Naunyn-Schmiedeberg's Arch Pharmacol (1992) 345 :452 - 460
Naunyn-Schmiedeberg's
Archivesof
Pharmacology © Springer-Verlag 1992
Withdrawal precipitation by benzodiazepine receptor antagonists in dogs chronically treated with diazepam or the novel anxiolytic and anticonvulsant -carboline abecarnil Wolfgang L6scher and Dagmar H6nack Department of Pharmacology, Toxicology and Pharmacy, School of Veterinary Medicine, Bfinteweg 17, W-3000 Hannover 71, Federal Republic of Germany Received October 16, 1991/Accepted November 29, 1991
Summary. The effects of the benzodiazepine (BZ) receptor antagonists flumazenil (Ro 15-1788) and the flcarboline ZK 93426 were compared in dogs before and after chronic treatment with diazepam or the novel BZ receptor ligand abecarnil (ZK 112119). Abecarnil, a flcarboline, is thought to act as partial (low efficacy) and/ or subtype selective agonist at central BZ receptors. Diazepam and abecarnil were administered at doses which, based on previous experiments on anticonvulsant activity, resulted in about equieffective drug concentrations during treatment. In dogs treated with diazepam, 6 mg/kg/day p.o., for 2 weeks, severe abstinence symptoms, including seizures, were precipitated in all animals by i. v. infusion of the BZ receptor antagonists, differences being found in the type of symptoms caused by flumazenil and ZK 93426. In dogs treated with abecarnil, 4 rag/ kg/d s.c., for 6 weeks, only relatively mild abstinence symptoms were precipitated by infusion of flumazenil or ZK 93426, although pharmacologically active plasma concentrations of abecarnil had been maintained throughout the period of treatment. This suggests that BZ receptor ligands which act as partial and/or selective agonists might be more favourable than traditional agonists, such as diazepam, regarding the induction of physical dependence. Key words: Benzodiazepines - fl-Carbolines - Dependence - Tolerance - Flumazenil -- Diazepam -Abecarnil - Precipitated withdrawal
Introduction Development of tolerance and dependence limits the therapeutic use of traditional benzodiazepine (BZ) receptor agonists, such as diazepam (Owen and Tyrer 1983; Woods et al. 1987; Haefely et al. 1990). Recent reports Send offprint requests to W. LSscher at the above address
have indicated that partial (low efficacy) BZ receptor agonists, i.e. compounds which induce smaller fractional responses in their target cells than do full agonists at the same fractional receptor occupancy, may have advantages in this respect, because low-efficacy agonism prevents overstimulation of a receptor population, thereby reducing overdose problems, desensitization responses and adaptation (Haigh and Feely 1988; Ha&ely et al. 1990). Indeed, repeated treatment of mice with the partial BZ receptor agonist bretazenil (Ro 16-6028) produced no significant tolerance to the anticonvutsant effect, in contrast to the effect of full agonists (Haigh and Feely 1988). Furthermore, again unlike full agonists, physical dependence could not be induced in squirrel monkeys after repeated very high doses of bretazenil as assessed by challenge with the BZ receptor antagonist flumazenil (Haefely et al. 1990). Although these data on bretazenil indicate that partial BZ receptor agonism might have important practical consequences, there are also studies on other partial BZ receptor agonists reporting less favorable data. Thus, during chronic treatment of dogs with clonazepam, which has a lower intrinsic efficacy than diazepam (Haefely et al. 1990), there was only slight reduction in anticonvulsant potency during chronic treatment, but severe withdrawal symptoms, including seizures, were seen upon abrupt termination of treatment (Scherkl et al. 1985; Scherkl and Frey 1986). Similarly, during chronic treatment of dogs with clorazepate, a prodrug of desmethyldiazepam, which acts as a partial agonist at BZ receptors (Frey and L6scher 1982; Gobbi et al. 1987), no tolerance developed to the anticonvulsant effect, but again severe withdrawal symptoms were observed upon discontinuation of treatment (Scherkl et al. 1989). This clearly demonstrates that, during treatment with BZs that act as partial BZ receptor agonists, physical dependence may occur without obvious tolerance. This apparent separation of BZ tolerance and dependence has been demonstrated also in rodents (Wilson et al. 1989). More recently, the preclinical pharmacological properties of abecarnil, an anxiolytic and anticonvulsant fl-carboline with high affinity for central BZ receptors,
