Neural Mechanisms of Drug Reinforcement" GEORGE F. KOOB lkprtment of Neumpbamacoology The Sm@s Resmtzh InstitUa: 10666 North Tmvey Pines Road Lajolla, calrfirrnia 92037 INTRODUCTION Definitions of dependence and addiction emphasize two important phenomena: a compulsion to take the drug with a loss of control in limiting intake-, and a chaxacteristic withdrawal syndrome that results in physical signs as well as motivational signs of discomfort when the drug is removed. The search fbr neurobiologicalsubstrates b r these phenomena has depended on the development of animal models, both fbr the acute reinforcing (or rewarding) effects of drugs, as well as for the withdrawal syndromes associated with the removal of the drug after chronic access or administration. The acute reinforcingproperties of drug can be considered to be subsumed under the theoretical construct of the positive reinbrcing action of the drug, while the motivational properties of drug withdrawal can be subsumed under the theoretical construct of the negative reinbrcing actions of the drug. Recent studies using animal models have brought the search fbr a neurobiological substrate for drug dependence to a fbcus on the medial fbrebnh and its connections with the region of the nucleus accumbens in the anterior part of the basal brebrain. The nucleus accumbens is well situated to integrate limbic hnction with the extrapyramidal motor system, and appears to play a critical role in mediating not only the acute reinforcing effects of drugs but also may be involved in the motivational aspects of drug withdrawal. The present manuscript will explore studies directed at elucidating the neural substrates b r cocaine, opiate, and ethanol reinforcement.

NEURAL-S

FOR THE ACUTE REINFORCING EFFECTS OF COCAINE

Animal models fbr the acute reinbrcing effects of psychomotor stimulants have included measures of the e5xts of psychomotor stimulants on reward thresholds using intracranial self-stimulation, measures of preference b r the environment paired with drug administration (place prekrence) and direct self-administration of the drug. In

Prrpmtion of this chapter was supporred in part by NIAAA Specialized Center Grant AA NIAAA Grant AA 08459, NIDA Grant DA 04043, and NIDA Grant DA 04398.

06420,

171

172

ANNALS

NEW YORK ACADEMY OF SCIENCES Total Number of Injections

.-

I 0

I 1 111111II I

gJ1

I I I

I 2

I I I I I

I l l I l l

I l l

I

3 36

1 1 I l l I l l I

F

Rat

I I I 1 I

36

I 1 1 1 1 1 1 1 I I I I I 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

37

I

0

I I I

I l l I 1

I I I I I I II I

I

I

2

1

I

3

Hours FIGURE 1. Rcpmentative response records for two rats self-administering cocaine. Test scssions were 3 h in duration. Each mark repments a response/infusion of intravenous drug (0.75 mgkghnjection). (liken with permission from Koob, 1991.96)

studies involving direct self-administrationof the drugs, rats with limited access to cocaine (3 h/day) will show a stable and regular drug intake over each daily session (FIG.1). In addition, no obvious tolerance or dependence develops in these limited access ( 0.05). m e n with permission from KDob ct d.,1987a.10)

SCH 23390

-

1

2

Hours

3

1

2

Hours

3

174

ANNALS NEW YORK ACADEMY OF SCIENCE3

the nudeus accumbens and the substantia innominata-ventral pallidum. To test the hypothesis that the processing of the reinhrcing propemes of cocaine may also involve the substantia innominata-vend pahdum, rats trained to intravenously selfadminister cocaine received bilateral ibotenic acid lesions of the +on of the substantia innominata-vend pallidum.l4 The substantia innominata-ventral pallidum lesions si@cantly deceased baseline cocaine self-administration,and when the rats were subjected to a pcognssive ratio procedure, these lesions produced a significant decrease in the highest ratio obtained h r cocaine.“ These results suggest that the substantia innominata-ventral pallidum may be an important site for the processing of the reinhrcing effects of cocaine.

NEURAL S U B S T R A . FOR THE NEGATIVE REINFORCING PROPERTIES OF COCAINE WITHDRAWAL Following prolongcd use of cocaine on a “bins” abstinence is characterized by severe depressive symptoms combined with irritability and anxiety.15J6 These symp toms characterize the “crash” associated with the abstinence fiom cocaine in cocaine dependence and may last several hours to several days. Anhedonia is one of the more salient depressive symptoms and can be defined as the inability to derive pleasure fiom normally pleasurable stimuli. A means of assessing anhedonia in animals has involved the use of measures of threshold fbr intracranial self-stimulation. Thresholds for intracranial self-stimulation have been hypothesized to reflect the hedonic state of an animal because animals will readily self-administer the stimulation to their own brains, and intracrand selfstimulation is thought to activate the same neural substrates that mediate the reinhrcing eficts of natural reinhrcers (cg.,water, food).I7 Cocaine injected acutely, as well as other psychomotor stimulants, has been well documented to lower self-stimulation thresholds in rats.18 To explore the possibility that prolonged self-administrationof cocaine may result in an increase in brain stimulation thresholds, animals were allowed to self-administer cocaine intravenously for long periods and reward thresholds were monitored during the coursc of cocaine withdrawal. Animals prepared with both chronic indwelling brain stimulation electrodes and chronic indwelling catheters were allowed to selfadminister cocaine fbr various time periods and tested fbr brain stimulation thresholds during cocaine withdrawal. During cocaine withdrawal brain stimulation reward thresholds were elevated compared to prcdrug baseline levels, and the magnitude and duration of the elevation in reward thmholds was proportional to the amount of cocaine self-administered (FIG.4). This elevation in reward threshold may reflect an “anhedonic” state and as such it may be homologous to the anhedonia reported by human drug users following a cocaine binge.19 In addition, these results suggcst that cocaine can alter the fbnction of the reward system@)in the medial hrebrain bundle during the coursc of a cocaine bout and withdrawal. A likely ncumhemid mechanism involved in this withdrawal state would be some hypoactivityof dopamine fimctioning.20 There is some direct support for this hypothesis in studies of in PiPo microdialysis during cocaine withdrawal (Weiss ct al. ,72 1992; this volume). However, there may also be separate neurochemical P’OCCSSCSthat o p

175

KOOB: NEUROBIOLOGY OF DRUG DEPENDENCE 48 Hours Cocaine

250

150 100

0. I

0 I

I

I

I

I

1

1

24 Hours Cocaine 150 100

12 Hours Cocaine

6 Hours Cocaine

2001

"'t oo

100

*oat

3 Hours Cocaine 0

Control Cocaine

100 0

I

1

I

1

I

I

L

1

3

6

12

24

40

72

Hours Post Cocaine FIGURE4. Intracranialself-stimulation thresholds following0,3,6,12,24 and 48 h of cocaine self-administrationa t several time points postcocaine (0, 1,3,6, 12,24,48 and 72 h). The results are expressed as percent change from baseline thrcshold levels. The mean f SEM baseline threshold for the experimental group was 37.4 f 2.5 pA and for the control group 35.9 f 3.1 pA. The asterisks indicate statistically significant diflkrences (p < 0.05) between control and experimental groups with Dunnett's tests, followinga significant group x hours interaction in an analysis of variance. @&en with permission from Markou and Kmb, 1991.19)

