Journal of Psychopharmacology
10(3) (1996) 206-213
m1996 British Associadon for
Psychopharinacology
Comparison of benzodiazepine (BZ) receptor agonists in two rodent activity tests P. J.
, D. Williamson and G. R. Dawson 1 Bayley, G. D. Bentley, A. Jackson
Merck Sharp & Dohme Research Laboratories, Neuroscience Research Centre, Terlings Park, Eastwick Road, Harlow, Essex CM20 2QR, UK Present address: Laboratory of Experimental Psychology, School of Biological Sciences, University of Sussex, Falmer, Brighton, Sussex 1 BN1 9QG, UK
The effects of four BZ receptor ligands in an operant test were compared with a rotarod test. In the operant test, rats were trained to pull a chain on a schedule that regulates the probability of delivery of food pellets to maintain a steady chain-pulling rate across a 1 h test session. For the rotarod test, mice were trained to remain on a rotarod for 2 min. Diazepam (0.1-3.0 mg/kg, i.p.), FG 8205 (0.1-3.0 mg/kg, i.p.), quazepam (3.0-60.0 mg/kg, i.p.) and zolpidem (0.3-10.0 mg/kg, i.p.) each produced dose-related impairments of performance in both the chainpulling test and the mouse rotarod test. Furthermore, the impairment in performance induced by FG 8205 (10.0 mg/kg, p.o.) was dose-dependently reversed by the BZ receptor antagonist, flumazenil (1.0-10.0 mg/kg, i.p.), indicating that the chain-pulling deficit was mediated via BZ receptor activation. Diazepam, FG 8205 and quazepam all had comparable potencies in both the rotarod assay and the chain-pulling test. However, zolpidem suppressed the chain-pulling rates at a dose 30-fold lower than that required to induce a significant deficit in the rotarod performance. As zolpidem is a preferentially sedative compound, this pattern of results is consistent with the hypothesis that the chain-pulling test is sensitive to sedation induced by BZ receptor
agonists. Key words: diazepam; FG 8205; flumazenil; quazepam; sedation; zolpidem
(1995)]. At least one a-subunit and one y-subunit required to form a fully functional receptor in vivo (Pritchett, Luddens and Seeburg, 1989). The properties of the BZ binding site depend on the combination of a- and ysubunits, but it is particularly important which a-subunit is McKernan
Introduction
are
A full receptor agonist at the benzodiazepine on the GABAA receptor complex, such as
(BZ) binding site diazepam, has a
range of
properties including anxiolysis, anticonvulsant, myorelaxant and sedative effects (Braestrup et al., 1984). Whilst all of these properties are clinically useful, they are rarely required simultaneously. For example, although diazepam is an effective treatment for anxiety, the accompanying sedation (a feeling of tiredness or drowsiness), ataxia (inability to coordinate voluntary movement) and myorelaxation (a decrease in muscle tone) all reduce the patient’s ability to perform their normal daily activities. Thus, in recent years much effort has been devoted to the development of BZ drugs that have anxiolytic effects, but are devoid of sedative, ataxic and myorelaxation effects. Two approaches have been taken to the development of anxioselective drugs. The first, the partial agonist approach, is to discover compounds that have sufficient efficacy at the BZ receptor to induce anxiolysis, but not enough to induce sedation (Haefely, Martin and Schoch, 1990). The second is to discover drugs that are selective for GABAA receptor subunits. Molecular cloning techniques have identified 14 GABAA receptor subunits in the mammalian brain [CX1-- 0.12]. FG 8205 dose dependently reduced the chain-pulling rate and an ANOVA of the overall mean chain-pulling rates showed that while a dose of 0.1mg/kg of FG 8205 did not significantly reduce the chainpulling rates at any time point, a dose of 0.3 mg/kg FG 8205 and above significantly reduced the chain-pulling rates compared to vehicle treated controls (Fig. 1 d, collapsed data). Post hoc tests showed that the mean chain-pulling rates of the groups given 0.3 and 1.0 mg/kg FG 8205 first =
=
differed from the vehicle-treated control group 20 min postand were significantly (p < 0.05) different from those of the vehicle-treated rats for the remainder of the session. The mean chain-pulling rates of the rats given 3.0 mg/kg FG 8205 differed significantly from vehicle-treated controls for the duration of the session (p < 0.05). The results for zolpidem across the chain-pulling test session are shown in Fig. 1 (e) and the mean chain-pulling rates for the collapsed data (expressed as a percentage of the baseline chainpulling rate) are shown in Fig. 1 (f). An ANOVA of the chainpulling rates obtained at 10 min intervals revealed a significant main effect of treatment [F(4,44) = 10.10, p < 0.01], but not of time [F(5,220) =1.77, p > 0.12]. However, there was a significant interaction between treatment and time [F(20,220) = 2.15, p < 0.01]. Zolpidem dose dependently reduced the mean chain-pulling rates. Post hoc tests showed that the chain-
injection
209
210
1.0 mg/kg differed from those of the vehicle-treated control group for only the first 20 min of the session (p < 0.05) and that rats given 3.0 mg/kg zolpidem differed from vehicle-treated controls for only the first 30 min of the session. Only the chain-pulling rates of the rats given 10.0 mg/kg zolpidem were significantly lower than those of the vehicle-treated controls for the duration of the session. The effect of quazepam on the mean chain-pulling rates across the test session is shown in Fig. 1 (g) and the mean chainpulling rates for the collapsed data (expressed as a percentage of the baseline chain-pulling rate) for the whole session are shown in Fig. 1 (h). An ANOVA of the mean chain-pulling rates obtained at 10 min intervals revealed a significant main and of time effect of treatment [F(4,45) = 8.90, p < 0.01] [F(5,225) = 9.38, p < 0.01]and a treatment by time interaction [F(20,225) = 3.54, p < 0.01]. An ANOVA of the overall mean percentage chain-pulling rates showed that the lowest dose of quazepam to reduce the chain-pulling rates was 10.0 mg/kg. Post hoc tests showed that the mean chain-pulling rates of the group given 60 mg/kg of quazepam differed from the vehicletreated rats for the duration of the session. By contrast, the mean chain-pulling rate of the group given 10.0 mg/kg quazepam differed significantly from that of the vehicletreated control group only 20 min after dosing.
pulling rates of rats given
Effect of flumazenil and FG 8205 on chain-pulling rates Figure 2(a) shows the effect of flumazenil and FG 8205 on the chain-pulling rates at 5 min intervals across the 30 min session. Figure 2(b) shows the mean chain-pulling rates (expressed as a percentage of the baseline chain-pulling rate) for the session as a whole. An ANOVA with factors of treatment and time revealed significant main effects of treatment [F(5,41) 25.72, =
p < 0.01]
] and of time
[F(5,205) = 20.79, p < 0.01] ] and
a
by time interaction [~25,205) = 3.58, p < 0.01]. Post hoc tests showed that there were no significant differences between the group given 10.0 mg/kg flumazenil/vehicle and the treatment
vehicle/vehicle control group (Fig.2b). However, the mean chain-pulling rate of the group given 10.0 mg/kg FG 8205/ vehicle was significantly lower than vehicle/vehicle controls throughout the session (Fig.2b). This reduction in chainpulling rates induced by 10.0 mg/kg FG 8205 was dose dependently reversed by flumazenil. Full reversal of the effects of 10.0 mg/kg FG 8205 was seen at 10.0 mg/kg Flumazenil since no significant differences in the chain-pulling rates were found between groups given 10.0 mg/kg FG 8205/10.0 mg/kg flumazenil or vehicle/vehicle controls at any time point across the session (Fig. 2b). A partial reversal of the effects of 10.0 mg/kg FG 8205 was seen at doses of flumazenil below 10.0 mg/kg. Hence, the chain-pulling rates of rats given 10.0 mg/kg FG 8205/3.0 mg/kg flumazenil first differed from the vehicle/ vehicle control group 1 S min from the start of the session and were significantly different from the control rats for the remainder of the session. The chain-pulling rates of rats given 10.0 mg/kg FG 8205/ 1.0 mg/kg flumazenil differed significantly from controls at all time points except 10 min from the start of the session.
