- Email: [email protected]
Potentiation by intraventricular muscimol of the anticonflict effect of benzodiazepines
Brain Research, 196 (1980) 447-453 © Elsevier/North-Holland BiomedicalPress
447
POTENTIATION BY I N T R A V E N T R I C U L A R M U S C I M O L OF THE ANTICONFLICT EFFECT OF BENZODIAZEPINES
A. R. CANANZI, E. COSTA* and A. GUIDOTTI Laboratory of Preclinical Pharmacology and ( A.R.C.) Laboratory of Clinical Psychopharmacology, National Institute of Mental Health, Saint Elizabeths Hospital, Washington, D.C. 20032 (U.S.A.)
(Accepted March 13th, 1980) Key words: muscimol -- benzodiazepines-- GABA -- punished behavior
SUMMARY The anticonflict action of muscimol (a potent GABA receptor agonist) and its ability to potentiate the anticonflict action of the benzodiazepines was studied in rats using Vogel's procedure ~1. Rats were injected intraventricularly with a dose of muscimol from 50 to 200 ng to study whether the anticonflict action was dose-related. The potency of a 200 ng dose of muscimol is comparable to the potency of an anticonflict dose of 0.5 mg i.v. of diazepam. THIP (a muscimol analogue with weaker intrinsic GABA-mimetic activity) is active in doses 10 times higher than muscimol. The threshold dose of diazepam (0.2 mg/kg i.v.) to elicit anticonflict action was markedly potentiated when it was injected 5 min after intraventricular muscimol (150 ng). These data support the concept that GABA receptors may be involved in the anticonflict effects of the benzodiazepines.
INTRODUCTION Recent studies on the synaptic pharmacology of benzodiazepines clearly show that these drugs when administered in anticonvulsant, muscle relaxant or anxiolytic doses facilitate GABAergic transmission in brain and spinal cord (for review see ref. 6). While the participation of GABA in the muscle relaxant and anticonvulsant properties of benzodiazepines is well documented, a role for this amino acid in mediating the anxiolytic action of these drugs has not as yet been established firmly. Several reasons account for this uncertainty among which are the difficulties one encounters in testing the anxiolytic action of drugs. * To w h o m correspondence should be addressed.
448 Today the most widely accepted animal model to study the anti-anxiety properties of benzodiazepines and to predict their clinical efficacy in psychoneurosis is an operant conditioning procedure which was termed 'conflict' or 'punishment' test 4,17. A participation of GABAergic mechanisms in the mediation of the anxiolytic action of benzodiazepines can be substantiated with these tests. Two GABA receptor antagonists, picrotoxin and bicuculline, antagonize the punishment-lessening effect of benzodiazepines3, ls,22, while the inhibitors of GABA degradation 22 potentiate the action of benzodiazepine on this test for anxiolytic activity. However, the systemic administration of muscimol (5-aminomethyl-3-isoxazolol), a potent and specific GABA receptor agonist, failed to potentiate the anticonflict activity of benzodiazepines19, 20. Studies on the fate of muscimol given systemically show that this drug is rapidly metabolized and that less than 0.01-0.02 ~o of the administered dose enters the brain1, a3. Probably the lack of behavioral effects elicited by parenteral muscimol could be attributed to its rapid metabolism and the poor penetration into the brain tissue. Since muscimol given into the cerebral ventricles (i.c.v.) elicits long-lasting pharmacological responses because its efflux rate from brain is quite slow 1, we examined whether i.c.v, muscimol, like parenteral benzodiazepines, could activate behavioral outputs that are inhibited by fear. MATERIALS AND METHODS Male Sprague-Dawley rats (Zivic Miller, Pa.) weighing between 200 and 240 g were kept in our facilities at 25 °C (with 12 h of light from 08.00 to 20.00 h) and were fed purina lab chow and water ad libitum for at least 1 week prior to experiments. To test the anticonflict action of muscimol and to compare it to that of benzodiazepines, we selected the conditioned suppression of drinking in thirsty rats as described by Vogel et al. 21. In each experiment at 17.00 h naive rats were placed in single cages in a sound-proof room and deprived of water for 72 h prior to the conflict test session. At the moment of the test the animals were placed in a clear plexiglass box (38 × 38 cm) with a stainless steel grid floor. A metal drinking tube connected with a water bottle was inserted into the box at a height of 2 cm above the grid. The grid floor was connected to a shock distributor programmed to deliver for 25 msec a 0.5 mA electric stimulus. Immediately prior to testing electrode creme was applied to the foreand hindpaws of the rat to ensure good grid contacts. The rat was then placed in the box, allowed to find the drinking tube and to drink for 10 sec, continuously, before receiving a shock. At the termination of the first shock a 3 min timer was started. During the 3-min period, shocks were delivered every time a 5-sec time period of continuous drinking was observed. In some experiments, in order to control whether the drug treatment was interfering with the learning process, thirsty rats were placed in a starting rectangular (l 5 × 10 cm) box that through a door led to a (20 × 20 cm) box containing 4 drinking bottles placed at the 4 corners; of these only one bottle contained water. When the rats reached the metal drinking tube of a bottle (an empty one or one full of water), they
