We examined the effects of pyrazoloacridine (PZA), an investigational anticancer agent in clinical trials, on cytotoxicity, DNA synthesis, and DNA damage in MCF-7 human breast carcinoma cells. With PZA concentrations ranging from 0.5 to 50 microM for durations of 3-72 hr, cytotoxicity increased in proportion to the total PZA exposure (concentration x time). Inhibition of DNA and RNA syntheses increased with increasing PZA concentration x time (microM.hr). A 24-hr exposure to 1 and 10 microM PZA reduced DNA synthesis to 62 and 5% of control, respectively, decreased the proportion of cells in S phase with accumulation of cells in G2 + M phase, and inhibited cell growth at 72 hr by 68 and 100%. Newly synthesized DNA was more susceptible to damage during PZA exposure, with subsequent induction of parental DNA damage. Significant damage to newly synthesized DNA as monitored by alkaline elution was evident after a 3-hr exposure to > or = 5 microM PZA. Longer PZA exposures (> or = 10 microM for 16 hr) were required to elicit damage to parental DNA. Induction of single-strand breaks in parental DNA correlated closely with induction of double-strand breaks and detachment of cells from the monolayer. PZA-mediated DNA fragmentation was not accompanied by the generation of oligonucleosomal laddering in MCF-7 cells, but induction of very high molecular weight DNA fragmentation (0.5 to 1 Mb) was detected by pulsed-field gel electrophoresis. In vitro binding of PZA to linear duplex DNA (1 kb DNA ladder) and closed, circular plasmid DNA was demonstrated by a shift in migration during agarose electrophoresis. PZA interfered with topoisomerase I- and II-mediated relaxation of plasmid DNA in a cell-free system, but the cytotoxic effects of PZA did not appear to involve a direct interaction with topoisomerase I or II (stabilization of the topoisomerase I- or II-DNA cleavable complex). PZA-mediated cytotoxicity correlated strongly with inhibition of DNA and RNA syntheses, and damage to both nascent and parental DNA. Neither the cytotoxicity of PZA nor induction of double-stranded DNA fragmentation was prevented by aphidicolin, indicating that PZA-mediated lethality occurred in the absence of DNA replication. Since free radical formation was not detected, induction of nascent and parental DNA damage appeared to be a consequence of the avid binding of PZA to DNA, presumably by interfering with the access of replication, repair, and transcription enzyme complexes.