Preview

PDF, English Download (73MB) | Terms of use

Abstract

In mammalians, the protein-coding sequences, which make up less than two percent of the genome, are separated from each other with large non-coding intervals. Genomic rearrangements in the developmental gene loci indicate that the genes are regulated by long-range enhancers from these non-coding regions. Recently, Genome Wide Association Studies (GWAS) have indicated that forty percent of all human genomic variations, which are associated with a phenotype, are exclusively in the non-coding regions. This suggests that the long-range control of gene expression is a widespread phenomenon in the mammalian genome. There are two main challenges to understand the long-range gene regulation. The first one is to discover the regulatory elements in the vast non-coding regions and to characterize their function. The second challenge is to identify the molecular mechanisms that enable and control the communication between the regulatory elements and their target genes. The genome may appear as a collection of elements but as shown by genomic rearrangements, its organization is important for the spatiotemporal regulation of gene expression. The processes, which convert the regulatory function of individual elements into collective spatiotemporal regulatory information, are poorly understood. In this study, I used mouse c-Myc/Pvt1 flanking locus as a model to understand the contribution of genome organization to endogenous gene expression. This locus is an evolutionarily conserved three megabase-long gene-poor region, with only one protein-coding gene: c-Myc. Retroviral insertions, chromosomal translocations and duplications, with breakpoints up to hundreds of kilobases far from c-Myc lead to various cancers both in mouse and in humans. Furthermore, GWAS showed that genetic variations in humans all along this locus are associated tissue and stage specific tumorigenic or developmental phenotypes. Moreover, a large-scale genome profile revealed by ENCODE project identified elements carrying signatures of enhancers in different cell types in this locus. These studies suggested that long-range regulatory activity is prominent in this locus. In this project, I generated tens mouse lines with a regulatory sensor at different positions to monitor the regulatory activity in the c-Myc/Pvt-1 flanking locus. I have revealed a long-range embryonic face enhancer, which overlaps with the linkage disequilibrium block in the orthologous human 8q24 locus associated with non-syndromic cleft lip and palate risk. In order to get insight into the biological role of this regulatory region, I have generated a series of genomic rearrangements and restricted the embryonic face specific regulatory elements to 250kb long interval. I have shown that this regulatory region acts on the c-Myc gene in a tissue specific manner over a megabase distance. Further analysis of c-Myc downregulation indicated deregulation of gene regulatory networks and metabolic pathways upon the deletion of face enhancer. These pathways may implicate the etiology of 8q24 dependent non-syndromic cleft lip and palate. In addition, in collaboration with Andreas Trumpp’s lab, we have investigated the effects of the deletions in the telomeric end of c-Myc/Pvt1 flanking locus on the hematopoietic system. We have identified that the most telomeric 350kb long region in this locus is critical for different stages of hematopoiesis. We have shown that the regulatory region at this locus acts on c-Myc gene despite being more than 1.4 megabase far. Finally, I investigated the elements that allow communication of distant regulatory regions with the promoter of c-Myc. I have shown that the regulatory landscape is confined in a Topologically Associated Domain (TAD) and the telomeric end of this TAD has dual functions for insulator and tethering activity.