Scientists have identified hundreds of genes that may increase the risk of developing Alzheimer's disease but the roles these genes play in the brain are poorly understood. This lack of understanding poses a barrier to developing new therapies, but in a recent study published in the American Journal of Human Genetics, researchers at Baylor College of Medicine and the Jan and Dan Duncan Neurological Research Institute (Duncan NRI) at Texas Children's Hospital offer new insights into how Alzheimer's disease risk genes affect the brain.
"We studied fruit fly versions of 100 human Alzheimer's disease risk genes," said first author Dr. Jennifer Deger, a neuroscience graduate in Baylor's Medical Scientist Training Program (M.D./Ph.D.), mentored by Drs. Joshua Shulman and Hugo Bellen. "We developed fruit flies with mutations that 'turned off' each gene and determined how this affected the fly's brain structure, function and stress resilience as the flies aged."
These tiny insects, known scientifically as Drosophila melanogaster, have long been used in genetic research and are invaluable for studying brain function. Fruit flies might seem far removed from humans, but scientists have found that most human genes have counterparts in fruit flies, allowing researchers to explore how these genes work in a living organism. In addition, the fly's short lifespan (only 10 weeks) makes it an ideal model to study human conditions of old age like Alzheimer's disease.
"We were very excited about the results," said Shulman, professor of neurology, neuroscience, and molecular and human genetics at Baylor and co-director of the Duncan NRI. He also is co-corresponding author of the work. "We found that most of the genes are expressed in the adult fly brain, including 24 specifically expressed in neurons and 13 in glia, another type of brain cell."
"Overall, we identified 50 candidate Alzheimer's disease risk genes in flies that were involved in both brain structure and function, including 18 that caused possible neurodegeneration when turned off," Deger said.
"One standout example was the gene Snx6, the fly version of human SNX32," Shulman said. "When this gene was turned off, the flies developed holes in their brain tissue - a sign of neurodegeneration."
In addition, the team found that 35 genes were required for proper electrical activity of neurons and eight for the ability of the flies to recover from stress. When these genes were turned off, the flies showed signs of seizures or paralysis after being exposed to heat or mechanical shock.
The researchers also tested whether the genes influenced the toxic effects of two proteins - amyloid-beta and tau - which build up in the brains of people with Alzheimer's. "Twenty-eight of the genes changed how the flies responded to amyloid-beta or tau, either making the damage worse or helping protect against it," Deger said.
Beyond identifying individual genes, the researchers looked for patterns. They grouped the genes according to the type of brain problem they caused - structural damage, functional impairment or poor stress recovery. Then, they compared these groups to genetic data from actual patients with Alzheimer's disease.
"Different people seemed to carry risk genes from different groups. Some had genetic changes linked to brain structure problems, while others had genetic variations tied to stress resilience," Shulman said. "This suggests that different individuals may develop Alzheimer's disease through distinct biological pathways. This idea - called 'causal heterogeneity' - could help explain why Alzheimer's looks different from person to person and why some treatments work for some people but not others."
To accelerate the study of Alzheimer's risk genes, the team created a portal called ALICE, or Alzheimer's Locus Integrative Cross-species Explorer, at https://alice.nrihub.org/, where users can access the functional data generated in this study for their own investigations.
Other contributors to this work include Shabab B. Hannan, Mingxue Gu, Colleen E. Strohlein, Lindsey D. Goodman, Sasidhar Pasupuleti, Zahid Shaik, Liwen Ma, Yarong Li, Jiayang Li, Morgan C. Stephens, Michal Tyrlík, Zhandong Liu, Ismael Al-Ramahi, Juan Botas, Chad A. Shaw, Oguz Kanca and Hugo Bellen. The authors are associated with Baylor College of Medicine and/or the Duncan NRI.
This work was supported by National Institutes of Health grants (T32GM136611, F31NS129062, R01AG074009, R01AG073260, R24OD031447, U01AG072439, T32NS043124, P50HD103555 and U54HD083092). Additional support was provided by Baylor Research Advocates for Student Scientists, the Robert and Janice McNair Foundation, Southern Star Medical Foundation and the BrightFocus Foundation (Postdoctoral Fellowship Program in Alzheimer's Disease Research, A2021008F).
