Astrocytes do so much more than any of us thought - they even communicate among specific brain regions across hemispheres!
The journey to this paper spanned nearly a decade, but I'm most excited to see what everyone else does with the tools we've built. There's so much for all of us to explore!
Posts by Misha Zilberter
Congrats, can’t wait to read it!
A study in Nature Aging shows that the strongest Alzheimer’s Disease risk gene, APOE4, can make certain brain cells in young mice hyperactive, which can predict later memory decline. The study also implicates Nell2 protein as a contributor to APOE4-driven dysfunction. #Alzsky 🧬🧪
Indeed!
I know that store, been many times when visiting! And their plombir (пломбир)
STAT: Trump administration drops court fight to cap NIH payments for research overhead costs
by Megan Molteni @statnews.com @mmolteni.bsky.social
bit.ly/4tVncB5
Single-Nuclei Transcriptomic Characterization of APOE4-Associated Alzheimer's Disease www.biorxiv.org/content/10.64898/2026.04...
True team effort. Huge credit to stellar postdocs Dennis Tabuena and Sung-Soo Jang, who led this work, Yadong Huang and his lab as close collaborators — and a special shout-out to @emilyaeryjones.bsky.social
for key in vivo recordings.
This work was supported by NIH/NIA. Fin.
Working model: neuronal APOE4 drives cell type–specific transcriptomic changes (incl. NELL2), leading to neuronal atrophy and hyperexcitability, resulting in network hyperactivity. With aging, this promotes interneuron loss and inhibitory deficits, ultimately resulting in cognitive decline. 15/15
This is supported by prior work from the Huang lab showing that suppressing network hyperexcitability in E4-KI mice with a 6-week pentobarbital treatment at 9 months protects DG interneurons and prevents later cognitive deficits at 16 months. 14/15
www.jneurosci.org/content/36/7...
Finally, we examined synaptic function and found progressive loss of inhibitory input in DG of E4-KI mice with aging. This suggests early hyperexcitability is not just a symptom, but a driver: sustained activity likely stresses vulnerable interneurons, leading to their dysfunction and loss. 13/15
Although understudied outside development, NELL2 is linked to neuronal growth and homeostasis and is part of a “neuron cellular homeostasis” GO pathway alongside AD genes (APP, PSEN1/2, etc). Nell2 is also elevated in AD brain and CSF and correlates with amyloid burden and cognitive decline. 12/15
We used CRISPRi to reduce Nell2 expression in young E4-KI mice and found that transduced neurons showed normalized size and excitability compared to neighboring cells, identifying Nell2 as a key mediator of APOE4-driven neuronal dysfunction. 11/15
To identify molecular drivers, we analyzed hippocampal snRNA-seq data, filtering for genes matching APOE-, cell type-, and age-dependent e-phys changes.
From hundreds of candidates, only a few passed filtering, and just one—Nell2—was upregulated in both CA3 and DG neurons in young E4-KI mice. 10/15
We next asked: what is the source of pathogenic APOE4? Most APOE is astrocyte-derived, but neurons upregulate it under stress. Deleting APOE4 from neurons (fE4/Syn1-Cre) normalized size and excitability; astrocyte deletion (fE4/GFAP-Cre) had no effect—supporting a neuron-intrinsic mechanism. 9/15
With aging, E3-KI neurons “caught up,” becoming smaller and more excitable. This suggests APOE4 may accelerate processes that normally occur with aging, consistent with epidemiological data showing earlier AD onset and higher risk in APOE4 carriers, with convergence to APOE3 at older ages. 8/15
Importantly, this was not uniform. Within the same regions, some neurons remained relatively normal, while others were strongly hyperexcitable, forming distinct functional clusters and revealing selective vulnerability to APOE4 rather than a global shift in function. 7/15
We then looked at what drives this at the single-neuron level. Excitatory neurons in CA3 and DG —but not CA1— were smaller and hyperexcitable in young E4-KI mice. Neuronal size strongly correlated with excitability in E4-KI but not E3-KI neurons, directly linking atrophy to hyperexcitability. 6/15
When we followed the same mice into older ages (14–18 mo), E4-KI animals developed learning and memory impairments.
Critically, E4-KI mice with the highest CA3/DG IIS rates when young performed the worst later — again consistent with human data. 5/15
We recorded hippocampal activity (CA3, DG, CA1) in young (5–10mo) and aged (12–18mo) mice, focusing on interictal spikes (IIS), a marker of hyperexcitability.
E4-KI mice showed increased IIS in CA3 and DG—but not CA1—already at young ages, mirroring hyperactivation seen in human APOE4 carriers.4/15
We used human APOE knock-in mice (E3-KI vs E4-KI), which recapitulate key features of human risk.
Even without amyloid or tau pathology, E4-KI mice develop cognitive deficits with aging. Notably, they also exhibit spontaneous seizures early in life. 3/15
APOE4 is the strongest genetic risk factor for late-onset AD. Even young, cognitively normal APOE4 carriers show hippocampal hyperactivation—and its extent predicts future cognitive decline. What drives this, and how do these early changes lead to later decline? 2/15
Our paper is now online at @nataging.nature.com!
Alzheimer’s disease begins decades before symptoms—but what drives those earliest changes is still not well understood.
We set out to understand how APOE4 disrupts neuronal and network function across aging. 🧵: 1/15
www.nature.com/articles/s43...
Sex-specific APOE4-dependent innate immunity regulates meningeal lymphatics, brain lipids, neuroinflammation, and cognition: @cp-neuron.bsky.social www.cell.com/neuron/fullt...
“Interrogation of data from cerebral organoids demonstrated that, although broad elements of in vivo development are recapitulated in vitro, many layer-specific transcriptomic programs in neuronal maturation are absent.”
Got questions about the brain dynamics, organoids are a dead end. Full stop.
Behind the scenes: postdoc, first author, and star patch-clamper Dennis Tabuena, captured in his native habitat during a press photoshoot.
*Dennis Tabuena
A half-century old question may have its final answer. Using high-resolution #Mini2P microscopes, we find no evidence of local topography in #PlaceCells. Place fields of neighbouring cells are no more similar than those of randomly selected cells. 🧠🗺️ Out now in @pnas.org www.pnas.org/doi/10.1073/...