Skip to main content
NCBI home page
British Journal of Cancer logoLink to British Journal of Cancer
. 2021 Oct 11;125(12):1647–1656. doi: 10.1038/s41416-021-01571-y

Combination of a synthetic retinoid and a DNA demethylating agent induced differentiation of neuroblastoma through retinoic acid signal reprogramming

Naoko Hattori 1,✉, Kiyoshi Asada 1, Nozomu Miyajima 1, Akiko Mori 1, Yoko Nakanishi 1, Kana Kimura 1, Mika Wakabayashi 1, Hideyuki Takeshima 1, Chika Nitani 2, Junichi Hara 2, Toshikazu Ushijima 1,✉
PMCID: PMC8651663  PMID: 34635821

Abstract

Background

The CpG island methylator phenotype of neuroblastoma (NBL) is strongly associated with poor prognosis and can be targeted by 5-aza-2’-deoxycytidine (5-aza-dC). Differentiation therapy is a standard maintenance therapy for high-risk NBLs. However, the in vivo effect of tamibarotene, a synthetic retinoic acid, and the efficacy of its combination with 5-aza-dC have not been studied. Here, we conducted a preclinical study to assess the in vivo tamibarotene effect and the combination.

Methods

Treatment effects were analysed by in vitro cell growth and differentiation state and by in vivo xenograft suppression. Demethylated genes were analysed by DNA methylation microarrays and geneset enrichment.

Results

Tamibarotene monotherapy induced neural extension and upregulation of differentiation markers of NBL cells in vitro, and tumour regression without severe side effects in vivo. 5-Aza-dC monotherapy suppressed tumour growth both in vitro and in vivo, and induced demethylation of genes related to nervous system development and function. Pre-treatment with 5-aza-dC in vitro enhanced upregulation of differentiation markers and genes involved in retinoic acid signaling. Pre-treatment with 5-aza-dC in vivo significantly suppressed tumour growth and reduced the variation in tumour sizes.

Conclusions

Epigenetic drug-based differentiation therapy using 5-aza-dC and TBT is a promising strategy for refractory NBLs.

Subject terms: Paediatric cancer, Cancer therapy, Cancer

Background

Neuroblastoma, the most common extracranial solid tumour in childhood, arises from primitive cells of the developing sympathetic nervous system in the neural crest and have only a limited number of driver mutations [1, 2]. Neuroblastoma cases exhibit diverse clinical courses, ranging from spontaneous regression to life-threatening progression. As one of the molecular mechanisms underlying poor prognosis of NBL cases, the presence of the CpG island methylator phenotype (CIMP) has been repeatedly confirmed [3–7], and its utility as a therapeutic target has also been suggested [8]. In addition, overexpression of EZH2, which mediates H3K27me3, has also been reported to be associated with poor prognosis [9–12]. Thus, epigenetic dysregulation is likely to be involved in tumorigenesis and progression of neuroblastoma, and DNA demethylating agents can be used to potentially target such unfavourable cases.

DNA demethylating agents, such as 5-aza-2’-deoxycytidine (5-aza-dC) and 5-azacytidine (5-aza-CR), are currently used in clinical practice for the treatment of hematological malignancies and are being developed for solid tumours [13, 14]. Treatment with 5-aza-dC induced upregulation of genes that were specifically expressed in neuroblastomas with favourable prognoses [15]. In addition, DNA demethylating agents enhanced the differentiation-inducing effects of all-trans-retinoic acid (ATRA) and 13-cis-retinoic acid (RA) [8, 16]. Clinically, treatment with 13-cis-RA in a phase III randomised trial was beneficial for patients and resulted in significant improvement of event-free survival in cases of high-risk neuroblastoma [17], and administration of 13-cis-RA is a standard therapy for high-risk NBL patients after stem cell transplantation in the US and Europe [18].

As a differentiation agent, tamibarotene (TBT: Am80) was synthessed to function as a specific agonist of the RARα and RARβ, and was shown to induce differentiation of NB-4 and HL-60 cells 10 times as potently as ATRA [19]. Treatment of NH-12 and SH-SY-5Y neuroblastoma cell lines with TBT induced morphological differentiation and expression of neuronal differentiation markers [20, 21]. Although the high potency of TBT on NBL differentiation compared with ATRA has been reported in vitro [19], its in vivo effect has not been studied. Recently, we conducted a phase I dose-escalation study to determine the recommended dose of TBT for pediatric patients with recurrent/refractory solid tumours (UMINID: 000017053) [22].

In this study, with the aim to establish epigenetic drug-based differentiation therapy for neuroblastoma using 5-aza-dC and TBT, we systematically analysed the differentiation ability of TBT in vitro and in vivo, tumour-suppressive and demethylation effects of 5-aza-dC in vitro and in vivo, and the priming effect of 5-aza-dC in vitro and in vivo. The in vivo effect of TBT and the priming effect of 5-aza-dC in vitro and in vivo are reported for the first time.

Methods

Neuroblastoma cell lines

NB-1, IMR-32 and TGW cell lines were purchased from the Japanese Collection of Research Bioresources (Tokyo, Japan). NB-39-nu cell line was obtained from the RIKEN Bio Resource Center (Tsukuba, Japan). KELLY, BE(2)-C and SH-SY5Y cell lines were obtained from the American Type Culture Collection (Manassas, VA). The cell lines were cultured in RPMI-1640 (NB-1, NB-39-nu and KELLY), in MEM (TGW), in MEM with 10% non-essential amino acid (IMR-32), or in medium consisting of a mixture of MEM and Ham’s F12 (BE(2)-C and SH-SY5Y) at 37 °C in a humidified atmosphere with 5% CO2. All media were supplemented with 10% fetal bovine serum and 1% penicillin/streptomycin. Information on MYCN amplification status was obtained from previous reports [23–26]. Cells were checked for mycoplasma infection using the MycoAlert Mycoplasma Detection Kit (Lonza; Basel, Switzerland).

Treatment of cells with 5-aza-dC and TBT

5-Aza-dC was purchased from Sigma–Aldrich (St Louis, MO), and dissolved in phosphate-buffered saline (PBS). TBT was purchased from FUJIFILM Wako Pure Chemical Corp. (Osaka, Japan), and dissolved in ethanol. All experiments using TBT were performed in dim light, and the final concentration of ethanol was 0.01%.

