- Email: [email protected]
Task-specific effects of nicotine in rats
Copyright
TASK-SPECIFIC INTRACRANIAL
EFFECTS
OF NICOTINE
SELF-STIMULATION
I
002%39OXtXh $3.00 + 0.00 19X6 Pergamon Press Ltd
IN RATS
AND LOCOMOTOR
ACTIVITY
G. J. SCHAEFER and R. P. MICHAEL Department of Psychiatry, Emory University School of Medicine, Atlanta, GA 30322 and Georgia Mental Health institute, 1256 Briarcliff Road. N.E. Atlanta, GA 30306, U.S.A.
(Acceptecl 9 June 1985)
Summa~-The acute effects of nicotine (0.03-l .Omgjkg) were studied in a locomotor activity procedure and in a series of intracranial self-stimulation (ICSS) paradigms. Nicotine produced a dose-dependent decrease in locomotor activity. When animals were trained to lever-press for intracranial self-stimulation on a continuous reinforcement schedule (CRF), the drug was ineffective except at the l.Omg/kg dose, which produced a moderate decrease in the rate of responding. However, when animals were tested in a fixed-ratio: 15 (FR: 15) paradigm, nicotine produced a steep, biphasic dos+response curve. At the 0. I mg/kg dose, the response rates were increased to approx. 60% above baseline, while at the I ,Omg,‘kg dose, response rates were decreased to approx. 90% below baseline values. The effects of nicotine were also studied in an auto-titration procedure which measured the rewarding value of the stimulus. There was a decrease in performance at larger doses similar to that observed in the continuous reinforcement procedure. but there were no significant changes in the threshold for reinforcement. Nicotine did not produce any change in the detection threshold for stimulation of the brain. In acute studies, therefore, nicotine produced both stimu~tion and djsruption of behavior, effects that were brought to light by the fixed-ratio schedule of reinforcement, and this may relate to the rewarding effects of nicotine. Key words: nicotine, locomotor activity, intracranial self-stimulation.
Nicotine is a psychoactive drug that may be responsible for the positive reinforcing effects of tobacco smoking. Since smokers typically engage in the behavior that results in the intake of nicotine at a high rate (approx. 70,000 puffs per year; Russeli, 1977), the “nicotine habit” should be regarded as an extremely strong one. In an attempt to extend present knowledge of the nature of the reinforcement produced by nicotine, the effects of this drug on motivated behavior have been examined. One procedure used is that of intracranial self-stimulation (ICSS), but previous reports on the effects of nicotine on lever-pressing for intracranial self-stimulation have been inconsistent. OIds and Domino (1969) observed in rats that nicotine produced an initial decrease, followed by an increase in response rates, but noted that depressant effects predominated. Pradhan and Bowling (1971) reported that nicotine increased response rates, pa~~culariy in animals with low baselines, and these results were confirmed by Newman (1972) who used a variable-interval rather than a continuous reinforcement schedule (CRF). The many and varied pharmacological effects of nicotine may tend to obscure the mechanisms responsible for its reinforcing characteristics, and the possibility should be kept in mind that changes in operant performance
may not reflect authentic shifts in the reinforcement value of the stimulus; this is of particular concern
when response rates are the sole dependent measure of the action of the drug. The purpose of these experiments was to examine the acute effects of nicotine on response rates and on the central reward mechanisms that are thought to maintain the smoking habit. The effects of nicotine were tested, therefore, in a group of rats trained to lever-press for intracranial self-stimulation on a continuous reinforcement schedule, the paradigm typically used to test effects of drugs. These animals were then retrained to lever-press for intracranial selfstimulation on a fixed-ratio: 15 (FR: 15) schedule of reinforcement, and the dose-response effects of nicotine were again tested. The second procedure, the auto-titration paradigm, was used to measure thresholds for reinforcement, and simultaneously provided a rate-dependent and a rate-independent measure of behavior due to intracranial self-stimulation (Schaefer and Holtzman, 1979). The reinforcement threshold was obtained by calculating the mean current intensity at which the animals re-set the stimulating current to a suprathreshold value. The third procedure was that recently developed to study the stimuius control properties of intracranial selfstimulation (Schaefer and Michael, 1986). This is a procedure independent of rate that allows the investigator to measure the threshold for detection of the stimulus and to determine the extent to which 125
126
G. J.
SCHAEFFKand
drugs alter this threshold. Finally, animals were tested for changes in spontaneous locomotor activity. Comparing data from the different paradigms in one laboratory increases the possibility of obtaining convergent evidence concerning the role of nicotine in positive rcinforccment mechanisms.
Malt rats (335420 g) were used. They were bred in the present authors’ colony from stock purchased from King Animal Laboratories, Inc. (Oregon, Wisconsin, U.S.A.). Between experimental sessions, animals were housed in group cages (24 per cage) in a colony room and food and water were available ad lihiturn. The colony room was artificially illuminated between 07:OO and l9:OO hr.
Locomotor activity was measured using an OmniTech Digiscan RXY activity monitor (Columbus, Ohio, U.S.A.). The device measured horizontal activity by counting the total number of interruptions of an infra-red beam. To measure activity, an animal acrylic inside a clear cage was placed (39.4 x 39.4 x 30.5 cm high, inside dimensions) which rested inside the “main frame” of the Digiscan monitor. The entire monitoring device was placed inside a sound-attenuating chamber which was equipped with a fdn and a 25 W red light bulb. In addition to the number of interruptions of the beam being displayed on a digital electronic counter, the output from the monitor was interfaced with a Beckman strip-chart recorder. Pen deflections from baseline increased in proportion to the speed at which the light beams were interrupted, thus producing an analogue measure of the speed of movement during the monitoring session. The operant test chamber (31 x 30 x 29 cm high) used to measure lever-pressing on a continuous reinforcement and a fixed-ratio: I5 schedule was constructed in this laboratory. The chamber was equipped with a single conventional lever (G6312, Ralph Gerbrands, Co., Arlington, Massachusetts, U.S.A.) which could trigger stimulation of the brain on either a continuous or partial reinforcement schedule. Electrical pulses were produced by a biphasic constant current stimulator constructed in this laboratory (Schaefer, Bonsall and Michael, 1982) and were passed through a two-channel commutator (Schaefer, Baumgardner and Michael, 1981) to the brain. by way of a length of spring-shielded hearing aid wire (Plastic Products Co., Roanoke, Virginia, U.S.A.). The stimuli consisted of 200 msec trains of biphasic square-wave pulses at 100 Hz with a pulse duration of 2 msec (+ I msec and -I msec). The range of current intensities used for the studies of the continuous reinforcement schedule was 27-50 PA, while the range of intensities for the fixed-ratio:15 was 35-250 {LA.
