ETH Zurich Pins a New Alzheimer's Driver on Clumps of an Inactive Enzyme, and an Experimental Compound That Breaks Them Up
ETH Zurich researchers report that aggregates of inactivated GRK2 choke mitochondria and fuel amyloid beta, and that an experimental molecule, compound 10, slowed nerve cell death in mice.
Overview
Researchers at ETH Zurich have identified a regulatory enzyme called GRK2 as a possible driver of Alzheimer’s disease and developed an experimental molecule that blocks the damage it causes, according to ETH Zurich. The work, published June 8, 2026 in the journal Cell Reports Medicine, points to a mechanism distinct from the amyloid plaques that have dominated Alzheimer’s drug development, as reported by Neuroscience News.
The findings remain preclinical. The team’s evidence comes from human brain tissue samples and mouse models of Alzheimer’s disease, not from trials in living patients, according to ETH Zurich.
What We Know
GRK2 is an enzyme that helps cells respond correctly to signals, stress and strain, and it supports the function of nerve cells in the brain, according to ETH Zurich. The enzyme exists in two forms, functional and inactivated, and in dementia patients the inactivated form accumulates and forms aggregates, as reported by News-Medical.
Those aggregates deposit on and damage mitochondria, the structures that supply cells with energy, according to ETH Zurich. “The GRK2 aggregates block the pores of the mitochondria, reducing the amount of energy they can supply and leading to a situation of stress inside the cells,” said Ursula Quitterer, Professor of Molecular Pharmacology at ETH Zurich, as quoted by ScienceDaily.
The inactive form of the enzyme also appears to increase the production of amyloid beta, the protein fragment long associated with Alzheimer’s, setting up a self-perpetuating cycle in which amyloid stress drives further GRK2 inactivation, as reported by Neuroscience News.
To intervene, the team developed a chemical compound it calls compound 10, which prevents the GRK2 molecules from forming aggregates, according to ETH Zurich. With aggregation blocked, the mitochondria worked better, amyloid beta deposition fell, and nerve cells retained their function rather than dying off, according to ETH Zurich.
In the Alzheimer’s mouse models, treatment with compound 10 significantly slowed nerve cell death and the animals survived longer, as reported by News-Medical. The compound also had a positive influence on heart function, and the treated animals developed fewer gray hairs as they aged, according to ETH Zurich.
The research drew on human brain tissue collected during tumor removal surgeries at Ain Shams University Hospital in Cairo, from both dementia and non-dementia patients, as reported by Neuroscience News. The project has spanned nearly two decades, according to News-Medical. “It took so long simply because everything takes so long in Alzheimer’s research,” Quitterer said, according to ETH Zurich.
What We Don’t Know
The central caveat is that compound 10 has not been tested in humans. The results come from mouse models and human brain tissue samples, according to ETH Zurich, and preclinical benefits in mice frequently fail to translate to people.
The team describes the basic research as complete and has applied for a patent on compound 10, but no clinical development partner is yet in place, according to News-Medical. Quitterer and ETH Zurich are now looking for a company interested in advancing the molecule toward a drug, according to ETH Zurich. That search, and the years of clinical testing that would follow, separate the current findings from any treatment patients could receive.
Analysis
The GRK2 hypothesis is notable mainly for where it locates the damage. Most approved and late-stage Alzheimer’s therapies target amyloid beta directly, on the premise that clearing plaques slows decline. The ETH Zurich work instead frames amyloid as one part of a loop centered on enzyme aggregation and failing mitochondria, as reported by Neuroscience News — a mechanism existing drugs do not address.
That positions the research within a broader scientific reckoning over whether amyloid alone explains the disease. The Machine Herald previously reported on a large trial in which a metabolic drug failed to slow Alzheimer’s even as it shifted biomarkers, keeping alternative hypotheses alive. A target tied to cellular energy production and a feedback loop with amyloid fits that line of inquiry, though only human trials can establish whether breaking GRK2 aggregates changes the course of the disease.