176

ANNALS NEW YORK ACADEMY OF SCIENCES

pose the acute action of the drug, such as changes in a nucleus accumbens output system that may be overactive during cocaine withdrawal. In summary, the acute reinforcing effects of cocaine appear to depend on the presynaptic release of dopamine within the region of the nucleus accumbens. Both D-1 and D-2 dopamine receptors may be important. Chronic access to cocaine produces a withdrawal state as reflected in increases in brain stimulation reward thresholds that appear to be opposite to the actions of the drug administered acutely. These effects are thought to reflect a change in the activity of neural elements in the medial forebrain bundle involved with the positive reinforcing effects of cocaine and thus may be responsible for the negative reinforcing state associated with the anhedonia of cocaine withdrawal.

NEURAL SUBSTRATES FOR THE ACUTE REINFORCING EFFECTS OF OPIATES Opiate drugs such as heroin, if provided in limited access, are readily selfadministered intravenously by rats much like cocaine (FIG. 5). Rats will maintain stable levels of drug intake on a daily basis without any major signs of physical dependence.21 This behavior probably best reflects the human equivalent of “chipping,”22 and has been used to study the neurobiologic basis of heroin reinforcement independent of the confounds of dependence. As with cocaine, decreases in the dose of heroin available to the animal will change the pattern of self-administrationsuch that the interinjection interval decreases and the number of injections increases8 Similar increases in the number of injections have been obtained by both systemic and central administration of competitive opiate antag0nists,4JlJ~-~~ suggesting that the animals attempt to compensate for the opiate antagonism by increasing the amount of drug injected. A systematicseries of studies exploring the opiate receptor subtype important for the reinforcing actions ofopiates using selectiveopiate agonists and antagonists suggest that p receptors play an important, if not critical, role in opiate reinforcement. Mu opioid agonists produce dose-dependent decreases in heroin self-administration and irreversible p selective antagonists dosedependently increase heroin self-administration.26 To explore the location of opioid receptors in the central nervous system important fbr the reinforcing properties of heroin, a series of studies was initiated using intracerebral injection ofa quaternary derivative of naloxone, methylnaloxonium. Methylnaloxonium is charged and hydrophilic and does not readily spread from the sites in the brain at which it is injected.27 Intracerebroventricular administration of methylnaloxonium dosedependently increased heroin self-administration in non-dependent rats (FIG. 6). The region of the nucleus accumbens appeared to be particularly sensitive to the effects of methylnaloxonium on heroin self-administration28 (FIG. 7). Injections of methylnaloxonium into the ventral tegmental area produced increases in heroin self-administration only at doses similar to those required by the intracerebroventricular route. These results suggested that neural elements in the region of the nucleus accumbens are responsible for both the reinforcing properties of opiates and cocaine. Rats will also self-administer opioid peptides in the region of the nucleus accumbens.29 The exact neurochemical mechanisms mediating the reinforcing effect of opiates in the nucleus accumbens is unknown but some evidence exists to suggest that it is in-

KOOB: NEUROBIOLOGY OF DRUG DEPENDENCE

ZEE# c

t2

8ZE#

177

178

ANNALS

.

400. A

I

NEW YOaK ACADEMY O F SCIENCES

.

300 * 200-

-

I

0.0

i P

FIGURE 6. The eflkcts of intracerebroven-

1

methylnaloxonium treatment on Rsponding for heroin over the first hour (A) and over the total 3-h self-administrationsession (B). Response rates weE expressed as the percentage ofbaseline responding.Asterisks indicate that the treatment dose was significantly &rrnt from the saline treatment, p < 0.05, Newman-Keuls test. S ix rats were tested acms all drug treatments. The day prior to ICV injections was used as the baseline day. m e n with permission from Vaccarino ct al., tncular

0.5 Dose (microgrsms)

300

I'1

0

mn Luu-

100-

1985.25)

I

rt

0.0

0.5 1.0 2.0 Dore (micrograms1

4.0

dependent of dopamine release. Rats trained to self-administer cocaine and heroin on alternate days and receiving 6-OHDA lesions of the nucleus accumbens showed a timedependent decrease or extinction of cocaine self-administmion, whereas heroin selfadministration returned to near normal 1evels.JO Similar results have been obtained using chronic dopamine receptor blockade and places preference 6 r heroin and selfadministrationJ1 (see this volume). However, opioids are self-administered directly into the source of the mesocorticolimbicdopamine system, the ventral tegmental area (VTA), and microinjections of opioids into the VTA lower brain stimulation reward thresholds and produce robust place prefrences.32.33 The place preferences produced by opioids appear to have a major dopaminergic component31~32(see both, this volume). Thus, the reinforcing actions of opiates may involve both a dopaminedependent (VIA) and dopamine-independent (nucleus accumbens) mechanism. There may be an interdependence of the neural substrates for the reinfbrcing stimuli of cocaine and heroin at the level of the nucleus accumbens post-synaptic to the dopamine innervation and at the level of the substantia innominata-ventral pallidum. Kainic acid lesions of the nudeus accumbens decrease both cocaine and heroin selfadministration,Mand ibotenic acid lesions of the substantia innominata-ventral pdlidum also decrease heroin self-administrati~n.~~ These results suggest that the substantia innominata-ventral pallidum may be an important part of the neural circuitry involved in the pcocessing of the reinforcing effects of both cocaine and heroin (FIG.8). The pcocessing of drug-reinforcing stimuli beyond the substantia innominataventral pallidum is unknown at this time. Evidence fiom locomotor activity studies

KOOB: NEUROBIOLOGY OP DRUG DEPENDENCE

179 1 Hour

PIGURE 7. Percent baseline (prei x IV heroin drug day) responding f during the first hour (ropppb)and for the total 3 hours (bottmnppb) of the heroin self-administration session followingmethylnaloxonium injections into the nucleus accumbens. Asterisks indicate a significant difference (p < 0.05)from saline vehicle (0.0 dose), Duncan Multiple Range II posmia' m. (Taken with permission from Vaccarino et d.,1985.26)

11

'I 0.0

f

,125

.25

-3.5

Methyl Naloxonium Chloride Dose ipgl

2-

3 Hours

-

-

160 140 120 -

111

'L 1

100 -

20

0.0

.125

0.25

Methyl Naloxonium Chloride Dose bg)

suggests roles for both the dorsomedial thalamus and the pedunculopontine nucleus in psychomotor stimulant activation (FIG. 8). Limbic afferents to the nucleus accumbens include major projections from the frontal cortex, amygdala and hippocampus.35 Thus, direct limbic infbrmation received by the nucleus accumbens may be directed to the appropriate motor groups to produce motivated behavior via the pallido-thalamic projections and perhaps thalamo-cortical projections (FIG. 8).

