- Email: [email protected]
Effects of nicotine on schedule-controlled behaviour
Neuropharmacology Vol. 24, No. 1, pp. 75-82, 1985 Printed in Great Britain. All rights reserved
EFFECTS
0028-3908/85 $3.00+ 0.00 Copyright 0 1985Pergamon Press Ltd
OF NICOTINE ON SCHEDULE-CONTROLLED BEHAVIOUR ROLE OF FIXED-INTERVAL LENGTH MODIFICATION BY MECAMYLAMINE CHLORPROMAZINE
AND AND
J. M. WHITE and C. C. GANGUZZA Department of Psychology, Monash University, Clayton, Victoria 3168, Australia (Accepted
11 May 1984)
Summary-The effects of nicotine, alone and in combination with mecamylamine and chlorpromazine, were studied in one group of rats exposed to a fixed-interval 30 set schedule of food reinforcement, and a second group exposed to a fixed-interval 120 set schedule. For both groups nicotine increased overall response rates in a dose-related fashion up to a maximum at 0.3 mg/kg. Examination of the within-interval response patterns showed that nicotine tended to increase the low level response rates in the early and middle parts of the interval and decrease, or increase to a proportionally smaller extent, the higher level response rates at the end of the interval. The response rate and pattern of the animal, rather than the schedule to which it was exposed, was found to be the main determinant of the effects of nicotine. Mecamylamine (l.Omg/kg) blocked most of the changes in rate produced by nicotine, although in the group exposed to the fixed-interval 30 set schedule, the increases in response rate tended to predominate after combined administration of mecamylamine and nicotine. Chlorpromazine (1 .Omg/kg) failed to block the effects of nicotine in either group. Instead, the effects of combined administration of nicotine and chlorpromazine were similar to the added effects of the two compounds given alone. It appears that the behavioural effects of nicotine are not mediated through catecholaminergic systems. Key words: nicotine, mecamylamine, chlorpromazine, schedule.
Investigation of the behavioural pharmacology of nicotine has shown that this compound can exert discriminative control over behaviour (e.g. Schechter and Rosecrans, 1972), act as a reinforcer (e.g.
Goldberg, Spealman and Goldberg, 1981) and modify rates of schedule-controlled behaviour (e.g. Morrison, 1967). While it shares these properties with a number of psychoactive compounds, comparison of the effects of nicotine on behaviour controlled by a variety of reinforcement schedules suggest that its actions are most similar to those of amphetamine. Like amphetamine, nicotine has been shown to increase responding under fixed-interval (FI; Pradhan, 1970), variable-interval (VI; Morrison, 1967) and differential-reinforcement-of-low-rate (DRL; Pradhan and Dutta, 1970) schedules, but to decrease responding under fixed-ratio (FR; Morrison, 1967) schedules. Although rats have been employed in most studies, the results obtained with squirrel monkeys have been very similar (Davis, Kersler and Dews, 1973; Spealman, Goldberg and Gardner, 1981). There is some evidence that a common mechanism of action may underly the similar behavioural effects of these two compounds. Amphetamine has been shown to induce release of catecholamines (CAs; Carlsson, 1970) and nicotine increases the turnover rate of noradrenaline (Bhagat, 1970). Orsingher and 75
schedule-controlled
behaviour, fixed-interval
Fulginiti (1971) found that both nicotine and amphetamine facilitated the acquisition of an avoidance response in rats in a shuttle-box, and that this action was reversed by pretreatment with the inhibitor of synthesis of catecholamines, a-methyl-p-tyrosine (cc-MPT). Schechter and Rosecrans (1972) demonstrated that this compound also blocked the discriminative stimulus effects of nicotine. However, Hirschhorn and Rosecrans (1974) failed to confirm this result, and suggested that the discrq_ancy may have been due to inconsistency in the elects of a-methyl-p-tyrosine or extraneous central effects of the compound. In a further study, Rosecrans and Chance (1977) did not obtain a clear effect of long-term depletion of catecholamines on the discriminative stimulus effects of nicotine. One purpose of the present study was to determine the role of release of catecholamines in mediating the behavioural effects of nicotine. Because of the difficulties and inconsistencies encountered in earlier studies, a different approach was adopted: chlorpromazine was tested for its ability to block the effects of nicotine on schedule-controlled behaviour. Chlorpromazine is a potent antagonist at dopamine receptors (both those associated with, and those independent of adenylyi cyclase; Kebabian, 1978) and at a-adrenergic receptors. It also has some antagonist
16
J. M. WHITEand C. C. GANGUZZA
activity at muscarinic cholinergic, tryptaminergic, histaminergic and a-adrenergic receptors (Baldessarini, 1980). Chlorpromazine has been shown to block a wide range of behavioural effects of amphetamine (e.g. Wilson and Schuster, 1972). This is most likely a result of its antagonist activity at dopaminergic and noradrenergic receptors, although antagonist activity at other sites may possibly play some role. If chlorpromazine blocks the effects of nicotine, as well as those of amphetamine, there would be strong evidence that a common mechanism, release of catecholamines, underlies the similar behavioural effects of the two compounds. If this is not the case, then it is more likely that these common effects are mediated in different ways. The actions of chlorpromazine were compared with those of the nicotinic cholinergic antagonist, mecamylamine. This compound has been shown to block the discriminative effects of nicotine (Hirschhorn and Rosecrans, 1974) and its effects on schedule-controlled behaviour (Stitzer, Morrison and Domino, 1970; Spealman et al., 1981). Whereas mecamylamine is thought to block the actions of nicotine at the nicotinic cholinergic receptor, chlorpromazine should block the effects of catecholamines released as a result of nicotine binding to the receptor. Fixed-interval schedules were used in the present study. Although moderate doses of nicotine have been shown to increase fixed-interval response rates in three studies (Pradhan, 1970; Davis et al., 1973; Spealman et al., 1981) only decreases in response rates were observed in a fourth (Stitzer et al., 1970). The within-interval patterns of responding were also affected differently in the latter study. One procedural difference was the use of a shorter fixed-interval (88 set vs 120-300 set) in this study. The possible importance of the length of the fixed-interval suggested by the different findings was examined in the present experiment. The effects of nicotine alone, and in combination with each of the antagonists, were examined in one group of rats trained on a fixed-interval 30 set schedule, and another group on a fixed-interval 120 set schedule. Data included both overall rates and within-interval patterns of responding.
METHODS Subjects Six adult male Wistar hooded rats, bred in the Psychology Department at Monash University, served as subjects. They were reduced to 85% of their free feeding weights (240-350g) prior to the commencement of training, and were retained at these weights throughout the period of the study. The animals were housed in individual cages, with free access to water, in a temperature-controlled environment.
