The brain is two organs, not one — the split goes back 600 million years

For generations, biologists assumed that every cell of the brain descended from a single ancestral founder — one progenitor that split and differentiated into every region. That assumption turns out to be wrong.

Researchers at Stanford Medicine have shown that the front and back of the brain grow from entirely separate cell lineages, each locked into its own fate before they ever communicate with each other. One population was committed to the forebrain and midbrain; the other was locked into the hindbrain. Both are separated by a molecular boundary that nothing in development ever crosses. They are not two branches of the same tree. They are two different trees that evolution planted side by side in the same skull.

The consequences reach from fundamental biology straight to the clinic. ALS and spinal muscular atrophy both specifically destroy neurons in the hindbrain, a region scientists have barely been able to study in the laboratory because they never knew how to grow those neurons outside the body. That barrier has now fallen.

Recommended ReadingThree brain systems collapse simultaneously between 50 and 75, a hippocampus map finds

Two travellers who never meet

The adult brain has three main regions: the forebrain, which handles thought, language, and consciousness; the midbrain, which relays signals; and the hindbrain (the brain stem), which runs everything that keeps you alive without asking permission — your heartbeat, your breathing, the muscles that let you swallow. A simple hierarchy, one founding progenitor cell populating all of it. So the standard model said.

When the Stanford team examined mouse embryos during gastrulation, the earliest stage when the body begins to organize itself, they found something that collapsed that model. Two different progenitor cell populations were already present, each expressing a different gene. One population, marked by a gene called Otx2, was committed exclusively to the forebrain and midbrain. The other, marked by Gbx2, was locked into the hindbrain. The two cell types never overlapped, never swapped roles, and never produced each other’s cell types.

The evidence came not just from tracking which cells went where, but from examining how the DNA inside each cell type was physically coiled. Chromatin — the packaging that determines which genes a cell can activate — was fundamentally different in each progenitor from their earliest existence. Each cell carried not just a directive to head in a different direction, but a structural inability to go anywhere else.

“Previous attempts to make hindbrain neurons likely tried to coax forebrain and midbrain progenitors into hindbrain cells, which our study shows is not possible,” said Rayyan Jokhai, a graduate student and co-first author of the research. Decades of laboratory frustration had a clean answer hiding at the very beginning of development: researchers had been trying to derive hindbrain neurons from the wrong starting cell.

600 million years of evidence

The finding was not a quirk of mouse embryos. The Stanford team looked back across 550 million years of evolutionary history and found the same two-origin pattern everywhere they looked: in chickens, in zebrafish, in acorn worms — tiny ocean-floor creatures that share a distant common ancestor with vertebrates. Jellyfish, which diverged from our lineage roughly 600 to 700 million years ago, already carry two distinct nervous systems at opposite ends of the body.

The conclusion became hard to avoid: evolution did not invent the hindbrain and forebrain as two branches of a single original brain. It brought two nervous systems that had been evolving independently and gradually placed them together inside the same organism.

“Our research suggests that evolution took two existing neural systems and pushed them together spatially,” said Kyle Loh, the study’s senior author and associate professor of developmental biology at Stanford Medicine. “Having the brain as one organ would probably be more efficient, but we rely on this primordial way to make the brain as two separate pieces.”

“I was surprised at our findings,” added Jokhai, “because the word ‘brain’ implies a contiguous organ that likely has a singular origin. But even 500 million years ago, there were these separate neural systems, which now almost operate as one.”

What it opens

For people living with ALS or spinal muscular atrophy, the practical shift is immediate. In both conditions, the hindbrain neurons controlling swallowing and breathing gradually stop working. Patients lose the ability to swallow safely, raising the risk of pneumonia from inhaled food or liquid, and eventually lose the ability to breathe independently. SMA is the leading genetic cause of death in children under one year of age. ALS is most commonly diagnosed between the ages of 40 and 70.

Scientists knew both conditions attacked the hindbrain. Studying them in a dish required growing hindbrain neurons, which proved nearly impossible — researchers had been trying to derive them from the wrong starting cell. Using the correct progenitor identity, the Stanford team has now for the first time coaxed human pluripotent stem cells into functional hindbrain motor neurons. Those lab-grown cells fired action potentials and expressed the proteins specific to the hindbrain segments that control the face, tongue, and throat.

There is also a less obvious connection: the hindbrain contains the circuits that regulate hunger, which is precisely where drugs like semaglutide, the active compound in Ozempic and Wegovy, exert their effect. The ability to grow and study these neurons in the lab may eventually clarify how those drugs produce their appetite suppression at the cellular level.

What remains open

The developmental split is confirmed. What is not settled is the evolutionary mechanism — exactly how and when two independently evolved nervous systems were brought into contact and knitted together into the organ we call the brain. That story spans hundreds of millions of years and is still being reconstructed.

The current study addressed three brain regions: the forebrain, midbrain, and hindbrain. The developmental origins of the spinal cord are the team’s next research target.

Some caveats apply. The primary work was done in mouse embryos; the team used human pluripotent stem cells to verify the hindbrain finding, but human gastrulation itself was not directly observed. The study did not pinpoint exactly when the Otx2 and Gbx2 populations first become distinct. The ability to grow hindbrain neurons in a dish is a research platform, not a treatment — years of additional work separate it from clinical application.

“Now we have a model to better understand these diseases, and work toward regenerative therapies for them,” said Jokhai.

Common questions about the brain’s two origins

Does this mean you have two separate brains?

Not in the sense that two independent minds share your skull. But developmentally: the hindbrain (brain stem) and the forebrain/midbrain arose from entirely different progenitor cell lineages, distinct from the earliest moments of embryonic development. They function as one organ in adulthood, but they were assembled from two ancient systems that evolution gradually brought together.

What does the hindbrain actually do?

The hindbrain, or brain stem, controls the automatic functions that keep you alive: breathing, heartbeat, sleep regulation, hunger signaling, and the muscles of the face, tongue, and throat that let you swallow and speak. The forebrain runs higher functions: language, reasoning, memory, and consciousness.

Why did it take this long to discover?

The single-progenitor model was never directly tested at gastrulation, the earliest phase of development. Previous laboratory work tried to derive brain stem neurons without first distinguishing between progenitor types and found the attempt unreliable. The Stanford study resolved the question by looking at the moment before any part of the brain had taken recognizable shape.

How does this affect ALS and SMA research?

Both diseases destroy hindbrain neurons that control breathing and swallowing. Scientists could not previously grow these cells reliably in a dish. Without a cellular model, research into both diseases at the molecular level was nearly impossible. The Stanford team has now grown functional human hindbrain neurons for the first time, opening the door to disease models and drug screens for both ALS and SMA.

The study appeared in Nature Neuroscience on September 18. The team’s next target is the developmental origins of the spinal cord.

The practical question is now open in both directions: if the hindbrain’s disease mechanisms can finally be studied in a dish, what else about the brain’s two-organ architecture has been invisible from the wrong starting point?

Reference: Dundes and Jokhai et al., “Separable developmental origins of the forebrain/midbrain and hindbrain,” Nature Neuroscience, 2026.

Tags: neuroscience, ALS, evolutionary biology, brain development, kyle loh, sma

More Like This

Lost brain support cells were thought irreplaceable — adult brains are replacing them

Scientists mapped a fruit fly’s 160,000 neurons — and its legs run themselves

The brain was walled off from immunity — its skull bone marrow had other plans

Human DNA hides two ghost ancestors — one from a species 1.8 million years old

Human Brain Project study offers insights into neurotransmitter receptor organisation

Breeding season inversion: how birds are overriding the ancestral clock