- Email: [email protected]
Similar effects of amphetamine and methylphenidate on the performance of complex operant tasks in rats
Behavioural Brain Research 109 (2000) 59 – 68 www.elsevier.com/locate/bbr
Research report
Similar effects of amphetamine and methylphenidate on the performance of complex operant tasks in rats Arthur J. Mayorga *, E. Jon Popke, C. Matthew Fogle, Merle G. Paule Food and Drug Administration, National Center for Toxicological Research, Di6ision of Neurotoxicology, HFT-132, Jefferson, AR 72079 -9502, USA Received 28 May 1999; received in revised form 5 November 1999; accepted 5 November 1999
Abstract Methylphenidate and D-amphetamine are central nervous system stimulants that have been suggested to share certain behavioral and neurochemical effects. The current study was undertaken to determine whether methylphenidate and D-amphetamine have similar effects on the performance of a battery of complex operant tasks in rats. Thus, the effects of amphetamine (0.1–6.0 mg/kg, i.p.) and methylphenidate (1.12–18.0 mg/kg, i.p) on the performance of rats in three complex food-reinforced operant tasks were examined. The tasks (and the brain functions they are intended to model) included: (1) conditioned position responding (auditory/visual/position discrimination); (2) incremental repeated acquisition (learning); and (3) temporal response differentiation (time estimation). In addition, each of these tasks was paired with a progressive ratio task to assess drug effects on the rats’ motivation to lever press for the food reinforcers used. Consistent with their effects in other behavioral paradigms, methylphenidate and D-amphetamine produced very similar patterns of disruption of the four tasks. Drug-induced changes in the endpoints of the progressive ratio task generally paralleled changes in the other three tasks, suggesting a major role for appetitive motivation in the effects of these agents. Several effects of these agents seen in the current study are consistent with their effects in children with attention-deficit-hyperactivity disorder. These data further validate the use of this battery of operant tasks for the characterization of pharmacological agents, and suggest that findings using these tasks may be predictive of what is seen in humans. © 2000 Elsevier Science B.V. All rights reserved. Keywords: Dopamine; Norepinephrine; Catecholamine; Cognitive; Behavior; Timing
1. Introduction Previous investigations comparing the effects of methylphenidate and D-amphetamine on behavior in rats have generally involved the study of non-operant behaviors, multiple simple operant behavioral tasks, or a single complex operant behavioral task [6,29,35,36,38,43]. The effects of methylphenidate and D-amphetamine on the performance of a battery of several complex operant behavioral tasks designed to model specific brain functions in rats has not been previously reported. Such a battery has been in use for * Corresponding author. Tel.: +1-870-543-7936; fax: + 1-870-5437745. E-mail address: [email protected] (A.J. Mayorga)
several years at the National Center for Toxicological Research to characterize the performance of complex operant tasks by rhesus monkeys and human children [32,33]. This battery consists of operant tasks designed to model color and position discrimination (conditioned position responding), learning (incremental repeated acquisition), time estimation (temporal response differentiation), memory (delayed-matching-to sample), and motivation (progressive ratio). These studies have demonstrated that task performance in human children and rhesus monkeys is very similar, and that children’s performance of several of the tasks correlates well with IQ scores [33,34]. Furthermore, correlations between representative endpoints of these different tasks in monkeys are generally not statistically significant; when endpoints do correlate, the correlation coefficients are
0166-4328/00/$ - see front matter © 2000 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 6 - 4 3 2 8 ( 9 9 ) 0 0 1 6 5 - 5
60
A.J. Mayorga et al. / Beha6ioural Brain Research 109 (2000) 59–68
small (5 0.4) [32]. These results suggest that each task is measuring aspects of brain function that are different from those measured in the other tasks. This is consistent with previous research suggesting that different brain areas are involved in the performance of each of these tasks [1,13,16,19]. Taken together, these data suggest that each of these complex operant tasks models a separate aspect of cognitive function. Thus, these tasks may be useful for determining the effects of pharmacological agents on specific aspects of brain function. The effects of several classical pharmacological agents on the performance of these complex operant tasks have been characterized in rhesus monkeys [4,5,8,9,11,12,40 –42]. The results of these studies suggest that an agent with a particular pharmacological or neurochemical mechanism of action produces a profile of effects on task performance that is unique from agents that act through different mechanisms. Recently, several of the complex operant tasks previously used to determine the effects of pharmacological agents in monkeys have been adapted for use in rats [27]. Thus, the effects of pharmacological agents on complex operant task performance in rats can be compared to the wealth of behavioral, pharmacological, and neurochemical data that exist for this species. Such comparisons may aid in the characterization of the mechanisms underlying complex operant task performance. Previous data suggest that methylphenidate and Damphetamine may have similar behavioral and neurochemical effects. Collectively, in vivo microdialysis data suggest that low to moderate doses of amphetamine ( B 1.0 mg/kg) and methylphenidate (B20.0 mg/kg) primarily elevate dopamine and norepinephrine levels in striatum, cortex, and hippocampus without affecting levels of serotonin [20 – 23]. Evidence suggesting that the two agents are behaviorally similar includes results from self-administration studies demonstrating that both rats and humans generalize methylphenidate to D-amphetamine [15,24,31]. Furthermore, the two agents produce common behavioral effects such as increases in locomotor activity, gnawing, and stereotypy [29,35]. Similar effects in experiments examining a single operant behavior have also been reported [6,36,43]. The current study investigated the effects of amphetamine (0.1–6.0 mg/kg, i.p.) and methylphenidate (1.12 –18.0 mg/kg, i.p.) on the performance of complex operant tasks designed to model auditory/visual/position discrimination, learning, and time estimation in rats. Since these two agents have very similar neurochemical and behavioral effects in other studies performed in rats, they were used as tools in the current study to validate the battery of operant tasks employed. In addition, these two agents are both used clinically in the treatment of attention-deficit-hyperactivity disorder [45]. Thus, it was also of interest to utilize these tasks to
predict which cognitive functions would be most sensitive to the effects of these agents. Accuracy and response rate were monitored for each task to permit comparison of drug effects across tasks. A time-constrained progressive ratio task was performed after each of the other tasks to assess the effects of drugs on the rats’ motivation to work for the food reinforcers used [46]. Time-constrained progressive ratio tasks have been used previously to facilitate comparison of drug effects on progressive ratio performance versus that of other time-constrained operant schedules [1,10,36]. For the progressive ratio task, the break point and the response rate were monitored.
2. Methods
2.1. Subjects A total of 24 male Sprague–Dawley rats obtained from the NCTR breeding colony were used. Subjects were individually housed at weaning (postnatal day 21) in standard Plexiglas cages lined with wood chips. Water was freely available, and the housing room was maintained on a 12:12-h light/dark cycle. Temperature and humidity were maintained at 21°C and 45–55%, respectively. These subjects were used in previous experiments to examine the effects of acute caffeine and nicotine administration on operant test battery performance. Behavioral baselines were similar before and after the administration of caffeine and nicotine, suggesting that these agents had reversible effects on behavior.
2.2. Apparatus All test sessions were conducted in operant behavior chambers with internal measurements of 24.8×22.9× 21.0 cm. Each chamber was located inside a sound-attenuating box and contained a front panel instrumented with three retractable response levers, each positioned below a stimulus light array. Reinforcer (45-mg dustless precision food pellets, Bioserve, Frenchtown, NJ) delivery into a trough below the center lever was accompanied by the noise of the pellet dispenser operation. Each chamber and panel were controlled by a microcomputer that administered the behavioral tasks and recorded the behavioral responses.
2.3. Operant training and testing procedure Beginning on postnatal day (PND) 70, rats were gradually food-deprived to 80–85% of their free-feeding weights and maintained at this weight throughout the experiment. On PND 90, three separate groups (n=8) of subjects began training to perform the condi-
A.J. Mayorga et al. / Beha6ioural Brain Research 109 (2000) 59–68
61
tioned position responding (CPR), incremental repeated acquisition (IRA), or the temporal response differentiation (TRD) tasks in 50-min daily sessions, 5 days per week. After stable performance (defined under Section 2.14 below) for each group on its respective task was achieved, the progressive ratio (PR) task was added. The PR task was added for all rats by PND 150. Each test session was 50 min long (40 min for the CPR/IRA/ TRD tasks followed immediately by the 10-min PR task). All sessions were conducted Monday through Friday at approximately the same time each day.
had elapsed. One of six serial position indicator lights signaled position in the response sequence, indicating the remaining number of correct responses required for reinforcer delivery. Responses on an incorrect lever resulted in the illumination of an incorrect response indicator light, no reinforcement, and initiation of a 2-s timeout period. Incorrect responses did not reset the response requirement, thus error correction was permitted. Correct responses were followed by illumination of the appropriate serial position indicator light, a correct response indicator light, and a 1-s timeout.
2.4. CPR (conditioned position responding)
2.6. TRD (temporal response differentiation)
In this task, rats were trained to discriminate between two auditory stimuli (a low-frequency tone (:350 Hz) vs. a high-frequency tone (:1450 Hz), both at :80 dB) or two visual stimuli (all lights of a 3 × 3 array illuminated vs. only the vertical center three lights illuminated). Each trial began with the presentation of either a tone or a pattern of lights on the array over the center lever. Subjects continued the trial by pressing the center lever (an observing response), which terminated the stimulus presentation after which the single central cue lights above the levers located to the left and the right of the center lever were immediately illuminated. If the stimulus had been either the low-frequency tone or the illumination of the center three lights in the 3× 3 array, a response (choice) on the left lever resulted in reinforcer delivery; if the stimulus was either the highfrequency tone or the illumination of all nine lights in the 3×3 array, a choice response on the right lever was reinforced. Incorrect responses resulted in a 10-s timeout period prior to initiation of the next trial (presentation of the next randomly presented stimuli; i.e. error correction was not permitted). Following correct responses, subjects immediately began a new trial.
