Transposable Elements in Neural Progenitor Cells

Abstract: More than 90% of DNA does not code for proteins and for a long time these sequences were referred to as “junk DNA” due to their unknown purpose. With the advent of new technologies it is now known, that the non-coding part of the genome is of great importance for regulating gene expression and is therefore indispensable. Transposable elements comprise about 50% of the genome and co-exist as symbionts regulated by epigenetic mechanisms - a highly defined machinery that controls gene expression and is mandatory for a proper development and maintenance of an organism. Although transposable elements are associated with diseases, their role in fine-tuning the host gene expression becomes more and more evident, which seems to justify the positive selection during evolution. A transposable element called Line-1 was found to be active in neural progenitor cells and in the brain. Several studies report Line-1 transcription and frequent retrotransposition during normal brain development, with further evidence that Line-1 induced retrotransposition can influence neuronal gene expression. Today, there is few published data focusing on epigenetic regulation of transposable elements in neural progenitor cells. In this thesis, I identify TRIM28 as key regulator of certain groups of transposable elements in mouse and human neural progenitor cells. This feature is unique compared to other somatic tissues, where DNA-methylation is prevalent. Here I demonstrate, that transposable elements MMERVK10C and IAP1 in mouse neural progenitor cells are repressed by the establishment of H3K9me3-associated heterochromatin. De-repressed MMERVK10C and IAP1 furthermore activate nearby genes and generate long non-coding RNAs. Homozygous TRIM28 knockout is lethal, while mice with mono-allelic TRIM28 expression are characterised with a distinct behavioural phenotype. Moreover we are also able to show that TRIM28 is regulating a fraction of young Alu-elements in human neural progenitor cells, which is not the case in human embryonic stem cells. Furthermore, we report that transcribed Alu-elements influence gene expression of close-by genes. Studying pluripotent cells revealed that TRIM28 modulates transposable elements in mouse embryonic stem cells. Activation of transposable elements upon TRIM28 depletion induces changes in gene expression of close-by genes and causes alteration of the repressive chromatin mark H3K9me3 at transposable element loci. Upregulated genes were shown to have bivalent promoters, characterised by H3K4me3 and H3K27me3 and lay close to H3K9me3 regulated transposable elements. These findings in mouse embryonic stem cells are highly relevant for the interpretation of my studies in neural progenitor cells. Taken together, this thesis demonstrates that the regulation of transposable elements in mouse and human neural progenitor cells is distinct compared to previous reports regarding somatic tissues. These results provide novel insights into why the brain has developed into such a complex organ with so many different cell types.