449 were allowed to lick the tube for 10 sec; immediately thereafter they were removed and placed again in the starting box. During 10 trials, normal thirsty rats selected the bottle containing water 70-80 ~ of the times. To study the changes in the motivation, rats were placed in the (38 x 38 cm) plexiglass box and allowed to drink ad libitum without shock. In some experiments the possible analgesic capacity of the drugs used in the conflict test situation was determined by using the tail-flick procedure described by D'Amour and Smith7. Drugs were injected i.c.v, by implanting a polyethylene cannula 15 in the lateral ventricle under slight ether anesthesia, 3 days before the experiment. RESULTS In satiated rats, the intraventricular injections of muscimol in doses ranging from 50 to 260 ng produced in 2-5 min compulsive eating behavior but not compulsive drinking. The food consumption reached maximal intensity 15-20 min post-injection. Higher doses of muscimol (300-600 ng i.c.v.) produced marked sedation, decreased mobility and an apparent decrease in the motivation for food intake. These results are in agreement with previous reports from this laboratory10. THIP, (4,5,6,7-tetrahydroisoxazolo-[5,4-c]pyridin 3-ol) an analogue of muscimol with weaker intrinsic GABA-mimetic activity, induced eating behavior in doses 10 times higher than that of muscimol. Isoguvacine (1,2,3,6 tetrahydropyridine-4-carboxylic acid), another weak GABA receptor agonist failed to induce significant behavioral responses up to doses of 2000 ng i.c.v. In normal rats, diazepam injected i.v. stimulated food intake, but failed to elicit compulsive drinking. Stimulation of food intake was dose-related up to 0.5 mg/kg i.v. (the maximal dose used in these experiments),; peak effect occurred between 5 and 10 rain after the intravenous injection. Naive thirsty animals placed in the plexiglass box with one bottle of water (see Methods), in absence of punishment, drank without interruption for the first 3 min (time spent drinking 150-170 sec). From the 3rd min to the 10th min mark they continued to drink water but at a slower rate (time spent drinking between 3 and 10 rain is 150-200 sec). Muscimol, diazepam and muscimol-plus-diazepam injected into thirsty animals in absence of punishment failed to change the latency or the total drinking time during the first 3 min. However, diazepam and muscimol in larger doses produced a slight (20 ~ ) but significant increase of drinking time during the time interval between 3 and 10 min after the thirsty animals were placed in the drinking cage. When thirsty rats were placed in the 20 x 20 cm box (see Methods) containing at the 4 corners 4 bottles, 3 empty and one containing water, the control rats after 2-3 trials invariably moved toward the side of the box that contained the bottle with water and started drinking. Muscimol, diazepam or the combination of these two drugs failed to influence this behavior. The effect of diazepam, muscimol, THIP and isoguvacine on thirsty rats subjected to conditioned suppression of drinking is illustrated in Table I. During the first 3-min session, rats treated with diazepam doses of 0.3 or 0.5 mg/kg i.v. approached the drinking bottle more often and therefore took a greater number of
450 TABLE I Ant±punishment activity o f diazepam and GA B.4-mimetic compounds"
Muscimol, THIP and isoguvacine were injected intracerebroventricularly (i.c.v.) 15 min before the test. Diazepam was injected intravenously (i.v.) 10 min before the test. n = number of animals. Compound
Control Diazepam
Muscimol
THIP Isoguvacine
Dose
n
0.2 mg/kg i.v. 0.3 mg/kg i.v. 0.5 mg/kg i.v. 50 ng 1.c.v. 150 ng l.c.v. 175 ng l.c.v. 200 ng 1.c.v. 300 ng l.c.v. 400 ng i.c.v. 1000 ng l.c.v. 2000 ng l.c.v. 2000 ng 1.c.v.
Shocks ± S.E.
25 10 5 3 3 3 4
3.5 ± 0.45 5.0 ± 0.75 19 ± 5.2* 28 ± 2.8* 3.8 ± 0.50 9.3 ± 1.4" 12 ~. 2.7*
11
22 ± 2.4*
3 2 5 3 5
14 ± 0.88* 11 ± (6-16) 5.0 ± 0.65 17 ~ 0.80* 5.0 :~ 0.82
* Statistically significant; P < 0.05 when compared with control values. shocks t h a n n o n - t r e a t e d rats. The total time spent d r i n k i n g water d u r i n g this 3-min session was approximately 15 sec in control rats a n d 140 sec in rats treated with 0.5 m g / k g i.v. diazepam. The latter time was close to the d r i n k i n g time spent by naive thirsty rats in a n o r m a l situation. Rats treated with m u s c i m o l a n d T H I P received a n u m b e r of shocks greater than solvent-treated rats (Table II), d u r i n g the first 3 min. The response was dose-related from 50 to 200 ng i.c.v, of muscimol. Doses of 300-400 ng were somehow less effective. After 600 ng i.c.v, of muscimol, the rats were incapable o f m o v i n g a n d they could n o t reach the tube of the water bottle. T H I P produced a significant increase in n u m b e r of TABLE II Interaction between diazepam and muscimol on punishment activity
Each group is the mean ± S.E. of 6 rats. Diazepam or solvent (1 ml/kg) was given i.v. 5 min after i.c.v. injection of muscimol or saline. The animals were tested 15 min after muscimol. Treatment
~ Shock ± S.E.
i.c.v,
i.v.
