Pharmaceutical Research, Vol. 17, No. 3, 2000 Research Paper has been conducted on the mechanisms involved in tolerance Mechanism-Based Modeling of development, focussing specifically on the functionality of the Functional Adaptation Upon GABA-benzodiazepine receptor complex (2). The extrapolation from in vitro test-systems to the in vivo situation remains a Chronic Treatment with Midazolam complex issue, however. It is therefore not fully understood which mechanisms explain, in a quantitative manner, functional tolerance development in vivo. Adriaan Cleton, 1 Jonas O ¨ dman, 1 Recently, we have successfully developed a new mecha- Piet Hein Van der Graaf, 1 Wim Ghijsen, 4 nism-based model (based on the operational model of agonism Rob Voskuyl, 2,3 and Meindert Danhof 1,5 (3)) to characterise functional adaptation of midazolam (4). As pharmacodynamic endpoint, the EEG was used, which reflects benzodiazepine-induced enhancement of GABA-ergic inhibi- Received September 22, 1999; accepted December 10, 1999 tion in a quantitative manner (5,6). The operational model of Purpose. A mechanism-based model is applied to analyse adaptive agonism is based on receptor theory and contains two important changes in the pharmacodynamics of benzodiazepines upon chronic parameters. The first is the parameter tissue maximum (E m ), that treatment in rats. represents the maximal primary response that can be achieved in Methods. The pharmacodynamics of midazolam was studied in rats the system by receptor activation. The second is an efficacy which received a constant rate infusion of the drug for 14 days, resulting parameter () that describes the efficiency of the transduction in a steady-state concentration of 102  8 ngml -1 . Vehicle treated rats of receptor occupation into the pharmacological response and were used as controls. Concentration-EEG effect data were analysed on that is closely related to the intrinsic efficacy of a given ligand basis of the operational model of agonism. The results were compared (3). Briefly in our previous investigation it was shown by appli- to data obtained in vitro in a brain synaptoneurosomal preparation. cation of a stepwise approach, that the observed reduction in the Results. The relationship between midazolam concentration and EEG maximum EEG effect (, i.e., intrinsic activity) of midazolam in effect was non-linear. In midazolam pre-treated rats the maximum EEG effect was reduced by 51  23 V from the original value of 109  the kindling model of epilepsy can be explained by a change 15 V in vehicle treated group. Analysis of this change on basis of in the parameter tissue maximum (E m ) rather than the efficacy the operational model of agonism showed that it can be explained by parameter (). These findings were confirmed by the results a change in the parameter tissue maximum (E m ) rather than efficacy from in vitro investigations, showing no changes in the density, (). In the in vitro studies no changes in density, affinity or functionality affinity and functionality of the benzodiazepine receptor in a of the benzodiazepine receptor were observed. synaptoneurosomal preparation. Furthermore, it was shown by Conclusions. It is concluded that the observed changes in the concen- simulation that the observed reduction in maximum EEG effect tration-EEG effect relationship of midazolam upon chronic treatment should also occur for other benzodiazepines with a different are unrelated to changes in benzodiazepine receptor function. intrinsic efficacy. KEY WORDS: benzodiazepines; pharmacokinetics; EEG; opera- In the present investigation we have applied the same tional model of agonism; receptor binding; muscimol-induced Cl - approach to analyse changes in the concentration-EEG effect uptake. relationship of midazolam following continuous exposure to the drug for 14 days by chronic infusion. On the basis of this INTRODUCTION model predictions are made regarding changes in the pharmaco- Benzodiazepines are widely prescribed for their anxiolytic, dynamics of other benzodiazepines with different intrinsic effi- muscle relaxant, sedative-hypnotic and anticonvulsant effects. cacy. These predictions show that the absolute change in Their clinical usefulness however is limited by functional toler- intrinsic activity upon chronic treatment will be smaller for ance development upon chronic treatment (1). Much research partial agonists than for full agonists. METHODS 1 Division of Pharmacology, Leiden Amsterdam Center for Drug Research, Leiden University, P.O. Box 9503, 2300 RA Leiden, The Animals Netherlands. Adult male Wistar rats (Harlan, C.P.B., Zeist, The Nether- 2 Stichting Epilepsie Instellingen Nederland, Achterweg 5, 2103 SW Heemstede, The Netherlands. lands), weighing 200–225 g, were used. The animals were 3 Department of Physiology, Leiden University, The Netherlands. housed individually in plastic cages, at 21°C and a 12-h light– 4 Institute of Neurobiology, University of Amsterdam, Kruislaan 320, dark cycle (lights on: 8:00 am to 8:00 pm). Food (Standard 1098 SM Amsterdam, The Netherlands. Laboratory Rat, Mouse and Hamster Diets, RMH-TM, Hope 5 To whom correspondence should be addressed. (e-mail: Farms, Woerden, The Netherlands) and tap water were available m.danhof@lacdr.leidenuniv.nl) ad libitum. The Committee on Animal Experimentation of ABBREVIATIONS: , upper asymptote of concentration-effect rela- Leiden University approved the protocol of this study. tionship; E 0 , no drug response; n H , midpoint slope of concentration- effect relationship; EC 50 , midpoint of concentration-effect relationship; E m , maximal achievable effect in the system; K A , agonist equilibrium/ EEG-Effect Measurement dissociation constant; n, slope index; , efficacy parameter; K E , concen- The pharmacodynamics of midazolam was characterised tration of agonist-receptor complex required to produce half-maximal in two groups of 8 rats on the basis of quantitative EEG monitor- effect; C.I., confidence interval; MVOF, minimum value of objec- tive function. ing as described previously (7). Briefly, one week before the 321 0724-8741/00/0300-0321$18.00/0  2000 Plenum Publishing Corporation