- Email: [email protected]
Anxiogenic aspects of diazepam withdrawal can be detected in animals
European Journal of Pharmacology, 92 (1983) 127-130
127
Elsevier
Short communication
ANXIOGENIC ASPECTS OF DIAZEPAM WITHDRAWAL CAN BE DETECTED IN ANIMALS MICHAEL EMMETT-OGLESBY *, DAVID SPENCER, JR., MARILYN LEWIS, FATMA ELMESALLAMY and HARBANS LAL
Department of Pharmacologo~', Texas College of Osteopathic Medicine, Fort Worth, Texas 76107, U.S.A. Received 7 June 1983, revised MS received 22 June 1983, accepted 27 June 1983
M. E M M E T T - O G L E S B Y ,
D.G. S P E N C E R , JR., M.W. LEWIS, F. E L M E S A L L A M Y a n d H. LAL, Anxiogenic
aspects of diazepam withdrawal can be detected in animals, E u r o p e a n J. Pharmacol. 92 (1983) 127-130.
Animals can be trained to discriminate the presence of pentylenetetrazol, and this discrimination has previously been proposed as an animal bioassay for anxiogenicity. In rats made dependent on diazepam, pentylenetetrazoi-like stimuli occurred during spontaneous or precipitated (with RO 15-1788) withdrawal; these stimuli were blocked by pentobarbital. These results demonstrate that the pentylenetetrazol-based animal model of anxiety can be used to objectively quantify a subjective aspect of benzodiazepine dependence/withdrawal. Anxiety
Pentylenetetrazol
Diazepam
R O 15-1788
1. Introduction
In man, often the earliest arising and most persistent sign of withdrawal from chronic benzodiazepine use is anxiety (Marks, 1978). Animal studies of benzodiazepine dependence, however, have focused on overt withdrawal signs such as autonomic arousal and convulsions, perhaps because experimental methodology to test subjective events in animals is not widely recognized. One approach to investigating such questions is to use discriminative stimulus methodology (Lal, 1977). Evidence from several studies over the last decade indicate that animals can be trained to use the internal discriminative stimuli (IDS) arising from drug injection as the basis for choosing which of several potential responses is correct. Where direct comparisons are available, animals and man classify stimulus properties (i.e., subjective effects) of drugs in parallel (Glennon and Rosecrans, 1981). * To whom all correspondence should be addressed: Department of Pharmacology, Texas College of Osteopathic Medicine, Camp Bowie at Montgomery, Fort Worth, Texas 76107, U.S.A. 0014-2999/83/$03.00 © 1983 Elsevier Science Publishers B.V.
Withdrawal
Dependence
Pentylenetetrazol, which is reported to be anxiogenic in man (Rodin and Calhoun, 1970), has IDS properties in rats. The IDS that pentylenetetrazol produces in rats are best correlated with the anxiogenic effects of this compound in man, since (1) clinically efficacious anxiolytics block them, (2) compounds such as the fl-carbolines that are anxiogenic in man generalize to them, and (3) although pentylenetetrazol is a convulsant, nonanxiolytic anti-convulsants (such as ethosuximide) do not block these IDS (for review see Lal and Shearman, 1980), We now report that subjective aspects of diazepam withdrawal can be detected in rats using the pentylenetetrazol model of anxiety. The present experiment examined the pentylenetetrazol-like stimuli produced by diazepam withdrawal. The specific questions addressed were: (1) does RO 15-1788 (ethyl 8-fluoro-5,6-dihydro5-methyl-6-oxo-4H-imidazo (1,5-a) (1,4)), a benzodiazepine receptor antagonist (M/3hler et al., 1981), produce dose-dependent generalization to pentylenetetrazol IDS in diazepam-dependent rats, and (2) are the subjective effects of diazepam withdrawal pharmacologically specific to the benzodiazepine receptor?