Saline (10/zl) Saline (10 ILl) Muscimol (150 ng) Muscimol (150 ng)
Solvent Diazepam (0.2 mg/kg) Solvent Diazepam (0.2 mg/kg)
3.1 4.3 9.2 22
± ± ± ±
0.69 0.56 1.6" 0.5**
* Statistically significant; P < 0.05 when compared with control values. Mean value of this group is statistically (P < 0.05) higher than the mean values of the group treated only with muscimol.
**
451 shocks when given only in doses of 2000 ng i.c.v. Isoguvacine up to doses of 2000 ng failed to increase the number of shocks that the rats took during the first 3 min. In a subsequent experiment the effect of the association of muscimol and diazepam was studied. In this experiment a dose of diazepam of 0.2 mg/kg i.v. was administered together with 150 ng muscimoh Although diazepam in the dose given failed to change the number of approaches to the water during the first 3 min and muscimol produced only marginal changes, a marked potentiation was observed when the two drugs were combined (Table II). None of the drugs, at the maximal doses used, altered the threshold to painful stimuli (tail-flick test). The data obtained for the flick reaction time in sec were: control, 6.5 ! 0.2, n = 5; muscimol (200 ng), 6.2 ± 0.5, n = 6; diazepam (0.5 mg/kg), 7.2 ± 0.4, n = 6; muscimol (200 ng) ÷ diazepam (0.2 mg/kg), 6.0 d: 0.08, n -- 6. DISCUSSION Similarities between the biochemical and neuropharmacological profile of muscimol and diazepam have been taken as crucial evidence to support the hypothesis that the anticonvulsant, muscle relaxant, and sedative action of benzodiazepines are mediated by a facilitation of GABAergic transmissionS,6,1t, 12. Obviously in ascertaining whether GABA mediates the anti-anxiety action of benzodiazepines, it is important to establish that muscimol and benzodiazepines produce a similar increase in the number of shocks that the animals take when exposed to the conflict situation. Here we report that muscimol and THIP, similar to diazepam, induce eating behavior but not drinking in satiated rats. In addition, intraventricular injections of muscimol potentiate the anticonflict action of diazepam in thirsty rats in which the drinking is suppressed by punishment. These data support the contention that GABA receptors may be involved in the anticonflict effects of benzodiazepines and agree with previous reports showing that pharmacological manipulations of GABA neurons can modify the anticonflict activity of benzodiazepines3,18,2L The data reported here also indicate for the first time that, in thirsty rats, the stimulation of brain GABA receptors with muscimol and THIP reduces the behavioral inhibition produced by punishment. This observation is in line with a report by Soubrie et al. 16 that picrotoxin potentiates the behavioral inhibition induced in rats by punishment. However, our observation is at apparent variance with previous reports on the failure of muscimol to mimic benzodiazepines in anticonflict procedures19,20. Most likely the reason for this discrepancy is due to the different rate of injection, parenterally in previous reports19, 20 and i.c.v, in the present report. In fact muscimol is rapidly metabolized and if injected i.v. or i.p., penetrates with difficulty the blood-brain barrier1,1~. The pharmacological action of muscimol metabolites is presently unknown but probably these metabolites do not enter in the brainL Our choice of the i.c.v, was motivated by our knowledge that muscimol injected i.c.v, is metabolized slowly and selectively accumulates in cortex, striatum, hippocampus and hypothalamus 1. These areas are rich in GABA 8 and benzodiazepine receptors ~4 and are endowed with a potential pivotal role in the control of animal behavior.
452 The action of muscimol in decreasing the behavioral restrictions determined by the fear of punishment is maximal for doses of 200 ng and the dose-effect relationship is very sharp. This should be expected for a drug that has an affinity for GABA receptors which is at least 10 times higher than that of GABA 9. Higher doses of muscimol (300-600 ng) per se decrease motility and cause general depression, hence they cannot be used for these behavioral studies. The anticonflict action of muscimol and diazepam cannot be due to an increase in motivation in drinking because both in control and thirsty animals, muscimol, diazepam and a combination of muscimol and diazepam fail to increase the drinking latency time or the total drinking time. In addition the effect of these drugs on punished behavior is very unlikely to be due to an increased motivation for drinking because in our experimental situation the animals were kept for 72 h without water, a condition under which motivation for drinking is practically maximal (see also Table I). The muscimol inhibition of behavioral restrictions caused by punished behavior is not due to a disruption of learning because muscimol in doses which increase punished behavior failed to influence the learning capacity of rats. It has been reported that muscimol, although devoid of analgesic activity can potentiate morphine analgesiaL We have explored whether muscimol in association with diazepam abolishes behavioral restrictions caused by electroshock because it increases pain threshold. Measurements of pain perception by the tail-flick test reveal that muscimol, diazepam and their association fail to change pain threshold. Stimulation of brain GABA receptors with muscimol elicits benzodiazepam-like anticonflict activity, which is dose-related. When the doses are increased above a certain limit, muscimol decreases motor activity and therefore fails to elicit anticonflict activity. These data are consistent with the idea that GABA receptors are involved in the control of behavioral inhibition induced by punishment. This hypothesis finds support in recent biochemical observations of the existence of a strict correlation between benzodiazepine and GABA receptors and with the concept that the GABA receptor is a functional unit which contains GABA receptors, benzodiazepine receptors, chloride ionophores along with modulatory factors all of which are closely associated6. The part of parts of the brain more directly involved in this behavioral action of muscimol and diazepam are now being investigated. ACKNOWLEDGEMENTS
We thank Prof. Povl Krosgaard-Larsen for making available to us THIP and isoguvacine and Dr. D. Della Bella (Zambon SpA, Milan) for a generous supply of muscimol.
