Decitabine induces cell differentiation and growth inhibition accompanied by upregulation of microRNA‑143 and MS4A3 in the myelodysplastic syndromes‑derived leukemic cell line MDS‑L

Introduction

Myelodysplastic syndromes (MDS) are a heterogeneous group of clonal hematopoietic malignancies characterized by cytopenia, dysplasia, ineffective hematopoiesis and an increased risk of progression to acute myeloid leukemia (AML). The annual incidence of MDS in the USA is ~4/100,000 individuals, according to the United States Surveillance, Epidemiology, and End Results database (1). A large proportion of patients with MDS are elderly, with a median age at diagnosis of ~70 years (1) and are often not eligible for stem cell transplantation because of comorbidities, advanced age and/or the lack of a suitable transplant donor (2,3). The hypomethylating agents (HMAs) azacytidine and decitabine (5-aza-2′-deoxycytidine; DAC) are used to treat high-risk MDS and AML in patients who are not eligible for allogeneic hematopoietic cell transplantation (4,5).

HMAs can inhibit DNA methyltransferase (DNMT) activity and prevent the methylation of nascent strands after DNA replication; this epigenetic modification is reversible, as HMAs do not affect de novo DNMT synthesis (6,7). DAC is a pyrimidine nucleoside analog that enters the cell via the human equilibrative nucleoside transporters, human equilibrate nucleoside transporter 1 (hENT1) and hENT2. Once inside the cell, DAC is phosphorylated by the deoxycytidine kinase to generate DAC 5′-monophoshate, which is then converted to 5-aza-2′-deoxycytidine triphosphate (5-aza-dCTP), the active metabolite of DAC. Subsequently, 5-aza-dCTP competes with and replaces cytosine in cytosine-guanosine dinucleotide (CpG) islands, inhibiting promoter methylation and thus contributing to DNMT degradation (8–13). Therefore, DAC may induce DNA demethylation and restore the expression levels of multiple tumor suppressor genes [such as BRCA1, cadherin 1 and FOXO family genes (including FOXO1 and 3A)] that were silenced by abnormal methylation (14,15). Numerous studies identify methylation patterns that predict response to DAC (16,17). A retrospective analysis of the DNMT3A mutation status demonstrates a high response rate in a small cohort (the complete response rate was 75% in 6/8 cases), but this requires further validation in larger studies (18–20). The pretreatment levels of microRNA-29b (miR29b) have been reported to be associated with the response to DAC (21,22). This is consistent as miR29b function to specifically inhibits DNMT1 expression; however, the association of miR29b levels with the response to HMAs could not be confirmed in another larger study (23). The exact mechanism of action of DAC is currently unknown, although several studies have been conducted to identify the molecular features that predict the response to DAC (24–26).

The present study investigated the mechanisms of DAC-induced demethylation and expression restoration of tumor suppressor genes via studies in the MDS-derived leukemic cell line MDS-L (27).

Materials and methods

Chemicals and reagents

DAC was purchased from Sigma-Aldrich (Merck KGaA). MDS-L cells, a blastic subline derived from bone marrow blasts isolated from a 50-year-old male Japanese patient with MDS that was established in a previous study (27), were used in the present study. The cells were cultured in RPMI 1640 (FUJIFILM Wako Pure Chemical Corporation) medium supplemented with 20% heat-inactivated fetal bovine serum (FUJIFILM Wako Pure Chemical Corporation) and maintained in a humidified atmosphere with 5% CO2 at 37°C. The cells were cultured with a non-toxic concentration (5 nM) of DAC at 37°C for 96 h and passaged for 6 months to generate the MDS-L-DAC cell line.

Cell line characteristics

A proliferation assay was performed using the trypan dye exclusion method to determine the effects of DAC on growth inhibition during 72 h of treatment. The cell suspension and 0.5 % trypan blue staining solution (Nacalai Tesque, Inc.) were mixed in a 1:3 ratio at room temperature for 5 min and the number of live cells was measured using a hemocytometer under a light microscope. The IC50 values were calculated from the growth inhibition curves generated with GraphPad Prism software (version 8.0; Dotmatics).

