Abstract
Runx1 is crucial for blood cell development and is one of the genes most frequently translocated or mutated in leukaemia, thereby contributing to disease pathogenesis. Runx1 is transcribed from two promoters, P1 and P2, and the resulting transcripts give rise to different protein isoforms. Runx1-P2 is required for the generation of definitive haematopoietic progenitor and stem cells, whereas Runx1-P1 plays a role in controlling their population size. The multisubunit protein complex cohesin is best known for its role in sister chromatid cohesion from G2 phase to mitosis. Cohesin also cooperates with the insulator protein CCCTC-binding factor (CTCF) to regulate expression of genes. Cohesin and CTCF spatially connect gene regulatory elements with promoters, thereby structuring the genome in three-dimensional space. Cohesin was previously shown to be essential for runx1 expression in haematopoietic progenitor cells in zebrafish. CTCF however, was not required for expression of runx1 during zebrafish development, but instead restricted its expression in multipotent cells of the tail bud. A large intronic region that harbours gene regulatory elements resides between the two promoters of Runx1 in all vertebrates characterized to date. In mouse, this region includes a haematopoietic-specific enhancer located 24 kilobase pair (kb) from the mouse Runx1-P1 transcriptional start site (TSS). In zebrafish, four regulatory elements located +13, +14, +39 and +89 kb downstream of the P1 TSS, were identified based on the binding of cohesin, CTCF or on the presence of enhancer histone modifications. However, their functionality in regulating runx1 expression was unknown. In my thesis, I investigated the mechanism by which cohesin regulates Runx1 expression. I hypothesised it does so by controlling the formation of chromatin interactions between regulatory elements and the Runx1 promoters.
I found that cohesin controls isoform-specific expression of runx1 during zebrafish development, and that the regulation of Runx1 expression by cohesin is conserved in mammalian cells. Three of the regulatory elements identified in zebrafish, termed +13, +14 and +89, were capable of functioning as insulators in a CTCF-dependent manner, suggesting they may be involved in insulating regulatory activity within certain genomic regions surrounding runx1. Determining the function of the +39 element requires further investigation. I used circular chromosome conformation capture (4C) to identify novel regulatory elements that interact with the Runx1 promoters and the +24 enhancer in a mouse haematopoietic progenitor cell line, HPC7. 4C interaction data showed that Runx1 is located within a topologically associated domain in which not many other genes are present, indicating that regulatory elements located in this region are likely involved in regulating transcription from Runx1. The active promoter in the HPC7 cell line, P1, interacted with two conserved non-coding regulatory elements, as well as with the +24 enhancer. Each regulatory element recruits multiple blood transcription factors in HPC7 cells, suggesting they function as haematopoietic-specific enhancers. The inactive P2 promoter did not interact with any conserved non-coding elements. Cohesin and CTCF did not bind to the putative enhancer elements, and do not seem to be directly involved in the formation of their interaction with Runx1-P1. Instead, cohesin and CTCF might organise the chromatin domain in which Runx1 resides, which in turn controls the availability of regulatory elements to interact with Runx1.
Overall, the findings of my study suggest that spatial and temporal regulation of Runx1 gene expression could be orchestrated by cohesin, through its role in organising chromatin in three-dimensional space.