Ethanol Effects on Dopaminergic ???Reward??? Neurons in the Ventral Tegmental Area and the Mesolimbic Pathway

Ethanol Effects on Dopaminergic “Reward” Neuronsin the Ventral Tegmental Area and the

Mesolimbic PathwaySarah B. Appel, William J. McBride, Marco Diana, Ivan Diamond, Antonello Bonci, and Mark S. Brodie

Dopaminergic (DA) neurons in the ventral tegmental area (VTA) provide the DA innervation of thenucleus accumbens. This mesolimbic DA pathway is important for the reinforcing effects of alcohol andplays a central role in alcohol-related behaviors. This Research Society on Alcoholism symposium includeda discussion of the acute and chronic effects of ethanol and ethanol withdrawal on DA VTA neurons. Theexperiments that were discussed ranged from studies in the freely moving behaving rat and electrophysi-ological studies in vivo, to electrophysiological studies in brain slices and acutely dissociated DA VTAneurons, to neurochemical studies that explored the cellular basis of ethanol’s actions. Because ethanol’seffects on this reinforcement pathway are critically important for voluntary intake of alcohol and alcoholaddiction, this symposium report may be of interest to both basic science and clinical researchers in thealcohol field. This symposium focused on effects of ethanol on the mesolimbic dopamine pathway, specif-ically the VTA and the nucleus accumbens. The organizer/co-chairs were Sarah B. Appel and Mark S.Brodie. The presentations were (1) Introduction, by Mark S. Brodie; (2) Reinforcing Actions of Alcohol inthe Ventral Tegmental Area: Intracranial Self-Administration Studies, by William J. McBride; (3) APossible Mechanism Mediating the Direct Excitation of Dopaminergic Ventral Tegmental Area Neuronsby Ethanol, by Sarah B. Appel; (4) Effect of Chronic Ethanol and Withdrawal on Dopaminergic VentralTegmental Area Neurons: In Vivo Electrophysiological Studies, by Marco Diana; (5) Ethanol InducesProtein Kinase A Translocation Into the Nucleus, Cyclic AMP Response Element Binding Protein Phos-phorylation, and Increases in Cyclic Adenosine Monophosphate–Dependent Gene Expression, by IvanDiamond; and (6) Co-activation of Dopamine D1 and D2 Receptors Is Necessary for Dopamine-MediatedIncreases in Firing Activity in Nucleus Accumbens Neurons, by Antonello Bonci.

Key Words: Alcohol, Dopamine, Nucleus Accumbens, Reinforcement, cAMP.

INTRODUCTION

Mark S. Brodie

This report describes data presented in a symposium atthe 2001 Research Society on Alcoholism meeting; the

introduction and the references have been updated to in-clude recent publications. The topic of this symposium wasethanol action on important components of the “reward”pathways in the central nervous system, specifically, theventral tegmental area (VTA), and its dopaminergic (DA)innervation of the nucleus accumbens. The VTA containsthe DA neurons named the A10 group by Dahlstrom andFuxe (Dahlstrom and Fuxe, 1964; Grenhoff et al., 1991)and is located in the ventral mesencephalon, medial to thesubstantia nigra and anterior to the pons. The dopamine-containing neurons of the VTA project to the nucleusaccumbens in the forebrain and also supply DA innervationto other structures, including the prefrontal cortex, por-tions of the amygdala, the olfactory tubercle, the lateralseptal nucleus, and the hippocampus (Oades and Halliday,1987).

The seminal electrical self-stimulation experiments ofJames Olds and subsequent pharmacological studies indi-cated that the DA neurons of the mesencephalon play animportant role in “reward” (Olds and Fobes, 1981; Oldsand Milner, 1954; Wise, 1987). Specifically, DA projectionsfrom the VTA to the nucleus accumbens are important formediating the reinforcing effects of drugs of abuse. Results

From the Department of Physiology and Biophysics (SBA, MSB), Univer-sity of Illinois at Chicago, College of Medicine, Chicago, Illinois; Departmentof Psychiatry (WJM), Institute of Psychiatric Research, Indiana UniversitySchool of Medicine, Indianapolis, Indiana; Department of Drug Sciences(MD), University of Sassari, Sassari, Italy; Ernest Gallo Clinic and ResearchCenter (AB, ID), Emeryville, California; Department of Neurology (AB, ID),University of California, San Francisco, California; Department of Cellularand Molecular Pharmacology, Neuroscience Graduate Program and Centerfor the Neurobiology of Addiction (ID), University of California, San Fran-cisco, San Francisco, California.

Received for publication April 14, 2004; accepted July 22, 2004.Sponsored by PHS grants AA05846 (SBA), AA09125 (MSB), AA07462,

AA07611, AA10721 and AA12262 (WJM), AA10039 (ID) and funds pro-vided by the State of California for medical research on alcohol and substanceabuse through the University of California, San Francisco (ID and AB).

Reprint requests: Sarah B. Appel, PhD, Department of Physiology andBiophysics (M/C 901) University of Illinois at Chicago 835 S. Wolcott Ave-nue, Chicago, IL 60612-7342; Fax: 312-996-1414; e-mail: [email protected].

Copyright © 2004 by the Research Society on Alcoholism.

DOI: 10.1097/01.ALC.0000145976.64413.21

0145-6008/04/2811-1768$03.00/0ALCOHOLISM: CLINICAL AND EXPERIMENTAL RESEARCH

Vol. 28, No. 11November 2004

1768 Alcohol Clin Exp Res, Vol 28, No 11, 2004: pp 1768–1778

of experiments implicating dopamine in the reinforcingeffects of drugs of abuse have been reviewed by Wise(1987) and by Koob and Bloom (1988). Lesions in orpharmacological blockade of the mesoaccumbens DApathway have been shown to block drug-induced reinforce-ment for psychomotor stimulants and opiates (De Wit andWise, 1977; Spyraki et al., 1982, 1983; Yokel and Wise,1976).

With regard to ethanol specifically, it has been shownthat DA agents alter the reinforcing effects of ethanol in anumber of different sophisticated behavioral paradigms. Ina series of papers by Pfeffer and Samson (1985, 1986, 1988),systemic administration of dopamine agonists or dopamineantagonists reduced ethanol intake in rats in limited accessor operant paradigms. Similar results were observed byWeiss et al. (1990), who found that systemic administrationof the dopamine agonist bromocriptine decreased ethanolpreference in a two-lever, free-choice task. In operant re-sponse studies performed by Samson’s group (Samson etal., 1990, 1992), selective administration of dopamine ago-nists locally into the nucleus accumbens increased respond-ing for ethanol, and dopamine antagonists into the nucleusaccumbens decreased such responding. In a different an-tagonist study, using a different behavioral paradigm, Levyet al. (1991) demonstrated that the dopamine antagonistsulpiride injected into the nucleus accumbens increaseddrinking in alcohol-preferring (P) rats. Findings such asthese from a number of laboratories provide strong evi-dence for an important role of dopamine in mediating thereinforcing action of ethanol.

More recent behavioral studies in the past 10 years havehelped to define more specifically the role of the mesolim-bic dopamine pathway in the rewarding/reinforcing effectsof natural reinforcers such as food (Salamone and Correa,2002) and of drugs of abuse, including ethanol (Di Chiara,2002; Salamone et al., 2003). New conceptualizations havepointed to two aspects of reward, subjective feelings ofpleasure or euphoria and the implicit reinforcing effects onmotivated behavior, which have been referred to as “liking”and “wanting,” respectively (Berridge and Robinson, 2003).Experiments in which the mesolimbic dopamine pathwaywas blocked in a variety of different ways, including mutantmice that make no dopamine as a result of the lack oftyrosine hydroxylase in their DA neurons, suggest that the“liking” aspect of reward may be independent of dopamine(Berridge and Robinson, 2003; Cannon and Palmiter,2003). The lack of mesolimbic dopamine, however, blocks“wanting” or the motivation to seek out and self-administerthe food reward (Cannon and Palmiter, 2003). Similarly,“hyperdopaminergic” mice who have reduced levels of thedopamine transporter show more “wanting” of a sweetreward as measured by enhanced runway performance butnot increased “liking” of the reward as measured by oro-facial reactions (Pecina et al., 2003). In experiments inwhich ethanol seeking (appetitive responding) and ethanolintake (consummatory responding) were procedurally sep-

arated, ethanol seeking was found to be more sensitive thanethanol intake to blockade of dopamine D2 receptors bythe D2 antagonists raclopride and remoxipride (Cza-chowski et al., 2001, 2002), although higher concentrationsof raclopride also decreased ethanol intake (Czachowski etal., 2001).