Analysis of neurite extension and cell viability

Cells were seeded at a density of 1 × 105 to 5 × 105 cells/well in BioCoat™ six-well plate coated with Poly-D-Lysine/Laminin (Corning, New York, NY; 354595) on Day 0 in triplicate, followed by treatment with a drug on Day 1. Cell morphology was analysed under a phase-contrast microscope, and at least two images were captured at high magnification using a BZ-X710 microscope system (Keyence, Osaka, Japan). For quantitative analysis of differentiation induction, the number and the mean length of neurites in a cell with at least one neurite were measured for 42–85 cells in each condition using the BZ-X710 software. Cell counts were determined using a microscope. The half maximum inhibitory concentration (IC50) was obtained using the non-linear regression analysis of log (inhibitor) versus the normalised response with a variable slope using GraphPad Prism software (GraphPad Software; La Jolla, CA).

Immunofluorescence staining

Immunofluorescence staining was performed as described previously [27]. Cells were stained with an anti-ßIII-tubulin mouse monoclonal antibody (1:500; Covance, Berkeley, CA; MMS-435P) in the same buffer for 1 h, and incubated with Alexa Fluor 488-conjugated goat anti-mouse IgG (1:1 000, Thermo Fisher Scientific; A32723) for 1 h.

Genome-wide DNA methylation analysis

Genome-wide DNA methylation analysis was performed using Infinium HumanMethylation450 BeadChip and MethylationEPIC BeadChip (Illumina; San Diego, CA) for NB-1 and TGW, respectively, as described previously [28, 29]. Methylation microarray analysis was conducted once for a sample because of its high reproducibility [30]. The methylation level of each CpG site was obtained as a β-value, which ranged from 0 (unmethylated) to 1 (fully methylated). To reduce the data size and obtain data that were easier to handle, the CpG probes were grouped into 296,494 genomic blocks (GBs) (Methylation450) and 548,546 GBs (MethylationEPIC), which consisted of probes within 500 bp [31]. The GBs were classified based on their locations relative to a transcription start site (TSS) [TSS200 (200 bp upstream region from TSS), TSS1500 (regions between 200 bp upstream and 1500 bp upstream from TSS), 5’-UTR, first exon, gene body, 3’-UTR and intergenic region] and their locations against a CpG island (CGI) (N Shelf, N Shore, CGI, S Shore, S Shelf and non-CGI). The definitions of an enhancer region and its target gene were based on the annotation for Methylation450 and MethylationEPIC, originated from ENCODE and FAMTOM5 projects.

Ingenuity pathways analysis (IPA)

Analysis of the biological functions of the microarray data was performed using Ingenuity Pathway Analysis (IPA) (Qiagen, Ingenuity H Systems, Redwood City, CA; http://www.ingenuity.com). A geneset was prepared from the genome-wide DNA methylation data by selecting genes whose β-value decreased (delta beta > 0.2) following 5-aza-dC treatment and that were methylated in mock-treated cells (β-value > 0.6). By performing a core analysis, modules of the “Physiological System Development and Function” module were generated from a geneset. The “upstream regulators” were also searched for. “Upstream regulators” were defined based on knowledge of relationships between upstream transcriptional regulators and their downstream targets obtained in published literature. P-values were calculated using a right-tailed Fisher’s exact test.

Quantitative reverse transcription PCR (RT-qPCR)

Complementary DNA (cDNA) was synthesised from total RNA using oligo-(dT)12-18 (Thermo Fisher Scientific) and SuperScript IV Reverse Transcriptase (for TGW) and SuperScript III Reverse Transcriptase (for the other cell lines) (Thermo Fisher Scientific). The number of cDNA molecules was quantified using real-time PCR. The primer sequences and PCR conditions are shown in Supplementary Table 1. The copy number of each sample was calculated by comparing the amplification curve with those of standard DNA samples with known copy numbers. The number of target cDNA molecules was normalised to that of GAPDH cDNA molecules. A Venn diagram showing overlapping genes upregulated by the treatment of 5-aza-dC, TBT and their combination was drawn using the R package (https://cran.rproject.org/).

Expression microarray experiments and data processing

Expression microarray experiments were conducted as described previously [32]. The 75th percentile of the signal intensity of all probes was normalised to be 0, and the signals were distributed from −7 to 10 in NB-1 treated with mock and 5-aza-dC, and −6 to 10 in NB-1 treated with TBT and combination drugs. Genes with signal intensities of −1 or more were regarded as having positive expression. The expression microarray experiment was conducted once for a sample based upon its reported high reproducibility [33].

Xenograft tumour formation assay in nude mice

Female nude mice (BALB/c-nu/nu; 6 week old; ~20 g) were purchased from Charles River Lab. Japan, Inc. (Yokohama, Japan), and were housed in specific pathogen-free and barrier conditions under a 12 h dark/light cycle with free access to food and water. They were used after an acclimation time of 7 days. NB-1 cells (5.0 × 106 cells per mouse) were subcutaneously injected into inguinal regions of mice. For monotherapy treatment of TBT or 5-aza-dC, once tumours reached an average volume of 100 mm3, mice bearing xenograft tumours were randomly divided into four groups and treated with vehicle (PBS in 10% ethanol for TBT; PBS for 5-aza-dC), TBT, and 5-aza-dC. TBT was orally administered at 2.0 mg/kg on days 1 to 7 of each 14-day treatment cycle in a total of three cycles. 5-Aza-dC was intraperitoneally administered at 5 mg/kg on days 1 to 3 of each 14-day treatment cycle in a total of three cycles. For combined treatment of TBT or 5-aza-dC, when tumours reached a volume of 100 mm3, the mice were randomised into four groups (PBS in 10% ethanol, 5-aza-dC, TBT and both 5-aza-dC and TBT). 5-Aza-dC was intraperitoneally administered daily at 2.5 mg/kg on days 1–3, days 15–17 and days 29–31, and TBT was orally administered daily at 1.0 mg/kg on days 8–14 and days 22–28. The length and width of tumours were measured with standard calipers twice per week, and tumour volumes were calculated using the formula: tumour volume = (length × width2) × 0.5. Bodyweight was measured twice per week. At 42 days (for TBT), 35 days (for 5-aza-dC), or 34 days (for combination) after the treatment, mice were anesthetised with isoflurane and sacrificed by bleeding under isoflurane anesthesia in the afternoon. Tumours and major organs were immediately collected. All experimental procedures were conducted in a laboratory. Total blood was collected to determine the number of leukocytes, erythrocytes and platelets. All experimental procedures in the present study were approved by the Committee of National Cancer Center.

