A randomization test-based method for risk assessment in neurotoxicology.

A current trend in risk assessment for systemic toxicity (noncancer) endpoints is to utilize the observable range of the dose-effect curve in order to estimate the likelihood of obtaining effects at lower concentrations. Methods to accomplish this endeavor are typically based on variability in either the effects of fixed doses (benchmark approaches), or on variability in the doses producing a fixed effect (probabilistic or tolerance-distribution approaches). The latter method may be particularly desirable because it can be used to determine variability in the effect of an agent in a population, which is an important goal of risk assessment. This method of analysis, however, has typically been accomplished using dose-effect data from individual subjects, which can be impractical in toxicology. A new method is therefore presented that can use traditional groups-design data to generate a set of dose-effect functions. Population tolerances for a specific effect can then be estimated from these model dose-effect functions. It is based on the randomization test, which assesses the generality of a data set by comparing it to a data set constructed from randomized combinations of single point estimates. The present article describes an iterative line-fitting program that generates such a data set and then uses it to provide risk assessments for two pesticides, triadimefon and carbaryl. The effects of these pesticides were studied on the locomotor activity of laboratory rats, a common neurobehavioral end point. Triadimefon produced dose-dependent increases in activity, while carbaryl produced dose-dependent decreases in activity. Risk figures derived from the empirical distribution of individual dose-effect functions were compared to those from the iterative line-fitting program. The results indicate that the method generates comparable risk figures, although potential limitations are also described.

Repeated exposure to inhaled toluene induces behavioral and neurochemical cross-sensitization to cocaine in rats.

RATIONALE: Toluene is a solvent found in many commercial products and is frequently abused by inhalation. Whether previous exposure to toluene alters subsequent responses to other drugs of abuse is not known. OBJECTIVES: This study determined the effects of repeated toluene exposure on the acute motor-stimulant response to cocaine and on cocaine-induced dopamine (DA) concentrations in the nucleus accumbens (NAc). METHODS: One week following bilateral cannulae implantation over the NAc, 27 adult, male Wistar rats began a daily 30-min exposure regimen to either toluene (8,000 ppm) or air for ten sessions. Approximately 24 h or 96 h after their last exposure, animals were injected with either saline or cocaine (15 mg/kg, i.p.) and locomotor activity and DA concentrations in the NAc were measured. RESULTS: Exposure to toluene rendered the rats immobile, and the time required for recovery of normal posture decreased across the ten sessions. In all animals tested, systemic cocaine administration enhanced both locomotor activity and DA concentrations in the NAc. These increases, however, were significantly greater in rats previously exposed to toluene. CONCLUSIONS: Overall, these findings show that repeated toluene exposure enhances behavioral and neurochemical responses to subsequent cocaine administration.

Morphine-induced conditioned taste aversions in the LEW/N and F344/N rat strains.

Previous reports have shown that the LEW/N and F344/N inbred rat strains display a differential sensitivity to cocaine in a number of preparations, with the LEW/N rats displaying an increased sensitivity to both the reinforcing and aversive effects of cocaine (relative to the F344/N rats). Given that the LEW/N rats are also more sensitive to the reinforcing effects of morphine than the F344/N strain, the present experiment examined the ability of morphine to condition taste aversions in the LEW/N and F344/N strains to determine if the general sensitivity to cocaine generalizes to another drug of abuse. Specifically, on four conditioning trials, 35 LEW/N and 33 F344/N female rats were allowed access to a novel saccharin solution and then injected with varying doses of morphine (0, 10, 32 and 56 mg/kg). On intervening recovery days, subjects were allowed 20-min access to water. Following the fourth trial, a final aversion test was administered. The F344/N rats, but not the LEW/N rats, rapidly acquired morphine-induced taste aversions at all doses of morphine. Pharmacokinetic differences between the strains were also assessed. Specifically, 10 mg/kg morphine (or vehicle) was administered to subjects of both strains and plasma morphine levels were analyzed at 0.5, 2 and 4 h postinjection. No differences in plasma levels between the strains were observed. Unlike with cocaine, the LEW/N rats do not seem generally sensitive to morphine (relative to the F344/N rats). Rather, the differential sensitivity of the two strains to these compounds seems to be preparation dependent. Possible mechanisms underlying the differential sensitivity evident in the strains were discussed.

A comparison of cocaine, GBR 12909, and phentermine self-administration by rhesus monkeys on a progressive-ratio schedule.

