INVESTIGATIONS OF THE NEUROBIOLOGICAL AND BEHAVIOURAL ACTIONSOF COCAINEByERIN EARL BROWNB.Sc. (Honours), University of Alberta, 1988A THESIS SUBMITTED IN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OFDOCTOR OF PHILOSOPHYinTHE FACULTY OF GRADUATE STUDIESDivision of Neurological SciencesWe accept this thesis as conformingto the required standardTHE UNIVERSITY OF BRITISH COLUMBIAAugust 1993© Erin Earl BrownIn presenting this thesis in partial fulfilment of the requirements for an advanceddegree at the University of British Columbia, I agree that the Library shall make itfreely available for reference and study. I further agree that permission for extensivecopying of this thesis for scholarly purposes may be granted by the head of mydepartment or by his or her representatives. It is understood that copying orpublication of this thesis for financial gain shall not be allowed without my writtenpermission.(Signature)Department of ^NJ E `-'424::, L—Orc I C AL SC( €-4 ce- sThe University of British ColumbiaVancouver, CanadaDate 5.---er- 2_4 ) igq DE-6 (2/88)iiAbstractAlthough once considered a benign recreational stimulant, cocaine is now recognizedto possess substantial abuse potential with considerable medical and social consequences.Accordingly, these experiments examined the behavioural and neurobiological effects ofcocaine in the rat.The behavioural and neurochemical interactions between cocaine and buprenorphinewere examined using a conditioned place preference (CPP) procedure and in vivomicrodialysis. Cocaine and buprenorphine both elicited CPP; moreover, these drugsinteracted to produce significantly larger CPPs when given in combination. Both cocaineand buprenorphine increased interstitial concentrations of dopamine in the nucleusaccumbens; the effect of cocaine was potentiated by the coadministration of buprenorphine.Taken as a whole, these results indicate that buprenorphine can interact with cocaine in asynergistic manner.The ability of stimuli previously paired with cocaine to elicit similar neurochemicalchanges as cocaine was assessed by in vivo microdialysis. Although acutely administeredcocaine produced a significant increase in interstitial dopamine concentrations in the nucleusaccumbens, the presentation of a cocaine-paired environment did not. Despite the absenceof a conditional neurochemical effect, significant conditioned locomotion was observed.These data do not support the hypothesis that stimuli paired with cocaine produce theirbehavioural effects by eliciting similar neurochemical effects as cocaine.To understand better the neurobiology of cocaine-induced environment-specificconditioning, expression of c-Fos, a putative marker of neuronal activity, was examined inthe forebrain of rats exposed to an environment in which they had previously receivedcocaine. Compared to saline-treated controls, cocaine produced an increase in locomotorbehaviour that was accompanied by an increase in c-Fos expression within specific limbicregions, as well as the basal ganglia. Exposure of rats to the cocaine -paired environmentiiialso produced an increase in locomotion that was associated with an increase in c-Fosexpression within specific limbic regions, but not within the basal ganglia. These findingssuggest that specific limbic regions exhibit increased neuronal activation during thepresentation of cocaine-paired cues and may be involved in the formation of associationsbetween cocaine's stimulant actions and the environment in which the drug administrationoccurred.Given the large body of evidence implicating the amygdaloid complex in the learningof stimulus-reward associations, the effects of quinolinic acid lesions of the amygdala oncocaine-induced conditional locomotion and CPP were examined. Although destruction ofthe amygdala did not affect basal locomotion, cocaine-induced locomotion or cocaine-induced conditional locomotion, cocaine-induced CPP was completely blocked by theamygdaloid lesions. These data demonstrate that cocaine-induced stimulus-rewardconditioning can be differentially affected by lesions of the amygdala.These studies provide a further understanding of the neurobiology of cocaine'sbehavioural actions. The implications for the treatment of cocaine abuse are discussed.ivTable of ContentsPageAbstract ^ iiTable of Contents ^ ivList of Tables viList of Figures ^ viiAcknowledgements xI. Introduction ^ 1II. Behavioural and Neurochemical Interactionsbetween Cocaine and Buprenorphine ^ 12(A) Introduction ^ 12(B) Materials and Methods ^ 14(C) Results ^ 19(D) Discussion 37(E)^Notes ^ 45III.^Cocaine-Induced Conditioned Locomotion:Absence of Associated Increases in Dopamine Release ^ 46(A) Introduction ^ 46(B) Materials and Methods ^ 47(C) Results ^ 52(D) Discussion 66VIV.^Evidence for Conditional Neuronal Activation Following Exposure to aCocaine-Paired Environment: Role of Forebrain Limbic Structures ^ 72(A) Introduction ^ 72(B) Materials and Methods ^ 73(C) Results ^ 77(D) Discussion 92V.^Differential Effects of Excitotoxic Lesions of the Amygdala on Cocaine-InducedConditioned Locomotion and Conditioned Place Preference ^ 99(A) Introduction ^ 99(B) Materials and Methods ^ 100(C) Results ^ 104(D) Discussion 117VI. General Discussion ^ 121VII. References ^ 132viList of TablesPageTable 1^Diagnostic criteria for psychoactive substance dependence ^ 3Table 2^Diagnostic criteria for psychoactive substance abuse 4vi iList of FiguresPageFigure 1^Schematic representation of the phases of thecocaine abstinence syndrome ^ 7Figure 2^The effects of cocaine on the time spent in the drug-pairedcompartment before and after conditioning ^ 21Figure 3^The effects of buprenorphine on the time spent in the drug-pairedcompartment before and after conditioning ^ 23Figure 4^^The effects of low doses of cocaine and buprenorphine aloneand in combination on the time spent in the drug-pairedcompartment before and after conditioning ^ 25Figure 5^The effects