In addition to these behavioral data, it has been shownthat dopamine overflow in the nucleus accumbens is in-creased by a number of drugs of abuse, including ethanol.This was shown in microdialysis experiments in anesthe-tized rats by Imperato and Di Chiara (1986). This findingwas significantly underscored in microdialysis experimentsin awake, behaving rats by Weiss et al. (1993). In that study,the authors reported that in rats that self-administeredethanol in a free-choice operant task, dopamine release inthe nucleus accumbens was increased. Furthermore, thedopamine release in the nucleus accumbens of P rats wasincreased more than that of nonselected Wistar rats. Arecent microdialysis study separated the bar-pressing pe-riod from the period of ethanol consumption and foundthat dopamine release in the nucleus accumbens seemed tobe associated with the stimulus properties of ethanol, as itwas greatest in the initial part of the ethanol consumptionperiod (Doyon et al., 2003).

Electrophysiological recordings from DA VTA neuronsin behaving animals by Schultz’s group indicate that theactivity of the DA neurons does not simply indicate theactual receipt of a reward but codes for higher level infor-mation about the predictability, novelty, and intensity ofthe reward (Schultz, 2001, 2002). In addition, omission ofan expected reward causes a phasic inhibition of the firingof DA neurons at the time the reward was expected.Schultz described the activity of the DA neurons as codinga “reward prediction error signal,” which is a “powerfulteaching signal” for learning and motivated behavior(Schultz, 2001). In fact, with training, the DA VTA neuronsbegin to fire upon presentation of the cue that predicts theoccurrence of the reward (conditioned stimulus). Wise(2002) pointed out that drugs of abuse (e.g., ethanol) candirectly activate or enhance activity in the DA pathway.This confers them with a powerful ability to make drug-predictive cues into conditioned reinforcers; that is, the cueitself becomes reinforcing (Wise, 2002, 2004). This process,coupled with sensitization of neural circuits upon repeatedexposure to the drug, can cause drug-associated cues totrigger excessive wanting, “craving,” or drug-seeking behav-ior (Robinson and Berridge, 2003). Robinson and Berridgedescribed the development of this motivational componentof addiction as “incentive sensitization.”

In summary, although the exact role of the mesolimbicDA pathway in the hedonic and reinforcing effects of foodand drugs of abuse is currently an active area of researchand debate, it seems clear that the DA pathway from theVTA to the nucleus accumbens is involved in various as-pects of drug and ethanol self-administration and addiction(Salamone et al., 2003). Specifically, this DA pathway

ETHANOL EFFECTS ON DA VTA NEURONS AND THE MESOLIMBIC PATHWAY 1769

seems to contribute to drug-seeking behavior, incentivemotivation for working to obtain the drug, incentive sensi-tization, and cue-related wanting or “craving,” which maylead to relapse (Robinson and Berridge, 2003; Salamone etal., 2003).

Because dopamine cell bodies of the VTA provide theDA innervation of the nucleus accumbens, some electro-physiological studies have focused on ethanol action onthe DA neurons of the VTA. It was shown by Gessa et al.(1985) that systemic administration of ethanol causes aconcentration-dependent excitation of DA neurons ofthe VTA. It should be noted that, in this study, DAneurons of the substantia nigra were also excited byethanol, but higher concentrations of ethanol were re-quired. Our laboratory demonstrated that ethanol in-creases the spontaneous firing rate of DA neurons inbrain slices of the VTA (Brodie et al., 1990). Thisethanol-induced excitation of DA VTA neurons oc-curred even under low-calcium, high-magnesium condi-tions that blocked synaptic transmission, suggesting to usthat ethanol had a direct effect on the firing rate of DAneurons of the VTA (see “A Possible Mechanism Medi-ating the Direct Excitation of Dopaminergic VentralTegmental Area Neurons by Ethanol” below).

The topic of this symposium was the action of ethanol onthe dopamine neurons in the VTA and the DA communi-cation between the VTA and the nucleus accumbens. Thesection of the symposium by William McBride describedstudies by his laboratory on self-administration of ethanoldirectly into the VTA in awake, behaving rats, which sup-port the idea that ethanol action on DA neurons of theVTA is important for its reinforcing effect. The section bySarah Appel described electrophysiological studies in ratbrain slices and isolated neurons by the Appel/Brodie lab-oratory, which focused on the mechanism by which ethanolexcites neurons in the VTA. Increased firing of DA neu-rons in the VTA would lead to an increase in dopaminerelease in the nucleus accumbens. The section by MarcoDiana described his electrophysiological studies in the in-tact rat, which investigated the effect of chronic ethanoland withdrawal from chronic ethanol on the firing of DAVTA neurons. Then, in Ivan Diamond’s section of thesymposium, the focus shifted to postsynaptic synergy be-tween dopamine D2 and adenosine A2 receptors in thenucleus accumbens and neurochemical studies on the in-teraction of dopamine D2 receptor agonists and ethanol ina model system designed to emulate this interaction in thenucleus accumbens. Finally, the section by Antonello Boncidescribed his studies on interactions between dopamine D1

and D2 receptor activation on neurons of the nucleus ac-cumbens. The work presented in the last two sections isrelevant to understanding the downstream effects ofethanol-induced increases in dopamine release in the nu-cleus accumbens.

REINFORCING ACTIONS OF ALCOHOL IN THE VENTRALTEGMENTAL AREA: INTRACRANIAL SELF-

ADMINISTRATION STUDIES

William J. McBride

Several reports have implicated the involvement of theVTA in mediating the actions of ethanol. It has beendemonstrated that ethanol increases the firing rate of DAneurons both in vitro (Brodie et al., 1990, 1995) and in vivo(Gessa et al., 1985). Systemic ethanol administration alsohas been shown to increase somatodendritic dopamine re-lease in the VTA (Campbell et al., 1996). Operant respond-ing for oral ethanol self-administration has been shown toincrease the release of dopamine in the nucleus accumbensof P rats (Weiss et al., 1993). More recently, it was dem-onstrated that activation of dopamine D2 autoreceptors inthe VTA reduced ethanol drinking of P rats (Nowak et al.,2000). The objectives of our studies were to (1) determinewhether there is a difference in the reinforcing effects ofethanol in the VTA of the high-alcohol-consuming P line ofrats and low-alcohol-consuming rats, (2) examine the re-gional heterogeneity within the VTA for the intracranialself-administration (ICSA) of ethanol, and (3) determinethe involvement of dopamine neurons in the VTA in me-diating the reinforcing effects of ethanol.

The ICSA procedure has been used to identify brainregions involved in mediating the rewarding effects ofdrugs of abuse (McBride et al., 1999; Wise and Hoffman,1992) and has been used to examine the reinforcing actionsof ethanol within the VTA. In the first study (Gatto et al.,1994), it was determined that the P line of rats wouldself-administer 50 to 200 mg/dl of ethanol directly into theVTA, whereas the alcohol nonpreferring (NP) line did not.These results suggested that the VTA is a site that supportsthe reinforcing effects of alcohol and that this action withinthe VTA is genetically influenced and may be related toinnate alcohol preference.

ICSA studies indicated that there may be differencesbetween the anterior and posterior VTA in mechanismsthat support reinforcement (Ikemoto et al., 1997b, 1998).Wistar rats readily self-administered a �-amino butyric acidtype A (GABAA) receptor antagonist into the anterior(�4.8 to �5.2 mm bregma) but not into the posterior (�5.3to �6.3 mm bregma) VTA, suggesting that GABAA recep-tor–regulated mechanisms that mediate reinforcement pro-cesses may be different between the anterior and the pos-terior VTA (Ikemoto et al., 1997b). For testing thispossibility, a second study (Ikemoto et al., 1998) establishedthat the GABAA receptor agonist muscimol was self-infused into the posterior but not the anterior VTA, sup-porting the idea that the anterior and the posterior VTAare involved in mediating the reinforcing process but bydifferent GABAA receptor–regulated mechanisms. Theoriginal study (Gatto et al., 1994) targeted the middle andmiddle-posterior portions of the VTA. Therefore, the ob-jective of the second study was to examine the ICSA of

1770 APPEL ET AL.

ethanol into the anterior and posterior portions of the VTA.This study established that 100 to 400 mg/dl of ethanol wasself-infused into the posterior but not the anterior portion ofthe VTA of Wistar rats (Rodd-Henricks et al., 2000), suggest-ing that the VTA is a functionally heterogeneous structurewith regard to ethanol reinforcement.

