useful for this purpose (Doty 1969; Colpaert et al. 1977;. Stutz and Maroli ~978; Kornetsky and Esposito 1981;. Schaefer and Michaei 1985; Druhan etal. 1987b).
Psychopharmacology
Psychopharmacology (].989) 97:331-338
9 Springer-Verlag 1989
Differential effects of cholinergic drugs on discriminative cues and self-stimulation produced by electrical stimulation of the ventral tegmental area J.P. Druhan ~, H.C. Fibiger 2, and A.G. Phil]gps ~ Department of Psychology and 2 Division of Neurological Sciences, Department of Psychiatry, University of British Columbia, Vancouver, B.C., Canada V6T 1Y7
Abstract. Cholinergic receptors have been shown to modulate a subset of discriminative cues produced by electrical stimulation of the ventral tegmental area (VTA) in rats. The present study identified the specific cholinergic receptor type modulating these electrical brain-stimu!ation (EBS) cues, and assessed whether these receptors also mediated the rewarding effects of VTA EBS. The EBS cues were enhanced by the acetylcholinesterase inhibitor physostigmine and the muscarinic receptor agonists pilocarpine and RS-86, whereas the nicotinic receptor agonist nicotine had no effect. The enhancing effects of pilocarpine or RS-86 were attenuated by the muscarinic antagonist scopolamine. The EBS cues were not affected when scopolamine was injected aione, although high doses disrupted discriminated responses. Intracranial self-stimulation (ICSS) rates were depressed by physostigmine and pilocarpine and increased by nicotine and scopolamine. These results indicated a facilitatory influence of muscarinic receptors cn the EBS cues, and an inhibitory role in VTA !CSS. Nicotinic receptor activation did not affect the EBS cues, but facilitated ICSS. These differential effects of cholinergic receptor activation point to a dissociation of the specific EBS cues measured in this study from the rewarding effects of VTA stimulation.
Key words: Cue properties - Electrical brain stimulation Self stimulation - Acetylcholine tinic receptors Rats
Muscarinic and nico-
Electrical brain-stimu]ation (EBS) has frequently been employed to investigate the neurobiological basis of motivation and emotion in laboratory animals. For example, numerous studies have exploited the rewarding effects of lateral hypothalamic (LH) and ventral tegmentat area (VTA) EBS to determine the roles of specific neurotransmitter systems in mediating positive affect (for reviews see: Wise 1978; Kornetsky and Esposito 1979; FiNger and Phillips ~987). In addition, it has been suggested that the discriminative stimulus properties of LH or VTA EBS also may be useful for this purpose (Doty 1969; Colpaert et al. 1977; Stutz and Maroli ~978; Kornetsky and Esposito 1981; Schaefer and Michaei 1985; Druhan etal. 1987b). Attempts to modulate LH or VTA EBS stimuli pharmacologically Offprint requests to." A.G. Phillips, Department of Psychology, 2136 West Mall, University of British Columbia, Vancouver, B.C., V6T ] Y7
may provide valuable insights into the neurochemical substrates underlying the perception of positive affect in animals. Recent pharmacological studies have demonstrated that VTA stimulation may produce at least two neurochemically distinct stimuli (Druhan etal. 1987 b, c), which can be measured independently with different discrimination procedures. For exampie, when rats were trained to discriminate between high and low intensities of prolonged, intermittent VTA stimulation, the indirect dopamine (DA) receptor agonist amphetamine enhanced '&e perceived intensities of the EBS cues, whereas the DA receptor antagonist, haloperidoi attenuated them (Drnhan et al. 1987 b). The acetylcholinesterase inhibitor physostigmine had no effect on the perception of these cues (Druhan et ai. 1987c). In contrast, when the EBS cues consisted of brief presentations of VTA stimu!ation, physostigmine enhanced the perceived intensities of the cues whereas amphetamine and haIoperidol were ineffective (Druhan etal. 1987 b, c). This pattern of drug effects indicated that the cues produced by prolonged intermittent stimulation of the VTA may be mediated by a dopaminergic substrate, whereas those associated with brief presentations of EBS may involve cholinergic processes. At present, it is not known whether the choiinergic processes associated with brief EBS cues involve muscarinic or nicotinic acetyicholine (ACH) receptors. Physostigmine increases the concentration of ACH, and therefore cannot be used to discriminate between cholinergic receptor types (Taylor J 980). In the present study, drugs that are selective for muscarinic and nicotinic receptors were administered to rats trained to discriminate between high and low intensity cues produced by brief presentations of VTA EBS. Specifically, the rats were injected with the muscarinic receptor agonists pilocarpine and RS-86, the muscarinic receptor antagonist scopolamine, and the direct nicotinic agonist nicotine (Clarke et at. 1985; Palacios et al. 1986). Modulation of the EBS cues by drugs specific to either the muscarinic or nicotinic cholinergic receptor types would imply a role for that receptor in either mediating or modulating the cues associated with brief stimulations of the VTA. Although stimulation of the VTA can produce strong rewarding effects, it is not clear whether these effects are responsible for producing alt of the discriminative stimulus properties of the EBS. It is conceivable that the rewarding effects of EBS served as stimuli for discriminated responding when the cues were produced by prolonged and intermittent stimulation (Druhan et al. 1987b, c). Dopamine-