R. P.
MICHAEL
The auto-titration procedure was used to measure reinforcement thresholds and experiments were conducted in an operant chamber similar to that used for the continuous reinforcement and fixed-ratio: I5 experiments In addition to a conventional lever, however, there was an omnidirectional lever (G6313, Gerbrands) suspended from the ceiling near the opposite wall of the chamber. The stimulator and titrator have been described in detail elsewhere (Schaefer, Baumgardner and Michael, 1979). The electrical stimuli were of the same parameters as those noted above except that the train duration was 100 msec. The current was programmed to drop 3 PA after each 15th lever-press and the current was held constant at each step. When the omnidirectional lever was pressed, the starting current was re-set and a new titration series started. The operant chambers for the experiments on intracranial self-stimulation were controlled by a combination of +28 and +5 VDC modules. In addition, cumulative recorders were interfaced with the programming equipment and these showed response rates and reinforcement delivery for the continuous reinforcement and the fixed-ratio: I5 experiments and also response rates and current drop for each titration series during the auto-titration experiment. The test chamber used in the detection threshold experiment was 30.5 cm long, 31 cm wide and 45 cm high (see Schaefer and Michael, 1986). On the front wall were positioned two conventional levers (G6312, Gerbrands), separated by a Plexiglas partition mounted perpendicular to the wall. These two levers were designated the “choice levers”. On the opposite wall an omnidirectional lever (G6313) was mounted and was designated the “initiating lever”. Programming for this study was controlled by +5 VDC modules. A biphasic constant current stimulator (Schaefer et al., 1982) modified to produce two different stimulus trains produced the electrical pulses. The stimulus produced by lever presses on the choice levers was a 500-msec train of 100 Hz with a pulse duration of 2 msec and a range of current intensities from 75 to 225 PA. The stimulus train produced by the initiating lever was the same as that produced by the choice lever except for the intensity, which was a proportional amount (O-100%) of that produced by the choice levers. Surgery and histology Rats used in experiments on brain stimulation were anesthetized with sodium pentobarbital (50 mg/kg, i.p.), and given atropine sulfate (0.25 mg, s.c.) to reduce any respiratory distress. The animals were then positioned in a stereotaxic device, and after exposing the skull, 4-5 stainless-steel screws were fixed to the skull. Next, a hole was drilled in the skull, the dura was incised, and a bipolar platinum electrode (tip diameter = 0. I25 mm, Plastic Products Co.) was lowered into the lateral hypothalamus (medial forebrain bundle) using coordinates: AP 5.2,
Nicotine
and brain
L 1.7, H -2.2 (Pellegrino, Pellegrino and Cushman, 1979). To form a secure anchor, crania-plastic cement was applied to the screws and electrodes. Finally, the animals were given an intramuscular injection of 100,000 U of benzathine penicillin G and procaine penicillin G. When the experiments were completed. the animals were killed with sodium pentobarbital and perfused via the heart with 10% formalin. After fixation, frozen sections of brain were cut at 50 p and alternate sections were stained with cresyl violet and W&l’s stain. From this material, the electrode tips were accurately located. Proc’edurcs To test locomotor activity, the animals were first habituated to the procedure for 5 days by injecting them with saline and I5 min later placing them in the locomotor apparatus for 12 min. Testing of drugs was then begun. On Mondays and Thursdays the animals were given saline, and on Tuesdays and Fridays they were given different doses of nicotine tartrate. Data from the first 2 min of the test were discarded and the total number of interruptions of the infra-red beam during the next 10 min were used for analysis. The doses of nicotine tartrate (0.03, 0. I, 0.3 and I .Omg/kg; ICN K&K Labs, Plainview, New York, U.S.A.) were dissolved in 0.9% saline and expressed as free base. They were administered subcutaneously in random sequence and locomotor activity was measured between 09:OO and 12:00 hr. Animals used in the experiments on brain stimulation were allowed at least a week to recover from surgery. Following this, one group was trained to lever-press for intracranial self-stimulation on a continuous reinforcement schedule until response rates stabilized. Animals were then given either saline (Mondays and Thursdays) or nicotine (Tuesdays and Fridays) 15 min prior to being tested for I5 min. Injections were administered subcutaneously in a random sequence. After completing the continuous reinforcement study, the animals were trained on a fixed-ratio schedule with a final ratio of 15. When the rates had stabilized. the effects of nicotine were re-tested during 15min sessions. One of the 8 rats used in the continuous reinforcement study could not be retrained on the fixed-ratio: 15 schedule. The animals tested in the auto-titration experiment were first trained to press the conventional lever which produced intracranial self-stimulation on a continuous reinforcement schedule. Every 15th leverpress reduced the current by 3 PA. When the current was decreased to a value that would not support lever-pressing, the animals were trained to move to the back of the cage and press an omnidirectional lever. This lever re-set the current to the initial, suprathreshold value and also generated a brief tone from a Sonalert loudspeaker. It did not, however, produce stimulation of the brain. Training continued until the re-set current or threshold for each animal had clearly stabilized. Starting currents ranged be-
self-stimulation
127
tween 93 and 126itA. Testing with saline and nicotine followed the same procedure as for the continuous reinforcement study described above. A detailed description of the training and testing procedures for the detection threshold study has been published elsewhere (Schaefer and Michael, 1986). Briefly, animals were trained in a discrete trial procedure to make a right or left lever press after a single response on the initiating lever. The first response on the initiating lever produced a I-set tone from a Sonalert loudspeaker and on 50% of trials an electrical stimulus. When stimulation occurred, the animal was required to press the right choice lever to obtain an additional stimulus; when stimulation did not occur, the animal was required to press the left choice lever to obtain a stimulus. The first press on either choice lever terminated the trail. Stimulation was, therefore, available on one of two choice levers on each trial and the animal learned to select the appropriate choice lever according to the outcome of the first press of the initiating lever; the position of the appropriate choice lever was reversed for half of the animals. The beginning of a trial was signalled by the onset of a light in the chamber and was terminated by the completion of the two-response chain or the end of a session. A 5 set intertrial interval was used and between trials the test chamber was dark. The animals received 80 training trials per day, until they reached 95% accuracy (76 out of 80) on 4 consecutive daily sessions. Tests with nicotine and saline were then conducted and these sessions differed from training sessions in two respects. First, both levers were activated. This prevented the first trial of a S-trial block from indicating which lever was “correct” for that block. Second, while the initiating lever produced either 0 or 100% current during training sessions, during test sessions the initiating lever produced either 0-IO-2040-60-80 or 100% of the current available on the choice lever. Each current was tested for 5 consecutive trials which constituted one test block and the order of current intensities was randomized. Animals were tested twice with each dose of nicotine and the results were averaged. Da& ana(ysis The data for locomotor activity were analyzed by averaging the scores for the 4 days on saline which preceded a test day with nicotine and expressing the scores for each dose of nicotine as a percentage of the mean score for saline. An analysis of variance was used to assess the significance of differences between activity scores (Kirk, 1968) followed by Dunnett’s test (two-tailed) to compare differences between activity scores after saline and after different doses of nicotine. The total number of lever presses made during the 15min test session provided the data for the continuous reinforcement and fixed-ratio: 15 experiments. These response rates were presented as a percentage of the rate of responding on days on
12x
G.
J.
and R. P.