The initial study with P and NP rats (Gatto et al., 1994)suggested that there may be a positive relationship betweenthe reinforcing effects of ethanol in the VTA and innatealcohol preference. For testing this hypothesis, the dose �response effects for the ICSA of 0 to 200 mg/dl of ethanolwere examined in the posterior VTA of Wistar rats (lowalcohol drinkers) and the P line of rats (Rodd et al., 2004a).Wistar rats self-administered 150 and 200 mg/dl ethanol butdid not self-infuse 50 to 100 mg/dl ethanol. In contrast, Prats self-administered ethanol at concentrations of 75 mg/dland higher, indicating that the dose-response curve for theP line is to the left of the curve for the Wistar rats. Theseresults support the above hypothesis and suggest that thereis a positive relationship between the reinforcing effects ofethanol in the posterior VTA and innate high alcoholpreference.

In the study that compared Wistar and P rats (Rodd etal., 2004a), data for the dose-response curves were ob-tained in the fourth acquisition session, after which allanimals were subjected to two extinction sessions and thena reinstatement session. Both P and Wistar rats readilyextinguished responding on the active lever when vehiclewas substituted for ethanol during the two extinction ses-sions, and both reinstated responding on the active leverwhen ethanol was restored. However, for the P rat, at dosesof 150 mg/dl and lower, the number of infusions during thereinstatement session was significantly higher than infu-sions during the fourth acquisition session, whereas for theWistar rats, there were no differences in the number ofinfusions between the fourth acquisition session and thereinstatement session at any of the ethanol concentrations.These results suggest that sensitization or tolerance to thereinforcing effects of ethanol may have developed withinthe posterior VTA of P but not Wistar rats. The develop-ment of sensitization or tolerance to the reinforcing effectsof ethanol may contribute to the high alcohol drinkingbehavior of P rats.

The electrophysiological (Brodie et al., 1990, 1995; Gessaet al., 1985) and microdialysis data (Campbell et al., 1996)suggest that ethanol can activate DA neurons in the VTA.Although the mechanisms underlying the ICSA of ethanolinto the posterior VTA are unknown, one hypothesis is thatethanol self-infusions are maintained by activating DA neu-rons within the VTA. To test this hypothesis, we tested theco-infusion of the D2/D3-like dopamine receptor agonistquinpirole with 200 mg/dl of ethanol into the posteriorVTA of Wistar rats (Rodd et al., 2004b). Infusion of quin-pirole into the VTA should activate dopamine D2 autore-ceptors and reduce the activity of DA neurons. In thisstudy, 200 mg/dl of ethanol alone was given during the first

four sessions, and stable responding on the active lever (60� 5 responses/session) and inactive lever (15 � 4 respons-es/session) was attained by Wistar rats. Addition of 10 �Mof quinpirole to the 200 mg/dl of ethanol solution resultedin a marked decrease in responding on the active lever (15� 2 responses/session). Responding on the active lever wasreinstated to control levels when 200 mg/dl of ethanol wasrestored in a subsequent session. These results suggest thatthe reinforcing effects of ethanol in the posterior VTA aremediated by DA neuronal activity.

In conclusion, the results suggest that (1) there is re-gional heterogeneity within the VTA for the reinforcingeffects of ethanol, (2) there is a positive relationship be-tween high alcohol preference and sensitivity to the rein-forcing effects of ethanol in the VTA, (3) sensitization ortolerance to the reinforcing effects of ethanol develops inthe VTA of P rats but not Wistar rats under the presentICSA conditions, and (4) the reinforcing effects of ethanolwithin the VTA require DA neuronal activity.

A POSSIBLE MECHANISM MEDIATING THE DIRECTEXCITATION OF DOPAMINERGIC VENTRAL TEGMENTAL

AREA NEURONS BY ETHANOL

Sarah B. Appel and Mark S. Brodie

The cell bodies of DA neurons in the VTA project theiraxons in the mesolimbic dopamine pathway to the nucleusaccumbens (Oades and Halliday, 1987). Ethanol increasesthe firing rate of DA neurons in the VTA measured in vivoin unanesthetized rats (Gessa et al., 1985), which leads torelease of dopamine from terminals in the nucleus accum-bens (Di Chiara and Imperato, 1988; Weiss et al., 1993);this activation of the mesolimbic dopamine pathway isimportant for the reinforcing effects of ethanol (Samson etal., 1990, 1992; Wise, 1987). We have shown that behavior-ally relevant concentrations of ethanol (20–120 mM) in-creased the firing rate of DA VTA neurons recorded inbrain slices in a concentration-dependent manner and thatthis effect persisted when Ca2�-dependent synaptic trans-mission was blocked with low-Ca2�/high-Mg2� media (Bro-die et al., 1990). This observation showed that ethanolexcitation did not require Ca2�-dependent neurotransmit-ter release and suggested that it was a direct postsynapticeffect on the DA neurons themselves. To control for thepossible contribution of other indirect influences such asCa2�-independent neurotransmitter release, we studied theeffect of ethanol on acutely dissociated DA VTA neurons(Brodie et al., 1999b). Neurons were dissociated by enzy-matic treatment and trituration of the VTA dissected frombrain slices. The dissociation procedure truncated the den-dritic trees, stripped off synaptic contacts, and widely dis-persed the individual neurons. The dissociated neuronswere identified as dopaminergic according to electrophys-iological, pharmacological, and immunohistochemical cri-teria as follows. Loose-patch cell attached recording fromthese dissociated neurons showed that they were spontane-

ETHANOL EFFECTS ON DA VTA NEURONS AND THE MESOLIMBIC PATHWAY 1771

ously active, with long-duration action potentials and apacemaker-like firing pattern (uniform interspike interval)typical of DA VTA neurons recorded in brain slices. Bathapplication of dopamine to these neurons decreased thefiring rate; this inhibition by dopamine is characteristic ofDA neurons in the VTA and is due to the presence of D2autoreceptors on these neurons. Immunohistochemicalstaining for tyrosine hydroxylase, the rate-limiting enzymefor synthesis of dopamine, was performed after recordingfrom dissociated neurons to confirm their identity as DAneurons. Acutely dissociated DA VTA neurons were ro-bustly excited by ethanol at behaviorally relevant concen-trations (20–120 mM), and this effect was concentrationdependent and reversible with washout (Brodie et al.,1999b). As described above, ICSA experiments in theMcBride laboratory have shown that rats will self-administer 200 mg/dl of ethanol directly into the VTA; thiscorresponds to a concentration of 44 mM. Application of40 mM ethanol to acutely dissociated DA VTA neuronscaused a clear, prompt increase in firing rate of ~30%above baseline (Brodie et al., 1999b). Higher concentra-tions of ethanol caused larger increases, with 120 mM ofethanol increasing the firing rate by �60%. This studyshows that ethanol can excite DA VTA neurons in theabsence of input from surrounding cells and providesstrong evidence that ethanol directly excites these neurons.

Our intracellular current clamp experiments in brainslices have shown that ethanol excitation is associated witha decrease in the afterhyperpolarization (AHP) that followsspontaneous action potentials in DA VTA neurons (Brodieand Appel, 1998). This observation suggested that ethanolmay be increasing the firing rate of DA VTA neurons bymodulating a current that contributes to the AHP, either byreducing a K� current or increasing the inward current Ih.To test this hypothesis, we used a number of blockers ofdifferent types of K� channels and of Ih to determinewhether any of these could prevent the ethanol excitationof DA VTA neurons. These studies used extracellularsingle-unit recording, intracellular recording, and whole-cell patch-clamp recording from DA VTA neurons in brainslices from Fischer 344 rats. Ethanol (40–120 mM) andchannel blockers were applied in the bath. Ethanol excita-tion of DA VTA neurons was not reduced by blockade ofIh with external cesium or the more selective Ih blockerZD7288 (Appel et al., 2003), indicating that enhancementof Ih is not a major factor mediating ethanol excitation inthese neurons. The SK-type of calcium-dependent K� cur-rent contributes to the AHP in DA VTA neurons, butblockade of this current with apamin or D-tubocurarine didnot block the ethanol excitation (Brodie et al., 1999a),which indicated that ethanol-induced excitation was notdue to a reduction in SK current. External tetraethylam-monium ion (TEA) was used to block the BK type ofcalcium-dependent K� current, and external barium wasused to block G-protein–coupled inwardly rectifying K�

channels; neither treatment prevented the ethanol excita-

tion of DA VTA neurons (Appel et al., 2003). It is notedthat at the concentrations tested, external cesium, barium,and TEA also block some types of delayed rectifier K�

channels.It is interesting that we found that the ethanol-induced

excitation of DA VTA neurons was blocked by quinidine(20–80 �M), a drug that blocks many types of delayedrectifier K� channels, including some that are insensitive toTEA and the other blockers tested (Appel et al., 2003).Therefore, we measured the effect of ethanol on the de-layed rectifier K� current in voltage-clamp experimentswith whole-cell patch recording from acutely dissociatedDA VTA neurons and DA VTA neurons in brain slicesfrom Fischer 344 rats. Voltage-gated sodium current wasblocked with tetrodotoxin. The holding potential was �40mV to inactivate IA, and sustained outward K� currentswere evoked with depolarizing voltage steps from �40 to 40mV (in 10 mV increments). These K� currents were eithernoninactivating or slowly inactivating. Ethanol (30–100mM) decreased this delayed rectifier K� current, and theeffect reversed with washout (Appel, 2004; Brodie et al.,2000).

