Skip to main content
. 2017 Mar 13;6:e22631. doi: 10.7554/eLife.22631

Figure 1. OCT4 binds distal regulatory sites in mouse embryonic stems cells to shape chromatin accessibility.

(A) Western blot analysis of OCT4cond (ZHBTC4) mouse ESCs before (UNT) and after 24 hr treatment with doxycycline (DOX). (B) Alkaline phosphatase staining of OCT4cond mouse ESCs before (UNT) and after 24 hr DOX treatment. (C) A metaplot of OCT4 ChIP-seq signal in OCT4cond ESCs before (UNT) and after 24 hr DOX treatment at OCT4 peaks (n = 15920). (D) Annotation of OCT4 peaks as promoters or distal regulatory elements using the relative enrichment of promoter-associated H3K4me3 or distal regulatory element-associated H3K4me1. (E) A genomic snapshot of ATAC-seq and OCT4 ChIP-seq signal in OCT4cond ESCs before (UNT) and after 24 hr DOX treatment at the Utf1 locus. The downstream OCT4-bound regulatory element is highlighted in the grey box. (F) A heatmap illustrating OCT4 targets (n = 15920) ranked by their loss of chromatin accessibility (ATAC-seq) after 24 hr DOX treatment of OCT4cond ESCs. Normalised read densities for ATAC-seq and OCT4 ChIP-seq are presented, with a heatmap indicating their annotation as either promoters or distal regulatory elements (right). (G) A metaplot of OCT4cond ATAC-seq signal before (UNT) and after 24 hr DOX treatment at OCT4 binding sites with significant reductions in ATAC-seq signal (OCT4-dependent; n = 11557) and those without significant changes (OCT4-independent; n = 4362). Tn5 control represents transposition of purified genomic DNA to control for potential sequence bias. (H) As in (G), profiling the changes in nucleosome occupancy before (UNT) and after (DOX) OCT4 depletion. Nucleosome signal was generated using the NucleoATAC package. (I) Piecharts identifying the proportion of OCT4-bound distal regulatory elements (left) or OCT4-bound promoters (right) that display significant changes in chromatin accessibility as measured by ATAC-seq. Changes were deemed to be significant with FDR < 0.05 and a fold change greater than 1.5-fold. (J) A metaplot depicting the OCT4cond ATAC-seq signal before (UNT) and after (DOX) treatment at the 25% of OCT4 peaks with the greatest changes in ATAC-seq signal following OCT4 depletion. (K) Gene ontology analysis for genes closest to OCT4 target sites depicted in (J). This reveals an enrichment for the pluripotency expression network (left) and biological processes associated with developmental gene regulation (right).

DOI: http://dx.doi.org/10.7554/eLife.22631.003

Figure 1.

Figure 1—figure supplement 1. Annotation and characterisation of OCT4 binding sites in OCT4cond ESCs.

Figure 1—figure supplement 1.

(A) Motif enrichment analysis for canonical motif sequences (top) or de novo motif sequences (bottom) reveals high enrichment for OCT4 (POU5F1) and similar motif sequences in the OCT4cond OCT4 binding sites (n = 15920). % reflects the percentage of OCT4 peaks identified to contain each motif sequence. (B) Central enrichment analysis performed by CentriMO reveals that canonical and de novo OCT4 motifs are centrally enriched at OCT4 peak intervals. (C) A metaplot of H3K4me1 and H3K4me3 at OCT4 binding sites annotated as distal regulatory elements or promoters. (D) A violin plot comparing the distance from OCT4 binding sites annotated as distal regulatory elements or promoters to nearest refGene TSS. p denotes significance value by Wilcoxon ranked-sign test.
DOI: http://dx.doi.org/10.7554/eLife.22631.004
Figure 1—figure supplement 2. Changes in chromatin accessibility at OCT4-bound sites following depletion of OCT4 in ESCs.

Figure 1—figure supplement 2.

(A) Genomic snapshots of OCT4-dependent chromatin accessibility (ATAC-seq) before (UNT) and after (DOX) OCT4 depletion. Distal OCT4 binding sites are highlighted in grey boxes. (B) Genomic snapshot of OCT4-independent chromatin accessibility at OCT4-bound promoter.
DOI: http://dx.doi.org/10.7554/eLife.22631.005
Figure 1—figure supplement 3. Chromatin accessibility profiling at OCT4 binding sites in somatic cell lines or tissues.

Figure 1—figure supplement 3.

(A) A metaplot of ENCODE DNase-seq profiles for eight mouse cell lines or tissues (Yue et al., 2014) at OCT4 targets most dependent upon OCT4 for normal chromatin accessibility (25% most affected in OCT4cond ESCs; as in Figure 1J). This reveals that these sites are completely inaccessible in somatic cell lines or tissues which lack OCT4 expression. (B) A metaplot analysis of ENCODE DNase-seq data in (A) profiled at DNase I hypersensitive sites identified from each cell line/tissue.
DOI: http://dx.doi.org/10.7554/eLife.22631.006