The Histone Methyltransferase DOT1L Promotes Neuroblastoma by Regulating Gene Transcription

Author Manuscript Published OnlineFirst on February 16, 2017; DOI: 10.1158/0008-5472.CAN-16-1663 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

The histone methyltransferase DOT1L promotes neuroblastoma by regulating gene transcription

Matthew K. K. Wong1, Andrew E. Tee1, Giorgio Milazzo2, Jessica L. Bell3, Rebecca C. Poulos4, Bernard Atmadibrata1, Yuting Sun1, Duohui Jing1, Nicholas Ho1, Dora Ling1, Pei Y. Liu1, Xu D. Zhang5, Stefan Hüttelmaier3, Jason W. H. Wong4, Jenny Wang1,6, Patsie Polly7, Giovanni Perini2, Christopher J. Scarlett8, and Tao Liu1,6*

1Children’s Cancer Institute Australia for Medical Research, University of New South Wales, Sydney, Australia. 2Department of Pharmacy and Biotechnology, University of Bologna, 40126 Bologna, Italy. 3Institute of Molecular Medicine, Martin Luther University, ZAMED, 06120 Halle, Germany. 4Prince of Wales Clinical School and Lowy Cancer Research Centre, UNSW Australia, Sydney, NSW 2052, Australia. 5School of Biomedical Sciences and Pharmacy, University of Newcastle, Newcastle, Australia. 6Centre for Childhood Cancer Research, UNSW Medicine, UNSW Australia, Sydney, Kensington, NSW 2052, Australia. 7Department of Pathology and Mechanisms of Disease and Translational Research, University of New South Wales, Kensington NSW 2052, Australia. 8School of Environmental & Life Sciences, University of Newcastle, Ourimbah, NSW, Australia.

Corresponding Author: Tao Liu, Children’s Cancer Institute Australia, Lowy Cancer Research Centre, UNSW Australia, Kensington, Sydney, NSW 2052, Australia. Phone: 61 2 9385 1935; Fax: 61 2 9662 6584; E-mail: [email protected]

Running title: DOT1L promotes gene transcription and neuroblastoma

Key words: DOT1L, N-Myc, neuroblastoma, H3K79 methylation

Grant Support: The authors were supported by National Health & Medical Research Council Australia. G. Perini is supported by a Grant of the Italian Association for Cancer Research (AIRC-IG15182) and T. Liu is the recipient of an Australian Research Council Future Fellowship.

Disclosure of Potential Conflicts of Interest The authors declare no competing financial interests.

Word Count: 4993 Total numbers of figures and tables: 7

Downloaded from cancerres.aacrjournals.org on September 29, 2021. © 2017 American Association for Cancer Research. Author Manuscript Published OnlineFirst on February 16, 2017; DOI: 10.1158/0008-5472.CAN-16-1663 Author manuscripts have been peer reviewed and accepted for publication but have not yet been edited.

Abstract

Myc oncoproteins exert tumorigenic effects by regulating expression of target oncogenes.

Histone H3 lysine 79 (H3K79) methylation at Myc-responsive elements of target gene

promoters is a strict prerequisite for Myc-induced transcriptional activation, and DOT1L is

the only known histone methyltransferase that catalyses H3K79 methylation. Here we show

that N-Myc upregulatsd DOT1L mRNA and protein expression by binding to the DOT1L

gene promoter. shRNA-mediated depletion of DOT1L reduced mRNA and protein

expression of N-Myc target genes ODC1 and E2F2. DOT1L bound to the Myc Box II domain

of N-Myc protein, and knockdown of DOT1L reduced histone H3K79 methylation and N-

Myc protein binding at the ODC1 and E2F2 gene promoters and reduced neuroblastoma cell

proliferation. Treatment with the small molecule DOT1L inhibitor SGC0946 reduced H3K79

methylation and proliferation of MYCN gene-amplified neuroblastoma cells. In mice

xenografts of neuroblastoma cells stably expressing doxycycline-inducible DOT1L shRNA,

ablating DOT1L expression with doxycycline significantly reduced ODC1 and E2F2

expression, reduced tumor progression, and improved overall survival. Additionally, high

levels of DOT1L gene expression in human neuroblastoma tissues correlated with high levels

of MYCN, ODC1, and E2F2 gene expression and independently correlated with poor patient

survival. Taken together, our results identify DOT1L as a novel co-factor in N-Myc-mediated

transcriptional activation of target genes and neuroblastoma oncogenesis. Furthermore, they

characterize DOT1L inhibitors as novel anticancer agents against MYCN-amplified

neuroblastoma.

Introduction

Neuroblastoma is the most commonly diagnosed paediatric solid tumor in early

childhood (1). Neuroblastoma arises from neural crest cells and is characterised by variable

clinical behaviour, from spontaneous regression to inexorable progression (2). Adverse

clinical prognostic factors include age > 18 months at diagnosis, advanced disease stage and

amplification of the MYCN oncogene which encodes the N-Myc oncoprotein (3-5).

N-Myc is expressed during embryogenesis in the nervous system (6, 7). N-Myc

induces neuroblastoma by regulating the expression of genes important for cell

differentiation, malignant transformation and cell proliferation (8). Like its analogue c-Myc

oncoprotein, N-Myc activates target gene transcription by forming a heterodimer with MAX,

and this heterodimer in turn recruits a range of co-factors that alter chromatin structure, such

as histone H3 lysine 79 (H3K79) di-methylation (9-11).

DOT1L is a unique histone methyltransferase as it is the only known histone

methyltransferase that catalyses mono-methylation (me), di-methylation (me2) and tri-

methylation (me3) at the H3K79 position (12, 13). Quantitative chromatin

immunoprecipitation studies of H3K79 methylation across the human genome reveal that

H3K79me and H3K79me2 are linked to active gene transcription (14-16). DOT1L is

involved in the oncogenesis of several leukaemia subtypes characterised by chromosomal

translocations involving the mixed lineage leukaemia (MLL) oncogene. DOT1L forms a

protein complex with the MLL fusion proteins, and DOT1L-mediated H3K79 methylation is

responsible for maintaining an open chromatin state around MLL fusion protein target

oncogenes (17, 18).

