Deconstruction of the Fgf5 enhancer cluster landscape

During embryonic development, cell fate decisions manifest by dramatic changes in the gene expression patterns. Gene expression of most developmental genes is controlled by genetic elements called enhancers, short DNA sequences that consist of multiple binding sites for transcription factors. These elements control the expression of the target gene from a distance which complicated the study of them. For a long time, enhancer location had been elusive until it was realized that putative active enhancers can be identified by a distinct chromatin signature through Next Generation sequencing. From these initial studies we learned that enhancer vastly outnumber actual genes. Furthermore, with identification of the location of enhancers within the genome and in distinct cell types it became clear that often multiple enhancers together control the expression of a target gene. It was proposed that these so called super-enhancers show unusual sensitivity to perturbations suggesting that the single enhancer elements might indeed be working together in a collaborative fashion. We have recently identified such a enhancer cluster that is activated during an early embryonic cell fate decision that can be followed in vitro. The work laid out in the proposal Deconstruction of the Fgf5 enhancer cluster landscape will shed light on how single elements are contributing to the expression of the target gene, Fgf5. By deleting single elements and performing very fine grained time courses we will determine even small contributions of single elements to the overall target gene expression. Furthermore, we will test for effects of enhancer loss not only on the promoter of the target gene but also on the other surrounding enhancer elements and their chromatin environment. In addition, it has been shown that the three-dimensional organization of the chromatin in the nucleus plays a major role in the regulation of gene expression. Therefore, we will test whether the loss of an enhancer elements leads to changes in the overall chromosomal three-dimensional architecture. Finally, we will analyze the underlying mechanism that drives enhancer activation and three-dimensional organization. Taken together, the proposed work will allow us to build the first comprehensive model of orchestrated enhancer action in mammals and thus advance our understanding of a fundamental aspect of gene regulation.

Enhancers are genetic elements that control the activity of a gene from a distance. Enhancers consist of multiple transcription factor binding sites that recruit additional co-activators to lead to the activation of the target gene. Enhancers can be identified through a combinatorial epigenomic signature: it was realized that many enhancers share the binding of co-activators such as p300 and specific modifications of the surrounding histones. Genome-wide analysis was carried out to map the active enhancer landscape in numerous cell types. Therefore, we know now which elements might be active in a given cell type. However, how enhancers function is still not very well understood. In this study, we focused on the interaction of multiple enhancers and how they affect each other. Many genes are regulated by not just one enhancer, but multiple enhancers that together form a cluster. It is unclear how these elements then work together to collaboratively activate the target gene. We studied one locus in detail to investigate how single enhancer elements individually and as group work in the activation of the Fgf5 gene. We demonstrated that one element in close proximity to the gene promoter amplifies the expression of the target gene Fgf5, but only when other enhancers are present. Alone and without other enhancers, this element has very minor effects on the activation of the target gene. These data suggest that enhancers that are identified through their epigenomic signature might not all function in the exact same way, but that there are differences between them. We showed that Polymerase II, one of the enzymes that transcribe DNA into RNA, accumulates at the enhancer of interest. It was previously thought that the strength of an enhancer correlates with the amount of Polymerase II accumulated. However, the element of interest did not have any strong intrinsic enhancer activity. Furthermore, we expanded our analysis to genome-wide comparison and showed that intrinsic enhancer activity does not correlate with the amount of Polymerase II at an enhancer. But what might determine Polymerase II levels? Here we showed that the distance of an enhancer to the next active promoter might be one driver of polymerase II accumulation at the enhancer: the further away an enhancer is from the next active promoter, the lower the amount of polymerase is at these elements. This study opens up numerous new avenues of research. On the one hand, we clearly showed that enhancers with low intrinsic activity can contribute to the expression of the target gene in the right genomic context. This will be of interest for the analysis of mutations in the future. Furthermore, it raises the question whether the accumulation of Polymerase II plays a role in enhancer function.