

2

as

D. Economidou et al. BIOL PSYCHIATRY 2011;69:266–274 269

ay to 110.2 � 6.0 cocaine infusions by the last day of self-adminis-ration (6-hour sessions). Atomoxetine reduced cue-induced re-apse to cocaine seeking after a 1-week period of abstinence (Figure). An ANOVA revealed a significant increase in responding on thective lever [F (1,17) � 83.3; p � .001], with all treatment groupsemonstrating significant relapse to cocaine seeking (p � .01 – .001)nd a significant interaction between relapse and ATO treatment

F (2,17) � 3.7; p � .05]. Newman–Keuls post hoc tests showed a signif-cant decrease in cue-induced relapse after ATO treatment with 1.0

g/kg (p � .01) (Figure 5). There was no significant effect of any of thereatments on inactive lever responding (data not shown).

iscussion

The results of this study show that selective NE reuptake inhibi-ion by ATO had the remarkable effect of greatly reducing bothocaine and heroin seeking as well as diminishing the propensity toelapse to cocaine seeking. By contrast, MPH had no effect on eitherocaine or heroin seeking and, as with ATO, at doses known toffect impulsivity (21). The effects of ATO were especially seennder the second-order schedule of reinforcement, in which seek-

igure 2. Effects of atomoxetine on heroin seeking under a second-order sche active (A, C) and inactive lever (B, D) during the first drug-free intervalompared with vehicle-treated animals.

able 1. Summary of the Effects of ATO or Its Vehicle

Veh

st Seeking IntervalActive lever 66.7 � 19.5 1Inactive lever 7.4 � .8

nd Seeking IntervalActive lever 135.9 � 2.5 1Inactive lever 13.1 � 2.6

A summary of the effects of atomoxetine (ATO) (.3, 1.0, and 3.0 mg/kg, Ifixed-interval 15-min schedule of cocaine self-administration. Data are m

econd 15-min intervals.

ing responses are maintained with vigor over time through thecontingent presentation of drug-associated conditioned reinforc-ers (16,17). Atomoxetine had no effect on responding for a high-incentive natural reinforcer (sucrose, FR1) and had only minor ef-fects when tested at the highest dose of 3.0 mg/kg under higherratios or longer intervals of reinforcement (FR7-, RI60-, FI15-min)(Experiment S1 in Supplement 1). The effect of ATO could not beattributed to general sedation, because it had no significant effecton locomotor activity in a novel environment (Experiment S2 inSupplement 1). These data make it clear that the effect of ATO toreduce cocaine and heroin seeking as well as relapse to cocaineseeking (in two very different procedures, the one used here andalso the abstinence-relapse procedure of Economidou et al. [6])cannot be attributed to its anti-impulsivity effects, which are well-documented in animals and humans, but to another mechanism.This unique effect of ATO suggests potentially significant therapeuticpotential in relapse prevention.

Few previous studies have investigated the involvement of NErgicmechanisms in cocaine and heroin seeking, save important datashowing the engagement of NErgic transmission in stress-induced

le of reinforcement. Data shown are mean (� SEM) number of responses on) and after heroin infusion, second interval (C, D). *p � .05, and **p � .01,

.3 ATO 1.0 ATO 3.0

24.3 70.6 � 14.5 60.1 � 26.81.6 6.2 � 1.5 5.2 � .2

45.6 146.1 � 25.9 108.2 � 18.76.7 12.4 � 2.2 14.6 � 2.3

ts vehicle (Veh) on active and inactive lever responses in rats trained under� SEM) values of the total number of lever responses during the first and

hedu(A, B

ATO

00.2 �4.1 �

75.9 �17.8 �

P) or iean (

www.sobp.org/journal

rsce

Fc ) num

Fo(

270 BIOL PSYCHIATRY 2011;69:266–274 D. Economidou et al.

w

elapse after extinction (32–34). Platt et al. (35) showed the effects ofelective NET inhibitors on cocaine seeking in the presence of co-aine-associated stimuli after extinction in squirrel monkeys. How-ver, the results of this study are not easily compared with the

igure 3. Effect of methylphenidate on cocaine seeking under a second-orocaine infusion, second interval (C, D). Data are presented as mean (� SEM

igure 4. Effect of methylphenidate on heroin seeking under a secondf responses on the active (A, C) and inactive lever (B, D) during the first

C, D).

ww.sobp.org/journal

present data, because cued relapse was not observed and becauseinstrumental responding was itself extinguished (35). To date, onlydopamine D1 receptor antagonists have been shown to reducecue-induced relapse to both cocaine and heroin seeking after ex-

hedule of reinforcement during the first drug-free interval (A, B) and afterber of responses on the active (A, C) and inactive lever (B, D).

