Electron paramagnetic resonance with spin trapping is a useful technique to detect reactive oxygen species, such as superoxide radical anion (O2*-), a key species in many biological processes. We evaluated the abilities of four spin traps in trapping cell-generated O2*-: 5-tert-butoxycarbonyl-5-methyl-1-pyrroline N-oxide (BMPO), 2-diethoxyphosphoryl-2-phenethyl-3,4-dihydro-2H-pyrrole N-oxide (DEPPEPO), 5-diethoxyphosphoryl-5-methyl-1-pyrroline N-oxide (DEPMPO), and 5,5-dimethyl-1-pyrroline N-oxide (DMPO). Optimal experimental conditions for obtaining maximal signal intensity of O2*- adduct in a cellular system were first studied. The maximal intensities of BMPO, DEPMPO, and DMPO adducts were similar while DEPPEPO did not trap cell-generated O2*- induced by 1,6-benzo[a]pyrene quinone in a human mammary epithelial cell line (MCF-10A). BMPO and DEPMPO adducts were more stable, considering the stability of their maximal signal, than DMPO adduct in the tested cellular systems. In addition, we observed that O2*- spin adducts were reduced to their corresponding hydroxyl adducts in the cellular system. The selection of optimal spin trap in trapping cell-generated O2*- is discussed.