The toxicity of benzene, a human leukemogen and ubiquitous environmental pollutant, is mediated in part by ring-hydroxylated metabolites including hydroquinone (HQ) and ring-opened metabolites including trans,trans-muconaldehyde (muconaldehyde, MUC), and their interactions. DNA-protein crosslinks (DNAPC) and DNA strand breaks (DNASB) are toxic lesions associated with the mechanism(s) of toxicity of carcinogenic compounds. In the present studies, we examined the hypothesis that individual and interactive effects of MUC and HQ are involved in the formation of DNAPC and DNASB. We extended our previous studies on DNAPC induction by MUC in HL-60 cells to HQ and mixtures of MUC and HQ, and determined DNASB levels, including 3'OH DNASB. Treatment of HL-60 cells with 25 to 100 microM HQ followed by incubation for 4 hours resulted in 1.3- to 2.8-fold increases in DNAPC levels compared with control, as determined by a K+/sodium dodecyl sulfate (SDS) precipitation assay. At 25 and 100 microM, MUC was 1.8 and 4.9 fold more effective at inducing DNAPC than HQ. Treatment with equimolar mixtures of 25 or 50 microM MUC and HQ resulted in higher DNAPC formation relative to the DNAPC levels expected if the effects were only additive. 3'OH DNASB levels as determined by the TUNEL assay showed a significant concentration-dependent increase 1 hour after treatment with 5 to 25 microM MUC, whereas HQ treatment had no effect. Cotreatment with 25 and 50 microM MUC/HQ mixtures resulted in significant decreases in TUNEL labeling relative to treatment with MUC alone. HL-60 cells treated with 1 to 50 microM MUC or HQ exhibited concentration- and time-dependent increases in DNASB as determined by the FADU assay, which measures a variety of single- and double-strand breaks and alkali labile sites. Exposure to 10 microM MUC gave QDNASB values (1 QDNASB approximately equals 100 DNASB/cell) of 7.5 +/- 1.2 and 15.4 +/- 1.4 at the 1- and 2-hour time points respectively, compared with 0.1 +/- 3.8 and 0.0 +/- 1.5 for the corresponding time controls. The QDNASB values after treatment with 10 microM HQ were 4.4 +/- 0.7 and 17.7 +/- 2.1 at the 1- and 2-hour time points, respectively, compared with 0.0 +/- 0.5 and 0.0 +/- 1.3 for the corresponding time controls. Induction of DNASB was additive 1 hour after treatment with equimolar MUC/HQ mixtures of 5 to 50 microM. These in vitro findings are significant in that DNAPC and DNASB lesions induced by MUC and HQ as well as their interactions could contribute to benzene-induced hematotoxicity and leukemogenesis.