Psychopharmacology (1986) 89 : 113-117 Psychopharmacology © Springer-Verlag 1986 Chronic treatment with Ro 15-1788 distinguishes between its benzodiazepine antagonist, agonist and inverse agonist properties Sandra E. File 1, J. Dingemanse 2, H.L. Friedman 3, and D.J. Greenblatt 3 1 MRC Neuropharmacology Research Group, Department of Pharmacology, The School of Pharmacy, University of London, Brunswick Square, London WC1N lAX, United Kingdom 2 Center for Bio-Pharmaceutical Sciences Sylvius Laboratories, Leiden, The Netherlands. 3 Division of Clinical Pharmacology, Tufts New England Medical Center, Boston, USA Abstract. Ro 15-1788 (fiumazepil) is an imidazodiazepine that is able to antagonise most of the behavioural actions of the benzodiazepines, as well as having some intrinsic effects. Acute administration of Ro 15-1788 (10 mg/kg) de- creases social interaction between male rats and elevates exploratory head-dipping. After 5 days of pretreatment there was tolerance to the former effect, although Ro 15-1788 retained its ability to antagonise the effects on so- cial interaction of the p-carboline, FG 7142. Ro 15-1788 also retained its ability to elevate head-dipping: addition- ally, the chronically-treated rats had elevated motor activity and rearing scores. The acute effects of lorazepam in the holeboard were unchanged by chronic pretreatment with Ro-15-1788. The plasma and brain concentrations after acute administration of lorazepam were unchanged follow- ing chronic administration of Ro 15-1788. After chronic treatment the brain concentrations of Ro 15-1788 were un- changed. It is unlikely that pharmacokinetic factors could underlie the different behavioural changes following chron- ic treatment. Key words: Benzodiazepine antagonist - Inverse agonist - Agonist - Ro 15-1788 - Tolerance - Exploration - Motor activity - Social interaction The benzodiazepines have a wide spectrum of behavioural effects, most of which can be antagonised by the imidazo- diazepine, Ro 15-1788 (flumazepil) (Hunkeler et al. 1981). Ro 15-1788 is also able to antagonise the effects of/%carbo- lines that are inverse agonists at the benzodiazepine binding sites (Nutt et al. 1982). In addition, Ro 15-1788 has intrinsic effects (for review see File and Pellow 1985a). It is anxio- genie in the social interaction test of anxiety (File et al. 1982a), in the Vogel punished drinking test (Brown et al. 1984; Corda et al. 1982; File and Pellow 1985b) and in food and water consumption tests in unfamiliar environ- ments (Hoffman and Britton 1983; File and Pellow 1985 b). These intrinsic effects are in the direction of those produced by inverse agonists (e.g., the fl-carboline, FG 7142). Ro 15-1788 also has the intrinsic action of elevating exploratory head-dipping (File et al. 1982b), which we have interpreted as being an agonist action. In the mouse, low doses of benzodiazepines elevate head-dipping (File and Pellow Offprint requests to. S.E. File 1985b; Sansone 1979). We suggest that this elevation is difficult to demonstrate in the rat because the sedative ef- fects of acutely administered benzodiazepines are so strong that the elevation in head-dipping is masked. With chronic treatment, tolerance develops to the sedative effects and locomotor activity is no longer reduced; at this point, a clear elevation of head-dipping can be seen in the rat (File 1984; for further discussion, see File and Pellow 1986). With chronic administration, tolerance develops at dif- ferent rates to the different behavioural effects of the benzo- diazepines (see File 1984, 1985). Tolerance to the decreases in locomotor activity and exploratory head-dipping occurs in 3-5 days (File 1981); whereas tolerance does not develop to the enhanced head-dipping, and there is no cross-toler- ance between the sedative and stimulant effects (File and Pellow 1985c). Tolerance to the anxiolytic effects occurs only after 21-22 days (Vellucci and File 1979; Brown et al. 1984). It was the purpose of the present study to examine whether tolerance develops to the intrinsic actions of Ro 15-1788 on social interaction and/or exploratory head-dip- ping. It also examined whether tolerance developed to the antagonist properties of Ro 15-1788, assessed by its ability to antagonise the effects of the /~-carboline FG 7142 in the social interaction test. Finally, it assessed whether the effects of the benzodiazepine, lorazepam, in the holeboard were modified by chronic pretreatment with Ro 15-1788. Unfortunately we were unable to find a dose of lorazepam that elevated head-dipping in the rat, so only a sedative dose could be used. However, in order to assess the possibil- ity that a stimulant effect might emerge after chronic treat- ment with Ro 15-1788 we selected a low dose that was half our sedative dose. The sedative dose selected was the lowest that reliably causes significant reductions in head- dipping. The dose of Ro 15-1788 which reliably produces intrin- sic inverse agonist effects is 10 mg/kg. In Experiment 1 this was therefore the dose chosen for both once and twice daily injections. A group with twice daily injections was included because of the short half-life of Ro 15-1788 (Lister et al. 1984). To equate the total daily dose a third chronic drug group received once daily injections of 20 mg/kg. Doses higher than 20 mg/kg were not selected because at this point Ro 15-1788 has benzodiazepine-like effects (Dantzer and Perio 1982). In Experiment 2, twice daily doses of 10 mg/kg were used since this dose gives clear agonist actions in the holeboard test.