Topoisomerase II-targeted drugs, such as etoposide, "poison" this enzyme and kill cells by increasing levels of covalent topoisomerase II-cleaved DNA complexes. In spite of the success of this drug in the treatment of human cancers, a significant proportion of patients treated with etoposide eventually develop secondary leukemias that are characterized by translocations at chromosome band 11q23. Since similar translocations are associated with primary leukemias in previously untreated infants, we questioned whether they could also be triggered by the actions of "endogenous topoisomerase II poisons". Recent studies, which demonstrated that several forms of spontaneous DNA damage stimulate cleavage mediated by Drosophila topoisomerase II, suggest that DNA lesions may act as these endogenous poisons. Therefore, to determine whether the ability to recognize spontaneous DNA damage has been conserved from this lower eukaryote to mammalian species, the effects of apurinic sites, apyrimidinic sites, and deaminated cytosine residues on human topoisomerase IIalpha were assessed. All three lesions were potent poisons of the human enzyme and stimulated cleavage when located within the four-base overhang generated by enzyme-mediated DNA scission. Furthermore, these lesions increased levels of cleavage at five sites proximal to 11q23 translocation breakpoints and did so with an efficacy that was comparable to or greater than that of therapeutic concentrations of etoposide. Although the physiological relevance of these findings has yet to be established, they suggest a potential role for endogenous topoisomerase II poisons in the initiation of leukemic chromosomal breakpoints.