People who carry APOE ε4, the gene variant most strongly associated with Alzheimer's, face an elevated risk of developing the disease. Yet some carriers never do, remaining cognitively healthy into old age. Two years ago, the Columbia team reported that some of these protected individuals possess a rare protective mutation in the fibronectin (FN1) gene. Fibronectin normally helps support tissue structure, but in people with APOE ε4 it can accumulate excessively around the brain's blood vessels, potentially compromising the blood-brain barrier. The rare mutation appears to limit that buildup.
The new study moves from that genetic clue to a detailed disease mechanism. Combining human brain tissue and cerebrospinal fluid samples, genetic and epigenetic analyses, single-cell sequencing, spatial imaging, stem-cell models, three-dimensional cultures of blood vessels, and experiments in zebrafish and mice, the researchers found that APOE ε4 promotes the production and accumulation of fibronectin around brain blood vessels. Mice engineered to carry human APOE ε4 had nearly twice as much fibronectin in the brain as mice with the common APOE ε3 variant, alongside evidence of a leaky blood-brain barrier.
The team also showed that fibronectin is not merely a marker of damage. When they increased human fibronectin specifically in astrocytes, the support cells that help maintain the barrier, the barrier became more permeable; reducing fibronectin restored its function. "The animal experiments allowed us to move beyond association," said Prabesh Bhattarai, co-first author and an associate research scientist in the Kizil lab. "When we increased fibronectin in astroglia, the barrier became permeable. When we reduced fibronectin, barrier function improved."
At the molecular level, excess fibronectin activates abnormal signals through integrins and an enzyme called focal adhesion kinase, disrupting communication between growth factors including VEGF, HBEGF and IGF1. Those signals help blood vessel cells and astrocytes coordinate to keep the barrier healthy. "We observed the same biological pattern across experimental models and human Alzheimer's tissue, making the mechanism especially compelling," said co-first author Elanur Yilmaz.
Signs of the same disturbance appeared in human samples: elevated fibronectin in brain tissue and cerebrospinal fluid was associated with inflammation in astrocytes, and genetic analyses linked FN1 to blood vessel disease and changes involving VEGFA, a growth factor that maintains healthy brain blood vessels. "The convergence of these independent human datasets strengthens the case that fibronectin is part of the APOE ε4-related vascular disease process," said Badri Vardarajan, associate professor of neurological science at Columbia, who collaborated on the discovery of the FN1 mutation.
The implications could be significant, because about one in five people carries at least one copy of APOE ε4. The findings suggest several possible approaches, including one that would prevent harmful amounts of fibronectin from accumulating around the brain's blood vessels. "This new study takes us from a genetic clue to a disease mechanism," said corresponding author Caghan Kizil, professor of neurological sciences at Columbia University Vagelos College of Physicians and Surgeons. "We knew that changes in fibronectin could protect against Alzheimer's, but we didn't know why excess fibronectin was harmful in the first place. This study gives us that detailed mechanism, and with it, ideas for how to reproduce that natural protection."