Artificial Syncing and the Role of Myelin in Learning

Artificial synchronization improves myelin-related learning deficits in mice

Posted Jan 12, 2020

Learning has long been a neuron-dominated domain—a process governed by the strengthening and weakening of the connection points between neurons. In the last decade, however, the essential role of myelin in the learning process has emerged. Studies where scientists impede a mouse’s ability to generate new myelin show they cannot learn complex motor tasks (like running on a wheel with irregularly spaced rungs) or learn and recall the location of a platform in a water maze, as well as their proper-myelin-forming counterparts.

A study published this month in the journal Glia, adds to the growing body of knowledge surrounding the role myelin plays in learning. The study, led by Daisuke Kato of the National Institute for Basic Biology in Japan, shows the effect myelin abnormalities have on neuron activity while a learned task is being performed. The group also shows that myelin abnormalities can be partially mitigated by artificially synchronizing the neural circuits involved in the task. Their findings are important for better understanding the role of myelin in learning and for developing therapies that can overcome myelin-related deficits in learning that may accompany diseases where myelin is affected. These include multiple sclerosis, Alzheimer’s disease, depression, and many others.

The authors of the Glia study used a mouse with a mutation in the proteolipid protein gene. Proteolipid protein (PLP) has two main functions: embedded in myelin, it holds concentric layers together to maintain its proper structure. The protein is also involved in maturing an oligodendrocyte precursor cell into an oligodendrocyte that can generate new myelin.

Myelin Deficit, Learning Deficit 

Kato and his colleagues first tested the hypothesis that mice with abnormal myelin (i.e. mice with mutated PLP) would not be able to learn a motor task as well as mice with normal myelin. In a learning session, mice had to learn how to pull a lever and hold it to get a drop of water (outside the short learning session, mice had free access to food and water). The researchers found that mice with the mutated myelin protein performed just as well as the others in the early training stages but fell behind mice with normal myelin later on.

When the scientists looked at the number of new oligodendrocytes and markers of myelin generation (e.g. MBP mRNA) in the brains of the mice in the study, they found mice with normal PLP had markers of myelin development that correlated with their improved performance in the task. “These results suggest the increase in new myelinating oligodendrocytes production and MBP mRNA expression were critical for the normal motor learning which was impaired in the [mice with mutated PLP],” the authors of the study concluded.

More Background Noise

Even a task as simple as pulling a lever to get water involves many different areas of the brain. The mouse has to want something and come up with a strategy for getting it, which requires the anterior frontal cortex. The mouse uses its temporal cortex to recall past strategies it has used that will help guide its current strategy. And the mouse has to know where it is in relation to the lever it will pull, information calculated in the parietal cortex. All this is happening at the same time. Neurons in these different areas of the brain are firing simultaneously, generating electrical impulses that travel along nerve fibers and converge on neurons in the motor cortex.

Activity in the neurons of the motor cortex is required for motor learning, they initiate the action of pulling the lever. The researchers wanted to know the effect abnormal myelin had on these neurons. They investigated by watching calcium. Whenever a neuron is active, the concentration of calcium increases as the ion rushes into the cell and is released from internal stores. The scientists inserted a protein into the motor neurons of the mice that would flash a color when calcium was present. By tracking the flashes of color, the researchers were able to see the activity of the motor neurons over time.

Kato and his colleagues noticed the activity of the motor neurons in mice with normal myelin was much more refined. Motor neurons in the mice with normal myelin fired with greater strength when they were performing the task and the motor neurons in these mice also had much less background noise (random activity) when the task wasn’t being performed.

It's All About Timing

The thalamus is the subcortical motor center. It has abundant connections with the motor cortex and other areas of the brain related to movement, it has a known ability to regulate cortical activity, and myelinated nerve fibers connecting the two regions ensure synchronization of cortical inputs during development. The authors of the Glia study hypothesized the source of the increased background noise in mice with abnormal myelin may be the nerve fibers connecting the thalamus and the motor cortex.

The group tested this hypothesis using a technique called optogenetics. Channelrhodopsin is a light-sensitive protein that helps certain species of algae move in response to light. When it’s inserted into neurons, however, it allows scientists to make neurons fire just by shining light on them. The scientists involved in the Glia study put channelrhodopsin into the neurons of the thalamus of mice with abnormal myelin and mice with normal myelin. By shining light on the axons coming from the thalamus, the researchers could measure features of the action potentials (i.e. electrical impulses) coming from neurons in the region. They found action potentials in mice with mutated PLP were slower and more variable.

When they shone light on the neurons in the thalamus and watched the activity of neurons in the motor cortex, they noticed the variable nature of the action potentials coming from the thalamus were causing the neurons of the motor cortex to remain slightly excited for longer. The prolonged excitation could be the reason for the spontaneous activity they had observed before.

Still a myelin deficit, but less of a learning deficit 

For the final series of experiments of the study, Kato and his colleagues reasoned that if asynchronous input into the motor cortex from the nerve fibers coming from the thalamus was responsible for impaired learning, then artificially synchronizing the input when the animals were pulling the lever would lead to an improvement.

They artificially synchronized the nerve fibers connecting the thalamus and the motor cortex by repetitively stimulating the neurons in the thalamus of mice immediately after the mice pulled the lever to get water during the training sessions. The artificial synchronization wasn’t able to completely rescue the learning deficit in mice with abnormal myelin, but it was partially restored. The mice that had been exposed to the repetitive light stimuli after the lever pull experienced more successes in the late-stage training sessions, where they had struggled before. “Optogenetic stimulation of thalamic cell bodies with lever pulling might improve thalamic activity by compensating dispersed inputs from basal ganglia and cerebellum that are crucial for motor learning,” the authors concluded in the paper.

The results of the study are important for better understanding the role myelin and synchronization has on learning and opens the door for therapeutic approaches that may have cognitive disabilities due to myelin deficits. “Our results revealed the pathological neuronal circuit activity with impaired myelin and suggest the possibility that pairing noninvasive brain stimulation with relevant behaviours may ameliorate cognitive and behavioural abnormalities in diseases with impaired myelination,” wrote the authors.

References

Kato, D. et al. Motor learning requires myelination to reduce asynchrony and spontaneity in neural activity. Glia 68, 193–210 (2020)