- Email: [email protected]
Combinations of clozapine and phencyclidine: effects on drug discrimination and behavioral inhibition in rats
Neuropharmacology 40 (2001) 289–297 www.elsevier.com/locate/neuropharm
Combinations of clozapine and phencyclidine: effects on drug discrimination and behavioral inhibition in rats Amelia D. Compton a, Jennifer E. Slemmer b, Michael R. Drew b, James M. Hyman c, Keith M. Golden c, Robert L. Balster c, Jenny L. Wiley c,* a
c
Department of Psychology, Virginia Commonwealth University, Richmond, VA 23284-2018, USA b Department of Psychology, University of Richmond, Richmond, VA, USA Department of Pharmacology and Toxicology, Virginia Commonwealth University, Richmond, VA 23298-0613, USA Received 4 May 2000; received in revised form 5 July 2000; accepted 7 July 2000
Abstract Phencyclidine (PCP) produces psychotomimetic effects in humans that resemble schizophrenia symptoms. In an effort to screen compounds for antipsychotic activity, preclinical researchers have investigated whether these compounds block PCP-induced behaviors in animals. In the present study, the atypical antipsychotic clozapine was tested in combination with an active dose of PCP in two-lever drug discrimination and mixed signalled–unsignalled differential-reinforcement-of-low-rates (DRL) procedures. PCP produced distinctive effects in each task: it substituted for the training dose in PCP discrimination and it increased the number of responses with short (⬍3 s) interresponse times as well as increasing overall response rates in the DRL schedule. Acute dosing with clozapine failed to alter the behavioral effects of PCP in either procedure even when tested up to doses that produced pharmacological effects alone. These results suggest that acute dosing with clozapine would not affect behaviors most closely associated with PCP intoxication. Further, they bring into question the utility of using PCP combination procedures in animals to screen for antipsychotic potential. Since chronic dosing is required for therapeutic efficacy of antipsychotics, future studies should focus on investigation of chronic dosing effects of these drugs in combination with PCP. 2000 Elsevier Science Ltd. All rights reserved. Keywords: Behavioral inhibition; Clozapine; DRL; Drug discrimination; Phencyclidine; Schizophrenia
1. Introduction The primary action of phencyclidine (PCP) is as an open channel blocker at N-methyl-d-aspartate (NMDA) receptors, a sub-class of ionotropic receptors for the excitatory amino acid neurotransmitter glutamate. In humans, PCP produces a subjective state that has been described as “psychotomimetic” (Krystal et al., 1994; Lahti et al., 1995). Although PCP obviously does not produce a syndrome that is identical to psychosis, similarities between this drug-induced state and the symptoms of schizophrenia have led to the hypothesis that glutamate hypoactivity may play a role in the development of schizophrenia (i.e., the PCP model of psychosis;
* Corresponding author. Tel.: +1-804-828-2067; fax: +1-804-8282117. E-mail address: [email protected] (J.L. Wiley).
Javitt and Zukin, 1991). Preclinical investigation has concentrated on the feasibility of using PCP-induced behaviors in animals to model certain aspects of human schizophrenia and on using blockade of these behaviors to screen novel compounds for potential antipsychotic efficacy. Traditionally, pharmacological models of schizophrenia have used acute dosing with dopamine agonists to mimic the presumed dopamine overactivity in schizophrenia; however, in humans the psychotomimetic effects of dopamine agonists, including amphetamine and other psychomotor stimulants, typically are observed only following chronic administration and are likely to resemble the positive symptoms of paranoid schizophrenia (Angrist and Gershon, 1970; Ellinwood and Kilbey, 1977). In contrast, the subjective state produced by PCP-like drugs occurs after acute administration and has elements that resemble both the positive and negative symptoms of schizophrenia (Krystal et al., 1994). Hence, PCP-induced behaviors may represent an
0028-3908/01/$ - see front matter 2000 Elsevier Science Ltd. All rights reserved. PII: S 0 0 2 8 - 3 9 0 8 ( 0 0 ) 0 0 1 2 6 - X
290
A.D. Compton et al. / Neuropharmacology 40 (2001) 289–297
alternative animal model of schizophrenia to dopaminergic agonist produced stereotypy. PCP-like drugs produce a number of behavioral effects that are most likely related to their ability to block the open channel of the NMDA receptor complex, including stimulation of locomotor activity (Ginski and Witkin, 1994), disruption of prepulse inhibition of the acoustic startle response (Mansbach and Geyer, 1989), discriminative stimulus effects (Balster and Willetts, 1988), and disruption of operant responding (Sanger and Jackson, 1989). In order to determine whether any of these PCP-induced behaviors are sensitive to attenuation by drugs used to treat schizophrenia, PCP has been tested in combination with typical and atypical antipsychotics in most of these procedures. [For the purposes of this paper, atypical antipsychotics (e.g., clozapine) are distinguished from typical antipsychotics (e.g., haloperidol) in that the former, but not the latter, lack high liability for producing extrapyramidal motor effects at therapeutic doses.] Whereas both typical and atypical antipsychotics attenuate locomotor activation induced by PCP-like NMDA open channel blockers (Hoffman, 1992), only atypical antipsychotics have been found to block their discriminative stimulus effects (Corbett, 1995) and their disruption of prepulse inhibition of acoustic startle (Bakshi and Geyer, 1995; Bakshi et al., 1994; Swerdlow et al., 1996), although other studies have reported contradictory results in these procedures (Hoffman et al., 1993; Smith et al., 1999; Wiley, 1994). Nevertheless, these findings lend support to the idea that behaviors induced by PCP-like drugs may be a useful model for differentiating novel atypical from typical antipsychotic drugs. In the present study, we have examined the ability of clozapine to alter discriminative stimulus effects of PCP and its disruptive effects on operant responding under a differential-reinforcement-of-low-rates (DRL) schedule. Drug discrimination represents an animal model of the subjective effects of psychoactive drugs in humans (Balster, 1990). It is hypothesized that the subjective effects modelled by PCP discrimination may include its psychotomimetic effects since there is a high correspondence between drugs with PCP-like discriminative stimulus effects and the propensity of these drugs to produce behaviors resembling psychosis in humans (Balster and Willetts, 1996). DRL schedules have been used to model behavioral inhibition, the ability to delay or inhibit a response. Inappropriate or impulsive behavior observed in certain psychiatric disorders or following ingestion of drugs of abuse may represent, in part, a failure of behavioral inhibition. Sanger (1992) reported that among the effects of PCP-like drugs under this type of schedule is an increase in the number of burst responses (i.e., responses with very short inter-response times). In order to determine whether increased stimulus control during signalled components would offset the disruption
that had previously been observed with PCP-like drugs, a signalled–unsignalled DRL schedule was used in the present study. These experiments were designed to determine whether acute doses of the atypical antipsychotic clozapine would block the discriminative stimulus effects of PCP or its disruption of operant responding under a DRL schedule. Previous reports of these experiments were published in abstract form (Compton et al., 1997; Slemmer et al., 1997). 2. Methods 2.1. Animals Adult male Sprague–Dawley (295–365 g) and Long– Evans hooded rats (285–350 g), obtained from Harlan (Dublin, VA), were used for the discrimination and operant behavior studies, respectively. They were maintained at indicated body weights by restriction of daily access to standard rodent chow. When sessions were not being conducted, the rats were individually housed in wire suspension cages in a temperature-controlled (20–22°C) vivarium environment with a 12-h light–dark cycle (lights on at 7 a.m.). Water was freely available in the home cages. Rats in each group were drug-naive at the beginning of the study. 2.2. Apparatus Rats in each procedure were trained and tested in standard operant conditioning chambers (BRS/LVE Inc., Laurel, MD and Lafayette Instruments Co., Lafayette, IN) housed in sound-attenuated cubicles. Pellet dispensers delivered 45-mg BIO SERV (Frenchtown, NJ) food pellets to a food cup on the front wall of the chamber beside a single response lever (DRL procedure) or between two response levers (discrimination procedure). Fan motors provided ventilation and masking noise for each chamber. House lights located above the food cup were illuminated during training and testing sessions for both procedures. In the chambers used for the DRL experiment, stimulus lights were located over the lever and in a corresponding position on the other side of the food cup. Sked-11 software (State Systems II Inc., Kalamazoo, MI) installed on a MicroPDP-11BA23 computer (Digital Equipment Corp., Maynard, MA) was used to control schedule contingencies and to record data for the DRL experiment. A micro-computer with Logic ‘1’ interface (MED Associates, Georgia, Vermont) and MED-PC software (MED Associates) was used for these tasks in the drug discrimination experiment. 2.3. Drugs PCP HCl (National Institute on Drug Abuse, Rockville, MD) was dissolved in saline. A stock solution of
A.D. Compton et al. / Neuropharmacology 40 (2001) 289–297