of moderate doses of cocaine and buprenorphine aloneand in combination on the time spent in the drug-pairedcompartment before and after conditioning ^ 27Figure 6^The effects of cocaine on dialysate concentrations ofdopamine and metabolites from the nucleus accumbens ^ 30Figure 7^The effects of buprenorphine on dialysate concentrations ofdopamine and metabolites from the nucleus accumbens ^ 32Figure 8^The effects of cocaine + buprenorphine on dialysate concentrationsof dopamine and metabolites from the nucleus accumbens ^ 34Figure 9^Summary of the effects of saline, cocaine, buprenorphine andcocaine + buprenorphine on dialysate concentrationsof dopamine from the nucleus accumbens ^ 36Figure 10^Photomicrograph of a coronal section through the nucleus accumbensof a rat implanted with a dialysis probe ^ 54viiiFigure 11^The effects of saline and cocaine on the dialysate concentrationsof dopamine and metabolites from the nucleus accumbens ^ 56Figure 12^Locomotor counts of control, pseudoconditioned and conditionedsubjects when exposed to the conditioned environment ^ 58Figure 13^Dialysate concentrations of dopamine and its metabolitesfrom the nucleus accumbens prior to and during exposureto an environment paired with cocaine or saline ^ 61Figure 14^Dialysate concentrations of dopamine and its metabolitesfrom the nucleus accumbens prior to and during exposureto an environment paired with cocaine or saline ^ 63Figure 15^Locomotor activity of conditioned and pseudoconditionedsubjects when exposed to the conditioned environment ^ 65Figure 16^Camera lucida drawings of representative sections used forthe counting of Fos-positive nuclei ^ 79Figure 17 Locomotor counts for control, pseudoconditioned and conditionedsubjects when exposed to the conditioned environment, as well asthe locomotor counts of saline- and cocaine-treated subjects 81Figure 18^Number of Fos-positive nuclei within the cingulate cortex, claustrum,piriform cortex and nucleus accumbens ^ 83Figure 19^Number of Fos-positive nuclei within the striatum, lateral septum,paraventricular nucleus of the thalamus and amygdala ^ 85Figure 20^Photomicrographs of Fos immunoreactivity in the cingulate cortex ^ 88Figure 21^Photomicrographs of Fos immunoreactivity in the lateral habenula ^ 90Figure 22^Photomicrograph of the amygdaloid region following infusionsof quinolinic acid ^ 106Figure 23^Schematic of a representative bilateral lesion of the amygdalafollowing infusions of quinolinic acid ^ 108ixFigure 24^Locomotor counts for control, pseudoconditioned and conditionedsubjects during 10 days of conditioning with cocaine ^ 112Figure 25^Locomotor counts for control, pseudoconditioned and conditionedsubjects when exposed to the conditioned environment ^ 114Figure 26^^The effects of cocaine on the time spent in the drug-pairedcompartment before and after conditioning for non-lesionedand lesioned subjects ^ 116xAcknowledgmentsI would like to acknowledge the invaluable assistance of Campbell Clark, LilliCollu, Geert Damsma, Jamie Day, Janet Finlay, Mark Kimmins, George Nomikos, GeorgeRobertson, Sandra Sturgeon, Chui-Si Tham, Danielle Wenkstern and Catriona Wilson. Theguidance, encouragement and friendship of the aforementioned people made this anenjoyable endeavor. The supervision and guidance of Chris Fibiger, The Big Guy", notonly aided in the completion of this thesis, but also helped mature and refine myunderstanding and appreciation of science. Finally, I would like to acknowledge the personwho saw me through both the good times and bad, my wife, Kim Capri. She tolerated yearsof being a "science widow", yet continued to provide me with encouragement and support.1I. INTRODUCTIONShortly after cocaine became commercially available in Europe and the United Statesin the 1880's it began to receive endorsements and praise for its action as a tonic for thebody and mind (Musto, 1992). At this time the stimulant properties of cocaine wereemphasized, while its addictive potential was generally dismissed. Within thirty years,however, the adverse effects of cocaine had become apparent and severe restrictions on itsuse were enacted. In the United States the recognition of the abuse potential of cocaineculminated in the passage of the Harrison Act of 1914. Cocaine use diminished during the1920's as the adverse consequences associated with its use became more widelyacknowledged (Musto, 1992). As cocaine use began to increase rapidly in the 1970's andearly 1980's (Anthony, 1992), it was again suggested by medical professionals that cocainewas a safe, nonaddicting stimulant (Grinspoon and Bakalar, 1980; Van Dyke and Byck,1982). These assertions were partly the result of an absence of strong empirical evidenceregarding the actions of cocaine. At present, clinical and epidemiological research hasclearly documented numerous severe consequences of cocaine use (Anthony, 1992; Benowitz,1992) and it is recognized that cocaine possesses substantial abuse potential. Given thatcocaine abuse remains a serious medical and social issue, considerable interest remains inelucidating the actions and effects of this powerful stimulant.Clinical Characteristics of Cocaine AddictionAcute administration of cocaine produces a sense of alertness, euphoria and well-being (Jaffe, 1989; Johanson and Fischman, 1989). There are also decreases in hunger andneed for sleep. In addition to these behavioural changes, a number of physiological signsoccur following cocaine, such as tachycardia, pupillary dilation and elevated blood pressure.Cocaine can also induce paranoia, suspiciousness and overt psychosis, especially following aprolonged "binge" (Gawin and Ellinnwood, 1988; Jaffe, 1989; Satel et al., 1991).2Of those individuals who try intranasal cocaine ("snorting") the National Institute ofDrug Abuse estimates that 10 to 15% become abusers (Gawin, 1991). Most cocaine usersseeking treatment report that their initial use was intermittent. Eventually, however,episodes