Since the discovery of insulin in 1922 (1), the skeletal muscle, adipose tissue, and liver are traditionally regarded as key insulin-sensitive organs (2). Evidence of insulin actions in the brain emerged more recently, approximately 35 years ago, with studies showing that the hypothalamic actions of insulin regulate peripheral energy homeostasis (reviewed in [3]). Interestingly, insulin receptors were also detected in high densities in other brain regions (4), suggesting that insulin’s role in the central nervous system (CNS) extends beyond hypothalamic control of energy homeostasis.
The hippocampus, a brain region key to memory and learning, was found to present particularly high levels of insulin receptors, suggesting that insulin could play a role in synaptic plasticity mechanisms and memory formation in rodents and humans. Indeed, studies using in vitro and in vivo experimental models indicated that insulin regulates neuronal survival, acts as a growth factor, and regulates circuit function and plasticity (reviewed in [5]). In cultured hippocampal neurons, insulin receptors present a punctate dendritic distribution that is consistent with the presence of insulin receptors at synaptic compartments (6). In harmony with the proposed role of insulin signaling in cognition, intranasal insulin treatment—a method achieving direct CNS delivery of insulin without invasiveness or major complications (7)—improves memory in healthy adults, without changing blood levels of insulin or glucose (reviewed in [8]). While these recent studies and others have started to unravel the actions of insulin signaling in the brain, the precise actions of insulin in brain areas responsible for memory still remain underexplored.
A recent study provided an important advance in this field by showing that specific deletion of brain insulin receptors in mice (NIRKO mice) led to augmented anxiety and depressive-like behavior (9). NIRKO mice further presented brain alterations in mitochondrial morphology and function, elevated oxidative stress, and decreased dopamine signaling (9). While an early study found that NIRKO mice do not show major learning and memory impairment (10), the study by Kleinridders et al. (9) demonstrated that deletion of insulin receptors led to important defects in brain function. Therefore, further studies contributing to the understanding of the role of insulin signaling in memory are definitely warranted.
In this issue of Diabetes, Grillo et al. (11) present their study of the role of hippocampal insulin receptor signaling in neuroplasticity in rats by using a lentiviral vector expressing an insulin receptor antisense sequence. Using this approach, the authors were able to downregulate insulin receptors in the hippocampus without affecting peripheral glucose homeostasis, thereby generating a rat model of hippocampal-specific impaired insulin signaling. Interestingly, they showed that these rats exhibited impaired hippocampal synaptic plasticity and hippocampus-dependent spatial learning. Furthermore, hippocampal levels of GluN2B subunit and phosphorylated GluA1 were reduced, providing molecular clues on how the deficits in synaptic transmission develop without proper hippocampal insulin signaling. While levels of insulin receptors, insulin-stimulated phosphorylation of the insulin receptors, and phosphorylated Akt were decreased in the hippocampus of the lentiviral-exposed rats, it remains to be established if other key components of the insulin signaling pathway are also compromised in this model.
Thus, while the brain had been once considered an insulin-insensitive organ, it is now clear that it is an important target for insulin actions. It is crucial to identify in detail the signaling pathways used by insulin in the brain, as failure of those signals has been associated with brain disorders, including Alzheimer disease (AD) (12). It is further possible that brain insulin resistance is a mechanism leading to cognitive decline in patients with diabetes, who are at a higher risk of developing dementia and AD (13). In fact, patients with type 1 and 2 diabetes show structural brain alterations and cognitive impairment (14), and it is becoming increasingly clear that diabetes is a threat to the brain. The study by Grillo et al. (11) contributes to the understanding of how hippocampal insulin resistance may negatively influence brain function and may be linked to cognitive decline (Fig. 1).
Proposed role of neuronal insulin signaling in synaptic plasticity and memory. This simplified scheme illustrates some of the mechanisms by which impaired insulin signaling impacts synapse plasticity and function. A: In physiological conditions, insulin receptors (InsRs) are present at synapses, and phosphorylation (P) of the InsRs and Akt reflects proper insulin signaling, which is linked to phosphorylation of GluA1 and presence of GluN2B at synapses, favoring synapse function and memory formation. B: In pathological conditions (e.g., diabetes, AD), decreased levels of InsRs and phosphorylation of the InsRs and Akt suggest reduced insulin signaling, which may be linked to decreased levels of GluN2B and GluA1 phosphorylation at synapses. Reduced insulin signaling ultimately leads to impaired synapse plasticity and function.
Proposed role of neuronal insulin signaling in synaptic plasticity and memory. This simplified scheme illustrates some of the mechanisms by which impaired insulin signaling impacts synapse plasticity and function. A: In physiological conditions, insulin receptors (InsRs) are present at synapses, and phosphorylation (P) of the InsRs and Akt reflects proper insulin signaling, which is linked to phosphorylation of GluA1 and presence of GluN2B at synapses, favoring synapse function and memory formation. B: In pathological conditions (e.g., diabetes, AD), decreased levels of InsRs and phosphorylation of the InsRs and Akt suggest reduced insulin signaling, which may be linked to decreased levels of GluN2B and GluA1 phosphorylation at synapses. Reduced insulin signaling ultimately leads to impaired synapse plasticity and function.
Impaired insulin sensitivity has been linked to cognitive deficits and structural and functional brain abnormalities in the elderly (15,16). Furthermore, lower levels of insulin and insulin receptors and altered levels of different components of the insulin signaling pathways were described in AD hippocampi, indicating that a scenario of brain insulin resistance develops in AD (17). Hippocampal insulin signaling impairments were also described in animal models of AD (5). As a result of these and other studies, defective brain insulin signaling is now regarded as an important characteristic of AD pathology. It is interesting to note that similar mechanisms leading to peripheral insulin resistance and overall health decline in type 2 diabetes have been found to cause brain insulin resistance and memory impairment in AD. Linking pathogenic mechanisms in the AD brain to mechanisms present in metabolic diseases provides a rationale for using antidiabetes agents as novel therapeutics in AD and as a strategy to prevent cognitive decline in diabetes. Supporting this view, recent studies in mice and monkey models of AD demonstrated that the use of exendin-4 and liraglutide, antidiabetes agents, protects the brain from defective insulin signaling and endoplasmic reticulum stress (18,19).
Collectively, the elegant study by Grillo et al. (11) supports the emerging notion that impaired insulin signaling in the brain plays a central role in synaptic plasticity and learning. The challenge now is to identify in detail the molecular mechanisms underlying brain insulin resistance and to determine if restoring brain insulin signaling might be helpful to circumvent synapse deterioration and memory failure in AD and related disorders.
See accompanying article, p. 3927.
Article Information
Funding. F.G.D.F. is supported by grants from the Human Frontier Science Program and the John Simon Guggenheim Memorial Foundation; the Brazilian funding agencies Conselho Nacional de Desenvolvimento Científico e Tecnológico and Fundação Carlos Chagas Filho de Amparo à Pesquisa do Estado do Rio de Janeiro; and the Canadian Institutes of Health Research (operating grant MOP-38854). C.B. is supported by grants from the Novo Nordisk Foundation and the Swedish Brain Foundation.
Duality of Interest. No potential conflicts of interest relevant to this article were reported.