Fig 2: Sox2 remains bound transiently at Oct4-sensitive enhancers upon Oct4 depletion.(A) Metagenes analysis of Oct4 and Sox2 occupancy at Oct4-sensitive, -insensitive, and -independent enhancers at 0 hr, data were obtained from King and Klose, 2017. y-Axis depicts ChIP-seq coverage density in reads per million (RPM). (B) Percentage of motif occurrence at Oct4-sensitive, -insensitive, and -independent enhancers for SoxOct composite motif and Oct4, Sox2, Esrrb, and Klf4 motifs. (C) Heatmap showing changes of Sox2 occupancy at Oct4-sensitive and -insensitive enhancers over the entire time course of doxycycline (DOX) treatment. Normalized read densities are shown and peaks were ranked accordingly. (D) Same as (C), but using boxplots to depict quantification of Sox2 occupancy changes at Oct4-sensitive and -insensitive enhancers. y-Axis represents Sox2 ChIP-seq density in reads per kilobase (RPKs). Black bars represent the median values for each group. Lower and upper boxes are the first and third quartiles, respectively. The ends of the whiskers extend the box by 1.5 times the interquartile range. Outliers are omitted. (E) Genome browser view illustrating changes of chromatin accessibility, Sox2 and Oct4 occupancy at Klf4 SE. Tracks from top to bottom: ATAC-seq coverages (green), ChIP-seq coverages for Sox2 (blue), and Oct4 (red) at 0, 3, 6, 9, 12, and 15 hr; ChIP-seq coverages for Oct4, Sox2, and Nanog (purple) from ZHBTc4 mouse ES cell at 0 hr (King and Klose, 2017); superenhancer (SE) annotation (Whyte et al., 2013). (F) Genome browser view illustrating changes of chromatin accessibility, Sox2 and Oct4 occupancy at Sox2 SE. Tracks were visualized and ordered in the same way as in (E). Note that data from two biological replicates were generated for all assays and that the two replicates were merged for illustration.

Fig 3: Ythdf1, Ythdf2, and Ythdf3 are redundant in mouse naïve ESC maintenance. (A) Immunostaining of Ythdf1, Ythdf2, and Ythdf3 in WT mESCs, showing a protein expression in the cytosolic compartment of the cell. (B) Cell growth curve of all KO lines and WT control. Cells were grown on mouse feeders, in serum/LIF conditions. (C) Bright field and immunostaining of Nanog (green), Esrrb (red), and DAPI (blue) in KO cells (single, triple, and Mettl3) and WT control, showing that all cell lines express Nanog and Esrrb.

Fig 4: USP21 is recruited to gene promoters and modulates H2A ubiquitination by interaction with Nanog.(a,b) Histone ubiquitination and methylation were analysed in E14 cells cultured in LIF-free condition (a) or in E14 cells stably expressing the control or shRNA targeting USP21 or Nanog (b). The anti-USP21 antibody is from Abcam. (c) Venn diagram depicting overlap between target genes of Usp21 and Nanog defined by their binding peaks (see the ‘ChIP-Seq data analysis' in the Methods section). Nanog ChIP-Seq data are referred to GEO Data Sets with project number GSE44764. (d) Binding of USP21 and its mutants to the Nanog-regulated genes. Data are means±s.d. (n=3). #P<0.05, ##P<0.01, ###P<0.001 versus control (Student's t-test); *P<0.05, **P<0.01, ***P<0.001 versus USP21 WT (Student's t-test). (e) The effect of USP21 and its mutants on the H2A ubiquitination at K119 at Nanog-regulated genes. Data are means±s.d. (n=3). ##P<0.01, ###P<0.001 versus control (two-way analysis of variance (ANOVA) test); *P<0.05, **P<0.01, ***P<0.001 versus USP21 WT (two-way ANOVA test). (f) USP21 affects the H3K4me3 level at Nanog-regulated genes. Data are means±s.d. (n=3). *P<0.05, **P<0.01 versus shCon (two-way ANOVA test). (g) Model showing that USP21 plays an important role in the maintenance of mESC self-renewal through regulating ubiquitination of Nanog and ubH2A.

Fig 5: Impact of AdOx on the expression of Nanog and Oct3/4 in P19 cells. A. Nanog and Oct3/4 expression in P19 cells with AdOx (AdOx) or without (Control) as determined by Western blot. GAPDH is a loading control. Numbers indicated quantification of the Nanog or Oct3/4 western signal relative to GAPDH respectively. Digits at the bottom of the first & second row indicate the relative darkness of each band compared with the control lane set as 1.0. B. Impact of AdOx on the relative luciferase activity of the nanog and oct4 promoters. pGL3-nanog-luc (driven by the -289/+117 fragment of the nanog gene) and pGL-oct3/4-luc (driven by the -2143/+30 fragment of the oct4 gene) were individually co-transfected with pRL-TK into P19 cells treated with AdOx or without. The relative promoter activities showed that AdOx has a different effect on the two genes in P19 cells. Annotations are as described in Figure 1G.