Only the left retractable lever was used in this task. Rats were required to press and hold the lever in the depressed position for at least 10 s but not more than 14 s. Releasing the lever outside of this 10–14-s ‘window’ terminated the trial, after which the rat could immediately begin a new trial.
2.5. IRA (incremental repeated acquisition) All three retractable levers were extended for this task. Rats were required to learn a new sequence of lever presses for each test session. The task began with the presentation of a one-lever sequence (IRA1). Each response on the correct one of the three levers resulted in reinforcer delivery. After 20 correct, but not necessarily consecutive, responses, a 1-min timeout period was followed by the presentation of an ‘incremented’ two-lever sequence (IRA2). In the IRA2 sequence, a response on a different lever than the correct lever for IRA1 was required before a response on the original IRA1 lever produced reinforcement. After 20 errorless (no errors made between the first and last correct lever presses of the sequence) two-lever sequences, the task was incremented to a three-lever sequence and so on, up to a six-lever sequence or until the allotted task time
2.7. PR (progressi6e ratio) Only the right retractable lever was used in this task. Subjects were initially reinforced for making one lever press, after which the subject had to increase by one the number of lever presses made for each subsequent reinforcer.
2.8. Drugs and dosing procedure Methylphenidate hydrochloride and D-amphetamine sulfate (Research Biochemicals, Natick, MA) were each dissolved in sterile bacteriostatic 0.9% saline solution for a final injection volume of 1.0 ml/kg. The two drug studies were performed sequentially as follows: methylphenidate (PND 415–450), D-amphetamine (PND 465–500), with 2 weeks separating each drug study where rats received only saline injections. Doses of methylphenidate (1.12, 2.25, 4.50, 9.00, 18.00 mg/kg) or D-amphetamine (0.10, 0.30, 1.00, 3.00, 6.00 mg/kg) calculated as the salt were administered by intraperitoneal injection in a semi-randomized order 15 min prior to operant testing on Tuesdays and Fridays of each week. All doses were given on each of 2 separate test days. Testing without prior injection (baseline) was conducted on Mondays and Wednesdays, and saline injections were administered on Thursdays.
2.9. Beha6ioral endpoints For the auditory/visual/position discrimination, learning, and time estimation tasks, accuracy and response rate were monitored. For the motivation task, the break point and the response rate were monitored. Observing and choice response latencies were also mon-
62
A.J. Mayorga et al. / Beha6ioural Brain Research 109 (2000) 59–68
itored for the auditory/visual/position discrimination task. These endpoints are defined below.
2.10. Accuracy Task accuracy was calculated by dividing the number of correct responses by the total number of responses and multiplying this quotient by 100.
2.11. Response rate Task response rate was calculated by dividing the total number of responses in a session by the total task time (in seconds) minus any timeout periods.
2.12. Response latency For the auditory/visual/position discrimination task, the average response latency (time, in seconds, before a response is made) was calculated separately for observing responses and choice responses. Observing response latency is defined as the time between the onset of the stimulus and the observing response. Choice response latency is defined as the time between the completion of the observing response and the completion of the choice response.
2.13. Breakpoint Progressive ratio break point is defined as the value of the last ratio completed for which the subject received a reinforcer.
2.14. Statistical analysis Subjects from the following groups failed to exhibit stable performance for the measure of percent task completed and were excluded from the analyses: methylphenidate CPR (n =1), TRD (n = 2); D-amphetamine CPR (n= 1), TRD (n =3). However, these subjects did exhibit stable performance for the PR task and were therefore included in the analysis of PR endpoints. As in previous studies from this laboratory, stable performance was defined by percent task completed (reinforcers earned/reinforcers possible× 100) data for vehicle observations that had a standard error of less than 15% of its mean. The values for the vehicle treatment data for each subject used in the analyses were the averages of the ten vehicle sessions conducted during and for 2 weeks after the dose-response determination. The values for the drug treatment data used in the analyses were the average of the two drug sessions conducted for each subject at each dose. The overall effect of drug treatment for each endpoint in a given task was analyzed using a repeated-measures analysis of variance (ANOVA). If ANOVA (P B 0.05) revealed a
significant overall effect, Dunnett’s post-hoc test was used to determine whether data at each drug dose differed significantly from the vehicle average. Accuracy data were not included in the analysis at a given dose of drug if the majority of subjects failed to complete three trials for the TRD and CPR tasks or ten trials for the IRA task. For analysis of TRD response duration distributions, response 53 s in duration were omitted and the remaining data were fitted to a Gaussian distribution. The peak location (mean of the distribution) and the peak spread (coefficient of variation) were then estimated, and the drug-induced changes in these parameters were analyzed using a repeated-measures ANOVA followed by Dunnett’s post-hoc test (PB 0.05). Peak location and spread data at doses of drug that produced a non-Gaussian distribution were excluded from the ANOVAs (Gaussian fit determined using the ‘corrected Akaike information criterion’ [3]) Similar analyses have been used previously in the analysis of interresponse-time greater-than-t (IRT\t) schedules [47].
3. Results
3.1. Auditory/6isual/position discrimination task (CPR) Under vehicle conditions, subjects completed an average of 147 (9 19) trials. D-Amphetamine and methylphenidate both increased response rate at lower doses and decreased response rate at higher doses, although the methylphenidate-induced changes failed to be significant due to large variability (Fig. 1A). Both agents also significantly decreased accuracy at the dose that caused the greatest increase in response rate (Fig. 1B). The effects of either drug on accuracy did not differ regardless of whether visual or auditory stimuli were used (data not shown). Due to the large decreases in response rate that occurred at the higher doses of both agents, there were large increases in both the observing and choice response latencies at these doses (data not shown). In order to characterize whether drug-induced increases in CPR response rate were due to changes in observing and/or choice response latencies, an additional analysis was performed that included only doses of drug that increased response rate. Over the dose range of D-amphetamine or methylphenidate that increased response rate, both agents significantly decreased the observing response latency without affecting choice response latency (Fig. 1C).
3.2. Learning task (IRA) Under vehicle conditions, subjects completed an average of 119 (9 20) trials. D-Amphetamine and methylphenidate both increased response rate at lower
A.J. Mayorga et al. / Beha6ioural Brain Research 109 (2000) 59–68
63
Fig. 1. Effects of methylphenidate and D-amphetamine on the (A) response rate, (B) accuracy, and (C) observing (ORL) and choice (CHRL) response latency for the auditory/visual/position discrimination task. * Represents significant difference from vehicle control (Dunnett’s post-hoc test).
64
A.J. Mayorga et al. / Beha6ioural Brain Research 109 (2000) 59–68
doses and decreased response rate at higher doses, although the increases in response rate were not significant for either agent (Fig. 2A). Drug-induced increases in response rate did not significantly affect accuracy. Accuracy was only decreased at doses that either decreased or had no effect on response rate (Fig. 2B).
3.3. Time estimation task (TRD) Under vehicle conditions, subjects completed an average of 111 (944) trials. D-Amphetamine and methylphenidate both produced an increase in the variability of the response rate at low doses and a significant decrease in the response rate at higher doses (Fig. 3a). Accuracy in this task was significantly decreased at doses of D-amphetamine and methylphenidate that failed to decrease accuracy in the other tasks (Fig. 3b). D-Amphetamine (Fig. 3c) and methylphenidate (Fig. 3d) had similar effects on the pattern of the response duration distribution. ANOVA indicated an overall significant effect of D-amphetamine (F(2,8) =5.923) and methylphenidate (F(3,15) =6.000) on the peak lo-
cation. The 0.3-mg/kg dose of D-amphetamine and the 4.50-mg/kg dose of methylphenidate each produced a significant decrease in the mean peak location. This was accompanied by a significant increase in the peak spread for methylphenidate but not for D-amphetamine. Both agents also increased bursts (durations 5 3 s) at low doses and decreased bursts at higher doses.
3.4. Appetiti6e moti6ation task (PR) The results of the analysis of the PR task endpoints were the same regardless of the task which preceded PR, thus PR data for all subjects were combined. D-Amphetamine and methylphenidate both produced an increase in the response rate and break point at lower doses and a decrease in these endpoints at higher doses (Fig. 4A and B). Significant decreases in response rate and break point were seen at the higher doses of D-amphetamine but not methylphenidate.
3.5. O6erall task sensiti6ity Accuracy for the time estimation task was significantly decreased at doses of amphetamine (0.3 mg/kg) and methylphenidate (1.12 mg/kg) that failed to decrease any endpoint of the other tasks. Slightly higher doses (1.0 mg/kg amphetamine and 4.50 mg/kg methylphenidate) decreased accuracy in the auditory/visual/position discrimination task. Response rate and accuracy of the learning task were only significantly decreased at the two highest doses of amphetamine and methylphenidate tested.
4. Discussion
Fig. 2. Effects of methylphenidate and D-amphetamine on the (A) response rate and (B) accuracy for the learning task. * Represents significant difference from vehicle control (Dunnett’s post-hoc test).
