Fig 2: Euchromatic histone methyltransferases (EHMTs) as inversely regulated genes with CXCL10 level in high-risk neuroblastomas and its association with MYCN amplification. (A) Genes correlating with CXCL10 expression in INSS 4 NBs ranked by increasing Pearson correlation coefficients. Epigenetic modifiers from Xu et al30 are highlighted. Red: positively correlated; blue: negatively correlated. (B) As in (A) but showing genes correlating with MYCN expression. (C) Boxplots showing EHMT2 and EHMT1 expression (RNA-seq, log2) in INSS 4 NBs by MYCN status. (D) Boxplots showing EHMT2 and EHMT1 expression (3′mRNA-seq, log2) in human NB cell lines by MYCN status. (E) Western blots for MYCN, EHMT1, EHMT2, EZH1, EZH2 and β-actin in siMYCN or non-targeting non-targeting siRNA (siNT)-treated SK-N-BE and IMR-32 cells. Representative blots of biological replicates (n=4). (F, G) Quantification of experiment described in (E) from biological replicates (n=4). Error bars; SD. (H) and (I) MYCN-binding peaks in the genomic regions of EHMT2 (upper panels) and EZH2 (lower panel) by ChIP-seq. Cell lines and data set as indicated. (J) and (K) EHMT2 and EHMT1 expression in NB cells versus other cancer cells from the CCLE database. Statistics: two-sided unpaired t-tests (C, D, J, K). Two-sided unpaired t-test with logarithmic values (F, G). Boxplots: Boxes indicate second and third quartile. Bars indicate first and fourth quartile. Horizontal line represents median. *p<0.05; **p<0.01; ***p<0.001. Otherwise p values as indicated. NB, neuroblastoma.

Fig 3: EZH2 is required for NEPC cells.a Expression profiles of PRC subunits in PAC and NEPC samples. P-value adjusted by Bonferroni method. **P < 0.01, ****P < 0.0001. (All values below 0 were close to 0 and the green color was therefore not distinguishable on the right bar). b qPCR analysis of EZH2 and PHF19 expression in our tumor specimen cohort including 40 PAC and eight NEPC samples. ****P < 0.0001. c qPCR analysis of expression of four PRC genes in control PC and NEPC cells. **P < 0.01, ****P < 0.0001. d Western blotting of EZH1 and EZH2 in control PC and NEPC cells. e ChIP enrichment analysis of EZH2-binding site on miR-708 in control PC cells and NEPC cells. f Effect of EZH2 knockdown and EZH2 inhibition by GSK-126 on NEPC markers expression. EZH2-knockdown PC cells were treated with NEPC-inducing medium for 6 days, or control cells were treated with NEPC-inducing medium with GSK126 (2.5 μM) for 6 days. NEPC phenotype was evaluated by qPCR of NEPC markers CgA and SCG3. g Effects of EZH2 inhibitor GSK126, docetaxel (DTX), PI3K/AKT pathway inhibitor LY294002 alone, combination of GSK126 with DTX or LY294002, or DTX + LY294002 on cell viability of NEPC cells derived from C4–2. NEPC cells were treated with vehicle, GSK126 (5 μM), DTX (10 nM), LY294002 (LY, 5 μM), DTX + LY294002, DTX + GSK126, or DTX + GSK126 + LY for 48 h, and the percentages of viable cells were counted and calculated. *P < 0.05, **P < 0.01. h miR-708 expression in NEPC cells derived from C4–2 treated with the indicated drugs for 24 h. **P < 0.01, ****P < 0.0001. i Effects of anti-miR-708 on NEPC markers expression when EZH2 was inhibited. C4–2 cells were transfected with anti-miR-708 and then treated with NEPC-inducing medium for 6 days with or without GSK126. **P < 0.01. Error bars represent the standard deviation of biological triplicates