The dopamine reuptake inhibitor GBR 12909 and the dopamine releaser phentermine may have potential for the treatment of cocaine abuse in humans. Pre-session treatment with either drug can decrease cocaine-maintained responding in rhesus monkeys while not affecting food-maintained responding. Both drugs are self-administered, but in some reports the patterns of responding they maintain differ from typical cocaine-reinforced responding. This study compared self-administration of cocaine (1--100 microg/kg/inj), GBR 12909 (3--100 microg/kg/inj), and phentermine (10--170 microg/kg/inj) in rhesus monkeys on a progressive-ratio schedule. Individual unit doses of each drug were available across several consecutive sessions. Cocaine self-administration was typical: the average number of ratios completed per session was a bitonic (increasing/decreasing) function of unit dose. Phentermine self-administration was variable across subjects (two of four monkeys self-administered reliably); one subject exhibited clear signs of behavioral toxicity. Self-administration of GBR 12909 was similarly variable across subjects. In the two subjects that self-administered GBR 12909 reliably, self-administration of small to mid-sized unit doses was enhanced following exposure to large unit doses. These data indicate that differences in self-administration of these drugs can be observed under progressive ratio procedures. Further, the data add to existing evidence suggesting that phentermine and GBR 12909 have at least moderate potential to be abused by humans.

Response requirements and unit dose modify the effects of GBR 12909 on cocaine-maintained behavior.

Previous studies found that GBR 12909 can decrease cocaine-maintained responding at doses that do not affect food-maintained responding. In this study, the effects of GBR 12909 (0.3-3.0 mg/kg) were further examined by varying the response requirement and unit dose of cocaine. Rhesus monkeys earned food or cocaine under a multiple fixed-ratio (FR) schedule. The FR for food was always 30, but the FR for cocaine was varied from 10-130 and the unit dose was varied from 5.6-56.0 microg/kg per injection. Doses of GBR 12909 were tested in an ascending order, for 5 consecutive sessions each. GBR 12909 selectively decreased cocaine-maintained responding in all monkeys in at least 1 condition. These effects were enhanced with large response requirements and/or small unit doses. The results demonstrate that environmental variables can influence the selectivity of GBR 12909's effects and contribute to a growing debate concerning the evaluation of potential pharmacotherapies for drug abuse.

Abuse liability of the anesthetic propofol: self-administration of propofol in rats under fixed-ratio schedules of drug delivery.

RATIONALE: Previous reports suggest that propofol (PPF) may have abuse potential in humans. Hence, we hypothesized that PPF could reinforce self-administration behavior in other species. Positive reinforcing effects of PPF could be interpreted as an index of abuse liability. OBJECTIVE: Acquisition and maintenance of i.v. PPF self-administration were examined in 12 rats. METHODS: Six rats were initially given access to methohexital (MHX, 2.0 mg/kg per infusion) under a fixed ratio (FR) 1 schedule, while the other six were initially given access to PPF (1.7 mg/kg per infusion). Once stable responding was established, various doses of PPF (0.56, 1.0, and 1.7 mg/kg per infusion) and vehicle (Intralipid 20%) were made available. RESULTS: The number of PPF infusions per session was an inverse function of dose, with 0.56 mg/kg and 1.0 mg/kg per infusion maintaining significantly more infusions per session than vehicle in most rats under the FR 1 schedule. For some rats, the number of vehicle infusions per session was equal to or greater than the number of PPF infusions. Increasing the response requirement to FR 5 decreased the number of vehicle infusions per session in these rats, while PPF maintained a higher number of infusions than vehicle under this FR value in six of seven rats. CONCLUSION: PPF served as a reinforcer in rats under FR schedules of i.v. drug delivery, adding to the extant evidence that it has abuse potential.

Regulation of benzodiazepine receptor binding and GABA(A) subunit mRNA expression by punishment and acute alprazolam administration.