SCHAEFER
vehicle and analyses of variance and Dunnett’s test were used to assess results. In the experiments on reward threshold, the data consisted of (I) the total number of lever presses on the stimulation (conventional) lever, (2) the total number of presses on the re-set (omnidirectional) lever and (3) the value of the current intensity at which it was re-set to a suprathreshold level for each titration series. The average intensity of current at which the re-set lever was activated during each titration series defined the reinforcement threshold. The value of the threshold for reinforcement produced by each dose of nicotine was compared with the threshold value for the previous day on saline. The same procedure was used to calculate the effects of nicotine on changes in lever pressing on the stimulation lever and on the re-set lever. Analyses of variance and Dunnett’s test were then performed on the threshold values, on the response rates and on the number of re-sets. In the detection threshold experiment, the number of trials completed on the intracranial selfstimulation appropriate lever was recorded for each block of 5 trials at each intensity of current on the initiating lever. For saline and each dose of nicotine the intensity of current was plotted against the number of trials completed on the intracranial selfstimulation correct lever. The method of Litchfield and Wilcoxon (1949) was then applied to determine whether any dose of nicotine altered the detection threshold as shown by a significant shift in the position of the curves from the control curve for saline. The cumulative latencies to complete the 5-trial test blocks as well as the number of intertrial interval responses on the choice levers were also analyzed by an analysis of variance. RESULTS
The acute effects of nicotine were studied over a 30-fold range of doses. In the locomotor activity paradigm (Fig. I), a dose-dependent decrease in activity occurred. A slight, non-significant decrease in activity occurred at the smallest dose, followed by a return to baseline at the next larger dose. As the dose was increased further, locomotor activity declined to 22% of the level with saline (F = 26.6, df= 4, 20, P < 0.001). Nicotine produced different effects on response the continuous reinforcement and rates in fixed-ratio: IS studies (Fig. 2). When reward was available for every lever-press, there were virtually no changes in lever pressing over the 0.03-0.3 mg/kg dose of nicotine. At the 1.0 mg/kg dose, a moderate decrease to 60% of control values was observed (F = 9.7, df’= 4, 28, P < 0.001). When intracranial self-stimulation was available only on every 15th lever press, a very different pattern of responding to nicotine moderate
occurred. increase
The
0.03 mg/kg
(30%)
in
dose
response
produced rates
which
a
MICHAEL
Dose
(mg/kg)
Fig. I. Effects of nicotine on spontaneous locomotor activity during IO-min test sessions (n = 6). Vertical bars give standard errors of means (SEM) in this and subsequent Figures. Horizontal interrupted lines give control values when the vehicle alone was administered. The mean value If: SEM for all control days for spontaneous locomotor activity counts was: 1477 & 97. Significantly different from control: *P < 0.05; **p < 0.01.
were further
increased to 56% above control levels at the 0.1 mg/kg dose. The 0.3 mg/kg dose produced a sharp drop in rates to below baseline levels and rates were further suppressed to only 7% of the control values at the I .O mg/kg dose (F = 20.4, df= 4, 24, P < 0.001). Thus, with a fixed-ratio schedule, a biphasic effect was clearly observed. The mean baseline reinforcement and the continuous rates in fixed-ratio: I5 experiments were not significantly different from each other (1230 + 97 vs 903 k 63, t = 1.77, NS), but it was observed that the animal with the smallest baseline rate showed the greatest increase in response rate with 0.1 mg/kg of nicotine. Figure 3 shows the data for the reinforcementthreshold experiment. Panel A shows that the changes in response rates were similar to those in the
(A)
(El
CRF
FR:15
I
-G z 160-
T
5 >
OL, 0.03
, 0.10
, 0.30
, 100 Dose
,
L
0.03 (mg/
,
,
1..
010
030
1.00
kg)
Fig. 2. Effects of nicotine on the rate of lever-pressing for intracranial self-stimulation during IS-min test sessions when reinforcement was provided on a continuous reinforcement schedule (CRF; A, n = 8). and on a fixed-ratio: 15 (FR: 15) schedule (B, n = 7). The mean value k SEM for all control days for lever-pressing was: CRF 1230 + 97, FR: 15 903 & 163. Other symbols as in Fig. I.
Nicotine and brain self-stimulation
(8)
129
the auto-titration
Threshold
experiment, and the 5 animals used
in the detection threshold material was not available
experiment.
Histological
for 2 animals. The electrode tips terminated in or near the lateral ‘hypothalamus or the zona incerta. Within the range of electrode placements found here, there were no differences in the intensities of current needed to maintain lever pressing or in the response to nicotine. DISCUSSION
L
OL 0.03
I 0.10
I 0.30
i 1.00
I 003
I 0.10
I 030
1 100
Dose (mg/kg) Fig. 3. Effects of nicotine on auto-titration, intracranial self-stimulation responding during 15-mitt test sessions. A shows the results for the rate of responding on the stimulation lever during the test sessions. B shows the results for the reinforcemeni threshold, which was defined as the average current intensity at which the animal pressed the re-set lever (n = 7). The mean value + SEM for all control days for responding on the stimulation lever was 2062 k 248, and the mean reinforcement threshold was 85. I + 5.8 PA. Other symbols as in Fig. I. continuous reinforcement study. A slight increase at the 0.1 mg/kg dose was followed by a decrease to 44% of baseline at the 1.O mg/kg dose (F = 4.6, d’ 4,
24, P < 0.01). The number of re-sets per session followed a similar pattern to the data for response rate and are not presented here. Panel B shows the effects of nicotine on the thresholds for reinforcement. It can be seen that these remained remarkably constant throughout; however, there was a small (8%) but non-significant decrease in the reinforcement threshold at the 1.Omg/kg dose of nicotine. The overall analysis of variance just reached significance (F = 2.8, <#; 4, 24, P < 0.05). Figure 4 shows the data for the detection threshold experiment. The response curves produced by each dose of nicotine were parallel to that produced by saline and none of the doses resulted in a significant shift in either direction from treatment with saline. Thus, the detection thresholds were not altered by nicotine. However, in this paradigm it was not possible to test the I .Omg/kg dose because the animals stopped responding. Therefore, an intermediate dose of 0.56 mg/kg was tested and an increase in responding on the intracranial self-stimulation correct lever occurred when the intensity of current on the initiating lever was 20 and 40% of the training current. This is represented by a slight shift to the left (square symbols) in Fig. 4, and a trend towards a lower detection threshold might have occurred. No dose of nicotine significantly altered either the response latencies or the number of intertrial interval responses and the data are not shown.
Figure 5 shows the sites of the electrode tips for 7 of the 8 animals used in the continuous reinforcement and fixed-ratio: 15 studies, 6 of the 7 animals used in
These studies have demonstrated that the acute effects of nicotine, like those of other psychoactive drugs, are dependent on the schedule of intracranial self-stimulation reinforcement employed. With continuous reinforcement, no effects on lever-pressing response rates were observed until the largest dose (1 .Omgjkg) was used. when a significant depression occurred; however, it was clear that this dose had a behaviorally-disrupting effect. In contrast, with the fixed-ratio : 15 schedule a marked stimulation (approx. 60%) of lever pressing occurred followed, at larger doses, by a steep decline to only 7% of baseline values (Fig. 2). What is the explanation for this clear-cut biphasic effect in the fixed-ratio schedule? There were no changes in locomotor activity at these small doses (Fig. I), and no changes in thresholds for reinforcement (Fig. 3), but the stimulus after acute administration of nicotine must, according to operant conditioning theory, be regarded as more reinforcing because the response rates increased by 60%. The continuous reinforcement schedule is apparently in-
z
I
1
1
0
10
20
I 40
I 60
1 80
-J 100
Current on initiating lever (X training current 1 Fig. 4. Effects of nicotine (O.l~.56mg/kg) and saline on choice responding in animals trained to discriminate between the presence and absence of stimulation of the brain. The ordinate shows the number of trials out of a maximum of five that were completed on the choice lever appropriate for brain stimulation. The abscissa indicates the current produced by the initiating lever which has been converted to a percentage of that available during training sessions. Each point represents the mean of two observations in each of the 5 animals: 0 = saline: 0. = 0.1 mg/kg; 0 = 0.3 mgjkg; m = 0.56 mg/kg. ICSS = i~tmc~dnial selfstimulation.
