A single gene mutation protects against Alzheimer’s disease in people destined to get the disease very young — and now we know why.
The gene mutation affects a protein called reelin that directs brain cells to shred the probable culprits in the disease — toxic amyloid plaques and tau tangles. The mutation makes reelin work much more efficiently, new research reveals.
Article continues below
A tale of two genes
Unpacking how the protective mutation worked began with a population that faced the opposite problem: an extraordinarily harmful mutation that accelerates Alzheimer’s disease.
For decades, people in the lush valleys near Medellin, Colombia, had faced premature memory loss.
Neurologist Dr. Francisco Lopera grew up in the region. While he was still a medical student, he came across his first case, a 47-year-old man displaying memory symptoms usually seen in elderly people with dementia. Lopera traveled across the region, determined to map where people were facing early memory loss. He ultimately identified thousands of people affected by a rare genetic form of Alzheimer’s disease. The condition was autosomal dominant, meaning that anyone carrying at least one copy of the mutated presenilin 1 (PSEN1) gene would, like clockwork, lose their memories in their mid-forties.
Lopera’s work in mapping this affected population was invaluable to dementia research, but his most important contribution came just a year before he died in 2024. He co-authored a paper in the journal Nature Medicine that detailed the case of a patient he met in his travels across Colombia. This patient had the PSEN1 gene mutation but lived well into his 60s before developing Alzheimer’s. This was the neurological equivalent of a house that stays standing for decades despite cracks in its foundations that should give way.
Lopera discovered that this man’s resilient brain was strengthened by another mutation, dubbed COLBOS after the research centers in Colombia and Boston that characterized it.
Improving efficiency
The new research, published in December 2025 in the Journal of the American Chemical Society, has identified exactly how the COLBOS mutation protected the patient’s brain for decades.
When the COLBOS variant was first identified in 2023, scientists noted that the mutation altered how a cell signaling protein called reelin functioned. The protein promotes the formation of new connections between brain cells, prevents the toxic tau protein from becoming activated, and prevents amyloid plaques from building up in the brain.
COLBOS altered how reelin binds to another signaling molecule called heparan sulfate, a sugar found on the cell surface of virtually all cell types in humans, including neurons. But how reelin’s binding ability affected Alzheimer’s progression remained unclear. In the new paper, molecular biologist Chunyu Wang and colleagues at the Rensselaer Polytechnic Institute in New York mapped this process.
Wang’s study relied on a technique called surface plasmon resonance, which reveals. how strongly a free-floating molecule — in this case, reelin — binds to a molecule anchored to a sensor surface — in this case, heparan sulfate. Wang’s team saw that the COLBOS mutation acted like a molecular glue, strengthening the bonds between the two molecules, suggesting it could cause reelin to accumulate at the surface of neurons in the brain.
This change explained why COLBOS was able to ward off Alzheimer’s. When reelin binds to heparan sulfate, the protein becomes localized to the surface of brain cells, where its anti-Alzheimer’s signaling is most effective. Here, reelin can more easily ward off cognitive decline by slowing key Alzheimer’s disease processes, such as the phosphorylation of the tau protein, said Wang. Phosphorylation destabilizes the normally ordered structure of tau, causing toxic tangles to build up in neurons.
Herz mapped out much of the reelin pathway 20 years ago in a series of papers. Herz’s work was in mice, and although his team predicted reelin mutations could be neuroprotective, it was only through Lopera’s tireless clinical work that the theory was proven.
Uphill battle
Unfortunately, as Lopera noted, the COLBOS mutation could only delay rather than prevent people with rare Alzheimer’s mutations from getting the disease. Herz’s theory is that patients with PSEN1 mutations exhibit malfunctions in organelles called the endolysosomal compartments. These are like cellular shredders that chop up troublesome proteins like tau and amyloid. He added that the COLBOS mutation makes the process of feeding these proteins into the shredders more effective.
But as the brain ages, Herz said, the shredder’s defects become harder to overcome, despite reelin’s effects, leading to Alzheimer’s.
The findings could have useful implications for future Alzheimer’s therapies that delay or prevent disease in the vast majority of patients, including ones without high-risk PSEN1 mutations. Wang pointed to recent research that showed reelin-producing neurons are some of the first to die in Alzheimer’s disease. Without these neurons, less reelin is made, toxic waste builds up, and Alzheimer’s disease accelerates.
Wang hypothesized that if they could get reelin to act more efficiently at brain cell surfaces, even with less reelin present, that could protect such people from further symptoms. Wang is currently discussing with a colleague at Rensselaer the development of a gene therapy that enhances reelin signaling based on these findings.
Wang pointed out that even if researchers could delay Alzheimer’s by a fraction of the two decades that the COLBOS mutation granted to the Colombian patients, it would be by far the biggest disease improvement ever realized with Alzheimer’s. Current drugs on the market for the condition may extend independent living by two to three years, tops, he said, so “20 years is amazing.”
Pan, L., Song, X., Su, G., Gandy, L. A., Fang, B., Buttaci, M., Gibson, J., Xia, K., Zhang, F., Liu, J., Wang, L., Temple, S., & Wang, C. (2025). N-Sulfated heparan sulfate promotes reelin signaling as a co-receptor. Journal of the American Chemical Society, 147(51), 46773–46779. https://doi.org/10.1021/jacs.5c15573
