The DNA damage induced by 7-chloro-3-[[(N,N-dimethylamino)propyl]amino]-2-quinoxalinecarbonitrile 1,4-di-N-oxide hydrochloride (Q-85 HCl) in Caco-2 cells under hypoxic and well-oxygenated conditions has been studied by using the comet assay. This compound has shown a good in vitro profile of high selective toxicity in hypoxia, but its mechanism of action is unknown. The DNA damage has been evaluated by performing the comet assay after a 2-h treatment with Q-85 HCl (0.1, 0.2, 0.4 microM in hypoxia; 20, 40 microM in well-oxygenated conditions). The number of cells in apoptosis has also been assessed by flow cytometry analysis of Annexin V-FITC staining. The capability of the cells to repair the DNA damage and the proliferation rate was evaluated at different times after the treatment (24-168 h). Under hypoxic conditions, a clear dose-dependent increase in the number of nuclei with a comet was observed (comet score: 132 +/- 13, 343 +/- 30 and 399 +/- 1; control comet score: 42 +/- 14). Under well-oxygenated conditions, the number of nuclei with comet increased significantly with respect to the control (comet score: 273 +/- 14 and 312 +/- 9; control comet score: 27 +/- 4). Cells in apoptosis were not detected by the comet assay nor by flow cytometry. The recovery from DNA damage was time- and concentration-dependent in hypoxia (cells treated with the highest concentration still showed DNA damage after 72 h) and rather time-dependent in well-oxygenated conditions (DNA was completely repaired after 24 h). In conclusion, Q-85 HCl acts by DNA damage and not only the reduced intermediate is genotoxic but also some other derivatives and Q-85 HCl itself may be acting.