er schedule of reinforcement. Data shown are mean (� SEM) number-free interval (A, B) and after heroin self-administration, second interval

der sc

-orddrug

titwrrlbchanbtoosr

apwf1fphtetsMDctoccA

Fcowvm

D. Economidou et al. BIOL PSYCHIATRY 2011;69:266–274 271

inction of the drug-taking response (36,37). However, in employ-ng extinction-reinstatement procedures, it is difficult to rule outhe possibility that these potentially effective treatments interactith the neural mechanisms underlying extinction learning and

etention—for example mediated by the infralimbic cortex (38)—ather than cued mechanisms of relapse that are of the most trans-ational interest, because extinction of instrumental drug-takingehavior is not a feature of relapse-prevention treatments in thelinic (39,40). A metabotropic glutamatergic 2/3 receptor agonistas been shown to reduce cued reinstatement of cocaine seekingnd context-induced reinstatement of heroin seeking after absti-ence (41– 43). The �-aminobutyric acid (GABA)B receptor agonistaclofen was also shown to decrease heroin seeking to some ex-

ent, in addition to a larger effect on cocaine seeking under a sec-nd-order schedule of reinforcement. However, it also affectedther measures of responding, suggesting that in addition to itshort elimination half-life treatment with baclofen would noteadily translate to the clinic (44).

Despite their common use in the treatment of ADHD (7–9), ATOnd MPH have quite distinct pharmacological properties. Methyl-henidate has a rather mixed pharmacology—it binds to the DATith tenfold higher affinity than to the NET but has very low affinity

or the serotonin transporter (SERT), with Ki values of 34,339, and �0,000 nmol/L, respectively, as well as having substantial affinities

or the serotonin 5-HT1A and 5-HT2B receptors. By contrast, ATO is aotent and selective NET inhibitor with a Ki value of 5 nmol/L butas a 15- and 290-fold lower affinity for the SERT and DAT, respec-

ively (24 –29,45,46). These data might indicate that the behavioralffects of ATO to reduce cocaine and heroin seeking depend uponhe inhibition of NE reuptake and not its relatively lesser effects onerotonin and dopamine reuptake. The complete lack of effect ofPH on cocaine and heroin seeking indicates that inhibition of theAT has no effect on drug seeking, but this itself requires additional

omment. As a stimulant drug, it might even have been expectedhat MPH might increase responding for cocaine under a second-rder schedule of reinforcement, because self-administered co-aine has this effect (14). However, the baseline seeking rates forocaine were significantly higher than in our earlier studies (e.g.,

igure 5. Effect of atomoxetine treatment on cue-induced relapse to co-aine seeking after abstinence. Data are mean (� SEM) number of responsesn the active lever during the last cocaine self-administration day (cocaine;hite bar) and at relapse day (relapse; dark bars). **p � .01, and ***p � .001,

ersus cocaine, ##p � .01, difference from vehicle-treated (.0 mg/kg) ani-als.

rroyo et al. [14]), and thus it might have been more difficult to see

increases in responding with MPH pretreatment under the condi-tions of this experiment. Indeed, the current data show that self-administered cocaine only marginally increased the high rate ofresponding seen in the first drug-free seeking interval.

The major effects of ATO were seen when drug seeking wasmaintained over time by drug-associated stimuli acting as condi-tioned reinforcers. The effect to decrease relapse was quantitativelyless impressive than that to reduce cocaine and heroin seeking, butresponse rates in the relapse procedure are much lower, and thetesting is carried out in extinction— unlike responding under sec-ond-order schedules of drug reinforcement. Because ATO had min-imal or no effects under fixed and random interval schedules andmore effortful ratios of reinforcement, it is unlikely that decreasingresponding that must be maintained over long time periods or athigher rates for cocaine or heroin alone provides an explanation forits action. Instead it might reflect interactions between increased cen-tral NE transmission and the neural mechanisms underlying the con-trol over drug seeking by cocaine- or heroin-associated conditionedreinforcers.

Clinical trials with tricyclic antidepressant agents, such as desi-pramine, which preferentially enhance NErgic transmission (bind-ing affinity NET � SERT � DAT) (24,28,47) also suggest this behav-ioral locus of action. Thus, in cocaine-addicted individuals,desipramine decreased “craving” and resulted in longer abstinenceperiods in those that completed the treatment trial. However, therewere high rates of early treatment termination because of adverseside effect events (48 –51), perhaps indicating further the treatmentpotential of ATO, because it is clinically well-tolerated in humans(52–54). Atomoxetine has also been shown to reduce the physio-logical effects of cocaine and amphetamine but only to attenuatethe subjective effects of amphetamine in human subjects (55–57).In addition, in ADHD patients with comorbid alcohol abuse, treat-ment with ATO resulted in a reduction of cumulative heavy drinkingdays as well as significantly improving ADHD symptoms (58).