453 h a v e been d e s c r i b e d ( S t e p h e n s et al. 1990; T u r s k i et al. 1990). L i k e bretazenil, a b e c a r n i l is effective in l o w e r doses t h a n d i a z e p a m in tests p r e d i c t i v e o f anxi01ytic a n d antiepileptic activity, b u t is clearly less p o t e n t t h a n d i a z e p a m in tests o f s e d a t i o n a n d m u s c l e r e l a x a t i o n a n d even a n t a g o n i z e s the m o t o r i m p a i r i n g effects o f t r a d i t i o n a l BZs, a p h a r m a c o l o g i c a l profile c h a r a c t e r i s t i c for p a r t i a l B Z r e c e p t o r a g o n i s m ( S t e p h e n s et al. 1990; T u r s k i et al. 1990). D u r i n g c h r o n i c t r e a t m e n t o f d o g s w i t h a b e c a r n i l , there was o n l y a slight r e d u c t i o n o f the a n t i c o n v u l s a n t p o t e n c y , a n d n o w i t h d r a w a l s y m p t o m s were o b s e r v e d after d i s c o n t i n u a t i o n o f t r e a t m e n t ( L 6 s c h e r et al. 1990). F u r t h e r m o r e , i.v. injection o f the B Z r e c e p t o r a n t a g o n i s t flumazenil after 5 weeks of chronic treatment with a b e c a r n i l d i d n o t p r e c i p i t a t e a n y severe w i t h d r a w a l effects ( L 6 s c h e r et al. 1990). H o w e v e r , the l a c k o f withdrawal symptoms after termination of chronic treatment m i g h t be r e l a t e d to the v e r y slow decline in d r u g levels o b s e r v e d in the a n i m a l s . F u r t h e r m o r e , f l u m a z e n i l was o n l y injected a t a r e l a t i v e l y l o w d o s a g e (1 m g / k g i.v.). T h e p o s s i b i l i t y t h a t BZs a n d fl-carbolines b i n d , at least in p a r t , to different s u b p o p u l a t i o n s o f the B Z r e c e p t o r c o u l d also e x p l a i n the l a c k o f w i t h d r a w a l p r e c i p i t a t i o n b y flumazenil. I n this respect, it is interesting to n o t e t h a t Z K 93426 (ethyl 5 - i s o p r o p o x y - 4 - m e t h y l - f i - c a r b o l i n e - 3 c a r b o x y l a t e ) , a B Z a n t a g o n i s t o f the fl-carboline g r o u p , h a s b e e n s h o w n to p r e c i p i t a t e a n a b s t i n e n c e s y n d r o m e in d i a z e p a m - d e p e n d e n t d o g s t h a t differed f r o m the synd r o m e p r e c i p i t a t e d b y f l u m a z e n i l ( L 6 s c h e r et al. 1990). I n the p r e s e n t s t u d y , d o g s were t r e a t e d c h r o n i c a l l y w i t h d i a z e p a m o r a b e c a r n i l w i t h a c r o s s - o v e r design at d o s e s which, b a s e d o n p r e v i o u s e x p e r i m e n t s o n a n t i c o n v u l s a n t efficacy in d o g s ( F r e y et al. 1984; L 6 s c h e r et al. 1990), result in a b o u t equieffective d r u g c o n c e n t r a t i o n s during chronic treatment. For withdrawal precipitation, either f l u m a z e n i l o r Z K 93426 were given, u s i n g a n i n t r a v e n o u s i n f u s i o n t e c h n i q u e p r e v i o u s l y d e v e l o p e d for withd r a w a l p r e c i p i t a t i o n in d i a z e p a m - d e p e n d e n t rats a n d d o g s ( W i l s o n a n d G a l l a g e r 1988; L 6 s c h e r et al. 1989).
Materials and methods Animals. Six female Beagle dogs (Winkelmann Versuchstierzucht GmbH, Borchen, FRG), aged 3 years and weighing 10 to 11 kg, were used for the experiments. Acute experiments. For testing of precipitated withdrawal, solutions of the BZ antagonists flumazenil or ZK 93426 were infused intravenously at a constant rate until the occurrence of severe abstinence symptoms, such as seizures. This technique of inducing withdrawal has been recently described by Wilson and Gallager (1988) for rats and by L6scher et al. (1989) for dogs. In order to differentiate withdrawal symptoms from effects induced by the BZ antagonists or the vehicle used for solutions of antagonists, it was necessary to study the effects of vehicle and BZ antagonists in diazepam-naive dogs prior to onset of chronic treatment. Flumazenil and ZK 93426 were dissolved in 50% glycofurol in water at a concentration of 10 mg/ml, which was the maximum amount of drug which could be dissolved. Either vehicle (50% glycofurol in water) or vehicle with the solubilized antagonists were infused with an infusion pump (Braun, Melsungen, FRG) through a polyethylene catheter into the cephalic vein of one foreleg. Infusion rate, total infusion volume of
vehicle and total dose of BZ antagonist were chosen on the basis of previous experiments in dogs (L6scher et al. 1989). From these experiments, an infusion rate of 3 ml/min up to a total volume of 2 ml/kg (equivalent to a total dose of 20 mg/kg of the antagonist) was chosen for the chronic experiments with diazepam and abecaruil. Higher infusion rates resulted in too marked effects of vehicle alone (mainly ataxia). The maximum dose of antagonists chosen (20 mg/kg) corresponded to the maximum dose infused intravenously for withdrawal precipitation in diazepam-dependent rats by Wilson and Gallager (1988) and diazepam-dependent dogs by L6scher et al. (1989). For final evaluation, all 6 diazepam-naive dogs received vehicle infusion at the chosen rate and total volume and, 3 - 5 days later, infusion of vehicle plus antagonist (3 dogs flumazenil and 3 dogs ZK 93426). The animals were closely examined for changes in general behaviour during several hours