332 containing neurons appear to mediate both these cue properties and the rewarding effects of VTA EBS (Druhan et al. 1987b; Fibiger et al. 1987; Phillips et al. 1987). In contrast, little is known about the relation between VTA EBS reward and the cues produced by brief presentations of EBS. This uncertainty arises partly from the paucity of information concerning the involvement of cholinergic neurons in ICSS obtained with VTA electrodes. Accordingly, a second goal of the present study was to assess the effects of the cholinergic drugs described above on VTA ICSS. Parallel effects of these drugs on both ICSS rate-intensity functions and the perceived intensity of the brief EBS cues would point to a common cholinergic substrate for some of the rewarding and stimulus properties of VTA EBS. Alternatively, differential effects of cholinergic drugs on ICSS and EBS discriminations would suggest that these two properties of the brain-stimulation have independent substrates.
(CRF) schedule during five daily 30-rain sessions. Rats that self-stimulated were then food deprived to 90% of their free-feeding weight and given additional ICSS rate-intensity sessions during which 11 different current intensities (6-26 gA, in 2 gA steps) were made available, each for 5 rain in ascending, then descending order. Each change in the current level was signalled by the free delivery of ten stimulation trains (two trains/s) at the new intensity. Rate-intensity functions measured over the final 4 days of this phase were averaged to obtain a single function from which one low current intensity (8-12 ~tA) and one high current intensity (10 gA higher than the low intensity) were chosen for use as discriminative stimuli. Selection of the stimuli in this manner ensured that the rats would discriminate across a range of intensities that were explicitly rewarding.
Discrimination training. Initially, all rats were trained to Experiment 1
Methods Subjects. The subjects were male hooded rats (Charles River, Long Evans strain) weighing 30(~350 g at the time of surgery. Throughout the experiment, the rats were housed individually in stainless steel cages with tap water available ad lib. A 12 h light/dark cycle was maintained in the animal colony, and all rats were tested in the light phase of this cycle. Surgery and histology. Each rat was anesthetized with 65 mg/kg sodium pentobarbital, and a bipolar electrode (Plastic Products MS303/2) was implanted stereotaxically into the VTA. With the incisor-bar set at - 3 . 2 mm below the interaural line, the coordinates from stereotaxic zero were: A . P . = + 2 . S m m ; L . = 0.6ram; D . V . = + l . 7 m m . The electrode was anchored to the skull with jeweller's screws and dental cement. Upon completion of the experiment, all rats were killed and their brains were sectioned and stained with cresyl-violet for verification of electrode placements.
bar-press for 45 mg Noyes food pellets on a CRF schedule during two daily 30-rain sessions with only one lever inserted into the chamber. The location of the lever was alternated daily to prevent development of side preferences. Subsequently, the rats were given daily two-lever discrimination sessions consisting of 90 trials delivered an average of 20 s apart (range = 15 25 s). Each trial began with the delivery of four 200-ms trains (spaced 200 ms apart) of either the high or low intensity EBS. The pellet dispenser was made operable 1 s later and the rats were then able to respond on the lever appropriate for the cue on that trial to receive one 45 mg food pellet. The appropriate lever for each current intensity was counterbalanced between rats. All trials were terminated (the houselight was turned off and the pellet dispenser made inoperable) either following the first correct response or after 10 s had elapsed without a response. During the first five training sessions incorrect responses initiated a further 10-s period in which the first appropriate response was rewarded. In subsequent sessions, incorrect responses resulted in nonrewarded termination of the trial. When the rats were responding with a high rate of accuracy, only 75% of correct responses were reinforced, Responses during the ITI were recorded separately for each lever, but had no programmed consequences.
