In a search for suitable biomarkers for human dietary exposure to heterocyclic aromatic amines (HAAs), we have investigated the concentrations of three common fried food mutagens in food and urine after consumption of a fried meat meal. In this connection we developed a method for the determination of HAAs and have investigated the common fried red meat HAAs 2-amino-1-methyl-6-phenylimidazo[4,5-beta]pyridine (PhIP), 2-amino-3,8-dimethylimidazo[4,5-f]quinoxaline (MeIQx) and 2-amino-3,4,8-trimethylimidazo[4,5-f]quinoxaline (DiMeIQx). Eight volunteers participated in the study, each consuming a meal of fried minced beef patties (295 g), boiled potatoes, and a green salad. Urine was collected for two 12-hr periods prior to and following the meal. HAAs were determined in cooked meat and in untreated and acid hydrolysed urine by a series of liquid/liquid extractions, followed by Blue cotton adsorption and finally by a novel derivatized technique for gas chromatography-mass spectrometry (GC-MS). The primary amino groups were derivatized by acylation with heptafluorobutyric acid anhydride, and the resulting amide methylated using diazomethane. Phenolic hydroxyl groups were also methylated by this procedure, making it possible to detect hydroxylated HAAs, possible metabolites or constituents of the fried meat. 4'-Hydroxy-PhIP ¿2-amino-1-methyl-6-(4-hydroxyphenyl)imidazo[4,5-beta]pyridine¿ (4'-OH-PhIP) was indeed found in meat as well as in urine. The contents of PhIP, MeIQx and DiMeIQx in meat were 4.0 +/- 2.6, 3.5 +/- 0.9 and 0.3 +/- 0.1 ng g-1 (mean +/- SD, n = 4), from which the mean amounts ingested were calculated to be 1180, 1030 and 90 ng, respectively. Total amounts of HAAs in the 0-24-hr post-meal untreated urine (and percent of ingested dose) were 6-23 ng PhIP (0.5-2%) and 10-63 ng MeIQx (1-6%). In hydrolysed urine, the levels of HAAs were higher, totalling 24-100 ng PhIP (2-8.5%) and 133-329 ng of MeIQx (13-32%). DiMeIQx was below detection limit in all urine samples. Judged from our study, there were rather large inter-individual variations in the amounts of excreted HAAs, possibly caused by variations in the activities of enzymes taking part in HAA metabolism.