Subscribe to eTOC

In the Pipeline-Alzheimer's Disease
New Evidence for Brain Glucose Dysregulation in Alzheimer's Disease: Abnormalities Evident Years Before Clinical Symptoms


Based on an analysis of postmortem tissue samples, researchers found that higher brain tissue glucose concentrations, reduced glycolytic flux, and lower protein levels of neuronal glucose transporters were associated with the severity of Alzheimer's pathology and symptoms.

Abnormalities in the brain's glycolytic and glucose transport systems may signal early pathological signs of Alzheimer's disease (AD) years before the onset of clinical symptoms, according to an analysis of postmortem tissue from a long-running study of community-dwelling participants in Baltimore.

The study, published online ahead of print on October 19 in the Alzheimer's & Dementia journal, found that higher brain tissue glucose concentrations, reduced glycolytic flux, and lower protein levels of neuronal glucose transporters were associated with the severity of AD pathology and symptoms.

The findings provide evidence that brain glucose dysregulation may be “intrinsic to the pathogenesis” of AD, said the senior study author Madhav Thambisetty, MD, PhD, investigator and chief of the unit of clinical and translational neuroscience in the laboratory of behavioral neuroscience at the National Institute on Aging. These data set the stage for future studies of “therapies that target the glycolytic pathway to prevent or slow the pathological process and reduce the risk of developing symptoms,” he said.

Dr. Thambisetty and colleagues noted that earlier studies have shown abnormalities in insulin signaling machinery in AD patients compared to controls. But although imaging studies have shown reduced brain glucose uptake in regions vulnerable to AD pathology, it has not been clear whether the problems with regulating brain glucose metabolism is a key pathogenic factor in AD or whether abnormalities of brain glucose homeostasis in AD are related to peripheral glucose concentration.

To address these questions, Dr. Thambisetty and his colleagues turned to the Baltimore Longitudinal Study of Aging (BLSA), which has since 1958 collected medical and health data on Baltimore residents, and since 1986, collected postmortem tissue samples.


For the current study, the investigators conducted quantitative metabolomics assays of frozen brain tissue samples from 43 BLSA participants: 14 patients with AD, 14 controls, and 15 asymptomatic AD patients (those with brain pathology but no clinical disease during their life).

They looked at glucose levels in three different brain regions: the middle frontal gyrus that is vulnerable to amyloid, the inferior temporal cortex that is vulnerable to tau and neurofibrillary tangles, and the cerebellum that served as a control region because it seems resistant to the classical AD pathology.

The researchers found that brain glucose levels were significantly higher in the middle frontal gyrus and the inferior temporal cortex in patients versus controls. Those with no clinical signs of AD but who had amyloid and tau pathology fell somewhere in the middle. There were no differences in glucose concentrations in the cerebellum between the three groups.


DR. MADHAV THAMBISETTY said the findings provide evidence that brain glucose dysregulation may be “intrinsic to the pathogenesis” of AD.

The brain tissue glucose concentrations also correlated with severity in both the plaque and tangle burdens.

Were the AD brains having a harder time breaking down glucose? To answer this question, the investigators looked at the glycolytic pathway itself, which breaks down glucose to pyruvate, the metabolite used to form ATP, the energy that fuels the brain. They found the lowest rates of glycolysis in the AD patients and the highest rates in controls. Those with AD pathology but no clinical signs of cognitive problems again fell somewhere in the middle.

The researchers analyzed protein levels of two primary glucose transporters in the brain, neuronal glucose transporter-3 (GLUT3) and glucose transporter-1 (GLUT-1). These transporters work in astrocytes (GLUT1) and also within the vascular endothelial cells of the blood-brain barrier (GLUT3). They reported that GLUT3 protein levels were lower in AD and asymptomatic AD patients compared to controls. There were no differences in protein levels of GLUT1 in the three groups.

The conclusion, said Dr. Thambisetty, is that “neurons are unable to break down glucose through glycolysis.”

The researchers also looked at 445 plasma-fasting glucose levels collected over a 19-year period in the 43 participants to determine whether these levels were related to brain tissue glucose concentrations. The average time between death and the last measurement was about five years. Interestingly, they did find a strong relationship to the last fasting blood glucose and the brain glucose levels at death in all three brain regions.

“Higher levels as well as greater increases over time in plasma-fasting glucose are associated with higher brain tissue glucose concentrations,” the authors wrote in the study. There was, however, no relationship between the higher plasma fasting glucose concentrations and AD pathology noted at death, they said.