of high-dosage use become more frequent until "runs" or binges occur (DSM-III-R;Gawin and Ellinwood, 1988; Jaffe, 1989). Binges, which can last several days, arecharacterized by the user compulsively readministering cocaine every 10 to 30 minutes.During the binge, the user experiences periods of extreme euphoria, the memories of whichwill later be contrasted with the mood of the undrugged state and can produce cravings forthe drug (Gawin and Ellinwood, 1988; Gawin and Kleber, 1986). Binges tend to beinterrupted only when drug toxicity occurs, the user collapses from physical exhaustion orthe cocaine supplies are depleted. The termination of the binge is generally followed by anintense and unpleasant "crash".Clinical findings appear to suggest that cocaine smoking or intravenousadministration tend to produce a more rapid progression from infrequent use to cocaineabuse or dependence than intranasal use of cocaine (DSM-III-R; Jaffe, 1989; Gawin andEllinwood, 1988). This result may reflect the fact that these routes of administrationproduce a more rapid rise in blood and brain concentrations of cocaine and an intense"rush", followed by a rapid decline in blood and brain concentrations (Jones, 1990).Psychoactive substance disorders, as defined in the DSM-III-R, are divided intopsychoactive substance dependence, and a residual category, psychoactive substance abuse,which is reserved for individuals who exhibit pathological drug use, but fail to meet thecriteria for substance dependence. The criteria for psychoactive substance dependence andabuse are presented in Tables 1 and 2, respectively. Most individuals who exhibit high-dose, binge use of cocaine would meet the criteria for cocaine dependence. Unfortunately,not all authors use these terms with reference to the DSM-III-R diagnostic criteria, or theyuse other terms, such as cocaine addiction, making it difficult to make comparisons betweenthe results of different investigators.3Table 1.Diagnostic Criteria for Psychoactive Substance Dependence A.^At least three of the following:1). substance often taken in larger amounts or over a longer period than theperson intended.2). persistent desire or one or more unsuccessful efforts to cut down or controlsubstance use.3). a great deal of time spent in activities necessary to get the substance (e.g.theft), taking the substance, or recovering from its effects.4). frequent intoxication or withdrawal symptoms when expected to fulfill majorrole obligations at work, school, or home (e.g. does not go to work becausehung over, goes to school or work "high", intoxicated while taking care of hisor her children), or when substance use is physically hazardous (e.g. drivingwhile intoxicated).5). important social, occupational, or recreational activities given up or reducedbecause of substance use.6). continued substance use despite knowledge of having a persistent or recurrentsocial, psychological, or physical problem that is caused or exacerbated by theuse of the substance.Note: The following items may not apply to cannabis, hallucinogens or phencyclidine8). characteristic withdrawal symptoms (e.g. fatigue, insomnia or hypersomnia,and/or psychomotor agitation for cocaine withdrawal).9). substance often taken to relieve or avoid withdrawal symptoms.B.^Some symptoms of the disturbance have persisted for at least 1 month, or haveoccurred repeatedly over a longer period of time.Adapted from DSM-III-R Diagnostic and Statistical Manual of Mental Disorders, ed 3,revised. American Psychiatric Association, Washington, DC, 1987.4Table 2.Diagnostic Criteria for Psychoactive Substance Abuse A.^A maladaptive pattern of psychoactive substance use indicated by at least one of thefollowing:1). continued use despite knowledge of having a persistent or recurrent social,occupational, psychological, or physical problem that is caused or exacerbatedby the use of the psychoactive substance.2). recurrent use in situations in which use is physically hazardous (e.g. drivingwhile intoxicated).B.^Some symptoms of the disturbance have persisted for at least 1 month, or haveoccurred repeatedly over a longer period of time.C.^Never met the criteria for psychoactive substance dependence for this substance.Adapted from DSM-III-R Diagnostic and Statistical Manual of Mental Disorders, ed 3,revised. American Psychiatric Association, Washington, DC, 1987.5A triphasic syndrome associated with the abstinence from cocaine has been describedby Gawin and Kleber (1986). This syndrome is represented schematically in Figure 1. Thefirst phase, the crash, immediately follows the cessation of a cocaine binge, and ischaracterized by symptoms that vary throughout the crash phase. During the initial periodof phase 2, withdrawal, the individual experiences improved mood, normalized sleep andlow levels of cocaine craving; however, as this phase proceeds anhedonia, anergia, anxietyand increased cocaine craving, especially in response to stimuli previous associated withcocaine use, become apparent. The third phase, extinction, appears to represent a period ofextended vulnerability to relapse, although mood and withdrawal anhedonia havenormalized.A clear understanding of the clinical characteristics of cocaine abuse or dependenceis useful for a number of reasons. First, the informed clinician will have greater predictivepower regarding the probable course of cocaine abuse for individuals seeking treatment, andhence be capable of formulating more rational treatment programs that specifically addressproblems associated with the abstinence from cocaine. Second, the relative dirth ofempirical clinical data regarding cocaine's abuse potential that contributed to the belief thatcocaine was a harmless stimulant will be avoided in the future. Finally, an awareness of thesalient features of cocaine abuse will allow basic researchers investigating this topic toconsider their findings in a broader context.Neurobiology