Induction of delayed mutations by benzene and ethylene dibromide in Drosophila

Genotoxicity of benzene and its metabolites

The potential role of genotoxicity in human leukemias associated with benzene (BZ) exposures was investigated by a systematic review of over 1400 genotoxicity test results for BZ and its metabolites. Studies of rodents exposed to radiolabeled BZ found a low level of radiolabel in isolated DNA with no preferential binding in target tissues of neoplasia. Adducts were not identified by 32P-postlabeling (equivalent to a covalent binding index <0.002) under the dosage conditions producing neoplasia in the rodent bioassays, and this method would have detected adducts at 1/10,000th the levels reported in the DNA-binding studies. Adducts were detected by 32P-postlabeling in vitro and following high acute BZ doses in vivo, but levels were about 100-fold less than those found by DNA binding. These findings suggest that DNA-adduct formation may not be a significant mechanism for BZ-induced neoplasia in rodents. The evaluation of other genotoxicity test results revealed that BZ and its metabolites did not produce reverse mutations in Salmonella typhimurium but were clastogenic and aneugenic, producing micronuclei, chromosomal aberrations, sister chromatid exchanges and DNA strand breaks. Rodent and human data were compared, and BZ genotoxicity results in both were similar for the available tests. Also, the biotransformation of BZ was qualitatively similar in rodents, humans and non-human primates, further indicating that rodent and human genotoxicity data were compatible. The genotoxicity test results for BZ and its metabolites were the most similar to those of topoisomerase II inhibitors and provided less support for proposed mechanisms involving DNA reactivity, mitotic spindle poisoning or oxidative DNA damage as genotoxic mechanisms; all of which have been demonstrated experimentally for BZ or its metabolites. Studies of the chromosomal translocations found in BZ-exposed persons and secondary human leukemias produced by topoisomerase II inhibitors provide some additional support for this mechanism being potentially operative in BZ-induced leukemia.

Genetic damage by bifunctional agents in repair-active pre-meiotic stages of Drosophila males

Most of our understanding of germline mutagenesis in Drosophila is based on the DNA repair-inactive, haploid post-meiotic stages. The diploid, repair-active pre-meiotic stages are more relevant to the situation encountered in somatic cells. DNA mono-adducts induced by agents like methyl methanesulphonate (MMS) and ethylene oxide (EO) are well repaired in the pre-meiotic cell stages, and these agents show therefore, no or considerable lower mutagenic activity in these stages. In contrast, in this study the two bifunctional nitrogen mustards chlorambucil (CAB) and mechlorethamine (MEC) show significantly elevated mutant frequencies of both post- and pre-meiotic germ cells. Results were similar for the X-chromosomal and the autosomal (2nd) recessive lethal (RL) test. CAB and MEC were also active in stem cells, but in comparison with post-stem cell stages they seem to be better protected. The germ cell specific response in post- and pre-meiotic cell stages was for both nitrogen mustards comparable to mutagenic activity patterns observed in the specific locus test in the mouse. It was reported that for diepoxybutane (DEB), another cross-linking agent, the ratio of the RL frequency for the 2nd- and the X-chromosome was increased from 2.1 for post-meiotic stages to 9.5 for pre-meiotic stages. In own experiments aiming to confirm this observation, a high ratio was indeed found. The induction of large deletions by DEB could be the reason for this difference, since such lesions might include both a sex-linked lethal and a vital gene required for the development of spermatocytes into mature sperm. Similar differences were expected for CAB and MEC since they are also inducers of large deletions. But unexpectedly, no differences in 2nd/X RL ratio between post- and pre-meiotic cell stages were found for the nitrogen mustards. Possible causes such as distinct proportions of multi-locus deletions (MLDs), mitotic recombination and the formation of persistent lesions, are discussed.