Epigenetic regulation of neural stem cell differentiation

In 1942, Conrad Waddington coined the term, "epigenetics". Historically, the word "epigenetics" was used to describe events that could not be explained by genetic mechanisms. Now, epigenetics is generally regarded as the molecular interface between genotype and phenotype---a collection of mechanisms that changes the gene expression or cellular phenotype without alterations in DNA sequence. For instance, almost all cell types in a multicellular organism contain identical genetic sequences; however, embryogenesis produces a vast diversity of cell types with stable gene expression and different morphology/functions. Thus, developmental processes, which are mediated by differentiation of pluripotent and tissue specific stem/progenitor cells, is profoundly influenced and potentially initiated by changes in epigenome, rather than the genome itself. Major epigenetic mechanisms include covalent modifications of histones and DNA. DNA cytosine methylation, the predominant chemical modification of the mammalian genome, is essential for mammalian development. However, our knowledge about their distribution and function in the genome remains limited. The main goal of my thesis research is to understand how de novo DNA methyltransferases, which is required for establishing new DNA methylation patterns, regulates gene expression and differentiation in neural stem/progenitor cells. By investigating the genome-wide occupancy of the de novo DNA methyltransferase 3a (Dnmt3a) and its role in gene regulation in multipotent neural stem/progenitor cells (NPCs), I found that Dnmt3a preferentially occupies and methylates genomic regions flanking proximal promoters of many moderately transcribed genes. Dnmt3a antagonizes binding of the polycomb repression complex 2 (PRC2) to these transcriptionally active/permissive Dnmt3a bound genes in a DNA methylation-dependent manner. Genetic ablation of Dnmt3a in NPCs causes a marked increase of PRC2-mediated repressive histone modification in genomic regions encompassing promoters and reduced transcription of a cohort of Dnmt3a bound genes, many of which are critical for proper differentiation of NPCs. We propose that the unexpected antagonism between de novo DNA methyltransferases and polycomb complexes may play a general role in epigenetic regulation of neural development.