Disease Focus

Editors Note: Disease Focus articles provide brief overviews of a neural disease or syndrome, emphasizing potential links to basicneuralmechanisms.They arepresented in thehopeofhelping researchers identify clinical implicationsof their research. Formoreinformation, see http://www.jneurosci.org/misc/ifa_minireviews.dtl.

Drug-Evoked Synaptic Plasticity Causing Addictive Behavior

Christian LuscherDepartments of Basic and Clinical Neurosciences, University Hospital and University of Geneva, 1211 Geneva, Switzerland

IntroductionThe clinical manifestations of drug addic-tion are the subject of medical literatureand fiction alike. In Novel with Cocaine,first published almost 80 years ago under apseudonym and only recently rediscov-ered, Mark Levi provides a disturbing ac-count of a young Russian man becomingaddicted to cocaine. Vadim, the novelsantihero reflects: Before I came in con-tact with cocaine I assumed that happi-ness was an entity, while in fact all humanhappiness consists of a clever fusion oftwo elements: the physical feeling of hap-piness, and the external event providingthe psychic impetus for that feeling.

This statement eloquently describes acore element of the addiction process. Aninitially neutral stimulus becomes attrac-tive when associated with drug consump-tion, and even after prolonged periods ofabstinence this cue may trigger cravingand cause the subject to relapse (Stewart etal., 1984; Childress et al., 1988). The riskfor relapse remains high after years of ab-stinence, which constitutes a major chal-lenge for treating drug addiction.

Therefore, many researchers have arguedthat the secret to understanding addiction liesin the elucidation of the memory trace thatlinks the cue to the compulsive drug use. The

implicit underlying hypothesis is that addic-tive drugs generate an inappropriate learningsignal that leads to the encoding of a uniquetrace, which, when reactivated, has a strongbehavioral impact(Redishetal.,2008;Schultz,2011).

Here we will review emerging evidencethat the cellular correlates of the drugmemory trace are the various forms ofsynaptic plasticity, mainly of glutamater-gic transmission, in the circuits of themesolimbic system, which we have previ-ously called drug-evoked synaptic plas-ticity (Luscher and Malenka, 2011). Wewill review the literature reporting drug-evoked synaptic plasticity in animal mod-els of addiction and argue for a stagedremodeling of the mesolimbic circuitry(Kalivas and OBrien, 2008) that is even-tually responsible for relapse and compul-sive drug use.

Animal modelsThe study of the neurobiological mecha-nisms underlying addictive behavior re-quires animal models. While the basis ofbehavioral pharmacology was laid out us-ing nonhuman primates and rats, morerecently genetically modified mice havebecome extremely useful tools. There isgeneral agreement that no animal modelfully reflects the complexity of the humandisease. Many behavioral studies, however,reproduce core components of addiction,startingwith the self-administration (SA) ofthe substance (Clark et al., 1961). De-pending on the substance involved, SAcan be oral (e.g., ethanol or benzodiaz-epines; Licata and Rowlett, 2008; Venge-liene et al., 2009), intravenous (e.g.,cocaine through a jugular catheter in ratsor mice; Gerber and Wise, 1989; Jackson

et al., 1998), or even directly into a specificbrain region [e.g., morphine into the ven-tral tegmental area (VTA); Bozarth andWise, 1981]. While SA is a necessary con-dition to demonstrate the reinforcing na-ture of a given substance, it is by nomeanssufficient (Collins et al., 1984). SA trans-lates well to clinical reports of drug liking;however, used in its most basic form, itdoes not fully mimic core features of ad-diction, including loss of control or re-lapse after periods of extendedwithdrawal(OConnor et al., 2011). What are neededfor mechanistic investigations of the syn-aptic drug trace are tests that demonstratedrug-adaptive behavior associated withan external stimulus. Locomotor sensiti-zation, often used as a proxy of incentivesaliency, refers to the enhanced locomo-tor effect of cocaine and other psycho-stimulants, which can be observed alreadyat the second injection of the drug (Rob-inson and Berridge, 2001). Locomotorsensitization also has an associative com-ponent, as its magnitude depends on theenvironment (the effect is enhanced in anovel environment; Badiani et al., 1995).Importantly, locomotor sensitization isalready observed several days after a singleinjection (Valjent et al., 2010) and manyweeks after repeated injections. A directmeasure of the association with the envi-ronment can be obtained in the condi-tioned place preference (CPP) test. In thisparadigm, using an apparatus with twodistinct compartments, one side is pairedwith the drug injection, the other with sa-line. Already after a first session, a prefer-ence for the compartment where the drugwas received can be measured. Both loco-motor sensitization and CPP, however,have the disadvantage that the drug is

Received Aug. 7, 2013; revised Sept. 13, 2013; accepted Sept. 19, 2013.My laboratory is supported by the Swiss National Science Foundation,

the National Competence Center Synapsy, and European Research Coun-cil Advanced Grant MeSSI. I thank Meaghan Creed, Kelly Tan, and EoinOConnor for many helpful comments on the manuscript. I apologize to allcolleagues whose work I was unable to cite due to space limitations.