The neural basis of drug seeking under a second-order scheduleof reinforcement has primarily been defined in studies with cocaine(16). Acquisition depends upon the basolateral amygdala, nucleusaccumbens core, and serial interactions between these structures(59 – 61) as well as the orbital prefrontal cortex (PFC) (62). Cocaine-seeking responses are potentiated by cocaine (14) and decreasedby GABAergic inhibition of the ventral tegmental area dopamineneurons (63), an effect that is mediated by dopamine transmissionin the nucleus accumbens shell (64). A similar if not identical neuralsystem underlies conditioned reinforcement, because lesions orinactivation of the basolateral amygdala, orbital PFC, and nucleusaccumbens core all impair conditioned reinforcement (65– 67),whereas increasing dopamine transmission in the nucleus accum-bens shell potentiates the control over behavior by conditionedreinforcers (68). The NErgic influences on conditioned reinforce-ment are not well-understood. Cador et al. (69) showed that neithernucleus accumbens NErgic depletion nor intra-accumbens NE infu-sions affected the acquisition of responding with conditioned rein-forcement, but the conditions under which this lack of effect wereseen differ considerably from the cocaine and heroin seeking pro-cedures used here. Thus, although ATO reduced drug seeking andrelapse precisely under conditions when conditioned reinforce-ment processes were critically engaged, these effects might notdepend directly upon diminishing conditioned reinforcement pro-cesses per se.

However, in well-trained animals such as those in the presentstudies, the control over cocaine seeking has been shown to de-pend on dominant control by the dorsal striatum as the behavior

emerges as a stimulus-response habit (70). This dorsal striatal

www.sobp.org/journal

mlt(dAdWdSrpt

ctecsmtomatptHcrcoau

pmirmcriwcitdrtgmv

gitffl

272 BIOL PSYCHIATRY 2011;69:266–274 D. Economidou et al.

w

echanism is itself engaged by ventral striatal areas via their regu-ation of the dopaminergic innervation of the dorsal striatumhrough serial interactions with midbrain dopamine neurons71,72) as revealed by the effects of disconnection of the ventral andorsal striatum (31). Thus, another possible mechanism by whichTO might decrease drug seeking is by disrupting these dopamine-ependent interactions between ventral and dorsal striatum.ithin the ventral striatum, it is the shell region that has a high

ensity of the NET and receives a rich NErgic innervation (73–75).everal studies have demonstrated that NE modulates dopamineelease in the nucleus accumbens (76 –78) and in this way mightrofoundly diminish the ventral striatal recruitment of dorsal stria-

al processes underlying drug-seeking habits.Treatment with ATO has been also shown to facilitate inhibitory

ontrol in a variety of settings—for example in a stop-signal reac-ion time task in which MPH and a selective DAT inhibitor had noffect (21,79). This observation might be viewed in the context ofontemporary theories of drug addiction, which attribute the per-istence of drug-seeking habits to deficits in inhibitory control

echanisms (80 – 84) that are localized to PFC areas and might arisehrough the long-term effects of self-administered drugs (85– 89),r represent pre-existing vulnerability traits (5,90 –92). The NErgicodulation of the PFC has further been shown to enable behavioral

daptation and to facilitate flexibility of responding in a variety ofasks (93–96). Manipulations of PFC areas markedly influence theerformance of drug-seeking behavior (97) and also the propensity

o relapse (38,98,99) through interactions with striatal areas (38).ypoactivity of the PFC is also seen in individuals addicted to co-aine, heroin, and other drugs (100 –104). Thus, the effect of ATO toeduce both cocaine and heroin seeking as well as relapse to co-aine seeking might reflect the impact of increased NE transmissionn the inflexible and habitual pursuit of drugs that is occasionednd maintained by drug-associated CS, to which addicted individ-als are especially susceptible (82,105,106).

The results of these experiments show that, in addition to com-letely preventing relapse to cocaine seeking in high-impulsive ani-als (6), NE reuptake blockade by ATO results in a marked decrease

n both cocaine and heroin seeking as well as significantly reducingelapse to cocaine seeking after abstinence. These effects of ATO

ight reflect NErgic modulation of conditioned reinforcement pro-esses that are key in each of the models of drug seeking andelapse studied here. But they might also reflect modulation ofnhibitory control of drug-seeking habits that enable animals to

ithhold responding that is driven by drug-associated CS. The psy-hological mechanism of action of ATO might be elucidated by

nvestigation of its central sites of action, such as limbic corticostria-al circuitries that are modulated by dopamine and serially engageduring the performance of drug-seeking habits (92) and during

elapse (83,98,107), as well as inhibitory control functions localizedo PFC areas (94,108 –110). However, the present data strongly sug-est that ATO, a clinically approved and well-tolerated medication,ight have great utility in the promotion of abstinence and the pre-

ention of relapse in addicted individuals seeking treatment.

This work was funded by a UK Medical Research Council (MRC)rant (G9536855) and was completed within the Behavioral and Clin-

cal Neuroscience Institute, which is supported by a joint award fromhe MRC and the Wellcome Trust. The authors thank David Theobaldor technical assistance. The authors of this manuscript report no con-icts of interest.

Supplementary material cited in this article is available online.

ww.sobp.org/journal

1. O’Brien CP (2005): Anticraving medications for relapse prevention: Apossible new class of psychoactive medications. Am J Psychiatry 162:1423–1431.

2. Coffey SF, Gudleski GD, Saladin ME, Brady KT (2003): Impulsivity andrapid discounting of delayed hypothetical rewards in cocaine-depen-dent individuals. Exp Clin Psychopharmacol 11:18 –25.

