The free radical theory of aging proposes that reactive oxygen species (ROS) cause oxidative damage over the lifetime of the subject. It is the cumulative and potentially increasing amount of accumulated damage that accounts for the dysfunctions and pathologies seen in normal aging. We have previously demonstrated that both normal rodent brain aging and normal human brain aging are associated with an increase in oxidative modification of proteins and in changes in plasma membrane lipids. Several lines of investigation indicate that one of the likely sources of ROS is the mitochondria. There is an increase in oxidative damage to the mitochondrial genome in aging and a decreased expression of mitochondrial mRNA in aging. We have used a multidisciplinary approach to the characterization of the changes that occur in aging and in the modeling of brain aging, both in vitro and in vivo. Exposure of rodents to acute normobaric hyperoxia for up to 24 h results in oxidative modifications in cytosolic proteins and loss of activity for the oxidation-sensitive enzymes glutamine synthetase and creatine kinase. Cytoskeletal protein spin labeling also reveals synaptosomal membrane protein oxidation following hyperoxia. These changes are similar to the changes seen in senescent brains, compared to young adult controls. The antioxidant spin-trapping compound N-tert-butyl-alpha-phenylnitrone (PBN) was effective in preventing all of these changes. In a related study, we characterized the changes in brain protein spin labeling and cytosolic enzyme activity in a series of phenotypically selected senescence-accelerated mice (SAMP), compared to a resistant line (SAMR1) that was derived from the same original parents. In general, the SAM mice demonstrated greater oxidative changes in brain proteins. In a sequel study, a group of mice from the SAMP8-sensitive line were compared to the SAMR1-resistant mice following 14 days of daily PBN treatment at a dose of 30 mg/kg. PBN treatment resulted in an improvement in the cytoskeletal protein labeling toward that of the normal control line (SAMR1). The results of these and related studies indicate that the changes in brain function seen in several different studies may be related to the progressive oxidation of critical brain proteins and lipids. These components may be critical targets for the beneficial effects of gerontotherapeutics both in normal aging and in disease of aging.