Depression May Stall the Brain’s Production of New Neurons

It’s a pattern Maura Dupont sees again and again. A patient goes out with a friend for lunch. The friend is tired and doesn’t talk much. With an intact hippocampus, the patient remembers this as a unique event. With depression, it becomes tangled with every previous memory of feeling rejected. “They’re upset with me,” the patient concludes. In the hippocampus, the brain’s memory hub, a stalled process of creating new neurons may be why depressed people have trouble separating a friend’s quiet mood from a history of rejection.

The finding, published in Nature Medicine, comes from an extraordinary dataset: nearly half a million cell nuclei extracted from the hippocampus of people with major depressive disorder and healthy controls with no history of psychiatric illness, collected within hours of death. Using advanced techniques that measured which genes were active in each cell, researchers at Columbia University Vagelos College of Physicians and Surgeons have for the first time mapped the molecular programs that control adult neurogenesis — the trickle of new neurons created in the hippocampus — and shown that the process appears to stall in depression.

What Pattern Separation Means For Your Lunch Date

The hippocampus is one of the few places in the adult brain that creates new neurons. Most of our roughly 100 billion neurons are laid down before birth, but this slow trickle into the hippocampus was long thought to be vestigial in humans, a remnant of development that served no real purpose. That view has been shifting. Studies in mice have shown that new neurons help distinguish between similar but different memories, a function called pattern separation. The same appears true in people: patients who received radiation to the hippocampus to treat a benign brain tumor, ablating new neuron formation, showed transient deficits in separating similar experiences, as documented in a 2025 study of hippocampal irradiation.

“The newborn neurons seem to enhance pattern separation because they are especially responsive to new experiences,” Dupont says. “They can be incorporated into new memory circuits more easily, allowing new memories to be stored separately from the old ones.”

The pattern separation framework explains something clinicians have long observed in depression: a negative memory bias that colours every new experience. Without new neurons, the hippocampus cannot keep old emotional memories from bleeding into the present. “When this ability is impaired, memories, together with their emotional value, become less distinct and more likely to blend together,” Dupont says. “I see this a lot in my patients, where they can only retrieve negative information from their memories.”

A Stalled Process, Not A Depleted One

The Columbia team profiled 495,037 nuclei from the hippocampus of 11 people with MDD and 19 controls, combining single-nucleus RNA sequencing with chromatin accessibility assays. They identified a full neurogenic lineage — from quiescent neural stem cells through activated progenitors and neuroblasts to immature and mature granule cells — confirming that adult human hippocampal neurogenesis is a real, ongoing process, consistent with recent findings that immature neurons persist throughout life.

In the depressed brains, something had gone wrong. The distribution of cells along this lineage shifted: more quiescent stem cells, fewer neuroblasts. “Turning neurogenesis back on may be a way to treat depression in some people by rewiring their hippocampus circuit,” Dupont says.

The study went beyond cell counts. The researchers recorded which genes were active in each cell and where each cell sat in the hippocampus circuit. They found that the stalled neurogenesis was accompanied by widespread molecular changes across the entire trisynaptic circuit — the hippocampus’s primary means of establishing new emotional memories. Genes involved in creating new connections, cross-talk between neurons, energy supply and intracellular transport were all altered. The tissue showed signs of inflammation and cellular stress. Some genes were disrupted by epigenetic changes reflecting life experiences such as stress, learning and aging. Earlier work from the same group had already shown that human hippocampal neurogenesis persists into old age; the new study reveals what happens when that process breaks down.

“We do not yet know the complete mechanism, particularly in humans,” Dupont emphasises. But the pattern is clear: the whole circuit suffers, not just the stem-cell niche.

The researchers also found that some of the dysregulated genes have known genetic variants associated with major depression, while others were affected by environmental factors. This suggests both inherited and acquired components. “These are like dimmer switches that control how active genes are,” Dupont says, “and they are affected by life experiences such as stress, learning, aging, chemicals, etc.”

A substantial fraction of the altered genes overlapped with those implicated in neurodegenerative diseases, autoimmune conditions and neurodevelopmental disorders. Interferon-related genes, typically associated with antiviral defence, were elevated in the neurogenic populations of depressed brains, a finding that echoes the growing recognition of neuroinflammation in psychiatric illness.

Classifying Depression Like Cancer

Historically, depression was thought to be a deficiency of neurotransmitters, especially serotonin. That view drove a generation of antidepressant development, drugs that provide relief for some but leave many others treatment-resistant. The new study suggests a different picture: depression at the cellular level looks like multiple diseases with overlapping but distinct mechanisms.

“The wide range of effects we found could reflect different pathogenetic mechanisms, perhaps indicating that depression is not just one disease,” Dupont says.

The ambition is to reclassify depression the way cancer has been reclassified: not by location or symptom cluster, but by molecular features. “Classifying cancers based on their cellular characteristics, not their locations, has led to new and improved treatments,” Dupont says. “We hope the same will be true for depression and other psychiatric or brain diseases.”

That’s still a long way off. The study is cross-sectional and postmortem, so it cannot prove that stalled neurogenesis causes depression or results from it. Most of the depressed donors died by suicide, making it difficult to untangle depression from suicide pathology. And the exact role of new neurons in human memory remains an active question. But for the first time, researchers have a detailed molecular map of what goes wrong in the depressed hippocampus.

What would it mean to turn neurogenesis back on? The answer is not about growing a new brain. It is about rewiring the one you have: giving the hippocampus fresh cells that can encode new experiences without the emotional baggage of the old ones. “Without the ability to create new neurons, people with depression may not have the resilience to effectively adapt to the environment,” Dupont says. The next step is figuring out how to restart the process — and who, among the many faces of depression, might benefit most.

FAQ