20 mg/ml clozapine (Research Biochemicals International, Natick, MA) was mixed in sterile water with 5–10 drops of lactic acid. Lower concentrations were obtained by dilution of this stock. All injections were administered intraperitoneally at a volume of 1 ml/kg. 2.4. Drug discrimination procedure Rats were trained to press one lever following administration of 2 mg/kg PCP and to press another lever after injection with saline, each according to a fixed-ratio 20 schedule of food reinforcement. Completion of 20 consecutive responses on the injection-appropriate lever resulted in delivery of a food pellet. Each response on the incorrect lever reset the ratio requirement on the correct lever. The daily injections for each rat were administered in a double alternation sequence of 2 mg/kg PCP and saline. Rats were injected and returned to their home cages until the start of the experimental session 15 min later. Training occurred during sessions conducted five days a week (Monday–Friday) until the rats had met three criteria during eight of ten consecutive sessions: (1) first completed fixed ratio 20 on the correct lever; (2) percentage of correct-lever responding ⬎80%; and (3) response rate ⬎0.4 responses/s. Following successful acquisition of the discrimination, substitution and antagonism tests were conducted on Tuesdays and Fridays during 15-min test sessions. Training continued on Mondays, Wednesdays, and Thursdays. During test sessions, responses on either lever delivered reinforcement according to a fixed ratio 20 schedule. In order to be tested, rats must have completed the first FR and made at least 80% of all responses on the injection-appropriate lever on the preceding day’s training session. In addition, the rat must have met these same criteria during at least one of the training sessions with the alternate training compound (PCP or saline) earlier in the week. A dose–effect curve for PCP was determined first. Then, rats were tested with different doses of clozapine (0–5.6 mg/kg, i.p.) in combination with the 2 mg/kg training dose of PCP. Clozapine or its vehicle was administered 60 min before the start of the session; PCP or saline was injected 40 min later (i.e., 10 min presession). Doses of both compounds were administered in ascending order. Before the combination dose–effect curve determination, control tests with combinations of vehicle and saline and vehicle and 2 mg/kg PCP were conducted. 2.5. Mixed signalled–unsignalled DRL procedure In this experiment, rats initially were trained to leverpress for food reinforcement. Once the lever-press response was acquired, a mixed DRL schedule was instituted. In order to obtain a reinforcer under the DRL
291
schedule, the rat had to wait a specified number of seconds between lever presses. Premature responses reset the DRL clock. During the first component of the mixed DRL schedule (signalled component), each reinforced response resulted in the onset of the stimulus lights on either side of the food cup and the end of the delay was signalled by the offset of these lights. During the second component (unsignalled component), the stimulus lights were not illuminated; hence, there was no exteroceptive signal at the end of the delay. The house light remained illuminated during both types of components. Over the course of training, the length of the delay was increased to 15 s for each component. Under the final schedule, the signalled and unsignalled components alternated for a total of six components. Each component lasted 8 min with a 1-min time out separating components for a total session length of 54 min. During the time out, the house and stimulus lights were extinguished and lever presses did not have programmed consequences. Testing began when responding was stable across sessions and components. A dose–effect curve for clozapine alone (0–10 mg/kg) was determined first. Then, rats were tested with the same doses of clozapine in combination with 2 mg/kg PCP. Clozapine or its vehicle was administered 60 min before the start of the session; PCP or saline was injected 45 min later (i.e., 15 min pre-session). For each dose– effect curve determination, doses or dose combinations were administered in randomized order according to a Latin Square design. 2.6. Data analysis For each drug discrimination test session, percentage of responses on the PCP lever and response rate (responses/s) were calculated. The ED50 (with 95% confidence intervals) for PCP was calculated using leastsquares linear regression on the linear part of the dose– effect curves (Tallarida and Murray, 1987) for percentage of drug-lever responding, plotted against log10 transformation of the dose. Since rats that responded less than 20 times during a test session did not press either lever a sufficient number of times to earn a reinforcer, their lever selection data were excluded from data analysis. For data analysis of sessions in the DRL experiment, responses during the three signalled components were analyzed together as were responses during the three unsignalled components. For each type of component, responses that occurred within 30 s of the last response were recorded separately in ten 3-s bins. Responses with interresponse times (IRT) ⬎30 s were recorded in an eleventh bin for each component. The temporal characteristics of the rats’ response patterns were analyzed with a method adapted from Richards and Seiden (1991). Responses with IRTs of 3 s or less were termed “burst responses” and were analyzed separately for signalled
292
A.D. Compton et al. / Neuropharmacology 40 (2001) 289–297
and unsignalled components with two-factor (PCP vs saline and clozapine dose) repeated measures ANOVAs. The mean of the total number of responses (burst plus non-burst responses) and the mean number of reinforcers for each treatment condition were also analyzed separately for signalled and unsignalled components with two-factor repeated measures ANOVAs. Differences between means revealed by significant ANOVAs were further compared through the use of Tukey post hoc tests (a=0.05). In addition, the median IRT of non-burst responses for each treatment condition was calculated for signalled and unsignalled components based on the distribution of IRTs of non-burst responses for each animal. Means of these median IRTs were subjected to separate two-way repeated measures ANOVAs for signalled and unsignalled components. This peak value of IRTs is known to be sensitive to drug effects under simple DRL schedules (e.g., Sanger, 1992). Since the meaningfulness of the median IRT of non-burst responses is of dubious value in instances where the total number of responses is very low, this measure was omitted from analysis in cases where a rat earned less than 10 reinforcers during signalled or unsignalled components over the entire session for a particular dose or dose combination. Finally, the mean number of responses during time-out periods throughout the session was calculated for each treatment condition and analyzed with repeated measures ANOVA.
3. Results 3.1. Effects of clozapine on PCP discrimination Phencyclidine produced full dose-dependent substitution for the PCP training dose (Fig. 1, upper panel), indicating that rats had successfully acquired the discrimination. The ED50 for PCP substitution was 0.94 mg/kg (95% CL: 0.62–1.42 mg/kg). Clozapine (0.3–5.6 mg/kg) did not block substitution of 2 mg/kg PCP (Fig. 1, upper panel), although the 3 and 5.6 mg/kg doses of clozapine in combination with 2 mg/kg PCP decreased overall response rates compared to vehicle control test results (Fig. 1, lower panel). 3.2. Effects of PCP alone and in combination with clozapine in DRL procedure The main differences between baseline performance in the signalled and unsignalled components were that subjects were more accurate and earned more reinforcers in the signalled components and the distribution of responses showed much less variability (i.e., more peaked distribution) in these components. Because of these differences in baseline values across type of component, drug effects were analyzed and are presented
Fig. 1. Percentage of PCP-lever responding (top panel) and response rate (bottom panel) in rats trained to discriminate 2 mg/kg PCP from saline. Before the start of the session, rats were injected with PCP alone (䊐) or with clozapine followed by an injection of 2 mg/kg dose of PCP (䊏). Points above VEH represent the results of a control test with saline. The points above the 0 dose represent a test with the clozapine vehicle and 2 mg/kg PCP. For both of the dose–effect curve determinations, each value represents the mean (±SEM) for 6–8 rats, except for percentage of PCP-lever responding for the 2 mg/kg PCP plus 5.6 mg/kg clozapine combination (n=3).
separately for signalled vs. unsignalled components. The profile of main effects of PCP (2 mg/kg) was identical across both types of components (filled symbols above vehicle): PCP significantly increased the number of burst responses as well as overall rates of responding (Figs. 2 and 3, respectively), but did not affect the number of reinforcers received (Fig. 4). PCP also increased response rate during time-out periods (data not shown). Further, 2 mg/kg of PCP shifted the mean of the distributions of median IRTs of non-burst responses to the left (shorter IRTs) in the unsignalled component and produced a trend (p=0.07) toward a leftward shift in the signalled component (Fig. 5 and Table 1). Clozapine at doses up to 10 mg/kg did not affect this shift (Fig. 5 and Table 1). Clozapine administered alone produced significant
A.D. Compton et al. / Neuropharmacology 40 (2001) 289–297
Fig. 2. Effects of clozapine alone (䊐) and clozapine plus 2 mg/kg PCP (䊏) on the number of burst responses (consecutive responses with IRT⬍3 s) during signalled (top panel) and unsignalled (bottom panel) components of a mixed signalled–unsignalled DRL procedure. Each value represents the mean (±SEM) for 12 rats. # indicates a main effect for PCP or for clozapine dose.
effects in the DRL procedure. The 10 mg/kg dose of clozapine decreased overall rates of responding (Fig. 3) and the 5 and 10 mg/kg doses of clozapine reduced the number of reinforcers earned in both components (Fig. 4). In contrast, none of the doses of clozapine alone affected the number of burst responses (Fig. 2). Because PCP alone increased the number of non-burst responses (an effect opposite of that of the high dose of clozapine), the effects of clozapine were less pronounced when rats also received PCP than when they received only clozapine. Although the effects of a combination of PCP and clozapine significantly decreased response and reinforcer rates compared to the effects of PCP alone, these effects occurred only at doses where clozapine alone produced similar decreases, suggesting that clozapine did not specifically alter the effects of PCP. Clozapine alone did not affect mean of the median IRT distributions (Fig. 5 and Table 1).
