Hyperoxia, used therapeutically in the treatment of respiratory insufficiencies, can cause lung injury, probably through the actions of reactive oxygen species. The present studies were designed to test the hypothesis that oxidation of specific proteins would provide useful biomarkers of the onset of tissue injury, and thereby provide clues as to the mechanisms responsible. We exposed adult male Sprague-Dawley rats to room air or to greater than 95% O2 for 60 h and examined proteins in pleural effusion and broncho-alveolar lavage (BAL) fluids, and in lung tissue homogenates and subfractions. Oxidation of protein thiols was assessed by derivatization with monobromobimane, separation by electrophoresis, and visualization of the fluorescent thioether derivatives. Derivatization with 2,4-dinitrophenylhydrazine (DNPH), electrophoresis, and western analysis was employed to assess a different class of oxidative modifications, frequently termed 'protein carbonyls'. In addition, we investigated the effects of the 21-aminosteroid U-74389G, 10 mg/kg, given intraperitoneally every 12 h, on biomarkers of protein oxidation and on manifestations of lung injury. Hyperoxia caused lung injury evidenced by pleural effusions, increases in BAL protein concentrations, and pulmonary edema; U-74389G attenuated the first two indices of lung injury, but did not alter edema. Protein thiol status of the fractions studied were not affected notably by hyperoxia, or by the aminosteroid. The formation of DNPH-reactive sites on a limited number of proteins by hyperoxia was observed, and some of these effects were attenuated in the animals given U-74389G. Histological examination of lung tissues showed accumulation of intra-alveolar protein exudates in hyperoxic rats, and a significant attenuation of this effect was observed in the animals treated with U-74389G. In conclusion, studies of shifts in protein thiol status that may be caused by hyperoxia will require increasingly specific methods of analysis, and characterization of the specific DNPH-reactive proteins formed in hyperoxia may provide critical insights into the mechanisms of lung injury. Administration of U-74389G offers some degree of protection against hyperoxia and attenuation of these biomarkers of oxidation, but the precise mechanisms by which this protection is effected will require additional study.