after the infusion as described in detail elsewhere (L6scher et al. 1989). Rectal body temperature was measured at 5, 15, 30, 60, 90, and 120 rain after infusion. Chronic experiments. For the chronic experiments with diazepam, doses and dosing intervals were chosen on the basis of previous experiments in dogs in order to use a treatment protocol which induces physical dependence in this species (L6scher et al. 1989). In all experiments, those dogs which had received flumazenil or ZK 93426 in the acute experiments prior to diazepam treatment received the same antagonist after diazepam treatment to allow a direct comparison of effects. The 6 dogs were orally treated with diazepam at a dose of 2 mg/kg administered daily at 7: 00, 15: 00 and 23 : 00 for 2 weeks. Blood (2 ml) for drug analysis was sampled 1, 2 and 3 h after the first dose as well as immediately before, and 1 and 3 h after the morning doses at day 3, 8 and 11. On day 15, blood was sampled prior to and I h after the last dose of diazepam in the morning. Following blood sampling after 1 h, the respective antagonist was infused at a rate of 3 ml/min with a concentration of 10 mg/ml as described above. In all dogs, the respective antagonist was infused up to the maximum dose of 20 mg/kg, because no severe abstinence symptoms (e.g. seizures) were precipitated during the time of infusion (see Results). Observation of changes in general behaviour and determination of body temperature were carried out as described for the acute experiments. In addition, on the days after the last day of diazepam treatment (and antagonist infusion), dogs were closely examined for behavioural alterations, and body weight and temperature were measured each morning for up to 1 week. Three months after this diazepam trial, the same dogs were used for chronic experiments with daily subcutaneous (s.c.) injection of the partial BZ receptor agonist abecarnil. Previous experiments with this drug in dogs had shown that oral administration of abecarnil is unsuitable for maintenance treatment because of too low bioavailability and too rapid elimination of the fl-carboline after this route of administration (L6scher et al. 1990). However, active drug concentrations can be maintained during chronic treatment by once dally s.c. injection of the drug suspended in peanut oil (L6scher et al. 1990). Indeed, when injected s.c. as suspension in peanut oil, abecarnil is absorbed only slowly so that a once daily injection results in accumulation of plasma drug concentrations during chronic treatment. In the present experiments, abecaruil was daily injected s.c. in the morning at a dose of 4mg/kg for 46 days. This dose was chosen on the basis of previous experiments in dogs (L6scher et al. 1990), indicating that pharmacodynamic (e. g. anticonvulsant) effects obtained with once daily administration of this dosage are similar to those obtained with the protocol chosen for diazepam (see also discussion). The longer treatment period (46 days) chosen for abecarnil as compared with that of diazepam (2 weeks) was derived from previous experiments with partial BZ receptor agonists in dogs (Scherkl et al. 1989), indicating that marked physical dependence can develop with such drugs not before 5 - 6 weeks of chronic treatment. Suspensions of abecarnil in peanut oil (25 mg of drug per ml of oil) were freshly prepared each day. Blood for plasma drug analysis was sampled before and 1, 2, 4 and 8 h after s.c. injection on day 1, 15 and 30 of the chronic treatment
454 period. On day 46, blood was sampled prior to and 2 h after the last dose of abecarnil and the respective antagonist was then infused up to the maximum dose of 20 mg/kg as in the foregoing experiments with diazepam. Observation of behavioural changes and determination of body temperature and body weight were carried out as described above. All observations of behavioural effects produced by BZ receptor antagonists in diazepam-naive dogs and dogs after chronic treatment with diazepam or abecarnil were carried out by the same two observers (W. L. and D. H.). A detailed description of the procedure used to evaluate behavioural effects has been published (L6scher et al. 1989). In addition to describing the behavioural alterations of diazepam- or abecarnil-pretreated dogs in response to BZ receptor antagonists, the signs of withdrawal were weighted according to the "Diazepam Withdrawal Abstinence Scale" generated by McNicholas et al. (1983) for comparing withdrawal syndromes after chronic treatment with different BZ receptor agonists. Weighting factors of this scale are I for gross tremor, 10 for tonic-clonic seizures and 1.4 for disturbed gait, such as stiff-legged walking. Withdrawal scores were obtained by weighing the observed withdrawal signs by these factors and then summing up the weighted scores of the respective group of dogs (separately calculated for diazepam and abecarnil treatment).