Ci.
130
J.
SCHAEFEKand
R. P.
MICHAEL
5.6
4.6
Fig. 5. Reconstruction of the location of the electrode placements for the animals used in the intracranial self-stimulation experiments. The symbols refer to the electrode placements for animals used in each of the three paradigms: A = continuous reinforcement and fixed-ratio: IS; l = auto-titration; n = detection thresholds. These sections were adopted from Pellegrino et al. (1979). Numbers to the left of the sections indicate the anterior-posterior location of the section relative to interaural plane. Abbreviations: RE-nucleus reunies thalami; RH-nucleus rhomboideus thalami; MT-mamillothalamic tract; FXfornix; VMH-ventromedial nucleus of hypothalamus; D~H~orsomedial nucleus of hypothalamus; PMV-ventral premamillary nucleus; PH-posterior nucleus of hypothalamus; ARH-arcuate nudeus of the hypothalamus.
sensitive to the effect of nicotine on self-stimulation in the brain in this paradigm which must clearly be quite a subtle one. With other types of reinforcement, such as food and water, the effects of nicotine also depend heavily on the schedule of reinforcement employed (Morrison, 1967; Morrison and Armitage, 1967; Hendry and Rosecrans, 1982). It should be noted that baseline values were some 25% lower in the fixed-ratio: I5 group than in the continuous reinforcement group, which might account for some part of the greater increase occurring in the fixed-ratio: I5 group, but it is felt that this is unlikely to be the whole explanation, particularly because this group also showed a greater decrease in response rates at larger doses; indeed, the whole pattern of responding was different in the two groups. However, the different frequencies with which rewards are provided has also been found to be a critical factor in determining the response patterns to narcotic antagonists such as naloxone (West, Schaefer and Michael, 1983). The importance of the role played by the frequency of reward may help in understanding the strength of the nicotine habit itself and this has been emphasized by studies on the self-administration of nicotine in squir-
rel monkeys (Goldberg, Spealman and Goldberg, 1981). In the auto-titration procedure the lever pressing of the animals was slowed down at larger doses, but this could not be accounted for in terms of a change in the threshold for reward. However, it could be accounted for by the generalized depression in locomotor activity at these doses. This contrasts with the effects of another CNS stimulant, namely, rtamphetamine, which produces marked decreases in reward thresholds (Schaefer and Holtzman, 1979). The lack of an acute effect of nicotine on reinforcement thresholds in no way precludes chronic effects similar to those described by Clarke and Kumar (1984). Since there were no changes in thresholds for reward (except perhaps at the 1.0 mg/kg dose), the present authors did not anticipate any changes in detection thresholds and it seems that nicotine did not in fact alter sensory mechanisms in such a way as to change the discriminative properties of intracranial self-stimulation, confirming the results of Clarke and Kumar (1983). These acute studies have now laid the ground-work for a comparison with the effects of the chronic administration of nicotine on intracranial self-stimulation which it is anticipated will have
Nicotine and brain self-stimulation predominantly stimulatory of behavior.
effects
and
cause
less
disruption
Acknow,led~ement-General by the Georgia Department is gratefully acknowledged.
muscarinic stimulatton
131 and nicotinic cholinergic behavior. J. fhcrrmuc.
agonists on selfesp. Ther. 166:
189-204. research support was provided of Human Resources and this
REFERENCES
Clarke P. B. S. and Kumar R. (1983) Nicotine does not improve discrimination of brain stimulation reward by rats. Psychopharmacology 19: 271-277. Clarke P. B. S. and Kumar R. (1984) Effects of nicotine and d-amphetamine on intracranial self-stimulation in a shuttle box test in rats. Psychopharmacology 84: 109-I 14. Goldberg S. R.. Spealman R. D. and Goldberg D. M. (I 98 I) Persistent behavior at high rates maintained by intravenous self-administration of nicotine. Science 214: 5733575. Hendry J. S. and Rosecrans J. A. (1982) Effects of nicotine on conditioned and unconditioned behaviors in experimental animals. Pharmac. Ther. 17: 431-454. Kirk R. E. (1968) Ecperimental Design: Procedures for the Behavioral Sciences. Brooks/Cole, Belmont, Calif. Litchfield J. T. Jr and Wilcoxon F. (1949) A simplified dose-effect experiments. J. method of evaluating Pharmac:. e-up. Ther. 96: 99-l 13. Morrison C. F. (1967) Effects of nicotine on operant behavior of rats. fnf. J. Neuroph~rmac, 6: 229-240. Morrison C. F. and Armitaee A. K. (1967) Effects of nicotine upon the free operant behavior of rats and spontaneous motor activity of mice. Ann. N. Y. Acad. Sci. 142: 268-276. Newman L. M. (1972) Effects of cholinergic agonists and antagonists on self-stimulation behavior in the rat. J. camp. physiol. Psychol. 19: 394413. Olds M. E. and Domino E. F. (1969) Comparison of
Pellegrino L. J.. Pellegrino A. S. and Cushman A. _I. (1979) A Stereoruxic Atlas qf rhe Rut Bruin. Plenum Press, New York. Pradhan S. N. and Bowling 6. (1971) Effects of nicotine on self-st~mulatjon in rats. J. Phurmu~. e_‘;p. Ther. 176: 2299243. Russell M. A. H. (1977) Smoking problems: An overview. In: Research on Smoking Behavior (Jarvik M. E., Cullen J. W., Gritz E. R., Vogt T. M. and West L. J., Eds), pp, 13-34. NIDA Research Monograph 17, Dept HEW, Washington, D.C. Schaefer G. J. and Holtzman S. G. (1979) Free-operant and auto-titration brain self-stimulation procedures in the rat: A comparison of drug effects, Pharmac. Binchem. fiehur. to: 127-135. Schaefer G. J. and Michael R. P. (1986) The di~riminative stimulus properties of and detection thresholds of intracranial self-stimulation: Effects of d-amphetamine, morphine, and haloperidol. Psychopharmacology. In press, Schaefer G. J., Baumgardner D. G. and Michael R. P. (1979) Constant-current, biphasic titrating stimulator for brain self-stimulation. Phy&/. Behav. 25 1217-1219. Schaefer G. J., Baumaardner D. G. and Michael R. P. (1981) A rugged an;d simple commutator for electrical stimulation of the brain of unrestrained animals. Physiol. Behav. 26: 319-321. Schaefer G. J., Bonsafi R. W. and Michael R. P. (1982) An easily constructed biphasic constant-current stimulator for intracranial self-stimulation. Ph~~ssiol. Behnr. 29: 163-165. West C. H. K., Schaefer G. J. and Michael R. P. (1983) Increasing the work requirements lowers the threshold of naloxone for reducing self-stimulation in the mid-brain of rats. Pharmac. Biochem. Behav. 18: 705-710.