Crucially, H3K79me2 is essential for c-Myc binding to its target gene promoters (10,

11). Here we investigated the role of DOT1L-mediated H3K79 methylation in N-Myc-

induced target gene transcription, neuroblastoma cell proliferation in vitro and tumor

progression in vivo, and examined whether DOT1L expression in human neuroblastoma

tissues independently predicted patient prognosis.

Materials and Methods

Cell culture

Neuroblastoma BE(2)-C, NBL-S, SK-N-FI and HEK-293 cells were cultured in Dulbecco's

modified Eagle's medium supplemented with 10% fetal calf serum. Kelly cells were grown in

RPMI 1640 supplemented with 10% fetal calf serum. BE(2)-C and NBL-S cells were

provided by Barbara Spengler (Fordham University, New York, NY) and Dr Susan Cohn

(Northwestern University, Chicago, IL) respectively 20 years ago. HEK-293 cells were

obtained from the American Type Culture Collection 20 years ago. Kelly and SK-N-FI cells

were obtained from the European Collection of Cell Cultures and Sigma Aldrich in 2010. The

identity of cell lines was verified in 2010, 2014 and 2015 by small tandem repeat profiling

conducted at the Garvan Institute of Medical Research or Cellbank Australia.

Protein co-immunoprecipitation assays

HEK293 human embryonic cells were transiently co-transfected with an N-Myc-

expression pcDNA3.1 construct, together with an empty vector or Flag-DOT1L-expression

pcDNA3.1 construct for 48 hours. Protein was extracted from the cells and co-

immunoprecipitation was carried out using an anti-FLAG antibody (Abcam) or a mouse IgG

as a negative control, followed by immunoblot analysis.

GST pull-down assays

Different N-Myc protein fragments were cloned into the pGEX-2T construct, in frame

with N-terminal GST. The constructs were transformed into BL-21 E.Coli and IPTG was

used for induction of T7-driven transcription. After purification, equal amount of GST-N-

Myc protein fragments were immobilized onto glutathione-agarose beads (Sigma). HEK-

293T cells were transfected with Flag-DOT1L expression construct, and nuclear protein from

the cells was incubated with equal amount of different GST-N-Myc protein fragments

immobilized onto glutathione agarose beads. Pulled-down complexes were analysed by

immunolot with a monoclonal anti-Flag antibody (Santa Cruz Biotech), and ponceau staining

detected by ChemiDoc MP (BIORAD) was used as loading controls.

Chromatin immunoprecipitation (ChIP) assays

ChIP assays were performed with mouse anti-N-Myc antibody (Merck Millipore,

Billerica, MA), rabbit anti-H3K79me2 antibody (Abcam), rabbit and mouse control IgG

(Santa Cruz Biotech), followed by PCR with primers designed to cover the core promoter

regions of the DOT1L, ODC1 and E2F2 genes containing Myc responsive element E-Boxes

or remote negative control regions. Fold enrichment of the DOT1L, E2F2 and ODC1 gene

core promoter regions by the anti-N-Myc or anti-H3K79me2 antibody was calculated by

dividing cycle threshold of PCR products from the DOT1L, E2F2 and ODC1 gene core

promoter regions by cycle threshold of PCR products from the negative control region,

relative to input.

Luciferase reporter assays

Modulation of DOT1L promoter activity by N-Myc was analysed by luciferase assays.

The DOT1L gene core promoter containing the E-Box (-389bp to +50bp relative to

transcription start site) was cloned into the pLightSwitch_Prom construct (SwitchGear

Genomics, Menlo Park, CA). BE(2)-C cells were co-transfected with control siRNA or N-

Myc siRNA-1, together with Go Clone Promoter Control construct plus the

pLightSwitch_Prom construct expressing empty vector or the DOT1L core promoter.

Luciferase activities were measured with a LightSwitch Dual Assay System Kit (SwitchGear

Genomics), and normalized according to Go Clone Promoter Control construct according to

the manufacturer`s instructions.

Establishment of DOX-inducible control shRNA and DOT1L shRNA expression

constructs and neuroblastoma cell lines stably expressing the constructs

The lentiviral DOX-inducible GFP-IRES-shRNA FH1tUTG construct from Dr Marco

Herold (Walter and Eliza Hall Institute of Medical Research, Melbourne, Australia) (21) was

used to generate control shRNA or DOT1L shRNA expressing construct and neuroblastoma

cell lines. DOT1L shRNA and scrambled control shRNA sequences were selected from

previously published work (22) and cloned into the FH1tUTG construct. The DOX-inducible

GFP-IRES-control shRNA or DOT1L shRNA FH1tUTG construct was transfected into 293T

cells. The viral media were collected and used to infect BE(2)-C and Kelly neuroblastoma

cells over 72 hours with polybrene (Santa Cruz). GFP-based cell sorting was performed for

selecting cells with high GFP protein expression.

In vivo mouse experiments

Animal experiments were approved by the Animal Care and Ethics Committee of

UNSW Australia, and the animals’ care was in accord with institutional guidelines. Female

Balb/c nude mice aged 5 to 6 weeks were injected subcutaneously while under anesthesia

with either 2×106 DOX-inducible DOT1L shRNA BE(2)-C cells or 8×106 DOX-inducible

DOT1L shRNA Kelly cells into the right flank. Mice were fed with 10% sucrose water with

or with DOX at 2mg/ml when tumors reached 0.005cm3 in volume. Tumor development was

monitored and tumor volume calculated using (length × width × height)/2. Mice were culled

when tumor volume reached 1cm3, and tumor tissues were snap-frozen and analyzed by

immunoblotting for DOT1L, ODC1 and E2F2 protein expression.

Patient tumor sample analysis

DOT1L, N-Myc, ODC1 and E2F2 mRNA expression was examined in 88 (Versteeg

dataset) (23) and 476 (Kocak dataset) (24, 25) human neuroblastoma samples in the

publically available gene expression datasets (http://r2.amc.nl). Correlation between DOT1L

and N-Myc, ODC1 as well as E2F2 expression in human neuroblastoma tissues was analysed

with Pearson’s correlation. Probability of survival was investigated according to the method

of Kaplan and Meier and two-sided log-rank tests (26). Multivariable Cox regression

analyses were performed after inclusion of disease stage, age at diagnosis, MYCN

amplification status, N-Myc and DOT1L expression levels. Probabilities of survival and

hazard ratios (HRs) were provided with 95% confidence intervals. Proportionality was

confirmed by visual inspection of the plots of log(2log(S(time))) versus log(time), which

were observed to remain parallel (27).

