A

sofptniacea©

K

C

1

cCr

NC

1d

Pharmacological Research 56 (2007) 418–427

Review

Endocannabinoid regulation of relapse mechanisms

Liana Fattore a,b,∗, Paola Fadda b,c, Walter Fratta a,b,c

a Institute of Neuroscience, CNR, National Research Council, Italyb Centre of Excellence “Neurobiology of Addiction”, Italy

c Department of Neuroscience, University of Cagliari, Italy

Accepted 5 September 2007

bstract

Addiction involves a complex neuropharmacologic behavioural cycle, in which positive reinforcement exerted by the drug and the negativetate of withdrawal drive the user to extremes to obtain the drug. Comprehensive studies have established that relapse is the most commonutcome of recovery programs treating addictive behaviours. Several types of anticraving medication are available nowadays, such as naltrexoneor the treatment of alcoholism, bupropion for nicotine, methadone or buprenorphine for heroin. This review focuses on recent behavioural dataroviding a rationale for an endocannabinoid mechanism underlying reinstatement of compulsive drug seeking. Studies supporting the contentionhat reinstatement of extinguished drug self-administration behaviour may be generated by cannabinoid CB1 receptor agonists and attenuated, ifot blocked, by CB1 receptor antagonists, are here reviewed. In support to these findings, conditioned place preference studies substantiate thenvolvement of the endocannabinoid system in recidivism mechanisms by demonstrating that motivation to relapse can be triggered by CB1 receptorctivation while blockade of such receptors may prevent reinstatement of place conditioning induced by either drug primings or drug-associated

ues. Finally, biochemical studies evaluating changes in endocannabinoid levels, CB1 receptor density and CB1 mRNA expression during re-xposure to drug following extinction are also examined. Taken together, the evidence available has important implications in the understandingnd treatment of relapsing episodes in patients undergoing detoxification. 2007 Elsevier Ltd. All rights reserved.

eywords: CB1 receptor; Reinstatement; Self-administration; Conditioned place preference; Addiction

ontents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4182. Self-administration studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4193. Conditioned place preference studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4214. Changes in endocannabinoid contents, CB1 receptor density and function, and mRNA levels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4215. Discussion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 423

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 424

. Introduction behaviours, such as gambling [1,2], binge eating [3,4], and drugaddiction [5], support the conceptualization of addiction as a

Drug addiction is defined in behavioural terms as craving,ompulsive drug use and relapse or recurrent use of substances.onsistently high relapse rates (75–90%) which have been

eported throughout the literature for a variety of compulsive

∗ Corresponding author at: CNR Institute of Neuroscience c/o Department ofeuroscience, Cittadella Universitaria, 09042 Monserrato (CA), University ofagliari, Italy. Tel.: +39 070 6754327; fax: +39 070 6754312.

E-mail address: lfattore@in.cnr.it (L. Fattore).

flporsurn

043-6618/$ – see front matter © 2007 Elsevier Ltd. All rights reserved.oi:10.1016/j.phrs.2007.09.004

amily of social problem behaviours which share essential simi-arities. To date, pharmacologic approaches have been less thanromising in preventing relapse, with only a few exceptionsf patients with long histories of heroin use and subsequentehabilitation on a maintenance program or highly motivated

mokers. Indeed, although smoking cessation rates have contin-ed to increase, the vast majority of smokers who quit eventuallyelapse. Typically, available medications may be corrective butot curative for severely addicted persons, as the majority expe-

ical R

rt

ewrimhlffq[ttieaertefaiad

eaporie[agminn

ndpucsvmptabc

oaccnbrd

2

wivamt[caaa[s

tneaat0is5CHLtpripimrisdsm

L. Fattore et al. / Pharmacolog

ience a return of symptoms after withdrawal of maintenancereatment.

An established method applied in study of the reinforcingffects of a drug entails the use of experimental animal models inhich drug acquisition is contingent upon a specific behavioural

esponse: for example, an animal will soon learn that a drugnjection is given every time it presses a particular lever [6]. To

odel human relapse in laboratory animals, similar protocolsave been modified in “reinstatement” procedures, where estab-ishment of responding is maintained by drug reinforcer andollowed by its extinction. Once the behaviour has decreased inrequency, experimental manipulations are imposed and the fre-uency of the previously reinforced behaviour is then reassessed7]. Advancing knowledge in the field of neuroscience have iden-ified conditioned cueing, drug priming and stress exposure ashe three main types of stimuli capable of eliciting relapse, point-ng to the learned responses to drug-related stimuli as criticallements contributing towards the chronic relapsing nature ofddiction [8]. The long-lasting value of contextual stimuli inliciting drug seeking behaviour in animal models of relapseesembles the survival of conditioned cue reactivity in humans,hus confirming the pivotal part played by learning factors in thenduring addictive potential of drugs. Vulnerability to relapseollowing withdrawal is proposed to be the result of neuroad-ptive processes within the brain which lead to impairmentn mechanisms mediating hedonic perception of the drug andppearance of affective changes such as dysphoria, anxiety andepression during withdrawal.

Considerable evidence exists implicating perturbations inndocannabinoid signalling as potential substrates for theseffective states. Cannabinoid receptors belong to a family of G-rotein-coupled receptors and are densely distributed in areasf the brain related to motor control, cognition, emotionalesponses, motivated behaviour and homeostasis [9]. Accord-ngly, the endocannabinoid system can be modulated to enhancemotional learning, such as extinction of conditioned fear10,11], or to ameliorate anxiety and depression [12,13]. Bycting as retrograde signalling messengers in GABAergic andlutamatergic synapses, endocannabinoids are likely to exert aodulatory control of post-synaptic transmission, thus interact-

ng with the majority of brain neurotransmitters [14]. Cannabi-oid receptors are also critical modulators of the autonomicervous system, the immune system and microcirculation.

Recently, pre-clinical studies have shown how endocannabi-oid signalling is crucially implicated in the re-initiation ofrug seeking and taking behaviours following even a prolongederiod of drug abstinence. Here, we review data from studiessing pre-clinical models in rats and mice on the effects ofannabinoid system manipulation on the re-occurrence of drugeeking episodes following extinction, as measured by the intra-enous drug self-administration protocol, the most widely usedodel in the study of addictive behaviour, and conditioned place

reference procedure, a behavioural model of incentive motiva-

ion. These two methodologies have been used traditionally tossess the positive reinforcing properties of a drug, but have alsoeen applied in examining brain reward mechanisms, includingraving and relapse.

usi

esearch 56 (2007) 418–427 419

While summarizing behavioural studies evaluating the effectsf selective CB1 receptor agonists, CB1 antagonists and fattycid amide hydrolase inhibitors on extinguished responding oronditioning, we will also examine recent works investigatingellular and molecular mechanisms by which the endocannabi-oid system may impinge on relapsing neural circuitry. Theroader implications of the data reviewed here for futureesearch and for the maintenance of addicted patients underetoxification are briefly discussed.

. Self-administration studies

Drug self-administration (SA) paradigms have long beenidely used as a tool in exploring behavioural and neuroanatom-

cal features of addiction, providing essential information onulnerability factors and brain circuits involved in drug abusend dependence [15]. Both acute SA paradigm in drug-naıveice and chronic SA in trained rats have been used in the past

o disclose the rewarding properties of synthetic cannabinoids16–18], as subsequently confirmed for THC and endo-annabinoids in trained squirrel monkeys [19,20]. Importantly,daptations of these paradigms to “extinction/reinstatement”nimal models of relapse have allowed investigation of the mech-nisms underlying resumption of drug use in abstinent subjects8], thereby increasing our knowledge of the neuroanatomicalubstrates involved in relapse [21–23].

Over the past decade, data collected in our and other labora-ories have demonstrated that stimulation of the CB1 receptorsot only affects relapse to cannabinoids [24], but also influ-nces relapse to other drugs of abuse (i.e. heroin), thus exertingmore general control on drug seeking behaviour [25]. Indeed,n acute intraperitoneal (i.p.) or subcutaneous (s.c.) administra-ion of a CB1 receptor agonist, such as WIN 55,212-2 (0.25 and.5 mg kg−1), triggers relapse to cannabinoid seeking behaviourn long-abstinent Long Evans rats with a previous history oftable SA [24]. Likewise, CB1 receptor stimulation by WIN5,212-2 (0.15 and 0.3 mg kg−1 i.p.) as well as by two otherB1 receptor agonists CP 55,940 (0.05 and 0.1 mg kg−1 i.p.) andU 250 (20 �g kg−1 s.c.) elicits relapse to heroin in abstinentister Hooded or Wistar rats following 3 weeks of extinc-

ion [26,27]. Importantly, cannabinoid-induced reinstatement isrevented by pre-treatment with the CB1 receptor antagonistimonabant, at doses (0.3–3.0 mg kg−1) not affecting respond-ng per se, thus demonstrating to be CB1 receptor-mediatedrocesses [24,26,28]. These findings extended previous stud-es demonstrating how endocannabinoid transmission crucially

odulates the reinforcing properties of opioids (reviewed inefs. [25,29–31]). However, besides its well established rolen relapse to cannabinoids and opioids, the endocannabinoidystem has been reported to be implicated in craving for andrug-seeking resumption of many other widely used addictiveubstances, both legal (alcohol, nicotine) and illegal (cocaine,ethamphetamine).