Rotarod Table 1 shows the mean latency to fall from the rotarod for each dose for the four drug groups. FSD, defined significantly as the lowest dose at which the latency to fall differed significantly from vehicle treated controls (p < 0.05), was found for each drug. Both diazepam and FG 8205 had an FSD of 3.0 mg/kg. Zolpidem and quazepam were found to have FSDs of 30.0 mg and 10.0 mg/kg respectively.
Figure 2 The reversal of the effects of FG 8205 by flumazenil in the rat chain-pulling task. The chain-pulling rates are shown as a percentage of the mean chain-pulling rate of the previous baseline day and the overall standard error of mean (SEM) chain-pulling rate is shown. The mean chain-pulling rates for each 5 min period are shown in the left-hand panel for each treatment group. The right-hand panel shows the overall session mean chain-pulling rate. An asterisk above a bar indicates a significant difference from vehicle-treated control group (Student-Newman-Keuls test, p < 0.05)
211 Table 1 rotarod
zolpidem
latency ( f SE) in seconds for mice to fall from a administration of either diazepam, FG 8205, quazepam
The
mean
following or
An asterisk indicates a significant difference from the vehicle-treated control group (Dunnetts post hoc test, p < 0.05). NT, not tested.
Discussion The aim of the present study was to measure the effects of several compounds known to act at the BZ receptor in a rat operant chain-pulling test and to compare the results obtained to the mouse rotarod task, a standard measure of motor disruption (Haefeley et al., 1981). The results for the full BZ receptor agonist diazepam were consistent with previous findings and diazepam reduced the latency to fall from a rotarod. In addition, diazepam was found to be more potent in reducing operant responding in the rat chain-pulling test. The partial agonist FG 8205 had similar effects to diazepam and was found to be more potent in the rat chain-pulling test than the mouse rotarod. Deficits in the rat chain-pulling performance induced by FG 8205 were reversed by the BZ receptor antagonist, flumazenil. Finally, two full agonists, zolpidem and quazepam, with selectivity for BZI receptors were tested. Both zolpidem and quazepam reduced the rat chain pulling at lower doses than those which affected the rotarod performance. Taken together, these results support the validity of the chainpulling test as a measure of the suppressant effects of BZ receptor agonists and partial agonists. Diazepam was found to induce performance deficits in the mouse rotarod test at a similar dose to those previously reported to be effective in reducing mouse spontaneous locomotor activity and rotarod performance (e.g. 1.0 mg/kg; Bourin et al., 1992). Diazepam induced a deficit in the rat chain-pulling test at 1.0 mg/kg for the first half of the 60 min test session while at the highest dose, diazepam (3.0 mg/kg) reduced the chainpulling rates across the entire session which is consistent with the short half-life of 1.1 h for diazepam in the rat (Klotz, Antonin and Bieck, 1976). Diazepam is generally reported to reduce rat locomotor activity and reduce the time to fall from a rotarod at approximately 1O.Omg/kg (Haefely, 1991). The BZ receptor partial agonist FG 8205 reduced the latency of mice to fall from a rotarod at 3.0 mg/kg, a lower dose than that previously reported (30.0 mg/kg) to affect mouse rotarod performance (Tricklebank et al., 1990). A performance deficit occurred in the rat chain-pulling test at doses of FG 8205 which are not effective in other tests of motor performance. Thus, 0.3 mg/kg FG 8205 induced a deficit in the chain-pulling performance, whereas a higher dose
of 3.0 mg/kg of FG 8205 was required before deficits in the rat rotarod and locomotor activity tests were seen (unpublished results). The deficit in the rat chain-pulling test, seen at higher doses of FG 8205, occurred soon after the start of the session and lasted for the duration of the session, an observation that is consistent with the plasma concentration curve for FG 8205 which reaches a peak 10 min after oral administration and has a brain half-life of 3.6 h in the rat (unpublished results). Taken together with the results for diazepam, these findings show that the rat chain-pulling test compares favourably with established tests of motor performance in its ability to detect the adverse motor effects of BZ receptor agonists and partial agonists. In a separate experiment, we also demonstrated that the FG 8205 deficit could be reversed by the BZ receptor antagonist, flumazenil. The highest dose of flumazenil ( 10.0 mg/kg) antagonized the effects of 10.0 mg/kg FG 8205 throughout the session; however lower doses of flumazenil (1.0-3.0 mg/kg) only blocked the effects of FG 8205 in the early part of the test session. The lack of effect of flumazenil in the latter part of the session probably reflected the relatively short half-life of flumazenil of 8.3 min in the rat (Mandema, Gubbens-Stibbe and Danhof, 1991) compared to that of FG 8205. The BZ agonist zolpidem has known selectivity for the BZ1 receptor subtype (Benavides et al., 1988), a property which is now thought to confer selectivity for a,-containing BZ receptors (Pritchett, Luddens and Seeburg, 1989). Zolpidem was found to reduce the chain-pulling rates at a dose of 1.0 mg/ kg. The results obtained with zolpidem were clearly different to the non-selective drugs as the chain-pulling rates were reduced at a dose 30-fold lower than that found to induce a deficit in the mouse rotarod. In comparison, FG 8205 reduced the chain-pulling rates at a dose 10-fold lower than that found to induce a rotarod deficit and diazepam was found to be equipotent in both tests. Low doses of zolpidem only affected chain pulling for the first half of the session which is consistent with the short half-life of approximately 30 min for zolpidem in the rat (Arbilla et al., 1985). The rotarod results were consistent with previous experiments which report that unlike other BZ receptor agonists such as diazepam, zolpidem does not produce a rotarod deficit until relatively high doses are reached when given 30 min before testing in both mice (EDso =18.0 mg/kg; Perrault et al., 1990) and rats (ED50 6.0 mg/kg; Sanger et al., 1987). However, lower doses of zolpidem have been reported to be behaviourally active and zolpidem has been found to reduce mouse locomotor activity (EDso =1.2 mg/kg; Perrault et al., 1990) and rat open field locomotor activity (1.0 mg/kg; Sanger and Zivkovic, 1988) when given 30 min before testing and Sanger and Zivkovic (1987) have also reported that 0.3 mg/kg of zolpidem reduced rat lever pressing in an operant test. Considering the short half-life of zolpidem, differences in the effective dose of zolpidem between studies could simply reflect the different pretreatment times used. However, the studies cited above all used a pre-treatment time of 30 min for zolpidem and the same pretreatment time was used in the present mouse rotarod experiment. Rather, the discrepancy between the high dose of zolpidem needed to produce a rotarod deficit and the low effective doses in other tests of motor performance suggests that zolpidem may induce specific depressant effects. Considering the potent hypnotic effect of zolpidem in humans (Nicholson =
