Fig 2: Decreased chromatin accessibility of stemness genes in BRPF1−/− H1 hESCs(A) Heatmap and signal densities of ATAC-seq data for OC accessible chromatin regions in WT and BRPF1−/− H1 hESCs (D12, D30). The signal around a ±3 kb window centered on OC peaks center in the indicated cells. OC, open to closed, indicates regions with reduced accessibility in BRPF1−/− H1 hESCs.(B) Box-plot of signal densities of the ATAC-seq data for OC accessible chromatin regions in the indicated cells.(C) GO analysis of genes associated with OC regions is shown.(D) Heatmap and signal densities of ATAC-seq data for CO accessible chromatin regions in WT and BRPF1−/− H1 hESCs (D12, D30). The signal around a ±3 kb window centered on CO peaks center in the indicated cells. CO, close to open, indicates regions with increased accessibility in BRPF1−/− H1 hESCs.(E) Box-plot of signal densities of the ATAC-seq data for CO accessible chromatin regions in the indicated cells.(F) GO analysis of genes associated with CO regions is shown.(G) Box-plot of transcription levels of selected stemness genes in H1 hESCs, H1 hESC-derived NPCs, and trophoblast stem (TS) cells.(H) Heatmap and signal densities of ATAC data in regions associated with selected stemness genes. ATAC-seq signal around a ±5 kb window centered on OC peaks in the indicated cells.(I and J) Genomic view of ATAC-seq data on selected genes with different functions in H1 and BRPF1−/− hESCs. These selected genes include (I) hESCs stemness genes and (J) cell fate commitment genes.

Fig 3: Loss of histone H3 lysine 23 acetylation in BRPF1−/− hESCs(A) Morphology of WT and BRPF1−/− H1 hESCs cultured at different time points. Scale bar, 200 μm. qRT-PCR analysis of the expression level of BRPF1 in WT and BRPF1−/− H1 hESCs cultured at different time points. The significance level was determined by unpaired two-tailed Student’s t tests. ∗∗, p < 0.01. The data represent the mean ± SD (standard deviation) from three independent repeats (n = 3).(B and C) Examination of the expression of the selected (B) pluripotent and (C) lineage genes in BRPF1−/− H1 hESCs cultured at different time points. The significance level was determined by unpaired two-tailed Student’s t tests. ∗∗, p < 0.01. The data represent the mean ± SD (standard deviation) from three independent repeats (n = 3).(D) Examination of H3K23ac, H3K14ac, H3K4me3 or OCT4, NANOG in BRPF1−/− H1 hESCs cultured at different time points by western blot. All error bars throughout the figure represent the SD (standard deviation) from three independent replicates (n = 3).

Fig 4: Co-localization of BRPF1, H3K4me3, and H3K23ac on fundamental and stemness genes in human ESCs(A) Signal densities for input, H3K4me3, BRPF1 (FLAG), and H3K23ac ChIP-seq data around ±5 kb centered on BRPF1 (FLAG) associated peaks in KI hESCs.(B) BRPF1 (FLAG) associated peaks classification, H3K4me3/H3K23ac/FLAG co-localization peaks (yellow); H3K4me3/FLAG co-localization peaks (indigo); H3K23ac/FLAG co-localization peaks (red); and only FLAG localization peaks (none, gray).(C) Upper panel, all genes from the H1 RNA-sep in Figure 1 broken down by decile of expression (1st being lowest 10% and 10th being highest 10%). lower panel, The average signal intensity of H3K4me3, BRPF1 (FLAG), and H3K23ac ± 5 kb around the TSS of 1st, 3st, 5st, 7st, and 10st.(D) Signal densities for H3K4me3, BRPF1 (FLAG), and H3K23ac ChIP-seq data ±5 kb around the TSS of selected stemness genes in KI hESCs.(E) Signal densities of H3K4me3, H3K23ac, and ATAC-seq data ±3 kb around the peak center of open to close (O-C) and close to open (C-O) in WT hESCs.(F and G) Genomic views of signal densities of ChIP-seq data of H3K4me3 (red), BRPF1 (FLAG) (blue), and H3K23ac (purple) on selected genes in KI hESCs. These selected genes include (F) hESCs stemness genes and (G) cell fate commitment genes.

Fig 5: Functional analysis of BRPF1 with the deletion of different histone binding modules(A) Overview of wild-type BRPF1 and the BRPF1 mutants with domain architecture. Wild-type BRPF1 (B_WT); BRPF1 with N terminal deletion (B_N−); BRPF1 with PWWP module deletion (B_PWWP−); BRPF1 with PZP module deletion (B_PZP−). The dotted line indicates the deleted regions.(B) qRT-PCR analysis of the expression with domain-specific primers. The significance level was determined by unpaired two-tailed Student’s t tests. ∗∗, p < 0.01. The data represent the mean ± SD (standard deviation) from three independent repeats (n = 3).(C) Western blot analysis of FLAG-tagged of overexpression of wild-type BRPF1 and each BRPF1 mutant in BRPF1−/− H1 hESCs. Overexpression of the negative control in BRPF1−/− H1 hESCs (B_3FLAG).(D) Morphology of BRPF1−/− H1 hESCs overexpressing wild-type BRPF1 and each BRPF1 mutant. Scale bar, 200 μm.(E) FACS analysis of OCT4 in BRPF1−/− H1 hESCs overexpressing wild-type BRPF1 and each BRPF1 mutant. The significance level was determined by unpaired two-tailed Student’s t tests. ∗∗, p < 0.01. The data represent the mean ± SD (standard deviation) from three independent repeats (n = 3).(F) qRT-PCR analysis of the expression of the pluripotent genes OCT4, NANOG, and cell fate commitment genes PAX6, RUNX1, and SOX17 in BRPF1−/− H1 hESCs overexpressing wild-type BRPF1 and each BRPF1 mutant. The significance level was determined by unpaired two-tailed Student’s t tests. ∗∗, p < 0.01. The data represent the mean ± SD (standard deviation) from three independent repeats (n = 3).(G) Western blot analysis of H3K4me3 and H3K23ac in BRPF1−/− H1 hESCs overexpressing wild-type BRPF1 and each BRPF1 mutant. All error bars throughout the figure represent the SD (standard deviation) from three independent replicates (n = 3).