In vivo genotoxicity of EMS: statistical assessment of the dose response curves.

EMS induced micronuclei and lacZ mutations in in vivo studies in mice with a clearly sublinear dose dependency. As reported elsewhere in this issue, NOEL dose values of between 25 mg/kg/day and 80 mg/kg/day were observed for the different endpoints and tissues analysed. Here we show that statistical assessment of the data provides solid support that the induction of mutagenic and clastogenic effects after in vivo treatment with the directly DNA damaging mutagen EMS adheres to a thresholded dose response relation. These data corroborate similar evidence obtained in in vitro studies. We conclude that cells are fully capable of repairing large amounts of DNA ethylations induced by EMS without experiencing elevated mutation frequencies. The stochastic, linear risk assessment model generally employed for DNA damaging genotoxins can therefore be refuted for EMS. While presently this conclusion cannot be generalized to other genotoxins a change of paradigm appears to be indicated at least for alkylating agents inducing a comparable type and spectrum of DNA lesions as EMS.