212
and Pascoe, 1986), it is reasonable to assume that the deficits in locomotor activity, lever pressing and chain pulling seen at relatively low doses of zolpidem are due to sedation. The other BZI selective agonist, quazepam, was equipotent in its effects on the mouse rotarod and rat chain-pulling tests and was active at 10.0 mg/kg in both tests. The rotarod results were consistent with previous findings which reported that in mice 4.9 mg/kg of quazepam reduced the latency to fall from a rotarod (Perrault et al., 1990) and 2.5 mg/kg of quazepam reduced spontaneous locomotor activity (Ongini et al., 1982). In the rat chain-pulling test, 10.0 mg/kg of quazepam reduced performance, although its effects and that of 30.0 mg/kg disappeared rapidly, indicating that the compound has a short duration of action. Quazepam has previously been reported to impair locomotor activity and rotarod performance in the rat at a dose 10-fold higher than that found to disrupt chain pulling
(>,I 00 mg/kg; Corradi et al., 1985). Since chain pulling was monitored across the entire test session, behavioural disruption could be observed at low doses. When data were collapsed across the test session however, quazepam was found to induce significant deficits only at the highest dose tested (60.0 mg/kg), a dose closer to those reported to impair locomotor activity and rotarod performance in the rat. Since quazepam induced maximal behavioural disruption soon after dosing, it seems likely that the compound would have reduced mouse rotarod performance at shorter pre-treatment times than used in the present study. It is also interesting to note that although zolpidem was 30-fold more potent in the chain-pulling test than in the rotarod test, no such separation in activity was found for quazepam. These results are similar to those reported by Perrault et al. (1990) who found that while zolpidem and quazepam both induced sedation at similar doses, as measured by exploratory activity in mice, zolpidem was 4-fold less potent than quazepam at inducing ataxic effects in the mouse rotarod. Since zolpidem and quazepam have a similar affinity for CX1containing GABAA receptors, it is not clear why the two compounds should show different behavioural profiles. Perrault et al. (1990) suggested that quazepam has a lower intrinsic activity than zolpidem and thus may be a partial agonist. Relative to full BZ receptor agonists, partial agonists at the BZ receptor produce less muscle relaxation, sedation and ataxia while retaining anxiolytic and anticonvulsant activity (Haefely, Martin and Schoch, 1990; Haefely, 1991). If quazepam were a partial agonist, this would account for the finding that quazepam was less potent than zolpidem in disrupting chain pulling. However, quazepam would also be expected to be less potent than zolpidem in the rotarod test, which was not the case. While the reason for the differences in the behavioural effects of zolpidem and quazepam remain to be determined, the results indicate that the chain pulling and rotarod tests may be sensitive to different properties of certain compounds. Clinical results from a number of trials involving BZ receptor partial agonists are now accumulating and confirm that partial agonists are indeed anxiolytic, but are also sedative at therapeutic doses (Potokar and Nutt, 1994). While the sedative effects of the BZ receptor partial agonists may be species specific (Haefely, 1988), from the pre-clinical perspective it is imperative that rodent tests have sufficient sensitivity to detect low levels of sedation. Tricklebank et al. (1990) reported that 0.5 mg/kg of the BZ receptor partial
agonists FG 8205 was needed to induce anxiolytic-like effects in rats. In the present study, a dose of 0.3 mg/kg of FG 8205 was found to reduce chain pulling which implies that the separation between the anxiolytic and motor disrupting effects of partial agonists, such as FG 8205, may be less than initially hoped. Attention has also been given to the hypothesis that different BZ receptor subtypes which occur in different regions of the brain have different physiological functions, with the possibility that subtype-selective compounds may show only part of the behavioural effects seen with non-selective BZ receptor agonists (Nielsen and Braestrup, 1980). For instance, it is known that the properties of the BZ binding site depend largely on which a-subunit is present in the GABAA receptor (Pritchett, Luddens and Seeburg, 1989) and both quazepam and zolpidem, which are selective for a,-containing BZ receptors, have been suggested to be sedative at doses which do not induce ataxia (Miller et al., 1992). Behavioural findings that both quazepam and zolpidem are more potent in reducing spontaneous locomotor activity than in reducing rotarod performance have been used to lend support to the idea that a,-selective BZ agonists are ’hypnoselective’. Indeed, the present results, which show that quazepam and zolpidem reduced the chain-pulling rates are not inconsistent with a predominantly sedative behavioural profile of ai-selective BZ receptor agonists. Unfortunately, there are currently few BZ receptor ligands with sufficient selectivity for BZ subtypes to allow meaningful correlations to be made between behaviour and BZ receptor subtypes [for a review see Doble and Martin (1992)]. However, the finding that the chain-pulling test may detect low levels of motor disruption may be useful when assessing subtype-selective BZ receptor ligands when such compounds become available. The rat chain-pulling test is likely to measure a number of factors, only one of which may be sedation and in this respect it may be similar to other rodent measures of motor performance. The rotarod test is perhaps the most widely used such measure. Despite the popularity of the rotarod test, little consensus exists as to what the test measures. Thus, it has been described as an index of sedation, muscle relaxation, ataxia, fatigue or even ’neurotoxicity’ and ’neuronal deficit’ (Haefely et al., 1981). The spontaneous locomotor activity test, which is also commonly used to assess the motor effects of BZ ligands has been criticized on the basis of the complexity of BZ agonist effects (Haefely et al., 1981). A number of investigators (e.g. Sansone, 1979) have reported an increase in locomotor activity using low doses of BZ agonists which may reflect an increase in exploratory activity due to the anxiolytic or behavioural disinhibitory properties of BZ agonists and a corresponding decrease in activity at higher doses, due to a complex of motor effects. While the rat chain-pulling test does not overcome all of the problems associated with tests of motor performance, it successfully detected the activity of BZ ligands at doses that were generally equal to or lower than those reported using other measures of motor performance, including the mouse rotarod test. Of course, discrepancies between the effective dose of compounds in the mouse rotarod and rat chain-pulling tests may represent species differences in drug metabolism rather than primary differences in sensitivity between the two tests. While this remains a possibility, the rat chain-pulling test was shown to have advantages over the
213
