Click
here to close Hello! We notice that
you are using Internet Explorer, which is not supported by Echinobase
and may cause the site to display incorrectly. We suggest using a
current version of Chrome,
FireFox,
or Safari.
Sci Adv
2022 Dec 02;848:eabn2258. doi: 10.1126/sciadv.abn2258.
Show Gene links
Show Anatomy links
Active DNA demethylation of developmental cis-regulatory regions predates vertebrate origins.
Skvortsova K
,
Bertrand S
,
Voronov D
,
Duckett PE
,
Ross SE
,
Magri MS
,
Maeso I
,
Weatheritt RJ
,
Gómez Skarmeta JL
,
Arnone MI
,
Escriva H
,
Bogdanovic O
.
???displayArticle.abstract???
DNA methylation [5-methylcytosine (5mC)] is a repressive gene-regulatory mark required for vertebrate embryogenesis. Genomic 5mC is tightly regulated through the action of DNA methyltransferases, which deposit 5mC, and ten-eleven translocation (TET) enzymes, which participate in its active removal through the formation of 5-hydroxymethylcytosine (5hmC). TET enzymes are essential for mammalian gastrulation and activation of vertebrate developmental enhancers; however, to date, a clear picture of 5hmC function, abundance, and genomic distribution in nonvertebrate lineages is lacking. By using base-resolution 5mC and 5hmC quantification during sea urchin and lancelet embryogenesis, we shed light on the roles of nonvertebrate 5hmC and TET enzymes. We find that these invertebrate deuterostomes use TET enzymes for targeted demethylation of regulatory regions associated with developmental genes and show that the complement of identified 5hmC-regulated genes is conserved to vertebrates. This work demonstrates that active 5mC removal from regulatory regions is a common feature of deuterostome embryogenesis suggestive of an unexpected deep conservation of a major gene-regulatory module.
Fig. 1. Evolutionary conservation of TET protein sequence, catalytic domain structure, and developmental expression.(A) Sea urchin, lancelet, and zebrafish TET protein domain structure predicted using SWISS-MODEL and HMMER. Sea urchin sTet, Florida lancelet bTet, and zebrafish Tet3 harbor an N-terminal DNA binding CxxC domain and a C-terminal catalytic DSBH domain containing a large low-complexity insert. (B) Multiple sequence alignments of human, mouse, and zebrafish TET1, TET2, and TET3 as well as sea urchin, lancelet, and fruit fly TET core catalytic domain. A part of DSBH is shown. The color of each amino acid indicates percentage identity (PID) with darker blue depicting higher PID and lighter blue depicting lower PID. (C) Three-dimensional models of methylcytosine dioxygenase domains of sea urchin sTet, lancelet bTet, and zebrafish Tet3 performed using SWISS-MODEL. Crystal structure of the human TET2-5fC complex (5d9y.1.A) was ranked as a top template in the template search and was used for model building. ɑ Helix is depicted in blue, and β sheet is depicted in green. (D) TET gene expression dynamics during sea urchin, lancelet, and zebrafish development. a.a., amino acids; 2-OG, 2-oxoglutarate. CPM, counts per million.
Fig. 2. DNA methylation dynamics of sea urchin and lancelet development.(A) Average DNA methylation (mCG/CG) levels during sea urchin and lancelet development. (B) Average gene expression (CPM + 1) levels during sea urchin and lancelet development. (C) Number of hypomethylated DMRs (hypoDMRs) identified between embryonic and adult stages in sea urchin and lancelet genomes. (D) Heatmaps depicting DNA methylation (mCG/CG) at adult hypoDMRs in sea urchin and lancelet genomes.
Fig. 3. DNA hydroxymethylation landscape of sea urchin and lancelet development.(A) Percentage of methylated and hydroxymethylated CpGs in embryonic and adult tissues of sea urchin, lancelet, and zebrafish. CpG sites with at least 80% methylation and CpG sites with at least 10% hydroxymethylation (P adj. < 0.05) were included in the analysis. Only CpG sites with a minimum of 10× coverage were considered. (B) Concordance of hydroxymethylation levels between biological replicates. Sea urchin 48-hpf and lancelet 60-hpf samples are shown. (C and D) Ternary plots depicting the relationship between methylated (mC), hydroxymethylated (hmC), and unmethylated (C) CpG states in sea urchin (C) and lancelet (D) embryos and adult tissues. CpG sites hydroxymethylated at 48 hpf in sea urchin (C) and 60 hpf in lancelet (D) (P adj. < 0.05, proportion test) are shown. (E and F) Heatmaps depicting DNA hydroxymethylation dynamics (ACE-seq) of open chromatin regions (ATAC-seq) in sea urchin (E) and lancelet (F). K-means clustering (k = 2) of ACE-seq and ATAC-seq signal over ATAC-seq peaks. Boxplots of DNA hydroxymethylation and DNA methylation dynamics during sea urchin (E) and lancelet (F) development. (G) Integrative genomic viewer (IGV) browser tracks depicting DNA hydroxymethylation (ACE-seq and hMeDIP-seq) enrichment at open chromatin regions (ATAC-seq) coinciding with DNA demethylation (MethylC-seq) in sea urchin and lancelet genomes.
Fig. 4. Developmental expression of 5hmC-marked genes.(A) Gene Ontology (GO) analysis of genes harboring 5hmC-marked ATAC-seq peaks at their promoters or gene bodies in sea urchin and lancelet. (B) Percentage of 5hmC-marked ATAC-seq peaks at different genomic regions. (C) Length of the genes harboring promoter-associated 5hmC-marked ATAC-seq peaks (promoter atac hmC+) and non-5hmC ATAC-seq peaks (promoter atac hmC−). (D) HOMER motif enrichment analysis of atac hmC+ regions associated with either gene promoters or gene bodies. (E) Expression dynamics of promoter atac hmC+ genes as compared to promoter atac hmC- genes in sea urchin and lancelet embryos and adult tissues.
Fig. 5. Conservation of developmental regulatory logic of 5hmC-marked genes in zebrafish.(A) Expression of 2R/3R-ohnolog zebrafish genes associated with either 5hmC-marked ATAC-seq peaks overlapping phylo-DMRs (5hmC/phylo-DMR), 5hmC-marked ATAC-seq peaks not overlapping phylo-DMRs (5hmC/no phylo-DMR), or non-5hmC ATAC-seq peaks not overlapping phylo-DMRs (no 5hmC/no phylo-DMR) in embryos and adult tissues. (B) Uniform manifold approximation and projection (UMAP) projection of 7738 individual cells from the 24-hpf zebrafish embryo. Cells are color-coded by tissue types. (C) Z-transformed module scores of three gene sets: 5hmC/phylo-DMR, 5hmC/no phylo-DMR, and no 5hmC/no phylo-DMR in 24-hpf zebrafish embryo. Module scores depict the difference between the average gene expression of each gene set and random control genes. (D) Heatmaps showing DNA methylation profiles of 5hmC/phylo-DMR, 5hmC/no phylo-DMR, and no 5hmC/no phylo-DMR ATAC-seq peaks in 1k-cell and 24-hpf zebrafish embryos as well as multiple adult tissues.