TASK-SPECIFIC INTRACRANIAL
EFFECTS
OF NICOTINE
SELF-STIMULATION
I
002%39OXtXh $3.00 + 0.00 19X6 Pergamon Press Ltd
IN RATS
AND LOCOMOTOR
ACTIVITY
G. J. SCHAEFER and R. P. MICHAEL Department of Psychiatry, Emory University School of Medicine, Atlanta, GA 30322 and Georgia Mental Health institute, 1256 Briarcliff Road. N.E. Atlanta, GA 30306, U.S.A.
(Acceptecl 9 June 1985)
Summa~-The acute effects of nicotine (0.03-l .Omgjkg) were studied in a locomotor activity procedure and in a series of intracranial self-stimulation (ICSS) paradigms. Nicotine produced a dose-dependent decrease in locomotor activity. When animals were trained to lever-press for intracranial self-stimulation on a continuous reinforcement schedule (CRF), the drug was ineffective except at the l.Omg/kg dose, which produced a moderate decrease in the rate of responding. However, when animals were tested in a fixed-ratio: 15 (FR: 15) paradigm, nicotine produced a steep, biphasic dos+response curve. At the 0. I mg/kg dose, the response rates were increased to approx. 60% above baseline, while at the I ,Omg,‘kg dose, response rates were decreased to approx. 90% below baseline values. The effects of nicotine were also studied in an auto-titration procedure which measured the rewarding value of the stimulus. There was a decrease in performance at larger doses similar to that observed in the continuous reinforcement procedure. but there were no significant changes in the threshold for reinforcement. Nicotine did not produce any change in the detection threshold for stimulation of the brain. In acute studies, therefore, nicotine produced both stimu~tion and djsruption of behavior, effects that were brought to light by the fixed-ratio schedule of reinforcement, and this may relate to the rewarding effects of nicotine. Key words: nicotine, locomotor activity, intracranial self-stimulation.
Nicotine is a psychoactive drug that may be responsible for the positive reinforcing effects of tobacco smoking. Since smokers typically engage in the behavior that results in the intake of nicotine at a high rate (approx. 70,000 puffs per year; Russeli, 1977), the “nicotine habit” should be regarded as an extremely strong one. In an attempt to extend present knowledge of the nature of the reinforcement produced by nicotine, the effects of this drug on motivated behavior have been examined. One procedure used is that of intracranial self-stimulation (ICSS), but previous reports on the effects of nicotine on lever-pressing for intracranial self-stimulation have been inconsistent. OIds and Domino (1969) observed in rats that nicotine produced an initial decrease, followed by an increase in response rates, but noted that depressant effects predominated. Pradhan and Bowling (1971) reported that nicotine increased response rates, pa~~culariy in animals with low baselines, and these results were confirmed by Newman (1972) who used a variable-interval rather than a continuous reinforcement schedule (CRF). The many and varied pharmacological effects of nicotine may tend to obscure the mechanisms responsible for its reinforcing characteristics, and the possibility should be kept in mind that changes in operant performance
may not reflect authentic shifts in the reinforcement value of the stimulus; this is of particular concern
when response rates are the sole dependent measure of the action of the drug. The purpose of these experiments was to examine the acute effects of nicotine on response rates and on the central reward mechanisms that are thought to maintain the smoking habit. The effects of nicotine were tested, therefore, in a group of rats trained to lever-press for intracranial self-stimulation on a continuous reinforcement schedule, the paradigm typically used to test effects of drugs. These animals were then retrained to lever-press for intracranial selfstimulation on a fixed-ratio: 15 (FR: 15) schedule of reinforcement, and the dose-response effects of nicotine were again tested. The second procedure, the auto-titration paradigm, was used to measure thresholds for reinforcement, and simultaneously provided a rate-dependent and a rate-independent measure of behavior due to intracranial self-stimulation (Schaefer and Holtzman, 1979). The reinforcement threshold was obtained by calculating the mean current intensity at which the animals re-set the stimulating current to a suprathreshold value. The third procedure was that recently developed to study the stimuius control properties of intracranial selfstimulation (Schaefer and Michael, 1986). This is a procedure independent of rate that allows the investigator to measure the threshold for detection of the stimulus and to determine the extent to which 125
126
G. J.
SCHAEFFKand
drugs alter this threshold. Finally, animals were tested for changes in spontaneous locomotor activity. Comparing data from the different paradigms in one laboratory increases the possibility of obtaining convergent evidence concerning the role of nicotine in positive rcinforccment mechanisms.
Malt rats (335420 g) were used. They were bred in the present authors’ colony from stock purchased from King Animal Laboratories, Inc. (Oregon, Wisconsin, U.S.A.). Between experimental sessions, animals were housed in group cages (24 per cage) in a colony room and food and water were available ad lihiturn. The colony room was artificially illuminated between 07:OO and l9:OO hr.
Locomotor activity was measured using an OmniTech Digiscan RXY activity monitor (Columbus, Ohio, U.S.A.). The device measured horizontal activity by counting the total number of interruptions of an infra-red beam. To measure activity, an animal acrylic inside a clear cage was placed (39.4 x 39.4 x 30.5 cm high, inside dimensions) which rested inside the “main frame” of the Digiscan monitor. The entire monitoring device was placed inside a sound-attenuating chamber which was equipped with a fdn and a 25 W red light bulb. In addition to the number of interruptions of the beam being displayed on a digital electronic counter, the output from the monitor was interfaced with a Beckman strip-chart recorder. Pen deflections from baseline increased in proportion to the speed at which the light beams were interrupted, thus producing an analogue measure of the speed of movement during the monitoring session. The operant test chamber (31 x 30 x 29 cm high) used to measure lever-pressing on a continuous reinforcement and a fixed-ratio: I5 schedule was constructed in this laboratory. The chamber was equipped with a single conventional lever (G6312, Ralph Gerbrands, Co., Arlington, Massachusetts, U.S.A.) which could trigger stimulation of the brain on either a continuous or partial reinforcement schedule. Electrical pulses were produced by a biphasic constant current stimulator constructed in this laboratory (Schaefer, Bonsall and Michael, 1982) and were passed through a two-channel commutator (Schaefer, Baumgardner and Michael, 1981) to the brain. by way of a length of spring-shielded hearing aid wire (Plastic Products Co., Roanoke, Virginia, U.S.A.). The stimuli consisted of 200 msec trains of biphasic square-wave pulses at 100 Hz with a pulse duration of 2 msec (+ I msec and -I msec). The range of current intensities used for the studies of the continuous reinforcement schedule was 27-50 PA, while the range of intensities for the fixed-ratio:15 was 35-250 {LA.