3. Moeller FG, Dougherty DM, Barratt ES, Schmitz JM, Swann AC,Grabowski J (2001): The impact of impulsivity on cocaine use andretention in treatment. J Subst Abus Treat 21:193–198.

4. Doran N, Spring B, McChargue D (2007): Effect of impulsivity on cravingand behavioral reactivity to smoking cues. Psychopharmacology (Berl)194:279 –288.

5. Dalley JW, Fryer TD, Brichard L, Robinson ESJ, Theobald DEH, Lääne K,et al. (2007): Nucleus accumbens: D2/3 receptors predict trait impulsiv-ity and cocaine reinforcement. Science 315:1267–1270.

6. Economidou D, Pelloux Y, Robbins TW, Dalley JW, Everitt BJ (2009):High impulsivity predicts relapse to cocaine-seeking after punish-ment-induced abstinence. Biol Psychiatry 65:851– 856.

7. Slatkoff J, Greenfield B (2006): Pharmacological treatment of attention-deficit/hyperactivity disorder in adults. Expert Opin Investig Drugs 15:649 – 667.

8. Gibson AP, Bettinger TL, Patel NC, Crismon ML (2006): Atomoxetineversus stimulants for treatment of attention deficit/hyperactivity dis-order. Ann Pharmacother 40:1134 –1142.

9. Heal DJ, Cheetham SC, Smith SL (2009): The neuropharmacology ofADHD drugs in vivo: Insights on efficacy and safety. Neuropharmacol-ogy 57:608 – 618.

10. Childress AR, Hole AV, Ehrman RN, Robbins SJ, McLellan AT, O’Brien CP(1993): Cue reactivity and cue reactivity interventions in drug depen-dence. NIDA Res Monogr 137:73–95.

11. O’Brien CP, Childress AR, McLellan AT, Ehrman R (1992): Classical con-ditioning in drug-dependent humans. Ann N Y Acad Sci 654:400 – 415.

12. Garavan H, Pankiewicz J, Bloom A, Cho JK, Sperry L, Ross TJ, et al. (2000):Cue-induced cocaine craving: Neuroanatomical specificity for drugusers and drug stimuli. Am J Psychiatry 157:1789 –1798.

13. Goldberg SR, Kelleher RT, Morse WH (1975): Second-order schedules ofdrug injection. Fed Proc 34:1771–1776.

14. Arroyo M, Markou A, Robbins TW, Everitt BJ (1998): Acquisition, main-tenance and reinstatement of intravenous cocaine self-administrationunder a second-order schedule of reinforcement in rats: Effects ofconditioned cues and continuous access to cocaine. Psychopharma-cology (Berl) 140:331–344.

15. Alderson HL, Robbins TW, Everitt BJ (2000): Heroin self-administrationunder a second-order schedule of reinforcement: Acquisition andmaintenance of heroin-seeking behaviour in rats. Psychopharmacol-ogy (Berl) 153:120 –133.

16. Everitt BJ, Robbins TW (2000): Second-order schedules of drug rein-forcement in rats and monkeys: Measurement of reinforcing efficacyand drug-seeking behaviour. Psychopharmacology (Berl) 153:17–30.

17. Schindler CW, Panlilio LV, Goldberg SR (2002): Second-order schedulesof drug self-administration in animals. Psychopharmacology (Berl) 163:327–344.

18. Shaham Y, Shalev U, Lu L, De Wit H, Stewart J (2003): The reinstatementmodel of drug relapse: History, methodology and major findings. Psy-chopharmacology (Berl) 168:3–20.

19. Lu L, Grimm JW, Hope BT, Shaham Y (2004): Incubation of cocainecraving after withdrawal: A review of preclinical data. Neuropharma-cology 47(suppl 1):214 –226.

20. van Gaalen MM, van Koten R, Schoffelmeer ANM, Vanderschuren LJMJ(2006): Critical involvement of dopaminergic neurotransmission in im-pulsive decision making. Biol Psychiatry 60:66 –73.

21. Eagle DM, Tufft MRA, Goodchild HL, Robbins TW (2007): Differentialeffects of modafinil and methylphenidate on stop-signal reaction timetask performance in the rat, and interactions with the dopamine recep-tor antagonist cis-flupenthixol. Psychopharmacology (Berl) 192:193–206.

22. Puumala T, Ruotsalainen S, Jäkälä P, Koivisto E, Riekkinen P, Sirviö J(1996): Behavioral and pharmacological studies on the validation of anew animal model for attention deficit hyperactivity disorder. Neuro-biol Learn Mem 66:198 –211.

23. Bizarro L, Patel S, Murtagh C, Stolerman IP (2004): Differential effects of

psychomotor stimulants on attentional performance in rats: Nicotine,

D. Economidou et al. BIOL PSYCHIATRY 2011;69:266–274 273

amphetamine, caffeine and methylphenidate. Behav Pharmacol15:195–206.

