Coumarin is the basic structure of numerous naturally occurring compounds with important and diverse physiological activities. More than a thousand coumarin derivatives have been described, varying from simple coumarins containing alkyl and hydroxyl side chains to complex coumarins with benzoyl, furanoyl, pyranoyl, or alkylphosphorothionyl substituents. Coumarin and 3,4-dihydrocoumarin were nominated by the Food and Drug Administration and the National Cancer Institute for study because of the widespread use of coumarin in perfumes, cosmetics, and other products as a fragrance, continued interest in coumarin compounds as flavor-enhancing agents for foods, and the interest in structure-activity relationships of this important group of compounds. Coumarin is believed to be metabolized to a 3,4-epoxide intermediate, which may be responsible for its toxic effects, while 3,4-dihydrocoumarin, which lacks the 3,4-double bond, is not considered likely to form an epoxide intermediate.
Toxicity and carcinogenicity studies were conducted by administering coumarin (97% pure) in corn oil by gavage to groups of male and female F344/N rats and B6C3F1 mice for 16 days, 13 weeks, and 2 years. Genetic toxicology studies were conducted in Salmonella typhimurium, cultured Chinese hamster ovary cells, Drosophila melanogaster, and B6C3F1 mice.
Sixteen-day study in rats
Groups of five male and five female rats received coumarin in corn oil by gavage at doses of 0, 25, 50, 100, 200, or 400 mg per kg body weight, 5 days a week for a total of 12 doses in a 16-day period. All female rats and four male rats receiving 400 mg/kg died. The mean body weight gains and final mean body weights of surviving dosed male and female rats were similar to those of the controls. There were no clinical signs of organ-specific toxicity, and there was no evidence of impaired blood coagulation from measurements of capillary clotting time or prothrombin and activated partial thromboplastin time.
Sixteen-day study in mice
Groups of five male and five female mice received coumarin in corn oil by gavage at doses of 0, 40, 75, 150, 300, or 600 mg per kg body weight, 5 days a week for a total of 12 doses in a 16-day period. All mice receiving 600 mg/kg, two male mice receiving 300 mg/kg, and one male mouse receiving 75 mg/kg died. The mean body weight gains and final mean body weights of surviving dosed male and female mice were similar to those of the controls. Clinical findings of inactivity, excessive lacrimation, piloerection, bradypnea, ptosis, or ataxia were observed in some mice from the 300 and 600 mg/kg groups within the first several hours after dosing. Capillary clotting time and platelet counts of dosed mice were similar to those of controls.
Thirteen-week study in rats
Groups of 10 male and 10 female rats received coumarin in corn oil by gavage at doses of 0,19, 38, 75,150, or 300 mg per kg body weight. Three male and three female rats receiving 300 mg/kg died. The mean body weight gains and final mean body weights of male rats that received 150 and 300 mg/kg were significantly lower than those of the controls. There were no clinical signs related to specific organ toxicity.
Male and female rats receiving coumarin exhibited dose-related decreases in mean erythrocyte volume and mean erythrocyte hemoglobin, and dose-related increases in erythrocyte counts. Serum levels of total bilirubin and one or more cytoplasmic enzymes including alanine aminotransferase, aspartate aminotransferase, ornithine carbamoyltransferase, and/or sorbitol dehydrogenase in males and females receiving 300 mg/kg were higher than those of controls.
The absolute and relative liver weights of male and female rats that received 150 and 300 mg/kg were significantly greater than those of the controls. Centrilobular hepatocellular degeneration and necrosis, chronic active inflammation, and bile duct hyperplasia were observed in the liver of rats receiving 150 or 300 mg/kg.
The high dose selected for the 2-year study was 100 mg/kg, which was just below the level at which mortality, lower final mean body weights, and treatment-related liver lesions were observed in the 13-week study.
Thirteen-week study in mice
Groups of 10 male and 10 female mice received coumarin in corn oil by gavage at doses of 0, 19, 38, 75, 150, or 300 mg per kg body weight. Two male mice receiving 300 mg/kg died. The mean body weight gain and final mean body weight of surviving male mice that received 300 mg/kg were significantly lower than those of the controls. No clinical signs of toxicity were observed.
Male and female mice receiving coumarin exhibited dose-related decreases in mean erythrocyte volume and mean erythrocyte hemoglobin. The absolute and relative liver weights of males and females that received 150 and 300 mg/kg were significantly greater than those of the controls. Centrilobular hepatocellular hypertrophy was observed in male and female mice receiving 300 mg/kg.
The high dose selected for the 2-year study was 200 mg/kg, which was just below the level at which mortality and liver lesions were observed in the 13-week study.