293
Fig. 3. Effects of clozapine alone (䊐) and clozapine plus 2 mg/kg PCP (䊏) on response rate (total number of responses/min) during signalled (top panel) and unsignalled (bottom panel) components of a mixed signalled–unsignalled DRL procedure. Each value represents the mean (±SEM) for 12 rats. # indicates a main effect for PCP or for clozapine dose. * indicates that the interaction was significant and that the value for the dose or dose combination is significantly different from the corresponding vehicle or PCP+vehicle dose. @ indicates that the interaction was significant and that the value for the PCP+clozapine dose is significantly different from the corresponding clozapine alone dose.
4. Discussion Using a discrete trial shock-avoidance discrimination procedure, Corbett (1995) found that acute doses of the prototypic atypical antipsychotic clozapine, but not the typical antipsychotic haloperidol, antagonized the discriminative stimulus effects of the NMDA open channel blocker dizocilpine. In contrast, other researchers (present study; Smith et al., 1999) have reported that clozapine did not block the discriminative stimulus effects of PCP-like drugs in drug discrimination procedures that used food reinforcement to motivate responding. Although the reasons for this disparity are not entirely clear, it is likely that they result from differences in the type of reinforcement or other experimental
294
A.D. Compton et al. / Neuropharmacology 40 (2001) 289–297
Fig. 4. Effects of clozapine alone (䊐) and clozapine plus 2 mg/kg PCP (䊏) on the number of reinforcers earned during signalled (top panel) and unsignalled (bottom panel) components of a mixed signalled–unsignalled DRL procedure. Each value represents the mean (±SEM) for 12 rats. # indicates a main effect for PCP or for clozapine dose. * indicates that the interaction was significant and that the value for the dose or dose combination is significantly different from the corresponding vehicle or PCP+vehicle dose. @ indicates that the interaction was significant and that the value for the PCP+clozapine dose is significantly different from the corresponding clozapine alone dose.
parameters. For example, schedule of reinforcement and training dose can influence results in PCP discrimination studies (Mansbach and Balster, 1991; McMillan and Wenger, 1984). While it is possible that clozapine might have blocked a lower training dose of PCP, it is unlikely. Clozapine produced even less attenuation of the discriminative stimulus effects of PCP under conditions of food reinforcement than did the typical antipsychotic haloperidol against even larger training doses of PCP (Beardsley and Balster, 1988; Browne, 1986). Haloperidol consistently has been found to partially attenuate the discriminative stimulus effects of PCP and PCP-like drugs, but usually only at doses that also severely suppress response rates (Wiley, 1995). In the present study, clozapine did not attenuate the discriminative stimulus effects of PCP, although substantial clozapine-induced response-rate decreases were observed. Considered together, these results suggest that clozapine would not
be any more effective as a treatment for acute PCP intoxication than are typical antipsychotics. Further, these results suggest that PCP discrimination, at least one using food reinforcement, does not provide a reliable preclinical method of distinguishing novel atypical antipsychotics from traditional neuroleptics. Clozapine also did not block the effects of a 2 mg/kg dose of PCP on responding under the mixed signalled– unsignalled DRL schedule. The main effect of PCP was to increase the number of responses that did not result in reinforcement. Notably, PCP increased the number of burst and time-out responses. Given that DRL schedules generally engender response rates that are greatly reduced compared to those under ratio schedules, these results may reflect the rate-dependent effects of PCP, an interpretation that is supported by the fact that PCP increased overall rates of responding. With a DRL schedule, a drug with rate-dependent effects would increase burst responses because of greater opportunity for responses with shorter IRTs during the fixed session length. In the present study, these effects occurred during both types of components, suggesting that the increased stimulus control present during signalled components was not enough to overcome the disruption of behavioral inhibition produced by PCP. Further, these results are consistent with those of previous researchers who have reported that PCP-like and other NMDA antagonists disrupt responding in unsignalled DRL procedures (Sanger, 1992; Stephens and Cole, 1996). As mentioned previously, clozapine did not alter any of the main effects of PCP, even at doses where clozapine began to produce its own effects. Hence, clozapine failed to affect PCPinduced behavior in either of the two operant procedures included in this study. In contrast, the locomotor activating effects of PCP and its disruption of prepulse inhibition of acoustic startle are reported to be attenuated by clozapine (Bakshi et al., 1994; Hoffman, 1992), although these results have not been consistently replicated across all studies (Hoffman et al., 1993; Wiley, 1994). These findings suggest that, while clozapine may attenuate the effects of PCP on spontaneous or reflexive behavior, it may not alter the effects of PCP on more complicated, learned behaviors such as drug discrimination and schedule-controlled behavior. The results of tests with clozapine alone are also of interest. In previous studies, acute dosing with clozapine had been shown to disrupt the pattern of operant responding under a variety of schedule conditions, including fixed interval (Kaempf and Porter, 1987), fixed consecutive number (Picker, 1988), and unsignalled DRL (Bruhwyler et al., 1993). In addition, Compton and Porter (1995) reported that clozapine dose-dependently disrupted matching behavior in a multiple variable interval schedule. Many of the rats in the Compton and Porter study failed to alter their response pattern to match signalled changes in reinforcement contingencies (i.e., cloz-
A.D. Compton et al. / Neuropharmacology 40 (2001) 289–297
295
Fig. 5. Effects of clozapine alone (left panels) and clozapine plus 2 mg/kg PCP (right panels) on the distributions of IRTs during signalled (top panels) and unsignalled (bottom panels) components of a mixed signalled–unsignalled DRL procedure. Each value represents the mean for 12 rats.
Table 1 Effects of clozapine alone and clozapine+2 mg/kg PCP on median IRT in a mixed DRL 15-s schedule Drug
Clozapine
Clozapine dose Mean (±SEM) (mg/kg) of signalled median IRTs 0 1.25 2.5 5
Clozapine+2 mg/kg PCPa
10 0 1.25 2.5 5 10
a
Mean (±SEM) of unsignalled median IRTs
17.0 (0.34), 17.2 (0.54), n=12 n=12 16.7 (0.45), 16.0 (0.81), n=12 n=12 15.7 (0.39), 14.5 (0.43), n=12 n=12 16.2 (0.63), 15.7 (1.08), n=11 n=11 13.5 (3.00), n=3 not determined 16.0 (0.34), 15.2 (0.69), n=12 n=12 15.0 (0.69), 14.2 (0.84), n=12 n=12 15.7 (0.65), 14.0 (1.03), n=12 n=12 14.7 (0.78), 13.0 (0.81), n=12 n=12 14.8 (0.88), n=9 13.5 (1.00), n=9
The main effect of PCP was to decrease the mean of the median IRTs. This effect was significant for the unsignalled components (p=0.03), but was only a trend (p=0.07) for the signalled components. Clozapine did not produce a main effect for either component and there were no interactions.
apine produced response rates across the schedule that were similar despite changes in reinforcement availability in the different components). In the matching study, all changes in reinforcement availability were signalled which presented difficulty in determining whether failure to match response rates with reinforcement density resulted from drug effects on the timing of responses or on stimulus control. In the present study, use of both signalled and unsignalled components in the same schedule afforded the ability to determine whether inappropriate responding resulted from disruption of the animal’s timing capacity or from the inability to use external stimuli to regulate behavior, or both. In this mixed signalled– unsignalled DRL schedule, baseline rates of overall responding during vehicle tests were lower in the unsignalled than in the signalled, components, resulting in responses that were spread over a larger number of IRT bins (i.e., a flatter IRT frequency distribution). In contrast, the IRT frequency distribution showed a more pronounced peak during signalled components after vehicle administration. As expected, more reinforcers were earned during signalled components; hence, responding was more efficient. The main effect of clozapine in this procedure was a dose-dependent decrease in the total number of responses. Visual inspection reveals that this rate-decreasing effect resulted in a flattening of the IRT distributions at higher clozapine doses during both components (see Fig. 5). Mean IRTs were not significantly affected by clozapine. Hence, rather than simply
296
A.D. Compton et al. / Neuropharmacology 40 (2001) 289–297
shifting the rats’ distribution of IRTs toward shorter latencies, clozapine instead dose-dependently caused responding to become more randomly distributed in time in both types of tasks (i.e., flattening of the IRT curves with increasing dose vs. shifts in the location of the peak IRT). This finding suggests that clozapine affects brain systems involved both with temporal ordering and with stimulus control, although the effect was less pronounced under conditions of high stimulus control in the signalled components. In summary, PCP produces distinctive effects in drug discrimination and mixed signalled–unsignalled DRL procedures. The atypical antipsychotic clozapine failed to meaningfully alter the effects of PCP in either of these tests. Hence, although clozapine has been shown to attenuate the behavioral effects of PCP in a number of procedures (e.g., startle and locomotor), it does not alter all of the effects of PCP. Importantly, clozapine does not consistently attenuate the behaviors that may be most closely associated with PCP intoxication; i.e., its discriminative stimulus effects and its disruption of operant behavior. These results bring into question the utility of distinguishing atypical from typical antipsychotics using PCP combination procedures. Further, clozapine produces distinct effects of its own on the timing of responses and on stimulus control. While these results suggest that acute dosing with clozapine may not offer an effective antidote to the consequences of PCP abuse, further studies need to be done in order to discover whether chronic dosing with clozapine might represent an effective treatment for the psychotomimetic effects of PCP. In human schizophrenia patients, repeated oral dosing with clozapine is necessary for clinical effectiveness. Future research should concentrate on determination of the effects of repeated dosing with clozapine on the behavioral effects of PCP.