Quantitative autoradiography of benzodiazepine (BZ) receptors and competitive reverse transcription-polymerase chain reaction were used to characterize changes in BZ binding and GABA(A) receptor subunit transcription levels associated with the anxiolytic effects of alprazolam. Effects were assessed on punished and non-suppressed water consumption using a lick suppression (Vogel) paradigm. Alprazolam had no effect on non-suppressed licking, [(3)H]Ro 15-1788 binding or receptor subunit transcript levels, compared to non-drug controls. When each fifth lick produced a shock (0-0.5 mA), responding was suppressed in an intensity-related manner. The highest intensity significantly decreased licking (85%), [(3)H]Ro 15-1788 binding (12%) and alpha1 transcript levels (63%) in the basolateral nucleus of the amygdala, and [(3)H]Ro 15-1788 binding in the mediodorsal thalamic nucleus (15%), compared to non-punished controls. Punishment increased the ratio of gamma2L/S transcripts in the basolateral nucleus of the amygdala. Alprazolam blocked or reversed each of these effects. These results show that punishment has similar effects on BZ binding and GABA(A) receptor subunit expression and that alprazolam can block or reverse those effects. Such changes may be related to the anxiolytic effects of alprazolam.

Oxygenated analogues of 1-[2-(Diphenylmethoxy)ethyl]- and 1-[2-[Bis(4-fluorophenyl)methoxy]ethyl]-4-(3-phenylpropyl)piperazines (GBR 12935 and GBR 12909) as potential extended-action cocaine-abuse therapeutic agents.

An investigation into the preparation of potential extended-release cocaine-abuse therapeutic agents afforded a series of compounds related to 1-[2-(diphenylmethoxy)ethyl]-4-(3-phenylpropyl)piperazine (1a) and 1-[2-[bis(4-fluorophenyl)methoxy]ethyl]-4-(3-phenylpropyl)piperazine (1b) (GBR 12935 and GBR 12909, respectively), which were designed, synthesized, and evaluated for their ability to bind to the dopamine transporter (DAT) and to inhibit the uptake of [(3)H]-labeled dopamine (DA). The addition of hydroxy and methoxy substituents to the benzene ring on the phenylpropyl moiety of 1a-1d resulted in a series of potent and selective ligands for the DAT (analogues 5-28). The hydroxyl groups were included to incorporate a medium-chain carboxylic acid ester into the molecules, to form oil-soluble prodrugs, amenable to "depot" injection techniques. The introduction of an oxygen-containing functionality to the propyl side chain provided ketones 29 and 30, which demonstrated greatly reduced affinity for the DAT and decreased potency in inhibiting the uptake of [(3)H]DA, and benzylic alcohols 31-36, which were highly potent and selective at binding to the DAT and inhibiting [(3)H]DA uptake. The enantiomers of 32 (34 and 36) were practically identical in biological testing. Compounds 1b, 32, 34, and 36 all demonstrated the ability to decrease cocaine-maintained responding in monkeys without affecting behaviors maintained by food, with 34 and 36 equipotent to each other and both more potent in behavioral tests than the parent compound 1b. Intramuscular injections of compound 41 (the decanoate ester of racemate 32) eliminated cocaine-maintained behavior for about a month following one single injection, without affecting food-maintained behavior. The identification of analogues 32, 34, and 36, thus, provides three potential candidates for esterification and formulation as extended-release cocaine-abuse therapeutic agents.

Preclinical research on cocaine self-administration: environmental determinants and their interaction with pharmacological treatment.

It has been asserted that any comprehensive understanding of cocaine abuse and its treatment will require attention to both behavioral and pharmacological variables. Although the preclinical literature evaluating the effects of pharmacological variables on cocaine self-administration has been extensively reviewed, no comprehensive review of the effects of environmental variables on cocaine self-administration has been published. The present review summarizes and critiques the preclinical findings on environmental determinants of cocaine self-administration. The influence of environmental variables on the effects of pharmacological interventions on cocaine self-administration are also described. Several environmental variables have been shown to affect cocaine self-administration, including unit dose, schedule of cocaine delivery, schedules of nondrug stimuli, behavioral history, conditioned stimuli, food deprivation, exposure to stress, and rearing environment. Among these variables, unit dose, schedule of cocaine delivery, availability of alternative nondrug reinforcers, food deprivation, and rearing environment have also been shown to alter pharmacological treatment effects on cocaine self-administration. Thus, drug effects on cocaine self-administration are malleable and dependent upon the environmental context within which they occur. Suggestions for future research on the effects of these and other environmental variables on cocaine self-administration and its pharmacological treatment are presented.

Alterations of GABAA receptor subunit mRNA levels associated with increases in punished responding induced by acute alprazolam administration: an in situ hybridization study.

