UCLA studies find brain-building stem cells respond to fuel supply and physical touch

Two studies published this week by researchers at the David Geffen School of Medicine at UCLA describe previously unrecognized ways that the human brain's master-builder stem cells decide what kind of neurons to make. One paper, in the journal Cell, shows that these stem cells shift their output depending on how they metabolize sugar. The other, in Science, finds that physical contact from a deep brain structure called the thalamus pushes the same stem cells toward making a different set of neurons entirely. Taken together, the two papers recast the immediate surroundings of these cells — their fuel supply and their physical neighbors — as active participants in building the cortex, rather than passive scenery.

The cells at the center of both studies are radial glia, a population of stem cells that generates most of the neurons in the cerebral cortex, the outer layer of the brain responsible for language, reasoning and sensory processing. How radial glia decide which of several possible neuron types to produce, and when, has been a long-standing question in developmental neuroscience, with direct bearing on conditions ranging from autism to brain cancer.

What the two teams found

The Cell paper, led by co-first authors Jessenya Mil and Jose Soto in the labs of Aparna Bhaduri and Heather Christofk, built what the authors call a metabolic atlas of the early human cortex, drawing on donated human tissue and brain organoids grown from stem cells. The team found that radial glia depend heavily on the pentose phosphate pathway, a metabolic route that processes glucose into the raw materials rapidly dividing cells need. When the researchers lowered glucose availability or chemically and genetically blocked the pathway, the stem cells changed course, producing more inhibitory neurons and other cell types that normally appear later in cortical development. Supplying the sugar ribose was enough to reverse several of those changes — restoring gene expression patterns, cell-type proportions in the organoids, and cellular energy levels.

The companion study in Science, led by first author Claudia Nguyen in the Bhaduri lab, took a different approach, fusing lab-grown cortical and thalamic organoids into structures called assembloids to observe how the two brain regions interact as they develop. The thalamus is a relay structure deep in the brain that channels sensory information to the cortex. The researchers found that thalamic projections make direct physical contact with radial glia, and that this contact — not just chemical signaling at a distance — prompts the stem cells to generate more excitatory neurons, particularly the upper-layer neurons that are disproportionately expanded in the human cortex compared with other species. The team traced the effect to a gene called NRXN1, which is best known for helping neurons form connections with one another. When the researchers built assembloids using cells carrying an NRXN1 mutation previously linked to autism spectrum disorder, the thalamic signals no longer behaved normally, throwing off the balance between stem cells and the neurons they produced.

Why this line of research matters

Both findings extend a body of work, some of it from the same UCLA labs, mapping how the developing human brain is instructed at the molecular level. Errors in the timing or balance of neuron production during fetal cortical development have long been implicated in neurodevelopmental and neuropsychiatric conditions, and radial glia themselves are the presumed cell of origin for some pediatric brain cancers, since a stem cell that keeps dividing rather than maturing into a neuron can become a tumor precursor. Because both new studies used tissue and stem-cell models specific to humans, they also speak to a persistent limitation in neuroscience: mice, the standard laboratory model for brain development, appear to lack the thalamic-contact mechanism the Science paper describes, meaning some human-specific features of cortical growth cannot be studied in rodents at all.

The research is not a study of patients or a clinical trial; it relies on organoids and assembloids, three-dimensional clusters of cells grown from stem cells that approximate, but do not fully replicate, a developing brain, along with donated human tissue samples. Findings in these systems can diverge from what happens inside a living fetus, and neither paper claims to have identified a treatment or a diagnostic test. The immediate audience for the work is researchers studying autism spectrum disorder, other neurodevelopmental conditions, and pediatric brain tumors, along with scientists building more accurate organoid models of the human brain.

Researchers describe the findings as a shift in framing

Bhaduri, an assistant professor of biological chemistry and a senior author on both papers, said the metabolic finding surprised her team because it assigns an active role to a process usually treated as housekeeping.

"What was surprising is that metabolism isn't just a passive thing that happens in the background. It can really control how stem cells make decisions," Bhaduri said, according to a UCLA news release describing the two studies.

On the thalamic contact finding, Bhaduri said the physical nature of the interaction was the new piece. "We already knew that these projections influence how the cortex develops," she said. "What we specifically found is that this influence comes through an actual physical connection" — one that, she noted, very likely does not exist in rodents, underscoring why human-tissue and organoid models were necessary to see it at all. The findings were also covered independently by ScienceDaily and by Medical Xpress, both of which summarized the two papers as complementary evidence that a developing stem cell's environment, not just its internal genetic program, steers what kind of brain cell it becomes.

What comes next

Both teams say their next step is to test whether manipulating metabolism or thalamic contact in organoid models can correct neuron-production imbalances seen in models of autism spectrum disorder or other neurodevelopmental conditions, and whether the same pentose phosphate pathway dependency shows up in radial-glia-derived brain tumors, where it could eventually be a target for treatment. The work was funded by the National Institutes of Health, the National Science Foundation, the Brain & Behavior Research Foundation, the Alfred P. Sloan Foundation, the Rose Hills Foundation, the Esther A. and Joseph Klingenstein Fund, the Simons Foundation, the Chan Zuckerberg Initiative, the UCLA Broad Stem Cell Research Center and the UCLA Health Jonsson Comprehensive Cancer Center, according to the university. Neither paper reports human clinical data, and researchers outside the two labs will need to confirm the mechanisms in independent tissue samples and organoid lines before the findings can be considered established science rather than a single group's result.