Apparatus. The rats were tested in six separate chambers (24 x 29 • 30 cm), each having Plexiglas walls and ceiling, and a wire grid floor. Two levers (4.5 • 7 cm) were mounted 3.5 cm above the floor on opposite ends of the chamber. A 28 V house-light was positioned external to the chamber at the center of a side wall and a food-hopper positioned directly below it (3 cm above the floor). Each chamber was located within a sound attenuating enclosure (55 x 55 x 60 cm), and ventilation fans masked extraneous noise. The electrode leads (Plastic Products 303-302) were suspended from Mercotac commutators and passed through an opening in the ceiling. Constant current stimulation was delivered to the rats by a ten-channel, programmable sine-wave stimulator. A Data General Nova 3 computer with MANX software was used to control the experimental events and record responses made by the rats. Stimulation currents were monitored continuously on a Telequipment D54R oscilloscope,
Selection of EBS cue intensities. One week after surgery each rat was trained to lever-press for 200 ms trains of 20 pA, 60 Hz sine wave EBS on a continuous reinforcement
Stimulus generalization. Following training, rats that acquired the task (10 out of 11) were given three stimulus generalization tests (100 trials per test) that alternated daily with regular training sessions. During these tests responses were cued by either the regular training intensities (40 trials per intensity) or by four ,intermediate current levels (2 ~tA apart, five trials per intensity). Responses made following intermediate intensities were never reinforced, whereas correct responses to the two training currents were rewarded on 85% of the trials. Stimulus generalization functions were determined for each rat by calculating the percentage of responses that were emitted on the lever appropriate for the high training intensity at each current level.
Effects of cholinergic agonists on EBS generalization. The rats were given separate generalization tests after receiving physostigmine sulfate (0.20, 0.35 and 0.50 mg/kg, IP), pilocarpine nitrate (1.25, 2.5 and 5,0 mg/kg, IP), RS-86 hydrobromide (0.5, 1.0 and 1.5 mg/kg, IP) or l-nicotine bitartrate (0.2, 0.4 and 0.8 mg/kg, SC). Pilocarpine and RS-86 were injected 15 min before testing, whereas physostigmine and
333
nicotine were administered 20 rain before testing. Scopolamine methylbromide (0.50 mg/kg, IP) was injected 15 rain prior to piiocarpine and RS-86, and 10 rain before physostigmine, to reduce the peripherai effects of these drugs. Scopolamine methylbromide has no effects on the discrimination of brief VTA EBS cues (Druhan et al. 1987 b). Separate tests were conducted after saline injections for comparison with each drug.
PLATE #
45b
46b
Effects of scopolamine. Generalization tests were conducted 15 min after injections of scopolamine hydrobromide (&20, 0.35, 0.50, 0.80 and 1.10 mg/kg, IP) or saline. In subsequent tests, pilocarpine (5.0 mg/kg, IP) or RS-86 (1.5 mg/kg, IP) was administered after injections of either scopolamine methyIbromide (0.50 mg/kg, IP) or scopolamine hydrobromide (0.20 mg/kg, IP). Pilocarpine and RS-86 were administered 5 rain after the scopolamine compounds and the discrimination trials began ~5 min later. Separate saline tests were given for comparison with the agonist/antagonist tests involving pitocarpine and RS-86.
Drugs. Physostigmine sulfate (Sigma, St. Louis, MO), pilocarpine nitrate (Sigma), RS-86 hydrobromide (Sandoz, Basle, Switzerland), scopolamine hydrobromide (Sigma), l-nicotine bitartrate (BDH, Poole, Dorset, UK) and scopolamine methylbromide (Sigma) were diluted in 0.9% saline solution with the concentration varied to maintain an injection volume of 1 mi/kg. N a O H (0.05 M) was added to the nicotine solutions to bring the pH to 7.2. All doses were expressed in terms of the salt except for nicotine, which was expressed as the base. The drug sessions were conducted on every 3rd day, with the order of drug administration counterbalanced across animals. Regular training days were interposed between the drug tests. Statistical analyses. Independent two-way repeated-measures analyses of variance (ANOVA) were used to analyse the generalization functions obtained for each drug condition. Drug dose and current intensity served as independent variables, whereas per cent responses on the high-intensity stimulation (HS) lever was the dependent variable. The number of trials completed, the ITI response rates and the ITI response biases were analysed by a one-way ANOVA with drug dose as the independent factor. Except where indicated otherwise, Newman-KeuI's test (P