The central nervous system stimulants D-amphetamine and methylphenidate produced a similar profile of effects on several complex operant tasks over the dose ranges tested. Response rates for all tasks were increased at low doses and decreased at higher doses of both agents. The increase in response rate for the auditory/visual/position discrimination task produced by both agents was characterized by a significant decrease in the observing but not the choice response latency. Accuracy of the time estimation task was significantly decreased at doses that did not significantly decrease the accuracy of any other task. Significant changes in the endpoints of the motivation task were also produced by both agents, suggesting that changes in the rats’ motivation to lever press for food may have affected performance of the other tasks. One advantage of examining drug effects on the performance of complex operant tasks is the potential to compare changes in response rate in different tasks
A.J. Mayorga et al. / Beha6ioural Brain Research 109 (2000) 59–68
65
Fig. 3. Effects of methylphenidate and D-amphetamine on the (a) response rate and (b) accuracy for the time estimation task. * Represents significant difference from vehicle control (Dunnett’s post-hoc test). Effects of (c) D-amphetamine and (d) methylphenidate on the relative distribution of response durations is also shown. Responses with corresponding durations of 0.01 – 1 s are shown in bin 1, 1.01 – 2 s in bin 2, etc., and all responses with durations \ 19.01 s in bin 20.
with changes in accuracy. For example, in the auditory/ visual/position discrimination task, doses of D-amphetamine that significantly increased response rate significantly decreased accuracy (Fig. 1). Although the increase in response rate induced by 4.5 mg/kg methylphenidate in that task was not statistically significant, it also coincided with a decrease in accuracy (Fig. 1). Interestingly, the increase in response rate was characterized by a decrease in the latency to make an observing response but not a choice response (Fig. 1C). The lack of a significant change in choice response latency accompanied by the decrease in accuracy suggests that some effect of the two agents other than the increased response rate interfered with the ability of the rats to make a correct choice. The increased variability in response rate seen for both agents in the time estimation task (Fig. 3a) was likely due to the drug-induced increase in the tendency of rats to engage in bursting (response durations 5 3 s; Fig. 3c and d). This tendency leads to a greater number of lever press durations that are not reinforced and a corresponding
decrease in task accuracy. For the learning task, however, methylphenidate-induced decreases in accuracy occurred at doses that did not affect response rate. Taken together, these data suggest that the effects of D-amphetamine and methylphenidate on complex operant task accuracy cannot always be explained simply by changes in response rate. The decrease in accuracy observed in the time estimation task is consistent with the effects of methylphenidate and D-amphetamine on the accuracy of performance of other behavioral tasks that require temporal discrimination [6,38,43]. However, the unique finding of the current work was that time estimation task accuracy was more sensitive to drug-induced decreases in performance than any endpoint of any task tested. This drug-induced decrease in accuracy was characterized by a significant left-shift in the peak location of the response duration distribution. Similar left-shifts in the response distributions of other timing tasks in response to methamphetamine have been reported previously [25,26,28]. These left-shifts have been
66
A.J. Mayorga et al. / Beha6ioural Brain Research 109 (2000) 59–68
suggested to result from an increase in the speed of an hypothesized ‘internal clock’ that is regulated by dopaminergic systems [30]. The current data supports this hypothesis, since drug-induced left-shifts in the TRD response duration distributions were observed at doses of D-amphetamine and methylphenidate similar to those that have been demonstrated to increase dopamine release in vivo [20,21]. Although it is tempting to speculate that the decrease in accuracy was entirely due to a change in the rats’ ability to estimate the passage of time, examination of the response duration distributions suggests that an increase in bursting responses (B3 s duration) may also have contributed to the decrease in correctly-timed responses (Fig. 3c and d). This may be due to competing behaviors induced by the two agents that interfered with the performance of the time estimation task. Methylphenidate and D-amphetamine have been demonstrated to produce increases in locomotion, exploration, and stereotypy over the dose ranges tested [29,35]. It is therefore possible
Fig. 4. Effects of methylphenidate and D-amphetamine on the (A) response rate and (B) break point for the progressive ratio task. * Represents significant difference from vehicle control (Dunnett’s post-hoc test).
that the increased tendency to engage in these competing behaviors reduced the ability of the rats to maintain lever hold of sufficient duration to achieve reinforcement (Fig. 3c and d). These competing behaviors may not have had as profound an effect on the other tasks tested, since they require lever presses of much shorter duration. In addition, other time estimation tasks that utilize presses of short duration (e.g. differential-reinforcement-of-low-rates of responding and peak interval) may also be less sensitive to competing behaviors. It may be useful to examine changes in the response rate and break point of the progressive ratio task to determine whether performance changes in other tasks are secondary to changes in appetitive motivation. Indeed, changes in the response rate and break point of the progressive ratio task were seen at doses of methylphenidate and D-amphetamine that produced similar changes in the response rates of some of the other tasks. For example, significant increases in the response rate of the auditory/visual/position discrimination task were observed at doses that also increased progressive ratio response rate and break point. This is consistent with a previous report demonstrating increases in time-constrained progressive ratio response rate and break point with administration of D-amphetamine or methylphenidate at similar doses [36]. In addition, the significant decreases in progressive ratio task endpoints observed at the two highest doses of D-amphetamine were extremely similar to the effects on response rate seen in the other three tasks. There were also notable exceptions, however. Significant increases in response rate for the auditory/visual/position discrimination task were seen at the 0.3-mg/kg dose of amphetamine, a dose that did not affect the progressive ratio endpoints. Similarly, the highest dose of methylphenidate significantly decreased response rate in the time estimation and learning tasks, but had no effect on progressive ratio response rate or break point. Drug-induced changes in progressive ratio break point have been suggested to represent changes in the ‘rewarding value’ of the reinforcer [17]. Thus, although there are important exceptions, these results suggest that changes in appetitive motivation play a major role in the effects of these agents on the performance of the other tasks tested. Alternatively, progressive ratio break point can also be affected by factors other than appetitive motivation, such as changes in the kinetic requirements of the response [44]. Thus, drug-induced changes in motor function may also have contributed to the effects seen in the current study. Although it is premature to speculate that specific neurochemical changes induced by methylphenidate and D-amphetamine are responsible for the changes seen in task performance, these agents have been reported to have similar effects on certain neurochemical endpoints within the dose range tested. In vivo microdi-
A.J. Mayorga et al. / Beha6ioural Brain Research 109 (2000) 59–68
alysis studies in rats indicate that methylphenidate increases synaptic levels of norepinephrine and dopamine in striatum, hippocampus, and cortex over the dose range currently tested [21]. D-Amphetamine has similar effects, except that significant increases in serotonin are also seen at doses \1.0 mg/kg [21,22]. The relative potency of these two agents with regard to their neurochemical effects is roughly consistent with their relative potency in producing behavioral effects in the current study. Conclusions regarding relative potency based on the current data must be made with caution, however, since phenomena such as tolerance and sensitization inherent to repeated dosing with these agents may be confounding factors. Nevertheless, the similar effects of these agents on the performance of the complex operant tasks tested may be related to their common neurochemical mechanisms of action. This notion is consistent with data generated in monkeys which demonstrate that agents with different pharmacological mechanisms have different patterns of effects on complex operant task performance. Furthermore, interesting behavioral differences seen in the current study may also reflect the neurochemical profiles of these agents. For example, the much greater efficacy of D-amphetamine at doses \1.0 mg/kg to decrease endpoints of the appetitive motivation task may reflect serotonergic involvement, since increased serotonin has been implicated in appetite suppression [14]. In light of this data, future studies should be conducted to examine neurochemical endpoints in rats performing complex operant tasks in order to determine the neurochemical mechanisms or specific brain areas underlying drug-induced changes in task performance. The current study did not examine the effects of D-amphetamine and methylphenidate in an animal model of attention-deficit-hyperactivity disorder (ADHD), although models of this disorder have recently been proposed [18,37]. Nevertheless, it may be interesting to compare the current results with the effects of these two agents on cognitive functions in children with ADHD. Time estimation is disrupted in ADHD children, and this disruption is not improved by methylphenidate administration [2]. This is consistent with the current results, which demonstrate a disruption of time estimation accuracy by methylphenidate. Also consistent with the current results, performance of learning tasks in ADHD children is disrupted at higher doses of stimulant medication [39,48]. Furthermore, the changes in progressive ratio endpoints seen in the current study are consistent with reports of changes in appetite in ADHD children treated with D-amphetamine and methylphenidate [7]. It would therefore be interesting to conduct future studies examining the effects of D-amphetamine and methylphenidate on cognitive task performance using a rat model of ADHD.
