Urinary metabolites as biomarkers of acrylamide exposure in mice following dietary crisp bread administration or subcutaneous injection.

Heat-treated carbohydrate-rich foods may contain high levels of acrylamide (AA). Crisp bread is a significant dietary AA source in the Nordic countries. We studied whether urinary metabolites of AA could be candidate biomarkers of AA intake and internal dose in mice following dietary crisp bread administration or sc injection. The crisp bread was experimentally baked to contain three different concentrations of AA: 0.19, 1.02, and 2.65 mg/kg, giving dietary exposures to AA of 0.024 +/- 0.002, 0.14 +/- 0.02, and 0.29 +/- 0.04 mg/kg bodyweight (bw)/day (mean +/- SD), respectively. A linear relationship was found between dietary AA exposure and urinary AA metabolites. On average, 55% of the ingested dose was recovered as urinary AA metabolites, and the molar proportions between the urinary metabolites showed similar proportions for the different doses. Urine AA metabolites were measured after sc injection of AA at doses of 0.05, 0.5, 5, and 50 mg/kg bw, and the urinary recovery for the three lowest doses was 54%. With the highest dose, 80% was recovered in urine, and the changed pattern of urinary metabolites indicated saturation of the metabolic conversion of AA to glycidamide. These results indicate that urinary metabolites of AA are good biomarkers of AA intake and internal dose up to 5 mg/kg bw/day. After sc injection of [(14)C]AA, 92% of the radioactivity was found in the urine and 2% in feces, liver, blood, and intestinal content (6% was not detected), demonstrating that sc AA was highly systemically available, that the major part AA metabolites was excreted, and that a significant portion of urinary AA metabolites (most likely glyceramide) was not accounted for by the present analytical method. Since the urinary recovery of AA after crisp bread feeding and sc injection was practically identical, an indicative "bioavailability" of AA from crisp bread was suggested to be approximately complete.