Cellular morphology was investigated using May-Grunwald-Giemsa staining. Using 100 µl of cell culture (2×106 cells/ml), samples were centrifuged in a cytospin at 60 × g for 2 min at 4°C. May-Grunwald's stain solution (Nakalai Tesque, Inc.) was diluted two-fold with 100% methanol and added to the specimens, which were then fixed and stained at room temperature for 5 min. Giemsa's stain solution was diluted 20-fold with M/15 phosphate buffer (pH 7.4) (Muto Pure Chemicals Co., Ltd.) added to the preparation, and stained at room temperature for 20 min. Specimens were observed using a light microscope (BX51; Oly) and images were captured using cellSens Standard software (version 1.16; Evident Scientific).

Cell surface markers were detected by SRL, Inc. (H.U. Group Holdings, Inc.). Cells (5×105 cells) were collected and aliquoted for flow cytometric analysis and were incubated with fluorochrome-conjugated monoclonal antibodies against the indicated surface makers for 25 min at room temperature in the dark. Following antibody staining, PBS was added to the samples, which were then gently mixed and incubated for 7 min at room temperature. After incubation, cells were washed using centrifugation at 4°C at 620 × g for 5 min, and the supernatant was discarded. The cell pellets were then resuspended in cold PBS and analyzed using fluorescence-activated cell sorting FACSCanto™ II flow cytometry (BD Biosciences). Data acquisition was carried out using FACSDiva™ software (version 6.1.3; BD Biosciences) and leukocyte populations were identified based on the expression of CD45. The expression of surface markers was identified using the following antibodies: CD11b (cat. no. 347557; 1:20), CD15 (cat. no. 567961; 1:20), CD13 (cat. no. 565124; 1:20), CD19 (cat. no. 562441; 1:20), CD33 (cat. no. 572166; 1:20), glycophorin A (expressed in erythrocyte progenitor cells and erythrocytes) (cat. no. 561775; 1:20), CD7 (cat. no. 659124; 1:20), CD34 (cat. no. 659123; 1:20), CD10 (cat. no. 563508; 1:20) and HLA-DR (cat. no. 569674; 1:20) all obtained from BD Biosciences antigen was evaluated. The activity of DNMT was measured using a DNMT assay kit according to the manufacturer's protocol (Active Motif, Inc.).

DNA methylation array analysis

DNA was extracted from cells using a QIAamp DNA Mini kit (Qiagen Sciences, Inc.). All DNA samples were quantified using the fluorometric method and assessed for purity with a NanoDrop2000c (Thermo Fisher Scientific, Inc.). DNA methylation array analysis was performed using the Infinium Human Methylation 450 BeadChip (Illumina, Inc.) and included evaluation of DNA methylation at 485,576 CpG sites. The BeadChips were scanned using an Illumina HiScan SQ scanner. The intensities of the images were extracted using GenomeStudio (version 2010.3) and Methylation module (version 1.8.5) software (Illumina, Inc.). Hierarchical clustering of differentially methylated genomic regions was performed using RnBeads (version 2.14.0; http://rnbeads.org/) for MDS-L and MDS-L-DAC cells using the 100 genes and promoters identified with the most significant methylation changes.

