Alzheimer’s Risk Gene APOE4 Overexcited Young Mouse Memory Circuits Before Memory Declined

TL;DR: Young mice carrying the human APOE4 gene, the biggest genetic risk factor for Alzheimer’s, had overexcited memory circuits months before any memory trouble showed up. Removing APOE4 from their neurons returned those cells to normal.

Source: Nature Aging (2026) | Tabuena et al.

About 60 to 75% of people with Alzheimer’s carry at least one copy of the APOE4 gene variant. In people, carriers show memory changes years before diagnosis, and people with early Alzheimer’s often have hidden seizure-like activity in the brain.

A team at the Gladstone Institutes in San Francisco asked whether APOE4 overexcites the brain’s memory hub early in life, and whether that sets the stage for later decline.

Mice With Human APOE Genes

The researchers compared two groups of mice whose own APOE gene was swapped for a human version:

• APOE4 mice: Carrying the high-risk human APOE4 variant.
• APOE3 mice: Carrying APOE3, the common, neutral variant.

These mice do not build up amyloid plaques, so any changes come from APOE4 itself rather than plaques.

Early Spikes Came Before Memory Trouble

At 5 to 10 months old, APOE4 mice had more abnormal electrical spikes (brief bursts seen between seizures in epilepsy) in two hippocampus areas, CA3 and the dentate gyrus. At 6 to 8 months, they still learned a water maze normally. By 14 months, the same mice had clear trouble learning where a hidden platform was.

In a group followed over time, APOE4 mice with more spikes when young tended to take longer to find the platform when older. That link did not show up in APOE3 mice.

Shrunken Neurons That Fire Too Easily

Recording from single cells, the researchers found subgroups of smaller neurons in young APOE4 mice that fired more readily than normal. With age, more problems piled up in the dentate gyrus: overexcited granule cells, weakening inhibitory (braking) signals and an imbalance between excitation and inhibition.

Removing APOE4 From Neurons Fixed the Early Problem

APOE is made mainly by support cells called astrocytes, but neurons make it too under stress. The team used mice in which APOE4 could be deleted from neurons only. In those mice, the early cell shrinkage and overexcitability were fully normalized.

Gene activity analysis pointed to candidates behind the problem. One, Nell2, was more active in the affected neurons. Using a CRISPR tool to turn Nell2 down in those neurons returned their size and excitability toward normal.

What Mouse Circuits Can’t Tell Us Yet

• Mice, not people: This is not a screening test or treatment.
• A link, not proof: Early spikes tracked later maze trouble in only 11 APOE4 mice.
• Age and sex matter: Findings varied by cell type, age and sex.
• Nell2 is one candidate: The CRISPR test rescued excitability in recorded cells, not memory.

Could Calming Early Overactivity Help Carriers?

The study fits a growing idea that APOE4 harms the brain decades before Alzheimer’s symptoms, partly by overexciting memory circuits. The open question is whether reducing that early overactivity, through neuron-targeted treatments or genes like Nell2, would protect memory later in life.