A broken enzyme may explain how dementia starts in the brain

Scientists have traced a rare form of early-onset dementia to a single point mutation—a tiny flaw in an enzyme called GPX4 that leaves neurons defenseless against a specific type of cell death.

The discovery, made across 14 years of international collaboration, suggests this same mechanism might underlie more common forms of dementia too. It's a shift in how researchers think about what triggers neurodegeneration: not just the protein clumps we've focused on for decades, but damage to the cell membrane itself.

The Surfboard That Protects Neurons

Think of GPX4 as a surfboard with a fin. That fin dips into the cell membrane and rides along the inner surface, neutralizing harmful molecules called lipid peroxides as it goes. In children with this rare dementia, a single letter change in their genetic code alters the fin's shape. The enzyme can no longer anchor properly into the membrane. Without GPX4 doing its job, lipid peroxides accumulate. The buildup weakens the membrane, triggers a process called ferroptosis, and neurons rupture and die.

Researchers confirmed this mechanism in mice engineered to carry the same mutation. As GPX4 became impaired, the animals slowly developed serious movement problems, lost neurons in key brain regions like the cerebral cortex and cerebellum, and showed signs of brain inflammation. The pattern matched what was seen in the affected children—and resembled patterns typical of Alzheimer's disease.

When the team analyzed protein changes in these mice, they found something striking: many proteins that increase or decrease in Alzheimer's disease showed similar changes when GPX4 failed. This suggests ferroptosis—this specific type of cell death—might not be unique to this rare childhood disorder. It could be a common thread running through multiple types of dementia.

A New Angle on an Old Problem

For years, dementia research has centered on amyloid-beta plaques—sticky protein deposits that accumulate in the brains of Alzheimer's patients. But this finding points to an earlier problem: damage to the cell membrane that sets neurodegeneration in motion. "Our data indicate that ferroptosis can be a driving force behind neuronal death—not just a side effect," says researcher Svenja Lorenz.

Early experiments are encouraging. In cell cultures and mice, blocking ferroptosis slowed the cell death caused by loss of GPX4. But researchers are careful to note this is still basic research. No therapy yet.

What's remarkable is how long this took to uncover. Marcus Conrad, one of the lead researchers, notes it took 14 years to connect a tiny structural detail of a single enzyme to a severe human disease. That timeline underscores why basic research needs sustained funding and why dementia—complex, multifaceted—demands teams that span genetics, structural biology, stem cell research, and neuroscience across continents.

With ferroptosis now on the radar as a potential driver of neurodegeneration, researchers have a new target to pursue. The next phase: understanding whether ferroptotic stress contributes to dementia in the broader population, and whether therapies that block this pathway might slow or prevent cognitive decline.