In a first-of-its-kind study, EMBL researchers have shown how DNA methylation across the genome contributes to the precise regulation of gene expression
If one imagines the genome as an instruction manual for the functioning of a cell, every page of this manual is covered with annotations, highlights, and bookmarks. The role of some of these marks remains mysterious - do they actively direct the reader to the right place at the right time, or do they merely indicate the pages the reader has already visited?
This subtle distinction in the language of the cell can play an important role in its survival and function. As researchers from the Krebs group at EMBL Heidelberg have now shown, one such annotation - DNA methylation - exerts a highly selective layer of control on the expression of genes, one that varies according to cell type and fate.
What is DNA methylation?
DNA methylation is a repressive epigenetic modification. In mammalian cells, it is found at specific DNA sequences where enzymes add a small chemical group called 'methyl' to the cytosine base, subtly modifying its structure. It plays a key role in the control of genome activity and is essential for mammalian development. One important function is to prevent the expression of genes, a process called silencing. DNA methylation is a stable mark that can be inherited through several generations of cells, and even organisms.
In the analogy above, the annotations, highlights, and bookmarks represent what scientists call 'epigenetic marks', whereas the 'reader' is usually the complex molecular machinery responsible for gene expression. The latter includes specialised proteins known as transcription factors.
When a particular region of DNA needs to be expressed, the area surrounding it undergoes physical and chemical changes, making it more accessible to such molecular machines. While DNA methylation is found across the genome, whether and how it affects this accessibility at specific genomic regions remains relatively unexplored.
"Our group is interested in the fundamental mechanisms that regulate gene expression," said Arnaud Krebs, Group Leader at EMBL Heidelberg. "We are particularly interested in cis-regulatory elements like enhancers - DNA regions that control the activity of genes."
Krebs' team was intrigued by the fact that while DNA methylation is often reduced at active enhancers, the cause-effect relationship between the two remains unclear. Does the activation of these DNA regions lead to a removal of methylation? Or does the reduction in methylation itself drive the activation?
To investigate this, the team used a high-resolution technique developed in their lab - single-molecule footprinting. This method allowed them to simultaneously measure DNA methylation, accessibility, and transcription factor binding, at the level of single DNA molecules. They applied this across the whole genome in multiple cell types, including mouse embryonic stem cells and differentiated cells. This combination of scale and resolution allowed the scientists to gain a deeper understanding of DNA methylation's role in gene regulation in a living cell.
The team found that while the accessibility of ~97% of the enhancers they studied was insensitive to DNA methylation, about 3% required the absence of DNA methylation to get activated. At these sites, methylation reduced DNA accessibility and directly prevented the binding of transcription factors. The identity of these methylation-sensitive enhancers varied across cell types and stages.
"The 3% of enhancers that seem to be regulated by DNA methylation are enriched for cell-type specific enhancers. We think they are connected to genes that are important for cellular identity," said Elisa Kreibich, PhD student in the Krebs group and first author of the study, now published in Molecular Cell.
"By making our measurements at the level of single molecules, we can figure out the connections and interactions between the layers of gene regulation that exist in a cell," added Krebs. "While DNA methylation has often been used as a marker for cellular processes, including those involved in cancer, our study shows where it is truly instructive, rather than simply indicative."
The team has now made the tools and analysis codes they used openly available for scientists across the world to use. The research was also supported by funding from the German Research Foundation (DFG).
