Conditioned Place Preference 85 E. Sensory Modalities Visual Cues — Vision cues are a poor choice for albino rodents, but pigeons are an excellent subject for visual cues. Rats prefer a dark environment vs. a light environment. The use of different shapes and colors are good variables for pigeons or primates, while simple light and dark choices are appropriate for rodents. The preferred chamber design for rodents is a patterned design, not a solid black or white configuration which increases the strong preference for the dark environment. Olfaction — Olfaction is one of the best exteroceptive cues (no odor, female or male scent, and food) for rodents. It is worth noting that failure to clean the chambers after each test session, during habituation and during training, has the potential to confound results. Auditory Cues — Developmental rodent studies utilizing powerful exterocep- tive cues often use recordings of ultrasonic chatter between mother and pup. Tones and clicking sounds provide means of creating distinct auditory environments. Tactile Cues — One should allow for a wide range and degree of specificity between chambers by using different flooring materials (solid floor [smooth, tex- tured], spaced bars, wire mesh, wood chips). IV. Experimental Designs A. Number and Duration of Conditioning Trials Drugs with powerful reinforcing properties (amphetamine, cocaine, and morphine) require fewer and shorter conditioning trials then a weaker reinforcing drug (nico- tine). Drugs with a very short half-life may allow a researcher to run a drug session in the morning and a vehicle session in the afternoon. The number of drug and vehicle training sessions should be equal. Number of testing sessions in CPP or CPA is one. The test session is in a non-drugged state. Comment — A criticism of CPP or CPA is testing in a non-drug state when conditioning in a drugged state creates the potential of confounding by state depen- dent learning. State dependent learning takes place when an animal is conditioned to the interaction between exteroceptive cues and interoceptive cues, but then tested with only exteroceptive cues. This dilemma potentially arises for weak reinforcing drugs. Adding a test session in the drugged state will determine if state dependent learning was a factor in the conditioning. Additional test sessions for evaluation of agonist and antagonist are helpful in developing a complete behavioral and pharma- cological profile of the test drugs. B. Biased and Unbiased Designs In addition to the above considerations, two conditioning techniques can be used in CPP and CPA design, biased and unbiased training. In biased conditioning the 0704/C05/frame Page 85 Monday, July 17, 2000 5:06 PM © 2001 by CRC Press LLC 86 Methods of Behavior Analysis in Neuroscience subject’s baseline preference is first determined for the chosen apparatus before experimental manipulation. In unbiased conditioning the subject is trained to asso- ciate the experimental manipulation with a particular environment. Both the biased and unbiased conditioning techniques in a two-chamber apparatus require a forced choice decision by the animal. Incorporating a three-chamber, four-chamber, or open field apparatus eliminates the potential confound of forced choice. The subject of choice for years was the rat, but labs frequently now use the mouse. The Syrian hamster and primates have also been used in CPP and CPA. 11,12 Thus, there are a variety of approaches that can be used to investigate CPP and the reader is referred to reviews by Schechter 13,14 for additional views of the field. V. Training and Testing A. Training Several days of free access to all environments allow the animal to habituate to the apparatus, eliminating novelty as a confounding variable. Baseline data should be determined as an average of time spent in each chamber over 3 to 5 days. The length of time necessary to determine baselines depends on the environmental differences between apparatus chambers. Distinct environments require fewer baseline sessions while less distinct or ambiguous environments require more baseline sessions. Durazzo et al. 1 found 20-min habituation sessions just as statistically significant as Single Alternation Drug Schedule Drug days Vehicle days WEEK 1 Monday Tuesday Wednesday Thursday Friday WEEK 2 Tuesday Monday Thursday Wednesday Friday Double Alternation Drug Schedule Drug days Vehicle days WEEK 1 Monday Wednesday Tuesday Thursday Friday WEEK 2 Monday Tuesday Thursday Wednesday Friday 0704/C05/frame Page 86 Monday, July 17, 2000 5:06 PM © 2001 by CRC Press LLC Conditioned Place Preference 87 40-min habituation