With respect to alcohol abuse, when drug intake was eval-

ated by means of a two-bottle free-choice paradigm afterubsequent period of alcoholization and alcohol deprivation,t was found that rimonabant (0.3–3.0 mg kg−1 i.p.) prevented

4 colog

tvrdipta

iioaenleawrnhNevchlistotriopwHas

hdtnbnruvbt(ttf

ohuibseantiesarp

rlil(Sa(icte[ai2dsti5r

ordaabldrs

n

20 L. Fattore et al. / Pharma

he acquisition of alcohol drinking behaviour and decreasedoluntary ethanol intake in Sardinian alcohol-preferring (sP)ats genetically selected for alcohol preference [32,33]. It alsoecreases voluntary consumption of ethanol along with sucrosentake [34,35] and water consumption without altering alcoholreference [36], pointing to an endocannabinoid mechanism inhe control of operant behaviour motivated by both drug (i.e.lcohol) and natural rewards (i.e. sucrose).

More recently, it has been found that alcohol–cannabinoidnteractions extended to relapse processes, as THC (1.0 mg kg−1

.p.) was shown to significantly reinstate in trained animalsperant responding previously reinforced with alcohol, butlso responding reinforced by beer or sucrose [37], providingvidence of a relatively non-specific effect of the cannabi-oid priming. Moreover, CB1 receptor stimulation produces aong-lasting increase of the alcohol deprivation effect, as ratsxposed to WIN 55,212-2 during alcohol deprivation showedpersistent increase in responding for alcohol lasting over 2eeks. In contrast, control animals only displayed a tempo-

ary (1–2 days) increase in alcohol consumption, suggesting thaton-contingent chronic exposure to cannabinoids during alco-ol deprivation may potentiate relapse into alcohol use [38].otably, unanimous consensus has been reached by studies

valuating the effect of CB1 receptor blockade on alcohol depri-ation effect, i.e. the temporary increase in voluntary alcoholonsumption occurring following a period of abstinence, whichas been proposed to resemble increased alcohol intake withoss of control over drinking frequently associated with relapsen human alcoholics [39]. In fact, a complete abolishment ofuch an effect in sP rats has been described following pre-reatment with either 0.3–3.0 mg kg−1 i.p. of rimonabant [40]r 0.3–3.0 mg kg−1 i.p. of the newly synthesized CB1 recep-or antagonist, SR147778 [41]. In line with these observations,imonabant (0.3–3.0 mg kg−1 i.p.) has proved to be effectiven reducing reinstatement of alcohol seeking triggered by anrange extract odour previously associated with the contingentresentation of the drug, i.e. a drug-associated cue [35,42],hile not affecting foot-shock stress-induced reinstatement [43].owever, the anandamide transport inhibitor AM404 (0.4, 2.0

nd 10.0 mg kg−1 i.p.) fails to modify reinstatement of alcoholeeking triggered by a drug-associated cue [44].

Behavioural interactions between cannabinoids and nicotineave also been proposed by several independent investigations,emonstrating that rimonabant (0.3–3.0 mg kg−1 i.p.) is ableo block nicotine-induced dopamine release in the shell of theucleus accumbens, reduce nicotine intake in an SA procedureut not substitute for nicotine or antagonize the nicotine cue in aicotine discrimination procedure [45]. In turn, the �7 nicotiniceceptor antagonist methyllycaconitine reduced cannabinoid SAnder both FR1 and FR5 schedules of reinforcement [46], pro-iding further support to the existence of a functional cross-talketween the CB1 and nicotinic receptors in modulating addic-ive behaviour. On the other hand, mice lacking the CB1 receptor

CB1 knockout) do self-administer nicotine at a level similar tohat of their corresponding wild type littermate [47], suggestinghat integrity of CB1 receptors is not an essential prerequisiteor the expression of nicotine reinforcing effects.

retA

ical Research 56 (2007) 418–427

In the past years, relapse studies focussed mainly on heroinr cocaine seeking reinstatement and, to a lesser extent, to alco-ol seeking, thus leaving many issues related to nicotine relapsenanswered. Only a few years ago, rimonabant (1.0 mg kg−1

.p.) was shown to attenuate the reinstatement of nicotine seekingehaviour evoked by non-contingent exposure to a drug-pairedtimulus several weeks after responding for nicotine had beenxtinguished [48], thus proving to be of help in maintainingbstinence from nicotine by diminishing the reinforcing value oficotine-related stimulus. Accordingly, a second study showedhat blockade of CB1 receptors by rimonabant (1.0–3.0 mg kg−1

.p.) attenuated responding for stimuli previously associated withither nicotine or sucrose delivery in a similar fashion [49],trengthening the hypothesis that endocannabinoids may playmore general role in modulating cue reactivity or conditioned

einforcement of both drug and natural reinforcers even afterrolonged abstinence.

Establishment of a role of the endocannabinoid system inelapse to cocaine has been reached only recently, as ear-ier studies using a within-session protocol seemed to excludets participation in the reinitiating of cocaine seeking fol-owing extinction. Indeed, although a priming with THC0.3–3.0 mg kg−1 i.p.) does not trigger relapse to cocaine inprague–Dawley rats 3 h from the last access to the drug [50],n acute priming with the more potent CB1 agonist HU2100.02–0.1 mg kg−1 s.c.) elicits reinstatement of cocaine-seekingn Wistar rats after 2 weeks of extinction [51]. Accordingly,ocaine-primed relapse in cocaine-trained rats has been reportedo be inhibited by pre-treatment with rimonabant, administeredither at 0.03–3.0 mg kg−1 s.c. [51] or at 5.0–10 mg kg−1 i.p.52], as well as by the novel highly selective CB1 receptorntagonist AM251 at doses ranging from 1.0 to 10 mg kg−1

.p. [53]. Intriguingly, subchronic treatment with WIN 55,212-(0.3–3.0 mg kg−1 i.p.) following interruption of cocaine SA

ose-dependently attenuated cue-induced cocaine seeking rein-tatement [54], supporting a cannabinoid mechanism in relapseo cocaine. However, priming injection of cocaine (10 mg kg−1

.p.) was not able to reinstate responding for cannabinoid in WIN5,212-2-trained rats, thus excluding a cocaine mechanism in theesumption of cannabinoid seeking [24].

Finally, relatively little is known at present about the effectf CB1 receptor stimulation or blockade on the propensity toelapse to methamphetamine (METH) following cessation ofrug intake. Several years ago, a pioneering SA study revealedfunctional interaction between the endocannabinoid system

nd the reinforcing properties of METH by demonstrating thatoth anandamide (0.1–2.0 mg kg−1 i.p.) and its synthetic ana-ogue R-(+)-methanandamide (1.0–5.0 mg kg−1 i.p.) tended toose-dependently increase METH self-administration in trainedats, while the CB1 antagonist AM251 (1.0–5.0 mg kg−1 i.p.)ignificantly decreased responding [55].

More recently, a second study showed how the endocannabi-oid system may also modulate relapse to METH seeking,

imonabant (3.2 mg kg−1 i.p.) being able to block the reinstatingffects of either an acute METH priming or a single re-exposureo a stimulus previously paired with drug infusion delivery [56].s in this study the cyclooxigenase inhibitor diclofenac was

ical R

aMdp[ai

omouohsici

3

iatletwoo(aiiemipifBCtwcldpput

civf

C3ndiccC[tf

etthorbtea3tctptcCTwasi

4d

isdrrnb[eCsbA

L. Fattore et al. / Pharmacolog

lso found to attenuate cue- and drug-induced reinstatement ofETH seeking, authors argued on the possibility that the arachi-

onic acid cascade is likely to mediate endocannabinoid systemarticipation in the reinstatement of METH seeking behaviour56]. However, at doses ranging from 0.032 to 0.32 mg kg−1,n acute intravenous priming of AM251 fails to affect METH-nduced reinstatement of METH seeking [57].

