Fig 2: N-Myc and c-Myc upregulate JMJD6 expression by binding to the JMJD6 promoter. a, b CHP134 cells were transfected with control siRNA, N-Myc siRNA-1 or N-Myc siRNA-2, and SK-N-AS cells were transfected with control siRNA, c-Myc siRNA-1, or c-Myc siRNA-2. Forty-eight hours later, RNA and protein were extracted for RT-PCR (a) and immunoblot (b) analyses. Error bars represent normalized standard errors from three independent experiments (**p < 0.01, ***p < 0.001, one-way ANOVA). c SHEP Tet/21N cells were treated with DOX (2 µg/ml) or vehicle control for 48 h. RT-PCR and immunoblot analyses were conducted. Error bars represent normalized standard errors from three independent experiments (***p < 0.001, two-tailed unpaired Student’s t test). d Schematic representation of the JMJD6 gene promoter. TSS indicates transcription start site. Amplicons A, B, and C represented the sites for ChIP PCR primers. e ChIP assays were performed with a control IgG, anti-N-Myc or anti-c-Myc antibody (Ab), followed by PCR with primers targeting a remote negative control region (amplicon A), the E-box region (amplicon B), or the exon 2 region (amplicon C) of the JMJD6 gene in CHP134 and SK-N-AS cells. Fold enrichment of the JMJD6 gene region was calculated as the difference in cycle thresholds obtained with the anti-N-Myc or anti-c-Myc Ab compared with the control IgG, relative to input. Error bars represent normalized standard errors from three independent experiments (*p < 0.05, **p < 0.01, two-tailed unpaired Student’s t test). Source data are provided as a Source Data file

Fig 3: The growth-progression model.(A) Scheme of the growth-progression model with heritable variables relating to cell size s and cycle progression timing p. (B) Measured and simulated cell-cycle length distributions (upper). Model distribution resolved by the division-limiting process (lower). (C) Measured and modeled correlation pattern with Spearman rank correlation coefficient and bootstrap 95%-confidence bounds. (D) Proportion of simulated cells limited by growth or progression. (E) Correlation of simulated mother-daughter cycle lengths colored by their division limitation: both by τg (black), both by τp (green), mother τp – daughter τg (magenta), mother τg – daughter τp (cyan). Percentage of cells in each subgroup and their correlation coefficients are shown. (F) Correlation of simulated cousin-cousin cycle length colored by the limitation of the common grandmother: by τg (orange) or τp (blue). (G) Autocorrelations along ancestral line of cycle length τ, growth time τg and the progression time τp, and the cross-correlation τp⁢τg.Figure 3—source data 1.Best-fit parameter values of the growth-progression model for all experiments shown, obtained from ABC-simulations.Cell size was assumed to increase exponentially except for MYCN-inhibited cells, which were modelled by a logistic growth process with a fixed normalised maximum cell size sm⁢a⁢x equal to 20 times the threshold division size. The bottom two rows show the parameters used for simulated perturbations of growth cycle progression, respectively.

Fig 4: Silencing TFAP4 selectively inhibits growth of MYCN-amplified neuroblastoma cell lines. a Schema of cell growth competition assay. SHEP21N cells infected with either shTFAP4-GFP or an empty vector control pGIPZ-GFP were mixed with equal number of uninfected SHEP21N cells. The infected cells to WT cells ratio is measured by FACS. b TFAP4 gene expression (qRT-PCR) in SHEP21N-pGIPZ control cells and SHEP21N expressing shTFAP4. c %GFP ratio measured at day 4, 7, 10, and 14. Experiments were performed in triplicate. Mean ± std dev. d % of neuroblastoma cell survival 96 h after siRNA-mediated downregulation of TFAP4. Neuroblastoma cell lines were transfected with four different constructs of siRNA against TFAP4. The experiments were performed in quadruplicate. Mean ± std dev. *, P < 0.05; **, P < 0.01; ***, P < 0.001. e Proliferation assays showing neuroblastoma cell growth after silencing TFAP4 by doxycycline-inducible shRNAs. Neuroblastoma cell lines were infected with two different dox-inducible shRNAs against TFAP4 as well as the empty vector control pTRIPZ. shRNA was induced by 1 μg/ml doxycycline at day 0. Cells were counted on day 0, 2, 4, and 6. Experiments were performed in triplicate. Mean ± std dev. *, P < 0.05; TFAP4 silencing was confirmed by western blot. f Colony formation assay. Neuroblastoma cells infected with doxycycline-inducible shRNAs were plated in semisolid agar media for 21 days and stained with MTT. Experiments were done in triplicate. Mean ± std dev. *, P < 0.05; **, P < 0.01

Fig 5: N-Myc and Rb1 loss tumors adopt NEPC methylation programs.a Log2 fold change in expression of DNA methylation regulators which were significantly deregulated in poorly differentiated GEM tumors. Benjamini-Hochberg adjusted two-sided t-test *p < 0.05, **p < 0.01 (p = 0.0437, 0.0267, 0.0089, 0.0164, 0.0031, 0.0582, 0.0013, 0.0131). b Normalized N-Myc binding peaks from ChIP-seq in poorly differentiated and adenocarcinoma tumors near the transcription start site of each gene. c Flowchart of the downstream analysis with differentially methylated genes of poorly differentiated tumors compared to adenocarcinomas. d Heatmap with methylation scores of hypermethylated genes (n = 102) and hypomethylated (n = 258) in two pairs of tumor tissues. e, f N-Myc binding and methylation status of upregulated genes Foxa2 (e) or Sox2 (f) in the two histological subtypes of mouse tumors. Red indicates the % of methylated C at each site for each tumor type. g Cluster map of CRPC (n = 18) and NEPC (n = 10) samples using methylation scores of 360 genes (distance method: Correlation, clustering method: Ward D2). P values indicated on each of the major clusters are based on two hypergeometric tests for the enrichment of NEPC (p = 1.45 × 10−5) or CRPC (p = 1.45 × 10−5) samples in either cluster, respectively. h Cluster map of CRPC (n = 73) and NEPC (n = 36) samples using expression of 360 genes (distance method: Correlation, clustering method: Ward D2). P values indicated on each of the major clusters are based on two hypergeometric tests for the enrichment of NEPC (p = 1.8 × 10−10) or CRPC (p = 3.6 × 10−10) samples in either cluster, respectively. i GSEA results using hypermethylated genes (top panel) from mouse tumors (n = 102) and LNCaP cells (n = 57) and hypomethylated genes (bottom panel) from mouse tumors (n = 258) and LNCaP cells (n = 185) (bottom panel). The number on each horizontal bar indicates the number of genes overlapping with the respective gene set.