sessions. Test sessions are generally 30 to 40 min in length. In CPP a drug or environmental manipulation thought to have reinforcing properties is paired with the less preferred environment and vehicle, or no environmental manipulation is paired with the preferred environment. B. Testing The testing of subjects is performed in a non-manipulated state. The percentage of time spent in each chamber is tabulated during the test session. If the animal spends more time in the less preferred (baseline data) chamber associated with the experi- mental manipulation, the researcher can infer the experimental manipulation had a rewarding effect on the affective state of the animal. In CPA a drug or environmental manipulation thought to have aversive properties is paired with the preferred envi- ronment and vehicle or no environmental manipulation is paired with the less preferred environment. If the animal spends more time in the preferred (baseline data) chamber associated with vehicle, or no environmental manipulation, the researcher can infer the experimental manipulation had a punishing effect on the affective state of the animal. Designs that include a novel chamber do not allow free access to the novel chamber during the habituation period. C. Concurrent Measurements A variety of other measures can be used to obtain information about the drug being studied. Spontaneous activity and location of subject within a chamber can be determined via infrared beams and computer technology. 15 In addition, video taping can also be utilized to determine other behavioral characteristics of the subjects under drug and non-drug states. VI. Research Objectives: Pharmacological and Neuroscience Objectives A. Mechanisms of Drug Action The CPP design is a simple procedure that has the potential for learning more about the positive reinforcing properties of a specific drug. This is especially important in research where the goal is to determine mechanisms of drug action using neuro- science approaches. It is extremely difficult using brain site techniques (lesions, microdialysis, and other neurophysiolgical procedures) in animals self-administering a drug or other operant procedures. The CPP procedure in many cases is a useful and manageable tool in evaluating drug action at different brain sites. 0704/C05/frame Page 87 Monday, July 17, 2000 5:06 PM © 2001 by CRC Press LLC 88 Methods of Behavior Analysis in Neuroscience B. Drug Design and Discovery The CPP design has the potential to evaluate new compounds suspected of having the same effects as a given agent. Thus, animal subjects trained with morphine should generalize to meperdine or methadone. C. Antagonist Studies A major issue in substance abuse research is the ability to detect new drugs that may be useful as antagonists to drugs such as cocaine that may help cocaine addicts in their rehabilitation. Thus, rats trained to cocaine can be challenged with other agents to determine the relative ability to attenuate cocaine’s reward stimulus. In addition, antagonism studies can provide additional information as to how a specific drug is acting (receptor classification). VII. Conclusion CPP and CPA are behavioral paradigms that are best used with experimental manip- ulations that produce a distinct discriminative stimulus. As mentioned previously, a drug may have both rewarding and punishing potential. Are you evaluating the rewarding or punishing effect of the drug? Preliminary studies are necessary when a literature review fails to provide pharmacokinetic information about the test article. When deciding upon the “biased” or “unbiased” conditioning method, a researcher might consider combining the two methods, first using the “biased” method where the preferred environment is determined and then using the “unbiased” method where the subjects are split between both environments. References 1. Durazzo, T.C., Gauvin, D.V., Goulden, K.L., Briscoe, R.J., and Holloway, F.A., Cocaine- induced conditioned place approach in rats: The role of dose and route of administration, Pharmacol. Biochem. Behav. 49, 1001, 1994. 2. Hasenohrl, R.U., Oitzl, M.S., and Huston, J.P., Conditioned place preference in the corral: A procedure for measuring reinforcing properties of drugs, J. Neurosci. Methods , 30, 141, 1989. 3. Parker, L.A., Place conditioning in a three- or four-choice apparatus: Role of stimulus novelty in drug-induced place conditioning, Behav. Neurosci. , 106, 294, 1992. 4. Klebaur, J.E. and Bardo, M.T., The effects of anxiolytic drugs on novelty-induced place preference, Behav. Brain Res. , 101, 51, 1999. 