particularly in the ability to monitor the effects of compounds on behaviour across time. This feature of the chain-pulling test enabled the time course of drug-induced motor impairments to be measured and was particularly useful when demonstrating the effects of short-acting compounds such as flumazenil in reversing the effects of FG 8205. In conclusion, two disadvantages of the rat chain-pulling test compared to other tests of motor performance are that it requires a comparatively long training period before testing can begin and it necessitates a considerable amount of equipment. Nevertheless, the test appears to be sensitive to a variety of BZ ligands and can detect the onset and offset of drug effects which may make the test particularly suitable for inclusion as part of a programme of assessing a wide range of BZ receptor ligands. From a practical point of view many compounds can be rapidly tested once rats are trained to perform the test and if the same rats are used to test more than one compound, direct comparisons between the effects of different compounds are possible. rotarod test,
References S, Depoortere H, George P, Langer S (1985) Pharmacological profile of the imidazopyridine zolpidem at benzodiazepine receptors and electrocorticogram in rats. Naunyn-Schmiedeberg’s Arch Pharmacol 330: 248-251 Benavides J, Peny B, Dubois A, Perrault G, Morel E, Zivkovic B, Scatton B (1988) In vivo interaction of zolpidem with central benzodiazepine (BZD) binding sites (as labelled by [3H]Ro 151788) in the mouse brain. Preferential affinity of zolpidem for the : 1033-1041 1 (BZD1) subtype. J Pharmacol Exp Ther 245 W Bourin M, Hascoet M, Mansouri B, Colombel MC, Bradwejn J (1992) Comparison of behavioural effects after single and repeated administrations of four benzodiazepines in three mice : 72-77 behavioural models. J Psychiatr Neurosci 17 Braestrup C, Hororé T, Nielsen M, Petersen E N, Jensen L H (1984) Ligands for benzodiazepine receptors with positive and negative 859-862 efficacy. Biochem Pharmacol 33: Burt DA, Kamatchi G L (1991) GABA A receptor subtypes: from pharmacology to molecular biology. FASEB J 5: 2916-2923 Corradi CL, Bennet DA, Diverio J, Wilson DE (1985) The differential sedative/muscle relaxant profiles of several classical and novel anxiomodulators in four behavioural paradigms. Soc Neurosci Abstracts 1(11): 425 Dawson G R, Dickinson A (1990) Performance on ratio and interval schedules with matched reinforcement rates. Q J Exp Psychol 42B(3): 225-239 Dawson G R, Crawford S P, Stanhope K J, Iversen S D, Tricklebank M D (1995) One-trial tolerance to the effects of chlordiazepoxide on the elevated plus maze may be due to locomotor habituation, : 570-572 not repeated drug exposure. Psychopharmacology 113 Doble A, Martin I L (1992) Multiple benzodiazepine receptors: no : 76-81 reason for anxiety. Trends Pharm Sci 13 Dunham N W, Miya T S (1957) A note on a simple apparatus for detecting neurological deficit in rats and mice. J Am Pharm : 208-209 Assoc (Sci Edition) 46 W Partial Haefely (1988) agonists of the benzodiazepine receptor: from animal data to results in patients. In Biggio G, Costa E (eds), Chloride channels and their modulation by neurotransmitters and drugs. Raven Press, New York, pp. 275-292 Haefely W (1991) Comparative pharmacology of benzodiazepine receptor ligands with differing intrinsic efficacy. In Barnard E, Costa E (eds), Transmitter amino acid receptors: structures, transduction and models for drug development, Fidia Research
Arbilla
Foundation Series Volume 6. Thieme Medical Publishers Inc., New York, pp. 91-111 Haefely W, Pieri L, Polc P, Schaffner R (1981) General pharmacology and neuropharmacology of benzodiazepine derivatives. In Hoffmeister F, Stille G (eds), Psychotropic agents part II:
anxiolytic sup. Springer-Verlag, Berlin, pp. 13-283 Haefely W, Martin J R, Schoch P (1990) Novel anxiolytics that act as partial agonists at benzodiazepine receptors. Trends Pharm Sci : 452-456 11
Hunkeler W, Mohler H, Pieri L, Polc P, Bonetti E P, Cumin R, Schaffner R, Haefely W (1981) Selective antagonists of benzodiazepines. Nature 290: 514-516 Klotz U, Antonin K H, Bieck P R (1976) Pharmacokinetics and plasma binding of diazepam in man, dog, rabbit, guinea pig and rat. J Pharmacol Exp Ther 199: 67-73 Macdonald R L, Olsen R W (1994) GABA A receptor channels. Ann Rev Neurosci 17: 569-602 Mandema JA, Gubbens-Stibbe J M, Danhof M (1991) Stability and pharmacokinetics of flumazenil in the rat. Psychopharmacology 103: 384-387 Merlotti L, Roehrs T, Koshorek G, Zorick F, Lamphere J, Roth T (1989) The dose effects of zolpidem on the sleep of healthy normals. J Clin Psychopharmacol : 9-14 9 Miller LG, Galpern WR, Byrnes JJ, Greenblatt DJ (1992) Benzodiazepine receptor binding of benzodiazepine hypnotics: : 413-416 receptor ligand specificity. Pharmacol Biochem Behav 43 Nicholson A N, Pascoe DA (1986) Hypnotic activity of an imidazopyridine (zolpidem). Br J Clin Pharmacol 21: 205-211 Nielsen M, Braestrup C (1980) Ethyl β-carboline-3-carboxylate shows differential benzodiazepine receptor interaction. Nature 606-607 286:
Ongini E, Parravicini L, Bamonte F, Guzzon V, Iorio L C, Barnett A (1982) Pharmacological studies with quazepam, a new : 458-1462 benzodiazepine hypnotic. Arzneim-Forsch/Drug Res 32 Perrault G, Morel E, Sanger D J, Zivkovic B (1990) Differences in pharmacological profiles of a new generation of benzodiazepine : 487-494 187 and non-benzodiazepine hypnotics. Eur J Pharmacol Platt J R (1973) Percentile reinforcement: paradigms for experimental analysis of response shaping. In Bower G (ed.), The psychology of learning and motivation, Vol.7. Academic Press, New York, pp. 271-296 Potokar J, Nutt D J (1994) Anxiolytic potential of bendzodiazepine partial agonists. CNS Drugs 1: 305-315 Pritchett D B, Seeburg PH (1990) γ-Aminobutyric acid A receptor α5-subunit creates novel type II benzodiazepine receptor : 1802-1804 pharmacology. J Neurochem 54 Pritchett DB, Luddens H, Seeburg PH (1989) Type I and type II A benzodiazepine receptors produced in transfected cells. GABA 1389-1392 Science 245:
Sanger D J, Zivkovic B (1987) Investigation of the development of tolerance to the actions of zolpidem and midazolam. :1513-1518 Neuropharmacology 26 Sanger D J, Zivkovic B (1988) Further behavioural evidence for the selective sedative action of zolpidem. Neuropharmacology 27: 1125-1130
Perrault G, Morel E, Joly D, Zivkovic B (1987) The behavioural profile of zolpidem, a novel hypnotic drug of imidazopyridine structure. Physiol Behav 41: 235-240 Sansone M (1979) Effects of repeated administration of chlordiazepoxide on spontaneous locomotor activity in mice. : 109-110 Psychopharmacology 66 Tricklebank M D, Honoré T, Iversen S D, Kemp JA, Knight A R, Marshall G R, Rupniak N M J, Singh L, Tye S, Watjen F, Wong E H F (1990) The pharmacological properties of the imidazobenzodiazepine, FG 8205, a novel partial agonist at the benzodiazepine receptor. Br J Pharmacology 101: 753-761 Whiting P W, Wafford K, McKernan R M (1995) GABA A receptors in the central nervous system. Int Rev Neurobiol in press
Sanger DJ,
10(3) (1996) 206-213
m1996 British Associadon for
Psychopharinacology
Comparison of benzodiazepine (BZ) receptor agonists in two rodent activity tests P. J.