Apparatus Three operant chambers of identical dimensions (25 x 25 x 25 cm) were used. In each, there was a small recessed area in the middle of one wall. A 1.5 cm hole in this area allowed a dipper to deliver 0.15 ml of liquid. To the left of the dipper was a standard rat lever. Each chamber was illuminated by a 4 W fluorescent light and was enclosed in a soundand light-attenuating cubicle. Control of the experiment and collection of the data were performed by a small computer. Cumulative records were also obtained. Procedure All animals were trained to press the lever to obtain access for 3.5 set to a 25% solution of sweetened condensed milk (Nestle) with tap water. Three animals (RN7, RN17, RN18) were then assigned to the fixed-interval 30 set group, and three (RNl, RN5, RN1 1) to the fixed-interval 120 set group. Over three sessions they were exposed to fixed-interval schedules of gradually increasing length until the target value had been reached. Training continued at these values for a further 37 sessions, by which time there was no consistent directional change in the overall response rates. The medians of the last three b!ocks of four sessions were used to evaluate the change. Sessions ended when 30 reinforcements had been delivered and were conducted for 5 days a week. Testing was begun following stabilization. Each rat received nicotine hydrogen tartrate in doses of 0.01, 0.03, 0.1, 0.3 and 1.0 mg/kg; doses of nicotine in the range 0.03-1.0 mg/kg in combination with mecamylamine hydrochloride (1 .O mg/kg) and in combination with chlorpromazine hydrochloride (1 .O mg/kg). Mecamylamine and chlorpromazine were also administered together with saline. However, two animals (RN5 and RN7) became ill before the effects of the mecamylamine-saline combination could be determined. Drugs were administered on Tuesdays and Fridays and saline on Thursdays; normal training sessions continued on Mondays and Wednesdays. Injections of nicotine were given 5 min prior to the start of the session, while chlorpromazine and mecamylamine were administered 15 min before the session. The order of doses was randomized for each rat. Session length and total number of lever presses were recorded and the overall response rate calculated. Each interval was divided into five equal sections and the number of bar presses in each section recorded. Averaging over the whole session gave a measure of response rate in each fifth. All animals were exposed to each dose of nicotine and each combination of nicotine and antagonist twice. The data presented are averages of the two determinations. Drugs Nicotine
hydrogen(
+)tartrate
(BDH
Chemicals
Behavioural effects of nicotine
‘A
FI 30%~
FI 120%~ I s
0.03
001
Dose
of nicotine
I
I
I
0.1
0.3
1.0
1 mg
/ kg )
Fig. 1. The effects of graded doses of nicotine (NIC), alone and in combination with mecamylamine (MEC; 1.Omg/kg) and chlorpromazine (CHLOR; 1.0 mg/kg) on overall rates of fixed-interval (FI) responding. Each point is a mean of data for three subjects Upper panel: fixed-interval 30 set group. Lower panel: fixed-interval 120 set group. (0) Nicotine, (m) nicotine + mecamylamine (1 .O), (A) nicotine + chlorpromazine (1 .O).
Ltd, Poole, England), mecamylamine hydrochloride [a gift of Merck, Sharpe & Dohme (Australia) Pty Ltd] and chlorpromazine hydrochloride (Sigma Chemical Company, St Louis, Missouri, U.S.A.) were dissolved in 0.9% saline. Solutions of drugs and saline were administered subcutaneously in a volume of 1.0 ml/kg body weight. All doses of drugs are expressed in terms of the free base. RESULTS Nicotine alone Figure 1 shows the mean change in overall response rate following administration of nicotine for both fixed-interval 30 set and fixed-interval 120 set
groups. The values are percentages of the mean rate in control sessions on Thursdays, conducted during the relevant series of doses. These means, together with standard deviations, are shown in Table 1. At doses of 0.1 and 0.3 mg/kg, nicotine increased the response rates above control levels. After administration of 0.3 mg!kg of nicotine the mean increase was 87% in the fixed-interval 30 set group and 59% in the fixed-interval 120 set group. Although the
II
magnitude varied between animals, there was an increase for all six at this dose. The 0.01 and 0.03 mg/kg doses had little effect on the response rate, as did the largest dose tested, 1.0 mg/kg. Administration of doses above this level resulted in response rates near zero as the toxic effects of the drug became more pronounced. A more detailed picture of the effects of nicotine can be gained from Fig. 2. It shows the effects of the three largest doses on response rates in each fifth of the fixed interval. As the two smallest doses of nicotine had little effect on response rates, these data have been omitted for clarity. In the upper panels are the results for the subjects exposed to the fixed-interval 30 set schedule. Subject RN7 showed a characteristic fixed-interval pattern with a very low level of response rate early in the interval, gradually and then sharply increasing to a much higher level at the end. Nicotine increased the response rates early in the interval, although the change in the first fifth was often small. The 0.1 mg/kg dose decreased the response rates later in the interval, whereas the 0.3 mg/kg dose produced only increases in response rate. The largest dose (1 .Omg/kg) substantially decreased the rates at the end of the interval. The results from RN18 are similar, particularly at the largest dose of nicotine. However, RN17 had a different fixed-interval pattern and the effects of nicotine also varied from the others. For this animal the increase in rate through the interval was much less steep and the rate in the final fifth was considerably smaller. The principle effect of the drug was to increase response rate through most of the interval. Even after administration of the drug the maximum response rate was only about 18 responses per minute. Figure 2 (lower panel) shows that for subjects exposed to the fixed-interval 120 set schedule, 0.3 mg/kg of nicotine produced increases in response rate across most of the interval. However, 1.Omg/kg of nicotine had only slight effects, mostly increasing the rate. Only in the data from RN1 1 was there any evidence of decreasing effects on rate in the final fifth of the interval. The 1.0 mg/kg dose had slight effect on increasing rate through most of the interval for RN5, and in the middle fifths for RNl. Like RN17, RN5 responded at a lower level of rate and the increases produced by the nicotine were proportionally greater than they were for the other two animals in the group. For this animal the largest
Table 1, Mean and standard deviation (in parentheses) of response rates in saline control sessions Fixed-interval 30 set Drugs Nicotine Nicotine + Mecamylamine Nicotine + Chlorpromazine
7 25.7 (4.2) 39.9 (7.8) 41.4 (5.2)
17 3.0 0:;) (0.8) 1.7 (2.6)
Fixed-interval
120 set
18
1
5
11
17.8 (7.8) 28.2 (8.0) 32.5 (6.9)
13.1 (1.2) 12.3 (3.5) 10.8 (1.8)
6.4 (2.2) 13.3 (8.2) 33.2 (7.1)
24.1 (3.2) 25.2 (2.4) 15.7 (4.8)
78
J. M. 100
and C. C.
WHITE
r
r
.
GANGUZZA
FI 30sec
1 RN17
.z
F
100
r
r
FI 120 set
Fifth
of
fixed
-
interval
Fig. 2. The effects of 0.1, 0.3 and 1.0 mg/kg of nicotine (NIC) on response rates in each fifth of the fixed-interval. Upper panels: subjects in fixed-interval 30 set group. Lower panels: subjects in fixed-interval 120set group. (0) Saline, (0) nicotine (O.l), (m) nicotine (0.3), (A) nicotine (1.O).
increases in rate occurred after administration of 0.1 mg/kg of nicotine. These data do not reveal any consistent differences in the effects of nicotine with changes in the value of the fixed-interval schedule. Nicotine and mecamylamine The effects on the overall rates of 1.0 mg/kg meca-
mylamine, administered together with graded doses of nicotine, are also shown in Fig. 1. Mecamylamine failed to block the rate-increasing effects of nicotine on fixed interval 30 set responding. Marked increases in response rate were still evident through the range 0.034.3 mg/kg. However, mecamylamine produced increases in response rate when administered together FI 30 set
c
60
::
120
El 2
100
c
RN7
RN 17
FI 120%~
t
80
RN1 60
1
Fifth
of
fixed-
2
3
4
5
interval
Fig. 3. The effectsof 0.1,0.3 and 1.0 mg/kg of nicotine (NIC) in combination with mecamylamine (MEC; l.Omgjkg) on response rate in each fifth of the fixed-interval. The effects of saline together with mecamylamine (1 .Omg/kg) are also shown for all subjects except RN5 and RN 17. Upper panels: subjects in the fixed-interval 30 set group. Lower panels: subjects in the fixed-interval 120 set group. (0) Saline, (0) saline + mecamylamine (l.O), (0) nicotine (0.1) + mecamylamine (1 .O), (m) nicotine (0.3) + mecamylamine (1 .O), (A) nicotine (1 .O)+ mecamylamine (1 .O).