Statistical analysis

Differences in tumour volume, bodyweight and blood count were analysed using an unpaired Student’s t-test. Gene expression was compared using a paired Student’s t-test. The results were considered significant when a P-value < 0.05 was obtained by a two-sided test. All calculations were performed using the Microsoft Excel software (Microsoft Corp, Seattle, WA).

Results

TBT-induced differentiation in vitro and tumour-growth suppression in vivo

To evaluate the effects of TBT on the differentiation of neuroblastoma cell lines in vitro, we treated seven cell lines with 0.01, 0.1, 1 and 10 µM TBT for four days (Supplementary Fig. S1a). Cellular growth was suppressed with 0.1–10 µM TBT in five of the seven cell lines (Supplementary Fig. S1b). Morphological differentiation was estimated by neurite outgrowth, a reliable marker for neuronal differentiation of neuroblastoma cells [18]. Immunofluorescence staining clearly showed that the neurite induced by TBT treatment expressed ßIII-tubulin, a marker for differentiated neurons (Fig. 1a). Quantitative analysis of the number and length of neurites also showed that TBT treatment significantly increased neurite length in six cell lines (NB-1, NB-39-nu, IMR-32, KELLY, BE(2)-C and SH-SY5Y) (Fig. 1b, and Supplementary Fig. S1c, S1d).

Fig. 1. Effects of TBT on neuroblastoma differentiation and tumour growth.

Fig. 1

Open in a new tab

a Representative images of neural outgrowth and immunofluorescence staining. Neural extension and increased expression of ßIII-tubulin were observed in NB-1 cells treated with 1 µM of TBT. Scale bar = 100 µm. b Quantification of neurite length and number. The average neurite length and the number of neurites in a cell were increased in NB-1 cells by treatment with 1 µM of TBT. n.s. non-significant. c Induction of six differentiation marker genes in seven cell lines treated with TBT. Heatmap was drawn using the maximum and minimum values of expression levels of each gene. Information on MYCN amplification status was obtained from previous reports. With the criterion of upregulation of three genes or more among the six genes evaluated, five of the seven cell lines showed differentiation by TBT. d Xenograft growth curves of the TBT- and mock-treated groups. Results are shown as mean ± SE, and statistical significance was tested using an unpaired Student’s t-test (n = 8/vehicle group; n = 6 /TBT group). e Macroscopic view of the xenografts in the two treatment groups. f The therapeutic effect assessed by tumour weight at necropsy. Horizontal bars show the average. Although there was no significant difference in the average tumour weight between mock- and TBT-treated groups, the macroscopic sizes of xenograft tumours in the TBT-treated group were smaller than those of the vehicle-administered group.

To further estimate the differentiation of neuroblastoma cells, we selected six genes (ARHGEF3, GAP43, HOXD4, NGFR, NTRK1 and NTRK2) as differentiation markers based on results from previous studies [9, 20, 21, 34–36]. Induction of the marker genes was analysed in the seven cell lines treated with 0.01, 0.1, 1.0 and 10 µM of TBT (Fig. 1c, and Supplementary Fig. S1e). Five (NB-1, IMR-32, NB-39-nu, SH-SY5Y and KELLY) of the seven cell lines showed upregulation of three or more of the six marker genes. SH-SY5Y showed a strong increase in NTRK2 expression and a slight increase in NTRK1 expression after TBT treatment as reported previously [20, 21]. KELLY also exhibited upregulation of NTRK2, which has been reported to be induced by ATRA treatment [37]. From these data, we concluded that TBT could induce differentiation of neuroblastoma cells in vitro.

Next, the in vivo anti-tumour effect of TBT was evaluated by transplanting NB-1 cells into the inguinal region of nude mice. We administered 2.0 mg/kg of TBT on days 1–7 of each 14-day treatment cycle for a total of three cycles, and found significant suppression of xenograft growth (Fig. 1d). Although there was no significant difference in the average tumour weight between mock- and TBT-treated groups (P = 0.15), the macroscopic size of xenograft tumours in the TBT-treated group was smaller than that of the vehicle-administered group at the time of sacrifice (Fig. 1e and f). As for the adverse effects, there was no significant difference in the bodyweight loss between vehicle and TBT groups (Supplementary Fig. S2a). The white blood cell (WBC) count was decreased in the TBT-treated group (31% of control mice) while the levels of red blood cells (RBCs), hemoglobin and hematocrit were not affected (Supplementary Fig. S2). These findings demonstrated that TBT exerted an anti-tumour effect in vivo.

5-Aza-dC-induced tumour-growth suppression in vitro and in vivo

The effect of 5-aza-dC on tumour growth in vitro was analysed by changes in cell numbers following 5-aza-dC treatment. Treatment with 5-aza-dC on days 1, 2 and 3 showed growth-suppressive effects in a dose-dependent manner (Fig. 2a). The IC50 values of the cell lines were less than 0.4 µM (Fig. 2a). Further, in vivo anti-tumour effect of 5-aza-dC was analysed in mice bearing the NB-1 xenograft. 5-Aza-dC was intraperitoneally administered at 5 mg/kg, in accordance with a previous study on in vivo tolerability [38], on days 1 to 3 of each 14-day treatment cycle for a total of three cycles. Treatment with 5-Aza-dC inhibited tumour growth (Fig. 2b and c) and the tumour size was consistently small even though there were no significant differences in the average tumour weight between vehicle- and 5-aza-dC-treated groups (P = 0.20) (Fig. 2d). Moderate hematotoxicity, such as decreased WBC count (33% of control mice), was observed in the 5-aza-dC-treated mice (Supplementary Fig. S2).

Fig. 2. Effects of 5-aza-dC on tumour growth.

Fig. 2

Open in a new tab

a Dose-response curves of the seven neuroblastoma cell lines to 5-aza-dC. Data are shown as the mean ± SD obtained from three experiments (left). The IC50 values of the seven neuroblastoma cell lines to 5-aza-dC (right). Treatment with 5-aza-dC induced a growth-suppressive effect in a dose-dependent manner. b Xenograft growth curves of the 5-aza-dC- and mock-treated groups. Results are shown as mean ± SE, and statistical significance was tested using an unpaired Student’s t-test (n = 5/vehicle group; n = 6/5-aza-dC group). c Macroscopic view of the xenografts in the two treatment groups. d The therapeutic effect assessed by tumour weight at necropsy. Horizontal bars show the average. The tumour size was consistently small even though there were no significant differences in the average tumour weight between vehicle- and 5-aza-dC-treated groups.

