Retinoids often exhibit a complex metabolic pattern and differential transplacental kinetics, which make it difficult to pinpoint the proximate compound responsible for the observed teratogenic effect. We have therefore studied the pharmacokinetics and metabolism of all-trans-retinoic acid (all-trans-RA) in cynomolgus monkeys following application of a nonteratogenic dosing regimen and compared the results with corresponding data from a previous study with a teratogenic dosing regimen with 13-cis-RA [Hummler et al. (1994) Teratology 50:184-193]. All-trans-RA was administered to pregnant cynomolgus monkeys (Macaca fascicularis) by nasogastric intubation at a dose of 5 mg/kg body wt once daily from gestational day (GD) 16 to 26 and twice daily at 8-h intervals from GD 27 to 31. Examination of the fetuses of four dams on GD 100 +/- 2 showed no embryotoxic or teratogenic effects of the applied dosing regimen (Experiment 1). Maternal plasma retinoid pharmacokinetics on GD 16, 26, and 31 as well as embryonic retinoid profiles after the last drug administration on GD 31 were determined in thirteen further dams (Experiment 2). All-trans-RA reached much lower plasma concentrations after the last two treatments on GD 31 than after the first one on GD 16 and the eleventh one on GD 26 (0-24-h area-under-the-concentration-time-curve (AUC) values: 104 +/- 59 ng x h/ml (after the last treatment on GD 31), 189 +/- 110 (GD 16) and 393 +/- 305 ng x h/ml (GD 26). The predominant plasma metabolites of all-trans-RA were its beta-glucuronide and the beta-glucuronide of all-trans-4-oxo-RA. Both of these retinoids accumulated in the plasma during the period of treatment and displayed AUC values 5- to 30-fold higher than those of all-trans-RA. Embryonic concentrations of all-trans-RA were not increased over endogenous levels after the last administration on GD 31 when plasma concentrations were low. To evaluate the placental transport of all-trans-RA in the presence of high plasma concentrations, a further experiment was performed, in which a single dose of all-trans-RA (10 mg/kg body wt) was given to four pregnant monkeys on GD 31, and plasma pharmacokinetics as well as embryonic concentrations of retinoids at 4 h post-treatment were determined (Experiment 3). This dosing schedule yielded high plasma concentrations of all-trans-RA, while embryonic concentrations were about 40% of plasma levels. Based on the plasma AUC values on GDs 16 and 26 obtained in Experiment 2 and the degree of placental transfer, as determined on GD 31 in the presence of high plasma levels in Experiment 3, we estimated embryonic AUC values for the 24-h period following the nonteratogenic doses on GDs 16 and 26 in Experiment 2. These AUC values were similarly high to the embryonic AUC value of all-trans-RA obtained after application of the teratogenic dosing regimen with 13-cis-RA [Hummler et al. (1994) Teratology 50:184-193]. In addition, plasma AUC values of all-trans-RA were 2- to 7-fold higher after all-trans-RA administration (present study) than after dosing with the teratogenic dose of 13-cis-RA. These results strengthen our recent suggestion that the teratogenic effects induced in cynomolgus monkeys by 13-cis-RA treatment cannot solely result from the action of all-trans-RA, but may involve 13-cis-RA and 13-cis-4-oxo-RA, which could act directly or function as transport vehicle.