Altogether, these studies clearly demonstrate involvementf an endocannabinoid mechanism in drug-induced reinstate-ent, although discrepant results have been obtained depending

n experimental protocol, range of drug doses or animal strainsed. Notably, CB1 receptor blockade likely produces no effectn stress-induced reinstatement of either cocaine [51] or alco-ol [35] seeking, sustaining the notion that CB1 antagonistsuch as rimonabant may be effective in contrasting drug seek-ng reinstatement elicited by drug primings or presentation ofues predictive of drug availability, rather than preventing thatnduced by a stressor (for recent reviews see refs. [58,59]).

. Conditioned place preference studies

Conditioned place preference (CPP) or aversion (CPA) test-ng is a behavioural method believed capable of measuring theffective (positive, neutral or negative) properties of psychoac-ive drugs and is a commonly used measure of positive incentiveearning [60,61]. CPP is used in evaluation of motivational prop-rties of substances, consisting in repeated association betweenhe primary unconditioned properties of a stimulus (i.e. drug)ith a distinct environment, and a neutral stimulus (i.e. saliner drug vehicle) with a second environment. After a numberf such associations, during which the unconditioned stimulusUCS) acquires properties of conditioned stimulus (CS), animalsre given access to both contexts in a drug-free state. A CPPs demonstrated if the animal spends a greater amount of timen the drug-paired environment than the saline/vehicle-pairednvironment. The preference demonstrates that the CS environ-ent associated with the drug effects has acquired conditioned

ncentive properties [62]. Place-preference learning follows therinciples of classical conditioning, including extinction, dur-ng which animals display a reduced approach to the CS contextollowing its experience in the absence of the UCS drug [63].ased on theoretical formulations of Pavlovian conditioning,PP appears to reflect a preference for a context due to the con-

iguous association between the context and a drug stimulus:ithin this theoretical framework, it seems clear that although

ommonly used for assessing rewarding properties and abuseiability of a drug, CPP measures a learning process that is fun-amentally distinct from drug SA. However, CPP paradigm hasroved to be particularly helpful in unmasking the motivationalroperties of Cannabis derivates, although following repeatednsuccessful attempts [64], discrepant results [65–71], or withhe aid of particular experimental conditions [72,73].

A massive body of evidence has revealed that the endo-

annabinoid system, although not fundamental, constitutes anmportant component in animal perception of the motivationalalue of several drugs of abuse and various other kinds of rein-orcers. In fact, cannabinoids produce a significant effect on

in

t

esearch 56 (2007) 418–427 421

PP induced by cocaine [64], morphine [74–77], nicotine [78],,4 methylenedioxymethamphetamine (MDMA) [79] and theaturally occurring hallucinogen salvinorin A, the active ingre-ient of Salvia divinorum [80]. Moreover, they also affect CPPnduced by standard and palatable food, i.e. sucrose and standardhow pellets. Indeed, consonant with a general role for endo-annabinoid transmission in appetitive motivational processes,B1 receptor antagonists block the acquisition of food-CPP

64], indicative of the ability of the endocannabinoid systemo enhance perception of the rewarding value of different rein-orcers.

As for SA paradigms, typical CPP protocols have beenxtended and tailored into animal models of “extinc-ion/reinstatement” of a previously drug-induced place condi-ioning. By this means, it has been possible to demonstrateow CB1 receptor blockade impairs both the establishmentf nicotine-CPP (as pre-pairing injections of 3.0 mg kg−1 i.p.imonabant prevents acquisition of nicotine-CPP) and the short-ut not long-term expression of nicotine-CPP (as a single pre-est administration of the same dose of rimonabant abolished thexpression of nicotine-CPP when the test session took place 24 hfter the last conditioning session but not when it was conductedor 12 weeks after conditioning) [81,82]. Therefore motiva-

ional effects of nicotine appear to be controlled by endogenousannabinoids, although the long-term expression of such incen-ive learning seems to be independent of endocannabinoidrocesses. Furthermore, when administered during extinctionraining, low doses of THC and the non-psychoactive componentannabidiol (CBD), potentiate the extinction of both cocaine-PP and amphetamine-CPP, at doses (0.25–1.0 mg kg−1 i.p.CH and 5 mg kg−1 i.p. CBD) not affecting learning or retrievalhen administered singly [83]. Besides fear conditioning [84]

nd spatial learning [85], the endocannabinoid system thuseems to play a critical role in the extinction of conditionedncentive learning (Table 1).

. Changes in endocannabinoid contents, CB1 receptorensity and function, and mRNA levels

Within the rat brain, endocannabinoid transmission is signif-cantly altered during chronic drug exposure, as described byeveral studies reporting marked changes in the content of anan-amide (AEA) and 2-arachidonyl-glycerol (2-AG) in cerebralegions related to reward and addiction. Specifically, not onlyepeated administration of THC but also chronic alcohol andicotine leads to decreased AEA and 2-AG contents in the mid-rain and to an increased AEA formation in the limbic forebrain86–88], supporting the existence of region-dependent differ-nces in the regulation of the two major endocannabinoids.onversely, chronic cocaine exhibited only a small, although

ignificant, decrease in the 2-AG content in the limbic fore-rain [88], while morphine induced no effect in AEA level [89].ccordingly, chronic ethanol was reported to increase AEA

n human SK-N-SH cells [90] and 2-AG in cerebellar granuleeurons [91].

Importantly, a very recent paper by Caille et al. [92] addressedhe issue as to whether changes in brain endocannabinoid con-

422 L. Fattore et al. / Pharmacological Research 56 (2007) 418–427

Table 1Effect of drug primings on the reinstatement of extinguished drug self-administration (SA) behaviour

Previous SA Priming Effect References

WIN 55,212-2 WIN 55,212-2 Reinstatement* Spano et al. [24]WIN 55,212-2 Heroin Reinstatement* Spano et al. [24]WIN 55,212-2 Cocaine No effect Spano et al. [24]Heroin WIN 55,212-2 Reinstatement* Fattore et al. [27]Heroin CP 55,940 Reinstatement* Fattore et al. [27]Heroin THC No effect Fattore et al. [27]Heroin Rimonabant/Heroin Antagonism Fattore et al. [28]Heroin HU210 Reinstatement* De Vries et al. [26]Alcohol THC Reinstatement McGregor et al. [37]Alcohol Drug-associated cue Reinstatement* Economidou et al. [35]Alcohol Rimonabant/Alcohol Antagonism Serra et al. [40]Nicotine Drug-associated cue Reinstatement* Cohen et al. [48]; De Vries et al. [49]Nicotine Rimonabant/nicotine Not investigated –Cocaine THC No effect Schenk and Partridge [50]Cocaine HU210 Reinstatement* De Vries et al. [41]Cocaine Rimonabant/cocaine Antagonism De Vries et al. [41]; Filip et al. [52]M *

M

A onist

tidsaSbciw

eidtwaa

iaa[taroCibiped[t

wlpsco

rtoCtoH5pwb

aeirarfil

litl

ETH Drug-associated cueETH Rimonabant/METH

sterisks indicate an effect reversible by pre-treatment with CB1 receptor antag

ent also occur during voluntary drug intake by providing the firstn vivo evidence of variations in limbic endocannabinoid levelsuring drug operant responding. This paper skillfully demon-trates that ethanol SA and heroin SA differentially alter AEAnd 2-AG levels in the rat nucleus accumbens, whereas cocaineA does not. Accordingly, intra-accumbens injection of rimona-ant (1.0–3.0 mg kg−1) reduced both ethanol and heroin, but notocaine SA [92,93], in keeping with our previous studies show-ng that CB1 receptor antagonism or deletion does not interfereith cocaine SA [47,94].Regardless of the evidence of significant changes in brain

ndocannabinoid signalling as a consequence of drug admin-strations and more importantly of voluntary drug intake, toate only one study has been published on possible variations inhe endocannabinoid content during relapse to drug seeking, inhich the authors describe a significant reduction of both AEA

nd 2-AG contents in the brains of rats allowed to relapse tolcohol after abstinence [36].

