Year: 2021
Type: dissertation
Abstract: The basal ganglia circuits are critically involved in the acquisition, learning and consolidation of motor skills. The striatum is the major input region of the basal ganglia and is mainly composed of medium spiny neurons (MSNs). MSN dendritic spines represent a major site of synaptic plasticity in the basal ganglia. While MSN functional synaptic plasticity has been shown to occur in the context of motor learning, it remains unknown whether it is accompanied by structural plasticity, i.e., by changes in the number, size and/or spatial pattern of dendritic spines. In this study, we aimed to characterize motor performance and striatal activity upon motor skill learning, with the ultimate goal of studying structural plasticity upon motor learning. We trained mice in a self-paced operant task where a reward is obtained after pressing a lever four times. As training progresses, the position and retractability of the lever impose laterality, allowing the study of movement sequences performed by a single forelimb. During training, mice improved their performance and learned to perform the task with only one forepaw, increasing the total number of lever presses per minute and organizing their behavior in sequences of lever presses. Using video-based paw trajectory detection, we further dissected movement kinematics during task learning, and found that paw trajectory variability during lever press did not significantly decrease throughout training. To identify cells that were active in the last session of the motor task, and therefore more likely to have undergone synaptic plasticity, we performed immunostaining against c-Fos, an immediate early gene commonly used as a neuronal plasticity marker. Using a whole-brain cell detection pipeline, we were able to achieve unbiased cell detection and atlas mapping of c-Fos expressing cells, using fixed brain slices of mice sacrificed 1h after the last training session. Our preliminary results suggest that task-trained animals have subtle changes in c-Fos expression in the contralateral hemisphere to the trained paw. These changes appeared to be consistent across regions involved in motor learning and performance in task-trained animals that include the striatum, primary motor cortex layer 5, and upper limb region of the primary somatosensory cortex. Finally, we sought to establish the tools and techniques needed to study structural plasticity in fixed brain slices. We optimized a novel viral approach to achieve sparse labelling of MSNs, allowing the visualization and reconstruction of the whole dendritic arbor of single MSNs, including dendritic spines. Using high-resolution confocal microscopy, together with deconvolution and spine analysis software, we were able to image and reconstruct MSN dendritic spines and characterize spine morphology. We found that D1-MSNs have a preponderance of mushroom spines, followed by thin and stubby spines, with filopodia making up just a small fraction of overall dendritic protrusions. Our work has established the basic techniques and methodologies that will allow future studies on dendritic spine density, volume and distribution in neurons that recently underwent plasticity upon motor learning. Those experiments will, in turn, advance our understanding of how MSN input computation changes during motor learning, with important implications for the study and treatment of movement disorders.
Author: Ana Margarida Marques Pinto
Language: English
Topic: Motor Control and Adaptation
Subfield: Cognitive Neuroscience
Field: Neuroscience
Domain: Life Sciences
Open Access status: green
OpenAlex ID: W3171807051