Correspondence shouldbeaddressed toChristian Luscher, Departmentsof Basic and Clinical Neurosciences, University Hospital and University ofGeneva, 1211 Geneva, Switzerland. E-mail: christian.luscher@unige.ch.

DOI:10.1523/JNEUROSCI.3406-13.2013Copyright 2013 the authors 0270-6474/13/3317641-06$15.00/0

The Journal of Neuroscience, November 6, 2013 33(45):1764117646 17641

injected by the experimenter, whichstrongly contrasts with the clinical reality.A more accurate model therefore is cue-induced cocaine seeking (De Wit andStewart, 1981; Shaham et al., 2003), wherelever pressing or nose poking for a druginjection is paired with a cue (light orsound). After acquisition of SA, the sim-ple presentation of the cue will prompt aseeking behavior even though the drug isno longer available. This effect can be ob-served up to several weeks after the last SAsession in rats (Kruzich et al., 2001) andmice (Highfield et al., 2002).One interest-ing feature is that the seeking behavior be-comes stronger over the first few weeks ofwithdrawal, an observation called incu-bation of craving (Grimm et al., 2001).

Dopamine hypothesisWhat is the pharmacological effect defin-ing addictive drugs that is eventually re-sponsible for the adaptive behavior? Theyall increase mesolimbic dopamine (DA)levels, which mediate their reinforcing ef-fect (Wise, 1987; Di Chiara and Imperato,1988). While this statement is supportedby many experiments, particularly earlybehavioral pharmacology studies, it hasbeen challenged, most prominently by ge-netic manipulations. For example, eventhough cocaine increases mesolimbic DAby inhibiting the DA transporter (DAT);mice lacking the DAT still readily self-administer cocaine (Rocha et al., 1998).Arguably, this observation is incompatiblewith the DA hypothesis. However, subse-quent work during the last decade has dem-onstrated that in these DAT-lacking mice,abnormally high levels of DA during devel-opment induce adaptations in themesolim-bic DA system. Consequently, the results ofcocaine self-administration in thesemice re-quire reinterpretation. Rather than a com-plete knockout, amore subtlemanipulationconsisting of point mutations renders theDAT insensitive to cocaine but induces onlyminimal effects on DA reuptake (Chen etal., 2006). With the cocaine unable to bindthis mutated DAT, these knock-in mice nolonger self-administered cocaine. This ob-servation supports the hypothesis that DAincreases are necessary for the reinforcingeffects of cocaine. As a side note, a simi-lar knock-in approach has identified theGABAA receptor subtype containing the 1subunit inVTAGABAneurons as the initialtarget responsible for the addictive proper-ties of benzodiazepines (Tan et al., 2010).Amplification of GABAA receptors in VTAGABA neurons leads to the disinhibition ofDA neurons, a cellular mechanism sharedwith opioids, -hydroxybutyrate, and can-

nabinoids [Fig. 1 (modified from Luscherand Ungless, 2006)]. By contrast, nicotinecandirectly stimulateDAneurons, andpsy-chostimulants exert their effect by interfer-ing with DA reuptake, as discussed above.The DA hypothesis, therefore, is still rele-vant and has received evenmore direct sup-port from optogenetic manipulations ofVTA DA neurons, which confirmed thattheir activation is sufficient to support self-administration and conditioned place pref-erence (Tsai et al., 2009; Adamantidis et al.,2011; Lammel et al., 2012). Whether DAneuron activity is also sufficient to drivelong-lasting adaptive behavior remains tobe explored.

Drug-evoked synaptic plasticityWhile the requirement for increased DAmay explain the acute reinforcing effectsof addictive drugs, it does not provide anexplanation for craving and relapse. Afterall, these key features of addiction mani-fest long after the drugs have been clearedfrom the brain. Given the establishedmodulatory role of DA on glutamatergicand GABAergic transmission, the synaptictrace of addictive drugs may actually be ex-pressedat these synapses andnotnecessarilyinvolve a change inDAmodulation.The ex-perimental demonstration of such a tracewas achieved over a decade ago using aclever ex vivo approach. Briefly, slices of theVTA were prepared 1 d following a singleinjection of cocaine and glutamatergic

transmission at AMPA receptor (AMPAR)and NMDA receptor (NMDAR) was re-corded in VTA DA neurons. Since thenumber of synapses stimulated in thispreparation is beyond the control of theexperimenter, the ratio of AMPAR- andNMDAR-mediated synaptic currents wascalculated and found to be increased afterthe cocaine treatment (Ungless et al., 2001).

