The overarching strategy of the lab is to generate several genome-wide data in a cell-based system that we can then utilize to pin point potential regions of interest, where DNA methylation may be playing a role of gene activator. We utilize mouse embryonic stem cells, which can be cultured such that they exhibit global DNA methylation levels similar to the bona fide in vivo embryonic cells. We performed a technique where we can assess the methylation state of every cytosine in the genome. Furthermore, we can differentiate the cells in a manner akin to the natural embryonic progression, and perform the same technique to determine where DNA methylation is gained. Our favored hypothesis for DNA methylation-dependent gene activation is that the methyl-mark does not activate genes, per se, but actually antagonizes another epigenetic mark that maintains the silent genic state. This epigenetic mark is found on the histone proteins around which DNA is wrapped. In this manner, we can determine the regions where the histone mark is lost, while the DNA mark is gained. We discovered nearly 3,000 such regions. By incorporating the genome-wide gene expression data we generated, we discovered over 100 genes that appear activated. In order to prove that what we have observed is a genuine effect of DNA methylation, we are employing what is known as “epigenome editing”. With epigenome editing, we use a Cas9 enzyme that can no longer cut the DNA. However, it still binds to the DNA, and we engineered it such that it recruits DNA modifiers: either an enzyme that deposit the DNA methylation mark, or one that removes it. By using this precision technology, we can now show that depositing DNA methylation on an otherwise unmethylated region can tell the adjacent gene to turn on. These experiments are ongoing.
In parallel, we are also keen to understand how the wave of DNA methylation that occurs during this early window of embryonic development impacts the way the DNA is folded, and how in turn this folding affects gene expression and embryonic progression. The motivation for this aspect is the fact that a key regulator of genome folding is sometimes sensitive to the presence of DNA methylation on its binding site. Therefore, using our differentiation system, we can use genome-wide techniques to ascertain when and where this factor disappears from its binding site when the genome becomes DNA methylated. Additionally, we can use specific protocols that can tell us how 3D genome organization changes in this period. Mammalian genes are often dependent on long range physical contacts with so-called “enhancer” elements. Thus, if the DNA folding does not occur properly, these contacts can be negatively affected. If normal DNA methylation pattern are impaired, what is the number of these contacts that fail to occur? And in turn, do they prevent embryonic cells from differentiating correctly? We have generated, and our currently analyzing the data that will help us answer these fundamental questions of epigenetic regulation of embryonic development. Finally, as we are doing for DNA methylation-dependent activation, we are currently using our epigenome editing system to prove that the DNA methylation plays an important role in facilitating proper genome regulation via its impact on the 3D genome structure.
In sum, we have generated a number of high-quality datasets, which we believe demonstrate the non-conventional ways by which DNA methylation affects the very early stages of embryonic development. We have described these results in a number of international conferences (eg, EMBO Awakening of the Genome in Vienna), as well as an invited speaker in various European institutes. We are hoping to publish our findings in the next reporting period.
