LOS ANGELES – Growing evidence from basic science and preclinical studies demonstrates that insulin plays a key role in brain synaptic function viability, vascular function, amyloid/tau regulation, and cerebral glucose metabolism.
In addition, brain insulin resistance in Alzheimer’s disease (AD) is associated with increased cerebral hyperglycemia, reduced cerebral glucose utilization, reduced blood flow, and reduced accumulation of amyloid and tau.
Metabolic interventions such as the administration of intranasal insulin and a modified ketogenic diet have beneficial effects on cognition and biomarkers of pathology and may be valuable therapeutic tools, Suzanne Craft, PhD, said at the World Congress on Insulin Resistance, Diabetes & Cardiovascular Disease. “Whether insulin is made in the brain de novo remains a matter of controversy, but it clearly crosses the blood-brain barrier every time that insulin is raised in the periphery,” said Dr. Craft, director of the Alzheimer’s disease core center at Wake Forest University, Winston-Salem, N.C. “It crosses the blood-brain barrier and binds to receptors distributed in the hippocampus, entorhinal cortex, and frontal cortex. Once it binds to its receptors, it carries out a plethora of activities: It modulates glucose utilization, neurotransmitter levels, neuronal firing, brain cholesterol metabolism, synaptic viability, and memory,” she said.
When that process goes awry, several pathologic processes linking insulin resistance (IR) and Alzheimer’s disease occur, including impaired proteostasis (oligomeric beta amyloid, microtubule-associated tau, and oligomeric insulin); hyperglycemia-induced toxicity and reduced glucose utilization; mitochondrial dysfunction; and vascular dysfunction, Dr. Craft continued.
In the past 5 years, researchers have developed ways to measure expression levels of insulin resistance markers like insulin receptor substrate 1 (IRS-1) pSer–positive neurons, Dr. Craft said. Others have found that increased IRS-1 pSer is associated with paired helical filaments (PHFs) tau in mild cognitive impairment (MCI) and AD, and that increased IRS-1 pSer in neutrally derived plasma exomes increased in AD years before onset and in type 2 diabetes. “We can look at the neurons themselves and what we see is that this IR marker colocalizes with tau,” Dr. Craft said. “It’s not very common in normal folks, but as you progress through the stages of MCI to AD, it becomes more common. So there’s a progressive increase in IR markers that associates with neurons in tau.” Other imaging studies have shown that homeostatic model assessment IR predicts gray matter atrophy, reduced blood flow, and amyloid deposition in middle-aged adults.
One way to overcome IR in AD is to increase insulin availability in the brain. Intranasal administration of insulin is a novel method being tested by Dr. Craft and her associates. “This is not inhaled insulin; it does not target the lungs,” she explained. “It’s insulin administered with a very specialized device that targets the olfactory cleft in the upper nasal passages. Virtually none of the insulin is deposited in the lungs or nasopharyngeally.” The approach is modeled on the notion that there are pathways from the olfactory perivascular spaces to the brain by which peptides can travel readily by bulk flow. “They reach the brain within minutes,” she said. “It’s a way of delivering peptides to the brain that bypasses the blood-brain barrier.”
In a published study, she and her associates randomized 104 adults with MCI or AD to receive 20 IU insulin, 40 IU insulin, or placebo twice daily for 4 months (Arch Neurol. 2012;61:29-38). Tests performed at baseline and at 4 months included cognitive evaluation based on story recall and the ADAS-Cog (Alzheimer’s Disease Assessment Scale–Cognitive subscale); function based on the Dementia Rating Severity Scale, FDG-PET (positron emission tomography with 18fluorodeoxyglucose), and cerebrospinal fluid biomarkers. “We showed that the 20-IU dose of intranasal insulin improved memory quite substantially (P less than .05),” Dr. Craft said. “It also improved glucose utilization as assessed by FDG-PET. We also saw changes in spinal fluid biomarkers of amyloid in a favorable direction. Most recently, we looked at the exosomal indicator of IR (IRS-1 pSer), and what we saw quite remarkably was a reduction in the same condition that the memory improved. This gives us hope that we have a marker of whether or not we’re having a positive impact.”