Changes in the mRNA encoding alpha1, alpha2, beta2 and gamma2 subunits of the GABAA receptor associated with the anxiolytic effects of alprazolam were measured in 20 brain regions using in situ hybridization techniques. Compared to non-punished controls, punishment decreased alpha1 mRNA levels in two nuclei of the amygdala, the cerebral cortex, and the mediodorsal thalamic nucleus and decreased alpha2 mRNA levels in the hippocampus. Punishment increased beta2 mRNA levels in ventroposterior thalamic nucleus and gamma2 mRNA levels in the CA2 area of the hippocampus. All of these effects were reversed when alprazolam increased punished responding, while alprazolam alone had no effect on either non-punished responding or GABAA receptor subunit regulation in these brain regions. Some brain regions that were unaffected by punishment were altered by alprazolam plus punishment. These results demonstrate that punishment and alprazolam can produce reciprocal changes in the mRNA levels for some subunits of the GABAA receptor. These changes may alter GABAergic synaptic inhibition by altering the density of GABAA receptors or their efficacy to bind drugs. They suggest that the underlying mechanisms by which drugs affect behavior can depend upon the conditions under which behavior is assessed.

Effects of phentermine on responding maintained under multiple fixed-ratio schedules of food and cocaine presentation in the rhesus monkey.

Drugs that decrease drug-maintained responding at doses that do not decrease other behaviors in animals may be suitable candidates for development as medications to treat drug abuse in humans. The present study examined whether this effect could be obtained with phentermine, a drug that has been reported to decrease cocaine intake in humans. Rhesus monkeys were trained under multiple fixed-ratio 30-response schedules of food and i.v. cocaine delivery. Phentermine was always given as a slow, i.v. infusion. Acute treatment with phentermine (0.3-10 mg/kg) decreased cocaine-maintained responding at doses that did not decrease, or decreased less, food-maintained responding for each of three unit doses of cocaine (10-100 microg/kg/injection). Subacute treatment with phentermine (3 or 5.6 mg/kg, daily) also decreased cocaine-maintained responding more than food-maintained responding. After subacute treatment was terminated, rates of cocaine-maintained responding generally recovered to levels comparable to those seen during untreated control sessions. Phentermine (0.3-3 mg/kg) did not generally increase responding associated with a very low (1 microg/kg/injection) unit dose of cocaine, suggesting that the decrease in cocaine-maintained responding at higher unit doses was not the result of a leftward shift in the cocaine unit dose-effect function. Phentermine (0.1-3 mg/kg) decreased responding maintained by 1-[2-[bis(4-fluorophenyl) methoxy]ethyl]-4-[3-phenylpropyl] piperazine (GBR 12909) (30 microg/kg/injection) at doses similar to those that decreased food-maintained responding. These results show that phentermine is effective in decreasing cocaine self-administration and suggest that it may be an effective medication for cocaine abuse.

Effects of phentermine on responding maintained by progressive-ratio schedules of cocaine and food delivery in rhesus monkeys.

Previous reports indicate that intravenous pretreatment with phentermine can decrease cocaine-maintained responding without affecting food-reinforced responding under fixed-ratio schedules. The present experiments were designed to explore the generality of this effect using progressive-ratio schedules of reinforcement and different routes of phentermine administration. Unit doses of cocaine and food-pellet magnitudes were identified that maintained similar breaking points, and the effects of acute exposure to phentermine were assessed. In Experiment 1, a 'conventional' (one-trial) progressive-ratio schedule was used, in which response requirements increased after each reinforcer delivery; in Experiment 2, a 'modified' (five-trial) progressive-ratio schedule was used, in which response requirements increased after every five reinforcer deliveries. In one group of monkeys, responding was maintained by food; in another, cocaine infusions maintained responding. Phentermine (0.1-5.6mg/kg, intramuscularly (i.m.)) dose-dependently decreased breakpoints on both progressive-ratio schedules. There were no differences in phentermine's effects on cocaine- and food-maintained behavior. In Experiment 3, intravenous administration of phentermine had largely similar effects. Taken together with results from previous reports, these data suggest that the effects of phentermine pretreatment are influenced by the behavioral procedure used to maintain responding and/or by the efficacy of the food and cocaine reinforcers.

Progressive-ratio schedules of drug delivery in the analysis of drug self-administration: a review.