Subsequent investigations revealedthat this early drug trace can be observedwith all addictive drugs tested to date, in-cluding ethanol, morphine, and nicotine,but also amphetamines and benzodia-zepines (Saal et al., 2003; Das et al., 2008;Heikkinen et al., 2009). The plasticity wasalso observed after a 2-h-long optogeneticstimulation of the VTA DA neurons,mimicking the increased firing typicallyobserved with a single dose of morphineor nicotine (Brown et al., 2010). This in-creased AMPA/NMDA ratio is due to anenhanced AMPA transmission in conjunc-tion of a decreased NMDA transmission.The molecular expression mechanismtherefore involves a redistribution ofboth AMPARs and NMDARs. While theabsolute number of AMPARs remainsunchanged (Mameli et al., 2007), theGluA2-containing receptors at baselineare exchanged for GluA2-lacking recep-tors (Bellone and Luscher, 2006; Argilli etal., 2008).

At the same time, GluN3A-containingNMDARs appear in exchange forGluN2A

Figure 1. The mesocorticolimbic dopamine system as an initial target of addictive drugs. The VTA, at the origin of the meso-corticolimbic system, is composed of dopamine projection neurons that are under inhibitory control of GABA interneurons(somtimesdescribedas an independentnucleus at the tail of theVTA, the rostromedial tegmentum). Themain targets are theNAcand the mPFC. Addictive drugs cause an increase in mesocorticolimbic dopamine through three distinct cellular mechanisms:direct activation of dopamine neurons (e.g., nicotine); indirect disinhibition of dopamine neurons [opioids, gamma-hydroxybutyric acid (GHB), cannabinoids, and benzodiazepines]; as well as interference with dopamine reuptake (cocaine, ec-stasy, and amphetamines). Note that the latter group also increases dopamine in the VTA itself, owing to the perturbation of thereuptake of dendritically released dopamine (modified from Luscher and Malenka, 2011, with permission).

17642 J. Neurosci., November 6, 2013 33(45):1764117646 Luscher Drug-Evoked Synaptic Plasticity

containing NMDARs, again leaving thetotal number of NMDARs unchanged(Yuan et al., 2013). After a single injection,this plasticity persists for 1 week, afterwhich transmission is normalized. Thisnormalization requires the presence ofmetabotropicglutamatereceptor1(mGluR1)receptors,whichare locatedperisynapticallyand are most efficiently activated with ashort burst of activity. mGluR1 activationengages mammalian target of rapamycin-dependent local protein synthesis (Mameliet al., 2007). Interestingly, mGluR1-driven reinsertion of GluA2-containingAMPARshas alsobeenobservedat synapsesinmultiple brain areas (Clem andHuganir,2010; McCutcheon et al., 2011), suggestinga conserved mechanism for neurons toeliminateGluA2-lackingAMPARs fromthesynapse.

The case for a permissive metaplasticityThe complexity of the expression mecha-nisms of drug-evoked synaptic plasticityin the VTA indicates that the effect ofaddictive drugs on the mesolimbic circuitis not a straightforward potentiation of

transmission, but suggests a shift of syn-aptic calcium signaling fromNMDARs toAMPARs (Creed and Luscher, 2013). Innaive mice, glutamatergic transmissioncan cause synaptic entry of calcium, pro-vided the DA cell is depolarized, whichrelieves the NMDARs from the magne-sium block. In contrast, after cocainetreatment, GluN3A-containingNMDARshave a very low permeability for divalentcations and are inefficient conductors ofcalcium. In contrast, the GluA2-lackingAMPARs readily conduct calcium. In fact,owing to the inward rectification, thegreater the hyperpolarization, the morecalcium enters the cell. In other words, innaive animals, the postsynaptic neuronsmust be depolarized to engage calciumsignaling, whereas after cocaine adminis-tration, hyperpolarization is required. Asa consequence, the rules for activity-dependent potentiation of synaptic trans-mission are inverted (Mameli et al., 2011).This can be experimentally demonstratedby using a spike timing-dependent potenti-ation (STDP) protocol, where the postsyn-aptic neuron receives either a depolarizing

or a hyperpolarizing current injection (Fig.2). In slices from cocaine-treated animals,depolarizing STDP was no longer able topotentiate transmission; however, the po-tentiationcouldbe rescuedby switching toahyperpolarizing STDP. Together, thesefindings strongly suggest a permissive rolefor this early form of drug-evoked synapticplasticity. In this sense, the AMPA andNMDA receptor alterations inducedby drug exposure can be considered ametaplasticity, enabling specific formsof subsequent activity-dependent syn-aptic plasticity.