Reverse transcription (RT)-PCR

RNA was extracted from cells using a RNeasy Mini Kit (Qiagen Sciences, Inc.). Total RNA was reverse transcribed to cDNA using a PrimeScript™ RT-PCR kit (Takara Bio, Inc.) and TaqMan™ MicroRNA Reverse Transcription kit (cat. no. 4366597; Thermo Fisher Scientific, Inc.). Quantitative PCR was carried out using a TaqMan Fast Advanced Master Mix kit and Step One Plus real-time PCR systems (Applied Biosystems; Thermo Fisher Scientific, Inc.) with the following conditions: An initial 95°C for 20 sec, followed by 40 cycles of 95°C for 1 sec and 60°C for 20 sec. The transcript levels of Wilms' tumor 1 (WT1) were measured via RT-PCR with the Otsuka WT1 mRNA OneStep Assay (WT1, Hs01103751_m1; Thermo Fisher Scientific, Inc.) according to the manufacturer's protocol. The expression levels of membrane spanning 4-domains A3 (MS4A3), KRAS and miR143 were measured using TaqMan® Gene Expression Assays (Applied Biosystems™; Thermo Fisher Scientific, Inc.). The expression levels were evaluated using the 2−ΔΔCq method (28). Assay IDs used were as follows: MS4A3, ID no. Hs00960994_m1; KRAS, ID no. Hs00364284_g; GAPDH, ID no. Hs02786624_g1; hsa-miR-143, ID no. 477912_mir (miR-143 sequence, 5′-UGAGAUGAAGCACUGUAGCUC-3′); and U6 snRNA, ID no. 001973 (Applied Biosystems™; Thermo Fisher Scientific, Inc.), respectively.

Western blot analysis

Western blot analysis was performed to measure the protein expression levels of demethylation genes. After culturing the cells at 37°C for 48 h, theses were washed with PBS, and protein lysates were extracted from the cells (1×107 cells) using the Qproteome Mammalian Protein Prep kit (Qiagen, Inc.). Protein quantification was carried out using the BCA method. The 20 µg lysates were electrophoresed in a 10–20% e-PAGEL gel (ATTO Corporation) and were and transferred onto a Immun-Blot PVDF Membrane for Protein Blotting (Bio-Rad Laboratories, Inc.). Blocking was carried out using an undiluted solution of Blocking One (Nacalai Tesque, Inc.) and incubated at room temperature for 30 min. The membranes were probed with primary antibodies overnight at 4°C and the membrane was washed with TBST containing 0.001% Tween20. Subsequently, the membrane was probed with secondary antibodies for 1 h at room temperature and the membrane was washed with TBST containing 0.01% Tween20. Western lighting ECL Pro (PerkinElmer, Inc.) and FUSION-SOLO.7S.EDGE (Vilber Lourmat) were used to visualize and quantify protein signals. Mouse polyclonal anti-MS4A3 antibody (cat. no. ab68207; 1:500; Abcam) was used as a primary antibody. The primary antibodies caspase 3 (cat. no. 9662; 1:1,000), cleaved caspase 3 (cat. no. 9661; 1:1,000), caspase 8 (cat. no. 9746; 1:1,000), cleaved caspase 3 (cat. no. 9496; 1:1,000), heat shock protein 70 (Hsp70; cat. no. 4872; 1:1,000) and β-actin (cat. no. 3700 1:1,000) were from Cell Signaling Technology, Inc. The Bcl-2 antibody (cat. no. ab182858; 1:1,000) was from Abcam and heat shock protein 90 (Hsp90; cat. no. AC88; 1:1,000) was from Enzo Life Sciences, Inc. The secondary antibody used was either, Peroxidase AffiniPure Goat Anti-Rabbit IgG (cat. no. 111-035-144; 1:10,000) or Peroxidase-AffiniPure Goat Anti-Mouse IgG, F(ab')2 Fragment Specific (cat. no. 115-035-072; 1:10,000) from Jackson ImmunoResearch Laboratories, Inc. Band density was detected using a FUSION-SOLO.7S.EDGE (Vilber Lourmat). Unpaired Student's t tests were carried out using GraphPad Prism software (version 5.0; Dotmatics). One-way analysis of variance followed by Tukey's test was used to compare multiple groups.

