Skip to content

Latest commit

 

History

History
68 lines (62 loc) · 5.58 KB

EnhancerEvolution.md

File metadata and controls

68 lines (62 loc) · 5.58 KB
Raw

Enhancer Evolution across 20 Mammalian Species

Profiling Promoter and Enhancer Regulatory Evolution in Mammalian Liver

  • liver as a representative adult somatic tissue from 20 species of mammals
  • Genome-wide ChIP-seq: H3K4me3 and H3K27ac
  • A total of 30–45,000 regions per species were enriched in liver, and these separated into H3K27ac, H3K4me3&H3K27ac, and H3K4me3-marked element.
  • For over 400 of our human liver enhancers (typically 2 kb in length), the transgenic activities of overlapping 145 bp segments were assayed in liver cancer cells. Over 65% showed activity in transgenic assays in a cancer cell line. Over 90% of the enhancers not active in transgenic assays were nevertheless bound in human liver by at least one liver-specific TF
  • In adult liver, a typical mammalian genome contains on average 12,500 H3K4me3 locations (representing active promoter elements) and 22,500 H3K27ac-enriched regions (representing active enhancers)

Enhancer Evolution Is Appreciably More Rapid Than Proximal Promoter Evolution

  • We defined the maximum detectable conservation of activity as the number of species in which the DNA could be aligned
  • the transcription initiation sites driving gene expression in liver are highly conserved
  • Our data reveal that rapid enhancer evolution, often involving exaptation of ancestral DNA, is active and widespread across all the mammalian clades in our study

Quantifying the Divergence Rates of Enhancers, Promoters, and TF Binding in a Cross-Section of Mammals

  • We first identified, by pairwise analysis of all 20 species, whether regions called as enhancers and promoters were pre- sent in the same location between two mammalian genomes
  • Inter-species conservation of promoters and enhancers could be plausibly described as a function of time-of-divergence by fitting an exponential decay curve
  • Our results are consistent with a model where the increased size and sequence heterogeneity of regions with promoter or enhancer activity could buffer evolutionary changes more robustly than can site-specific TF binding alone

Highly Conserved Regulatory Regions Are Largely Proximal Promoters

  • A total of 2,151 genomic regions appeared highly conserved (present in, at a minimum, all ten of the highest-quality placental genomes), representing 5% of all human regions active in liver.
  • 1,871/2,151 elements (87%) were enriched for both H3K27ac and H3K4me3, consis- tent with acting as promoters
  • Although nearly three times as common as promoters, the activity of only 1% of human enhancers is highly conserved. In contrast, the activity of 16% of promoters is highly conserved.
  • Three independent lines of evidence support the functionality of the sequences we identify as highly conserved regulatory regions in liver.
  • all show enhanced sequence constraint
  • genes near highly conserved enhancers are strongly enriched for liver-specific functions, and genes near conserved proximal promoters are enriched for house-keeping functions
  • highly conserved enhancers are enriched for TF binding motifs for liver-specific regulators such as CEBPA and PBX1, whereas highly conserved proximal promoters appear dominated by transcriptional initiation regulatory sequences

Recently Evolved Regulatory Activity Is Pervasive in Mammals

  • From each placental order, we selected a representative species (human, mouse, cow, dog) and then identified a set of newly evolved or, more formally, apomorphic active promoters and enhancers in liver
  • We found that a typical mammalian liver deploys between 1,000 to 2,000 promoters and 10,000 enhancers not found in any other study species; we henceforth refer to these enhancers and promoters as recently evolved.
  • Especially for enhancers, recently evolved regions are 10–20 times more abundant than those conserved across placentals or shared across multiple species in a particular lineage

Exaptation Drives Recently Evolved Enhancer, but Not Promoter, Activity

  • we tested whether functional exaptation of ancestral DNA, recently reported for human-specific enhancers active in embry- onic limb (Cotney et al., 2013), is a prevalent mechanism in mammalian genome evolution.
  • We first asked whether recently evolved proximal promoters are primarily found in ancestral DNA sequences older than 100 Ma. across four orders of mammals, the recent evolution of promoters occurred within evolutionarily younger DNA segments (i.e., not shared with other study species) about three to four times as often as occurred by exaptation of ancestral DNA.
  • Within the ancestral DNA commandeered into new promoters, and regardless of species interrogated, diverse ERV repeat ele- ments are over-represented, consistent with previous reports that ERVs are pre-primed to transcriptional initiation (Fort et al., 2014).
  • In contrast, the vast majority of enhancers in liver are recently evolved—as well as far more likely to exapt ancestral DNA. Of the typically 10,000 recently evolved enhancers in a given species,
  • 52%–77% contained sequences of ancestral DNA over 100 Ma old.
  • The remaining recently evolved enhancers were found in younger DNA, and enriched for mobile repetitive element families, including LTRs in all lineages and lineage-specific SINEs and DNA transposons exclusive to primates, carnivores, or ungulates
  • [Summary] In a typical mammalian species, the 1,000 to 2,000 recently evolved liver promoters occur predominantly in younger DNA typically less than 40 Ma old, whereas the 10,000 recently evolved enhancers are formed predominantly by exaptation of ancestral DNA.