Statistical analysis

For statistical analysis, experiments were conducted three times. Data were analysed

with Prism 6 software (GraphPad) and presented as mean ± standard error. Differences were

analyzed for significance using ANOVA among groups or two-sided t-test for two groups.

All statistical tests were two-sided. A p value of less than 0.05 was considered statistically

significant.

Results

N-Myc up-regulates DOT1L expression by binding to the DOT1L gene promoter

Myc proteins induce gene transcription by binding to canonical and non-canonical E-

boxes at target gene promoters (9). Our bioinformatics analysis identified a non-canonical E-

box (CACGCG) located -288bp upstream of the DOT1L gene transcription start site. We

therefore examined whether N-Myc modulated DOT1L gene expression in neuroblastoma

cells. As shown in Fig. 1A and B, transfection of N-Myc overexpressing BE(2)-C and Kelly

neuroblastoma cells with N-Myc siRNA-1 or N-Myc siRNA-2 efficiently knocked down N-

Myc mRNA and protein expression, and reduced DOT1L mRNA and protein expression.

Consistently, withdrawal of tetracycline from cell culture media of SHEP-Tet/21N cells,

which are stably transfected with a tetracycline withdrawal-inducible N-Myc expression

construct, increased N-Myc and DOT1L mRNA and protein expression (Fig. 1C).

We next performed ChIP assays with an anti-N-Myc antibody or control IgG and real-

time PCR with primers targeting a negative control region, the E-box region located -288bp

upstream of the DOT1L transcription start site (Amplicon B), 5’ un-transcribed region

located approximately -600 to -800bp upstream (Amplicon A), or intron 1 regions located

approximately 100 to 180bp (Amplicon C) or 350 to 500bp (Amplicon D) downstream (Fig.

1D). The ChIP assays showed that the N-Myc antibody immunoprecipitated the E-box region

(Amplicon B) seven-fold higher than the negative control region (Fig. 1E). Amplicons A, C

and D located to either side of amplicon B displayed much lower enrichment by the N-Myc

antibody (Fig. 1E). Importantly, luciferase assays showed that transfection with the

pLightSwitch_Prom construct encoding the DOT1L gene core promoter containing the E-

Box led to high luciferase activity, compared with the empty vector pLightSwitch_Prom

construct, and that knocking down N-Myc largely blocked the luciferase activity (Fig. 1F).

Taken together, the data suggest that N-Myc up-regulates DOT1L gene expression via

binding to the DOT1L gene promoter.

DOT1L-mediated H3K79 methylation is required for N-Myc protein binding to target

gene promoters

DOT1L forms protein complexes with MLL fusion proteins, and DOT1L-mediated

histone H3K79 methylation is essential for MLL fusion protein-mediated leukaemogenic

gene transcription and leukaemogenesis (28, 29). H3K79me2 has been shown to be essential

for c-Myc protein binding at c-Myc target gene promoters (10, 11). We therefore examined

whether DOT1L formed a protein complex with N-Myc, and whether DOT1L is required for

histone H3K79me2 and N-Myc protein binding at N-Myc target gene promoters.

HEK293 cells were co-transfected with an N-Myc-expression construct, together with

a construct encoding empty vector or Flag-tagged DOT1L. Immunoprecipitation with an anti-

Flag antibody pulled-down N-Myc protein in cells co-transfected with the DOT1L and the N-

Myc expression constructs, but did not pull-down N-Myc protein in cells transfected with the

N-Myc expression construct alone (Fig. 2A). Furthermore, different N-Myc protein

fragments (1-86, 82-254, 249-358, 336-400 and 400-464 amino acids) were cloned into the

pGEX-2T construct, in frame with N-terminal GST. GST pull-down assays showed that

DOT1L protein bound to the N-Myc 82-254 amino acid fragment (Fig. 2B). The data

demonstrate that DOT1L protein directly binds to the Myc Box II region of N-Myc protein.

We next examined whether knocking-down DOT1L expression reduced H3K79me2

and N-Myc protein binding at the gene promoters of ODC1 and E2F2, two known Myc target

genes (30-32). BE(2)-C cells were transfected with control siRNA, DOT1L siRNA-1 or

DOT1L siRNA-2 for 72 hours. ChIP assays were performed with an anti-H3K79me2

antibody, anti-N-Myc antibody or control IgG, followed by PCR with primers targeting a

negative control region or E-Box regions of the ODC1 or E2F2 gene promoters. The assays

showed that the H3K79me2 antibody and the anti-N-Myc antibody immunoprecipitated the

E-Box regions of both the ODC1 and E2F2 gene promoters in cells transfected with

scrambled control siRNA. Compared with the scrambled control siRNA, both DOT1L

siRNAs reduced the presence of H3K79me2 and N-Myc protein binding at the E-Box regions

(Fig. 2C and D). These findings together suggest that DOT1L and N-Myc form a protein

complex at the N-Myc target gene promoters, and that DOT1L-mediated H3K79 methylation

is required for N-Myc binding to Myc binding motifs present at N-Myc target gene

promoters.

DOT1L activates the expression of the N-Myc target genes ODC1 and E2F2

We next examined whether DOT1L regulated the expression of the Myc target genes

ODC1 and E2F2 in human neuroblastoma cells. BE(2)-C and Kelly cells were transfected

with control siRNA, N-Myc siRNA-1, N-Myc siRNA-2, DOT1L siRNA-1 or DOT1L

siRNA-2. Consistent with literature (30-32), RT-PCR and immunoblot analyses showed that

transfection with N-Myc siRNAs reduced ODC1 and E2F2 mRNA and protein expression

(Fig. S1). RT-PCR and immunoblot analyses showed that transfection with DOT1L siRNAs

efficiently knocked down DOT1L mRNA and protein expression, and significantly reduced

ODC1 and E2F2 mRNA and protein expression (Fig. 3A and B).