Two-year study in rats
Groups of 60 male and 60 female rats were administered coumarin in corn oil by gavage at doses of 0, 25, 50, or 100 mg per kg body weight. After 15 months, 10 animals from each group were evaluated.
Survival, body weights, and clinical findings
None of the male rats receiving 100 mg/kg and only two males receiving 50 mg/kg survived until the end of the study (vehicle control, 28/50; 25 mg/kg, 9/50; 50 mg/kg, 2/51; 100 mg/kg, 0/50). Survival of dosed female rats was similar to that of the controls (29/50, 38/50, 36/50, 30/50). The reduced survival in dosed male rats was primarily attributed to chemical-related exacerbation of spontaneously occurring renal disease. Final mean body weights of female rats that received 100 mg/kg and all dosed groups of male rats were lower than those of the controls. There were no clinical signs of toxicity in rats, other than nonspecific signs relating to debilitation as a result of renal or other spontaneous disease.
Hematology and clinical chemistry
At the 15-month interim evaluation, the values for one or more hematologic parameters including mean erythrocyte volume, mean erythrocyte hemoglobin in 50 and 100 mg/kg rats, and hematocrit or hemoglobin in 100 mg/kg rats were significantly lower than those of controls. Activated partial thromboplastin times were also significantly lower in 50 and 100 mg/kg males, while platelet counts were significantly higher. Activities of alanine aminotransferase, sorbitol dehydrogenase, or g-glutamyltransferase in 50 and 100 mg/kg male and 100 mg/kg female rats were significantly higher than those of the controls at the 15-month interim evaluation.
Pathology findings
The principal lesions associated with the administration of coumarin to rats for up to 2 years occurred in the liver, kidney, and forestomach. While the hepatic lesions were seen in all groups of males, they occurred only in the 50 and 100 mg/kg females. The lesions consisted of a spectrum of changes including hepatocellular necrosis, fibrosis, cytologic alteration, and increased severity of bile duct hyperplasia. The incidences of hepatocellular neoplasms were not increased in dosed rats.
There was a chemical-related increase in the average severity of nephropathy in all groups of dosed male and female rats. There were corresponding increased incidences of parathyroid gland hyperplasia in all groups of dosed males, probably as a result of compromised renal function. In the standard evaluation of single kidney sections, a low incidence of renal adenomas was seen in all groups of males and in 100 mg/kg females (males: vehicle control, 1/49; 25 mg/kg, 2/50; 50 mg/kg, 2/51; 100 mg/kg, 1/50; females: 0/49, 0/50, 0/50, 2/49). An evaluation of step sections identified additional individuals with renal tubule focal hyperplasia (males: 2/49, 12/50, 10/51, 6/50; females: 1/49, 0/50, 4/50, 2/49) and adenoma (males: 0/49, 4/50, 5/51, 4/50; females: 0/49, 0/50, 1/50,1/49) in the dosed groups.
The incidences of forestomach ulcers in all groups of dosed male rats and in 100 mg/kg female rats were significantly greater than those of the controls (males: 7/48, 24/50, 35/51, 34/50; females: 1/48, 1/49, 6/50, 9/48).
Stop-exposure evaluation
A group of 40 male rats received 100 mg/kg coumarin in corn oil by gavage for 9 months, when 20 of the animals were necropsied and evaluated. The remainder of the male rats received only the corn oil vehicle during the 15-month recovery period. Similarly, a group of 30 male rats received 100 mg/kg coumarin in corn oil by gavage for 15 months, when 10 of the rats were necropsied and evaluated. The remaining 20 rats received only corn oil during the 9-month recovery period. A group of 20 vehicle control male rats were necropsied at 9 months, and another 10 vehicle control male rats were necropsied at 15 months.
While chemical-related hepatic lesions were seen at both the 9- and 15-month interim evaluations, the incidences and severities of these lesions following the recovery period were generally similar to controls. Thus, the hepatic lesions produced by 9 or 15 months of exposure were reversible. In contrast to the liver lesions, the severity of nephropathy in male rats following the recovery period was significantly greater than that of males examined at the 9- and 15-month interim evaluations. This is not unexpected, since nephropathy is a progressive degenerative disease that naturally increases in severity with age.
The incidence of renal tubule hyperplasia in the 15-month stop-exposure group (dosed for 15 months followed by the recovery period) and the incidence of renal tubule adenoma in the 9-month stop-exposure group were significantly greater than those of the control group.
Two-year study in mice
Groups of 70 male and 70 female mice were administered coumarin in corn oil by gavage at doses of 0, 50, 100, or 200 mg per kg body weight for up to 2 years. After 15 months, 19 or 20 mice from each group were evaluated.