Acknowledgements Research supported by National Institute on Drug Abuse grant DA-01442.
References Angrist, B., Gershon, S., 1970. The phenomenology of experimentally induced amphetamine psychosis: preliminary observations. Biological Psychiatry 2, 95–107. Bakshi, V.P., Geyer, M.A., 1995. Antagonism of phencyclidineinduced deficits in prepulse inhibition by the putative atypical antipsychotic olanzapine. Psychopharmacology 122, 198–201. Bakshi, V.P., Swerdlow, N.R., Geyer, M.A., 1994. Clozapine antagonizes phencyclidine-induced deficits in sensorimotor gating of the startle response. Journal of Pharmacology and Experimental Therapeutics 271, 787–794. Balster, R.L., 1990. Perception of drug effects. In: Berkley, M.A.,
Stebbins, W.C. (Eds.), Comparative Perception. Wiley, New York, pp. 127–154. Balster, R.L., Willetts, J., 1988. Receptor mediation of the discriminative stimulus properties of phencyclidine and sigma-opioid agonists. In: Colpaert, F.C., Balster, R.L. (Eds.), Transduction Mechanisms of Drug Stimuli. Springer, Berlin, pp. 122–135. Balster, R.L., Willetts, J., 1996. Phencyclidine: A drug of abuse and a tool for neuroscience research. In: Schuster, C.R., Kuhar, M.J. (Eds.), Pharmacological Aspects of Drug Dependence: Towards an Integrated Neurobehavioral Approach. Springer, Berlin, pp. 233– 262. Beardsley, P.M., Balster, R.L., 1988. Evaluation of antagonists of the discriminative stimulus and response rate effects of phencyclidine. Journal of Pharmacology and Experimental Therapeutics 244, 34–40. Browne, R., 1986. Discriminative stimulus properties of PCP mimetics. In: Clouet, D.H. (Ed.), Phencyclidine: An Update. National Institute on Drug Abuse, Rockville, MD, pp. 134–147. Bruhwyler, J., Chleide, E., Houbeau, G., Waegeneer, N., Mercier, M., 1993. Differentiation of haloperidol and clozapine using a complex operant schedule in the dog. Pharmacology Biochemistry Behavior 44, 181–189. Compton, A.D., Porter, J.H., 1995. Effects of clozapine and pimozide on the parameters of Herrnstein’s matching equation applied to multiple operant schedule responding in rats. Society for Neuroscience Abstracts 21, 467. Compton, A.D., Drew, M.R., Golden, K.M., Wiley, J.L., 1997. Clozapine impairs temporal perception and stimulus control in rats. Schizophrenia Research 24, 75. Corbett, R., 1995. Clozapine but not haloperidol antagonizes an MK801 discriminative stimulus cue. Pharmacology Biochemistry Behavior 51, 561–564. Ellinwood, E.H. Jr, Kilbey, M.M., 1977. Chronic stimulant intoxication models of psychosis. In: Hanin, I., Usdin, E. (Eds.), Animal Models in Psychiatry and Neurology. Pergamon Press, Oxford, pp. 61–74. Ginski, M.J., Witkin, J.M., 1994. Sensitive and rapid behavioral differentiation of N-methyl-d-aspartate receptor antagonists. Psychopharmacology 114, 573–582. Hoffman, D.C., 1992. Typical and atypical neuroleptics antagonize MK-801-induced locomotion and stereotypy in rats. Journal of Neural Transmission 89, 1–10. Hoffman, D.C., Donovan, H., Cassella, J.V., 1993. The effects of haloperidol and clozapine on the disruption of sensorimotor gating induced by the noncompetitive glutamate antagonist MK-801. Psychopharmacology 111, 339–344. Javitt, D.C., Zukin, S.R., 1991. Recent advances in the phencyclidine model of schizophrenia. American Journal of Psychiatry 148, 1301–1308. Kaempf, G.L., Porter, J.H., 1987. Differential effects of pimozide and clozapine on schedule-controlled and schedule-induced behaviors after acute and chronic administration. Journal of Pharmacology and Experimental Therapeutics 243, 437–445. Krystal, J.H., Karper, L.P., Seibyl, J.P., Freeman, G.K., Delaney, R., Bremner, J.D., Heninger, G.R., Bowers, M.B. Jr., Charney, D.S., 1994. Subanesthetic effects of the noncompetitive NMDA antagonist, ketamine, in humans. Archives of General Psychiatry 51, 199–214. Lahti, A.C., Koffel, B., LaPorte, D., Tamminga, C.A., 1995. Subanesthetic doses of ketamine stimulate psychosis in schizophrenia. Neuropsychopharmacology 13, 9–19. Mansbach, R.S., Balster, R.L., 1991. Pharmacological specificity of the phencyclidine discriminative stimulus in rats. Pharmacology Biochemistry Behavior 39, 971–975. Mansbach, R.S., Geyer, M.A., 1989. Effects of phencyclidine and phencyclidine biologs on sensorimotor gating in the rat. Neuropsychopharmacology 2, 299–308.
A.D. Compton et al. / Neuropharmacology 40 (2001) 289–297
McMillan, D.E., Wenger, G.R., 1984. Bias of phencyclidine discrimination by the schedule of reinforcement. Journal of Experimental Analysis of Behavior 42, 51–66. Picker, M., 1988. Effects of clozapine on fixed-consecutive-number responding in rats: a comparison to other neuroleptic drugs. Pharmacology Biochemistry Behavior 30, 603–612. Richards, J.B., Seiden, L.S., 1991. A quantitative interresponse–time analysis of DRL performance differentiates similar effects of the antidepressant desipramine and the novel anxiolytic gepirone. Journal of Experimental Analysis of Behavior 56, 173–192. Sanger, D.J., 1992. NMDA antagonists disrupt timing behaviour in rats. Behavioural Pharmacology 3, 593–600. Sanger, D.J., Jackson, A., 1989. Effects of phencyclidine and other Nmethyl-d-aspartate antagonists on the schedule-controlled behavior of rat. Journal of Pharmacology and Experimental Therapeutics 248, 1215–1221. Slemmer, J.E., Hyman, J.M., Wiley, J.L., Compton, A.D., Balster, R.L., 1997. Clozapine fails to attenuate phencyclidine’s discriminative stimulus effects in rats. Virginia Journal of Science 48, 145.
297
Smith, J.A., Boyer-Millar, C., Goudie, A.J., 1999. Does MK-801 discrimination constitute an animal model of schizophrenia useful for detecting atypical antipsychotics? Pharmacology Biochemistry Behavior 64, 429–433. Stephens, D.N., Cole, B.J., 1996. AMPA antagonists differ from NMDA antagonists in their effects on operant DRL and delayed matching to position tasks. Psychopharmacology 126, 249–259. Swerdlow, N.R., Bakshi, V., Geyer, M.A., 1996. Seroquel restores sensorimotor gating in phencyclidine-treated rats. Journal of Pharmacology and Experimental Therapeutics 279, 1290–1299. Tallarida, R.J., Murray, R.B., 1987. Manual of Pharmcologic Calculations with Computer Programs, 2nd ed. Springer, New York. Wiley, J.L., 1994. Clozapine’s effects on phencyclidine-induced disruption of prepulse inhibition of the acoustic startle response. Pharmacology Biochemistry Behavior 49, 1025–1028. Wiley, J.L., 1995. Effect of repeated haloperidol administration on phencyclidine discrimination in rats. Progress in Neuro-Psychopharmacology and Biological Psychiatry 19, 699–711.