0704/C05/frame Page 88 Monday, July 17, 2000 5:06 PM © 2001 by CRC Press LLC Conditioned Place Preference 89 5. Kosten, T.A., Miserendino, M.J., Chi, S., and Nestler, E.J., Fischer and Lewis rat strains show differential cocaine effects in conditioned place preference and behavioral sen- sitization but not in locomotor activity or conditioned taste aversion, J. Pharmacol. Exp. Ther. , 269, 137, 1994. 6. Tzschentke, T.M., Measuring reward with the conditioned place preference paradigm: A comprehensive review of drug effects, recent progress and new issues, Prog. Neu- robiol. , 56, 613, 1998. 7. Fudala, P.J., Teoh, K.W., and Iwamoto, E.T., Pharmacologic characterization of nico- tine-induced conditioned place preference, Pharmacol. Biochem. Behav. , 22, 237, 1985. 8. Fudala, P.J. and Iwamoto, E.T., Further studies on nicotine-induced conditioned place preference in the rat, Pharmacol. Biochem. Behav. , 25, 1041, 1986. 9. Clarke, P.B. and Fibiger, H.C., Apparent absence of nicotine-induced conditioned place preference in rats, Psychopharmacol. , 92, 84, 1987. 10. Jorenby, D.E., Steinpreis, R.E., Sherman, J.E., and Baker, T.B., Aversion instead of preference learning indicated by nicotine place conditioning in rats, Psychopharmacol. , 101, 533, 1990. 11. Meisel, R.L., Joppa, M.A., and Rowe, R.K., Dopamine receptor antagonists attenuate conditioned place preference following sexual behavior in female Syrian hamsters, Eur. J. Pharmacol. , 309, 21, 1996. 12. Pomerantz, S.M., Wertz, I., Hepner, B., Wasio, L., and Piazza, I., Cocaine-induced conditioned place preference in rhesus monkeys, Soc. Neurosci. , Abstr., 18, 1572, 1992. 13. Schechter, M.D. and Calcagnetti, D.J., Trends in place preference conditioning with a cross-indexed bibliography; 1957-1991, Neurosci. Biobehav. Rev. , 17, 21, 1993. 14. Schechter, M.D. and Calcagnetti, D.J., Continued trends in the conditioned place preference literature from 1992 to 1996, inclusive, with a cross-indexed bibliography, Neurosci. Biobehav. Rev. , 22, 827, 1998. 15. Brockwell, N.T., Ferguson, D.S., and Beninger, R.J., A computerized system for the simultaneous monitoring of place conditioning and locomotor activity in rats, J. Neu- rosci. Meth. , 64, 227, 1996. 0704/C05/frame Page 89 Monday, July 17, 2000 5:06 PM © 2001 by CRC Press LLC - © 2001 by CRC Press LLC 6 Chapter Intravenous Drug Self-Administration in Nonhuman Primates Leonard L. Howell and Kristin M. Wilcox Contents I. Introduction II. Surgical Procedures III. Schedules of Drug Delivery A. Initial Training B. Fixed-Ratio Schedules C. Fixed-Interval Schedules D. Second-Order Schedules E. Progressive-Ratio Schedules IV. Research Application and Data Interpretation V. Discussion References I. Introduction The abuse of psychoactive drugs such as cocaine and heroin has spanned several decades and continues to be widespread in the U.S. Currently, research efforts have focused on the development of therapeutics to treat drug abuse. Drug self-adminis- tration studies have done much to help us understand the behavioral and pharmaco- logical mechanisms underlying drug abuse. An understanding of these mechanisms will in turn aid in the development of effective therapeutic agents. 0704/C06/frame Page 91 Monday, July 17, 2000 5:08 PM 92 Methods of Behavior Analysis in Neuroscience Important to the study of drug effects on behavior is the understanding that drugs function as stimuli to control behavior. 1 Based on the principles of operant conditioning, presentation of a stimulus as a consequence of behavior may either increase or decrease the probability that a behavior will occur again. 2 If a stimulus increases the probability that a behavior will recur, then that stimulus is defined as a positive reinforcer. 2 Stimuli such as food and water function as positive reinforcers, and data from self-administration studies indicate that drugs also function as positive reinforcers. Drug self-administration procedures in animals have been used exten- sively to evaluate the reinforcing effects of drugs. The first studies examining the reinforcing effects of drugs in the 1950s and 1960s focused on morphine self- administration in morphine-dependent animals. 3,4 Later studies demonstrated that dependence was not necessary to initiate self-administration. 