Download pdf - An Endocannabinoid Hypothesis of Drug Reward and Drug Addiction

An Endocannabinoid Hypothesis of DrugReward and Drug Addiction

Emmanuel S. Onaivi

Department of Biology, William Paterson University, Wayne, New Jersey, USA

Pharmacologic treatment of drug and alcohol dependency has largely been disappoint-ing, and new therapeutic targets and hypotheses are needed. There is accumulatingevidence indicating a central role for the previously unknown but ubiquitous endo-cannabinoid physiological control system (EPCS) in the regulation of the rewardingeffects of abused substances. Thus an endocannabinoid hypothesis of drug reward ispostulated. Endocannabinoids mediate retrograde signaling in neuronal tissues andare involved in the regulation of synaptic transmission to suppress neurotransmitterrelease by the presynaptic cannabinoid receptors (CB-Rs). This powerful modulatoryaction on synaptic transmission has significant functional implications and interactionswith the effects of abused substances. Our data, along with those from other investiga-tors, provide strong new evidence for a role for EPCS modulation in the effects of drugsof abuse, and specifically for involvement of cannabinoid receptors in the neural basis ofaddiction. Cannabinoids and endocannabinoids appear to be involved in adding to therewarding effects of addictive substances, including, nicotine, opiates, alcohol, cocaine,and BDZs. The results suggest that the EPCS may be an important natural regulatorymechanism for drug reward and a target for the treatment of addictive disorders.

Key words: marijuana; capsaicin; endocannabinoids; cannabinoid receptors; dopa-mine; drug dependency; reward; addiction; withdrawal anxiogenesis

Introduction

In reviewing the history of human drugaddiction, one finds previous misconceptionsthat people addicted to drugs lacked willpowerand were morally weak. But we now knowthat drug addiction is a chronic relapsingbrain disease characterized by the compulsiveuse of addictive substances despite adverseconsequences to the individual and society.For more than 50 years, it has been assumedthat all drugs of abuse release dopamine in thebrain’s reward system to produce pleasure andeuphoria, consequently leading to addictionin vulnerable individuals.1,2 However, manyagents, such as inhalants, barbiturates, orbenzodiazepines, do not activate midbrain

Address for correspondence: Department of Biology, William PatersonUniversity, 300 Pompton Road, Wayne, NJ 07470. Voice: +973-720-3453;fax: +973-720-2338. [email protected]

dopamine-mediated transmission consistently,despite the fact that these drugs have rewardingproperties and are heavily abused.1 Thereforedopamine is not a simple marker of reward orhedonia and it might no longer be tenable tosuggest that drugs of abuse are simply activatingthe brain’s “natural reward system.”2 In thisarticle I propose that the dopamine hypothesisis another misconception. The dopamineprojections in the brain do not convey a spe-cific “reward” signal since dopamine releaseoccurs as a response not only to all drugsof abuse, but also to stress, foot shock, andaversive and salient stimuli.3,4 Mice thatcannot make dopamine (DD mice) have beenused to test the hypothesis that dopamine isnecessary for reward. The results show thatdopamine is not required for natural reward5

and morphine-induced reward.6 Thus, nu-merous problems are associated with thedopamine hypothesis of reward (Table 1). For

Ann. N.Y. Acad. Sci. 1139: 412–421 (2008). C© 2008 New York Academy of Sciences.doi: 10.1196/annals.1432.056 412

Onaivi: Endocannabinoids in Addiction 413

TABLE 1. Problems Associated with Dopamine Hypothesis of Reward: Beyond the Nucleus Accumbensand Dopamine Hypothesis of Reward

• Not all studies point to a unitary role for dopamine in one brain circuit as the most relevant system in drug abuse.• Dopamine may not be involved in brain reward mechanisms, as was previously thought.• Reward centers in the brain consist of multiple systems and neuroanatomic sites other than the mesoaccumbens

dopamine circuitry.• Dopamine-independent mechanisms involving other neurotransmitters like glutamate, GABA, serotonin,

endocannabinoids, stress hormones, and dynorphin are potential substrates for the rewarding effects of abusedsubstances.

