From chromatin fibers to lamina-associated domains: what are the recognition determinants?

ended

"Lamina-associated domains (LADs) are large chromatin domains anchored to a protein meshwork coating the inner nuclear envelope, the nuclear lamina (NL). They cover 40% of the human genome and hold thousands of transcriptionally repressed genes. While some LADs can be locally detached from the NL upon gene activation, others are constitutively at the NL. This implies that LAD recognition and anchoring to the NL has to be tightly regulated to ensure acute gene expression. However, little is known about mechanisms governing LAD recognition and targeting to the NL. LADs are known to be A and T nucleotide-rich, but other sequence determinants have remained poorly characterized so far. In this project I will (1) investigate whether LAD recognition is directly linked to their high A/T content and (2) develop a novel tool to search for additional sequence determinants. First, I will insert in the mouse genome long DNA stretches of varying A/T percentages and determine whether A/T-rich sequences preferentially associate with the NL. Second, I will develop an approach to randomly scramble a LAD sequence organization. This method relies on the recombination between LoxP sites randomly inserted by local ""hopping"" of a transposable element. After determining the effects of the resulting rearrangements on NL association and expression of the surrounding genes, I will identify sequences that drive LAD formation. Besides, the cohesin complex partitions the genome into topologically associated domains and hence has a major impact in genome 3D folding. At last, I will investigate cohesin’s role in LAD formation and remodeling after recombination. Altogether, this project will result in an unprecedented characterization of LAD biology, a deeper understanding of cohesin’s role in genome folding and set up a novel genome-scrambling technique with large applications in genome biology."