Behavioural effects of nicotine
xi^
60
.’E \ 2
60
:
20
E e
79
FI 30 set
40
O FI 120 set
r 60
60
RN 1
t
RN 11
1234512345
Fifth
of fixed - interval
Fig. 4. The effects of 0.1, 0.3 and 1.0 mg/kg of nicotine (NIC) in combination with chlorpromazine (CHLOR; 1.0mg/kg) on response rate in each fifth of the fixed-interval. The effectsof saline together with chlorpromazine (1.Omg/kg) are also shown. Upper panels: subjects in the fixed-interval 30 set group. Lower panels: subjects in the fixed-interval 120 set group. (0) Saline, (0) saline + chlorpromazine (1 .O), (a) nicotine (0.1) + chlorpromazine (1 .O), (W) nicotine (0.3) + chlorpromazine (1.O), (A) nicotine (1.0) + chlorpromazine (1.O).
with saline. The largest increase, at 0.3 mg/kg, was somewhat smaller when mecamylamine was administered together with the nicotine than when nicotine was administered alone. These results can be seen in more detail in Fig. 3 (upper panel), which shows the patterns of responding after administration of mecamylamine and nicotine (0. l-l .Omg/kg). The controls show that mecamylamine itself had some effect, but changed the effects of nicotine considerably. Whereas 1.Omg/kg nicotine alone decreased response rates toward the end of the interval for both RN7 and RN18 (Fig. 2), there were no rate-decreasing effects at all when mecamylamine was also administered. In addition, the rate-increasing effects of nicotine for RN17 were markedly reduced by mecamylamine. Comparison of Fig. 2 (upper panel) and Fig. 3 (upper panel) clearly shows that mecamylamine narrowed the range of changes in response rate (both increases and decreases) brought about by nicotine. Increases in response rate still predominated thereby accounting for the elevated response rates after combined administration of mecamylamine and nicotine. Figure 1 (lower panel) shows the effect of mecamylamine (1.0 mg/kg) and graded doses of nicotine on the fixed-interval 120 set response rate. These data demonstrate almost complete blockade of the effects of nicotine by mecamylamine. Deviations from control levels of responding were only slight and there was no clear dose-related trend. The more detailed picture presented in Fig. 3 (lower panels) suggests that the presence of mecamylamine did not completely abolish the effects of nicotine. The data from
RN1 and RN5 showed increases in response rate in the second half of the interval at certain doses. For RN5 there were also decreases in response rate after large doses of nicotine (0.3 and 1.0 mg/kg) in combination with mecamylamine. Indeed, RN5 was somewhat atypical in that the effects of nicotine seemed to be considerably different (predominantly decreasing rate rather than increasing rate) when it was administered together with mecamylamine. Otherwise, the data from this group are also consistent with the observation that mecamylamine reduced the magnitude of the changes in response rate produced by nicotine. Nicotine and chlorpromazine
The effects of 1.Omg/kg of chlorpromazine, administered together with graded doses of nicotine, are also shown in Fig. 1. It is important to note that chlorpromazine alone had decreasing effects on rate. It reduced response rates to 82% (FI 30sec group) and 70% (FI 120 set group) of control levels. When chlorpromazine was given with nicotine the increases in response rate were smaller than when nicotine was given alone. The average curve of the fixed-interval 30 set group showed a small increase in response rate after administration of 0.3 mg/kg nicotine and chlorpromazine, and then a substantial decrease with the largest dose of nicotine (1.0 mg/kg) and chlorpromazine. Figure 4 (upper panel) shows that the decreasing effects of chlorpromazine alone on rate occurred predominantly with RN1 7 and RN1 8 and
80
J. M, RN?
WHITE
and C. C.
GANGUZZA
RNq
Soline
Saline
NICll.O)
NIC(l.O)+
NIC(1.0)
MECI1.O1
NIC (1.0)
+MEC
(1.0)
+CHLOR(l.O) NIC(!.O)+
CHLOAll.01
5 min
Fig. 5. ~umuiative records illustrating the effects of 1.Omg/kg of nicotine (NIC) alone and in combination with 1.Omg/kg of mecamylamine (MEC) and 1.Omg/kg of chlorpromazine (CHLOR). The subjects were RN7 (fixed-interval 30 see; left panels) and RN1 (fixed-inte~al 120 see; right panels). The ordinate shows time and the abscissa cumulative lever-presses. The pen rests after each reinforcement.
resulted from decreases in the higher level of response rates toward the end of the interval. When administered with nicotine, chlorpromazine substantially reduced the increasing effects on rate of this compound, but accentuated its decreasing effects on rate in two subjects (RN17 and RN18). Subject RN17 seemed much less sensitive to the effects of chlorpromazine than the other two, but with the largest dose of nicotine the level of response rate during the middle of the interval was somewhat lower when chlorpromazine was also administered. These actions of chlorpromazine could not be considered to be antagonism. Rather, the effects of chlorpromazine appeared to combine in an additive manner with those of nicotine. Chlorpromazine had a similar effect on the overall response rates of the animals exposed to the fixed-interval 12Osec schedule (Fig. 1). Marked increasing effects on rate were evident after administration of ~hlorproma~ne together with 0.3 mg/kg of nicotine. Decreasing effects on rate occurred with the smallest dose of nicotine (0.03 mg,/kg) but not with the largest dose (1.0 mg/kg). Figure 4 (lower panels) shows that, as with the fixed-interval 3Osec group, the effect of chlorpromazine alone was to decrease the high level of response rates later in the interval. Again, one subject (RNS) seemed less sensitive to the effects of chlorpromazine than the other two. The data for all subjects shows that chlorpromazine atten-
uated the increasing effects of nicotine on rate, confirming the major finding from the fixed-interval 30 set group. Time course Comparison of the results from the two groups is complicated by the difference in length of session. This was a consequence of equating the number of reinforcers. Figure 5 shows some cumulative records of an animal exposed to the fixed-interval 30 set schedule and one exposed to the fixed-interval 120 set schedule. The effects of nicotine alone, the ability of mecamylamine to reverse these and the failure of chlorpromazine to block the actions of nicotine are clearly illustrated. Both nicotine alone and the combination produced nicotine-chlorpromazine strong decreasing effects on rate early in the session. The duration of these effects as a proportion of session length is approximately constant. These data suggest that the duration of action of the drugs was determined both by the properties of the drug itself and the interaction between the drug and the schedule of reinforcement. Thus, comparison between the two groups can be valid despite a difference in session length. DISCUSSION
The results of the present study confirm earlier findings that nicotine has the effect of increasing rate