DNA demethylation of neuronal differentiation-associated genes by 5-aza-dC

The demethylation effect of 5-aza-dC was analysed in NB-1 and TGW using an Infinium microarray. Compared with mock-treated cells, 5-aza-dC-treated cells showed decreased β-values at virtually all CpG sites, demonstrating that the demethylation ability of 5-aza-dC was achieved in the tested neuroblastoma cell lines (Fig. 3a). Even when analysis was limited to CpG sites in CpG islands (CGIs), the demethylation effect was similar.

Fig. 3. Effects of 5-aza-dC on DNA methylation.

Fig. 3

Open in a new tab

a Genome-wide DNA demethylation by 5-aza-dC in NB-1 and TGW cell lines. Genome-wide DNA methylation levels were analysed using an Infinium Human Methylation BeadChip (450 K for NB-1, EPIC for TGW) in non-treated cells and 5-aza-dC-treated cells (0.3 µM) and compared. Methylation levels are shown as β-values (0.0–1.0). Comparison between non-treated cells and treated cells showed that 5-aza-dC induced strong DNA demethylation at CpG sites and at CpG islands (CGIs). b IPA analysis of the biological functions of hypomethylated CGIs located at TSS200 or enhancer regions. “Embryonic development” was significantly enriched in both cell lines, and “Nervous system development and function” was enriched in NB-1 cells. c Identification of upstream regulators by IPA. Transcription factors related to neuron differentiation were identified as upstream regulators of hypomethylated genes. d DNA methylation status of the target genes of ZFHX3 and ADRB2 in NB-1 and TGW, respectively. The genes involved in neural differentiation and RA signaling were demethylated by 5-aza-dC treatment.

To identify demethylation sites that resulted in changes in gene expression, we focused on CGIs that were located within 200 bp from a transcription start site (TSS200) or in enhancer regions [39], highly methylated in non-treated cells (β-values ≥ 0.6), and demethylated by 5-aza-dC treatment (Δ β-values ≥ 0.2) (marked by a trapezoid with a red broken line, Fig. 3a). The biological functions of the genes with demethylated promoter or enhancer CGIs were analysed using IPA. In the “physiological system development and function” module, the “embryonic development” category was the most significantly enriched in NB-1 and TGW, and the “nervous system development and function” category was also one of the top five significantly enriched categories in NB-1 (Fig. 3b).

Furthermore, the upstream regulators enriched for the demethylated genes were examined by IPA (Fig. 3c and d). In NB-1, putative target genes of ZFHX3, a transcription factor involved in neuron differentiation, were preferentially demethylated. Interestingly, ZFHX3 itself was demethylated by 5-aza-dC treatment. In TGW, the putative target genes of ADRB2, a transcription factor related to neuron differentiation, were demethylated. These data showed that 5-aza-dC treatment reprogramed neuron differentiation-related genes, and suggested that the treatment may enhance the differentiation induction by differentiation agents.

5-Aza-dC treatment enhanced differentiation by TBT

To investigate the effects of 5-aza-dC on enhancement of TBT-induced differentiation in vitro, seven cell lines were first treated with 0.1, 0.3 or 1 µM 5-aza-dC for 3 days, and then treated with 1 or 10 µM TBT for an additional 4 days (Fig. 4a). In NB-1, neurite length increased after TBT treatment alone, but both neurite length and number were increased when 5-aza-dC pre-treatment was introduced (Figs. 4b and 4c). NB-39nu, IMR-32 and KELLY showed only a mild effect by TBT treatment alone, but they showed increased neurite number and length following 5-aza-dC pre-treatment (Supplementary Fig. S3b and S3c). Expression microarray analysis of NB-1 showed that the number of upregulated genes by the combined treatment was larger than those by a single treatment with 5-aza-dC or TBT (Fig. 4d) (511 by the combination, 373 by 5-aza-dC and 184 by TBT). Few genes upregulated by TBT treatment overlapped those by 5-aza-dC monotherapy or by combination treatment (Fig. 4e). Only half of genes upregulated by the combination overlapped those upregulated by 5-aza-dC monotherapy, supporting that the combination treatment was able to induce genes not induced by monotherapy via its synergistic effect (Fig. 4e). The expression of differentiation marker genes (ARHGEF3, HOXD4, GAP43 and NTRK1) were highly induced when TBT and 5-aza-dC were combined in the seven cell lines (Fig. 4f).

Fig. 4. Effects of the combination of 5-aza-dC and TBT on neuroblastoma differentiation.

Fig. 4

Open in a new tab

a Schedule of the combined treatment of 5-aza-dC (0.1, 0.3 or 1 µM) and TBT (1 or 10 µM). b Representative immunofluorescence staining data. Expression of ßIII-tubulin was enhanced in NB-1 cells by pre-treatment with 1 µM of 5-aza-dC for three days prior to exposure to 1 µM of TBT. Scale bar = 100 µm. c Quantification of neurite length and number in NB-1 cells. The number of neurites per cell increased by 5-aza-dC treatment prior to TBT treatment. d Induction of gene expression by 5-aza-dC, TBT and combined treatment. Expression levels (log2 values) of 50,739 probes were obtained by an expression microarray, and drug-treated cells (y-axis) were compared with non-treated cells (x-axis). Genes with no change were plotted on the black dashed line, and genes upregulated two-fold or more and with expression > −1.0 after a drug treatment are surrounded by a trapezoid with red dashed lines. The numbers of upregulated genes for 5-aza-dC, TBT and combined treatment were 373, 184 and 511, respectively. e Numbers of genes overlappingly upregulated by the treatment of 5-aza-dC, TBT and their combination. f Expression of neuronal markers in the seven cell lines. Additional or synergistic effects were observed by the combination of 5-aza-dC and TBT (1.0 µM 5-aza-dC and 10 µM TBT for NB-1; 0.3 µM and 1.0 µM for NB-39nu; 0.3 µM and 10 µM for IMR-32, SH-SY5Y, BE(2)-C and TGW; 0.1 µM and 1.0 µM for KELLY). Each RT-qPCR analysis was performed in triplicate, and the results are presented as mean ± SE. n.s. non-significant. g Expression of the genes involved in RA signaling in seven cell lines. The expression of CYP26B1, S100A6, HIC1 and CDH11 were additively or synergistically upregulated by the combination of 5-aza-dC and TBT treatment (CYP26B1 expression: 0.3 µM 5-aza-dC and 1.0 µM TBT for NB-1, NB-39nu, IMR-32 and TGW; 0.3 µM and 10 µM for SH-SY5Y and BE(2)-C; 0.1 µM and 1.0 µM for KELLY) (Other genes: 1.0 µM 5-aza-dC and 10 µM TBT for NB-1; 0.3 µM and 1.0 µM for NB-39nu and SH-SY5Y; 0.3 µM and 10 µM for IMR-32 and TGW; 0.1 µM and 1.0 µM for KELLY). Values represent the mean ± SE of three experiments. n.s. non-significant.

