Effects of GLT-1 loss on central nervous system insulin signaling and implications for Alzheimer’s disease pathogenesis

Thesis (Ph.D.)--University of Washington, 2015Regulation of glutamatergic neurotransmission in the central nervous system (CNS) by glutamate transporters allows for the fine balance between tightly regulated signaling and prevention of glutamate-induced neurotoxicity. In forebrain regions, glutamate is essential for cognition along with regulation of metabolism. The glutamate transporter GLT-1 (also known as EAAT2) is responsible for the majority of glutamate uptake in the forebrain and in keeping with its dominant role in maintaining glutamate homeostasis, loss or dysfunction of GLT-1 has been implicated in multiple CNS disorders, including Alzheimer’s disease (AD). In AD, GLT-1 levels begin falling early and are reduced by as much as 50% in later stages of the disease. These findings, along with others, suggest a role of glutamate dyshomeostasis in AD pathogenesis. In conjunction with glutamatergic disturbances in AD, there has been a wealth of recent evidence identifying insulin signaling disturbances in the brains of individuals with AD. The insulin signaling changes identified are believed to be indicative of a state of insulin resistance in the brain, which have been found primarily in forebrain regions such as the cortex and hippocampus. Along with regulating metabolic changes in both the CNS and periphery, there is evidence that insulin action in the brain is critical for modulating cognitive processes. Similar to GLT-1 loss in AD, insulin signaling changes have been identified early in the course of the disease. Interestingly, GLT-1 expression levels have also been found to be regulated by insulin signaling. Taken collectively, these findings suggest that glutamatergic and insulin signaling share several similarities in the brain, particularly that both are important for metabolic and cognitive processes, which when disturbed may each play a role in AD pathogenesis. The goal of this thesis was to examine the relationship between GLT-1 loss and insulin signaling disturbances in the context of AD. Previous work from the Cook laboratory has shown that partial GLT-1 loss (to levels consistent with those identified in AD cases) causes deficits in cognitive function in a mouse model of AD (Mookherjee et al., 2011). However, partial loss of GLT-1 resulted in only modest changes to amyloid processing in these mice suggesting that increased amyloid pathology was not responsible for the accelerated onset of cognitive deficits. Given the interactions between glutamatergic and insulin signaling and their similar functions in the brain, we examined if partial GLT-1 loss resulted in disturbances to CNS insulin signaling in these same animals. We found alterations to several components of the pathway, including decreased insulin receptor and IRS-1 activation along with increased Akt activation, indicative of an overall reduction in insulin signaling in the brain. These changes mirrored the onset of cognitive deficits previously identified in these mice and were similar to insulin signaling disturbances identified in AD brains. As insulin signaling changes in AD have been identified predominantly in neurons and neuronal GLT-1 is responsible for a significant portion of glutamate uptake even though its expression levels are low, we utilized primary cortical neurons to determine the mechanistic relationship between GLT-1 loss and insulin signaling alterations. Loss of neuronal GLT-1 function in vitro resulted in a significant decrease in insulin-evoked phosphorylation of the insulin receptor along with significant reductions in both the insulin-evoked and basal phosphorylation states of other insulin signaling proteins including Akt, GSK-3β, and mTOR. Total IRS-1 levels were also found to be significantly reduced by loss of GLT-1 function. Insulin signaling changes induced by GLT-1 inhibition were reversed by scavenging of extracellular glutamate and inhibition of NMDA-type glutamate receptors. Collectively, these results suggest that loss of GLT-1 led to dyshomeostasis of glutamatergic signaling thereby disturbing insulin signaling in the brain, which was accompanied by deficits in cognitive function. Furthermore, these changes occurred early, similar to their appearance in AD. Thus, this study links two previously distinct components of AD, which may together play a role in AD pathogenesis