Toxicity studies of Caramel Colour III and 2-acetyl-4(5)-tetrahydroxybutylimidazole in F344 rats

Abstract

Caramel Colour III is used as a colour additive in beers and a variety of foods. Beer is the most important single source of Caramel Colour III in the diet although consumption of dark beers has been decreasing in recent years. The Joint FAO/WHO Expert Committee on Food Additives (JECFA) has established an acceptable daily intake of 200 mg/kg/day for Caramel Colour III. The safety of Caramel Colour III has been questioned during recent years following feeding studies in the rat that were associated with reduced white cell and lymphocyte counts. These effects have been attributed to the presence of 2-acetyl-4(5)-tetrahydroxybutylimidazole (THI) in this class of caramel colour. Short-term oral toxicity studies were conducted on low-THI and high-THI samples of Caramel Colour III (13 wk) and on a sample of THI (28 days). In both studies, the test materials were mixed with demineralized water and the solutions were given to the animals ad lib. in the drinking fluid. In the 13-wk subchronic toxicity study of Caramel Colour III, groups of 20 rats/sex were given concentrations of caramel colour equivalent to intakes of 0, 10, 15 or 20 g low-THI caramel colour/kg body weight/day or 20 g/kg of a high-THI caramel colour. In the 4-wk toxicity study with THI, groups of 20 rats/sex were given 0, 8 or 64 ppm THI (equivalent to approx. 0, 0.9 or 7.2 mg/kg/day) and 10 rats/sex were given 1, 2, 4, 16 or 32 ppm THI (equivalent to approx. 0.1, 0.2, 0.5, 1.9 or 3.7 mg/kg/day) for 4 wk followed by a 2-wk recovery phase for 10 rats/sex in the 0, 8 and 64 ppm groups. Rats given Caramel Colour III had soft faeces; there were no other treatment-related clinical observations and no treatment-related deaths occurred. All treated groups given Caramel Colour III had lower food and fluid consumption than controls. Males given 15 or 20 g low-THI caramel colour/kg or 20 g high-THI caramel colour/kg and females given 20 g/kg of either type had lower body weights than controls. In the 4-wk toxicity study with THI, there were no treatment-related ante-mortem observations, and no effects on body weights or food consumption. Fluid consumption by males and females treated with 64 ppm THI was lower than that of controls. Haematological changes noted in both studies included lower total white blood cell and lymphocyte (absolute and percentage) counts, and higher absolute and percentage neutrophil counts; these changes returned to normal by study termination in the caramel colour study, and by the third day post-treatment in the THI study. There were no consistent treatment-related alterations in blood chemistry or urinalysis variables for either study, and any changes noted were not associated with macroscopic or microscopic pathological alterations. There were no toxicologically important pathological findings for either study. Based on these studies the no-observed-adverse-effect level (NOAEL) for low-THI Caramel Colour III was considered to be 20 g/kg body weight. The acceptable daily intake for this caramel colour established by JECFA is 200 mg/kg body weight. The NOAEL for THI was found to be 0.38 mg/kg body weight for male rats and 0.12 mg/kg body weight for females.