67
In conclusion, methylphenidate and D-amphetamine produced similar patterns of effects on the performance of four complex operant tasks designed to model cognitive functions in rats. This confirms and extends previous work demonstrating similar behavioral effects of these two agents in both rats and humans. Furthermore, comparison of the current results with published neurochemistry data suggests that the effects of these two agents on complex operant task performance reflect their neurochemical mechanisms of action. In addition, several effects of these agents on task performance reflect what is seen clinically in the treatment of attention-deficit-hyperactivity disorder in children. Thus, the current studies provide evidence that this battery of operant tasks is a valid tool to characterize the effects of drugs, and that the effects of agents in these tasks may be predictive of effects seen in humans. Acknowledgements Arthur J. Mayorga is supported by the Oak Ridge Institute for Science and Education Postgraduate Research Program. Expert statistical advice was provided by Brett Thorn. References [1] Aberman JE, Ward SJ, Salamone JD. Effects of dopamine antagonists and accumbens dopamine depletions on time-constrained progressive ratio performance. Pharmacol Biochem Behav 1998;61:341 – 8. [2] Barkley RA, Koplowitz S, Anderson T, McMurray MB. Sense of time in children with ADHD: effects of duration, distraction, and stimulant medication. J Int Neuropsychol Soc 1997;3:359– 69. [3] Bozdogan H. Model selection and Akaike’s Information Criterion (AIC): the general theory and its analytical extensions. Psychometrika 1987;52:345 – 70. [4] Buffalo EA, Gillam MP, Allen RR, Paule MG. Acute effects of caffeine on several operant behaviors in rhesus monkeys. Pharmacol Biochem Behav 1993;46:733 – 7. [5] Buffalo EA, Gillam MP, Allen RR, Paule MG. Acute behavioral effects of MK-801 in rhesus monkeys: assessment using an operant test battery. Pharmacol Biochem Behav 1994;48:935–40. [6] Eckerman DA, Segbefia D, Manning S, Breese GS. Effects of methylphenidate and D-amphetamine on timing in the rat. Pharmacol Biochem Behav 1987;27:513 – 5. [7] Efron D, Jarman F, Barker M. Side effects of methylphenidate and dexamphetamine in children with attention deficit hyperactivity disorder: a double-blind, crossover trial. Pediatrics 1997;100:662 – 6. [8] Ferguson SA, Paule MG. Acute effects of chlorpromazine in a monkey operant behavioral test battery. Pharmacol Biochem Behav 1992;42:333 – 41. [9] Ferguson SA, Paule MG. Acute effects of pentobarbital in a monkey operant behavioral test battery. Pharmacol Biochem Behav 1993;45:107 – 16. [10] Ferguson SA, Paule MG. Effects of chlorpromazine and diazepam on time estimation behavior and motivation in rats. Pharmacol Biochem Behav 1996;53:115 – 22.
68
A.J. Mayorga et al. / Beha6ioural Brain Research 109 (2000) 59–68
[11] Frederick DL, Schulze GE, Gillam MP, Paule MG. Acute effects of physostigmine on complex operant behavior in rhesus monkeys. Pharmacol Biochem Behav 1995;50:641–8. [12] Frederick DL, Gillam MP, Lensing S, Paule MG. Acute effects of LSD on rhesus monkey operant test battery performance. Pharmacol Biochem Behav 1997;57:633–41. [13] Friedman HR, Goldman-Rakic PS. Activation of the hippocampus and dentate gyrus by working memory: a 2-deoxyglucose study of behaving rhesus monkeys. J Neurosci 1988;8:4693 – 706. [14] Gibson EL, Kennedy AJ, Curzon J. D-Fenfluramine and D-norfenfluramine-induced hypophagia: differential mechanisms and involvement of postsynaptic 5-HT receptors. Eur J Pharmacol 1993;242:83 – 90. [15] Heishman SJ, Henningfield JE. Discriminative stimulus effects of D-amphetamine, methylphenidate, and diazepam in humans. Psychopharmacology 1991;103:436–42. [16] Hinton SC. Functional magnetic resonance imaging and pharmacological investigation of interval timing in humans. Neurotoxicol Teratol 1999;21:202. [17] Hodos W. Progressive ratio as a measure of reward strength. Science 1961;134:943–4. [18] Kohlert JG, Bloch GJ. A rat model for attention-deficit-hyperactivity disorder. Physiol Behav 1993;53:1215–8. [19] Kojima S, Kojima M, Goldman-Rakic PS. Operant behavioral analysis of memory loss in monkeys with prefrontal lesions. Brain Res 1982;248:51–9. [20] Kuczenski R, Segal D. Concomitant characterization of behavioral and striatal neurotransmitter response to amphetamine using in vivo microdialysis. J Neurosci 1989;9:2051–65. [21] Kuczenski R, Segal D. Effects of methylphenidate on extracellular dopamine, serotonin, and norepinephrine: comparison with amphetamine. J Neurochem 1997;68:2032–7. [22] Kuczenski R, Segal DS, Leith NJ, Applegate CD. Effects of amphetamine, methylphenidate, and apomorphine on regional brain serotonin and 5-hydroxyindole acetic acid. Psychopharmacology 1987;93:329 –35. [23] Kuczenski R, Segal DS, Cho AK, Melega W. Hippocampus norepinephrine, caudate dopamine and serotonin, and behavioral responses to the stereoisomers of amphetamine and methamphetamine. J Neurosci 1995;15:1308–17. [24] Leith NJ, Barrett RJ. Self-stimulation and amphetamine: tolerance to D and L isomers and cross tolerance to cocaine and methylphenidate. Psychopharmacology 1981;74:23–8. [25] Maricq AV, Church RM. The differential effects of haloperidol and methamphetamine on time estimation in the rat. Psychopharmacology 1983;79:10–5. [26] Maricq AV, Roberts S, Church RM. Methamphetamine and time estimation. J Exp Psychol Anim Behav Proc 1981;7:18 – 30. [27] Mayorga AJ, Fogle CM, Paule MG. Adaptation of a primate operant test battery to the rat: effects of chlorpromazine. Neurotoxicol Teratol 2000;22:31–9. [28] McClure GYH, Wenger GR, McMillan DE. Effects of drugs on response duration differentiation. V: Differential effects under temporal response differentiation schedules. J Pharmacol Exp Ther 1997;281:1357 –67. [29] McNamara CG, Davidson ES, Schenk S. A comparison of the motor-activating effects of acute and chronic exposure to amphetamine and methylphenidate. Pharmacol Biochem Behav 1993;45:729 – 32.
.
[30] Meck WH. Selective adjustment of the speed of internal clock and memory processes. J Exp Psychol Anim Behav Proc 1983;9:171 – 201. [31] Nielsen JA, Duda NJ, Mokler DJ, Moore KE. Self-administration of central stimulants by rats: a comparison of the effects of D-amphetamine, methylphenidate, and McNeil 4612. Pharmacol Biochem Behav 1984;20:227 – 32. [32] Paule MG. Use of the NCTR operant test battery in non-human primates. Neurotoxicol Teratol 1990;12:413 – 8. [33] Paule MG, Forrester TM, Maher MA, Cranmer JM, Allen RR. Monkey versus human performance in the NCTR operant test battery. Neurotoxicol Teratol 1990;12:503 – 7. [34] Paule MG, Chelonis JJ, Buffalo EA, Blake DJ, Casey PH. Operant test battery performance in children: correlation with IQ. Neurotoxicol Teratol 1999;21:223 – 30. [35] Pechnick R, Janowsky DS, Judd L. Differential effects of methylphenidate and D-amphetamine on stereotyped behavior in the rat. Psychopharmacology 1979;65:311 – 5. [36] Poncelet M, Chermat R, Soubrie P, Simon P. The progressive ratio schedule as a model for studying the psychomotor stimulant activity of drugs in the rat. Psychopharmacology 1983;80:184 – 9. [37] Puumala T, Ruotsalainen S, Jakala P, Koivisto E, Riekkinen P Jr, Sirvio J. Behavioral and pharmacological studies on the validation of a new animal model for attention deficit hyperactivity disorder. Neurobiol Learn Mem 1996;66:198 – 211. [38] Rapp DL, Robbins TW. The effects of D-amphetamine on temporal discrimination in the rat. Psychopharmacology 1976;51:91 – 100. [39] Risser MG, Bowers TG. Cognitive and neuropsychological characteristics of attention deficit hyperactivity disorder children receiving stimulant medications. Percept Mot Skills 1993;77:1023 – 31. [40] Schulze GE, Paule MG. Effects of morphine sulfate on operant behavior in rhesus monkeys. Pharmacol Biochem Behav 1991;38:77 – 83. [41] Schulze GE, McMillan DE, Bailey JR, Scallet AC, Ali SF, Slikker W Jr, et al. Acute effects of delta-9-tetrahydrocannabinol in rhesus monkeys as measured by performance in a battery of cognitive function tests. J Pharmacol Exp Ther 1988;245:178–86. [42] Schulze GE, Slikker W Jr, Paule MG. Multiple behavioral effects of diazepam in rhesus monkeys. Pharmacol Biochem Behav 1989;34:29 – 35. [43] Seiden LS, Andresen J, MacPhail RC. Methylphenidate and D-amphetamine: effects and interactions with alphamethyltyrosine and tetrabenzene on DRL performance in rats. Pharmacol Biochem Behav 1979;10:577 – 84. [44] Skjoldager P, Pierre PJ, Mittleman G. Reinforcer magnitude and progressive ratio responding: effects of increased effort, prefeeding, and extinction. Learn Motiv 1993;24:303 – 43. [45] Solanto MV. Neuropsychopharmacological mechanisms of stimulant drug action in attention-deficit hyperactivity disorder: a review and integration. Behav Brain Res 1998;94:127 –52. [46] Stewart WJ. Progressive reinforcement schedules: a review and evaluation. Aust J Psychol 1974;27:9 – 22. [47] Tannock R, Schachar R. Methylphenidate and cognitive perseveration in hyperactive children. J Child Psychol Psychiatry 1992;33:1217 – 28. [48] Wogar MA, Bradshaw CM, Szabadi E. Does the effect of central 5-hydroxytryptamine depletion on timing depend on motivational change? Psychopharmacology 1993;112:86 – 92.
