Little is currently known about the number, proportion, or lineage of distinct cell types in the developing human fetal brain. Knowledge of such a component list and its functional genomic foundations is crucial for understanding the function of this complex system, its evolution, and how it is disrupted in disease. We hypothesize that comprehensive single-cell mRNA expression profiles provide an accurate and efficient rubric for a first generation classification schema that can be integrated with lineage, morphology and connectivity. We will use unsupervised learning algorithms to cluster 10,000 single cell transcriptomes derived from RNAseq of the human fetal cortical anlage, providing an unbiased model to identify and understand the resultant cell classes. We will validate these cell class determinations using in situ hybridization. We will use marker genes identified in this analysis to perform lineage tracing using cell-type specific reporters engineered via genome editing, and obtain single cell transcriptional profiles at different stages of development to determine lineage relationships. Finally, we will analyze intact fetal cerebral hemispheres and mouse-human chimeras processed with the CLARITY technique to provide 3D localization and morphology of a subset of cell types. To markedly improve scalability of this approach, we will apply enhanced Bessel Beam Tomography for spectral image acquisition, increasing imaging speed by 2-3 orders of magnitude while maintaining high resolution. In parallel, we will develop automated image analysis algorithms to enable cell detection and comprehensive morphological assessment in an automated and correctable fashion. Overall, completion of these studies will permit, among many future advances: (1) The generation of an unbiased rubric of cell class based on integration of morphological, transcriptomic, and connectivity data used to understand cell type diversity and function;(2) Connection of cellular function, morphology, and connectivity to specific genes and proteins;(3) Refinement of the cellular basis of transcriptomic disease signatures and understanding of the individual cellular mechanisms of disease (4) The construction of connectivity diagrams based on discovered circuit components allowing development of realistic computational models of brain function and dysfunction;(5) Optimization of cell types generated by in vitro stem cell models based on gold standard in vivo cell class definition;and (6) Advancement of light microscopy techniques for rapid, spectral high resolution imaging of postmortem human and mouse brain including automated processing, which in addition to permitting high throughput brain mapping, will significantly advance mechanistic studies in mouse models and pathological studies in human. Completion of these aims will provide a proof of principle, multiple tools, and a core framework on which to build a more comprehensive knowledge of cell classes in the human brain.
Growing evidence implicates abnormalities in the development and architecture of the human cerebral cortex in neuropsychiatric illnesses such as autism and schizophrenia. Understanding the pathological disruption of a complex system such as the human cortex first requires an understanding of its components: individual cells. We will develop novel methods and tools to identify the individual cell classes that compose the human fetal cortex, determine the lineage of progenitor cell classes, and map the location, morphology, and connectivity of each cell class.
|Ecker, Joseph R; Geschwind, Daniel H; Kriegstein, Arnold R et al. (2017) The BRAIN Initiative Cell Census Consortium: Lessons Learned toward Generating a Comprehensive Brain Cell Atlas. Neuron 96:542-557|
|Won, Hyejung; de la Torre-Ubieta, Luis; Stein, Jason L et al. (2016) Chromosome conformation elucidates regulatory relationships in developing human brain. Nature 538:523-527|