Plasma level determinations. Diazepam and its metabolites desmethyl-diazepam and oxazepam were determined in plasma by gas chromatography with electron capture determination as described previously (L6scher and Frey 1981). Abecarnil was determined in plasma by means of a highly sensitive HPLC-method with fluorescence detection as recently described (L6scher et al. 1987; Krause et al. 1989). Drugs. For chronic treatment, diazepam was used in the form of commercial tablets containing 2, 5 or 10 mg of diazepam (Valium; Hoffmann-La Roche, Grenzach-Whylen, FRG). Flumazenil (Ro 15-1788) was a gift from Hoffmann-La Roche (Basle, Switzerland). Abecarnil (ZK 112119; isopropyl-6-benzyloxy-4-methoxymethyl-flcarboline-3-carboxylate) and ZK 93426 (ethyl 5-isopropoxy-4methyl-fl-carboline-3-carboxylate) were obtained from Schering AG (Berlin, FRG). Pure samples of diazepam, desmethyldiazepam, oxazepam and medazepam (used as internal standard) for gas chromatography were obtained from Hoffmann-La Roche. Statistics. Significance of differences between abecarnil and diazepam treated dogs in the incidence of withdrawal signs precipitated by BZ receptor antagonists was calculated by the Chi-square test.
Results
Acute effects of B Z receptor antagonists In the acute experiments with i. v. infusion of BZ antagonists or vehicle alone in diazepam-naive dogs it was found that, at the infusion rate (3 ml/min) and maximum infusion volume (2 ml/kg) chosen, the glycofurol vehicle alone caused transient ataxia (slight weakness in hind legs for about 3 0 - 1 2 0 s after termination of infusion) in 5 of 6 dogs examined and an increase in body temperature in 2 dogs (Table 1). After infusion o f flumazenil (20 mg/kg) ataxia was much more marked and o f longer duration than ataxia observed with vehicle alone and the dogs showed sedation and vocalization, indicating that flumazenil exerted effects of its own (Table 1). In contrast, ataxia occurring after infusion of Z K 93426 (20 mg/kg)
was not different from that observed in the same dogs with vehicle alone, indicating that Z K 93426, at least in dogs, is a more neutral antagonist than is flumazenil.
Chronic experiments with diazepam During three times daily treatment of dogs with 2 mg/ kg diazepam for 2 weeks, diazepam did not cause any observable side-effects nor did it change body temperature or body weight. Plasma concentrations of diazepam and its metabolites during the period of treatment are shown in Table 2. Diazepam was extensively metabolized in the animals so that, even after the first dose, levels of desmethyldiazepam and oxazepam were several times higher than those of the parent drug. After repeated dosing, maximum drug and metabolite levels were higher than those determined after the first dosing with diazepam, indicating cumulation. Immediately prior to infusion o f antagonists, i.e. 1 h after the last dose of diazepam at day 15 of the trial, average levels of diazepam, oxazepam and desmethyldiazepam were 75, 67 and 1200 ng/ml, respectively. Infusion of antagonists precipitated marked withdrawal symptoms in all dogs tested, but differences were observed between the 2 antagonists (Table 1). With flumazenil, all 3 dogs showed agitation or hyperexcitation and generalized tremor during the i.v. infusion, which, however, was continued up to the maximum dose of 20 mg/kg. After the infusion, the dogs were immobile with rigid postures (2 dogs in a prone position) and increased muscle tone in limbs (as determined by palpation) and exhibited generalized tremor, and twitches and jerks of head, limbs or body. In addition, two dogs showed pronounced hyperventilation. Immobility lasted for 10 to 30 rain after which 2 of the animals still had a disturbed gait with retarded setting of paws and/or rigid walking. In i dog, transient circling was seen. Twitches and jerks were observed for up to 3 h after the infusion. With Z K 93426, one dog exhibited tremor and 2 dogs myoclonic jerks during the last minute of infusion. After the infusion, in contrast to flumazenil the dogs were able to run around normally (although hot foot walking with tentative placement and rapid lifting of paws was observed in one animal), but showed twitches (1 dog) or marked generalized myoclonic jerks (2 dogs) during forward locomotion, which resembled the typical generalized epileptic myoclonic jerks of head or body which can be induced by pentylenetetrazol in diazepam-naive dogs (L6scher 1983). Three hours after infusion, one dog exhibited a generalized tonic-clonic seizure, while another dog still showed myoclonic jerks. On the days after the last day of diazepam treatment, all 6 dogs behaved normally and showed no alterations in body weight or body temperature, indicating that the decline in diazepam and metabolite levels did not cause withdrawal symptoms in addition to those already precipitated by the antagonist at times of maximum drug levels. In the experiments with chronic administration of diazepam, there were only minimal differences in drug and metabolite levels between the dogs used for
455
!