Taken together, our studies suggest that ethanol excita-tion of DA VTA neurons is due to reduction of a quinidine-sensitive K� current of the delayed rectifier type. Ongoingwork in our laboratory is directed at identifying the nativeethanol-sensitive K� channel in terms of cloned K� chan-nels of known structure. There are five gene families thatcode for delayed rectifier channel proteins that are blockedby quinidine (Appel et al., 2003). Pharmacological experi-ments with more selective channel blockers, single-cell reversetranscriptase–polymerase chain reaction, and immunohisto-chemistry are being used to identify the ethanol-sensitivechannel protein in DA VTA neurons. Identification of theethanol-sensitive K� channel protein in DA VTA neuronsand the gene that codes for it could have important implica-tions for understanding genetic factors in alcohol addictionand may provide a new target for controlling the reinforcingeffects of ethanol and for novel pharmacotherapies foralcoholism.

EFFECT OF CHRONIC ETHANOL AND WITHDRAWAL ONDOPAMINERGIC VENTRAL TEGMENTAL AREA NEURONS:

IN VIVO ELECTROPHYSIOLOGICAL STUDIES

Marco Diana

Among the major neurotransmitter systems involved inthe central actions of ethanol are dopamine-containingneurons, whose cell bodies are located in the midbrain.These neuronal ensembles, which have been implicated inmotor regulation, affect-related disorders, and cognitivefunctions, are typically subdivided into three major com-partments: the nigrostriatal, the mesocortical, and the me-solimbic systems. The last pathway is known to project tothe subdivisions of the nucleus accumbens (i.e., shell andcore) through which it influences motivational and affect-

1772 APPEL ET AL.

related actions of drugs of abuse, including ethanol. Ac-cordingly, acute administration of ethanol in vivo is knownto increase DA neuronal activity in the VTA (Gessa et al.,1985) and dopamine outflow in the nucleus accumbens, asmeasured by microdialysis (Di Chiara and Imperato, 1988).In addition, in vitro intracellular recording studies from DAneurons of the VTA in brain slices (Brodie and Appel,1998) and acutely dissociated neurons have shown a directaction of ethanol on the somatic membrane of VTA neu-rons (Brodie et al., 1999b). This direct excitation can bepotentiated by blockade of a Ca2�-activated potassium con-ductance (Brodie et al., 1999a). All of these studies haveprovided important information about the cellular site ofaction of ethanol in the rodent central nervous system andform the basis for understanding the acute effects of etha-nol on DA neurons. However, when considered in light ofthe general phenomenon of drug addiction (alcoholism),acute studies remain of limited utility and difficult forinterpretation (Pulvirenti and Diana, 2001) as this braindisorder is considered by many (Nestler, 1992, 1993, 2001;Pulvirenti and Diana, 2001) to be a chronic drug-inducedalteration in neuronal plasticity at various (molecular, cel-lular, and system) levels.

For these reasons, we began a series of studies, spanningalmost a decade, to investigate some of the effects ofchronic administration of ethanol and its withdrawal on thefunctionality and pharmacological responsiveness of DAneurons projecting to the nucleus accumbens. To this end,we used mostly electrophysiological techniques, such asextracellular recordings coupled with antidromic identifica-tion from terminal fields, and biochemical investigations,such as the microdialysis method and behavioral observa-tions, when possible, to compare various indices suggestiveof chronic ethanol intoxication.

Among the first studies was the demonstration that ratsthat were chronically treated with ethanol, although show-ing tolerance to the ethanol-induced loss of righting reflex,did not show a comparable reduction in the ethanol-activating properties of DA neurons projecting to the nu-cleus accumbens (Diana et al., 1992). Consistently, ethanol-induced stimulation of dopamine outflow in the ventralstriatum was not reduced when rats were pretreated withcomparable doses of chronic ethanol (Rossetti et al., 1993),suggesting different mechanisms for behavioral effects suchas loss of righting reflex and neuronal effects on DA sys-tems with their relevance to the phenomenon of drug de-pendence. Subsequently, the withdrawal syndrome thatemerged after suspension of chronic ethanol treatment wasinvestigated with the intention of dissociating and/or cor-relating DA involvement with behavioral manifestation ofethanol withdrawal and the temporal relationship of thesechanges. It was observed (Diana et al., 1993) that firing rateand burst firing of DA neurons (together with an elonga-tion of the refractory periods) are reduced well belowcontrol levels in parallel with a reduction of dopamineoutflow in the nucleus accumbens, at the time rats display

marked somatic withdrawal signs. Furthermore, the reduc-tion in DA cell firing rate and burst firing does not seem tobe related to somatic withdrawal as electrophysiologicalparameters are still reduced 72 hr after the last ethanoladministration, whereas, at this time, behavioral evaluationdoes not detect differences between ethanol-dependentand control rats (Diana et al., 1996). In addition, an ethanolwithdrawal–induced depolarization block, affecting DAneurons, has been suggested (Shen and Chiodo, 1993).Experiments aimed at clarifying this point, however, haveprovided contrasting results and different interpretations(Diana et al., 1995a).

Collectively, the results suggest that the hypodopaminer-gic status, as indexed by a reduced firing rate, burst firing,and spikes/burst, induced by withdrawal from chronic eth-anol administration, is related to the psychological and notsomatic discomfort that accompanies and outlasts suspen-sion of chronic ethanol administration, as well as otherdrugs of abuse (Diana, 1996, 1998; Diana et al., 1995b).Indeed, the reduction in electrophysiological activity of DAneurons is accompanied by a parallel decrease in extracel-lular concentrations of dopamine, as measured by micro-dialysis (Diana et al., 1993) in the nucleus accumbens. Inaddition, whereas the onset of electrophysiological andbiochemical changes signals the emergence of the somaticsigns of withdrawal, firing rate and burst firing remainreduced well after somatic manifestations have subsided(Diana et al., 1996). Finally, pharmacological interventionsaimed at restoring the functionally deficient DA mesolim-bic system may prove useful in reducing the most harmfulconsequences of alcohol abuse, such as alcohol craving, andthus prevent relapse.

ETHANOL INDUCES PROTEIN KINASE ATRANSLOCATION INTO THE NUCLEUS, CYCLIC AMP

RESPONSE ELEMENT BINDING PROTEINPHOSPHORYLATION, AND INCREASES IN CYCLIC

ADENOSINE MONOPHOSPHATE–DEPENDENT GENEEXPRESSION

Ivan Diamond

The cellular and molecular mechanisms that mediateneural responses to ethanol and sustained ethanol con-sumption include many neurotransmitter systems (Dia-mond and Gordon, 1997). Adenosine is an inhibitory neu-romodulator that also seems to mediate many acute andchronic responses to ethanol (Mailliard and Diamond,2004). Our laboratory has investigated the role of adenosin-ergic signaling in this process. We find that adenosineuptake via an equilibrative nucleoside transporter, ENT1,is inhibited by ethanol. Ethanol-induced increases in extra-cellular adenosine activate A2 receptors to stimulate adenylylcyclase activity, increasing cyclic adenosine monophosphate(cAMP) levels; cAMP activates protein kinase A (PKA).

