I.am.a.LAD · From chromatin fibers to lamina-associated domains: what are the recognition determinants?

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

About 2 meters of genomic DNA are confined in a 10 µm cell nucleus. Genome folding needs to be controlled at multiple levels to achieve such a level of compaction. One such mechanism involves contacts of the genome with the nuclear lamina (NL), a proteinaceous meshwork that coats the inner nuclear envelope. Genome-NL interactions occur through long stretches of genomic sequences named lamina-associated domains (LADs). Conversely, inter-LADs (iLADs) represent genomic sequences not in contact with the NL. In most human and mouse cell types, there are about 1,000 of these LADs, each spanning 0.1 - 10 Mb and collectively covering about 40% of the genome. LADs are massive structures: a LAD of 3 Mb consists of ~1 mm of linear DNA or roughly 150 µm of nucleosomal 10 nm fibre. However, how the tethering of such a gigantic molecule to the NL is organized, and how it is instructed by its own DNA sequence remains a mystery. The goal of the study is to identify new LAD determinants, driving LAD recognition and targeting to the NL. Genome organization deregulation is often associated with cancer development. By elucidating fundamental principles governing genome organization, I hope that my research will enable the identification of new targets that could play a role in cancer development or progression. Moreover, a non-functional NL can lead to the development of diseases called laminopathies. However, the extent to which LADs are affected and how they contribute to disorder development remains unknown. By unveiling mechanisms triggering LAD recognition and targeting to the NL, this project will also open new perspectives towards the understanding of laminopathy development. To gain insight into the forces driving genome–nuclear lamina interactions, I first developed a technique to locally scramble LAD and iLAD sequences. This method consists in 1) randomly relocating loxP sites in the genome by hopping of the Sleeping Beauty (SB) transposable element, and 2) creating local recombinations between them (i.e. deletions and inversions) by Cre-lox recombination. I generated 12 cell lines, each harbouring distinct LAD-iLAD recombinations, and then explored their new LAD pattern. My results show that LAD-NL interactions are multivalent. Interestingly, I identified tethering elements as more potent than others. Those have an autonomous affinity for the NL and can boost the NL association of flanking sequences. Finally, I also show that neighbouring LADs can cooperate, if close enough in the linear space, to boost their association with the NL.

Data: CORDIS, © European Union