Combinations of clozapine and phencyclidine: effects on drug discrimination and behavioral inhibition in rats Amelia D. Compton a, Jennifer E. Slemmer b, Michael R. Drew b, James M. Hyman c, Keith M. Golden c, Robert L. Balster c, Jenny L. Wiley c,* a
c
Department of Psychology, Virginia Commonwealth University, Richmond, VA 23284-2018, USA b Department of Psychology, University of Richmond, Richmond, VA, USA Department of Pharmacology and Toxicology, Virginia Commonwealth University, Richmond, VA 23298-0613, USA Received 4 May 2000; received in revised form 5 July 2000; accepted 7 July 2000
Abstract Phencyclidine (PCP) produces psychotomimetic effects in humans that resemble schizophrenia symptoms. In an effort to screen compounds for antipsychotic activity, preclinical researchers have investigated whether these compounds block PCP-induced behaviors in animals. In the present study, the atypical antipsychotic clozapine was tested in combination with an active dose of PCP in two-lever drug discrimination and mixed signalled–unsignalled differential-reinforcement-of-low-rates (DRL) procedures. PCP produced distinctive effects in each task: it substituted for the training dose in PCP discrimination and it increased the number of responses with short (⬍3 s) interresponse times as well as increasing overall response rates in the DRL schedule. Acute dosing with clozapine failed to alter the behavioral effects of PCP in either procedure even when tested up to doses that produced pharmacological effects alone. These results suggest that acute dosing with clozapine would not affect behaviors most closely associated with PCP intoxication. Further, they bring into question the utility of using PCP combination procedures in animals to screen for antipsychotic potential. Since chronic dosing is required for therapeutic efficacy of antipsychotics, future studies should focus on investigation of chronic dosing effects of these drugs in combination with PCP. 2000 Elsevier Science Ltd. All rights reserved. Keywords: Behavioral inhibition; Clozapine; DRL; Drug discrimination; Phencyclidine; Schizophrenia
1. Introduction The primary action of phencyclidine (PCP) is as an open channel blocker at N-methyl-d-aspartate (NMDA) receptors, a sub-class of ionotropic receptors for the excitatory amino acid neurotransmitter glutamate. In humans, PCP produces a subjective state that has been described as “psychotomimetic” (Krystal et al., 1994; Lahti et al., 1995). Although PCP obviously does not produce a syndrome that is identical to psychosis, similarities between this drug-induced state and the symptoms of schizophrenia have led to the hypothesis that glutamate hypoactivity may play a role in the development of schizophrenia (i.e., the PCP model of psychosis;
* Corresponding author. Tel.: +1-804-828-2067; fax: +1-804-8282117. E-mail address: [email protected] (J.L. Wiley).
Javitt and Zukin, 1991). Preclinical investigation has concentrated on the feasibility of using PCP-induced behaviors in animals to model certain aspects of human schizophrenia and on using blockade of these behaviors to screen novel compounds for potential antipsychotic efficacy. Traditionally, pharmacological models of schizophrenia have used acute dosing with dopamine agonists to mimic the presumed dopamine overactivity in schizophrenia; however, in humans the psychotomimetic effects of dopamine agonists, including amphetamine and other psychomotor stimulants, typically are observed only following chronic administration and are likely to resemble the positive symptoms of paranoid schizophrenia (Angrist and Gershon, 1970; Ellinwood and Kilbey, 1977). In contrast, the subjective state produced by PCP-like drugs occurs after acute administration and has elements that resemble both the positive and negative symptoms of schizophrenia (Krystal et al., 1994). Hence, PCP-induced behaviors may represent an
0028-3908/01/$ - see front matter 2000 Elsevier Science Ltd. All rights reserved. PII: S 0 0 2 8 - 3 9 0 8 ( 0 0 ) 0 0 1 2 6 - X
290
A.D. Compton et al. / Neuropharmacology 40 (2001) 289–297
alternative animal model of schizophrenia to dopaminergic agonist produced stereotypy. PCP-like drugs produce a number of behavioral effects that are most likely related to their ability to block the open channel of the NMDA receptor complex, including stimulation of locomotor activity (Ginski and Witkin, 1994), disruption of prepulse inhibition of the acoustic startle response (Mansbach and Geyer, 1989), discriminative stimulus effects (Balster and Willetts, 1988), and disruption of operant responding (Sanger and Jackson, 1989). In order to determine whether any of these PCP-induced behaviors are sensitive to attenuation by drugs used to treat schizophrenia, PCP has been tested in combination with typical and atypical antipsychotics in most of these procedures. [For the purposes of this paper, atypical antipsychotics (e.g., clozapine) are distinguished from typical antipsychotics (e.g., haloperidol) in that the former, but not the latter, lack high liability for producing extrapyramidal motor effects at therapeutic doses.] Whereas both typical and atypical antipsychotics attenuate locomotor activation induced by PCP-like NMDA open channel blockers (Hoffman, 1992), only atypical antipsychotics have been found to block their discriminative stimulus effects (Corbett, 1995) and their disruption of prepulse inhibition of acoustic startle (Bakshi and Geyer, 1995; Bakshi et al., 1994; Swerdlow et al., 1996), although other studies have reported contradictory results in these procedures (Hoffman et al., 1993; Smith et al., 1999; Wiley, 1994). Nevertheless, these findings lend support to the idea that behaviors induced by PCP-like drugs may be a useful model for differentiating novel atypical from typical antipsychotic drugs. In the present study, we have examined the ability of clozapine to alter discriminative stimulus effects of PCP and its disruptive effects on operant responding under a differential-reinforcement-of-low-rates (DRL) schedule. Drug discrimination represents an animal model of the subjective effects of psychoactive drugs in humans (Balster, 1990). It is hypothesized that the subjective effects modelled by PCP discrimination may include its psychotomimetic effects since there is a high correspondence between drugs with PCP-like discriminative stimulus effects and the propensity of these drugs to produce behaviors resembling psychosis in humans (Balster and Willetts, 1996). DRL schedules have been used to model behavioral inhibition, the ability to delay or inhibit a response. Inappropriate or impulsive behavior observed in certain psychiatric disorders or following ingestion of drugs of abuse may represent, in part, a failure of behavioral inhibition. Sanger (1992) reported that among the effects of PCP-like drugs under this type of schedule is an increase in the number of burst responses (i.e., responses with very short inter-response times). In order to determine whether increased stimulus control during signalled components would offset the disruption
that had previously been observed with PCP-like drugs, a signalled–unsignalled DRL schedule was used in the present study. These experiments were designed to determine whether acute doses of the atypical antipsychotic clozapine would block the discriminative stimulus effects of PCP or its disruption of operant responding under a DRL schedule. Previous reports of these experiments were published in abstract form (Compton et al., 1997; Slemmer et al., 1997). 2. Methods 2.1. Animals Adult male Sprague–Dawley (295–365 g) and Long– Evans hooded rats (285–350 g), obtained from Harlan (Dublin, VA), were used for the discrimination and operant behavior studies, respectively. They were maintained at indicated body weights by restriction of daily access to standard rodent chow. When sessions were not being conducted, the rats were individually housed in wire suspension cages in a temperature-controlled (20–22°C) vivarium environment with a 12-h light–dark cycle (lights on at 7 a.m.). Water was freely available in the home cages. Rats in each group were drug-naive at the beginning of the study. 2.2. Apparatus Rats in each procedure were trained and tested in standard operant conditioning chambers (BRS/LVE Inc., Laurel, MD and Lafayette Instruments Co., Lafayette, IN) housed in sound-attenuated cubicles. Pellet dispensers delivered 45-mg BIO SERV (Frenchtown, NJ) food pellets to a food cup on the front wall of the chamber beside a single response lever (DRL procedure) or between two response levers (discrimination procedure). Fan motors provided ventilation and masking noise for each chamber. House lights located above the food cup were illuminated during training and testing sessions for both procedures. In the chambers used for the DRL experiment, stimulus lights were located over the lever and in a corresponding position on the other side of the food cup. Sked-11 software (State Systems II Inc., Kalamazoo, MI) installed on a MicroPDP-11BA23 computer (Digital Equipment Corp., Maynard, MA) was used to control schedule contingencies and to record data for the DRL experiment. A micro-computer with Logic ‘1’ interface (MED Associates, Georgia, Vermont) and MED-PC software (MED Associates) was used for these tasks in the drug discrimination experiment. 2.3. Drugs PCP HCl (National Institute on Drug Abuse, Rockville, MD) was dissolved in saline. A stock solution of
A.D. Compton et al. / Neuropharmacology 40 (2001) 289–297