128
,,,- 100]
2. Materials and methods
Subjects were male Long-Evans hooded rats, trained in conventional behavioral chambers to press ten times on one of two levers for food reward. The lever producing reward was dependent on injection condition: one lever was always correct after receiving pentylenetetrazol (20 mg/kg); the other lever was correct after injection of saline (for methodology and experimental control technology see Lal, 1977; Emmett-Oglesby et al., 1982). No cue other than the effect of the injection substance was permitted to guide the selection of the correct lever. On test sessions, independent of the substance injected, the lever upon which ten responses were first emitted was considered selected. Prior to diazepam dependence, rats were tested to determine which drugs generalized to or antagonized the pentylenetetrazol stimulus. Generalization tests were conducted 15 min following i.p. injection of the substance to be tested. Antagonism tests were conducted by first injecting the substance to be tested, followed 15 min later by pentylenetetrazol (20 mg/kg), and then testing for lever selection 15 min later. After these tests were completed, pentylenetetrazol-saline discrimination-training was halted, and diazepam, 20 mg/kg, was injected every 8 h. On the fifth and sixth days of chronic diazepam, the ability of RO 15-1788 to produce pentylenetetrazol-like stimuli was determined: 15 min after diazepam administration, rats received either 2,5, 10 or 40 mg/kg RO 15-1788 and were tested for lever selection 15 min later. The effects of saline and pentobarbital (10 mg/kg) on the appearance of pentylenetetrazol-like stimuli at 8 h spontaneous withdrawal from diazepam were tested on the seventh and eleventh days of chronic diazepam: they were injected 15 rain prior to discrimination testing. For the pentobarbital test, immediately following lever selection, each subject was injected with RO 151788 (10 mg/kg) and retested 15 min later. On the eighth day of chronic diazepam, the reliability of pentylenetetrazol-IDS detection was reassessed: only those rats selecting the saline lever after 20 mg/kg diazepam injection (16 of 18) were included in the study.
80 ..IN m< Inllc
~
60-
~.
40"
0. 20-
O-
@ ,
//
,'
C
Saline
0
0
215 10 4'0 RO 15-1788 (mg/kg)
Fig. 1. Dose-dependent generalization of RO 15-1788 to the pentylenetetrazol stimulus prior to ( O ) and during ( × ) diazepam dependence. Animals were allowed to make a lever choice 15 min after injection with the indicated dose of RO 15-1788. Each point represents data from at least seven subjects.
3. Results
Before chronic exposure to diazepam, all subjects selected the correct lever following either TABLE 1 Generalization of various treatments to the pentylenetetrazol stimulus before and after repeated diazepam injection. Diazepam Test dependence treatment Before
During c
Saline PTZ Diazepam + PTZ
a
Dose N u m b e r of % Selecting ( m g / k g ) subjects PTZ lever b 20 l0 20
Saline Pentobarbital 10 Pentobarbital 10 + R O 1 5 - 1 7 8 8 10
16 16
0 100
9
11
16 6
56 0
6
0
a Drugs are listed in the order given. The interval between the last injection and testing was 15 min. In tests involving drug combinations, the first substance was administered 30 min prior to testing and 15 rain prior to the second drug. b Percent of subjects completing the first ten responses to obtain reinforcement by selecting the pentylenetetrazol lever. c During the last stage of prolonged diazepam exposure (after 7-11 days of 60 m g / k g diazepam per day), injection and testing was performed 8 h after the most recent diazepam (20 m g / k g ) injection.
129 pentylenetetrazol or saline injection, and diazepam antagonized the pentylenetetrazol cue (table 1, upper portion). In diazepam-dependent rats, RO 151788 produced a dose-dependent selection of the pentylenetetrazol lever with an EDs0 value of 7.1 m g / k g (fig. 1). Rats spontaneously withdrawn for 8-h from diazepam also selected the pentylenetetrazol-lever (table 1, lower portion; X2 = 12.5, P < 0.001). Whether pentylenetetrazol-lever selection was induced by spontaneous or RO 151788-precipitated withdrawal, pentobarbital pretreatment blocked selection of the pentylenetetrazol lever (table 1, lower portion; X2 for both pentobarbital tests = 5.71, P < 0.025).