• In schizophrenics, dopamine excess in mesoaccumbens causes heightened state of arousal but not pleasure.• Smokers and cocaine addicts continue to take hits long after the cigarettes become distasteful or after the effects

of cocaine have worn off.• Addictions arise from a complex pattern of pathogenetic and environmental situations.• Differential effects of abused substances on the complex network of genes, hormones, neurotransmitters, and

modulators do not support the concept of a single reward transmitter.• Manipulation of dopamine circuitry as a pharmacologic target should provide medication for substance abuse.• There is no causal relationship with regard to dopamine as a pleasure or reward transmitter triggered by abused

substances.• Activation of dopamine pathways is not involved in brain-stimulation reward of all brain sites relevant to

addiction.• Electrolytic lesions and 6-OH dopamine lesion studies of dopamine cell bodies in the ventral tegmental area and

other brain sites did not attenuate brain-stimulation reward.

example, self-administration of opiates andalcohol occurs even when the mesolimbicdopamine system is lesioned7 and interferencewith accumbens dopamine transmission doesnot substantially blunt the primary motivationfor natural rewards.2 Although we cannot un-derestimate the role of dopamine in the centralnervous system, recent studies in schizophre-nia, for which the dopamine hypothesis hasdominated treatment approaches, now showsthat the glutamate receptor offers promise fora new class of anti-psychotic agents (see 2007report in Nature Medicine at <http://ealerts.nature.com/cgi-bin24/DM/y/ef6D0SoYzX0HjT0Bbiv0EH>.)

The activation of the “natural reward sys-tem,” supposedly mediated by the accumbensdopamine, cannot reasonably be used as a gen-eral explanation for drug abuse and addiction.2

It is timely that there is accumulating evidenceindicating a central role for the endocannabi-noid physiological control system (EPCS) in theregulation of the rewarding effects of abusedsubstances. Such recent studies have shown thatthis endocannabinoid system is involved in thecommon neurobiological mechanism underly-

ing drug addiction9–12 (Table 2). Therefore,an endocannabinoid hypothesis of drug re-ward is postulated from data from our stud-ies and those of others. Endocannabinoids me-diate retrograde signaling in neuronal tissuesand are involved in the regulation of synaptictransmission to suppress neurotransmitter re-lease by the presynaptic cannabinoid receptors(CB-Rs). This powerful modulatory action onsynaptic transmission has significant functionalimplications and interactions with the effects ofabused substances. Additional support for theendocannabinoid hypothesis of drug reward isderived from an action of cannabinoids or mar-ijuana use on brain reward pathways that issimilar to that of other abused substances. Fur-thermore, administration of cannabinoids orthe use of marijuana exert numerous pharma-cologic effects through their interactions withvarious neurotransmitters and neuromodula-tors (Table 2). In our preliminary studies test-ing the endocannabinoid hypothesis of drug re-ward, we investigated the interaction betweenvanilloid and cannabinoid agonists and an-tagonists in the mouse model of aversion us-ing the elevated plus-maze test. The vanilloid

414 Annals of the New York Academy of Sciences

TABLE 2. Framework for an Endocannabinoid Hypothesis of Drug Reward

• An endocannabinoid physiological system control system (EPCS) exists and has a central role in the regulationof the rewarding effects of abused substances.

• The EPCS is intricately involved in almost all the biological processes of the human body and brain.• The EPCS appears to exert a powerful modulatory action on retrograde signaling associated with cannabinoid

inhibition of synaptic transmission.• The retrograde signaling appears to be involved in the modulation of neurotransmitter release by cannabinoids

and endocannabinoids.• The abundant distribution of the cannabinoid receptors in the brain provides the EPCS with limitless signaling

capabilities of crosstalk within, and possibly between, receptor families.• A missense mutation in human fatty acid amide hydrolase may be associated with problem drug use in

vulnerable individuals.• Cannabinoids induce alterations in brain disposition and pharmacologic actions of drugs of abuse.• There are changes in endocannabinoid contents in the brain of rats chronically exposed to nicotine, ethanol, or

cocaine.• “Runners high,” the sense of euphoric well-being that come with vigorous exercise, stimulates the release and

elevated levels of endocannabinoids• The mechanisms of dependence on different substances appear to be different in terms of their impact on the