To reveal the molecular mechanisms involved in enhancement of differentiation by 5-aza-dC pre-treatment, we focused on the genes involved in RA signaling. The expression of CYP26B1, which contributes to the metabolism of RA and is also a target gene of RARα [40], was synergistically induced in all the cell lines, especially in KELLY (Fig. 4f). The expression of the RAR target genes, including S100A6, a glial cell marker [41] and CDH11, was additively upregulated by 5-aza-dC pre-treatment. HIC1, a direct target of RAR and a tumour suppressor gene silenced by DNA methylation [42], was upregulated by 5-aza-dC pre-treatment in IMR-32 and KELLY (Fig. 4g). These data supported the usefulness of reprograming by 5-aza-dC before TBT treatment.

Combination of 5-aza-dC and TBT induced tumour-growth suppression in vivo

To show the effect of combined treatment with 5-aza-dC and TBT in vivo, we applied a half concentration of 5-aza-dC and TBT used for monotherapy because monotherapy with 5-aza-dC and TBT induced strong tumour suppressive effects, and lower doses are more likely to avoid serious adverse effects. 5-Aza-dC was intraperitoneally administered daily at 2.5 mg/kg on days 1–3, days 15–17 and days 29–31, and TBT was orally administered daily at 1.0 mg/kg on days 8–14 and days 22–28. Combined treatment of 5-aza-dC and TBT significantly suppressed tumour growth (P < 0.05), while 5-aza-dC monotherapy showed no effect (Fig. 5a and b). The xenograft tumours in mock and 5-aza-dC monotherapy groups showed large variation in their size, but the combination of 5-aza-dC and TBT reduced the variation and showed small sizes (Fig. 5c). As for the adverse effects, there were no significant differences in bodyweight loss in any groups (Fig. 5d). WBC and RBC counts, along with hemoglobin level, were moderately suppressed by 5-aza-dC, but no further suppression was induced by the combination (Fig. 5e). These findings showed that 5-aza-dC pre-treatment and TBT synergistically suppressed tumour growth in vivo.

Fig. 5. Effects of a combination of 5-aza-dC and TBT on tumour growth.

Fig. 5

Open in a new tab

a Xenograft growth curves of the mock-, 5-aza-dC-, TBT- and combination-treated groups. A pre-treatment with 5-aza-dC induced significant suppression of tumour growth (P < 0.05). Results are shown as mean ± SE, and statistical significance was tested using an unpaired Student’s t-test (n = 5/vehicle group; n = 4/5-aza-dC group, n = 3/TBT group, n = 4/combination group). n.s. non-significant. b Macroscopic view of the xenografts in the four treatment groups. c The therapeutic effect assessed by tumour weight at necropsy. Horizontal bars show the average. The xenograft tumours in mock and 5-aza-dC monotherapy groups showed large variation in their size, but the combination of 5-aza-dC and TBT reduced the variation and showed small sizes. d Changes in bodyweight. Bodyweight was measured twice per week. There was no significant difference in bodyweight among the four treatment groups. e Effects on blood cell counts. WBC and RBC counts, along with hemoglobin level, were suppressed moderately by 5-aza-dC, but no further suppression was induced by the combination. Results are shown as mean ± SE (n = 5/vehicle group; n = 3/5-aza-dC group, n = 3/TBT group, n = 4/combination group). n.s. non-significant.

Discussion

This study showed that pre-treatment with 5-aza-dC enhanced the differentiation-inducing effects of TBT in multiple neuroblastoma cell lines in vitro, regardless of the MYCN amplification status (Table 1). Both TBT and 5-aza-dC administration suppressed the growth of xenograft tumours with moderate adverse effects. Mechanistically, demethylation of the genes involved in neuronal differentiation and RA target genes, namely epigenetic reprograming of these genes, was shown. The in vivo toxicity of the individual drugs was tolerable, and it is expected that the doses of individual drugs can be further reduced by the combination strategy. The combination therapy using a half concentration of TBT and 5-aza-dC of monotherapy significantly inhibited tumour growth while the same dose of 5-aza-dC or TBT showed no effect on tumour-growth suppression. However, the level of adverse effects was not improved by the combination therapy, which may be due to the effect of 5-aza-dC. There is a possibility that improvement of the pharmacokinetics of 5-aza-dC, for example by the use of oral formula or prodrugs, will improve clinical response without increasing adverse effects [43, 44]. Based on the findings from this study, clinical trials are warranted to treat neuroblastoma patients with a combination of a DNA demethylating agent and TBT.

Table 1.

Neuroblastoma cell lines and their responses: MYCN status, IC50 for 5-aza-dC and differentiation status.

Cell line MYCN status IC50 for
5-aza-dC
(µM)		 Differentiation status
Neurite extension Marker gene expression*1
5-aza-dC (1 µM) TBT (1 µM) Combination 5-aza-dC (1 µM) TBT (1 µM) Combination
NB-1 + 0.30 + + + + + +
NB-39-nu + 0.30 + + + + + +
IMR-32 + 0.10 − + + − + +
SH-SY5Y − 0.05 NA + NA _ − +
KELLY + 0.03 − + +*2 − − +
BE(2)-C + 0.11 NA + NA − − +
TGW + 0.07 NA − NA − − +
Open in a new tab

+ indicates positive, − indicates negative. *1, + significant upreguation; − no significant upregulation. *2 0.5 µM 5-aza-dC.

NA not analysed.