We have reported that chronic exposure to ethanolcauses sustained translocation of the catalytic subunit of

ETHANOL EFFECTS ON DA VTA NEURONS AND THE MESOLIMBIC PATHWAY 1773

PKA (C�) into the nucleus of NG108-15 cells (Dohrman etal., 1996). We investigated the regulation of ethanol-induced PKA translocation. Ethanol-induced PKA C�translocation occurs in two phases (Dohrman et al., 2002).First, there is a remarkable increase in PKA C� transloca-tion from the Golgi to the cytoplasm and nucleus withinminutes of exposure to ethanol. Translocation peaks at ~10min, and PKA exits the nucleus after 1 hr. This early phaseof PKA translocation is regulated by adenosine A2 recep-tors and requires cAMP activity. Thus, adenosine A2 re-ceptor blockade or Rp-cAMPS blocks ethanol-inducedPKA C� translocation in the first few minutes. This firstphase is followed by a second and more prolonged trans-location, which begins at 6 to 12 hr during chronic exposureto ethanol. Now PKA C� remains in the nucleus as long asethanol is present. If ethanol is withdrawn, then PKA C�leaves the nucleus 12 hr later (Dohrman et al., 1996). Thissecond phase of PKA translocation does not require aden-osine receptors; adenosine A2 receptor blockade is withouteffect. However, the second phase of translocation requiresRNA and protein synthesis. Sustained nuclear localizationof PKA C� is highly unusual and seems to be a character-istic feature of ethanol-induced translocation. This suggeststhat ethanol may induce the expression of a specific bindingprotein that keeps PKA C� in the nucleus.

PKA C� translocated into the nucleus is functionallyactive. cAMP-dependent cAMP response element bindingprotein (CREB) phosphorylation can be demonstrated inthe early phase and reaches a peak from 1 to 3 hr (Con-stantinescu et al., 1999). This is followed by a reduction inCREB phosphorylation, but phosphorylated CREB levelsalways remain higher than controls even after 24 hr ofexposure to ethanol. Blockade of adenosine A2 receptors inthe early phase of PKA C� translocation prevents ethanol-induced CREB phosphorylation. However, adenosine A2receptor blockade has no effect on forskolin-stimulatedadenylyl cyclase activity because forskolin activation by-passes the A2 receptor.

CREB phosphorylation often but not always increasescAMP response element (CRE)-mediated gene expression.NG108-15 cells were transfected with firefly luciferasedriven by CRE as a reporter for CRE-mediated gene ex-pression. When exposed to ethanol, CRE-mediated lucif-erase expression was easily detected hours later (Asher etal., 2002). At 4 hr, the increase in CRE-mediated geneexpression was blocked by an adenosine receptor antago-nist. At 14 hr, however, ethanol-induced CRE-mediatedgene expression was no longer affected by adenosine re-ceptor blockade. However, both early and later phases ofCRE-mediated gene expression required PKA activity.Taken together, these results indicate that ethanol activatesan adenosine A2 receptor-dependent G�s-mediated cAMPsignaling pathway that results in translocation of PKA C�into the nucleus, phosphorylation of CREB, and activationof CRE-mediated gene expression.

The nucleus accumbens is implicated in reinforcement of

alcohol-drinking behavior. Neurons in the nucleus accum-bens characteristically express adenosine A2 and dopamineD2 receptors on the same cells. We created a stable cell lineexpressing adenosine A2 and dopamine D2 receptors(NG108-15/D2) to model nucleus accumbens mediumspiny neurons. Studies with these cells indicate that earlyactivation of dopamine D2 receptors mimics the early phaseof ethanol-induced PKA translocation and gene expression.In addition, there is synergy between dopamine D2 receptorand ethanol/adenosine A2 receptor activation of PKA sig-naling. Subthreshold concentrations of a D2 receptor ago-nist or ethanol, which are ineffective alone, when addedtogether activate PKA translocation (Yao et al., 2002).Furthermore, synergistic activation of PKA translocationinto the nucleus is followed by synergistic activation ofcAMP-dependent gene expression (Yao et al., 2002). Syn-ergy is mediated by the release of �� dimers from Gi, whichstimulate adenylyl cyclase II and IV, setting up a cascade ofmolecular events that persist long after the transient in-crease in cAMP has dissipated.

Taken together, our results suggest that an ethanol-induced increase in cAMP is followed by an early phase ofPKA C� translocation associated with cAMP-dependentgene expression. There is synergy between D2 receptor andethanol/A2 receptor activation of this pathway. Synergybetween dopamine D2 and adenosine A2 receptors requires�� dimers released from Gi/o and adenosine. This may berelated to the role of the nucleus accumbens in ethanolreinforcement and consumption because medium spinyneurons in this region are characterized by the expressionof dopamine D2 and adenosine A2 receptors in the samecells. Thus, inhibiting the action of �� dimers in the nucleusaccumbens reduces voluntary ethanol consumption in ratsin a two-bottle choice paradigm (Yao et al., 2002). Furtherstudies are under way to develop new therapeutic interven-tions to inhibit synergy between dopamine D2 and adeno-sine A2 receptors for the prevention or treatment ofalcoholism.

CO-ACTIVATION OF DOPAMINE D1 AND D2 RECEPTORSIS NECESSARY FOR DOPAMINE-MEDIATED INCREASES

IN FIRING ACTIVITY IN NUCLEUS ACCUMBENSNEURONS

Antonello Bonci

Dopamine in the nucleus accumbens has long been con-sidered an important mediator of addictive behaviors(Robinson and Berridge, 1993; Spanagel and Weiss, 1999).The shell region of the nucleus accumbens seems particu-larly implicated in a number of cellular and behavioralphenomena, especially in relation to addictive drugs (Deu-tch and Zahm, 1992; Zahm, 1999). However, the details onthe functional effects of dopamine in the nucleus accum-bens remain controversial, with different studies findingexcitation or inhibition of medium spiny neurons by dopa-mine (Nicola et al., 2000). Thus, understanding the intra-

1774 APPEL ET AL.

cellular signaling events that mediate DA effects in thenucleus accumbens is crucial for understanding how dopa-mine can produce changes in both cellular and behavioralactivity.

Dopamine receptors are generally grouped into two sub-families, the D1-like receptors and the D2-like receptors(Missale et al., 1998). Opposing influences of dopamine D1and D2 receptors on cAMP-dependent signaling have beenreported in many studies (Stoof and Kebabian, 1981), withD1 receptors acting through Gs-like Golf, and D2 receptorsacting through Gi/o proteins. However, a number of behav-ioral studies have observed that a cooperative interaction ofD1 and D2 receptors can occur in the nucleus accumbens,because animals will self-administer D1 and D2 receptoragonists directly into the nucleus accumbens in combina-tion but not alone (Ikemoto et al., 1997a).

Our experiments were designed to examine the effects ofDA receptor agonists on action potential (spike) firing inmedium spiny neurons in the nucleus accumbens shell andthe cellular mechanisms that might mediate these effects(Hopf et al., 2003). For mimicking the up-state in mediumspiny neurons in our in vitro slice preparation, a series ofcurrent pulses were delivered to medium spiny neurons.Our data show that bath application of dopamine signifi-cantly elevated spike firing. To determine the dopaminereceptor subtypes that are responsible for this increase inspike firing, we applied selective dopamine receptor ago-nists. Bath application of a selective D1 receptor agonistalone (SKF 81297 or SKF 82957) or quinpirole, a selectiveD2 receptor agonist, alone did not alter spike firing. Thesedata using selective agonists suggest that D1 and D2 recep-tors act cooperatively in the nucleus accumbens shell toincrease spike firing. Dopamine-mediated enhancement ofspike firing was prevented by pre-exposure to the D1 re-ceptor antagonist SCH 23390 or the D2 receptor antagonisteticlopride. An important issue is whether D1 and D2 re-ceptor signaling occurs within the same cell or through D1and D2 receptors located on different cells. For addressingthis question, synaptic release was inhibited by the irrevers-ible antagonists of N- and P/Q-type calcium channels cono-toxin and agatoxin, as well as continuous exposure to theL-type antagonist nifedipine. Such exposure completely in-hibited evoked release even 1 hr after exposure to toxins.Nonetheless, exposure to dopamine led to a significantincrease in spike rate. These data suggest that thedopamine-mediated signaling events did not require synap-tic transmission, and thus the effects of D1/D2 receptorco-activation on spike firing likely occurred within the samecell. Several studies suggest that PKA plays a major role inDA signaling (Greengard et al., 1999). To test for a role ofthe cAMP system during the dopamine-induced increase inspike rate, we applied Rp-cAMPS, an inhibitor of cAMP-dependent processes (Hopf et al., 2003). Addition of Rp-cAMPS during the dopamine response significantly re-duced the spike firing rate to near baseline levels. Incontrast, in neurons that were not exposed to Rp-cAMPS,

the dopamine-induced increase in spike rate was main-tained during the 20 min of exposure. Also, Rp-cAMPS hadno effect on basal firing activity, suggesting that the cAMPsystem did not regulate spike rate under basal conditionsbut was required for expression of the increased firing rateduring exposure to dopamine. In support of this possibility,forskolin, an activator of adenylyl cyclases, increased spikefiring to a similar degree as DA receptor agonists, whereasdideoxy-forskolin, an inactive analog of forskolin, had noeffect. Hence, receptor-independent stimulation of the ad-enylyl cyclase signaling cascade was sufficient to increasefiring rate.