24. Bymaster FP, Katner JS, Nelson DL, Hemrick-Luecke SK, Threlkeld PG,Heiligenstein JH, et al. (2002): Atomoxetine increases extracellular lev-els of norepinephrine and dopamine in prefrontal cortex of rat: Apotential mechanism for efficacy in attention deficit/hyperactivity dis-order. Neuropsychopharmacology 27:699 –711.

25. Barkley RA (1977): A review of stimulant drug research with hyperac-tive children. J Child Psychol Psychiatry 18:137–165.

26. Gatley SJ, Pan D, Chen R, Chaturvedi G, Ding YS (1996): Affinities ofmethylphenidate derivatives for dopamine, norepinephrine and sero-tonin transporters. Life Sci 58:231–239.

27. Gehlert DR, Schober DA, Hemrick-Luecke SK, Krushinski J, Howbert JJ,Robertson DW, et al. (1995): Novel halogenated analogs of tomoxetinethat are potent and selective inhibitors of norepinephrine uptake inbrain. Neurochem Int 26:47–52.

28. Tatsumi M, Groshan K, Blakely RD, Richelson E (1997): Pharmacologicalprofile of antidepressants and related compounds at human mono-amine transporters. Eur J Pharmacol 340:249 –258.

29. Wong DT, Threlkeld PG, Best KL, Bymaster FP (1982): A new inhibitor ofnorepinephrine uptake devoid of affinity for receptors in rat brain.J Pharmacol Exp Ther 222:61– 65.

30. Robinson ESJ, Eagle DM, Mar AC, Bari A, Banerjee G, Jiang X, et al.(2008): Similar effects of the selective noradrenaline reuptake inhibitoratomoxetine on three distinct forms of impulsivity in the rat. Neuropsy-chopharmacology 33:1028 –1037.

31. Belin D, Everitt BJ (2008): Cocaine seeking habits depend upon dopam-ine-dependent serial connectivity linking the ventral with the dorsalstriatum. Neuron 57:432– 441.

32. Brown ZJ, Tribe E, D’Souza NA, Erb S (2009): Interaction between nor-adrenaline and corticotrophin-releasing factor in the reinstatement ofcocaine seeking in the rat. Psychopharmacology (Berl) 203:121–130.

33. Shaham Y, Highfield D, Delfs J, Leung S, Stewart J (2000): Clonidineblocks stress-induced reinstatement of heroin seeking in rats: An effectindependent of locus coeruleus noradrenergic neurons. Eur J Neurosci12:292–302.

34. Mantsch JR, Weyer A, Vranjkovic O, Beyer CE, Baker DA, Caretta H(2010): Involvement of noradrenergic neurotransmission in the stress-but not cocaine-induced reinstatement of extinguished cocaine-in-duced conditioned place preference in mice: role for beta-2 adrenergicreceptors. Neuropsychopharmacology 35:2165–78.

35. Platt DM, Rowlett JK, Spealman RD (2007): Noradrenergic mechanismsin cocaine-induced reinstatement of drug seeking in squirrel monkeys.J Pharmacol Exp Ther 322:894 –902.

36. See RE (2009): Dopamine D1 receptor antagonism in the prelimbiccortex blocks the reinstatement of heroin-seeking in an animal modelof relapse. Int J Neuropsychopharmacol 12:431– 436.

37. Ciccocioppo R, Sanna PP, Weiss F (2001): Cocaine-predictive stimulusinduces drug-seeking behavior and neural activation in limbic brainregions after multiple months of abstinence: Reversal by D(1) antago-nists. Proc Natl Acad Sci U S A 98:1976 –1981.

38. Peters J, LaLumiere RT, Kalivas PW (2008): Infralimbic prefrontal cortexis responsible for inhibiting cocaine seeking in extinguished rats.J Neurosci 28:6046 – 6053.

39. Tiffany ST, Conklin CA (2002): The promise and pitfalls of animal andhuman models of relapse: Comment on Leri and Stewart. Exp ClinPsychopharmacol 10:361–363, discussion 364 –366.

40. O’Brien CP, Gardner EL (2005): Critical assessment of how to studyaddiction and its treatment: Human and non-human animal models.Pharmacol Ther 108:18 –58.

41. Baptista MAS, Martin-Fardon R, Weiss F (2004): Preferential effects ofthe metabotropic glutamate 2/3 receptor agonist LY379268 on condi-tioned reinstatement versus primary reinforcement: Comparison be-tween cocaine and a potent conventional reinforcer. J Neurosci 24:4723– 4727.

42. Lu L, Uejima JL, Gray SM, Bossert JM, Shaham Y (2007): Systemic andcentral amygdala injections of the mGluR(2/3) agonist LY379268 at-tenuate the expression of incubation of cocaine craving. Biol Psychiatry61:591–598.

43. Bossert JM, Busch RF, Gray SM (2005): The novel mGluR2/3 agonistLY379268 attenuates cue-induced reinstatement of heroin seeking.

Neuroreport 16:1013–1016.

44. Di Ciano P, Everitt BJ (2003): The GABA(B) receptor agonist baclofenattenuates cocaine- and heroin-seeking behavior by rats. Neuropsy-chopharmacology 28:510 –518.

