Abstract: Objective To evaluate the in vivo genotoxic risk of emodin administered by oral gavage to mice for 28 consecutive days. Methods KM mice were repeatedly administered with emodin for 28 consecutive days by oral gavage, followed by a recovery period of 28 days. Peripheral blood was collected before administration, 14, 28, 42 and 56 days after the first administration to detect the mutation rate of reticulocytes and total red blood cells as well as the percentage of reticulocytes in total red blood cells. Liver, kidney and peripheral blood was collected after the last administration to carry out the comet test. The percentage of tail DNA of at least 100 cells in each animal was analyzed. Samples of bone marrow cells were prepared after the last administration, and polychromatic erythrocytes as well as the incidence of micronuclei in polychromatic erythrocytes were calculated. Results Compared with the vehicle control group (0.5% carboxymethylcellulose sodium), the Pig-a gene mutation rate in groups dosed with emodin was not increased after 28 consecutive days of administration of 300 mg·kg^-1 or under. Compared with the vehicle control group (0.5% carboxymethylcellulose sodium), the percentage of tail DNA of kidney cells in groups dosed with emodin was significantly increased with a dose-response pattern (P<0.01, P<0.001）. Data on bone marrow micronucleus test showed no increase in the rate of micronuclei within the current dose range. Conclusion After 28 consecutive days of oral administration of emodin to mice, the main action site of emodin in mice is found to be the kidney. Under current conditions, emodin does not increase in the rate of Pig-a gene mutation or in the incidence of bone marrow micronuclei in mice, but it could lead to DNA damage to kidney cells.

Key words: Polygonum multiflorum, emodin, genotoxicity, KM mice, Pig-a, comet assay, micronucleus assay, risk assessment