EPCS.• The endocannabinoid transmission is a component of the brain reward system and appears to play a role in

dependence/withdrawal of abused substances.• CB1 knockout mice have a reduced sensitivity to reward. However, mice that cannot make dopamine (mice

lacking tyrosine hydroxylase) respond to rewarding stimuli, indicating reward without dopamine.• Overeating, alcohol, and sucrose consumption is decreased in CB1 receptor–deleted mice.• The endocannabinoid system is involved in the neural circuitry regulating alcohol consumption and motivation

to consume alcohol.• Evidence for the existence of a functional link between the cannabinoid and opioid receptor systems in the

control of alcohol intake and motivation to consume alcohol.• Alcohol self-administration is decreased and alcohol sensitivity and withdrawal in CB1 receptor knockout mice

is increased.• The endocannabinoid system modulates opioid rewarding and addictive effects by crosstalk between

endogenous opioid and endocannabinoid systems.• Involvement of endocannabinoid and glutamate neurotransmission in brain circuits is linked to reward and

mnemonic processes. Long-term potentiation (LTP) is abolished in mice lacking mGluR5 receptors andenhanced LTP and memory in mice lacking cannabinoid CB1 receptors.

• Memory-related plasticity may be a common mechanism in the endocannabinoid system in the control bycannabinoids of conditioned drug seeking.

agonist used is a natural product, capsaicin, theactive ingredient in hot chili peppers, whichis known to be habit forming. In a follow-upstudy, we determined the effect of the CB1 an-tagonist, rimonabant, on withdrawal aversionsfrom chronic treatment with abused drugs.Our results suggest that the endocannabi-noid physiological control system may be animportant natural regulatory mechanism forreward.

Endocannabinoid Physiological ControlSystem and Reward, Drug Abuse,

and Addictions

Many studies now show that the endo-cannabinoid system13,14 is involved as themajor player and most likely common neu-robiological mechanism underlying drug re-ward. There is substantial evidence supportinga role for the endocannabinoid system as


a modulator of dopaminergic activity in thebasal ganglia.15 The endocannabinoid sys-tem therefore participates in the primaryrewarding effects of alcohol, opioids, nico-tine, cocaine, amphetamine, cannabinoids, andbenzodiazepines through the release of endo-cannabinoids that act as retrograde messen-gers to inhibit classical transmitters, includ-ing dopamine, serotonin, GABA, glutamate,acetylcholine, and norepinephrine.13 Further-more the endocannabinoid system is intricatelyinvolved in the common mechanisms underly-ing relapse to drug-seeking behavior by medi-ating the motivational effects of drug-relatedenvironmental stimuli and drug re-exposure.11

Therefore substantial data now point to a rolefor the endocannabinoid system in trigger-ing and/or preventing reinstatement of drug-seeking behavior.12 It appears that the effectsof perturbation of the endocannabinoid sys-tem by drugs of abuse can be ameliorated byrestoring the perturbed system using cannabi-noid ligands. It is not surprising that prelimi-nary studies with cannabinoid antagonists areshowing promise in the reduction of drug use,in smoking cessation, and reduction in alco-hol consumption, and rimonabant has been ap-proved in Europe for treating obesity. It is hopedthat these encouraging positive results will leadto new therapeutic agents in the treatment ofdrug dependency. The promiscuous action anddistribution of cannabinoid receptors in mostrelevant biological systems provide the EPCSwith limitless signaling capabilities for crosstalkwithin, and possibly between, receptor fami-lies, which may explain the myriad behavioraleffects associated with smoking marijuana. TheEPCS therefore appears to play a central rolein regulating the neural substrate underlyingmany aspects of drug addiction, including crav-ing and relapse.8 The findings that the EPCS isinvolved in the reinstatement model providedevidence of the EPCS in the neural machineryunderlying relapse. Relapse, the resumption ofdrug taking after a period of drug abstinence, isconsidered the main hurdle in treating drug ad-

diction and pharmacologic modulation of theendocannabinoid tone with rimonabant gavepositive results in human trials. A summaryof data from recent studies of the efficacies ofcannabinoid antagonist in mutant mice has re-cently been reviewed.10 As the usefulness of thepharmacotherapy of substance abuse has beenlimited, there is now sufficient preclinical evi-dence for clinical trials to evaluate the efficacyof cannabinoid-based drugs in the treatment ofdrug dependency.