At cellular level, attenuation of both CB1 receptor bind-ng (down-regulation) and CB1 agonist-stimulated G-proteinctivation (desensitization) occur after long-term cannabinoiddministration and are believed to contribute to tolerance95,96]. Decreased CB1 receptor level and efficiency take placehroughout the brain, although the magnitude of changes variesmong regions [97–99]. For example, a down-regulation of CB1eceptors was observed following chronic treatment with THCr CP 55,940 [100–102] but not cocaine [103]. Moreover, whileB1 receptor-mediated G-protein signalling remained unaltered

n various brain areas of rats given either rimonabant or rimona-ant plus nicotine for 3-weeks CPP [82], CB1 receptor binding isncreased and decreased by chronic morphine in limbic and hip-ocampal structures, respectively [103]. Finally, chronic alcohol

xposure resulted in a decreased number of CB1 receptors and aesensitization of the cannabinoid-activated signal transduction87]. Accordingly, CB1 receptor binding was downregulated inhe prefrontal cortex of alcohol-preferring Alko Alcohol (AA)

dOmc

Reinstatement Anggadiredja et al. [56]Antagonism Anggadiredja et al. [56]

s.

hen compared to non-preferring Alko Non-Alcohol (ANA) ratines [104] and in the brain membranes of C57BL/6 (alcohol-referring) with respect to DBA/2 (alcohol-avoiding) mice [91],uggesting an overactive endocannabinoid transmission and aompensatory downregulation of CB1 signalling in the brainsf animals genetically selected for their preference for alcohol.

However, when CB1 receptor binding is measured in operantats incongruous results have been observed, with CB1 recep-or density and functionality varying differentially dependingn the specific drug used and the length of SA training orPP conditioning. We recently demonstrated that CB1 recep-

or binding and functionality are altered in a differential, seldompposite, manner in several reward-related brain areas of Listerooded rats trained to intravenously self-administer either WIN5,212-2 (12.5 �g kg−1 per infusion) or heroin (0.03 mg kg−1

er infusion), providing a possible biochemical basis for theell-known interactions between opioid and CB1 receptors atehavioural level [105].

Again, despite the body of evidence pointing to importantlterations in CB1 receptor number and efficiency followingither passive or active drug exposure, only one study to date hasnvestigated possible variations in CB1 receptor binding duringelapse to drug seeking after abstinence [36]. In this paper, inddition to changes in endocannabinoid content, authors alsoeported CB1 receptor binding as being only minimally modi-ed during relapse to alcohol, with a significant increase being

imited to the superficial layer of the cerebral cortex.Changes in CB1 receptor mRNA occur in the rat brain fol-

owing chronic exposure to THC, which has been shown toncrease CB1 mRNA levels above vehicle control animals inhe cerebellum and hippocampus, and to severely reduce theseevels in the striatum and caudate-putamen, the magnitude and

irection of changes varying with treatment duration [106,107].n the other hand, both acute and chronic administration oforphine decreases the expression of the CB1 mRNA in the

audate-putamen, cerebellum and the shell portion of the nucleus

L. Fattore et al. / Pharmacological R

Table 2Changes in endocannabinoid contents, CB1 receptor density and efficiencyduring relapse to alcohol

Brain Changes Relapse to Alcohol Reference

AEA content ↓ Gonzalez et al. [36]2-AG content ↓ Gonzalez et al. [36]CB1 binding ↑ Gonzalez et al. [36]C

Nr

asggiartpCaanmaoid

cbibisdtd

5

ias[nprnc[Dtit

as

aqdpupnart

octtimhurb[oatfacbttttn(

mhcimApatdstbn

B1 mRNA ↑ Gonzalez et al. [36]

o other studies exist to date on possible variations of such parameters duringelapse to other drugs of abuse.

ccumbens [76,103,108], while repeated treatment with theerotonin reuptake inhibitor fluoxetine decreases CB1 receptorene expression by approximately 45–50% from vehicle, sug-esting that these neuroplastic changes may be involved, at leastn part, in the clinical efficacy of cannabinoids in neuropsychi-tric disorders [109]. Accordingly, chronic ethanol consumptioneduced CB1 receptor gene expression in the caudate-putamen,he ventromedial nucleus of the hypothalamus and in the hip-ocampus [110]. Conversely, an increased expression of theB1 receptor has been described in reward-relevant brainreas of drug-naıve genetically selected Marchigian-Sardinianlcohol-preferring (msP) rats in comparison with drug-naıveon-selected Wistar rats [42], pointing to a link between CB1RNA levels and distribution with genetic predisposition to

lcohol abuse. However, this proposal has been hampered bypposite findings, showing marked reductions of CB1 mRNAn dorsal striatum of AA rats, which was not accompanied byecreased binding [104].

Once again, as previously described for endocannabinoidontent and CB1 receptor binding, only one research study haseen conducted so far to evaluate CB1 mRNA expression dur-ng re-exposure to the drug after a period of abstinence. In therains of animals relapsing to alcohol, the latter study detected anncreased mRNA expression in the medial caudate-putamen, theuperficial and deep layers of the cerebral cortex, the ventrome-ial hypothalamic nucleus and the hippocampus [36], indicatinghat mRNA levels during relapse did not change in the sameirection as during voluntary alcohol intake (Table 2).

. Discussion

Although in recent years enormous progress has been maden delineating biological features, neurocircuitry componentsnd signalling mechanisms mediating craving, conditioned drugeeking behaviour and long-lasting susceptibility to relapse111], interventions directed at relapse prevention are unfortu-ately still problematic and only effective in highly-motivatedatients supported by stable social and affective conditions. Byegulating the brain reward circuits [112,113], endocannabi-oid signalling is involved in almost all stages of addictionycle, from maintenance of stably acquired addictive behaviour114] to relapse to drug seeking following abstinence [59].

ata collected to date from behavioural studies strongly support

he notion that the endocannabinoid system mediates relaps-ng episodes following cessation of drug intake, highlightinghe importance of considering a CB1 antagonist-based ther-

Grnt

esearch 56 (2007) 418–427 423

py for drug and context-induced promotion of relapse to drugeeking.

The exact brain location at which CB1 receptors medi-te these effects on drug taking and seeking remains an openuestion. CB1 receptors belong to the seven transmembraneomain family of G-protein-coupled receptors, are linked to Gi/oroteins, inhibit adenylate cyclase [115,116], and are almostbiquitary within the brain where they regulate a variety ofhysiological functions [9]. For example, they are found in highumbers in the ventral tegmental area and in the basolateralmygdala [117,118], part of the corticolimbic circuit mediatingespectively drug-primed reinstatement and drug-cue associa-ions (i.e. conditioning) [119,120].

Among the neurotransmission pathways in which modulationf relapse to drug seeking is likely to be hampered by the endo-annabinoid system, particular emphasis has been focused onhe dopaminergic system, as pharmacological manipulations ofhe endocannabinoid system are believed to attenuate cocaine-nduced reinstatement via inhibition of drug priming-induced

esocorticolimbic dopamine release [8,121]. Accordingly, itas been proposed that blockade of CB1 receptors might mod-late the motivational effects of drugs of abuse or naturalewards by blocking the disinhibitory effects indirectly exertedy endocannabinoids on mesolimbic dopaminergic neurons122]. This could explain, for example, why in CB1 knock-ut mice the dopamine-releasing effects of morphine [123] andlcohol [124] are significantly reduced or why dopamine fluc-uations in the nucleus accumbens, known to accompany sex,ood and drug seeking behaviour [125–127], can all be attenu-ted by rimonabant [128]. However, based on the evidence for aannabinoid–dopamine interaction in the striatum [129], it coulde hypothesised that activation of the endocannabinoid sys-em downstream of the dopamine synapse may also contributeowards cocaine and heroin seeking [26–28,51]. Nonetheless,he finding that rimonabant blocks the potentiation of relapseo food seeking induced by the dopamine D3 receptor ago-ist [130] suggested an impact of CB1 blockade on dopamineD3)-mediated reward-related processes.

On the other hand, N-methyl-d-aspartate (NMDA) andetabotropic glutamate (Glu) receptor (mGluR5) antagonists

ave also been shown to prevent cue-induced reinstatement ofocaine and nicotine seeking, respectively [131,132], thus point-ng to potential glutamate–endocannabinoids interactions in

odulating discrete cue-induced reinstatement of drug seeking.ctivation of post-synaptic Glu receptors is known to suppressre-synaptic functions through the endocannabinoid system inretrograde manner [133,134], lending support to the notion

hat endocannabinoids may act as retrograde messengers fromepolarised post-synaptic neurons to pre-synaptic terminals touppress the release of glutamate. In this view, one could arguehat CB1 antagonists may attenuate drug seeking reinstatementy inhibiting glutamate release [135], even though endocannabi-oids also act in a retrograde manner on axon terminals of

ABAergic interneurons to reduce GABA release [136]. The

espective intensities of these opposing Glu-GABA effects haveot yet been assessed, however some authors have proposed thathe opposite actions exerted by dopaminergic drugs (i.e. nico-

4 colog

tcpd

cctssvpaoi

R

24 L. Fattore et al. / Pharma

ine) and endocannabinoids on glutamatergic and GABAergicontrol over dopamine neurons in limbic area may representossible mechanisms through which rimonabant interferes withrug-induced neuronal plasticity [82].

