Alzheimer's disease (AD) is the most common cause of dementia, accounting for 60–80% of cases, although there is growing awareness that AD is often mixed with other dementia causes. At present, approximately 33.9 million people worldwide have AD, and the prevalence is expected to triple over the next 40 years. Recent increasing evidence has suggested that diabetes mellitus in the elderly is a risk factor for cognitive impairment and dementia. Many epidemiological studies have shown that the prevalence of not only vascular dementia (VD), but also AD in diabetic older adults is increased as compared with control groups. In several population-based cohort studies, relative risk (RR) for VD is estimated to be 1.8~3.4, whereas RR for AD is estimated to be 1.4~2.3 in diabetic patients. From accumulated evidence, Biessels et al. reviewed the risk of dementia in diabetes mellitus, indicating that atherosclerosis such as brain infarction, microvascular disease due to insidious ischemia, advanced protein glycation and oxidative stress due to glucose toxicity, and insufficient insulin action were major factors, and the added involvement of aging and genetic factors leads to dementia[1]. However, the precise mechanisms, especially the association between diabetes mellitus and AD, have not been elucidated.

One attractive mechanism reported by Craft et al. is that hyperinsulinemia caused by insulin resistance seems to be associated with the pathology of AD[2]. Peripheral hyperinsulinemia is induced to downregulate the transfer of insulin through the blood–brain barrier, then insulin levels are decreased in the brain. Because insulin receptors, located in astrocytes and neuronal synapses, are highly concentrated in the olfactory bulb, hypothalamus and hippocampus, brain insulin signaling is particularly important for learning and memory, therefore, suggesting that insulin resistance might contribute to cognitive deficits in AD. Among many ongoing clinical therapeutic strategies for AD, both insulin-sensitizing compounds to improve insulin resistance and intranasal administration of insulin to accumulate insulin in the brain seem to be effective in improving cognitive function.

Currently available evidence strongly supports the position that the initiating event in AD is related to abnormal processing of amyloid-β peptide (Aβ), ultimately leading to formation of Aβ plaques in the brain. Recent studies have shown that soluble oligomers of the Aβ (AβOs) induces AD-like pathology, including neuronal tau hyperphosphorylation, oxidative stress, synapse deterioration and inhibition of synaptic plasticity. Thus, AβOs are small, diffusible aggregates that accumulate in the AD brain, and are recognized as potent synaptotoxins. In a more recently published study, Bomfim et al. examined the mechanism of dysregulated insulin signaling and the protective effect of an antidiabetic agent on AD-associated AβOs in The Journal of Clinical Investigation in April 2012[3]. In both experimental conditions of cultured hippocampal neurons and intracerebroventricular injection of AβOs in monkeys, Bomfim et al. showed elevated levels of serine phosphorylation of insulin receptor substrate-1 (IRS-1) and activated Jun N-terminal kinase (JNK), as shown in brain tissue from humans with AD. They also found that AβOs activated the JNK/tumor necrosis factor (TNF)-α pathway, induced IRS-1 phosphorylation at multiple serine residues and inhibited physiological IRS-1 phosphorylation at tyrosine residue, an essential step in the IR-stimulated signaling pathway.

Previous studies have linked IRS-1 serine phosphorylation to JNK activation in type 2 diabetes and in obesity-related insulin resistance. Furthermore, JNK activation is known to be stimulated by TNF-α, and TNF-α levels are elevated in AD. Bomfim et al.'s [3] present results clearly showed that AβOs-induced elevation in pro-inflammatory TNF-α levels triggers aberrant activation of JNK and ultimately serine phosphorylation of IRS-1. In an additional set of experiments, Bomfim et al. [3] also showed that the involvement of double-stranded ribonucleic acid-dependent protein kinase (PKR) and IκB kinase (IKK)which are two stress-sensitive kinases that mediate serine phosphorylation of IRS-1 and are critical regulators of peripheral insulin resistance, were also activated by AβOs (Figure 1a).


Figure 1. Proposed mechanism underlying disrupted brain insulin signaling in Alzheimer's disease (AD). (a) Oligomers of amyloid-β peptide (AβOs) stimulate tumor necrosis factor (TNF)-α signaling, which activates the Jun N-terminal kinase (JNK) pathway and, possibly, ribonucleic acid-dependent protein kinase (PKR) and IκB kinase (IKK) pathways. Activation of these stress-sensitive kinases results in serine phosphorylation of insulin receptor substrate-1 (IRS-1) and blocks downstream insulin signaling. (b) Stimulation of insulin and glucagon-like peptide 1 receptors (GLP1R) blocks AβOs-induced defects in insulin signaling. In both panels, red arrows indicate inhibitory pathways and green arrows indicate stimulatory pathways of insulin signaling. Reproduced from Bomfim et al.[3] with permission from the American Society for Clinical Investigation. AKT, protein kinase B (AKT/PKB); Ex-4, exendin-4; I, insulin; IR, insulin receptor; P, phosphate; PI3K, phosphoinositide 3-kinase; pSER, serine phosphate; pTyr465, tyrosine465 phosphate; TNFR, TNF receptors.

Download figure to PowerPoint

Stimulation of brain insulin signaling has been suggested as a promising approach to prevent memory decline in AD. Bomfim et al. [3] next examined the effects of insulin and exendin-4 (exenatide), an incretin hormone analog that activates the insulin signaling pathway through glucagon-like peptide 1 receptor (GLP1R) stimulation. GLP1Rs are present and functional in the human brain, and emerging evidence indicates that GLP1R stimulation regulates neuronal plasticity and cell survival[4]. The results showed that both insulin and exendin-4 prevented the increase in IRS-1 serine phosphorylation and the decrease in IRS-1 tyrosine phosphorylation induced by AβOs (Figure 1b). Exendin-4 might be more effective than insulin in protecting neurons from AβOs, as brain insulin signaling can decline with aging and AD.

It was finally found that systemic administration of exendin-4 decreased levels of hippocampal IRS-1 serine phosphorylation and activated JNK, and improved behavioral measures of cognition in AD transgenic mice. Based on these results, it is likely that enhancing brain signaling through the use of exendin-4 or another GLP1R agonist, rather than insulin, might be a key alternative to blocking insulin resistance and memory impairment in AD.

Because cognitive impairment is a most crucial factor that causes frailty in diabetic older adults, diabetic control is very important in preventing cognitive decline, as well as vascular events. In order to investigate adequate levels of blood glucose control in diabetic older adults, a large-scale randomized controlled trial, the Japanese elderly diabetes intervention trial (J-EDIT), examined the 6-year decline related to cognitive function among diabetic older adults aged 65–85 years. A current analysis has shown that low high-density lipoprotein cholesterol and higher diastolic blood pressure are significantly associated with cognitive decline in older diabetic patients. A higher glycated hemoglobin level had a tendency toward an association with cognitive decline[5]. Thus, the risk factors and mechanisms that drive the association between diabetes and accelerated cognitive decline and dementia need to be further examined. However, comprehensive management of diabetes, including dyslipidemia and hypertension, might contribute to the prevention of decline in cognitive function in older diabetic patients.


  1. Top of page
  2. References