The scientists are now designing studies to identify triggers for these glycolytic defects. They are looking at the enzymatic steps in this pathway to assess whether they can uncover genes or environmental factors that alter glycolysis.

Dr. Thambisetty added that it is premature to change the recommendations for AD treatments or prevention. “I will not tell my patients anything different than what I tell them now: Control blood sugar, exercise regularly, and eat healthy,” he said.


“This is an exceptional study,” said George Perry, PhD, dean of the College of Sciences and professor of biology at the University of Texas in San Antonio. “The link between metabolism and Alzheimer's disease has been growing stronger with observational, genetic, health disparity, and lifestyle intervention studies.”

He noted that almost two decades ago, scientists led by Suzanne de la Monte, MD, MPH, professor of pathology and laboratory medicine at Brown University, coined the term type 3 diabetes for the unique glucose utilization abnormalities seen in AD.

“Since then, unraveling the nature of that abnormality has gained attention, but has not revealed its exact mechanism,” Dr. Perry said. “This study advances the field significantly by examining a well-controlled autopsy sample, and showing that reduced glycolytic and glucose transport activity is an early change of AD. These findings clearly place metabolic abnormality as a primary driver of AD.”

Brown University's Dr. de la Monte said “the study nicely demonstrates impaired glucose uptake in AD brains, corresponding with FDG-PET studies demonstrating reductions in brain glucose metabolism from the early stages of neurodegeneration. The authors emphasize the alterations in GLUT expression; GLUTs regulate glucose uptake in most cells. Since glucose is the overwhelmingly dominant fuel for brain metabolism, failure of neurons to transport glucose essentially causes them to starve. In the process, the cells undergo oxidative stress, membrane injury, and compromise of various functions including memory, plasticity, neurotransmitter function, cell survival, and cellular homeostasis. The overall process can trigger inflammation, further stress, injury, and death of cells.” She explained that GLUTs are regulated by insulin, and in insulin-resistant states, GLUT expression and function are impaired.

Although the authors did not specifically investigate brain insulin resistance, their data illustrate the effects, she added. “A major strength of their work was to demonstrate that the key fuel, glucose, was not being utilized — this is the same problem that occurs in type 2 diabetes. People with type 2 diabetes have at least a two-fold higher risk of developing cognitive impairment or AD. Altogether, this is very exciting because it provides further evidence that brain metabolic dysfunction related to glucose utilization begins prior to the onset of AD, and further suggests specific treatment and prevention strategies for cognitive impairment and AD.”

Suzanne Craft, PhD, professor of medicine and director of the Alzheimer's Disease Core Center and co-director of the J. Paul Sticht Center for Healthy Aging and Alzheimer's Prevention at Wake Forest School of Medicine, noted that study participants “were very well-characterized with respect to cognition and with respect to peripheral glucose levels.”

But, she said, while the study addresses “underlying mechanisms that might be associated with reduced glucose utilization, such as reduced activity of glycolytic enzymes, and reduced levels of glucose transporters, we still don't know if the AD pathology causes the reduced glucose utilization or vice versa, or if some third pathology causes both.”

“I read this paper with interest, because glycolysis has been a major interest of mine,” said Marcus E. Raichle, MD, PhD, a professor in the department of radiology, neurology, neurobiology, and biomedical engineering at Washington University School of Medicine. In a study, published in the Proceedings of the National Academy of Sciences, he and colleagues found that the area of the brain that is very glycolytic is in the default mode network, the area that his group initially described. It is one of the earliest networks to be damaged in AD, he said, adding: “This glycolysis pathway is definitely not working right.” “How does this fit into our understanding of why someone gets Alzheimer's? That's the $64,000 question,” he said. “We've documented an important pathway and now we need to accumulate more information to prove that it is right.”

Russell H. Swerdlow, MD, professor of neurology, molecular & integrative physiology, and biochemistry & molecular biology at the University of Kansas School of Medicine and director of the University of Kansas Alzheimer's Disease Center, has also been interested in energy networks in AD. “Glucose flux is an important component of that,” he said. “There's got to be something fairly fundamental that is not brain-limited. We argue that it is energy metabolism. This study is further proof that there is a problem with energy metabolism.”


•. An Y, Varma VR, Varma S, et al. Evidence for brain glucose dysregulation in Alzheimer's disease Alzheimers Dement 2017; Epub 2017 Oct 19.