Behavioural effects of nicotine in animals exposed to fixed-interval schedules. The increases occurred irrespective of whether the interval was 30 or 120 set in duration. Thus, the size of schedule did not appear to account for the failure to observe increases in response rate in an earlier study (Stitzer et al., 1970). The two also differed in the type of reinforcer used: food in the present study and water in the one by Stitzer et al. (1970). Whether this or some other factor can account for the discrepancy is not known. The results obtained here differed from those of Stitzer et al. (1970) in another respect. In the latter study nicotine mainly decreased the low levels of rates of responding occurring early in the interval, and decreased or had little effect on the high levels of response rates at the end. Here, irrespective of size of the schedule, response rates early in the interval increased after administration of nicotine. Small doses tended to increase response rates at the end of the interval, whereas large doses sometimes decreased these. Like Stitzer et al. (1970) most earlier studies did not show details of the changes in the pattern of fixed-interval responding produced by nicotine, but the reports were generally consistent with the data obtained here. One difference was noted: previous studies found little evidence of increases in response rate toward the end of the interval, but such increases were regularly observed here. Most were small, but for those animals with low level baseline rates the increases were proportionately quite large. The failure of earlier studies to observe increases in response rate in the latter part of the interval may simply have been because all the animals were responding at relatively high rates. Unfortunately, the data are not available to confirm this. Although mecamylamine blocked most of the effects of nicotine, some stimulation of responding was observed when the two compounds were administered together. The graphs of individual response patterns show this to be due almost entirely to one animal (RN17) in the fixed-interval 30 set group. This particular animal had a very low level baseline rate and certain combinations of nicotine and mecamylamine approximately doubled this rate. Stitzer et al. (1970) also observed such stimulation of fixed-interval responding in one animal after administration of nicotine and mecamylamine. As baseline rates were not reported in that study it is not clear whether this stimulation occurs only when the baseline rate is low, or whether some other factor is important. Other animals here did show some increase in rate with combinations of nicotine and mecamylamine, but the deviations from control levels were usually small and not clearly related to the dose of nicotine. Although chlorpromazine diminished the increasing effects of nicotine on rate (more so for the FI 30 set group than for the FI 120 set group) it clearly is not a nicotine antagonist. For both groups the effect on overall response rates of combinations
81
of nicotine and chlorpromazine closely followed the added effects of each drug alone. Furthermore, chlorpromazine failed to consistently attenuate the changes in the pattern of fixed-interval responding brought about by nicotine. These data are strong evidence against the hypothesis that the behavioural effects of nicotine are mediated by catecholamine systems. They support the results of earlier studies in which neither long-term depletion of catecholamines (Rosecrans and Chance, 1977), nor inhibition of synthesis of catecholamines (Hirschhorn and Rosecrans, 1974) substantially modified the discriminative effects of nicotine. In contrast, other experiments have shown that dibenamine, an cc-adrenergic antagonist, blocked changes in avoidance responding induced by nicotine (Orsingher and Fulginiti, 1971) and that inhibition of synthesis of catecholamines blocked both changes in avoidance responding induced by nicotine (Orsingher and Fulginiti, 1971) and the discriminative effects of nicotine (Schechter and Rosecrans, 1972). A study by Orsingher and Fulginiti (1973) may help resolve some of these apparent contradictions. They demonstrated that the locus of the interaction between dibenamine and nicotine was peripheral rather than central. It was suggested that the combination of the two compounds produced a vascular effect which reversed the central effects of nicotine. If it is assumed that a-methyl-p-tyrosine reverses the actions of nicotine in a similar manner, then these contrary results could be accounted for without assuming that nicotine releases catecholamines centrally. It is important to note that in the study of Schechter and Rosecrans (1972) cc-methyl-p-tyrosine attenuated, but did not completely block the discriminative effects of nicotine. In the later study (Hirschhorn and Rosecrans, 1974) where a different training procedure was used, no attenuation was found. It is plausible to suggest that nicotine causes release of catecholamines in the periphery but not in the central nervous system and that any behavioural effects of nicotine which can be shown to be mediated by catecholamines are due to peripheral actions of the drug. In the present study, nicotine has been shown to exert a strong effect on schedule-controlled behaviour. The direction and magnitude of the changes in response rate were related to the baseline rate of the animal as well as the dose of nicotine; the size of the schedule per se did not appear to be important. The ability of mecamylamine to block these effects suggests that nicotine has its primary action at nicotinic cholinergic receptors. The failure of chlorpromazine to block the changes in response rate induced by nicotine confirms earlier evidence suggesting that changes in central catecholamine systems do not underly the behavioural effects of nicotine. work was supported by a grant from the Australian Tobacco Research Foundation.
Acknowledgement-This
82
J. M. WHITE and C. C. GANGUZZA REFERENCES
Baldessarini R. J. (1980) Drugs and the treatment of psychiatric disorders. In: The Pharmacological Basis of Therapeutics (Gilman A. G., Goodman L. S. and Gilman A., Eds), 6th edn, pp. 391-447. Macmillan, New York. Bhagat B. (1970) Effects of chronic administration of nicotine on storage and synthesis of noradrenaline in rat brain. Br. J. Pharmac. 38: 8692. Carlsson A. (1970) Amphetamine and brain catecholamines. In: Amphetamine and Related Compounds (Costa E. and Garattini S., Eds), pp. 289-300. Raven Press, New York. Davis T. R. A., Kersler C. J. and Dews P. B. (1973) of behavioral effects of nicotine, Comparison d-amphetamine, caffeine and dimethylheptyl tetrahydrocannabinol in squirrel monkeys. Psychopharmacologia 32: 51-65. Goldberg S. R., Spealman R. D. and Goldberg D. M. (1981) Persistent behavior at high rates maintained by intravenous self-administration of nicotine. Science 214: 573-575. Hirschhom I. D. and Rosecrans J. A. (1974) Studies on the time course and the effect of cholinergic and adrenergic receptor blockers on the stimulus effect of nicotine. Psychopharmacology 40: 109-120. Kebabian J. W. (1978) Multiple classes of dopamine receptors in mammalian central nervous system: the involvement of dopamine-sensitive adenylyl cyclase. Life Sci. 23: 479484. Morrison C. F. (1967) Effects of nicotine on operant behaviour of rats. Int. J. Neuropharmac. 6: 229-240. Orsingher 0. A. and Fulginiti S. (1971) Effects of alphamethyl tyrosine and adrenergic blocking agents on the
facilitating action of amphetamine and nicotine on learning in rats. Psychopharmacologia 1% 231-240. Orsingher 0. A. and Fulginiti S. (1973) Influence of peripheral mechanisms on the facilitatory learning action of amphetamine and nicotine in rats. Pharmaco1og.y 9: 138-144. Pradhan S. N. (1970) Effects of nicotine on several schedules of behavior in rats. Archs int. Pharmacodyn. Ther. 183: 127-138. Pradhan S. N. and Dutta S. N. (1970) Comparative effects of nicotine and amphetamine on timing behavior in rats. Neuropharmacology 9: 9-l 6. Rosecrans J. A. and Chance W. T. (1977) Cholinergic and non-cholinergic aspects of the discriminative stimulus properties of nicotine. In: Discriminative Stimulus Properfies of Drugs (La1 H., Ed.), pp. 155-185. Plenum Press, New York. Schechter M. D. and Rosecrans J. A. (1972) Nicotine as a discriminative stimulus in rats depleted of norepinephrine or 5-hydroxytryptamine. Psychopharmacologia 24: 417429. Spealman R. D., Goldberg S. R. and Gardner M. L. (1981) Behavioral effects of nicotine: schedule controlled responding by squirrel monkeys. J. Pharmac. exp. Ther. 216: 484491. Stitzer M., Morrison J. and Domino E. F. (1970) Effects of nicotine on fixed interval behavior and their modification by cholinergic antagonists. J. Pharmac. exp. Ther. 171: 166177. Wilson M. C. and Schuster C. R. (1972) The effects of chlorpromazine on psychomotor stimulant selfadministration in the rhesus monkey. Psychopharmacologia 26: 115-126.