4. Discussion These results show that internal stimuli produced by withdrawal from diazepam can be objectively assessed in a drug discrimination paradigm. Thus, the discrimination of pentylenetetrazol, an animal assay for subjective drug effects related to anxiety, detects withdrawal from diazepam dependence that has been induced by as few as 5 days of diazepam exposure. The significance of such rapid development of physical dependence to diazepam is unclear; however, it is consistent with clinical reports of anxiety appearing after terminating relatively short-term administration of diazepam. There has been clinical debate as to whether such rapidly developing anxiety is a true sign of drugwithdrawal or is only a return of the patient's pre-existing condition (Marks, 1978). The present data suggest that diazepam dependence and withdrawal in animals generates internal stimuli with pharmacological relevance to the production of anxiety in man during benzodiazepine withdrawal. RO 15-1788 produced dose-dependent generalization to pentylenetetrazol IDS in diazepam dependent rats. In contrast, in non-dependent rats, RO 15-1788 does not generalize to pentylenetetrazol IDS (Gherezghiher and Lal, 1982). These results agree with data suggesting that RO 15-1788 is a benzodiazepine receptor antagonist (M6hler et al., 1981) capable of precipitating signs of withdrawal in diazepam-dependent animals, including subjective signs.
An additional interesting finding of this study is that even though diazepam and its active metabolites have long elimination half-lives (Harvey, 1980), significant withdrawal signs were detected 8 h after the last diazepam dose. These spontaneous withdrawal signs, as well as those precipitated by RO 15-1788, were blocked by pentobarbital. Pentobarbital is a non-benzodiazepine drug with anxiolytic properties (Harvey, 1980), and there is substantial evidence that its effects are not mediated by benzodiazepine receptors (Davis and Ticku, 1981). The present findings support the hypothesis that a non-benzodiazepine receptor mechanism is responsible for the pentobarbital alleviation of benzodiazepine withdrawal. Therefore, the pentylenetetrazol anxiety model provides a powerful new approach for studying the neuropharmacological bases underlying the subjective aspects of dependence and withdrawal.
Acknowledgements We gratefully acknowledge the gifts of diazepam from Hoffman-LaRoche,Nutley, N.J., and of RO 15-1788 from W. Haefely, Hoffman-LaRoche,Basel, Switzerland. This research supported in part by American OsteopathicAssociation Grant 70460 and by Faculty Research Grants 34910 and 34940.
References Davis, W.C. and M.K. Ticku, 1981,Picrotoxininand diazepam bind to two distinct proteins: Further evidencethat pentobarbital may act at the picrotoxinin site, J. Neurosci. 1, 1036. Emmett-Oglesby, M., D. Spencer and D. Arnoult, 1982, A TRS-80-based system for the control of behavioral experiments, Pharmacol. Biochem.Behav. 17, 583. Gherezghiher, T. and H. Lal, 1982, RO 15-1788 selectively reverses antagonism of pentylenetetrazol-induced discriminative stimuli by benzodiazepines but not by barbiturates, Life Sci. 31, 2955. Glennon, R. and J. Rosecrans, 1981, Speculations on the mechanism of action of hallucinogenic idolealkylamines, Biochem. Behav. Rev. 5, 197. Harvey, S.C., Hypnotics and sedatives, in: The Pharmacological Basis of Therapeutics,eds. A.G. Gilman, LS. Goodman and A. Gilman (MacMillan Publishing, New York) p. 339. Lal, H., 1977, Drug-induced discriminable stimuli: past research and future perspectives,in: DiscriminativeStimulus Properties of Drugs, ed. H. Lal (Plenum Press, New York) p. 207.
130 Lal, H. and G. Shearman, 1980, Interoceptive discriminative stimuli in the development of CNS drugs and a case of an animal model of anxiety, Ann. Rev. Med. Chem. 15, 51. Marks, J., 1978, The Benzodiazepines: Use, Overuse, Misuse. Abuse. MTP Press, Lancaster, England. M/3hler, H., W. Burkard, H. Keller, J. Richards and W. Haefely,
1981, Benzodiazepine antagonist RO 15-1788: binding characteristics and interaction with drug-induced changes in dopamine turnover and cerebellar cGMP levels, J. Neurochem. 37, 714. Rodin, E. and H. Calhoun, 1970, Metrazol in a 'normal' volunteer population, J. Nerv. Ment. Dis. 150, 438.