, D. Williamson and G. R. Dawson 1 Bayley, G. D. Bentley, A. Jackson
Merck Sharp & Dohme Research Laboratories, Neuroscience Research Centre, Terlings Park, Eastwick Road, Harlow, Essex CM20 2QR, UK Present address: Laboratory of Experimental Psychology, School of Biological Sciences, University of Sussex, Falmer, Brighton, Sussex 1 BN1 9QG, UK
The effects of four BZ receptor ligands in an operant test were compared with a rotarod test. In the operant test, rats were trained to pull a chain on a schedule that regulates the probability of delivery of food pellets to maintain a steady chain-pulling rate across a 1 h test session. For the rotarod test, mice were trained to remain on a rotarod for 2 min. Diazepam (0.1-3.0 mg/kg, i.p.), FG 8205 (0.1-3.0 mg/kg, i.p.), quazepam (3.0-60.0 mg/kg, i.p.) and zolpidem (0.3-10.0 mg/kg, i.p.) each produced dose-related impairments of performance in both the chainpulling test and the mouse rotarod test. Furthermore, the impairment in performance induced by FG 8205 (10.0 mg/kg, p.o.) was dose-dependently reversed by the BZ receptor antagonist, flumazenil (1.0-10.0 mg/kg, i.p.), indicating that the chain-pulling deficit was mediated via BZ receptor activation. Diazepam, FG 8205 and quazepam all had comparable potencies in both the rotarod assay and the chain-pulling test. However, zolpidem suppressed the chain-pulling rates at a dose 30-fold lower than that required to induce a significant deficit in the rotarod performance. As zolpidem is a preferentially sedative compound, this pattern of results is consistent with the hypothesis that the chain-pulling test is sensitive to sedation induced by BZ receptor
agonists. Key words: diazepam; FG 8205; flumazenil; quazepam; sedation; zolpidem
(1995)]. At least one a-subunit and one y-subunit required to form a fully functional receptor in vivo (Pritchett, Luddens and Seeburg, 1989). The properties of the BZ binding site depend on the combination of a- and ysubunits, but it is particularly important which a-subunit is McKernan
Introduction
are
A full receptor agonist at the benzodiazepine on the GABAA receptor complex, such as
(BZ) binding site diazepam, has a
range of
properties including anxiolysis, anticonvulsant, myorelaxant and sedative effects (Braestrup et al., 1984). Whilst all of these properties are clinically useful, they are rarely required simultaneously. For example, although diazepam is an effective treatment for anxiety, the accompanying sedation (a feeling of tiredness or drowsiness), ataxia (inability to coordinate voluntary movement) and myorelaxation (a decrease in muscle tone) all reduce the patient’s ability to perform their normal daily activities. Thus, in recent years much effort has been devoted to the development of BZ drugs that have anxiolytic effects, but are devoid of sedative, ataxic and myorelaxation effects. Two approaches have been taken to the development of anxioselective drugs. The first, the partial agonist approach, is to discover compounds that have sufficient efficacy at the BZ receptor to induce anxiolysis, but not enough to induce sedation (Haefely, Martin and Schoch, 1990). The second is to discover drugs that are selective for GABAA receptor subunits. Molecular cloning techniques have identified 14 GABAA receptor subunits in the mammalian brain [CX1-- 0.12]. FG 8205 dose dependently reduced the chain-pulling rate and an ANOVA of the overall mean chain-pulling rates showed that while a dose of 0.1mg/kg of FG 8205 did not significantly reduce the chainpulling rates at any time point, a dose of 0.3 mg/kg FG 8205 and above significantly reduced the chain-pulling rates compared to vehicle treated controls (Fig. 1 d, collapsed data). Post hoc tests showed that the mean chain-pulling rates of the groups given 0.3 and 1.0 mg/kg FG 8205 first =
=
differed from the vehicle-treated control group 20 min postand were significantly (p < 0.05) different from those of the vehicle-treated rats for the remainder of the session. The mean chain-pulling rates of the rats given 3.0 mg/kg FG 8205 differed significantly from vehicle-treated controls for the duration of the session (p < 0.05). The results for zolpidem across the chain-pulling test session are shown in Fig. 1 (e) and the mean chain-pulling rates for the collapsed data (expressed as a percentage of the baseline chainpulling rate) are shown in Fig. 1 (f). An ANOVA of the chainpulling rates obtained at 10 min intervals revealed a significant main effect of treatment [F(4,44) = 10.10, p < 0.01], but not of time [F(5,220) =1.77, p > 0.12]. However, there was a significant interaction between treatment and time [F(20,220) = 2.15, p < 0.01]. Zolpidem dose dependently reduced the mean chain-pulling rates. Post hoc tests showed that the chain-
injection
209
210
1.0 mg/kg differed from those of the vehicle-treated control group for only the first 20 min of the session (p < 0.05) and that rats given 3.0 mg/kg zolpidem differed from vehicle-treated controls for only the first 30 min of the session. Only the chain-pulling rates of the rats given 10.0 mg/kg zolpidem were significantly lower than those of the vehicle-treated controls for the duration of the session. The effect of quazepam on the mean chain-pulling rates across the test session is shown in Fig. 1 (g) and the mean chainpulling rates for the collapsed data (expressed as a percentage of the baseline chain-pulling rate) for the whole session are shown in Fig. 1 (h). An ANOVA of the mean chain-pulling rates obtained at 10 min intervals revealed a significant main and of time effect of treatment [F(4,45) = 8.90, p < 0.01] [F(5,225) = 9.38, p < 0.01]and a treatment by time interaction [F(20,225) = 3.54, p < 0.01]. An ANOVA of the overall mean percentage chain-pulling rates showed that the lowest dose of quazepam to reduce the chain-pulling rates was 10.0 mg/kg. Post hoc tests showed that the mean chain-pulling rates of the group given 60 mg/kg of quazepam differed from the vehicletreated rats for the duration of the session. By contrast, the mean chain-pulling rate of the group given 10.0 mg/kg quazepam differed significantly from that of the vehicletreated control group only 20 min after dosing.
pulling rates of rats given
Effect of flumazenil and FG 8205 on chain-pulling rates Figure 2(a) shows the effect of flumazenil and FG 8205 on the chain-pulling rates at 5 min intervals across the 30 min session. Figure 2(b) shows the mean chain-pulling rates (expressed as a percentage of the baseline chain-pulling rate) for the session as a whole. An ANOVA with factors of treatment and time revealed significant main effects of treatment [F(5,41) 25.72, =
p < 0.01]
] and of time
[F(5,205) = 20.79, p < 0.01] ] and
a
by time interaction [~25,205) = 3.58, p < 0.01]. Post hoc tests showed that there were no significant differences between the group given 10.0 mg/kg flumazenil/vehicle and the treatment
vehicle/vehicle control group (Fig.2b). However, the mean chain-pulling rate of the group given 10.0 mg/kg FG 8205/ vehicle was significantly lower than vehicle/vehicle controls throughout the session (Fig.2b). This reduction in chainpulling rates induced by 10.0 mg/kg FG 8205 was dose dependently reversed by flumazenil. Full reversal of the effects of 10.0 mg/kg FG 8205 was seen at 10.0 mg/kg Flumazenil since no significant differences in the chain-pulling rates were found between groups given 10.0 mg/kg FG 8205/10.0 mg/kg flumazenil or vehicle/vehicle controls at any time point across the session (Fig. 2b). A partial reversal of the effects of 10.0 mg/kg FG 8205 was seen at doses of flumazenil below 10.0 mg/kg. Hence, the chain-pulling rates of rats given 10.0 mg/kg FG 8205/3.0 mg/kg flumazenil first differed from the vehicle/ vehicle control group 1 S min from the start of the session and were significantly different from the control rats for the remainder of the session. The chain-pulling rates of rats given 10.0 mg/kg FG 8205/ 1.0 mg/kg flumazenil differed significantly from controls at all time points except 10 min from the start of the session.