Apoptosis rate by cell cycle

Cells (5×105 cells) were washed with PBS. The mixture was centrifuged at 4°C at 200 × g for 5 min before the supernatant was removed and ice-cold 70% methanol was added to the cell pellet and then incubated on ice for 30 min. The mixture was centrifuged again at 4°C at 400 × g for 5 min, and the supernatant was removed and the cell pellet was resuspended in PBS. Subsequently, 10 mg/ml RNnase A solution (Sigma-Aldrich; Merck KGaA) was added and incubated at 37°C for 20 min. The mixture was centrifuged at 4°C at 400 × g for 5 min, and the supernatant was removed and the cell pellet was resuspended in PBS. Subsequently, 500 µg/ml PI solution (Nacalai Tesque, Inc.) was added and the cells were incubated on ice for 20 min. Analysis was carried out using a FACSCanto™ II flow cytometer (BD Biosciences). Unstained cells and isotype controls were used to set up background fluorescence and gating. Cells were first gated using forward scatter. The main cell population was then analyzed. Data analysis was carried out using BD FACSDiva™ software (version 6.1.3; BD Biosciences).

Quantification of β-catenin positivity

The percentage of β-catenin positive cells was determined using flow cytometry and the primary anti-β-catenin antibody (cat. no. 8480; 1:50; Cell Signaling Technology, Inc.) and a secondary antibody anti-rabbit IgG (H+L), F(ab')2 Fragment (Alexa Fluor® 488 conjugate) (cat. no. 4412; 1:250; Cell Signaling Technology, Inc.). Cells (1×106 cells) were washed with PBS and after centrifugation at 200 × g for 5 min at 4°C, the supernatant was removed and fixed with 4% formaldehyde for 10 min at 37°C. The sample was centrifuged at 200 × g for 5 min at 4°C, and the fixative was removed. Then, ice-cold 90% methanol was added and incubated on ice for 30 min. The mixture was centrifuged at 200 × g and 4°C for 5 min, and the permeate was removed. The cells were washed with PBS, and the mixture was centrifuged at 200 × g for 5 min at 4°C to remove the supernatant. The anti-β-catenin antibody was added and incubated at room temperature for 1 h. The cells were washed with PBS, and the mixture was centrifuged at 200 × g for 5 min at 4°C to remove the supernatant. The secondary antibody was added and incubated at room temperature for 30 min. The cells were washed with PBS and the mixture was centrifuged at 200 × g for 5 min at 4°C to remove the supernatant. This was resuspended in PBS and was then analyzed using a FACSCanto™ II flow cytometer (BD Biosciences). Unstained cells and isotype controls were used to set up background fluorescence and gating. Cells were first gated using forward scatter. The main cell population was then analyzed. Data analysis was carried out using BD FACSDiva™ software (version 6.1.3; BD Biosciences).

Statistical analyses

All statistical analyses were performed using Microsoft Excel (Microsoft Corporation). The unpaired Student's t-test was conducted to compare the differences between two groups. One-way analysis of variance followed by Tukey's test was used to compare multiple groups. All graphs, linear regression lines and curves were generated using the GraphPad Prism software (version 8.0; Dotmatics). P<0.05 was considered to indicate a statistically significant difference.

Results

DAC induces the differentiation of MDS-L cells

The IC50 value of DAC in MDS-L cells was 64 nM (Fig. S1). As the efficacy of DAC is most notable in patients with MDS after the fourth cycle of 6-weekly dosing (6 months) in clinical practice, MDS-L cells were incubated with a 5 nM of DAC, the lowest concentration that could inhibit DNMT activity without inducing cytotoxicity, for 6 months, which generated the MDS-L-DAC cell line. Compared with that in the parental MDS-L cells, the enzymatic activity of DNMT1 in MDS-L-DAC cells was significantly inhibited by DAC treatment (Fig. 1A). There was no significant difference in the ability of DAC incorporation into cellular DNA between the MDS-L and MDS-L-DAC cells (Fig. 1B). Morphologically, the MDS-L-DAC cells were mature and differentiated, with lobulated nuclei and cytoplasmic granulation (Fig. 1C). Compared with the MDS-L cells, the MDS-L-DAC cells had a significantly reduced positivity for CD34 and MHC class II DR antigen, and an increased positivity for CD15 (Figs. 1D and S2) and partial loss of chromosomal abnormalities (Fig. S3).

