Discovery in mice sheds light on how the brain learns to adjust how we walk (video)

Written by Dr. Ambika Tewari Edited by Dr. Sriram Jayabal

New research identifies the cell type in the cerebellum that is vital for a specific form of motor learning

Locomotion – the process of moving oneself from one place to another – is highly adaptive. Depending on our current needs, we can alter the way we walk (known as our gait) without much trouble. For instance, we might increase our speed to get to a meeting in time or, if we have time for a more relaxing stroll, reduce our speed. This locomotor adaptation may seem effortless, but it actually involves a high level of coordination. It is quite apparent when witnessing a toddler trying to walk that figuring out our adaptive mechanisms plays an important role in fine-tuning movements. Because of this, determining how locomotor adaptation works has become a focus of research in the field of rehabilitative therapy, especially with patient populations that experiences gait deficits.

Young toddler boy learning how to walk and balance
Photo of a toddler how to walk. How does of brain learn how to fine tune our movements to help us balance? Photo by Aleksandr Balandin on Pexels.com

In an effort to better understand the intricate details of locomotor adaptation, researchers at the University of Portugal recently performed a study using adult mice. In this study, mice performed a task on a specialized piece of equipment called a split-belt treadmill, which consists of two separate belts running parallel to each other. The speed of these belts can be independently controlled, allowing researchers to impose different demands to the limbs on the right side of the body versus the left. Though split-belt treadmills are used in rehabilitative therapy for patients with post-stroke hemiparesis (where one side of the body is weakened after stroke), this was the first study that adapted the use of this treadmill for mice.

Using the mouse as a model system allowed the researchers to take advantage of additional tools to investigate the precise role of the cerebellum in locomotor adaptation. The walking task consisted of three phases: 1) a baseline period where the two belt speeds were equal, 2) an adaptation phase where one belt was moving at twice the speed as the other, producing asymmetry in the way the mice walked, and 3) a washout phase where the two belt speeds were returned to baseline conditions to determine whether this gait asymmetry effect persisted. The results showed that, depending on the speed applied to each limb, mice learned to adapt their step length (the difference in position between both front limbs or both back limbs) but not stride length (the difference in position when a single limb lifts off to when it touches the ground). When these adaptations in step length occurred, mice exhibited pronounced after-effects that slowly dissipated when the belts were returned to equal speeds during the washout phase. In addition, the data showed that both spatial (how far, as in “space”) and temporal (how fast, as in “time”) components of gait contribute to step length symmetry.

What part of the brain is needed for this learning? Previous studies have implicated the cerebellum in split-belt treadmill learning in humans. Altered cerebellar function typically results in ataxia or loss of muscle coordination. Notably, the movement abnormalities seen in human ataxic patients are translatable to other animal species, including mice, and can therefore be studied in detail. Two established mouse models of ataxia, Purkinje cell degeneration and reeler mice (which both have an abnormal cerebellum), were subjected to the same split-belt treadmill protocol to test their locomotor adaptation. Both mice showed severe impairment in motor learning: both spatial and temporal components of learning completely disappeared. Mice with problems in their cerebral cortex, another part of the brain that controls aspects of movement, showed no impairments in spatial or temporal adaptation. This suggests that the driver of locomotor adaptive learning is the cerebellum and not the cerebral cortex.

To further localize the precise region of the cerebellum that drives motor learning, these researchers turned to chemo-genetics, a method of targeting a precise cell population using specific drugs. Using this technique, they turned off individual cell populations in the cerebellum one-by-one to evaluate their contribution to locomotor adaptive learning. The overall structure of the cerebellum consists of the cerebellar cortex and the deep cerebellar nuclei. The cerebellar cortex contains Purkinje cells, the neurons that act as the “protagonist” of the cerebellar cortex, which process sensory information from the environment. Purkinje cells then communicate this information to the deep cerebellar nuclei, which provide the major output from the cerebellum to coordinate movement. The deep cerebellar nuclei include the fastigial nucleus, the interposed nucleus, and the dentate nucleus. Using chemo-genetics, researchers silenced the Purkinje cells that communicate with one of the three aforementioned nuclei of the deep cerebellar nuclei, one at a time, and tested their locomotor adaptive behavior using the split-belt treadmill. These elegant experiments showed that only the Purkinje cells targeting the interposed nucleus of the deep cerebellar nuclei were necessary for step length adaptive learning.

This study sheds light on a specific form of locomotor learning – step length adaptation – which is controlled by a specific cell type in cerebellum – the Purkinje cells that target the interposed nucleus. Though the researchers made this finding in mice, it is worth noting that the process of mouse locomotor learning is very similar to human locomotor learning. Because of this, they day might soon come when these findings make an impact in the clinic, specifically in the design of effective new rehabilitative therapies to improve the locomotor ability of patients with gait disorders.

We highly recommend watching this video abstract by Darmohray and colleagues where they explain the main findings of their paper.

Key Terms

Cerebellum: A primary area of pathology in the spinocerebellar ataxias. This brain region sits toward the back of the skull and, though small in stature, contains the majority of the nerve cells (neurons) in the central nervous system. Contains the circuits that fine-tune our movements, giving us the ability to move with precision

Cerebral Cortex: The wrinkly top layer which surrounds the brain, made up of tightly packed groups of neurons. The cerebral cortex is thought to play a key role in making decisions, higher order thinking, and interpreting information from our senses.

Locomotion: The process of moving from one place to another

Purkinje Neurons: A type of neuron in the cerebellum. They are some of the largest cells in the brain. They help regulate fine movement.  Purkinje cell loss/pathology is a common feature in cerebellar ataxia.

Reeler Mouse: A type of mouse model with an underdeveloped cerebellum. They were named after its characteristic staggering gait which is described as “reeling”.

Split-Belt Treadmill: A treadmill where two belts run next to each other, one for the left leg and one for the right. The belts can be run at different speeds from each other. For example: the left belt could be running faster than the right.

Conflict of Interest Statement

The authors and editor declare no conflict of interest.

Citation of Article Reviewed

Darmohray, D.M., Jacobs, J.R., Marques, H.G. and Carey, M.R., 2019. Spatial and temporal locomotor learning in mouse cerebellum. Neuron. https://www.ncbi.nlm.nih.gov/pubmed/30795901