127
Elsevier
Short communication
ANXIOGENIC ASPECTS OF DIAZEPAM WITHDRAWAL CAN BE DETECTED IN ANIMALS MICHAEL EMMETT-OGLESBY *, DAVID SPENCER, JR., MARILYN LEWIS, FATMA ELMESALLAMY and HARBANS LAL
Department of Pharmacologo~', Texas College of Osteopathic Medicine, Fort Worth, Texas 76107, U.S.A. Received 7 June 1983, revised MS received 22 June 1983, accepted 27 June 1983
M. E M M E T T - O G L E S B Y ,
D.G. S P E N C E R , JR., M.W. LEWIS, F. E L M E S A L L A M Y a n d H. LAL, Anxiogenic
aspects of diazepam withdrawal can be detected in animals, E u r o p e a n J. Pharmacol. 92 (1983) 127-130.
Animals can be trained to discriminate the presence of pentylenetetrazol, and this discrimination has previously been proposed as an animal bioassay for anxiogenicity. In rats made dependent on diazepam, pentylenetetrazoi-like stimuli occurred during spontaneous or precipitated (with RO 15-1788) withdrawal; these stimuli were blocked by pentobarbital. These results demonstrate that the pentylenetetrazol-based animal model of anxiety can be used to objectively quantify a subjective aspect of benzodiazepine dependence/withdrawal. Anxiety
Pentylenetetrazol
Diazepam
R O 15-1788
1. Introduction
In man, often the earliest arising and most persistent sign of withdrawal from chronic benzodiazepine use is anxiety (Marks, 1978). Animal studies of benzodiazepine dependence, however, have focused on overt withdrawal signs such as autonomic arousal and convulsions, perhaps because experimental methodology to test subjective events in animals is not widely recognized. One approach to investigating such questions is to use discriminative stimulus methodology (Lal, 1977). Evidence from several studies over the last decade indicate that animals can be trained to use the internal discriminative stimuli (IDS) arising from drug injection as the basis for choosing which of several potential responses is correct. Where direct comparisons are available, animals and man classify stimulus properties (i.e., subjective effects) of drugs in parallel (Glennon and Rosecrans, 1981). * To whom all correspondence should be addressed: Department of Pharmacology, Texas College of Osteopathic Medicine, Camp Bowie at Montgomery, Fort Worth, Texas 76107, U.S.A. 0014-2999/83/$03.00 © 1983 Elsevier Science Publishers B.V.
Withdrawal
Dependence
Pentylenetetrazol, which is reported to be anxiogenic in man (Rodin and Calhoun, 1970), has IDS properties in rats. The IDS that pentylenetetrazol produces in rats are best correlated with the anxiogenic effects of this compound in man, since (1) clinically efficacious anxiolytics block them, (2) compounds such as the fl-carbolines that are anxiogenic in man generalize to them, and (3) although pentylenetetrazol is a convulsant, nonanxiolytic anti-convulsants (such as ethosuximide) do not block these IDS (for review see Lal and Shearman, 1980), We now report that subjective aspects of diazepam withdrawal can be detected in rats using the pentylenetetrazol model of anxiety. The present experiment examined the pentylenetetrazol-like stimuli produced by diazepam withdrawal. The specific questions addressed were: (1) does RO 15-1788 (ethyl 8-fluoro-5,6-dihydro5-methyl-6-oxo-4H-imidazo (1,5-a) (1,4)), a benzodiazepine receptor antagonist (M/3hler et al., 1981), produce dose-dependent generalization to pentylenetetrazol IDS in diazepam-dependent rats, and (2) are the subjective effects of diazepam withdrawal pharmacologically specific to the benzodiazepine receptor?