20 mg/ml clozapine (Research Biochemicals International, Natick, MA) was mixed in sterile water with 5–10 drops of lactic acid. Lower concentrations were obtained by dilution of this stock. All injections were administered intraperitoneally at a volume of 1 ml/kg. 2.4. Drug discrimination procedure Rats were trained to press one lever following administration of 2 mg/kg PCP and to press another lever after injection with saline, each according to a fixed-ratio 20 schedule of food reinforcement. Completion of 20 consecutive responses on the injection-appropriate lever resulted in delivery of a food pellet. Each response on the incorrect lever reset the ratio requirement on the correct lever. The daily injections for each rat were administered in a double alternation sequence of 2 mg/kg PCP and saline. Rats were injected and returned to their home cages until the start of the experimental session 15 min later. Training occurred during sessions conducted five days a week (Monday–Friday) until the rats had met three criteria during eight of ten consecutive sessions: (1) first completed fixed ratio 20 on the correct lever; (2) percentage of correct-lever responding ⬎80%; and (3) response rate ⬎0.4 responses/s. Following successful acquisition of the discrimination, substitution and antagonism tests were conducted on Tuesdays and Fridays during 15-min test sessions. Training continued on Mondays, Wednesdays, and Thursdays. During test sessions, responses on either lever delivered reinforcement according to a fixed ratio 20 schedule. In order to be tested, rats must have completed the first FR and made at least 80% of all responses on the injection-appropriate lever on the preceding day’s training session. In addition, the rat must have met these same criteria during at least one of the training sessions with the alternate training compound (PCP or saline) earlier in the week. A dose–effect curve for PCP was determined first. Then, rats were tested with different doses of clozapine (0–5.6 mg/kg, i.p.) in combination with the 2 mg/kg training dose of PCP. Clozapine or its vehicle was administered 60 min before the start of the session; PCP or saline was injected 40 min later (i.e., 10 min presession). Doses of both compounds were administered in ascending order. Before the combination dose–effect curve determination, control tests with combinations of vehicle and saline and vehicle and 2 mg/kg PCP were conducted. 2.5. Mixed signalled–unsignalled DRL procedure In this experiment, rats initially were trained to leverpress for food reinforcement. Once the lever-press response was acquired, a mixed DRL schedule was instituted. In order to obtain a reinforcer under the DRL
291
schedule, the rat had to wait a specified number of seconds between lever presses. Premature responses reset the DRL clock. During the first component of the mixed DRL schedule (signalled component), each reinforced response resulted in the onset of the stimulus lights on either side of the food cup and the end of the delay was signalled by the offset of these lights. During the second component (unsignalled component), the stimulus lights were not illuminated; hence, there was no exteroceptive signal at the end of the delay. The house light remained illuminated during both types of components. Over the course of training, the length of the delay was increased to 15 s for each component. Under the final schedule, the signalled and unsignalled components alternated for a total of six components. Each component lasted 8 min with a 1-min time out separating components for a total session length of 54 min. During the time out, the house and stimulus lights were extinguished and lever presses did not have programmed consequences. Testing began when responding was stable across sessions and components. A dose–effect curve for clozapine alone (0–10 mg/kg) was determined first. Then, rats were tested with the same doses of clozapine in combination with 2 mg/kg PCP. Clozapine or its vehicle was administered 60 min before the start of the session; PCP or saline was injected 45 min later (i.e., 15 min pre-session). For each dose– effect curve determination, doses or dose combinations were administered in randomized order according to a Latin Square design. 2.6. Data analysis For each drug discrimination test session, percentage of responses on the PCP lever and response rate (responses/s) were calculated. The ED50 (with 95% confidence intervals) for PCP was calculated using leastsquares linear regression on the linear part of the dose– effect curves (Tallarida and Murray, 1987) for percentage of drug-lever responding, plotted against log10 transformation of the dose. Since rats that responded less than 20 times during a test session did not press either lever a sufficient number of times to earn a reinforcer, their lever selection data were excluded from data analysis. For data analysis of sessions in the DRL experiment, responses during the three signalled components were analyzed together as were responses during the three unsignalled components. For each type of component, responses that occurred within 30 s of the last response were recorded separately in ten 3-s bins. Responses with interresponse times (IRT) ⬎30 s were recorded in an eleventh bin for each component. The temporal characteristics of the rats’ response patterns were analyzed with a method adapted from Richards and Seiden (1991). Responses with IRTs of 3 s or less were termed “burst responses” and were analyzed separately for signalled
292
A.D. Compton et al. / Neuropharmacology 40 (2001) 289–297
and unsignalled components with two-factor (PCP vs saline and clozapine dose) repeated measures ANOVAs. The mean of the total number of responses (burst plus non-burst responses) and the mean number of reinforcers for each treatment condition were also analyzed separately for signalled and unsignalled components with two-factor repeated measures ANOVAs. Differences between means revealed by significant ANOVAs were further compared through the use of Tukey post hoc tests (a=0.05). In addition, the median IRT of non-burst responses for each treatment condition was calculated for signalled and unsignalled components based on the distribution of IRTs of non-burst responses for each animal. Means of these median IRTs were subjected to separate two-way repeated measures ANOVAs for signalled and unsignalled components. This peak value of IRTs is known to be sensitive to drug effects under simple DRL schedules (e.g., Sanger, 1992). Since the meaningfulness of the median IRT of non-burst responses is of dubious value in instances where the total number of responses is very low, this measure was omitted from analysis in cases where a rat earned less than 10 reinforcers during signalled or unsignalled components over the entire session for a particular dose or dose combination. Finally, the mean number of responses during time-out periods throughout the session was calculated for each treatment condition and analyzed with repeated measures ANOVA.
3. Results 3.1. Effects of clozapine on PCP discrimination Phencyclidine produced full dose-dependent substitution for the PCP training dose (Fig. 1, upper panel), indicating that rats had successfully acquired the discrimination. The ED50 for PCP substitution was 0.94 mg/kg (95% CL: 0.62–1.42 mg/kg). Clozapine (0.3–5.6 mg/kg) did not block substitution of 2 mg/kg PCP (Fig. 1, upper panel), although the 3 and 5.6 mg/kg doses of clozapine in combination with 2 mg/kg PCP decreased overall response rates compared to vehicle control test results (Fig. 1, lower panel). 3.2. Effects of PCP alone and in combination with clozapine in DRL procedure The main differences between baseline performance in the signalled and unsignalled components were that subjects were more accurate and earned more reinforcers in the signalled components and the distribution of responses showed much less variability (i.e., more peaked distribution) in these components. Because of these differences in baseline values across type of component, drug effects were analyzed and are presented
Fig. 1. Percentage of PCP-lever responding (top panel) and response rate (bottom panel) in rats trained to discriminate 2 mg/kg PCP from saline. Before the start of the session, rats were injected with PCP alone (䊐) or with clozapine followed by an injection of 2 mg/kg dose of PCP (䊏). Points above VEH represent the results of a control test with saline. The points above the 0 dose represent a test with the clozapine vehicle and 2 mg/kg PCP. For both of the dose–effect curve determinations, each value represents the mean (±SEM) for 6–8 rats, except for percentage of PCP-lever responding for the 2 mg/kg PCP plus 5.6 mg/kg clozapine combination (n=3).
separately for signalled vs. unsignalled components. The profile of main effects of PCP (2 mg/kg) was identical across both types of components (filled symbols above vehicle): PCP significantly increased the number of burst responses as well as overall rates of responding (Figs. 2 and 3, respectively), but did not affect the number of reinforcers received (Fig. 4). PCP also increased response rate during time-out periods (data not shown). Further, 2 mg/kg of PCP shifted the mean of the distributions of median IRTs of non-burst responses to the left (shorter IRTs) in the unsignalled component and produced a trend (p=0.07) toward a leftward shift in the signalled component (Fig. 5 and Table 1). Clozapine at doses up to 10 mg/kg did not affect this shift (Fig. 5 and Table 1). Clozapine administered alone produced significant
A.D. Compton et al. / Neuropharmacology 40 (2001) 289–297
Fig. 2. Effects of clozapine alone (䊐) and clozapine plus 2 mg/kg PCP (䊏) on the number of burst responses (consecutive responses with IRT⬍3 s) during signalled (top panel) and unsignalled (bottom panel) components of a mixed signalled–unsignalled DRL procedure. Each value represents the mean (±SEM) for 12 rats. # indicates a main effect for PCP or for clozapine dose.
effects in the DRL procedure. The 10 mg/kg dose of clozapine decreased overall rates of responding (Fig. 3) and the 5 and 10 mg/kg doses of clozapine reduced the number of reinforcers earned in both components (Fig. 4). In contrast, none of the doses of clozapine alone affected the number of burst responses (Fig. 2). Because PCP alone increased the number of non-burst responses (an effect opposite of that of the high dose of clozapine), the effects of clozapine were less pronounced when rats also received PCP than when they received only clozapine. Although the effects of a combination of PCP and clozapine significantly decreased response and reinforcer rates compared to the effects of PCP alone, these effects occurred only at doses where clozapine alone produced similar decreases, suggesting that clozapine did not specifically alter the effects of PCP. Clozapine alone did not affect mean of the median IRT distributions (Fig. 5 and Table 1).
