Transcriptional identities of major CNS cell types in humans with insights into astrocyte diversity

Understanding the molecular basis of cellular identity in the central nervous system (CNS) is essential for ascertaining the functions of different cell types and their susceptibilities to diverse pathologies. It is assumed that diversity in cellular identity is needed for the assortment of CNS functional outputs that are key for behavior and cognition. Yet this information remains challenging to acquire in humans. In this dissertation, two studies are presented that provide a new outlook on cellular identity with functional insights into astrocyte diversity. In the first study, analysis of gene coexpression relationships from >7000 neurotypical adult human samples identified consensus transcriptional signatures of astrocytes, oligodendrocytes, microglia, and neurons. A novel metric called gene expression 'fidelity' was created to quantify the extent to which a gene is sensitively and specifically expressed by a given cell type and provides a bridge between the analysis of gene expression data from bulk tissue samples and single cells. Predictive modeling of transcriptomes using high-fidelity genes identified cell type-specific transcriptional differences in disease, among brain regions, and between species. The second study addresses the functional significance of the astrocyte molecular diversity revealed in the first study by focusing on the potassium channel Kir4.1 in mouse spinal cord astrocytes. In the spinal cord, functionally distinct sub-classes of motor neurons control movement through innervation of skeletal muscle. The largest of these, fast α-motor neurons, target fast-twitch muscle fibers that generate peak strength and are selectively vulnerable in amyotrophic lateral sclerosis (ALS). We found that astrocytes adjacent to motor neurons express a VGlutT1-dependent developmental upregulation of Kir4.1, with maximal expression around fast α−motor neurons. Astrocyte Kir4.1 was selectively required for the maintenance of fast α−motor neuron size and electrophysiological properties. In addition, animals displayed markedly reduced peak strength and loss of large fast-twitch muscle fibers. Downregulation of Kir4.1 was observed in ALS mice but was not found to causally alter motor neuron survival in this model. These findings suggest that astrocyte-neuron interactions are essential for establishing CNS diversity. More broadly this work demonstrates that a focus on molecular cellular identity leads to numerous insights into CNS function