of Cocaine-induced RewardThe investigation of the neurobiology of cocaine abuse has proven to be aparticularly fruitful endeavor, as a number of animal models have been developed toinvestigate this topic. For example, the self-administration paradigm (Weeks, 1962) hasbeen used by numerous investigators to examine the neurochemical and neuroanatomicalsubstrates that mediate cocaine's ability to function as a positive reinforcer. The results of alarge number of studies using various pharmacological challenges and specific brain lesionsFigure 1. Schematic representation of the phases of the cocaine abstinence syndrome.Duration and intensity of symptoms vary on the basis of binge characteristics andpatient history and diagnosis. Binges range from under four hours to six or moredays. [Modified from Gawin and Kleber, 1986]6^Phase 1^ Phase 2^ Phase 3Crash Withdraw! Extinction9 hours to 4^days^1 to 10 weeks^Indefinite...111110.Cocaine BingeRelapseEarlyAgitationDepressionAnorexiaHigh Cocaine CravingMiddleFatigueDepressionNo Cocaine CravingInsomnia WithIncreasing Desirefor SleepLateExhaustionHypersomnolenceHyperphagiaNo Cocaine CravingEarlySleep NormalizedEuthymic MoodLow Cocaine CravingLow AnxietyMiddle and LateAnhedoniaAnergiaAnxietyHigh Cocaine CravingConditioned CuesExacerbateCravingNormal HedonicResponseEuthymic MoodEpisodic CravingConditioned CuesTrigger CravingAbstinence8strongly implicate the mesolimbic dopaminergic pathway in the rewarding effects of cocaine(Fibiger et al., 1992; Fibiger and Phillips, 1987; Johanson and Fischman, 1989). Forexample, Ritz and colleagues (1987) have reported that although cocaine blocks the reuptakeof noradrenaline, dopamine (DA) and serotonin, its action at the DA uptake site appearsmost directly related to its rewarding effects. This conclusion is in agreement with studiesthat have demonstrated that low doses of specific DA receptor antagonists increase the rateof responding for intravenous cocaine (De Wit and Wise, 1977; Ettenberg et al., 1982;Roberts and Vickers, 1984; Koob et a/., 1987a), while neither a- or /3-adrenergic receptorantagonists affect cocaine self-administration (De Wit and Wise, 1977; Wilson and Schuster,1974; Woolverton, 1987). These results have been interpreted by some investigators toindicate that low doses of DA receptor antagonists produce a partial DA receptor blockadethat results in an increase in the rate of responding to maintain the previous level ofpostsynaptic activation. However, it has also been proposed that "dopamine (receptor)antagonists exert their effect by antagonizing effects of cocaine unrelated to itsreinforcement" (Johanson and Fischman, 1989). Recent findings, however, lend additionalsupport for the hypothesis that DA receptor antagonists can affect the self-administration ofcocaine in a reward-related manner. Roberts et al. (1989) found that a low dose ofhaloperidol, which had previously been shown to produce an increase in responding forcocaine on a continuous reinforcement schedule (Roberts and Vickers, 1984), produced asignificant reduction in the break point for cocaine in the progressive ratio self-administration paradigm, a measure which varies directly with the strength of the rewardingeffects of the self-administered drug. Moreover, Bergman and colleagues (1990) havereported that the dose-response curve for cocaine self-administration is shifted to the rightby selective D1 and D2 receptor antagonists. These data, while consistent with thehypothesis that DA receptor antagonists attenuate the rewarding properties of cocaine, arenot readily accounted for by the simple rate-reduction antagonism hypothesis (Johanson andFischman, 1989). It should be noted, however, that rate of responding has been shown to be9affected by factors independent of the rewarding effects of cocaine, and that conclusionsbased on changes in the simple rate of responding must be interpreted with caution(Johanson and Fischman, 1989).The previously discussed results are further supported by studies that haveestablished that 6-hydroxydopamine (6-OHDA) lesions of the mesolimbic dopaminergicpathway, which produce extensive depletions of DA in the nucleus accumbens, producedramatic suppression in the rate of lever pressing for cocaine (Pettit et a/., 1984; Roberts etal., 1977, 1980). This finding does not appear to be the result of a generalized motor deficitor other nonspecific effects, as Dworkin and Smith (1988) have demonstrated that 6-OHDAlesions of the nucleus accumbens suppress the ascending limb of the cocaine self-administration dose-effect curve, while responding for food and water was unaffected.Furthermore, Koob et al. (1987b) have reported that the break point for cocaine in theprogressive ratio self-administration paradigm was significantly reduced by 6-OHDA lesionsof the nucleus accumbens.Although the aforementioned results suggest that the dopaminergic projection to thenucleus accumbens plays a critical role in the reinforcing effects of cocaine, otherobservations suggest that additional factors are involved in the self-administration of thispsychomotor stimulant. For example, Roberts and Koob (1982) reported that while 6-OHDA lesions of the ventral tegmental area, the origin of the dopaminergic projections tothe nucleus accumbens and other limbic structures, produced significant reductions in therate of cocaine self-administration, there was no significant correlation between themagnitude of the DA depletion in the nucleus accumbens and the change in respondingfrom prelesion rates. Moreover, several rats with considerable depletions of accumbens DAresponded at near normal pre-lesion rates for cocaine. The authors speculated that thedopaminergic innervation of other structures, such as the olfactory bulb, frontal cortex oramygdala, or a combination of structures may be critical in the self-administration ofcocaine. To this end, Goeders and Smith (1983) reported that rats will