Rotarod Table 1 shows the mean latency to fall from the rotarod for each dose for the four drug groups. FSD, defined significantly as the lowest dose at which the latency to fall differed significantly from vehicle treated controls (p < 0.05), was found for each drug. Both diazepam and FG 8205 had an FSD of 3.0 mg/kg. Zolpidem and quazepam were found to have FSDs of 30.0 mg and 10.0 mg/kg respectively.
Figure 2 The reversal of the effects of FG 8205 by flumazenil in the rat chain-pulling task. The chain-pulling rates are shown as a percentage of the mean chain-pulling rate of the previous baseline day and the overall standard error of mean (SEM) chain-pulling rate is shown. The mean chain-pulling rates for each 5 min period are shown in the left-hand panel for each treatment group. The right-hand panel shows the overall session mean chain-pulling rate. An asterisk above a bar indicates a significant difference from vehicle-treated control group (Student-Newman-Keuls test, p < 0.05)
211 Table 1 rotarod
zolpidem
latency ( f SE) in seconds for mice to fall from a administration of either diazepam, FG 8205, quazepam
The
mean
following or
An asterisk indicates a significant difference from the vehicle-treated control group (Dunnetts post hoc test, p < 0.05). NT, not tested.
Discussion The aim of the present study was to measure the effects of several compounds known to act at the BZ receptor in a rat operant chain-pulling test and to compare the results obtained to the mouse rotarod task, a standard measure of motor disruption (Haefeley et al., 1981). The results for the full BZ receptor agonist diazepam were consistent with previous findings and diazepam reduced the latency to fall from a rotarod. In addition, diazepam was found to be more potent in reducing operant responding in the rat chain-pulling test. The partial agonist FG 8205 had similar effects to diazepam and was found to be more potent in the rat chain-pulling test than the mouse rotarod. Deficits in the rat chain-pulling performance induced by FG 8205 were reversed by the BZ receptor antagonist, flumazenil. Finally, two full agonists, zolpidem and quazepam, with selectivity for BZI receptors were tested. Both zolpidem and quazepam reduced the rat chain pulling at lower doses than those which affected the rotarod performance. Taken together, these results support the validity of the chainpulling test as a measure of the suppressant effects of BZ receptor agonists and partial agonists. Diazepam was found to induce performance deficits in the mouse rotarod test at a similar dose to those previously reported to be effective in reducing mouse spontaneous locomotor activity and rotarod performance (e.g. 1.0 mg/kg; Bourin et al., 1992). Diazepam induced a deficit in the rat chain-pulling test at 1.0 mg/kg for the first half of the 60 min test session while at the highest dose, diazepam (3.0 mg/kg) reduced the chainpulling rates across the entire session which is consistent with the short half-life of 1.1 h for diazepam in the rat (Klotz, Antonin and Bieck, 1976). Diazepam is generally reported to reduce rat locomotor activity and reduce the time to fall from a rotarod at approximately 1O.Omg/kg (Haefely, 1991). The BZ receptor partial agonist FG 8205 reduced the latency of mice to fall from a rotarod at 3.0 mg/kg, a lower dose than that previously reported (30.0 mg/kg) to affect mouse rotarod performance (Tricklebank et al., 1990). A performance deficit occurred in the rat chain-pulling test at doses of FG 8205 which are not effective in other tests of motor performance. Thus, 0.3 mg/kg FG 8205 induced a deficit in the chain-pulling performance, whereas a higher dose
of 3.0 mg/kg of FG 8205 was required before deficits in the rat rotarod and locomotor activity tests were seen (unpublished results). The deficit in the rat chain-pulling test, seen at higher doses of FG 8205, occurred soon after the start of the session and lasted for the duration of the session, an observation that is consistent with the plasma concentration curve for FG 8205 which reaches a peak 10 min after oral administration and has a brain half-life of 3.6 h in the rat (unpublished results). Taken together with the results for diazepam, these findings show that the rat chain-pulling test compares favourably with established tests of motor performance in its ability to detect the adverse motor effects of BZ receptor agonists and partial agonists. In a separate experiment, we also demonstrated that the FG 8205 deficit could be reversed by the BZ receptor antagonist, flumazenil. The highest dose of flumazenil ( 10.0 mg/kg) antagonized the effects of 10.0 mg/kg FG 8205 throughout the session; however lower doses of flumazenil (1.0-3.0 mg/kg) only blocked the effects of FG 8205 in the early part of the test session. The lack of effect of flumazenil in the latter part of the session probably reflected the relatively short half-life of flumazenil of 8.3 min in the rat (Mandema, Gubbens-Stibbe and Danhof, 1991) compared to that of FG 8205. The BZ agonist zolpidem has known selectivity for the BZ1 receptor subtype (Benavides et al., 1988), a property which is now thought to confer selectivity for a,-containing BZ receptors (Pritchett, Luddens and Seeburg, 1989). Zolpidem was found to reduce the chain-pulling rates at a dose of 1.0 mg/ kg. The results obtained with zolpidem were clearly different to the non-selective drugs as the chain-pulling rates were reduced at a dose 30-fold lower than that found to induce a deficit in the mouse rotarod. In comparison, FG 8205 reduced the chain-pulling rates at a dose 10-fold lower than that found to induce a rotarod deficit and diazepam was found to be equipotent in both tests. Low doses of zolpidem only affected chain pulling for the first half of the session which is consistent with the short half-life of approximately 30 min for zolpidem in the rat (Arbilla et al., 1985). The rotarod results were consistent with previous experiments which report that unlike other BZ receptor agonists such as diazepam, zolpidem does not produce a rotarod deficit until relatively high doses are reached when given 30 min before testing in both mice (EDso =18.0 mg/kg; Perrault et al., 1990) and rats (ED50 6.0 mg/kg; Sanger et al., 1987). However, lower doses of zolpidem have been reported to be behaviourally active and zolpidem has been found to reduce mouse locomotor activity (EDso =1.2 mg/kg; Perrault et al., 1990) and rat open field locomotor activity (1.0 mg/kg; Sanger and Zivkovic, 1988) when given 30 min before testing and Sanger and Zivkovic (1987) have also reported that 0.3 mg/kg of zolpidem reduced rat lever pressing in an operant test. Considering the short half-life of zolpidem, differences in the effective dose of zolpidem between studies could simply reflect the different pretreatment times used. However, the studies cited above all used a pre-treatment time of 30 min for zolpidem and the same pretreatment time was used in the present mouse rotarod experiment. Rather, the discrepancy between the high dose of zolpidem needed to produce a rotarod deficit and the low effective doses in other tests of motor performance suggests that zolpidem may induce specific depressant effects. Considering the potent hypnotic effect of zolpidem in humans (Nicholson =
