Research has revealed that astrocytes play a leading role in eliminating unnecessary neural connections during the process of acquiring motor skills. The study suggests that astrocytes read dopamine signals to selectively organize synapses, and that this process is directly linked to actual motor learning ability.

The research team led by Deputy Director Won-Seok Jung of the Center for Vascular Research at the Institute for Basic Science (IBS) announced that the results were published online in the international academic journal 'Nature Communications' on February 23. Researchers from KAIST and UNIST participated in the joint study.

The creation and elimination of synapses have previously been explained primarily through neurons. Although it has recently been revealed that astrocytes and microglia are involved in synapse regulation, the specific operating principles in actual motor learning have not been clearly elucidated.

The research team tracked the repetitive motor learning process in mice and observed synaptic changes using high-resolution imaging. As a result, phagocytosis by astrocytes increased as learning progressed, and this was mediated by the phagocytic receptor MEGF10. Other glial cells were hardly involved in synaptic clearance during motor learning.

In addition, increasing neural activity in the cerebral motor cortex also increased synaptic removal in astrocytes. When dopamine signals were regulated, opposing responses were observed in medium spiny neurons of the striatum. In D1 receptor cells, synaptic removal decreased, strengthening connections, while in D2 receptor cells, removal increased, reorganizing circuits. This demonstrates that astrocytes are involved in selective remodeling in response to dopamine signals.

Deputy Research Director Won-seok Jeong explained that learning involves the process of creating new synapses as well as the process of rewiring to remove unnecessary connections, stating that this study has identified the central mechanism. The research team explained that these results provide foundational data for understanding dopamine-related diseases such as Parkinson's disease.