The role of the 3d architecture of the nucleus in shaping vertebrate transcriptional regulation

Resumen

Animal morphological diversity is astonishing and it is partially due to differences in gene expres- sion between different species during development. Recently, the genome folding in Topologically Associated Domains (TADs) found in most animals has been shown to be critical in the control of transcription during development. Distal enhancers are able to interact with the promoters of developmental genes only when they belong to the same 3D environment or TAD. Therefore, we investigated how changes in the 3D folding of the genome could have impacted the changes in gene regulation responsible for the evolution of vertebrates. In order to do so we combined syntenic analysis with Chromatin Conformation Capture experiments such as 4C-seq and HiChIP. First we compared the chromatin folding around the zebrafish HoxD and the amphioxus Hox loci using 4C-seq experiments coupled to computational modelling. The chromatin architecture around the vertebrate HoxD locus is peculiar, with al the HoxD genes located at the boundary between two TADs allowing them to switch to respond to distal enhancers located in either of the two TADs during the patterning of the limbs. In contrast, all the amphioxus Hox genes belong to the same TAD. However, the region located downstream from Hox1 is homologous to the vertebrate anterior TAD and is wired both in 3D and functionally to the regulation of Hox genes also in amphioxus. This suggests a stepwise evolution of the chromatin folding in two TADs found in extant vertebrates, with the anterior TAD being already wired to Hox genes in the last common ancestor of chordates. Second we performed a genome wide comparison of the chromatin folding between zebrafish and amphioxus using HiChIP and antibodies against different histone modifications. Using H3K4me3 HiChIP experiments we were able to identify the Regulatory Landscapes (RLs) of all active devel- opmental promoters using a single experiment. By doing so we were able to identify almost four hundred cases of chromosomal rearrangements that potentially altered the boundaries of a TAD and were susceptible to generate regulatory novelties in the vertebrate lineage. Also, we found that the two events of whole genome duplication that occurred at the root of vertebrates allowed some of the paralog genes originated to increase their RLs both in size and in number of enhancers.