Finally, potential involvement of the arachidonic acid cas-ade, known to regulate numerous intracellular actions ofannabinoids, in the role played by the endocannabinoid sys-em in relapse should also be taken into account [137]. Inupport of the latter, THC and AEA have been shown totimulate arachidonic acid mobilization [138,139], induce acti-ation of phospholipase [140,141], and increase brain levels ofrostaglandin E2, one of the final products of the arachidoniccid cascade [142,143], while behavioural suppressant effectsf THC and HU 210 are antagonised by the cyclooxigenasenhibitor diclofenac [144].

eferences

[1] Hodgins DC, el-Guebaly N. Retrospective and prospective reports ofprecipitants to relapse in pathological gambling. J Consult Clin Psychol2004;72:72–80.

[2] Ledgerwood DM, Petry NM. What do we know about relapse in patho-logical gambling? Clin Psychol Rev 2006;26:216–28.

[3] Mitchell JE, de Zwaan M, Roerig JL. Drug therapy for patients witheating disorders. Curr Drug Targets CNS Neurol Disord 2003;2:17–29.

[4] Pederson KJ, Roerig JL, Mitchell JE. Towards the pharmacotherapy ofeating disorders. Expert Opin Pharmacother 2003;4:1659–78.

[5] O’Brien CP. Anticraving medications for relapse prevention: a possi-ble new class of psychoactive medications. Am J Psychiatry 2005;162:1423–31.

[6] Iwamoto E, Martin W. A critique of drug self-administration as amethod for predicting abuse potential of drugs. NIDA Res Monogr1988;81:457–65.

[7] Katz JL, Higgins ST. The validity of the reinstatement model of cravingand relapse to drug use. Psychopharmacology 2003;168:21–30.

[8] Shaham Y, Shalev U, Lu L, De Wit H, Stewart J. The reinstatement modelof drug relapse: history, methodology and major findings. Psychophar-macology 2003;168:3–20.

[9] Ameri A. The effects of cannabinoids on the brain. Prog Neurobiol1999;58:315–48.

[10] Chhatwal JP, Davis M, Maguschak KA, Ressler KJ. Enhancing cannabi-noid neurotransmission augments the extinction of conditioned fear.Neuropsychopharmacology 2005;3:516–24.

[11] Lin HC, Mao SC, Gean PW. Effects of intra-amygdala infusion of CB1receptor agonists on the reconsolidation of fear-potentiated startle. LearnMem 2006;13:316–21.

[12] Kathuria S, Gaetani S, Fegley D, Valino F, Duranti A, Tontini A, et al.Modulation of anxiety through blockade of anandamide hydrolysis. NatMed 2003;9:76–81.

[13] Gobbi G, Bambico FR, Mangieri R, Bortolato M, Campolongo P, SolinasM, et al. Antidepressant-like activity and modulation of brain monoamin-ergic transmission by blockade of anandamide hydrolysis. Proc Natl AcadSci 2005;102:18620–5.

[14] Piomelli D. The molecular logic of endocannabinoid signalling. Nat RevNeurosci 2003;4:873–84.

[15] Gardner EL. What we have learned about addiction from animal modelsof drug self-administration. Am J Addict 2000;9:285–313.

[16] Martellotta MC, Cossu G, Fattore L, Gessa GL, Fratta W. Self-

administration of the cannabinoid receptor agonist WIN 55,212-2 indrug-naive mice. Neuroscience 1998;85:327–30.

[17] Fattore L, Cossu G, Martellotta CM, Fratta W. Intravenous self-administration of the cannabinoid CB1 receptor agonist WIN 55,212-2in rats. Psychopharmacology 2001;156:410–6.

ical Research 56 (2007) 418–427

[18] Fattore L, Cossu G, Fratta W. Functional interaction between cannabi-noids and opioids in animal models of drug addiction. In: Proceedings ofthe “Frontiers in Addiction Research”. NIDA Symposium Satellite at theSfN Meeting. 2002.

[19] Justinova Z, Tanda G, Redhi GH, Goldberg SR. Self-administration ofdelta9-tetrahydrocannabinol (THC) by drug naive squirrel monkeys. Psy-chopharmacology 2003;169:135–40.

[20] Justinova Z, Solinas M, Tanda G, Redhi GH, Goldberg SR.The endogenous cannabinoid anandamide and its synthetic analogR(+)-methanandamide are intravenously self-administered by squirrelmonkeys. J Neurosci 2005;25:5645–50.

[21] Aston-Jones G, Harris GC. Brain substrates for increased drug seekingduring protracted withdrawal. Neuropharmacology 2004;47:167–79.

[22] Self DW. Regulation of drug-taking and -seeking behaviors by neu-roadaptations in the mesolimbic dopamine system. Neuropharmacology2004;47:242–55.

[23] Kalivas PW, Hu XT. Exciting inhibition in psychostimulant addiction.Trends Neurosci 2006;29:610–6.

[24] Spano MS, Fattore L, Cossu G, Deiana S, Fadda P, Fratta W. CB1 recep-tor agonist and heroin, but not cocaine, reinstate cannabinoid-seekingbehaviour in the rat. Br J Pharmacol 2004;143:343–50.

[25] Fattore L, Deiana S, Spano MS, Cossu G, Fadda P, Scherma M, et al. Endo-cannabinoid system and opioid addiction: behavioural aspects. PharmacolBiochem Behav 2005;81:343–59.

[26] De Vries TJ, Homberg JR, Binnekade R, Raaso H, Schoffelmeer AN.Cannabinoid modulation of the reinforcing and motivational proper-ties of heroin and heroin-associated cues in rats. Psychopharmacology2003;168:164–9.

[27] Fattore L, Spano MS, Cossu G, Deiana S, Fratta W. Cannabinoid mecha-nism in reinstatement of heroin-seeking after a long period of abstinencein rats. Eur J Neurosci 2003;17:1723–6.

[28] Fattore L, Spano MS, Cossu G, Deiana S, Fadda P, Fratta W.Cannabinoid CB(1) antagonist SR 141716A attenuates reinstatement ofheroin self-administration in heroin-abstinent rats. Neuropharmacology2005;48:1097–104.

[29] Corchero J, Manzanares J, Fuentes JA. Cannabinoid/opioid crosstalk inthe central nervous system. Crit Rev Neurobiol 2004;16:159–72.

[30] Fattore L, Cossu G, Spano MS, Deiana S, Fadda P, Scherma M, et al.Cannabinoids and reward: interactions with the opioid system. Crit RevNeurobiol 2004;16:147–58.

[31] Parolaro D, Vigano D, Rubino T. Endocannabinoids and drug dependence.Curr Drug Targets CNS Neurol Disord 2005;4:643–55.

[32] Colombo G, Agabio R, Fa M, Guano L, Lobina C, Loche A, et al.Reduction of voluntary ethanol intake in ethanol-preferring sP ratsby the cannabinoid antagonist SR-141716. Alcohol Alcohol 1998;33:126–30.

[33] Serra S, Carai MA, Brunetti G, Gomez R, Melis S, Vacca G, etal. The cannabinoid receptor antagonist SR 141716 prevents acquisi-tion of drinking behavior in alcohol-preferring rats. Eur J Pharmacol2001;430:369–71.

[34] Arnone M, Maruani J, Chaperon F, Thiebot MH, Poncelet M, Soubrie P,et al. Selective inhibition of sucrose and ethanol intake by SR 141716, anantagonist of central cannabinoid (CB1) receptors. Psychopharmacology1997;132:104–6.

[35] Economidou D, Mattioli L, Cifani C, Perfumi M, Massi M, Cuomo V,et al. Effect of the cannabinoid CB1 receptor antagonist SR-141716Aon ethanol self-administration and ethanol-seeking behaviour in rats.Psychopharmacology 2006;183:394–403.

[36] Gonzalez S, Valenti M, de Miguel R, Fezza F, Fernandez-Ruiz J, Di MarzoV, et al. Changes in endocannabinoid contents in reward-related brainregions of alcohol-exposed rats, and their possible relevance to alcoholrelapse. Br J Pharmacol 2004;143:455–64.

[37] McGregor IS, Dam KD, Mallet PE, Gallate JE. Delta9-THC rein-

states beer- and sucrose-seeking behaviour in abstinent rats: comparisonwith midazolam, food deprivation and predator odour. Alcohol Alcohol2005;40:35–45.