NEURAL SUBSTRATES FOR THE NEGATIVE REINFORCING PROPERTIES OF OPIATE WITHDRAWAL Dependence on opiate drugs is defined by a characteristic withdrawal syndrome that appears with the abrupt termination of opiate administration or can be precipitated with administration of competitive opiate antagonists. Opiate withdrawal in humans is characterized by both physical and motivational symptoms such as nausea, gastrointestinal disturbances, chills, sympathetic reactions, and a painful flu-like dysphoric state.% In rats, opiate physical dependence has been characterized by an abstinence

180

ANNALS NEW YORK ACADEMY OF SCIENCES

FIGURE 8. Schematic model of brain sites and circuitry that participate in the reinforcing and adaptive opposing actions of opiates and psychostimulants. The region of the nucleus accumbens (N ACC)is the target of a dopaminergic projection from the ventral tegmental a m (VIA) and of afferents from olfactory cortex (Olf Ctx) and from limbic cortex. The nucleus accumbens projects, among other targets, to the ventral pallidum (V Pall), and also sends a reciprocal connection, believed to be GABA-mediated, back to the VTA. From the ventral pallidum, connections project to the pedunculopontine nucleus (PPN)and to the dorsal medial thalamus (DMT), which have been proposed as being functionally important in motor activation in the rat.97 T h e ventral pallidum may also regulate responsivenessof neurons in the frontal cortex (FC), a site from which psychostimulant reinforcement has also been observed. Also illustrated as potentially important for the implementation of the adaptive opposing responses to the behavioral e&cu of these drugs, is the locus coeruleus (LC); although its connections are not shown, the LC projects to the amygdala and to olfactory, frond, and limbic cortices. (Taken with permission from Koob and Bloom, 1988.84)

syndrome that includes the appearance of ptosis, teeth chattering, wet dog shakes, and diarrhea,37 and these symptoms can be dramatically precipitated in dependent animals by systemic injections of opiate antagonists.37 More motivational measures have included the disruption of trained operant behavior for food reward or the develop ment of place aversions following precipitated withdrawal with systemic opiate antagonist administration.38 Studies ofthe neural substrates of physical dependence on opiates have revealed multiple sites responsible for the classical opiate abstinence syndrome. Early studies implicated the periaqueductal gray39.40 and dorsal thalamus.41 More recent work using intracerebral microinjections of methylnaloxonium in rats dependent on morphine has revealed that the locus coeruleus is particularly important fbr the activating effects of opiate withdrawal.” Other physical signs appear to depend on a more widespread activation of opiate receptors. The sites in the brain responsible for the negative reinforcing propexties of opiate withdrawal appear to be more selective. The response-disruptive effects of local intracerebral administration of methylnaloxonium to rats physically dependent on morphine were explored by measuring performance on a fixed ratio-15 (FR-15) schedule during precipitated opiate withdrawal. Rats were implanted with cannulas aimed at the lateral ventricle periaqueductal gray, medial dorsal thalamus and nucleus accumbens,

KOOB: NEUROBIOLOGY OF DRUG DEPENDENCE

181

Intracerebroventrlculrr

Nucleur Accumbenr

'pT!-

500

FIGURE 9. The e&a of intracerebd methylnaloxonium p a k d with a particular environment on the amount of time spent in that environment during an injection-fire test session. Values represent the median difkrence between the postconditioning score and the preconditioning score. Dots refer to the interquartile range of this distribution. Darkened bars represent those doscs where the conditioning scorn were significantly difkrent from the preconditioning scorn using the non-parametric Wdcoxon Matched Pairs Signed-Rank test. S i cance was set at p < 0.02 to convol for multiple comparisons. (Taken with permission from Stinus ct a[., 1990.45)

I

Poriaquductrl Gray

"?-ll

500

L Ventral Togmental Area I

M d l r l Doraal Thalamw 0

U 0

~mmlloool~ Dow MN (ngh? pi 8allne)

and were trained on an FR-15schedule of reinforcement for bod, and then made dependent on morphine using subcutaneous morphine pellets. Very low doses of methylnaloxonium (4-64 rig) injected into the nucleus accumbens produced a disruption of hod motivated operant responding, whereas injections of methylnaloxonium into the periaqueductal gray or dorsal thalamus produced a dose effect function similar to that following inmcerebroventricular inje~tion.'~ These results suggested that during the development of morphine dependence, neural elements in the region of the nudeus accumbens may have become sensitized to opiate antagonists and may be responsible hr the negative stimulus eflkcts of opiate withdrawal. Confirmation of a negative reinhning effect for the withdrawal induced by intracerebral methylnaloxonium was shown using the place aversion pmcedure,M and the nucleus accumbens also proved to be the most sensitive site for inmcerebd injections of methylnaloxonium to produce place aversions in dependent rats45 (FIG. 9).

182

ANNALS NEW YORK ACADEMY OF SCIENCES

NEURAL-S

POR THE ACUTE REINFORCING EFFECTS OF ETHANOL

Several brain neurotransmitter systems such as opioid peptides, serotonin, norepinephrine, dopamine, and GAR4 have been implicated in the reinforcing properties of ethanol based on pharmacological studies using either neurotransmitter agonists or antagonists. Direct intravenous self-administrationof drugs has been an e&ctive tool for the study of the neuropharmacologyof the reinfbrcing actions of drugs such as cocaine and opiates (see above). However, intravenous self-administration of ethanol, and most sedative hypnotics, is not readily obtained in rats. The alternative model, oral ethanol self-administration in the rat, has been Fraught with problems of taste, consummatorybehavior contbunds, and lack of reliable blood alcohol determinations, but several recent studies have provided reliable procedures for the initiation and maintenance of alcohol intake using taste adulteration procedures (sucrose or saccharine substitution). Reliable, sustained operant responding fbr 10%ethanol can be obtained in free-feeding and drinking rats (nondeprived), even after complete removal of sweetener, provided that the sweetener is slowly withdrawn.The opiate antagonists naloxone and naltrexone have consistently been shown to decrease ethanol drinking,"-53 and opiate agonists will enhance ethanol intake in limited access situations.*" However, only relatively large opiate antagonist doses reduce ethanol drinking, in contrast to the small doses required fbr altering intravenous opiate self-administration.50Also, systemically administered naloxone and other opiate antagonists suppress fbod and water intake over a wide dose range,57-59and inhibitory effects of naloxone on ingestive behavior have been seen in non-deprived as well as deprived animals and extend to stimuli which are normally potent reinfbrcers (eg., sucrose and sweetened milk) .a361 Thus, there is considerable evidence to suggest that opiate antagonists inhibit consummatory behavior in general. Similar conclusions resulted from a recent study where non-motivationally constrained rats (Wistar and alcohol-preferring-P rats) were trained on a free-choice operant procedure to lever press fbr ethanol and were injected systemically with low doses of naloxone. Naloxone (0.125-0.5 mglkg) produced dosedependent reductions in responding for both ethanol and water, and consequently decreased the total amount of fluid intake.a However, ethanol preference was not altered in either strain of rats since the water-ethanol ratios remained constant across naloxone doses. The decreases in operant responding for both water and ethanol do not seem to support a selective role fbr opiate receptors in the reinfbrcing actions of ethanol, but appear more consistent with the well-documented inhibitory e&cts on consummatory behaviors of this opiate antagonist. Neuropharmacological manipulation of serotonin systems has been shown to alter ethanol consumption in numerous studies. Treatments designed to increase the synaptic availability of serotonin such as a precursor loading (5-hydroxytryptophan), administration of serotonin re-uptake blockers, or central injection of serotonin itself reduce voluntary ethanol intake.624 The results of neuropharmacologicaldecreases in serotonin fbnction on ethanol self-administration,however, are more difKcult to interpret. Serotonin synthesis inhibitors, serotonin neurotoxins, and serotonin antagonists have been shown to decrease rather than increase voluntary ethanol drinl~ing.~5"~ Several studies have suggested that brain dopamine systems may be involved in the reinfbrcing properties of ethanol. Dopamine receptor antagonists have been shown to reduce lever-pressing fbr ethanol in nondeprived ratsa.69 and also reduce home cage

