Intravenous self-administration of the cannabinoid CB1 receptor agonist WIN 55,212-2 in rats

Psychopharmacology (2001) 156:410 416 DOI 10.1007/s002130100734 ORIGINAL INVESTIGATION Liana Fattore Gregorio Cossu Cristina M. Martellotta Walter Fratta Intravenous self-administration of the cannabinoid CB1 receptor agonist WIN 55,212-2 in rats Received: 10 August 2000 / Accepted: 12 February 2001 / Published online: 11 May 2001 Springer-Verlag 2001 Abstract Rationale: 9 -Tetrahydrocannabinol ( 9 -THC), the main psychoactive ingredient of marijuana, as well as synthetic cannabinoid (CB1) receptor agonists, has led to negative or equivocal results when tested with the intravenous self-administration procedure, the best validated behavioural model for evaluating abuse liability of drugs in experimental animals. We recently reported, however, that the synthetic CB1 receptor agonist WIN 55,212-2 is intravenously self-administered by drugnaive mice and that its self-administration is blocked by the cannabinoid CB1 receptor antagonist SR 141716A. Objective: To assess a reliable model of cannabinoid intravenous self-administration in rats. Long Evans male rats were allowed the opportunity to self-administer WIN 55,212-2 at doses ranging from 6.25 to 50 µg/kg per injection, under a fixed-ratio 1 (FR1) schedule of reinforcement and nose-pokes as the operant responses. The effect of either a change in the unit drug dose available or a pretreatment with the specific CB1 receptor antagonist SR 141716A were then investigated (maintenance phase). Finally, the extinction of the self-administration behaviour was evaluated. Results: Response rate depended on the drug dose available, with maximum rates occurring at 12.5 µg/kg per injection. Response rate increased following pretreatment with the specific CB1 receptor antagonist, SR 141716A. Moreover, operant behaviour rapidly extinguished following both the substitution of saline or vehicle for cannabinoid and the disconnection of the drug delivery pumps. Conclusion: Rats will intravenously self-administer the synthetic CB1 receptor agonist WIN 55,212-2 under specific experimental conditions, thus allowing further investigation of the L. Fattore W. Fratta ( ) B.B. Brodie Department of Neuroscience, University of Cagliari and Centre for Neuropharmacology, CNR, 09124 Cagliari, Italy e-mail: wfratta@unica.it Tel.: +39-70-6754313, Fax: +39-70-6754312 G. Cossu M.C. Martellotta Neuroscience S.c.a.r.l., 09100 Cagliari, Italy neurobiological mechanisms underlying cannabinoidtaking behaviour. Keywords Cannabinoid CB1 receptor WIN 55,212-2 Intravenous self-administration SR 141716A Rewarding properties Operant behaviour Rat Introduction Cannabis is one of the oldest abused drugs and still represents the most widely used illicit recreational drug (Rouse 1996; Chen et al. 1997). However, there have not been reliable behavioural models for elucidation of the mechanisms mediating the subjective effects of cannabinoids seen in humans. To date, controversial data have been reported on the ability of cannabinoids to induce in the rat both place and taste preference (Switzman et al. 1981; Lepore et al. 1995; Parker and Gillies 1995; McGregor et al. 1996; Sañudo-Peña et al. 1997; Chaperon et al. 1998; Mallet and Beninger 1998; Valjent and Maldonado 2000) and arousal and brain self-stimulation (Bhattacharyya et al. 1980; Stark and Dews 1980; Kucharski et al. 1983; Gardner et al. 1988; Lepore et al. 1996). Indeed, the numerous efforts to study the reinforcing properties of cannabinoids by means of intravenous self-administration (IVSA) techniques, the most reliable measure of drug addictive liability, have provided varying results (Harris et al. 1974; van Ree et al. 1978; Takahashi and Singer 1979; Mansbach et al. 1994). Moreover, rats have been shown to be reluctant to selfadminister Cannabis sativa extracts orally (Corcoran and Amit 1974; Leite and Carlini 1974). Since almost all drugs abused by humans are easily self-administered by rats (Collins et al. 1984), drug selfadministration studies in rats have greatly contributed towards our understanding of central mechanisms involved in drug-seeking behaviour (Koob and Goeders 1989). Although cannabinoids, similarly to other abused drugs, serve as positive reinforcers in several animal species including humans (Chait and Zacny 1992), it has been rather