Figure 1. DAC induces the differentiation of MDS-L cells. (A) The enzymatic activity of DNMT1 was measured in both MDS-L and cell lines. MDS-L 11.17±1.25; MDS-L-DAC 4.55±1.46, P=0.004, using Student's t-test. (B) Incorporation of DAC into the DNA of MDS-L cells and MDS-L-DAC cells. Data shown as mean ± SD of ≥3 independent experiments. (C) Morphological changes in MDS-L and MDS-L-DAC cells observed by May-Grunwald Giemsa staining. Upper panel (scale bar, 100 µm): 100-fold of MDS-L cells (left) and MDS-L-DAC cells (right). Lower panel (scale bar, 20 µm): 400-fold of MDS-L cells (left) and MDS-L-DAC cells (right). (D) Flow cytometric analysis of CD11b, CD13, CD15, CD33, CD34 and HLA-DR expression. **P<0.01. DNMT1, DNA methyltransferase 1; MDS, myelodysplastic syndromes; MDS-L, MDS cell line; DAC, decitabine; N.S., not significant; HLA-DR, MHC class II DR antigen.

Profiling of DNA demethylation upon DAC exposure

Comprehensive genome-wide profiling of DNA methylation in MDS-L and MDS-L-DAC cells was performed to investigate the epigenetic changes induced by DAC in MDS-L cells. Clustering analysis was performed on the 100 CpG sites, CpG islands, tiling regions (evaluated over a large region containing up to 5,000 bp), genes and promoters with the greatest differences in β values between MDS-L and MDS-L-DAC cells. DAC treatment resulted in the demethylation of 80 CpG sites and 79 tiling regions (Fig. S4). Regarding gene regions and promoter regions with methylation changes, 75 gene regions and 76 promoter regions were demethylated in MDS-L-DAC cells (Fig. 2A and B). Among these gene regions and promoter regions, excluding pseudogenes and long non-coding RNAs, 24 gene regions and 35 promoter regions were found to exhibit significant demethylation (Tables SI and SII).

Figure 2. DNA methylation array analysis. Hierarchical clustering of differentially methylated genomic regions in MDS-L and MDS-L-DAC cells on the basis of the 100 genes and promoters with the most significant methylation changes. A gradient color scale ranging between blue (completely unmethylated) to red (completely methylated) is shown. (A) Genes and (B) promoters. MDS, myelodysplastic syndromes; MDS-L, MDS cell line; DAC, decitabine.

Restoration of MS4A3 expression and induction of apoptosis in MDS-L-DAC cells

In MDS-L-DAC cells, increased expression levels of MS4A3 were observed at both the transcriptional and translational levels (Fig. 3A and B). DAC treatment induced apoptosis in 20% of the MDS-L-DAC cells, compared with only 5% of the control MDS-L cells (Figs. 3C and S5). Additionally, 20% of cells treated with 50 nM decitabine underwent apoptosis, compared with only 10% of the control cells, accompanied by markedly increased cleavage of caspase-3 and caspase-8 (Fig. 3D), compared with that of MDS-L cells, which suggested the involvement of MS4A3-mediated apoptosis (29). Expression of MS4A3 is also known to induce cell differentiation (30), and the induction of differentiation in MDS-L-DAC cells as aforementioned also suggested the restoration of MS4A3 expression.

Figure 3. Restoration of MS4A3 expression levels and induction of apoptosis in MDS-L-DAC cells. (A) Reverse transcription-quantitative PCR was performed to evaluate the expression of MS4A3. Relative expression value of MDS-L-DAC was 118±37.3 compared with the MDS-L group. (B) Western blot analysis was performed to evaluate the protein level of MS4A3. (C) Percentages of apoptotic cells treated with various concentrations of DAC. The proportions of sub-G1 group of each were 5.6±0.14 in MDS-L and 12.3±6.25 in MDS-L-DAC groups (P=0.04, Student's t-test). The proportion of apoptotic cells was 7.13±0.06 in MDS-L and 24.5±10.6 in MDS-L-DAC in the control groups, 11.0±0.86 and 21.3±5.06 when treated with 50 nM DAC, and 14.5±2.15 and 21.7±0.1 when treated with 100 nM DAC, respectively. Statistical analysis was performed using Student's t-test. (D) Western blot analysis was carried out to evaluate the protein levels of caspase 3, cleaved caspase 3, caspase 8 and cleaved caspase 8. *P<0.05, **P<0.01, ***P<0.001 and ****P<0.0001. MDS, myelodysplastic syndromes; MDS-L, MDS cell line; DAC, decitabine; MS4A3, membrane spanning 4-domains A3.