Causal role of drug-evoked synapticplasticity in behaviorEstablishing a causal relationship betweendrug-evoked synaptic plasticity and drug-adaptive behavior has been challenging, inpart because of the temporal discrepancybetween these two phenomena. A singleinjection already leaves a synaptic trace(discussed above) that lasts for 1 weekbut may not drive many behavioral adap-tations. Interestingly, a recent study usingretrograde tracing techniques has provided

Figure 2. Drug-evoked synaptic plasticity in dopamine neurons of the ventral tegmental area. Addictive drugs or strong stimulation of dopamine neurons causes a synaptic plasticity on theexcitatory afferents. This plasticity is expressed by the dual exchange of AMPA and NMDA receptors: GluA2-lacking AMPA receptors and GluN3A-containing NMDA receptors are inserted. As aconsequence calcium enters the postsynaptic cell more readily when themembrane is hyperpolarized, which inverts the rules for activity-dependent synaptic plasticity. At baseline, LTP is inducedwhenglutamate is released onto depolarized dopamine neurons (Hebbian LTP that is NMDARdependent),while after cocaine exposure LTP can be observedwhenglutamate transmission coincideswith a hyperpolarization (anti-Hebbian LTP that is AMPAR dependent). Normal transmission is restored by the activation of mGluR1. Together, addictive drugs evoked a permissive metaplasticityat excitatory afferents onto VTA dopamine neurons, which gate subsequent adaptations downstream (see text).

Luscher Drug-Evoked Synaptic Plasticity J. Neurosci., November 6, 2013 33(45):1764117646 17643

information that the VTADA neurons thatproject to the nucleus accumbens (NAc) areparticularly prone to undergo drug-evokedsynaptic plasticity (Lammel et al., 2012).These early forms of drug-evoked plasticityaremerely buildingblocks of circuit remod-eling,which, after repetitive exposure, even-tually lead to behavioral alterations. In linewith this interpretation, in mice whereNMDA receptors were selectively abol-ished in DA neurons (DAT-Cre ERT2GluN1flox; Zweifel et al., 2008), the behav-ioral repercussions of drug exposure areobserved only after a delay and followingseveral injections. Specifically, reinstate-ment of drug seeking was reduced, whichinvolves an extended circuitry including theNAc (Engblom et al., 2008).

Further evidence for a stepwise remod-eling of mesolimbic circuitry comes fromexperiments suggesting a hierarchical or-ganization of drug-evoked synaptic plas-ticity between the VTA and the NAc.Quick and efficient reversal of cocaine-evoked plasticity in the VTA by local orsystemic application of a positive alloste-ric modulator of the mGluR1 preventedsome drug-evoked synaptic adaptationsin the NAc (Mameli et al., 2009). Lesionexperiments and pharmacologicalmanip-ulations implicate connections to andfrom the NAc in the integration of re-warding drug effects and contextual envi-ronmental cues (Sesack and Grace, 2010).The crucial plasticity that presumablymediates this integration occurs at theexcitatory afferents onto medium spinyneurons (MSNs), which represent 95% ofneurons in the NAc and are an interfacebetween limbic and motor functions(Mogenson et al., 1980). There are twopopulations ofMSNs that are divided intotwo equal-sized classes depending on thetype of dopamine receptors (eitherD1R orD2R) they express and their projectionsites (Gong et al., 2007). Both D1R MSNsand D2R MSNs receive excitatory afferentsfrom the medial prefrontal cortex (mPFC),the ventral hippocampus (vHippo), and thebasolateral amygdala (BLA; Papp et al.,2012). Previous studies have attributed theemotional valence to the BLA input (Cadoret al., 1989), which, when selectively stimu-lated, is reinforcing (Stuber et al., 2011).Theactivation of the mPFC-to-NAc projectionhas been implicated in the initiation of ac-tion and the actual seeking behavior, andthere is evidence linking the hypoactivity ofcortical projection neurons to compulsivecocaine consumption (Chen et al., 2013).Finally, the afferents from the vHippowould code for the context (Sesack andGrace, 2010). Regardless of the origin, syn-

aptic plasticity at glutamatergic transmis-sion onto MSNs has been correlated withcue-induced relapse to cocaine seeking(Conrad et al., 2008). Given the proposedfunctions of these specific inputs, cocainemay evoke distinct forms of plasticity at ex-citatory synapses onto MSNs, which wouldcontribute to specific components of adap-tive behavior. What will be needed toaddress this hypothesis is the selective ma-nipulations at identified synapses (Fig. 3).With the advance of cell type-specific opto-genetic controlofneuronal activity, suchex-periments are now possible. The blueprintfor this approach consists of exposing theanimal to the drug, such that a drug-adaptive behavior can be observed (e.g., re-petitive injections of cocaine to causelocomotor sensitization), and characteriz-ing drug-evoked synaptic plasticity ex vivo(e.g., potentiation of mPFC afferents ontoD1R MSNs). Since most forms of plasticitycanbe reversedby an appropriate stimulationprotocol, the next step consists of finding aprotocol that can normalize transmission invitro(e.g.,1HzNMDAR-dependentdepoten-tiation). This protocol can then be applied invivo after transfecting the appropriate up-streamneurons andactivating the afferent ax-onsby terminal illumination (e.g., injectionofoptogenetic effector channelrhodopsin-2 intothe mPFC shining light into the NAc) to testthe effect of depotentiation on drug-adaptive behavior (e.g., erasure of sensitiza-tion). A proof of principle study (Pascoli etal., 2011) has shown that this approach isfeasible. Another example using this ap-proach links incubation of craving to thepotentiation of excitatory transmissionfrom the amygdala to the NAc (Lee et al.2013). Yet further experiments must beperformed to understand the role ofdrug-evoked synaptic plasticity at iden-

tified synapses in other behavior (e.g.,cue-induced cocaine seeking modelingrelapse).

