Toxicity and carcinogenicity studies of Caramel Colour IV in F344 rats and B6C3F1 mice

Food Chem Toxicol. 1992 May;30(5):431-43. doi: 10.1016/0278-6915(92)90071-r.


Caramel Colour IV, a type of caramel colour used in the manufacture of cola soft drinks, was evaluated for subchronic and chronic toxicity in rats, and carcinogenicity in Fischer-344 (F344) rats and B6C3F1 mice. In each of the studies, Caramel Colour IV was mixed with demineralized water and the solutions given to the animals ad lib. in the drinking fluid. The concentrations of Caramel Colour IV in the drinking fluid were adjusted periodically to achieve the desired caramel colour intake per kg body weight. In the range-finding studies, groups of 30 rats/sex were given Caramel Colour IV at levels of 0, 15, 20, 25 or 30 g/kg for 13 wk, and groups of 10 male rats were given levels of 0, 2.5, 5, 10 or 15 g/kg for 6 wk followed, for some dose groups, by a 2-wk withdrawal period, and then re-initiation of dosing for another 2 wk. In the rat chronic toxicity study, levels of Caramel Colour IV of 0, 2.5, 5, 7.5 or 10 g/kg were given to groups of 25 rats/sex for 12 months. The test groups in the rat and mouse carcinogenicity studies were composed of 50 animals/sex and each species was given the caramel colour at levels of 0, 0, 2.5, 5 or 10 g/kg for 24 months. In each of the studies, treated animals tended to have dose-related lower water consumption than controls. This was attributed to poor palatability of the drinking fluid, and was generally associated with decreased food consumption and body weights. Rats given caramel colour often had soft or liquid malodorous faeces although there were no treatment-related ante-mortem observations in mice. Blood biochemical changes in the rat (i.e. reduced blood urea nitrogen, alkaline phosphatase and total serum protein) appeared to be related to dietary influences and were not considered toxicologically significant. There were no treatment-related alterations in haematological variables or treatment-related differences in survival or in the incidence of benign or malignant tumours among treated and control groups and no toxicologically important pathological findings. On the basis of these studies, Caramel Colour IV was not toxic or carcinogenic in F344 rats or B6C3F1 mice. The highest dose level tested in the long-term studies (10 g/kg) was considered to be the no-observed-adverse-effect level (NOAEL).

MeSH terms

  • Administration, Oral
  • Animals
  • Blood Proteins / analysis
  • Blood Urea Nitrogen
  • Body Weight / drug effects
  • Candy
  • Carbohydrates
  • Carcinogenicity Tests
  • Digestive System / drug effects
  • Drinking / drug effects
  • Eating / drug effects
  • Female
  • Food Coloring Agents / administration & dosage
  • Food Coloring Agents / toxicity*
  • Kidney / drug effects
  • Male
  • Mice
  • Neoplasms / chemically induced*
  • Organic Chemicals
  • Random Allocation
  • Rats
  • Rats, Inbred F344
  • Urine / chemistry


  • Blood Proteins
  • Carbohydrates
  • Food Coloring Agents
  • Organic Chemicals
  • caramel coloring