EFFECTS
0028-3908/85 $3.00+ 0.00 Copyright 0 1985Pergamon Press Ltd
OF NICOTINE ON SCHEDULE-CONTROLLED BEHAVIOUR ROLE OF FIXED-INTERVAL LENGTH MODIFICATION BY MECAMYLAMINE CHLORPROMAZINE
AND AND
J. M. WHITE and C. C. GANGUZZA Department of Psychology, Monash University, Clayton, Victoria 3168, Australia (Accepted
11 May 1984)
Summary-The effects of nicotine, alone and in combination with mecamylamine and chlorpromazine, were studied in one group of rats exposed to a fixed-interval 30 set schedule of food reinforcement, and a second group exposed to a fixed-interval 120 set schedule. For both groups nicotine increased overall response rates in a dose-related fashion up to a maximum at 0.3 mg/kg. Examination of the within-interval response patterns showed that nicotine tended to increase the low level response rates in the early and middle parts of the interval and decrease, or increase to a proportionally smaller extent, the higher level response rates at the end of the interval. The response rate and pattern of the animal, rather than the schedule to which it was exposed, was found to be the main determinant of the effects of nicotine. Mecamylamine (l.Omg/kg) blocked most of the changes in rate produced by nicotine, although in the group exposed to the fixed-interval 30 set schedule, the increases in response rate tended to predominate after combined administration of mecamylamine and nicotine. Chlorpromazine (1 .Omg/kg) failed to block the effects of nicotine in either group. Instead, the effects of combined administration of nicotine and chlorpromazine were similar to the added effects of the two compounds given alone. It appears that the behavioural effects of nicotine are not mediated through catecholaminergic systems. Key words: nicotine, mecamylamine, chlorpromazine, schedule.
Investigation of the behavioural pharmacology of nicotine has shown that this compound can exert discriminative control over behaviour (e.g. Schechter and Rosecrans, 1972), act as a reinforcer (e.g.
Goldberg, Spealman and Goldberg, 1981) and modify rates of schedule-controlled behaviour (e.g. Morrison, 1967). While it shares these properties with a number of psychoactive compounds, comparison of the effects of nicotine on behaviour controlled by a variety of reinforcement schedules suggest that its actions are most similar to those of amphetamine. Like amphetamine, nicotine has been shown to increase responding under fixed-interval (FI; Pradhan, 1970), variable-interval (VI; Morrison, 1967) and differential-reinforcement-of-low-rate (DRL; Pradhan and Dutta, 1970) schedules, but to decrease responding under fixed-ratio (FR; Morrison, 1967) schedules. Although rats have been employed in most studies, the results obtained with squirrel monkeys have been very similar (Davis, Kersler and Dews, 1973; Spealman, Goldberg and Gardner, 1981). There is some evidence that a common mechanism of action may underly the similar behavioural effects of these two compounds. Amphetamine has been shown to induce release of catecholamines (CAs; Carlsson, 1970) and nicotine increases the turnover rate of noradrenaline (Bhagat, 1970). Orsingher and 75
schedule-controlled
behaviour, fixed-interval
Fulginiti (1971) found that both nicotine and amphetamine facilitated the acquisition of an avoidance response in rats in a shuttle-box, and that this action was reversed by pretreatment with the inhibitor of synthesis of catecholamines, a-methyl-p-tyrosine (cc-MPT). Schechter and Rosecrans (1972) demonstrated that this compound also blocked the discriminative stimulus effects of nicotine. However, Hirschhorn and Rosecrans (1974) failed to confirm this result, and suggested that the discrq_ancy may have been due to inconsistency in the elects of a-methyl-p-tyrosine or extraneous central effects of the compound. In a further study, Rosecrans and Chance (1977) did not obtain a clear effect of long-term depletion of catecholamines on the discriminative stimulus effects of nicotine. One purpose of the present study was to determine the role of release of catecholamines in mediating the behavioural effects of nicotine. Because of the difficulties and inconsistencies encountered in earlier studies, a different approach was adopted: chlorpromazine was tested for its ability to block the effects of nicotine on schedule-controlled behaviour. Chlorpromazine is a potent antagonist at dopamine receptors (both those associated with, and those independent of adenylyi cyclase; Kebabian, 1978) and at a-adrenergic receptors. It also has some antagonist
16
J. M. WHITEand C. C. GANGUZZA
activity at muscarinic cholinergic, tryptaminergic, histaminergic and a-adrenergic receptors (Baldessarini, 1980). Chlorpromazine has been shown to block a wide range of behavioural effects of amphetamine (e.g. Wilson and Schuster, 1972). This is most likely a result of its antagonist activity at dopaminergic and noradrenergic receptors, although antagonist activity at other sites may possibly play some role. If chlorpromazine blocks the effects of nicotine, as well as those of amphetamine, there would be strong evidence that a common mechanism, release of catecholamines, underlies the similar behavioural effects of the two compounds. If this is not the case, then it is more likely that these common effects are mediated in different ways. The actions of chlorpromazine were compared with those of the nicotinic cholinergic antagonist, mecamylamine. This compound has been shown to block the discriminative effects of nicotine (Hirschhorn and Rosecrans, 1974) and its effects on schedule-controlled behaviour (Stitzer, Morrison and Domino, 1970; Spealman et al., 1981). Whereas mecamylamine is thought to block the actions of nicotine at the nicotinic cholinergic receptor, chlorpromazine should block the effects of catecholamines released as a result of nicotine binding to the receptor. Fixed-interval schedules were used in the present study. Although moderate doses of nicotine have been shown to increase fixed-interval response rates in three studies (Pradhan, 1970; Davis et al., 1973; Spealman et al., 1981) only decreases in response rates were observed in a fourth (Stitzer et al., 1970). The within-interval patterns of responding were also affected differently in the latter study. One procedural difference was the use of a shorter fixed-interval (88 set vs 120-300 set) in this study. The possible importance of the length of the fixed-interval suggested by the different findings was examined in the present experiment. The effects of nicotine alone, and in combination with each of the antagonists, were examined in one group of rats trained on a fixed-interval 30 set schedule, and another group on a fixed-interval 120 set schedule. Data included both overall rates and within-interval patterns of responding.
METHODS Subjects Six adult male Wistar hooded rats, bred in the Psychology Department at Monash University, served as subjects. They were reduced to 85% of their free feeding weights (240-350g) prior to the commencement of training, and were retained at these weights throughout the period of the study. The animals were housed in individual cages, with free access to water, in a temperature-controlled environment.