KOOB: NEUROBIOLOGY OF DRUG DEPENDENCE

183

ethanol drinking.70 To test the hypothesis that dopamine receptors in the nucleus accumbens have a role in ethanol self-administration, non-motivationally constrained male Wistar rats were trained to orally self-administer ethanol in a two-lever, lirechoice self-administration task. The animals were then prepared with chronic indwelling guide cannulas aimed above the nucleus accumbens. Fluphenazine decreased ethanol self-administration at doses of 2 and 4 pg; water self-administration was unaltered at 2 pg but slightly decreased at 4 pg.71 These data, combined with observations of a release of dopamine as measured by in mW, microdialysis in the nucleus accumbens during ethanol self-administration72 (see this volume), suggest that dopamine receptors in the region of the nucleus accumbens may be involved in ethanol reinforcement in the non-dependent rat. GABA has long been hypothesized to have a role in the intoxicating effects of ethanol based on the ability of GABAergic antagonists to reverse many of the behavioral effects of ethanol, and GABAmimetic drugs can potentiate some of ethanol actions. For example, GABA antagonists decrease the ability of ethanol to produce ataxia, anesthesia and the release of punished responding (anti-conflict effect^).^^-^^ At a biochemical level, ethanol in the 10-50 mM range potentiates stimulation by GABA of C1- uptake in synaptosomes from the cerebral c0rtex7~and cerebellum.78 Further support for a role of brain GABA is the observation that the partial inverse benzodiazepine agonist, RO 154513, which has been shown to reverse some of the behavioral effects of e t h a n 0 1 , ~ ~produces ~l a dose-dependent reduction of oral ethanol (10%) selfadministration in nondeprived rats81 and in an operant lire-choice situation82 (see this volume) ( ~ L E1). The neuropharmacological data reviewed above provide some evidence for the actions of four major neurotransmitter systems in ethanol reinforcement: opioid pep-

TABLE 1.

Effects of Low Doses of Neurotransmitter Antagonists on Responding for 10%Ethanol and Water Using a Free-Choice Operant Task in Non-Deprived Rats"

Treatment

Ethanol Responding Water Responding

Systemic GABA complex-inverse agonist RO 15-4513 DA receptor antagonist Calcium diacetyl homotaurine-glutamate antagonist Isopropylbicyclophosphate-GABAl picrotoxinin antagonist Naloxone, opiate antagonist Methysergide, serotonin antagonist

decrease decrease

n o change n o change

decrease

n o change

decrease decrease no change

no change decrease no change

decrease decrease

no change

no change

no change

Nucleus accumbens Dopamine receptor antagonist NMDA antagonist-APV

no change

Responding for 0.2%saccharin and water GABA complex-inverse agonist RO 154513 a

Results supporting these conclusions can be found in Weiss ct al., 199048 and Rassnick et a[.,

199071 and 1992.82

184

ANNALS

NEW YORK ACADEMY OF SCIENCES

tides, serotonin, dopamine and GABA. Other approachesinvolving biochemical measures, lesions and genetics also provide significant evidence fbr opioid peptides, serotonin, dopamine and These neurotransmitters do brm an intimate part of the brain circuitry hypothesized to be part of the systems involved in drug reinbrcement in general (FIG.8), but the exact site or sites fbr the reinfbrcingactions ofethanol will require further investigation."

NEURAL SUaETRATEs FOR THE NEGATIVE REINFORCING PROPERTIES OF ETHANOL WITHDRAWAL Ethanol withdrawal in humans is characterized in its early stages (first 1-2 days) by anxiety, anorexia, insomnia, tremor, some mild disorientation, and possibly hallucinations. This syndrome is accompanied by a major sympathetic hyperactivity including elevated blood pressure, heart rate and body temperature. Tonic-clonic seizures similar to those of grand mal epilepsy can be observed in the first 1-2 days. In later stages of severe withdrawal, a syndrome called delirium tremens may become manifkt which is characterized by marked tremor, anxiety, insomnia, and autonomic hyperactivity. Subjects can become totally disoriented with respect to time and place, with vivid hallucinations and outbursts of irrational behavior. The pharmacological treatment for ethanol withdrawal has included barbiturates, phenothiazines, and antihistamines, but benzodiazepines are considered &r and more effective. Ethanol withdrawal in animals is characterized by central nervous system hyperexcitability that results in both physical and motivational Signs of dependence. Physical signs include tremor, lack of a venmmedial distal flexion, weight loss, and audiogenic or stress-induced seizures. More motivational measures have included disruption of operant behavior,85 increased responsiveness in acoustic startle tests,86 and increased sensitivity in the behavioral tests of anxiety such as the elevated plus maze Studies of the neurochemical bases fbr the physical signs of ethanol withdrawal have suggested a functional role fbr GABA. GABA agonists decrease the central nervous system hyperexcitability during ethanol withdrawal and ethanol withdrawal-induced convulsions.88J9 GABA antagonists exacerbate many of the symptoms of ethanol withdmwal,W and the partial inverse benzodiazepine agonist RO 15-4513 has been shown to increase the incidence of seizures during ethanol withdrawal.91 GABA has also been implicated in more motivational measures of ethanol withdrawal. Using a drug discrimination procedure, ethanol withdrawal as a stimulus produces stimulus characteristics similar to injection of pentylenetetrazol (PTZ), an anxiogenic drug.92 This Przlike interoceptive stimulus produced by ethanol withdrawal is potentiated by bicucultine and picrotoxin, suggesting that the anxiogenic-like response produced by ethanol withdrawal may be related to an ethanol-induced alteration in the function of the GABA-benzodiazepine ionophore complex.93 Another candidate possibly involved in the motivational aspects of ethanol withdrawal is the brain stress hormone, corticotropin-releasingfictor ( 0 . CRF is a neuropeptide widely distributed in the central nervous system and has been hypothesized to have a functional role in behavioral responses to stress. CRF itself has anxiogenic actions and a CRF antagonist, a-helical CRF can reverne some behavioral responses to stress.% Rats made dependent on ethanol and then subjected to abrupt withdrawal from chronic ethanol show an "anxiogenic-like" response in several "anxiety" tests, such as the elevated plus m a ~ . To ~ 5explore the role of endogenous brain CRF sys-