difficult to demonstrate their rewarding properties in most commonly used animal models of addictive behaviour. By means of an acute IVSA model in drug-naive mice, we recently demonstrated that the two synthetic cannabinoid CB1 receptor agonists, WIN 55,212-2 and CP 55,940 are self-administered according to a concentration-dependent, bell-shaped curve (Fratta et al. 1998; Martellotta et al. 1998). This particular behavioural model, where animals are given the opportunity to self-administer the drug only once ( drug-naive mice), has proved to be particularly useful in evaluating the effect of the first exposure to a drug of abuse. Nevertheless, the development of chronic IVSA of cannabinoids and the time course of such behaviour still remained to be determined. In the present study, we investigated those experimental conditions and technical factors which lead to IVSA of cannabinoids by rats, using a nose-poking like-response operandum and a fixed ratio (FR1) schedule of reinforcement, each nose-poke resulting in the delivery of the reinforcer. In an effort to develop an animal model of chronic IVSA of cannabinoids, we determined the best experimental conditions (food deprivation, restrained drug-access condition, fixed ratio schedule of reinforcement, drug unit dose and animal strain) under which rats self-administer cannabinoids. Moreover, the effects of both changes in the unit dose of drug available and pretreatment with the specific CB1 receptor antagonist SR 141716A were investigated. Finally, extinction of drug-seeking behaviour was also analysed. Thus, this study develops a reliable animal model of cannabinoids intravenous self-administration in rats, providing an important tool for the better evaluation of the neurobiological mechanisms through which cannabinoids produce their rewarding effects. Materials and methods Animals Seventy-five male Long Evans rats (Charles-River, Italy) weighing 250 280 g at the beginning of the study were housed four per cage under a 12-h reversed light/dark cycle (lights on 8:00 p.m.), with food and water ad libitum, and handled daily for approximately 10 min during the first week after arrival. Experimental procedures started the following week and took place at the same time each day during the dark phase of the cycle (between 8:30 and 12:30 a.m.). At the beginning of the IVSA sessions, food restriction conditions were applied (see below). All procedures were approved by the local Animal Care Committee and the EC regulations for animal use in research (86/609/EEC). 411 Surgery After anaesthesia with chloral hydrate (400 mg/kg IP), rats were surgically implanted with a chronic catheter, as previously described by Weeks (1972). Briefly, under sterile conditions, a silastic catheter (L-155; Silmedic medical grade, Gessil, France) was inserted into the right jugular vein, tunnelled under the skin and exited in the midscapular region, where it was bonded to a larger diameter extension (L-205) and was anchored to the back of the neck with two sutures. After surgery, each animal recovered in its home cage with food and water freely available and for the following 3 days received a daily infusion of 0.2 ml of a sterile solution containing heparin (1%) and gentamicin (0.08 mg/ml). Free passage of 0.2 ml of heparinized (1%) saline solution through the catheter was checked before and after each IVSA session throughout all the experiments. Drugs WIN 55,212-2 (RBI, Natick, Mass., USA) was dissolved in 1 drop of Tween 80 and diluted in heparinized (1%) saline solution (volume of injection=100 µl). SR 141716A (Sanofi, Montpellier, France) was dissolved in 1 drop of Tween 80, diluted in physiological solution and administered intraperitoneally (IP) 30 min before the session in a volume of 5 ml/kg. To ensure sterility, fresh WIN 55,212-2 solutions were