Spatial learning deficit in transgenic mice that conditionally over-express GSK-3β in the brain but do not form tau filaments


Address correspondence and reprint requests to José J. Lucas, Centro de Biología Molecular ‘Severo OchoaCSIC/UAM, Fac. Ciencias, Universidad Autónoma de Madrid., Cantoblanco, 28049 Madrid, Spain. E-mails:


Deregulation of glycogen synthase kinase-3 (GSK-3) activity in neurones has been postulated as a key feature in Alzheimer's disease (AD) pathogenesis. This was further supported by our recent characterization of transgenic mice that conditionally over-express GSK-3β in hippocampal and cortical neurones. These mice, designated Tet/GSK-3β, showed many of the biochemical and cellular aspects of AD neuropathology such as tau hyperphosphorylation and somatodendritic localization, decreased nuclear β-catenin, neuronal death and reactive gliosis. Tet/GSK-3β mice, however, did not show tau filament formation up to the latest tested age of 3 months at least. Here we report spatial learning deficits of Tet/GSK-3β mice in the Morris water maze. In parallel, we also measured the increase in GSK-3 activity while further exploring the possibility of tau filament formation in aged mice. We found a significant increase in GSK-3 activity in the hippocampus of Tet/GSK-3β mice whereas no tau fibrils could be found even in very old mice. These data reinforce the hypothesis of GSK-3 deregulation in AD pathogenesis, and suggest that Tet/GSK-3β mice can be used as an AD model and, most remarkably, can be used to test the therapeutic potential of the selective GSK-3 inhibitors that are currently under development. Additionally, these experiments suggest that destabilization of microtubules and alteration of intracellular metabolic pathways contribute to AD pathogenesis independent of toxicity triggered by the aberrant tau deposits.