Study of 90,000 spinal neurons links a small cell group to walking speed

Walking looks effortless, but the spinal cord must coordinate the timing and position of moving limbs with remarkable precision. A map of nearly 90,000 mouse neurons now points to a small group with a particularly focused job: helping set movement speed.

Scientists at St. Jude Children’s Research Hospital created a single-cell atlas of V1 interneurons, inhibitory spinal neurons involved in motor control. Their analysis identified 14 molecular clusters and connected one subgroup with the speed of rhythmic locomotion.

The findings suggest that controlling movement speed and controlling limb flexion may involve different V1 neurons.

“Understanding the role of different neurons gives us better insights into how these brain-to-muscle neuronal circuits work, which may be important for helping aid recovery after spinal cord damage,” said corresponding author Jay Bikoff of the St. Jude Department of Developmental Neurobiology.

V1 interneurons run within spinal circuits that process signals from the brain and peripheral nervous system before influencing motor neurons.

Earlier experiments showed that removing the V1 population slows rhythmic locomotion and causes excessive limb flexion. What remained unclear was whether one V1 cell type handled both effects.

The team approached that question with single-nucleus RNA sequencing, which measures patterns of gene activity inside individual cell nuclei.

Researchers analyzed V1 interneurons from mice at birth and at 14, 28, and 56 days of age. After quality control, their developmental dataset contained 89,494 V1 nuclei.

The cells separated into 14 clusters based on gene-expression patterns. The largest cluster contained 22.9 percent of the neurons, while the smallest represented 2.2 percent.

Four previously recognized major groups were marked by Foxp2, Pou6f2, Sp8, and MafA. Only 39 of the 89,494 nuclei expressed markers from more than one major group.

The atlas also identified another group marked by Rnf220. Together, five molecular clades accounted for more than 95 percent of V1 interneurons.

Even neurons within the same major group carried different molecular signatures.

Researchers found distinct combinations of transcription factors, proteins that help regulate which genes cells use. Ion channels also varied considerably between clusters.

Among the most strongly differing ion-channel genes were 13 from voltage-gated potassium channel families and 10 ionotropic glutamate receptors. Several calcium and TRP channel genes also differed.

One V1Pou6f2 subgroup stood apart especially strongly.

That cluster expressed high levels of Piezo2, a mechanoreceptor connected with sensing mechanical information. Researchers confirmed Piezo2 expression in V1 neurons using genetically labeled mice.

The atlas therefore provided more than a list of neuron types. It highlighted molecular features that could contribute to differences in how those neurons respond to signals.

Researchers also asked whether these neuron categories disappeared or reorganized as mice grew.

They did not.

Cells from all four ages appeared throughout the 14 clusters in similar proportions. No major new V1 populations emerged with age, and the diversity present near birth persisted into adulthood.

Individual genes still changed substantially.

The largest shifts occurred during the first two weeks after birth. Genes connected with neuron development and synapse organization appeared more strongly early in life, while genes involved in synaptic transmission became more prominent later.

For example, Sema6d, which is associated with guiding developing nerve fibers, declined with age. Snap25, involved in neurotransmitter release, increased.

Those developmental changes occurred within cellular identities that otherwise remained stable.

The team next compared the atlas with V1 neurons from mice lacking Engrailed1, or En1.

Removing En1 barely changed the overall number of V1 cells. Control tissue contained 268 plus or minus nine V1 neurons per section, compared with 265 plus or minus seven in knockout mice.

But the composition changed.

Sequencing 7,553 nuclei from control and En1-knockout animals produced 11 clusters. Ten remained represented at roughly normal levels.

One cluster dropped sharply.

That population belonged to the V1Pou6f2 group and expressed another marker, Nr5a2. Tissue measurements showed V1Nr5a2 neurons falling from 6.3 percent of V1 cells in controls to 0.1 percent after En1 loss.

V1Pou6f2 neurons overall declined from 13.9 percent to 8.4 percent.

“We found that slowed locomotor speed and hyperflexion were separable, suggesting they may be controlled by different cells,” Bikoff said.

The movement experiments clarified why that selective loss mattered.

When researchers removed V1 neurons broadly, locomotor-like rhythm slowed dramatically. Frequencies of roughly 0.38 to 0.44 hertz under control conditions fell to 0.13 hertz.

Those animals also developed severe hindlimb hyperflexion.

Researchers tracked ankle and knee positions and recorded significantly reduced joint angles, confirming that broad V1 loss affected both movement rhythm and flexion.

Removing En1 produced a different result.

Animals lacking the gene showed slower locomotor-like activity. Yet mice with En1 removed specifically from the spinal cord did not develop the same hindlimb hyperflexion.

Their ankle and knee angles remained similar to controls.

That separation implicated the En1-dependent V1Pou6f2 population in regulating locomotor speed while indicating that other V1 cells contribute to flexion and extension.

“Intriguingly, we found that loss of one small subpopulation of V1 interneurons, V1Pou6f2, was associated only with slowed speed, implicating these neurons in controlling the speed of our body’s rhythmic movements but not flexion or extension,” Bikoff said.

These studies examine spinal-cell diversity, connections from the brain and the role of inhibitory neurons in movement and disease.

A brain-wide map of descending inputs onto spinal V1 interneurons: This study maps brain inputs to V1 neurons and compares connections reaching molecularly defined subsets. (Neuron, 2025)

Spinal V1 inhibitory interneuron clades differ in birthdate, projections to motoneurons, and heterogeneity: This research examines differences in developmental timing, motor-neuron connections and sensory inputs across V1 groups. (eLife, 2024)

Spinal inhibitory neurons degenerate before motor neurons and excitatory neurons in a mouse model of ALS: The study investigates changes in inhibitory spinal neurons during disease progression in an ALS mouse model. (Science Advances, 2024)

A harmonized atlas of mouse spinal cord cell types and their spatial organization: This broader atlas places molecularly defined spinal-cell populations within their anatomical organization. (Nature Communications, 2021)

Regulation of locomotor speed and selection of active sets of neurons by V1 neurons: Experiments in zebrafish investigate how V1 neurons influence locomotor speed and recruitment of motor circuitry. (Nature Communications, 2019)

Research findings are available online in the journal PubMed Central.

The original story "Study of 90,000 spinal neurons links a small cell group to walking speed" is published in The Brighter Side of News.

Like these kind of feel good stories? Get The Brighter Side of News' newsletter.