The pro-Alzheimer’s allele APOE4 makes hippocampal neurons in mice smaller and hyperexcitable. This effect, which resembles epilepsy and accelerated aging, can be mitigated by manipulating a neuronal protein [1].
Before symptoms arise
Alzheimer’s disease begins long before symptoms appear, building silently for decades. The single strongest genetic risk factor for the common, late-onset form of Alzheimer’s is the ε4 variant of the apolipoprotein (APOE) gene, APOE4. Carrying a single copy of this variant (being heterozygous) roughly triples your Alzheimer’s risk; having two copies increases it about 12-fold.
Scientists have long known that years before visible symptoms appear, the brain’s hub for learning and memory (the hippocampus) becomes abnormally overactive [2]. This hyperexcitability manifests as interictal spikes (IIS): brief, spontaneous bursts of synchronized neuronal firing, similar to what occurs in epilepsy.
These spikes are common in preclinical and early Alzheimer’s, and their frequency predicts the rate of cognitive decline [3]. Young, cognitively normal APOE4 carriers exhibit hippocampal hyperactivation, and APOE4 is associated with higher epilepsy risk and earlier onset [4].
However, the mechanisms behind this APOE4-related hippocampal hyperexcitability have been largely unknown. A new study from Gladstone Institutes, published in Nature Aging, aimed to bridge this gap.
Don’t get too excited
The authors analyzed in vivo local field potential (LFP) recordings: electrical signals recorded from brain probes implanted in freely moving mice, which had one of two human APOE alleles knocked-in: APOE4 (E4-KI) or the less Alzheimer’s-associated variant APOE3 (E3-KI). Data was collected at young (5-10 months) and aged (12-18 months) timepoints.
Young E4-KI mice showed elevated IIS rates in certain hippocampal regions (specifically, in CA3 and dentate gyrus, but not in CA1), compared to age-matched E3-KI animals. Aged E3-KI mice also showed some increase in IIS rates, suggesting that this particular Alzheimer’s feature resembles accelerated aging.
Next, a cohort of E3-KI and E4-KI mice underwent LFP recordings when they were young, and they were tested on the standard Morris water maze test at both young and old ages. This was to see whether early IIS rates correlate with later learning performance in the same individual animals.
Young E4-KI mice performed normally on the water maze, but by 14 months, the same animals developed significant spatial learning deficits. Early IIS rates in young E4-KI mice significantly predicted how poorly those same mice would perform on the water maze in old age, while no such correlation existed in E3-KI mice.
“To the best of our knowledge, this is the first study that has directly examined what APOE4 does to the function of neurons at different ages,” said Misha Zilberter, Ph.D., principal staff research scientist at Gladstone and a senior author of the study. “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.”
Using whole-cell patch-clamp recordings, in which a tiny glass pipette is sealed onto a cell, the researchers precisely measured the electrical properties of single neurons. CA3 pyramidal neurons in young E4-KI mice turned out to be hyperexcitable compared to E3-KI animals. E4-KI neurons were also smaller, which seemed to directly contribute to their hyperexcitability.
By old age, E3-KI CA3 cells became smaller and more excitable as well, eliminating the genotype difference and again hinting at accelerated aging. No significant differences were found in CA1 pyramidal neurons between genotypes at either age, confirming the phenomenon’s region-specific nature.
APOE is mainly produced by astrocytes in the brain, though stressed neurons can also make it. To determine which cellular source drives the phenotype, the authors used E4-KI mice with APOE4 deleted either from astrocytes or from neurons. Removing APOE4 from neurons completely rescued all morphological and electric abnormalities in CA3 pyramidal neurons, while removing APOE4 from astrocytes had no effect.
The researchers found two robust neuronal clusters: one characterized by smaller size and higher excitability (the “hyperexcitable” cluster) and one with normal properties. Young E4-KI mice had significantly more neurons in the hyperexcitable cluster, and aging shifted E3-KI neurons into this cluster. Deleting APOE4 in neurons depleted the hyperexcitable cluster back to E3-KI levels.
Identifying the target
Single-nucleus RNA sequencing produced several genes differentially expressed between E4-KI and E3-KI mice. Neural epidermal growth factor-like protein 2 (Nell2) stood out. The authors then used an interference technique called CRISPRi to knock Nell2 down, which led to reduced excitability and larger cell size.
“This study is a big breakthrough for the field of Alzheimer’s research,” said Yadong Huang, MD, Ph.D., associate director of the Gladstone Institute of Neurological Disease and a senior author of the study. “What’s exciting about Nell2 is that we were able to reverse the disease manifestations in adult mice by lowering its level. That tells us the damage is not irreversible, and that there may be a window for intervention even after disease processes have been triggered.”
While these findings are indeed encouraging, it is still unknown how Nell2 does what it does or whether manipulating it would translate into reduced network hyperexcitability in vivo or improved learning and memory. Setting up such an experiment is a complex task likely to be tackled in a separate follow-up study.
Literature
[1] Tabuena, D. R., Jang, S. S., Grone, B., Yip, O., Aery Jones, E. A., Blumenfeld, J., … & Zilberter, M. (2026). Neuronal APOE4-induced early hippocampal network hyperexcitability in Alzheimer’s disease pathogenesis Nature Aging, 1-19.
[2] Putcha, D., Brickhouse, M., O’Keefe, K., Sullivan, C., Rentz, D., Marshall, G., … & Sperling, R. (2011). Hippocampal hyperactivation associated with cortical thinning in Alzheimer’s disease signature regions in non-demented elderly adults. Journal of Neuroscience, 31(48), 17680-17688.
[3] Vossel, K. A., Tartaglia, M. C., Nygaard, H. B., Zeman, A. Z., & Miller, B. L. (2017). Epileptic activity in Alzheimer’s disease: causes and clinical relevance. The Lancet Neurology, 16(4), 311-322.
[4] Briellmann, R. S., Torn–Broers, Y., Busuttil, B. E., Major, B. J., Kalnins, R. M., Olsen, M., … & Berkovic, S. F. (2000). APOE ε4 genotype is associated with an earlier onset of chronic temporal lobe epilepsy. Neurology, 55(3), 435-437.
View the article at lifespan.io
