17q contains neuroblastoma dependency genes

a. CRISPR score for 17q genes in 10 neuroblastoma cell lines. Score <-0.4 is defined as neuroblastoma dependency genes. Data are derived from Avana sgRNA library screening63.

b. STRING protein interaction network showing 17q essential genes with various biological functions.

c. Heatmap by K-means clustering analysis showing 17q essential genes are highly expressed in high-risk neuroblastomas based on RNA-seq data (SEQC dataset).

d. Kaplan-Meier survival curve showing 17q essential gene signature is correlated with worse event-free survival (SEQC dataset).

e. Kaplan-Meier survival curve showing 17q essential gene signature is correlated with worse overall survival (SEQC dataset).

JMJD6 is required for neuroblastoma growth and facilitates MYC-mediated cellular transformation.

a. Copy number of genes encoding JmjC domain proteins in St Jude neuroblastoma cohort.

b. Kaplan-Meier survival curve showing high JMJD6 is correlated with worse event-free survival (SEQC RNA-seq dataset).

c. Crystal violet showing the colony staining on day 7 after JMJD6 shRNA knockdown in neuroblastoma cell lines validated by western blot (harvested at 72h).

d. Xenograft tumor growth of BE2C (right) models with lentiviral JMJD6 shRNA knockdown. P-value calculated by multiple unpaired t-test across each row. n=5 per group.***p<0.001, **p<0.01.

e. Xenograft tumor growth of SK-N-AS models with lentiviral JMJD6 shRNA knockdown. P-value calculated by multiple unpaired t-test across each row. ***p<0.001, **p<0.01.

f. Western blot validating the expression of retroviral based MYCN and JMJD6 in JoMa1 cells.

g. Cell proliferation of JoMa1 cells transduced with indicated constructs, GFP, JMJD6, MYCN, JMJD6+MYCN.

h. Colony formation JoMa1 cells transduced with indicated constructs, GFP, JMJD6, MYCN, JMJD6+MYCN. Top panel showing photos taken under light microscope. Bottom panel showing cell colonies stained with crystal violet. P value calculated by multiple unpaired t-test across each row. ***p<0.001, **p<0.01.

i. Xenograft tumor growth of JoMa1 cells transduced with indicated constructs, GFP, JMJD6, MYCN, JMJD6+MYCN. n=5 per group. P value calculated by multiple unpaired t-test across each row. ***p<0.001, **p<0.01.

JMJD6 regulates pre-mRNA splicing of genes involved in metabolism

a. Pathway enrichment for JMJD6 co-dependency genes whose knockout exhibits similar phenotype with JMJD6 knockout based on re-analysis of DepMAP data.

b. Pathway enrichment for genes whose knockout exhibits opposite phenotype with JMJD6 knockout based on re-analysis of DepMAP data.

c. Chromosomal location enrichment for JMJD6 co-dependency genes whose knockout exhibits similar phenotype with JMJD6 knockout based on re-analysis of DepMAP data.

d. Chromosomal location enrichment for genes whose knockout exhibits opposite phenotype with JMJD6 knockout based on re-analysis of DepMAP data.

e. Pathways analysis for genes downregulated and upregulated by JMJD6 knockdown commonly shared in SK-NAS and BE2C cells.

f. Alternative splicing events altered by JMJD6 knockdown in BE2C and SK-N-AS cells.

g. Pathway enrichment for each splicing event commonly shared in BE2C and SK-N-AS cells after JMJD6 knockdown.

h. Isoform identification based on splicing events in BE2C and SK-N-AS cells, followed by pathway enrichment for commonly shared alterations in both cell lines.

JMJD6 regulates alternative splicing of glutaminolysis gene, GLS

a. Sashimi plot showing the alternative splicing of GLS after JMJD6 knockdown in BE2C cells in duplicates. The number indicates the RNA-seq read counts of exon junction.

b. Real time PCR assessing the relative expression of GAC and KGA isoforms after JMJD6 knockdown in BE2C cells in triplicates.

c. Western blot showing the expression of GAC and KGA isoforms in SK-N-AS, BE2C, SIMA after JMJD6 knockdown for 72 hours.

d. KGA and GAC specific reporter analysis showing only KGA-driven luciferase activity is significantly upregulated by JMJD6 knockdown.

e. RNA-immunoprecipitation showing JMJD6 interaction with GLS RNA. Top panel shows the western blot analysis of Flag tagged JMJD6 in input, immunoprecipitation (IP) and flowthrough (FT)_fractions. Bottom panel shows RT PCR analysis of enrichment of GAC/KGA bound by JMJD6 in IP and FT fractions.

f. Spearman correlation analysis of JMJD6 and GAC/KGA expression levels in two neuroblastoma cohorts GSE45547(left) and GSE120572 (right).

g. Kaplan-Meier curve showing the association of high or low GAC expression levels with event-free survival in a cohort of neuroblastoma (GSE45547).

h. Kaplan-Meier curve showing the association of high or low KGA expression levels with event-free survival in a cohort of neuroblastoma (GSE45547).