Interaction between CB1 and CB2Receptors in Drug Abuse

and Addiction

The expression of CB1 cannabinoid recep-tors (CB1-Rs) in the brain and its periphery hasbeen well studied, but the brain neuronal ex-pression of CB2-Rs had been ambiguous andcontroversial and its role in substance abuse isunknown. There is now ample evidence for thefunctional presence of CB2-Rs in mammalianbrain neurons.16–18 We have investigated theinvolvement of CB2-Rs in alcohol preferencein mice and alcoholism in humans.19 So we at-tempted to determine whether CB2-Rs in thebrain play a role in alcohol abuse/dependencein an animal model and then examined an as-sociation between the CB2 gene polymorphismand alcoholism in Japanese population. Thereis an association between the Q63R polymor-phism of the CB2 gene and alcoholism in theJapanese population. Our data therefore re-vealed that CB2-Rs are functionally expressedin brain neurons and play a role in substanceabuse and dependency.17–19 Now that we knowthat CB2-Rs are present in the brain, the nextquestion is: what is the nature of and con-tribution to the known effects of CB1-Rs inthe rewarding effects of drug abuse? In viewof the recent definitive demonstration of neu-ronal CB2-Rs in brain, one possible explana-tion may be that CB2-Rs and CB1-Rs workindependently and/or cooperatively in differ-ent neuronal populations to regulate a number


Figure 1. Interaction of cannabinoids and vanilloids in the mouse plus-maze test. (A)Acute effects of administration of vanilloid receptor agonist, capsaicin (0.1–1.0 mg/kg);CB-Ragonist, WIN55212-2 (0.1–2.5 mg/kg); vanilloid receptor antagonist, capsazepine (0.1–2.0 mg/kg); and CB1-R antagonist, rimonabant (0.1–3.0 mg/kg). All drugs were admin-istered i.p. for 30 min, except for capsaicin, which was administered for 10 min beforeplacement of the animals in the plus-maze for the 5-min test session. (B) Acute effects ofselected dose of WIN55212-2 (1.0 mg/kg) alone and capsaicin (0.5 mg/kg alone and incombination with WIN55212-2) on the performance of mice in the plus-maze test. For thecombination, mice received WIN55212-2 20 min prior to the administration of capsaicin andthen tested in the plus-maze 10 min after the administration of capsaicin. (C) Acute effects ofcapsaicin (0.5 mg/kg) in male and female mice of different strains (C57BL/6, BALBc, andDBA/2). Capsaicin was administered for 10 min and the mice were tested in the plus-mazefor 5 min. (D) The effect of pretreatment with capsazepine (2.0 mg/kg) and/or rimonabant(3.0 mg/kg) 30 min prior to the administration of capsaicin (0.5 mg/kg) or WIN55212-2(1.0 mg/kg) on mouse aversive behavior in the plus-maze test. Data in Figure 1A–D representtime spent in the open arms of the plus-maze ± SEM (N = 10 animals per group). Significantdifference from vehicle-treated controls is indicated as ∗ = P < 0.05 or + = P < 0.05; ANOVAwith Dunnett’s multiple comparison tests were done.


Figure 1. Continued.

Figure 2. Antagonism of withdrawal reactionfrom drug abuse treatment by CB1 receptor antago-nism, rimonabant, in the plus-maze test. Modificationof withdrawal aversions of mice from chronic cocaine(1.0 mg/kg) (C), diazepam (1.0 mg/kg) (D), ethanol(8% w/v) (E), and methanandamide (10 mg/kg)(A) by rimonabant (3 mg/kg) in the plus-maze test.(V=vehicle; SR=rimonabant)

of physiological activities influenced by drugsof abuse, cannabinoids, endocannabinoids, andmarijuana. Nevertheless, our studies and thoseof others demonstrate the functional expressionof CB2-Rs in brain that may provide novel tar-gets for the effects of cannabinoids in substanceabuse disorders beyond neuroimmunocannabi-noid activity.