To further demonstrate that DOT1L regulates Myc target gene expression, we

established BE(2)-C and Kelly cells stably expressing DOX-inducible control shRNA or

DOT1L shRNA using the GFP-IRES-shRNA expression FH1tUTG construct (21). The cells

were then treated with vehicle control or DOX or 72 hours, followed by histone protein

extraction. Immunoblot analysis confirmed that DOX treatment considerably reduced histone

H3K79 di-methylation in DOX-inducible DOT1L shRNA, but not control shRNA, BE(2)-C

and Kelly cells (Fig. 3C and D). We then treated the DOX-inducible control shRNA or

DOT1L shRNA BE(2)-C and Kelly cells with vehicle control or DOX for 96 hours, followed

by RNA and protein extraction. RT-PCR and immunoblot analyses showed that treatment

with DOX efficiently knocked down DOT1L mRNA and protein expression, and also

effectively reduced ODC1 and E2F2 mRNA and protein expression in both BE(2)-C and

Kelly cells (Fig. 3E and F). In comparison, knocking down N-Myc or DOT1L did not have

an effect on the expression of the non-Myc target genes HIF1α and VEGF (Fig. S2A and B).

We next performed differential gene expression studies with Affymetrix microarray in DOX-

inducible control shRNA and DOT1L shRNA BE(2)-C cells after treatment with vehicle

control or DOX. Gene set enrichment analysis (GSEA) showed that genes with E-Boxes at

promoters were possibly enriched among those down-regulated by DOT1L shRNA

(Supplementary Table S1).

Taken together, the data suggest that DOT1L induces histone H3K79 di-methylation

and activates the transcription of the N-Myc target genes ODC1 and E2F2.

DOT1L is required for MYCN-amplified neuroblastoma cell proliferation

Up-regulation of ODC1 has been well-documented to contribute to N-Myc-mediated

neuroblastoma cell proliferation (31). Next we examined the effect of DOT1L and its target

E2F2 upon neuroblastoma cell proliferation. BE(2)-C and Kelly neuroblastoma cells were

transfected with control siRNA, DOT1L siRNA-1, DOT1L siRNA-2, E2F2 siRNA-1 or E2F2

siRNA-2 for 96 hours. Alamar blue assays showed that transfection with DOT1L siRNAs or

E2F2 siRNAs reduced the numbers of BE(2)-C and Kelly cells (Fig. 4A and B).

To further demonstrate that DOT1L is required for neuroblastoma cell proliferation,

we treated DOX-inducible control shRNA or DOT1L shRNA BE(2)-C and Kelly cells with

vehicle control or DOX. Alamar blue assays showed that DOX treatment had minimal effect

on cell proliferation in DOX-inducible control shRNA BE(2)-C and Kelly cells, but

significantly reduced cell proliferation in DOX-inducible DOT1L shRNA BE(2)-C and Kelly

cells (Fig. 4C). In addition, cell cycle analysis showed that knocking down DOT1L

expression with DOX did not increase the percentage of cells at the sub-G1 phase in DOX-

inducible DOT1L shRNA BE(2)-C and Kelly cells (Supplementary Fig. S3A and B).

Small molecule DOT1L inhibitors such as SGC0946 are promising novel anticancer

agents (33-35). We next treated MYCN gene-amplified BE(2)-C and Kelly cells with vehicle

control or 1.25µM SGC0946 for 72 hours, 96 hours or 7 days. Immunoblot analyses

confirmed that treatment with SGC0946 led to reduction in H3K79me2 72 hours post-

treatment and more significantly 7 days post-treatment (Fig. 4D). RT-PCR analysis showed

that SGC0946 reduced the expression of the DOT1L and N-Myc target gene E2F2 (Fig. 4E),

and Alamar blue assays confirmed that SGC0946 dose-dependently reduced BE(2)-C and

Kelly cell proliferation (Fig. 4F). By contrast, treatment with SGC0946 did not reduce E2F2

gene expression and did not have an effect on cell proliferation in MYCN gene non-amplified

NBL-S and SK-N-FI neuroblastoma cells (Fig. 4G and H). Taken together, the data suggest

that DOT1L is required for MYCN-amplified neuroblastoma cell proliferation, and that

DOT1L inhibitors are effective anticancer agents against MYCN-amplified neuroblastoma.

DOT1L is required for neuroblastoma tumor growth in vivo

We next examined whether DOT1L is required for neuroblastoma tumor growth in

vivo. DOX-inducible DOT1L shRNA BE(2)-C and Kelly cells were injected into the flanks

of Balb/c nude mice. Once tumors were palpable, the mice were divided into DOX or vehicle

control treatment sub-groups, and fed with water with or without DOX. Tumor growth was

monitored and mice culled once tumor volume reached 1cm3.

The DOX treatment sub-group displayed slower tumor growth in comparison to the

control treatment sub-group for both DOX-inducible DOT1L shRNA BE(2)-C and Kelly cell

xenografts (Fig. 5A). Kaplan-Meier survival analysis showed that DOX treatment, compared

with control treatment, resulted in a longer median survival in mice xenografted with DOX-

inducible DOT1L shRNA BE(2)-C or Kelly cells by two fold (Fig. 5B).

At the conclusion of the in vivo experiment, mice were culled, tumor tissues snap-

frozen and protein extracted. Immunoblot analysis showed that both the DOX-inducible

DOT1L shRNA BE(2)-C and Kelly tumors displayed a decrease in DOT1L protein

expression as a result of DOX treatment compared to the control treatment (Fig. 5C and D for

immunoblot gels and Fig. S4A and B for protein quantification). Protein expression of the N-

Myc target genes, ODC1 and E2F2, in both DOX-inducible DOT1L shRNA BE(2)-C and

Kelly xenograft tumors, was also reduced in the DOX treatment group, compared to the

control treatment group (Fig. 5C and D for immunoblot gels and Fig. S4A and B for protein

quantification). Taken together, knocking down DOT1L gene expression with DOX reduces

ODC1 and E2F2 protein expression, reduces neuroblastoma tumor progression and improves

mouse survival in vivo.