One mechanism by which D1 and D2 receptors couldinteract is via stimulation of adenylyl cyclases. G-protein ��subunits (G��), released during activation of Gi/o (e.g., viathe D2 receptor) could act synergistically with G�s-typesubunits (e.g., from the D1 receptor) at specific subtypes ofadenylyl cyclase to activate cAMP-dependent signaling (Su-nahara et al., 1996; Watts and Neve, 1997; Yao et al., 2002).To address whether G�� was required for the activation ofspike firing by DA receptor agonists, we used SP��, aninhibitory peptide that interferes with binding of G�� toseveral targets, and FVIII, an inactive control peptide (Maet al., 1997). Dialysis of neurons with SP�� prevented theincrease of spike firing elicited by the combination of a D1and a D2 receptor agonist, whereas 200 �M of the inactivepeptide FVIII had no effect (Hopf et al., 2003). These datasupport the possibility that the inhibition mediated by SP��is upstream of adenylyl cyclase, because activation of spikefiring by forskolin was not prevented by either SP�� orFVIII. Thus, although effector sites for G�� other thanadenylyl cyclases could be involved in mediating such aresponse, these data indicate that G�� and activation of thecAMP system are required for the increase in spike firingobserved after cooperative activation of D1 and D2 recep-tors. Taken together, these results provide evidence for anovel cellular mechanism through which D1 and D2 recep-tors in the nucleus accumbens could mediate dopamine-dependent behaviors.

CONCLUSIONS

The mesolimbic dopamine pathway is important forthe reinforcing properties of ethanol (Salamone et al.,2003; Weiss et al., 1993; Wise, 1987). The mesolimbicpathway originates from the cell bodies of DA neurons inthe VTA, which project their axons to and release dopa-mine in the nucleus accumbens and other parts of thelimbic system. The first three presentations in this sym-posium all dealt with studies of ethanol effects in theVTA but utilized very different experimental prepara-tions. McBride and colleagues studied intracranial self-administration of ethanol directly into the VTA by freelymoving, behaving rats. These studies showed that thereinforcing effects of ethanol required activity of the DAneurons, as ethanol self-administration was blocked

ETHANOL EFFECTS ON DA VTA NEURONS AND THE MESOLIMBIC PATHWAY 1775

when ethanol was co-infused with quinpirole, a DA re-ceptor agonist that acts at D2 autoreceptors on DA VTAneurons to inhibit them. In addition, greater sensitivityto the reinforcing effects of ethanol in the VTA wascorrelated with high alcohol preference, and P rats alsoshowed behavioral changes, which could indicate thatsensitization or tolerance developed to the reinforcingeffects of ethanol. Such development of sensitization ortolerance to the reinforcing effects of ethanol within theVTA may be a factor that contributes to high alcoholdrinking. In this regard, it is interesting that a recentelectrophysiological study found sensitization to theacute excitatory effect of ethanol on DA VTA neurons inbrain slices from C57BL/6J mice, which had receivedchronic intermittent treatment with ethanol (Brodie,2002).

The in vitro electrophysiological studies by Brodie andAppel showed that acute application of ethanol directlyexcites the DA neurons in the VTA and that the mecha-nism that underlies this excitatory effect seems to be re-duction of current through delayed rectifier K� channels,which open during the action potential AHP. The ethanolexcitation of DA VTA neurons could be blocked by quin-idine. These studies point to a native ethanol-sensitive K�

channel on DA VTA neurons as a novel target formodulating the reinforcing properties of ethanol andfor new pharmacotherapeutic approaches to the treat-ment of alcoholism.

The in vivo electrophysiological studies by Diana showedthat chronic ethanol treatment of rats did not cause toler-ance to ethanol activation of DA VTA neurons projectingto the nucleus accumbens or to dopamine outflow in theventral striatum measured by microdialysis. During thewithdrawal period after chronic treatment, there was areduction in the spontaneous firing rate and burst firing ofDA VTA neurons. The hypodopaminergic status of themesolimbic pathway, which accompanies and outlasts eth-anol withdrawal after chronic treatment in these animalstudies, could underlie the dysphoria and psychologicaldiscomfort seen in human alcoholics during and afterdetoxification.

DA VTA neurons project their axons in the mesolimbicpathways to release dopamine in the nucleus accumbens,and this release is increased by ethanol excitation in theVTA. The last two presentations in this symposium dealtwith different aspects of the postsynaptic effects of do-pamine on the target cells in the nucleus accumbens, themedium spiny neurons. Bonci’s electrophysiologicalstudies in brain slices showed that dopamine increasesthe firing rate of medium spiny neurons in the nucleusaccumbens; it is interesting that co-activation of both D1and D2 receptors is needed for this effect. The increasedfiring rate was due to increased cAMP resulting from acooperative effect of D1 and D2 receptors on adenylylcylase. These studies indicate a novel mechanism inwhich D1 receptors act through G�s subunits and D2

receptors act through G�� subunits to cooperativelyactivate the cAMP system in nucleus accumbens mediumspiny neurons.

The studies by Diamond looked at effects of ethanol,adenosine, and dopamine on signaling and gene expres-sion downstream from cAMP. Their work in NG108-15cells showed that chronic ethanol exposure causes asustained translocation of the catalytic subunit of PKAinto the nucleus, phosphorylation of CREB, and activa-tion of CRE-mediated gene expression. In a stable cellline expressing adenosine A2 and dopamine D2 receptors(NG108-15/D2) as a model of nucleus accumbens me-dium spiny neurons, Diamond demonstrated synergy be-tween D2 receptor and ethanol/A2 receptor activation ofPKA translocation and CRE-mediated gene expression.�� dimers released from Gi/o are required for synergy inDiamond’s studies and for cooperativity between dopa-mine D1 and D2 receptors in Bonci’s studies. Thesefindings may be related to Diamond’s observation thatexpression of �� inhibitors in the nucleus accumbenscauses a reduction in voluntary ethanol consumption.This work could be relevant to understanding the roleof the nucleus accumbens in the reinforcing effects ofethanol. Furthermore, it is tempting to hypothesizethat G�� might also play a role in modulating neuro-physiological and behavioral mechanisms underlying thepathogenesis of craving for alcohol and other addictiveagents.

In conclusion, DA neurons of the VTA project to thenucleus accumbens. The work presented in this symposiumhas shown that acutely administered ethanol directly ex-cites DA VTA neurons through a specific mechanism andthat the maintenance of intracranial self-administration ofethanol into the VTA depends on activation of those DAneurons. Furthermore, chronic ethanol induces changes inthe physiology of DA VTA neurons, which may be relatedto relapse to drinking, as these changes outlast signs ofphysical dependence. The result of ethanol-induced activa-tion of DA VTA neurons is an increased output of dopa-mine onto medium spiny neurons of the nucleus accum-bens, where dopamine exerts a cooperative action at D1

and D2 receptors to increase the firing of the medium spinyneurons; this action involves G�� dimers. Ethanol anddopamine activation of D2 receptors synergize in a modelof medium spiny neurons to cause PKA translocation andCRE-mediated gene expression, an action that also in-volves �� dimers. Plastic changes in second messengersystems in the nucleus accumbens may play a role in relapseto drinking, as �� inhibitors in the nucleus accumbensreduce voluntary ethanol intake. Because of the impor-tance of the VTA–nucleus accumbens axis to the reinforc-ing effects of ethanol, any point on this circuit may serve asa potential target for therapeutic intervention for the treat-ment of alcoholism.

1776 APPEL ET AL.

REFERENCES

Appel SB (2004) Role of voltage-gated potassium channels in mediationof the direct excitatory action of ethanol on mesolimbic dopamineneurons. World J Biol Psychiatry (Suppl 1) 5:19.

Appel SB, Liu ZP, McElvain MA, Brodie MS (2003) Ethanol excitation ofdopaminergic ventral tegmental area neurons is blocked by quinidine.J Pharmacol Exp Ther 306:437–446.