Apparatus Three operant chambers of identical dimensions (25 x 25 x 25 cm) were used. In each, there was a small recessed area in the middle of one wall. A 1.5 cm hole in this area allowed a dipper to deliver 0.15 ml of liquid. To the left of the dipper was a standard rat lever. Each chamber was illuminated by a 4 W fluorescent light and was enclosed in a soundand light-attenuating cubicle. Control of the experiment and collection of the data were performed by a small computer. Cumulative records were also obtained. Procedure All animals were trained to press the lever to obtain access for 3.5 set to a 25% solution of sweetened condensed milk (Nestle) with tap water. Three animals (RN7, RN17, RN18) were then assigned to the fixed-interval 30 set group, and three (RNl, RN5, RN1 1) to the fixed-interval 120 set group. Over three sessions they were exposed to fixed-interval schedules of gradually increasing length until the target value had been reached. Training continued at these values for a further 37 sessions, by which time there was no consistent directional change in the overall response rates. The medians of the last three b!ocks of four sessions were used to evaluate the change. Sessions ended when 30 reinforcements had been delivered and were conducted for 5 days a week. Testing was begun following stabilization. Each rat received nicotine hydrogen tartrate in doses of 0.01, 0.03, 0.1, 0.3 and 1.0 mg/kg; doses of nicotine in the range 0.03-1.0 mg/kg in combination with mecamylamine hydrochloride (1 .O mg/kg) and in combination with chlorpromazine hydrochloride (1 .O mg/kg). Mecamylamine and chlorpromazine were also administered together with saline. However, two animals (RN5 and RN7) became ill before the effects of the mecamylamine-saline combination could be determined. Drugs were administered on Tuesdays and Fridays and saline on Thursdays; normal training sessions continued on Mondays and Wednesdays. Injections of nicotine were given 5 min prior to the start of the session, while chlorpromazine and mecamylamine were administered 15 min before the session. The order of doses was randomized for each rat. Session length and total number of lever presses were recorded and the overall response rate calculated. Each interval was divided into five equal sections and the number of bar presses in each section recorded. Averaging over the whole session gave a measure of response rate in each fifth. All animals were exposed to each dose of nicotine and each combination of nicotine and antagonist twice. The data presented are averages of the two determinations. Drugs Nicotine
hydrogen(
+)tartrate
(BDH
Chemicals
Behavioural effects of nicotine
‘A
FI 30%~
FI 120%~ I s
0.03
001
Dose
of nicotine
I
I
I
0.1
0.3
1.0
1 mg
/ kg )
Fig. 1. The effects of graded doses of nicotine (NIC), alone and in combination with mecamylamine (MEC; 1.Omg/kg) and chlorpromazine (CHLOR; 1.0 mg/kg) on overall rates of fixed-interval (FI) responding. Each point is a mean of data for three subjects Upper panel: fixed-interval 30 set group. Lower panel: fixed-interval 120 set group. (0) Nicotine, (m) nicotine + mecamylamine (1 .O), (A) nicotine + chlorpromazine (1 .O).
Ltd, Poole, England), mecamylamine hydrochloride [a gift of Merck, Sharpe & Dohme (Australia) Pty Ltd] and chlorpromazine hydrochloride (Sigma Chemical Company, St Louis, Missouri, U.S.A.) were dissolved in 0.9% saline. Solutions of drugs and saline were administered subcutaneously in a volume of 1.0 ml/kg body weight. All doses of drugs are expressed in terms of the free base. RESULTS Nicotine alone Figure 1 shows the mean change in overall response rate following administration of nicotine for both fixed-interval 30 set and fixed-interval 120 set
groups. The values are percentages of the mean rate in control sessions on Thursdays, conducted during the relevant series of doses. These means, together with standard deviations, are shown in Table 1. At doses of 0.1 and 0.3 mg/kg, nicotine increased the response rates above control levels. After administration of 0.3 mg!kg of nicotine the mean increase was 87% in the fixed-interval 30 set group and 59% in the fixed-interval 120 set group. Although the
II
magnitude varied between animals, there was an increase for all six at this dose. The 0.01 and 0.03 mg/kg doses had little effect on the response rate, as did the largest dose tested, 1.0 mg/kg. Administration of doses above this level resulted in response rates near zero as the toxic effects of the drug became more pronounced. A more detailed picture of the effects of nicotine can be gained from Fig. 2. It shows the effects of the three largest doses on response rates in each fifth of the fixed interval. As the two smallest doses of nicotine had little effect on response rates, these data have been omitted for clarity. In the upper panels are the results for the subjects exposed to the fixed-interval 30 set schedule. Subject RN7 showed a characteristic fixed-interval pattern with a very low level of response rate early in the interval, gradually and then sharply increasing to a much higher level at the end. Nicotine increased the response rates early in the interval, although the change in the first fifth was often small. The 0.1 mg/kg dose decreased the response rates later in the interval, whereas the 0.3 mg/kg dose produced only increases in response rate. The largest dose (1 .Omg/kg) substantially decreased the rates at the end of the interval. The results from RN18 are similar, particularly at the largest dose of nicotine. However, RN17 had a different fixed-interval pattern and the effects of nicotine also varied from the others. For this animal the increase in rate through the interval was much less steep and the rate in the final fifth was considerably smaller. The principle effect of the drug was to increase response rate through most of the interval. Even after administration of the drug the maximum response rate was only about 18 responses per minute. Figure 2 (lower panel) shows that for subjects exposed to the fixed-interval 120 set schedule, 0.3 mg/kg of nicotine produced increases in response rate across most of the interval. However, 1.Omg/kg of nicotine had only slight effects, mostly increasing the rate. Only in the data from RN1 1 was there any evidence of decreasing effects on rate in the final fifth of the interval. The 1.0 mg/kg dose had slight effect on increasing rate through most of the interval for RN5, and in the middle fifths for RNl. Like RN17, RN5 responded at a lower level of rate and the increases produced by the nicotine were proportionally greater than they were for the other two animals in the group. For this animal the largest
Table 1, Mean and standard deviation (in parentheses) of response rates in saline control sessions Fixed-interval 30 set Drugs Nicotine Nicotine + Mecamylamine Nicotine + Chlorpromazine
7 25.7 (4.2) 39.9 (7.8) 41.4 (5.2)
17 3.0 0:;) (0.8) 1.7 (2.6)
Fixed-interval
120 set
18
1
5
11
17.8 (7.8) 28.2 (8.0) 32.5 (6.9)
13.1 (1.2) 12.3 (3.5) 10.8 (1.8)
6.4 (2.2) 13.3 (8.2) 33.2 (7.1)
24.1 (3.2) 25.2 (2.4) 15.7 (4.8)
78
J. M. 100
and C. C.
WHITE
r
r
.
GANGUZZA
FI 30sec
1 RN17
.z
F
100
r
r
FI 120 set
Fifth
of
fixed
-
interval
Fig. 2. The effects of 0.1, 0.3 and 1.0 mg/kg of nicotine (NIC) on response rates in each fifth of the fixed-interval. Upper panels: subjects in fixed-interval 30 set group. Lower panels: subjects in fixed-interval 120set group. (0) Saline, (0) nicotine (O.l), (m) nicotine (0.3), (A) nicotine (1.O).
increases in rate occurred after administration of 0.1 mg/kg of nicotine. These data do not reveal any consistent differences in the effects of nicotine with changes in the value of the fixed-interval schedule. Nicotine and mecamylamine The effects on the overall rates of 1.0 mg/kg meca-
mylamine, administered together with graded doses of nicotine, are also shown in Fig. 1. Mecamylamine failed to block the rate-increasing effects of nicotine on fixed interval 30 set responding. Marked increases in response rate were still evident through the range 0.034.3 mg/kg. However, mecamylamine produced increases in response rate when administered together FI 30 set
c
60
::
120
El 2
100
c
RN7
RN 17
FI 120%~
t
80
RN1 60
1
Fifth
of
fixed-
2
3
4
5
interval
Fig. 3. The effectsof 0.1,0.3 and 1.0 mg/kg of nicotine (NIC) in combination with mecamylamine (MEC; l.Omgjkg) on response rate in each fifth of the fixed-interval. The effects of saline together with mecamylamine (1 .Omg/kg) are also shown for all subjects except RN5 and RN 17. Upper panels: subjects in the fixed-interval 30 set group. Lower panels: subjects in the fixed-interval 120 set group. (0) Saline, (0) saline + mecamylamine (l.O), (0) nicotine (0.1) + mecamylamine (1 .O), (m) nicotine (0.3) + mecamylamine (1 .O), (A) nicotine (1 .O)+ mecamylamine (1 .O).