We also showed the efficacy of TBT in vivo as a single agent in neuroblastoma xenograft model. Although the high potency of TBT on differentiation compared with ATRA has been reported in vitro [19], its in vivo effects have not been well studied. This study showed that TBT was able to suppress tumour growth without loss of bodyweight or severe hematotoxicity. To note, Ro 13-6307, another synthetic retinoid, induced much more severe bodyweight loss in rats compared with 13-cis-RA [45], and both ATRA and 13-cis-RA have shown retinoid toxicity and resistance in clinical cases [46]. TBT may overcome the current issues with the retinoid therapy, and is considered worthy of being developed as a neuroblastoma therapy.

As a mode of action of DNA demethylating agents, the enhanced immune reaction resulting from the activation of an endogenous retrovirus has recently attracted a lot of attention, in addition to the reactivation of multiple tumour suppressor genes, namely epigenetic reprograming [47]. Herein, our findings showed that the enhancement of differentiation by epigenetic reprograming of the promoters or enhancers of neuron differentiation-associated genes and RA-signaling genes was at least involved in the modes of action in neuroblastoma. Previous studies also reported that 5-aza-dC enhanced the expression of genesets in favourable neuroblastoma cases [15], and that hypoxia signaling was upregulated by a combination of AZA and RA [16]. Taken together, in neuroblastoma, DNA demethylating agents appear to reprogram multiple signals, leading to a beneficial effect in combination with a differentiation agent.

Not all of the neuroblastoma cell lines responded to the differentiation induction by TBT (Table 1), suggesting that stratification of patients who are likely to respond may be necessary. In acute myeloid leukemia (AML), it is known that patients with super-enhancer at RARα are likely to respond to TBT [48]. Further preclinical and clinical studies for the development of biomarkers may be necessary.

In conclusion, a DNA demethylating agent and TBT, especially when used in combination, were found to be effective in suppressing neuroblastoma growth through the upregulation of RA-signaling genes.

Supplementary information

Supplementary information file (1.3MB, docx)

Acknowledgements

We thank Dr. H. Kawamoto for his critical discussion. We also thank Dr. T. Imai, Dr. N. Uchiya and Ms. Y. Shiotani, and the National Cancer Center Research Core Facility for their assistance in the analysis of xenografts.

Author contributions

TU, NH, KA, CN and JH designed the study; NH, NM, AM, YN, KK, MW and HT performed most of the experiments; NH and TU wrote the manuscript; and all authors read and approved the manuscript.

Funding

The Core Facility was supported in part by National Cancer Center Research and Development Fund (2020-J-2). This study was supported by the AMED under Grant Number JP20ck0106421 to TU, NH, CN and JH, and JSPS KAKENHI Grant Number JP18H02704 to TU and NH.

Data availability

The datasets used in this study are available at the Gene Expression Omnibus (GEO) database (accession no. GSE180645, GSE180660 and GSE180929, https://www.ncbi.nlm.nih.gov/geo/).

Competing interests

The authors declare no competing interests.

Ethics approval and consent to participate

All animal experiments were approved by the Committee of National Cancer Center and were conducted according to the institutional guidelines for animal care.

Consent for publication

Not applicable.

Footnotes

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Contributor Information

Naoko Hattori, Email: nhattori@ncc.go.jp.

Toshikazu Ushijima, Email: tushijim@ncc.go.jp.

Supplementary information

The online version contains supplementary material available at 10.1038/s41416-021-01571-y.