self-administer10cocaine directly into the medial prefrontal cortex, but not into the nucleus accumbens. Toexamine the role of this dopaminergic projection area in the role of systemicallyadministered cocaine, Martin-Iverson et al. (1986) investigated the effect of 6-OHDAlesions of the medial prefrontal cortex on the rate and pattern of cocaine self-administration. Although 6-OHDA lesions produced approximately a 95% depletion in DAin the medial prefrontal cortex, no significant effect of these lesions on either the rate orpattern of cocaine self-administration was observed. These results strongly suggest that thedopaminergic innervation of the medial prefrontal cortex is not necessary for the reinforcingeffects of intravenously administered cocaine. Given the previously discussed shortcomingsof using rate of responding as a measure of the reinforcing properties of cocaine,reexamination of the importance of the medial prefrontal cortex and other DA projectionareas is warranted.The rewarding properties of cocaine have also been extensively investigated using theconditioned place preference (CPP) paradigm. This paradigm is based on the principle thatif rewarding or appetitive stimuli are reliably associated with a given environment thenthere will be an increased tendency to approach or maintain contact with that environment(Carr et al., 1989). Although the findings of self-administration and CPP studies aregenerally in agreement, it should be noted that these paradigms have divergent theoreticalbackgrounds and behavioural demands and cannot be assumed necessarily to examineidentical phenomena (Wise, 1989). Cocaine has been shown to produce a CPP by numerousinvestigators (Bardo et al., 1984; Mackey and van der Kooy, 1985; Morency and Beninger,1986; Nomikos and Spyraki, 1988; Spyraki et al., 1982a, 1987). However, it appears thatmultiple mechanisms are potentially involved in cocaine-induced CPP, as this effect hasbeen reported to be dependent on dopaminergic transmission under some conditions(Morency and Beninger, 1986; Spyraki et al., 1987), while independent of dopaminergictransmission in other studies (Mackey and van der Kooy, 1985; Morency and Beninger,1986; Spyraki et al., 1982a). Although these data suggest that DA is not necessary for11cocaine-induced CPP following i.p. or s.c. administration, they cannot be assumed toindicate that dopaminergic transmission is not normally involved in this behaviour. Clearly,an understanding of the apparent DA-independent actions of cocaine in the CPP paradigmwill assist in further addressing these discrepancies.The advent of in vivo microdialysis has dramatically increased current understandingof the neurochemical actions of cocaine. The ability of behaviourally relevant doses ofcocaine to increase interstitial concentrations of DA in the rat nucleus accumbens has beenwell established (Bradberry and Roth, 1989; Di Chiara and Imperato, 1988a; Moghaddamand Bunney, 1989a; Pettit and Justice, 1989, 1991). Moreover, Pettit and Justice (1989,1991) have demonstrated that cocaine self-administration produces significant dose-relatedincreases in interstitial DA in the nucleus accumbens. Interestingly, cocaine has beenreported to preferentially increase interstitial DA in the nucleus accumbens, as compared toits effect in the striatum (Di Chiara and Imperato, 1988a) and the medial prefrontal cortex(Moghaddam and Bunney, 1989a). When considered together with the previously discussedbehavioural studies, recent microdialysis findings provide further support for the importanceof the mesolimbic dopaminergic projection to the nucleus accumbens in the rewardingproperties of cocaine.SummaryCocaine is presently recognized as a powerful stimulant possessing significant abusepotential, as well as a specific pattern of abuse and withdrawal (DSM-III-R; Gawin, 1991;Jaffe, 1989). A large body of evidence indicates that the mesolimbic DA system plays afundamental role in the reinforcing properties of cocaine (Di Chiara and Imperato, 1988;Fibiger et al., 1992; Fibiger and Phillips, 1987; Johanson and Fischman, 1989; Roberts et al.,1977; 1989). Despite the growth of both the clinical understanding of cocaine abuse and theneurobiological actions of cocaine, few attempts have been made to integrate the findings ofthese fields of investigation in a meaningful way. The following studies are a modestattempt to approach this objective.12II. BEHAVIOURAL AND NEUROCHEMICAL INTERACTIONS BETWEEN COCAINEAND BUPRENORPHINE(A) IntroductionBuprenorphine (BUP) is a synthetic opioid with the pharmacological profile of amixed agonist-antagonist. It has potent antinociceptive actions (Cowan et al., 1977; Dumand Herz, 1981) with a limited capacity for producing physical dependence (Cowan et al.,1977; Jasinski et al., 1978; Mello and Mendelson, 1980; Yanagita et al., 1981). BUP can alsoattenuate opiate self-administration in humans (Mello and Mendelson, 1980; Mello et al.,1982) and in non-human primates (Mello et al., 1983).Recently, it has been reported that BUP suppresses cocaine self-administration byrhesus monkeys (Mello et al., 1989). Based on these data, it was suggested that BUP mightbe useful in the pharmacotherapy of cocaine abuse. This proposal is supported by apreliminary report by Kosten and colleagues (1989) that indicated that heroin addicts treatedwith BUP had a significant reduction in cocaine-positive urines, compared to those subjectstreated with methadone. However, BUP is a potential drug of abuse, as indicated byepidemiological (Chowdhury and Chowdhury, 1990; Lewis, 1985; O'Connor et al., 1988;Strang, 1985) and primate self-administration studies (Lukas et al., 1986; Mello et al., 1981;Woods, 1977; Young et al., 1984). Moreover, as is the case for a variety of drugs of abuse,BUP decreases the threshold for intracranial self-stimulation (Hubner and Kornetsky, 1988).Cocaine or morphine pretreatment also suppresses responding for cocaine