The challenge ofindividual vulnerabilityOne of the striking features of addiction isonly poorly understood: the individualvulnerability. At equal levels of drug con-sumption, some people readily lose con-trol, while others maintain a controlledrecreational use formany years. Just thinkof alcohol, a drug that is legal in manycountries that the overwhelming majorityof people can enjoy recreationally duringan entire adult lifetime without ever los-ing control. This is also well documentedfor nicotine and cannabis, and even holdstrue for harder drugs such as amphet-amines and cocaine, the latter leading toaddiction in20% of regular users over aspan of10 years (Wagner and Anthony,2002). While several genetic variationshave been associatedwith drug use (Gold-man et al., 2005), a causal relationship re-mains to be established. Moreover, moststudies described above focus on genericmechanisms that underlie the behavioralchanges, and little is known about theneurobiology underlying individual vul-nerability. One requirement is certain; in-dividual vulnerability can be observedonly after prolonged drug exposure. Thisrequirement is not met in most animalsmodels presented above, which is whybehavioral pharmacology screens showvery little variability. However, pioneer-ing work using SA for several months inrats have demonstrated that animal-to-animal variability can also be observed inrodents (Deroche-Gamonet et al., 2004;Vanderschuren and Everitt, 2004). In asubsequent study, it has been argued that

Cocaine exposureto induce drug-adaptive

behavior

Characterize drug-evoked synaptic plasticity ex vivo

Design reversal protocol in vitro

Apply reversal protocol in vivo

Validate normalizationof synaptic

transmission ex vivo

LTD induced by 1Hzstimulation

Optogenetic stimulation of NAc afferents (1 Hz, 10 min)

Potentiation of excitatorytransmission onto D1R-MSNs

Locomotor sensitization

Depotentiation of

Figure 3. Experimental blueprint to establish causal relationship between drug-evoked synaptic plasticity and drug-adaptivebehavior. In gray example experiments are from a proof-of-principle study (Pascoli et al., 2011). For further explanation see text.

17644 J. Neurosci., November 6, 2013 33(45):1764117646 Luscher Drug-Evoked Synaptic Plasticity

high impulsivity predicts the develop-ment of addiction-like behavior in rats(Belin and Everitt, 2008).

Emerging ideas for rationaladdiction treatmentsA more comprehensive understanding ofthe precise nature of circuit remodelingcaused by addictive drugs resulting fromspecific formsof drug-evoked synaptic plas-ticity may help us to design novel rationaltreatments for the human disease. Suchtreatments may be pharmacological in na-ture or use neuromodulatory approachessuch as deep brain stimulation or transcra-nial magnetic stimulation (TMS). The for-mer may target specific types of receptorsknown to appear only after drug exposure,such as the GluA2-lacking AMPARs orthe GluN3-containingNMDARs, but thesetypes of pharmacological interventionsare really only in their infancy. It may alsobe possible to exploit and enhance thephysiological mechanism of reversingdrug-evoked synaptic plasticity.One area offocus may be on the mGluR1 receptors,which can reverse the redistribution of AM-PARs and NMDARs.

Perhaps a more selective interventionwould be to target specific areas of the re-ward circuitry, such as the NAc, with deepbrain stimulation. However, novel proto-cols would have to be developed, as cur-rent high-frequency stimulation (100Hz) is unlikely to affect plasticity. TMSmay have the appeal of being noninvasive.If we succeed in determining the corticalafferents undergoing drug-evoked synap-tic plasticity to develop efficient reversalprotocols, TMS may emerge as a promis-ing treatment option.

Ultimately, the understanding of drug-induced plasticity at specific synapsesmust be furthered to develop a therapy thatreverses plasticity and associated drug-induced adaptive behavior. The goal is toprovide a cure for addiction and to proveAaronSorkinwrongwhenherecounted ina2012 interview with the W magazine: . . . the hardest thing I do every day is nottake cocaine. You dont get cured of addic-tionyoure just in remission.