Behavioural effects of nicotine
xi^
60
.’E \ 2
60
:
20
E e
79
FI 30 set
40
O FI 120 set
r 60
60
RN 1
t
RN 11
1234512345
Fifth
of fixed - interval
Fig. 4. The effects of 0.1, 0.3 and 1.0 mg/kg of nicotine (NIC) in combination with chlorpromazine (CHLOR; 1.0mg/kg) on response rate in each fifth of the fixed-interval. The effectsof saline together with chlorpromazine (1.Omg/kg) are also shown. Upper panels: subjects in the fixed-interval 30 set group. Lower panels: subjects in the fixed-interval 120 set group. (0) Saline, (0) saline + chlorpromazine (1 .O), (a) nicotine (0.1) + chlorpromazine (1 .O), (W) nicotine (0.3) + chlorpromazine (1.O), (A) nicotine (1.0) + chlorpromazine (1.O).
with saline. The largest increase, at 0.3 mg/kg, was somewhat smaller when mecamylamine was administered together with the nicotine than when nicotine was administered alone. These results can be seen in more detail in Fig. 3 (upper panel), which shows the patterns of responding after administration of mecamylamine and nicotine (0. l-l .Omg/kg). The controls show that mecamylamine itself had some effect, but changed the effects of nicotine considerably. Whereas 1.Omg/kg nicotine alone decreased response rates toward the end of the interval for both RN7 and RN18 (Fig. 2), there were no rate-decreasing effects at all when mecamylamine was also administered. In addition, the rate-increasing effects of nicotine for RN17 were markedly reduced by mecamylamine. Comparison of Fig. 2 (upper panel) and Fig. 3 (upper panel) clearly shows that mecamylamine narrowed the range of changes in response rate (both increases and decreases) brought about by nicotine. Increases in response rate still predominated thereby accounting for the elevated response rates after combined administration of mecamylamine and nicotine. Figure 1 (lower panel) shows the effect of mecamylamine (1.0 mg/kg) and graded doses of nicotine on the fixed-interval 120 set response rate. These data demonstrate almost complete blockade of the effects of nicotine by mecamylamine. Deviations from control levels of responding were only slight and there was no clear dose-related trend. The more detailed picture presented in Fig. 3 (lower panels) suggests that the presence of mecamylamine did not completely abolish the effects of nicotine. The data from
RN1 and RN5 showed increases in response rate in the second half of the interval at certain doses. For RN5 there were also decreases in response rate after large doses of nicotine (0.3 and 1.0 mg/kg) in combination with mecamylamine. Indeed, RN5 was somewhat atypical in that the effects of nicotine seemed to be considerably different (predominantly decreasing rate rather than increasing rate) when it was administered together with mecamylamine. Otherwise, the data from this group are also consistent with the observation that mecamylamine reduced the magnitude of the changes in response rate produced by nicotine. Nicotine and chlorpromazine
The effects of 1.Omg/kg of chlorpromazine, administered together with graded doses of nicotine, are also shown in Fig. 1. It is important to note that chlorpromazine alone had decreasing effects on rate. It reduced response rates to 82% (FI 30sec group) and 70% (FI 120 set group) of control levels. When chlorpromazine was given with nicotine the increases in response rate were smaller than when nicotine was given alone. The average curve of the fixed-interval 30 set group showed a small increase in response rate after administration of 0.3 mg/kg nicotine and chlorpromazine, and then a substantial decrease with the largest dose of nicotine (1.0 mg/kg) and chlorpromazine. Figure 4 (upper panel) shows that the decreasing effects of chlorpromazine alone on rate occurred predominantly with RN1 7 and RN1 8 and
80
J. M, RN?
WHITE
and C. C.
GANGUZZA
RNq
Soline
Saline
NICll.O)
NIC(l.O)+
NIC(1.0)
MECI1.O1
NIC (1.0)
+MEC
(1.0)
+CHLOR(l.O) NIC(!.O)+
CHLOAll.01
5 min
Fig. 5. ~umuiative records illustrating the effects of 1.Omg/kg of nicotine (NIC) alone and in combination with 1.Omg/kg of mecamylamine (MEC) and 1.Omg/kg of chlorpromazine (CHLOR). The subjects were RN7 (fixed-interval 30 see; left panels) and RN1 (fixed-inte~al 120 see; right panels). The ordinate shows time and the abscissa cumulative lever-presses. The pen rests after each reinforcement.
resulted from decreases in the higher level of response rates toward the end of the interval. When administered with nicotine, chlorpromazine substantially reduced the increasing effects on rate of this compound, but accentuated its decreasing effects on rate in two subjects (RN17 and RN18). Subject RN17 seemed much less sensitive to the effects of chlorpromazine than the other two, but with the largest dose of nicotine the level of response rate during the middle of the interval was somewhat lower when chlorpromazine was also administered. These actions of chlorpromazine could not be considered to be antagonism. Rather, the effects of chlorpromazine appeared to combine in an additive manner with those of nicotine. Chlorpromazine had a similar effect on the overall response rates of the animals exposed to the fixed-interval 12Osec schedule (Fig. 1). Marked increasing effects on rate were evident after administration of ~hlorproma~ne together with 0.3 mg/kg of nicotine. Decreasing effects on rate occurred with the smallest dose of nicotine (0.03 mg,/kg) but not with the largest dose (1.0 mg/kg). Figure 4 (lower panels) shows that, as with the fixed-interval 3Osec group, the effect of chlorpromazine alone was to decrease the high level of response rates later in the interval. Again, one subject (RNS) seemed less sensitive to the effects of chlorpromazine than the other two. The data for all subjects shows that chlorpromazine atten-
uated the increasing effects of nicotine on rate, confirming the major finding from the fixed-interval 30 set group. Time course Comparison of the results from the two groups is complicated by the difference in length of session. This was a consequence of equating the number of reinforcers. Figure 5 shows some cumulative records of an animal exposed to the fixed-interval 30 set schedule and one exposed to the fixed-interval 120 set schedule. The effects of nicotine alone, the ability of mecamylamine to reverse these and the failure of chlorpromazine to block the actions of nicotine are clearly illustrated. Both nicotine alone and the combination produced nicotine-chlorpromazine strong decreasing effects on rate early in the session. The duration of these effects as a proportion of session length is approximately constant. These data suggest that the duration of action of the drugs was determined both by the properties of the drug itself and the interaction between the drug and the schedule of reinforcement. Thus, comparison between the two groups can be valid despite a difference in session length. DISCUSSION
The results of the present study confirm earlier findings that nicotine has the effect of increasing rate