High levels of DOT1L gene expression in human neuroblastoma tissues independently

predict poor patient prognosis

To assess clinical relevance of DOT1L expression in human neuroblastoma tissues,

we examined DOT1L gene expression in human neuroblastoma tissues in the publically

available Versteeg (23) and Kocak (24, 25) microarray gene expression-patient prognosis

datasets downloaded from the R2 platform (http://r2.amc.nl) (last accessed on August 26,

2014). Two-sided Pearson’s correlation study showed that DOT1L mRNA expression

positively correlated to N-Myc mRNA expression in the 88 human neuroblastoma tissues of

the Versteeg dataset and in the 476 human neuroblastoma tissues of the Kocak dataset (Fig.

6A). In addition, DOT1L mRNA expression positively correlated to ODC1 and E2F2 mRNA

expression in human neuroblastoma tissues (Fig. S5).

Using the median and the upper quartile DOT1L mRNA expression as the cut-off

points, Kaplan-Meier survival analysis showed that high DOT1L mRNA expression levels in

neuroblastoma tissues were associated with poor patient prognosis in both the Versteeg and

the Kocak datasets (Fig. 6B and C). Additionally, high levels of DOT1L mRNA expression in

the 72 MYCN-amplified neuroblastoma tissues were positively associated with poor patient

overall survival in the large Kocak dataset (Fig. 6D). Importantly, multivariable Cox

regression analysis showed that high levels of DOT1L expression in neuroblastoma tissues

strongly associated with reduced patient overall survival and event-free survival, independent

of disease stage, age at diagnosis, N-Myc mRNA expression level and MYCN amplification

status, the current most important prognostic markers for neuroblastoma patients (1), using

the median or upper quartile of DOT1L mRNA expression as the cut-off points (Table 1) or

using the exact value of N-Myc and DOT1L mRNA expression levels (Table S2). Taken

together, the data suggest that high levels of DOT1L expression in human neuroblastoma

tissues independently predict poor patient prognosis.

Discussion

Like its analogue c-Myc, N-Myc exerts tumorigenic effects in part by binding to Myc-

responsive element E-boxes at target gene promoters and consequently activating oncogenic

gene expression (9). In this study, we have identified a non-canonical E-Box upstream of the

DOT1L gene transcription start site, and demonstrated that N-Myc directly binds to the

DOT1L gene core promoter region containing the E-Box and upregulates DOT1L promoter

activity, DOT1L mRNA and protein expression in neuroblastoma cells. The data suggest that

N-Myc up-regulates DOT1L gene expression by binding to its gene promoter.

By analyzing 35 histone marks, Guccione et al have shown that histone H3K4 tri-

methylation and H3K79 di-methylation at Myc-responsive elements of target gene promoters

are strict prerequisites for Myc-induced transcriptional activation (10). Thomas et al and we

have recently shown that the histone H3K4 tri-methylation presenter WDR5 binds to N-Myc

and c-Myc proteins, and that WDR5-mediated histone H3K4 tri-methylation plays an

essential role in Myc-mediated target gene transcription (36, 37). However, the mechanism

through which histone H3K79 is di-methylated during Myc-induced transcriptional

activation, is unknown.

DOT1L is the only histone methyltransferase that catalyses methylation at the H3K79

position (12, 13). DOT1L is well-known to form protein complexes with MLL fusion

proteins and plays a critical role in MLL-mediated transcriptional activation and

leukaemogenesis (28, 29). Most recently, DOT1L has been found to complex with c-Myc,

leading to transcriptional activation of the epithelial–mesenchymal transition regulator genes

(38). In this study, we have found that knocking down DOT1L gene expression with siRNAs

or DOX-inducible shRNA reduces the expression of well-known Myc target genes, including

ODC1 and E2F2 (30-32). Knocking down DOT1L gene expression reduces H3K79 di-

methylation at ODC1 and E2F2 gene promoter regions, and reduces N-Myc protein binding

at the ODC1 and E2F2 gene promoter regions containing Myc-responsive element E-Boxes.

Protein co-immunoprecipitation and GST pull-known assays demonstrate that DOT1L

protein directly binds to N-Myc protein. Taken together, our data indicate that DOT1L and

N-Myc form a protein complex at N-Myc target gene promoters, resulting in H3K79 di-

methylation and transcriptional activation of N-Myc target genes including ODC1 and E2F2.

Up-regulation of ODC1 contributes to N-Myc-mediated neuroblastoma cell

proliferation and tumorigenesis (31). E2F2 is involved in DNA synthesis (39), and loss of

E2F2 results in reduced tumor incidence in a mouse model of c-Myc-induced breast cancer

(40). In this study, we have demonstrated that suppression of E2F2 or DOT1L gene

expression reduces neuroblastoma cell proliferation. The data suggest that DOT1L is required

for neuroblastoma cell proliferation through activating the transcription of N-Myc target

genes, such as ODC1 and E2F2.

In this study, we have found that DOT1L expression correlates with N-Myc, ODC1

and E2F2 expression in primary human neuroblastoma tissues, and that a high level of

DOT1L expression in tumor tissues correlates with poor neuroblastoma patient survival,

independent of disease stage, age at the time of diagnosis and MYCN amplification status, the

current most important prognostic markers for neuroblastoma patients (1). Importantly,

knocking down DOT1L gene expression in mice xenografted with neuroblastoma cells

reduces ODC1 and E2F2 expression in tumor tissues and reduces tumor growth. The data

suggest that DOT1L plays a critical role in N-Myc-mediated neuroblastoma by increasing the

expression of oncogenic genes, such as ODC1 and E2F2, and that DOT1L is a therapeutic

target for neuroblastoma.

Several small molecular DOT1L inhibitors, including EPZ004777, EPZ5676 and

SGC0946 have recently been reported (33-35). EPZ004777 displays specificity against

DOT1L compared with a panel of eight other histone methyltransferases, and reduces MLL-

rearranged leukaemogenesis (28, 34). Modifications to EPZ00477 lead to the synthesis of

EPZ5676 (33) and SGC0946 (35) with improved efficacy and in vivo availability. In this

study, we have found that treatment with SGC0946 reduces H3K79 methylation and DOT1L

target gene expression, and results in growth inhibition in MYCN-amplified, but not MYCN

non-amplified neuroblastoma cells. The data suggest that DOT1L inhibitors are potential

anticancer agents against MYCN-amplified neuroblastoma.

