Dual perspectives: Cellular differentiation in health and disease

The process of cellular differentiation is essential for maintaining homeostasis in regenerative tissues, including the human gut mucosa. Human small intestinal organoids (hSIOs) have been used to study this process in vitro. However, our understanding of the mechanisms regulating the specification and differentiation of secretory cell types in the human intestinal epithelium remains incompletely understood. Furthermore, current techniques for culturing human small intestinal organoids (hSIOs) fail to fully recapitulate the 3D architecture and cell type diversity of the human gut epithelium. To address these issues, in Chapter 2 we detail an optimized 2-step method for culturing hSIOs through a nuanced modulation of the Wnt pathway and the addition of the cytokine IL-22. This enables the establishment of hSIOs with extensive crypt-like bud structures that closely resemble the organization of the human gut epithelium in vivo. In exploring the role of IL-22 we find that, contrary to previous studies, it does not drive ISC proliferation, but instead induces Paneth cell differentiation through downstream activation of the PI3K/AKT/mTOR signaling axis. Then, in Chapter 3 we perform a CRISPR screen in hSIOs to identify transcriptional regulators of secretory lineage differentiation. In addition to known regulators, this establishes the zinc finger transcription factor ZNF800 as a critical regulator of EEC cell fate. Specifically, ZNF800 directly represses enterochromaffin-biased EEC differentiation through a network of downstream transcription factors. Together, these results highlight mechanisms by which signaling pathways and transcription factors regulate cellular differentiation in healthy tissue.

Pediatric cancer is thought to arise through the interruption of normal cellular differentiation during development. Our understanding of how this occurs in pediatric soft tissue sarcomas is not well understood. Work in the second half of this thesis investigates the role of cellular differentiation in pediatric sarcomagenesis and seeks to identify novel therapeutic vulnerabilities using single-cell RNA-sequencing and patient-derived tumor organoid models. In Chapter 4, we compile a single-cell transcriptomic atlas of pediatric RMS tumors and tumor organoid models. Using this data, we perform a survey of the immune microenvironment of these tumors, finding evidence of cellular dysfunction and the suppression of tumor immune responses. We then assess the transcriptional heterogeneity of malignant cells, which reveals subtype-specific cell states mirroring normal myogenic differentiation trajectories. Using bulk transcriptomic data, we show that the differentiation state of RMS tumors is associated with differences in patient outcomes. In Chapter 5, we seek to define RMS-specific genetic vulnerabilities by performing a genome-wide CRISPR viability screen in a tumor organoid model of FP-RMS. This reveals the putative susceptibility of RMS cells to perturbation of the oxidative phosphorylation and de novo nucleotide biosynthesis pathways. We then assess the potency and specificity of kinase inhibitors targeting screen hits in a panel of RMS tumor organoids. In Chapter 6, we conduct a single-cell transcriptomic survey of rare pediatric mesenchymal tumors. We use this data to define tumor-specific transcriptional signatures and characterize infiltrating immune cells. Furthermore, we investigate the putative origins of these tumors by comparing malignant cells to normal fetal developmental trajectories.