Materials and Methods

The elevated plus-maze test is a validatedmethod of measuring the performance of ro-dents, which are nocturnal in nature, andobtaining an index of anxiety when controland experimental animals are exposed to the


maze. The plus-maze consists of two openarms and two enclosed arms linked by a cen-tral platform and arranged in a plus sign(+). The elevated plus-maze system was usedto study withdrawal anxiogenesis from se-lected drugs with abuse potential, including co-caine, diazepam, and ethanol. The syntheticmethanandamide was included in the study.Adult C57Bl/6 mice were evaluated in theelevated plus-maze test of anxiety followingabrupt cessation from chronic treatment withselected doses of cocaine 1.0 mg/kg, diazepam,1.0 mg/kg, ethanol, 8% w/v, and methanan-damide 10 mg/kg. In a separate group of thesemice, the ability of the CB1 cannabinoid an-tagonist, rimonabant (1.0–3.0 mg/kg), to blockthe withdrawal aversions of mice from alcoholand selected drugs with abuse potential wasdetermined. The in vivo pharmacologic interac-tion was evaluated by the co-administration ofcannabinoids and vanilloid ligands. Capsaicin,known to activate CB-Rs and vanilloid recep-tors (VR1-Rs) in the CNS, was used to study theinvolvement of the EPCS in the habit-formingproperties of capsaicin (0.1–2.0 mg/kg) in mice.Testing the interaction between the vanilloidand cannabinoid systems was performed usingselected agonists and antagonists. The ability ofthe antagonist drug to block agonist drug effectwas also evaluated in this paradigm.

Results

Interaction of Cannabinoids andVanilloids: A Biological Basis of WhySome Like it Hot and Others Do Not

We tested the hypothesis that the en-docannabinoid physiological control system(EPCS) may play a role in the habit-formingproperties of capsaicin and may also be in-volved in the rewarding effects of abused sub-stances. In the first set of studies, the crosstalkbetween cannabinoids and vanilloids was de-termined by evaluating the interactions of cap-saicin (0.1–2.0 mg/kg), a vanilloid receptor 1

agonist, and WIN5512-2 (0.1–2.5 mg/kg), acannabinoid agonist (Fig. 1A–D). The effects ofthe vanilloid receptor antagonist, capsazepine,and the cannabinoid antagonist, rimonabant,in the actions of capsaicin and WIN55212-2were determined (Fig. 1D). The aversive be-havior induced in mice after treatment withcapsaicin, which was dependent on gender andstrain of mice used, was enhanced by pretreat-ment with WIN55212-2 (Fig. 1B). While the30-min pretreatment with WIN55212-2 en-hanced aversions induced by capsaicin, cap-sazepine or rimonabant blocked the aversionsinduced by capsaicin and WIN55212-2. Theeffect of capsaicin (0.5 mg/kg) on the perfor-mance of mice in the plus-maze test was de-pendent on the gender and mouse strain, withC57BL/J male mice being more susceptible tothe behavioral aversions induced by capsaicin(Fig. 1C).

Antagonism of Withdrawal Reactionfrom Alcohol and Drug Abuse by CB1

Receptor Antagonism in thePlus-Maze Test

The effects of rimonabant on the withdrawalaversions from cessation with chronic treatmentof mice with selected addictive drugs are shownin Figure 2. Cannabinoid CB1 receptor antag-onism reduced behavioral aversions after with-drawal from alcohol, cocaine, and diazepam. Itis important to note that rimonabant, the CB1-R antagonist alone, produced variable effects,characterized by reduced aversion at low doseand increased aversions to the open arms of themaze at higher doses.

Discussion

The major goal of this study is to understandthe role of endocannabinoids (the marijuana-like substances produced by the human body) indrug reward and addiction. Cannabinoids, theconstituents of marijuana, mediate their effectsby acting on the endocannabinoids and their


Figure 3. A hypothetical illustration of the crosstalk between the VR1 ion channel recep-tor and the G-protein-coupled CB1-Rs at the cellular and molecular level. Top panel showsthe natural products chili peppers and the cannabis plant. Endocannabinoids released bycapsaicin or marijuana use as byproducts of phospholipid metabolism are able to activateboth receptors (lower panel).

receptors (CB-Rs), while capsaicin, the pungentchemical in hot chili peppers, a habit-formingfood substance, activate the vanilloid type 1(VR1) receptors to produce its effects. Mar-ijuana and capsaicin enhance appetite, taste,and perhaps addiction and habit-forming prop-erties through CB-Rs and VR1 receptors(Fig. 3). The data obtained reveal differentialsensitivities of capsaicin in mice, which weregender- and strain-dependent, which may in-dicate why some like hot chili peppers and oth-ers do not. Mice naturally do not like capsaicinand these data are supported by the finding thatgenetically modified mice lacking the VR1 re-ceptor no longer avoid spicy peppers.20 Thismay be due to the interaction between thecannabinoid and vanilloid systems: both cap-sazepine and rimonabant (antagonists at vanil-loid and cannabinoid receptors, respectively)blocked the aversions induced by WIN55212-2 and capsaicin, indicating crosstalk betweencannabinoid and vanilloid systems. Althoughthe interaction between the EPCS and thevanilloid system is not well established and stud-