293
Fig. 3. Effects of clozapine alone (䊐) and clozapine plus 2 mg/kg PCP (䊏) on response rate (total number of responses/min) during signalled (top panel) and unsignalled (bottom panel) components of a mixed signalled–unsignalled DRL procedure. Each value represents the mean (±SEM) for 12 rats. # indicates a main effect for PCP or for clozapine dose. * indicates that the interaction was significant and that the value for the dose or dose combination is significantly different from the corresponding vehicle or PCP+vehicle dose. @ indicates that the interaction was significant and that the value for the PCP+clozapine dose is significantly different from the corresponding clozapine alone dose.
4. Discussion Using a discrete trial shock-avoidance discrimination procedure, Corbett (1995) found that acute doses of the prototypic atypical antipsychotic clozapine, but not the typical antipsychotic haloperidol, antagonized the discriminative stimulus effects of the NMDA open channel blocker dizocilpine. In contrast, other researchers (present study; Smith et al., 1999) have reported that clozapine did not block the discriminative stimulus effects of PCP-like drugs in drug discrimination procedures that used food reinforcement to motivate responding. Although the reasons for this disparity are not entirely clear, it is likely that they result from differences in the type of reinforcement or other experimental
294
A.D. Compton et al. / Neuropharmacology 40 (2001) 289–297
Fig. 4. Effects of clozapine alone (䊐) and clozapine plus 2 mg/kg PCP (䊏) on the number of reinforcers earned during signalled (top panel) and unsignalled (bottom panel) components of a mixed signalled–unsignalled DRL procedure. Each value represents the mean (±SEM) for 12 rats. # indicates a main effect for PCP or for clozapine dose. * indicates that the interaction was significant and that the value for the dose or dose combination is significantly different from the corresponding vehicle or PCP+vehicle dose. @ indicates that the interaction was significant and that the value for the PCP+clozapine dose is significantly different from the corresponding clozapine alone dose.
parameters. For example, schedule of reinforcement and training dose can influence results in PCP discrimination studies (Mansbach and Balster, 1991; McMillan and Wenger, 1984). While it is possible that clozapine might have blocked a lower training dose of PCP, it is unlikely. Clozapine produced even less attenuation of the discriminative stimulus effects of PCP under conditions of food reinforcement than did the typical antipsychotic haloperidol against even larger training doses of PCP (Beardsley and Balster, 1988; Browne, 1986). Haloperidol consistently has been found to partially attenuate the discriminative stimulus effects of PCP and PCP-like drugs, but usually only at doses that also severely suppress response rates (Wiley, 1995). In the present study, clozapine did not attenuate the discriminative stimulus effects of PCP, although substantial clozapine-induced response-rate decreases were observed. Considered together, these results suggest that clozapine would not
be any more effective as a treatment for acute PCP intoxication than are typical antipsychotics. Further, these results suggest that PCP discrimination, at least one using food reinforcement, does not provide a reliable preclinical method of distinguishing novel atypical antipsychotics from traditional neuroleptics. Clozapine also did not block the effects of a 2 mg/kg dose of PCP on responding under the mixed signalled– unsignalled DRL schedule. The main effect of PCP was to increase the number of responses that did not result in reinforcement. Notably, PCP increased the number of burst and time-out responses. Given that DRL schedules generally engender response rates that are greatly reduced compared to those under ratio schedules, these results may reflect the rate-dependent effects of PCP, an interpretation that is supported by the fact that PCP increased overall rates of responding. With a DRL schedule, a drug with rate-dependent effects would increase burst responses because of greater opportunity for responses with shorter IRTs during the fixed session length. In the present study, these effects occurred during both types of components, suggesting that the increased stimulus control present during signalled components was not enough to overcome the disruption of behavioral inhibition produced by PCP. Further, these results are consistent with those of previous researchers who have reported that PCP-like and other NMDA antagonists disrupt responding in unsignalled DRL procedures (Sanger, 1992; Stephens and Cole, 1996). As mentioned previously, clozapine did not alter any of the main effects of PCP, even at doses where clozapine began to produce its own effects. Hence, clozapine failed to affect PCPinduced behavior in either of the two operant procedures included in this study. In contrast, the locomotor activating effects of PCP and its disruption of prepulse inhibition of acoustic startle are reported to be attenuated by clozapine (Bakshi et al., 1994; Hoffman, 1992), although these results have not been consistently replicated across all studies (Hoffman et al., 1993; Wiley, 1994). These findings suggest that, while clozapine may attenuate the effects of PCP on spontaneous or reflexive behavior, it may not alter the effects of PCP on more complicated, learned behaviors such as drug discrimination and schedule-controlled behavior. The results of tests with clozapine alone are also of interest. In previous studies, acute dosing with clozapine had been shown to disrupt the pattern of operant responding under a variety of schedule conditions, including fixed interval (Kaempf and Porter, 1987), fixed consecutive number (Picker, 1988), and unsignalled DRL (Bruhwyler et al., 1993). In addition, Compton and Porter (1995) reported that clozapine dose-dependently disrupted matching behavior in a multiple variable interval schedule. Many of the rats in the Compton and Porter study failed to alter their response pattern to match signalled changes in reinforcement contingencies (i.e., cloz-
A.D. Compton et al. / Neuropharmacology 40 (2001) 289–297
295
Fig. 5. Effects of clozapine alone (left panels) and clozapine plus 2 mg/kg PCP (right panels) on the distributions of IRTs during signalled (top panels) and unsignalled (bottom panels) components of a mixed signalled–unsignalled DRL procedure. Each value represents the mean for 12 rats.
Table 1 Effects of clozapine alone and clozapine+2 mg/kg PCP on median IRT in a mixed DRL 15-s schedule Drug
Clozapine
Clozapine dose Mean (±SEM) (mg/kg) of signalled median IRTs 0 1.25 2.5 5
Clozapine+2 mg/kg PCPa
10 0 1.25 2.5 5 10
a
Mean (±SEM) of unsignalled median IRTs
17.0 (0.34), 17.2 (0.54), n=12 n=12 16.7 (0.45), 16.0 (0.81), n=12 n=12 15.7 (0.39), 14.5 (0.43), n=12 n=12 16.2 (0.63), 15.7 (1.08), n=11 n=11 13.5 (3.00), n=3 not determined 16.0 (0.34), 15.2 (0.69), n=12 n=12 15.0 (0.69), 14.2 (0.84), n=12 n=12 15.7 (0.65), 14.0 (1.03), n=12 n=12 14.7 (0.78), 13.0 (0.81), n=12 n=12 14.8 (0.88), n=9 13.5 (1.00), n=9
The main effect of PCP was to decrease the mean of the median IRTs. This effect was significant for the unsignalled components (p=0.03), but was only a trend (p=0.07) for the signalled components. Clozapine did not produce a main effect for either component and there were no interactions.
apine produced response rates across the schedule that were similar despite changes in reinforcement availability in the different components). In the matching study, all changes in reinforcement availability were signalled which presented difficulty in determining whether failure to match response rates with reinforcement density resulted from drug effects on the timing of responses or on stimulus control. In the present study, use of both signalled and unsignalled components in the same schedule afforded the ability to determine whether inappropriate responding resulted from disruption of the animal’s timing capacity or from the inability to use external stimuli to regulate behavior, or both. In this mixed signalled– unsignalled DRL schedule, baseline rates of overall responding during vehicle tests were lower in the unsignalled than in the signalled, components, resulting in responses that were spread over a larger number of IRT bins (i.e., a flatter IRT frequency distribution). In contrast, the IRT frequency distribution showed a more pronounced peak during signalled components after vehicle administration. As expected, more reinforcers were earned during signalled components; hence, responding was more efficient. The main effect of clozapine in this procedure was a dose-dependent decrease in the total number of responses. Visual inspection reveals that this rate-decreasing effect resulted in a flattening of the IRT distributions at higher clozapine doses during both components (see Fig. 5). Mean IRTs were not significantly affected by clozapine. Hence, rather than simply
296
A.D. Compton et al. / Neuropharmacology 40 (2001) 289–297
shifting the rats’ distribution of IRTs toward shorter latencies, clozapine instead dose-dependently caused responding to become more randomly distributed in time in both types of tasks (i.e., flattening of the IRT curves with increasing dose vs. shifts in the location of the peak IRT). This finding suggests that clozapine affects brain systems involved both with temporal ordering and with stimulus control, although the effect was less pronounced under conditions of high stimulus control in the signalled components. In summary, PCP produces distinctive effects in drug discrimination and mixed signalled–unsignalled DRL procedures. The atypical antipsychotic clozapine failed to meaningfully alter the effects of PCP in either of these tests. Hence, although clozapine has been shown to attenuate the behavioral effects of PCP in a number of procedures (e.g., startle and locomotor), it does not alter all of the effects of PCP. Importantly, clozapine does not consistently attenuate the behaviors that may be most closely associated with PCP intoxication; i.e., its discriminative stimulus effects and its disruption of operant behavior. These results bring into question the utility of distinguishing atypical from typical antipsychotics using PCP combination procedures. Further, clozapine produces distinct effects of its own on the timing of responses and on stimulus control. While these results suggest that acute dosing with clozapine may not offer an effective antidote to the consequences of PCP abuse, further studies need to be done in order to discover whether chronic dosing with clozapine might represent an effective treatment for the psychotomimetic effects of PCP. In human schizophrenia patients, repeated oral dosing with clozapine is necessary for clinical effectiveness. Future research should concentrate on determination of the effects of repeated dosing with clozapine on the behavioral effects of PCP.