R. P.
MICHAEL
The auto-titration procedure was used to measure reinforcement thresholds and experiments were conducted in an operant chamber similar to that used for the continuous reinforcement and fixed-ratio: I5 experiments In addition to a conventional lever, however, there was an omnidirectional lever (G6313, Gerbrands) suspended from the ceiling near the opposite wall of the chamber. The stimulator and titrator have been described in detail elsewhere (Schaefer, Baumgardner and Michael, 1979). The electrical stimuli were of the same parameters as those noted above except that the train duration was 100 msec. The current was programmed to drop 3 PA after each 15th lever-press and the current was held constant at each step. When the omnidirectional lever was pressed, the starting current was re-set and a new titration series started. The operant chambers for the experiments on intracranial self-stimulation were controlled by a combination of +28 and +5 VDC modules. In addition, cumulative recorders were interfaced with the programming equipment and these showed response rates and reinforcement delivery for the continuous reinforcement and the fixed-ratio: I5 experiments and also response rates and current drop for each titration series during the auto-titration experiment. The test chamber used in the detection threshold experiment was 30.5 cm long, 31 cm wide and 45 cm high (see Schaefer and Michael, 1986). On the front wall were positioned two conventional levers (G6312, Gerbrands), separated by a Plexiglas partition mounted perpendicular to the wall. These two levers were designated the “choice levers”. On the opposite wall an omnidirectional lever (G6313) was mounted and was designated the “initiating lever”. Programming for this study was controlled by +5 VDC modules. A biphasic constant current stimulator (Schaefer et al., 1982) modified to produce two different stimulus trains produced the electrical pulses. The stimulus produced by lever presses on the choice levers was a 500-msec train of 100 Hz with a pulse duration of 2 msec and a range of current intensities from 75 to 225 PA. The stimulus train produced by the initiating lever was the same as that produced by the choice lever except for the intensity, which was a proportional amount (O-100%) of that produced by the choice levers. Surgery and histology Rats used in experiments on brain stimulation were anesthetized with sodium pentobarbital (50 mg/kg, i.p.), and given atropine sulfate (0.25 mg, s.c.) to reduce any respiratory distress. The animals were then positioned in a stereotaxic device, and after exposing the skull, 4-5 stainless-steel screws were fixed to the skull. Next, a hole was drilled in the skull, the dura was incised, and a bipolar platinum electrode (tip diameter = 0. I25 mm, Plastic Products Co.) was lowered into the lateral hypothalamus (medial forebrain bundle) using coordinates: AP 5.2,
Nicotine
and brain
L 1.7, H -2.2 (Pellegrino, Pellegrino and Cushman, 1979). To form a secure anchor, crania-plastic cement was applied to the screws and electrodes. Finally, the animals were given an intramuscular injection of 100,000 U of benzathine penicillin G and procaine penicillin G. When the experiments were completed. the animals were killed with sodium pentobarbital and perfused via the heart with 10% formalin. After fixation, frozen sections of brain were cut at 50 p and alternate sections were stained with cresyl violet and W&l’s stain. From this material, the electrode tips were accurately located. Proc’edurcs To test locomotor activity, the animals were first habituated to the procedure for 5 days by injecting them with saline and I5 min later placing them in the locomotor apparatus for 12 min. Testing of drugs was then begun. On Mondays and Thursdays the animals were given saline, and on Tuesdays and Fridays they were given different doses of nicotine tartrate. Data from the first 2 min of the test were discarded and the total number of interruptions of the infra-red beam during the next 10 min were used for analysis. The doses of nicotine tartrate (0.03, 0. I, 0.3 and I .Omg/kg; ICN K&K Labs, Plainview, New York, U.S.A.) were dissolved in 0.9% saline and expressed as free base. They were administered subcutaneously in random sequence and locomotor activity was measured between 09:OO and 12:00 hr. Animals used in the experiments on brain stimulation were allowed at least a week to recover from surgery. Following this, one group was trained to lever-press for intracranial self-stimulation on a continuous reinforcement schedule until response rates stabilized. Animals were then given either saline (Mondays and Thursdays) or nicotine (Tuesdays and Fridays) 15 min prior to being tested for I5 min. Injections were administered subcutaneously in a random sequence. After completing the continuous reinforcement study, the animals were trained on a fixed-ratio schedule with a final ratio of 15. When the rates had stabilized. the effects of nicotine were re-tested during 15min sessions. One of the 8 rats used in the continuous reinforcement study could not be retrained on the fixed-ratio: 15 schedule. The animals tested in the auto-titration experiment were first trained to press the conventional lever which produced intracranial self-stimulation on a continuous reinforcement schedule. Every 15th leverpress reduced the current by 3 PA. When the current was decreased to a value that would not support lever-pressing, the animals were trained to move to the back of the cage and press an omnidirectional lever. This lever re-set the current to the initial, suprathreshold value and also generated a brief tone from a Sonalert loudspeaker. It did not, however, produce stimulation of the brain. Training continued until the re-set current or threshold for each animal had clearly stabilized. Starting currents ranged be-
self-stimulation
127
tween 93 and 126itA. Testing with saline and nicotine followed the same procedure as for the continuous reinforcement study described above. A detailed description of the training and testing procedures for the detection threshold study has been published elsewhere (Schaefer and Michael, 1986). Briefly, animals were trained in a discrete trial procedure to make a right or left lever press after a single response on the initiating lever. The first response on the initiating lever produced a I-set tone from a Sonalert loudspeaker and on 50% of trials an electrical stimulus. When stimulation occurred, the animal was required to press the right choice lever to obtain an additional stimulus; when stimulation did not occur, the animal was required to press the left choice lever to obtain a stimulus. The first press on either choice lever terminated the trail. Stimulation was, therefore, available on one of two choice levers on each trial and the animal learned to select the appropriate choice lever according to the outcome of the first press of the initiating lever; the position of the appropriate choice lever was reversed for half of the animals. The beginning of a trial was signalled by the onset of a light in the chamber and was terminated by the completion of the two-response chain or the end of a session. A 5 set intertrial interval was used and between trials the test chamber was dark. The animals received 80 training trials per day, until they reached 95% accuracy (76 out of 80) on 4 consecutive daily sessions. Tests with nicotine and saline were then conducted and these sessions differed from training sessions in two respects. First, both levers were activated. This prevented the first trial of a S-trial block from indicating which lever was “correct” for that block. Second, while the initiating lever produced either 0 or 100% current during training sessions, during test sessions the initiating lever produced either 0-IO-2040-60-80 or 100% of the current available on the choice lever. Each current was tested for 5 consecutive trials which constituted one test block and the order of current intensities was randomized. Animals were tested twice with each dose of nicotine and the results were averaged. Da& ana(ysis The data for locomotor activity were analyzed by averaging the scores for the 4 days on saline which preceded a test day with nicotine and expressing the scores for each dose of nicotine as a percentage of the mean score for saline. An analysis of variance was used to assess the significance of differences between activity scores (Kirk, 1968) followed by Dunnett’s test (two-tailed) to compare differences between activity scores after saline and after different doses of nicotine. The total number of lever presses made during the 15min test session provided the data for the continuous reinforcement and fixed-ratio: 15 experiments. These response rates were presented as a percentage of the rate of responding on days on
12x
G.
J.
and R. P.
