New mouse research from Gladstone Institutes suggests that APOE4 Nell2 brain hyperactivity could be detectable years before any memory symptoms emerge, and, crucially, may be reversible. The findings, published in Nature Aging, map a molecular chain from a well-known genetic risk factor to early neuronal changes, and point towards a protein that has not previously been studied in this context.
What APOE4 Nell2 brain hyperactivity looks like in practice
APOE4 is one of three common variants of the APOE gene. Roughly one in four people carry it, and the variant is estimated to appear in 60 to 75 per cent of people with Alzheimer’s disease. The research team studied young mice carrying APOE4 and found excessive neuronal activity in two areas of the hippocampus, the brain region central to memory. Those same hippocampal regions have previously been found to be hyperactive in human APOE4 carriers as well.
The mechanism, the researchers argue, runs through a protein called Nell2. APOE4 increased Nell2 production, which in turn caused neurons to become smaller. Smaller neurons respond more easily to stimulation, making them more likely to fire excessively. The mice with the greatest hippocampal hyperactivity when young later performed worst on spatial learning and memory tests.
‘We found fundamental changes in brain circuits occurring in young mice that still had normal learning and memory, and importantly, that those changes predicted the development of cognitive deficits at older ages,’ says Misha Zilberter, PhD, principal staff research scientist at Gladstone and a senior author of the study.
The comparison with APOE3 mice (carrying a variant associated with lower Alzheimer’s risk) is instructive. Neurons in APOE3 mice did eventually become more excitable, but only in old age. Yadong Huang, MD, PhD, associate director of the Gladstone Institute of Neurological Disease and a senior author of the study, says this pattern ‘suggests APOE4 accelerates a process that resembles normal aging, and could explain why people with the gene variant are more likely to develop Alzheimer’s disease earlier in life.’
Nell2 as a potential target, and the limits of mouse data
Before attributing this mechanism to astrocytes, the support cells that produce most APOE4 in a healthy brain, the researchers tested the idea directly. Deleting the APOE4 gene from astrocytes produced no change. Deleting it from neurons restored normal cell size and firing behaviour. The hyperactivity, the team concluded, was driven entirely by APOE4 produced within neurons themselves.
Nell2 had not previously been investigated specifically in connection with APOE4, though earlier work had found elevated levels of the protein in the brains of Alzheimer’s patients, with higher amounts associated with poorer cognitive function. Using CRISPRi (a method that reduces gene activity without permanently altering DNA) the researchers lowered Nell2 in hippocampal neurons from adult APOE4 mice. The neurons became larger and less excitable. ‘That tells us the damage is not irreversible, and that there may be a window for intervention even after disease processes have been triggered,’ Huang says.
What the study does not yet show is whether any of this translates to humans. The word ‘may’ is doing a lot of work in statements about future therapies targeting Nell2. The researchers themselves frame their conclusions carefully: the finding ‘raises the possibility’ that such drugs could help people who carry APOE4. First author Dennis R. Tabuena, PhD, a scientist co-mentored by Zilberter and Huang, notes that the degree of hyperactivity in young mice predicted later memory decline, but mouse models of Alzheimer’s have a long history of producing results that do not survive translation into clinical trials.
Where Nell2 sits in a crowded pipeline
The broader context is a drug development landscape that, by most measures, is growing rapidly. According to the Alzheimer’s Association, there are currently 192 clinical trials and 158 novel agents in the Alzheimer’s disease drug development pipeline in 2026. Eight Phase 3 trials are expected to reach their primary completion date this year, along with 29 Phase 2 trials. Of the agents currently in trials, disease-targeting therapies account for 73 per cent, cognition-enhancing symptom-targeted therapies contribute 18 per cent, and drugs targeting neuropsychiatric symptoms make up 10 per cent.
One recent addition to the approved end of that pipeline: in April 2026, the US Food and Drug Administration approved AUVELITY® for the treatment of agitation associated with dementia due to Alzheimer’s disease, according to Delve Insight. That approval addresses a neuropsychiatric symptom rather than the underlying disease process, a reminder that most approvals to date have targeted downstream effects rather than causes.
A Nell2-targeting therapy, if it reaches trials at all, would sit firmly in the disease-targeting category. With eight Phase 3 trials completing this year, the field will learn a good deal more about what translates and what does not before any Nell2 programme would reach that stage. The Nature Aging paper is a beginning, not a result.