KOOB: NEUROBIOLOGY OF DRUG DEPENDENCE

a

185

f

t

I

1

-r

5 FIGURE 10. Effect of ICV administration of a-helical CRF to mts tested in the elevated plus-maze after ethanol withdrawal. The y)hand bar in each panel contains data from rats tested 8 h after withdrawal from control diet and 30 min after ICV vehicle administration (“controls”). The three @t-hand bars in each panel contain data from rats tested 8 h after withdrawal from ethanol diet and 30 min after ICV a-helical CRF administration (0,5and 25 mg). The toppancl shows mean E ti (* SEM) %of time spent on the open arms. The mu#& pnel shows the mean (k SEM) % number of entries onto the open arms. 0 The hum p c l shows the mean (* SEM) total number of arm entries. Difference from “controls”: p < 0.05 (ANOVA). Significantly diarent from p u p receiving ICV vehicle after ethanol withdrawal: t p < 0.05 (Newman-Keul’s tests after ANOVA). (Taken with permission from Baldwin ct al.,

i

0

25

t

I

60

5

0

T

40

20

a 0

0

0

0

5

25

191.87)

5

25

dose of alpha-helical CRF (pg)

terns in ethanol withdrawal, the effect of a CRF antagonist on the behavioral response of rats during withdrawal was examined. Rats chronically maintained (2-3 weeks) on a liquid diet showed a significant anxiogenic-like response in an elevated plus maze 8 hours into withdrawal. This anxiogenic-like response was reversed with intracerebroventricular administration of the CRF antagonist, a-helical CRF (FIG. lO).S7 These results suggest that CRF in the central nervous system may also be involved in some of the more motivational aspects of ethanol withdrawal.

SUMMARY AND CONCLUSIONS The brain substrates involved in the effect of cocaine on brain stimulation reward, in the psychomotor activation associated with cocaine, and in cocaine selfadministration appear to be fbcused on the medial forebrain bundle and its connec-

186

ANNALS

NEW YORK ACADEMY OF SCIENCES

tions with the basal fbrebrain, notably the nucleus accumbens. Chronic access to cocaine produces a withdrawal state as reflected in increases in brain stimulation w a r d thresholds, and this change in reward threshold appears to be opposite to the actions of the drug administered acutely. These effects are thought to reflect a change in the activity of reward elements in the medial fbrebrain bundle and may be responsible for the negative reinfbrcing state associated with the anhedonia of cocaine withdrawal. Opiate receptors particularly sensitive to the reinforcing eficts of heroin also appear to be located in the region of the nucleus accumbens and the ventral tegmental area. There is good evidence fbr both dopamine-dependent and dopamine-independent opioid interactions in the ventral tegmental-nucleus accumbens connection. In addition, the opiate receptors in the region of the nucleus accumbens may become sensitized during the course of opiate withdrawal and thus become responsible for the aversive stimulus effects of opiate dependence. Reliable measures of the acute reinfbrcing effects of ethanol have been established in rat models, and substantial evidence exists to show that non-deprived rats will orally self-administer pharmacologically relevant amounts of ethanol in lever-press choice situations. Neuropharmacological studies of ethanol reinhrcement in non-dependent rats suggest important roles for serotonin, GABA and dopamine. A role for opioid p e p tides in ethanol reinforcement may reflect more general actions of opioid peptides in consummatory behavior. Studies of ethanol dependence have implicated brain GABAergic and CRF systems in the more motivational aspects of withdrawal. Future studies will need to fbcus on the common neurobiologic changes associated with all these drugs, particularly re@ng their hedonic and motivational properties. ACKNOWLEDGMENTS

Results discussed in this chapter are derived from studies with the following individuals: Stephen Negus, Patricia Robledo, Rahel Maldonado, Steve Heinrichs, Athina Markou, Stephanie Rassnick, B a d Caine, Helen Baldwin, Luigi Pulvirenti, Karen Britton, Friedbert Weiss, Luis Stinus and Floyd E. Bloom.I thank them fbr their help and collaboration. I thank Molecular and Experimental Medicine's Word Processing Center for their help in manuscript preparation. REFERENCES 1. DAVIS, W. M. & S. G. SMITH.1975. E&ct of haloperidol on (+)-amphetamine selfadministration.J. Pharm. Pharmacol. 27: 540-542. 2. YOKEL, R A. & R A. WISE.1975. Increased lever prrssing for amphetamine after pimozide in rats: Implications for a dopamine theory of nward. Science 187: 547-549. 3. YOKEL,R A. & R A. WISE.1976. Attenuation of intravenous amphetamine reinforcement by c e n d dopamine blockade in rats. Psychopharmacology 48: 311-318. 4. EITBNBERG, A., H. 0. PBITIT, F. E. BLOOM & G. F. KOOB.1982. Heroin and cocaine intravenousself-administration in rats: Mediation by separate neural systems. Psychopharmacology 78: 204-209. 5. ROBERTS, D. C. S., M. E. CORCORAN& H . C. FIBIGER. 1977. On the role of ascending catecholaminecgicsystems in intravenous self-administration of cocaine. Pharmacol. Biochem. Behav. 6 615-620. 6. ROBERTS, D. C. S.,G . F. KOOB,P. KLONOPP& H. C. FIBIGER. 1980. Extinction and re-

KOOB: NEUROBIOLQGY OF DRUG DEPENDENCE

7. 8.

9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27.