o,.o
~o~
o o
Naunyn-Schmiedeberg's
Archivesof
Pharmacology © Springer-Verlag 1992
Withdrawal precipitation by benzodiazepine receptor antagonists in dogs chronically treated with diazepam or the novel anxiolytic and anticonvulsant -carboline abecarnil Wolfgang L6scher and Dagmar H6nack Department of Pharmacology, Toxicology and Pharmacy, School of Veterinary Medicine, Bfinteweg 17, W-3000 Hannover 71, Federal Republic of Germany Received October 16, 1991/Accepted November 29, 1991
Summary. The effects of the benzodiazepine (BZ) receptor antagonists flumazenil (Ro 15-1788) and the flcarboline ZK 93426 were compared in dogs before and after chronic treatment with diazepam or the novel BZ receptor ligand abecarnil (ZK 112119). Abecarnil, a flcarboline, is thought to act as partial (low efficacy) and/ or subtype selective agonist at central BZ receptors. Diazepam and abecarnil were administered at doses which, based on previous experiments on anticonvulsant activity, resulted in about equieffective drug concentrations during treatment. In dogs treated with diazepam, 6 mg/kg/day p.o., for 2 weeks, severe abstinence symptoms, including seizures, were precipitated in all animals by i. v. infusion of the BZ receptor antagonists, differences being found in the type of symptoms caused by flumazenil and ZK 93426. In dogs treated with abecarnil, 4 rag/ kg/d s.c., for 6 weeks, only relatively mild abstinence symptoms were precipitated by infusion of flumazenil or ZK 93426, although pharmacologically active plasma concentrations of abecarnil had been maintained throughout the period of treatment. This suggests that BZ receptor ligands which act as partial and/or selective agonists might be more favourable than traditional agonists, such as diazepam, regarding the induction of physical dependence. Key words: Benzodiazepines - fl-Carbolines - Dependence - Tolerance - Flumazenil -- Diazepam -Abecarnil - Precipitated withdrawal
Introduction Development of tolerance and dependence limits the therapeutic use of traditional benzodiazepine (BZ) receptor agonists, such as diazepam (Owen and Tyrer 1983; Woods et al. 1987; Haefely et al. 1990). Recent reports Send offprint requests to W. LSscher at the above address
have indicated that partial (low efficacy) BZ receptor agonists, i.e. compounds which induce smaller fractional responses in their target cells than do full agonists at the same fractional receptor occupancy, may have advantages in this respect, because low-efficacy agonism prevents overstimulation of a receptor population, thereby reducing overdose problems, desensitization responses and adaptation (Haigh and Feely 1988; Ha&ely et al. 1990). Indeed, repeated treatment of mice with the partial BZ receptor agonist bretazenil (Ro 16-6028) produced no significant tolerance to the anticonvutsant effect, in contrast to the effect of full agonists (Haigh and Feely 1988). Furthermore, again unlike full agonists, physical dependence could not be induced in squirrel monkeys after repeated very high doses of bretazenil as assessed by challenge with the BZ receptor antagonist flumazenil (Haefely et al. 1990). Although these data on bretazenil indicate that partial BZ receptor agonism might have important practical consequences, there are also studies on other partial BZ receptor agonists reporting less favorable data. Thus, during chronic treatment of dogs with clonazepam, which has a lower intrinsic efficacy than diazepam (Haefely et al. 1990), there was only slight reduction in anticonvulsant potency during chronic treatment, but severe withdrawal symptoms, including seizures, were seen upon abrupt termination of treatment (Scherkl et al. 1985; Scherkl and Frey 1986). Similarly, during chronic treatment of dogs with clorazepate, a prodrug of desmethyldiazepam, which acts as a partial agonist at BZ receptors (Frey and L6scher 1982; Gobbi et al. 1987), no tolerance developed to the anticonvulsant effect, but again severe withdrawal symptoms were observed upon discontinuation of treatment (Scherkl et al. 1989). This clearly demonstrates that, during treatment with BZs that act as partial BZ receptor agonists, physical dependence may occur without obvious tolerance. This apparent separation of BZ tolerance and dependence has been demonstrated also in rodents (Wilson et al. 1989). More recently, the preclinical pharmacological properties of abecarnil, an anxiolytic and anticonvulsant fl-carboline with high affinity for central BZ receptors,
