Simple model reproduces patterns of toxic protein buildup across multiple neurodegenerative diseases
by Jessica P. Johnson
Fatal neurodegenerative diseases including Alzheimer’s, Parkinson’s, and amyotrophic lateral sclerosis (ALS) begin as tiny pockets of misfolded proteins that evade the body’s normal detritus-removal systems. They spread throughout the brain and clog neural pathways. But exactly how these proteins propagate and spread, and how they relate to brain atrophy, has been notoriously difficult to pin down.
A recent study published in brief in Physical Review Letters, and in greater depth in the Journal of the Mechanics and Physics of Solids, suggests that basic principles of physics can explain the movement of toxic proteins through the brains of people with different forms of neurodegenerative disease. “Each neurodegenerative disease is different, but there are only so many mechanisms by which they can propagate, and that’s something that our model showed,” says study coauthor Johannes Weickenmeier, an engineer at the Stevens Institute of Technology in Hoboken, New Jersey. A better understanding of how neurodegenerative proteins spread and accumulate during different stages of disease could help improve diagnosis, better predict disease progression, and inform novel treatments.
Weickenmeier and colleagues built a generalized model capable of describing the unique progression of multiple neurodegenerative diseases. The model combined an anatomical reconstruction of a human brain, clinical data on the originating locations of toxic protein buildup in Alzheimer’s, Parkinson’s, and ALS patients, as well as a physics-based reaction-diffusion equation that simulates the growth and spread of misfolded proteins throughout brain tissue.
When the researchers seeded their model brain with toxic model proteins, the simulated proteins spread in patterns that matched clinical observations of Alzheimer’s, Parkinson’s, or ALS. Different patterns emerged for each disease even though the model assumed the same mechanism of spread. This suggests that the proteins all use similar modes of transport.
“Despite the apparent complexity of these neurodegenerative diseases, their impact on the brain can be captured in this relatively simple model,” says Lary Walker, an experimental neuropathologist at the Emory University School of Medicine in Atlanta who was not involved in the study.
The model also predicted patterns of atrophy, which occur in brain regions where large aggregations of toxic proteins accumulate. Understanding the links between neurodegenerative disease at the cellular or protein levels and disease symptoms at the anatomical or whole brain level could help elucidate the requisite biochemical pathways, and may one day enable diagnosis before symptoms develop. And the model’s apparent non-specificity could make it a useful tool for modeling other neurodegenerative diseases, including CTE.
But those achievements are still a long way off. First, Weickenmeier says, he intends to validate the model’s predictive capabilities using longitudinal, annual MRI scans of Alzheimer’s patients. “It remains to be seen how well the simulations involving individual brain regions coincide with what is known about the formation of protein assemblies in those regions,” says Michel Goedert, neuroscientist at the University of Cambridge in the UK who was not involved in the study.
Weickenmeier also intends to refine the model’s predictive abilities by incorporating biological data specific to a particular neurodegenerative disease. “It’s definitely too early to apply it to prediction in a given case,” Walker says. “But as I see it, it’s a foundation for using mathematical modeling as a way of predicting disease progression.”
Originally published at blog.pnas.org on October 26, 2018.