187

covery of cocaine self-administration following 6-hydmxydopamine lesions of the nucleus accumbens. Pharmacol. Biochem. Behav. 12: 781-787. LYNESS,W. H., N. M. FRIEDLE & K. E. MOORE.1979. Destruction of dopaminergic nerve terminals in nucleus accumbens: S c t of d-amphetamine self-administration. Pharmacol. Biochem. Behav. lk 663-666. KOOB,G. F., F. J. VACCARINO, M. AMURIC& F. E. BLOOM.1987. Positive reinforcement properties of drugs: Search for neural substrates. In Brain Rnvard Systems and Abuse. J. Engel & L. Oreland, Eds.: 35-50. Raven Prcss. New York. MARTIN-~VERSON, M. T., C. SZOSTAK& H. C. FIBIGER.1986. 6-Hydroxydopamine lesions of the medial prefrontal cortex Fail to influence intravenous self-administration of cocaine. Psychopharmacology 88: 310-314. KOOB, G. F., H. T. LE & J. CRBESE.1987. D-1 and D-2 dopamine antagonists increase cocaine self-administration in the rat. Ncurosci. Lett. 79: 315-320. CAINE, S. B., M. BBRHOW,M. AMALRIC & G. F. KWB. 1990. The D1 antagonist SCH 23390 and the D2 a n m n i s t raclooride selectivelv decrease behavior maintained bv cocaine or food in the r a t Soc. Neubsci. Abstr. 16: 25a. RDBLEDO,P., R MALDONALW-LOPEZ & G. F. KOOB.1992. Role of dopamine receptors in the nucleus accumbens in the r e w d n g properties of cocaine. Ann. NY Acad. Sci. 654: 509-512. This volume. SWERDLOW, N. R, L. W.SWANSON& G. F. KOOB.1984. Substantia innominata: Critical link in the behavioral expression of mesolimbic dopamine stimulation in the rat. Neurosci. Lett. 50: 19-24. HUBNBR, C. B. & G. F. KOOB.1990. The ventral pallidurn plays a role in mediating cocaine and heroin self-administration in the rat. Brain Rrs. 508: 20-29. GAWIN,F. H. & H. D. KLEBER.1986. Abstinence symptomatology and psychiatric dragnasis in cocaine abusers. Arch. Gen. Psychiatry 43: 107-113. ELLINWOOD, E. H. & W. M. PETRIE. 1977. Dependence on amphetamine, cocaine and other stimulants. In Drug Abuse: Clinical and Basic Aspects. S. N. Pradhan, Ed.: 248-262. Maby. St. Louis. STELLAR, J. R & E. STELLAR.1985. The Neurobiology of Reward and Motivation. SpringerVerlag. New York. Esposm, R U., A. H. D. MOTOLA& C. KORNRSICY. 1978. Cocaine: Acute e&cts on reinforcement thresholds for self-stimulation behavior to the medial forebrain bundle. Pharmacol. Biochcm. Behav. 8: 437439. MARKOIJ,M. A. & G. F. KOOB.1991. Pastcocaine anhedonia: An animal model of cocaine withdnwal. Neuropsychopharmacology 4 17-26. C. A. & M. S. GOLD.1985. New concepts in cocaine addiction: The dopamine DACKIS, depletion hypothesis. N e u m i . Biobehav. Rcv. 9 469477. KOOB,G. F., H. 0.PElm,A. EITJMBERG & F. E. BLOOM.1984. Efkcts of opiate antagonists and their quaternary derivativeson heroin self-administrationin the rat. J. Pharmacol. Exp. Ther. 229: 481-486. ZINBERG,N. E. & R C. JACOBSON. 1976. The natural history of "chipping." Am. J. Psychiatry 133: 37-40. GOLDBERG, S. R , J. H. WOODS & C. R S C H U ~ T E1971. ~ Nalorphine-induced c h a n p in morphine self-administration rhesus monkeys.J. Pharmacol. Exp. Ther. 176:464-471. WEEKS,J. R & R J. COLLINS.1976. Changes in morphine self-administration in rats induced by prostaglandin E and naloxone. Prostaglandins 12: 11-19. VACCARINO, F. J., H. 0. Parrrr, F. E. BLOOM& G. F. KOOB. 1985. El%cts of inuacerebrovenmcular administration of methylnaloxonium chloride on heroin self-administration in the rat. Pharmacol. Biochem. Behav. 23: 495-498. NEGIJS, S.S., N. B. WEINGEB, S.J. HENRICKSON & G,F. KOOB.1991. El%ct of mu,delta, and kappa opioid agonists on heroin maintained mponding in the rat. Abstract Committee on Problems of Drug Dependence Meeting, SCHROEDER, R L., M. B. WEINGEB,L. VAHASSIAN & G. F. KOOB. 1991. Methylnaloxonium diffuses out of the brain more slowly than naloxone after d k c t intracerebral injection. N e u m i . Lett. l2: 173-177.

188

ANNALS NEW YORK ACADEMY OF SCIENCES

28. VACCARINO,F. J., F. E. BLOOM& G. F. KOOB. 1985. Blockade of nucleus accumbens opiate receptors attenuates intravenous heroin reward in the rat. Psychopharmacology (Berlin) 85: 37-42. 29. GOEDERS, N. E., J. D. LANE& J. E. SMITH.1984. Self-administration of methionine enkephalin into the nucleus accumbens. Pharmacol. Biochem. Behav. 2 0 451-455. 30. P m , H . O., A. EITBNBERG,F. E. BLOOM& G. F. KOOB.1984. Destruction of dopa-

mine in the nudeus accumbens selectively attenuates cocaine but not heroin selfadministration in rats. Psychopharmacology 84: 167-173. 31. STINUS, L., M. CADOR& M. LE MOAL. 1992. Interaction between endogenous opioids and dopamine within the nucleus accumbens. Ann. N.Y. Acad. Sci. 654:254-273. This volume. T. S., R SPANAGEL & A. HEW. 1992. Conditioning of opioid reinforce32. SHIPPENBERG, ment: Neurochemical and neuroanatomical substrates. Ann. N.Y. Acad. Sci. 654:347356. This volume. 33. WISE,R A. 1992. Self-stimulation and drug reward mechanisms. Ann. N.Y. Acad. Sci. 654: 192-198. This volume. 34. DE WIED, D. & B. Bows. 1979. Modulation of memory processes by neuropeptides of hypothalamic-neurohyphyseal origin. In Brain Mechanisms in Memory and Learning: From the Single Neuron to Man. M. A. B. Brazier, Ed.: 139-149. Raven Press.New Yo&. 35. KELLEY,A. E. & L. S m s . 1984. The distribution of the projection from the parataenial nucleus of the thalamus to the nucleus accumbens in the rat: An autoradiographic study. Exp. Brain Res. 54: 499-512. 36. JAFFE,J. H. 1987. Drug addiction and drug abuse. In Goodman and Gilman’s The Pharmacological Basis of Therapeutics. A. G. Gilman, L. S. Goodman & T. W. Rall, Eds. 7th Edit.: 532-580. MacMillan Publishing Co.New York. 37. WAY, E. L., H. H. LOH & F. H. SHEN. 1969. Simultaneous quantitative assessment of morphine tolerance and physical dependence. J, Pharmacol. Exp. Ther. 167: 1-8. V. F. & S. B. SPABBER.1977. A comparison of the effects of naloxone upon body 38. GELLERT, weight loss and suppression of fixed-ratio operant behavior in morphinedependent rats. J. Pharmacol. Exp. Ther. 201: 44-54. 39. WEI, E., H. H . LOH & E. L. WAY. 1972. Neuroanatomical correlates of morphine dependence. Science 177: 616617. 40. WEI, E., H . H. LOH & E. L. WAY.1973. Brain sites ofprccipitated abstinence in morphinedependent rats. J. Pharmacol. Exp. Ther. 185: 108-115. M. A. & R A. WISE.1984. Anatomically distinct opiate receptor fields mediate 41. B~ZARTH, reward and physical dependence. Science 22.4: 516-517. R, L. GOLD, L. STINUS& G. F. KOOB. 1992. Morphine withdrawal syn42. MALDONALDO, drome after local administration of methylnaloxonium in several brain structure. J. Pharmacol. Exp. Ther. In press. 43. KOOB, G. F., T. L. WALL& F. E. BLOOM.1989. Nucleus accumbens as a substrate for the aversive stimulus effects of opiate withdrawal. Psychopharmacology (Berlin) 98: 530-534.

