Glycogen synthase kinase‐3 (GSK3) is an important signalling protein in the brain and modulates different forms of synaptic plasticity. Neuronal functions of GSK3 are typically attributed to one of its two isoforms, GSK3β, simply because of its prevalent expression in the brain. Consequently, the importance of isoform‐specific functions of GSK3 in synaptic plasticity has not been fully explored. We now directly address this question for NMDA receptor‐dependent long‐term depression (LTD) in the hippocampus. Here, we specifically target the GSK3 isoforms with shRNA knock‐down in mouse hippocampus and with novel isoform‐selective drugs to dissect their roles in LTD. Using electrophysiological and live imaging approaches, we find that GSK3α, but not GSK3β, is required for LTD. The specific engagement of GSK3α occurs via its transient anchoring in dendritic spines during LTD induction. We find that the major GSK3 substrate, the microtubule‐binding protein tau, is required for this spine anchoring of GSK3α and mediates GSK3α‐induced LTD. These results link GSK3α and tau in a common mechanism for synaptic depression and rule out a major role for GSK3β in this process.

Highlights d KIF5A/C and protrudin are required for long-term depression (LTD) d KIF5A/C, protrudin, and AMPA receptors associate upon activation of NMDA receptors d Protrudin prevents AMPA receptor recycling to synapses after internalization

A significant number of individuals with type 2 diabetes (T2DM) develop cognitive deficits over time that in some cases could lead to dementia. It remains to be identified the diabetes-related factors or comorbid conditions that drive the association and how they work. In this manuscript, we show that 14-15 month-old hAPP NL/F mice, a knock‐in mouse model of preclinical Alzheimer’s disease (AD) (that is, they generate Aβ42 at an early age but do not show symptoms of the disease until they are old), but not wild type, develop behavioural deficits that correlate with electrophysiological and dendritic spine changes when affected by T2DM. The deficits are not paralleled by higher levels of toxic forms of amyloid beta-peptide (Aβ) or by neuroinflammation but by a reduction in γ-secretase activity, in turn accountable, among other consequences, for a reduction in the levels of the synaptogenesis regulator protein PSD95, and by increased phosphorylation of Tau at epitopes of dendritic spine targeting. RNAseq analysis of the cerebral cortex of wild type and hAPP NL/F, with and without T2DM, revealed the significant downregulation of hypoxia inducible factor 3 (HIF-3α) in the hAPP NL/F diabetic mice only, a decrease that was also observed at the protein level. The main conclusion of this work is that T2DM will trigger or accelerate cognitive illness fundamentally in those individuals who already have a genetic predisposition to AD. A second conclusion is that the cognitive disorders produced by diabetes in these individuals are not due to the excess of toxic forms of Aβ but to other causes, including loss of function of γ-secretase. The third conclusion is that an alteration of the response to hypoxia through Hif3α would be mediating the negative effects of T2DM in the brain.