453 h a v e been d e s c r i b e d ( S t e p h e n s et al. 1990; T u r s k i et al. 1990). L i k e bretazenil, a b e c a r n i l is effective in l o w e r doses t h a n d i a z e p a m in tests p r e d i c t i v e o f anxi01ytic a n d antiepileptic activity, b u t is clearly less p o t e n t t h a n d i a z e p a m in tests o f s e d a t i o n a n d m u s c l e r e l a x a t i o n a n d even a n t a g o n i z e s the m o t o r i m p a i r i n g effects o f t r a d i t i o n a l BZs, a p h a r m a c o l o g i c a l profile c h a r a c t e r i s t i c for p a r t i a l B Z r e c e p t o r a g o n i s m ( S t e p h e n s et al. 1990; T u r s k i et al. 1990). D u r i n g c h r o n i c t r e a t m e n t o f d o g s w i t h a b e c a r n i l , there was o n l y a slight r e d u c t i o n o f the a n t i c o n v u l s a n t p o t e n c y , a n d n o w i t h d r a w a l s y m p t o m s were o b s e r v e d after d i s c o n t i n u a t i o n o f t r e a t m e n t ( L 6 s c h e r et al. 1990). F u r t h e r m o r e , i.v. injection o f the B Z r e c e p t o r a n t a g o n i s t flumazenil after 5 weeks of chronic treatment with a b e c a r n i l d i d n o t p r e c i p i t a t e a n y severe w i t h d r a w a l effects ( L 6 s c h e r et al. 1990). H o w e v e r , the l a c k o f withdrawal symptoms after termination of chronic treatment m i g h t be r e l a t e d to the v e r y slow decline in d r u g levels o b s e r v e d in the a n i m a l s . F u r t h e r m o r e , f l u m a z e n i l was o n l y injected a t a r e l a t i v e l y l o w d o s a g e (1 m g / k g i.v.). T h e p o s s i b i l i t y t h a t BZs a n d fl-carbolines b i n d , at least in p a r t , to different s u b p o p u l a t i o n s o f the B Z r e c e p t o r c o u l d also e x p l a i n the l a c k o f w i t h d r a w a l p r e c i p i t a t i o n b y flumazenil. I n this respect, it is interesting to n o t e t h a t Z K 93426 (ethyl 5 - i s o p r o p o x y - 4 - m e t h y l - f i - c a r b o l i n e - 3 c a r b o x y l a t e ) , a B Z a n t a g o n i s t o f the fl-carboline g r o u p , h a s b e e n s h o w n to p r e c i p i t a t e a n a b s t i n e n c e s y n d r o m e in d i a z e p a m - d e p e n d e n t d o g s t h a t differed f r o m the synd r o m e p r e c i p i t a t e d b y f l u m a z e n i l ( L 6 s c h e r et al. 1990). I n the p r e s e n t s t u d y , d o g s were t r e a t e d c h r o n i c a l l y w i t h d i a z e p a m o r a b e c a r n i l w i t h a c r o s s - o v e r design at d o s e s which, b a s e d o n p r e v i o u s e x p e r i m e n t s o n a n t i c o n v u l s a n t efficacy in d o g s ( F r e y et al. 1984; L 6 s c h e r et al. 1990), result in a b o u t equieffective d r u g c o n c e n t r a t i o n s during chronic treatment. For withdrawal precipitation, either f l u m a z e n i l o r Z K 93426 were given, u s i n g a n i n t r a v e n o u s i n f u s i o n t e c h n i q u e p r e v i o u s l y d e v e l o p e d for withd r a w a l p r e c i p i t a t i o n in d i a z e p a m - d e p e n d e n t rats a n d d o g s ( W i l s o n a n d G a l l a g e r 1988; L 6 s c h e r et al. 1989).
Materials and methods Animals. Six female Beagle dogs (Winkelmann Versuchstierzucht GmbH, Borchen, FRG), aged 3 years and weighing 10 to 11 kg, were used for the experiments. Acute experiments. For testing of precipitated withdrawal, solutions of the BZ antagonists flumazenil or ZK 93426 were infused intravenously at a constant rate until the occurrence of severe abstinence symptoms, such as seizures. This technique of inducing withdrawal has been recently described by Wilson and Gallager (1988) for rats and by L6scher et al. (1989) for dogs. In order to differentiate withdrawal symptoms from effects induced by the BZ antagonists or the vehicle used for solutions of antagonists, it was necessary to study the effects of vehicle and BZ antagonists in diazepam-naive dogs prior to onset of chronic treatment. Flumazenil and ZK 93426 were dissolved in 50% glycofurol in water at a concentration of 10 mg/ml, which was the maximum amount of drug which could be dissolved. Either vehicle (50% glycofurol in water) or vehicle with the solubilized antagonists were infused with an infusion pump (Braun, Melsungen, FRG) through a polyethylene catheter into the cephalic vein of one foreleg. Infusion rate, total infusion volume of
vehicle and total dose of BZ antagonist were chosen on the basis of previous experiments in dogs (L6scher et al. 1989). From these experiments, an infusion rate of 3 ml/min up to a total volume of 2 ml/kg (equivalent to a total dose of 20 mg/kg of the antagonist) was chosen for the chronic experiments with diazepam and abecaruil. Higher infusion rates resulted in too marked effects of vehicle alone (mainly ataxia). The maximum dose of antagonists chosen (20 mg/kg) corresponded to the maximum dose infused intravenously for withdrawal precipitation in diazepam-dependent rats by Wilson and Gallager (1988) and diazepam-dependent dogs by L6scher et al. (1989). For final evaluation, all 6 diazepam-naive dogs received vehicle infusion at the chosen rate and total volume and, 3 - 5 days later, infusion of vehicle plus antagonist (3 dogs flumazenil and 3 dogs ZK 93426). The animals were closely examined for changes in general behaviour during several hours after the infusion