A single-cell hippocampus atlas finds neurogenic cells and a SuperAger signature
Researchers profiled nearly 356,000 nuclei from post-mortem hippocampi across five groups. Chromatin accessibility in neurogenic cells was already altered in people with preclinical Alzheimer's pathology.
Whether the adult human hippocampus makes new neurons has been disputed for decades, with individual papers reaching opposite conclusions and the field splitting along methodological lines [s1]. A study published in Nature on 25 February does not settle the dispute by argument. It builds an atlas.
Researchers analysed 355,997 nuclei isolated from post-mortem human hippocampus samples using multiomic single-cell sequencing — single-nucleus RNA sequencing combined with single-nuclei assay for transposase-accessible chromatin sequencing — and identified neural stem cells, neuroblasts and immature granule neurons [s1].
The five groups
The design is the study's most interesting feature, because it includes a comparison group that most work of this kind does not.
Post-mortem hippocampi came from five cohorts [s1]:
- young adults with intact memory;
- aged adults with no cognitive impairment;
- aged adults with extraordinary memory capacity, referred to as SuperAgers;
- adults with preclinical intermediate pathology;
- adults with Alzheimer's disease.
The SuperAger group and the preclinical group are what give the analysis directionality. One represents the upper tail of cognitive ageing; the other represents people with pathology but not yet disease. Between them, they let the researchers ask not only what Alzheimer's looks like but what precedes it and what resists it.
What the multiomics showed
The central result concerns chromatin rather than transcription. Dysregulated neurogenesis was largely associated with changes in chromatin accessibility [s1].
Analysis of transcription factors and target gene signatures distinguishing each group revealed early alterations in chromatin accessibility of neurogenic cells in individuals with preclinical Alzheimer's disease, and such changes were more evident in samples from individuals with Alzheimer's disease [s1].
"Early" is the operative word. The alterations appear in people with intermediate pathology who had not progressed to disease — meaning the change is present at a stage where the clinical picture is not yet Alzheimer's.
In the SuperAgers, the researchers identified a distinct profile of neurogenesis that they suggest may reflect a resilience signature [s1]. The authors' framing is appropriately provisional: it is a profile that may reflect resilience, not a demonstrated cause of it.
Beyond the neurogenic lineage, the study reports that alterations in the profile of astrocytes and CA1 neurons govern cognitive function in the ageing hippocampus [s1]. The authors summarise the work as pointing to a multiomic molecular signature of the hippocampus that distinguishes cognitive resilience from deterioration with ageing [s1].
Where this sits in a long argument
The paper's own framing acknowledges the contested history: the existence of human hippocampal neurogenesis has long been disputed, and its relevance to cognition remains unknown [s1]. Recent work had established the presence of proliferating progenitors and immature neurons, and a reduction in immature neurons in Alzheimer's disease, but their origin and the molecular networks regulating neurogenesis and its function were poorly understood [s1].
This study addresses the second gap more than the first. It identifies the cell types and then asks what regulatory state distinguishes them across cognitive outcomes.
What the design cannot tell you
Every limitation here flows from the same source: this is post-mortem tissue.
No time course. Each brain contributes a single snapshot at death. Statements about "early" alterations rest on comparing groups at different disease stages, not on watching individuals change. Cross-sectional group differences are consistent with a temporal sequence but do not demonstrate one.
No causal direction. Whether altered chromatin accessibility in neurogenic cells contributes to cognitive decline, results from the same process that causes it, or reflects agonal and post-mortem conditions is not resolved by association. The same applies to the SuperAger signature: it may be a cause of resilience, a consequence of it, or a correlate of something else that differs between these donors.
Small tail groups. SuperAgers are, by definition, rare, and the paper's cohort structure means the most interesting comparisons rest on the fewest brains. The abstract does not report per-group sample sizes.
Function is inferred. Nuclei sequencing identifies cells with the molecular characteristics of neural stem cells, neuroblasts and immature neurons. It does not observe those cells maturing, integrating into circuits, or contributing to memory. The relevance of human hippocampal neurogenesis to cognition, which the authors note remains unknown, is not established by this paper either [s1].
Why it still matters
Two reasons.
The dispute over adult human neurogenesis has repeatedly turned on whether the cells detected are real or artefacts of tissue handling and marker choice. A 355,997-nucleus multiomic dataset that resolves distinct neural stem cell, neuroblast and immature neuron populations is a substantially harder observation to attribute to artefact than a single immunohistochemical marker.
And the location of the signal — in chromatin accessibility rather than in cell counts — reframes what the question is. A field that has spent years counting immature neurons is being pointed toward the regulatory state of the cells that produce them.
What to watch
Whether the chromatin findings replicate in independent brain banks with different collection protocols. Post-mortem interval and tissue handling are the historical confounders in this literature.
Whether the SuperAger signature holds when more such brains are profiled, and whether any element of it corresponds to something measurable in living people.
And whether the transcription factor networks the study identifies turn out to be manipulable in model systems. That is the step between a descriptive atlas and a mechanism.
Sources
- [s1] Disouky A, Sanborn MA, Sabitha KR, et al., "Human hippocampal neurogenesis in adulthood, ageing and Alzheimer's disease," Nature, published online 25 February 2026. https://doi.org/10.1038/s41586-026-10169-4
Sources
- Human hippocampal neurogenesis in adulthood, ageing and Alzheimer's disease — Nature , February 25, 2026
More on
A sugar-processing defect in the brain, and a supplement that made it worse in mice
University of Florida researchers identify excess glycan production as a driver of Alzheimer's pathology, and find oral glucosamine worsened cognition in mice and tracked with faster decline.
Alzheimer's blood tests exist. In people with no symptoms, they answer the wrong question.
The approved tests are for patients already showing cognitive decline. Used on the worried well, a positive result mostly signals a risk that may never become disease — and no treatment follows from it.
A tau antibody drove a tangle marker down 90% — in seven people
The biomarker effect reported at CTAD is large and sustained over two years. The sample is seven, the design is open dose-escalation, and imaging was available for three.
An FDA-cleared Alzheimer's blood test missed badly in one memory clinic cohort
At a Mayo Clinic cohort of 252 patients, 40% of amyloid-negative people tested positive. The manufacturer says it has found a manufacturing problem and put the product on hold.