by non-humanprimates (Balster et al., 1992; Herling et al., 1979; Stretch, 1977), indicating that decreases inthe rate of responding in self-administration paradigms can occur for reasons other thandecreases in the rewarding properties of the reinforcer. These observations raise thepossibility that the BUP-induced decrease in cocaine self-administration reported by Melloet al. (1989) was due to a summation of the rewarding properties of these agents. In view13of these considerations, the conditioned place preference (CPP) paradigm was utilized toaddress two specific questions. First, can BUP produce CPPs? Second, is cocaine-inducedCPP affected by BUP?Although cocaine blocks the reuptake of a variety of biogenic amines, its action atthe DA uptake site appears to be most directly related to its rewarding effects (Ritz et al.,1987). This conclusion is consistent with self-administration studies that utilized specificneurotoxic lesions to illustrate that the integrity of the mesolimbic DA pathway to thenucleus accumbens is necessary for cocaine self-administration (Pettit et al., 1984; Roberts etal., 1977; Roberts et al., 1980). Moreover, low doses of specific DA receptor antagonistsproduce predictable alterations in self-administration of cocaine (De Wit & Wise, 1977;Ettenberg et al., 1982; Roberts & Vickers, 1984; Roberts et al., 1989), and local infusions ofthe DA receptor antagonist spiroperidol into the nucleus accumbens disrupts cocaine self-administration (Phillips & Broekkamp, 1980). In vivo microdialysis studies further suggestthat the rewarding properties of cocaine are mediated by increases in interstitial DA in thenucleus accumbens (Di Chiara & Imperato, 1988a; Moghaddam & Bunney, 1989a; Pettit &Justice, 1989, 1991).Although the association between the rewarding properties of cocaine and the releaseof DA in the nucleus accumbens is well established, the importance of mesolimbic DA inthe rewarding properties of opioids is unclear. Ettenberg et al. (1982) reported thatpretreatment with the neuroleptic a-flupenthixol produced a compensatory increase incocaine self-administration, but had no effect on heroin self-administration. In agreementwith this finding, 6-OHDA lesions, which reduced cocaine self-administration, had nolasting effect on heroin self-administration (Pettit et al., 1984). However, other findingssuggest that the DA projections from the ventral tegmental area (VTA) to the nucleusaccumbens may play a role in the rewarding effects of opioids. First, it has been reportedthat infusions of opioids directly into the VTA are rewarding, as assessed by self-administration (Bozarth and Wise, 1981a; Welzl et al., 1989) and CPP (Bals-Kubik et al.,141993; Phillips and LePiane, 1980, 1982). Second, heroin-induced CPPs are attenuated byneuroleptics (Bozarth & Wise, 1981b; Spyraki et al., 1983). Third, infusions of opioidreceptor antagonists directly into the VTA decrease the rewarding properties of systemicheroin (Britt and Wise, 1983). Finally, results from in vivo microdialysis studies demonstratethat p receptor agonists, such as morphine, methadone, fentanyl and ED-Ala2 , N-methyl-Phe4 , G1y5 -oli-enkephalin (DAMGO), increase interstitial DA concentrations in the nucleusaccumbens (Di Chiara and Imperato, 1988b; Spanagel et al., 1990). Because BUP increasesDA turnover at behaviourally relevant doses (Cowan et al., 1976), it is possible that theeffect of BUP on cocaine self-administration (Mello et al., 1989) may be the result ofinteractions between these agents on DA release in the nucleus accumbens. To determine ifBUP alters the effects of cocaine on interstitial DA concentrations, in vivo microdialysisstudies were also undertaken.(B) Materials and MethodsSubjectsSubjects were 268 male Long Evans rats (Charles River, Quebec), weighing 270-350g at the beginning of the experiment. The rats were group housed (four per cage), on a12:12 h light:dark cycle (lights on 07:00), with food and water available ad libitum. Allsubjects were handled daily for one week prior to initiation of the experiment. Allexperimental procedures were conducted at approximately the same time each day, duringthe animal's light phase.15ApparatusPlace preference conditioning was conducted in four identical shuttle boxes (78 x 25cm, 35 cm high). Each box was divided into two compartments (36 x 25 cm) joined by atunnel (6 x 8 x 8 cm) that could be closed at both ends by guillotine doors. The twocompartments differed in the appearance of the walls and the type of floor (solid brownwalls and a 1.2 cm wire mesh floor versus walls with black and white strips 1 cm wide anda floor of parallel bars spaced 1.2 cm apart). Translucent Plexiglas lids allowed a moderateamount of light from overhead incandescent lights to enter each compartment. Each shuttlebox was mounted on a fulcrum, allowing its position to be detected by microswitches. Thetime spent in each compartment and the number of crosses between compartments wasrecorded with dedicated electronic equipment.ProcedureThe procedure for CPP consisted of three phases: habituation, conditioning and test.During the three day habituation phase, rats were placed in one of the compartments of theshuttle box (hereafter referred to as the start side), following a counterbalanced design.Following placement in the start side, and with both guillotine doors raised, the rats weregiven access to both compartments during the 900 s trial. During each trial, the time spentin each compartment and the number of crosses between compartments were recorded. Theconditioning phase was conducted over the next eight days. On days 1, 3, 5 and 7, ratswere given drug injections and immediately confined to the non-start side for 30 min. Onalternate days, rats were injected with saline and confined to the start side for 30 min. Onthe test day, each rat was placed in the start-side and given access to both compartments;the time spent in each compartment