Glycogen synthase kinase (GSK)-3β is a multifunctional protein that has been implicated in the pathological characteristics of Alzheimer's disease (AD), including the heightened levels of neurofibrillary tangles, amyloid-beta (Aβ), and neurodegeneration. In this study we used 12-month-old SAMP8 mice, an AD model, to examine the effects GSK-3β may cause regarding the cognitive impairment and oxidative stress associated with AD. To suppress the level of GSK-3β, SAMP8 mice were treated with an antisense oligonucleotide (GAO) directed at this kinase. We measured a decreased level of GSK-3β in the cortex of the mice, indicating the success of the antisense treatment. Learning and memory assessments of the SAMP8 mice were tested post-antisense treatment using an aversive T-maze and object recognition test, both of which observably improved. In cortex samples of the SAMP8 mice, decreased levels of protein carbonyl and protein-bound HNE were measured, indicating decreased oxidative stress. Nuclear factor erythroid-2-related factor 2 (Nrf2) is a transcription factor known to increase the level of many antioxidants, including glutathione-S transferase (GST), and is negatively regulated by the activity of GSK-3β. Our results indicated the increased nuclear localization of Nrf2 and level of GST, suggesting the increased activity of the transcription factor as a result of GSK-3β suppression, consistent with the decreased oxidative stress observed. Consistent with the improved learning and memory, and consistent with GSK-3b being a tau kinase, we observed decreased tau phosphorylation in brain of GAO-treated SAMP8 mice compared to that of RAO-treated SAMP8 mice. Lastly, we examined the ability of GAO to cross the blood-brain barrier and determined it to be possible. The results presented in this study demonstrate that reducing GSK-3 with a phosphorothionated antisense against GSK-3 improves learning and memory, reduces oxidative stress, possibly coincident with increased levels of the antioxidant transcriptional activity of Nrf2, and decreases tau phosphorylation. Our study supports the notion of GAO as a possible treatment for AD.