A panel of 26 metabolites that appear in the blood and brains of people with Alzheimer’s disease point to a potential new biomarker path to be pursued as future therapeutic targets.
A team of researchers from the NIA’s Intramural Research Program (IRP), led by Dr. Madhav Thambisetty, Chief of the Clinical and Translational Neuroscience Unit of the NIA IRP, sought to clarify, with the use of advanced machine learning techniques, if brain metabolites associated with Alzheimer’s pathology could also be detected in blood and were related to progression of the disease.
Their results were published in the January 25, 2018, issue of PLoS Medicine. Previous biomarker studies for Alzheimer’s disease had not connected markers in the blood to those in the brain, as scientists have been trying to learn more about the long timeline between the start of Alzheimer’s brain pathology and the development of outward symptoms.
To attack this problem, first author Dr. Vijay Varma and colleagues performed metabolic analyses of brain and blood tissue samples from autopsies of three groups of participants in the Baltimore Longitudinal Study of Aging (BLSA): people who had Alzheimer’s disease, a normal control group, and a smaller group who had asymptomatic Alzheimer’s disease, meaning they had significant disease pathology at autopsy but showed no signs of cognitive impairment during life.
The team employed a computer system with machine learning techniques, which allowed them to use algorithms to more rapidly analyze metabolites and narrow them down to the most likely targets. With this approach, they used data patterns to identify a panel of 26 metabolites out of 180 studied that could accurately differentiate Alzheimer’s disease from control brain samples.
They then measured the same 26 metabolites in about 700 blood samples from the Alzheimer’s Disease Neuroimaging Initiative and the BLSA, testing for relationships with known signs of Alzheimer’s, such as brain shrinkage on MRI scans, cerebrospinal fluid measures of abnormal proteins linked to Alzheimer’s, and cognitive test results.
The researchers found that altered blood concentrations of some of the 26 metabolites were consistently associated with brain atrophy, cerebrospinal fluid measures of Alzheimer’s pathology, cognitive performance and risk of Alzheimer’s disease before established symptoms developed.
The analysis zeroed in specifically on sphingolipids - common parts of cell membranes known to play a part in a variety of human diseases. These were found to be connected to several known Alzheimer’s disease biological mechanisms, such as tau phosphorylation (the precursor of tau tangles in the brain), beta-amyloid metabolism, problems with calcium regulation in neurons, production of the neurotransmitter acetylcholine, and brain cell death. While sphingolipids emerged from this study as one of the major types of metabolites for further exploration, the researchers also want to do more in-depth analyses of other metabolite classes.
Reference:
Varma VR, et al. Brain and blood metabolite signatures of pathology and progression in Alzheimer disease: A targeted metabolomics study. PLoS Medicine. 2018 Jan 25;15(1):e1002482.