ReferencesAdamantidis AR, Tsai HC, Boutrel B, Zhang F,

Stuber GD, Budygin EA, Tourino C, Bonci A,DeisserothK, de Lecea L (2011) Optogeneticinterrogation of dopaminergic modulation ofthe multiple phases of reward-seeking behav-ior. J Neurosci 31:1082910835. CrossRefMedline

Argilli E, Sibley DR, Malenka RC, England PM,Bonci A (2008) Mechanism and time courseof cocaine-induced long-term potentiation in

the ventral tegmental area. J Neurosci 28:90929100. CrossRef Medline

Badiani A, Anagnostaras SG, Robinson TE (1995)The development of sensitization to the psy-chomotor stimulant effects of amphetamineis enhanced in a novel environment. Psy-chopharmacology (Berl) 117:443452.CrossRef Medline

Belin D, Everitt BJ (2008) Cocaine seeking hab-its depend upon dopamine-dependent serialconnectivity linking the ventral with the dor-sal striatum. Neuron 57:432441. CrossRefMedline

Bellone C, Luscher C (2006) Cocaine triggeredAMPA receptor redistribution is reversed invivo bymGluR-dependent long-term depres-sion. Nat Neurosci 9:636641. CrossRefMedline

Bozarth MA, Wise RA (1981) Intracranial self-administration of morphine into the ventraltegmental area in rats. Life Sci 28:551555.CrossRef Medline

Brown MT, Bellone C, Mameli M, Laboue`be G,Bocklisch C, Balland B, Dahan L, Lujan R, De-isseroth K, Luscher C (2010) Drug-drivenAMPA receptor redistribution mimicked byselective dopamine neuron stimulation. PLoSOne 5:e15870. CrossRef Medline

CadorM, Robbins TW, Everitt BJ (1989) Involve-ment of the amygdala in stimulus-reward asso-ciations: interaction with the ventral striatum.Neuroscience 30:7786. Medline

Chen BT, Yau HJ, Hatch C, Kusumoto-Yoshida I,Cho SL, Hopf FW, Bonci A (2013) Rescuingcocaine-induced prefrontal cortex hypoactiv-ity prevents compulsive cocaine seeking. Na-ture 496:359362. CrossRef Medline

ChenR,TilleyMR,WeiH,ZhouF,ZhouFM,ChingS,QuanN, StephensRL,Hill ER,Nottoli T,HanDD, GuHH (2006) Abolished cocaine rewardin mice with a cocaine-insensitive dopaminetransporter. ProcNatlAcadSciU S A103:93339338. CrossRefMedline

Childress AR,McLellan AT, Ehrman R, OBrien CP(1988) Classically conditioned responses inopioid and cocaine dependence: a role in re-lapse. NIDA ResMonogr 84:2543. Medline

Clark R, Schuster CR, Brady JV (1961) Instru-mental conditioning of jugular self-infusionin the rhesus monkey. Science 133:18291830. CrossRef Medline

Clem RL, Huganir RL (2010) Calcium-permeableAMPAreceptor dynamicsmediate fearmemoryerasure. Science 330:11081112. CrossRefMedline

CollinsRJ,Weeks JR,CooperMM,GoodPI, RussellRR (1984) Prediction of abuse liability ofdrugs using IV self-administration by rats. Psy-chopharmacology (Berl) 82:613. Medline

Conrad KL, Tseng KY, Uejima JL, Reimers JM,Heng LJ, Shaham Y, Marinelli M, Wolf ME(2008) Formation of accumbens GluR2-lacking AMPA receptors mediates incubationof cocaine craving. Nature 454:118121.CrossRef Medline

CreedMC, Luscher C (2013) Drug-evoked synap-tic plasticity: beyond metaplasticity. Curr OpinNeurobiol 23:553558. CrossRefMedline

Das P, Lilly SM, Zerda R, GunningWT 3rd, Alva-rez FJ, Tietz EI (2008) Increased AMPA re-ceptor GluR1 subunit incorporation in rathippocampal CA1 synapses during benzodi-

azepine withdrawal. J Comp Neurol 511:832846. CrossRef Medline

Deroche-Gamonet V, Belin D, Piazza PV (2004)Evidence for addiction-like behavior in the rat.Science 305:10141017. CrossRefMedline

De Wit H, Stewart J (1981) Reinstatement ofcocaine-reinforced responding in the rat. Psy-chopharmacology (Berl) 75:134143. CrossRefMedline

Di Chiara G, Imperato A (1988) Drugs abusedby humans preferentially increase synapticdopamine concentrations in the mesolimbicsystem of freely moving rats. Proc Natl AcadSci U S A 85:52745278. CrossRef Medline

EngblomD, BilbaoA, Sanchis-SeguraC,Dahan L,Perreau-Lenz S, Balland B, Parkitna JR, LujanR, Halbout B, Mameli M, Parlato R, SprengelR, Luscher C, Schutz G, Spanagel R (2008)Glutamate receptors on dopamine neuronscontrol the persistence of cocaine seeking.Neuron 59:497508. CrossRef Medline

Gerber GJ, Wise RA (1989) Pharmacologicalregulation of intravenous cocaine and heroinself-administration in rats: a variable doseparadigm. Pharmacol Biochem Behav 32:527531. CrossRef Medline