and the number of crosses between compartments wererecorded.16Cocaine-Induced CPPThe first experiment was conducted to determine the dose-response relationship forcocaine in the CPP paradigm. Animals were randomly assigned to one of four groups (n=12per group), that received 0, 1.25, 2.5 or 5.0 mg/kg cocaine.BUP-Induced CPPThe dose response relationship for BUP in the CPP paradigm was determined in thisexperiment. Animals were randomly assigned to one of nine groups (n=12 per group) thatreceived 0, 0.005, 0.01, 0.03, 0.075, 0.15, 0.3, 0.6 or 0.9 mg/kg BUP.Cocaine & BUP, CPP ExperimentsBased on the dose-response data obtained in the first two experiments, two studieswere conducted to examine the effects of combinations of cocaine and BUP in the CPPparadigm. In the first of these studies, animals were randomly assigned to one of fourgroups (n=12 per group) that received either vehicle + vehicle, cocaine + vehicle, vehicle +BUP or cocaine + BUP. Cocaine and BUP were given in doses of 1.5 and 0.01 mg/kg,respectively. The second study of this experiment used the same design as the first, exceptthat cocaine and BUP were given in doses of 5.0 and 0.075 mg/kg, respectively.Cocaine & BUP, Dialysis ExperimentsRats were anaesthetized with sodium pentobarbital (50 mg/kg i.p.), mounted in astereotaxic instrument, and a vertical microdialysis probe was implanted into the nucleusaccumbens (AP: +3.6 mm; ML: -1.5 mm; DV: -8.2 mm from dura; relative to bregma;Pellegrino et al., 1979). The microdialysis probe was a variant of the concentric verticaldesign (outer diameter = 250 pm, molecular weight cut-off = 6000 Dalton; Spectra Pordialysis fibre). The active surface was 2 mm in length, and was 0.2 mm from the tip of theprobe. Recovery of DA, 3,4-dihydroxyphenylacetic acid (DOPAC), homovanillic acid17(HVA) and 5-hydroxyindoleacetic acid (5-HIAA) was measured in vitro at 37°C and at aflow rate of 5 pl/min. The in vitro recoveries (%) for this probe were (mean ± SD, n=6):3.5 ± 0.4 (DA), 4.0 ± 0.5 (DOPAC), 3.7 ± 0.4 (HVA) and 4.3 ± 0.5 (5-HIAA).Following surgery, the rats were individually housed for 48 h prior to testing. Onthe test day, after a stable baseline had been established (not more than 5% variation for 3consecutive samples), each rat was injected with either cocaine + saline, BUP + saline,cocaine + BUP or saline + saline. Microdialysis samples were taken for five hours post-injection. Histological verification of the probe placement was conducted following testing.Microdialysis experiments were conducted on-line (Damsma et al., 1990), such thatthe microdialysis probe was perfused at 5 p1/min (Harvard Apparatus perfusion pump); theresulting dialysate was directed to the sample loop (100 pl) of an injector (Valco) throughthe outlet tubing (PE-10, Clay Adams). Samples were automatically injected into a highperformance liquid chromatographic (HPLC) system with electrochemical detection every 20min. The perfusion fluid contained: NaCl (147 mM), KC1 (3.0 mM), CaCl2 (1.3 mM),MgC12 (1.0 mM), and sodium phosphate (1.5 mM, pH 7.3).The concentrations of DA, DOPAC, HVA and 5-HIAA in the dialysate weredetermined with HPLC in conjunction with electrochemical detection. Separation of DAand the acid metabolites was achieved by reversed-phase liquid chromatography (150 x 4.6mm, Nucleosil 5 C18, Chrompack). The flow rate of the mobile phase [0.1 M acetic acidadjusted to pH 4.1 with solid sodium acetate, 0.01 mM EDTA, 0.35 - 0.5 mM octanesulfonicacid (Kodak) and 10 % methanol] was 1.85 ml/min. Detection of the amines was achievedby the sequential oxidation and reduction of the samples by a coulometric detection system(coulometric electrode = +0.4 V; amperometric electrode = -0.2 V; ESA Inc.). The acidmetabolites were quantified by their oxidation at the coulometric electrode, while DA wasmeasured at the subsequent amperometric electrode. The detection limit of this assay wasapproximately 5 fmol/injection for DA and 20 fmol/injection for DOPAC, 5 - HIAA andHVA.18DrugsCocaine hydrochloride (Sigma) and buprenorphine hydrochloride (Schering) weredissolved in isotonic saline and were injected intraperitoneally (i.p.). In both the cocaineand BUP CPP experiments, drugs were injected in a volume of 3 ml/kg; in the combinationexperiments (cocaine and BUP), drugs were injected in a volume of 1 ml/kg. Doses ofcocaine are expressed as weight of salt, while those of buprenorphine are expressed as thebase.Statistical AnalysisTrend analysis was used to evaluate the results from the cocaine and BUP CPPexperiments. Data were weighted based on the dose of the drug. Within-group comparisonswere then conducted using the Bonferroni t statistic to assess which doses producedsignificant CPP. An initial analysis of variance illustrated that the preconditioning valuesfor the time spent on the non-start side did not differ between groups [F(20,231)=0.30,p=0.999], justifying the use of within-subject comparisons. The results from thecombination CPP experiments (cocaine and BUP) were also analyzed using the Bonferroni tstatistic. A Reverse Helmert analysis was conducted to further examine some of the CPPdata. The data from the dialysis experiments (% values), from time 0 to 300 min, wereevaluated using a two-way analysis (Treatment X Time) of variance with repeated measures(Geisser-Greenhouse adjustment of d.f.). Percentage values of dialysate concentrations werebased on an average of the three samples prior to the injection of drug. Comparisons ofdifferences between groups at specific time points were made using the Bonferroni tstatistic. All statistics were performed using SPSS:X version.3 software.