Upregulation of miR143 in MDS-L-DAC cells

The expression of miR143 significantly increased in MDS-L-DAC cells compared with that of MDS-L cells, consistent with the DNA methylation profile (Fig. 4A). miR143 expression is typically downregulated in hematopoietic malignancies (30), thus, changes in the expression levels of molecules downstream of the restored miR143 expression in MDS-L-DAC cells were examined. The expression levels of KRAS and WT1, which are components of the downstream signaling pathways of miR143 (31,32), were significantly decreased in MDS-L-DAC cells compared with that of MDS-L cells (Fig. 4B). In addition, MDS-L-DAC cells presented decreased expression levels of β-catenin, Bcl-2, Hsp70 and Hsp90 (Fig. 4C and D; Fig. S6), which suggested a decrease in leukemogenicity due to the restoration of miR143 expression.

Figure 4. Upregulation of miR143 in MDS-L-DAC cells. (A) RT-PCR was performed to evaluate the expression of miR143. Relative expression value of MDS-L-DAC was 6.96±2.67 compared with the MDS-L group. Statistical analysis was performed using Student's t-test. (B) RT-PCR was performed to evaluate the expression of KRAS, MAPK1 and WT1. Relative expression value of MDS-L-DAC was 0.33±0.14 in KRAS and 0.43±0.15 in MAPK1 compared with the MDS-L group. Expression of WT1 mRNA was 33.0±2.0 in MDS-L and 17.7±0.58 in MDS-L-DAC. Statistical analysis was performed using Student's t-test. (C) The expression of β-catenin was evaluated using flow cytometry. The proportions of β-catenin positivity were 20.5±4.79 in MDS-L and 1.64±1.90 in the MDS-L-DAC group. Statistical analysis was performed using Student's t-test. (D) Western blot analysis was performed to evaluate the protein levels of Bcl-2, Hsp70 and Hsp90. **P<0.01, ***P<0.001 and ****P<0.0001. MDS, myelodysplastic syndromes; MDS-L, MDS cell line; DAC, decitabine; Hsp, heat shock protein; RT, reverse transcription; miR, microRNA; WT, Wilms tumor 1.

Discussion

DAC is an effective HMA used for the treatment of MDS and AML (5,24). However, in contrast to other cytotoxic agents such as hydroxycarbamide and cytarabine, the mechanism underlying the demethylation effect of DAC remains unclear. Since DAC is rapidly eliminated by the enzyme cytidine deaminase, with a half-life of <20 min in vivo and 5–16 h in vitro (33), the present study explored the genes involved in the DAC response in MDS-L cells with long-term exposure to DAC. In MDS-L cells, long-term culture with DAC significantly reduced DNMT1 activity and altered the methylation profiles of multiple genes, including MS4A3 and miR143, leading to cell differentiation and gene expression. These results may suggest a mechanism for HMA-induced expression recovery and differentiation in MDS-L cells.

DNA methylation array analysis was used to evaluate the methylation status of CpG sites, CpG islands, tiling regions, promoter regions and gene regions in MDS-L and MDS-L-DAC cells, which were compared to assess methylation changes comprehensively. In MDS-L-DAC cells, demethylation of both MS4A3 and miR143 was observed, further confirming the restoration of the expression of these genes. Among them, MS4A3 is a member of the MS4 protein family, whose members possess four transmembrane domains, and are known to function as cell surface signaling receptors and intracellular adapter proteins (34). MS4A3 is expressed in specific subsets of hematopoietic cells, including myeloid precursors, basophilic granulocytes and CD34-positive hematopoietic stem and progenitor cells (30,34,35). Increased expression levels of MS4A3 in hematopoietic stem cells induces differentiation into granulocytes (30); conversely, suppression of MS4A3 in AML cells is involved in tumor aggressiveness (29). In MDS-L-DAC cells, a tendency toward differentiation and the induction of apoptosis were observed, which may involve the restoration of MS4A3 expression levels. The present study was not able to validate this observation due to the lack of other MDS cell lines for examination; thus, a limitation is that the present findings were exclusively observed in MDS-L cells. Future work is necessary to validate the restoration of MS4A3 expression levels and therapeutic effects in cells derived from patients with MDS.

