Evolutionary dynamics of genomic regulatory blocks: HCNEs and HARs, the content and the boundaries

摘要

Until recently, conserved gene synteny was considered to be a consequence of insufficient time of separation between the observed genomes, preventing gene order shuffling by chromosomal rearrangements. New evidence shows that vertebrate chromosomes contain large territories of long-range regulation,termed Genomic Regulatory Blocks (GRBs).GRBs contain target genes that respond to long-range enhancers, which can typically be recognized as highly conserved noncoding elements between species (HCNEs). Additionally, they often contain functionally unrelated bystander genes, which are kept in synteny with the target gene by the regulatory elements contained in their introns or beyond.Here we attempt to answer two basic questions about GRBs. First, why do they contain bystander genes, and second, what are the underlying mechanisms that shape the extent and boundaries of GRBs?Given that HCNEs typically represent regulatory inputs that need to be present in cis to their corresponding target gene to function; is the principal force that causes retention of bystander genes in GRBs through evolution that they harbor HCNEs? For about 200 GRBs we defined previously, we checked the ortholog status (retained/lost) of all bystander genes,and the HCNEs presence or absence in each.We found that the retention of bystander genes in cis with target genes was significantly associated with the harboring of intragenic HCNEs in the bystanders (chi2 test, p'value:4.633e-5).Recent studies in Drosophila suggest that HCNEs distant to the target gene (elsewhere referred to as so-called shadow enhancers)evolve more rapidly than the enhancers mapping near or within target genes. If this principle also applies to vertebrate GRBs, it might explain the observed distribution of HCNEs around target genes and the constraints they impose on genome rearrangements during evolution. To test the hypothesis, we examined the distribution of distances between target genes and human-accelerated regions (HARs)from two recent independent studies, and compared it to the total set of HCNEs in the corresponding GRBs. The results show that HARs tend to occupy peripheral positions in GRBs, close to the very edges as defined conserved synteny. This might point to GRB edges as sites of high developmental regulatory innovation.