212
and Pascoe, 1986), it is reasonable to assume that the deficits in locomotor activity, lever pressing and chain pulling seen at relatively low doses of zolpidem are due to sedation. The other BZI selective agonist, quazepam, was equipotent in its effects on the mouse rotarod and rat chain-pulling tests and was active at 10.0 mg/kg in both tests. The rotarod results were consistent with previous findings which reported that in mice 4.9 mg/kg of quazepam reduced the latency to fall from a rotarod (Perrault et al., 1990) and 2.5 mg/kg of quazepam reduced spontaneous locomotor activity (Ongini et al., 1982). In the rat chain-pulling test, 10.0 mg/kg of quazepam reduced performance, although its effects and that of 30.0 mg/kg disappeared rapidly, indicating that the compound has a short duration of action. Quazepam has previously been reported to impair locomotor activity and rotarod performance in the rat at a dose 10-fold higher than that found to disrupt chain pulling
(>,I 00 mg/kg; Corradi et al., 1985). Since chain pulling was monitored across the entire test session, behavioural disruption could be observed at low doses. When data were collapsed across the test session however, quazepam was found to induce significant deficits only at the highest dose tested (60.0 mg/kg), a dose closer to those reported to impair locomotor activity and rotarod performance in the rat. Since quazepam induced maximal behavioural disruption soon after dosing, it seems likely that the compound would have reduced mouse rotarod performance at shorter pre-treatment times than used in the present study. It is also interesting to note that although zolpidem was 30-fold more potent in the chain-pulling test than in the rotarod test, no such separation in activity was found for quazepam. These results are similar to those reported by Perrault et al. (1990) who found that while zolpidem and quazepam both induced sedation at similar doses, as measured by exploratory activity in mice, zolpidem was 4-fold less potent than quazepam at inducing ataxic effects in the mouse rotarod. Since zolpidem and quazepam have a similar affinity for CX1containing GABAA receptors, it is not clear why the two compounds should show different behavioural profiles. Perrault et al. (1990) suggested that quazepam has a lower intrinsic activity than zolpidem and thus may be a partial agonist. Relative to full BZ receptor agonists, partial agonists at the BZ receptor produce less muscle relaxation, sedation and ataxia while retaining anxiolytic and anticonvulsant activity (Haefely, Martin and Schoch, 1990; Haefely, 1991). If quazepam were a partial agonist, this would account for the finding that quazepam was less potent than zolpidem in disrupting chain pulling. However, quazepam would also be expected to be less potent than zolpidem in the rotarod test, which was not the case. While the reason for the differences in the behavioural effects of zolpidem and quazepam remain to be determined, the results indicate that the chain pulling and rotarod tests may be sensitive to different properties of certain compounds. Clinical results from a number of trials involving BZ receptor partial agonists are now accumulating and confirm that partial agonists are indeed anxiolytic, but are also sedative at therapeutic doses (Potokar and Nutt, 1994). While the sedative effects of the BZ receptor partial agonists may be species specific (Haefely, 1988), from the pre-clinical perspective it is imperative that rodent tests have sufficient sensitivity to detect low levels of sedation. Tricklebank et al. (1990) reported that 0.5 mg/kg of the BZ receptor partial
agonists FG 8205 was needed to induce anxiolytic-like effects in rats. In the present study, a dose of 0.3 mg/kg of FG 8205 was found to reduce chain pulling which implies that the separation between the anxiolytic and motor disrupting effects of partial agonists, such as FG 8205, may be less than initially hoped. Attention has also been given to the hypothesis that different BZ receptor subtypes which occur in different regions of the brain have different physiological functions, with the possibility that subtype-selective compounds may show only part of the behavioural effects seen with non-selective BZ receptor agonists (Nielsen and Braestrup, 1980). For instance, it is known that the properties of the BZ binding site depend largely on which a-subunit is present in the GABAA receptor (Pritchett, Luddens and Seeburg, 1989) and both quazepam and zolpidem, which are selective for a,-containing BZ receptors, have been suggested to be sedative at doses which do not induce ataxia (Miller et al., 1992). Behavioural findings that both quazepam and zolpidem are more potent in reducing spontaneous locomotor activity than in reducing rotarod performance have been used to lend support to the idea that a,-selective BZ agonists are ’hypnoselective’. Indeed, the present results, which show that quazepam and zolpidem reduced the chain-pulling rates are not inconsistent with a predominantly sedative behavioural profile of ai-selective BZ receptor agonists. Unfortunately, there are currently few BZ receptor ligands with sufficient selectivity for BZ subtypes to allow meaningful correlations to be made between behaviour and BZ receptor subtypes [for a review see Doble and Martin (1992)]. However, the finding that the chain-pulling test may detect low levels of motor disruption may be useful when assessing subtype-selective BZ receptor ligands when such compounds become available. The rat chain-pulling test is likely to measure a number of factors, only one of which may be sedation and in this respect it may be similar to other rodent measures of motor performance. The rotarod test is perhaps the most widely used such measure. Despite the popularity of the rotarod test, little consensus exists as to what the test measures. Thus, it has been described as an index of sedation, muscle relaxation, ataxia, fatigue or even ’neurotoxicity’ and ’neuronal deficit’ (Haefely et al., 1981). The spontaneous locomotor activity test, which is also commonly used to assess the motor effects of BZ ligands has been criticized on the basis of the complexity of BZ agonist effects (Haefely et al., 1981). A number of investigators (e.g. Sansone, 1979) have reported an increase in locomotor activity using low doses of BZ agonists which may reflect an increase in exploratory activity due to the anxiolytic or behavioural disinhibitory properties of BZ agonists and a corresponding decrease in activity at higher doses, due to a complex of motor effects. While the rat chain-pulling test does not overcome all of the problems associated with tests of motor performance, it successfully detected the activity of BZ ligands at doses that were generally equal to or lower than those reported using other measures of motor performance, including the mouse rotarod test. Of course, discrepancies between the effective dose of compounds in the mouse rotarod and rat chain-pulling tests may represent species differences in drug metabolism rather than primary differences in sensitivity between the two tests. While this remains a possibility, the rat chain-pulling test was shown to have advantages over the