MicroRNA-143 is known to serve important roles in cell proliferation, cell death and metastasis in various types of cancer, including AML, osteosarcoma and bladder cancer, and is known to function as a tumor suppressor (36–39). In the present study, a decrease in the expression levels of KRAS and WT1, as well as in the downstream signaling molecules β-catenin and Hsp70, were observed in MDS-L-DAC cells. In colon cancer cells, miR143 has been shown to bind to the 3′-untranslated region of the KRAS and inhibit KRAS expression (31). It has also been shown that TGF-β1-mediated induction of miR143 reduces WT1 expression levels in human podocytes (32). Through these mechanisms, the restoration of miR143 expression levels in MDS-L-DAC cells may have reduced WT1 expression levels. Furthermore, by evaluating molecules downstream of miR143 and WT1 (40), Hsp70 and β-catenin were downregulated in MDS-L-DAC cells, compared with that of MDS-L cells. The expression levels of Bcl-2 and Hsp90, key molecules of WT1 signaling were evaluated, and it was found Bcl-2 and Hsp90 expression levels were also decreased in MDS-L-DAC cells. Upregulation of Hsp70 and β-catenin is associated with a poor therapeutic response and adverse prognosis in AML (41,42), and therefore, decreased expression of these predictors of adverse outcomes in MDS-L-DAC cells potentially indicated that hsp70 and β-catenin may be future therapeutic targets via miR143. In addition to these changes in gene expression, it is suggested that exposure to DAC may cause the demethylation of genes, such as CDNK2B and TP73 (43,44), which may be involved in cell differentiation. Further evaluation using patient samples derived from MDS may clarify this point in the future.

In conclusion, the expression of MS4A3 and miR143 was epigenetically silenced in the MDS-L cell line, and long-term culture of the MDS-L cell line with DAC resulted in the restoration of MS4A3 and miR143 expression. As a result, the restoration of MS4A3 expression, along with suppression of the expression of KRAS, WT1 and Wnt/β-catenin pathway components downstream of miR143, induced cell differentiation and apoptosis was induced. The changes in methylation profiles upon long-term DAC treatment suggest that the MS4A3 and miR143 genes and their downstream signaling pathways may be potential therapeutic targets in AML in the future.

Supplementary Material

Supporting Data

Supporting Data

Acknowledgements

The authors would like to thank Mr. Daisuke Minegishi of Tohoku Chemical Co., Ltd., (Iwate, Japan) for his guidance in DNA methylation array analysis.

Funding

The present work was supported in part by a Grant-in-Aid for Scientific Research from the Ministry of Education, Culture, KAKENHI (grant nos. 23501307 and 19K07637), the NAITO foundation (grant no. 20-007) and a grant from Japan Gout Research Foundation (2008, 2009 and 2010).

Availability of data and materials

The data generated in the present study may be found in the National Center for Biotechnology Information Gene Expression Omnibus database under the accession no. GSE293053 or at the following URL: https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE293053.

Authors' contributions

NH and TY designed and conducted the present study. RN carried out the high-throughput sequencing experiments and performed the examination of flow cytometry, PCR and western blotting. NH, RN and TY analyzed the experimental results and considered and conducted necessary additional experiments. RN and NH contributed to manuscript writing. KT generated the MDS-L cell lines. NH and TY confirm the authenticity of all the raw data. All authors read and approved the final version of the manuscript.

Ethics approval and consent to participate

Not applicable.

Patient consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

References