as described in detail elsewhere (L6scher et al. 1989). Rectal body temperature was measured at 5, 15, 30, 60, 90, and 120 rain after infusion. Chronic experiments. For the chronic experiments with diazepam, doses and dosing intervals were chosen on the basis of previous experiments in dogs in order to use a treatment protocol which induces physical dependence in this species (L6scher et al. 1989). In all experiments, those dogs which had received flumazenil or ZK 93426 in the acute experiments prior to diazepam treatment received the same antagonist after diazepam treatment to allow a direct comparison of effects. The 6 dogs were orally treated with diazepam at a dose of 2 mg/kg administered daily at 7: 00, 15: 00 and 23 : 00 for 2 weeks. Blood (2 ml) for drug analysis was sampled 1, 2 and 3 h after the first dose as well as immediately before, and 1 and 3 h after the morning doses at day 3, 8 and 11. On day 15, blood was sampled prior to and I h after the last dose of diazepam in the morning. Following blood sampling after 1 h, the respective antagonist was infused at a rate of 3 ml/min with a concentration of 10 mg/ml as described above. In all dogs, the respective antagonist was infused up to the maximum dose of 20 mg/kg, because no severe abstinence symptoms (e.g. seizures) were precipitated during the time of infusion (see Results). Observation of changes in general behaviour and determination of body temperature were carried out as described for the acute experiments. In addition, on the days after the last day of diazepam treatment (and antagonist infusion), dogs were closely examined for behavioural alterations, and body weight and temperature were measured each morning for up to 1 week. Three months after this diazepam trial, the same dogs were used for chronic experiments with daily subcutaneous (s.c.) injection of the partial BZ receptor agonist abecarnil. Previous experiments with this drug in dogs had shown that oral administration of abecarnil is unsuitable for maintenance treatment because of too low bioavailability and too rapid elimination of the fl-carboline after this route of administration (L6scher et al. 1990). However, active drug concentrations can be maintained during chronic treatment by once dally s.c. injection of the drug suspended in peanut oil (L6scher et al. 1990). Indeed, when injected s.c. as suspension in peanut oil, abecarnil is absorbed only slowly so that a once daily injection results in accumulation of plasma drug concentrations during chronic treatment. In the present experiments, abecaruil was daily injected s.c. in the morning at a dose of 4mg/kg for 46 days. This dose was chosen on the basis of previous experiments in dogs (L6scher et al. 1990), indicating that pharmacodynamic (e. g. anticonvulsant) effects obtained with once daily administration of this dosage are similar to those obtained with the protocol chosen for diazepam (see also discussion). The longer treatment period (46 days) chosen for abecarnil as compared with that of diazepam (2 weeks) was derived from previous experiments with partial BZ receptor agonists in dogs (Scherkl et al. 1989), indicating that marked physical dependence can develop with such drugs not before 5 - 6 weeks of chronic treatment. Suspensions of abecarnil in peanut oil (25 mg of drug per ml of oil) were freshly prepared each day. Blood for plasma drug analysis was sampled before and 1, 2, 4 and 8 h after s.c. injection on day 1, 15 and 30 of the chronic treatment
454 period. On day 46, blood was sampled prior to and 2 h after the last dose of abecarnil and the respective antagonist was then infused up to the maximum dose of 20 mg/kg as in the foregoing experiments with diazepam. Observation of behavioural changes and determination of body temperature and body weight were carried out as described above. All observations of behavioural effects produced by BZ receptor antagonists in diazepam-naive dogs and dogs after chronic treatment with diazepam or abecarnil were carried out by the same two observers (W. L. and D. H.). A detailed description of the procedure used to evaluate behavioural effects has been published (L6scher et al. 1989). In addition to describing the behavioural alterations of diazepam- or abecarnil-pretreated dogs in response to BZ receptor antagonists, the signs of withdrawal were weighted according to the "Diazepam Withdrawal Abstinence Scale" generated by McNicholas et al. (1983) for comparing withdrawal syndromes after chronic treatment with different BZ receptor agonists. Weighting factors of this scale are I for gross tremor, 10 for tonic-clonic seizures and 1.4 for disturbed gait, such as stiff-legged walking. Withdrawal scores were obtained by weighing the observed withdrawal signs by these factors and then summing up the weighted scores of the respective group of dogs (separately calculated for diazepam and abecarnil treatment).