filtered by 0.22 µm syringe filters prior to use. Apparatus The intravenous self-administration apparatus consisted of eight Plexiglas operant cages (30 30 30 cm). Two holes, provided with photobeam detectors, were made 2 cm above the floor, 15 cm apart. At the time of the IVSA session, the catheter was connected to a swivel system through a metal spring, which was in turn connected to an infusion pump via a plastic tube; the swivel system allowed the animal to move freely in the operant cage. All equipment was obtained from Med Associates (Georgia, Vt., USA). Nose-poking in one of the holes (defined as active) switched on the infusion pump, injecting the drug solution into the animal s venous system. Nose-poking in the other hole (defined as passive) had no effect on the pump. Assessment of experimental schedule and data collection was programmed through PC software. By using dual-hole operant chambers, we could assess specificity of responding on the drug-reinforced hole. Indeed, nose-pokes in the passive hole, which did not deliver drug infusions, were always recorded in order to verify whether self-administered cannabinoid produces a non-specific effect in animals. Intravenous self-administration procedure As shown in Fig. 1, 3 5 days following surgery, animals were first deprived of food for 24 h and then allowed 3-h daily access to WIN 55,212-2 (6.25, 12.5, 25 and 50 µg/kg per injection) under a continuous reinforcement (FR1) schedule. Each nose-poke resulted in an intravenous infusion of 0.1 ml of drug solution delivered over a period of 5 s. Coincident with onset of the infusion, a stimulus light was turned on for 10 s, during which time nose-pokes were recorded but had no consequences (time-out period). Each rat was given a priming infusion in the experimental chamber before the start of the daily session. IVSA sessions occurred once daily Monday through Saturday. Phase 1: acquisition of self-administration behaviour During this phase, food supply was restricted (15 g/day) to reduce the animal s body weight by 20%, thereby facilitating the initiation of self-administration behaviour (Carroll et al. 1979; de la Garza and Johanson 1987; de Vaca and Carr 1998). Food was given as a single meal immediately after each daily session. In animals that acquired self-administration behaviour, a substantial drug-maintained response was usually established by the end of the third week, remaining stable as long as catheters were patent. Animals failing to acquire stable responding for the CB1 agonist after 3 weeks of training were counted but dropped from the experiments. Phase 2: maintenance of self-administration behaviour Once trained, rats were kept in food-restriction conditions and fed with approximately 20 g of chow each day for the entire duration of

412 Fig. 1 Diagram outlining the experimental conditions and procedure (acquisition, maintenance and extinction phases) of WIN 55,212-2 intravenous self-administration in rats. For each experimental phase, corresponding length of time and food conditions (ad libitum, deprivation and restriction) are indicated Fig. 2 Patterns over the acquisition phase of intravenous cannabinoid self-administration for all WIN 55,212-2 doses tested. Mean numbers of infusions self-injected by rats in a 3-h session are indicated (n=12 14) the experiments. Only rats that developed a stable pattern of WIN 55,212-2 intake were allowed to continue the daily IVSA sessions. Stability was defined as a stable trend in the number of injections self-administered over five consecutive sessions. Thus, as the daily drug intake had stabilised with a range of less than 15%, rats were switched to a different unit dose of reinforcer or pretreated with the specific CB1 antagonist SR 141716A (1.5 and 3 mg/kg IP). Phase 3: extinction of self-administration behaviour After stable and reliable responding behaviour had been reached, an extinction period was introduced, both by replacing drug solution with saline or vehicle solution and disconnecting the syringe pump (De Wit and Stewart 1983; Schenk et al. 1996), allowing nose-pokes