Orateur
Description
In the nucleus of eukaryotic cells, the DNA is folded in a structure called chromatin. It has been shown that there is a strong connection between the three-dimensional organisation of the genome and the expression of genes. The domain of 3D genomics is now moving forward to find fundamental principles of genome organisation. Many methods exist in the domain to reveal the three-dimensional contacts inside DNA. I think they can be classified in two types.
Firstly, the method that rely on cross-linking (HiC, 3C, 4C…). These methods are the most used and the most developed. Despite being very useful in revealing the structure of the genome at different scales, they measure the contact after performing a fixation and cross-linking. This chemistry might have a strong impact on genome organisation at small scale.
Secondly, there are methods relying on enzyme reaction on the genome. The later has the advantage that the contacts are marked in vivo. DamId and DamC are examples of techniques of this class. Both rely in a bacterial enzyme called DNA Adenine Methylase that binds on GATC quadruplets and catalyses methyl group transfer from S-adenosyl-methionine (AdoMet) to the nitrogen atom at the sixth position of adenine (m6A). In DamId technique, DAM enzyme is bound to a protein of interest, forming a X-DAM protein (for example rTetR-DAM). Then, the m6A positions in the genome reveal the preferential binding position of X. This technique has been shown to be efficient for rTetR and LexA protein binding. The Dam-C technique relies on the same logic. But Dam is bound to a protein whose binding sites are known and the methylation signal is used to identify trans-contacts between the addressing loci and distant loci in the genome. DamId and DamC are useful in eukaryotes because they do not have constitutive m6A methylation.
The team of Gael Yvert at LBMC is developing a modified version of Dam-C technique that they called Light Inducible DAM (LiDAM). It is a DAM enzyme fused with an asLov potein such that the reactive site of DAM is blocked by the asLov protein in closed conformation. When the sample is exposed to light in the visible spectrum, the asLov conformation changes to an open state, letting the reactive site of Dam accessible and allowing the binding and methylation to take place at GATC loci. In the current setting, a plasmid expressing rTetR-LiDam is used. We are using saccharomyces cerevisiae (yeast) cells that have been engineered to contain a docker of TetO sites on chromosome 10. TetO is the known binding site of rTetR. This system is the same as in the original Dam-C and we are currently in the track of reproducing the original DamC study with this system.
The originality of our study relies (1) on the use of nanopore sequencing which allow for a big simplification of the experimental protocol, (2) on the use light induction. The light induction allows to gain time resolution. One can control when the reaction of methylation can occur. This leads us to imagine a scenario where the cells are cultured in the dark. Then the cell might be synchronised in a given state. At time $t_0$ after synchronisation a perturbation is made to the cells (for instance inducing a double strand break, DSB). The cells evolve for a time $τ_{probe}=t_{on}-t_0$ and then they are illuminated during $τ_{exp}= t_{on}-t_{off}$. We would obtain a recording of all the contacts that occurs during the duration τ_d. To make the analogy with a photography, $τ_{exp}$ plays the role of exposure time while $τ_{probe}$ plays the role of a delay between a pump (the trigger of the DSB) and a probe which measure the contacts. Scanning pump-probe delay would give access to a “film” of the repair of the double strand break. The method is still under development and we are not yet there.
In my talk I will present the current status of this project. I will first talk about the experimental setup that have been developed by Mohammed Baddaze in the team of Gael Yvert. Secondly, I will explain which data analysis strategy I settled. Finally, I will present conclusion we are able to draw now. Especially I will show comparison of methylation patterns we observed in different conditions and HiC data and accessibility data in yeast genome.