45. Markowitz JS, DeVane CL, Pestreich LK, Patrick KS, Muniz R (2006): Acomprehensive in vitro screening of d-, l-, and dl-threo-methylpheni-date: An exploratory study. J Child Adolesc Psychopharmacol 16:687–698.

46. Han DD, Gu HH (2006): Comparison of the monoamine transportersfrom human and mouse in their sensitivities to psychostimulant drugs.BMC Pharmacol 6:6.

47. Béïque JC, Lavoie N, de Montigny C, Debonnel G (1998): Affinities ofvenlafaxine and various reuptake inhibitors for the serotonin and nor-epinephrine transporters. Eur J Pharmacol 349:129 –132.

48. Gawin FH, Byck R, Kleber HD (1986): Desipramine augmentation ofcocaine abstinence: Initial results. Clin Neuropharmacol 9(suppl4):202–204.

49. Gawin FH, Kleber HD, Byck R, Rounsaville BJ, Kosten TR, Jatlow PI, et al.(1989): Desipramine facilitation of initial cocaine abstinence. Arch GenPsychiatry 46:117–121.

50. Kolar AF, Brown BS, Weddington WW, Haertzen CC, Michaelson BS,Jaffe JH (1992): Treatment of cocaine dependence in methadonemaintenance clients: A pilot study comparing the efficacy of desipra-mine and amantadine. Int J Addict 27:849 – 868.

51. Arndt IO, Dorozynsky L, Woody GE, McLellan AT, O’Brien CP (1992):Desipramine treatment of cocaine dependence in methadone-main-tained patients. Arch Gen Psychiatry 49:888 – 893.

52. Spencer T, Biederman J, Heiligenstein J, Wilens T, Faries D, Prince J, et al.(2001): An open-label, dose-ranging study of atomoxetine in childrenwith attention deficit hyperactivity disorder. J Child Adolesc Psycho-pharmacol 11:251–265.

53. Quintana H, Cherlin EA, Duesenberg DA, Bangs ME, Ramsey JL, FeldmanPD, et al. (2007): Transition from methylphenidate or amphetamine toatomoxetine in children and adolescents with attention-deficit/hyperac-tivity disorder—a preliminary tolerability and efficacy study. Clin Ther 29:1168–1177.

54. Jasinski DR, Faries DE, Moore RJ, Schuh LM, Allen AJ (2008): Abuseliability assessment of atomoxetine in a drug-abusing population.Drug Alcohol Depend 95:140 –146.

55. Stoops WW, Blackburn JW, Hudson DA, Hays LR, Rush CR (2008): Safety,tolerability and subject-rated effects of acute intranasal cocaine ad-ministration during atomoxetine maintenance. Drug Alcohol Depend92:282–285.

56. Sofuoglu M, Poling J, Hill K, Kosten T (2009): Atomoxetine attenuatesdextroamphetamine effects in humans. Am J Drug Alcohol Abus 35:412– 416.

57. Sofuoglu M, Sewell RA (2009): Norepinephrine and stimulant addic-tion. Addict Biol 14:119 –129.

58. Wilens TE, Adler LA, Weiss MD, Michelson D, Ramsey JL, Moore RJ, et al.(2008): Atomoxetine treatment of adults with ADHD and comorbidalcohol use disorders. Drug Alcohol Depend 96:145–154.

59. Whitelaw RB, Markou A, Robbins TW, Everitt BJ (1996): Excitotoxiclesions of the basolateral amygdala impair the acquisition of cocaine-seeking behaviour under a second-order schedule of reinforcement.Psychopharmacology (Berl) 127:213–224.

60. Ito R, Robbins TW, Everitt BJ (2004): Differential control over cocaine-seeking behavior by nucleus accumbens core and shell. Nat Neurosci7:389 –397.

61. Di Ciano P, Everitt BJ (2001): Dissociable effects of antagonism of NMDAand AMPA/KA receptors in the nucleus accumbens core and shell oncocaine-seeking behavior. Neuropsychopharmacology 25:341–360.

62. Hutcheson DM, Everitt BJ (2003): The effects of selective orbitofrontalcortex lesions on the acquisition and performance of cue-controlledcocaine seeking in rats. Ann NY Acad Sci 1003:410 – 411.

63. Di Ciano P, Everitt BJ (2004): Contribution of the ventral tegmental areato cocaine-seeking maintained by a drug-paired conditioned stimulusin rats. Eur J Neurosci 19:1661–1667.

64. Ito R, Dalley JW, Howes SR, Robbins TW, Everitt BJ (2000): Dissociationin conditioned dopamine release in the nucleus accumbens core andshell in response to cocaine cues and during cocaine-seeking behaviorin rats. J Neurosci 20:7489 –7495.

65. Parkinson JA, Olmstead MC, Burns LH, Robbins TW, Everitt BJ (1999):Dissociation in effects of lesions of the nucleus accumbens core and

shell on appetitive Pavlovian approach behavior and the potentia-

www.sobp.org/journal

274 BIOL PSYCHIATRY 2011;69:266–274 D. Economidou et al.

w

tion of conditioned reinforcement and locomotor activity byD-amphetamine. J Neurosci 19:2401–2411.

