In the first study, the authors used positron emission tomography (PET) to measure the metabolic rates for oxygen (in three scans using [15O]-labelled water, carbon monoxide or oxygen) and glucose (using [18F]-labelled fluorodeoxglucose) in young adults. They found high rates of aerobic glycolysis in the prefrontal cortex, lateral parietal cortex, precuneus and posterior cingulate cortex, lateral temporal cortex, gyrus rectus, and caudate nuclei. These regions correspond with areas comprising the 'default mode network' — a set of regions that are most active when an individual is awake but not engaged in a task — and cognitive control networks. The cerebellum and the inferior temporal gyrus, including the hippocampus, showed low rates of aerobic glycolysis.
Amyloid-β plaques are found in regions of the default mode network early in Alzheimer's disease. In the second study, Raichle and colleagues therefore investigated whether the reliance on aerobic glycolysis in these areas is related to amyloid-β deposition. They used PET to assess amyloid-β deposition in patients with Alzheimer's disease and in cognitively normal people who were known to have elevated levels of the protein, and compared the resulting brain images with the spatial distribution maps of aerobic glycolysis levels from the first study. The authors divided the brain images into 167 non-overlapping cubes and, in a second approach, picked 60 small spherical areas in several regions of interest. For each cube and sphere, they assessed the levels of amyloid-β and aerobic glycolysis. Both methods revealed a striking spatial correlation between levels of amyloid-β and aerobic glycolysis, and this correlation was stronger in people with Alzheimer's disease than in cognitive normal people with elevated amyloid-β levels.
This is a preview of subscription content, access via your institution