Fig 4: MCV LT activates EZH2 gene expression by targeting the pRb-E2F pathway. A. qRT-PCR analysis for Ezh1 and Ezh2 mRNA expression in BJ cells stably transduced with empty vector (Empty), HA-E6, HA-E7, and HA-E7LYCDK pRb binding mutant. Relative mRNA expression to empty vector control was determined by the 2-ΔΔCt method with GAPDH as a normalization control. Error bars represent standard deviation. B. EZH2, EZH1, and E2F-1 protein expression were examined by immunoblot using stable cells from Fig. 1A. Hsp70 was used as a loading control. C. qRT-PCR analysis for Ezh1 and Ezh2 mRNA expression in BJ cells stably transduced with empty vector (Empty), MCV sT, tumor-derived 339 LT, and 339LTLFCDK pRb binding mutant. Relative mRNA expression to empty vector control was determined by the 2-ΔΔCt method with 18S as a normalization control. Error bars represent standard deviation. D. BJ cells used in Fig. 1C were immunoblotted for EZH2, EZH1, and E2F-1 expression. Hsp70 was used as a loading control.

Fig 5: Combination EZHi/DNMTi represses oncogenic programs independently of H3K27ac driven gene activationA) Gene set enrichment analysis (GSEA) for VD (VAL+DAC)-associated transcription versus REST (VEH/DAC/VAL) across HALLMARK and KEGG gene sets.B) Scatterplot of gene expression change (relative to VEH) between VD treatment (y-axis) versus VD + the p300/CBP inhibitor A485 (VDA) treatment (x-axis). Each dot represents an individual gene with coordinates derived from designated drug treatment gene expression change (Log2FC relative to VEH).C) Supervised clustered heatmap of relative gene expression across drug treatments (row z-score, each row is a gene) for genes significantly upregulated by VD treatment. Roughly 83% of genes are sensitive to A485 treatment.D) Scatterplot of GSEA results between VD treatment (y-axis) versus VDA treatment (x-axis). Each dot represents an individual gene set Normalized Enrichment Score (NES) for the designated drug treatment within the HALLMARK/KEGG gene sets.E) Supervised clustered heatmap of NES across drug treatments for gene sets of interest.F) Outgrowth measurements (% confluency) of RKO cells treated once with DAC (300 nM) and/or valemetostat (VAL, 1 μM) with or without A485 (10 μM) and observed for 10 days. Data are the mean ±SEM of technical replicates from a single experiment (n=12 images per timepoint and treatment) and are representative of three biological replicates.G) MA plots for level of H3K27ac (in VEH) versus change in H3K27ac signal (relative to VEH) across the drug treatments. H3K27ac peaks within 10kb of a gene promoter were considered. Differentially expressed genes for each drug treatment are highlighted for their association with H3K27ac dynamics.H) Scatterplot of H3K27me1 change in single-agent EZHi treatment (x-axis) compared to the H3K27ac change in EZHi/DNMTi combinations treatments for H3K27ac loss that occurred within 20kb (+/−) of H3K27me1 signal. H3K27ac peaks are colored by their condensed chromHMM state. Large dots represent significant H3K27ac loss with FC ≤ 0.5 (log2FC ≤ −1) relative to VEH.I) Scatterplot of H3K27me1 change (within 20 kb of H3K27ac loss) between single-agent EZH1/2i (VAL) and EZH2i (TAZ) treatments at VD-associated significant H3K27ac loss (bottom left quadrant of VD plot in H). H3K27ac peaks are colored by the condensed chromHMM state of the nearest H3K27me1 peak.J) Average profiles of gene body H3K27me1 signal (siQ-ChIP) from genes described in I. The average profiles are presented on top with the change in the modification relative to VEH presented as heatmaps on bottom.K) Average profiles of gene body H3K27ac signal (siQ-ChIP) from genes described in I for single-agent EZHi (left) and EZHi/DNMTi (right). The average profiles are presented on top with the change in the modification relative to VEH presented as heatmaps on bottom.L) Average profiles of gene body DNA methylation (EPIC array) from genes described in I for single-agent EZHi (left) and EZHi/DNMTi (right) The average profiles are presented on top with the change in the modification relative to VEH presented as heatmaps on bottom.M) Boxplot of the change in gene expression (relative to VEH) for each drug treatment for genes described in I. Dots represent individual genes.See also Figure S8.