GAC and KGA are both important for cell survival

a. Western blotting analysis of expression of KGA and GAC in BE2C cells with indicated antibodies.

b. Colony formation assay for BE2C cells overexpressing KGA and GAC for 7 days (left =crystal violet staining, right = quantification of cell density).). N=3 per group. ***p<0.001.

c. Western blotting analysis of expression of KGA and GAC in SK-N-AS cells with indicated antibodies.

d. Colony formation assay for SK-N-AS cells overexpressing KGA and GAC for 7 days (left =crystal violet staining, right = quantification of cell density). N=3 per group. **p<0.01. ***p<0.001.

e. Bubble blot showing the pathways significantly upregulated and downregulated by both KGA and GAC in BE2C cells.

f. Bubble blot showing the pathways significantly upregulated and downregulated by both KGA and GAC in SKNAS cells.

g. Whole cell lysates (on 72h) subject to Western blot showing the knockdown of GLS (both GAC and KGA), GAC alone and KGA alone in BE2C cells.

h. Colony forming assay (on day 7) od BE2C cells with knockdown of GLS (both GAC and KGA), GAC alone and KGA alone. two independent experiments.

i. Quantification of colonies in h using ImageJ software.

JMJD6 forms an interaction network consists of proteins involved in splicing and protein synthesis

a. Flag tagged JMJD6 transduced into SK-N-AS cells for immunoprecipitation with anti-Flag followed by protein identification by mass spectrometry. The interacting protein partners of JMJD6 are analyzed by STRING protein network.

b. Immunoprecipitation followed by western blot to validate the JMJD6 interacting partners in SK-N-AS cells. IP= immunoprecipitation, FT= flowthrough.

c. Click-IT AHA labeling showing the newly synthesized proteins after overexpression of JMJD6 in SK-N-AS cells

d,e. Western blot showing the expression of GAC and KGA isoforms in BE2C (d) SK-N-AS (e), after U2AF2 and CPSF6 knockdown for 72hours.

JMJD6 regulates production of citric acid cycle intermediates and NTP

a. Heatmap showing the metabolites differentially expressed in SK-N-AS cells (n=5) after JMJD6 knockout (n=5) based on LC-MS/MS analysis.

b. Pathway analysis of metabolites downregulated by JMJD6 knockout.

c. Pathway cartoon showing the connections of TCA, glycolysis, glutaminolysis, and β-oxidation.

d. Correlation of metabolite abundance with JMJD6 dependency. The positive correlation indicates that the higher the abundance of metabolites, the more resistance of cells to JMJD6 knockout. On the contrary, the negative correlation indicates the higher the abundance of metabolites, the more sensitive of cells to JMJD6 knockout.

JMJD6-GAC pathway regulates the response of neuroblastoma cells to indisulam treatment

a. Spearman correlation of effects of JMJD6 knockout and RBM39 knockout demonstrating the co-dependency of JMJD6 and RBM39 from DepMAP CRISPR screening data (n=1086). Each dot represents one cell line.

b. GSEA analysis for indisulam sensitive vs resistant neuroblastoma cell lines based on CTD2 (Cancer Target Discovery and Development) data showing histone lysine demethylase gene signature is the one that is significantly associated with indisulam response.

c. Heatmap from GSEA (b) showing the individual genes in indisulam sensitive and resistant cells.

d. JMJD6 expression in indisulam sensitive and resistant neuroblastoma cells. p value calculated by student t test.

e. Western blot showing JMJD6 knockout in SK-N-AS cells using indicated antibodies.

f. Colony formation of SK-N-AS cells in triplicates with or without JMJD6 knockout treated with different concentrations of indisulam for 7 days, stained with crystal violet. N=3 per group.

g. Quantification of cell density by using ImageJ software from f. ns=not significant. ** p<0.001, ***p<0.0001

h. Western blot showing JMJD6 knockout in BE2C cells using indicated antibodies.

i. Colony formation of BE2C cells in triplicates with or without JMJD6 knockout treated with 100nM of indisulam for 5 days, stained with crystal violet. N=3 per group.

j. Quantification of cell density by using ImageJ software from f. ns=not significant. ** p<0.001, ***p<0.0001

k. Colony formation of BE2C cells in triplicates with KGA and GAC overexpression treated with 250nM of indisulam for 5 days, stained with crystal violet. N=3 per group.

l. Colony formation of SK-N-AS cells in triplicates with KGA and GAC overexpression treated with 100nM of indisulam for 7 days, stained with crystal violet.

m. Quantification of cell density by using ImageJ software from k. * p<0.01, **p<0.001

n. Quantification of cell density by using ImageJ software from l. **p<0.001

Working mechanism of JMJD6 in MYC-driven neuroblastoma.

Overactive MYC drives high-load of gene transcription, enhanced protein synthesis and high rate of metabolism, leading to detrimental cellular stresses and consequent cell death (Model a). However, when 17q is amplified, high levels of JMJD6 and other proteins encoded by 17q genes physically interacts with the splicing and translational machineries, enhancing pre-mRNA splicing of metabolic genes such as GLS and inhibiting global protein synthesis, respectively, leading to reduced detrimental stresses and enhanced cancer cell survival and tumorigenesis (Model b). The high levels of JMJD6 predicts high dependency of RBM39, which are more sensitive to indisulam treatment.