Asher O, Cunningham TD, Yao L, Gordon AS, Diamond I (2002) Eth-anol stimulates cAMP-responsive element (CRE)-mediated transcrip-tion via CRE-binding protein and cAMP-dependent protein kinase.J Pharmacol Exp Ther 301:66–70.

Berridge KC, Robinson TE (2003) Parsing reward. Trends Neurosci 26:507–513.

Brodie MS (2002) Increased ethanol excitation of dopaminergic neuronsof the ventral tegmental area after chronic ethanol treatment. AlcoholClin Exp Res 26:1024–1030.

Brodie MS, Appel SB (1998) The effects of ethanol on dopaminergicneurons of the ventral tegmental area studied with intracellular record-ing in brain slices. Alcohol Clin Exp Res 22:236–244.

Brodie MS, Liu ZP, Yu B, Appel SB (2000) Ethanol decreases a sustainedpotassium current in rat dopaminergic neurons of the ventral tegmentalarea. Soc Neurosci Abstr 26:785.

Brodie MS, McElvain MA, Bunney EB, Appel SB (1999a) Pharmacolog-ical reduction of small conductance calcium-activated potassium current(SK) potentiates the excitatory effect of ethanol on ventral tegmentalarea dopamine neurons. J Pharmacol Exp Ther 290:325–333.

Brodie MS, Pesold C, Appel SB (1999b) Ethanol directly excites dopami-nergic ventral tegmental area reward neurons. Alcohol Clin Exp Res23:1848–1852.

Brodie MS, Shefner SA, Dunwiddie TV (1990) Ethanol increases thefiring rate of dopamine neurons of the rat ventral tegmental area invitro. Brain Res 508:65–69.

Brodie MS, Trifunovic RD, Shefner SA (1995) Serotonin potentiatesethanol-induced excitation of ventral tegmental area neurons in brainslices from three different rat strains. J Pharmacol Exp Ther 273:1139–1146.

Campbell AD, Kohl RR, McBride WJ (1996) Serotonin-3 receptor andethanol-stimulated somatodendritic dopamine release. Alcohol 13:569–574.

Cannon CM, Palmiter RD (2003) Reward without dopamine. J Neurosci23:10827–10831.

Constantinescu A, Diamond I, Gordon AS (1999) Ethanol-induced trans-location of cAMP-dependent protein kinase to the nucleus. Mechanismand functional consequences. J Biol Chem 274:26985–26991.

Czachowski CL, Chappell AM, Samson HH (2001) Effects of raclopride inthe nucleus accumbens on ethanol seeking and consumption. AlcoholClin Exp Res 25:1431–1440.

Czachowski CL, Santini LA, Legg BH, Samson HH (2002) Separatemeasures of ethanol seeking and drinking in the rat: Effects of remo-xipride. Alcohol 28:39–46.

Dahlstrom A, Fuxe K (1964) Evidence for the existence of monoaminecontaining neurons in the central nervous system. I. Demonstration ofmonoamines in the cell bodies of brain stem neurons. Acta PhysiolScand 62:1–55.

Deutch AY, Zahm DS (1992) The current status of neurotensin-dopamineinteractions. Issues and speculations. Ann N Y Acad Sci 668:232–252.

De Wit H, Wise RA (1977) Blockade of cocaine reinforcement in rats withthe dopamine receptor blocker pimozide, but not with the noradrener-gic blockers phentolamine and phenoxybenzamine. Can J Pharmacol31:195–203.

Diamond I, Gordon AS (1997) Cellular and molecular neuroscience ofalcoholism. Physiol Rev 77:1–20.

Diana M (1996) Dopaminergic neurotransmission and drug withdrawal:Relevance to drug craving, in Advances in Behavioral Biology 47: TheBasal Ganglia V (Ohye C, Kimura M, McKenzie J eds), pp 123–130.Plenum Press, New York.

Diana M (1998) Drugs of abuse and dopamine cell activity, in Advances inPharmacology 42: Catecholamine Bridging Basic Science With ClinicalMedicine (Goldstein DS, Eisenhofer G, McCarty R eds), pp 998–1001.Academic Press, San Diego.

Diana M, Gessa GL, Rossetti ZL (1992) Lack of tolerance to ethanol-induced stimulation of mesolimbic dopamine system. Alcohol Alcohol27:329–333.

Diana M, Pistis M, Carboni S, Gessa GL, Rossetti ZL (1993) Profounddecrement of mesolimbic dopaminergic neuronal activity during etha-nol withdrawal syndrome in rats: Electrophysiological and biochemicalevidence. Proc Natl Acad Sci U S A 90:7966–7969.

Diana M, Pistis M, Muntoni A, Gessa G (1996) Mesolimbic dopaminergicreduction outlasts ethanol withdrawal syndrome: Evidence of pro-tracted abstinence. Neuroscience 71:411–415.

Diana M, Pistis M, Muntoni AL, Gessa GL (1995a) Ethanol withdrawaldoes not induce a reduction in the number of spontaneously activedopaminergic neurons in the mesolimbic system. Brain Res 682:29–34.

Diana M, Rossetti ZL, Gessa GL (1995b) Central dopaminergic mecha-nisms of alcohol and opiate withdrawal syndromes, in Drug Addictionand Related Clinical Problems (Tagliamonte A, Maremmani I eds), pp19–26. Springer-Verlag Wien, New York.

Di Chiara G (2002) Nucleus accumbens shell and core dopamine: Differ-ential role in behavior and addiction. Behav Brain Res 137:75–114.

Di Chiara G, Imperato A (1988) Drugs abused by humans preferentiallyincrease synaptic dopamine concentrations in the mesolimbic dopaminesystem of freely moving rats. Proc Natl Acad Sci U S A 85:5274–5278.

Dohrman DP, Chen HM, Gordon AS, Diamond I (2002) Ethanol-inducedtranslocation of protein kinase A occurs in two phases: Control bydifferent molecular mechanisms. Alcohol Clin Exp Res 26:407–415.

Dohrman DP, Diamond I, Gordon AS (1996) Ethanol causes transloca-tion of cAMP-dependent protein kinase catalytic subunit to the nucleus.Proc Natl Acad Sci U S A 93:10217–10221.

Doyon WM, York JL, Diaz LM, Samson HH, Czachowski CL, GonzalesRA (2003) Dopamine activity in the nucleus accumbens during con-summatory phases of oral ethanol self-administration. Alcohol Clin ExpRes 27:1573–1582.

Gatto GJ, McBride WJ, Murphy JM, Lumeng L, Li TK (1994) Ethanolself-infusion into the ventral tegmental area by alcohol-preferring rats.Alcohol 11:557–564.

Gessa GL, Muntoni F, Collu M, Vargiu L, Mereu G (1985) Low doses ofethanol activate dopaminergic neurons in the ventral tegmental area.Brain Res 348:201–203.

Greengard P, Allen PB, Nairn AC (1999) Beyond the dopamine receptor:The DARPP-32/protein phosphatase-1 cascade. Neuron 23:435–447.

Grenhoff J, Janson AM, Svensson TH, Fuxe K (1991) Chronic continuousnicotine treatment causes decreased burst firing of nigral dopamineneurons in rats partially hemitransected at the mesodiencephalic junc-tion. Brain Res 562:347–351.

Hopf FW, Cascini MG, Gordon AS, Diamond I, Bonci A (2003) Coop-erative activation of dopamine D1 and D2 receptors increases spikefiring of nucleus accumbens neurons via G-protein �� subunits. J Neu-rosci 23:5079–5087.

Ikemoto S, Glazier BS, Murphy JM, McBride WJ (1997a) Role of dopa-mine D1 and D2 receptors in the nucleus accumbens in mediatingreward. J Neurosci 17:8580–8587.

Ikemoto S, Murphy JM, McBride WJ (1997b) Self-infusion of GABA(A)antagonists directly into the ventral tegmental area and adjacent re-gions. Behav Neurosci 111:369–380.

Ikemoto S, Murphy JM, McBride WJ (1998) Regional differences withinthe rat ventral tegmental area for muscimol self-infusions. PharmacolBiochem Behav 61:87–92.

Imperato A, Di Chiara G (1986) Preferential stimulation of dopaminerelease in the nucleus accumbens of freely moving rats by ethanol.J Pharmacol Exp Ther 239:219–228.

Koob GF, Bloom FE (1988) Cellular and molecular mechanisms of drugdependence. Science 242:715–723.