Research report
Similar effects of amphetamine and methylphenidate on the performance of complex operant tasks in rats Arthur J. Mayorga *, E. Jon Popke, C. Matthew Fogle, Merle G. Paule Food and Drug Administration, National Center for Toxicological Research, Di6ision of Neurotoxicology, HFT-132, Jefferson, AR 72079 -9502, USA Received 28 May 1999; received in revised form 5 November 1999; accepted 5 November 1999
Abstract Methylphenidate and D-amphetamine are central nervous system stimulants that have been suggested to share certain behavioral and neurochemical effects. The current study was undertaken to determine whether methylphenidate and D-amphetamine have similar effects on the performance of a battery of complex operant tasks in rats. Thus, the effects of amphetamine (0.1–6.0 mg/kg, i.p.) and methylphenidate (1.12–18.0 mg/kg, i.p) on the performance of rats in three complex food-reinforced operant tasks were examined. The tasks (and the brain functions they are intended to model) included: (1) conditioned position responding (auditory/visual/position discrimination); (2) incremental repeated acquisition (learning); and (3) temporal response differentiation (time estimation). In addition, each of these tasks was paired with a progressive ratio task to assess drug effects on the rats’ motivation to lever press for the food reinforcers used. Consistent with their effects in other behavioral paradigms, methylphenidate and D-amphetamine produced very similar patterns of disruption of the four tasks. Drug-induced changes in the endpoints of the progressive ratio task generally paralleled changes in the other three tasks, suggesting a major role for appetitive motivation in the effects of these agents. Several effects of these agents seen in the current study are consistent with their effects in children with attention-deficit-hyperactivity disorder. These data further validate the use of this battery of operant tasks for the characterization of pharmacological agents, and suggest that findings using these tasks may be predictive of what is seen in humans. © 2000 Elsevier Science B.V. All rights reserved. Keywords: Dopamine; Norepinephrine; Catecholamine; Cognitive; Behavior; Timing
1. Introduction Previous investigations comparing the effects of methylphenidate and D-amphetamine on behavior in rats have generally involved the study of non-operant behaviors, multiple simple operant behavioral tasks, or a single complex operant behavioral task [6,29,35,36,38,43]. The effects of methylphenidate and D-amphetamine on the performance of a battery of several complex operant behavioral tasks designed to model specific brain functions in rats has not been previously reported. Such a battery has been in use for * Corresponding author. Tel.: +1-870-543-7936; fax: + 1-870-5437745. E-mail address: [email protected] (A.J. Mayorga)
several years at the National Center for Toxicological Research to characterize the performance of complex operant tasks by rhesus monkeys and human children [32,33]. This battery consists of operant tasks designed to model color and position discrimination (conditioned position responding), learning (incremental repeated acquisition), time estimation (temporal response differentiation), memory (delayed-matching-to sample), and motivation (progressive ratio). These studies have demonstrated that task performance in human children and rhesus monkeys is very similar, and that children’s performance of several of the tasks correlates well with IQ scores [33,34]. Furthermore, correlations between representative endpoints of these different tasks in monkeys are generally not statistically significant; when endpoints do correlate, the correlation coefficients are
0166-4328/00/$ - see front matter © 2000 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 6 - 4 3 2 8 ( 9 9 ) 0 0 1 6 5 - 5
60
A.J. Mayorga et al. / Beha6ioural Brain Research 109 (2000) 59–68
small (5 0.4) [32]. These results suggest that each task is measuring aspects of brain function that are different from those measured in the other tasks. This is consistent with previous research suggesting that different brain areas are involved in the performance of each of these tasks [1,13,16,19]. Taken together, these data suggest that each of these complex operant tasks models a separate aspect of cognitive function. Thus, these tasks may be useful for determining the effects of pharmacological agents on specific aspects of brain function. The effects of several classical pharmacological agents on the performance of these complex operant tasks have been characterized in rhesus monkeys [4,5,8,9,11,12,40 –42]. The results of these studies suggest that an agent with a particular pharmacological or neurochemical mechanism of action produces a profile of effects on task performance that is unique from agents that act through different mechanisms. Recently, several of the complex operant tasks previously used to determine the effects of pharmacological agents in monkeys have been adapted for use in rats [27]. Thus, the effects of pharmacological agents on complex operant task performance in rats can be compared to the wealth of behavioral, pharmacological, and neurochemical data that exist for this species. Such comparisons may aid in the characterization of the mechanisms underlying complex operant task performance. Previous data suggest that methylphenidate and Damphetamine may have similar behavioral and neurochemical effects. Collectively, in vivo microdialysis data suggest that low to moderate doses of amphetamine ( B 1.0 mg/kg) and methylphenidate (B20.0 mg/kg) primarily elevate dopamine and norepinephrine levels in striatum, cortex, and hippocampus without affecting levels of serotonin [20 – 23]. Evidence suggesting that the two agents are behaviorally similar includes results from self-administration studies demonstrating that both rats and humans generalize methylphenidate to D-amphetamine [15,24,31]. Furthermore, the two agents produce common behavioral effects such as increases in locomotor activity, gnawing, and stereotypy [29,35]. Similar effects in experiments examining a single operant behavior have also been reported [6,36,43]. The current study investigated the effects of amphetamine (0.1–6.0 mg/kg, i.p.) and methylphenidate (1.12 –18.0 mg/kg, i.p.) on the performance of complex operant tasks designed to model auditory/visual/position discrimination, learning, and time estimation in rats. Since these two agents have very similar neurochemical and behavioral effects in other studies performed in rats, they were used as tools in the current study to validate the battery of operant tasks employed. In addition, these two agents are both used clinically in the treatment of attention-deficit-hyperactivity disorder [45]. Thus, it was also of interest to utilize these tasks to
predict which cognitive functions would be most sensitive to the effects of these agents. Accuracy and response rate were monitored for each task to permit comparison of drug effects across tasks. A time-constrained progressive ratio task was performed after each of the other tasks to assess the effects of drugs on the rats’ motivation to work for the food reinforcers used [46]. Time-constrained progressive ratio tasks have been used previously to facilitate comparison of drug effects on progressive ratio performance versus that of other time-constrained operant schedules [1,10,36]. For the progressive ratio task, the break point and the response rate were monitored.
2. Methods
2.1. Subjects A total of 24 male Sprague–Dawley rats obtained from the NCTR breeding colony were used. Subjects were individually housed at weaning (postnatal day 21) in standard Plexiglas cages lined with wood chips. Water was freely available, and the housing room was maintained on a 12:12-h light/dark cycle. Temperature and humidity were maintained at 21°C and 45–55%, respectively. These subjects were used in previous experiments to examine the effects of acute caffeine and nicotine administration on operant test battery performance. Behavioral baselines were similar before and after the administration of caffeine and nicotine, suggesting that these agents had reversible effects on behavior.
2.2. Apparatus All test sessions were conducted in operant behavior chambers with internal measurements of 24.8×22.9× 21.0 cm. Each chamber was located inside a sound-attenuating box and contained a front panel instrumented with three retractable response levers, each positioned below a stimulus light array. Reinforcer (45-mg dustless precision food pellets, Bioserve, Frenchtown, NJ) delivery into a trough below the center lever was accompanied by the noise of the pellet dispenser operation. Each chamber and panel were controlled by a microcomputer that administered the behavioral tasks and recorded the behavioral responses.
2.3. Operant training and testing procedure Beginning on postnatal day (PND) 70, rats were gradually food-deprived to 80–85% of their free-feeding weights and maintained at this weight throughout the experiment. On PND 90, three separate groups (n=8) of subjects began training to perform the condi-
A.J. Mayorga et al. / Beha6ioural Brain Research 109 (2000) 59–68
61
tioned position responding (CPR), incremental repeated acquisition (IRA), or the temporal response differentiation (TRD) tasks in 50-min daily sessions, 5 days per week. After stable performance (defined under Section 2.14 below) for each group on its respective task was achieved, the progressive ratio (PR) task was added. The PR task was added for all rats by PND 150. Each test session was 50 min long (40 min for the CPR/IRA/ TRD tasks followed immediately by the 10-min PR task). All sessions were conducted Monday through Friday at approximately the same time each day.
had elapsed. One of six serial position indicator lights signaled position in the response sequence, indicating the remaining number of correct responses required for reinforcer delivery. Responses on an incorrect lever resulted in the illumination of an incorrect response indicator light, no reinforcement, and initiation of a 2-s timeout period. Incorrect responses did not reset the response requirement, thus error correction was permitted. Correct responses were followed by illumination of the appropriate serial position indicator light, a correct response indicator light, and a 1-s timeout.
2.4. CPR (conditioned position responding)
2.6. TRD (temporal response differentiation)
In this task, rats were trained to discriminate between two auditory stimuli (a low-frequency tone (:350 Hz) vs. a high-frequency tone (:1450 Hz), both at :80 dB) or two visual stimuli (all lights of a 3 × 3 array illuminated vs. only the vertical center three lights illuminated). Each trial began with the presentation of either a tone or a pattern of lights on the array over the center lever. Subjects continued the trial by pressing the center lever (an observing response), which terminated the stimulus presentation after which the single central cue lights above the levers located to the left and the right of the center lever were immediately illuminated. If the stimulus had been either the low-frequency tone or the illumination of the center three lights in the 3× 3 array, a response (choice) on the left lever resulted in reinforcer delivery; if the stimulus was either the highfrequency tone or the illumination of all nine lights in the 3×3 array, a choice response on the right lever was reinforced. Incorrect responses resulted in a 10-s timeout period prior to initiation of the next trial (presentation of the next randomly presented stimuli; i.e. error correction was not permitted). Following correct responses, subjects immediately began a new trial.
