Researchers find evidence connecting a specific blood protein to brain tissue loss

Higher levels of a blood protein called growth differentiation factor 15 during midlife were associated with approximately twice the rate of dementia over the following twenty years. This protein is released by cells during aging, inflammation, and other forms of biological stress. The research, published in Science Advances, provides evidence linking this protein to future cognitive decline.

Previous research identified the protein as a possible early warning sign for dementia. However, scientists were uncertain whether it merely reflected poor health or actually contributed to damaging brain processes. To find out, researchers investigated whether blood levels of this protein predicted different forms of dementia. They also examined whether it was connected to brain changes, inherited risk, and immune activity.

Led by Cassandra O. Blew at the US National Institutes of Health, the team combined evidence from six large studies that tracked participants over time. The principal midlife analysis included 11,595 American adults with an average age of fifty-seven. These participants were followed for about twenty years. The scientists also analyzed data from older American adults, British participants, and Icelandic adults, with average ages ranging from sixty to seventy-six.

The researchers measured the protein in blood samples and examined subsequent diagnoses of all forms of dementia, including Alzheimer’s disease and vascular dementia. The statistical models accounted for factors like age, sex, kidney function, smoking, obesity, diabetes, and genetic susceptibility. Results revealed that in the midlife American sample, each doubling of the protein’s levels was associated with a 55 percent greater risk of developing dementia.

Among participants with protein levels in the top half of the group, 7.5 percent developed dementia over twenty years, compared with 3.9 percent of those in the bottom half. A similar pattern emerged in the older American group over a seven-year period. In that older group, 18.7 percent of individuals with high levels developed dementia, compared to 9.5 percent of those with lower levels.

Free daily newsletter

The relationship was stronger for vascular dementia than for Alzheimer’s disease. Vascular dementia develops through reduced blood flow and damage to the brain’s blood vessels, while Alzheimer’s is associated with abnormal protein buildups. In the British sample, each doubling of the protein was associated with a 101 percent higher vascular dementia risk but only a 20 percent higher Alzheimer’s risk. The Icelandic group produced similar estimates, showing a 106 percent higher risk for vascular dementia compared with a 24 percent increase for Alzheimer’s.

Brain scans provided matching evidence. Higher levels of the protein were associated with smaller brain volumes, thinner brain tissue, and greater damage to white matter, which is the network of nerve fibers connecting different brain regions. The elevated protein also predicted increased odds of small strokes and microscopic brain bleeds. The findings suggest the protein relates more strongly to blood vessel damage, general brain tissue loss, and inflammation rather than the specific protein pathways of Alzheimer’s disease.

Genetic analyses provided evidence suggesting the protein might play a direct role in causing brain changes. To explore this, the team conducted an experiment using macrophages, which are immune cells that engulf and destroy cellular debris. The scientists exposed macrophages from six adults to the growth differentiation factor 15 protein. This exposure altered immune and energy-related pathways in the cells, such as reducing antiviral signaling and disrupting the removal of free heme, an iron-rich molecule that can be toxic when released from red blood cells.

The authors concluded that their findings support the role of this circulating protein as an early warning sign, particularly for vascular dementia and brain inflammation. They noted that the results also help identify the biological mechanisms by which the protein might drive dementia risk.

A few limitations restrict the study’s conclusions. The blood protein added only modest predictive information for dementia risk beyond a person’s age. The research design also cannot conclusively establish that the protein causes the disease. Finally, the association between the protein and dementia was not statistically significant in a smaller Japanese group of 340 people, suggesting further research is needed across different populations.

The paper, “Plasma GDF15 affects long-term dementia risk and alters neuroimmune signaling,” was authored by Cassandra O. Blew, Michael R. Duggan, Dimitrios Tsitsipatis, Gabriela T. Gomez, Zulema Rodriguez-Hernandez, Luke C. Pilling, Jingsha Chen, Eva Jacobsen, Heather E. Dark, Yifei Lu, Shannon M. Drouin, Cassandra M. Joynes, Minhao Yao, Murat Bilgel, Abhay Moghekar, Qu Tian, Julián Candia, Mary Kaileh, Aditi Gupta, Krystyna Mazan-Mamczarz, Myriam Gorospe, Alexey Lyashkov, Yevgeniya Lukyanenko, Mika Kivimaki, Philipp Frank, Lori L. Jennings, Valborg Gudmundsdottir, Vilmundur Gudnason, Lenore J. Launer, Naoto Kaneko, Shintaro Kato, Makio Furuichi, Masaki Shibayama, Masahisa Katsuno, Keita Hiraga, Yukiko Nishita, Rei Otsuka, James R. Pike, Mary R. Rooney, Pascal Schlosser, Yuhan Cui, Guray Erus, Christos Davatzikos, Rebecca F. Gottesman, Iwao Waga, Priya Palta, Christie Ballantyne, Michael Griswold, Zhonghua Liu, Luigi Ferrucci, Allison B. Herman, and Keenan A. Walker.