A sulphite-oxidase-deficient rat model: Subchronic toxicology

Toxicity resulting from exposure to sulphite originating both endogenously and exogenously was investigated in normal rats and in rats made sulphite-oxidase-deficient by molybdenum deficiency abetted by administration of tungstate. The sulphite-oxidase-deficient rats were outwardly as healthy as controls and exhibited normal weight gain and maintenance over the 9-wk test period. The systemic sulphite exposures of normal and deficient rats resulting from various sulphite treatments could be compared by determining the concentrations of tissue S-sulphonate (RS-SO 3 −) metabolites formed. In general, relatively low intakes of exogenous sulphite (0–3·5 mmol/kg/day) by sulphite-oxidase-deficient rats produced systemic sulphite exposures equivalent to those produced by the ingestion by normal rats of highly sulphited diets (intakes of 13–25 mmol/kg/day). The advantages of the sulphite-oxidase-deficient rat compared to the normal rat as a model for human exposure are discussed. Using these two animal models, it was demonstrated that anaemia and thiamine deficiency, which have been produced previously in sulphite-feeding studies, result solely from the action of high concentrations of sulphite in the diet and/or gut and are not attributable to systemic sulphite exposure. Likewise, prothrombin time and erythrocyte concentrations of glutathione were not affected by high systemic sulphite concentrations in these experiments. A 4 149 incidence of mammary adenocarcinoma was observed in sulphite-oxidase-deficient rats, all in rats aged less than 5 months, compared to 0 143 observed in age-matched rats with normal sulphite oxidase. Although this result was not statistically significant, the rarity of spontaneous tumours of this type among rats of this age suggests that these carcinomas may, in fact, have been treatment related. If indicated, further investigation will be undertaken to determine the role of sulphite-oxidase-deficiency, sulphite and/or tungstate, as well as other elements of the model, in the aetiology of these tumours.