5,6 Since these early studies, self-administration procedures have been used as a model of drug taking that can be studied under controlled laboratory conditions and applied to human drug use. Drug self-administration studies in animals have contributed substantially to our knowledge of the neuropharmacological mechanisms controlling drug abuse. For example, studies with opioids have shown that drugs with high affinity for µ opioid receptors function as positive reinforcers, 7,8 whereas opioids with high affinity for κ opioid receptors generally do not. 8,9 Additionally, the self-administration of psychomotor stimulants and opioids has been found to be affected by the adminis- tration of antagonists either systemically or centrally (cf, 10,11 ). If administration of an antagonist shifts the dose-response function for a self-administered drug to the right, then it can be assumed that the site of action of the antagonist is important for the reinforcing effects of the drug. 12-15 In addition to the administration of antagonists, self-administration of drugs has been affected by lesions of certain brain neurotransmitter systems (cf, 10,11 ). Studies have also found that animals will self- administer drugs directly into certain brain areas, suggesting a neuropharmacological mechanism for their reinforcing effects (cf, 10 ). Over the years, drug self-administration procedures in animals have been found to be valid and reliable for determining the abuse liability of drugs in humans. It is well established that animals will self-administer most drugs that are abused by humans. 16,17 In particular, studies in nonhuman primates have made a significant contribution to the field of drug abuse research. Nonhuman primates are ideal subjects since they are phylogenetically more closely related to humans than are other species. Thus, we can apply information obtained from nonhuman primates to problems of human drug abuse with greater accuracy. This chapter will discuss methods of self-administration in nonhuman primates, including different prepara- tions and schedules of drug reinforcement. While the focus will be primarily on self-administration of psychoactive stimulants such as cocaine, the methodology and general principles apply to other pharmacological classes including opiates, benzo- diazepines, and alcohol. 0704/C06/frame Page 92 Monday, July 17, 2000 5:08 PM © 2001 by CRC Press LLC 96 Methods of Behavior Analysis in Neuroscience FIGURE 6.1 A cumulative record of lever pressing maintained in three squirrel monkeys under a second-order FI 600- second schedule of cocaine (0.1 mg/injection) intravenous self-administration with FR 20 components. A red light illuminated the test chamber during the 600-second FI, and every 20th response during the FI changed the red light to white for 2 seconds. The first FR completed after the 600-second FI elapsed produced a drug injection and changed the red light to white for 15 seconds. Abscissa: time. Ordinate: cumulative number of responses. Pen deflection on the time axis indicates the end of the interval and the beginning of a timeout. The response pen reset vertically upon completion of each FI or when the pen reached the top of the paper. 500 Responses C C C C C C C C C C C C C C C S-104 15 Minutes S-106 S-102 COCAINE (0.1 mg/infusion) 0704/C06/frame Page 96 Monday, July 17, 2000 5:08 PM © 2001 by CRC Press LLC 98 Methods of Behavior Analysis in Neuroscience response rate increases with drug dose. In contrast, the descending limb of the curve reflects a nonspecific disruption of operant behavior as excessive drug accumulates over the session, and response rate decreases with drug dose. In addition, an inverse relationship has been obtained between infusion duration and reinforcing effects. 35,36 The longer the infusion time required to deliver a constant volume of drug solution, the less effective the drug functions as a reinforcer. A study by Glowa et al. 37 illustrates the use of an FR schedule of drug self- administration to characterize the effectiveness of a dopamine reuptake inhibitor to alter the reinforcing effects of cocaine in rhesus monkeys. A standard tethered- catheter, home-cage system was used for intravenous drug delivery. 5 Animals were trained to self-administer cocaine under an FR 10 schedule of drug delivery during 90-minute daily sessions. Pretreatment with the high-affinity dopamine reuptake inhibitor, GBR 12909, dose-dependently decreased rates of cocaine-maintained responding, and the effect was larger when lower doses of cocaine were used to maintain responding. Moreover, the rate-decreasing effects of GBR 12909 were greater on cocaine-maintained responding than on food-maintained responding under a multiple schedule of drug and food delivery. The results obtained were consistent with previous reports demonstrating that drugs with dopamine agonist effects can FIGURE 6.2 A cumulative record of lever pressing maintained in a rhesus monkey under a progressive-ratio schedule of cocaine (0.1 mg/kg/injection) self-administration over a daily session. The daily session consisted of five components, each made up of four trials at a particular response requirement. The response require- ment began at an FR of 120 and doubled in subsequent components (i.e., 120, 240, 480, 960, 1920). A trial ended with a drug injection or the expiration of a limited hold. The session ended if the limited hold expired two consecutive times. Abscissa: time. Ordinate: cumulative number of responses. The response pen reset vertically upon completion of the FR or when the pen reached the top of the paper. 