ied, the results from the investigation whetherthe interaction between endocannabinoid andendovanilloid systems induced by the naturalligands could be a basis for why some like hotchili peppers and others do not are intriguing.Therefore it is tempting to speculate and extendthe known regulation of the neural substratesaltered by abused substances to an interactionor a crosstalk between the EPCS and the en-dovanilloid (VR1) system. The existence andinvolvement of the EPCS in this in vivo modelis presented as additional evidence that manip-ulating the EPCS could be exploited in reduc-ing the behavioral consequences of withdrawalfrom drug dependency. Finally we tested thehypothesis that the EPCS is an integral com-ponent of the reward circuit using in vivo tests.We determined the influence of CB-R antago-nism on withdrawal anxiogenesis from chronicalcohol, cocaine, and diazepam treatment in

vivo using the mouse plus-maze test. The CB1-R antagonist rimonabant blocked the behav-ioral aversions to the open arms of the plus-maze, which was precipitated from withdrawal


from cocaine, diazepam, and alcohol. Similarto the withdrawal aversions in the plus-mazetest, the conditioned place preference (CPP)paradigm has been used to investigate the re-warding effects of cannabinoids. Rimonabanthas been shown to counteract the CPP sup-ported by classical reinforcers including food,cocaine, and morphine.21 This is in agreementwith data that demonstrate the antagonistic ac-tivity of rimonabant against disruption of cog-nition or reward-enhancing properties of mor-phine, amphetamine, and cocaine,22 which wehave extended to ethanol and diazepam. Theblockade of the behavioral aversions by thecannabinoid antagonist after chronic admin-istration with alcohol, cocaine, and diazepamis in agreement with data obtained duringcannabinoid-induced alterations in brain dis-positions of drugs of abuse that correlated withbehavioral alterations in mice.23 These results,taken together, suggest that in the CNS theEPCS may be a directly important natural reg-ulatory mechanism for reward in the brain andmay also contribute to reduction in aversiveconsequences of abused substances. Thus, thedata presented along with those of others sup-port not only the existence of the EPCS, butalso an endocannabinoid hypothesis of drugreward and addiction that may go beyond thedopamine hypothesis of drug reward.

Conclusions

The preliminary data on the association be-tween activation of cannabinoid and vanilloidreceptors show that the interaction of abusedsubstances with the endocannabinoid systemis pivotal in habit forming and a neural basisof drug abuse and dependency. Cannabinoidstherefore appear to be involved in adding tothe rewarding effects of addictive substances in-cluding, cocaine, alcohol, and benzodiazepines.These results, taken together, suggest that in theCNS the EPCS may be a directly importantnatural regulatory mechanism for reward in thebrain and also contribute to reduction in aver-

sive consequences of abused substances. Theexistence and involvement of an endocannabi-noid physiological control system in this in vivo

model is presented as additional evidence thatmanipulating the EPCS could be exploited inreducing the behavioral consequences of with-drawal from alcohol and drug dependency. Itis a good thing that controversy is one of thefuels of science. Thus, there is a lot more re-search to be done to better understand the na-ture and neurobiology of the endocannabinoidphysiological control system in health and dis-ease. In the end, the source of eternal bliss maynot be dopamine but endocannabinoids—thebrain and body’s marijuana and beyond.13

Acknowledgments

This work would not have been possiblewithout the support in part by the WilliamPaterson University’s Center for Research, theDean’s (Dr. DeYoung) Student Worker Fundfor maintenance of research animals, and theProvost’s Office for release time for research.I thank my colleagues in the Biology De-partment, and especially the Chairperson, Dr.Gardner, and Norman Schanz in the mouselaboratory. I acknowledge the assistance andsupport of Dr. Patricia Tagliaferro and mycollaborators, including Drs. Hiroki Ishiguro,B. Emmanuel Akinshola, and Sanika Chirwa.I am grateful for the technical assistance ofpast and present students working with me, es-pecially Zoila Mora, Alex Perchuk, DanielleColas, and Chiara Brandoni for direct in-volvement with some part of these studies. Iwould also like to thank Drs. John McPart-land, Richard Rothman, Mike Baumann, andWilliam Freed for constructive comments. Iam indebted to Dr. George Uhl, the Chiefof Molecular Neurobiology Branch at NIDA-NIH, where I am a Guest Scientist.