to be recorded without a drug consequence. Food-restriction conditions continued, with rats being fed 20 g of chow immediately after each session. Under these conditions, a nosepoke in the active hole resulted respectively either in an infusion of 100 µl of saline or vehicle or in the sole presentation of the stimulus light previously paired with drug delivery. Statistical analysis Due to the fact that not all animals completed the entire set of experiments (due to catheter blockages or leakage), data were computed as for independent rather than correlated samples. For animals tested with more than one drug treatment (upper/lower unit dose, pretreatment with the CB1 antagonist), a minimum of 5 days separated each treatment and the order of treatment presentation was counterbalanced. Each drug dose tested included a minimum of 12 animals, whereas SR 141716A pretreatment or changed unit dose groups were four to six each. The number of reinforcements earned during the 180-min session was recorded and statistical analysis of data was evaluated by one-way analysis of variance (ANOVA). Individual between-group comparisons were carried out using the Newman-Keuls post-hoc test when appropriate. Since nose-poke activity in the passive hole was virtually absent once a stable baseline of drug intake had been reached and throughout the experiments, this factor was not included in data analysis. Statistical significance was set at P<0.05. Results Acquisition of self-administration behaviour The mean number of sessions required to reach the first stable responding day (16±4 daily sessions) did not differ among the doses tested. Acquisition phase was characterised by an initial low level of intake, typically followed mostly by a gradual increase in the mean number of infusions or weak peaks in the average rate intake. Figure 2 shows the daily course of the mean responding

rate for each treatment group during the acquisition phase. The number of rats acquiring IVSA was higher at the doses of 12.5 and 25 µg/kg per injection, being 86.67 and 72.22%, respectively. When the highest dose (50 µg/kg per injection) was presented, fewer animals (57.14%) displayed operant behaviour, whereas at the lowest dose tested (6.25 µg/kg per injection) only one animal out of 12 supported IVSA. It is noteworthy that among animals not acquiring IVSA, some began self-administering WIN 55,212-2 but never stabilised drug intake and others did not initiate such operant behaviour at all. However, once stabilised, responding was maintained by each rat throughout the duration of the experiments. Fig. 3 Intravenous WIN 55,212-2 self-administration in rats. Each bar represents mean±sem of the mean cumulative numbers of reinforcers during 6 consecutive sessions immediately after the first 3 days of stable responding (n=12 14). Doses are expressed as µg/kg per injection. **P<0.01 and *P<0.05 significant difference from vehicle group (ANOVA followed by Newman-Keuls test) Maintenance of self-administration behaviour 413 As shown in Fig. 3, during the following maintenance phase, after stable levels of drug intake had been reached, rats exhibited significantly higher rate of operant behaviour when allowed contingent free access to WIN 55,212-2 (12.5, 25 and 50 µg/kg per injection) than to vehicle solution [F(3,22)=4.82, *P<0.05 and **P<0.01]. On the contrary, no significant differences in the rate of self-administration were observed at the lowest dose tested (6.25 µg/kg per injection). The average amount of daily drug intake was 277.4±14.5, 280.75±18.75 and 318.5±13.5 µg/kg for the CB1 receptor agonist doses of 12.5, 25 and 50 µg/kg per injection, respectively. However, daily intake did not exceed 26.9±2.44 µg/kg when the unit dose of 6.25 µg/kg per injection of cannabinoid was made available. The mean values of drug intake were reached within four to six sessions after the acquisition phase. Moreover, the number of infusions was inversely proportional to the selfadministered dose, a feature of the descending limb of dose-effect function: thus, higher doses of WIN 55,212-2 decreased rates