References

  • 1.Brodeur GM. Neuroblastoma: biological insights into a clinical enigma. Nat Rev Cancer. 2003;3:203–16. doi: 10.1038/nrc1014. [DOI] [PubMed] [Google Scholar]
  • 2.Cheung NK, Dyer MA. Neuroblastoma: developmental biology, cancer genomics and immunotherapy. Nat Rev Cancer. 2013;13:397–411. doi: 10.1038/nrc3526. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Abe M, Ohira M, Kaneda A, Yagi Y, Yamamoto S, Kitano Y, et al. CpG island methylator phenotype is a strong determinant of poor prognosis in neuroblastomas. Cancer Res. 2005;65:828–34. [PubMed] [Google Scholar]
  • 4.Abe M, Westermann F, Nakagawara A, Takato T, Schwab M, Ushijima T. Marked and independent prognostic significance of the CpG island methylator phenotype in neuroblastomas. Cancer Lett. 2007;247:253–8. doi: 10.1016/j.canlet.2006.05.001. [DOI] [PubMed] [Google Scholar]
  • 5.Banelli B, Brigati C, Di Vinci A, Casciano I, Forlani A, Borzi L, et al. A pyrosequencing assay for the quantitative methylation analysis of the PCDHB gene cluster, the major factor in neuroblastoma methylator phenotype. Lab Invest. 2012;92:458–65. doi: 10.1038/labinvest.2011.169. [DOI] [PubMed] [Google Scholar]
  • 6.Banelli B, Merlo DF, Allemanni G, Forlani A, Romani M. Clinical potentials of methylator phenotype in stage 4 high-risk neuroblastoma: an open challenge. PLoS ONE. 2013;8:e63253. doi: 10.1371/journal.pone.0063253. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Kiss NB, Kogner P, Johnsen JI, Martinsson T, Larsson C, Geli J. Quantitative global and gene-specific promoter methylation in relation to biological properties of neuroblastomas. BMC Med Genet. 2012;13:83. doi: 10.1186/1471-2350-13-83. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Abe M, Watanabe N, Mcdonell N, Takato T, Ohira M, Nakagawara A, et al. Identification of genes targeted by CpG island methylator phenotype in neuroblastomas, and their possible integrative involvement in poor prognosis. Oncology. 2008;74:50–60. doi: 10.1159/000139124. [DOI] [PubMed] [Google Scholar]
  • 9.Wang C, Liu Z, Woo CW, Li Z, Wang L, Wei JS, et al. EZH2 Mediates epigenetic silencing of neuroblastoma suppressor genes CASZ1, CLU, RUNX3, and NGFR. Cancer Res. 2012;72:315–24. doi: 10.1158/0008-5472.CAN-11-0961. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Bate-Eya LT, Gierman HJ, Ebus ME, Koster J, Caron HN, Versteeg R, et al. Enhancer of zeste homologue 2 plays an important role in neuroblastoma cell survival independent of its histone methyltransferase activity. Eur J Cancer. 2017;75:63–72. doi: 10.1016/j.ejca.2016.12.019. [DOI] [PubMed] [Google Scholar]
  • 11.Li Z, Takenobu H, Setyawati AN, Akita N, Haruta M, Satoh S, et al. EZH2 regulates neuroblastoma cell differentiation via NTRK1 promoter epigenetic modifications. Oncogene. 2018;37:2714–27. doi: 10.1038/s41388-018-0133-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Qadeer ZA, Valle-Garcia D, Hasson D, Sun Z, Cook A, Nguyen C, et al. ATRX in-frame fusion neuroblastoma is sensitive to EZH2 inhibition via modulation of neuronal gene signatures. Cancer Cell. 2019;36:512–27. doi: 10.1016/j.ccell.2019.09.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Diesch J, Zwick A, Garz AK, Palau A, Buschbeck M, Gotze KS. A clinical-molecular update on azanucleoside-based therapy for the treatment of hematologic cancers. Clin Epigenetics. 2016;8:71. doi: 10.1186/s13148-016-0237-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Linnekamp JF, Butter R, Spijker R, Medema JP, Van Laarhoven HWM. Clinical and biological effects of demethylating agents on solid tumours—a systematic review. Cancer Treat Rev. 2017;54:10–23. doi: 10.1016/j.ctrv.2017.01.004. [DOI] [PubMed] [Google Scholar]
  • 15.Tang XX, Robinson ME, Riceberg JS, Kim DY, Kung B, Titus TB, et al. Favorable neuroblastoma genes and molecular therapeutics of neuroblastoma. Clin Cancer Res. 2004;10:5837–44. doi: 10.1158/1078-0432.CCR-04-0395. [DOI] [PubMed] [Google Scholar]
  • 16.Westerlund I, Shi Y, Toskas K, Fell SM, Li S, Surova O, et al. Combined epigenetic and differentiation-based treatment inhibits neuroblastoma tumor growth and links HIF2alpha to tumor suppression. Proc Natl Acad Sci USA. 2017;114:E6137–E6146. doi: 10.1073/pnas.1700655114. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Matthay KK, Villablanca JG, Seeger RC, Stram DO, Harris RE, Ramsay NK, et al. Treatment of high-risk neuroblastoma with intensive chemotherapy, radiotherapy, autologous bone marrow transplantation, and 13-cis-retinoic acid. Children’s Cancer Group. N Engl J Med. 1999;341:1165–73. doi: 10.1056/NEJM199910143411601. [DOI] [PubMed] [Google Scholar]
  • 18.Reynolds CP, Matthay KK, Villablanca JG, Maurer BJ. Retinoid therapy of high-risk neuroblastoma. Cancer Lett. 2003;197:185–92. doi: 10.1016/s0304-3835(03)00108-3. [DOI] [PubMed] [Google Scholar]
  • 19.Ohnishi K. PML-RARalpha inhibitors (ATRA, tamibaroten, arsenic troxide) for acute promyelocytic leukemia. Int J Clin Oncol. 2007;12:313–7. doi: 10.1007/s10147-007-0694-6. [DOI] [PubMed] [Google Scholar]
  • 20.Shiohira H, Kitaoka A, Enjoji M, Uno T, Nakashima M. Am80 induces neuronal differentiation via increased tropomyosin-related kinase B expression in a human neuroblastoma SH-SY5Y cell line. Biomed Res. 2012;33:291–7. doi: 10.2220/biomedres.33.291. [DOI] [PubMed] [Google Scholar]
  • 21.Shiohira H, Kitaoka A, Shirasawa H, Enjoji M, Nakashima M. Am80 induces neuronal differentiation in a human neuroblastoma NH-12 cell line. Int J Mol Med. 2010;26:393–9. [PubMed] [Google Scholar]
  • 22.Nitani C, Hara J, Kawamoto H, Taguchi T, Kimura T, Yoshimura K, et al. Phase I study of tamibarotene monotherapy in pediatric and young adult patients with recurrent/refractory solid tumors. Cancer Chemother Pharmacol. 2021;88:99–107. doi: 10.1007/s00280-021-04271-9. [DOI] [PubMed] [Google Scholar]
  • 23.Kato H, Okamura K, Kurosawa Y, Kishikawa T, Hashimoto K. Characterization of DNA rearrangements of N-myc gene amplification in three neuroblastoma cell lines by pulsed-field gel electrophoresis. FEBS Lett. 1989;250:529–35. doi: 10.1016/0014-5793(89)80790-2. [DOI] [PubMed] [Google Scholar]
  • 24.Suzuki T, Hirota M, Iwabuchi M, Nomura N, Saito F. A new human neuroblasoma cell line: biological characteristics, cytogenetics and N-myc oncogene analysis. Tissue Cult Res Commun. 1990;8:6–15. [Google Scholar]
