Maybe it was a macrophage that took a wrong turn at the bloodstream. Maybe it was a microglia having a mid-life identity crisis. Maybe somebody sneezed on the microscope slide. For years, whenever a scientist peered into a zebrafish brain and spotted an immune cell that did not quite look like the usual residents, there was a comfortable list of excuses ready to go. The boring answer was almost always "eh, probably just microglia being weird."

The real answer, it turns out, is that the fish brain was running a whole second station in its immune kitchen, and nobody had read the full menu. (Yes, I am going to baste this post in cooking metaphors until it is golden brown. I apologize in advance, but not enough to stop.)

The brain's immune kitchen had a secret sous-chef

Here is the setup. Your brain, and a fish's brain, has a head chef for immune cleanup: microglia. These are the cells that simmer away in the background, tasting for trouble, clearing out debris, and generally keeping the neural broth from spoiling. For a long time, the zebrafish brain was assumed to be a one-cook operation.

A team using single-cell transcriptomics decided to actually inventory every immune cell in the adult zebrafish brain, reading the genetic "recipe card" each cell was following. Single-cell sequencing is basically eavesdropping on what every individual cell is making for dinner, one cell at a time. When they did the headcount, they found microglia, sure, but also a distinct population of myeloid cells carrying a gene signature that looked suspiciously like mammalian dendritic cells (Rovira et al., 2025).

Dendritic cells are the immune system's tasting-and-tattling crew. They sample what is around, then run off to alert the rest of the immune system about what they found. Finding resident dendritic-like cells living inside the brain tissue itself, not just loitering at the borders, was the plot twist.

Same recipe, different cooks

The genuinely neat part is the evolution angle. These zebrafish dendritic-like cells depend on a gene called batf3, which is the exact same ingredient mammals (including us) need to bake conventional type 1 dendritic cells, the cDC1 crew (Grajales-Reyes et al., 2015, DOI: 10.1038/ni.3197). Fish and mammals split off from a common ancestor something like 400 million years ago, which is a long time to keep using the same family recipe.

So the immune system did not invent brain-resident dendritic cells separately in fish and in us. It seems to have hung onto a blueprint that predates basically everything you would recognize as an animal. The CNS immune compartment in mammals is already known to be a crowded, heterogeneous kitchen full of microglia, border macrophages, and dendritic cells (Mildenberger et al., 2023, DOI: 10.1038/s41591-023-02673-1). This work says: that complexity is not a mammal thing, it is an old thing.

Knock out one ingredient, see what burns

To prove these were truly different cells and not microglia in a clever apron, the team built fluorescent reporter fish that light up dendritic-like cells separately from microglia. Then they looked at popular mutant fish (the irf8, csf1ra, and csf1rb knockouts) that researchers have used for years to study brain immunity.

Surprise: microglia and the dendritic-like cells reacted differently to losing those genes. Pull irf8 out of the recipe and the two cell types respond like two separate dishes, not one. That matters because a lot of past zebrafish studies treated "brain myeloid cells" as a single blob. If your knockout is quietly hitting two different cell types in two different ways, your conclusions get a little undercooked.

Why fish-brain bookkeeping matters to you

You might reasonably ask why anyone should care about a tiny tattletale cell in a fish. Fair. But zebrafish are see-through as embryos, breed fast, and let scientists watch immune cells move in a living brain in real time, which is roughly impossible in a mouse and very illegal in a human. The more accurately we can map the fish brain's immune lineup, the better the fish works as a stand-in for studying neuroinflammation, brain injury, and disease in us. Knowing there is a second cook in the kitchen means we stop blaming the head chef for everything that comes out of it.

The brain, as usual, was more crowded and more interesting than the simple story suggested. The cells were always there. We just had not bothered to count the place settings.

References

• Rovira, M., Ferrero, G., Miserocchi, M., Montanari, A., Lattuca, R., & Wittamer, V. (2025). A single-cell transcriptomic atlas reveals resident dendritic-like cells in the zebrafish brain parenchyma. eLife. DOI: 10.7554/eLife.91427 (PMCID: PMC12698085)
• Grajales-Reyes, G. E., et al. (2015). Batf3 maintains autoactivation of Irf8 for commitment of a CD8alpha+ conventional DC clonogenic progenitor. Nature Immunology. DOI: 10.1038/ni.3197
• Mildenberger, W., et al. (2023). Multiomic spatial landscape of innate immune cells at human central nervous system borders. Nature Medicine. DOI: 10.1038/s41591-023-02673-1
• Reizis, B. (2026). Dendritic cell heterogeneity tailored to their function. Immunological Reviews. DOI: 10.1111/imr.70123

Disclaimer: The image accompanying this article is for illustrative purposes only and does not depict actual experimental results, data, or biological mechanisms.