Stanford researchers show that the human forebrain and hindbrain develop from two distinct progenitor cell types and grow human hindbrain neurons that control swallowing, speech and facial expression for the first time

Stanford researchers show that the human forebrain and hindbrain develop from two distinct progenitor cell types and grow human hindbrain neurons that control swallowing, speech and facial expression for the first time

Researchers at Stanford Medicine, Caltech and the University of California, San Francisco traced the fate of hundreds of cells in mouse embryos. Almost all their descendants stayed within their original lineage, and attempts to redirect the cells with signaling cues failed. The findings were published on September 18 in Nature Neuroscience.

For decades, the brain’s anatomical continuity was taken to mean that it developed from a single shared progenitor cell type. Laboratories learned to turn stem cells into forebrain and midbrain neurons, but attempts to produce hindbrain neurons repeatedly failed.

Researchers already knew that the genes Otx2 and Gbx2 become active in different regions of neural tissue at day 7,5 of mouse embryonic development. What remained uncertain was whether these cells were already committed to distinct fates or simply occupied different locations while retaining the ability to become any cell type. Resolving this required direct observation: Kyle Loh, Carolyn Dundes, Ryan Jockhai and colleagues randomly labeled hundreds of progenitor cells with different colors and tracked each cell’s descendants. Among 494 clones in 16 embryos, almost every clone remained entirely within either the forebrain and midbrain or the hindbrain. In a separate experiment, the researchers used a brief 12-hour pulse of the same substance to label cells with active Gbx2 specifically. Their descendants contributed exclusively to the hindbrain, without mixing with the Otx2 lineage.

Experiments in cell culture confirmed the same pattern: signals intended to redirect the progenitors failed both when the cells were cultured separately and when they were cultured together. An analysis of DNA packaging revealed why: by the second day, the same genomic regions already differed in accessibility between the two cell types. Their commitment was reflected both in gene activity and in the physical accessibility of DNA to the machinery that reads it. This explains why earlier protocols, which directed cells toward a forebrain fate by default, could not subsequently redirect them toward a hindbrain fate.

Using the newly identified progenitor, the team grew human hindbrain motor neurons for the first time. The neurons produced acetylcholine and responded to stimulation by firing action potentials. In the body, these neurons control the muscles of the face, tongue and throat, supporting swallowing, speech, facial expression and eye movements.

“The loss of hindbrain motor neurons likely impairs the ability to eat and swallow in spinal muscular atrophy and amyotrophic lateral sclerosis, leading to choking, pneumonia and, in some cases, death,” the authors write in the paper.

The same principle also explains a childhood cancer. Diffuse pontine glioma is the most common cause of cancer death in children, with a five-year survival rate of about 1%. Study coauthor Michelle Monje received The Brain Prize, one of the most prestigious awards in neuroscience, in 2025 for discovering that healthy neurons form synapses with glioma cells and accelerate their growth through electrical activity. The new map of progenitor cells explains why a single mutation in H3F3A produces different tumor types depending on the cell lineage in which it occurs.

The authors found the same separation in chickens, zebrafish, macaques and acorn worms, invertebrates whose ancestors diverged from those of vertebrates 550–600 million years ago.

“Our results surprised me: the word ‘brain’ itself suggests a single organ with a shared origin. Yet these separate nervous systems already existed 500 million years ago and now function almost as a single whole,” said Ryan Jockhai.