Drugs, like other reinforcers, can vary in their relative abilities to support operant responding. Considerable research has been designed to obtain useful measures of a given drug's or dose's "reinforcing efficacy" and to identify the ways in which a variety of behavioral and pharmacological variables impact these measures. Progressive-ratio schedules of drug delivery generate an index of a drug's or dose's reinforcing efficacy (the breaking point) and are being used increasingly as tools in the analysis of drug self-administration. Progressive-ratio schedules of drug delivery have been used to characterize the effects of pretreatment drugs, lesions, drug deprivation, physical dependence, and repeated non-contingent drug exposure on breaking points. Behavioral factors, including food restriction and electric shock, and organismic factors, including gender and strain, have also been investigated using progressive-ratio schedules of drug delivery. To the extent that breaking points provide an index of reinforcing efficacy, these studies demonstrate that a wide range of variables can influence the reinforcing efficacy of self-administered drugs. The objectives of this review are to critique existing research themes, outline potential limitations of progressive-ratio procedures, and to suggest potentially fruitful uses of these procedures in future research.

Studies of the biogenic amine transporters. VII. Characterization of a novel cocaine binding site identified with [125I]RTI-55 in membranes prepared from human, monkey and guinea pig caudate.

[125I]RTI-55 is a cocaine analog with high affinity for dopamine (DA) and serotonin (5-HT) transporters. Quantitative ligand binding studies revealed a novel high affinity [125I]RTI-55 binding site assayed under 5-HT transporter (SERT) conditions which has low affinity for almost all classic biogenic amine transporter ligands, including high affinity 5-HT transporter inhibitors such as paroxetine, but which retains high affinity for cocaine analogs. This site, termed SERT(site2) for its detection under 5-HT transporter conditions (not for an association with the SERT) occurs in monkey caudate, human caudate, and guinea pig caudate membranes, but not in rat caudate membranes. SERT(site2) is distinguished from the DA transporter (DAT) and SERT by several criteria, including a distinct ligand-selectivity profile, the inability to detect SERT(site2) in cells stably expressing the cloned human DAT, and insensitivity to irreversible ligands which inhibit [125I]RTI-55 binding to the DAT and SERT. Perhaps the most striking finding about SERT(site2) is that a wide range of representative antidepressant agents have very low affinity for SERT(site2). The affinity of cocaine for this site is not very different from the concentration cocaine achieves in the brain at pharmacological doses. Viewed collectively with the observation that ligands with high affinity for SERT(site2) are mostly cocaine analogs, these data lead us to speculate that actions of cocaine which differ from those of classic biogenic amine uptake inhibitors may be mediated in part via SERT(site2).

Behavioral reactivity to aversive stimuli in a transgenic mouse model of impaired glucocorticoid (type II) receptor function: effects of diazepam and FG-7142.

Transgenic mice with impaired type II-glucocorticoid receptor mediated feedback inhibition of hypthalamic-pituitary-adrenal activity were assessed in three different tests assessing behavioral reactivity to aversive stimuli, the elevated plus maze, the Thatcher-Britton novelty-conflict paradigm, and the startle paradigm. Transgenic mice more frequently entered and spent more time in the open arms of the elevated plus in comparison to B6C/3F1 mice. Transgenic mice took significantly longer to begin eating in the Thatcher-Britton novelty conflict paradigm, and displayed increased reactivity in the startle paradigm. Administration of 1 or 2 mg/kg diazepam reversed the behavioral effects observed in all three tests. Administration of the benzodiazepine receptor inverse agonist N-methyl-beta-carboline-3 carboxamide (FG-7142, 10 mg/kg) reduced the ratio of open to total arm entries and the time spent in the open arms of the plus maze in transgenic, but not B6C/3F1, mice. This dose of FG-7142 did not influence performance of either strain in the Thatcher-Britton or startle paradigms. These results are discussed in terms of the hypothesis that the transgenic mice are more sensitive to the aversive properties of novel stimuli, and that they may have difficulty discriminating between signals of relative safety and danger.

Effects of dopaminergic drugs on food- and cocaine-maintained responding. IV: Continuous cocaine infusions.

