How does amyloid beta protein (A-beta) harm cells in Alzheimer’s disease? Why is this harm concentrated in certain brain regions? Researchers don’t know, but two studies published in the Oct. 12, 2010, edition of the Proceedings of the National Academy of Sciences suggest a plausible explanation.
According to the studies, the regions where A-beta deposits are seen in the brains of people with Alzheimer’s closely match the regions that normally rely heavily on less-efficient but faster processes of energy production in cells. The studies’ authors propose that A-beta in its disease-driving forms might impair these processes, and thus might principally harm the brain regions that most depend on them.
“It really opens up a lot of questions,” says Pierre Magistretti, a neurobiologist and brain metabolism researcher at the Ecole Polytechnique Federal in Lausanne, Switzerland, who wasn’t involved in the research. Magistretti is a vice-chairman of the European Dana Alliance for the Brain.
“It suggests that we should expand our view of what the cell-biological problem in Alzheimer’s might be,” says Marcus Raichle, a neurologist and neurobiologist at Washington University at St. Louis who was senior author of one of the two papers. Raichle is a member of the Dana Alliance for Brain Initiatives.
Glycolysis for speed
The basic energy-molecule used by living cells is adenosine triphosphate (ATP). Adult cells usually make it in a multi-step process that includes the simple sugar glucose and oxygen and leaves water and carbon dioxide as byproducts. But there are faster, less-efficient ways of turning glucose into ATP, and some cellular processes in the brain depend on them. These faster processes, which don’t require oxygen, account for only 10–15 percent the adult brain’s use of glucose and are used more extensively by fetal cells and cancer cells, and by muscle cells during intense exercise.Raichle has been researching brain metabolism for several decades, including developing functional imaging technology that tracks the brain’s use of glucose and oxygen. In a paper in Science in 1988, he and his colleagues found that these non-oxygen-consuming uses of glucose in the brain increase temporarily when brain activity increases. “The question of what is really going on there has been lingering in the back of my mind since then,” he says.
Several years ago, Raichle informally examined brain-metabolism data taken during functional imaging experiments and noticed that these alternative uses of glucose seemed to vary considerably from region to region in ordinary brains. Also piquing his interest was the observation that the regions that relied the most on these alternate energy processes appeared to be the ones that make up the “default mode network,” a set of brain regions that are relatively active when a person is not engaged in any specific task.
Raichle and his fellow metabolism researchers pioneered research on the default mode network, but in recent years Alzheimer’s researchers have taken an interest too, because the regions that make up the network are also the ones that gather the most A-beta deposits. That connection prompted Raichle and his colleagues to set up a formal set of studies.
In one study of 33 young adults, the researchers mapped the levels of these non-oxygen burning uses of glucose in resting brains, using positron emission tomography (PET) scans. They found that the levels of “aerobic glycolysis”—a catchall term for these alternative processes—did vary throughout the brain, and corresponded closely to the default mode network. In the second study, the researchers found a strong correspondence between the more aerobic-glycolysis-dependent regions in these 33 young brains and the regions that had accumulated signs of A-beta plaques in brain-imaging of 39 elderly people. The plaques largely spared the regions that showed average or below-average aerobic glycolysis—even if their overall energy use was high, such as in the visual cortex.
One possible explanation for the finding is that these alternate glucose uses are especially vulnerable to disruption by A-beta. For example, Raichle points out that glucose-fuelled processes are used by helper cells known as astrocytes to keep concentrations of the neurotransmitter glutamate below toxic levels in the synapses of cortical neurons. In principle, disruption of these processes by A-beta could lead to the deaths of the neurons. Magistretti’s group recently showed that A-beta in its disease-causing forms does alter the metabolism of astrocytes, apparently putting them under stress and ultimately weakening the neurons they are meant to protect. “One of the things that astrocytes do is to remove A-beta from the extracellular space, and then somehow they have to degrade it, but this is an extra burden for them,” says Magistretti.
Recent studies also have linked Alzheimer’s to G3PD, an enzyme that among other functions is needed for glucose-fuelled glutamate management: People with ordinary late-onset Alzheimer’s are more likely to have certain variants of G3PD, which may be less functional than normal; and A-beta clusters have been reported to cause G3PD to aggregate and become dysfunctional.
Magistretti suggests that one way to investigate further would be “to follow from an early age, in transgenic mice that overexpress A-beta, their glucose utilization and their A-beta deposits, and see how they develop over time.”
William Powers, a neurologist at the University of North Carolina whose own research has uncovered evidence of a similar abnormality in glycolysis in Huntington’s disease, suggests testing to see whether glucose use affects A-beta deposition rather than vice-versa. “This could be done in an animal model of Alzheimer’s by feeding a diet high in fat and low in carbohydrates, which will reduce glucose availability to the brain and decrease cerebral glucose metabolism,” he says.