128
,,,- 100]
2. Materials and methods
Subjects were male Long-Evans hooded rats, trained in conventional behavioral chambers to press ten times on one of two levers for food reward. The lever producing reward was dependent on injection condition: one lever was always correct after receiving pentylenetetrazol (20 mg/kg); the other lever was correct after injection of saline (for methodology and experimental control technology see Lal, 1977; Emmett-Oglesby et al., 1982). No cue other than the effect of the injection substance was permitted to guide the selection of the correct lever. On test sessions, independent of the substance injected, the lever upon which ten responses were first emitted was considered selected. Prior to diazepam dependence, rats were tested to determine which drugs generalized to or antagonized the pentylenetetrazol stimulus. Generalization tests were conducted 15 min following i.p. injection of the substance to be tested. Antagonism tests were conducted by first injecting the substance to be tested, followed 15 min later by pentylenetetrazol (20 mg/kg), and then testing for lever selection 15 min later. After these tests were completed, pentylenetetrazol-saline discrimination-training was halted, and diazepam, 20 mg/kg, was injected every 8 h. On the fifth and sixth days of chronic diazepam, the ability of RO 15-1788 to produce pentylenetetrazol-like stimuli was determined: 15 min after diazepam administration, rats received either 2,5, 10 or 40 mg/kg RO 15-1788 and were tested for lever selection 15 min later. The effects of saline and pentobarbital (10 mg/kg) on the appearance of pentylenetetrazol-like stimuli at 8 h spontaneous withdrawal from diazepam were tested on the seventh and eleventh days of chronic diazepam: they were injected 15 rain prior to discrimination testing. For the pentobarbital test, immediately following lever selection, each subject was injected with RO 151788 (10 mg/kg) and retested 15 min later. On the eighth day of chronic diazepam, the reliability of pentylenetetrazol-IDS detection was reassessed: only those rats selecting the saline lever after 20 mg/kg diazepam injection (16 of 18) were included in the study.
80 ..IN m< Inllc
~
60-
~.
40"
0. 20-
O-
@ ,
//
,'
C
Saline
0
0
215 10 4'0 RO 15-1788 (mg/kg)
Fig. 1. Dose-dependent generalization of RO 15-1788 to the pentylenetetrazol stimulus prior to ( O ) and during ( × ) diazepam dependence. Animals were allowed to make a lever choice 15 min after injection with the indicated dose of RO 15-1788. Each point represents data from at least seven subjects.
3. Results
Before chronic exposure to diazepam, all subjects selected the correct lever following either TABLE 1 Generalization of various treatments to the pentylenetetrazol stimulus before and after repeated diazepam injection. Diazepam Test dependence treatment Before
During c
Saline PTZ Diazepam + PTZ
a
Dose N u m b e r of % Selecting ( m g / k g ) subjects PTZ lever b 20 l0 20
Saline Pentobarbital 10 Pentobarbital 10 + R O 1 5 - 1 7 8 8 10
16 16
0 100
9
11
16 6
56 0
6
0
a Drugs are listed in the order given. The interval between the last injection and testing was 15 min. In tests involving drug combinations, the first substance was administered 30 min prior to testing and 15 rain prior to the second drug. b Percent of subjects completing the first ten responses to obtain reinforcement by selecting the pentylenetetrazol lever. c During the last stage of prolonged diazepam exposure (after 7-11 days of 60 m g / k g diazepam per day), injection and testing was performed 8 h after the most recent diazepam (20 m g / k g ) injection.
129 pentylenetetrazol or saline injection, and diazepam antagonized the pentylenetetrazol cue (table 1, upper portion). In diazepam-dependent rats, RO 151788 produced a dose-dependent selection of the pentylenetetrazol lever with an EDs0 value of 7.1 m g / k g (fig. 1). Rats spontaneously withdrawn for 8-h from diazepam also selected the pentylenetetrazol-lever (table 1, lower portion; X2 = 12.5, P < 0.001). Whether pentylenetetrazol-lever selection was induced by spontaneous or RO 151788-precipitated withdrawal, pentobarbital pretreatment blocked selection of the pentylenetetrazol lever (table 1, lower portion; X2 for both pentobarbital tests = 5.71, P < 0.025).