[38] Lopez-Moreno JA, Gonzalez-Cuevas G, Rodriguez de Fonseca F. NavarroM. Long-lasting increase of alcohol relapse by the cannabinoid recep-

ical R

L. Fattore et al. / Pharmacolog

tor agonist WIN 55,212-2 during alcohol deprivation. J Neurosci2004;24:8245–52.

[39] Spanagel R, Holter SM. Pharmacological validation of a new animalmodel of alcoholism. J Neural Transm 2000;107:669–80.

[40] Serra S, Brunetti G, Pani M, Vacca G, Carai MA, Gessa GL, et al.Blockade by the cannabinoid CB(1) receptor antagonist, SR 141716,of alcohol deprivation effect in alcohol-preferring rats. Eur J Pharmacol2002;443:95–7.

[41] Gessa GL, Serra S, Vacca G, Carai MA, Colombo G. Suppressing effectof the cannabinoid CB1 receptor antagonist, SR147778, on alcohol intakeand motivational properties of alcohol in alcohol-preferring sP rats. Alco-hol Alcohol 2005;40:46–53.

[42] Cippitelli A, Bilbao A, Hansson AC, del Arco I, Sommer W, HeiligM, et al. Cannabinoid CB1 receptor antagonism reduces conditionedreinstatement of ethanol-seeking behavior in rats. Eur J Neurosci2005;21:2243–51.

[43] Economidou D, Mattioli L, Ubaldi M, Lourdusamy A, Soverchia L,Hardiman G, et al. Role of cannabinoidergic mechanisms in ethanolself-administration and ethanol seeking in rat adult offspring follow-ing perinatal exposure to Delta(9)-tetrahydrocannabinol. Toxicol ApplPharmacol 2007;15:73–85.

[44] Cippitelli A, Bilbao A, Gorriti MA, Navarro M, Massi M, Piomelli D,et al. The anandamide transport inhibitor AM404 reduces ethanol self-administration. Eur J Neurosci 2007;26:476–86.

[45] Cohen C, Perrault G, Voltz C, Steinberg R, Soubrie P. SR141716, a cen-tral cannabinoid (CB(1)) receptor antagonist, blocks the motivationaland dopamine-releasing effects of nicotine in rats. Behav Pharmacol2002;13:451–63.

[46] Solinas M, Scherma M, Fattore L, Stroike J, Wertheim C, Tanda G, et al.Nicotinic �7 receptors as a new target for treatment of cannabis abuse. JNeurosci 2007;27:5615–20.

[47] Cossu G, Ledent C, Fattore L, Imperato A, Bohme GA, Parmentier M,et al. Cannabinoid CB1 receptor knockout mice fail to self-administermorphine but not other drugs of abuse. Behav Brain Res 2001;118:61–5.

[48] Cohen C, Perrault G, Griebel G, Soubrie P. Nicotine-associated cuesmaintain nicotine-seeking behavior in rats several weeks after nico-tine withdrawal: reversal by the cannabinoid (CB1) receptor antagonist,rimonabant (SR141716). Neuropsychopharmacology 2004;30:145–55.

[49] De Vries TJ, de Vries W, Janssen MC, Schoffelmeer AN. Suppression ofconditioned nicotine and sucrose seeking by the cannabinoid-1 receptorantagonist SR141716A. Behav Brain Res 2005;161:164–8.

[50] Schenk S, Partridge B. Cocaine-seeking produced by experimenter-administered drug injections: dose-effect relationships in rats. Psy-chopharmacology 1999;147:285–90.

[51] De Vries TJ, Shaham Y, Homberg JR, Crombag H, Schuurman K, DiebenJ, et al. A cannabinoid mechanism in relapse to cocaine seeking. Nat Med2001;7:1151–4.

[52] Filip M, Golda A, Zaniewska M, McCreary AC, Nowak E, KolasiewiczW, et al. Involvement of cannabinoid CB(1) receptors in drug addic-tion: effects of rimonabant on behavioral responses induced by cocaine.Pharmacol Rep 2006;58:806–19.

[53] Xi ZX, Gilbert JG, Peng XQ, Pak AC, Li X, Gardner EL. CannabinoidCB1 receptor antagonist AM251 inhibits cocaine-primed relapse in rats:role of glutamate in the nucleus accumbens. J Neurosci 2006;26:8531–6.

[54] Gonzalez-Cuevas G, Aujla H, Martin-Fardon R, Lopez-Moreno JA,Navarro M, Weiss F. Subchronic cannabinoid agonist/WIN 55,212-2)treatment during cocaine abstinence alters subsequent cocaine seekingbehaviour. Neuropsychopharmacology 2007 [epub ahead of print, March28, 2007].

[55] Vinklerova J, Novakova J, Sulcova A. Inhibition of methamphetamineself-administration in rats by cannabinoid receptor antagonist AM 251. JPsychopharmacol 2002;16:139–43.

[56] Anggadiredja K, Nakamichi M, Hiranita T, Tanaka H, Shoyama Y,

Watanabe S, et al. Endocannabinoid system modulates relapse to metham-phetamine seeking: possible mediation by the arachidonic acid cascade.Neuropsychopharmacology 2004;29:1470–8.

[57] Boctor SY, Martinez Jr JL, Koek W, France CP. The cannabinoid CB(1)receptor antagonist AM251 does not modify methamphetamine reinstate-

esearch 56 (2007) 418–427 425

ment of responding. Eur J Pharmacol 2007 [epub ahead of print, June 13,2007].

[58] De Vries TJ, Schoffelmeer AN. Cannabinoid CB1 receptors control con-ditioned drug seeking. Trends Pharmacol Sci 2005;26:420–6.

[59] Fattore L, Spano MS, Deiana S, Melis V, Cossu G, Fadda P, et al. An endo-cannabinoid mechanism in relapse to drug seeking: a review of animalstudies and clinical perspectives. Brain Res Rev 2007;53:1–16.

[60] Cunningham CL. Drug conditioning and drug-seeking behavior. In:Donohue W, editor. Learning and behavior therapy. Boston, MA: Allyn& Bacon; 1998. p. 518–44.

[61] Tzschentke TM. Measuring reward with the conditioned place preferenceparadigm: a comprehensive review of drug effects, recent progress andnew issues. Prog Neurobiol 1998;56:613–72.

[62] Robinson TE, Berridge KC. The neural basis of drug craving: anincentive-sensitization theory of addiction. Brain Res Rev 1993;18:247–91.

[63] Parker LA, McDonald RV. Reinstatement of both a conditioned placepreference and a conditioned place aversion with drug primes. PharmacolBiochem Behav 2000;66:559–61.

[64] Chaperon F, Soubrie P, Puech AJ, Thiebot MH. Involvement of centralcannabinoid (CB1) receptors in the establishment of place conditioningin rats. Psychopharmacology 1998;135:324–32.

[65] Lepore M, Vorel SR, Lowinson J, Gardner EL. Conditioned place prefer-ence induced by delta 9-tetrahydrocannabinol: comparison with cocaine,morphine, and food reward. Life Sci 1995;56:2073–80.

[66] Parker LA, Gillies T. THC-induced place and taste aversions in Lewisand Sprague–Dawley rats. Behav Neurosci 1995;109:71–8.

[67] McGregor IS, Issakidis CN, Prior G. Aversive effects of the syntheticcannabinoid CP 55,940 in rats. Pharm Biochem Behav 1996;53:657–64.

[68] Sanudo-Pena MC, Tsou K, Delay ER, Hohman AG, Force M, Walker JM.Endogenous cannabinoids as an aversive or counter-rewarding system inthe rat. Neurosci Lett 1997;223:125–8.

[69] Mallet PE, Beninger RJ. Delta9-tetrahydrocannabinol, but not the endoge-nous cannabinoid receptor ligand anandamide, produces conditionedplace avoidance. Life Sci 1998;62:2431–9.

[70] Cheer JF, Kendall DA, Marsden CA. Cannabinoid receptors and rewardin the rat: a conditioned place preference study. Psychopharmacology2000;151:25–30.

[71] Braida D, Iosue S, Pegorini S, Sala M. Delta9-tetrahydrocannabinol-induced conditioned place preference and intracerebroventricularself-administration in rats. Eur J Pharmacol 2004;506:63–9.

[72] Valjent E, Maldonado R. A behavioural model to reveal place pref-erence to delta 9-tetrahydrocannabinol in mice. Psychopharmacology2000;147:436–8.

[73] Bortolato M, Campolongo P, Mangieri RA, Scattoni ML, Frau R, TrezzaV, et al. Anxiolytic-like properties of the anandamide transport inhibitorAM404. Neuropsychopharmacology 2006;31:2652–9.

[74] Braida D, Pozzi M, Cavallini R, Sala M. Conditioned place preferenceinduced by the cannabinoid agonist CP 55,940: interaction with the opioidsystem. Neuroscience 2001;104:923–6.

