The complex hierarchy of three-dimensional patterns that characterize the 3D folding of mammalian chromosomes appears as an important element in controlling gene expression. At the megabase scale, chromosomes are partitioned into domains that define two main compartments, corresponding to transcriptionally active and inactive regions, respectively. Each compartment domain is itself composed of distinct domains characterized by increased self-interactions called topological domains (TADs). Recent high-resolution Hi-C approaches revealed a finer-scale organisation of the genome in smaller “contact domains”, often associated with loops linking specific points. At these different scales, the spatial organisation of the genome shows tight correlation with its chromatin structure and its transcriptional activity. However, while steady progress is being made in describing the 3D folding of the genome at increased resolution, the mechanisms that determine this folding, its dynamic properties and the functional implications of these emerging features are still poorly understood.
We use advanced genome tagging and engineering strategies, as well as targeted inactivation of factors involved in chromosomal folding to unravel the elements and mechanisms that drive the folding of large loci in specific yet dynamic conformations and their influence on gene expression. Our recent results show that the complex patterns of vertebrate HiC maps result from the superimposition of two distinct mechanisms: 1) a cohesin-independent mechanism which brings together regions of similar chromatin states 2) a cohesin-dependent folding that associate different small compartments into TADs. Within TADS, we show as well that enhancers are not acting in a homogeneous manner, but that their influence is distributed in complex patterns, partially guided by the underlying structure. I will discuss the different implications of these findings for our views of genome organisation.