(C) ResultsCocaine-induced CPPThe results of this experiment indicate that the time spent in the drug-paired (non-start) side was directly related to the dose of cocaine (Figure 2). Specifically, a significantlinear dose-response function was observed [F(1,44)=7.26, p<0.02]. Within-groupcomparisons illustrated that only the 5.0 mg/kg group spent significantly more time on thedrug-paired side of the apparatus [t(11)=2.6, p<0.05].BUP-induced CPPBUP elicited a dose-related CPP (Figure 3). The data were best fit by a cubic dose-response function [F(1,99)=13.88, p<0.001], although a linear dose-response function was alsoobserved [F(1,99)=4.83, p<0.05]. Within-group comparisons indicated that at 0.03 mg/kg,0.075 mg/kg, 0.15 mg/kg and 0.3 mg/kg BUP produced significant CPP [t(11)=5.28, 4.54,5.78, 3.61, respectively, p<0.001].Cocaine & BUP, CPP ExperimentsConsistent with the data from the previous experiments, low doses of BUP (0.01mg/kg) or cocaine (1.5 mg/kg) did not by themselves produce CPP (Figure 4). However, awithin-group comparison indicated that animals pretreated with a combination of BUP andcocaine did exhibit CPP [t(11)=2.63, p<0.05]. This finding was replicated using anothergroup of animals [t(I1)=2.55, p<0.05].CPP was produced by moderate doses of cocaine (5.0 mg/kg) [z(11)=2.89, p<0.01] andBUP (0.075 mg/kg) [t(11)=4.09, p<0.001], as well as by a combination of these drugs[t(11)=6.25, p<0.001] (Figure 5). A Reverse Helmert analysis of the three treatmentconditions indicated that, while the cocaine and BUP treatment groups did not differ1920Figure 2. The effects of cocaine on the time spent in the drug-paired (non-start)compartment before (^) and after (u) conditioning (n = 12/group). Values representthe mean ± SEM. * indicates a significant within-group difference (p<0.05) of pre-versus post-conditioning scores.21EI Pre—conditioningI. Post—conditioning*7;a) 600 —cawa 550 -(75cw 500 -cc7,7cCL(91 450 -=cca 400 -zzI— 350 -wCL00.0^1.25^2.5COCAINE (mg/kg)MIw 300 -MI7-5.022Figure 3. The effects of buprenorphine on the time spent in the drug-paired (non-start)compartment before (0) and after (•) conditioning (n = 12/group). Values representthe mean ± SEM. * indicates a significant within-group difference (p<0.05) of pre-versus post-conditioning scores.2324Figure 4. The effects of saline, cocaine (1.5 mg/kg), BUP (0.01 mg/kg), and cocaine (1.5mg/kg) + BUP (0.01 mg/kg) on the time spent in the drug-paired (non-start)compartment before (^) and after (■) conditioning (n = 12/group). Values representthe mean ± SEM. * indicates a significant within-group difference (p<0.05) of pre-versus post-conditioning scores.257;650-Nw0 600-E50w 550-cc47(^-a.01^500-ncca 450-z}-- 400-Zwo_coO Pre—conditioning• Post—conditioning*0.0+Coc 1.5+Coc 0.0 +C oc 1.5+Coc0.0 Bup 0.0 Bup 0.01 Bup 0.01 Bup26Figure 5. The effects of saline, cocaine (5.0 mg/kg), BUP (0.75 mg/kg), and cocaine (5.0mg/kg) + BUP (0.75 mg/kg) on the time spent in the drug-paired (non-start)compartment before (0) and after (■) conditioning (n = 12/group). Values representthe mean ± SEM. * indicates a significant within-group difference (p<0.05) of pre-versus post-conditioning scores. f indicates a significant difference (p<0.05) betweenthe combination group (cocaine + BUP) and the cocaine and the BUP treatmentgroups, as determined by a reverse Helmert analysis.270.0+Coc 5.0+ Coc 0.0 Coc+5.0+Coc0.0 Bup 0.0 Bup 0.075 Bup 0.075 Bup1U• 700-co.......w0 650-(7)0 600-wcc•Tia. 550-1^-0ccD 500-0z 450-I—z 400-wa_(/)350-wMI=^ Pre—conditioningIN Post—conditioning**significantly from each other [F(1,33)=0.75], the combined treatment was significantlydifferent from the individual treatments [F(1,33)=5.29, p<0.05].Cocaine & BUP, Dialysis ExperimentsThe average baseline output of DA (± SEM, n=16) was 1.84 fmol/min ± 0.48, anddid not differ significantly between groups [F(3,12)=0.68]. The average basal outputs (±SEM, n=16) for DOPAC, HVA and 5-HIAA were 533 ± 65, 297 ± 42, and 245 ± 54fmol/min, respectively. Basal values of DOPAC [F(3,12)=1.53], HVA [F(3,12)=0.18] and 5-HIAA [F(3,12)=0.69] did not differ significantly between the groups.Cocaine produced a rapid increase in interstitial concentrations of DA[F(3.29,19.74)=11.76, p<0.001] that returned to baseline within 120 to 160 min (Figure 6).Cocaine also significantly decreased DOPAC [F(2.70,16.23)=4.85, p<0.05] and HVA[F(2.77,16.65)=4.68, p<0.05] without affecting the serotonin metabolite, 5-HIAA[F(2.41,12.05)=1.45]. All statistical comparisons are made against saline injected controls(data not shown).Interstitial DA was gradually increased by BUP [F(3.99,23.92)=9.80, p<0.001],reaching approximately 200% of baseline values five hours post injection (Figure 7).Interstitial concentrations of DOPAC [F(3.17,19.04)=5.21, p<0.01] and HVA[F(3.32,19.92)=15.73, p<0.001] were also increased by BUP, although the effect on DOPACwas not sustained over the five hour test period. Interstitial 5-HIAA was unaffected byBUP [F(2.60, 15.62)=1.63].Co-administration of cocaine and BUP produced a rapid increase in the interstitialconcentration of DA that returned to baseline within 120 to 160 min [F(2.45,14.73)=8.80,p<0.005] (Figure 8). DA concentrations showed a further modest increase after 180 min.DOPAC [F(3.12,18.72)=12.18, p<0.001] and HVA [F(2.96,17.76)=14.89, p<0.001] were alsosignificantly altered by the drug combination, as compared to saline injected controls. TheDA metabolites showed a transient decrease, followed by a gradual increase over the2829Figure 6. The effects of cocaine (5.0 mg/kg) + saline on microdialysis output of DA (o),DOPAC (^), HVA (,n,) and 5-HIAA (0). Cocaine was administered at time 0. Valuesrepresent the group mean (n = 4) + SEM. Percentage values of dialysateconcentrations were based on an average of the three samples prior to the injectionof drug.200 -13180 -cliiw 160 -ft(UEl95 —TIME (min)Figure 15. Locomotor activity (horizontal distance) of conditioned (Cs+, n=6) andpseudoconditioned (Cs - , n=6) subjects from Experiment 4. Values represent thegroup mean ± SEM. Inset: Total horizontal distance for the conditioned andpseudoconditioned subjects for the 30 min test period. * p < 0.05 compared topseudoconditioned controls.64- 7500E0- 5000 0z4CF-cna- 2500 _1