Acknowledgements Research supported by National Institute on Drug Abuse grant DA-01442.
References Angrist, B., Gershon, S., 1970. The phenomenology of experimentally induced amphetamine psychosis: preliminary observations. Biological Psychiatry 2, 95–107. Bakshi, V.P., Geyer, M.A., 1995. Antagonism of phencyclidineinduced deficits in prepulse inhibition by the putative atypical antipsychotic olanzapine. Psychopharmacology 122, 198–201. Bakshi, V.P., Swerdlow, N.R., Geyer, M.A., 1994. Clozapine antagonizes phencyclidine-induced deficits in sensorimotor gating of the startle response. Journal of Pharmacology and Experimental Therapeutics 271, 787–794. Balster, R.L., 1990. Perception of drug effects. In: Berkley, M.A.,
Stebbins, W.C. (Eds.), Comparative Perception. Wiley, New York, pp. 127–154. Balster, R.L., Willetts, J., 1988. Receptor mediation of the discriminative stimulus properties of phencyclidine and sigma-opioid agonists. In: Colpaert, F.C., Balster, R.L. (Eds.), Transduction Mechanisms of Drug Stimuli. Springer, Berlin, pp. 122–135. Balster, R.L., Willetts, J., 1996. Phencyclidine: A drug of abuse and a tool for neuroscience research. In: Schuster, C.R., Kuhar, M.J. (Eds.), Pharmacological Aspects of Drug Dependence: Towards an Integrated Neurobehavioral Approach. Springer, Berlin, pp. 233– 262. Beardsley, P.M., Balster, R.L., 1988. Evaluation of antagonists of the discriminative stimulus and response rate effects of phencyclidine. Journal of Pharmacology and Experimental Therapeutics 244, 34–40. Browne, R., 1986. Discriminative stimulus properties of PCP mimetics. In: Clouet, D.H. (Ed.), Phencyclidine: An Update. National Institute on Drug Abuse, Rockville, MD, pp. 134–147. Bruhwyler, J., Chleide, E., Houbeau, G., Waegeneer, N., Mercier, M., 1993. Differentiation of haloperidol and clozapine using a complex operant schedule in the dog. Pharmacology Biochemistry Behavior 44, 181–189. Compton, A.D., Porter, J.H., 1995. Effects of clozapine and pimozide on the parameters of Herrnstein’s matching equation applied to multiple operant schedule responding in rats. Society for Neuroscience Abstracts 21, 467. Compton, A.D., Drew, M.R., Golden, K.M., Wiley, J.L., 1997. Clozapine impairs temporal perception and stimulus control in rats. Schizophrenia Research 24, 75. Corbett, R., 1995. Clozapine but not haloperidol antagonizes an MK801 discriminative stimulus cue. Pharmacology Biochemistry Behavior 51, 561–564. Ellinwood, E.H. Jr, Kilbey, M.M., 1977. Chronic stimulant intoxication models of psychosis. In: Hanin, I., Usdin, E. (Eds.), Animal Models in Psychiatry and Neurology. Pergamon Press, Oxford, pp. 61–74. Ginski, M.J., Witkin, J.M., 1994. Sensitive and rapid behavioral differentiation of N-methyl-d-aspartate receptor antagonists. Psychopharmacology 114, 573–582. Hoffman, D.C., 1992. Typical and atypical neuroleptics antagonize MK-801-induced locomotion and stereotypy in rats. Journal of Neural Transmission 89, 1–10. Hoffman, D.C., Donovan, H., Cassella, J.V., 1993. The effects of haloperidol and clozapine on the disruption of sensorimotor gating induced by the noncompetitive glutamate antagonist MK-801. Psychopharmacology 111, 339–344. Javitt, D.C., Zukin, S.R., 1991. Recent advances in the phencyclidine model of schizophrenia. American Journal of Psychiatry 148, 1301–1308. Kaempf, G.L., Porter, J.H., 1987. Differential effects of pimozide and clozapine on schedule-controlled and schedule-induced behaviors after acute and chronic administration. Journal of Pharmacology and Experimental Therapeutics 243, 437–445. Krystal, J.H., Karper, L.P., Seibyl, J.P., Freeman, G.K., Delaney, R., Bremner, J.D., Heninger, G.R., Bowers, M.B. Jr., Charney, D.S., 1994. Subanesthetic effects of the noncompetitive NMDA antagonist, ketamine, in humans. Archives of General Psychiatry 51, 199–214. Lahti, A.C., Koffel, B., LaPorte, D., Tamminga, C.A., 1995. Subanesthetic doses of ketamine stimulate psychosis in schizophrenia. Neuropsychopharmacology 13, 9–19. Mansbach, R.S., Balster, R.L., 1991. Pharmacological specificity of the phencyclidine discriminative stimulus in rats. Pharmacology Biochemistry Behavior 39, 971–975. Mansbach, R.S., Geyer, M.A., 1989. Effects of phencyclidine and phencyclidine biologs on sensorimotor gating in the rat. Neuropsychopharmacology 2, 299–308.
A.D. Compton et al. / Neuropharmacology 40 (2001) 289–297
McMillan, D.E., Wenger, G.R., 1984. Bias of phencyclidine discrimination by the schedule of reinforcement. Journal of Experimental Analysis of Behavior 42, 51–66. Picker, M., 1988. Effects of clozapine on fixed-consecutive-number responding in rats: a comparison to other neuroleptic drugs. Pharmacology Biochemistry Behavior 30, 603–612. Richards, J.B., Seiden, L.S., 1991. A quantitative interresponse–time analysis of DRL performance differentiates similar effects of the antidepressant desipramine and the novel anxiolytic gepirone. Journal of Experimental Analysis of Behavior 56, 173–192. Sanger, D.J., 1992. NMDA antagonists disrupt timing behaviour in rats. Behavioural Pharmacology 3, 593–600. Sanger, D.J., Jackson, A., 1989. Effects of phencyclidine and other Nmethyl-d-aspartate antagonists on the schedule-controlled behavior of rat. Journal of Pharmacology and Experimental Therapeutics 248, 1215–1221. Slemmer, J.E., Hyman, J.M., Wiley, J.L., Compton, A.D., Balster, R.L., 1997. Clozapine fails to attenuate phencyclidine’s discriminative stimulus effects in rats. Virginia Journal of Science 48, 145.
297
Smith, J.A., Boyer-Millar, C., Goudie, A.J., 1999. Does MK-801 discrimination constitute an animal model of schizophrenia useful for detecting atypical antipsychotics? Pharmacology Biochemistry Behavior 64, 429–433. Stephens, D.N., Cole, B.J., 1996. AMPA antagonists differ from NMDA antagonists in their effects on operant DRL and delayed matching to position tasks. Psychopharmacology 126, 249–259. Swerdlow, N.R., Bakshi, V., Geyer, M.A., 1996. Seroquel restores sensorimotor gating in phencyclidine-treated rats. Journal of Pharmacology and Experimental Therapeutics 279, 1290–1299. Tallarida, R.J., Murray, R.B., 1987. Manual of Pharmcologic Calculations with Computer Programs, 2nd ed. Springer, New York. Wiley, J.L., 1994. Clozapine’s effects on phencyclidine-induced disruption of prepulse inhibition of the acoustic startle response. Pharmacology Biochemistry Behavior 49, 1025–1028. Wiley, J.L., 1995. Effect of repeated haloperidol administration on phencyclidine discrimination in rats. Progress in Neuro-Psychopharmacology and Biological Psychiatry 19, 699–711.