Collaborative study of thresholds for mutagens: Hormetic responses in the micronucleus test and gene induction by mutagenic treatments

The linear no-threshold model (LNT) asserts that the genetic effects and carcinogenicity of radiation are proportional to the dose. LNT is also applied to carcinogens and mutagens. However, most experimental data show that the dose-response curve is not linear but rather a J-shaped curve, known as a hormetic response. LNT and hormesis are mutually exclusive. Which is correct? In this study, we investigated dose-response curves of mutagens in the micronucleus test using rodent cells. Since the frequency of background micronuclei was low, detecting a further decrease was difficult. When we conducted a challenge test, where cells were pre-treated with a low dose and post-treated with a high dose, clear hormetic responses were observed. Additionally, during a cross-reaction test, where cells were pre-treated with a low dose of one mutagen and post-treated with a high dose of another mutagen, unequivocal hormetic responses were detected. To investigate gene expression patterns, human lymphoma TK6 cells were treated with mitomycin C, ethyl methanesulfonate, and hydrogen peroxide, and the expression of six genes was examined by RT-PCR. Both GADD45A and p21 genes were induced in a time- and dose-dependent manner. In conclusion, the mutagens used here exhibit hormesis, indicating that the LNT model is invalid.