The purpose of our investigation was to characterize the relationships between the pharmacodynamics of synthetic opioids in vivo and the interaction at the mu-opioid receptor. The pharmacokinetics and pharmacodynamics were determined in vivo after a single i.v. infusion of 3.14 mg/kg alfentanil (A), 0.15 mg/kg fentanyl (F) or 0.030 mg/kg sufentanil (S) in rats. Amplitudes in the 0.5 to 4.5 Hz frequency band of the electroencephalogram (EEG) was used as pharmacodynamic endpoint. The EEG effect intensity was related to the (free) concentration in blood (A) or in a hypothetical effect compartment (F, S) on basis of the sigmoidal Emax pharmacodynamic model. The interaction at the mu-opioid receptor was determined in vitro on basis of the displacement of [3H]-naloxone binding in washed rat brain membranes. The value of the sodium shift was used as a measure of in vitro intrinsic efficacy. For the EEG effect the in vivo potencies based on free drug concentrations (EC50,u) were 4.62 +/- 0.66 ng/ml (A), 0.69 +/- 0.05 ng/ml (F) and 0.29 +/- 0.06 ng/ml (S). In the receptor binding studies the affinities at the mu-opioid receptor (Kl) were 47.4 +/- 6.6 nM (A), 8.6 +/- 4.1 nM (F) and 2.8 +/- 0.2 nM (S). For each opioid the ratio between EC50,u and Kl was the same with a value of 0.23-0.25, indicating the existence of receptor reserve for the EEG effect. The intrinsic activity (Emax) of the three opioids in vivo was similar with values of 111 +/- 10 microV (A), 89 +/- 11 microV (F) and 104 +/- 4 microV (S). However, the values of the sodium shift varied between 2.8 (S) and 19.1 (A). Further analysis of the in vivo pharmacodynamic data on basis of an operational model of agonism provided evidence for a large receptor reserve, which explains why compounds with different values of the sodium shift all behave as full agonists in vivo.