Goldman D, Oroszi G, Ducci F (2005) The ge-netics of addictions: uncovering the genes.Nat Rev Genet 6:521532. CrossRef Medline

Gong S, Doughty M, Harbaugh CR, Cummins A,Hatten ME, Heintz N, Gerfen CR (2007)Targeting Cre recombinase to specific neuronpopulations with bacterial artificial chromo-some constructs. J Neurosci 27:98179823.CrossRef Medline

Grimm JW, Hope BT, Wise RA, Shaham Y(2001) Neuroadaptation. Incubation of co-caine craving after withdrawal. Nature 412:141142. CrossRef Medline

Heikkinen AE, Moykkynen TP, Korpi ER (2009)Long-lasting modulation of glutamatergictransmission in VTA dopamine neurons after asingle dose of benzodiazepine agonists. Neuro-psychopharmacology 34:290298. CrossRefMedline

Highfield DA, Mead AN, Grimm JW, Rocha BA,Shaham Y (2002) Reinstatement of cocaineseeking in 129X1/SvJ mice: effects of cocainepriming, cocaine cues and food deprivation.Psychopharmacology (Berl) 161:417424.CrossRef

Jackson A, Mead AN, Rocha BA, Stephens DN(1998) AMPA receptors and motivation fordrug: effect of the selective antagonist NBQXon behavioural sensitization and on self-administration in mice. Behav Pharmacol9:457467. CrossRef Medline

Kalivas PW,OBrienC (2008) Drug addiction asa pathology of staged neuroplasticity. Neuro-psychopharmacology 33:166180. CrossRefMedline

Kruzich PJ, Congleton KM, See RE (2001) Condi-tioned reinstatement of drug-seeking behaviorwith a discrete compound stimulus classi-cally conditioned with intravenous cocaine.Behav Neurosci 115:10861092. CrossRefMedline

Lammel S, LimBK, RanC,HuangKW, BetleyMJ,Tye KM, Deisseroth K, Malenka RC (2012)Input-specific control of reward and aversionin the ventral tegmental area. Nature 491:212217. CrossRef Medline

Luscher Drug-Evoked Synaptic Plasticity J. Neurosci., November 6, 2013 33(45):1764117646 17645

Lee BR, Ma YY, Huang YH, Wang X, Otaka M,Ishikawa M, Neumann PA, Graziane NM,Brown TE, Suska A, Guo C, Lobo MK, Ses-ack SR, Wolf ME, Nestler EJ, Shaham Y,Schluter OM, Dong Y (2013) Maturationof silent synapses in amygdala-accumbensprojection contributes to incubation of co-caine craving. Nat Neurosci. Advance on-line publication retrieved September 29,2013. doi:10.1038/nn.3533.

Licata SC, Rowlett JK (2008) Abuse and depen-dence liability of benzodiazepine-type drugs:GABA(A) receptor modulation and beyond.Pharmacol Biochem Behav 90:7489. CrossRefMedline

Luscher C, Malenka RC (2011) Drug-evokedsynaptic plasticity in addiction: frommolecu-lar changes to circuit remodeling. Neuron 69:650663. CrossRef Medline

Luscher C, Ungless MA (2006) The mechanisticclassification of addictive drugs. PLoS Med3:e437. CrossRef Medline

Mameli M, Balland B, Lujan R, Luscher C (2007)Rapid synthesis and synaptic insertion of GluR2for mGluR-LTD in the ventral tegmental area.Science 317:530533. CrossRefMedline

Mameli M, Halbout B, Creton C, Engblom D,Parkitna JR, Spanagel R, Luscher C (2009)Cocaine-evoked synaptic plasticity: persis-tence in the VTA triggers adaptations in theNAc. Nat Neurosci 12:10361041. CrossRefMedline

Mameli M, Bellone C, Brown MT, Luscher C(2011) Cocaine inverts rules for synapticplasticity of glutamate transmission in theventral tegmental area. Nat Neurosci 14:414416. CrossRef Medline

McCutcheon JE, Loweth JA, Ford KA,Marinelli M,Wolf ME, Tseng KY (2011) Group I mGluRactivation reverses cocaine-induced accumula-tion of calcium-permeable AMPA receptors innucleusaccumbens synapsesvia aproteinkinaseC-dependentmechanism. JNeurosci31:1453614541. CrossRefMedline

Mogenson GJ, Jones DL, Yim CY (1980) Frommotivation to action: functional interfacebetween the limbic system and the motorsystem. Prog Neurobiol 14:6997. CrossRefMedline

OConnor EC, Chapman K, Butler P, Mead AN

(2011) The predictive validity of the rat self-administrationmodel for abuse liability. Neu-rosci Biobehav Rev 35:912938. CrossRefMedline