The effects of cocaine (continuously infused during the session with and without a are-loading dose) on responding maintained under multiple fixed-ratio 30 schedules of limited access to food and cocaine delivery (10 or 56 micrograms/kg per injection) were studied in nine rhesus monkeys. When responding was maintained by 10 micrograms/kg per injection, cocaine decreased rates of cocaine-maintained responding in a dose-related manner while having no effect on food-maintained responding. When responding was maintained by 56 micrograms/kg per injection, cocaine decreased both cocaine- and food-maintained responding. These results show that the effects of continuous infusions of cocaine depend upon both the unit dose of cocaine and the event that maintains responding. As such, the effects of continuous infusions of cocaine are similar to those of other, longer-acting dopamine agonists. Such results support the development of long-acting agonist approaches to the pharmacotherapeutic treatment of cocaine abuse.

Phentermine/fenfluramine decreases cocaine self-administration in rhesus monkeys.

Dopaminergic agonists can decrease cocaine self-administration at doses that do not decrease food-maintained responding, a pre-clinical effect indicative of a potential treatment for human cocaine abuse. To assess whether similar effects could be obtained with medications currently used to treat substance abuse, phentermine and fenfluramine were given alone and in combination to rhesus monkeys responding under schedules of food and cocaine delivery. Phentermine decreased cocaine-maintained responding with no effect on food-maintained responding. Fenfluramine also selectively decreased cocaine-maintained responding, but only at the highest dose. Combining a lower dose of fenfluramine with phentermine selectively decreased cocaine-maintained responding, but not more than with phentermine alone. These results suggest that phentermine, as well as its combination with fenfluramine, may be useful in the treatment of cocaine abuse.

Heteroaromatic analogs of 1-[2-(diphenylmethoxy)ethyl]- and 1-[2-[bis(4-fluorophenyl)methoxy]ethyl]-4-(3-phenylpropyl)piperazines (GBR 12935 and GBR 12909) as high-affinity dopamine reuptake inhibitors.

A new series of heteroaromatic GBR 12935 [1-[2-(diphenylmethoxy)ethyl]-4-(3-phenylpropyl)-piperazine] (I) and GBR 12909 [1-[2-[bis(4-fluorophenyl)methoxy]ethyl]-4-(3-phenylpropyl)piperazine] (2) analogs was synthesized and evaluated as dopamine transporter (DAT) ligands. Analogs 5-16, in which the benzene ring in the phenylpropyl side chain of the GBR molecule had been replaced with a thiophene, furan, or pyridine ring, exhibited high affinity and selectivity for the DAT vs serotonin transporter (SERT) and stimulated locomotor activity in rats in a manner similar to the parent compound 2. In cocaine and food self-administration studies in rhesus monkeys, both thiophene-containing (6 and 8) and pyridine-containing (14 and 16) derivatives displayed potency comparable to 2 in decreasing the cocaine-maintained responding at the doses tested (0.8, 1.7, and 3 mg/kg). However, these compounds did not produce the degree of separation between food- and cocaine-maintained responding that was seen with 2. Among the bicyclic fused-ring congeners 17-38, the indole-containing analog of 2, 22, showed the greatest affinity for binding to the DAT, with IC50 = 0.7 nM, whereas the corresponding indole-containing derivative of 1, 21, displayed the highest selectivity (over 600-fold) at this site vs the SERT site.

Effects of drugs on food- and cocaine-maintained responding, III: Dopaminergic antagonists.

The effects of three dopamine (DA) antagonists (SCH23390, pimozide, and chlorpromazine), with various degrees of selectivity for D1 and D2 receptors, and an agonist (the cocaine analog, CFT) were studied on responding maintained under a multiple fixed-ratio (FR) 30 food, FR30 cocaine (1-100 micrograms/kg per injection) delivery, with an interposed 10-min time-out (TO), schedule in rhesus monkeys. The effects of each drug depended upon the unit dose of cocaine. With an intermediate (10 micrograms/kg per injection) unit dose of cocaine, each antagonist decreased rates of responding maintained by either event in a dose-related manner. At higher (56-100 micrograms/kg per injection) unit doses of cocaine, antagonists generally increased and then decreased both food- and cocaine-maintained responding in a dose-related manner. These increases appeared to result from the blockade of non-specific rate-decreasing effects of self-administered cocaine, questioning their relevance to the reinforcing effects of cocaine. The results failed to support a role for pharmacological selectivity in this rate-decreasing effect of cocaine, as both D1 and D2 antagonists were able to reverse the effect. In contrast, CFT decreased cocaine-maintained responding at doses less than those that decreased food-maintained responding, and failed to shift the cocaine dose-effect function to the left. These results, together with previous work, suggest that agonists can selectively decrease drug-seeking behavior.