SCHAEFER
vehicle and analyses of variance and Dunnett’s test were used to assess results. In the experiments on reward threshold, the data consisted of (I) the total number of lever presses on the stimulation (conventional) lever, (2) the total number of presses on the re-set (omnidirectional) lever and (3) the value of the current intensity at which it was re-set to a suprathreshold level for each titration series. The average intensity of current at which the re-set lever was activated during each titration series defined the reinforcement threshold. The value of the threshold for reinforcement produced by each dose of nicotine was compared with the threshold value for the previous day on saline. The same procedure was used to calculate the effects of nicotine on changes in lever pressing on the stimulation lever and on the re-set lever. Analyses of variance and Dunnett’s test were then performed on the threshold values, on the response rates and on the number of re-sets. In the detection threshold experiment, the number of trials completed on the intracranial selfstimulation appropriate lever was recorded for each block of 5 trials at each intensity of current on the initiating lever. For saline and each dose of nicotine the intensity of current was plotted against the number of trials completed on the intracranial selfstimulation correct lever. The method of Litchfield and Wilcoxon (1949) was then applied to determine whether any dose of nicotine altered the detection threshold as shown by a significant shift in the position of the curves from the control curve for saline. The cumulative latencies to complete the 5-trial test blocks as well as the number of intertrial interval responses on the choice levers were also analyzed by an analysis of variance. RESULTS
The acute effects of nicotine were studied over a 30-fold range of doses. In the locomotor activity paradigm (Fig. I), a dose-dependent decrease in activity occurred. A slight, non-significant decrease in activity occurred at the smallest dose, followed by a return to baseline at the next larger dose. As the dose was increased further, locomotor activity declined to 22% of the level with saline (F = 26.6, df= 4, 20, P < 0.001). Nicotine produced different effects on response the continuous reinforcement and rates in fixed-ratio: IS studies (Fig. 2). When reward was available for every lever-press, there were virtually no changes in lever pressing over the 0.03-0.3 mg/kg dose of nicotine. At the 1.0 mg/kg dose, a moderate decrease to 60% of control values was observed (F = 9.7, df’= 4, 28, P < 0.001). When intracranial self-stimulation was available only on every 15th lever press, a very different pattern of responding to nicotine moderate
occurred. increase
The
0.03 mg/kg
(30%)
in
dose
response
produced rates
which
a
MICHAEL
Dose
(mg/kg)
Fig. I. Effects of nicotine on spontaneous locomotor activity during IO-min test sessions (n = 6). Vertical bars give standard errors of means (SEM) in this and subsequent Figures. Horizontal interrupted lines give control values when the vehicle alone was administered. The mean value If: SEM for all control days for spontaneous locomotor activity counts was: 1477 & 97. Significantly different from control: *P < 0.05; **p < 0.01.
were further
increased to 56% above control levels at the 0.1 mg/kg dose. The 0.3 mg/kg dose produced a sharp drop in rates to below baseline levels and rates were further suppressed to only 7% of the control values at the I .O mg/kg dose (F = 20.4, df= 4, 24, P < 0.001). Thus, with a fixed-ratio schedule, a biphasic effect was clearly observed. The mean baseline reinforcement and the continuous rates in fixed-ratio: I5 experiments were not significantly different from each other (1230 + 97 vs 903 k 63, t = 1.77, NS), but it was observed that the animal with the smallest baseline rate showed the greatest increase in response rate with 0.1 mg/kg of nicotine. Figure 3 shows the data for the reinforcementthreshold experiment. Panel A shows that the changes in response rates were similar to those in the
(A)
(El
CRF
FR:15
I
-G z 160-
T
5 >
OL, 0.03
, 0.10
, 0.30
, 100 Dose
,
L
0.03 (mg/
,
,
1..
010
030
1.00
kg)
Fig. 2. Effects of nicotine on the rate of lever-pressing for intracranial self-stimulation during IS-min test sessions when reinforcement was provided on a continuous reinforcement schedule (CRF; A, n = 8). and on a fixed-ratio: 15 (FR: 15) schedule (B, n = 7). The mean value k SEM for all control days for lever-pressing was: CRF 1230 + 97, FR: 15 903 & 163. Other symbols as in Fig. I.
Nicotine and brain self-stimulation
(8)
129
the auto-titration
Threshold
experiment, and the 5 animals used
in the detection threshold material was not available
experiment.
Histological
for 2 animals. The electrode tips terminated in or near the lateral ‘hypothalamus or the zona incerta. Within the range of electrode placements found here, there were no differences in the intensities of current needed to maintain lever pressing or in the response to nicotine. DISCUSSION
L
OL 0.03
I 0.10
I 0.30
i 1.00
I 003
I 0.10
I 030
1 100
Dose (mg/kg) Fig. 3. Effects of nicotine on auto-titration, intracranial self-stimulation responding during 15-mitt test sessions. A shows the results for the rate of responding on the stimulation lever during the test sessions. B shows the results for the reinforcemeni threshold, which was defined as the average current intensity at which the animal pressed the re-set lever (n = 7). The mean value + SEM for all control days for responding on the stimulation lever was 2062 k 248, and the mean reinforcement threshold was 85. I + 5.8 PA. Other symbols as in Fig. I. continuous reinforcement study. A slight increase at the 0.1 mg/kg dose was followed by a decrease to 44% of baseline at the 1.O mg/kg dose (F = 4.6, d’ 4,
24, P < 0.01). The number of re-sets per session followed a similar pattern to the data for response rate and are not presented here. Panel B shows the effects of nicotine on the thresholds for reinforcement. It can be seen that these remained remarkably constant throughout; however, there was a small (8%) but non-significant decrease in the reinforcement threshold at the 1.Omg/kg dose of nicotine. The overall analysis of variance just reached significance (F = 2.8, <#; 4, 24, P < 0.05). Figure 4 shows the data for the detection threshold experiment. The response curves produced by each dose of nicotine were parallel to that produced by saline and none of the doses resulted in a significant shift in either direction from treatment with saline. Thus, the detection thresholds were not altered by nicotine. However, in this paradigm it was not possible to test the I .Omg/kg dose because the animals stopped responding. Therefore, an intermediate dose of 0.56 mg/kg was tested and an increase in responding on the intracranial self-stimulation correct lever occurred when the intensity of current on the initiating lever was 20 and 40% of the training current. This is represented by a slight shift to the left (square symbols) in Fig. 4, and a trend towards a lower detection threshold might have occurred. No dose of nicotine significantly altered either the response latencies or the number of intertrial interval responses and the data are not shown.
Figure 5 shows the sites of the electrode tips for 7 of the 8 animals used in the continuous reinforcement and fixed-ratio: 15 studies, 6 of the 7 animals used in
These studies have demonstrated that the acute effects of nicotine, like those of other psychoactive drugs, are dependent on the schedule of intracranial self-stimulation reinforcement employed. With continuous reinforcement, no effects on lever-pressing response rates were observed until the largest dose (1 .Omgjkg) was used. when a significant depression occurred; however, it was clear that this dose had a behaviorally-disrupting effect. In contrast, with the fixed-ratio : 15 schedule a marked stimulation (approx. 60%) of lever pressing occurred followed, at larger doses, by a steep decline to only 7% of baseline values (Fig. 2). What is the explanation for this clear-cut biphasic effect in the fixed-ratio schedule? There were no changes in locomotor activity at these small doses (Fig. I), and no changes in thresholds for reinforcement (Fig. 3), but the stimulus after acute administration of nicotine must, according to operant conditioning theory, be regarded as more reinforcing because the response rates increased by 60%. The continuous reinforcement schedule is apparently in-
z
I
1
1
0
10
20
I 40
I 60
1 80
-J 100
Current on initiating lever (X training current 1 Fig. 4. Effects of nicotine (O.l~.56mg/kg) and saline on choice responding in animals trained to discriminate between the presence and absence of stimulation of the brain. The ordinate shows the number of trials out of a maximum of five that were completed on the choice lever appropriate for brain stimulation. The abscissa indicates the current produced by the initiating lever which has been converted to a percentage of that available during training sessions. Each point represents the mean of two observations in each of the 5 animals: 0 = saline: 0. = 0.1 mg/kg; 0 = 0.3 mgjkg; m = 0.56 mg/kg. ICSS = i~tmc~dnial selfstimulation.
