Inactivation of Cu,Zn-superoxide dismutase (Cu,Zn-SOD) by its own reaction product H2O2 is a well-known phenomenon. Generation of the hydroxyl radical has been a matter of great concern, and the target molecule has been regarded as its own active site histidine residues, whose oxidized forms have not yet been identified (Hodgson, E.K., and Fridovich, I. (1975) Biochemistry 14, 5294-5299). Here we report on the identification of oxidized histidines generated at the active site of Cu,Zn-SOD by reaction with H2O2. When bovine erythrocyte Cu,Zn-SOD (0.5 mg/ml) was treated with 5 mM H2O2 in 50 mM sodium phosphate buffer (pH 7.2), histidine was significantly lost; however, except for a significant increase in aspartate and glutamate, nothing new appeared in the amino acid analysis of oxidized Cu,Zn-SOD. On the other hand, the hydrolysates of oxidized Cu,Zn-SOD involved an unknown product that was detectable by reverse-phase high performance liquid chromatography with electrochemical detection. The product was found to be identical to 2-oxo-histidine that had been discovered as the major oxidation product of histidine and its peptides treated with a copper/ascorbate-free radical generating system. The main product present in the hydrolysate of H2O2-treated Cu,Zn-SOD was 2-oxo-histidine. Approximately 0.66 mol/mol subunit was formed when Cu,Zn-SOD was treated with 5 mM H2O2 for 30 min. Both metal chelators and the hydroxyl radical scavengers only slightly inhibited the 2-oxo-histidine formation (10-39%), suggesting that the active species were produced mainly inside the ligands of the Cu2+ in the enzyme. Trypsin digestion of H2O2-treated Cu,Zn-SOD showed selective reactions at the sequences of Gly24-Lys67 and Thr114-Arg126, in that histidine residues locate at the active center. Two new products derived from those peptides appeared in the tryptic map. Amino acid analysis of both products demonstrated the loss of only histidine. One of them derived from Thr114-Arg126 contained an equimolar amount of 2-oxo-histidine, indicating that His-118 was converted selectively to 2-oxo-histidine; however, another product derived from Gly24-Lys67 contained only 0.085 mol of 2-oxo-histidine/mol of peptide, suggesting that the product is a mixture consisting of unidentified forms of oxidized histidine. Taken together, the present study provided direct evidence that 2-oxo-histidine was generated in the Cu,Zn-SOD exposed to H2O2 and that its generation was selective at histidine 118 of the active site of the enzyme.