Connectivity of the Posterior Cerebellum: Transsynaptic Viral Tracing with Light-Sheet Imaged Whole-Mouse Brains

Abstract

The cerebellum is conserved across vertebrate evolution, representing a constant mammalian brain fraction. Classically, the cerebellum has been associated only with motor functions. However, the nonmotor role of the cerebellum has recently begun to attract greater attention; currently, its role in brain processing is poorly understood. Cerebellar disruption has consequences beyond motor coordination. Injuries to the posterior cerebellum produces a cognitive-affective syndrome characterized by nonmotor impairment in executive function and affect.

The cerebellum receives substantial input from nonmotor cortical regions and projects back to the neocortex. Much evidence has come from neuroanatomical viral tracing studies in primates, suggesting polysynaptic loops join forebrain areas with the cerebellum. Enriched understanding of nonmotor cerebellar function can be achieved by identifying cerebellar polysynaptic connections to nonmotor neocortical areas. Anatomical studies of nonmotor cerebello-cortical loops are instructive of the cerebellum’s role in cognition and affect-function derived from structure.

Transsynaptic tracing, histology, and subsequent analysis have challenges including low detection signal, tissue sectioning damage, and biases from manual quantification and anatomical assignment cells. Primary contributions of the thesis involve combining viral tracing with histological techniques for automated quantification of transsynaptic connectivity. I have optimized tissue clearing techniques for detection of virally-infected cells, allowing for volumetric imaging of whole-mouse brains with light-sheet microscopy. Using computational alignment software, I can automatedly and reproducibly register brain volumes to an atlas, removing error-prone manual structure identification. I have generated a local atlas with a complete cerebellum, which maintains translatability with the field standard that lacks the posterior cerebellum. This approach serves as a template solution anatomical commutability across research groups. Using neural networks, we have developed a three-dimensional cell detection pipeline capable of efficiently and accurately detecting polysynaptic connectivity in large brain volumes. Using this pipeline, I present findings showing the cerebellum densely projects to and can influence nonmotor brain regions with functions typically associated with autism spectrum disorder.