66. Cador M, Robbins TW, Everitt BJ (1989): Involvement of the amygdalain stimulus-reward associations: Interaction with the ventral striatum.Neuroscience 30:77– 86.

67. Pears A, Parkinson JA, Hopewell L, Everitt BJ, Roberts AC (2003): Lesionsof the orbitofrontal but not medial prefrontal cortex disrupt condi-tioned reinforcement in primates. J Neurosci 23:11189 –11201.

68. Taylor JR, Robbins TW (1984): Enhanced behavioural control by condi-tioned reinforcers following microinjections of d-amphetamine intothe nucleus accumbens. Psychopharmacology (Berl) 84:405– 412.

69. Cador M, Taylor JR, Robbins TW (1991): Potentiation of the effects ofreward-related stimuli by dopaminergic-dependent mechanisms inthe nucleus accumbens. Psychopharmacology (Berl) 104:377–385.

70. Vanderschuren LJMJ, Di Ciano P, Everitt BJ (2005): Involvement of thedorsal striatum in cue-controlled cocaine seeking. J Neurosci 25:8665–8670.

71. Haber SN, Fudge JL, McFarland NR (2000): Striatonigrostriatal path-ways in primates form an ascending spiral from the shell to the dorso-lateral striatum. J Neurosci 20:2369 –2382.

72. Ikemoto S (2007): Dopamine reward circuitry: two projection systemsfrom the ventral midbrain to the nucleus accumbens-olfactory tuber-cle complex. Brain Res Rev 56:27–78.

73. Berridge CW, Stratford TL, Foote SL, Kelley AE (1997): Distribution ofdopamine beta-hydroxylase-like immunoreactive fibers within theshell subregion of the nucleus accumbens. Synapse 27:230 –241.

74. Delfs JM, Zhu Y, Druhan JP, Aston-Jones GS (1998): Origin of noradrener-gic afferents to the shell subregion of the nucleus accumbens: Antero-grade and retrograde tract-tracing studies in the rat. Brain Res 806:127–140.

75. McKittrick CR, Abercrombie ED (2007): Catecholamine mapping withinnucleus accumbens: Differences in basal and amphetamine-stimulatedefflux of norepinephrine and dopamine in shell and core. J Neurochem100:1247–1256.

76. Guiard BP, El Mansari M, Blier P (2008): Cross-talk between dopaminer-gic and noradrenergic systems in the rat ventral tegmental area, locusceruleus, and dorsal hippocampus. Mol Pharmacol 74:1463–1475.

77. Mizoguchi N, Saigusa T, Aono Y, Sekino R, Takada K, Oi Y, et al. (2008):The reboxetine-induced increase of accumbal dopamine efflux is in-hibited by l-propranolol: A microdialysis study with freely moving rats.Eur J Pharmacol 601:94 –98.

78. Nurse B, Russell VA, Taljaard JJ (1984): Alpha 2 and beta-adrenoceptoragonists modulate [3H]dopamine release from rat nucleus accumbensslices: Implications for research into depression. Neurochem Res9:1231–1238.

79. Bari A, Eagle DM, Mar AC, Robinson ESJ, Robbins TW (2009): Dissociableeffects of noradrenaline, dopamine, and serotonin uptake blockade onstop task performance in rats. Psychopharmacology (Berl) 205:273–283.

80. Kalivas PW, Volkow ND (2005): The neural basis of addiction: A pathol-ogy of motivation and choice. Am J Psychiatry 162:1403–1413.

81. Everitt BJ, Robbins TW (2005): Neural systems of reinforcement fordrug addiction: From actions to habits to compulsion. Nat Neurosci8:1481–1489.

82. Robbins TW, Everitt BJ (1999): Drug addiction: Bad habits add up.Nature 398:567–570.

83. Kalivas PW (2008): Addiction as a pathology in prefrontal cortical reg-ulation of corticostriatal habit circuitry. Neurotox Res 14:185–189.

84. Volkow ND, Li T (2004): Drug addiction: The neurobiology of behaviourgone awry. Nat Rev Neurosci 5:963–970.

85. Volkow ND, Fowler JS, Wang G, Swanson JM, Telang F (2007): Dopa-mine in drug abuse and addiction: Results of imaging studies andtreatment implications. Arch Neurol 64:1575–1579.

86. Porrino LJ, Smith HR, Nader MA, Beveridge TJR (2007): The effects ofcocaine: A shifting target over the course of addiction. Prog Neuropsy-chopharmacol Biol Psychiatry 31:1593–1600.

87. Beveridge TJ, Gill KE, Hanlon CA, Porrino LJ (2008): Review. Parallelstudies of cocaine-related neural and cognitive impairment in humansand monkeys. Philos Trans R Soc Lond B Biol Sci 363:3257–3266.