Only the left retractable lever was used in this task. Rats were required to press and hold the lever in the depressed position for at least 10 s but not more than 14 s. Releasing the lever outside of this 10–14-s ‘window’ terminated the trial, after which the rat could immediately begin a new trial.
2.5. IRA (incremental repeated acquisition) All three retractable levers were extended for this task. Rats were required to learn a new sequence of lever presses for each test session. The task began with the presentation of a one-lever sequence (IRA1). Each response on the correct one of the three levers resulted in reinforcer delivery. After 20 correct, but not necessarily consecutive, responses, a 1-min timeout period was followed by the presentation of an ‘incremented’ two-lever sequence (IRA2). In the IRA2 sequence, a response on a different lever than the correct lever for IRA1 was required before a response on the original IRA1 lever produced reinforcement. After 20 errorless (no errors made between the first and last correct lever presses of the sequence) two-lever sequences, the task was incremented to a three-lever sequence and so on, up to a six-lever sequence or until the allotted task time
2.7. PR (progressi6e ratio) Only the right retractable lever was used in this task. Subjects were initially reinforced for making one lever press, after which the subject had to increase by one the number of lever presses made for each subsequent reinforcer.
2.8. Drugs and dosing procedure Methylphenidate hydrochloride and D-amphetamine sulfate (Research Biochemicals, Natick, MA) were each dissolved in sterile bacteriostatic 0.9% saline solution for a final injection volume of 1.0 ml/kg. The two drug studies were performed sequentially as follows: methylphenidate (PND 415–450), D-amphetamine (PND 465–500), with 2 weeks separating each drug study where rats received only saline injections. Doses of methylphenidate (1.12, 2.25, 4.50, 9.00, 18.00 mg/kg) or D-amphetamine (0.10, 0.30, 1.00, 3.00, 6.00 mg/kg) calculated as the salt were administered by intraperitoneal injection in a semi-randomized order 15 min prior to operant testing on Tuesdays and Fridays of each week. All doses were given on each of 2 separate test days. Testing without prior injection (baseline) was conducted on Mondays and Wednesdays, and saline injections were administered on Thursdays.
2.9. Beha6ioral endpoints For the auditory/visual/position discrimination, learning, and time estimation tasks, accuracy and response rate were monitored. For the motivation task, the break point and the response rate were monitored. Observing and choice response latencies were also mon-
62
A.J. Mayorga et al. / Beha6ioural Brain Research 109 (2000) 59–68
itored for the auditory/visual/position discrimination task. These endpoints are defined below.
2.10. Accuracy Task accuracy was calculated by dividing the number of correct responses by the total number of responses and multiplying this quotient by 100.
2.11. Response rate Task response rate was calculated by dividing the total number of responses in a session by the total task time (in seconds) minus any timeout periods.
2.12. Response latency For the auditory/visual/position discrimination task, the average response latency (time, in seconds, before a response is made) was calculated separately for observing responses and choice responses. Observing response latency is defined as the time between the onset of the stimulus and the observing response. Choice response latency is defined as the time between the completion of the observing response and the completion of the choice response.
2.13. Breakpoint Progressive ratio break point is defined as the value of the last ratio completed for which the subject received a reinforcer.
2.14. Statistical analysis Subjects from the following groups failed to exhibit stable performance for the measure of percent task completed and were excluded from the analyses: methylphenidate CPR (n =1), TRD (n = 2); D-amphetamine CPR (n= 1), TRD (n =3). However, these subjects did exhibit stable performance for the PR task and were therefore included in the analysis of PR endpoints. As in previous studies from this laboratory, stable performance was defined by percent task completed (reinforcers earned/reinforcers possible× 100) data for vehicle observations that had a standard error of less than 15% of its mean. The values for the vehicle treatment data for each subject used in the analyses were the averages of the ten vehicle sessions conducted during and for 2 weeks after the dose-response determination. The values for the drug treatment data used in the analyses were the average of the two drug sessions conducted for each subject at each dose. The overall effect of drug treatment for each endpoint in a given task was analyzed using a repeated-measures analysis of variance (ANOVA). If ANOVA (P B 0.05) revealed a
significant overall effect, Dunnett’s post-hoc test was used to determine whether data at each drug dose differed significantly from the vehicle average. Accuracy data were not included in the analysis at a given dose of drug if the majority of subjects failed to complete three trials for the TRD and CPR tasks or ten trials for the IRA task. For analysis of TRD response duration distributions, response 53 s in duration were omitted and the remaining data were fitted to a Gaussian distribution. The peak location (mean of the distribution) and the peak spread (coefficient of variation) were then estimated, and the drug-induced changes in these parameters were analyzed using a repeated-measures ANOVA followed by Dunnett’s post-hoc test (PB 0.05). Peak location and spread data at doses of drug that produced a non-Gaussian distribution were excluded from the ANOVAs (Gaussian fit determined using the ‘corrected Akaike information criterion’ [3]) Similar analyses have been used previously in the analysis of interresponse-time greater-than-t (IRT\t) schedules [47].
3. Results
3.1. Auditory/6isual/position discrimination task (CPR) Under vehicle conditions, subjects completed an average of 147 (9 19) trials. D-Amphetamine and methylphenidate both increased response rate at lower doses and decreased response rate at higher doses, although the methylphenidate-induced changes failed to be significant due to large variability (Fig. 1A). Both agents also significantly decreased accuracy at the dose that caused the greatest increase in response rate (Fig. 1B). The effects of either drug on accuracy did not differ regardless of whether visual or auditory stimuli were used (data not shown). Due to the large decreases in response rate that occurred at the higher doses of both agents, there were large increases in both the observing and choice response latencies at these doses (data not shown). In order to characterize whether drug-induced increases in CPR response rate were due to changes in observing and/or choice response latencies, an additional analysis was performed that included only doses of drug that increased response rate. Over the dose range of D-amphetamine or methylphenidate that increased response rate, both agents significantly decreased the observing response latency without affecting choice response latency (Fig. 1C).
3.2. Learning task (IRA) Under vehicle conditions, subjects completed an average of 119 (9 20) trials. D-Amphetamine and methylphenidate both increased response rate at lower
A.J. Mayorga et al. / Beha6ioural Brain Research 109 (2000) 59–68
63
Fig. 1. Effects of methylphenidate and D-amphetamine on the (A) response rate, (B) accuracy, and (C) observing (ORL) and choice (CHRL) response latency for the auditory/visual/position discrimination task. * Represents significant difference from vehicle control (Dunnett’s post-hoc test).
64
A.J. Mayorga et al. / Beha6ioural Brain Research 109 (2000) 59–68
doses and decreased response rate at higher doses, although the increases in response rate were not significant for either agent (Fig. 2A). Drug-induced increases in response rate did not significantly affect accuracy. Accuracy was only decreased at doses that either decreased or had no effect on response rate (Fig. 2B).
3.3. Time estimation task (TRD) Under vehicle conditions, subjects completed an average of 111 (944) trials. D-Amphetamine and methylphenidate both produced an increase in the variability of the response rate at low doses and a significant decrease in the response rate at higher doses (Fig. 3a). Accuracy in this task was significantly decreased at doses of D-amphetamine and methylphenidate that failed to decrease accuracy in the other tasks (Fig. 3b). D-Amphetamine (Fig. 3c) and methylphenidate (Fig. 3d) had similar effects on the pattern of the response duration distribution. ANOVA indicated an overall significant effect of D-amphetamine (F(2,8) =5.923) and methylphenidate (F(3,15) =6.000) on the peak lo-
cation. The 0.3-mg/kg dose of D-amphetamine and the 4.50-mg/kg dose of methylphenidate each produced a significant decrease in the mean peak location. This was accompanied by a significant increase in the peak spread for methylphenidate but not for D-amphetamine. Both agents also increased bursts (durations 5 3 s) at low doses and decreased bursts at higher doses.
3.4. Appetiti6e moti6ation task (PR) The results of the analysis of the PR task endpoints were the same regardless of the task which preceded PR, thus PR data for all subjects were combined. D-Amphetamine and methylphenidate both produced an increase in the response rate and break point at lower doses and a decrease in these endpoints at higher doses (Fig. 4A and B). Significant decreases in response rate and break point were seen at the higher doses of D-amphetamine but not methylphenidate.
3.5. O6erall task sensiti6ity Accuracy for the time estimation task was significantly decreased at doses of amphetamine (0.3 mg/kg) and methylphenidate (1.12 mg/kg) that failed to decrease any endpoint of the other tasks. Slightly higher doses (1.0 mg/kg amphetamine and 4.50 mg/kg methylphenidate) decreased accuracy in the auditory/visual/position discrimination task. Response rate and accuracy of the learning task were only significantly decreased at the two highest doses of amphetamine and methylphenidate tested.
4. Discussion
Fig. 2. Effects of methylphenidate and D-amphetamine on the (A) response rate and (B) accuracy for the learning task. * Represents significant difference from vehicle control (Dunnett’s post-hoc test).