[75] Mas-Nieto M, Pommier B, Tzavara ET, Caneparo A, Da Nascimento S,Le Fur G, et al. Reduction of opioid dependence by the CB(1) antagonistSR141716A in mice: evaluation of the interest in pharmacotherapy ofopioid addiction. Br J Pharmacol 2001;132:1809–16.

[76] Navarro M, Carrera MR, Fratta W, Valverde O, Cossu G, Fattore L, et al.Functional interaction between opioid and cannabinoid receptors in drugself-administration. J Neurosci 2001;21:5344–50.

[77] Singh ME, Verty AN, McGregor IS, Mallet PE. A cannabinoid receptorantagonist attenuates conditioned place preference but not behaviouralsensitization to morphine. Brain Res 2004;1026:244–53.

[78] Valjent E, Mitchell JM, Besson MJ, Caboche J, Maldonado R.Behavioural and biochemical evidence for interactions between Delta9-tetrahydrocannabinol and nicotine. Br J Pharmacol 2002;135:564–78.

[79] Braida D, Iosue S, Pegorini S, Sala M. 3,4 Methylenedioxy-methamphetamine-induced conditioned place preference (CPP) is medi-ated by endocannabinoid system. Pharmacol Res 2005;51:177–82.

[80] Braida D, Limonta V, Pegorini S, Zani A, Guerini-Rocco C, Gori E, et al.Hallucinatory and rewarding effect of salvinorin A in zebrafish: kappa-

4 colog

26 L. Fattore et al. / Pharma

opioid and CB1-cannabinoid receptor involvement. Psychopharmacology2007;190:441–8.

[81] Forget B, Hamon M, Thiebot MH. Cannabinoid CB1 receptors areinvolved in motivational effects of nicotine in rats. Psychopharmacology2005;181:722–34.

[82] Forget B, Barthelemy S, Saurini F, Hamon M, Thiebot MH. Differen-tial involvement of the endocannabinoid system in short- and long-termexpression of incentive learning supported by nicotine in rats. Psy-chopharmacology 2006;189:59–69.

[83] Parker LA, Burton P, Sorge RE, Yakiwchuk C, Mechoulam R. Effect oflow doses of delta9-tetrahydrocannabinol and cannabidiol on the extinc-tion of cocaine-induced and amphetamine-induced conditioned placepreference learning in rats. Psychopharmacology 2004;175:360–6.

[84] Marsicano G, Wotjak CT, Azad SC, Bisogno T, Rammes G, Cascio MG,et al. The endogenous cannabinoid system controls extinction of aversivememories. Nature 2002;418:530–4.

[85] Varvel SA, Lichtman AH. Evaluation of CB1 receptor knockout mice inthe Morris water maze. J Pharm Exp Ther 2002;301:915–24.

[86] Di Marzo V, Berrendero F, Bisogno T, Gonzalez S, Cavaliere P, RomeroJ, et al. Enhancement of anandamide formation in the limbic forebrainand reduction of endocannabinoid contents in the striatum of delta9-tetrahydrocannabinol-tolerant rats. J Neurochem 2000;74:1627–35.

[87] Basavarajappa BS, Hungund BL. Neuromodulatory role of the endo-cannabinoid signaling system in alcoholism: an overview. ProstaglandinsLeukot Essent Fatty Acids 2002;66:287–99.

[88] Gonzalez S, Cascio MG, Fernandez-Ruiz J, Fezza F, Di Marzo V, RamosJA. Changes in endocannabinoid contents in the brain of rats chronicallyexposed to nicotine, ethanol or cocaine. Brain Res 2002;954:73–81.

[89] Gonzalez S, Schmid PC, Fernandez-Ruiz J, Krebsbach R, Schmid HH,Ramos JA. Region-dependent changes in endocannabinoid transmissionin the brain of morphine-dependent rats. Addict Biol 2003;8:159–66.

[90] Basavarajappa BS, Saito M, Cooper TB, Hungund BL. Stimulation ofcannabinoid receptor agonist 2-arachidonylglycerol by chronic ethanoland its modulation by specific neuromodulators in cerebellar granuleneurons. Biochim Biophys Acta 2000;1535:78–86.

[91] Basavarajappa BS, Hungund BL. Chronic ethanol increases the cannabi-noid receptor agonist anandamide and its precursor N-arachidonoyl-phosphatidylethanolamine in SK-N-SH cells. J Neurochem 1999;72:522–8.

[92] Caille S, Alvarez-Jaimes L, Polis I, Stouffer DG, Parsons LH. Specificalterations of extracellular endocannabinoid levels in the nucleus accum-bens by ethanol, heroin, and cocaine self-administration. J Neurosci2007;27:3695–702.

[93] Caille S, Parsons LH. Cannabinoid modulation of opiate reinforcementthrough the ventral striatopallidal pathway. Neuropsychopharmacology2006;31:804–13.

[94] Fattore L, Martellotta MC, Cossu G, Mascia MS, Fratta W. CB1 cannabi-noid receptor agonist WIN 55,212-2 decreases intravenous cocaineself-administration in rats. Behav Brain Res 1999;104:141–6.

[95] Breivogel CS, Childers SR, Deadwyler SA, Hampson RE, Vogt LJ, Sim-Selley LJ. Chronic delta9-tetrahydrocannabinol treatment produces atime-dependent loss of cannabinoid receptors and cannabinoid receptor-activated G proteins in rat brain. J Neurochem 1999;73:2447–59.

[96] Martin BR, Sim-Selley LJ, Selley DE. Signaling pathways involvedin the development of cannabinoid tolerance. Trends Pharmacol2004;25:325–30.

[97] Sim LJ, Hampson RE, Deadwyler SA, Childers SR. Effects of chronictreatment with �9-tetrahydrocannabinol on cannabinoid-stimulated[35S]GTP�S autoradiography in rat brain. J Neurosci 1996;16:8057–66.

[98] Sim-Selley LJ, Martin BR. Effect of chronic administration of WIN55,212-2 or �9-tetrahydrocannabinol on cannabinoid receptor adaptationin mice. J Pharmacol Exp Ther 2002;303:36–44.

[99] Romero J, Garcia-Palomero E, Castro JG, Garcia-Gil L, Ramos JA,

Fernandez-Ruiz JJ. Effects of chronic exposure to �9-tetrahydro-cannabinol on cannabinoid receptor binding and mRNA levels in severalrat brain regions. Mol Brain Res 1997;46:100–8.

[100] de Fonseca FR, Gorriti MA, Fernandez-Ruiz JJ, Palomo T, Ramos JA.Downregulation of rat brain cannabinoid binding sites after chronic

ical Research 56 (2007) 418–427

delta 9-tetrahydrocannabinol treatment. Pharmacol Biochem Behav1994;47:33–40.

[101] Breivogel CS, Scates SM, Beletskaya IO, Lowery OB, Aceto MD, MartinBR. The effects of delta9-tetrahydrocannabinol physical dependence onbrain cannabinoid receptors. Eur J Pharmacol 2003;459:139–50.

[102] Rubino T, Massi P, Patrini G, Venier I, Giagnoni G, Parolaro D. ChronicCP-55,940 alters cannabinoid receptor mRNA in the rat brain: an in situhybridization study. Neuroreport 1994;5:2493–6.

[103] Gonzalez S, Fernandez-Ruiz J, Sparpaglione V, Parolaro D, Ramos JA.Chronic exposure to morphine, cocaine or ethanol in rats produced dif-ferent effects in brain cannabinoid CB(1) receptor binding and mRNAlevels. Drug Alcohol Depend 2002;66:77–84.

[104] Hansson AC, Bermudez-Silva FJ, Malinen H, Hyytia P, Sanchez-Vera I,Rimondini R, et al. Genetic impairment of frontocortical endocannabinoiddegradation and high alcohol preference. Neuropsychopharmacology2007;32:117–26.

[105] Fattore L, Vigano D, Fadda P, Rubino T, Fratta W, Parolaro D.Bidirectional regulation of Mu-opioid and CB1-cannabinoid receptorin rats self-administering heroin or WIN 55,212-2. Eur J Neurosci2007;25:2191–200.

[106] Corchero J, Romero J, Berrendero F, Fernandez-Ruiz J, Ramos JA,Fuentes JA, et al. Time-dependent differences of repeated administra-tion with Delta9-tetrahydrocannabinol in proenkephalin and cannabinoidreceptor gene expression and G-protein activation by mu-opioid and CB1-cannabinoid receptors in the caudate-putamen. Brain Res Mol Brain Res1999;67:148–57.