500 Responses 90-31 15 Minutes 0704/C06/frame Page 98 Monday, July 17, 2000 5:08 PM © 2001 by CRC Press LLC Intravenous Drug Self-Administration in Nonhuman Primates 99 decrease cocaine-maintained responding. 38,39 This type of drug interaction has been attributed to a satiation of cocaine-maintained responding by pretreatment with a drug having dopaminergic effects. The latter approach to cocaine medication devel- opment has been referred to as substitute agonist pharmacotherapy. 40 Hence, response-independent delivery of a dopamine reuptake inhibitor may have decreased cocaine self-administration by substituting for the reinforcing effects of response- produced cocaine. This interpretation is supported by studies showing that GBR 12909 will substitute for cocaine as a reinforcer in squirrel monkeys. 41-43 Note that Glowa et al. 37 incorporated several important design features in their self-administration study. First, multiple unit doses of cocaine were self-administered on separate occasions in order to establish a complete dose-effect curve for cocaine. Hence, pretreatment effects of GBR 12909 could be assessed over a broad range of cocaine doses. Second, multiple pretreatment doses of GBR 12909 were adminis- tered in order to establish dose-dependency of pretreatment effects and to identify the optimal pretreatment dose that lacked overt behavioral toxicity. Lastly, the specificity of pretreatment effects on cocaine-maintained behavior was assessed by comparing drug effects on food-maintained behavior. The multiple schedule which alternated cocaine and food as maintaining events during separate components was well-suited for this application. Moreover, cocaine dose was manipulated to match response rate to that obtained during the food component of the multiple schedule. The finding that GBR 12909 suppressed cocaine-maintained responding at doses that had little or no effect on food-maintained responding under identical schedules and comparable response rates provides convincing evidence that GBR 12909 selec- tively attenuated the reinforcing effects of cocaine. A study by Woolverton 44 provides another example of cocaine self-administration under an FR schedule in rhesus monkeys. The objective was to characterize the effectiveness of dopamine antagonists to alter the reinforcing effects of cocaine. A standard tethered-catheter, home-cage system was used for intravenous drug delivery. Animals were trained to self-administer cocaine under an FR 10 schedule of drug delivery during 2-hour daily sessions. When responding was stable, the animals were pretreated with the D 1 antagonist, SCH 23390, or the D 2 antagonist, pimozide. Inter- mediate doses of pimozide generally increased cocaine self-administration, whereas SCH 23390 either had no effect or decreased cocaine self-administration. High doses of both antagonists decreased the rate of cocaine self-administration, but also produced pronounced catalepsy. Hence, the latter effects could not be attributed to a selective interaction with the reinforcing effects of cocaine. The author concluded that the selective increase in responding maintained by cocaine following pimozide pretreat- ment suggested a role for the D 2 -receptor in cocaine self-administration. Strengths of the Woolverton 44 study design included multiple unit doses of cocaine and multiple pretreatment doses of both dopamine antagonists. Extinction of cocaine self-administration when saline was substituted for cocaine was also characterized. Note that response rate for cocaine increased following pretreatment with the D 2 -selective antagonist. The latter effect is interpreted as a behavioral compensation to overcome the attenuation of the reinforcing effects of cocaine by pimozide. Since drug intake is a direct function of response rate under FR schedules, an increase in rate will result in greater session intake of cocaine which may 0704/C06/frame Page 99 Monday, July 17, 2000 5:08 PM © 2001 by CRC Press LLC © 2001 by CRC Press LLC [...]