Pharmacological inhibition of GABA uptake transporters provides a mechanism for increasing GABAergic transmission, which may be useful in the treatment of various neurological disorders. The purpose of our investigations was to develop an integrated pharmacokinetic-pharmacodynamic (PK/PD) model for the characterization of the pharmacological effect of tiagabine, R-N-(4,4-di-(3-methylthien-2-yl)but-3-enyl)nipecotic acid, in individual rats in vivo. The tiagabine-induced increase in the amplitude of the EEG 11.5-30 Hz frequency band (beta), was used as pharmacodynamic endpoint. Chronically instrumented male Wistar rats were randomly allocated to four groups which received an infusion of 3, 10, or 30 mg kg-1 of tiagabine or vehicle over 10 min. The EEG was continuously recorded in conjunction with frequent arterial blood sampling. The pharmacokinetics of tiagabine could be described by a biexponential equation. The pharmacokinetics of tiagabine were not dose dependent, and the pooled values for clearance, volume of distribution at steady state and terminal half-life were (mean +/- SE, n 23) 96 +/- 9 ml min-1 kg-1, 1.5 +/- 0.1 L kg-1 and 20 +/- 0.2 min. A time delay was observed between the occurrence of maximum plasma drug concentrations and maximal response. A physiological PK/PD model has been used to account for this time delay, in which a biophase was postulated to account for tiagabine available to the GABA uptake carriers in the synaptic cleft and the increase in EEG effect was considered an indirect response due to inhibition of GABA uptake carriers. The population values for the pharmacodynamic parameters characterizing the delay in pharmacological response relative to plasma concentrations were keo = 0.030 min-1 and kout = 81 min-1, respectively. Because of the large difference in these values the PK/PD model was simplified to the effect compartment model. Population estimates (mean +/- SE) were E0 = 155 +/- 6 microV, Emax = 100 +/- 5 microV, EC50 = 287 +/- 7 ng ml-1, Hill factor = 1.8 +/- 0.2 and keo = 0.030 +/- 0.002 min-1. The results of this analysis show that for tiagabine the combined "effect compartment-indirect response" model can be simplified to the classical "effect compartment" model.