The regulatory capacities of epigenetic mechanisms including DNA methylation, histone modifications, and non-coding RNAs, have seen a rising interest in recent years. These epigenetic marks are pervasive and non-randomly distributed across the genome, raising intriguing questions on how epigenetics contributes to genomic features that define cellular identity and function. Unlike fixed genetic information that is shared between all cell types, epigenetics involve multiple layers of regulation and can vary dramatically across different cell types and genomic contexts. Thus, much more effort is required to procure a complete perspective of the manifold epigenetic landscape. The body of work in this dissertation focuses on epigenetic studies in the mammalian pluripotent stem cell model system. We utilize high-throughput technologies such as microarrays and next-generation sequencing (NGS) as well as leverage existing epigenetic maps to address a wide range of molecular questions on a comprehensive global scale. This dissertation is organized into three overarching themes:
First, we employed genome-wide gene expression and DNA methylation profiling tools to determine whether different cell types display unique biomarkers that can be used to distinguish them from other cell types (Chapters 2-5). We found that human embryonic stem cells (hESCs) and induced pluripotent stem cells (hiPSCs) carry distinct features in both gene expression and DNA methylation patterns, arguing in favor of the idea that these two pluripotent cell types are different. Furthermore, we compared pluripotent stem cell derived retinal pigmented epithelium (hESC-RPE and hiPSC-RPE) with fetal and adult RPE and found that all RPE cells share a core set of signature genes that distinguishes them from all other cell types. We propose these signature genes will be useful for evaluating the quality of stem-cell derived RPE. Finally, novel corneal endothelial cells (CECs) biomarkers were identified through comparing 12 other tissue types, paving the way for future studies to evaluate properties of stem-cell derived CECs.
We next examined how molecular features of pluripotent stem cells are altered during the differentiation process of stem cells (Chapters 6-8). Using the RPE differentiation paradigm, we profiled both microRNA and DNA methylation patterns in intermediate stages between pluripotent stem cells and mature RPE. These two separate studies identified subsets of dynamically regulated epigenetic marks, some of which are associated with RPE signature gene expression. Furthermore, we used a highly innovative and powerful single-cell RNA-sequencing approach to profile transcriptional changes in the early embryo beginning from mature oocyte to morula stages. This study identified a conserved genetic program describing a highly dynamic transcriptional architecture during early embryogenesis.
Finally, we took a focused analysis on how DNA methyltransferases contribute to shaping the pluripotent stem cell epigenome (Chapters 9-10). Using mouse ESCs null of DNA methylation, we determined DNA methylation regulates a large set of genes through action with H3K27me3. Furthermore, we determined shared and unique genomic targets of each DNA methyltransferase, including novel de novo methylation activity for Dnmt1 in vivo. In the human model system, we generated iPSCs from ICF Syndrome patient fibroblasts which carry double heterozygous mutations in DNMT3B. We found DNMT3B is involved in a wave of de novo methylation during the reprogramming process and has unique genomic targets.