Community Research and Development Information Service - CORDIS

Final Report Summary - ENHANCEME (Elucidating the role of DNA methylation in enhancer priming in vivo)

When a zygote develops into a full-grown organism different cell types need to be specified. Every different cell type requires a distinct transcriptional program and when transcriptional programs are not properly maintained cells can die, or for example, can obtain cancerous characteristics. Hence the establishment and the maintenance of gene expression profiles is extremely important during development as well as during later stages in life. The different transcriptional profiles in different cell populations are all based on one and the same genomic sequence. The mechanisms that regulate differential gene expression in different cell types are often referred to as ‘epigenetics’. The most direct epigenetic mechanisms regulate gene expression transcriptionally by covalently modifying the DNA itself, mainly through methylation, or proteins bound to the DNA, most notably the histones.
Enhancers are non-coding regions in the genome that can influence the activity of regions in the genome that are protein coding. In recent years enhancers have been the topic of extensive research. This research, fuelled by the development of techniques like ChIP-Seq and 4C-Seq, has resulted in the identification of multiple different epigenetic states in which enhancers can reside. Inactive, but epigenetically marked, enhancers are called primed enhancers and have been described to reside on hypermethylated DNA. This is in contract to active enhancers, which generally reside on hypomethylated DNA. The functional role of DNA methylation at enhancers is at present not fully resolved, but it is has been described that hypomethylation correlates with fast induction of expression and that some proteins preferentially bind to unmethylated or methylated DNA.

The main goal of this study is the epigenetic characterization of enhancers throughout zebrafish development. Preliminary analysis already revealed a previously undescribed separation of enhancers based on DNA methylation characteristics in zebrafish. In the case of zebrafish active enhancers are predominantly on hypermethylated DNA (hyper-enhancers) and primed enhancers are primarily on hypomethylated DNA (hypo-enhancers). In this study I aimed to further characterize these two enhancers types using different experimental strategies. For instance I have generate novel ChIP-Seq data sets, for instance for proteins and histone modification associated with poised enhancer states. This is interesting as the hypo-enhancers are close to key developmental genes and these genes have been shown in other organisms to be poised for activation. Furthermore, at 4HPF hypo-enhancers are already epigenetically marked, many cell divisions before they become active, but it is unclear when they become primed. To study this, special effort will be put into the generation of data sets that can address this question prior the activation of the zygotic genome. I also noticed that the majority of the hypo-enhancers are never in an epigenetically active state during embryogenesis and hypothesised that they are only active in adult tissues. To address this question I will generate data sets that interrogate the (epigenetic) activity of enhancers in two adult tissues.
Enhancers need to be in physical proximity of their target promoter to activate them. To determine the 3D organization of the genome a technique called 4C-Seq can be used. I will use this technique to map the 3D organization surrounding hypo-enhancers. I will perform 4C-Seq at different developmental timepoints and in adult tissue. In this way I can study the dynamics of the enhancer-target interactions and I can also study if the enhancer is already in proximity of its target-gene prior the activation of the enhancer or the target-gene.
I will perform in vivo enhancer assay to establish if the identified hypo-enhancers are indeed bona fide enhancers. If this is the case I will study the spatial-temporal activity of the hypo-enhancers and see if the expression patterns are reproducible. This is the golden standard to really prove that a genomic region has that capacity to act as an enhancer.
Lastly the aim is to finish this story, write a article about the main findings and get it published in a peer-reviewed journal

In contrast to conventional epigenetic interactions at enhancers, I found that during early zebrafish development, DNA methylation has little influence on enhancer activity and vice versa, and that hypo-methylation is a unique feature of primed enhancers, whereas active enhancers are generally hyper-methylated. Integrating the hypo-and hyper-enhancers with datasets that identify genomic regions poised for activation I have found that hypo-enhancers are selectively enriched to associate with the poised state. Enrichment based analysis revealed that hypo-methylated enhancers are enriched close to important proteins that act later in development. Following the DNA methylation state of the hypo-enhancers throughout early development revealed that they lose DNA methylation after fertilisation and before the embryo starts to express genes. Further analysis of the hypo- and hyper-enhancers with different genomic data sets revealed that the hypo-enhancers are also unique in other features. For instance they specifically show enrichment for histone modifications normally found at the beginning of protein coding genes and reside in an open chromatin environment. Finally, I demonstrate that hypo-enhancers are active at later developmental stages and that they are physically associated with target-genes, irrespective of target-gene activity using 4C-Seq experiments. Further analysis of data sets generated from the adult intestine and brain that interrogate the activity of enhancers revealed that hypo-enhancers are active in these two tissues showing that hypo-enhancers are truly primed for activation in adult tissues. This is also in agreement with the in vivo enhancers assays I performed, which in all cases revealed highly specific expression patterns late in development. In summary I demonstrate that early development in zebrafish embodies a time-window uniquely characterized by non-canonical DNA methylation-enhancer relationships, including global DNA hypo-methylation of inactive enhancers and DNA hyper-methylation of active enhancers.


Franziska Hornig, (Grants Manager)
Tel.: +49 6131 3921453
Fax: +49 61313921421
Record Number: 188917 / Last updated on: 2016-09-14