Scientists Say the Brain Is Two Organs, Not One

Grow a dish of human stem cells and coax them, patiently, into brain cells, and you’ll get cortex without much trouble. Neurons that think, remember, plan: those cells oblige. Ask the same dish for a neuron from the hindbrain, the ancient stalk at the base of the skull that keeps a person breathing and swallowing, and for over a decade the answer has been a shrug. Stanford University researchers have now traced why: the forebrain and midbrain descend from one progenitor cell in the earliest embryo, and the hindbrain descends from an entirely separate one, and the two lines have never been interchangeable. The brain, in other words, is stitched together from two different starting materials.

That split explains a specific, stubborn failure. It also opens a door that has stayed shut through years of trying.

The team, led by Kyle Loh and reported by graduate students Rayyan Jokhai and Carolyn Dundes as co-first authors, went looking in mouse embryos barely a week old, at the exact moment the ectoderm layer first commits to becoming brain rather than skin or gut. They found two mutually exclusive populations sitting side by side: cells marked by a gene called Otx2, fated for the forebrain and midbrain, and cells marked by a different gene, Gbx2, fated only for the hindbrain. Genetic tracing that labeled the Gbx2 cells with a fluorescent tag, then followed those cells for days, found their descendants restricted entirely to the hindbrain at every stage checked, all the way out to birth. Sparse-labeling of nearly five hundred individual cell clusters across sixteen mice confirmed the pattern held almost without exception.

A label is not proof of commitment, though. A cell can wear the right badge and still be talked into changing careers if the right signal shows up.

So the researchers tried to talk the cells into changing careers. They took the hPS cell-derived version of each progenitor and forced it to sit in signals meant for the other fate: forebrain-inducing chemistry poured onto hindbrain-bound cells, and vice versa. Neither budged. The hindbrain progenitor resisted forebrain signals and built a hindbrain-like chromatin landscape anyway; the forebrain progenitor did the mirror image. Even mixed together in the same dish, sharing a culture medium, each type stuck to its assigned fate. The commitment, it turned out, was already written into which genes sat open and accessible in the chromatin, before either cell had made a single neuron.

“We’ve shown for the first time that the front of the brain arises from a totally different progenitor cell than the back of the brain,” says Loh, an associate professor of developmental biology at Stanford. “Our discovery means that we can now grow neurons from the back of the brain, the hindbrain, in a petri dish and study their functions.”

Having identified the correct starting cell, the team pushed it forward. Human pluripotent stem cells, guided first into the posterior neural ectoderm and then into hindbrain progenitors, went on to become motor neurons carrying the molecular fingerprints of rhombomeres 5 and 6, the hindbrain’s own internal segments, the ones whose neurons reach out through the cranial nerves to run the muscles of the face, tongue, and throat. These weren’t just correctly labeled cells sitting inert under a microscope. They fired action potentials when current was injected. They lit up with calcium signals on their own, unprompted, the electrical chatter of a working neuron. They could even be switched on with a pulse of light once fitted with a light-sensitive protein, the same optogenetic trick neuroscientists use to control neurons in a living brain. Rhombomere 5 and 6 neurons specifically matter because they are the ones that degenerate in spinal muscular atrophy and in some presentations of amyotrophic lateral sclerosis, leaving patients unable to swallow safely, at risk of choking and pneumonia, and eventually unable to breathe unassisted.

What the Data Cannot Rule Out

None of this proves the split is permanent under every possible condition. The authors are careful to note what their experiment cannot rule out: a vanishingly brief, unlabeled pan-brain progenitor might exist for a matter of hours before the fork appears, a shared root so short-lived it slipped past every method they used to look for it. And a challenge assay run in a dish is not the same as testing a cell inside the intact, signal-soaked chaos of a living embryo, which the authors admit is far harder to attempt directly.

There is also a piece still missing entirely. Nobody yet knows where the cerebellum, tucked at the very front edge of the hindbrain, actually comes from. It might belong to either lineage, or it might turn out to need a separate, still-undiscovered progenitor. Loh’s team leaves that question open rather than force an answer their data doesn’t support.

An Architecture Older Than the Brain Itself

What holds up, and holds up unexpectedly far, is how old the split appears to be. The researchers found the same two separate progenitor populations, marked by the same two genes, in gastrulating chicken embryos, in zebrafish, and in macaque, then followed the trail into a hemichordate: the acorn worm, a creature that looks nothing like a vertebrate and shared a common ancestor with us roughly 550 to 600 million years ago. “Our research suggests that evolution took two existing neural systems and pushed them together spatially,” Loh says. “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.” The architecture predates the brain that architecture eventually builds.

For a disease with no cure and a five-year median survival after diagnosis, and for infants whose lives are shortened by a single missing protein, a dish of real, electrically alive hindbrain neurons is not itself a treatment. It is a laboratory bench that did not exist before, one where a scientist can now watch rhombomere 5 and 6 motor neurons degenerate in real time, test a compound against them directly, and ask why these particular cells, and not others, are the ones that fail. “Now we have a model to better understand these devastating diseases, and work toward regenerative therapies for them,” says Jokhai. “This is a very exciting new frontier in brain research.”

Two progenitors, hundreds of millions of years apart in the fossil record and a few dozen microns apart in a week-old embryo, still keep to their own lane. The brain we experience as seamless was never built as one thing to begin with, and knowing that now changes what a dish of stem cells is finally able to become.

FAQ