44. HAND,T. H . , G. F. KOOB, L. STINUS& M. LE MOAL.1988. Aversive properties of opiate receptor blockade are centrally mediated and are potentiated by ptwious exposure to opiates. Brain Res. 4 7 4 364-368. 45. STINUS,L., M. LE MOAL& G. F. KOOB.1990. The nudeus accumbens and amygdala as possible substrates for the aversive stimulus e&cts of opiate withdrawal. Neuroscience 37: 767-773.

46. SAMSON,H. S. 1986. Initiation of ethanol reinforcement using a sucrose-substitution procedure in food-and water-sated rats. Alcoholism: Clinical and Experimental Research 10: 436442. 47. GRANT,K. A. & H. H. SAMSON. 1985. Induction and maintenance of ethanol selfadministration without food deprivation in the rat. Psychopharmacology 86: 475479. 48. WEISS,F., M. MIKHINLUL,F. E. BLOOM & G.F. Koon. 1990. Free-choice mponding for ethanol versus water in Alcohol-Prckming (P) and unselected Wistar rats is difkrentially

KOOB: NEUROBIOLOGY OF DRUG DEPENDENCE

49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59.

60. 61. 62. 63. 64.

65.

66. 67. 68. 69. 70. 71. 72.

189

altered by naloxone, bromocriptine and methysergide. Psychopharmacology 101: 178-186. P~LVIRENTI, L. & A. J. KASTIN. 1988. Naloxone, but not Tyr-MIF-1, reduces volitional ethanol drinking in rats: Correlation with degree ofspontaneous p&rences. Pharmacol. Biochem. Behav. 31: 129-134. REID, L. D. & G . A. HUNTER. 1984. Morphine and naloxone modulate intake of ethanol. Alcohol 1: 33-37. SANDI, C., J. BOXELL& C. GWZAZ. 1988. Naloxone decreases ethanol consumption within a free choice pm+ in rats. Pharmacol. Biochem. Behav. 29: 39-43. R, M. A. DAVIS& J. E. OLGIN.1986. Nalmone blocks the post-shock inVOLPICELLI, crease of ethanol consumption. L& Sci. 38: 841-847. MARPAINGJALLAT, P., D. MICBLI & J. L B ~ G N E 1983. N . Decin ethanol consumption by naloxone in naive and dependent rats. Pharmacol. Biochem. Behav. 18: 5355-5395. HUNTER, G. A., C. M. BEAMAN,L. L. Du” & L. D. RED. 1984. Selected opioids, ethanol and intake of alcohol. Alcohol 1: 43-46. C. L., S.A. CZIRR& L. D. REID. 1987. Persistence and spedcity of small doses HUBBELL, of morphine on intake of alcoholic beverages. Alcohol 4 149-156. REID, L. D., S. A. CZIRR,C. C. BENSINGER, C. L. HUBBELL & A. J. VOLANTH. 1987. Morphine and diprenorphine together potentiate intake of alcoholic beverages. Alcohol 4 161-168. BROWN,D. R & S. G. HOLTZMAN. 1979. Suppnssion of deprivation-induced food and water intake in rats and mice by naloxone. Pharmacol. Biochem. Behav. 1l: 567-573. S. G. 1979. Suppnssion of appetitive behavior in the rat by naloxone: Lack HOLTZMAN, of e&ct of prior morphine dependence. L& Sci. 24: 219-226. HYNBS,M. A., M. GALLAGHER & K. V. Y A ~ 1981. . Systemic and intraventricular naloxone administration:F,fikcts on food and water intake. Behav. N e d Biol. 32: 334-342. COOPER,S. J. 1980. Naloxone: E&cts on food and water consumption in the nondeprivcd and deprived rat. Psychopharmacology 7l(l):1-15. J. M., N. L. osraOws~~, V. J. M~FUUMAN, M. D. L m & L . D. RED. 1979. STAPLETON, Naloxone Educes fluid consumption in deprived and nondeprived rats. Bull. Psychon. Soc. 13: 237-239. AMIT, S.,E. A. SUTHBBLAND, K. BILL& S. 0. OGRBN. 1984. Zimelidine: A roliew of its etfects on ethanol consumption. Neurosci. Biobehav. RN. 8: 35-54. I. 1973. E&cts of pata-chlorophenylalanineand 5-hydroxyayptophane on alcohol GELLER, intake in rats. Pharmacol. Biochem. Behav. 1: 361-365. LA^, M. O . , C. A. NNUNJO & E. M. S~uaas.1986. Identi6cation of new drugs for modulating alcohol consumption. Psychopharmacol. Bull. 22: 1020-1025. FEY, H-H., M. P. MAGNUSSEN& C. KAERGMRDNIBLSBN.1970. The etfect of p-chloroamphetamineon the consumption of ethanol by rats. Arch. Int. Pharmacodyn. Ther. 183: 165-172. MYERS, R D. & W. L. VBALB. 1968. Alcohol prrkrence in the rats: Rcduction following depletion of brain serotonin. Science 160: 1469-1471. VBALB, W. L. & R D. MYERS.1970. Decrrasc in ethanol incake in rats following administration of p-chlorophenylalanine.Neuropharmacology9: 317-326. A. 0.& H. H. SAMSON. 1988. Haloperidol and apomorphine e&as on ethanol PPEPPER, reinforcement in free-keding rats. Pharmacol. Biochem. Behav. 2 9 343-350. ~ E P P B R ,A. 0. & H. H. SAMSON. 1985. ethanol reinfoirements: Interactive etkcts of amphetamine, pimozide and food restriction. Alcohol Drug Rrs. 6: 37-48. PPEPPER,A. 0.& H. H. SAMSON. 1986. Eflkct ofpimozide on home cagc ethanol dmkmg in the rat: Dependence on dmkmg session length. D q Alcohol Depend. 17: 47-55. & G. F. WB.1990. Modulation of ethanol selfRASSNICK,S., L. PIJLVXUBNTI administration by dopamine and GABA mechanisms within the nudeus accumbens. Abstract Rrscarch Society on Alcoholism, Toronto. WEISS,F., Y. L. HURD,U. UNGERSTEDT, A. MARKOW,P. M. PLOTSKY& G. F. KOOB.