REFERENCES 1 Baraldi, M., Grandison, L. and Guidotti, A., Distribution and metabolism of muscimol in the brain and other tissues of the rat, Neuropharmacology, 18 (1979) 57-62.
453 2 Biggio, G., Della Bella, D., Frigeni, V. and Guidotti, A., Potentiation of morphine analgesia by muscimol, Neuropharmacology, 16 (1977) 149-150. 3 Billingsley, M. L. and Kubena, R. K., The effects of naloxone and picrotoxin on the sedative and anticonflict effects of benzodiazepines, Life Sci., 22 (1978) 897-906. 4 Cook, L. and Davidson, A. B., Effects of behaviorally active drugs in a conflict-punishment procedure in rats. In S. Garattini, E. Mussini and L. O. Randall (Eds.), The Benzodiazepines, Raven Press, New York, 1973, pp. 327-345. 5 Costa, E., Guidotti, A., Map, C. C. and Suria, A., New concepts on the mechanism of action of benzodiazepines, Life Sci., 17 (1975) 167-186. 6 Costa, E. and Guidotti, A., Molecular mechanisms in the receptor action of benzodiazepines, Ann. Rev. Pharmacol. Toxicol., 19 (1979) 531-545. 7 D'Amour, F. E. and Smith, D. L., A method of determining loss of pain sensation, J. Pharmacol. exp. Ther., 72 (1941) 74-79. 8 Enna, S. J. and Snyder, S. H., Influence of ions, enzymes and detergents on gamma-aminobutyric acid receptor binding in synaptic membranes of rat brain, Molec. Pharmacol., 13 (1977) 442--453. 9 Enna, S. J., Collins, J. F. and Snyder, S. H., Stereospecificity and structure-activity requirements of GABA receptor binding in rat brain, Brain Research, 124 (1977) 184-190. 10 Grandison, L. and Guidotti, A., Stimulation of food intake by muscimol and beta endorphin, Neuropharmacology, 16 (1977) 533-536. 11 Guidotti, A., Synaptic mechanisms in the action of benzodiazepines. In M. A. Lipton, A. Di Mascio and K. F. Killam (Eds.), Psychopharmacology: a Generation of Progress, Raven Press, New York, 1978, pp. 1349-1357. 12 Haefely, W. E., Behavioral and neuropharmacological aspects of drugs used in anxiety and related states. In M. A. Lipton, A. Di Mascio and K. F. Killam (Eds.), Psychopharmacology: a Generation of Progress, Raven Press, New York, 1978, pp. 1359-1374. 13 Maggi, A. and Enna, S. J., Characteristics of muscimol accumulation in mouse brain after systemic administration, Neuropharmacology, 18 (1979) 361-366. 14 Mohler, H. and Okada, T., Benzodiazepine receptor: demonstration in the central nervous system, Science, 198 (1977) 849-851. 15 Robinson, C. A., Hengeveld, C. A. and Balbian Verster, F.de., Improved polyethylene cannulation techniques, Physiol. Behav., 4 (1969) 123-t24. 16 Soubrie, Ph., Thiebot, M. H. and Simon, P., Enhanced suppressive effects of aversive events induced in rats by picrotoxin: possibility of a GABA control on behavioral inhibition, PharmacoL Biochem. Behav., 10 (1979) 463-469. 17 Stein, L., Wise, C. D. and Berger, B. D., Antianxietyaction of benzodiazepines: decrease in activity of serotonin neurons in the punishment system. In S. Garattini, E. Mussini and L. O. Randall (Eds.), The Benzodiazepines, Raven Press, New York, 1973, pp. 299-326. 18 Stein, L., Wise, C. D. and Belluzzi, J. D., Effects of benzodiazepines on central serotonergic mechanisms. In E. Costa and P. Greengard (Eds.), Mechanisms of Action of Benzodiazepines, Raven Press, New York, 1975, pp. 29-44. 19 Sullivan, J. W., Sepinwall, J. and Cook, L., Anticonflictevaluation of muscimol, a GABA receptor agonist, alone and in combination with diazepam, Fed. Proc., 37 (1978) 2143. 20 Thiebot, M. H., Jobert, A. and Soubrie, P., Effets compares du muscimol et du diazepam sur les inhibitions du comportement induites chez le rat par la nouveaute, la punition etie non-renforcement, Psychopharmacologia (BerL), 61 (1979) 85-89. 21 Vogel, J. R., Beer, B. and Clody, D. E., A simple and reliable conflict procedure for testing antianxiety agents, Psychopharmacologia (Berl.), 21 (1971) 1-7. 22 Zakusov, V. V., Ostrovskaya, R. V., Kozhechkin, S. N., Markovic, V. V., Molodavkin, G. M. and Voronina, T. A., Further evidence for GABAergic mechanisms in the action of benzodiazepines, Arch. int. Pharmacodyn., 229 (1977) 313-326.
