Reduction of misleading ("false") positive results in mammalian cell genotoxicity assays. III: sensitivity of human cell types to known genotoxic agents.

We have demonstrated previously that the seemingly high rate of "false" or "misleading" positive results from in vitro micronucleus assays (MNvit) was greater when rodent derived cell lines and certain toxicity measures, such as relative cell count or replication index, were used. These studies suggested that the use of a human cell type with functional p53 and a toxicity measure that included a function of cell proliferation could dramatically reduce the detection of misleading positive results. A reduced "false positive rate" should not be at the expense of a loss of sensitivity of the assay. Therefore, we have investigated the sensitivity of the MNvit assay to known genotoxic agents using three cell types shown previously to be less prone to misleading positives, namely human lymphocytes (HuLy), TK6 and HepG2 cells. The 17 chemicals are well characterised and are from a list of chemicals known to produce positive results in in vitro mammalian cell assays. These data demonstrated a high sensitivity of the assay in which TK6 and HuLy cells were employed, such that 15 out of the 17 chemicals were correctly identified. By contrast, the use of HepG2 cells resulted in far fewer than expected positive responses. In conclusion, using TK6 and HuLy cells in preference to long established rodent cell lines in order to improve specificity does not compromise the sensitivity of the MNvit to detect known genotoxic agents.