213
particularly in the ability to monitor the effects of compounds on behaviour across time. This feature of the chain-pulling test enabled the time course of drug-induced motor impairments to be measured and was particularly useful when demonstrating the effects of short-acting compounds such as flumazenil in reversing the effects of FG 8205. In conclusion, two disadvantages of the rat chain-pulling test compared to other tests of motor performance are that it requires a comparatively long training period before testing can begin and it necessitates a considerable amount of equipment. Nevertheless, the test appears to be sensitive to a variety of BZ ligands and can detect the onset and offset of drug effects which may make the test particularly suitable for inclusion as part of a programme of assessing a wide range of BZ receptor ligands. From a practical point of view many compounds can be rapidly tested once rats are trained to perform the test and if the same rats are used to test more than one compound, direct comparisons between the effects of different compounds are possible. rotarod test,
References S, Depoortere H, George P, Langer S (1985) Pharmacological profile of the imidazopyridine zolpidem at benzodiazepine receptors and electrocorticogram in rats. Naunyn-Schmiedeberg’s Arch Pharmacol 330: 248-251 Benavides J, Peny B, Dubois A, Perrault G, Morel E, Zivkovic B, Scatton B (1988) In vivo interaction of zolpidem with central benzodiazepine (BZD) binding sites (as labelled by [3H]Ro 151788) in the mouse brain. Preferential affinity of zolpidem for the : 1033-1041 1 (BZD1) subtype. J Pharmacol Exp Ther 245 W Bourin M, Hascoet M, Mansouri B, Colombel MC, Bradwejn J (1992) Comparison of behavioural effects after single and repeated administrations of four benzodiazepines in three mice : 72-77 behavioural models. J Psychiatr Neurosci 17 Braestrup C, Hororé T, Nielsen M, Petersen E N, Jensen L H (1984) Ligands for benzodiazepine receptors with positive and negative 859-862 efficacy. Biochem Pharmacol 33: Burt DA, Kamatchi G L (1991) GABA A receptor subtypes: from pharmacology to molecular biology. FASEB J 5: 2916-2923 Corradi CL, Bennet DA, Diverio J, Wilson DE (1985) The differential sedative/muscle relaxant profiles of several classical and novel anxiomodulators in four behavioural paradigms. Soc Neurosci Abstracts 1(11): 425 Dawson G R, Dickinson A (1990) Performance on ratio and interval schedules with matched reinforcement rates. Q J Exp Psychol 42B(3): 225-239 Dawson G R, Crawford S P, Stanhope K J, Iversen S D, Tricklebank M D (1995) One-trial tolerance to the effects of chlordiazepoxide on the elevated plus maze may be due to locomotor habituation, : 570-572 not repeated drug exposure. Psychopharmacology 113 Doble A, Martin I L (1992) Multiple benzodiazepine receptors: no : 76-81 reason for anxiety. Trends Pharm Sci 13 Dunham N W, Miya T S (1957) A note on a simple apparatus for detecting neurological deficit in rats and mice. J Am Pharm : 208-209 Assoc (Sci Edition) 46 W Partial Haefely (1988) agonists of the benzodiazepine receptor: from animal data to results in patients. In Biggio G, Costa E (eds), Chloride channels and their modulation by neurotransmitters and drugs. Raven Press, New York, pp. 275-292 Haefely W (1991) Comparative pharmacology of benzodiazepine receptor ligands with differing intrinsic efficacy. In Barnard E, Costa E (eds), Transmitter amino acid receptors: structures, transduction and models for drug development, Fidia Research
Arbilla
Foundation Series Volume 6. Thieme Medical Publishers Inc., New York, pp. 91-111 Haefely W, Pieri L, Polc P, Schaffner R (1981) General pharmacology and neuropharmacology of benzodiazepine derivatives. In Hoffmeister F, Stille G (eds), Psychotropic agents part II:
anxiolytic sup. Springer-Verlag, Berlin, pp. 13-283 Haefely W, Martin J R, Schoch P (1990) Novel anxiolytics that act as partial agonists at benzodiazepine receptors. Trends Pharm Sci : 452-456 11
Hunkeler W, Mohler H, Pieri L, Polc P, Bonetti E P, Cumin R, Schaffner R, Haefely W (1981) Selective antagonists of benzodiazepines. Nature 290: 514-516 Klotz U, Antonin K H, Bieck P R (1976) Pharmacokinetics and plasma binding of diazepam in man, dog, rabbit, guinea pig and rat. J Pharmacol Exp Ther 199: 67-73 Macdonald R L, Olsen R W (1994) GABA A receptor channels. Ann Rev Neurosci 17: 569-602 Mandema JA, Gubbens-Stibbe J M, Danhof M (1991) Stability and pharmacokinetics of flumazenil in the rat. Psychopharmacology 103: 384-387 Merlotti L, Roehrs T, Koshorek G, Zorick F, Lamphere J, Roth T (1989) The dose effects of zolpidem on the sleep of healthy normals. J Clin Psychopharmacol : 9-14 9 Miller LG, Galpern WR, Byrnes JJ, Greenblatt DJ (1992) Benzodiazepine receptor binding of benzodiazepine hypnotics: : 413-416 receptor ligand specificity. Pharmacol Biochem Behav 43 Nicholson A N, Pascoe DA (1986) Hypnotic activity of an imidazopyridine (zolpidem). Br J Clin Pharmacol 21: 205-211 Nielsen M, Braestrup C (1980) Ethyl β-carboline-3-carboxylate shows differential benzodiazepine receptor interaction. Nature 606-607 286:
Ongini E, Parravicini L, Bamonte F, Guzzon V, Iorio L C, Barnett A (1982) Pharmacological studies with quazepam, a new : 458-1462 benzodiazepine hypnotic. Arzneim-Forsch/Drug Res 32 Perrault G, Morel E, Sanger D J, Zivkovic B (1990) Differences in pharmacological profiles of a new generation of benzodiazepine : 487-494 187 and non-benzodiazepine hypnotics. Eur J Pharmacol Platt J R (1973) Percentile reinforcement: paradigms for experimental analysis of response shaping. In Bower G (ed.), The psychology of learning and motivation, Vol.7. Academic Press, New York, pp. 271-296 Potokar J, Nutt D J (1994) Anxiolytic potential of bendzodiazepine partial agonists. CNS Drugs 1: 305-315 Pritchett D B, Seeburg PH (1990) γ-Aminobutyric acid A receptor α5-subunit creates novel type II benzodiazepine receptor : 1802-1804 pharmacology. J Neurochem 54 Pritchett DB, Luddens H, Seeburg PH (1989) Type I and type II A benzodiazepine receptors produced in transfected cells. GABA 1389-1392 Science 245:
Sanger D J, Zivkovic B (1987) Investigation of the development of tolerance to the actions of zolpidem and midazolam. :1513-1518 Neuropharmacology 26 Sanger D J, Zivkovic B (1988) Further behavioural evidence for the selective sedative action of zolpidem. Neuropharmacology 27: 1125-1130
Perrault G, Morel E, Joly D, Zivkovic B (1987) The behavioural profile of zolpidem, a novel hypnotic drug of imidazopyridine structure. Physiol Behav 41: 235-240 Sansone M (1979) Effects of repeated administration of chlordiazepoxide on spontaneous locomotor activity in mice. : 109-110 Psychopharmacology 66 Tricklebank M D, Honoré T, Iversen S D, Kemp JA, Knight A R, Marshall G R, Rupniak N M J, Singh L, Tye S, Watjen F, Wong E H F (1990) The pharmacological properties of the imidazobenzodiazepine, FG 8205, a novel partial agonist at the benzodiazepine receptor. Br J Pharmacology 101: 753-761 Whiting P W, Wafford K, McKernan R M (1995) GABA A receptors in the central nervous system. Int Rev Neurobiol in press
Sanger DJ,