Ci.
130
J.
SCHAEFEKand
R. P.
MICHAEL
5.6
4.6
Fig. 5. Reconstruction of the location of the electrode placements for the animals used in the intracranial self-stimulation experiments. The symbols refer to the electrode placements for animals used in each of the three paradigms: A = continuous reinforcement and fixed-ratio: IS; l = auto-titration; n = detection thresholds. These sections were adopted from Pellegrino et al. (1979). Numbers to the left of the sections indicate the anterior-posterior location of the section relative to interaural plane. Abbreviations: RE-nucleus reunies thalami; RH-nucleus rhomboideus thalami; MT-mamillothalamic tract; FXfornix; VMH-ventromedial nucleus of hypothalamus; D~H~orsomedial nucleus of hypothalamus; PMV-ventral premamillary nucleus; PH-posterior nucleus of hypothalamus; ARH-arcuate nudeus of the hypothalamus.
sensitive to the effect of nicotine on self-stimulation in the brain in this paradigm which must clearly be quite a subtle one. With other types of reinforcement, such as food and water, the effects of nicotine also depend heavily on the schedule of reinforcement employed (Morrison, 1967; Morrison and Armitage, 1967; Hendry and Rosecrans, 1982). It should be noted that baseline values were some 25% lower in the fixed-ratio: I5 group than in the continuous reinforcement group, which might account for some part of the greater increase occurring in the fixed-ratio: I5 group, but it is felt that this is unlikely to be the whole explanation, particularly because this group also showed a greater decrease in response rates at larger doses; indeed, the whole pattern of responding was different in the two groups. However, the different frequencies with which rewards are provided has also been found to be a critical factor in determining the response patterns to narcotic antagonists such as naloxone (West, Schaefer and Michael, 1983). The importance of the role played by the frequency of reward may help in understanding the strength of the nicotine habit itself and this has been emphasized by studies on the self-administration of nicotine in squir-
rel monkeys (Goldberg, Spealman and Goldberg, 1981). In the auto-titration procedure the lever pressing of the animals was slowed down at larger doses, but this could not be accounted for in terms of a change in the threshold for reward. However, it could be accounted for by the generalized depression in locomotor activity at these doses. This contrasts with the effects of another CNS stimulant, namely, rtamphetamine, which produces marked decreases in reward thresholds (Schaefer and Holtzman, 1979). The lack of an acute effect of nicotine on reinforcement thresholds in no way precludes chronic effects similar to those described by Clarke and Kumar (1984). Since there were no changes in thresholds for reward (except perhaps at the 1.0 mg/kg dose), the present authors did not anticipate any changes in detection thresholds and it seems that nicotine did not in fact alter sensory mechanisms in such a way as to change the discriminative properties of intracranial self-stimulation, confirming the results of Clarke and Kumar (1983). These acute studies have now laid the ground-work for a comparison with the effects of the chronic administration of nicotine on intracranial self-stimulation which it is anticipated will have
Nicotine and brain self-stimulation predominantly stimulatory of behavior.
effects
and
cause
less
disruption
Acknow,led~ement-General by the Georgia Department is gratefully acknowledged.
muscarinic stimulatton
131 and nicotinic cholinergic behavior. J. fhcrrmuc.
agonists on selfesp. Ther. 166:
189-204. research support was provided of Human Resources and this
REFERENCES
Clarke P. B. S. and Kumar R. (1983) Nicotine does not improve discrimination of brain stimulation reward by rats. Psychopharmacology 19: 271-277. Clarke P. B. S. and Kumar R. (1984) Effects of nicotine and d-amphetamine on intracranial self-stimulation in a shuttle box test in rats. Psychopharmacology 84: 109-I 14. Goldberg S. R.. Spealman R. D. and Goldberg D. M. (I 98 I) Persistent behavior at high rates maintained by intravenous self-administration of nicotine. Science 214: 5733575. Hendry J. S. and Rosecrans J. A. (1982) Effects of nicotine on conditioned and unconditioned behaviors in experimental animals. Pharmac. Ther. 17: 431-454. Kirk R. E. (1968) Ecperimental Design: Procedures for the Behavioral Sciences. Brooks/Cole, Belmont, Calif. Litchfield J. T. Jr and Wilcoxon F. (1949) A simplified dose-effect experiments. J. method of evaluating Pharmac:. e-up. Ther. 96: 99-l 13. Morrison C. F. (1967) Effects of nicotine on operant behavior of rats. fnf. J. Neuroph~rmac, 6: 229-240. Morrison C. F. and Armitaee A. K. (1967) Effects of nicotine upon the free operant behavior of rats and spontaneous motor activity of mice. Ann. N. Y. Acad. Sci. 142: 268-276. Newman L. M. (1972) Effects of cholinergic agonists and antagonists on self-stimulation behavior in the rat. J. camp. physiol. Psychol. 19: 394413. Olds M. E. and Domino E. F. (1969) Comparison of
Pellegrino L. J.. Pellegrino A. S. and Cushman A. _I. (1979) A Stereoruxic Atlas qf rhe Rut Bruin. Plenum Press, New York. Pradhan S. N. and Bowling 6. (1971) Effects of nicotine on self-st~mulatjon in rats. J. Phurmu~. e_‘;p. Ther. 176: 2299243. Russell M. A. H. (1977) Smoking problems: An overview. In: Research on Smoking Behavior (Jarvik M. E., Cullen J. W., Gritz E. R., Vogt T. M. and West L. J., Eds), pp, 13-34. NIDA Research Monograph 17, Dept HEW, Washington, D.C. Schaefer G. J. and Holtzman S. G. (1979) Free-operant and auto-titration brain self-stimulation procedures in the rat: A comparison of drug effects, Pharmac. Binchem. fiehur. to: 127-135. Schaefer G. J. and Michael R. P. (1986) The di~riminative stimulus properties of and detection thresholds of intracranial self-stimulation: Effects of d-amphetamine, morphine, and haloperidol. Psychopharmacology. In press, Schaefer G. J., Baumgardner D. G. and Michael R. P. (1979) Constant-current, biphasic titrating stimulator for brain self-stimulation. Phy&/. Behav. 25 1217-1219. Schaefer G. J., Baumaardner D. G. and Michael R. P. (1981) A rugged an;d simple commutator for electrical stimulation of the brain of unrestrained animals. Physiol. Behav. 26: 319-321. Schaefer G. J., Bonsafi R. W. and Michael R. P. (1982) An easily constructed biphasic constant-current stimulator for intracranial self-stimulation. Ph~~ssiol. Behnr. 29: 163-165. West C. H. K., Schaefer G. J. and Michael R. P. (1983) Increasing the work requirements lowers the threshold of naloxone for reducing self-stimulation in the mid-brain of rats. Pharmac. Biochem. Behav. 18: 705-710.