447
POTENTIATION BY I N T R A V E N T R I C U L A R M U S C I M O L OF THE ANTICONFLICT EFFECT OF BENZODIAZEPINES
A. R. CANANZI, E. COSTA* and A. GUIDOTTI Laboratory of Preclinical Pharmacology and ( A.R.C.) Laboratory of Clinical Psychopharmacology, National Institute of Mental Health, Saint Elizabeths Hospital, Washington, D.C. 20032 (U.S.A.)
(Accepted March 13th, 1980) Key words: muscimol -- benzodiazepines-- GABA -- punished behavior
SUMMARY The anticonflict action of muscimol (a potent GABA receptor agonist) and its ability to potentiate the anticonflict action of the benzodiazepines was studied in rats using Vogel's procedure ~1. Rats were injected intraventricularly with a dose of muscimol from 50 to 200 ng to study whether the anticonflict action was dose-related. The potency of a 200 ng dose of muscimol is comparable to the potency of an anticonflict dose of 0.5 mg i.v. of diazepam. THIP (a muscimol analogue with weaker intrinsic GABA-mimetic activity) is active in doses 10 times higher than muscimol. The threshold dose of diazepam (0.2 mg/kg i.v.) to elicit anticonflict action was markedly potentiated when it was injected 5 min after intraventricular muscimol (150 ng). These data support the concept that GABA receptors may be involved in the anticonflict effects of the benzodiazepines.
INTRODUCTION Recent studies on the synaptic pharmacology of benzodiazepines clearly show that these drugs when administered in anticonvulsant, muscle relaxant or anxiolytic doses facilitate GABAergic transmission in brain and spinal cord (for review see ref. 6). While the participation of GABA in the muscle relaxant and anticonvulsant properties of benzodiazepines is well documented, a role for this amino acid in mediating the anxiolytic action of these drugs has not as yet been established firmly. Several reasons account for this uncertainty among which are the difficulties one encounters in testing the anxiolytic action of drugs. * To w h o m correspondence should be addressed.
448 Today the most widely accepted animal model to study the anti-anxiety properties of benzodiazepines and to predict their clinical efficacy in psychoneurosis is an operant conditioning procedure which was termed 'conflict' or 'punishment' test 4,17. A participation of GABAergic mechanisms in the mediation of the anxiolytic action of benzodiazepines can be substantiated with these tests. Two GABA receptor antagonists, picrotoxin and bicuculline, antagonize the punishment-lessening effect of benzodiazepines3, ls,22, while the inhibitors of GABA degradation 22 potentiate the action of benzodiazepine on this test for anxiolytic activity. However, the systemic administration of muscimol (5-aminomethyl-3-isoxazolol), a potent and specific GABA receptor agonist, failed to potentiate the anticonflict activity of benzodiazepines19, 20. Studies on the fate of muscimol given systemically show that this drug is rapidly metabolized and that less than 0.01-0.02 ~o of the administered dose enters the brain1, a3. Probably the lack of behavioral effects elicited by parenteral muscimol could be attributed to its rapid metabolism and the poor penetration into the brain tissue. Since muscimol given into the cerebral ventricles (i.c.v.) elicits long-lasting pharmacological responses because its efflux rate from brain is quite slow 1, we examined whether i.c.v, muscimol, like parenteral benzodiazepines, could activate behavioral outputs that are inhibited by fear. MATERIALS AND METHODS Male Sprague-Dawley rats (Zivic Miller, Pa.) weighing between 200 and 240 g were kept in our facilities at 25 °C (with 12 h of light from 08.00 to 20.00 h) and were fed purina lab chow and water ad libitum for at least 1 week prior to experiments. To test the anticonflict action of muscimol and to compare it to that of benzodiazepines, we selected the conditioned suppression of drinking in thirsty rats as described by Vogel et al. 21. In each experiment at 17.00 h naive rats were placed in single cages in a sound-proof room and deprived of water for 72 h prior to the conflict test session. At the moment of the test the animals were placed in a clear plexiglass box (38 × 38 cm) with a stainless steel grid floor. A metal drinking tube connected with a water bottle was inserted into the box at a height of 2 cm above the grid. The grid floor was connected to a shock distributor programmed to deliver for 25 msec a 0.5 mA electric stimulus. Immediately prior to testing electrode creme was applied to the foreand hindpaws of the rat to ensure good grid contacts. The rat was then placed in the box, allowed to find the drinking tube and to drink for 10 sec, continuously, before receiving a shock. At the termination of the first shock a 3 min timer was started. During the 3-min period, shocks were delivered every time a 5-sec time period of continuous drinking was observed. In some experiments, in order to control whether the drug treatment was interfering with the learning process, thirsty rats were placed in a starting rectangular (l 5 × 10 cm) box that through a door led to a (20 × 20 cm) box containing 4 drinking bottles placed at the 4 corners; of these only one bottle contained water. When the rats reached the metal drinking tube of a bottle (an empty one or one full of water), they