Plasma level determinations. Diazepam and its metabolites desmethyl-diazepam and oxazepam were determined in plasma by gas chromatography with electron capture determination as described previously (L6scher and Frey 1981). Abecarnil was determined in plasma by means of a highly sensitive HPLC-method with fluorescence detection as recently described (L6scher et al. 1987; Krause et al. 1989). Drugs. For chronic treatment, diazepam was used in the form of commercial tablets containing 2, 5 or 10 mg of diazepam (Valium; Hoffmann-La Roche, Grenzach-Whylen, FRG). Flumazenil (Ro 15-1788) was a gift from Hoffmann-La Roche (Basle, Switzerland). Abecarnil (ZK 112119; isopropyl-6-benzyloxy-4-methoxymethyl-flcarboline-3-carboxylate) and ZK 93426 (ethyl 5-isopropoxy-4methyl-fl-carboline-3-carboxylate) were obtained from Schering AG (Berlin, FRG). Pure samples of diazepam, desmethyldiazepam, oxazepam and medazepam (used as internal standard) for gas chromatography were obtained from Hoffmann-La Roche. Statistics. Significance of differences between abecarnil and diazepam treated dogs in the incidence of withdrawal signs precipitated by BZ receptor antagonists was calculated by the Chi-square test.
Results
Acute effects of B Z receptor antagonists In the acute experiments with i. v. infusion of BZ antagonists or vehicle alone in diazepam-naive dogs it was found that, at the infusion rate (3 ml/min) and maximum infusion volume (2 ml/kg) chosen, the glycofurol vehicle alone caused transient ataxia (slight weakness in hind legs for about 3 0 - 1 2 0 s after termination of infusion) in 5 of 6 dogs examined and an increase in body temperature in 2 dogs (Table 1). After infusion o f flumazenil (20 mg/kg) ataxia was much more marked and o f longer duration than ataxia observed with vehicle alone and the dogs showed sedation and vocalization, indicating that flumazenil exerted effects of its own (Table 1). In contrast, ataxia occurring after infusion of Z K 93426 (20 mg/kg)
was not different from that observed in the same dogs with vehicle alone, indicating that Z K 93426, at least in dogs, is a more neutral antagonist than is flumazenil.
Chronic experiments with diazepam During three times daily treatment of dogs with 2 mg/ kg diazepam for 2 weeks, diazepam did not cause any observable side-effects nor did it change body temperature or body weight. Plasma concentrations of diazepam and its metabolites during the period of treatment are shown in Table 2. Diazepam was extensively metabolized in the animals so that, even after the first dose, levels of desmethyldiazepam and oxazepam were several times higher than those of the parent drug. After repeated dosing, maximum drug and metabolite levels were higher than those determined after the first dosing with diazepam, indicating cumulation. Immediately prior to infusion o f antagonists, i.e. 1 h after the last dose of diazepam at day 15 of the trial, average levels of diazepam, oxazepam and desmethyldiazepam were 75, 67 and 1200 ng/ml, respectively. Infusion of antagonists precipitated marked withdrawal symptoms in all dogs tested, but differences were observed between the 2 antagonists (Table 1). With flumazenil, all 3 dogs showed agitation or hyperexcitation and generalized tremor during the i.v. infusion, which, however, was continued up to the maximum dose of 20 mg/kg. After the infusion, the dogs were immobile with rigid postures (2 dogs in a prone position) and increased muscle tone in limbs (as determined by palpation) and exhibited generalized tremor, and twitches and jerks of head, limbs or body. In addition, two dogs showed pronounced hyperventilation. Immobility lasted for 10 to 30 rain after which 2 of the animals still had a disturbed gait with retarded setting of paws and/or rigid walking. In i dog, transient circling was seen. Twitches and jerks were observed for up to 3 h after the infusion. With Z K 93426, one dog exhibited tremor and 2 dogs myoclonic jerks during the last minute of infusion. After the infusion, in contrast to flumazenil the dogs were able to run around normally (although hot foot walking with tentative placement and rapid lifting of paws was observed in one animal), but showed twitches (1 dog) or marked generalized myoclonic jerks (2 dogs) during forward locomotion, which resembled the typical generalized epileptic myoclonic jerks of head or body which can be induced by pentylenetetrazol in diazepam-naive dogs (L6scher 1983). Three hours after infusion, one dog exhibited a generalized tonic-clonic seizure, while another dog still showed myoclonic jerks. On the days after the last day of diazepam treatment, all 6 dogs behaved normally and showed no alterations in body weight or body temperature, indicating that the decline in diazepam and metabolite levels did not cause withdrawal symptoms in addition to those already precipitated by the antagonist at times of maximum drug levels. In the experiments with chronic administration of diazepam, there were only minimal differences in drug and metabolite levels between the dogs used for
455
!
o,.o
~o~
o o