ETHANOL EFFECTS ON DA VTA NEURONS AND THE MESOLIMBIC PATHWAY 1777

Levy AD, Murphy JM, McBride WJ, Lumeng L, Li TK (1991) Microin-jection of sulpiride into the nucleus accumbens increases ethanol drink-ing in alcohol-preferring (P) rats. Alcohol Alcohol Suppl 1:417–420.

Ma JY, Catterall WA, Scheuer T (1997) Persistent sodium currentsthrough brain sodium channels induced by G protein �� subunits.Neuron 19:443–452.

Mailliard WS, Diamond I (2004) Recent advances in the neurobiology ofalcoholism: The role of adenosine. Pharmacol Ther 101:39–46.

McBride WJ, Murphy JM, Ikemoto S (1999) Localization of brain rein-forcement mechanisms: Intracranial self-administration and intracra-nial place-conditioning studies. Behav Brain Res 101:129–152.

Missale C, Nash SR, Robinson SW, Jaber M, Caron MG (1998) Dopa-mine receptors: From structure to function. Physiol Rev 78:189–225.

Nestler EJ (1992) Molecular mechanisms of drug addiction. J Neurosci12:2439–2450.

Nestler EJ (1993) Cellular responses to chronic treatment with drugs ofabuse. Crit Rev Neurobiol 7:23–39.

Nestler EJ (2001) Molecular basis of long-term plasticity underlying ad-diction. Nat Rev Neurosci 2:119–128.

Nicola SM, Surmeier J, Malenka RC (2000) Dopaminergic modulation ofneuronal excitability in the striatum and nucleus accumbens. Annu RevNeurosci 23:185–215.

Nowak KL, McBride WJ, Lumeng L, Li TK, Murphy JM (2000) Involve-ment of dopamine D2 autoreceptors in the ventral tegmental area onalcohol and saccharin intake of the alcohol-preferring P rat. AlcoholClin Exp Res 24:476–483.

Oades RD, Halliday GM (1987) Ventral tegmental (A10) system: Neuro-biology. 1. Anatomy and connectivity. Brain Res Rev 12:117–165.

Olds J, Milner P (1954) Positive reinforcement produced by electricalstimulation of septal area and other regions of rat brain. J Comp PhysiolPsychol 47:419–427.

Olds ME, Fobes JL (1981) The central basis of motivation: Intracranialself-stimulation studies. Annu Rev Psychol 32:523–574.

Pecina S, Cagniard B, Berridge KC, Aldridge JW, Zhuang X (2003)Hyperdopaminergic mutant mice have higher “wanting” but not “lik-ing” for sweet rewards. J Neurosci 23:9395–9402.

Pfeffer AO, Samson HH (1985) Oral ethanol reinforcement: Interactiveeffects of amphetamine, pimozide and food-restriction. Alcohol DrugRes 6:37–48.

Pfeffer AO, Samson HH (1986) Effect of pimozide on home cage ethanoldrinking in the rat: Dependence on drinking session length. DrugAlcohol Depend 17:47–55.

Pfeffer AO, Samson HH (1988) Haloperidol and apomorphine effects onethanol reinforcement in free feeding rats. Pharmacol Biochem Behav29:343–350.

Pulvirenti L, Diana M (2001) Drug dependence as a disorder of neuralplasticity: Focus on dopamine and glutamate. Rev Neurosci 12:141–158.

Robinson TE, Berridge KC (1993) The neural basis of drug craving: Anincentive-sensitization theory of addiction. Brain Res Brain Res Rev18:247–291.

Robinson TE, Berridge KC (2003) Addiction. Annu Rev Psychol 54:25–53.

Rodd ZA, Bell RL, Melendez RI, Kuc KA, Lumeng L, Li T-K, Murphy JM,McBride WJ (2004a) Comparison of intracranial self-administration ofethanol within the posterior ventral tegmental area between alcohol-preferring and Wistar rats. Alcohol Clin Exp Res 28:1212–1219.

Rodd ZA, Melendez RI, Bell RL, Kuc KA, Zhang Y, Murphy JM,McBride WJ (2004b) Intracranial self-administration of ethanol withinthe ventral tegmental area of male Wistar rats: Evidence for involve-ment of dopamine neurons. J Neurosci 24:1050–1057.

Rodd-Henricks ZA, McKinzie DL, Crile RS, Murphy JM, McBride WJ(2000) Regional heterogeneity for the intracranial self-administration ofethanol within the ventral tegmental area of female Wistar rats. Psy-chopharmacology (Berl) 149:217–224.

Rossetti ZL, Hmaidan Y, Diana M, Gessa GL (1993) Lack of tolerance toethanol-induced dopamine release in the rat ventral striatum. EurJ Pharmacol 231:203–207.

Salamone JD, Correa M (2002) Motivational views of reinforcement:Implications for understanding the behavioral functions of nucleusaccumbens dopamine. Behav Brain Res 137:3–25.

Salamone JD, Correa M, Mingote S, Weber SM (2003) Nucleus accum-bens dopamine and the regulation of effort in food-seeking behavior:Implications for studies of natural motivation, psychiatry, and drugabuse. J Pharmacol Exp Ther 305:1–8.

Samson HH, Tolliver GA, Haraguchi M, Hodge CW (1992) Alcoholself-administration: Role of mesolimbic dopamine. Ann N Y Acad Sci654:242–253.

Samson HH, Tolliver GA, Schwarz-Stevens K (1990) Oral ethanol self-administration: A behavioral pharmacological approach to CNS controlmechanisms. Alcohol 7:187–191.

Schultz W (2001) Reward signaling by dopamine neurons. Neuroscientist7:293–302.

Schultz W (2002) Getting formal with dopamine and reward. Neuron36:241–263.

Shen R-Y, Chiodo LA (1993) Acute withdrawal after repeated ethanoltreatment reduces the number of spontaneously active dopaminergicneurons in the ventral tegmental area. Brain Res 622:289–293.

Spanagel R, Weiss F (1999) The dopamine hypothesis of reward: Past andcurrent status. Trends Neurosci 22:521–527.

Spyraki C, Fibiger HC, Phillips AG (1982) Dopaminergic substrates ofamphetamine-induced place preference conditioning. Brain Res 253:185–193.

Spyraki C, Fibiger HC, Phillips AG (1983) Attenuation of heroin rewardin rats by disruption of the mesolimbic dopamine system. Psychophar-macology (Berl) 79:278–283.

Stoof JC, Kebabian JW (1981) Opposing roles for D-1 and D-2 dopaminereceptors in efflux of cyclic AMP from rat neostriatum. Nature 294:366–368.

Sunahara RK, Dessauer CW, Gilman AG (1996) Complexity and diversityof mammalian adenylyl cyclases. Annu Rev Pharmacol Toxicol 36:461–480.

Watts VJ, Neve KA (1997) Activation of type II adenylate cyclase by D2and D4 but not D3 dopamine receptors. Mol Pharmacol 52:181–186.

Weiss F, Lorang MT, Bloom FE, Koob GF (1993) Oral alcohol self-administration stimulates dopamine release in the rat nucleus accum-bens: Genetic and motivational determinants. J Pharmacol Exp Ther267:250–258.

Weiss F, Mitchiner M, Bloom FE, Koob GF (1990) Free-choice respond-ing for ethanol versus water in alcohol preferring (P) and unselectedWistar rats is differentially modified by naloxone, bromocriptine, andmethysergide. Psychopharmacology (Berl) 101:178–186.

Wise RA (1987) The role of reward pathways in the development of drugdependence. Pharmacol Ther 35:227–263.

Wise RA (2002) Brain reward circuitry: Insights from unsensed incentives.Neuron 36:229–240.

Wise RA (2004) Dopamine, learning and motivation. Nat Rev Neurosci5:483–494.

Wise RA, Hoffman DC (1992) Localization of drug reward mechanismsby intracranial injections. Synapse 10:247–263.

Yao L, Arolfo MP, Dohrman DP, Jiang Z, Fan P, Fuchs S, Janak PH,Gordon AS, Diamond I (2002) �� Dimers mediate synergy of dopamineD2 and adenosine A2 receptor-stimulated PKA signaling and regulateethanol consumption. Cell 109:733–743.

Yokel RA, Wise RA (1976) Attenuation of intravenous amphetaminereinforcement by central dopamine blockade in rats. Psychopharmacol-ogy (Berl) 48:311–318.

Zahm DS (1999) Functional-anatomical implications of the nucleusaccumbens core and shell subterritories. Ann N Y Acad Sci 877:113–128.

1778 APPEL ET AL.