Design principles of promoter and enhancer activity in mammalian genomes

Abstract

Correct gene expression patterns are central for cellular function and the development of organisms. This is controlled by regulatory elements such as enhancers and promoters. In this thesis, I present work from two projects with the goal to identify design principles of promoter and enhancer activity in mammalian genomes. In the first part of the thesis, I focused on CpG island promoters. This promoter type represents the majority of mammalian promoters and is characterised by a high density of the CpG dinucleotide. However, to what extent and how this characteristic dinucleotide contributes to promoter activity is still unclear and is one central question of this project. By monitoring binding of transcription factors (TFs) assumed to play a role in CpG island activity and quantifying the activity of promoter mutants and artificial promoters, we gained insight into the role of CpGs in transcriptional activity. The generated data suggests that high CpG density is not sufficient for transcriptional activity, yet necessary when combined with more complex TF binding motifs. We could further show that DNA methylation decreases activity of promoter mutants with low CpG density. Our experiments led us to hypothesise that high CpG density is required to generate a chromatin environment permissive for transcriptional activity. In the second part of the thesis, I focused on cell type and tissue specific regulatory elements. To illustrate an experimental workflow to identify and test regulatory elements for transcriptional activity in specific cell types, we used the mouse retina, a very specialised tissue comprised of ~50 cell types. To identify regulatory elements, we combined transcriptome and epigenome profiling to map the regulatory landscape of four distinct cell types isolated from mouse retinas (rods, cones, horizontal and starburst amacrine cells). This data also revealed sequence determinants and candidate TFs that control cellular specialisation. We tested previously identified regulatory regions using a parallelised reporter assay for their ability to autonomously control transcriptional activity in the four cell types. We were able to generate a catalogue of cis-regulatory regions active in retinal cell types and further demonstrate their utility as a potential resource for cellular tagging and manipulation. Taken together, the work presented here advances our knowledge about location and regulation of regulatory regions that function in specialised cell types and also provides insight into the regulation of CpG island promoters that tend to be ubiquitously expressed.