| 520 | |
▼a Animal development is one of the greatest sources of wonders in science. Development of multicellular organism is characterized by the differentiation of a fertilized egg into diverse cell types of the body in a programed temporal spatial order. The process of development includes fertilization, cleavage, gastrulation, organogenesis, metamorphosis, regeneration, and senescence. Characterizing cell differentiation in each step, by resolving the cell state diversity and cell fate dynamics, is the key to fully understanding developmental process. The expression levels of mRNA species are readily linked to cellular function, and therefore profiling the transcriptome of individual cells has emerged as a powerful strategy for resolving cell state heterogeneity. However, current methods for single cell RNA sequencing all rely on the isolation of individual cells within physical compartments and thus have problems such as low throughput, high cost, and information lost from other molecular layers. During the three and half years of my graduate study, I developed four novel high-throughput single cell genomic techniques to get over these limitations and applied them to profiling cell state heterogeneity and dynamics in development at single cell resolution.To resolve cellular state heterogeneity, I developed a combinatorial indexing strategy to profile the transcriptome across tens of thousands of single cells (sci-RNA-seq: Single cell Combinatorial Indexing RNA sequencing), and applied sci-RNA-seq to generate the first catalog of single cell transcriptomes at the scale of whole organism Caenorhabditis elegans (Cao. J., Jonathan. P., et al, Comprehensive single-cell transcriptional profiling of a multicellular organism, Science, 2017). I profiled over 50,000 cells from the nematode C. elegans at the L2 stage, which is over 50-fold "shotgun cellular coverage" of its somatic cell composition. This is the first study to show that single cell transcriptome alone is sufficient to separate all major cell types from whole animal. Cell type specific genes for 27 distinct cell types are identified, including for some fine-grained cell types that are present in only one or two cells per individual. Given that C.elegans is the only organism with a fully mapped cellular lineage, these data represent a rich resource for future research aimed at defining cell types and states. The dataset will advance our understanding of developmental biology, and constitute a major step towards a comprehensive, single-cell molecular atlas of a whole animal.To further characterize cellular state across multiple molecular layers, I developed sci-CAR, the first high throughput single cell genomic approach that can jointly profile epigenome (chromatin accessibility) and transcriptome in each of 1000s of single cells (Cao. J. et al, Joint profiling of chromatin accessibility and gene expression in thousands of single cells, Science, 2018). I applied sci-CAR to 11,233 cells from whole mouse kidney and linked cis-regulatory sites to their putative target genes based on the covariance of chromatin accessibility and transcription at the single-cell level. To the best of our knowledge, this represents the first joint profiling of the epigenome and transcriptome in individual cells at the scale and complexity of a whole mammalian organ.One critical challenge in development is to characterize the cell differentiation path for all major cell types forming our body. During mammalian organogenesis, the cells of the three germ layers transform into an embryo that includes most major internal and external organs. The key regulators of developmental defects can be studied during this critical window, but conventional approaches lack the throughput and resolution to obtain a global view of the molecular states and trajectories of a rapidly diversifying and expanding number of cell types. To investigate cell state dynamics in this critical window, I developed another single cell transcriptome profiling technique (sci-RNA-seq3), the first single cell RNA-seq technique capable of profiling millions of single cells in a single experiment, with over one hundred times higher throughput and lower cost compared with conventional approaches. I applied sci-RNA-seq3 to profiling ~ 2 million cells derived from 61 mouse embryos staged between 9.5 and 13.5 days of gestation (Cao. J., Spielmann. M., et al, The single-cell transcriptional landscape of mammalian organogenesis, Nature, 2018). This is by far the most comprehensive cell atlas of mammalian development as well as the largest single cell RNA-seq data set in the world. By unsupervised clustering analysis, I characterized hundreds of expanding, contracting and transient cell types, many of which are only detected because of the depth of cellular coverage obtained here, and defined the corresponding sets of cell type-specific marker genes, several of which are validated by whole mount in situ hybridization. With a new single cell RNA-seq analysis package Monocle 3, I further delineated and annotated 56 single cell developmental trajectories of mouse organogenesis, spanning all major systems such as central nervous system and reproductive system. The dynamics of cell proliferation and key gene regulators within each cell lineage are further identified. These data comprise a foundational resource for single cell genomic field and mammalian developmental biology.To further characterizing the mechanism regulating cell state dynamics, I developed sci-fate, the first strategy to recover whole transcriptome temporal dynamics across thousands of single cells (Cao. J., et al, Characterizing single cell temporal dynamics with sci-fate, manuscript in preparation, 2019). I applied sci-fate to a model system of cortisol response and developed a computation strategy to identify key driving transcription factors regulating cell state changes. Based on the data, I built a cell state transition network for future cell state prediction, and illustrate key factors regulating cell state transition dynamics. This is the first study to quantitatively characterize cell state dynamics at whole transcriptome level and constitutes a major step to fully understanding mechanisms in cell fate determination. |