Maternal immune activation during pregnancy has been associated with an increased risk of neurodevelopmental conditions in offspring. But an important question remains: how do inflammatory signals actually affect the developing human brain?
Researchers have now developed “cerebroids”, small 3D pieces of developing human cerebral cortex maintained outside the body, to investigate this question directly (Figure 1).
Figure 1: Establishment and validation of a 3D ex vivo culture of human fetal brain. a, Schematic illustration of the experimental paradigm. The DLPFC was dissected, cultured and processed for analysis. Illustration was partly generated using Servier Medical Art, provided by Servier, licensed under a Creative Commons license CC BY 4.0. b, Representative immunofluorescence images of cerebroids cultured for 8 d and 15 d, showing the layered organization of developing brain tissue. Dotted lines indicate the ventricular surface. c, Left: wholemount live image of adenovirus-GFP-infected cerebroids after 3 days in culture. Right: immunofluorescence image of a section from day 8 cerebroids showing GFP+ NSCs. d, TEM images of day 2 cerebroids illustrating NSCs lining the ventricular surface (top); higher-magnification view of the apical surface (bottom) showing mitochondria (mt), centrioles (*), apical junction (white arrow) and secretory vesicles (black arrow) e, Immunofluorescence image of proliferating (Ki67+) NSCs in cerebroids cultured for 8 d. f, Quantification of the number of Ki67+ cells in fetal cortex (n = 3 donors) versus day 8 cerebroids (n = 9 cerebroids from 3 donors). Data are presented as mean ± s.e.m. Unpaired t-test (two-sided). g–j, EdU pulse–chase experiment on cerebroids. Schematic of the EdU pulse–chase experimental design (g), with maximum intensity projection images showing mitotic (pH3+), proliferating (Ki67+) and EdU+ cells (h), EdU+TBR2+ intermediate progenitors (i) and EdU+CTIP2+ and MAP2+ neurons (arrows) in cerebroids cultured for 2 d (j). k, Immunofluorescence images of IBA1+ and CD68+ microglia in cerebroids cultured for 8 d. l, Magnified image of a CD68+ microglial cell. m, TEM image of a microglial cell, with pseudocoloring of the nucleus (brown) and cell body (yellow) in cerebroids cultured for 8 d. n, Immunofluorescence images of PDGFRβ+ pericytes and CD31+ endothelial cells in cerebroids cultured for 8 d. o, TEM image of a blood vessel in day 2 cerebroids, showing endothelial cells (EC), tight junction (arrow), pericytes (Per), pericyte peg-sockets (*), radial glia processes (p) and mitochondria (mt) (also see Extended Data Fig. 2h–j for higher-magnification images of the same). Scale bars: 500 μm (c (left)), 100 μm (b, c (right), e, h–k, n), 5 μm (d (top), l, m, o) and 500 nm (d, bottom). Ad-GFP, adenovirus-GFP; CP, cortical plate; D, day; RGC, radial glial cell; VZ, ventricular zone.
Their findings identify a potential pathway linking the inflammatory cytokine IL-17A to altered cortical development through sustained activation of NF-κB signalling.
Brain organoids have become an important tool for studying human neurodevelopment, but they are simplified systems and may not fully reproduce the cellular organisation and developmental architecture of the human brain.
Cerebroids take a different approach.
Rather than generating brain-like tissue entirely from stem cells, the researchers maintained ex vivo human fetal cortical tissue in a controlled 3D culture system.
The researchers used tissue from the dorsolateral prefrontal cortex and exposed these cerebroids to IL-17A, an inflammatory cytokine strongly associated with maternal immune activation.
Exposure to IL-17A caused the developing cortical tissue to:
- Fold prematurely
- Become thicker
- Produce neurons more rapidly
- Accelerate neuronal maturation
These changes suggest that inflammatory signalling doesn’t simply cause nonspecific cellular stress, it can potentially alter the timing and trajectory of neurodevelopmental programmes.
The researchers found evidence that IL-17A acts directly on neural stem cells within the developing cortex.
One of the key pathways activated was NF-κB, a major inflammatory transcriptional signalling pathway. NF-κB is best known for coordinating inflammatory gene expression, but it also has important roles in cell survival, differentiation and development.
In the cerebroids, IL-17A produced sustained NF-κB activation, providing a potential molecular bridge between maternal inflammation and altered neural development. And importantly, when the researchers blocked this pathway, many of the developmental abnormalities were reversed.
That provides experimental evidence that NF-κB signalling isn’t merely associated with the observed changes, it may be an important mediator of them.
The developing cerebral cortex is an extraordinarily tightly regulated system. Inflammatory signals arriving during this period could potentially shift that balance. Accelerating neuronal production or maturation may sound beneficial, but in a developing brain, developmental timing is itself a critical biological variable.
A process occurring too early can be just as disruptive as one occurring too late. The study therefore raises the possibility that inflammatory exposure could alter not simply how many neurons are produced, but when and how they develop and integrate into cortical architecture.
The researchers combined several approaches, including microscopy, cell labelling, RNA sequencing and proteomics, to investigate the effects of IL-17A.
This allowed them to move from an anatomical observation:
IL-17A → abnormal cortical folding
towards a potential molecular mechanism:
IL-17A → neural stem cell signalling → sustained NF-κB activation → altered neuronal production/maturation → disrupted cortical development
This mechanistic framework is particularly important because maternal immune activation involves many inflammatory mediators, cell types and systemic changes. IL-17A may therefore represent one component of a much larger network rather than the sole explanation.
The findings do not demonstrate that maternal inflammation causes autism or other neurodevelopmental conditions. The researchers studied human fetal cortical tissue in an experimental ex vivo system and exposed it directly to IL-17A.
That is very different from reproducing the complex physiological environment of pregnancy, where cytokine concentrations, placental signalling, maternal metabolism, fetal responses and immune-cell interactions are all interconnected. Instead, the study provides something more fundamental: a tractable human model for investigating how specific inflammatory signals can directly influence developing neural tissue.
Rather than asking only whether inflammation is associated with altered neurodevelopment, cerebroids provide a way to ask:
Which inflammatory signals act on which cells, through which pathways, and at what developmental stage?
Understanding the molecular pathways that make the developing brain vulnerable to inflammation could eventually help identify biomarkers of developmental risk or potential strategies for protecting neurodevelopment, although such applications remain firmly in the future.
Maternal immune activation may influence the developing brain through direct effects of inflammatory cytokines on neural developmental programmes.
Cerebroids could therefore become a powerful bridge between simplified organoid models and the complexity of the developing human brain, helping us understand how immune signals intersect with neurodevelopment long before disease becomes clinically apparent.
Journal article: Assir M.Z.K. et al. 2026. Modeling maternal immune activation in 3D ex vivo human fetal brain cerebroids reveals IL-17A-driven disruption of cortical development. Nature Neuroscience.
Summary by Stefan Botha
