The root of Alzheimer's disease, most researchers agree, is the accumulation of certain proteins (amyloid and tau) in the brain. But there is mounting evidence that the hormone insulin has an important role as well. Evidence for the connection includes the bidirectional association between type 2 diabetes (T2D) and AD, and the increased risk of AD within 3 years of diagnosed insulin resistance.
In a recent symposium at the New York Academy of Science, speakers agreed on the link between insulin resistance and Alzheimer’s, but beyond that, consensus was elusive. Is AD essentially “diabetes of the brain,” in which glucose utilization is impaired—or is it more a question of insulin signaling processes not directly related to energy management?
For example, insulin and amyloid beta (Aβ) are degraded by the same enzyme, and either too much or too little insulin in the brain could reduce amyloid clearance and promote its accumulation, several speakers suggested.
Their discussions of the mechanisms, molecules involved, and implications for intervention were sometimes complementary, often at odds. And a recurring motif was the possibility that drugs already approved for diabetes might be beneficial in AD.
Konrad Talbot of University of Pennsylvania described the core features of AD as a progression from accumulating soluble amyloid beta to cognitive decline, through steps including glial activation, release of cytokines, synaptic dysfunction, tangle formation, and neuronal death.
Insulin is in the midst of these changes, he said. “Virtually all the functions disrupted in AD are affected by insulin signaling … amyloid beta clearance, cell survival, inflammatory control, lipid metabolism, vascular function, and synaptogenesis and plasticity,” Talbot said. “Anything that can interrupt insulin signaling can contribute to the pathology associated with AD.”
Much of the insulin involved in these processes is produced in the brain itself, not the pancreas. “There seems to be a separate pool… You can decrease or increase plasma insulin [in the body] by enormous amounts, and have no effect on levels of insulin the brain,” he said. “The insulin resistance I’ll be telling you about is to local insulin.”
It is not clear what role amyloid beta plays in brain insulin resistance, but Talbot described research indicating a process that short-circuits the hormone’s effects within the neuron. Binding to its receptor, insulin activates a signaling molecule, insulin receptor substrate-1 (IRS-1), that starts a cascade of intracellular events. Feedback mechanisms in the cell limit IRS-1 activity through phosphorylation. Phosphorylation of IRS-1 is amplified in AD, presumably by amyloid, making neurons effectively unresponsive to insulin, Talbot said.
In post-mortem analysis of Alzheimer’s brain tissue, elevated levels of phosphorylated IRS-1 were strongly inversely correlated with patients’ episodic memory scores. “It accounted for 46% of variance,” he said.
Talbot suggested the possibility of mitigating brain insulin resistance and AD pathology with a glucagon-like peptide-1 analogue, such as the drug liraglutide used for diabetes. In mouse models of the disease, the drug reduced plaque load and microglial activation, promoted neurogenesis, and improved memory. Levels of phosophorylated IRS-1 normalized, suggesting the drug restored brain insulin signaling, he said.
Ewan C. McNay, of University at Albany of the State University of New York, agreed with Talbot on the importance of insulin-amyloid interactions in AD, but argued that glucose is very much involved. He cited evidence that insulin acts a “cognitive modulator,” by enabling neuronal glucose utilization: Experiments have shown that glucose metabolism in the brain rises when rats perform cognitive tasks, and adding insulin enhances both metabolism and performance. Blocking insulin by introducing an anti-insulin peptide intrahippocampally impairs their memory.
According to McNay, amyloid has an acute effect on insulin signaling: Injecting soluble amyloid intrahippocampally abolished the cognitive task-associated glucose utilization increase, and impaired the rats’ performance on the task.
The interference with glucose metabolism appears to involve the glucose transporter GluT4, which responds to insulin by rising to the cell membrane to bring glucose in. In the presence of amyloid, GluT4 remains in its storage vesicles, McNay said.
“Both insulin and beta amyloid regulate glucose metabolism, and interact with each other,” he said. “At the center of these interactions is a fourth molecule: GluT4. It could be a potential target for therapeutic approaches.”
Suzanne M. de la Monte, of Brown University, stressed the complexity of processes linking insulin resistance and AD, and suggested a prominent role for an additional set of molecules, toxic lipid ceramides.
AD is more than plaques and tangles: Like other neurodegenerative diseases, it entails a scarring process that affects both grey and white matter. In this neurodegenerative cascade, de la Monte said, ceramides may be important.
She noted an association between peripheral insulin resistance, as seen in obesity, metabolic syndrome, and type 2 diabetes, and the production of ceramides, which promote inflammation, oxidative stress, and cell death, as well as inhibit insulin signaling. These toxic lipids can cross the blood-brain barrier; the brains of people with AD also produce their own ceramides.
“I put insulin resistance at the center, but we can’t just talk about insulin resistance: Once the cascade gets going, it’s a vicious cycle of inflammation and oxidative stress,” she said. Insulin resistance in the rest of the body and in the brain are associated, but not yoked together, she observed. “AD occurs in people without diabetes, and vice versa,” she said. “But I think T2D is pushing AD rates like crazy: Patients with low-level AD may be pushed to clinical disease.”
New targets for therapy
José A. Luchsinger, of Columbia University Medical Center, cited data from a longitudinal study of older adults in Northern Manhattan showing an association between high blood levels of insulin—a common manifestation of peripheral insulin resistance in connection with early type 2 diabetes or obesity—and dementia some years later.
One possible mechanism, he suggested, is that elevated peripheral insulin could, via feedback at the blood-brain barrier, reduce brain levels of the hormone; this would downregulate insulin-degrading enzymes, allowing amyloid to accumulate.
The drug metformin, which is widely prescribed for type 2 diabetes to reduce glucose and serum insulin levels, might be useful in slowing Alzheimer’s progression.
In a pilot study, people with mild cognitive impairment were given metformin or placebo for 12 months. Although results are as yet unpublished, Luchsinger said that they were encouraging: Significant improvement in memory in the metformin group, and a trend toward increased glucose uptake in brain areas involved in AD. Phase III trials and further investigation into possible mechanisms are needed, he said.
William H. Frey, II, of the Alzheimer’s Research Center, Regions Hospital, St. Paul, MN, proposed delivering insulin directly to the brain—intranasally. Small molecules travel to the brain from the nasal cavity via the olfactory and trigeminal nerves, bypassing the blood-brain barrier, he said. In animal studies, intranasal drugs have been shown to reduce brain injury after experimentally induced stroke, and to improve memory to youthful levels in aged rodents with experimentally-induced diabetes.
Frey described a series of small studies where intranasal insulin improved memory in people with AD or mild cognitive impairment, without changing serum glucose or insulin. In the most recent, which involved 104 patients, glucose uptake and utilization increased in various brain regions along with memory and cognitive function gains. Larger clinical trials are coming soon, he said. [Frey holds a patent for intranasal insulin.]
“Theoretically, if patients are given intranasal insulin or a drug to increase insulin sensitivity at the first sign of reduced glucose uptake and utilization, it might delay or prevent onset and progression of the disease,” he said.