The central nervous system stimulants D-amphetamine and methylphenidate produced a similar profile of effects on several complex operant tasks over the dose ranges tested. Response rates for all tasks were increased at low doses and decreased at higher doses of both agents. The increase in response rate for the auditory/visual/position discrimination task produced by both agents was characterized by a significant decrease in the observing but not the choice response latency. Accuracy of the time estimation task was significantly decreased at doses that did not significantly decrease the accuracy of any other task. Significant changes in the endpoints of the motivation task were also produced by both agents, suggesting that changes in the rats’ motivation to lever press for food may have affected performance of the other tasks. One advantage of examining drug effects on the performance of complex operant tasks is the potential to compare changes in response rate in different tasks
A.J. Mayorga et al. / Beha6ioural Brain Research 109 (2000) 59–68
65
Fig. 3. Effects of methylphenidate and D-amphetamine on the (a) response rate and (b) accuracy for the time estimation task. * Represents significant difference from vehicle control (Dunnett’s post-hoc test). Effects of (c) D-amphetamine and (d) methylphenidate on the relative distribution of response durations is also shown. Responses with corresponding durations of 0.01 – 1 s are shown in bin 1, 1.01 – 2 s in bin 2, etc., and all responses with durations \ 19.01 s in bin 20.
with changes in accuracy. For example, in the auditory/ visual/position discrimination task, doses of D-amphetamine that significantly increased response rate significantly decreased accuracy (Fig. 1). Although the increase in response rate induced by 4.5 mg/kg methylphenidate in that task was not statistically significant, it also coincided with a decrease in accuracy (Fig. 1). Interestingly, the increase in response rate was characterized by a decrease in the latency to make an observing response but not a choice response (Fig. 1C). The lack of a significant change in choice response latency accompanied by the decrease in accuracy suggests that some effect of the two agents other than the increased response rate interfered with the ability of the rats to make a correct choice. The increased variability in response rate seen for both agents in the time estimation task (Fig. 3a) was likely due to the drug-induced increase in the tendency of rats to engage in bursting (response durations 5 3 s; Fig. 3c and d). This tendency leads to a greater number of lever press durations that are not reinforced and a corresponding
decrease in task accuracy. For the learning task, however, methylphenidate-induced decreases in accuracy occurred at doses that did not affect response rate. Taken together, these data suggest that the effects of D-amphetamine and methylphenidate on complex operant task accuracy cannot always be explained simply by changes in response rate. The decrease in accuracy observed in the time estimation task is consistent with the effects of methylphenidate and D-amphetamine on the accuracy of performance of other behavioral tasks that require temporal discrimination [6,38,43]. However, the unique finding of the current work was that time estimation task accuracy was more sensitive to drug-induced decreases in performance than any endpoint of any task tested. This drug-induced decrease in accuracy was characterized by a significant left-shift in the peak location of the response duration distribution. Similar left-shifts in the response distributions of other timing tasks in response to methamphetamine have been reported previously [25,26,28]. These left-shifts have been
66
A.J. Mayorga et al. / Beha6ioural Brain Research 109 (2000) 59–68
suggested to result from an increase in the speed of an hypothesized ‘internal clock’ that is regulated by dopaminergic systems [30]. The current data supports this hypothesis, since drug-induced left-shifts in the TRD response duration distributions were observed at doses of D-amphetamine and methylphenidate similar to those that have been demonstrated to increase dopamine release in vivo [20,21]. Although it is tempting to speculate that the decrease in accuracy was entirely due to a change in the rats’ ability to estimate the passage of time, examination of the response duration distributions suggests that an increase in bursting responses (B3 s duration) may also have contributed to the decrease in correctly-timed responses (Fig. 3c and d). This may be due to competing behaviors induced by the two agents that interfered with the performance of the time estimation task. Methylphenidate and D-amphetamine have been demonstrated to produce increases in locomotion, exploration, and stereotypy over the dose ranges tested [29,35]. It is therefore possible
Fig. 4. Effects of methylphenidate and D-amphetamine on the (A) response rate and (B) break point for the progressive ratio task. * Represents significant difference from vehicle control (Dunnett’s post-hoc test).
that the increased tendency to engage in these competing behaviors reduced the ability of the rats to maintain lever hold of sufficient duration to achieve reinforcement (Fig. 3c and d). These competing behaviors may not have had as profound an effect on the other tasks tested, since they require lever presses of much shorter duration. In addition, other time estimation tasks that utilize presses of short duration (e.g. differential-reinforcement-of-low-rates of responding and peak interval) may also be less sensitive to competing behaviors. It may be useful to examine changes in the response rate and break point of the progressive ratio task to determine whether performance changes in other tasks are secondary to changes in appetitive motivation. Indeed, changes in the response rate and break point of the progressive ratio task were seen at doses of methylphenidate and D-amphetamine that produced similar changes in the response rates of some of the other tasks. For example, significant increases in the response rate of the auditory/visual/position discrimination task were observed at doses that also increased progressive ratio response rate and break point. This is consistent with a previous report demonstrating increases in time-constrained progressive ratio response rate and break point with administration of D-amphetamine or methylphenidate at similar doses [36]. In addition, the significant decreases in progressive ratio task endpoints observed at the two highest doses of D-amphetamine were extremely similar to the effects on response rate seen in the other three tasks. There were also notable exceptions, however. Significant increases in response rate for the auditory/visual/position discrimination task were seen at the 0.3-mg/kg dose of amphetamine, a dose that did not affect the progressive ratio endpoints. Similarly, the highest dose of methylphenidate significantly decreased response rate in the time estimation and learning tasks, but had no effect on progressive ratio response rate or break point. Drug-induced changes in progressive ratio break point have been suggested to represent changes in the ‘rewarding value’ of the reinforcer [17]. Thus, although there are important exceptions, these results suggest that changes in appetitive motivation play a major role in the effects of these agents on the performance of the other tasks tested. Alternatively, progressive ratio break point can also be affected by factors other than appetitive motivation, such as changes in the kinetic requirements of the response [44]. Thus, drug-induced changes in motor function may also have contributed to the effects seen in the current study. Although it is premature to speculate that specific neurochemical changes induced by methylphenidate and D-amphetamine are responsible for the changes seen in task performance, these agents have been reported to have similar effects on certain neurochemical endpoints within the dose range tested. In vivo microdi-
A.J. Mayorga et al. / Beha6ioural Brain Research 109 (2000) 59–68
alysis studies in rats indicate that methylphenidate increases synaptic levels of norepinephrine and dopamine in striatum, hippocampus, and cortex over the dose range currently tested [21]. D-Amphetamine has similar effects, except that significant increases in serotonin are also seen at doses \1.0 mg/kg [21,22]. The relative potency of these two agents with regard to their neurochemical effects is roughly consistent with their relative potency in producing behavioral effects in the current study. Conclusions regarding relative potency based on the current data must be made with caution, however, since phenomena such as tolerance and sensitization inherent to repeated dosing with these agents may be confounding factors. Nevertheless, the similar effects of these agents on the performance of the complex operant tasks tested may be related to their common neurochemical mechanisms of action. This notion is consistent with data generated in monkeys which demonstrate that agents with different pharmacological mechanisms have different patterns of effects on complex operant task performance. Furthermore, interesting behavioral differences seen in the current study may also reflect the neurochemical profiles of these agents. For example, the much greater efficacy of D-amphetamine at doses \1.0 mg/kg to decrease endpoints of the appetitive motivation task may reflect serotonergic involvement, since increased serotonin has been implicated in appetite suppression [14]. In light of this data, future studies should be conducted to examine neurochemical endpoints in rats performing complex operant tasks in order to determine the neurochemical mechanisms or specific brain areas underlying drug-induced changes in task performance. The current study did not examine the effects of D-amphetamine and methylphenidate in an animal model of attention-deficit-hyperactivity disorder (ADHD), although models of this disorder have recently been proposed [18,37]. Nevertheless, it may be interesting to compare the current results with the effects of these two agents on cognitive functions in children with ADHD. Time estimation is disrupted in ADHD children, and this disruption is not improved by methylphenidate administration [2]. This is consistent with the current results, which demonstrate a disruption of time estimation accuracy by methylphenidate. Also consistent with the current results, performance of learning tasks in ADHD children is disrupted at higher doses of stimulant medication [39,48]. Furthermore, the changes in progressive ratio endpoints seen in the current study are consistent with reports of changes in appetite in ADHD children treated with D-amphetamine and methylphenidate [7]. It would therefore be interesting to conduct future studies examining the effects of D-amphetamine and methylphenidate on cognitive task performance using a rat model of ADHD.
