Abstract
Humans are continuously exposed to furan, a hepatotoxic and carcinogenic process-related food contaminant. Considerable uncertainties associated with current exposure estimates based on the content of furan in food underscore the need to explore biomarker monitoring as an alternative or complementary approach to furan exposure assessment. Previous work in rats demonstrated a linear correlation between furan dose and several urinary furan metabolites, whereby N -[4-carboxy-4-(3-mercapto-1 H -pyrrol-1-yl)-1-oxobutyl]-L-cysteinylglycine (GSH-BDA), the glutathione (GSH) conjugate of the reactive furan metabolite cis -2-butene-1,4-dial (BDA), was identified as a specific biomarker of exogenous furan exposure. To further test the validity of GSH-BDA as a biomarker of furan exposure, the present study was designed to monitor urinary excretion of furan metabolites in human volunteers after consumption of diets with low vs. high furan content using stable isotope dilution LC–MS/MS, and to investigate if analysis of GSH-BDA in human urine is suitable for translating biomarker levels into probable daily intakes. Ten healthy volunteers ( n = 5/sex) consumed a low-furan diet for three days (day 1–3), followed by consumption of foods with high furan content for three days (day 4–6) and returned to a low-furan diet for a further three days (7–9). Urinary GSH-BDA excretion significantly increased during periods of a high-furan diet, and rapidly declined upon returning to a low-furan diet. Probable daily intakes estimated from GSH-BDA excretion in non-smoking subjects and excretion rates previously determined in F344/DuCrl rats ranged from 0.05 to 0.31 µg/kg bw/d during periods of low-furan diet and increased to 0.18–1.20 µg/kg bw/d during consumption of a high-furan diet. These estimates are in good agreement with exposures reported by the European Food Safety Authority (EFSA), which range between 0.11 to 0.75 µg/kg bw per day (minimum LB to maximum UB) for the average adult and 0.20 to 1.22 µg/kg bw per day (minimum LB to maximum UB) for highly exposed adult consumers (95th percentile). Interestingly, higher excretion of GSH-BDA was observed in smoking compared to non-smoking individuals, highlighting tobacco smoke as a significant source of furan exposure and confounding factor when estimating furan exposure via diet. In contrast to GSH-BDA, high urinary background levels of lysine adducts of BDA and BDA-derived cysteine-lysine crosslinks limit their suitability as biomarkers of exogenous furan exposure.