Fig 1: Bap1 null mesothelioma growth requires PCGF3/5-dependent H2AK119ub1 accumulation(A) Western blot using the indicated antibodies in whole-cell lysates of the BAP1 null IST-MES2 mesothelioma cell line with either parental or BAP1 WT/C91S overexpression.(B) RNA-seq heatmap of those genes differentially expressed in +BAP1 WT IST-MES2 versus parental IST-MES2.(C) Metaplots and heatmaps representing normalized ChIP-seq intensity for H2AK119ub1 in the indicated cell lines.(D) Boxplots representing H2AK119ub1 (left) or H3K27me3 (right) ChIP-seq RPKM levels in the indicated cell lines at intergenic sites.(E) Western blot using the indicated antibodies in IST-MES2 cell line whole-cell lysates following knockout of PCGF3 or PCGF5 in the indicated combinations.(F) Growth curves measured using crystal violet staining (λ = 590 nm) of the indicated cell lines. Data are represented as mean ± SD.(G) Model of the dual role of BAP1 mode of action on transcription. BAP1 is essential for the spatial maintenance of H2AK119ub1 and H3K27me3. Spurious redistribution of these in the absence of BAP1, directed by the PCGF3/5-PRC1 and PRC2.2 complexes, promotes chromatin compaction and a general repression of transcription (Trithorax phenotype) while simultaneously allowing derepression of selected Polycomb target genes (Polycomb phenotype).
Fig 2: Diffuse H2AK119ub1 accumulation causes global chromatin compaction(A) Ice-normalized HiC contact matrix of the entire chromosome 11 in WT, Bap1 KO, and log2 fold change (BAP1 KO/WT) at 250 kbp resolution using two pooled HiC replicates.(B) Boxplot of contact frequency of log2 fold change (Bap1 KO/WT) ratios divided into quartiles using two pooled HiC replicates.(C) Boxplot of the log2 fold change (Bap1 KO/WT) ratio of the indicated histone modifications within the quartiles defined in (B). Wilcoxon test was used to ascertain significance.(D) Top: Log2 fold change (Bap1 KO/WT) HiC contact matrix of the indicated region of chromosome 12 (54.8–59.15 Mb) at 10 kb resolution. Bottom: Genome browser snapshot of indicated ChIP-seq tracks at the same region of chromosome 12.(E) Representative stochastic optical reconstruction microscopy (STORM) images of the indicated cell lines stained with histone 3. Scale bars of 1 μm (left column) and 0.5 μm (right column) are shown.(F) STORM quantifications of percentage clustered (top) and median points per cluster (bottom) of H3. Data are represented as mean ± SD.See also Figure S5.
Fig 3: BAP1 catalytic activity maintains stable PRC2 target association and spatial distribution of H3K27me3(A) Metaplots and heatmaps representing normalized ChIP-seq intensity for H3K27me3 (left) or SUZ12 (right).(B) Boxplot of normalized intensity profiles for H3K27me3 and SUZ12 ChIP-seq in the indicated cell lines.(C) Metaplots and heatmaps representing normalized ChIP-seq intensity for CBX7 (left) and RYBP (right) in the indicated cell lines.(D) Boxplots representing H3K27me3 ChIP-seq RPKM levels in the indicated cell lines at intergenic sites (n = 38,068).(E) Genome-wide XY scatterplot of normalized H3K27me3 ChIP-seq intensities at a resolution of 5 kb in Bap1 KO compared to WT ESCs. Each point represents one 5 kb window.(F) Genome browser snapshot of indicated ChIP-seq tracks at the HOXA locus in WT and Bap1 KO ESCs.See also Figure S3.
Fig 4: BAP1 binds active gene promoters and is excluded from Polycomb repressive domains(A) Heatmaps representing ChIP-seq intensity of the indicated proteins in wild-type ESCs.(B) Venn diagram of HA-BAP1, SUZ12, and RING1B target genes in ESCs.(C) Genome-wide functional annotation of peaks generated from the indicated ChIP-seq analyses.(D) Boxplots showing the expression levels obtained from RNA-seq analyses in WT mESCs for the clusters of target genes generated in (A).(E) Genome browser snapshot of ChIP-seq tracks showing an example of mutual exclusivity of PRC1/2 and PR-DUB target genes.(F) Western blot analysis with the indicated antibodies on total protein extracts from the indicated rescue ESC cell lines (E14 WT + empty vector, Bap1 KO + empty vector, Bap1 KO + BAP1 WT, Bap1 KO + BAP1 C91S).(G) Volcano plots of −log10 (p value) against log2 fold change representing the differences in gene expression in the indicated cell lines.(H) Percentage overlap of differentially expressed genes (DEGs) from (G) with either HA-BAP1, RING1B, or SUZ12 ChIP-seq targets.See also Figure S1 and Tables S1, S2, and S3.
Fig 5: BAP1 loss causes global increases in H2AK119ub1 and displacement of PRC1 from target loci(A) Metaplots and heatmaps representing normalized ChIP-seq intensity for H2AK119ub1 or RING1B in the indicated cell lines.(B) Boxplot of normalized intensity profiles for H2AK119ub1 and RING1B ChIP-seq in the indicated cell lines.(C) Boxplot representing H2AK119ub1 ChIP-seq RPKM levels in the indicated cell lines at intergenic sites (n = 38,068).(D) Representation of the log2 fold change CPM in H2AK119ub1 ChIP-seq signal in the indicated cell lines across chromosome 19 using 10 kb windows.(E) Schematic of experimental plan to biochemically characterize PR-DUB in chromatin and nucleosol fractions.(F) Western blot of the indicated cell lines in either nucleosol or chromatin fractions.(G) Comparison of stoichiometry (IBAQ relative to BAP1) of PR-DUB subunits in FLAG/HA-BAP1 IP mass spectrometry purifications from nucleosol and chromatin fractions. Data are represented as mean ± SD.(H) Model of activity of PR-DUB complex at both its bound target genes and in “hit-and-run” model of highly mobile nucleosolic complex throughout the genome.See also Figure S2.
Supplier Page from Abcam for Anti-RING2 / RING1B / RNF2 (phospho S168) antibody [EPR20902-58]