[107] Zhuang S, Kittler J, Grigorenko E, Kirby MT, Sim LJ, Hampson RE, et al.Effects of long-term exposure to delta9-THC on expression of cannabi-noid receptor (CB1) mRNA in different rat brain regions. Mol Brain Res1998;62:141–9.

[108] Rubino T, Tizzoni L, Vigano D, Massi P, Parolaro D. Modulation of ratbrain cannabinoid receptors after chronic morphine treatment. Neurore-port 1997;8:3219–23.

[109] Oliva JM, Uriguen L, Perez-Rial S, Manzanares J. Time course of opi-oid and cannabinoid gene transcription alterations induced by repeatedadministration with fluoxetine in the rat brain. Neuropharmacology2005;49:618–26.

[110] Ortiz S, Oliva JM, Perez-Rial S, Palomo T, Manzanares J. Chronic ethanolconsumption regulates cannabinoid CB1 receptor gene expression inselected regions of rat brain. Alcohol Alcohol 2004;39:88–92.

[111] Weiss S. Neurobiology of craving, conditioned reward and relapse. CurrOpin Pharmacol 2005;5:9–19.

[112] Lupica CR, Riegel AC, Hoffman AF. Marijuana and cannabinoid regula-tion of brain reward circuits. Br J Pharmacol 2004;143:227–34.

[113] Gardner EL. Endocannabinoid signaling system and brain reward: empha-sis on dopamine. Pharmacol Biochem Behav 2005;81:263–84.

[114] Le Foll B, Goldberg SR. Cannabinoid CB1 receptor antagonists aspromising new medications for drug dependence. J Pharmacol Exp Ther2005;312:875–83.

[115] Howlett AC, Johnson MR, Melvin LS, Milne GM. Nonclassical cannabi-noid analgetics inhibit adenylate cyclase: development of a cannabinoidreceptor model. Mol Pharmacol 1988;33:297–302.

[116] Howlett AC, Bidaut-Russell M, Devane WA, Melvin LS, Johnson MR,Herkenham M. The cannabinoid receptor: biochemical, anatomical andbehavioral characterization. Trends Neurosci 1990;13:420–3.

[117] Herkenham M, Lynn AB, Little MD, Johnson MR, Melvin LS, de CostaBR, et al. Cannabinoid receptor localization in brain. Proc Natl Acad Sci1990;87:1932–6.

[118] Glass M, Dragunow M, Faull RLM. Cannabinoid receptors in the humanbrain: a detailed anatomical and quantitative autoradiographic study inthe fetal, neonatal and adult human brain. Neuroscience 1997;77:299–318.

[119] Kalivas PW, McFarland K. Brain circuitry and the reinstatement of

cocaine-seeking behavior. Psychopharmacology 2003;168:44–56.

[120] See RE, Fuchs RA, Ledford CC, McLaughlin J. Drug addiction, relapse,and the amygdala. Ann NY Acad Sci 2003;985:294–307.

[121] Shalev U, Grimm JW, Shaham Y. Neurobiology of relapse to heroin andcocaine seeking: a review. Pharmacol Rev 2002;54:1–42.

ical R

levels in rodents by delta 1-tetrahydrocannabinol. Prostaglandins LeukotEssent Fatty Acids 1991;43:185–90.

L. Fattore et al. / Pharmacolog

[122] Schlicker E, Kathmann M. Modulation of transmitter release via presy-naptic cannabinoid receptors. Trends Pharmacol Sci 2001;22:565–72.

[123] Mascia MS, Obinu MC, Ledent C, Parmentier M, Bohme GA, Imperato A,et al. Lack of morphine-induced dopamine release in the nucleus accum-bens of cannabinoid CB(1) receptor knockout mice. Eur J Pharmacol1999;383:R1–2.

[124] Hungund BL, Szakall I, Adam A, Basavarajappa BS, Vadasz C. Cannabi-noid CB1 receptor knockout mice exhibit markedly reduced voluntaryalcohol consumption and lack alcohol-induced dopamine release in thenucleus accumbens. J Neurochem 2003;84:698–704.

[125] Robinson DL, Heien ML, Wightman RM. Frequency of dopamine con-centration transients increases in dorsal and ventral striatum of malerats during introduction of conspecifics. J Neurosci 2002;22:10477–86.

[126] Phillips PE, Stuber GD, Heien ML, Wightman RM, Carelli RM. Subsec-ond dopamine release promotes cocaine seeking. Nature 2003;422:61–8.

[127] Roitman MF, Stuber GD, Phillips PE, Wightman RM, Carelli RM.Dopamine operates as a subsecond modulator of food seeking. J Neurosci2004;24:1265–71.

[128] Cheer JF, Wassum KM, Sombers LA, Heien ML, Ariansen JL, AragonaBJ, et al. Phasic dopamine release evoked by abused substances requirescannabinoid receptor activation. J Neurosci 2007;27:791–5.

[129] Giuffrida A, Parsons LH, Kerr TM, Rodriguez de Fonseca F, Navarro M,Piomelli D. Dopamine activation of endogenous cannabinoid signalingin dorsal striatum. Nat Neurosci 1999;2:358–63.

[130] Duarte C, Alonso R, Bichet N, Cohen C, Soubrie P, Thiebot MH. Blockadeby the cannabinoid CB1 receptor antagonist, rimonabant (SR141716), ofthe potentiation by quinelorane of food-primed reinstatement of food-seeking behavior. Neuropsychopharmacology 2004;29:911–20.

[131] Bespalov AY, Zvartau EE, Balster RL, Beardsley PM. Effects ofN-methyl-D-aspartate receptor antagonists on reinstatement of cocaine-seeking behavior by priming injections of cocaine or exposures tococaine-associated cues in rats. Behav Pharmacol 2000;11:37–44.

[132] Bespalov AY, Dravolina OA, Sukhanov I, Zakharova E, Blokhina E,Zvartau E, et al. Metabotropic glutamate receptor (mGluR5) antagonistMPEP attenuated cue- and schedule-induced reinstatement of nicotineself-administration behavior in rats. Neuropharmacology 2005;49:167–78.

esearch 56 (2007) 418–427 427

[133] Maejima T, Hashimoto K, Yoshida T, Aiba A, Kano M. Presynapticinhibition caused by retrograde signal from metabotropic glutamate tocannabinoid receptors. Neuron 2001;31:463–75.

[134] Maejima T, Ohno-Shosaku T, Kano M. Endogenous cannabinoid as a ret-rograde messenger from depolarized postsynaptic neurons to presynapticterminals. Neurosci Res 2001;40:205–10.

[135] Melis M, Pistis M, Perra S, Muntoni AL, Pillola G, Gessa GL.Endocannabinoids mediate presynaptic inhibition of glutamatergic trans-mission in rat ventral tegmental area dopamine neurons through activationof CB1 receptors. J Neurosci 2004;24:53–62.

[136] Szabo B, Siemes S, Wallmichrath I. Inhibition of GABAergic neurotrans-mission in the ventral tegmental area by cannabinoids. Eur J Neurosci2002;15:2057–61.

[137] Yamamoto T, Anggadiredja K, Hiranita T. New perspectives in thestudies on endocannabinoid and cannabis: a role for the endocannabinoid-arachidonic acid pathway in drug reward and long-lasting relapse to drugtaking. J Pharmacol Sci 2004;96:382–8.

[138] Reichman M, Nen W, Hokin LE. Delta 9-tetrahydrocannabinol increasesarachidonic acid levels in guinea pig cerebral cortex slices. Mol Pharma-col 1988;34:823–8.

[139] Hunter SA, Burstein SH. Receptor mediation in cannabinoid stimu-lated arachidonic acid mobilization and anandamide synthesis. Life Sci1997;60:1563–73.

[140] Hunter SA, Burstein S, Renzulli L. Effects of cannabinoids on the activ-ities of mouse brain lipases. Neurochem Res 1986;11:1273–88.

[141] Wartmann M, Campbell D, Subramanian A, Burstein SH, Davis RJ. TheMAP kinase signal transduction pathway is activated by the endogenouscannabinoid anandamide. FEBS Lett 1995;359:133–6.

[142] Bhattacharya SK. delta-9-tetrahydrocannabinol (THC) increases brainprostaglandins in the rat. Psychopharmacology 1986;90:499–502.

[143] Hunter SA, Audette CA, Burstein S. Elevation of brain prostaglandin E2

[144] Yamaguchi T, Shoyama Y, Watanabe S, Yamamoto T. Behavioral sup-pression induced by cannabinoids is due to activation of the arachidonicacid cascade in rats. Brain Res 2001;889:149–54.