  • 25.Kang JH, Rychahou PG, Ishola TA, Qiao J, Evers BM, Chung DH. MYCN silencing induces differentiation and apoptosis in human neuroblastoma cells. Biochem Biophys Res Commun. 2006;351:192–7. doi: 10.1016/j.bbrc.2006.10.020. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Harenza JL, Diamond MA, Adams RN, Song MM, Davidson HL, Hart LS, et al. Transcriptomic profiling of 39 commonly-used neuroblastoma cell lines. Sci Data. 2017;4:170033. doi: 10.1038/sdata.2017.33. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Hattori N, Niwa T, Kimura K, Helin K, Ushijima T. Visualization of multivalent histone modification in a single cell reveals highly concerted epigenetic changes on differentiation of embryonic stem cells. Nucleic Acids Res. 2013;41:7231–9. doi: 10.1093/nar/gkt528. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Yamashita S, Kishino T, Takahashi T, Shimazu T, Charvat H, Kakugawa Y, et al. Genetic and epigenetic alterations in normal tissues have differential impacts on cancer risk among tissues. Proc Natl Acad Sci USA. 2018;115:1328–33. doi: 10.1073/pnas.1717340115. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Ishihara H, Yamashita S, Liu YY, Hattori N, El-Omar O, Ikeda T, et al. Genetic and epigenetic profiling indicates the proximal tubule origin of renal cancers in end-stage renal disease. Cancer Sci. 2020;111:4276–87. [DOI] [PMC free article] [PubMed]
  • 30.Bibikova M, Barnes B, Tsan C, Ho V, Klotzle B, Le JM, et al. High density DNA methylation array with single CpG site resolution. Genomics. 2011;98:288–95. doi: 10.1016/j.ygeno.2011.07.007. [DOI] [PubMed] [Google Scholar]
  • 31.Iida N, Okuda Y, Ogasawara O, Yamashita S, Takeshima H, Ushijima T. MACON: a web tool for computing DNA methylation data obtained by the Illumina Infinium Human DNA methylation BeadArray. Epigenomics. 2018;10:249–58. doi: 10.2217/epi-2017-0093. [DOI] [PubMed] [Google Scholar]
  • 32.Moro H, Hattori N, Nakamura Y, Kimura K, Imai T, Maeda M, et al. Epigenetic priming sensitizes gastric cancer cells to irinotecan and cisplatin by restoring multiple pathways. Gastric Cancer. 2020;23:105–15. doi: 10.1007/s10120-019-01010-1. [DOI] [PubMed] [Google Scholar]
  • 33.Consortium M. Shi L, Reid LH, Jones WD, Shippy R, Warrington JA, et al. The MicroArray Quality Control (MAQC) project shows inter- and intraplatform reproducibility of gene expression measurements. Nat Biotechnol. 2006;24:1151–61. doi: 10.1038/nbt1239. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Fernandes ND, Sun Y, Price BD. Activation of the kinase activity of ATM by retinoic acid is required for CREB-dependent differentiation of neuroblastoma cells. J Biol Chem. 2007;282:16577–84. doi: 10.1074/jbc.M609628200. [DOI] [PubMed] [Google Scholar]
  • 35.Takenobu H, Shimozato O, Nakamura T, Ochiai H, Yamaguchi Y, Ohira M, et al. CD133 suppresses neuroblastoma cell differentiation via signal pathway modification. Oncogene. 2011;30:97–105. doi: 10.1038/onc.2010.383. [DOI] [PubMed] [Google Scholar]
  • 36.Frumm SM, Fan ZP, Ross KN, Duvall JR, Gupta S, Verplank L, et al. Selective HDAC1/HDAC2 inhibitors induce neuroblastoma differentiation. Chem Biol. 2013;20:713–25. doi: 10.1016/j.chembiol.2013.03.020. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Dedoni S, Marras L, Olianas MC, Ingianni A, Onali P. Downregulation of TrkB expression and signaling by valproic acid and other histone deacetylase inhibitors. J Pharmacol Exp Ther. 2019;370:490–503. doi: 10.1124/jpet.119.258129. [DOI] [PubMed] [Google Scholar]
  • 38.Chuang JC, Warner SL, Vollmer D, Vankayalapati H, Redkar S, Bearss DJ, et al. S110, a 5-Aza-2’-deoxycytidine-containing dinucleotide, is an effective DNA methylation inhibitor in vivo and can reduce tumor growth. Mol Cancer Ther. 2010;9:1443–50. doi: 10.1158/1535-7163.MCT-09-1048. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Henrich KO, Bender S, Saadati M, Dreidax D, Gartlgruber M, Shao C, et al. Integrative genome-scale analysis identifies epigenetic mechanisms of transcriptional deregulation in unfavorable neuroblastomas. Cancer Res. 2016;76:5523–37. doi: 10.1158/0008-5472.CAN-15-2507. [DOI] [PubMed] [Google Scholar]
  • 40.Isoherranen N, Zhong G. Biochemical and physiological importance of the CYP26 retinoic acid hydroxylases. Pharmacol Ther. 2019;204:107400. doi: 10.1016/j.pharmthera.2019.107400. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Acosta S, Mayol G, Rodriguez E, Lavarino C, De Preter K, Kumps C, et al. Identification of tumoral glial precursor cells in neuroblastoma. Cancer Lett. 2011;312:73–81. doi: 10.1016/j.canlet.2011.08.004. [DOI] [PubMed] [Google Scholar]
  • 42.Hassan HM, Kolendowski B, Isovic M, Bose K, Dranse HJ, Sampaio AV, et al. Regulation of active DNA demethylation through RAR-mediated recruitment of a TET/TDG complex. Cell Rep. 2017;19:1685–97. doi: 10.1016/j.celrep.2017.05.007. [DOI] [PubMed] [Google Scholar]
  • 43.Hattori N, Sako M, Kimura K, Iida N, Takeshima H, Nakata Y, et al. Novel prodrugs of decitabine with greater metabolic stability and less toxicity. Clin Epigenetics. 2019;11:111. doi: 10.1186/s13148-019-0709-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Karahoca M, Momparler RL. Pharmacokinetic and pharmacodynamic analysis of 5-aza-2’-deoxycytidine (decitabine) in the design of its dose-schedule for cancer therapy. Clin Epigenetics. 2013;5:3. doi: 10.1186/1868-7083-5-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Ponthan F, Johnsen JI, Klevenvall L, Castro J, Kogner P. The synthetic retinoid RO 13-6307 induces neuroblastoma differentiation in vitro and inhibits neuroblastoma tumour growth in vivo. Int J Cancer. 2003;104:418–24. doi: 10.1002/ijc.10954. [DOI] [PubMed] [Google Scholar]
  • 46.Dobrotkova V, Chlapek P, Mazanek P, Sterba J, Veselska R. Traffic lights for retinoids in oncology: molecular markers of retinoid resistance and sensitivity and their use in the management of cancer differentiation therapy. BMC Cancer. 2018;18:1059. doi: 10.1186/s12885-018-4966-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Jones PA, Ohtani H, Chakravarthy A, De Carvalho DD. Epigenetic therapy in immune-oncology. Nat Rev Cancer. 2019;19:151–61. doi: 10.1038/s41568-019-0109-9. [DOI] [PubMed] [Google Scholar]
  • 48.Mckeown MR, Corces MR, Eaton ML, Fiore C, Lee E, Lopez JT, et al. Superenhancer analysis defines novel epigenomic subtypes of Non-APL AML, including an RARalpha dependency targetable by SY-1425, a potent and selective RARalpha agonist. Cancer Discov. 2017;7:1136–53. doi: 10.1158/2159-8290.CD-17-0399. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary information file (1.3MB, docx)

Data Availability Statement

The datasets used in this study are available at the Gene Expression Omnibus (GEO) database (accession no. GSE180645, GSE180660 and GSE180929, https://www.ncbi.nlm.nih.gov/geo/).

Articles from British Journal of Cancer are provided here courtesy of Cancer Research UK

RESOURCES