449 were allowed to lick the tube for 10 sec; immediately thereafter they were removed and placed again in the starting box. During 10 trials, normal thirsty rats selected the bottle containing water 70-80 ~ of the times. To study the changes in the motivation, rats were placed in the (38 x 38 cm) plexiglass box and allowed to drink ad libitum without shock. In some experiments the possible analgesic capacity of the drugs used in the conflict test situation was determined by using the tail-flick procedure described by D'Amour and Smith7. Drugs were injected i.c.v, by implanting a polyethylene cannula 15 in the lateral ventricle under slight ether anesthesia, 3 days before the experiment. RESULTS In satiated rats, the intraventricular injections of muscimol in doses ranging from 50 to 260 ng produced in 2-5 min compulsive eating behavior but not compulsive drinking. The food consumption reached maximal intensity 15-20 min post-injection. Higher doses of muscimol (300-600 ng i.c.v.) produced marked sedation, decreased mobility and an apparent decrease in the motivation for food intake. These results are in agreement with previous reports from this laboratory10. THIP, (4,5,6,7-tetrahydroisoxazolo-[5,4-c]pyridin 3-ol) an analogue of muscimol with weaker intrinsic GABA-mimetic activity, induced eating behavior in doses 10 times higher than that of muscimol. Isoguvacine (1,2,3,6 tetrahydropyridine-4-carboxylic acid), another weak GABA receptor agonist failed to induce significant behavioral responses up to doses of 2000 ng i.c.v. In normal rats, diazepam injected i.v. stimulated food intake, but failed to elicit compulsive drinking. Stimulation of food intake was dose-related up to 0.5 mg/kg i.v. (the maximal dose used in these experiments),; peak effect occurred between 5 and 10 rain after the intravenous injection. Naive thirsty animals placed in the plexiglass box with one bottle of water (see Methods), in absence of punishment, drank without interruption for the first 3 min (time spent drinking 150-170 sec). From the 3rd min to the 10th min mark they continued to drink water but at a slower rate (time spent drinking between 3 and 10 rain is 150-200 sec). Muscimol, diazepam and muscimol-plus-diazepam injected into thirsty animals in absence of punishment failed to change the latency or the total drinking time during the first 3 min. However, diazepam and muscimol in larger doses produced a slight (20 ~ ) but significant increase of drinking time during the time interval between 3 and 10 min after the thirsty animals were placed in the drinking cage. When thirsty rats were placed in the 20 x 20 cm box (see Methods) containing at the 4 corners 4 bottles, 3 empty and one containing water, the control rats after 2-3 trials invariably moved toward the side of the box that contained the bottle with water and started drinking. Muscimol, diazepam or the combination of these two drugs failed to influence this behavior. The effect of diazepam, muscimol, THIP and isoguvacine on thirsty rats subjected to conditioned suppression of drinking is illustrated in Table I. During the first 3-min session, rats treated with diazepam doses of 0.3 or 0.5 mg/kg i.v. approached the drinking bottle more often and therefore took a greater number of
450 TABLE I Ant±punishment activity o f diazepam and GA B.4-mimetic compounds"
Muscimol, THIP and isoguvacine were injected intracerebroventricularly (i.c.v.) 15 min before the test. Diazepam was injected intravenously (i.v.) 10 min before the test. n = number of animals. Compound
Control Diazepam
Muscimol
THIP Isoguvacine
Dose
n
0.2 mg/kg i.v. 0.3 mg/kg i.v. 0.5 mg/kg i.v. 50 ng 1.c.v. 150 ng l.c.v. 175 ng l.c.v. 200 ng 1.c.v. 300 ng l.c.v. 400 ng i.c.v. 1000 ng l.c.v. 2000 ng l.c.v. 2000 ng 1.c.v.
Shocks ± S.E.
25 10 5 3 3 3 4
3.5 ± 0.45 5.0 ± 0.75 19 ± 5.2* 28 ± 2.8* 3.8 ± 0.50 9.3 ± 1.4" 12 ~. 2.7*
11
22 ± 2.4*
3 2 5 3 5
14 ± 0.88* 11 ± (6-16) 5.0 ± 0.65 17 ~ 0.80* 5.0 :~ 0.82
* Statistically significant; P < 0.05 when compared with control values. shocks t h a n n o n - t r e a t e d rats. The total time spent d r i n k i n g water d u r i n g this 3-min session was approximately 15 sec in control rats a n d 140 sec in rats treated with 0.5 m g / k g i.v. diazepam. The latter time was close to the d r i n k i n g time spent by naive thirsty rats in a n o r m a l situation. Rats treated with m u s c i m o l a n d T H I P received a n u m b e r of shocks greater than solvent-treated rats (Table II), d u r i n g the first 3 min. The response was dose-related from 50 to 200 ng i.c.v, of muscimol. Doses of 300-400 ng were somehow less effective. After 600 ng i.c.v, of muscimol, the rats were incapable o f m o v i n g a n d they could n o t reach the tube of the water bottle. T H I P produced a significant increase in n u m b e r of TABLE II Interaction between diazepam and muscimol on punishment activity
Each group is the mean ± S.E. of 6 rats. Diazepam or solvent (1 ml/kg) was given i.v. 5 min after i.c.v. injection of muscimol or saline. The animals were tested 15 min after muscimol. Treatment
~ Shock ± S.E.
i.c.v,
i.v.