Acknowledgments

We thank Dr Marco Herold at Walter and Eliza Hall Institute of Medical Research,

Melbourne, Australia, for providing the FH1tUTG construct. Children's Cancer Institute

Australia is affiliated with UNSW Australia and Sydney Children’s Hospitals Network.

References

1. Brodeur GM. Neuroblastoma: biological insights into a clinical enigma. Nat Rev Cancer 2003;3:203-16. 2. Matthay KK, O'Leary MC, Ramsay NK, Villablanca J, Reynolds CP, Atkinson JB, et al. Role of myeloablative therapy in improved outcome for high risk neuroblastoma: review of recent Children's Cancer Group results. Eur J Cancer 1995;31a:572-5. 3. Janoueix-Lerosey I, Schleiermacher G, Michels E, Mosseri V, Ribeiro A, Lequin D, et al. Overall genomic pattern is a predictor of outcome in neuroblastoma. J Clin Oncol 2009;27:1026-33. 4. Maris JM. Recent advances in neuroblastoma. N Engl J Med 2010;362:2202-11. 5. Tomioka N, Oba S, Ohira M, Misra A, Fridlyand J, Ishii S, et al. Novel risk stratification of patients with neuroblastoma by genomic signature, which is independent of molecular signature. Oncogene 2008;27:441-9. 6. Mugrauer G, Alt FW, Ekblom P. N-myc proto-oncogene expression during organogenesis in the developing mouse as revealed by in situ hybridization. J Cell Biol 1988;107:1325-35. 7. Hirvonen H, Sandberg M, Kalimo H, Hukkanen V, Vuorio E, Salmi TT, et al. The N- myc proto-oncogene and IGF-II growth factor mRNAs are expressed by distinct cells in human fetal kidney and brain. J Cell Biol 1989;108:1093-104. 8. Maris JM, Matthay KK. Molecular biology of neuroblastoma. J Clin Oncol 1999;17:2264-79. 9. Eilers M, Eisenman RN. Myc's broad reach. Genes Dev 2008;22:2755-66. 10. Guccione E, Martinato F, Finocchiaro G, Luzi L, Tizzoni L, Dall' Olio V, et al. Myc- binding-site recognition in the human genome is determined by chromatin context. Nat Cell Biol 2006;8:764-70. 11. Martinato F, Cesaroni M, Amati B, Guccione E. Analysis of Myc-induced histone modifications on target chromatin. PLoS One 2008;3:e3650. 12. Feng Q, Wang H, Ng HH, Erdjument-Bromage H, Tempst P, Struhl K, et al. Methylation of H3-lysine 79 is mediated by a new family of HMTases without a SET domain. Curr Biol 2002;12:1052-8. 13. Min J, Feng Q, Li Z, Zhang Y, Xu RM. Structure of the catalytic domain of human DOT1L, a non-SET domain nucleosomal histone methyltransferase. Cell 2003;112:711-23. 14. Barski A, Cuddapah S, Cui K, Roh TY, Schones DE, Wang Z, et al. High-resolution profiling of histone methylations in the human genome. Cell 2007;129:823-37. 15. Morillon A, Karabetsou N, Nair A, Mellor J. Dynamic lysine methylation on histone H3 defines the regulatory phase of gene transcription. Mol Cell 2005;18:723-34. 16. Schubeler D, MacAlpine DM, Scalzo D, Wirbelauer C, Kooperberg C, van Leeuwen F, et al. The histone modification pattern of active genes revealed through genome-wide chromatin analysis of a higher eukaryote. Genes Dev 2004;18:1263-71. 17. Nguyen AT, Taranova O, He J, Zhang Y. DOT1L, the H3K79 methyltransferase, is required for MLL-AF9-mediated leukemogenesis. Blood 2011;117:6912-22. 18. Zeisig DT, Bittner CB, Zeisig BB, Garcia-Cuellar MP, Hess JL, Slany RK. The eleven-nineteen-leukemia protein ENL connects nuclear MLL fusion partners with chromatin. Oncogene 2005;24:5525-32. 19. Shahbazi J, Liu PY, Atmadibrata B, Bradner JE, Marshall GM, Lock RB, et al. The bromodomain inhibitor JQ1 and the histone deacetylase inhibitor panobinostat synergistically reduce N-Myc expression and induce anticancer effects. Clin Cancer Res 2016;22:2534-44.