Papp E, Borhegyi Z, Tomioka R, Rockland KS,Mody I, Freund TF (2012) Glutamatergicinput from specific sources influences the nu-cleus accumbens-ventral pallidum informa-tion flow. Brain Struct Funct 217:3748.CrossRef Medline

Pascoli V, Turiault M, Luscher C (2011) Rever-sal of cocaine-evoked synaptic potentiationresets drug-induced adaptive behaviour. Na-ture 481:7175. CrossRef Medline

Redish AD, Jensen S, Johnson A (2008) A uni-fied framework for addiction: vulnerabilitiesin the decision process. Behav Brain Sci 31:415437. CrossRef Medline

Robinson TE, Berridge KC (2001) Incentive-sensitization and addiction. Addiction 96:103114. CrossRef Medline

Rocha BA, Fumagalli F, Gainetdinov RR, Jones SR,Ator R, Giros B,Miller GW, CaronMG (1998)Cocaine self-administration in dopamine-transporter knockout mice. Nat Neurosci1:132137. CrossRef Medline

Saal D, Dong Y, Bonci A, Malenka RC (2003)Drugs of abuse and stress trigger a commonsynaptic adaptation in dopamine neurons.Neuron 37:577582. CrossRef Medline

Schultz W (2011) Potential vulnerabilities ofneuronal reward, risk, and decision mecha-nisms to addictive drugs.Neuron 69:603617.CrossRef Medline

Sesack SR, Grace AA (2010) Cortico-basal gangliareward network: microcircuitry. Neuropsycho-pharmacology 35:2747. CrossRefMedline

Shaham Y, Shalev U, Lu L, De Wit H, Stewart J(2003) The reinstatement model of drug re-lapse: history, methodology and major find-ings. Psychopharmacology (Berl) 168:320.CrossRef Medline

Stewart J, de Wit H, Eikelboom R (1984) Role ofunconditioned and conditioned drug effects inthe self-administration of opiates and stim-ulants. Psychol Rev 91:251268. CrossRefMedline

Stuber GD, Sparta DR, Stamatakis AM, van Leeu-wen WA, Hardjoprajitno JE, Cho S, Tye KM,Kempadoo KA, Zhang F, Deisseroth K, Bonci

A (2011) Excitatory transmission from theamygdala to nucleus accumbens facilitates re-ward seeking. Nature 475:377380. CrossRefMedline

Tan KR, Brown M, Laboue`be G, Yvon C, CretonC, Fritschy JM, RudolphU, Luscher C (2010)Neural bases for addictive properties of ben-zodiazepines. Nature 463:769774. CrossRefMedline

Tsai HC, Zhang F, Adamantidis A, Stuber GD,Bonci A, de Lecea L, Deisseroth K (2009)Phasic firing in dopaminergic neurons is suf-ficient for behavioral conditioning. Science324:10801084. CrossRef Medline

Ungless MA, Whistler JL, Malenka RC, Bonci A(2001) Single cocaine exposure in vivo induceslong-term potentiation in dopamine neurons.Nature 411:583587. CrossRefMedline

Valjent E, Bertran-Gonzalez J, Aubier B, GreengardP, Herve D, Girault JA (2010) Mechanisms oflocomotor sensitization to drugs of abuse in atwo-injection protocol. Neuropsychopharma-cology 35:401415. CrossRefMedline

Vanderschuren LJ, Everitt BJ (2004) Drug seekingbecomes compulsive after prolonged cocaineself-administration. Science 305:10171019.CrossRefMedline

Vengeliene V, Celerier E, Chaskiel L, Penzo F, Spa-nagel R (2009) Compulsive alcohol drinkingin rodents. Addict Biol 14:384396. CrossRefMedline

Wagner FA, Anthony JC (2002) From first druguse to drug dependence; developmental peri-ods of risk for dependence upon marijuana,cocaine, and alcohol. Neuropsychopharma-cology 26:479488. CrossRef Medline

Wise RA (1987) The role of reward pathways inthe development of drug dependence. Phar-macol Ther 35:227263. CrossRef Medline

YuanT,MameliM,OConnorEC,DeyPN,VerpelliC, Sala C, Perez-Otao I, Luscher C, Bellone C(2013) Expression of cocaine-evoked synapticplasticity by GluN3A-containing NMDA re-ceptors. Neuron, in press.

Zweifel LS, Argilli E, Bonci A, Palmiter RD(2008) Role ofNMDAreceptors in dopamineneurons for plasticity and addictive behaviors.Neuron 59:486496. CrossRef Medline

17646 J. Neurosci., November 6, 2013 33(45):1764117646 Luscher Drug-Evoked Synaptic Plasticity

Drug-Evoked Synaptic Plasticity Causing Addictive BehaviorIntroductionAnimal modelsDopamine hypothesisDrug-evoked synaptic plasticityThe case for a permissive metaplasticityCausal role of drug-evoked synaptic plasticity in behaviorThe challenge of individual vulnerabilityEmerging ideas for rational addiction treatments

References