Survival, body weights, and clinical findings
Survival of dosed male and female mice was similar to that of the controls (males: vehicle control, 43/50; 50 mg/kg, 47/50; 100 mg/kg, 42/50; 200 mg/kg, 37/51; females: 33/50, 40/50, 42/51, 28/51). The mean body weights of 200 mg/kg male and female mice were lower than those of controls throughout much of the study. There were no clinical findings related to chemical administration.
Hematology and clinical chemistry
Mean erythrocyte volume, mean erythrocyte hemoglobin, and hematocrit of 200 mg/kg males and mean erythrocyte volume of 200 mg/kg females were significantly lower than those of the controls. Blood platelet counts of 200 mg/kg males and females were significantly higher than those of controls. There were no biologically significant differences in enzyme activities between dosed and control mice.
Pathology findings
The principal toxic lesions associated with the administration of coumarin to mice occurred in the liver. The incidences of centrilobular hypertrophy in 100 and 200 mg/kg males and 200 mg/kg females were significantly greater than those of controls. The incidences of syncytial alteration in all male dose groups and in 200 mg/kg females were also significantly greater than controls.
The incidences of eosinophilic foci, a putative preneoplastic lesion, and of hepatocellular adenoma were significantly greater in the 50 and 100 mg/kg females. Hepatocellular carcinomas occurred with low incidences in the dosed females, but none occurred in the controls. The overall incidence of hepatocellular neoplasms (benign and malignant combined) in the 50 and 100 mg/kg females (control, 8/50; 50 mg/kg, 27/49; 100 mg/kg, 31/51; 200 mg/kg, 13/50) exceeds the range in historical controls (range 2%-34%; 129/898, 14.4%) from recent NTP studies. The reason for a lack of liver response in 200 mg/kg female mice is not known, but may be due in part to the decrease in body weight. While the incidences of eosinophilic foci were marginally greater in dosed male mice, the incidences of hepatocellular neoplasms were similar among the dosed and control groups.
The incidences of alveolar/bronchiolar adenomas were significantly greater in 200 mg/kg male and female mice than in the controls. Further, the incidence of alveolar/bronchiolar carcinoma in 200 mg/kg females was also significantly greater than in controls. The overall incidence of pulmonary neoplasms (benign and malignant combined) in the 200 mg/kg groups (males: 14/50, 9/50,15/50, 25/51; females: 2/51, 5/49, 7/49, 27/51) exceeds the range in historical controls (males: range 6%-28%; 166/900, 18.4%; females: range 0%-14%; 58/899, 6.5%) from recent NTP studies.
The incidence of squamous cell papilloma of the forestomach in 50 mg/kg males was greater than that of the controls (2/50, 8/50, 2/50, 0/51) and also exceeds the range of this neoplasm in control male mice from recent NTP studies (range 0%-14%; 27/902, 3.0%). The incidence of squamous cell papilloma of the forestomach in 50 mg/kg female mice was also slightly increased (1/52, 5/50, 2/51, 2/51); however, the incidence did not exceed the NTP historical range (27/901, 3%; range, 0%-10%).
Genetic toxicology
Coumarin induced gene mutations in Salmonella typhimurium strain TA100 in the presence, but not in the absence, of exogenous metabolic activation (S9); no mutations were induced in strains TA98, TA1535, or TA1537, with or without S9. In Chinese hamster ovary cells, coumarin induced sister chromatid exchanges in the absence of S9, and chromosomal aberrations in the presence of S9. Coumarin did not induce sex-linked recessive lethal mutations in germ cells of male Drosophila melanogaster treated either as adults by feeding or injection, or as larvae by feeding. No increase in the frequency of micronucleated erythrocytes was observed in peripheral blood of male and female B6C3F1 mice administered coumarin by gavage for 13 weeks.
Conclusions
Under the conditions of these 2-year gavage studies there was some evidence of carcinogenic activity of coumarin in male F344/N rats based on increased incidences of renal tubule adenomas. There was equivocal evidence of carcinogenic activity of coumarin in female F344/N rats based on a marginally increased incidence of renal tubule adenomas. There was some evidence of carcinogenic activity of coumarin in male B6C3F1 mice based on the increased incidence of alveolar/bronchiolar adenomas. There was clear evidence of carcinogenic activity of coumarin in female B6C3F1 mice based on increased incidences of alveolar/bronchiolar adenomas, alveolar/bronchiolar carcinomas, and hepatocellular adenomas. The marginally increased incidences of squamous cell papillomas of the forestomach in male and female mice receiving 50 mg/kg may have been related to coumarin administration.
The administration of coumarin to rats was also associated with an increased severity of nephropathy in the kidney and of bile duct hyperplasia in the liver, increased incidences of ulcers of the forestomach, and necrosis, fibrosis, and cytologic alteration of the liver. Administration of coumarin to mice was also associated with centrilobular hypertrophy, syncytial alteration, and eosinophilic focus in the liver.