Insulin resistant neurons arisen from peripheral hyperinsulinemia are senescent and correlate with memory impairment and cognitive decline: implications for Alzheimer's disease

[Background] The risk of developing Alzheimer's disease (AD) is almost double in patients with insulin resistance. Prolong hyperinsulinemia is a major cause of insulin resistance. We presented in last year that peripheral hyperinsulinemia and insulin resistance alters the biology and function in neurons at early stages. Mechanistically, we showed that failure of prompt glucose metabolism due to loss of hexokinase‐2 (HK2) impairs a multi‐step pathway that results in reactivation of cell cycle machinery, as well as accumulation of signals of phosphorylated tau and amyloid. Despite all these cellular findings, it is important to address these series of events in vivo, and their consequences on animal behavior. [Methods] Assays including behavioral analyses in young (3‐month) versus aged (22‐44 month) wild type mice (N=24) were performed. Stereotaxic injections of adeno‐associated virus (AAV) that carries shRNA against HK2 were performed in hippocampal CA1 region of young non‐insulin resistant wildtype mice. Behavioral, immunohistochemistry and a variety of cellular and biochemistry methods were performed to analyze changes occurred in these animals. [Results] Here we report that insulin resistant neurons eventually become senescent, a consequence of aberrant reactivation of cell cycle machinery. At behavioral level, insulin resistant animals showed cognitive decline and memory impairment, both were inter‐correlated to the severity of peripheral insulin resistance and the senescent load in the brain. Mechanistically, the importance of neuronal glycolysis was validated in vivo. Similar behavioral changes were recapitulated in young non‐insulin resistant animals with HK2 knockdown in CA1 neurons. Notably, these neurons also showed aberrant cell cycle re‐activation and expressed a senescent phenotype. [Conclusions] These findings therefore extended our understanding on how peripheral hyperinsulinemia and insulin resistance alters the biology and function of the brain, particularly at behavioral levels. At basic science level, our data also offer a mechanism on how post‐mitotic neurons could become senescent—an outcome of aberrant cell cycle reactivation. At translational level, this study re‐emphasizes that metabolic syndrome like peripheral insulin resistance is a modifiable risk factor for AD. Better management of insulin resistance is therefore a important strategy to prevent related neurodegeneration events and potential initiation of AD.