of self-administration behaviour. The response pattern for the CB1 receptor agonist during the 3 h of daily IVSA session was further examined. As shown in Fig. 4, the average level of drug intake tends to be slightly greater at the beginning than at the end of the session, reaching its maximum at the unit dose of 12.5 µg/kg per injection. At this dose, rats typically self-administered five to eight infusions in the first 30 min and then maintained a rather regular intake per hour throughout the entire session, although characterised by longer intra-reinforcement intervals. At all three doses of WIN 55,212-2 which maintain self-administration behaviour, each rat responded to both Fig. 4 Individual representative rats cumulative response records for WIN 55,212-2 during the maintenance phase. Each record represents a separate 3-h session and each small vertical mark represents an intravenous infusion of the drug. Unit dose of self-administrated WIN 55,212-2 is indicated on the left of each record, whereas number of injection per session is shown on the right side of each record

414 Table 1 Effect of both pretreatment with the selective CB1 receptor antagonist SR 141716A and shift in the unit drug dose available over the maintenance phase of cannabinoid IVSA. Data are expressed as mean number of CB1 receptor agonist infusions (12.5 and 25 µg/kg per injection) self-injected in a 3-h session (n=4 6) Baseline WIN 12.5 µg/kg per injection SR 3 mg/kg PRETREATMENT Shift to WIN 25 µg/kg per injection 22.192±1.16 34.7±1.87* 10.03±1.05* Baseline WIN 25 µg/kg per injection SR 3 mg/kg PRETREATMENT Shift to WIN 12.5 µg/kg per injection 11.23±0.75 17.5±1.11* 20±1.33* *P<0.05 significant difference from basal values (ANOVA followed by Newman-Keuls test) Fig. 5 Effect of substitution of WIN 55,212-2 by saline solution in rats (n=6) stable self-administrating the CB1 receptor agonist at the unit drug dose of 12.5 µg/kg per injection (extinction phase). **P<0.01 and *P<0.05 significant difference from basal values (ANOVA followed by Newman-Keuls test) lowering and increase of the unit dose of drug available by increasing and decreasing nose-poking activity, respectively. Specifically, a shift to higher unit dose induced a significant reduction (55%) in injection rate [F(3,87)=13, *P<0.05]. On the contrary, a shift to lower unit dose led to an augmented (78%) injection rate [F(5,22)=3.99, *P<0.05]. Similarly, pretreatment with the specific CB1 receptor antagonist SR 141716A administered 30 min prior to IVSA session resulted in a significant (56%) increase of the responding ratio [F(3,16)=5.29, *P<0.05], with no difference between the two doses tested (1.5 and 3 mg/kg IP). Table 1 illustrates the compensatory behaviour of animals with stable intake of WIN 55,212-2 in response to both a shifted unit drug dose available and CB1 receptor antagonist pretreatment. Figure 5 shows the daily pattern course of the extinction phase in rats stable self-administering WIN 55,212-2 at the dose of 12.5 µg/kg per injection. The highest response rates occurred during the first 1 h of extinction, reaching a maximum total of 87 nose-pokes by the end of the session. The number of responses to extinction did not differ significantly among any dose of WIN 55,212-2, but extinction latency was slightly greater for the dose of 50 µg/kg per injection. Furthermore, no significant difference in inducing extinction of the IVSA behaviour by substituting drug solution with saline (or vehicle) solution or by disconnecting the syringe pump was observed. In both cases, response rates greatly increased over the first 25 30 min and declined to lower levels at the end of the session. Extinction of self-administration behaviour Extinction phase started after 8 10 days of stable selfadministration behaviour and was characterised by an initial burst of responding followed by response cessation. Indeed, after a markedly significant increase in nose-poking activity [F(44,15)=10.93, *P<0.05 and **P<0.01] for four to six consecutive sessions, responding behaviour gradually declined to very low rates and