4. Discussion These results show that internal stimuli produced by withdrawal from diazepam can be objectively assessed in a drug discrimination paradigm. Thus, the discrimination of pentylenetetrazol, an animal assay for subjective drug effects related to anxiety, detects withdrawal from diazepam dependence that has been induced by as few as 5 days of diazepam exposure. The significance of such rapid development of physical dependence to diazepam is unclear; however, it is consistent with clinical reports of anxiety appearing after terminating relatively short-term administration of diazepam. There has been clinical debate as to whether such rapidly developing anxiety is a true sign of drugwithdrawal or is only a return of the patient's pre-existing condition (Marks, 1978). The present data suggest that diazepam dependence and withdrawal in animals generates internal stimuli with pharmacological relevance to the production of anxiety in man during benzodiazepine withdrawal. RO 15-1788 produced dose-dependent generalization to pentylenetetrazol IDS in diazepam dependent rats. In contrast, in non-dependent rats, RO 15-1788 does not generalize to pentylenetetrazol IDS (Gherezghiher and Lal, 1982). These results agree with data suggesting that RO 15-1788 is a benzodiazepine receptor antagonist (M6hler et al., 1981) capable of precipitating signs of withdrawal in diazepam-dependent animals, including subjective signs.
An additional interesting finding of this study is that even though diazepam and its active metabolites have long elimination half-lives (Harvey, 1980), significant withdrawal signs were detected 8 h after the last diazepam dose. These spontaneous withdrawal signs, as well as those precipitated by RO 15-1788, were blocked by pentobarbital. Pentobarbital is a non-benzodiazepine drug with anxiolytic properties (Harvey, 1980), and there is substantial evidence that its effects are not mediated by benzodiazepine receptors (Davis and Ticku, 1981). The present findings support the hypothesis that a non-benzodiazepine receptor mechanism is responsible for the pentobarbital alleviation of benzodiazepine withdrawal. Therefore, the pentylenetetrazol anxiety model provides a powerful new approach for studying the neuropharmacological bases underlying the subjective aspects of dependence and withdrawal.
Acknowledgements We gratefully acknowledge the gifts of diazepam from Hoffman-LaRoche,Nutley, N.J., and of RO 15-1788 from W. Haefely, Hoffman-LaRoche,Basel, Switzerland. This research supported in part by American OsteopathicAssociation Grant 70460 and by Faculty Research Grants 34910 and 34940.
References Davis, W.C. and M.K. Ticku, 1981,Picrotoxininand diazepam bind to two distinct proteins: Further evidencethat pentobarbital may act at the picrotoxinin site, J. Neurosci. 1, 1036. Emmett-Oglesby, M., D. Spencer and D. Arnoult, 1982, A TRS-80-based system for the control of behavioral experiments, Pharmacol. Biochem.Behav. 17, 583. Gherezghiher, T. and H. Lal, 1982, RO 15-1788 selectively reverses antagonism of pentylenetetrazol-induced discriminative stimuli by benzodiazepines but not by barbiturates, Life Sci. 31, 2955. Glennon, R. and J. Rosecrans, 1981, Speculations on the mechanism of action of hallucinogenic idolealkylamines, Biochem. Behav. Rev. 5, 197. Harvey, S.C., Hypnotics and sedatives, in: The Pharmacological Basis of Therapeutics,eds. A.G. Gilman, LS. Goodman and A. Gilman (MacMillan Publishing, New York) p. 339. Lal, H., 1977, Drug-induced discriminable stimuli: past research and future perspectives,in: DiscriminativeStimulus Properties of Drugs, ed. H. Lal (Plenum Press, New York) p. 207.
130 Lal, H. and G. Shearman, 1980, Interoceptive discriminative stimuli in the development of CNS drugs and a case of an animal model of anxiety, Ann. Rev. Med. Chem. 15, 51. Marks, J., 1978, The Benzodiazepines: Use, Overuse, Misuse. Abuse. MTP Press, Lancaster, England. M/3hler, H., W. Burkard, H. Keller, J. Richards and W. Haefely,
1981, Benzodiazepine antagonist RO 15-1788: binding characteristics and interaction with drug-induced changes in dopamine turnover and cerebellar cGMP levels, J. Neurochem. 37, 714. Rodin, E. and H. Calhoun, 1970, Metrazol in a 'normal' volunteer population, J. Nerv. Ment. Dis. 150, 438.