The Developing Brain Appears to Remove Neurons With Heavily Altered Genomes
A peer-reviewed Nature study of 2,357 human neurons finds widespread genomic copy-number changes are common before birth but far rarer afterward, pointing to a perinatal quality-control process.
Quick summary
- A peer-reviewed study examined 2,357 single neurons from human brains spanning fetal development to after birth.
- About 25.9% of prenatal cortical neurons carried large copy-number changes across their genomes, with a 4.5-fold lower proportion after birth.
- The authors infer that the developing brain selectively removes many heavily altered neurons around birth, but the postmortem data do not directly track individual cells over time.
The human cerebral cortex appears to run a surprisingly strict quality-control step around birth. In a study published in Nature on October 28, researchers found that neurons carrying extensive gains or losses of DNA segments were common before birth but much rarer in postnatal brains.
The result does not mean that a quarter of a baby's brain is defective. It shows that genetic variation between individual brain cells is broader during development than scientists had appreciated, and that many of the most heavily altered cells may be removed as neural circuits mature.
What the researchers measured
The team analyzed 2,357 individual neurons from human postmortem tissue, using two single-cell whole-genome sequencing methods across separate donor groups. They looked for copy-number variants, or CNVs, in which large stretches of DNA are duplicated or deleted within a cell.
Across prenatal cortical samples, an average of 25.9% of neurons carried widespread CNVs, although the proportion varied from 12.7% to 46% between samples. After birth, that proportion was roughly 4.5 times lower. A separate multiomic analysis of 44,861 nuclei connected the genomic changes with altered gene activity, particularly in pathways involved in synapses and neural communication.
The evidence points to a developmental filtering process: neurons with the most disruptive genome changes may be less able to join functioning circuits and are therefore more likely to disappear.
This builds on an earlier preprint, but the October publication is the peer-reviewed version with a larger single-neuron dataset and additional analyses. That distinction matters because the new paper is best read as strengthened evidence for the mechanism, not the first public disclosure of the idea.
Why birth may be a turning point
Developing brains routinely produce more neurons and connections than they retain. Cells compete for survival signals, while programmed cell death and circuit refinement remove cells that do not integrate successfully. The study proposes that this familiar developmental process may also act as a form of genome quality control.
Neurons with extensive CNVs showed unusual expression of genes linked to synaptic function. That could place them at a competitive disadvantage when circuits become more active around birth. The exact biological route remains unresolved, however, and the researchers did not observe the same neuron before and after birth.
The finding complements earlier NewTqnia coverage of how signals from the thalamus can influence the production of cortical neurons. Read how a deep-brain signal changed neuron production in human stem-cell models for another view of the mechanisms shaping the cortex.
What this could mean for brain disorders
Somatic mosaicism, in which cells in one person carry different genetic changes, has been studied in epilepsy, autism and other neurological conditions. The new work raises a focused question: do some conditions involve altered production or incomplete removal of neurons with disruptive CNVs?
That is a research direction, not a clinical conclusion. The study did not diagnose disease, identify a treatment or show that the surviving altered neurons cause a particular disorder. Its samples were also cross-sectional postmortem specimens, so the proposed elimination process is inferred from differences across developmental ages.
Reality check
This is a strong cellular and genomic study, but it does not prove how each altered neuron dies, whether the same filtering rate applies across all people, or whether changing that process would be safe or beneficial. The measured percentages describe cells in sampled tissue, not a prediction for an individual child. Future work will need longitudinal model systems and disease-specific samples to test the mechanism.
The larger picture
The study recasts genomic variation in the developing brain as a dynamic population problem. Early neural development may tolerate, or simply generate, many cells with substantial DNA changes, then narrow that diversity as functional circuits assemble.
If confirmed, that would add genome integrity to the list of properties the developing brain appears to test during circuit formation. It also offers a more precise framework for investigating why some genetically unusual neurons persist, and when their survival matters.
Verified topics and entities
Sources and citations4 sources
External references used to support the reporting in this article.
Published by
NewTqnia Science Desk
An institutional editorial team within NewTqnia