rapidly extinguished, reducing to zero within 7 10 days depending on individual variability. Specifically, the mean number of injections self-administered by rats was about 3.7-, 3.3- and 2.8-fold enhanced when saline substituted the CB1 receptor agonist doses of 12.5, 25 and 50 µg/kg per injection, respectively. Discussion The present study was designed to assess a chronic model of intravenous cannabinoid self-administration in rats. Interest in drug IVSA has been strongly stimulated by its relationship to drug abuse in humans; nevertheless, Cannabis sativa is to date the only abused drug for which self-administration by laboratory animals has proved difficult to obtain. The first studies on cannabinoid self-administration were performed in 1974 by Harris and colleagues (Harris et al. 1974), who suggested that rats did not orally self-administer a suspension of Cannabis sativa extracts. Over recent decades, different groups have tried to replicate these experiments and extend investigations on cannabinoids rewarding properties using several

behavioural paradigms, obtaining controversial results (van Ree et al. 1978; Takahashi and Singer 1979; Mansbach et al. 1994). Only recently, it has been demonstrated that the CB1 receptor agonist WIN 55,212-2 serves as a positive reinforcer of IVSA behaviour in drug-naive mice (Martellotta et al. 1998). As already known, drug-seeking behaviour is controlled by a number of variables (Meisch 1987), mostly genetic (Meisch and George 1987) and/or environmental (Carroll and Meisch 1984), whereas others (i.e. animal strain, unit dose of drug available, route of administration, session length and schedule of reinforcement) are strictly methodological and therefore can be manipulated with greater ease. Since Long Evans rats, unlike Sprague-Dawley, Wistar and Lewis rats, proved to be sensitive to the cannabinoid rewarding properties in behavioural paradigm of abuse liability (Lepore et al. 1995; Parker and Gillies 1995; McGregor et al. 1996; Sañudo Peña et al. 1997; Mallet and Beninger 1998), we decided to use this rat strain in the present study. In pilot studies previously carried out in our laboratory, we observed that the size of the unit dose itself is of crucial importance in determining the cannabinoid IVSA. A rather low range of doses has been demonstrated to be optimal in maintaining cannabinoid IVSA in rats, whereas higher doses do not favour such behaviour. We found an upper limit in the amount of WIN 55,212-2 that subjects self-administer in a given time: in fact, rats did not acquire self-administration behaviour with higher doses of the CB1 agonist, even when it was made available in substitution of more classical drugs of abuse, such as cocaine or heroin (data not shown). In addition, although animals were allowed a longer session to obtain WIN 55,212-2 (i.e. 6 h), there was no significant increase in the number of infusions obtained. Similar results have been observed even though the timeout period was altered. As it is extremely difficult to demonstrate the reinforcing properties of cannabinoids in experimental animals, we employed the fixed ratio (FR1) schedule, where one nose-poke is required to obtain a single injection of drug. The progressive ratio (PR) schedule, in which each subsequent drug injection requires more nose-pokes than the previous one, would be a further important step in determining the maximal effort (the breaking point ) the animal will perform to receive the drug. Moreover, it has been demonstrated that manipulation and food-restriction facilitate the acquisition and maintenance of self-administration of most abused drugs. Indeed, drugs appear to be more effective as reinforcers when animals are food deprived (de Vaca and Carr 1998). In our study, chronic food restriction has been shown to be a necessary condition, since unrestricted rats did not acquire cannabinoid self-administration behaviour. Based both on our previous data and investigations showing THC intravenous self-administration at low doses in rats (van Ree et al. 1978; Takahashi and Singer 1979), we used low