Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2014 Feb;24(2):260-6.
doi: 10.1101/gr.157750.113. Epub 2013 Nov 27.

Canonical Nucleosome Organization at Promoters Forms During Genome Activation

Affiliations
Free PMC article

Canonical Nucleosome Organization at Promoters Forms During Genome Activation

Yong Zhang et al. Genome Res. .
Free PMC article

Abstract

The organization of nucleosomes influences transcriptional activity by controlling accessibility of DNA binding proteins to the genome. Genome-wide nucleosome binding profiles have identified a canonical nucleosome organization at gene promoters, where arrays of well-positioned nucleosomes emanate from nucleosome-depleted regions. The mechanisms of formation and the function of canonical promoter nucleosome organization remain unclear. Here we analyze the genome-wide location of nucleosomes during zebrafish embryogenesis and show that well-positioned nucleosome arrays appear on thousands of promoters during the activation of the zygotic genome. The formation of canonical promoter nucleosome organization is independent of DNA sequence preference, transcriptional elongation, and robust RNA polymerase II (Pol II) binding. Instead, canonical promoter nucleosome organization correlates with the presence of histone H3 lysine 4 trimethylation (H3K4me3) and affects future transcriptional activation. These findings reveal that genome activation is central to the organization of nucleosome arrays during early embryogenesis.

Figures

Figure 1.
Figure 1.
Well-positioned nucleosome arrays are established at promoters during genome activation. (A) Sample profile of nucleosome organization around the promoter of supt6h at the 256-cell (blue, before the maternal-zygotic transition) and dome (green, after the maternal-zygotic transition) stages. Gaussian smoothing of midpoints of nucleosomal DNA was used to create the nucleosome profile. (B) Relationship between the position of adjacent nucleosomes before (blue) and after (green) the maternal-zygotic transition. The total number of nucleosomal sequencing reads with a start-to-start distance equal to the value on the x-axis is shown. (C) Genomic distribution of well-positioned nucleosome arrays at the 256-cell and dome stages. Promoters: regions from 2 kb upstream of the TSS to 2 kb downstream from the TSS. Gene bodies: regions from 2 kb downstream from the TSS to the transcription termination sites (TTS). Intergenic regions: all other genomic regions. (D) Average promoter nucleosome organization across all genes before (blue) and after (green) the maternal-zygotic transition. All reads were extended to 147 bp; the middle 73 bp were taken, piled up, and normalized by sequencing coverage. The +1, +2, +3, and −1 nucleosomes around TSS at the dome stage are indicated.
Figure 2.
Figure 2.
Well-positioned nucleosome arrays and H3K4me3 co-occur at promoters at the dome stage. (A) Sample profile of nucleosome organization (green) and H3K4me3 (red) around mrpl16 and mrpl39. Gaussian smoothing of the midpoints of nucleosomal DNA was used to represent the nucleosome profile. Identified well-positioned nucleosome arrays are indicated by black bars. (B) Venn diagram showing the overlap of well-positioned nucleosome arrays and H3K4me3 peaks at promoters. A 100-bp overlap was used as a minimal required cutoff. (C) Heatmaps of nucleosome organization; H3K4me3, H3K27me3, and RNA Pol II signals around TSSs; as well as the H3K36me3 signal in the last 2/3 regions of the concatenated exons. Genes are ranked by the H3K4me3 signal at their promoter. Each horizontal line represents the average signal for 100 genes. Color represents RPKM value. Genes were evenly grouped into 20 bins based on H3K4me3 density at their promoters, and the associated distribution of nucleosome array values at these promoters is given in the boxplot. In heatmaps, short genes (<1 kb) were excluded, and if one gene has multiple annotations, only the one with the strongest H3K4me3 signal was kept. A total of 18,890 genes was used. (D) The correlation between H3K4me3 density and nucleosome array value at the promoters. H3K4me3 density is calculated as the log2 transformed RPKM + 1 for each promoter.
Figure 3.
Figure 3.
Canonical nucleosome organization at promoters in the absence of transcriptional elongation. (A) Average promoter nucleosome organization for transcribed (red) and nontranscribed (blue) genes. Transcription status is inferred from H3K36me3 ChIP-seq enrichment in exons and RNA Pol II ChIP-seq enrichment in gene bodies. (B) Average promoter nucleosome organization for three types of nontranscribed genes: bivalent (marked by both H3K4me3 and H3K27me3 at promoters; red), monovalent (marked by H3K4me3, but not by H3K27me3 at promoters; blue) and nonmarked (marked neither by H3K4me3 nor by H3K27me3 at promoters; black) genes. (C) Average promoter nucleosome organization for three types of genes: with elongating Pol II (red), with nonelongating Pol II (blue), and with H3K4me3 at promoters but without Pol II (black). (AC) All reads were extended to 147 bp; the middle 73 bp were taken, piled up, and normalized by sequencing coverage. Number of genes used to draw an average profile was indicated for each class. Short genes (<1 kb) were excluded. The accompanying nucleosome profiles for each class of genes in the normalized version and at the 256-cell stage are shown in Supplemental Figure S10.
Figure 4.
Figure 4.
Nucleosome organization predicts H3K4me3 and transcription. Genes are grouped evenly into five bins based on promoter nucleosome array value at the 256-cell stage. G1 represents the lowest nucleosome array value, G5 the highest. The sum of the fractions of five groups is 1. (A) The fraction of genes marked by H3K4me3 at promoters at the oblong (black) and dome (gray) stage for each group. (B) The fraction of genes transcribed at the dome stage in each group. Transcription status is inferred from H3K36me3 ChIP-seq enrichment in exons and RNA Pol II ChIP-seq enrichment in gene bodies. Likelihood ratio test of linear correlation is used to calculate P-value.

Similar articles

See all similar articles

Cited by 36 articles

See all "Cited by" articles

Publication types

Associated data

LinkOut - more resources

Feedback