... standard tethered-catheter, home-cage system was used for intravenous drug delivery Animals were trained to self-administer cocaine under an FI 5-minute schedule during 4- hour daily sessions Under this schedule, the first response after 5 minutes produced a 10-second cocaine injection Pretreatment with the high-affinity dopamine reuptake inhibitor, 2β-propanoyl-3 -( 4- tolyl)-tropane (PTT), dose-dependently... dopamine reuptake inhibitors on food- and cocainemaintained responding I: Dependence on unit dose of cocaine, Exp Clin Psychopharmacol., 3, 219, 1995 38 Caine, S B and Koob, G F., Modulation of cocaine self-administration in the rat through D3 dopamine receptors, Science, 260, 18 14, 1993 39 Skjoldager, P., Winger, G., and Woods, J H., Effects of GBR 12909 and cocaine on cocaine-maintained behavior in. .. decreased rates of cocaine-maintained responding and total session intake of cocaine The reinforcing effects of PTT were also evaluated in a separate group of animals When substituted for cocaine, PTT maintained response rates that were similar to those maintained by saline and significantly lower than rates maintained by cocaine The results demonstrated that a long-acting dopamine reuptake inhibitor could... cocaine self-administration in nonhuman primates Also, failure of PTT to maintain rates of self-administration greater than those obtained during extinction conditions suggested that PTT may have limited abuse liability The study of Nader et al .45 was consistent with previous findings using dopamine reuptake inhibitors to decrease cocaine self-administration under fixed-ratio schedules of drug self-administration.37... Self-Administration in Nonhuman Primates 109 44 Woolverton, W L., Effects of a D1 and D2 dopamine antagonist in the self-administration of cocaine and piribedil by rhesus monkeys, Pharmacol Biochem Behav., 24, 351, 1986 45 Nader, M A., Grant, K A., Davies, H M L., Mach R H., and Childers, S R., The reinforcing and discriminative stimulus effects of the novel cocaine analog 2β-propanoyl-3 -( 4- tolyl)-tropane... selective interactions with the reinforcing properties of the self-administered drug In contrast, a general sedative effect will likely suppress drug- and food-maintained responding to a comparable extent Lastly, the reinforcing properties and abuse potential of the medication should be evaluated by substituting a range of doses of the medication for the self-administered drug Since reinforcing effects in. .. However, alternative interpretations were acknowledged, largely because specificity of pretreatment effects on cocaine-maintained behavior was not assessed by comparing drug effects on behavior maintained by nondrug reinforcers Nader et al .45 used an FI schedule of drug self-administration to characterize the effectiveness of a novel cocaine analog to alter the reinforcing effects of cocaine in rhesus monkeys... pretreated with saline Asterisks indicate a significant (p < 0.05) effect of RTI-113 pretreatment self-administration, and the effect was not surmounted by increasing the unit dose of cocaine (Figure 6 .4) The latter findings are consistent with previous studies using dopamine reuptake inhibitors to decrease cocaine self-administration under FR37 and FI45 schedules of drug self-administration Hence, the... self-administered drugs Drugs can be rank-ordered based on their relative reinforcing efficacy as determined by breakpoint in the progressive ratio.30,33 For example, cocaine has been found to maintain higher breakpoint values than diethylpropion, chlorphentermine, and fenfluramine in baboons.30 More recently, cocaine has been found to maintain higher breakpoint values than procaine in rhesus monkeys (Figure 6.5).33... Progressive-ratio performance maintained by drug infusions: Comparison of cocaine, diethylproprion, chlorphentermine, and fenfluramine, Psychopharmacology, 56, 5, 1978 31 Griffiths, R R., Bradford, L D., and Brady, J V., Progressive-ratio and fixed-ratio schedules of cocaine-maintained responding in baboons, Psychopharmacology, 65, 125, 1979 32 Rowlett, J K and Woolverton, W L., Self-administration of cocaine . high-affinity dopamine reuptake inhibitor, 2 β -propanoyl-3 β -( 4- tolyl)-tropane (PTT), dose-dependently decreased rates of cocaine-maintained responding and total session intake of cocaine the selective increase in responding maintained by cocaine following pimozide pretreat- ment suggested a role for the D 2 -receptor in cocaine self-administration. Strengths of the Woolverton 44 . doses of cocaine were used to maintain responding. Moreover, the rate-decreasing effects of GBR 12909 were greater on cocaine-maintained responding than on food-maintained responding under a multiple