She and her research team recently finished a phase 2 pilot study of regular insulin vs. long-acting insulin detemir, to determine if a longer-acting agent with longer exposure would have greater efficacy (J Alzheimers Dis. 2017;57:1325–34). In all, 36 participants were randomized to receive placebo, 40 IU of insulin detemir, or 40 IU of regular insulin daily for 4 months, administered with a nasal delivery device. The investigators found that only the group treated with 40 IU regular insulin had better memory after 2 and 4 months, compared with placebo (P less than .03). Regular insulin treatment was also associated with preserved volume on MRI. “The normal pattern is for AD patients to lose brain volume rather rapidly,” Dr. Craft commented. “We see that abolished by the insulin treatment, which suggests to us that we’re able to stave off this disease-related mechanism.” She and her associates are currently conducting a phase 3 clinical trial with regular insulin and a phase 3 trial with rapid-acting insulin that are expected to be completed in the summer of 2018.
Dr. Craft spent the last few minutes of her presentation discussing the ketogenic diet as a nonpharmacologic approach to preventing or treating brain insulin resistance and AD. “I think the power of diet is underestimated, both in terms of causing disease and potentially modulating it,” she said. Her research team just completed a study of 87 middle-aged adults who were randomized to a Western diet or a healthy diet. The Western diet was high in saturated fat, sugar, and salt. The healthy diet was low in saturated fat, sugar, and salt, but the macronutrient composition of both diets was the same. “It was a eucaloric diet with normal calorie intake; no weight change, so trying to understand the integrated effect of the Western diet,” she said. “All food was prepared by us and delivered to the patients two times per week.” Patients with type 2 diabetes, patients with hypertension, and those who were on statins were excluded from the study.
The researchers observed pronounced diet-induced changes in cerebral blood flow, all which favored the healthy diet group. “The Western diet reduced blood flow, and the healthy diet increased blood blow in the hippocampus, which is critical for memory, as in some other regions that are known to be affected in AD,” Dr. Craft said. “We saw an effect on memory as well, with the healthy diet improving memory and the Western diet reducing it. Both of these effects were significant, so 4 weeks on a diet such as this is sufficient to modulate key aspects of brain function.”
More recently, Dr. Craft and her colleagues have been evaluating the effects of what they term the modified Mediterranean ketogenic diet (MMKD). “It does allow for higher carbohydrate consumption, compared with a traditional ketogenic diet, but they still have to stay under 10% a day,” she said. “We have an emphasis on healthy fats. We send everybody home with extra virgin olive oil. We think it gives us extra compliance and the potential for long-term nutrition.” She explained that the diet increases plasma and CNS ketone bodies, beta-hydroxybutyrate, acetoacetate, and acetone, which serves as preferred alternative fuel for the brain. “If the brain has a choice between glucose and ketones, it will choose ketones,” Dr. Craft said. “It can use them more easily.”
Ketone bodies are derived from hepatic fatty acid oxidation and readily diffuse across the blood-brain barrier into the brain. They are also synthesized in the brain by astrocytes, and they appear to have direct neuroprotective effects. “Ketone bodies may be beneficial because they may correct the hyperglycemic state and reduce glucose utilization in the brain in AD years prior to symptom onset,” Dr. Craft said. “They may correct neuronal hyperexcitability and preclinical seizures in presymptomatic and early stages of AD; they restore the balance between inhibitory and excitatory neurotransmitters like GABA [gamma-aminobutyric acid] and glutamate.”
In an unpublished, 16-week study, Dr. Craft and her associates randomized 16 patients to a Mediterranean ketogenic diet or to an American Heart Association low-fat diet. Lumbar punctures and brain imaging were performed before and after diet intervention. By the end of 6 weeks, they observed significant increases in ketones and in HDL cholesterol level in the MMKD group, compared with the AHA diet group, as well as significant decreases in trigylcerides and HbA1c level. “I would say that we improved the peripheral metabolic profile with the ketogenic diet,” Dr. Craft said. They also observed significant improvements from baseline in memory, spinal fluid AD biomarkers, and mitochondrial respiration.
“One of the things we’re appreciating is the role of insulin in a host of activities in the brain,” she concluded. “Disrupting those activities can have dire consequences on brain function that may lead to a neurological milieu that lends itself to pathological aging conditions like Alzheimer’s. Several large ongoing trials are poised to validate results of smaller studies, elucidate underlying mechanisms, and provide new therapeutic targets. It’s an exciting time.”
Dr. Craft’s research is supported by the National Institute on Aging and the Alzheimer’s Association Zenith Program. Intranasal delivery devices were provided by Kurve Technology.