Conflicts of Interest

The author declares no conflicts of interest.


References

1. Spanagel, R. & F. Weiss. 1999. The dopamine hy-pothesis of reward: past and current status. Trends

Neurosci. 22: 521–527.2. Salamone, J.D. et al. 2005. Beyond the reward hy-

pothesis: alternative functions of nucleus accumbensdopamine. Curr. Opin. Pharmacol. 5: 34–41.

3. Horvitz, J.C. 2000. Mesolimbocortical and nigros-triatal dopamine responses to salient non-rewardevents. Neuroscience 96: 651–656.

4. Roll, E.T. 2005. Emotion Explained. Oxford UniversityPress. New York.

5. Cannon, M.C. & R.D. Palmiter. 2003. Reward with-out dopamine. J. Neuroscience 23: 10827–10831.

6. Hnasko, T.S. et al. 2005. Morphine reward indopamine-deficient mice. Nature 438: 854–857.

7. Koob, G.F. & M. Le Moal. 1997. Drug abuse: hedo-nic homeostatic dysregulation. Science 278: 52–58.

8. Yamamoto T. et al. 2004. New perspectives in thestudies on endocannabinoid and cannabis: a role forthe endocannabinoid-arachidonic acid pathway indrug reward and long-lasting relapse to drug taking.J. Pharmacol. Sci. 96: 382–388.

9. De Vries, T.J. & A.N.M. Schoffelmeer. 2005.Cannabinoid CB1 receptors control conditioneddrug seeking. Trends Pharmacol. Sci. 26: 420–426.

10. Parolaro, D. et al. 2005. Endocannabinoid and drugdependence. Curr. Drug Target CNS Neurol. Disorders 4:643–655.

11. Maldonado, F. et al. 2006. Involvement of the endo-cannabinoid system in drug addiction. Trends Neurosci.

29: 225–232.12. Fattore, L. et al. 2007. An endocannabinoid mecha-

nism in relapse to drug seeking: a review of animalstudies and clinical perspectives. Brain Res. Rev. 53:1–16.

13. Onaivi, E.S., T. Sugiura, V. Di Marzo, Eds. 2006. The

Brain and Body’s Marijuana and Beyond. CRC Taylor andFrancis. Boca Raton, FL.

14. Di Marzo, V. 2005. Cannabinoids. Handbook of Ex-

perimental Pharmacology. Pertwee, Ed. Springer-Verlag.Heidelberg.

15. Giuffrida, A. & D. Piomelli. 2000. The endocannabi-noid system: a physiological perspective on its role inpsychomotor control. Chem. Phys. Lipids. 108: 151–158.

16. Van Sickle, M.D. et al. 2005. Identification and func-tional characterization of brainstem cannabinoidCB2 receptors. Science 310: 329–332.

17. Gong, J.P. et al. 2006. Cannabinoid CB2 receptors:immunohistochemical localization in rat brain. Brain

Res. 1071: 10–23.18. Onaivi, E.S. 2006. Neuropsychobiological evidence

for the functional presence and expression ofcannabinoid CB2 receptors in the brain. Neuropsy-

chobiology 54: 231–246.19. Ishiguro, H. et al. 2006. Involvement of cannabinoid

CB2 receptor in alcohol preference in mice and alco-holism in humans. Pharmacogenomics J. E-pub aheadof print.

20. Caterina, M.J. et al. 2000. Impaired nociception andpain sensation in mice lacking the capsaicin receptor.Science 288: 306–313.

21. Chaperon, F. et al. 1999. Involvement of centralcannabinoid (CB1) in the establishment of place con-ditioning in rats. Psychopharmacology 135: 324–332.

22. Poncelet, M. et al. 1999. Blockade of CB1 receptorsby 141716 selectively antagonizes drug-induced re-instatement of exploratory behavior in gerbils. Psy-

chopharmacology 144: 144–150.23. Reid, M.J. & L.M. Bornheim. 2001. Cannabinoid-

induced alterations in brain disposition of drugs ofabuse. Biochem. Pharmacol. 61: 1357–1367.