Behavioural effects of nicotine in animals exposed to fixed-interval schedules. The increases occurred irrespective of whether the interval was 30 or 120 set in duration. Thus, the size of schedule did not appear to account for the failure to observe increases in response rate in an earlier study (Stitzer et al., 1970). The two also differed in the type of reinforcer used: food in the present study and water in the one by Stitzer et al. (1970). Whether this or some other factor can account for the discrepancy is not known. The results obtained here differed from those of Stitzer et al. (1970) in another respect. In the latter study nicotine mainly decreased the low levels of rates of responding occurring early in the interval, and decreased or had little effect on the high levels of response rates at the end. Here, irrespective of size of the schedule, response rates early in the interval increased after administration of nicotine. Small doses tended to increase response rates at the end of the interval, whereas large doses sometimes decreased these. Like Stitzer et al. (1970) most earlier studies did not show details of the changes in the pattern of fixed-interval responding produced by nicotine, but the reports were generally consistent with the data obtained here. One difference was noted: previous studies found little evidence of increases in response rate toward the end of the interval, but such increases were regularly observed here. Most were small, but for those animals with low level baseline rates the increases were proportionately quite large. The failure of earlier studies to observe increases in response rate in the latter part of the interval may simply have been because all the animals were responding at relatively high rates. Unfortunately, the data are not available to confirm this. Although mecamylamine blocked most of the effects of nicotine, some stimulation of responding was observed when the two compounds were administered together. The graphs of individual response patterns show this to be due almost entirely to one animal (RN17) in the fixed-interval 30 set group. This particular animal had a very low level baseline rate and certain combinations of nicotine and mecamylamine approximately doubled this rate. Stitzer et al. (1970) also observed such stimulation of fixed-interval responding in one animal after administration of nicotine and mecamylamine. As baseline rates were not reported in that study it is not clear whether this stimulation occurs only when the baseline rate is low, or whether some other factor is important. Other animals here did show some increase in rate with combinations of nicotine and mecamylamine, but the deviations from control levels were usually small and not clearly related to the dose of nicotine. Although chlorpromazine diminished the increasing effects of nicotine on rate (more so for the FI 30 set group than for the FI 120 set group) it clearly is not a nicotine antagonist. For both groups the effect on overall response rates of combinations
81
of nicotine and chlorpromazine closely followed the added effects of each drug alone. Furthermore, chlorpromazine failed to consistently attenuate the changes in the pattern of fixed-interval responding brought about by nicotine. These data are strong evidence against the hypothesis that the behavioural effects of nicotine are mediated by catecholamine systems. They support the results of earlier studies in which neither long-term depletion of catecholamines (Rosecrans and Chance, 1977), nor inhibition of synthesis of catecholamines (Hirschhorn and Rosecrans, 1974) substantially modified the discriminative effects of nicotine. In contrast, other experiments have shown that dibenamine, an cc-adrenergic antagonist, blocked changes in avoidance responding induced by nicotine (Orsingher and Fulginiti, 1971) and that inhibition of synthesis of catecholamines blocked both changes in avoidance responding induced by nicotine (Orsingher and Fulginiti, 1971) and the discriminative effects of nicotine (Schechter and Rosecrans, 1972). A study by Orsingher and Fulginiti (1973) may help resolve some of these apparent contradictions. They demonstrated that the locus of the interaction between dibenamine and nicotine was peripheral rather than central. It was suggested that the combination of the two compounds produced a vascular effect which reversed the central effects of nicotine. If it is assumed that a-methyl-p-tyrosine reverses the actions of nicotine in a similar manner, then these contrary results could be accounted for without assuming that nicotine releases catecholamines centrally. It is important to note that in the study of Schechter and Rosecrans (1972) cc-methyl-p-tyrosine attenuated, but did not completely block the discriminative effects of nicotine. In the later study (Hirschhorn and Rosecrans, 1974) where a different training procedure was used, no attenuation was found. It is plausible to suggest that nicotine causes release of catecholamines in the periphery but not in the central nervous system and that any behavioural effects of nicotine which can be shown to be mediated by catecholamines are due to peripheral actions of the drug. In the present study, nicotine has been shown to exert a strong effect on schedule-controlled behaviour. The direction and magnitude of the changes in response rate were related to the baseline rate of the animal as well as the dose of nicotine; the size of the schedule per se did not appear to be important. The ability of mecamylamine to block these effects suggests that nicotine has its primary action at nicotinic cholinergic receptors. The failure of chlorpromazine to block the changes in response rate induced by nicotine confirms earlier evidence suggesting that changes in central catecholamine systems do not underly the behavioural effects of nicotine. work was supported by a grant from the Australian Tobacco Research Foundation.
Acknowledgement-This
82
J. M. WHITE and C. C. GANGUZZA REFERENCES
Baldessarini R. J. (1980) Drugs and the treatment of psychiatric disorders. In: The Pharmacological Basis of Therapeutics (Gilman A. G., Goodman L. S. and Gilman A., Eds), 6th edn, pp. 391-447. Macmillan, New York. Bhagat B. (1970) Effects of chronic administration of nicotine on storage and synthesis of noradrenaline in rat brain. Br. J. Pharmac. 38: 8692. Carlsson A. (1970) Amphetamine and brain catecholamines. In: Amphetamine and Related Compounds (Costa E. and Garattini S., Eds), pp. 289-300. Raven Press, New York. Davis T. R. A., Kersler C. J. and Dews P. B. (1973) of behavioral effects of nicotine, Comparison d-amphetamine, caffeine and dimethylheptyl tetrahydrocannabinol in squirrel monkeys. Psychopharmacologia 32: 51-65. Goldberg S. R., Spealman R. D. and Goldberg D. M. (1981) Persistent behavior at high rates maintained by intravenous self-administration of nicotine. Science 214: 573-575. Hirschhom I. D. and Rosecrans J. A. (1974) Studies on the time course and the effect of cholinergic and adrenergic receptor blockers on the stimulus effect of nicotine. Psychopharmacology 40: 109-120. Kebabian J. W. (1978) Multiple classes of dopamine receptors in mammalian central nervous system: the involvement of dopamine-sensitive adenylyl cyclase. Life Sci. 23: 479484. Morrison C. F. (1967) Effects of nicotine on operant behaviour of rats. Int. J. Neuropharmac. 6: 229-240. Orsingher 0. A. and Fulginiti S. (1971) Effects of alphamethyl tyrosine and adrenergic blocking agents on the
facilitating action of amphetamine and nicotine on learning in rats. Psychopharmacologia 1% 231-240. Orsingher 0. A. and Fulginiti S. (1973) Influence of peripheral mechanisms on the facilitatory learning action of amphetamine and nicotine in rats. Pharmaco1og.y 9: 138-144. Pradhan S. N. (1970) Effects of nicotine on several schedules of behavior in rats. Archs int. Pharmacodyn. Ther. 183: 127-138. Pradhan S. N. and Dutta S. N. (1970) Comparative effects of nicotine and amphetamine on timing behavior in rats. Neuropharmacology 9: 9-l 6. Rosecrans J. A. and Chance W. T. (1977) Cholinergic and non-cholinergic aspects of the discriminative stimulus properties of nicotine. In: Discriminative Stimulus Properfies of Drugs (La1 H., Ed.), pp. 155-185. Plenum Press, New York. Schechter M. D. and Rosecrans J. A. (1972) Nicotine as a discriminative stimulus in rats depleted of norepinephrine or 5-hydroxytryptamine. Psychopharmacologia 24: 417429. Spealman R. D., Goldberg S. R. and Gardner M. L. (1981) Behavioral effects of nicotine: schedule controlled responding by squirrel monkeys. J. Pharmac. exp. Ther. 216: 484491. Stitzer M., Morrison J. and Domino E. F. (1970) Effects of nicotine on fixed interval behavior and their modification by cholinergic antagonists. J. Pharmac. exp. Ther. 171: 166177. Wilson M. C. and Schuster C. R. (1972) The effects of chlorpromazine on psychomotor stimulant selfadministration in the rhesus monkey. Psychopharmacologia 26: 115-126.