Saline (10/zl) Saline (10 ILl) Muscimol (150 ng) Muscimol (150 ng)
Solvent Diazepam (0.2 mg/kg) Solvent Diazepam (0.2 mg/kg)
3.1 4.3 9.2 22
± ± ± ±
0.69 0.56 1.6" 0.5**
* Statistically significant; P < 0.05 when compared with control values. Mean value of this group is statistically (P < 0.05) higher than the mean values of the group treated only with muscimol.
**
451 shocks when given only in doses of 2000 ng i.c.v. Isoguvacine up to doses of 2000 ng failed to increase the number of shocks that the rats took during the first 3 min. In a subsequent experiment the effect of the association of muscimol and diazepam was studied. In this experiment a dose of diazepam of 0.2 mg/kg i.v. was administered together with 150 ng muscimoh Although diazepam in the dose given failed to change the number of approaches to the water during the first 3 min and muscimol produced only marginal changes, a marked potentiation was observed when the two drugs were combined (Table II). None of the drugs, at the maximal doses used, altered the threshold to painful stimuli (tail-flick test). The data obtained for the flick reaction time in sec were: control, 6.5 ! 0.2, n = 5; muscimol (200 ng), 6.2 ± 0.5, n = 6; diazepam (0.5 mg/kg), 7.2 ± 0.4, n = 6; muscimol (200 ng) ÷ diazepam (0.2 mg/kg), 6.0 d: 0.08, n -- 6. DISCUSSION Similarities between the biochemical and neuropharmacological profile of muscimol and diazepam have been taken as crucial evidence to support the hypothesis that the anticonvulsant, muscle relaxant, and sedative action of benzodiazepines are mediated by a facilitation of GABAergic transmissionS,6,1t, 12. Obviously in ascertaining whether GABA mediates the anti-anxiety action of benzodiazepines, it is important to establish that muscimol and benzodiazepines produce a similar increase in the number of shocks that the animals take when exposed to the conflict situation. Here we report that muscimol and THIP, similar to diazepam, induce eating behavior but not drinking in satiated rats. In addition, intraventricular injections of muscimol potentiate the anticonflict action of diazepam in thirsty rats in which the drinking is suppressed by punishment. These data support the contention that GABA receptors may be involved in the anticonflict effects of benzodiazepines and agree with previous reports showing that pharmacological manipulations of GABA neurons can modify the anticonflict activity of benzodiazepines3,18,2L The data reported here also indicate for the first time that, in thirsty rats, the stimulation of brain GABA receptors with muscimol and THIP reduces the behavioral inhibition produced by punishment. This observation is in line with a report by Soubrie et al. 16 that picrotoxin potentiates the behavioral inhibition induced in rats by punishment. However, our observation is at apparent variance with previous reports on the failure of muscimol to mimic benzodiazepines in anticonflict procedures19,20. Most likely the reason for this discrepancy is due to the different rate of injection, parenterally in previous reports19, 20 and i.c.v, in the present report. In fact muscimol is rapidly metabolized and if injected i.v. or i.p., penetrates with difficulty the blood-brain barrier1,1~. The pharmacological action of muscimol metabolites is presently unknown but probably these metabolites do not enter in the brainL Our choice of the i.c.v, was motivated by our knowledge that muscimol injected i.c.v, is metabolized slowly and selectively accumulates in cortex, striatum, hippocampus and hypothalamus 1. These areas are rich in GABA 8 and benzodiazepine receptors ~4 and are endowed with a potential pivotal role in the control of animal behavior.
452 The action of muscimol in decreasing the behavioral restrictions determined by the fear of punishment is maximal for doses of 200 ng and the dose-effect relationship is very sharp. This should be expected for a drug that has an affinity for GABA receptors which is at least 10 times higher than that of GABA 9. Higher doses of muscimol (300-600 ng) per se decrease motility and cause general depression, hence they cannot be used for these behavioral studies. The anticonflict action of muscimol and diazepam cannot be due to an increase in motivation in drinking because both in control and thirsty animals, muscimol, diazepam and a combination of muscimol and diazepam fail to increase the drinking latency time or the total drinking time. In addition the effect of these drugs on punished behavior is very unlikely to be due to an increased motivation for drinking because in our experimental situation the animals were kept for 72 h without water, a condition under which motivation for drinking is practically maximal (see also Table I). The muscimol inhibition of behavioral restrictions caused by punished behavior is not due to a disruption of learning because muscimol in doses which increase punished behavior failed to influence the learning capacity of rats. It has been reported that muscimol, although devoid of analgesic activity can potentiate morphine analgesiaL We have explored whether muscimol in association with diazepam abolishes behavioral restrictions caused by electroshock because it increases pain threshold. Measurements of pain perception by the tail-flick test reveal that muscimol, diazepam and their association fail to change pain threshold. Stimulation of brain GABA receptors with muscimol elicits benzodiazepam-like anticonflict activity, which is dose-related. When the doses are increased above a certain limit, muscimol decreases motor activity and therefore fails to elicit anticonflict activity. These data are consistent with the idea that GABA receptors are involved in the control of behavioral inhibition induced by punishment. This hypothesis finds support in recent biochemical observations of the existence of a strict correlation between benzodiazepine and GABA receptors and with the concept that the GABA receptor is a functional unit which contains GABA receptors, benzodiazepine receptors, chloride ionophores along with modulatory factors all of which are closely associated6. The part of parts of the brain more directly involved in this behavioral action of muscimol and diazepam are now being investigated. ACKNOWLEDGEMENTS
We thank Prof. Povl Krosgaard-Larsen for making available to us THIP and isoguvacine and Dr. D. Della Bella (Zambon SpA, Milan) for a generous supply of muscimol.