67
In conclusion, methylphenidate and D-amphetamine produced similar patterns of effects on the performance of four complex operant tasks designed to model cognitive functions in rats. This confirms and extends previous work demonstrating similar behavioral effects of these two agents in both rats and humans. Furthermore, comparison of the current results with published neurochemistry data suggests that the effects of these two agents on complex operant task performance reflect their neurochemical mechanisms of action. In addition, several effects of these agents on task performance reflect what is seen clinically in the treatment of attention-deficit-hyperactivity disorder in children. Thus, the current studies provide evidence that this battery of operant tasks is a valid tool to characterize the effects of drugs, and that the effects of agents in these tasks may be predictive of effects seen in humans. Acknowledgements Arthur J. Mayorga is supported by the Oak Ridge Institute for Science and Education Postgraduate Research Program. Expert statistical advice was provided by Brett Thorn. References [1] Aberman JE, Ward SJ, Salamone JD. Effects of dopamine antagonists and accumbens dopamine depletions on time-constrained progressive ratio performance. Pharmacol Biochem Behav 1998;61:341 – 8. [2] Barkley RA, Koplowitz S, Anderson T, McMurray MB. Sense of time in children with ADHD: effects of duration, distraction, and stimulant medication. J Int Neuropsychol Soc 1997;3:359– 69. [3] Bozdogan H. Model selection and Akaike’s Information Criterion (AIC): the general theory and its analytical extensions. Psychometrika 1987;52:345 – 70. [4] Buffalo EA, Gillam MP, Allen RR, Paule MG. Acute effects of caffeine on several operant behaviors in rhesus monkeys. Pharmacol Biochem Behav 1993;46:733 – 7. [5] Buffalo EA, Gillam MP, Allen RR, Paule MG. Acute behavioral effects of MK-801 in rhesus monkeys: assessment using an operant test battery. Pharmacol Biochem Behav 1994;48:935–40. [6] Eckerman DA, Segbefia D, Manning S, Breese GS. Effects of methylphenidate and D-amphetamine on timing in the rat. Pharmacol Biochem Behav 1987;27:513 – 5. [7] Efron D, Jarman F, Barker M. Side effects of methylphenidate and dexamphetamine in children with attention deficit hyperactivity disorder: a double-blind, crossover trial. Pediatrics 1997;100:662 – 6. [8] Ferguson SA, Paule MG. Acute effects of chlorpromazine in a monkey operant behavioral test battery. Pharmacol Biochem Behav 1992;42:333 – 41. [9] Ferguson SA, Paule MG. Acute effects of pentobarbital in a monkey operant behavioral test battery. Pharmacol Biochem Behav 1993;45:107 – 16. [10] Ferguson SA, Paule MG. Effects of chlorpromazine and diazepam on time estimation behavior and motivation in rats. Pharmacol Biochem Behav 1996;53:115 – 22.
68
A.J. Mayorga et al. / Beha6ioural Brain Research 109 (2000) 59–68
[11] Frederick DL, Schulze GE, Gillam MP, Paule MG. Acute effects of physostigmine on complex operant behavior in rhesus monkeys. Pharmacol Biochem Behav 1995;50:641–8. [12] Frederick DL, Gillam MP, Lensing S, Paule MG. Acute effects of LSD on rhesus monkey operant test battery performance. Pharmacol Biochem Behav 1997;57:633–41. [13] Friedman HR, Goldman-Rakic PS. Activation of the hippocampus and dentate gyrus by working memory: a 2-deoxyglucose study of behaving rhesus monkeys. J Neurosci 1988;8:4693 – 706. [14] Gibson EL, Kennedy AJ, Curzon J. D-Fenfluramine and D-norfenfluramine-induced hypophagia: differential mechanisms and involvement of postsynaptic 5-HT receptors. Eur J Pharmacol 1993;242:83 – 90. [15] Heishman SJ, Henningfield JE. Discriminative stimulus effects of D-amphetamine, methylphenidate, and diazepam in humans. Psychopharmacology 1991;103:436–42. [16] Hinton SC. Functional magnetic resonance imaging and pharmacological investigation of interval timing in humans. Neurotoxicol Teratol 1999;21:202. [17] Hodos W. Progressive ratio as a measure of reward strength. Science 1961;134:943–4. [18] Kohlert JG, Bloch GJ. A rat model for attention-deficit-hyperactivity disorder. Physiol Behav 1993;53:1215–8. [19] Kojima S, Kojima M, Goldman-Rakic PS. Operant behavioral analysis of memory loss in monkeys with prefrontal lesions. Brain Res 1982;248:51–9. [20] Kuczenski R, Segal D. Concomitant characterization of behavioral and striatal neurotransmitter response to amphetamine using in vivo microdialysis. J Neurosci 1989;9:2051–65. [21] Kuczenski R, Segal D. Effects of methylphenidate on extracellular dopamine, serotonin, and norepinephrine: comparison with amphetamine. J Neurochem 1997;68:2032–7. [22] Kuczenski R, Segal DS, Leith NJ, Applegate CD. Effects of amphetamine, methylphenidate, and apomorphine on regional brain serotonin and 5-hydroxyindole acetic acid. Psychopharmacology 1987;93:329 –35. [23] Kuczenski R, Segal DS, Cho AK, Melega W. Hippocampus norepinephrine, caudate dopamine and serotonin, and behavioral responses to the stereoisomers of amphetamine and methamphetamine. J Neurosci 1995;15:1308–17. [24] Leith NJ, Barrett RJ. Self-stimulation and amphetamine: tolerance to D and L isomers and cross tolerance to cocaine and methylphenidate. Psychopharmacology 1981;74:23–8. [25] Maricq AV, Church RM. The differential effects of haloperidol and methamphetamine on time estimation in the rat. Psychopharmacology 1983;79:10–5. [26] Maricq AV, Roberts S, Church RM. Methamphetamine and time estimation. J Exp Psychol Anim Behav Proc 1981;7:18 – 30. [27] Mayorga AJ, Fogle CM, Paule MG. Adaptation of a primate operant test battery to the rat: effects of chlorpromazine. Neurotoxicol Teratol 2000;22:31–9. [28] McClure GYH, Wenger GR, McMillan DE. Effects of drugs on response duration differentiation. V: Differential effects under temporal response differentiation schedules. J Pharmacol Exp Ther 1997;281:1357 –67. [29] McNamara CG, Davidson ES, Schenk S. A comparison of the motor-activating effects of acute and chronic exposure to amphetamine and methylphenidate. Pharmacol Biochem Behav 1993;45:729 – 32.
.
[30] Meck WH. Selective adjustment of the speed of internal clock and memory processes. J Exp Psychol Anim Behav Proc 1983;9:171 – 201. [31] Nielsen JA, Duda NJ, Mokler DJ, Moore KE. Self-administration of central stimulants by rats: a comparison of the effects of D-amphetamine, methylphenidate, and McNeil 4612. Pharmacol Biochem Behav 1984;20:227 – 32. [32] Paule MG. Use of the NCTR operant test battery in non-human primates. Neurotoxicol Teratol 1990;12:413 – 8. [33] Paule MG, Forrester TM, Maher MA, Cranmer JM, Allen RR. Monkey versus human performance in the NCTR operant test battery. Neurotoxicol Teratol 1990;12:503 – 7. [34] Paule MG, Chelonis JJ, Buffalo EA, Blake DJ, Casey PH. Operant test battery performance in children: correlation with IQ. Neurotoxicol Teratol 1999;21:223 – 30. [35] Pechnick R, Janowsky DS, Judd L. Differential effects of methylphenidate and D-amphetamine on stereotyped behavior in the rat. Psychopharmacology 1979;65:311 – 5. [36] Poncelet M, Chermat R, Soubrie P, Simon P. The progressive ratio schedule as a model for studying the psychomotor stimulant activity of drugs in the rat. Psychopharmacology 1983;80:184 – 9. [37] Puumala T, Ruotsalainen S, Jakala P, Koivisto E, Riekkinen P Jr, Sirvio J. Behavioral and pharmacological studies on the validation of a new animal model for attention deficit hyperactivity disorder. Neurobiol Learn Mem 1996;66:198 – 211. [38] Rapp DL, Robbins TW. The effects of D-amphetamine on temporal discrimination in the rat. Psychopharmacology 1976;51:91 – 100. [39] Risser MG, Bowers TG. Cognitive and neuropsychological characteristics of attention deficit hyperactivity disorder children receiving stimulant medications. Percept Mot Skills 1993;77:1023 – 31. [40] Schulze GE, Paule MG. Effects of morphine sulfate on operant behavior in rhesus monkeys. Pharmacol Biochem Behav 1991;38:77 – 83. [41] Schulze GE, McMillan DE, Bailey JR, Scallet AC, Ali SF, Slikker W Jr, et al. Acute effects of delta-9-tetrahydrocannabinol in rhesus monkeys as measured by performance in a battery of cognitive function tests. J Pharmacol Exp Ther 1988;245:178–86. [42] Schulze GE, Slikker W Jr, Paule MG. Multiple behavioral effects of diazepam in rhesus monkeys. Pharmacol Biochem Behav 1989;34:29 – 35. [43] Seiden LS, Andresen J, MacPhail RC. Methylphenidate and D-amphetamine: effects and interactions with alphamethyltyrosine and tetrabenzene on DRL performance in rats. Pharmacol Biochem Behav 1979;10:577 – 84. [44] Skjoldager P, Pierre PJ, Mittleman G. Reinforcer magnitude and progressive ratio responding: effects of increased effort, prefeeding, and extinction. Learn Motiv 1993;24:303 – 43. [45] Solanto MV. Neuropsychopharmacological mechanisms of stimulant drug action in attention-deficit hyperactivity disorder: a review and integration. Behav Brain Res 1998;94:127 –52. [46] Stewart WJ. Progressive reinforcement schedules: a review and evaluation. Aust J Psychol 1974;27:9 – 22. [47] Tannock R, Schachar R. Methylphenidate and cognitive perseveration in hyperactive children. J Child Psychol Psychiatry 1992;33:1217 – 28. [48] Wogar MA, Bradshaw CM, Szabadi E. Does the effect of central 5-hydroxytryptamine depletion on timing depend on motivational change? Psychopharmacology 1993;112:86 – 92.