© 1996 Oxford University Press
research-article |
Comparison of the Hepatotoxicity of Coumarin in the Rat, Mouse, and Syrian Hamster: A Dose and Time Response Study1
*BIBRA International Woodmansterne Road, Carshalton, Surrey SM5 4DS, England
Robens Institute, University of Surrey Guildford, Surrey GU2 5XH, England
Received April 18, 1996; accepted August 13, 1996
The effects of coumarin treatment have been compared in male Sprague-Dawley CD rats, male CD-1 mice, and male Syrian hamsters. Rats were fed 0–0.75% coumarin for 1 and 4 weeks and 0–0.5% coumarin for 13 weeks, whereas mice and Syrian hamsters were fed 0–0.5 and 0–1.0% coumarin, respectively, for periods of 1, 4, and 13 weeks. In the rat, coumarin produced dose-related hepatotoxic effects which included vacuolar degeneration, apoptosis, and bile duct proliferation. These effects were particularly marked at dose levels of 0.3 and 0.5%, where liver tumors have been observed in a chronic study. Coumarin administration to rats also increased serum bilirubin content and both serum and hepatic 
-glutamyltransferase activity. While levels of hepatic total glutathione were increased by coumarin administration, microsomal cytochrome P450 content and ethylmorphine N-demethylase activity were reduced. Such effects were either less marked or absent in the mouse and Syrian hamster. Replicative DNA synthesis was studied by implanting osmotic pumps containing 5-bromo-2'-deoxyuridine during Study Weeks 0–1, 3–4, and 12–13. In the rat, coumarin administration for 4 and 13 weeks at dose levels of 0.3 and 0.5% produced a sustained stimulation of hepatocyte replicative DNA synthesis. No such effects were observed in the mouse and Syrian hamster. These results demonstrate marked species differences in coumarin-induced hepatotoxicity. While tumor formation in the rat appears due to chronic hepatotoxicity associated with a sustained regenerative hyperplasia, such effects were not observed in the CD-1 mouse and Syrian hamster. In assessing the hazard of coumarin to humans, account needs to be taken of both levels of exposure and species differences in response.