REFERENCES 1 Baraldi, M., Grandison, L. and Guidotti, A., Distribution and metabolism of muscimol in the brain and other tissues of the rat, Neuropharmacology, 18 (1979) 57-62.
453 2 Biggio, G., Della Bella, D., Frigeni, V. and Guidotti, A., Potentiation of morphine analgesia by muscimol, Neuropharmacology, 16 (1977) 149-150. 3 Billingsley, M. L. and Kubena, R. K., The effects of naloxone and picrotoxin on the sedative and anticonflict effects of benzodiazepines, Life Sci., 22 (1978) 897-906. 4 Cook, L. and Davidson, A. B., Effects of behaviorally active drugs in a conflict-punishment procedure in rats. In S. Garattini, E. Mussini and L. O. Randall (Eds.), The Benzodiazepines, Raven Press, New York, 1973, pp. 327-345. 5 Costa, E., Guidotti, A., Map, C. C. and Suria, A., New concepts on the mechanism of action of benzodiazepines, Life Sci., 17 (1975) 167-186. 6 Costa, E. and Guidotti, A., Molecular mechanisms in the receptor action of benzodiazepines, Ann. Rev. Pharmacol. Toxicol., 19 (1979) 531-545. 7 D'Amour, F. E. and Smith, D. L., A method of determining loss of pain sensation, J. Pharmacol. exp. Ther., 72 (1941) 74-79. 8 Enna, S. J. and Snyder, S. H., Influence of ions, enzymes and detergents on gamma-aminobutyric acid receptor binding in synaptic membranes of rat brain, Molec. Pharmacol., 13 (1977) 442--453. 9 Enna, S. J., Collins, J. F. and Snyder, S. H., Stereospecificity and structure-activity requirements of GABA receptor binding in rat brain, Brain Research, 124 (1977) 184-190. 10 Grandison, L. and Guidotti, A., Stimulation of food intake by muscimol and beta endorphin, Neuropharmacology, 16 (1977) 533-536. 11 Guidotti, A., Synaptic mechanisms in the action of benzodiazepines. In M. A. Lipton, A. Di Mascio and K. F. Killam (Eds.), Psychopharmacology: a Generation of Progress, Raven Press, New York, 1978, pp. 1349-1357. 12 Haefely, W. E., Behavioral and neuropharmacological aspects of drugs used in anxiety and related states. In M. A. Lipton, A. Di Mascio and K. F. Killam (Eds.), Psychopharmacology: a Generation of Progress, Raven Press, New York, 1978, pp. 1359-1374. 13 Maggi, A. and Enna, S. J., Characteristics of muscimol accumulation in mouse brain after systemic administration, Neuropharmacology, 18 (1979) 361-366. 14 Mohler, H. and Okada, T., Benzodiazepine receptor: demonstration in the central nervous system, Science, 198 (1977) 849-851. 15 Robinson, C. A., Hengeveld, C. A. and Balbian Verster, F.de., Improved polyethylene cannulation techniques, Physiol. Behav., 4 (1969) 123-t24. 16 Soubrie, Ph., Thiebot, M. H. and Simon, P., Enhanced suppressive effects of aversive events induced in rats by picrotoxin: possibility of a GABA control on behavioral inhibition, PharmacoL Biochem. Behav., 10 (1979) 463-469. 17 Stein, L., Wise, C. D. and Berger, B. D., Antianxietyaction of benzodiazepines: decrease in activity of serotonin neurons in the punishment system. In S. Garattini, E. Mussini and L. O. Randall (Eds.), The Benzodiazepines, Raven Press, New York, 1973, pp. 299-326. 18 Stein, L., Wise, C. D. and Belluzzi, J. D., Effects of benzodiazepines on central serotonergic mechanisms. In E. Costa and P. Greengard (Eds.), Mechanisms of Action of Benzodiazepines, Raven Press, New York, 1975, pp. 29-44. 19 Sullivan, J. W., Sepinwall, J. and Cook, L., Anticonflictevaluation of muscimol, a GABA receptor agonist, alone and in combination with diazepam, Fed. Proc., 37 (1978) 2143. 20 Thiebot, M. H., Jobert, A. and Soubrie, P., Effets compares du muscimol et du diazepam sur les inhibitions du comportement induites chez le rat par la nouveaute, la punition etie non-renforcement, Psychopharmacologia (BerL), 61 (1979) 85-89. 21 Vogel, J. R., Beer, B. and Clody, D. E., A simple and reliable conflict procedure for testing antianxiety agents, Psychopharmacologia (Berl.), 21 (1971) 1-7. 22 Zakusov, V. V., Ostrovskaya, R. V., Kozhechkin, S. N., Markovic, V. V., Molodavkin, G. M. and Voronina, T. A., Further evidence for GABAergic mechanisms in the action of benzodiazepines, Arch. int. Pharmacodyn., 229 (1977) 313-326.