20. Liu T LP, Tee AE, Haber M, Norris MD, Gleave ME, Marshall GM. Over-expression of clusterin is a resistance factor to the anti-cancer effect of histone deacetylase inhibitors. Eur J Cancer 2009;45:1846-54. 21. Herold MJ, van den Brandt J, Seibler J, Reichardt HM. Inducible and reversible gene silencing by stable integration of an shRNA-encoding lentivirus in transgenic rats. Proc Natl Acad Sci U S A 2008;105:18507-12. 22. Onder TT, Kara N, Cherry A, Sinha AU, Zhu N, Bernt KM, et al. Chromatin- modifying enzymes as modulators of reprogramming. Nature 2012;483:598-602. 23. Molenaar JJ, Koster J, Zwijnenburg DA, van Sluis P, Valentijn LJ, van der Ploeg I, et al. Sequencing of neuroblastoma identifies chromothripsis and defects in neuritogenesis genes. Nature 2012;483:589-93. 24. Kocak H, Ackermann S, Hero B, Kahlert Y, Oberthuer A, Juraeva D, et al. Hox-C9 activates the intrinsic pathway of apoptosis and is associated with spontaneous regression in neuroblastoma. Cell Death Dis 2013;4:e586. 25. Oberthuer A, Juraeva D, Li L, Kahlert Y, Westermann F, Eils R, et al. Comparison of performance of one-color and two-color gene-expression analyses in predicting clinical endpoints of neuroblastoma patients. Pharmacogenomics J 2010;10:258-66. 26. Kaplan EL, Meier P. Nonparametric Estimation from Incomplete Observations. In: Kotz S, Johnson N, editors. Breakthroughs in Statistics: Springer New York; 1992. p. 319-37. 27. Liu PY, Erriquez D, Marshall GM, Tee AE, Polly P, Wong M, et al. Effects of a novel long noncoding RNA, lncUSMycN, on N-Myc expression and neuroblastoma progression. J Natl Cancer Inst 2014;106:pii: dju113. 28. Chen CW, Koche RP, Sinha AU, Deshpande AJ, Zhu N, Eng R, et al. DOT1L inhibits SIRT1-mediated epigenetic silencing to maintain leukemic gene expression in MLL- rearranged leukemia. Nat Med 2015;21:335-43. 29. Nguyen AT, Zhang Y. The diverse functions of Dot1 and H3K79 methylation. Genes Dev 2011;25:1345-58. 30. Beier R, Burgin A, Kiermaier A, Fero M, Karsunky H, Saffrich R, et al. Induction of cyclin E-cdk2 kinase activity, E2F-dependent transcription and cell growth by Myc are genetically separable events. EMBO J 2000;19:5813-23. 31. Hogarty MD, Norris MD, Davis K, Liu X, Evageliou NF, Hayes CS, et al. ODC1 is a critical determinant of MYCN oncogenesis and a therapeutic target in neuroblastoma. Cancer Res 2008;68:9735-45. 32. Leone G, Sears R, Huang E, Rempel R, Nuckolls F, Park CH, et al. Myc requires distinct E2F activities to induce S phase and apoptosis. Mol Cell 2001;8:105-13. 33. Daigle SR, Olhava EJ, Therkelsen CA, Basavapathruni A, Jin L, Boriack-Sjodin PA, et al. Potent inhibition of DOT1L as treatment of MLL-fusion leukemia. Blood 2013;122:1017-25. 34. Daigle SR, Olhava EJ, Therkelsen CA, Majer CR, Sneeringer CJ, Song J, et al. Selective killing of mixed lineage leukemia cells by a potent small-molecule DOT1L inhibitor. Cancer Cell 2011;20:53-65. 35. Yu W, Chory EJ, Wernimont AK, Tempel W, Scopton A, Federation A, et al. Catalytic site remodelling of the DOT1L methyltransferase by selective inhibitors. Nat Commun 2012;3:1288. 36. Sun Y, Bell JL, Carter D, Gherardi S, Poulos RC, Milazzo G, et al. WDR5 Supports an N-Myc Transcriptional Complex That Drives a Protumorigenic Gene Expression Signature in Neuroblastoma. Cancer Res 2015;75:5143-54. 37. Thomas LR, Wang Q, Grieb BC, Phan J, Foshage AM, Sun Q, et al. Interaction with WDR5 promotes target gene recognition and tumorigenesis by MYC. Mol Cell 2015;58:440- 52.

38. Cho MH, Park JH, Choi HJ, Park MK, Won HY, Park YJ, et al. DOT1L cooperates with the c-Myc-p300 complex to epigenetically derepress CDH1 transcription factors in breast cancer progression. Nature Commun 2015;6:7821. 39. Nevins JR. The Rb/E2F pathway and cancer. Hum Mol Genet 2001;10:699-703. 40. Fujiwara K, Yuwanita I, Hollern DP, Andrechek ER. Prediction and genetic demonstration of a role for activator E2Fs in Myc-induced tumors. Cancer Res 2011;71:1924-32.

Figure legends

Figure 1. N-Myc up-regulates DOT1L expression by binding to the DOT1L gene promoter.

A-B, BE(2)-C (A) and Kelly (B) neuroblastoma cells were transfected with control siRNA,

N-Myc siRNA-1 or N-Myc siRNA-2 for 48 hours, followed by RT-PCR and immunoblot

analyses of N-Myc and DOT1L mRNA and protein expression. C, SHEP-Tet/21N cells were

treated with tetracycline (2ug/ml) or vehicle control for 72 hours. RT-PCR and immunoblot

analyses were conducted on N-Myc and DOT1L expression. D, schematic representation of

the DOT1L gene promoter. TSS indicated transcription start site. E, ChIP assays were

performed with a control IgG or N-Myc antibody (Ab), followed by PCR with primers

targeting a remote negative control region, the 5’ un-transcribed region (Amplicon A), the E-

Box region (Amplicon B) or the intron 1 region (Amplicons C and D) of the DOT1L gene in

BE(2)-C cells. Fold enrichment of the DOT1L gene promoter regions was calculated as the

difference in cycle thresholds obtained with the anti-N-Myc Ab compared with the control

IgG, relative to input. F, luciferase assays were performed in BE(2)-C cells after co-

transfection with control siRNA or N-Myc siRNA-1, together with Go Clone Promoter

Control construct plus pLightSwitch_Prom construct expressing empty vector or DOT1L

gene core promoter. Error bars represented standard errors. *, ** and *** indicated p < 0.05,

0.01 and 0.001 respectively.

Figure 2. DOT1L-mediated H3K79 methylation is required for N-Myc protein binding to

target gene promoters. A, HEK293 cells were co-transfected with N-Myc-expression

pcDNA3.1 construct, together with empty vector or Flag-DOT1L-expression pcDNA3.1

construct for 48 hours. Protein was extracted from the cells and co-immunoprecipitation (IP)

was carried out using an anti-FLAG antibody (Ab) or a mouse IgG as a negative control.

Immunoblot analysis was carried out using an anti-N-Myc Ab and anti-DOT1L Ab. B, HEK-

293T cells were transfected with Flag-DOT1L expression construct. Nuclear protein extract

from the cells was incubated with equal amount of different GST-N-Myc protein fragments

immobilized onto glutathione agarose beads, followed by immunoblot with an anti-Flag

antibody. As loading controls, ponceau stained images were detected by ChemiDoc MP.

Numbers on the left refers to molecular weights. C-D, BE(2)-C cells were transfected with

control siRNA, DOT1L siRNA-1 or DOT1L siRNA-2 for 72 hours. ChIP assays were

performed with a control IgG, anti-histone H3K79me2 (C) or anti-N-Myc Ab (D), followed

by PCR with primers targeting the E-box regions of the ODC1 and E2F2 gene promoters or

the DNA region -4000bp upstream of the E2F2 gene transcription start site as the negative

control. Error bars represent standard error. * and ** indicate p < 0.05 and 0.01 respectively.