190

ANNALS NEW YORK ACADEMY OF SCIENCES

1992. Neurochemical correlates of cocaine and ethanol self-administration. Ann. N.Y. Acad. Sci. 654: 220-241. This volume. 73. LILJEQUIST, S. & J. A. ENGEL.1984. The efects of GABA and benzodiazepine receptor antagonists on the anti-conflict actions of diazepam or ethanol. Pharmacol. Biochem. Behav. 21: 521-525. 74. KOOB,G. F., W. B. MENDELSON, J. SCHAFER,T. L. WALL,K. THATCHER-BWITON& F. E. BLOOM. 1988. Picrotoxin receptor ligand blocks anti-punishment effects of alcohol. Alcohol 5: 437-443. 75. FRYE,G. D. & G. R BRESE. 1982. GABAergic modulation of ethanol-induced motor impairment. J. Pharmacol. Exp. Ther. 223: 750-756. S. & J. ENGEL.1982. Etfects of GABAergic agonists and mtagonists on various 76. LILJEQUIST, ethanol-induced behavioral changes. Psychopharmacology 78: 7’ .75. & S . M. PAUL.1906. Ethanol stimulates 77. SUZDAK,P. D., R D. SCHWARTZ,P. SKOLNICK gamma-aminobutyric acid receptor-mediated chloride transport in rat brain synaptoneurosome. Proc. Natl. Acad. Sci. USA 83: 4 0 7 1 4 7 5 . 78. ALLAN,A. M. & R A. HARRIS. 1987. Acute and chronic ethanol treatments alter GABA receptor-operated chloride channels. Pharmacol. Biochem. Behav. 27: 665-670. 79. SUZDAK,P. D., J. R GLOWA,J. N. CRAWLEY, R D. SCHWARTZ,P. SKOLNICK & S. M. PAUL.1986. A selective imidambenzodiazepine antagonist of ethanol in the rat. Science 236: 1243-1247. 80. BRITON, K. T., C. L. EHLERS& G. F. KOOB. 1988. Is ethanol antagonist RO 15-4513 selective for ethanol. Science 239: 648-650. A. 0. PFEPFER,K. G. SADKHI& F. G. MILLS. 1987. 81. SAMSON,H. H., G. A. TOLLIVER, Oral ethanol reinforcement in the rat: Effect ofthe partial inverse benzodiazepine agonist R015-4513. Pharmacol. Biochem. Behav. 27: 517-519. W. ZIEGLG~SBERGER& G. F. KOOB. 82. RASSNICK,S., L. DAMICO,E. RILEY, L. PULVIRENTI, 1992.GABA and nucleus accumbens glutamate neurotransmission modulate ethanol selfadministration in rats. Ann. N.Y. Acad. Sci. 654: 502-505. This volume. 83. DEITRICH,R A., T. V. DUNWIDDIE, R A. HARRIS& V. G. ERWIN.1989. Mechanism ofaction ofethanol: Initial central nervous system actions. Pharmacol. RN.41: 489-537. 84. KOOB,G. F. & F. E. BLOOM.1988. Cellular and molecular mechanisms of drug dependence. Science 242: 715-723. U. & H. BEGLEITER. 1976. Response suppression on a mixed schedule of rein85. DENOBLE, forcement during alcohol withdrawal. Pharmacol. Biochem. Behav. 5: 227-229. S . , G. F. KOOB& M. A. GEYER.1992. Responding to acoustic startle during 86. USSNICK, chronic ethanol intoxication and withdrawal. Psychopharmacology 106: 351-358. 87. BALDWIN,H. A., S. RASSNICK,J. ~ E R G., F. KOOB& K. T. BRI?TON.1991. CRF antagonist revelses the “anxiogenic” response to ethanol withdrawal in the rat. Psychopharmacology 103: 227-232. 88. COOPER,B. R, K. VIIK,R M. FERNS& H. L. WHITE.1979. Antagonism ofthe enhanced susceptibility to audiogenic seizures during alcohol withdrawal in the rat by gammaaminobutyric acid (GABA) and GABA-mimetics. J. Pharmacol. Exp. Ther. 209: 396408. 89. FRYE,G. D., T. J. M c C o w &G. R BREESE. 1983. D~rcntialsensitivityofethanolwithdrawal signs in the rat to gamma-aminobutyric acid (GABA) mimetics: Blockade of audiop i c seizures but not forelimb trcmors. JPET 2 2 6 720-723. 90. GOLDSTEIN, D. B. 1973. Alcohol withdrawal reaction in mice: Effects of drugs that modify neurotransmission. J. Pharmacol. Exp. Ther. 186. 1-9. 91. LISTER,R G. & J . W. -IAN. 1987. RO 154513 induces seizures in DBAl2 mice underp i n g alcohol withdrawal. Alcohol 4(5): 409411. 92. LAL, H., C. M. HARRIS, D. BENJAMIN, A. C. SPRINGFIELD, S . BLADRA& M. EMMEITOGLFSBY.1988. Characterization of a pentylcnetetrazol-like interoceptive stimulus produced by ethanol withdrawal. J. Pharmacol. Exp. Ther. 247: 508-518. 93. INDEMUDIA, S. O . , S . BLADRA & H. LAL.1989. The pentylenetetraml-like interoceptive stimulus produced by ethanol withdrawal is potentiated by bicuculline and picrotoxin. Neuropsychopharmacology 2: 115-122.

KOOB: NEUROBIOLOGY OF DRUG DEPENDENCE

191

94. BALDWIN,H.A., K. T.BNITON & G. F. KCIOB.1990. Behavioral eEcts of CRF.In Behavioral Aspects of Neuroendocrinology. D. W.P M & D. Ganten, Eds.: 1-14. Springer Vcrlag. 95. FILE, S. E.,H. A. BALDWIN& P. K. HITCHWIT.1989. Flumazenil but not nincndipine revexs the increased anxiety during ethanol withdrawal in the rat. Psychopharmacology 98: 252-264. 96. KOOB, G. F. 1991. Neurobiological mechanisms in cocaine and opiate dependence. In Advances in Understanding the Addictive State. Raven Press. New York. 97. WB, G. F. & N. R S w ~ a ~ m w 1988. . Functional output of the mesolimbic dopamine system. Ann. N.Y. Acad. Sci. 537: 216227.