doses of WIN 55,212 ranging from 6.25 to 50 µg/kg per injection and found an inverted U-shaped relationship between dose and injection frequency. A general feature of IVSA of drugs of abuse is that responding rate and post-reinforcement intervals, which occur immediately after each infusion, are extremely sensitive to changes in the unit injection dose (Koob 1993). An increase in injection dose produces a compensatory reduction in injection rate with longer intervals and vice versa. As shown in Fig. 5, the rate of nose-poking was significantly modified by altering the unit dose of WIN 55,212-2: an increase in responding occurred when the unit dose of drug was reduced (and vice versa), presumably due to the animal s attempt to compensate for the altered drug dose. Finally, in animals stably self-administering cannabinoids, substituting for saline readily elicited a significant increase in nose-poking responding, thus demonstrating behavioural evidence for cannabinoid-seeking behaviour. The main criteria required to ensure that cannabinoid CB1 receptor agonist acts as a reinforcer in rats, and that their operant behaviour is therefore specifically directed toward obtaining cannabinoid, have therefore been satisfied in the present work. To summarise, this study showed that WIN 55,212-2 intravenous self-administration could be established in food-restricted Long Evans rats with the CB1 receptor agonist available on an FR1 schedule. It should be of great interest to further evaluate cannabinoids intravenous self-administration behaviour using higher FR schedules. In this model, nose-poking behaviour was specifically aimed at obtaining the CB1 receptor agonist (reinforcer), since changes in the unit dose and pretreatment with the CB1 antagonist SR 141716A led to a compensatory change in response and significantly reduced drug intake, respectively. Accordingly, such behaviour was extinguished either when WIN 55,212-2 was substituted with saline or the delivering syringe pump was disconnected. Establishment of this intravenous self-administration model in rats provides a useful tool for further investigation of brain pathways involved in rewarding properties of cannabinoids and their possible interactions with other endogenous systems. References 415 Bhattacharyya AK, Aulakh CS, Pradhan S, Ghosh P, Pradhan SN (1980) Behavioral and neurochemical effects of 9 -tetrahydrocannabinol in rats. Neuropharmacology 19:87 95 Carroll ME, Meisch RA (1984) Increased drug-reinforced behaviour due to food deprivation. In: Thompson T, Dews PB, Barrett JE (eds) Advances in behavioural pharmacology, vol 4. Academic Press, New York, pp 47 88 Carroll ME, France CP, Meisch RA (1979) Food deprivation increases oral and intravenous drug intake in rats. Science 205:319 321 Chait LD, Zacny JP (1992) Reinforcing and subjective effects of oral delta 9-THC and smoked marijuana in humans. Psychopharmacology 107:255 262 Chaperon F, Soubrié P, Puech AJ, Thiébot M-H (1998) Involvement of central cannabinoid (CB 1 ) receptors in the establish-

416 ment of place conditioning in rats. Psychopharmacology 135:324 332 Chen K, Kandel DB, Davies M (1997) Relationship between frequency and quantity of marijuana use and last year proxy dependence among adolescent and adults in the United States. Drug Alcohol Depend 46:53 67 Collins RJ, Weeks JR, Cooper MM, Good PI, Russell RR (1984) Prediction of abuse liability of drugs using IV self-administration by rats. Psychopharmacology 82:6 13 Corcoran ME, Amit Z (1974) Reluctance of rats to drink hashish suspension: free-choice and forced consumption, and the effects of hypothalamic stimulation. Psychopharmacologia 35: 129 147 de la Garza R, Johanson CE (1987) The effect of food deprivation on the self-administration of psychoactive drugs. Drug Alcohol Depend 19:17 27 de Vaca SC, Carr KD (1998) Food restriction enhanced the central rewarding effect of abused drug. J Neurosci 18:7502 7510 de Wit H, Stewart J (1983) Drug reinstatement of heroin-reinforced responding in the rat. Psychopharmacology 79:29 31 Fratta W, Martellotta MC, Cossu G, Fattore L (1998) Self-administration of the cannabinoid agonist WIN 55,212-2 and morphine in mice: evidence for a common neurobiological mechanism. Eur Neuropsychopharmacol 8:541 542 Gardner EL, Paredes W, Smith D, Donner A, Milling C, Cohen D, Morrison D (1988) Facilitation of brain stimulation reward by 9 -tetrahydrocannabinol. Psychopharmacology 96:142 144 Harris RT, Waters W, McLondon D (1974) Evaluation of reinforcing capability of delta-9-tetrahydrocannabinol in rhesus monkeys. Psychopharmacologia 37:23 29 Koob GF (1993) In: Korenman SG, Barchas JD (eds) Biological basis of substance abuse. Oxford University Press, New York, pp 339 354 Koob GF, Goeders NE (1989) In: Liebman L, Cooper H (eds) The neuropharmacological basis of reward. Oxford University Press, New York Kucharski LT, Williams JEG, Kornetsky C (1983) The effect of levonantradol on rewarding brain stimulation thresholds in the rat. Pharmacol Biochem Behav 19:149 151 Leite JR, Carlini EA (1974) Failure to obtain cannabis-directed behavior and abstinence syndrome in rats chronically treated with cannabis sativa extracts. Psychopharmacologia 36:133 145 Lepore M, Vorel SR, Lowinson J, Gardner EL (1995) Conditioned place preference induced by 9 -tetrahydrocannabinol: comparison with cocaine, morphine, and food reward. Life Sci 56:2073 2080 Lepore M, Liu X, Savage V, Matalon D, Gardner EL (1996) Genetic differences in by 9 -tetrahydrocannabinol-induced facilitation of brain stimulation reward as measured by a rate-frequency curve-shift electrical brain stimulation paradigm in three different rat strains. Life Sci 58:PL 365 372 Mallet PE, Beninger RJ (1998) 9 -Tetrahydrocannabinol, but not the endogenous cannabinoid receptor ligand anandamide, produces conditioned place avoidance. Life Sci 62:2431 2439 Mansbach RS, Nicholson KL, Martin BR, Balster RL (1994) Failure of 9 -tetrahydrocannabinol and CP 55,940 to maintain intravenous self-administration under a fixed-interval schedule in rhesus monkeys. Behav Pharmacology 5:219 225 Martellotta MC, Cossu G, Fattore L, Gessa GL, Fratta W (1998) Self-administration of the cannabinoid receptor agonist WIN 55,212-2 in drug-naive mice. Neuroscience 85:327 330 McGregor IS, Issakidis CN, Prior G (1996) Aversive effects of the synthetic cannabinoid CP 55,940 in rats. Pharmacol Biochem Behav 53:657 664 Meisch RA (1987) Factors controlling drug reinforced behaviour. Pharmacol Biochem Behav 27:367 371 Meisch RA, George FR (1987) Influence of genetic factors on drug reinforced behaviour in animals. In: Pickens R, Svikis D (eds) Biological vulnerability to drug abuse. NIDR Research Monograph, National Institute on Drug Abuse, Rockville, Md., USA Parker LA, Gillies T (1995) THC-induced place and taste aversion in Lewis and Sprague-Dawley rats. Behav Neurosci 109: 71 78 Rouse BA (1996) Epidemiology of illicit and abused drugs in the general population, emergency department drug-related episodes, and arrestees. Clin Chem 42:1330 1336 Sañudo-Peña MC, Tsou K, Delay ER, Hohman AG, Force M, Walker JM (1997) Endogenous cannabinoids as an aversive or counter-rewarding system in the rat. Neurosci Lett 223:125 128 Schenk S, Worley CM, McNamara C, Valadez E (1996) Acute and repeated exposure to caffeine: effects on reinstatement of extinguished cocaine-taking behaviour in rats. Psychopharmacology 126:17 23 Stark P, Dews PB (1980) Cannabinoids. I: Behavioural effects. J Pharmacol Exp Ther 214:124 130 Switzman L, Fishman B, Amit Z (1981) Pre-exposure effects of morphine, diazepam and 9 -THC on the formation of conditioned taste aversions. Psychopharmacology 74:149 157 Takahashi RN, Singer G (1979) Self-administration of 9 -tetrahydrocannabinol by rats. Pharmacol Biochem Behav 11:737 740 Valjent E, Maldonado A (2000) Behavioural model to reveal place preference to delta9-tetrahydrocannabinol in mice. Psychopharmacology 147:436 438 van Ree JM, Slangen JL, de Wied D (1978) Intravenous self-administration of drugs in rats. J Pharmacol Exp Ther 204:547 557 Weeks JR (1972) Long-term intravenous infusion. In: Myers RD (ed) Methods in psychobiology. Academic Press, London, pp 155 168