Figure 3. DOT1L activates the expression of the N-Myc target genes ODC1 and E2F2. A-B,

BE(2)-C (A) and Kelly (B) neuroblastoma cells were transfected with control siRNA,

DOT1L siRNA-1 or DOT1L siRNA-2 for 96 hours, followed by RT-PCR and immunoblot

analyses of DOT1L, ODC1 and E2F2 mRNA and protein expression. C-D, DOX-inducible

control (Cont) shRNA or DOT1L shRNA BE(2)-C (C) and Kelly (D) cells were treated with

vehicle control or 2µg/ml DOX for 72 hours, followed by acid extraction of histone proteins

and immunoblot analyses with anti-histone H3 and anti-H3K79me2 antibodies. E-F, DOX-

inducible control shRNA or DOT1L shRNA BE(2)-C (E) and Kelly (F) cells were treated

with vehicle control or 2µg/ml DOX for 96 hours, followed by RT-PCR and immunoblot

analyses of DOT1L, ODC1 and E2F2 expression. Error bars represent standard error. **, ***

and **** indicate p < 0.01, 0.001 and 0.0001 respectively.

Figure 4. DOT1L is required for MYCN-amplified neuroblastoma cell proliferation. A-B,

BE(2)-C and Kelly cells were transfected with control siRNA, DOT1L siRNA-1, DOT1L

siRNA-2 (A), E2F2 siRNA-1 or E2F2 siRNA-2 (B) for 96 hours, followed by Alamar blue

assays. C, DOX-inducible control shRNA or DOT1L shRNA BE(2)-C and Kelly cells were

treated with vehicle control or 2µg/ml DOX for 96 hours, followed by Alamar blue assays. D.

BE(2)-C and Kelly cells were treated with control or 1.25µM SGC0946 for 72 hours, 96

hours or 7 days, followed by immunoblot analysis of H3K79me2 and total H3 proteins. E-H.

BE(2)-C and Kelly (E-F) and MYCN-non-amplified NBL-S and SK-N-FI (G-H) cells were

treated with control or SGC0946 for 7 days, followed by RT-PCR analysis of E2F2 gene

expression (E and G) or Alamar blue assays (F and H). Error bars represent standard error. *,

**, *** and **** indicate p < 0.05, 0.01, 0.001 and 0.0001 respectively.

Figure 5. DOT1L is required for neuroblastoma tumor growth in vivo. A-B, DOX-inducible

DOT1L shRNA BE(2)-C and Kelly cells were xenografted into nude mice. Once tumors

reached 0.005cm3 in volume, the mice were divided into DOX and vehicle control sub-

groups, and fed with 10% sucrose water with or without 2mg/ml DOX. A, tumor growth was

measured every two days using calipers, and mice culled when tumor volume reached 1cm3.

B, Kaplan-Meier survival curves showed the probability of overall survival of the mice. P

value was obtained from two-sided log-rank test. C-D, protein was extracted from the tumors

from the mice, and subjected to immunoblot analysis of DOT1L, ODC1 and E2F2 protein

expression.

Figure 6. High levels of DOT1L gene expression in human neuroblastoma tissues

independently predict poor patient prognosis. A, two-sided Pearson’s correlation was

employed to analyze correlation between DOT1L and N-Myc mRNA expression in 88 and

476 human neuroblastoma samples in the publically available microarray gene expression

Versteeg and Kocak datasets downloaded from the R2 platform (http://r2.amc.nl). B-C,

Kaplan–Meier curves showed the probability of overall survival of patients according to

DOT1L mRNA expression levels in the 88 and 476 neuroblastoma patients in the Versteeg

and Kocak datasets, using the median (B) or upper quartile (C) of DOT1L expression level as

the cut-off point respectively. D, Kaplan–Meier curves showed the probability of patient

overall survival according to the DOT1L mRNA expression level in the 72 MYCN-amplified

neuroblastoma samples in the large Kocak dataset, using the upper quartile of DOT1L

expression level as the cut-off point.

Table 1. Multivariable Cox regression analysis of DOT1L expression in tumor tissues as a

factor prognostic for outcome in 476 neuroblastoma patients*

Event-free survival Overall survival Factors HR (95%CI) P value HR (95%CI) P value High DOT1L expression 1.90 (1.315-2.753) .0001 1.81 (1.076-3.045) 0.025 (median level as cut-off) MYCN amplification 2.03 (1.397-2.949) .0002 4.52 (2.885-7.080) 4.5E-11 Age > 18 months 1.07 (0.749-1.529) .710 1.60 (1.000-2.563) .050 Stages 3 & 4† 1.05 (1.032-1.072) 2.1E-7 1.07 (1.041-1.108) 8.0E-6 High DOT1L expression 1.96 (1.329-2.903) .001 1.97 (1.180-3.281) 0.010 (upper quartile as cut-off) MYCN amplification 1.69 (1.110-2.575) .015 3.70 (2.230-6.139) 4.1E-7 Age > 18 months 1.02 (0.710-1.473) 0.904 1.54 (0.953-2.475) .078 Stages 3 & 4† 1.05 (1.034-1.074) 5.5E-8 1.08 (1.043-1.111) 4.0E-6

* DOT1L expression was considered high or low in relation to the median or upper quartile

DOT1L expression in all 476 tumors analyzed. Hazard ratios were calculated as the antilogs

of the regression coefficients in the proportional hazards regression. Multivariable Cox

regression analysis was carried out by including the above listed four factors into the Cox

regression model. P value was obtained using two-sided log-rank test.

† Tumor stage was classified as favorable (International Neuroblastoma Staging System

stages 1, 2 and 4S) or unfavorable (International Neuroblastoma Staging System stages 3 and

4).

The histone methyltransferase DOT1L promotes neuroblastoma by regulating gene transcription

Matthew Wong, Andrew EL Tee, Giorgio Milazzo, et al.

Cancer Res Published OnlineFirst February 16, 2017.

Updated version Access the most recent version of this article at: doi:10.1158/0008-5472.CAN-16-1663

Supplementary Access the most recent supplemental material at: Material http://cancerres.aacrjournals.org/content/suppl/2017/02/16/0008-5472.CAN-16-1663.DC1

Author Author manuscripts have been peer reviewed and accepted for publication but have not yet been Manuscript edited.

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/early/2017/02/16/0008-5472.CAN-16-1663. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.