88. Hyman SE, Malenka RC, Nestler EJ (2006): Neural mechanisms of addic-

tion: The role of reward-related learning and memory. Annu Rev Neu-rosci 29:565–598.

ww.sobp.org/journal

89. Calu DJ, Stalnaker TA, Franz TM, Singh T, Shaham Y, Schoenbaum G(2007): Withdrawal from cocaine self-administration produces long-lasting deficits in orbitofrontal-dependent reversal learning in rats.Learn Mem 14:325–328.

90. Bechara A (2005): Decision making, impulse control and loss of will-power to resist drugs: A neurocognitive perspective. Nat Neurosci8:1458 –1463.

91. Belin D, Mar AC, Dalley JW, Robbins TW, Everitt BJ (2008): High impul-sivity predicts the switch to compulsive cocaine-taking. Science 320:1352–1355.

92. Everitt BJ, Belin D, Economidou D, Pelloux Y, Dalley JW, Robbins TW(2008): Review. Neural mechanisms underlying the vulnerability todevelop compulsive drug-seeking habits and addiction. Philos Trans RSoc Lond B Biol Sci 363:3125–3135.

93. Aston-Jones G, Cohen JD (2005): An integrative theory of locus coer-uleus-norepinephrine function: Adaptive gain and optimal perfor-mance. Annu Rev Neurosci 28:403– 450.

94. Ramos BP, Arnsten AFT (2007): Adrenergic pharmacology and cogni-tion: Focus on the prefrontal cortex. Pharmacol Ther 113:523–536.

95. McGaughy J, Ross RS, Eichenbaum H (2008): Noradrenergic, but notcholinergic, deafferentation of prefrontal cortex impairs attentionalset-shifting. Neuroscience 153:63–71.

96. Newman LA, Darling J, McGaughy J (2008): Atomoxetine reverses at-tentional deficits produced by noradrenergic deafferentation of me-dial prefrontal cortex. Psychopharmacology (Berl) 200:39 –50.

97. Weissenborn R, Robbins TW, Everitt BJ (1997): Effects of medial pre-frontal or anterior cingulate cortex lesions on responding for cocaineunder fixed-ratio and second-order schedules of reinforcement in rats.Psychopharmacology (Berl) 134:242–257.

98. Koya E, Uejima JL, Wihbey KA, Bossert JM, Hope BT, Shaham Y (2009):Role of ventral medial prefrontal cortex in incubation of cocaine crav-ing. Neuropharmacology 56(suppl 1):177–185.

99. McLaughlin J, See RE (2003): Selective inactivation of the dorsomedialprefrontal cortex and the basolateral amygdala attenuates condi-tioned-cued reinstatement of extinguished cocaine-seeking behaviorin rats. Psychopharmacology (Berl) 168:57– 65.

100. Goldstein RZ, Volkow ND (2002): Drug addiction and its underlyingneurobiological basis: Neuroimaging evidence for the involvement ofthe frontal cortex. Am J Psychiatry 159:1642–1652.

101. Volkow ND, Fowler JS, Wang GJ, Hitzemann R, Logan J, Schlyer DJ, et al.(1993): Decreased dopamine D2 receptor availability is associated withreduced frontal metabolism in cocaine abusers. Synapse 14:169 –177.

102. Hester R, Garavan H (2004): Executive dysfunction in cocaine addic-tion: Evidence for discordant frontal, cingulate, and cerebellar activity.J Neurosci 24:11017–11022.

103. Liu H, Li L, Hao Y, Cao D, Xu L, Rohrbaugh R, et al. (2008): Disruptedwhite matter integrity in heroin dependence: A controlled study utiliz-ing diffusion tensor imaging. Am J Drug Alcohol Abus 34:562–575.

104. Liu H, Hao Y, Kaneko Y, Ouyang X, Zhang Y, Xu L, et al. (2009): Frontaland cingulate gray matter volume reduction in heroin dependence:Optimized voxel-based morphometry. Psychiatry Clin Neurosci 63:563–568.

105. Everitt BJ, Dickinson A, Robbins TW (2001): The neuropsychologicalbasis of addictive behaviour. Brain Res Brain Res Rev 36:129 –138.

106. Volkow ND, Wang G, Telang F, Fowler JS, Logan J, Childress A, et al.(2006): Cocaine cues and dopamine in dorsal striatum: Mechanism ofcraving in cocaine addiction. J Neurosci 26:6583– 6588.

107. See RE, Elliott JC, Feltenstein MW (2007): The role of dorsal vs ventralstriatal pathways in cocaine-seeking behavior after prolonged absti-nence in rats. Psychopharmacology (Berl) 194:321–331.

108. Dalley JW, Mar AC, Economidou D, Robbins TW (2008): Neurobehav-ioral mechanisms of impulsivity: Fronto-striatal systems and func-tional neurochemistry. Pharmacol Biochem Behav 90:250 –260.

109. Ragozzino ME (2007): The contribution of the medial prefrontal cortex,orbitofrontal cortex, and dorsomedial striatum to behavioral flexibil-ity. Ann N Y Acad Sci 1121:355–375.

110. Aron AR, Robbins TW, Poldrack RA (2004): Inhibition and the right

inferior frontal cortex. Trends Cogn Sci 8:170 –177.